REVIEW ARTICLE

Mitochondrial Ca2+ concentrations in live cells:

quantification methods and discrepancies
Celia Fernandez-Sanz, Sergio De la Fuente and Shey-Shing Sheu
Center for Translational Medicine, Department of Medicine, Thomas Jefferson University, Philadelphia, PA, USA

Correspondence Intracellular Ca2+ signaling controls numerous cellular functions. Mitochon-

S. De la Fuente or S.-S. Sheu, Department dria respond to cytosolic Ca2+ changes by adapting mitochondrial functions
of Medicine, Center for Translational
and, in some cell types, shaping the spatiotemporal properties of the cytosolic
Medicine, Thomas Jefferson University,
Ca2+ signal. Numerous methods have been developed to specifically and quan-
1020 Locust Street, Philadelphia, PA 19107,
USA titatively measure the mitochondrial-free Ca2+ concentrations ([Ca2+]m), but
Tel: 215 503 0445/215 503 5152 there are still significant discrepancies in the calculated absolute values of
E-mails: Sergiodelafuente.perez@ [Ca2+]m in stimulated live cells. These discrepancies may be due to the distinct
jefferson.edu (SDF); or properties of the methods used to measure [Ca2+]m, the calcium-free/bound
shey-shing.sheu@jefferson.edu (SSS) ratio, and the cell-type and stimulus-dependent Ca2+ dynamics. Critical pro-
cesses happening in the mitochondria, such as ATP generation, ROS home-
(Received 1 April 2019, revised 29 April
ostasis, and mitochondrial permeability transition opening, depend directly on
2019, accepted 2 May 2019)
the [Ca2+]m values. Thus, precise determination of absolute [Ca2+]m values is
doi:10.1002/1873-3468.13427 imperative for understanding Ca2+ signaling. This review summarizes the
reported calibrated [Ca2+]m values in many cell types and discusses the dis-
Edited by Barry Halliwell crepancies among these values. Areas for future research are also proposed.

Keywords: fluorescent Ca2+ indicators; genetically encoded Ca2+

indicators; live cells; mitochondrial Ca2+ concentrations

A general cellular response to extracellular stimulus is ryanodine receptor (RyR1) or by Ca2+- or IP3-induced
the rise in the cytosolic Ca2+ concentration ([Ca2+]c) Ca2+ release, depending on the cell type and tissue.
leading to intracellular Ca2+ signaling which plays a These pathways elevate the resting [Ca2+]c concentra-
key role in multiple cellular processes, including gene tion from ~100 nM to a level that varies vastly from
expression, differentiation, bioenergetics regulation, report to report (Table 1). Unlike other cellular second
contraction, exocytosis, neurotransmission, and cell messengers, the Ca2+ can neither be created nor
death [1]. The multiple pathway can lead to this destroyed; Ca2+ can only be transported, released, or
[Ca2+]c increase, including the Ca2+ entry from the bound. Due to the importance of the [Ca2+]c increases
extracellular medium though the plasma membrane in controlling cell functions, levels of [Ca2+]c need to
calcium channels and Ca2+ release from the intracellu- be tightly regulated. During a Ca2+ transient, mito-
lar Ca2+ stores like the endoplasmic/sarcoplasmic retic- chondria can uptake Ca2+ from the cytosol to modu-
ulum (ER/SR). The release of Ca2+ from the ER/SR late the spatiotemporal profiles of [Ca2+]c mainly via
can be mediated by depolarization–initiated mechani- the mitochondrial calcium uniporter complex (MCUC)
cal coupling of L-type Ca channels with the type 1 or through other alternative pathways [2,3]. However,

Abbreviations
AM, acetoxymethyl; CaM, calmodulin; CyP, cyclophilin; ER, endoplasmic reticulum; FP, fluorescent proteins; FRET, fostering resonance
energy transfer; GECI, genetically encoded Ca2+ indicators; GFP, green fluorescent protein; MCUC, mitochondrial calcium uniporter complex;
mPTP, mitochondrial permeability transition pore; OGDH, 2-oxoglutarate dehydrogenase; PDH, pyruvate dehydrogenase; PDHP, pyruvate
dehydrogenase phosphatase; PTP, permeability transition pore; SR, sarcoplasmic reticulum.

FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies 1

Mitochondrial Ca2+ concentrations in live cells C. Fernandez-Sanz et al.

the increase in the mitochondrial Ca2+-free concentra- inside the mitochondrial matrix upon stimulation in
tion ([Ca2+]m) is involved not only in controlling the live cells. It is possible that mitochondria from differ-
cytosolic Ca2+ signal but also in regulating mitochon- ent cell types may reach significantly different [Ca2+]m,
drial functions, such as energy metabolism [4]. Cellular due to their own biochemical and physical properties
activities are expected to change upon stimulation, a (including localization and/or relation with other cellu-
process that demands energy. Mitochondria utilize the lar organelles). However, under a given identical cir-
increase in [Ca2+]m to enhance ATP production and cumstance, different methods applied should report
subsequently the energy generation [5]. and confirm the same [Ca2+]m in the same cell type. In
Because of its physiological and pathological rele- this review, we summarize the importance of knowing
vance, mitochondrial Ca2+ dynamics have been studied absolute values of [Ca2+]m and most used methods to
intensively in the last decades. Methods to measure achieve the measurements; and we compile most of the
changes in [Ca2+]m have substantially evolved in recent calibrated [Ca2+]m values reported in the literature. We
years. These methods are essentially classified into two also point out the discrepancies between these values
major groups: fluorescent dyes and genetically encoded and what may underlie these differences. Areas for
Ca2+ indicators (GECIs). GECIs can be subdivided future research are also proposed.
into fluorescent or bioluminescent probes. Ideally, all
of these methods are suitable to accurately report
Mitochondrial Ca2+ and bioenergetics
[Ca2+]m values, independent of the cell type or the
stimulus, however each has its own advantages and Mitochondria are the main ATP producers through
disadvantages, and different methods are preferred to oxidative phosphorylation in eukaryotic cells. The
measure [Ca2+]m depending on the cell type, stimulus, reactions of the Krebs cycle and the utilization of the
or even on the equipment used to perform the mea- resulting reducing power by oxidative phosphorylation
surements. So, despite the advances in the number of generates up to 32–36 ATP molecules, whereas gly-
tools available for quantifying [Ca2+]m, discrepancies colytic metabolism generates only 2. These reactions
persist in the absolute values that can be reached are regulated by [Ca2+]m in a mechanism designed to

Table 1. Reported mitochondrial calcium concentrations.

Probe Kd (lM) Cell type Stimulus [Ca2+]m reported References

Rhod-2 0.57 HeLa Histamine 3.5 lM [93]

Thapsigargin 900 nM
CCE 1.6 lM
Glomerulosa cells Extracellular Ca2+ 750 nM Equilibrated with Cyto [94]
Fura-FF 5.50 RBL Ip3 10–20 lM [65,113]
Rat myocytes Electrical stimulation 10 nM [96]
Rhod-5N 320 HeLa Histamine 20–80 lM [95]
Yc2 1.26 HeLa Histamine 3.2–10 lM [98]
Yc3 3.98 HeLa Histamine 49.5 lM
Yc4 104 HeLa Histamine 106 lM
Mitycam 0.2 Rabbit cardiomyocytes Electrical Stimulation 40 nM [99]
4mtD3cpv 0.47 Rat cardiomyocytes Electrical Stimulation 0.3–0.9 lM simulated [100]
Extracellular Ca2+ 0.3–2.4 lM simulated
AEQ variants 260-1000 Hela Histamine 20–150 lM [66,69,114–116]
Adipocytes Bradykinin 2 lM [117]
TNBC Cells ATP 80 lM [118]
Human Fibroblast Histamine 45 lM [119]
Cromafin K+ 200–600 lM [101]
Cardiomyocytes Angiotensin 1 lM [120]
Pleura mesothelioma Bradykinin 50 lM [121]
SH-SY5Y neuroblastoma Bradykinin 50 lM [116]
N2a neuroblastoma Bradykinin 85 lM [122]
Smooth muscle cells 0 Na+ 3.5 lM [123]
ATP 2 lM
STH Huntington estriatial (virus) Extracellular Ca2+ 60–450 lM [124]
SOCE 25 lM

