Microbiology (2003), 149, 2513–2527
DOI 10.1099/mic.0.26341-0
Shigella-induced necrosis and apoptosis of U937
cells and J774 macrophages
Takashi Nonaka,13 Taku Kuwabara,1 Hitomi Mimuro,2 Asaomi Kuwae24
and Shinobu Imajoh-Ohmi1
Correspondence
Takashi Nonaka
Shinobu Imajoh-Ohmi
[email protected]
Division of Cell Biology and Biochemistry, Department of Basic Medical Sciences1 and
Division of Bacterial Infection, Department of Microbiology and Immunology2, The Institute
of Medical Science, University of Tokyo, 4-6-1, Shirokanedai, Minato-ku, Tokyo 108-8639,
Japan
[email protected]
Received 11 March 2003
Revised
27 May 2003
Accepted 4 June 2003
It is currently unclear whether Shigella kills its phagocytic host cells by apoptosis or necrosis. This
study shows that rapid necrosis ensues in macrophage-like cell lines (U937 cells differentiated
by all-trans-retinoic acid and J774 cells) infected with the Shigella flexneri strain YSH6000. The
infected cells rapidly lose membrane integrity, a typical feature of necrosis, as indicated by the
release of the cytoplasmic lactate dehydrogenase and the exposure of phosphatidylserine (PS)
associated with the rapid uptake of propidium iodide (PI). The infected cells exhibit DNA
fragmentation without nuclear condensation, and substantial involvement of either caspase-3/-7 or
caspase-1 was not detected, which is also contrary to what is normally observed in apoptosis.
Cytochalasin D potently inhibited Shigella-induced cell death, indicating that only internalized
Shigella can cause necrosis. Osmoprotectants such as polyethylene glycols could suppress cell
death, suggesting that insertion of a pore by Shigella into the host cell membrane induces the
necrosis. The pore was estimated to be 2?87±0?4 nm in diameter. Shigella was also found to
be able to induce apoptosis but only in one of the lines tested and under specific conditions,
namely U937 cells differentiated with interferon-c (U937IFN). Caspase-3/-7 but not caspase-1
activation was observed in these infected cells and the exposure of PS occurred without the
uptake of PI. An avirulent Shigella strain, wild-type Shigella killed with gentamicin, and even
Escherichia coli strain JM109, could also induce apoptosis in U937IFN cells, and cytochalasin D
could not prevent apoptosis. It appears therefore that Shigella-induced apoptosis of U937IFN
cells is unrelated to Shigella pathogenicity and does not require bacterial internalization. Thus,
Shigella can induce rapid necrosis of macrophage-like cells in a virulence-related manner by
forming pores in the host cell membrane while some cells can be killed through apoptosis in a
virulence-independent fashion.
INTRODUCTION
Apoptosis is an active process of cellular suicide that is
triggered by a variety of physiological and stress stimuli.
Apart from playing important roles in normal development
and tissue homeostasis, apoptosis is also involved in the
interaction between host cells and bacterial pathogens, as
a number of bacterial pathogens appear to be capable of
manipulating host cell apoptotic pathways (Gao & Abu
Kwaik, 2000). Whether these manipulations are to the
advantage of the host or the bacteria varies among pathogens. Obligate intracellular bacteria such as Chlamydia
trachomatis and Rickettsia rickettsii inhibit host cell
apoptosis and this may allow these organisms to grow
3Present address: Department of Molecular Neurobiology, Tokyo Institute of Psychiatry, Tokyo Metropolitan Organization for Medical Research, 2-1-8,
Kamikitazawa, Setagaya-ku, Tokyo, 156-8585, Japan.
4Present address: Laboratory of Bacterial Infection, Kitasato Institute for Life Sciences, Kitasato University and the Kitasato Institute, 5-9-1 Shirokane,
Minato-ku, Tokyo 108-8642, Japan.
Abbreviations: Ac-DEVD-CHO, acetyl-Asp-Glu-Val-Asp-aldehyde; Ac-DEVD-MCA, acetyl-Asp-Glu-Val-Asp-4-methylcoumaryl-7-amide; Ac-WEHDMCA, acetyl-Trp-Glu-His-Asp-4-methylcoumaryl-7-amide; Ac-YVKD-CHO, acetyl-Tyr-Val-Lys-Asp-aldehyde; AMC, 7-amino-4-methylcoumarin; Gm,
gentamicin; IFN, interferon-c; PARP, poly(ADP-ribose) polymerase; PI, propidium iodide; PS, phosphatidylserine; RA, all-trans retinoic acid; RBC, red
blood cell; stsp, staurosporine; TBS, Tris-buffered saline; TUNEL, terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labelling;
Z-VAD-fmk, carbobenzoxy-Val-Ala-Asp-fluoromethylketone.
