Atypical mitochondrial fission upon bacterial infection
We recently showed that infection by Listeria monocytogenes causes mitochondrial network fragmentation through the secreted pore-forming toxin listeriolysin O (LLO). Here, we examine factors involved in canonical fusion and fission. Strikingly, LLO-induced mitochondrial fragmentation does not require the traditional fission machinery, as Drp1 oligomers are absent from fragmented mitochondria following Listeria infection or LLO treatment, as the dynamin-like protein 1 (Drp1) receptor Mff is rapidly degraded, and as fragmentation proceeds efficiently in cells with impaired Drp1 function. LLO does not cause processing of the fusion protein optic atrophy protein 1 (Opa1), despite inducing a decrease in the mitochondrial membrane potential, suggesting a unique Drp1- and Opa1-independent fission mechanism distinct from that triggered by uncouplers or the apoptosis inducer staurosporine. We show that the ER marks LLO-induced mitochondrial fragmentation sites even in the absence of functional Drp1, demonstrating that the ER activity in regulating mitochondrial fission can be induced by exogenous agents and that the ER appears to regulate fission by a mechanism independent of the canonical mitochondrial fission machinery.
Atypical mitochondrial ﬁssion upon bacterial infection
, Amy E. Palmer
, Chunxin Wang
, Richard J. Youle
, and Pascale Cossart
Unité des Interactions Bactéries-Cellules, Institut Pasteur, 75015 Paris, France;
Institut National de la Santé et de la Recherche Médicale, Unité 604, 75015
Institut National de la Recherche Agronomique Unité Sous Contrat 2020, 75015 Paris, France; and
Biochemistry Section, Surgical
Neurology Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD 20892
Contributed by Pasca le Cossart, August 21, 2013 (sent for review July 16, 2013)
We recently showed that infection by Listeria monocytogenes
causes mitochondrial network fragmentation through the se-
creted pore-forming toxin listeriolysin O (LLO). Here, we examine
factors in volved in canonical fusion and ﬁssion.Strikingly,LLO-
induced mitochondrial fragmentation does not require the traditional
ﬁssion machinery, as Drp1 oligomers are absent from fragmented
mitochondria following Listeria infection or LLO treatment, as the
dynamin-like protein 1 (Drp1) receptor Mff is rapidly degraded, and
as fragmentation proceeds efﬁciently in cells with impaired Drp1
function. LLO does not cause processing of the fusion protein optic
atrophy protein 1 (Opa1), despite inducing a decrease in the mito-
chondrial membrane potential, sugg esting a unique Drp1- and
Opa1-independent ﬁssion mechanism distinct from that triggered
by uncouplers o r t he ap opto sis inducer stauros pori ne . We sh ow
that the ER marks LLO-induced mitochondrial fragmentation sites
even in the absence of functional Drp1, demonstrating that the
ER activity in regulating mitochondrial ﬁssion can be induce d by
exogenous agents and that the ER appears t o regulate ﬁssion by
a mechanism independent of the canonical mitochondrial ﬁssion
live cell imaging
itochondria are essential organelles that perform a multi-
tude of functions, ranging from the production of biosyn-
thetic intermediates and energy to innate immune signaling and
cellular calcium buffering or the sto rage of proapoptotic com-
ponents (1). To perform these di verse functions, mitochondria
respond to cellular cues and display a highly variable and dynamic
morphology, constantly undergoing fusion and ﬁssion. It is be-
coming increasingly clear that mitochondrial dynamics and func-
tion are deeply interconnected, and mitochondrial dysfunction is
associated with a range of diseases.
Wild-type mitochondrial morphology and function are main-
tained by a balance between mitochondrial fusion and ﬁssion.
Fusion allows exchange of genetic material between single mi-
tochondria and is mediated by two large guanosine triphosphate
phosphohydrolases (GTPases) embedded in the outer membrane
(mitofusin 1 and 2) and an inner membrane GTPase, Opa1 (2).
Deletion mutants affecting these three proteins accumulate dys-
functional mitochondria, leading to neurodegenerative phenotypes
and different forms of myopathy (1, 3).
Mitochondrial fusion is balanced by ﬁssion, which is essential
to ensure proper distribution of mitochondria and energy supply
to daughter cells in mitosis or within a single cell. This necessity
is particularly evident in neurons, where ﬁssion defects prevent
efﬁcient mitochondrial transport to synapses, the crucial sites of
energy consumption (4, 5). The physiological importance of
mitochondrial ﬁssion is further highlighted by its essential role in
embryonic development in mice and nematodes (6–8).
Mitochondrial ﬁ ssion is thought to be accomplished by the
dynamin-like prote in Drp1, a mainly cytosolic protei n that is
recruitedtofutureﬁssion sites, where it oligomerizes to form
spirals that constrict mitochondria. Mitochondrial ﬁssion is regulated
at several levels: by initial ER- and actin-mediated mitochondrial
constriction (9, 10), leading to the accumulation of the membrane-
bound Drp1 receptor Mff and by several posttranslational mod-
iﬁcations of Drp1, which modulate its activity (11).
