THE THEODOR BU
¨
CHER LECTURE
Mitochondrial calciumsignallingincell death
Delivered on1July2004atthe29thFEBSCongressin Warsaw
Sara Leo
1
, Katiuscia Bianchi
1
, Marisa Brini
2
and Rosario Rizzuto
1
1 Department of Experimental and Diagnostic Medicine, Section of General Pathology, and Interdisciplinary Center for the Study of
Inflammation (ICSI), University of Ferrara, Italy
2 Department of Biochemistry, University of Padova, Italy
In the last few decades, much information has been
obtained onthe role of calcium ions as ubiquitous sec-
ond messengers that translate the binding of signalling
molecules to plasmamembrane receptors into defined
cell activities [1]. Thanks to the development of highly
efficient probes (the intracellularly trappable fluores-
cent indicators developed by Tsien and coworkers) [2],
it was possible to investigate thecalcium signals elici-
ted in a wide variety of cell types, either in culture (pri-
mary cultures and immortalized cell lines) or in situ
(organotypic slices or even the intact tissue within a
living organism) by the opening of plasmamembrane
Ca
2+
channels or Ca
2+
channels of intracellular reser-
voirs, cytologically identifiable with the endoplasmic
reticulum (ER) and, more recently, with the Golgi
apparatus [3,4].
The use of Ca
2+
as a second messenger rests on the
maintenance of a low cytosolic Ca
2+
concentration,
through the energy-consuming pumping activity of
Ca
2+
ATPases located in ER ⁄ SR (SERCA) or
plasmamembrane (PMCA). As to the triggering mech-
anism of the [Ca
2+
] rise, a route involves either the
stimulation of G-protein coupled receptors (specifically
those coupled to a G(aq) protein, that activate phos-
pholipase Cb and thus produce inositol 1,4,5-trisphos-
phate (IP3) from the hydrolysis of the lipid
phosphatidyl-inositol 4,5-diphosphate) or growth fac-
tors receptors (also causing the production of IP3
through the activation of phospholipase Cc, containing
an SH2 domain that recruits it to the activated GF-R)
[1]. An alternative route for raising cytosolic Ca
2+
concentration ([Ca
2+
]
c
) depends onthe opening of
Keywords
apoptosis; calcium; mitochondria;
organelles; photoproteins; signal
transduction
Correspondence
R. Rizzuto, Department of Experimental and
Diagnostic Medicine, General Pathology
Section, University of Ferrara,
Via L. Borsari 46, 44100 Ferrara, Italy
Fax: +39 0532247278
Tel: +39 0532291361
E-mail: r.rizzuto@unife.it
(Received 1 June 2005, accepted 11 July 2005)
doi:10.1111/j.1742-4658.2005.04855.x
The development of targeted probes (based onthe molecular engineering
of luminescent or fluorescent proteins) has allowed the specific measure-
ment of [Ca
2+
] in intracellular organelles or cytoplasmic subdomains. This
approach gave novel information on different aspects of cellular Ca
2+
homeostasis. Regarding mitochondria, it was possible to demonstrate that,
upon physiological stimulation of cells, Ca
2+
is rapidly accumulated in the
matrix. We will discuss the basic characteristics of this process, its role in
modulating physiological and pathological events, such as the regulation of
aerobic metabolism and the induction of cell death, and new insight into
the regulatory mechanisms operating in vivo.
Abbreviations
AGC, aspartate ⁄ glutamate metabolite carrier; COX8, cytochrome c oxidase; CRAC, Ca
2+
release-activated current; ER, endoplasmic
reticulum; HBx, x protein of the hepatitis B virus; IP3, inositol 1,4,5-trisphosphate; PKC, protein kinase C; PMCA, plasmamembrane Ca
2+
ATPase; SERCA, sarcoplamic reticulum ⁄ ER Ca
2+
ATPase.
FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS 4013
various classes of plasma membrane Ca
2+
channels,
such as those directly opened by ligand binding (e.g.
the ionotropic glutamate receptors of neurons), those
opened by the depolarization of the plasmamembrane
(the wide number of voltage-dependent Ca
2+
channels)
or those opened by other intracellular signals (e.g. sec-
ond messengers or the depletion of intracellular Ca
2+
stores) [5,6].
The concerted action of channels with distinct spa-
tial distribution and kinetics of opening determines a
high spatio-temporal specificity of the signals elicited
by different agonists, which in turn are decoded into
radically different intracellular effects. This adds fur-
ther interest and complexity to thesignalling properties
of Ca
2+
: not only tissue-specific functions (e.g. endo-
crine and neuro-secretion, muscle contraction and fer-
tilization), but also decisions oncell fate (proliferation,
cell death by necrosis or apoptosis) are controlled by
Ca
2+
[7]. Thus, not surprisingly deregulations in intra-
cellular Ca
2+
homeostasis have been implicated in the
pathogenesis of genetic (e.g. familial migraine and skin
disorders, such as Darier’s and Hailey-Hailey diseases)
and multifactorial (e.g. hypertension and diabetes) dis-
eases [8,9].
The recognition of the spatio-temporal complexity
of calcium signals and of their multiple signalling roles
has ignited interest in clarifying the molecular mecha-
nisms that allow to specifically decode different signal-
ling patterns. Extensive work inthe past decades has
revealed the broad repertoire of Ca
2+
effectors, i.e.
enzymes, channels or structural proteins that modify
their activity upon binding of Ca
2+
. At first, these
included cytosolic proteins, such as the Ca
2+
-depend-
ent kinases [protein kinase C (PKC), CamK] or phos-
phatases (calcineurin) and their targets [1]. In recent
years, however, it became clear that also processes
occurring within intracellular organelles (gene tran-
scription, post-translational modification of proteins
and aerobic metabolism) are modulated by [Ca
2+
]
changes [10,11]. Thus, Ca
2+
-dependent effects within
organelles are now considered a significant component
of the ‘Ca
2+
symphony’, initiated by physiological or
pathological stimuli, which may influence its final out-
come.