0002-6341 G 2003 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
Printed in Great Britain
2513
Takashi Nonaka and others
and persist intracellularly (Fan et al., 1998; Clifton et al.,
1998). In contrast, infection with many facultative intracellular bacteria, such as species of Shigella (Zychlinsky
et al., 1992), Salmonella (Monack et al., 1996), Yersinia
(Ruckdeschel et al., 1997) and Mycobacterium (Fratazzi
et al., 1997), appears to cause the host cell apoptosis. The
pathways involved in this vary between bacteria. Yersinia
induces apoptosis by translocating soluble proteins, such
as YopJ or YopP, via the type III secretion system into
the host cell cytoplasm. These proteins then bind to the
mitogen-activated protein kinase kinase and thus inhibit
its activity (Orth et al., 1999), which serves to block the
activation of nuclear factor kB, a central regulatory molecule
involved in cell survival (Schesser et al., 1998).
With regard to intracellular Shigella and Salmonella, these
bacteria have been suggested to be able to induce host cell
apoptosis by secreting effector proteins into the host cytoplasm, via the type III secretion system, factors that directly
bind and activate caspase-1 (Hilbi et al., 1998; Hersh et al.,
1999). Shigella kills its host macrophage by releasing IpaB,
a secreted bacterial protein that is associated with virulence
and the ability to invade host cells (Chen et al., 1996). IpaB
directly binds to and activates caspase-1 and is thus
thought to be the key molecule in the induction of
caspase-1-dependent apoptosis by Shigella infection (Hilbi
et al., 1998; Chen et al., 1996). Similarly, Salmonella is
believed to induce macrophage caspase-1-dependent
apoptosis by secreting the SipB protein into the cytoplasm
of its host macrophage. The SipB protein is thought to be
the Salmonella homologue of IpaB and it also binds directly
to and activates pro caspase-1 (Hersh et al., 1999).
Despite the involvement of caspase-1 in the apparent
induction of apoptosis by Shigella and Salmonella, this
molecule is usually not involved in most apoptotic
processes. Kuida et al. (1995) and Li et al. (1995, 1997)
demonstrated that caspase-1-deficient mice fail to produce
interleukin (IL)-1b and thus could resist endotoxic shock,
indicating that caspase-1 plays a major role in cytokine
maturation, and the mouse cells did not exhibit significant
defects in apoptosis. In contrast, caspase-3 and/or caspase-7
are known to play a central role in driving the classical
apoptotic pathways triggered by a variety of stimuli.
However, in Shigella-induced macrophage cell death, the
activation of caspase-3/-7, or the cleavage of poly(ADPribose) polymerase (PARP), one of the specific substrates
for these caspases during apoptosis, has not been detected
(Hilbi et al., 1998; Chen et al., 1996). The lack of caspase3/-7 involvement in Shigella-induced cell death thus casts
doubt on whether the killing truly occurs through apoptosis.
Supporting this is the finding by Fernandez-Prada et al.
(1997, 2000) that human monocyte-derived macrophages
infected in vitro with Shigella flexneri undergo a rapid
cytolytic event that is similar to oncosis (necrotic cell
death) but not apoptosis. They observed that when human
monocyte-derived macrophages were infected with virulent
Shigella, this resulted in cell death that involved rupture
2514
of the plasma membrane, cell swelling, disintegration of
the cellular ultrastructure, and generalized karyolysis. Other
groups have also reported that the macrophage cell death
induced by Salmonella infection does not resemble typical
apoptosis (Brennan & Cookson, 2000; Watson et al., 2000),
as Salmonella-infected macrophages appear to rapidly lose
their membrane integrity in a manner similar to Shigellainfected macrophages (Fernandez-Prada et al., 1997, 2000;
Nonaka et al., 1999). Thus, the mechanism by which Shigella
and Salmonella kill their host macrophages is unclear and
controversial (Boise & Collins, 2001).
In this study, we report that the wild-type S. flexneri strain
YSH6000 can induce both types of cell death, necrosis and
apoptosis, depending on the type of host cell. We used
terminal deoxynucleotidyl transferase-mediated dUTPbiotin nick end labelling (TUNEL) staining and measurements of caspase activities to discriminate between
apoptotic and necrotic cell death and show that Shigella
can induce rapid necrotic cell death in specifically
macrophage-like cells. We also found that Shigella can
induce apoptosis of particular cells. The mechanism does
not require phagocytosis of the bacteria and is not
dependent on their virulence. Thus, Shigella can kill host
cells through either necrosis or apoptosis depending on
the host cell type.
METHODS
Bacterial strains and cell cultures. S. flexneri 2a strain YSH6000
is the wild-type and N1411 is an ipaB : : Tn5-induced avirulent
mutant of YSH6000 (Sasakawa et al., 1986, 1988). S. flexneri-derived
strains and the Escherichia coli strain JM109 were grown in brainheart infusion (BHI; Difco) broth or L broth (Difco) at 37 uC.