Listeria monocytogenes is a foodborne pathogen capable of
invading nonphagocytic cells, where it can replicate and spread.
The pathogenic potential of L. monocytogenes correlates with the
expression of several virulence genes (12). One of the most im-
portant virulence factors is listeriolysin O (LLO), a highly reg-
ulated secreted pore-forming toxin (reviewed in ref. 13). LLO
belongs to the family of cholesterol-dependent cytolysins (CDCs),
most of which are produced by extracellular bacteria such as
Streptococci or Clostridia. CDCs oligomerize on cholesterol-con-
taining membranes to form nonselective ion-permeable pores of
variable sizes (14) that act in concert with bacterial phospholipases
to allow bacterial escape from thephagosome.Morerecently,LLO
has been found to have several intracellular and extracellular roles
that extend beyond phagosomal escape. For example, we have
shown that infection with L. monocytogenes causes fragmentation of
the host mitochondrial network by action of its pore-forming toxin
LLO before bacterial entry (15).
In this study, we demonstrate that LLO-induced mitochondrial
fragmentation does not follow canonical pathways, because it is
independent of key fusion and ﬁssion components, such as Opa1
and Drp1. We demonstrate that the ER marks mitochondrial
fragmentation sites even in the absence of functional Drp1, and
that the actin cytoskeleton also facilitates fragmentation. LLO-
induced fragmentation is distinct from that observed upon treatment
with uncouplers [such as carbonyl cyanide m-chlorophenylhydrazone
(CCCP)] and apoptosis inducers (such as staurosporine), revealing
a unique pathway for mitochondrial fragmentation that can be in-
duced by an exogenous agent.
Mitochondria are dynamic organelles that constantly fuse and
fragment while acting as central hubs of energy production,
apoptosis regulation, and Ca
signaling, therefore emerging
as potential targets of pathogens. We previously showed that
the foodborne bacterial pathogen Listeria monocytogenes
interferes with the dynamics and function of the host cell mi-
tochondrial network via the bacterial toxin listeriolysin O (LLO).
In this study, we analyze the effects of LLO on key players
known to be involved in mitochondrial dynamics and show
that the ﬁssion protein dynamin-like protein 1 (Drp1) is not
essential for LLO-induced fragmentation of the mitochondrial
network, whereas the endoplasmic reticulum (ER) plays an
important role, suggesting a unique Drp1-independent and
ER-dependent mechanism that is different from the canonical
Author contributions: F.S., A.E.P., and P.C. designed research; F.S. and A.E.P. performed
research; C.W. and R.Y. contributed new reagents/analytic tools; F.S., A.E.P., and P.C.
analyzed data; and F.S., A.E.P., R.Y., and P.C. wrote the paper.
The authors declare no conﬂict of interest.
Freely available online through the PNAS open access option.
To whom correspondence may be addressed. E-mail: firstname.lastname@example.org or fabrizia.
Permanent address: Department of Chemistry and Biochemistry and BioFrontiers Insti-
tute, University of Colorado, Boulder, CO 80309.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
October 1, 2013
Results and Discussion
Opa1 Is Not Processed upon LLO-Induced Mitochondrial Fragmentation.
Mitochondrial fragmentation can result either from an increase in
ﬁssion activity, an inhibition of fusion, or both. One possible
pathway for mitochondrial fragmentation is rapid proteolytic
processing of the inner membrane GTPase Opa1, as observed
upon uncoupler [CCCP or triﬂuorocarbonyl cyanidephenylhy-
drazone ( FCCP)]-induced mitochondrial fragmentation, which
results in a fusion block and an ensuing mitochondrial frag-
mentation due to unopposed ﬁssion events (16– 18). In contrast
to CCCP, Opa1 is not processed upon LLO treatment (Fig. 1A),
indicating that LLO-induced mitochondrial fragmentation is not
caused by a block in inner membrane fusion. This ﬁnding is con-
sisten t with the rapid ﬁssion kinetics that we have described, which
suggests an active ﬁssion mechanism (15). We therefore addressed
whether LLO induces fragmentation by recruiting the key ﬁssion
Mitochondrial Drp1 Oligomers Dissociate upon Infection or LLO
Treatment. Fission-inducing agents such as uncouplers or the
apoptosis inducer staurosporine appear to act by recruiting Drp1
to fragmenting mitochondria (Fig. 1C; refs. 19–21). In contrast,
LLO treatment and L. monocytogenes infection were character-
ized by a decrease in Drp1 puncta associated with mitochondria
(Fig. 1 C and D and Fig. S1A). This phenomenon was unique to
LLO and L. monocytogenes infection because a number of other
ﬁssion-inducing agents (the protonophore FCCP, the potassium
ionophore valinomycin, and the detergent digitonin) did not
show a decrease in mitochondria-associated Drp1 (Fig. S1B). In-
terestingly, potassium accumulation into mitochondria was
invoked by Dimmer et al. to explain Drp1-independent frag -
mentation upon silencing of the inner membrane protein
LETM1, because the potassium ionophore nigericin could re-
store tubular morphology of mitochondria in these cells (22). LLO
may cause an intramitochondrial potassium imbalance that could
impinge on mitochondrial morphology, although it does not ex-
plain the dissociation of Drp1 from mitochondria. Western blot
analysis of cell extracts showed no decrease in total Drp1 levels in
LLO-treated (Fig. 1B and Fig. S1C) cells. Use of two different
antibodies against endogenous Drp1 on ﬁxed samples (Fig. S1A)
and live cell imaging of overexpressed Drp1-GFP (23) (Fig. S1D)
conﬁrmed a strong decrease in mitochondria-associated Drp1
staining upon LLO treatment.