In this context, measuring Ca
2+
concentrations
within organelles with accuracy and specificity has
become an important experimental task. This task has
largely been accomplished, thanks to the development
of a new class of probes that are based on Ca
2+
-sensi-
tive reporter proteins and take advantage of the highly
selective mechanisms that target cellular proteins to
the correct location. The first successful example has
been that of aequorin. Aequorin is a Ca
2+
-sensitive
photoprotein of the jellyfish Aequorea victoria, which
emits light upon binding of Ca
2+
to three high-affinity
sites present inthe protein sequence. The protein can
be purified from jellyfish extracts and microinjected in
cells. Using this classical indicator, seminal observa-
tions were made, such as the repetitive [Ca
2+
]
c
spiking
induced by agonist stimulation [12]. More recently, we
have taken advantage of molecular biology techniques
for developing a series of specifically targeted Ca
2+
probes. The rationale was that of fusing the aequorin
cDNA with DNA sequences encoding specific target-
ing signals, i.e. the protein sequences that are necessary
and sufficient for localizing a mammalian protein to
the correct subcellular location. Figure 1A shows an
example, mtAEQ, which is the recombinant protein
developed for measuring [Ca
2+
] within the mitochond-
rial matrix and was instrumental in gaining new insight
into mitochondrial Ca
2+
handling, i.e. the topic of this
review [13]. Inthe chimeric cDNA, an aequorin moiety
including an HA1 tag was fused in frame with the
N-terminal portion (including the 25 amino acids
cleavable presequence and the first eight amino acids
of the mature polypeptide) of subunit VIII of cyto-
chrome c oxidase (COX8). The fusion protein, when
expressed in mammalian cells, is entirely distributed
to the mitochondria (Fig. 1B). The localization of
mtAEQ is revealed by immunofluorescence using an
antibody that recognizes the HA1 domain.
In general, targeted aequorins (also developed for
other intracellular compartments using similar strat-
egies) have proved to be extremely valuable, and
allowed many new data and novel concepts in Ca
2+
signalling to be obtained. The most important ones
A
B
Fig. 1. Schematic map of the mtAEQ construct (aequorin targeted
to themitochondrial matrix) and its localization by immunofluores-
cence.
Mitochondrial calciumsignallingincelldeath S. Leo et al.
4014 FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS
not covered in this review are the estimates of ER
[Ca
2+
] inthe near-millimolar range ( 0.5 mm)
[14,15], the role of the agonist-sensitive Ca
2+
store
played by the Golgi apparatus (endowed with a resting
[Ca
2+
]of 0.3 mm and rapidly emptying after agonist
stimulation) [16], the rapid equilibration of cytosolic
and nuclear [Ca
2+
] [17], the estimates of resting and
stimulated [Ca
2+
]
c
under the plasmamembrane, well
above those of the bulk cytosol [18]. This paper
focuses on mitochondria, as Ca
2+
handling by these
organelles was not only significantly different from
that expected, but also identified them as critical
checkpoints, in which radically different effects can be
triggered by a rise in [Ca
2+
].
Mechanism and role of mitochondrial
Ca
2+
homeostasis
The participation of mitochondria in Ca
2+
homeostasis
is a concept that alternated periods of glory and com-
plete dismissal. Indeed, as soon as the chemiosmotic
theory was accepted as the basis of energy conservation
in mitochondria, it became obvious that these organ-
elles could, at least potentially, efficiently accumulate
Ca
2+
down the electrochemical gradient established
across the inner membrane by the activity of respirat-
ory complexes. This possibility was actually directly
demonstrated by an extensive body of work carried out
with isolated mitochondria. Respiring mitochondria
can rapidly accumulate Ca
2+
through an electrogenic
pathway, termed the ‘mitochondrial Ca
2+
uniporter’
(MCU) [19]. This route was (and still is) undefined at
the molecular level, although very recent work by Clap-
ham and coworkers demonstrated that it is a bona fide
Ca
2+
channel [20]. Ca
2+
is then re-extruded by electro-
neutral exchangers (Na
2+
⁄ Ca
2+
and H
+
⁄ Ca
2+
exchangers, mostly expressed in nonexcitable and excit-
able cells, respectively) [21]. Based on this evidence,
mitochondria were thought to dynamically change the
matrix Ca
2+
concentration ([Ca
2+
]
m
) in living cells
challenged with Ca
2+
-mobilizing agonists.
This possibility was severely questioned inthe 1980s,
when thesignalling pathways downstream of receptor
stimulation were clarified. It was then demonstrated
that G-protein coupled and growth factor receptors
mobilize Ca
2+
from an intracellular store that proved
be the endoplasmic reticulum, not mitochondria [22].
Moreover, the accurate measurement of [Ca
2+
]
c
levels
with fluorescent indicators suggested that mitochondria
did not receive the Ca
2+
released either. Indeed, both
at rest and after stimulation the [Ca
2+
]
c
values were
well below those necessary for rapid Ca
2+
uptake
through the MCU. Thus, the general consensus
became that mitochondria can accumulate a significant
amount of Ca
2+
only when large and sustained
[Ca
2+
]
c
increases occurred, such as those postulated to
occur in various pathological derangements (e.g. the
Ca
2+
overload during neuronal excitotoxicity).
This situation was completely reversed when the tar-
geted recombinant indicators (first aequorin, then the
more recent GFP-based fluorescent probes) clearly
demonstrated that a [Ca
2+
]
c
rise elicited by a physiolo-
gical stimulation is almost invariably paralleled by
a robust [Ca
2+
]
m
increase [23], that usually largely
exceeds the values observed inthe bulk cytosol and
reached values as high as 500 lm [24]. The apparent
discrepancy with the sluggish rate of Ca
2+
uptake
observed in isolated mitochondria upon exposure to
Ca
2+
concentrations similar to those measured in the
bulk cytosol was reconciled by postulating that mito-
chondria upon opening of the IP
3
-sensitive channels
are not exposed to those low [Ca
2+
], but rather to the
much higher values generated inthe proximity of the
channel (the microdomain hypothesis) [25]. In support
of this notion, organelle labelling of the ER and mito-
chondria showed closed appositions and an aequorin
chimera located onthemitochondrial membrane detec-
ted [Ca
2+
] values well above those of the bulk cytosol
[26].