Human monoblastic U937, myeloid HL-60, monocytic THP-1, and
murine macrophage-like J774 cells were cultured in RPMI1640
medium (Nissui Pharmaceutical) supplemented with 10 % (v/v)
fetal calf serum (JRH Biosciences), 2 mM Gln, 100 units penicillin G
ml21 and 200 mg streptomycin ml21 at 37 uC in a humidified atmosphere of 5 % (v/v) CO2.
Infection of host cells with Shigella. U937 cells were pretreated with 100 units interferon-c ml21 (IFN; a gift from Shionogi
Pharmaceuticals) or 3 mM all-trans-retinoic acid (RA; Sigma) for
48 h to induce them to differentiate into macrophage-like cells
before infection as described previously (Nonaka et al., 1999;
Kikuchi et al., 1996). The host cells (~106 cells per well in 12- or
24-well plates) were cultured with antibiotic-free medium for 1 h
before infection. To obtain highly invasive bacteria, cultures incubated overnight at 30 uC were diluted 1 : 50 with BHI broth and
incubated at 37 uC for 2 h before use. The cells were then infected
with bacteria at the indicated m.o.i., centrifuged at 700 g for
10 min, and incubated for 1 h at 37 uC in a CO2 incubator.
Subsequently, 100 mg gentamicin ml21 (Gm; Sigma) was added to
the medium and the cell/bacteria mixture was further incubated at
37 uC in a CO2 incubator for the indicated time to kill the extracellular bacteria.
Measurement of Shigella-induced cytotoxicity. Cytotoxicity
induced by Shigella infection was analysed with the CytoTox 96
Cytotoxicity Assay Kit (Promega). The percentage cytotoxicity was
calculated by quantifying the amount of cytoplasmic lactate
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
Microbiology 149
Macrophage-like cell death by Shigella infection
dehydrogenase (LDH) released by the dying cells according to the
following formula: [(experimental LDH activity 2 spontaneous LDH
activity)/(total LDH activity 2 spontaneous LDH activity)]6100.
Measurement of caspase activity in Shigella-infected cells.
J774 or undifferentiated and differentiated U937 cells were either
infected with Shigella or incubated with 0?5 mM of the apoptosisinducing agent staurosporine (stsp; Sigma) as a positive control for
apoptosis. After the indicated period, ~107 cells were harvested and
washed with 0?15 M NaCl. Cytosolic fractions were prepared as
follows. Cells were resuspended in 200 ml lysis buffer (20 mM
HEPES/KOH, pH 7?5, 1 mM EDTA), disrupted by sonication, and
insoluble material was removed by centrifugation at 15 000 g for
20 min at 4 uC. The supernatant was tested for the measurement of
caspase-3/-7 or caspase-1 activity by using the fluorescent synthetic
peptide substrates Ac-DEVD-MCA and Ac-WEHD-MCA, respectively (Peptide Institute) according to the established method
(Thornberry et al., 1997). In brief, the samples (5, 10, 20 ml) were
mixed with 0?5 ml of assay buffer (20 mM HEPES/KOH, pH 7?5,
1 mM EDTA, 100 mM NaCl and 10 mM 2-mercaptoethanol) and
water in a total volume of 1 ml. After pre-incubation at 37 uC for
15 min, 10 ml of 10 mM peptide substrate was added to the reaction
mixtures and incubated at 37 uC for 15 min. One millilitre of 2 %
(v/v) acetic acid was added to stop the enzymic reaction, and
7-amino-4-methylcoumarin (AMC) release was measured fluorometrically using an excitation wavelength of 365 nm and an emission wavelength of 460 nm. One unit was defined as the amount of
enzyme capable of generating 1 nmol AMC in 1 h.
Detection of the cleavage of PARP by immunoblotting.
Infected or stsp-treated cells (~106 cells) were harvested and
washed with 0?15 M NaCl. Trichloroacetic acid was added to the
cell suspension to a final concentration of 10 % (v/v), and the
sample was placed on ice for 20 min. Precipitates were recovered
by centrifugation at 15 000 g for 5 min at 4 uC, dissolved in 100 ml
SDS-sample buffer containing 2?3 % (w/v) SDS, 10 % (w/v) glycerol,
5 % (v/v) 2-mercaptoethanol and 10 mg bromophenol blue ml21 in
125 mM Tris/HCl, pH 6?8, and heated at 100 uC for 5 min. Samples
were run out on a 7?5 % (w/v) polyacrylamide gel containing SDS
and electroblotted onto a PVDF membrane (Immobilon, Millipore).
The membrane was then soaked for 2 h at room temperature in
Tris-buffered saline (TBS: 0?15 M NaCl in 20 mM Tris/HCl,
pH 7?5) containing 20 mg bovine serum albumin (BSA) ml21. The
membrane was subsequently incubated overnight at 4 uC with a
polyclonal antibody specific for the amino-terminal region of the
85 kDa fragment of human PARP that is generated by caspase-3/-7
(Nonaka et al., 1999; Kato et al., 2000). The antibody was diluted
1 : 500 with TBS containing 20 mg BSA ml21. After washing three
times in TBS containing 0?05 % (v/v) Tween 20 (Wako Pure
Chemical Industries) for 15 min, the membrane was incubated at
room temperature for 1 h with 1 : 7000 diluted anti-rabbit IgG
conjugated with alkaline phosphatase (Promega) and then again
washed three times in TBS containing 0?05 % Tween 20 for 15 min.