LLO Effect on the Drp1 Receptors Fis1 and Mff. Drp1 requires several
membrane-anchored receptor proteins to bind to the mitochondrial
surface. We thus investigated whether the LLO-induced decrease
in mitochondria-associated Drp1 staining was due to modulation of
its receptors Fis1 (24–26) and Mff (27) at the mitochondrial
Upon LLO treatment or infection with L. monocytogenes, there
was no detectable decrease in the association of Fis1 with mito-
chondria by immunoﬂuorescence or total Fis1 protein levels by
Western blotting (Fig. S2 A and B), suggesting that loss of mi-
tochondrial Drp1 does not depend on a decrease in Fis1. In
contrast, LLO treatment and L. monocytogenes infection induced
a decrease in Mff protein as observed in total cell extracts by
Western blotting and at the single-cell level by immunoﬂuores-
cence and live-cell imaging (Fig. 2). The level of Mff mRNA was
unaltered, suggesting posttranscriptional modulation of Mff (Fig.
S2C). Mff did not decrease upon uncoupling (Fig. 2 A and B)nor
when LLO treatment was performed in the absence of extracel-
lular calcium (Fig. S2D). However, ionophore-induced calcium
inﬂux was not sufﬁcient to induce Mff degradation (Fig. S2D),
suggesting that calcium inﬂux alone is not sufﬁcient to cause Mff
degradation upon LLO treatment. LLO-induced degradation of
Mff appeared proteasome independent, because it was not pre-
vented by treatment with MG132 and lactacystin (Fig. S2E),
Fig. 1. Analysis of Opa1 and Drp1 upon LLO-induced mitochondrial frag-
mentation. (A) Western blot of HeLa cells treated for 1, 5, or 10 min with
6 nM LLO or for 10 min with 10 μM CCCP and probed for Opa1, which is
readily processed upon CCCP, but not LLO treatment. Tubulin was used as
a loading control. (B) HeLa cells were treated for 30 min with 10 μM CCCP,
for 1.5 h with 1.2 μM staurosporine (sts), or with 6 nM LLO for the indicated
timepoints, and total cell lysates are analyzed by Western blotting, showing
no apparent decrease in total Drp1 levels upon LLO or drug treatment.
GAPDH was used as a loading control. (C) Immunoﬂuorescence labeling of
the same cells analyzed in B. Drp1 is shown in green (Drp1NT), mitochondria
in red (Tom20), and nuclei in blue (DAPI). Staining reveals that mitochondrial
Drp1 oligomers are lost upon LLO treatment (decrease in green puncta) but
are retained upon CCCP or staurosporine treatment. (D) HeLa cells were
infected for 1 h with L. monocytogenes or L.monocytogenes Δhly (LLO
an MOI of 50 and stained for Drp1 (green), mitochondria (CoxIV; red), and
nuclei/bacteria with DAPI (blue). Wild-type L. monocytogenes induces mi-
tochondrial fragmentation and loss of mitochondria-associated Drp1,
whereas infection with the Δhly mutant does not overtly affect Drp1. For C
and D, the white box indicates a 2× enlarged region, which is shown below.
(Scale bars: 10 μm.)
www.pnas.org/cgi/doi/10.1073/pnas.1315784110 Stavru et al.
consistent with previous reports suggesting that LLO induces pro-
tein d egradation through an as-yet-unknown pathway (28, 29).
Timelapse studies of Mff-GFP in HeLa cells treated with LLO in-
dicate that Mff degradation begins almost immediately after LLO
addition (Fig. 2D and Fig. S2F) and before mitochondrial frag -
mentation. Indeed, LLO induced mitochondrial fragmentation in
HCT116 cells (Fig. S2G). Given that Mff serves as one of the
primary receptors for anchoring Drp1 at mitochondria, this result
indicates Drp1 may be lost from mitochondria before fragmentation,
further suggesting that Drp1 is not required at fragmentation sites.