Numerous studies then followed that demonstrated,
both incell lines and in intact tissues, the occurrence
of rapid [Ca
2+
]
m
transients in cells as diverse as HeLa,
hepatocytes, cardiac and skeletal muscle and neurons
[27]. A striking example is that of cardiac muscle, in
which a [Ca
2+
]
m
transient was detected at every con-
tractile cycle [28]. This implies that both the uptake
and the release mechanism are highly efficient in situ,
and allow the completion of a Ca
2+
cycle within the
short time frame of a single contraction.
What is the role of mitochondrial Ca
2+
homeosta-
sis? A first obvious function stems from long-standing
biochemical evidence, i.e. the demonstration by Den-
ton, McCormack and Hansford inthe 1960s that three
key metabolic enzymes (the pyruvate, a-ketoglutarate
and isocitrate dehydrogenases) are activated by Ca
2+
,
by different mechanisms. Inthe case of pyruvate dehy-
drogenase, this is through a Ca
2+
-dependent dephos-
phorylation step, whereas inthe latter two cases this is
through the direct binding of Ca
2+
to the enzyme
complex [29,30]. Thus, a [Ca
2+
] rise inthe matrix may
allow the up-regulation of aerobic metabolism and
tuning of ATP production to the increased needs of
stimulated cells. This could be demonstrated by the
direct measurement of mitochondrial ATP levels with
a targeted chimera of the ATP-sensitive photoprotein
luciferase. Parallel measurements of Ca
2+
and ATP
S. Leo et al. Mitochondrialcalciumsignallingincell death
FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS 4015
levels showed that the Ca
2+
signal within the mito-
chondria is responsible for the enhanced ATP produc-
tion, an effect that lasts longer than the Ca
2+
signal
itself, highlighting a novel form of cellular ‘metabolic
memory’ [31].
Interestingly, recent work indicates that other Ca
2+
-
dependent metabolic checkpoints are operative.
Namely, the aspartate ⁄ glutamate metabolite carriers
(AGCs) were shown to include EF-hand domains, and
Ca
2+
binding to these sites was shown to increase their
activity [32]. In turn, recombinant expression of wild
type AGCs enhanced ATP production upon cell stimu-
lation, an effect that was not observed with truncated
mutants lacking the Ca
2+
-binding domain [33].
Substantial evidence has built up in recent years
indicating that metabolic regulation is only one of the
roles of themitochondrial Ca
2+
signal. It now appears
evident that massive Ca
2+
loading (as inthe case of
glutamate excitotoxicity of neurons) and ⁄ or the com-
bined action of apoptotic agents or pathophysiological
conditions (e.g. oxidative stress) can induce a profound
alteration of organelle structure and function [34–36].
As a consequence, bioenergetic dysfunction and ⁄ or
release into the cytosol of proteins acting as caspase
cofactors, such as cytochrome c [37,38], AIF [39] and
Smac ⁄ Diablo [40], may lead thecell to necrotic or
apoptotic cell death. In relation to this effect, the anti-
oncogene Bcl-2 was shown to reduce the steady state
Ca
2+
levels inthe ER (and thus dampen the pro-apop-
totic Ca
2+
signal) [41,42].
Some of these data, which refer to our work on Bcl-
2 overexpression in HeLa cells [41], are presented in
Fig. 2A–C. HeLa cells were transiently transfected
with a cytomegalovirus-driven expression plasmid for
Bcl-2 and Ca
2+
homeostasis atthe subcellular level
was investigated 36 h after transfection using organ-
elle-targeted aequorin chimeras. In Bcl-2 overexpress-
ing cells a steady-state [Ca
2+
]
er
level of 350 lm was
measured, compared to 450 lm of control cells
(Fig. 2A). Accordingly, when the cells were stimulated
with an IP
3
-generating agonist (ATP 100 lm), the
[Ca
2+
] increases evoked inthe cytoplasm and in mito-
chondria were significantly smaller (Fig. 2B,C). This
‘reduction’ of cellular Ca
2+
signals has a protective
effects toward a variety of inducers of cell death, such
as the lipid mediators ceramide or oxidative stress.
When ER was partially depleted of Ca
2+
independ-
ently of Bcl-2, e.g. by inhibiting the SERCA, over-
expressing the PMCA or reducing the [Ca
2+
] of the
extracellular medium, survival upon ceramide treat-
ment was markedly enhanced. As to the downstream
targets of the Ca
2+
effect, mitochondria seem to play
an important role. Ceramide treatment causes large
scale morphological rearrangements (fragmentation,
swelling), consistent with both apoptotic (release of
cytochrome c) and necrotic (bioenergetic dysfunction)
hallmarks and probably linked to the opening of the
permeability transition pore. Conversely, organelle
morphology is preserved if ceramide is applied to cells
in which [Ca
2+
]
er
is reduced by one of the experimen-
tal procedures described above. This is apparent from
the experiment shown in Fig. 2D, in which mitochon-
dria are visualized by transfected mtGFP in cells trea-
ted with ceramide while being maintained in KRB
A
B
C
D
Fig. 2. Upper panel: analysis of Ca
2+
homeostasis (using different targeted
aequorins. (A) erAEQ; (B) cytAEQ; (C)
mtAEQ) in control vs. Bcl-2 overexpressing
HeLa cells (modified from [41]). Lower
panel: effects of ceramide application on
mitochondrial structure at different
extracellular [Ca
2+
] (modified from [41]).
Mitochondrial calciumsignallingincelldeath S. Leo et al.
4016 FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS
supplemented with physiological (1 mm; left) or a
lower (0.05 mm; right) Ca
2+
concentration, which cau-
ses a partial Ca
2+
depletion of the ER. Overall, the
data indicate that the anti-apoptotic protein Bcl-2, by
reducing Ca
2+
signals evoked by physiological and
pathological stimuli, counteracts the efficacy of death
pathways acting through themitochondrial check-
point.