2 FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies

C. Fernandez-Sanz et al. Mitochondrial Ca2+ concentrations in live cells

ensure that the ATP synthesis is coupled with the shown that high ADP/ATP ratios decrease not only
energy demand of the cell [6]. the Km of the OGDH complex for the substrate, but
Due to the great capacity of mitochondria to uptake also the K0.5 for the Ca2+ [21]. It seems reasonable to
Ca2+, it was thought that these organelles were essential think that under sudden high workload conditions, the
reservoirs of mobilizable intracellular Ca2+ ions [7]. high ADP/ATP mitochondrial ratios and [Ca2+]m
Nevertheless, several independent studies from the increases over the same time scale [22,23] stimulate a
1980s [8–10] showed that total [Ca2+]m was low under reduced power supply for oxidative phosphorylation
basal conditions, but increases in [Ca2+]c in response to by increasing the activity of the OGDH.
extrinsic agents were mirrored by increases of [Ca2+]m
[11,12]. The higher [Ca2+]m in response to increases of
NAD+-isocitrate dehydrogenase (NAD-ICDH)
[Ca2+]c would lead to the activation of oxidative meta-
bolism and result in an increased supply of reducing NAD-ICDH is a hetero-octamer of 29 (2a,b,c) sub-
equivalents to drive respiratory chain activity and ATP units which catalyzed the oxidation of isocitrate to 2-
synthesis. Export of mitochondrial ATP in exchange for ketoglutarate and CO2 [24,25]. This enzyme, located
ADP will be anticipated to meet the higher ATP exclusively in the mitochondria in mammalian cells, is
demand to fuel energy-requiring processes in the cyto- a key regulatory enzyme in the TCA cycle. Calcium is
sol, such as ion pumping and contraction [13]. one of the main cofactors of the NAD-ICDH. Impor-
tantly, the presence of ADP, isocitrate, and Mg2+ are
required for Ca2+ to bind and regulate NAD-ICDH
Mitochondrial enzymes modulated by [Ca2+]m activity. The stoichiometry of Ca2+ binding to the
Pyruvate dehydrogenase phosphatase (PDHP) enzyme is  1 Ca2+ per tetramer (2a,b,c) [11]. After
binding to Ca2+ in the presence of ADP, the NAD-
The pyruvate dehydrogenase (PDH) is a 50 MDa multi- ICDH Km decreased  8-fold in mitochondrial
enzyme complex that catalyzes the irreversible reaction extracts with a K0.5 for Ca2+ of 5.4 lM [21,26,27]. In
of pyruvate, CoA, and NAD to obtain acetyl-CoA, the same direction, it has been observed that the K0.5
NADH2, and CO2. The product, acetyl-CoA then enters for Ca2+ decreased from 43 to 5 lM by increasing
into the citrate cycle or fatty acid synthesis. The regula- ADP/ATP ratios, in toluene-permeabilized mitochon-
tion of the PDH complex activity is achieved via end- dria [21]. The higher K0.5 of the NAD-ICDH com-
product inhibition and reversible inhibitory phosphory- pared to the other intramitochondrial dehydrogenases
lation by a PDH kinase [14,15]. A Mg2+/Ca2+-dependent (PDHP and OGDH) together with the thigh regulation
PDH phosphatase, in turn, dephosphorylates and reac- of its activity depending on the ADP/ATP ratio might
tivates PDH. It has been described that Ca2+ activates reflect here again how [Ca2+]m mediates the fine-tuning
the PDH phosphatase in heart mitochondrial extracts of the oxidative metabolism on these organelles in
with K0.5 of 0.7 lM [16,17]. Therefore, [Ca2+]m regulates response to the different energy demands [12].
the activity of the PDH complex, a rate-limiting step for
pyruvate oxidation to acetyl CoA that in turn is
reflected in the rate of ATP synthesis in mammalian F1–FO ATP synthase
tissues [6,18]. This molecular motor couples the ATP turnover with
H+ translocation through the inner mitochondrial
membrane. Therefore, the F1–FO ATP synthase utilizes
2-Oxoglutarate dehydrogenase (OGDH)
the electrochemical proton gradient generated by the
The OGDH is a sizeable multi–subunit complex, which mitochondrial electron transport chain to synthesize
is similar to the PDH complex, consists of multiple ATP from ADP+Pi. In tissues with high energy
copies of three different subunits (E1/E2/E3). The requirements like the myocardium, ATP synthesis has
complex catalyzes the multiple step reaction that trans- to match ATP consumption according to energy
forms the 2-oxoglutarate to succinyl-CoA, Co2 and demands. It has been demonstrated by others that the
NADH [19]. From the many cofactors involved in the regulation of the ATP production rate by the F1–FO
regulation of the OGDH complex, it has been found ATP synthase might be independent from changes in
that Ca2+ and adenine nucleotides are important mod- mitochondrial membrane potential (Dwm) or rate of O2
ulators that directly bind to the complex, most likely consumption [28,29], being the enzyme the direct
to the E1 subunit [20] with a described K0.5 of potential target of this regulation. The F1–F0 ATP
 1 lM in mitochondrial extracts and of  0.2 lM in synthase activity regulation by Ca2+ fluxes might
permeabilized mitochondria. Additionally, it has been explain the stabilizing of ATP production during

FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies 3

Mitochondrial Ca2+ concentrations in live cells C. Fernandez-Sanz et al.

workload that Ca2+ dehydrogenase regulation alone [Ca2+]m measurement methods

cannot explain. In this concern, Ca2+ has emerged as a
key regulator of the ATP production. It has been As reported above, [Ca2+]m handling is essential in cell
reported that Ca2+ directly binds to F1–FO ATP syn- life. [Ca2+]m is not only essential for mitochondrial bio-
thase, specifically to the F1–FO ATP synthase b sub- genetics and energy production but also crucial for
unit [30–32] with K0.5 for Ca2+ in the nM range [29]. buffering and shaping [Ca2+]c transients. Additionally,
Studies on isolated cardiac myocytes from rats have [Ca2+]m is directly involved in preventing or triggering
investigated [ATP] and [Ca2+] from both mitochondria cell death both by necrosis or apoptosis [56]. There-
and cytosol under different workload conditions. fore, it is important to accurately know the [Ca2+]m
These measurements were performed by adenoviral dynamics to fully understand all these physiological
infection on isolate cardiomyocytes with luciferase and pathological processes. Several methods are cur-
([ATP]) and aequorin ([Ca2+]) specifically targeted to rently available to measure [Ca2+], from fluorescent
each cellular compartment [22]. Upon different work- dyes and proteins to bioluminescent sensors and
load conditions, no changes in the free [ATP] were radioisotopes of Ca2+. Those methods that are most
observed in neither of the studied compartments. frequently used to monitor mitochondrial Ca2+ dynam-
Interestingly, when cardiomyocytes were subjected to a ics in live cells and their pros and cons will be
sudden workload after a resting period, an initial drop explained further in detail.
in [ATP]m was observed before reaching equilibrium
again. This drop in [ATP]m was accompanied with a Fluorescent dyes
simultaneous increase in [Ca2+]m indicating an ATP
synthesis regulation mediated by [Ca2+]m variations in A number of fluorescent dyes to measure intracellular
response to changes in myocardial workload [22]. [Ca2+] are currently available. All of them are based
on the first fluorescent Ca2+ probe, Quin-2, developed
back in the 1980s by Roger Tsien as a modification of
The mitochondrial permeability transition pore (mPTP) the Ca2+ chelator EGTA [57]. A diverse group of Ca2+
Mitochondrial PTP is a nonspecific channel that is fluorescent indicators has been developed since then,
known to allow the diffusion of molecules up to covering multiple combinations of spectral (single/dou-
1500 Da, but its molecular nature remains elusive [33– ble-wavelength or UV/visible excitation) and chemical
45]. The mPTP has important physiological [46,47] (Ca2+ affinity, Kd) properties [58]. Some of these dyes
and pathological roles [48,49]. Transient openings of are ratiometric indicators, like the Fura-2 and Indo-1
the mPTP could participate in ROS generation families. The Oregon Green family provides a resting
[46,50,51] and Ca2+ [52] handling. It has been postu- signal while others like the Fluo family have no resting
lated that the mPTP itself might also influence signal. Finally, the Rhod family is suitable to be used
[Ca2+]m. Mathematical models suggest that under high in the long wavelength range [59]. These dyes work
intracellular calcium cycling during periods of stress, with the principle of changing the fluorescence inten-
the mitochondrial calcium efflux pathway mediated by sity according to changes in [Ca2+]. Most of them can
the mNCX may be unable to prevent calcium over- be easily loaded into the cytosol. The noninvasive ace-
load, suggesting possible different mechanisms of toxymethyl (AM) ester loading technique is the most
[Ca2+]m extrusion [52,53]. The observation of the mito- popular to achieve this aim [60]. The acetoxymethyl
chondrial calcium efflux inhibition in rat and mouse ester group lends hydrophobic properties to the fluo-
cardiomyocytes treated with mPTP blockers (cy- rescent dyes, allowing them to cross the cytosolic and
closporine A) or lacking cyclophilin D (CyP-D) [54,55] mitochondrial membranes. To further promote mito-
supported the possible role of mPTP on the mitochon- chondrial accumulation, a positively charged molecule
drial calcium extrusion process. Furthermore, an ele- is then attached to the Ca2+ indicators (e.g., Rhod-2).
vated [Ca2+]m has been found to be related to a shift The cationic nature of these dyes results in Dwm-driven
in substrate utilization from fatty acid oxidation to uptake into the mitochondrial matrix. Once inside the
glycolysis in the working heart, suggesting that the matrix, mitochondrial esterases hydrolyze the AM
mPTP might constitute a control point that links mito- groups, releasing the acids groups, retaining the probe
chondrial metabolism with myocardial workload in the mitochondrial matrix and conferring it the capa-
through changes in the [Ca2+]m [53]. According to bility to respond to changes in the [Ca2+]m. The use of
these observations, mPTP can modulate and sense fluorescent dyes as Ca2+ sensors has many advantages.
changes in [Ca2+]m, acting as a mediator of cell energy All of them are commercially available, and they are
demand adaptation or triggering cell death. easy and fast to load in any type of cells, from