LLO-Induced Mitochondrial Fragmentation Is Drp1 Independent. Drp1
is described as the key and essential factor mediating mitochon-
drial ﬁssion. Given that we observed a decrease rather than an
augmentation of the Drp1 signal at mitochondria upon infection
or LLO treatment, we tested whether Drp1 is essential for LLO-
induced mitochondrial fragmentation by using dominant-negative
K38A Drp1 in HeLa cells, Drp1
mouse embryonic ﬁbroblasts
(MEFs) (6), and HCT116 Drp1
cells. Strikingly, there was
no difference in mitochondrial fragmentation compared with
controls in cells transfected with K38A Drp1, as well as both
Drp1 knockout cell lines (Fig. 3), indicating that LLO- and
L. monocytogenes-induced mitochondrial fragmentation is truly
Drp1 independent. In contrast, CCCP-induced mitochondrial
fragmentation was substantially reduced in the Drp1 knockout cell
lines. Staining of ﬁxed cells, and live cell imaging with ﬂuorescent
protein-tagged markers, with both matrix and outer-membrane
markers conﬁrmed complete abscission of mitochondria upon
LLO treatment (Fig. 3C and Fig. S3 A and B). Furthermore, upon
incubation of wild-type and knockout cells with LLO for different
amounts of time, similar percentages of cells with fragmented
mitochondria were scored, indicating that the loss of Drp1 did not
delay the kinetics of LLO-induced mitochondrial fragmentation
The Drp1 dependence of mitochondrial ﬁssion induced by
uncouplers or activators of apoptosis is a matter of debate. For
example, it has been demonstrated that transfection of K38A
Drp1 does not block uncoupling-induced formation of disk-
shaped mitochondria (30), and deletion of Drp1 does not affect
the number of cells with fragmented mitochondria upon in-
duction of apoptosis (8). However, several other groups found
that overexpression of K38A Drp1 or Drp1 depletion prevented
or diminished apoptosis- or CCCP-induced mitochondrial frag-
mentation (6, 31–33). It has been suggested that this discrepancy
may arise because fragmentation depends both on the cell type
and the physiological cue used to induce apoptosis (6). In con-
trast, we see no measureable perturbation of fragmentation upon
knock out of Drp1 or expression of dominant-negative Drp1,
indicating that fragmentation induced by LLO is truly independent
of functional Drp1. Moreover, LLO-induced fragmentation is not
associated with cytochrome c release and activation of canonical
apoptosis (15), suggesting that this mechanism of fragmentation
is distinct from the one that activates uncouplers or apoptosis
induc ers. Opening of the mitochondrial transition pore (MTP)
may be in voked to explain LLO-induced mitochondrial frag-
menta tion. We do not favor this hypothesis because of previous
experiments conﬁrming that the mitochondrial inner membrane
remains intact upon LLO treatment, as respiration can resume
(15). Furthermor e, we could not inhib it the fragmentation
phenotype with the MTP inhibitor cyclosporine A (15).
The ER Marks Sites of Mitochondrial Fragmentation Even in the
Absence of Functional Drp1.
It has recently been shown that mi-
tochondrial ﬁssion occurs at contact sites with the ER, and that
mitochondria are constricted at such contact sites (9). Although
the mechanism of how the ER regulates ﬁssion was not identi-
ﬁed, it was suggested that ER proteins may participate in ﬁssion
and/or that mitochondrial constriction by the ER facilitates re-
cruitment of the mitochondrial ﬁssion machinery (i.e., Drp1).
We thus explored whether LLO-induced and Drp1-independent
mitochondrial fragmentation was marked by ER-mitochondrial
contact sites. Timelapse studies of U2OS epithelial cells expressing
the ER marker Sec61β-GFP (34) upon treatment with LLO
revealed that the ER marked sites of mitochondrial fragmenta-
tion in 87 ± 8% of events (Fig. 4). Transfection of U2OS cells
with K38A Drp1 or siRNA to knockdown Drp1 did not prevent
LLO-induced mitochondrial fragmentation. Intriguingly, there
was no change in the percent of fragmentations marked by ER-
mitochondria contact sites (96 ± 6% for K38A; 82 ± 6% for
siRNA), demonstrating that the ER plays an active role in reg-
ulating mitochondria ﬁssion even in the absence of Drp1.
Actin Is Involved in LLO-Induced Mitochondrial Fragmentation. Actin
has been shown to play a role in drug-induced mitochondrial
ﬁssion in mammalian cells, although the effect appears to be
stimulus dependent because oligomycin and cyclosporine A, but
not CCCP, depended on F-actin for efﬁcient mitochondrial
network fragmentation (30). We tested whether the action of
Fig. 2. LLO treatment and L. monocytogenes infection cause a decrease in
the Drp1 receptor Mff. (A) HeLa cells were treated for 10 min with 6 nM LLO,
30 min with 10 μM CCCP, or infected at an MOI of 50 with the indicated
strains for 1 h. Representative Western blot of Mff and GAPDH (loading
control) on total cell lysates shows a decrease in Mff levels upon LLO treat-
ment or infection with wild-type L. mono cytogenes.(B) Quantiﬁcation of
Western blots from >3 independent experiments (***P < 0.001; *P < 0.05). (C)
Immunoﬂuorescence of HeLa cells treated for 10 min with LLO and stained for
Mff (green), mitochondria (Tom20; red), and DNA (DAPI; blue). Although
there is heterogeneity in staining of endogenous Mff, there is a signiﬁcant
decrease in Mff staining upon LLO treatment. White box indicates a 4× en-
larged region that is shown below. (D) Heterogeneous behavior of mito-
chondria-associated Mff puncta observed by live cell imaging. Timelapse
spinning disk confocal images of HeLa cells expressing Mff-GFP (green) and
TagBFP-mito (red). Data were collected every 2 s 2 nM LLO was added at time
t = 50 s White arrowheads mark Mff-GFP puncta that are rapidly lost from
mitochondria. Yellow arrowhead marks an Mff dot that disappears before
fragmentation and green arrowhead marks one that remains associated with
mitochondria over the time course. (Scale bars: C,10μm; D,2.6μm.)