Interestingly, the link between Ca
2+
signalling and
cell death has been reinforced by the study of an unre-
lated pro-apoptotic protein, the x-protein of the hepa-
titis B virus (HBx), which also conceptually extended
the molecular mechanisms through which the Ca
2+
effect can be tuned [43]. In this case, when HeLa and
liver-derived HepG2 cells were transfected with HBx, a
marked enhancement of the cytosolic Ca
2+
responses
evoked by cell stimulation was detected (Fig. 3A). This
alteration (the opposite effect of Bcl-2) was not due to
an alteration in ER Ca
2+
handling (both the steady
state levels and the release kinetics were the same in
HBx-transfected and control cells), but rather to the
caspase-dependent cleavage of PMCA, the most effect-
ive molecular route for rapidly returning [Ca
2+
]
c
to
basal values (Fig. 3B). The Ca
2+
signalling alteration
leads to major morphological alterations of mitochon-
dria (Fig. 3C) and spontaneous apoptosis. Indeed cell
death was blocked by treatment with caspase-3 inhibi-
tors (100 lm ZVAD-fmk), loading of a Ca
2+
buffer
(2 lm BAPTA-AM) or overexpression of Bcl-2
(Fig. 3D).
On the cytosolic side, mitochondrial Ca
2+
uptake
exerts two different effects. Inthe first, the spatial
clustering of mitochondria in a defined portion of the
cell represents a physiological ‘fixed spatial buffer’ that
prevents (or delays) the spread of cytoplasmic Ca
2+
waves. In this case, mitochondria act as a ‘firewall’
that shields some cell domains from the Ca
2+
signal
elicited by submaximal agonist stimulation. This was
clearly shown in pancreatic acinar cells, in which only
high agonist doses (that overwhelm the mitochondrial
firewall) induce a [Ca
2+
]
c
rise inthe basal portion of
the cell, containing the nucleus [44].
The second mechanism through which mitochondria
may modulate cytosolic Ca
2+
transients refers to
events occurring insignalling microdomains, in which
mitochondria are placed in close contact with Ca
2+
channels that are under feedback control by Ca
2+
itself. In this case, mitochondrial Ca
2+
accumulation
participates in clearing Ca
2+
from the microenviron-
ment of the channel, thus reducing the (positive or
negative) feedback on channel activity. The first
demonstration of such an effect was obtained by
Lechleiter and coworkers in Xenopus oocytes, in which
the energization state (and thus the capacity to accu-
mulate Ca
2+
) was shown to influence the spatio-tem-
poral pattern of the typical propagating Ca
2+
waves
induced by IP
3
[45]. In mammals, several similar exam-
ples have been reported. In permeabilized hepatocytes
the decrease in [Ca
2+
]
er
evoked by submaximal IP
3
was enhanced when mitochondrial Ca
2+
uptake was
blocked [46]. Ca
2+
uptake by mitochondria thus sup-
presses the local positive feedback effects of Ca
2+
on
the IP
3
R, giving rise to subcellular heterogeneity in IP
3
sensitivity and IP
3
R excitability. Similarly, in astro-
AB
CD
Fig. 3. Effects of HBx overexpression on
Ca
2+
homeostasis (A,B), mitochondrial mor-
phology (C) and cell viability (D). See text for
details (modified from [43]).
S. Leo et al. Mitochondrialcalciumsignallingincell death
FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS 4017
cytes inhibition of mitochondrial Ca
2+
uptake almost
doubled the rate of propagation of thecalcium wave
across thecell [47]. In contrast, in BHK cells inhibition
of mitochondrial Ca
2+
uptake resulted in reduction of
ER Ca
2+
release [48], indicating that in this case
mitochondria play a major role in preventing the
Ca
2+
-depended inhibition of the InsP
3
channel. This
effect is not limited to ER Ca
2+
channels. Indeed, sev-
eral papers from the groups of Lewis and Parekh have
demonstrated that Ca
2+
uptake by energized mito-
chondria relieves the Ca
2+
-dependent inhibition of
Ca
2+
release-activated current (CRAC) channels, i.e.
those activated by the emptying of intracellular Ca
2+
stores [49,50]. This notion explains the high buffering
(mimicking mitochondrial activity) required to observe
I
CRAC
in many experimental conditions. A complex
role, somewhat similar to that proposed for the modu-
lation of capacitative Ca
2+
influx, has been proposed
by Nicholls and coworkers for the modulation of
Ca
2+
activation-inhibition of glutamate and voltage
operated Ca
2+
channels in cerebellar granule cells [51].
Specific regulatory pathways for
mitochondrial Ca
2+
homeostasis
Mitochondria can thus be regarded as critical check-
points in Ca
2+
signalling, acting as membrane-bound
Ca
2+
buffers, in which Ca
2+
itself plays a regulatory
role. In this situation, the possibility that their uptake
capacity (kinetics, amplitude) is tuned by converging
signalling pathways may add further complexity (and
option for regulation) to Ca
2+
-mediated signal trans-
duction.
We investigated this possibility, focusing on two
aspects: the role of the broad family of PKC kinases and
of the three-dimensional structure of mitochondria, in
turn controlled by the activity of large GTPases indu-
cing organelle fusion or fission (mitofusins, dynamin-
related protein-1).
PKC comprises a family of serine-threonine kinases
that are involved inthe transduction of a wide number
of extracellular signals. Based on their biochemical
properties, they are divided into classical (e.g. a, bI,
bII, c), activated by Ca
2+
and diacylglycerol, novel
(e.g. d, e, g, h), activated by diacylglycerol, and atyp-
ical (e.g. f, k), insensitive to both Ca
2+
and diacylglyc-
erol. Different isozymes are coexpressed inthe various
cell types giving rise to a highly flexible molecular rep-
ertoire, which can mediate radically different intra-
cellular effects, e.g. isoforms belonging to the same
group, such as d and e, have been reported to play
opposite effects on apoptosis. To evaluate specific
effects of PKC isoforms on cellular Ca
2+
homeostasis,
we overexpressed PKC–GFP chimeras in HeLa cells
and investigated agonist-dependent Ca
2+
signals in
the cytosol and mitochondria, using organelle-targeted
aequorin chimeras [52]. An interesting scenario
emerged, with distinct roles for the various isoforms.