Immune complexes were detected by the enzymic reaction of
alkaline phosphatase with 5-bromo-4-chloro-3-indolyl phosphate
(Nacalai tesque) and nitro blue tetrazolium (Nacalai tesque).
poly-L-lysine (Sigma) pre-treated coverslip before infection. Cells on
the coverslips were infected with Shigella at an m.o.i. of 50, incubated for 30 min, and then 100 mg Gm ml21 was added to each
sample. After incubation for the indicated times, the infected cells
on the coverslips were fixed with 4 % (w/v) paraformaldehyde in
phosphate-buffered saline (PBS) for 30 min. The coverslips were
then incubated in 50 mM NH4Cl in PBS for 5 min and cell permeabilization was performed with 0?2 % (v/v) Triton X-100 in PBS for
5 min. After blocking for 30 min in 40 mg BSA ml21 in TBS, the
bacteria on the coverslips were stained by a rabbit anti-S. flexneri 2a
lipopolysaccharide (LPS) antibody (Watarai et al., 1997) followed
by the Cy5-labelled goat anti-rabbit IgG (Sigma) as the secondary
antibody. After washing, the apoptotic cells were labelled by using
the In Situ Cell Death Detection Kit (Roche Diagnostics), according
to manufacturer’s protocol. The stained cells were observed by
confocal laser scanning microscopy (Bio-Rad).
Flow cytometric analysis. Undifferentiated and differentiated
U937 and J774 cells were infected with YSH6000 at an m.o.i. of
50 or treated with 0?5 mM stsp, and then harvested and washed with
PBS. Adherent J774 cells were removed from 6-well plates with
PBS-EDTA and pooled with the non-adherent cells in the culture
medium. Cells were labelled with Annexin-V-FLUOS (Roche
Diagnostics) according to the manufacturer’s protocol. In brief,
~106 cells were washed with incubation buffer (10 mM HEPES/
NaOH, pH 7?4, 140 mM NaCl, 5 mM CaCl2), resuspended in
100 ml labelling solution [10 ml Annexin-V-FLUOS and 10 ml
50 mg ml21 propidium iodide (PI) in 1 ml incubation buffer] and
incubated at room temperature in the dark for 15 min. Cells were
immediately analysed on a FACScan (Becton Dickinson).
Osmoprotection assay. U937RA and J774 cells were pre-incubated
in RPMI1640 medium supplemented with 20 mM of various osmoprotectants, namely sucrose (Nacalai Tesque), PEG 600 (Tokyo
Kasei), PEG 1000 (Nacalai Tesque), and PEG 2000 (Nacalai Tesque)
made in PBS. The cells were then infected at an m.o.i. of 50 for 2 h,
and LDH release by the infected cells was measured as described
above. For the estimation of the functional diameter of inserted pores
into host membrane, we used the values for hydrodynamic diameters
of the non-electrolytes (Scherrer et al., 1971), where the hydrodynamic diameters were calculated on the viscosity of non-electrolyte
solutions.
Measurement of apoptosis-inducing potential of Shigella.
J774 or undifferentiated and differentiated U937 cells were incubated
with either the wild-type YSH6000 strain, the avirulent mutant
N1411 lacking ipaBCDA, the Gm-treated killed wild-type YSH6000,
or E. coli JM109 at the indicated m.o.i.. The Gm-treated YSH6000
were obtained by resuspension in RPMI1640 with 100 mg Gm ml21
and incubation at 37 uC for 30 min. As a positive control for apoptosis, the cells were incubated with 0?5 mM stsp. The bacterial
suspension was added to host cells at the indicated m.o.i., and incubated at 37 uC for 5 h. The cells were harvested and prepared for
measurements of caspase activity or immunoblot analyses as
described above.
Protein concentration. Protein concentrations were determined
Inhibition of caspase activity and phagocytosis. Host cells
were pre-treated with 100 mM Ac-DEVD-CHO, which inhibits
caspase-3/-7, Ac-YVKD-CHO, which inhibits caspase-1, or Z-VADfmk, a general caspase inhibitor (Peptide Institute, Osaka, Japan), or
with 5 mg ml21 of the phagocytosis inhibitor cytochalasin D (Sigma)
1 h prior to infection with Shigella. LDH assays were performed 2
and 5 h after infection at an m.o.i. of 50.
TUNEL staining and immunofluorescence microscopy. J774
cells were grown to ~106 cells on a coverslip (18618 mm). In the
case of U937IFN and U937RA, ~106 cells were incubated on a
http://mic.sgmjournals.org
by the method of Bradford (1976) using BSA as the standard.