Stavru et al. PNAS
October 1, 2013
LLO on mitochondria required the presence of F-actin and
found that actin depolymerization by cytochalasin D reduces the
number of cells that display a fragmented mitochondrial network
upon LLO treatment (Fig. 5). Given that Drp1 is dispensable for
LLO-induced mitochondrial fragmentation, we hypothesize that,
in addition to contributing to Drp1 recruitment to mitochondria
as observed by ref. 30, actin may also facilitate mitochondrial
fragmentation by preconstricting or physically stretching mito-
chondria independently of Drp1.
In summary, we have shown that in contrast to uncouplers and
apoptosis-inducing agents, mitochondrial fragmentation caused by
Listeria via its toxin LLO proceeds in a Drp1-independent manner.
LLO-induced mitochondrial fragmentation concurs with deg-
radation of the Drp1-receptor Mff, causing a loss of Drp1 at
mitochondria. Drp1 has been proposed to be a multifunctional
protein that contributes to peroxisomal and possibly also to
ER ﬁssion (35, 36), as well as possessing a putative GTPase-
independent role in cell death (37). Its modulation by bacterial
infection might therefore represent a means to act on several
organ elles and cellular functi ons in a coordinated fashion.
The data presented in this study suggests that actin and the ER
play an active role in regulating LLO-induced mitochondrial ﬁs-
sion. Recently, the ER-localized formin INF2 has been proposed as
a link between the ER and actin at mitochondrial constriction sites
(10), but it remains to be clariﬁed whether this protein plays a role
in LLO-induced mitochondrial fragmentation. Although LLO acts
on mitochondrial dynamics before bacterial invasion and we could
not detect overt LLO accumulation on intracellular structures by
confocal imaging (15), we cannot exclude that a fraction of LLO
inserts into mitochondria-associated membranes enriched in cho-
lesterol (38), where it could modulate mitochon dria-ER contact
sites or their protein components, such as Mfn2 (39). Upon LLO
treatment, ER-mediated mitochondria l ﬁssionoccurseveninthe
absence of Drp1, indicating that either ER proteins may contribute
to mitochondrial abscission or that the ER recruits another as-yet-
unidentiﬁed mitochondrial ﬁssion machinery that exists in mam-
Given that among the exogenous agents that induce mitochon-
drial f ragmentation LLO is unique in its ability to act rapidly,
without activating canonical apoptosis, and in a Drp1-independent
manner, we suggest it could serve as an unprecedented tool to
understand the regulation and function of mitochondrial dynamics
proteins. Indeed, efforts in this ﬁeld should aim toward t he
identiﬁcation of the molecule(s) that facilitate mitochond rial
fragmentation upon LLO treatment in the absence of Drp1
and the signaling cascade that activates such fragmentation.
Materials and Methods
Reagent references, detailed experimental protocols, image and statistical
analysis, and nonstandard abbreviations used in this study are provided in
SI Materials and Methods.
Fig. 3. LLO and L. monocytogenes induce mito-
chondrial-fragmentation in Drp1
fection of HeLa cells transfected with K38A-HA
Drp1 with L. monocytogenes for 1 h or treatment
with 6 nM LLO for 10 min induces mitochondrial
fragmentation. Immunoﬂuorescence images show
mitochondria (CoxIV; red) and K38A-HA Drp1 (anti-
HA; green). Asterisk marks transfected cells, as
identiﬁed by immuﬂuorescence. The white box
indicates the portion of the image enlar ged 4× and
shown below. (Scale bars: 20 μm.) (B) Immunoﬂuo-
rescence of mitochondria (CoxIV; red) and nuclei
(DAPI; blue) upon treatment of WT or Drp1
with either 6 nM LLO (10 min) or 10 μM CCCP
(30 min). (C) The percentage of cells displaying a frag-
mented mitochondrial network upon CCCP or LLO
treatment was quantiﬁed in ﬁve or six randomly
chosen ﬁelds of view (n > 150 cells) per experiment.