Overexpression of PKCe did not modify the amplitude
and the kinetics of either the cytosolic and mitochond-
rial Ca
2+
transients evoked by histamine stimulation,
indicating that this PKC isoform does not influence
either Ca
2+
signalling globally inthecell or mito-
chondrial Ca
2+
accumulation. Conversely, PKCa
overexpression greatly reduces the agonist-evoked
[Ca
2+
] increase both inthe cytosol and inthe mito-
chondria. The monitoring of ER [Ca
2+
] demonstrated
that this is due to the sharp reduction of Ca
2+
release
from the organelle. These data are in keeping with pre-
vious demonstration that phorbol esters inhibited ER
Ca
2+
release [53], suggesting that PKCa is the isoform
responsible for this effect. As to the target, the demon-
stration of the phosphorylation of the IP
3
receptor by
PKC [54] indicates that the ER Ca
2+
release channel
itself is a plausible site of action for the PKCa effect.
Interestingly, some isoforms appear to have an
effect on mitochondrial, but not on cytosolic, Ca
2+
handling. Indeed, in cells overexpressing PKCb (and
to a smaller extent PKCd), themitochondrial but not
the cytosolic [Ca
2+
] rise is reduced. Conversely, over-
expression of PKCf enhances themitochondrial Ca
2+
responses to agonist stimulation, while still leaving
cytosolic [Ca
2+
] increases unaffected. These effects do
not depend on alterations of mitochondrial structure
(monitored by mtRFP labelling), an important deter-
minant of organelle responsiveness, nor to significant
changes inthe driving force for Ca
2+
accumulation
(the mitochondrial membrane potential, DY
m
). A pos-
sibility is that the Ca
2+
uptake machinery of
the organelle is directly modulated, but the demonstra-
tion of this possibility awaits the molecular characteri-
zation of the uptake process. As to the functional
significance of this regulatory mechanism, ‘mitochond-
rial Ca
2+
desensitization’, i.e. the sharp reduction of
[Ca
2+
]
m
peaks upon repetitive cell stimulation, was
proposed to be responsible for phenomena, such as
the down-regulation of insulin secretion in pancreatic
b-cells [55]. Interestingly, inhibition of PKCb greatly
reduces the [Ca
2+
]
m
reduction upon repetitive agonist
stimulation, indicating that it could represent the
molecular route for Ca
2+
-dependent inhibition of cel-
lular responses.
The other physiological regulation of mitochondria
that has been studied in relation to Ca
2+
signalling is
the state of fusion and fission. The study of mito-
chondrial structure, and of the mechanisms that
Mitochondrial calciumsignallingincelldeath S. Leo et al.
4018 FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS
dynamically control it in living cells, is a field of study
that literally boomed inthe past decade. Indeed,
although the idea that mitochondria can form an inter-
connected reticulum was shown by reconstructing elec-
tron microscopy images inthe late 1970s [56] and
more recently by using fluorescent GFP-based probes
and high-resolution digital imaging systems [26], it was
only after the identification of the molecules involved
that it became clear that the morphology of mitochon-
dria is tightly controlled by a dedicated cellular
machinery. Large GTPases, such as dynamin-related
protein-1 (Drp-1; a mechanoenzyme involved in mem-
brane constriction and fission), OPA1 (a dynamin rela-
ted GTPase, involved in fusion and mostly located in
the inner mitochondrial membrane) and mitofusins
(also involved in fusion and homologous to the first
element described, the ‘fuzzy onion’, fzo, protein of
Drosophila melanogaster [57]), and docking proteins,
such as Fis-1 (a transmembrane protein of the outer
mitochondrial membrane that participates in recruiting
Drp-1 to the organelle during organelle fragmenta-
tion). It is beyond the scope of this review to describe
in detail this fascinating field, and we refer to excellent
reviews on this topic [58–60]. Recent work gives some
insight into the dynamic regulation of the process, and
suggests that Ca
2+
could be involved: (a) Ca
2+
release
from the ER promotes the translocation of Drp-1 from
the cytoplasm to the outer mitochondrial membrane
[61]; (b) treatment with the Ca
2+
ionophore, A23187,
triggers mitochondrial fission [62]; and (c) a novel
group of rho-GTPases have been described (mitoch-
ondrial rho 1 and 2; miro-1 and miro-2), that promote
mitochondrial fusion and include EF-hand Ca
2+
bind-
ing sites in their sequence [63].
In our work, we took advantage of the molecular
knowledge of the fusion ⁄ fission machinery for actively
modifying the three-dimensional structure of mitochon-
dria, and verified the impact on organelle Ca
2+
hand-
ling. Specifically, given that in HeLa cells mitochondria
mostly form an interconnected network, we over-
expressed the fission protein Drp-1. As expected, this
caused massive fragmentation of themitochondrial net-
work, while leaving the ER morphology unaffected
(Fig. 4A). Studies carried out with targeted aequorins
showed that Ca
2+
release from the ER and the ensuing
[Ca
2+
]
c
increases are not modified by Drp-1-dependent
mitochondrial fragmentation; conversely, the [Ca
2+
]
m
peak was drastically reduced (Fig. 4B–D). Single-cell
analysis of [Ca
2+
]
m
dynamics, using the GFP-based
periCaM probe [64,65] and a high-speed imaging system
showed that in control cells intramitochondrial Ca
2+
waves originate from focal points and gradually diffuse
through themitochondrial network. These waves were
blocked inthe Drp-1-fragmented network, excluding
some individual mitochondria from the [Ca
2+
] rise and
thus reducing the average [Ca
2+
]
m
response. Conse-
quently, in Drp-1 overexpressing cells the apoptotic
efficacy of ceramide, which causes a Ca
2+
-dependent
perturbation of mitochondrial structure and function,
was drastically reduced [66].
AB
C
D
Fig. 4. Effects of Drp-1 overexpression on
organelle morphology (A), and Ca
2+
homeo-
stasis (B–D). See text for details (modified
from [66]).