RESULTS
Shigella induces macrophage-like cell lines to
undergo rapid cell death
We previously reported (Nonaka et al., 1999) that infection
with the wild-type S. flexneri strain YSH6000 caused rapid
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
2515
Takashi Nonaka and others
and striking oncosis (necrosis) only in U937 cells that had
been differentiated by treatment with RA (U937RA) and
induced apoptosis in U937 cells that had been differentiated
by IFN (U937IFN). U937RA appear to be more macrophagelike than U937IFN cells in that they have a greater capacity
to generate superoxide (Kikuchi et al., 1994) and they
express higher levels of CD11b (Kikuchi et al., 1996).
To determine whether the rapid necrotic cell death induced
by infection with S. flexneri YSH6000 occurs only with
macrophage-like cells, we attempted to infect another
macrophage-like cell line (murine J774 cells), the myeloid
HL-60 line and the monocytic THP-1 line, as well as
undifferentiated and RA- or IFN-differentiated U937 cells.
Host cells were infected with S. flexneri YSH6000 at various
m.o.i. After incubation for 1, 3 or 5 h, the viability of
the infected cells was quantified by measuring the activity
of LDH released due to loss of cell membrane integrity.
As shown in Fig. 1(a), Shigella-induced death of the
macrophage-like U937RA and J774 cells at an m.o.i. of 50
was more rapid and striking than that seen with the other
cells. We also observed a dose-dependent effect of m.o.i.
(data not shown). The morphology of the infected U937RA
and J774 cells revealed plasma membrane damage and
swelling of the cytoplasm but not nuclear condensation
(Nonaka et al., 1999, and data not shown). Wild-type
Shigella could not effectively induce cell death in the less
phagocytic cell lines HL-60, THP-1, and undifferentiated
U937 cells (U937UD) (Fig. 1a). The U937IFN cells were
lysed by Shigella to a lesser degree than U937RA or J774 cells.
Similar experiments were performed with the N1411
strain, which is an avirulent mutant of S. flexneri
YSH6000 that lacks ipaBCDA. While this strain was
phagocytosed by U937RA and J774 cells at levels similar
to the wild-type bacteria (data not shown), it failed to cause
the rapid cell death observed with the wild-type (Fig. 1b).
These observations suggest that S. flexneri YSH6000 causes
rapid cell death accompanied by membrane damage only
in macrophage-like cell lines and that this cell death is
dependent on the virulence of S. flexneri and its expression
of the IpaB, C, D and/or A proteins.
Cell death of macrophage-like cells infected
with Shigella is not accompanied by caspase
activation
Hilbi et al. (1998) reported that Shigella infection induces
macrophage apoptosis accompanied by caspase-1 but not
caspase-3 activation. This is unusual because caspase-1 is
not involved in most apoptotic processes, being more
prominent in cytokine maturation. We thus characterized
Shigella-induced cell death in macrophage-like cell lines
further by directly examining the caspase activities in the
infected cells. Host cells (U937UD, U937IFN, U937RA
and J774) were infected with wild-type Shigella at an m.o.i.
of 50. After subsequent incubation in the presence of
Gm for several hours, the cytosolic fractions of the
infected cells were prepared and assayed for caspase-1 and
caspase-3/-7 activity using the synthetic peptide substrates
2516
Fig. 1. Infection with S. flexneri strain YSH6000 predominantly
kills macrophage-like cell lines. Undifferentiated U937
(U937UD) (#), U937 differentiated with IFN (U937IFN) (n) or
RA (U937RA) (&), J774 ($), HL-60 (e) and THP-1 (%) cells
were infected at an m.o.i. of 50 with either the wild-type S.
flexneri strain YSH6000 (a) or the avirulent N1411 mutant of
YSH6000 that lacks the ipaBCDA genes (b). Supernatants
were taken at the indicated time points and LDH released in
the supernatants was assayed to measure cytotoxicity. The
percentage cytotoxicity was calculated as described in
Methods. Each data point represents the mean±SD from triplicate measurements.
Ac-WEHD-MCA and Ac-DEVD-MCA, respectively. As
shown in Fig. 2(a), no cleavage of either Ac-DEVD-MCA
or Ac-WEHD-MCA by the cytosolic fractions of infected
U937RA and J774 cells was detected. In contrast, strong
activity against Ac-DEVD-MCA, but not Ac-WEHDMCA, was seen with the infected U937IFN cells (Fig. 2a)
as well as the stsp-induced apoptotic cells that were used
as a positive control for apoptosis (Fig. 2b).