Pooled data from three independent experiments is
shown (**P < 0.003, one-tailed Student t test). (D)
Immuno ﬂuorescence of mitochondrial matrix (cyto-
chrome c; red), mitochondrial outer membrane
(Tom20; green), and nuclei (DAPI; blue) upon treat-
ment of WT or Drp1
HCT116 cells with either 6 nM
LLO (10 min) or 10 μM CCCP (30 min). White box
indicates 2× enlarged region shown below. (Scale bars:
Fig. 4. The ER marks sites of mitochondrial fragmentation independent of
Drp1. (A) U2OS Sec61β-GFP cells were treated with 2 nM LLO and examined live
by confocal microscopy. Representative images of mitochondria (DsRed2-mito;
red) and ER (Sec61β-GFP; green) are shown. Arrowheads indicate mitochondrial
fragmentation sites marked by ER-mitochondria contact sites. Experiments
were repeated a minimum of three times. (Scale bars: 2.6 μm.) (B)Quantiﬁca-
tion of mitochondrial fragmentations marked by ER-mitochondrial contact
sites. Analysis was carried out as described in Materials and Methods.
www.pnas.org/cgi/doi/10.1073/pnas.1315784110 Stavru et al.
Bacterial Infections. For infection experiments, bacteria were grown over-
night in brain-heart infusion medium (Difco; BD Biosciences) at 37 °C, sub-
cultured at a 1:10 dilution until an optical density of 0.8–1 was reached,
washed three times in medium without serum, and added to cells for 1 h at
a multiplicity of infection (MOI) of 50. The strains used were as follows: wild-
type L. monocytogenes (EGD, BUG600), L. monocytogenes Δhly (EGD,
BUG2132), and L. innocua overexpressing InlB (BUG1642).
Cell Culture, Drug Treatments, and Transfection. HeLa CCL2 cells were cultured
according to American Type Culture Collection guidelines. MEF (kind gift
from K. Mihara, Kyushu University, Japan) and U2OS Sec61β-GFP cells (kind
gift from Yoko Shibata and Tom Rapoport, Harvar d Medical School, Bosto n;
for the Sec61β-GFP plasmid, see ref. 34) were maintained in DMEM supple-
mented with 10% (vol/vol) FCS. HCT116 cells were maintained in McCoys 5A
medium supplemented with 10% FCS, nonessential amino acids, and gluat-
amine. All cells were maintained under 10% (vol/vol) CO
were purchased from Invitrogen. Drugs were purchased from Sigma-Aldrich
and incubated in full medium unless otherwise stated. Drugs were incubated
at the following concentrations and times: 10 μM CCCP for 30 min, 0.1 μM
valinomycin for 30 min, 1.2 μM staurosporine for 90 min, 0.05% (wt/vol)
digitonin for 5 min, 20 μM MG132 and 20 μM lactacystine for 90 min, 2.5μM
cytochalasin D for 30 min (HeLa) and 1 μM for 20 min (MEF), and 2–6nMLLO
for 10 min (in absence of FCS). DNA transfection (3.5-cm wells) was performed
with FuGENE HD (Roche) according to the manufacturer’s instructions. The
following plasmids were used for live-cell imaging: pDsRed2-mito (Clontech;
0.2 μg), Mff-GFP (ref. 9; 0.125 μg), mCherry-Drp1 (ref. 9; 0.1 μg), Drp1-GFP (ref.
23; 0.5 μg), BFP-mito (ref. 9; 0.5 μg), OM-GFP (“Mito-GFP”,ref.40;1μg), and
K38A-HA Drp1 (ref. 23; 4 μg). siRNA transfection was performed with Oligo-
fectamine (Invi trog en) accor din g to th e man ufactur er’s instructions, using
Drp1-speciﬁc sequences described i n ref. 15.
After treatm ent, cells were either analyzed with a spinning disk micro-
scope for live-cell imaging or processed for immunoﬂuorescence with the
antibodies indicated in the respective ﬁgure legends (see SI Materials and
Methods for antibody catalog numbers and for detailed live cell imaging
Quantitative RT-PCR. Total RNA was extracted using the RNeasy kit (Qiagen)
and treated with TurboDNase (Ambion). cDNA was synthetized from 500 ng
of RNA with the RT
cDNA synthesis kit (Qiagen). Standard quantitative RT-
PCR was performed with the EvaGreen kit (Bio-Rad) on a Bio-Rad CFX ma-
chine, and gene expression was calculated by the 2^(-ΔΔ Ct) method. Two
independent primer sets were used for MFF. Data represent the mean of
two biological replicates and three technical replicates.
Western Blotting. Cells were lysed with 2× Laemmli loading buffer (124 mM
Tris·HCl at pH 6.8, 4% (vol/vol) SDS, 20% (wt/vol) glycerol, 0.02% bromo-
phenol blue, and 0.03% DTT), and geno mic DNA was digested with Ben-
zonase (Novagen). Proteins were separated by SDS/PAGE (Bio-Rad),
transferred by wet transf er to a nitrocellulose membrane (GE Healthcare)
that was blocked in 10% (wt/vol) milk; antibodies were incubated in milk at
dilutions indicated by the manufacturers. Quantiﬁcation of Western blots
was accomplished by using the Gel Analyzer plugin in ImageJ. The Mff signal
was normalized to GAPDH signal, and ≥3 independent experiments were
ACKNOWLEDGMENTS. We thank Dr. K. Mihara for Drp1 knockout MEFs,
Drs. Yoko Shibata and Tom Rapoport for U2OS-Sec61βGFP cells, and Drs. Gia
Voeltz and A. van der Bliek for constructs; Dr. Tham Tho-Nam for excellent
technical support and the Plate-Forme Imagerie Dynamique staff at Institut
Pasteur for support; Dr. Ascel Samba-Louaka for critical reading of the man-
uscript; and Drs. Janet Shaw and Aurélien Roux for insightful discussions.