4 FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies

C. Fernandez-Sanz et al. Mitochondrial Ca2+ concentrations in live cells

established cell lines to primary cultures. The loading need to be expressed by the cell’s own machinery,
conditions must be optimized for each fluorescent dye, meaning that the plasmid encoding for the probe needs
enabling high spatiotemporal resolution for imaging to be delivered within the cell. Many cell lines do not
experiments. Most of them show a high Ca2+ affinity present difficulties with transfection protocols, however
that goes from 0.1 to 1 lM and a few show Ca2+ affin- others require different techniques, such as infection or
ity at a low micromolar range (< 30 lM) and excep- electroporation. After the probe internalization, at
tionally at a micromolar range of > 30 lM. Despite of least 24–48 h of probe expression is required prior to
their versatility and easy handling, they also have the experimental procedure. The waiting time for the
strong disadvantages. Even with the AM technique, expression of the probe is not a problem in the cell
they are difficult to target and confine in the mito- line, nevertheless, it becomes critical when working
chondrial matrix. Incomplete AM ester hydrolysis, with primary cells. Some of the primary cultures, like
extracellular AM ester hydrolysis, and leakage are cardiomyocytes, are difficult to maintain without
common problems related to these fluorescence dyes. major structural changes for the time frame required
The most concerning issue is that the dye, still local- for the probe’s expression. Alternative approaches
ized in the cytosol, causes a strong signal contamina- exist to achieve expression of GECIs in primary cul-
tion. Some strategies can be used to optimize the tures. Especially in rodent models, adenovirus or
mitochondrial compartmentalization. The cold load- adeno-associated virus can be injected into the live ani-
ing–warm incubation protocol [61] can help to reduce mal to make the primary cell culture express the probe
the artifactual cytosolic signal, and so can manganese by the time the cell is isolated. Another disadvantage
treatment [62], cobalt treatment [63], and membrane of GECIs is that, unlike with fluorescent dyes, the
permeabilization [64]. Despite the number of fluores- probe may be not present in every cell, a factor that
cent dyes, Fura2-FF AM and Rhod-2 AM are the can limit single cell imaging experiments. Two major
default selections to measure [Ca2+]m. The Fura2-FF is types of GECIs have been developed, one type based
a ratiometric dye excited at 340/380 nm, with an emis- on bioluminescent proteins and another type based on
sion peak at 510 nm. Thanks to its ratiometric condi- fluorescent proteins. Further individual characteristics
tion, artifacts created during the acquisition can be and functional details will be provided in the following
avoided so that more reliable Ca2+ measurements can sections.
be obtained. According to the manufacturer, the Kd
of Fura2-FF is 5 lM, however, in vivo studies deter-
Bioluminescent based GECIs
mined the Kd is approximately 4 lM [65]. The Rhod-2
has a very high Ca2+ affinity (Kd = 570 nM), but some The aequorin was the first organelle-targeted engi-
of his family members have lower affinity like Rhod- neered Ca2+ indicator [67]. It was used for the first
FF Kd = 19 lM or the extreme low Ca2+ affinity time during the 1960s and 1970s [68], and it is the
Rhod-5N with Kd = 350 lM. Previous studies have most widely used bioluminescent Ca2+ probe.
reported that Rhod family members are toxic for the Aequorin emits light in a Ca2+-dependent reaction
cells when loaded at higher concentrations than 2 lM where the cofactor coelenterazine is oxidized, releasing
and suggest that Rhod-2 is unable to respond to two CO2 and a single photon (470 nm) per aequorin mole-
or more [Ca2+]m repetitive increases [66]. cule. Using a calibration curve, the amount of light
emitted can be easily transferred to [Ca2+]. The native
aequorin possesses 3 Ca2+ binding sites and a high
Genetically encoded calcium indicators (GECIs)
Ca2+ affinity. This native version can only measure
Unlike the fluorescent dyes, GECIs are protein-like with reliability [Ca2+] between 0.1 and 5 lM. Single
probes that can be easily targeted to different orga- point mutations were introduced in the first and sec-
nelles by including a target sequence before the probe ond Ca2+-binding sites, reducing its Ca2+ affinity by 10
sequence. Another general advantage of the GECIs is times (disabling only the second-binding site) or by
that the sequence of their Ca2+-bound domain can be 100 times (disabling both the first and second Ca2+-
modified to reduce or adjust the probe affinity for binding sites) [69]. The three aequorin variants
Ca2+. This feature has allowed the generation of (wtAEQ, mutAEQ, and 2mutAEQ) in combination
GECIs that cover a wide range of [Ca2+], from the low with different types of coelenterazine allow the mea-
nM to high mM, making them suitable to perform surement of [Ca2+] from 0.1 lM to > 1 mM. Because of
[Ca2+] measurements in every cell organelle, including the bioluminescent nature of the aequorin, excitation
the largest intracellular Ca2+ store, the ER/SR. The illumination is not required; therefore, there are no
most significant disadvantage of GECIs is that they autofluorescence or phototoxicity artefacts. However,

FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies 5

Mitochondrial Ca2+ concentrations in live cells C. Fernandez-Sanz et al.

the bioluminescent enzymatic reaction is irreversible in development of single-fluorescence Ca2+ sensors (Peri-
the experimental period, so the aequorin is “con- cams, Camgaroos, GCaMPs, Cepias, and GECOs), the
sumed”. This “consumption” was a problem when the implementation of circulated permuted FP to enhance
aequorin was exposed to high [Ca2+], because it could FRET-based indicators, and the remodeling the CaM-
lead to miscalculation of the real [Ca2+] values, how- M13 interface (Dcpv family) or the replacing of the
ever this problem was solved with the low Ca2+ affinity CaM as a Ca2+ sensor for troponin C (TN-indicators
version of aequorin. The main disadvantage to family) [81–87]. Most of the GECIs use the CaM-M13
aequorin is that the amount of light emitted is extre- as Ca2+ binding site, but two fluorescent GECIs, the
mely low, and the experiments should be performed GFP-aequorin (GA) and the GFP and apo-aequorin
most of the times in a cell population. On the other (GAP), bind the Ca2+ to the aequorin [88,89]. The
hand, working with cell populations eliminates the probes here work as a BRET system, where the pho-
cell-to-cell variability. Measurements of [Ca2+]m can be tons emitted by aequorin are transferred to the GFP,
performed in individual cells expressing aequorin, and as in jellyfish. Due to the massive list of fluorescence
although the spatial resolution is very poor, the results GECIs, detailed features of most of them have already
are reliable due to its precise intracellular location [70]. been compiled in recent reviews [75,90–92].
The three aequorin variants mentioned above have Fluorescent GECIs have some significant advantages
been targeted to mitochondria, although the most fre- over bioluminescent. The Ca2+ dynamics can be visual-
quently used is the mutAEQ in combination with the ized with high resolution at the subcellular level due to
coelenterazine n. All the mitochondrial-targeted their high fluorescence emission compared to the extre-
aequorins have been calibrated and report absolute mely low emission of photons by the bioluminescent
[Ca2+]m values. Although aequorin is the most com- probes. Another benefit over the bioluminescent
monly used bioluminescent Ca2+ probe, other chemilu- probes is that the fluorescent reaction is reversible and
minescent Ca2+ sensors are worth mentioning, does not require any cofactors. Despite recent
including obelin [71], mitrocomin, photina [72], and improvements in fluorescent GECIs, the Ca2+ affinity
clytin [73]. The recent advances in genetic engineering in most of them is relatively high. Only a few display a
allowed the generation of other bioluminescent Ca2+ Kd in the 1–10 lM range and even fewer > 10 lM.
indicators, such as the RLuc8-based indicator BRAC Many fluorescent GECIs from diverse families have
[73], and the Split RLuc8-based Nano-Lantern [74]. been targeted to mitochondria and found to report
Further details about these bioluminescent Ca2+ indi- reliable [Ca2+]m dynamics. Nevertheless, most of them
cators were described in a review [75]. have not been calibrated to absolute [Ca2+]m values.