Poly(ADP-ribose) polymerase (PARP) is one of the target
substrates cleaved by caspase-3/-7 during apoptosis. The
cleavage of this 116 kDa protein occurs at Asp-214 (Lazebnik
et al., 1994) and generates the regulatory 30 kDa and the
catalytic 85 kDa fragments. To confirm the lack of caspase3/-7 activity in S. flexneri YSH6000-infected U937RA and
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
Microbiology 149
Macrophage-like cell death by Shigella infection
Fig. 2. Neither caspase-3/-7 nor caspase-1 is activated during Shigella-induced macrophage-like cell death. U937UD (#),
U937IFN (n), U937RA (&) and J774 ($) cells were either infected with wild-type Shigella at an m.o.i. of 50 (a) or treated
with the apoptosis-inducing agent stsp (b). At the time points indicated, cytosolic fractions of the cells were prepared and
caspase activities were determined using the caspase-3/-7 substrate Ac-DEVD-MCA or the caspase-1 substrate Ac-WEHDMCA. AMC released due to caspase activity was measured fluorometrically. One unit is defined as the amount of enzyme
capable of generating 1 nmol AMC in 1 h. Data represent means±SD from three separate experiments.
J774 cells, we analysed the proteolysis of PARP by immunoblotting with an anti-PARP antibody that recognizes the
85 kDa fragment. Undifferentiated and differentiated U937
and J774 cells were infected with the YSH6000 strain at an
m.o.i. of 50 and samples for immunoblot analyses were
prepared at various time points. PARP was not cleaved in
wild-type Shigella-infected U937RA and J774 cells (Fig. 3a).
In contrast, time-dependent PARP cleavage was observed
in U937IFN cells infected with wild-type Shigella (Fig. 3a)
and stsp-treated cells as a positive control for apoptosis
(Fig. 3c).
To further assess the role of caspase(s) in Shigella-induced
cell death of macrophage-like cells, we examined the
effects of synthetic caspase inhibitors on the mortality of
U937RA cells infected with wild-type Shigella. The inhibitors used were the caspase-3/-7 inhibitor Ac-DEVD-CHO,
the caspase-1 inhibitor Ac-YVKD-CHO, and the general
caspase inhibitor Z-VAD-fmk. Ac-DEVD-CHO and ZVAD-fmk are potent inhibitors of apoptosis (Kato et al.,
http://mic.sgmjournals.org
2000), and these inhibitors had no effects on viabilities of
cells used here (data not shown). U937RA cells were pretreated with the caspase inhibitors followed by infection
with YSH6000 at an m.o.i. of 50. After incubation for 2 and
5 h, the viabilities of the infected cells were examined by
measuring their release of LDH. Z-VAD-fmk suppressed
Shigella-mediated U937RA cell death by 45 %, 2 h postinfection, while its inhibitory effect was almost undetectable at 5 h after infection (Fig. 4). The caspase-1-specific
inhibitor also decreased Shigella-induced U937RA and
J774 cell death 2 h post-infection by 24 %, but after 5 h of
incubation, its inhibitory effect completely disappeared
(Fig. 4). In contrast, the caspase-3/-7 inhibitor did not
suppress Shigella-induced cell death at either time point.
The suppression of Shigella-induced cell death by Z-VADfmk at 2 h post-infection indicates that caspase(s) other
than caspase-3/-7 may be involved in the early phase of
Shigella-induced cell death. However, the observation that
even the general caspase inhibitor could only partially delay
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
2517
Takashi Nonaka and others
(a)
U937IFN
U937UD
kDa
U937RA
J774
whole PARP
117
97
cleaved
66
0
(b)
kDa
1
3
5
0
U937UD
1
3
5
0
U937IFN
1
3
5
0
U937RA
1
3
5
Incubation time (h)
J774
117
97
whole PARP
cleaved
66
0
(c)
kDa
1
3
5
0
U937UD
1
3
5
0
1
U937IFN
3
5
0
U937RA
1
3
5
Incubation time (h)
J774
117
97
whole PARP
cleaved
66
0
1
3
5
0
1
3
5
0
1
3
5
0
1
3
5
Incubation time (h)
Fig. 3. PARP is not cleaved in Shigella-infected macrophage-like cells. U937UD, U937IFN, U937RA and J774 cells were
infected at an m.o.i. of 50 with either the wild-type YSH6000 Shigella strain (a) or the avirulent mutant N1411 that lacks the
ipaBCDA genes (b), or they were incubated with the apoptosis-inducing agent stsp (c) for the indicated times. Approximately
56104 cells per lane were separated by SDS-PAGE and the gel was immunoblotted with a polyclonal antibody specific for
the amino-terminal region of the 85 kDa fragment of human PARP that is generated by caspase-3/-7 cleavage of PARP.
Results are representative of at least three experiments.
cell death indicates that Shigella-induced cell death is only
partly caspase-dependent.
Shigella infection of macrophage-like cells
stimulates DNA fragmentation without nuclear
condensation
To further assess whether the rapid cell death caused by
wild-type Shigella infection is apoptosis or necrosis, we
stained infected cells by TUNEL labelling. U937RA and J774
cells infected with wild-type bacteria were fixed and
incubated with the TUNEL reaction mixtures, and observed
by confocal laser microscopy. As shown in Fig. 5(a, b), the
condensed nuclear morphology typical of apoptosis was
detected in both macrophage-like cell lines when they had
been treated with stsp as a positive control for apoptosis.