This work received ﬁnancial support from the Institut Pasteur, Institut Na-
tional de la Santé et de la Recherche Médicale Unité 604, Institut National de
la Recherche Agronomique Unité Sous Contrat 2020, Fondation Louis Jeantet,
European Research Council Advanced G rant 233348 MODELIST, and Agence
Nationale pour la Recherche (E RANET Pathogenomics Grant LISTRESS and
Grant Blanc MITOPATHO). F.S. was supported by postdoctoral fellowships
from European Molecular Biology Organization and the Fondation pour l a
Recherche Médicale. C.W. and R.Y. are supported in part by the Intramural
Research Program of the National I nstitutes of Neurological Disorders and
Strok e, National Institutes of Hea lth. P.C. is a Howard Hughes Medic al In-
stitute Senior International Research Scholar.
1. Nunnari J, Suomalainen A (2012) Mitochondria: In sickness and in health. Cell 148(6):
2. Chen H, et al. (2010) Mitochondrial fusion is required for mtDNA stability in skeletal
muscle and tolerance of mtDNA mutations. Cell 141(2):280–289.
3. Chan DC (2012) Fusion and ﬁssion: Interlinked processes critical for mitochondrial
health. Annu Rev Genet 46:265–287.
4. Hollenbeck PJ, Saxton WM (2005) The axonal transport of mitochondria. J Cell Sci
5. Li Z, Okamoto K, Hayashi Y, Sheng M (2004) The importance of dendritic mito-
chondria i n the morphogenesis a nd plasticity of spines and synapses. Cell 119(6 ):
6. Ishihara N, et al. (2009) Mitochondrial ﬁssion factor Drp1 is essential for embryonic
development and synapse formation in mice. Nat Cell Biol 11(8):958–966.
7. Labrousse AM, Zappaterra MD, Rube DA, van der Bliek AM (1999) C. elegans dyna-
min-related protein DRP-1 controls severing of the mitochondrial outer membrane.
Mol Cell 4(5):815–826.
8. Wakabayashi J, et al. (2009) The dynamin-related GTPase Drp1 is required for em-
bryonic and brain development in mice. J Cell Biol 186(6):805–816.
9. Friedman JR, et al. (2011) ER tubules mark sites of mitochondrial division. Science
10. Korobova F, Ramabhadran V, Higgs HN (2013) An actin-dependent step in mitochondrial
ﬁssion mediated by the ER-associated formin INF2. Science 339(6118):464–467.
11. Otera H, Ishihara N, Mihara K (2013) New insights into the function and regulation of
mitochondrial ﬁssion. Biochim Biophys Acta 1833(5):1256–1268.
12. Cossart P (2011) Illuminating the landscape of host-pathogen interactions with the
bacterium Listeria monocytogenes. Proc Natl Acad Sci USA 108(49):19484–19491.
13. Hamon MA, Ribet D, Stavru F, Cossart P (2012) Listeriolysin O: The Swiss army knife of
Listeria. (Translated from eng). Trends Microbiol 20(8):360–368.
14. Rosado CJ, et al. (2008) The MACPF/CDC family of pore-forming toxins. Cell Microbiol
15. Stavru F, Bouillaud F, Sartori A, Ricquier D, Cossart P (2011) Listeria monocytogenes
transiently alters mitochondrial dynamics during infection. Proc Natl Acad Sci USA
16. Ishihara N, Fujita Y, Oka T, Mihara K (2006) Regulation of mitochondrial morphology
through proteolytic cleavage of OPA1. EMBO J 25(13):2966–2977.
17. Duvezin-Caubet S, et al. (2006) Proteolytic processing of OPA1 links mitochondrial
dysfunction to alterations in mitochondrial morphology. J Biol Chem 281(49):
18. Griparic L, Kanazawa T, van der Bliek AM (2007) Regulation of the mitochondrial
dynamin-like protein Opa1 by proteolytic cleavage. J Cell Biol 178(5):757–764.
Fig. 5. Actin contribution to LLO-induced mitochondrial fragmentation. (A)
Immunoﬂuorescence images of HeLa cells mock-treated or preincubated
with 2.5 μM cytochalasin D for 30 min to depolymerize actin and then
treated for 10 min with 6 nM LLO or for 30 min with 10 μM CCCP, showing
phallodin staining of actin in the grayscale images, mitochondria (red;
CoxIV), and nuclei (blue; DAPI). Actin depolymerization caused partial peri-
nuclear aggregation of mitochondria, but mitochondria retained a tubular
morphology. (B) Pooled data from three independent experiments (n > 150
cells per experiment, randomly chosen ﬁelds of view) showing that cytoD
treatment decreases the number of cells with fragmented mitochondria
upon LLO treatment. *P < 0.01, **P < 0.001 (one-tailed Student t test).