Fluorescent-based GECIs Reported [Ca2+]m values with different

measurement methods
Fluorescent-based GECIs emerged from three critical
achievements: the discovery of the green fluorescent Despite the abundance of tools to measure the
protein (GFP) [76]; the development of color variants [Ca2+]m, few researchers have calibrated the probes, so
suitable for fostering resonance energy transfer records reveal only qualitative changes in the [Ca2+]m.
(FRET) between two fluorescent proteins [77]; and the So while absolute [Ca2+]m values are critical to our
finding that Ca2+-binding to the calmodulin (CaM) understanding of Ca2+-mediated regulatory mecha-
fused with the M13 peptide derived from the myosin nisms of mitochondrial activities, ironically, they can
light chain kinase [78]. The first developed FRET- barely be found in the published literature. The follow-
based GECIs were FIP-CBSM and Cameleon. FIP- ing paragraphs present the reported [Ca2+]m values
CBSM is characterized by the localization of the M13 found, the method used to measure [Ca2+]m, the cell
peptide between BGFP and FGFP. Cameleon pos- type in which the experiments were performed, and the
sesses yellow and cyan fluorescent proteins (FP) fused physiological stimulus applied to achieve the [Ca2+]m
directly to the CaM and M13 peptide [79,80]. In this rise. In most of these experiments, intact cells were
type of ratiometric GECI, Ca2+ binding causes a con- used. However, some studies in permeabilized cells are
formational change in the probe that alters the fluores- also mentioned. Although there are numerous reports
cence intensity of both the donor and the acceptor and on absolute values of [Ca2+]m in isolated mitochondria,
therefore the FRET efficiency. these results are not discussed in the present review.
Since their initial development, GECIs have been In 2001, Collings et al. [93] performed experiments
evolved in multiple directions to improve their detec- in intact HeLa cells using 1 lM Rhod-2 as a probe.
tion efficacy. These improvements include the Their work reported a maximum [Ca2+]m of 3.5 lM

6 FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies

C. Fernandez-Sanz et al. Mitochondrial Ca2+ concentrations in live cells

upon histamine (100 lM) stimulation. The maximum calibration curve, relating known [Ca2+] with normal-
[Ca2+]m reached was also measured upon ER Ca2+ leak ized Rhod-5N fluorescence intensity. The calibration
(using the addition of thapsigargin to inhibit the was done in the presence of FCCP, and the minimum
SERCA pump) and during the capacitative Ca2+ entry. fluorescent value was taken with 0.5 mM EGTA pre-
The values obtained were 0.4 and 1 lM, respectively. sent and the maximum fluorescence with [Ca2+]
These authors showed that even with prolonged stimu- 10 mM. The Kd reported in vivo for Rhod-5N in the
lation (20 min) with histamine, [Ca2+]m barely reached described conditions was 470 lM.
3 lM. It is worth mentioning that the mitochondrial In 2009, Andrienko et al. [96] used the high Ca2+
Rhod-2 fluorescent did not decrease significantly after affinity dye Fura-2 (10 lM) and Rhod-2 (concentration
2 min following the maximum mitochondrial Ca2+ not specified) to measure [Ca2+]m in permeabilized
peak. The Rhod-2 signal was reduced to resting levels adult rat cardiomyocytes. Spontaneous SR Ca2+
25 min after the mitochondrial peak, pointing to the release events occurring at regular frequency were used
possibility that Rhod-2 cannot follow the mitochon- as mitochondrial Ca2+ uptake stimulus. According to
drial Ca2+ dynamics properly. A mitochondrial calibra- the calibrated Fura-2 signals, [Ca2+]m increased to
tion curve is also reported in the manuscript, showing 110 nM after 8 waves and to 216 nM after 20 waves.
a Kd of 1.3 lM for the Rhod-2. To obtain the mini- The calibrated Rhod-2 signals were inferred from the
mum Rhod-2 fluorescence, 4 mM of EGTA was obtained Fura-2 data. Therefore, the authors con-
applied and a 10 lM Ca2+ pulse was used to reach the cluded that the largest [Ca2+]m rise during SR release
maximum Rhod-2 fluorescence. was 10 nM, with the average rise in the 2–4 nM range.
In 2002, Pitter et al. [94] also used Rhod-2 (2 lM) to The Fura-2 calibration was done using Ca2+-free solu-
measure [Ca2+]m in permeabilized glomerulosa cells tion and Ca2+ 50–100 lM for minimum and maximum
and in an INS-1/EK-3 cell line. The authors increased fluorescence, respectively; then the Grynkiewicz cali-
the [Ca2+]c progressively from 60 to 740 nM to deter- bration equation was used (no Kd was reported).
mine mitochondrial Ca2+ uptake. The calibrated In 2001, Csordas & Hajnoczky [65] utilized RBL-
[Ca2+]m values were always similar to or even smaller 2H3 mucosal mast cells loaded with Fura-FF (5 lM)
than the [Ca2+]c added, despite a large electrochemical to measure mitochondrial Ca2+ uptake evoked by IP3-
gradient favoring mitochondrial Ca2+ accumulation. mediated ER Ca2+ release. The Kd obtained by a Ca2+
This work used the calibration equation [Ca2+] = Kd calibration kit (Calcium calibration kit2&3, molecular
(F Fmin)/(Fmax F) [58]. The basal fluorescence emis- probes) was 4 lM, a similar value reported by others
sion of the buffer with no Ca2+ added (no Ca2+ chela- [97]. The imaging studies performed showed that in
tor present either) was considered as the minimum these conditions, [Ca2+]m reached 10–20 lM.
fluorescence. To obtain the maximum fluorescence, In the same year, Arnaudeau et al.[98] used three
500 lM Ca2+ was added to the buffer (in the presence different variants of the fluorescent GECI Camaleon
of ionomycin and the uncoupler FCCP). A Kd value to measure [Ca2+]m in intact HeLa cells. The following
of 490 nM was reported for Rhod-2 and then used in Kd variants were reported: YC2mit 1.26 lM, YC3.1mit
the equation. 3.98 lM, and YC4.1 104 lM. These Kd values were
In 2012, De la Fuente et al.[95] used the extremely determined by applying increasing [Ca2+] from 1 nM to
low Ca2+ affinity dye Rhod-5N to measure [Ca2+]m in 10 mM (1 lM thapsigargin, 1 lM ionomycin, and 1 lM
intact HeLa cells. The stimulus used was histamine CCCP were present in the calibration buffer). The
(100 lM) and histamine plus the MCUC activator Hela cells were stimulated with histamine 50 lM, and
Kaempferol. The [Ca2+]m values registered in individ- the [Ca2+]m peak as wells as the % of saturated pixels
ual cells (with histamine only) ranged from 10 to were recorded. The study showed that [Ca2+]m
40 lM, and the average of multiple cells was 30 lM. increased as the Ca2+ affinity of the probe decreased.
When the histamine was applied together with kaemp- The Ca2+ values were 3.19 lM for YC2mit, 49.4 lM
ferol, the average values were higher, reaching levels for YC3.1mit, and 106 lM for YC4.1mit. The satu-
around 80 lM. Since the spatial resolution of the fluo- rated pixels were 24.8%, 17.8%, and 3.19% respec-
rescent dyes is high enough, [Ca2+]m values were also tively, pointing to the fact that a low Kd affinity of
quantified in subcellular mitochondrial regions. The the probe is critical to correctly and accurately mea-
calibrated Rhod-5N signal showed values close to sure [Ca2+]m values.
120 lM, ~34 times higher than the value obtained by In 2012, Lu et al. [99] performed imaging experi-
Collings et al., in those mitochondria surrounding the ments in intact adult rabbit cardiomyocytes, using the
nucleus. Instead of using the Grynkiewicz calibration fluorescent GECI Mitycam (inverse Pericam). At the
equation, De la Fuente et al. created an intracellular end of every experiment, the Mitycam was calibrated

FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies 7

Mitochondrial Ca2+ concentrations in live cells C. Fernandez-Sanz et al.