With regard to the Shigella-infected cells, Shigella invasion
into both host cell lines could be observed even 1 h after
infection but DNA fragmentation was not detected at this
point. After 3 h incubation, however, DNA fragmentation
was observed, but unlike the condensed pattern seen in
the stsp-treated cells, the nicked chromatin in the infected
cells stained diffusely. This is in agreement with a previous
report on macrophage cell death induced by Salmonella
2518
infection (Brennan & Cookson, 2000). Moreover, the
frequency of TUNEL-positive infected J774 cells was not
reduced at all in the presence of either the caspase-1 or
caspase-3/-7 inhibitors (Ac-YVKD-CHO and Ac-DEVDCHO, respectively) (data not shown). Thus, Shigella
infection of macrophage-like cells induces DNA fragmentation that is distinguishable from that observed in typical
apoptotic cells.
Shigella infection of macrophage-like cell lines
induces rapid membrane damage
In order to characterize the cell death induced by Shigella,
we used a flow cytometric assay that discriminates
between apoptosis and necrosis. Here a fluorescencelabelled phospholipid-binding protein, annexin V, is used
to specifically detect phosphatidylserine (PS), a membrane
lipid that normally localizes to the inner leaflet of the
plasma membrane. Early apoptotic cells display PS on
their outer leaflet while maintaining membrane integrity.
Necrotic cells also expose PS but this is due to membrane
damage that can be detected by their uptake of membraneimpermeable dyes such as propidium iodide (PI). Apoptotic
and necrotic cells can thus be separated by flow cytometry,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
Microbiology 149
Macrophage-like cell death by Shigella infection
PI-double-positive necrotic region, indicating that the
infected U937RA and J774 cells are undergoing necrosis
with rapid membrane damage (Table 1).
Macrophage-like cell death caused by
internalized Shigella is due to osmotic lysis
with the insertion of pores into the membrane
We first examined whether the phagocytosis of YSH6000
was an important step in the cytolytic pathway in infected
macrophage-like host cells. Experiments were thus performed in the presence of cytochalasin D, which inhibits
actin polymerization and consequently blocks phagocytosis. We confirmed that cytochalasin D has no cytotoxicity
on cells used here (data not shown). U937RA and J774
cells were pre-incubated with cytochalasin D and then
infected with Shigella. In the presence of cytochalasin D,
the phagocytosis of the wild-type bacteria was dramatically
suppressed (data not shown) and the release of LDH
remained at the level of uninfected control cells (Fig. 4).
Necrotic cell death could be inhibited by cytochalasin D,
suggesting that bacterial internalization could be critical
in the induction of the necrotic response.
Fig. 4. Effects of caspase and phagocytosis inhibitors on
macrophage-like cell death induced by Shigella infection.
U937RA cells were pre-treated for 1 h with 100 mM of
caspase inhibitors (Ac-DEVD-CHO, which inhibits caspase-3/-7;
Ac-YVKD-CHO, which inhibits caspase-1; and Z-VAD-fmk, a
general caspase inhibitor) or 5 mg cytochalasin D ml”1. The
cells were then infected with wild-type Shigella at an m.o.i. of
50. LDH release assays were performed 2 and 5 h post-infection.
Data represent means±SD from three separate experiments.
with apoptotic cells falling in the annexin V-FITC-positive
and PI-negative region and necrotic cells appearing in the
annexin V-FITC- and PI-double-positive region (Fig. 6).
Stsp-treated U937 and J774 cells were used as a positive
control for apoptosis and almost 30–40 % of these cells
were detected in the annexin V-FITC-positive and PInegative region (Fig. 6). Of the cell lines infected with
wild-type YSH6000, U937IFN cells prominently exposed
PS without PI uptake in a time-dependent manner as
shown in Table 1 (see further discussion below). The
infected U937UD cells reacted similarly but to a lesser
degree. In contrast, less than 10 % of the Shigella-infected
U937RA and J774 cells were found in the annexin V-FITCpositive and PI-negative region. Instead, increasing numbers of cells were detected in the annexin V-FITC- and
http://mic.sgmjournals.org
Shigella is known to lyse red blood cells (RBCs) when they
are brought into close contact with the cells by centrifugation. This is known as contact haemolysis and it is
apparently mediated by the Shigella-mediated insertion of
a 2?5 nm pore into the red blood cell membrane (Blocker
et al., 1999). A similar mechanism is also observed with
Yersinia species. Here, the formation of the 1?2–3?2 nm pore
in RBCs requires the yopB gene product (Hakansson et al.,
1996). It is thus possible that the Shigella-mediated rapid
cell death of infected host cells may also be due to osmotic
lysis caused by the formation of a pore in the host cell
membrane. Polyethylene glycols (PEGs) with various molecular masses have been used to measure the size of pores
induced by several pathogens (Moran et al., 1992; Kirby
et al., 1998; Dacheux et al., 2001). We tested the ability of
various osmoprotectants to protect infected U937RA cells
from death by pre-incubating host cells with osmoprotectants and then monitoring LDH release 2 h after infection.