Stavru et al. PNAS
October 1, 2013
19. Frank S, et al. (2001) The role of dynamin-related protein 1, a mediator of mito-
chondrial ﬁssion, in apoptosis. Dev Cell 1(4):515–525.
20. Cereghetti GM, et al. (2008) Dephosphorylation by calcineurin regulates translocation
of Drp1 to mitochondria. Proc Natl Acad Sci USA 105(41):15803–15808.
21. Cereghetti GM, Costa V, Scorrano L (2010) Inhibition of Drp1-dependent mitochon-
drial fragmentation and apoptosis by a polypeptide antagonist of calcineurin. Cell
Death Differ 17(11):1785–1794.
22. Dimmer KS, et al. (2008) LETM1, deleted in Wolf-Hirschhorn syndrome is required for
normal mitochondrial morphology and cellular viability. Hum Mol Genet 17(2):201–214.
23. Smirnova E, Griparic L, Shurland DL, van der Bliek AM (2001) Dynamin-related protein
Drp1 is required for mitochondrial division in mammalian cells. Mol Biol Cell 12(8):
24. Mozdy AD, McCaffery JM, Shaw JM (2000) Dnm1p GTPase-mediated mitochondrial
ﬁssion is a multi-step process requiring the novel integral membrane component
Fis1p. J Cell Biol 151(2):367–380.
25. Yoon Y, Krueger EW, Oswald BJ, McNiven MA (2003) The mitochondrial protein hFis1
regulates mitochondrial ﬁssion in mammalian cells through an interaction with the
dynamin-like protein DLP1. Mol Cell Biol 23(15):5409–5420.
26. Yu T, Fox RJ, Burwell LS, Yoon Y (2005) Regulation of mitochondrial ﬁssion and apoptosis
by the mitochondrial outer membrane protein hFis1. JCellSci118(Pt 18):4141–4151.
27. Otera H, et al. (2010) Mff is an essential factor for mitochondrial recruitment of Drp1
during mitochondrial ﬁssion in mammalian cells. J Cell Biol 191(6):1141–1158.
28. Ribet D, et al. (2010) Listeria monocytogenes impairs SUMOylation for efﬁcient in-
fection. Nature 464(7292):1192–1195.
29. Samba-Louaka A, Stavru F, Cossart P (2012) Role for telomerase in Listeria mono-
cytogenes infection. Infect Immun 80(12):4257–4263.
30. De Vos KJ, Allan VJ, Grierson AJ, Sheetz MP (2005) Mitochondrial function and actin
regulate dynamin-related protein 1-dependent mitochondrial ﬁssion. Curr Biol 15(7):
31. Legros F, Lombès A, Frachon P, Rojo M (2002) Mitochondrial fusion in human cells is
efﬁcient, requires the inner membrane potential, and is mediated by mitofusins. Mol
Biol Cell 13(12):4343–4354.
32. Ishihara N, Jofuku A, Eura Y, Mihara K (2003) Regulation of mitochondrial mor-
phology by membrane potential, and DRP1-dependent division and FZO1-dependent
fusion reaction in mammalian cells. Biochem Biophys Res Commun 301(4):891–898.
33. Malka F, et al. (2005) Separate fusion of outer and inner mitochondrial membranes.
EMBO Rep 6(9):853–859.
34. Shibata Y, et al. (2008) The reticulon and DP1/Yop1p proteins form immobile
oligomers in the tubular endoplasmic reticulum. J Biol Chem 283(27):18892–18904.
35. Koch A, et al. (2003) Dynamin-like protein 1 is involved in peroxisomal ﬁssion. J Biol
36. Pitts KR, Yoon Y, Krueger EW, McNiven MA (1999) The dynamin-like protein DLP1 is
essential for normal distribution and morphology of the endoplasmic reticulum and
mitochondria in mammalian cells. Mol Biol Cell 10(12):4403–4417.
37. Bras M, et al. (2007) Drp1 mediates caspase-independent type III cell death in normal
and leukemic cells. Mol Cell Biol 27(20):7073–7088.
38. Area-Gomez E, et al. (2012) Upregulated function of mitochondria-associated ER
membranes in Alzheimer disease. EMBO J 31(21):4106–4123.
39. de Brito OM, Scorrano L (2008) Mitofusin 2 tethers endoplasmic reticulum to mito-
chondria. Nature 456(7222):605–610.
40. Stavru F, Nautrup-Pedersen G, Cordes VC, Görlich D (2006) Nuclear pore complex
assembly and maintenance in POM121- and gp210-deﬁcient cells. J Cell Biol 173(4):
www.pnas.org/cgi/doi/10.1073/pnas.1315784110 Stavru et al.