in situ with low and high Ca2+ (concentrations not spec- intensity would be extremely low even when changes
ified) to obtain the [Ca2+]m values. The results showed a in [Ca2+] are still high. This phenomenon could lead to
gradient of Ca2+ within an individual mitochondrion misinterpretation of the data since a nonsaturated fluo-
with higher [Ca2+]m in the Z-lines (next to the SR Ca2+ rescence signal does not necessarily translate to a cor-
releasing sites) than in M-lines. The reported [Ca2+]m rectly calibrated [Ca2+]. According to Arnaudeau et al.
were 37 nM in the Z-lines versus 26 nM in the M-lines. [98], higher [Ca2+]m values occurred when a lower
These values are higher but not significantly different Ca2+-affinity probe was used. Note that in this work
from the ones obtained previously by the same group none of the YC Camaleon signals were saturated at
using Fura-2 and Rhod-2 [96]. the [Ca2+]m peak. However, the variant with the lowest
In 2017, Wust et al. [100] delivered the GECI Kd resulted in the measurement of the highest [Ca2+]m.
4mtD3cpv Camaleon (MitoCam) into adult rat car- The kinetics of the probes can also lead to different
diomyocytes by viral infection to estimate [Ca2+]m. detecting signals, a factor that is especially critical for
Upon electrical stimulation (0.1–4 Hz), the [Ca2+]m a rapid Ca2+ change, such as in beating cardiac muscle
values ranged from 300 to 800 nM while in the pres- cells. If the probes are slower in picking up the real-
ence of the Na+/Ca2+ exchanger inhibitor CGP-37157, time Ca2+ changes, then the signals will be distorted.
the values vary from 580 to 1125 nM. The manuscript Another source of discrepancy is the probe calibration
shows a MitoCam calibration where the Kd (470 nM) performed by the different groups. Because calibration
was applied to a modified Grynkiewicz equation [86]. protocols are not standardized, every group carries out
Lastly, multiple publications in several cell types can their own calibrations, leading to slight variations
be found where the bioluminescent GECI aequorin among them. One of the primary cause of discrepan-
was used as a mitochondrial Ca2+ probe. The calibra- cies is that the Fmax and Fmin used for the Grynkie-
tion of the different aequorins was performed as wicz equation are not always obtained under the same
described by De la Fuente et al. [69]. Several Ca2+ conditions. For the Fmin, some groups only omit the
concentrations were added in the presence of thapsi- Ca2+ from the medium while others add the Ca2+
gargin, ionomycin, oligomycin, and FCCP to obtain chelator EGTA in different concentrations to avoid
the calibration curve values. Ca2+ 10 mM addition any Ca2+ contamination. For the Fmax, a similar situ-
achieved the release of the total luminescence. The cal- ation is faced. While some groups apply a [Ca2+]
ibrated [Ca2+]m values from the aequorin and coelen- believed to saturate the probe, others go further and
terazine combinations differ significantly depending on apply [Ca2+] in the 10 mM range to ensure the full sat-
the cell type used for the measurements. As such, the uration. Another possible explanation resides in the
reported values of [Ca2+]m range from 1 lM in car- nature of the cells by themselves. The published data
diomyocytes to 600 lM in bovine chromaffin cells suggest quite a stable rise in the [Ca2+]c values
[101]. A list of reported [Ca2+]m using aequorins, (~ 1 lM) independently on the cell type, so the cytoso-
including further details on the cell type, the stimulus, lic Ca2+ is not a probable source of inconsistency. The
and the [Ca2+]m reached, can be found in Table 1. high [Ca2+]c microdomains have not been as exten-
sively studied as the global cytosolic Ca2+ signal [102].
Nevertheless, it does not appear to be the source of
Possible explanations for the
the variability either [103]. On the other hand, mito-
discrepancies in [Ca2+]m reported
chondria from diverse cell types may uptake Ca2+ in a
Despite many new tools and efforts to measure the different manner. The total amount of MCUC
[Ca2+]m accurately, there is still no consensus about expressed in that particular cell type or the proximity
the absolute values of [Ca2+] reached in the mitochon- between mitochondria and the Ca2+ source may affect
drial matrix during physiological stimulation, even in the mitochondrial Ca2+ uptake capabilities substan-
cases where the same cell types with the same stimulus tially [104]. Here is where the expression and the speci-
were used. The discrepancies may result from a variety fic localization of the MCUC within the cell becomes
of factors. One of the most critical is the Kd of the critical, as does the concept of mitochondrial
probe for Ca2+. If the Kd is too low for the range of ‘‘contactology’’ with other organelles, especially with
[Ca2+]m changes, then the dye will be saturated or close the main intracellular Ca2+ reservoir, the ER/SR [105].
to the saturation; conversely, if the Kd is too high, Another controversial point that may affect the
then the changes may not be detected. This is because reported-free [Ca2+]m is the role of the mitochondrial
most of the probes follow a sigmoidal saturation Pi and its capacity to buffer and form Ca2+ phosphate
curve, meaning that, when the Ca2+ levels start to devi- precipitates [106]. It has been shown that the presence
ate from Kd significantly, changes in the fluorescence of phosphate facilitates the mitochondrial Ca2+ uptake

8 FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies

C. Fernandez-Sanz et al. Mitochondrial Ca2+ concentrations in live cells

leading to 10–30 times more Ca2+ accumulation [107]. reached upon physiological stimulation as well as the
However, this large Ca2+ mitochondrial load is simul- mitochondrial Ca2+ dynamics in the different cell
taneously buffered by Pi having as a consequence types. As mentioned, mitochondria from different cell
mitochondrial morphological changes and Ca2+ precipi- types may uptake Ca2+ in a quite variable range of
tation [69]. While specific studies suggest that free concentrations. It is not possible to cover the full
[Ca2+]m higher than 2 lM sufficiently generates calcium range of [Ca2+] changes (from < 10 nM to > 1 mM)
phosphate precipitates [107], others claim that at least with only one “ideal” sensor. The decision of which
1.5 mM Ca2+ is required to form the same precipitates one is better depends on all factors described above
[69]. These values align with the reported data on the and the cell type where the [Ca2+]m will be measured.
solubility of the product of [HPO42 ]•[Ca2+] [108]. Addi- When its expression is feasible, we would recommend
tionally, the same study showed that the presence of 1 the use of fluorescent GECIs due to its specific orga-
or 3 mM phosphate precludes [Ca2+]m to reach higher nelle targeting. Among these, the ratiometric ones are
concentrations because of the above-mentioned Ca2+ preferred over the single-wavelength to ensure the
buffering capacity of the phosphate. A third study also interference of experimental artifacts. Accurate knowl-
showed that the presence of Pi reduced by almost 50% edge in these Ca2+ values will resolve current discrep-
the [Ca2+]m reached (from an original value of 50 lM in ancies in the amount of [Ca2+]m required to control
the absence of Pi) [109]. The role of the Pi as a mito- essential cellular processes, including energy produc-
chondrial Ca2+ buffer as well as the still under debate tion and mPTP opening, and even to prevent the for-
calcium-free/bound ratio [107,110,111] creates an extre- mation of calcium phosphate precipitates.
mely complex scenario where accurate and precise mea-
surement of [Ca2+]m is critical. Lastly, we cannot forget Acknowledgements
that all Ca2+ sensors are Ca2+ buffers due to their Ca2+-
We thank Jennifer Wilson for English editing to this
binding capability. Therefore, for the sensors with a
manuscript. This study was supported by: NIH/NHLBI
lower quantum yield, the higher loading concentrations
(HL122124, HL093671, HL137266, HL137426,
could alter the original Ca2+ signals at some degree.
HL142864).
However, it is hard to find studies reporting, discussing
or making a comparison between the buffering capaci-
References
ties among the Ca2+ sensors [112].
1 Berridge MJ (2012) Calcium signalling remodelling and
disease. Biochem Soc Trans 40, 297–309.
Conclusions and perspectives 2 Campanella M, Pinton P and Rizzuto R (2004)
Mitochondrial Ca2+ uptake was first described almost Mitochondrial Ca2+homeostasis in health and disease.
40 years ago. Alternations in [Ca2+]m regulate crucial Biol Res 37, 653–660.
processes such as enzyme activities, bioenergetics, and 3 De la Fuente S and Sheu S-S (2019) SR-mitochondria
cell death. Despite the importance of the dynamic communication in adult cardiomyocytes: a close
range of [Ca2+]m, the absolute values that can be relationship where the Ca2+has a lot to say. Arch
Biochem Biophys 663, 259–268.
reached upon stimulation are still not well known. The
4 Giorgi C, Marchi S and Pinton P (2018) The
probes available for precisely measuring [Ca2+] inside
machineries, regulation and cellular functions of
the mitochondria have individual pros and cons. The
mitochondrial calcium. Nat Rev Mol Cell Biol 19, 713–
ideal probe to measure [Ca2+]m requires some essential
730.
characteristics. It needs to be specifically and efficiently
5 Glancy B and Balaban RS (2012) Role of
targeted to the mitochondrial matrix to avoid cytosolic mitochondrial Ca2+in the regulation of cellular
signal contaminations. It should have fast enough energetics. Biochemistry 51, 2959–2973.
kinetics, to be able to track both fast Ca2+ uptakes as 6 Tarasov AI, Griffiths EJ and Rutter GA (2012)
well as fast Ca2+ release. It should have high quantum Regulation of ATP production by mitochondrial Ca2+.
yields to avoid buffering. It should provide enough Cell Calcium 52, 28–35.
spatial resolution at both the cellular and subcellular 7 Lehninger AL, Carafoli E and Rossi CS (1967)
levels. It needs to have the appropriate Ca2+ affinity to Energy-linked ion movements in mitochondrial
not to suffer saturation upon high [Ca2+]. Finally, it systems. Adv Enzymol Relat Areas Mol Biol 29, 259–
should be easy to deliver to the mitochondrial matrix, 320.
independently on the cell type. Further investigations 8 Denton RM and McCormack JG (1980) On the role of
are required to obtain this “perfect” mitochondrial the calcium transport cycle in heart and other
Ca2+ sensor and to use it to elucidate the [Ca2+]m mammalian mitochondria. FEBS Lett 119, 1–8.

FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies 9

Mitochondrial Ca2+ concentrations in live cells C. Fernandez-Sanz et al.