S. Leo et al. Mitochondrialcalciumsignallingincell death
FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS 4019
Conclusions
A large body of experimental work shows that in vir-
tually every cell type the [Ca
2+
]
c
increases elicited in
the cytosol, by the opening of plasma membrane or
ER ⁄ SR Ca
2+
channels, is paralleled by a large
increase inthe [Ca
2+
] of themitochondrial matrix.
This process has two important functional conse-
quences. The first is that part of the Ca
2+
entering the
cytoplasm is rapidly removed. Mitochondria thus act
as Ca
2+
buffers, and this activity influences both the
local microdomain (hence affecting local inhibitory or
activatory effects of Ca
2+
itself onthe channel) or the
kinetics of the diffusion across the cytosol (thus acting
as a barrier limiting the [Ca
2+
]
c
rise to one portion of
the cell). Mitochondrial role is not limited, however, to
Ca
2+
buffering. Indeed, within the organelle Ca
2+
can
regulate functions as diverse as aerobic metabolism
(through Ca
2+
sensitive enzymes and metabolite trans-
porters) and celldeath (probably by activating the per-
meability transition pore). Given this high functional
plasticity, it is not surprising that we start to obtain
evidence that different mechanisms can finely tune
amplitude and kinetics of themitochondrial Ca
2+
responses. We have described a few intriguing exam-
ples (PKC, the fusion ⁄ fission machinery), but it is rea-
sonable to predict that inthe near future we will learn
much about these signalling routes and their cross-talk,
and get to know (finally!) themitochondrial targets,
namely the channels, the regulatory elements and the
scaffolding proteins.
Acknowledgements
Experimental work inthe authors’ laboratory was sup-
ported by grants from the Italian Ministry of Education
(FIRB, PRIN, local interest grants), the PRRIITT pro-
gram of the Emilia-Romagna region (ER-GenTech),
Telethon-Italy, the Italian Association for Cancer
Research (AIRC) and the Italian Space Agency (ASI).
References
1 Berridge MJ, Bootman MD & Roderick HL (2003) Cal-
cium signalling: dynamics, homeostasis and remodelling.
Nat Rev Mol Cell Biol 4 , 517–529.
2 Tsien RY, Pozzan T & Rink TJ (1982) Calcium homeo-
stasis in intact lymphocytes: cytoplasmic free calcium
monitored with a new, intracellularly trapped fluores-
cent indicator. J Cell Biol 94, 325–334.
3 Pozzan T, Rizzuto R, Volpe P & Meldolesi J (1994)
Molecular and cellular physiology of intracellular cal-
cium stores. Physiol Rev 74, 595–636.
4 Petersen OH, Tepikin A & Park MK (2001) The endo-
plasmic reticulum: one continuous or several separate
Ca(2+) stores? Trends Neurosci 24, 271–276.
5 Montell C (2001) Physiology, phylogeny, and functions
of the TRP superfamily of cation channels. Sci. STKE
2001, re1.
6 Parekh AB & Putney JW Jr (2005) Store-operated cal-
cium channels. Physiol Rev 85, 757–810.
7 Orrenius S, Zhivotovsky B & Nicotera P (2003) Regula-
tion of cell death: the calcium-apoptosis link. Nat Rev
Mol Cell Biol 4, 552–565.
8 Carafoli E (2004) Calcium-mediated cellular signals: a
story of failures. Trends Biochem Sci 29, 371–379.
9 Rizzuto R & Pozzan T (2003) When calcium goes
wrong: genetic alterations of a ubiquitous signaling
route. Nat Genet 34, 135–141.
10 Duchen MR (2000) Mitochondria and calcium: from
cell signalling to cell death. J Physiol 529 Part 1 , 57–68.
11 Dolmetsch R (2003) Excitation-transcription coupling:
signaling by ion channels to the nucleus. Sci STKE
2003, pe4.
12 Woods NM, Cuthbertson KS & Cobbold PH (1986)
Repetitive transient rises in cytoplasmic free calcium in
hormone-stimulated hepatocytes. Nature 319, 600–602.
13 Rizzuto R, Simpson AW, Brini M & Pozzan T (1992)
Rapid changes of mitochondrial Ca
2+
revealed by spe-
cifically targeted recombinant aequorin. Nature 358,
325–327.
14 Montero M, Brini M, Marsault R, Alvarez J, Sitia R,
Pozzan T & Rizzuto R (1995) Monitoring dynamic
changes in free Ca
2+
concentration inthe endoplasmic
reticulum of intact cells. EMBO J 14, 5467–5475.
15 Barrero MJ, Montero M & Alvarez J (1997) Dynamics
of [Ca
2+
] inthe endoplasmic reticulum and cytoplasm
of intact HeLa cells: a comparative study. J Biol Chem
272, 27694–27699.
16 Pinton P, Pozzan T & Rizzuto R (1998) The Golgi
apparatus is an inositol 1,4,5-trisphosphate-sensitive
Ca
2+
store, with functional properties distinct from
those of the endoplasmic reticulum. EMBO J 17, 5298–
5308.
17 Brini M, Marsault R, Bastianutto C, Pozzan T &
Rizzuto R (1994) Nuclear targeting of aequorin: a new
approach for measuring nuclear Ca
2+
concentration in
intact cells. CellCalcium 16, 259–268.
18 Marsault R, Murgia M, Pozzan T & Rizzuto R (1997)
Domains of high Ca
2+
beneath the plasma membrane
of living A7r5 cells. EMBO J 16, 1575–1581.
19 Carafoli E (2003) Historical review: mitochondria and
calcium: ups and downs of an unusual relationship.
Trends Biochem Sci 28, 175–181.
20 Kirichok Y, Krapivinsky G & Clapham DE (2004) The
mitochondrial calcium uniporter is a highly selective ion
channel. Nature 427, 360–364.
Mitochondrial calciumsignallingincelldeath S. Leo et al.
4020 FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS
21 Cox DA & Matlib MA (1993) Modulation of intramito-
chondrial free Ca
2+
concentration by antagonists of
Na
+
–Ca
2+
exchange. Trends Pharmacol Sci 14, 408–
413.
22 Streb H, Irvine RF, Berridge MJ & Schulz I (1983)
Release of Ca
2+
from a nonmitochondrial intracellular
store in pancreatic acinar cells by inositol-1,4,5-trisphos-
phate. Nature 306, 67–69.