9 Hansford RG and Castro F (1981) Effects of toluene-permeabilized rat heart mitochondria.

micromolar concentrations of free calcium ions on the Interactions with regulation by adenine nucleotides
reduction of heart mitochondrial NAD(P) by and NADH/NAD+ratios. Biochem J 252, 181–189.
2-oxoglutarate. Biochem J 198, 525–533. 22 Bell CJ, Bright NA, Rutter GA and Griffiths EJ
10 Crompton M (1985) The regulation of mitochondrial (2006) ATP regulation in adult rat cardiomyocytes:
calcium transport in heart. Curr Top Membr Transp time-resolved decoding of rapid mitochondrial calcium
25, 231–276. spiking imaged with targeted photoproteins. J Biol
11 Rutter GA and Denton RM (1989) The binding of Chem 281, 28058–28067.
Ca2+ions to pig heart NAD+-isocitrate 23 Griffiths EJ (2009) Mitochondrial calcium transport in
dehydrogenase and the 2-oxoglutarate dehydrogenase the heart: physiological and pathological roles. J Mol
complex. Biochem J 263, 453–462. Cell Cardiol 46, 789–803.
12 Rutter GA (1990) Ca2(+)-binding to citrate cycle 24 Ramachandran N and Colman RF (1980) Chemical
dehydrogenases. Int J Biochem 22, 1081–1088. characterization of distinct subunits of pig heart DPN-
13 Jouaville LS, Pinton P, Bastianutto C, Rutter GA and specific isocitrate dehydrogenase. J Biol Chem 255,
Rizzuto R (1999) Regulation of mitochondrial ATP 8859–8864.
synthesis by calcium: evidence for a long-term 25 Mart
ınez-Rivas JM and Vega JM (1998) Purification
metabolic priming. Proc Natl Acad Sci USA 96, and characterization of NAD-isocitrate dehydrogenase
13807–13812. from Chlamydomonas reinhardtii. Plant Physiol 118,
14 Sale GJ and Randle PJ (1982) Occupancy of 249–255.
phosphorylation sites in pyruvate dehydrogenase 26 Denton RM, Richards DA and Chin JG (1978)
phosphate complex in rat heart in vivo. Relation to Calcium ions and the regulation of NAD+-linked
proportion of inactive complex and rate of re- isocitrate dehydrogenase from the mitochondria of rat
activation by phosphatase. Biochem J 206, 221–229. heart and other tissues. Biochem J 176, 899–906.
15 Fries M, Chauhan HJ, Domingo GJ, Jung H-I and 27 Gabriel JL and Plaut GW (1984) Inhibition of bovine
Perham RN (2003) Site-directed mutagenesis of a loop heart NAD-specific isocitrate dehydrogenase by
at the active site of E1 (alpha2beta2) of the pyruvate reduced pyridine nucleotides: modulation of inhibition
dehydrogenase complex. A possible common sequence by ADP, NAD+, Ca2+, citrate, and isocitrate.
motif. Eur J Biochem 270, 861–870. Biochemistry 23, 2773–2778.
16 Denton RM, McCormack JG and Edgell NJ (1980) 28 Mildaziene V, Baniene R, Nauciene Z, Marcinkeviciute
Role of calcium ions in the regulation of A, Morkuniene R, Borutaite V, Kholodenko B and
intramitochondrial metabolism. Effects of Na+, Brown GC (1996) Ca2+stimulates both the respiratory
Mg2+ and ruthenium red on the Ca2+-stimulated and phosphorylation subsystems in rat heart
oxidation of oxoglutarate and on pyruvate mitochondria. Biochem J 320 (Pt 1), 329–334.
dehydrogenase activity in intact rat heart 29 Territo PR, Mootha VK, French SA and Balaban
mitochondria. Biochem J 190, 107–117. RS (2000) Ca(2+) activation of heart mitochondrial
17 Rutter GA, Midgley PJW and Denton RM (1989) oxidative phosphorylation: role of the F(0)/F(1)-
Regulation of the pyruvate dehydrogenase complex by ATPase. Am J Physiol Cell Physiol 278, C423–C435.
Ca2+within toluene-permeabilized heart mitochondria. 30 Hubbard MJ and McHugh NJ (1996) Mitochondrial
Biochim Biophys Acta Mol Cell Res 1014, 263–270. ATP synthase F1-b-subunit is a calcium-binding
18 Randle PJ (1995) Metabolic fuel selection: general protein. FEBS Lett 391, 323–329.
integration at the whole-body level. Proc Nutr Soc 54, 31 Giorgio V, Burchell V, Schiavone M, Bassot C,
317–327. Minervini G, Petronilli V, Argenton F, Forte M,
19 McLain AL, Szweda PA and Szweda LI (2011) Tosatto S, Lippe G et al. (2017) Ca2+binding
a-Ketoglutarate dehydrogenase: a mitochondrial redox to F-ATP synthase b subunit triggers the
sensor. Free Radic Res 45, 29–36. mitochondrial permeability transition. EMBO Rep
20 Nichols BJ, Rigoulet M and Denton RM (1994) 18, 1065–1076.
Comparison of the effects of Ca2+, adenine nucleotides 32 Williams GSB, Boyman L and Lederer WJ (2015)
and pH on the kinetic properties of mitochondrial Mitochondrial calcium and the regulation of
NAD(+)-isocitrate dehydrogenase and oxoglutarate metabolism in the heart. J Mol Cell Cardiol 78, 35–45.
dehydrogenase from the yeast Saccharomyces 33 Zoratti M and Szab
o I (1994) Electrophysiology of the
cerevisiae and rat heart. Biochem J 303 (Pt 2), 461– inner mitochondrial membrane. J Bioenerg Biomembr
465. 26, 543–553.
21 Rutter GA and Denton RM (1988) Regulation of 34 Crompton M, Virji S and Ward JM (1998)
NAD+-linked isocitrate dehydrogenase and Cyclophilin-D binds strongly to complexes of the
2-oxoglutarate dehydrogenase by Ca2+ions within voltage-dependent anion channel and the adenine

10 FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies

C. Fernandez-Sanz et al. Mitochondrial Ca2+ concentrations in live cells

nucleotide translocase to form the permeability (2011) The permeability transition pore controls
transition pore. Eur J Biochem 258, 729–735. cardiac mitochondrial maturation and myocyte
35 Baines CP (2010) The cardiac mitochondrion: nexus of differentiation. Dev Cell 21, 469–478.
stress. Annu Rev Physiol 72, 61–80. 48 Paradies G, Ruggiero FM, Petrosillo G and
36 Kokoszka JE, Waymire KG, Levy SE, Sligh JE, Cai J, Quagliariello E (1993) Age-dependent decrease in the
Jones DP, MacGregor GR and Wallace DC (2004) cytochrome c oxidase activity and changes in
The ADP/ATP translocator is not essential for the phospholipids in rat-heart mitochondria. Arch Gerontol
mitochondrial permeability transition pore. Nature Geriatr 16, 263–272.
427, 461–465. 49 Ruiz-Meana M, Abell
an A, Mir
o-Casas E and Garcia-
37 Santos AC, Uyemura SA, Lopes JL, Bazon JN, Dorado D (2007) Opening of mitochondrial
Mingatto FE and Curti C (1998) Effect of naturally permeability transition pore induces hypercontracture
occurring flavonoids on lipid peroxidation and in Ca2+overloaded cardiac myocytes. Basic Res
membrane permeability transition in mitochondria. Cardiol 102, 542–552.
Free Radic Biol Med 24, 1455–1461. 50 Zorov DB, Juhaszova M and Sollott SJ (2014)
38 Wang X-J, Wang Z-B and Xu J-X (2005) Effect of Mitochondrial Reactive Oxygen Species (ROS) and
salvianic acid A on lipid peroxidation and membrane ROS-induced ROS release. Physiol Rev 94, 909–950.
permeability in mitochondria. J Ethnopharmacol 97, 51 Brookes PS, Yoon Y, Robotham JL, Anders MW and
441–445. Sheu S-S (2004) Calcium, ATP, and ROS: a
39 Kuwana T and Newmeyer DD (2003) Bcl-2-family mitochondrial love-hate triangle. Am J Physiol Cell
proteins and the role of mitochondria in apoptosis. Physiol 287, C817–C833.
Curr Opin Cell Biol 15, 691–699. 52 Bernardi P and von Stockum S (2012) The
40 Basa~nez G, Soane L and Hardwick JM (2012) A new permeability transition pore as a Ca(2+) release
view of the lethal apoptotic pore. PLoS Biol 10, channel: new answers to an old question. Cell Calcium
e1001399. 52, 22–27.
41 He L and Lemasters JJ (2002) Regulated and 53 Elrod JW and Molkentin JD (2013) Physiologic
unregulated mitochondrial permeability transition functions of cyclophilin D and the mitochondrial
pores: a new paradigm of pore structure and function? permeability transition pore. Circ J 77, 1111–1122.
FEBS Lett 512, 1–7. 54 Altschuld RA, Hohl CM, Castillo LC, Garleb AA,
42 Giorgio V, von Stockum S, Antoniel M, Fabbro A, Starling RC and Brierley GP (1992) Cyclosporin
Fogolari F, Forte M, Glick GD, Petronilli V, Zoratti inhibits mitochondrial calcium efflux in isolated adult
M, Szab
o I et al. (2013) Dimers of mitochondrial ATP rat ventricular cardiomyocytes. Am J Physiol 262,
synthase form the permeability transition pore. Proc H1699–H1704.
Natl Acad Sci USA 110, 5887–5892. 55 Elrod JW, Wong R, Mishra S, Vagnozzi RJ,
43 Alavian KN, Beutner G, Lazrove E, Sacchetti S, Park Sakthievel B, Goonasekera SA, Karch J, Gabel S,
H-A, Licznerski P, Li H, Nabili P, Hockensmith K, Farber J, Force T et al. (2010) Cyclophilin D controls
Graham M et al. (2014) An uncoupling channel within mitochondrial pore-dependent Ca(2+) exchange,
the c-subunit ring of the F1FO ATP synthase is the metabolic flexibility, and propensity for heart failure in
mitochondrial permeability transition pore. Proc Natl mice. J Clin Invest 120, 3680–3687.
Acad Sci USA 111, 10580–10585. 56 Orrenius S, Gogvadze V and Zhivotovsky B (2015)
44 Bernardi P and Di Lisa F (2015) The mitochondrial Calcium and mitochondria in the regulation of cell
permeability transition pore: molecular nature and role death. Biochem Biophys Res Commun 460, 72–81.
as a target in cardioprotection. J Mol Cell Cardiol 78, 57 Tsien RY, Pozzan T and Rink TJ (1984) Calcium
100–106. activities and fluxes inside small intact cells as measured
45 Karch J, Bround MJ, Khalil H, Sargent MA, with intracellularly trapped chelators. Adv Cyclic
Latchman N, Terada N, Peixoto PM and Molkentin Nucleotide Protein Phosphorylation Res 17, 535–541.
JD (2018) Inhibition of mitochondrial permeability 58 Grynkiewicz G, Poenie M and Tsien RY (1985) A new
transition by deletion of the ANT family and CypD. generation of Ca2+indicators with greatly improved
bioRxiv506964. https//doi.org/10.1101/506964 fluorescence properties. J Biol Chem 260, 3440–3450.
[PREPRINT]. 59 Minta A, Kao JP and Tsien RY (1989) Fluorescent
46 Gong G, Liu X, Zhang H, Sheu S-S and Wang W indicators for cytosolic calcium based on rhodamine
(2015) Mitochondrial flash as a novel biomarker of and fluorescein chromophores. J Biol Chem 264, 8171–
mitochondrial respiration in the heart. Am J Physiol 8178.
Heart Circ Physiol 309, H1166–H1177. 60 Tsien RY (1981) A non-disruptive technique for
47 Hom JR, Quintanilla RA, Hoffman DL, de Mesy loading calcium buffers and indicators into cells.
Bentley KL, Molkentin JD, Sheu S-S and Porter GA Nature 290, 527–528.

FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies 11

Mitochondrial Ca2+ concentrations in live cells C. Fernandez-Sanz et al.

61 Trollinger DR, Cascio WE and Lemasters JJ (2000) 74 Saito K, Chang YF, Horikawa K, Hatsugai N,
Mitochondrial calcium transients in adult rabbit Higuchi Y, Hashida M, Yoshida Y, Matsuda T, Arai
cardiac myocytes: inhibition by ruthenium red and Y and Nagai T (2012) Luminescent proteins for high-
artifacts caused by lysosomal loading of Ca(2+)- speed single-cell and whole-body imaging. Nat
indicating fluorophores. Biophys J 79, 39–50. Commun 3, 1262.
62 Miyata H, Silverman HS, Sollott SJ, Lakatta EG, 75 Perez Koldenkova V and Nagai T (2013) Genetically
Stern MD and Hansford RG (1991) Measurement of encoded Ca(2+) indicators: properties and evaluation.
mitochondrial free Ca2+concentration in living single Biochim Biophys Acta 1833, 1787–1797.
rat cardiac myocytes. Am J Physiol 261, H1123– 76 Shimomura O, Johnson FH and Saiga Y (1962)
H1134. Extraction, purification and properties of aequorin, a
63 Davidson SM, Yellon D and Duchen MR (2007) bioluminescent protein from the luminous
Assessing mitochondrial potential, calcium, and redox hydromedusan, Aequorea. J Cell Comp Physiol 59,
state in isolated mammalian cells using confocal 223–239.
microscopy. Methods Mol Biol 372, 421–430. 77 Heim R and Tsien RY (1996) Engineering green
64 Sedova M, Dedkova EN and Blatter LA (2006) fluorescent protein for improved brightness, longer
Integration of rapid cytosolic Ca2+signals by wavelengths and fluorescence resonance energy
mitochondria in cat ventricular myocytes. Am J transfer. Curr Biol 6, 178–182.
Physiol Cell Physiol 291, C840–C850. 78 Porumb T, Yau P, Harvey TS and Ikura M (1994) A
65 Csordas G and Hajnoczky G (2001) Sorting of calcium calmodulin-target peptide hybrid molecule with unique
signals at the junctions of endoplasmic reticulum and calcium-binding properties. Protein Eng 7, 109–115.
mitochondria. Cell Calcium 29, 249–262. 79 Romoser VA, Hinkle PM and Persechini A (1997)
66 Fonteriz RI, de la Fuente S, Moreno A, Lobaton CD, Detection in living cells of Ca2+-dependent changes in
Montero M and Alvarez J (2010) Monitoring the fluorescence emission of an indicator composed of
mitochondrial [Ca(2+)] dynamics with rhod-2, two green fluorescent protein variants linked by a
ratiometric pericam and aequorin. Cell Calcium 48, calmodulin-binding sequence. A new class of
61–69. fluorescent indicators. J Biol Chem 272, 13270–13274.
67 Rizzuto R, Simpson AW, Brini M and Pozzan T 80 Miyawaki A, Llopis J, Heim R, McCaffery JM,
(1992) Rapid changes of mitochondrial Ca2+revealed Adams JA, Ikura M and Tsien RY (1997) Fluorescent
by specifically targeted recombinant aequorin. Nature indicators for Ca2+based on green fluorescent proteins
358, 325–327. and calmodulin. Nature 388, 882–887.
68 Shimomura O, Johnson FH and Saiga Y (1963) 81 Nagai T, Sawano A, Park ES and Miyawaki A (2001)
Microdetermination of calcium by aequorin Circularly permuted green fluorescent proteins
luminescence. Science 140, 1339–1340. engineered to sense Ca2+. Proc Natl Acad Sci USA 98,
69 de la Fuente S, Fonteriz RI, de la Cruz PJ, Montero 3197–3202.
M and Alvarez J (2012) Mitochondrial free [Ca(2+)] 82 Baird GS, Zacharias DA and Tsien RY (1999)
dynamics measured with a novel low-Ca(2+) affinity Circular permutation and receptor insertion within
aequorin probe. Biochem J 445, 371–376. green fluorescent proteins. Proc Natl Acad Sci USA 96,
70 Nunez L, Senovilla L, Sanz-Blasco S, Chamero P, 11241–11246.
Alonso MT, Villalobos C and Garcia-Sancho J (2007) 83 Nakai J, Ohkura M and Imoto K (2001) A high
Bioluminescence imaging of mitochondrial signal-to-noise Ca(2+) probe composed of a single
Ca2+dynamics in soma and neurites of individual green fluorescent protein. Nat Biotechnol 19, 137–141.
adult mouse sympathetic neurons. J Physiol 580, 385– 84 Suzuki J, Kanemaru K, Ishii K, Ohkura M, Okubo Y
395. and Iino M (2014) Imaging intraorganellar Ca2+at
71 Campbell AK (1974) Extraction, partial purification subcellular resolution using CEPIA. Nat Commun 5,
and properties of obelin, the calcium-activated 4153.
luminescent protein from the hydroid Obelia 85 Zhao Y, Araki S, Wu J, Teramoto T, Chang YF,
geniculata. Biochem J 143, 411–418. Nakano M, Abdelfattah AS, Fujiwara M, Ishihara T,
72 Bovolenta S, Foti M, Lohmer S and Corazza S (2007) Nagai T et al. (2011) An expanded palette of
Development of a Ca(2+)-activated photoprotein, genetically encoded Ca(2)(+) indicators. Science 333,
Photina, and its application to high-throughput 1888–1891.
screening. J Biomol Screen 12, 694–704. 86 Palmer AE, Giacomello M, Kortemme T, Hires SA,
73 Inouye S and Tsuji FI (1993) Cloning and sequence Lev-Ram V, Baker D and Tsien RY (2006)
analysis of cDNA for the Ca(2+)-activated Ca2+indicators based on computationally redesigned
photoprotein, clytin. FEBS Lett 315, 343–346. calmodulin-peptide pairs. Chem Biol 13, 521–530.

12 FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies

C. Fernandez-Sanz et al. Mitochondrial Ca2+ concentrations in live cells

87 Reiff DF, Ihring A, Guerrero G, Isacoff EY, Joesch Rapid frequency-dependent changes in free
M, Nakai J and Borst A (2005) In vivo performance mitochondrial calcium concentration in rat cardiac
of genetically encoded indicators of neural activity in myocytes. J Physiol 595, 2001–2019.
flies. J Neurosci 25, 4766–4778. 101 Montero M, Alonso MT, Carnicero E, Cuchillo-
88 Baubet V, Le Mouellic H, Campbell AK, Lucas- Ibanez I, Albillos A, Garcia AG, Garcia-Sancho J and
Meunier E, Fossier P and Brulet P (2000) Chimeric Alvarez J (2000) Chromaffin-cell stimulation triggers
green fluorescent protein-aequorin as bioluminescent fast millimolar mitochondrial Ca2+transients that
Ca2+reporters at the single-cell level. Proc Natl Acad modulate secretion. Nat Cell Biol 2, 57–61.
Sci USA 97, 7260–7265. 102 Csord
as G, V
arnai P, Golen
ar T, Roy S, Purkins G,
89 Rodriguez-Garcia A, Rojo-Ruiz J, Navas-Navarro P, Schneider TG, Balla T and Hajn
oczky G (2010)
Aulestia FJ, Gallego-Sandin S, Garcia-Sancho J and Imaging interorganelle contacts and local calcium
Alonso MT (2014) GAP, an aequorin-based dynamics at the ER-mitochondrial interface. Mol Cell
fluorescent indicator for imaging Ca2+in organelles. 39, 121–132.
Proc Natl Acad Sci USA 111, 2584–2589. 103 Rizzuto R and Pozzan T (2006) Microdomains of
90 Suzuki J, Kanemaru K and Iino M (2016) Genetically intracellular Ca2+: molecular determinants and
encoded fluorescent indicators for organellar calcium functional consequences. Physiol Rev 86, 369–408.
imaging. Biophys J 111, 1119–1131. 104 Fieni F, Lee SB, Jan YN and Kirichok Y (2012)
91 Alonso MT, Rodriguez-Prados M, Navas-Navarro P, Activity of the mitochondrial calcium uniporter varies
Rojo-Ruiz J and Garcia-Sancho J (2017) Using greatly between tissues. Nat Commun 3, 1317.
aequorin probes to measure Ca(2+) in intracellular 105 Csordas G, Weaver D and Hajnoczky G (2018)
organelles. Cell Calcium 64, 3–11. Endoplasmic reticulum-mitochondrial contactology:
92 Granatiero V, Patron M, Tosatto A, Merli G and Rizzuto structure and signaling functions. Trends Cell Biol 28,
R (2014) The use of aequorin and its variants for 523–540.
Ca2+measurements. Cold Spring Harb Protoc 2014, 9–16. 106 Nicholls DG (2005) Mitochondria and calcium
93 Collins TJ, Lipp P, Berridge MJ and Bootman MD signaling. Cell Calcium 38, 311–317.
(2001) Mitochondrial Ca(2+) uptake depends on the 107 Chalmers S and Nicholls DG (2003) The relationship
spatial and temporal profile of cytosolic Ca(2+) between free and total calcium concentrations in the
signals. J Biol Chem 276, 26411–26420. matrix of liver and brain mitochondria. J Biol Chem
94 Pitter JG, Maechler P, Wollheim CB and Spat A 278, 19062–19070.
(2002) Mitochondria respond to Ca2+already in the 108 Fryer MW, Owen VJ, Lamb GD and Stephenson DG
submicromolar range: correlation with redox state. (1995) Effects of creatine phosphate and P(i) on
Cell Calcium 31, 97–104. Ca2+movements and tension development in rat
95 de la Fuente S, Fonteriz RI, Montero M and Alvarez skinned skeletal muscle fibres. J Physiol 482 (Pt 1),
J (2012) Dynamics of mitochondrial [Ca(2+)] 123–140.
measured with the low-Ca(2+)-affinity dye rhod-5N. 109 Wei AC, Liu T and O’Rourke B (2015) Dual effect of
Cell Calcium 51, 65–71. phosphate transport on mitochondrial Ca2+dynamics.
96 Andrienko TN, Picht E and Bers DM (2009) J Biol Chem 290, 16088–16098.
Mitochondrial free calcium regulation during 110 Babcock DF, Herrington J, Goodwin PC, Park YB
sarcoplasmic reticulum calcium release in rat cardiac and Hille B (1997) Mitochondrial participation in the
myocytes. J Mol Cell Cardiol 46, 1027–1036. intracellular Ca2+network. J Cell Biol 136, 833–844.
97 Xu-Friedman MA and Regehr WG (1999) Presynaptic 111 de la Fuente S, Montenegro P, Fonteriz RI, Moreno
strontium dynamics and synaptic transmission. Biophys A, Lobat
on CD, Montero M and Alvarez J (2010)
J 76, 2029–2042. The dynamics of mitochondrial Ca2+fluxes. Biochim
98 Arnaudeau S, Kelley WL, Walsh JV Jr and Demaurex Biophys Acta 1797, 1727–1735.
N (2001) Mitochondria recycle Ca(2+) to the 112 McMahon SM and Jackson MB (2018) An
endoplasmic reticulum and prevent the depletion of inconvenient truth: calcium sensors are calcium
neighboring endoplasmic reticulum regions. J Biol buffers. Trends Neurosci 41, 880–884.
Chem 276, 29430–29439. 113 Csordas G, Thomas AP and Hajnoczky G (1999)
99 Lu X, Ginsburg KS, Kettlewell S, Bossuyt J, Smith Quasi-synaptic calcium signal transmission between
GL and Bers DM (2013) Measuring local gradients of endoplasmic reticulum and mitochondria. EMBO J 18,
intramitochondrial [Ca(2+)] in cardiac myocytes 96–108.
during sarcoplasmic reticulum Ca(2+) release. Circ Res 114 Cieri D, Vicario M, Vallese F, D’Orsi B, Berto P,
112, 424–431. Grinzato A, Catoni C, De Stefani D, Rizzuto R, Brini
100 Wust RC, Helmes M, Martin JL, van der Wardt TJ, M et al. (2018) Tau localises within mitochondrial
Musters RJ, van der Velden J and Stienen GJ (2017) sub-compartments and its caspase cleavage affects ER-

FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies 13

Mitochondrial Ca2+ concentrations in live cells C. Fernandez-Sanz et al.

mitochondria interactions and cellular Ca(2+) transients stimulate autophagy in human fibroblasts
handling. Biochim Biophys Acta Mol Basis Dis 1864, carrying the 13514A>G mutation of the ND5 subunit
3247–3256. of NADH dehydrogenase. Cell Death Differ 23, 231–
115 Patron M, Checchetto V, Raffaello A, Teardo E, 241.
Vecellio Reane D, Mantoan M, Granatiero V, Szabo 120 Robert V, Gurlini P, Tosello V, Nagai T, Miyawaki
I, De Stefani D and Rizzuto R (2014) MICU1 and A, Di Lisa F and Pozzan T (2001) Beat-to-beat
MICU2 finely tune the mitochondrial Ca2+uniporter oscillations of mitochondrial [Ca2+] in cardiac cells.
by exerting opposite effects on MCU activity. Mol EMBO J 20, 4998–5007.
Cell 53, 726–737. 121 Patergnani S, Giorgi C, Maniero S, Missiroli S,
116 Cali T, Ottolini D, Negro A and Brini M (2012) Maniscalco P, Bononi I, Martini F, Cavallesco G,
alpha-Synuclein controls mitochondrial calcium Tognon M and Pinton P (2015) The endoplasmic
homeostasis by enhancing endoplasmic reticulum- reticulum mitochondrial calcium cross talk is
mitochondria interactions. J Biol Chem 287, 17914– downregulated in malignant pleural mesothelioma cells
17929. and plays a critical role in apoptosis inhibition.
117 Wright LE, Vecellio Reane D, Milan G, Terrin A, Di Oncotarget 6, 23427–23444.
Bello G, Belligoli A, Sanna M, Foletto M, Favaretto 122 Coussee E, De Smet P, Bogaert E, Elens I, Van
F, Raffaello A et al. (2017) Increased mitochondrial Damme P, Willems P, Koopman W, Van Den Bosch
calcium uniporter in adipocytes underlies L and Callewaert G (2011) G37R SOD1 mutant
mitochondrial alterations associated with insulin alters mitochondrial complex I activity, Ca(2+)
resistance. Am J Physiol Endocrinol Metab 313, E641– uptake and ATP production. Cell Calcium 49, 217–
E650. 225.
118 Tosatto A, Sommaggio R, Kummerow C, Bentham 123 Poburko D, Potter K, van Breemen E, Fameli N, Liao
RB, Blacker TS, Berecz T, Duchen MR, Rosato A, CH, Basset O, Ruegg UT and van Breemen C (2006)
Bogeski I, Szabadkai G et al. (2016) The Mitochondria buffer NCX-mediated Ca2+-entry and
mitochondrial calcium uniporter regulates breast limit its diffusion into vascular smooth muscle cells.
cancer progression via HIF-1alpha. EMBO Mol Med Cell Calcium 40, 359–371.
8, 569–585. 124 Lim D, Bertoli A, Sorgato MC and Moccia F (2016)
119 Granatiero V, Giorgio V, Cali T, Patron M, Brini M, Generation and usage of aequorin lentiviral vectors for
Bernardi P, Tiranti V, Zeviani M, Pallafacchina G, De Ca(2+) measurement in sub-cellular compartments of
Stefani D et al. (2016) Reduced mitochondrial Ca(2+) hard-to-transfect cells. Cell Calcium 59, 228–239.

14 FEBS Letters (2019) ª 2019 Federation of European Biochemical Societies