23 Rizzuto R, Bastianutto C, Brini M, Murgia M &
Pozzan T (1994) Mitochondrial Ca
2+
homeostasis in
intact cells. J Cell Biol 126, 1183–1194.
24 Montero M, Alonso MT, Carnicero E, Cuchillo-Ibanez
I, Albillos A, Garcia AG, Garcia-Sancho J & Alvarez J
(2000) Chromaffin-cell stimulation triggers fast millimo-
lar mitochondrial Ca
2+
transients that modulate secre-
tion. Nat Cell Biol 2, 57–61.
25 Rizzuto R, Brini M, Murgia M & Pozzan T (1993)
Microdomains with high Ca
2+
close to IP3-sensitive
channels that are sensed by neighboring mitochondria.
Science 262, 744–747.
26 Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty
KE, Lifshitz LM, Tuft RA & Pozzan T (1998) Close
contacts with the endoplasmic reticulum as determinants
of mitochondrial Ca
2+
responses. Science 280, 1763–
1766.
27 Brini M, De Giorgi F, Murgia M, Marsault R,
Massimino ML, Cantini M, Rizzuto R & Pozzan T
(1997) Subcellular analysis of Ca
2+
homeostasis in
primary cultures of skeletal muscle myotubes. Mol Biol
Cell 8, 129–143.
28 Rudolf R, Mongillo M, Magalhaes PJ & Pozzan T
(2004) In vivo monitoring of Ca
2+
uptake into mito-
chondria of mouse skeletal muscle during contraction.
J Cell Biol 166, 527–536.
29 McCormack JG, Halestrap AP & Denton RM (1990)
Role of calcium ions in regulation of mammalian
intramitochondrial metabolism. Physiol Rev 70, 391–
425.
30 Hansford RG (1994) Physiological role of mitochondrial
Ca
2+
transport. J Bioenerg Biomembr 26, 495–508.
31 Jouaville LS, Pinton P, Bastianutto C, Rutter GA &
Rizzuto R (1999) Regulation of mitochondrial ATP
synthesis by calcium: evidence for a long-term metabolic
priming. Proc Natl Acad Sci USA 96, 13807–13812.
32 Palmieri L, Pardo B, Lasorsa FM, del Arco A,
Kobayashi K, Iijima M, Runswick MJ, Walker JE,
Saheki T, Satrustegui J et al. (2001) Citrin and aralar1
are Ca
2+
-stimulated aspartate ⁄ glutamate transporters in
mitochondria. EMBO J 20, 5060–5069.
33 Lasorsa FM, Pinton P, Palmieri L, Fiermonte G,
Rizzuto R & Palmieri F (2003) Recombinant expression
of the Ca
2+
-sensitive aspartate ⁄ glutamate carrier increa-
ses mitochondrial ATP production in agonist-stimulated
Chinese hamster ovary cells. J Biol Chem 278, 38686–
38692.
34 Szalai G, Krishnamurthy R & Hajnoczky G (1999)
Apoptosis driven by IP(3)-linked mitochondrial calcium
signals. EMBO J 18, 6349–6361.
35 Pinton P, Ferrari D, Di Virgilio F, Pozzan T & Rizzuto
R (2001) Molecular machinery and signalling events in
apoptosis. Drug Dev Res 52, 558–570.
36 Pinton P, Ferrari D, Rapizzi E, Di Virgilio FD, Pozzan
T & Rizzuto R (2001) The Ca
2+
concentration of the
endoplasmic reticulum is a key determinant of cera-
mide-induced apoptosis: significance for the molecular
mechanism of Bcl-2 action. EMBO J 20, 2690–2701.
37 Kluck RM, Bossy-Wetzel E, Green DR & Newmeyer
DD (1997) The release of cytochrome c from mitochon-
dria: a primary site for Bcl-2 regulation of apoptosis.
Science 275, 1132–1136.
38 Yang J, Liu X, Bhalla K, Kim CN, Ibrado AM, Cai J,
Peng TI, Jones DP & Wang X (1997) Prevention of
apoptosis by Bcl-2: release of cytochrome c from mito-
chondria blocked. Science 275, 1129–1132.
39 Susin SA, Lorenzo HK, Zamzami N, Marzo I, Snow
BE, Brothers GM, Mangion J, Jacotot E, Costantini P,
Loeffler M et al. (1999) Molecular characterization of
mitochondrial apoptosis-inducing factor. Nature 397,
441–446.
40 Du C, Fang M, Li Y, Li L & Wang X (2000) Smac,
a mitochondrial protein that promotes cytochrome
c-dependent caspase activation by eliminating IAP inhi-
bition. Cell 102, 33–42.
41 Pinton P, Ferrari D, Magalhaes P, Schulze-Osthoff K,
Di Virgilio F, Pozzan T & Rizzuto R (2000) Reduced
loading of intracellular Ca
2+
stores and downregulation
of capacitative Ca(2+) influx in Bcl-2-overexpressing
cells. J Cell Biol 148, 857–862.
42 Foyouzi-Youssefi R, Arnaudeau S, Borner C, Kelley
WL, Tschopp J, Lew DP, Demaurex N & Krause KH
(2000) Bcl-2 decreases the free Ca
2+
concentration
within the endoplasmic reticulum. Proc Natl Acad Sci
USA 97, 5723–5728.
43 Chami M, Ferrari D, Nicotera P, Paterlini-Brechot P
& Rizzuto R (2003) Caspase-dependent alterations of
Ca
2+
signaling inthe induction of apoptosis by hepa-
titis B virus X protein. J Biol Chem 278, 31745–
31755.
44 Tinel H, Cancela JM, Mogami H, Gerasimenko JV,
Gerasimenko OV, Tepikin AV & Petersen OH (1999)
Active mitochondria surrounding the pancreatic acinar
granule region prevent spreading of inositol trispho-
sphate-evoked local cytosolic Ca(2+) signals. EMBO J
18, 4999–5008.
45 Jouaville LS, Ichas F, Holmuhamedov EL, Camacho P
& Lechleiter JD (1995) Synchronization of calcium
waves by mitochondrial substrates in Xenopus laevis
oocytes. Nature 377, 438–441.
46 Hajnoczky G, Hager R & Thomas AP (1999) Mito-
chondria suppress local feedback activation of inositol
S. Leo et al. Mitochondrialcalciumsignallingincell death
FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS 4021
1,4, 5-trisphosphate receptors by Ca
2+
. J Biol Chem
274, 14157–14162.
47 Boitier E, Rea R & Duchen MR (1999) Mitochondria
exert a negative feedback onthe propagation of intra-
cellular Ca
2+
waves in rat cortical astrocytes. J Cell
Biol 145, 795–808.
48 Landolfi B, Curci S, Debellis L, Pozzan T & Hofer AM
(1998) Ca
2+
homeostasis inthe agonist-sensitive inter-
nal store: functional interactions between mitochondria
and the ER measured in situ in intact cells. J Cell Biol
142, 1235–1243.
49 Hoth M, Fanger CM & Lewis RS (1997) Mitochondrial
regulation of store-operated calcium signaling in T lym-
phocytes. J Cell Biol 137, 633–648.
50 Gilabert JA & Parekh AB (2000) Respiring mitochon-
dria determine the pattern of activation and inactivation
of the store-operated Ca
2+
current I (CRAC). EMBO J
19, 6401–6407.
51 Budd SL & Nicholls DG (1996) Mitochondria, calcium
regulation, and acute glutamate excitotoxicity in cultured
cerebellar granule cells. J Neurochem 67, 2282–2291.
52 Pinton P, Leo S, Wieckowski MR, Di Benedetto G &
Rizzuto R (2004) Long-term modulation of mitochon-
drial Ca
2+
signals by protein kinase C isozymes. J Cell
Biol 165, 223–232.
53 Montero M, Lobaton CD, Gutierrez-Fernandez S,
Moreno A & Alvarez J (2003) Modulation of histamine-
induced Ca
2+
release by protein kinase C. Effects on
cytosolic and mitochondrial [Ca
2+
] peaks. J Biol Chem
278, 49972–49979.
54 Vermassen E, Fissore RA, Nadif KN, Vanderheyden V,
Callewaert G, Missiaen L, Parys JB & De Smedt H
(2004) Regulation of the phosphorylation of the inositol
1,4,5-trisphosphate receptor by protein kinase C. Bio-
chem Biophys Res Commun 319, 888–893.
55 Maechler P, Kennedy ED, Wang H & Wollheim CB
(1998) Desensitization of mitochondrial Ca
2+
and insu-
lin secretion responses inthe beta cell. J Biol Chem 273,
20770–20778.
56 Bakeeva LE, Chentsov Y & Skulachev VP (1978)
Mitochondrial framework (reticulum mitochondriale) in
rat diaphragm muscle. Biochim Biophys Acta 501, 349–
369.
57 Hales KG & Fuller MT (1997) Developmentally regu-
lated mitochondrial fusion mediated by a conserved,
novel, predicted GTPase. Cell 90, 121–129.
58 Bossy-Wetzel E, Barsoum MJ, Godzik A,
Schwarzenbacher R & Lipton SA (2003) Mitochondrial
fission in apoptosis, neurodegeneration and aging. Curr
Opin Cell Biol 15, 706–716.
59 Scorrano L (2003) Divide et impera: Ca
2+
signals, mit-
ochondrial fission and sensitization to apoptosis. Cell
Death Differ 10, 1287–1289.
60 Perfettini JL, Roumier T & Kroemer G (2005) Mito-
chondrial fusion and fission inthe control of apoptosis.
Trends Cell Biol 15, 179–183.
61 Breckenridge DG, Stojanovic M, Marcellus RC & Shore
GC (2003) Caspase cleavage product of BAP31 induces
mitochondrial fission through endoplasmic reticulum
calcium signals, enhancing cytochrome c release to the
cytosol. J Cell Biol 160, 1115–1127.
62 Duncan CJ, Greenaway HC, Publicover SJ, Rudge MF
& Smith JL (1980) Experimental production of
‘septa’ and apparent subdivision of muscle
mitochondria. J Bioenerg Biomembr 12, 13–33.
63 Fransson A, Ruusala A & Aspenstrom P (2003) Atypi-
cal Rho GTPases have roles inmitochondrial homeosta-
sis and apoptosis. J Biol Chem 278, 6495–6502.
64 Nagai T, Sawano A, Park ES & Miyawaki A (2001)
Circularly permuted green fluorescent proteins engi-
neered to sense Ca
2+
. Proc Natl Acad Sci USA 98,
3197–3202.
65 Filippin L, Magalhaes PJ, Di Benedetto G, Colella M
& Pozzan T (2003) Stable interactions between mito-
chondria and endoplasmic reticulum allow rapid accu-
mulation of calciumin a subpopulation of
mitochondria. J Biol Chem 278, 39224–39234.
66 Szabadkai G, Simoni AM, Chami M, Wieckowski MR,
Youle RJ & Rizzuto R (2004) Drp-1-dependent division
of themitochondrial network blocks intraorganellar
Ca
2+
waves and protects against Ca
2+
-mediated apop-
tosis. Mol Cell 16, 59–68.
Mitochondrial calciumsignallingincelldeath S. Leo et al.
4022 FEBS Journal 272 (2005) 4013–4022 ª 2005 FEBS
. THE THEODOR BU ¨ CHER LECTURE Mitochondrial calcium signalling in cell death Delivered on 1 July 2004 at the 29th FEBS Congress in Warsaw Sara Leo 1 , Katiuscia Bianchi 1 , Marisa Brini 2 and. 2005 FEBS 4 017 cytes inhibition of mitochondrial Ca 2+ uptake almost doubled the rate of propagation of the calcium wave across the cell [47]. In contrast, in BHK cells inhibition of mitochondrial. map of the mtAEQ construct (aequorin targeted to the mitochondrial matrix) and its localization by immunofluores- cence. Mitochondrial calcium signalling in cell death S. Leo et al. 4 014 FEBS Journal