Although sucrose (0?9 nm diameter) did not prevent the
lysis of infected cells, PEGs with a molecular mass of 600
(1?6 nm diameter) and 1000 (2 nm diameter) reproducibly
suppressed LDH release, and PEG 2000 (2?8 nm diameter)
was the most effective osmoprotectant, as shown in Fig. 7.
From a plot of the relative cell death in the presence of
osmoprotectants versus their diameter (not shown), we
estimated that the functional diameter of the pore is
2?87 nm (±0?4 nm, standard deviation, n=4). In analyses
of Shigella-infected J774 cells, the similar size of pores was
also determined (data not shown).
Shigella infection of U937 cells differentiated
with IFN induces apoptosis that is not
dependent on Shigella pathogenicity
Our previous report (Nonaka et al., 1999) showed that
S. flexneri YSH6000 induces apoptosis in U937 cells that
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
2519
Takashi Nonaka and others
2520
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
Microbiology 149
Macrophage-like cell death by Shigella infection
Fig. 5. Infection of macrophage-like cells with S. flexneri YSH6000 infection generates DNA fragmentation without nuclear
condensation. U937RA (a) and J774 (b) adhering to glass coverslips were either uniformly infected with wild-type YSH6000
at an m.o.i. of 50 for 1 and 3 h post-infection or treated with the apoptosis-inducing agent stsp for 5 h. Adherent cells were
stained with Cy5-labelled anti-S. flexneri 2a LPS antibody and by the TUNEL method, which detects fragmented DNA. The
cells were observed by confocal laser microscopy (606 objective). Results are representative of at least three experiments.
have been differentiated with IFN but not in U937 cells
differentiated with RA. Shigella-infected U937IFN showed
the morphological features of typical apoptosis, namely
cytoplasmic shrinkage and nuclear fragmentation (Nonaka
et al., 1999). To explore the caspases involved in infected
U937IFN cell death, we prepared a cytosolic fraction
of infected U937IFN cells and assayed its caspase activities using several peptide-MCA substrates. Although the
cytotoxicity of U937IFN by wild-type Shigella is not as
high as that of infected U937RA and J774 cells (Fig. 1a),
caspase-3/-7 activity was observed in infected U937IFN
lysates, much higher than observed for infected U937RA
and J774 cells (Fig. 2a). The cleavage of PARP was also
clearly detected only in infected U937IFN cells (Fig. 3a). In
contrast, caspase-1 activity was not detected at all (Fig. 2a).
U937IFN cells infected with Shigella also showed the most
Fig. 6. Shigella-infected macrophage-like cells expose PS as a result of rapid membrane damage. Undifferentiated and
differentiated U937 (a) and J774 (b) cells were infected with wild-type bacteria at an m.o.i. of 50 or treated with 0?5 mM stsp.
Infected cells were then stained with annexin V-FITC and the vital dye PI and immediately analysed by FACScan (Becton
Dickinson). Apoptotic cells are annexin V-FITC-positive and PI-negative (lower right region in the FACS panels), while necrotic
infected cells are annexin V-FITC- and PI-double positive (upper right region in the FACS panels). Results are representative
of at least three experiments.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
2521
Takashi Nonaka and others
Table 1. Flow cytometric analyses of infected and stsp-treated cells treated with annexin V/PI
The percentages of annexin V- and PI-double-positive (necrosis) or annexin V-positive and PI-negative (apoptosis) cells in infected cells
were calculated with the software CELLQuest (Becton Dickinson). A minimum of 20 000 cells were evaluated per sample. Each value represents the mean±SD from triplicate measurements.
Time post-infection (h)
Necrosis (%)
U937UD
0
1
3
5
Stsp for 3 h
2?1±0?3
3?1±0?5
6?4±0?7
8?7±0?8
1?8±0?7
U937IFN
U937RA
J774
U937UD
U937IFN
U937RA
J774
3?5±0?8
5?0±1?1
11?3±1?3
19?2±0?8
3?2±0?9
3?0±0?2
13?7±0?6
18?4±0?7
24?2±1?0
2?6±0?3
6?7±1?1
17?8±0?4
32?8±1?3
43?7±1?9
4?7±1?0
1?9±0?5
10?5±0?9
18?5±1?2
16?2±1?1
34?5±1?5
1?7±0?9
21?4±1?1
27?6±0?7
24?5±1?0
37?4±1?2
1?9±0?3
3?3±0?5
7?6±0?4
8?7±0?5
32?2±2?1
3?0±0?8
4?1±1?0
5?1±0?8
6?7±0?9
29?0±0?9
noticeable exposure of PS in the absence of PI uptake
compared to the other cell lines tested (Table 1). Thus,
Shigella-induced U937IFN apoptosis is accompanied by
activation of caspase-3/-7, but not caspase-1, and exposes
PS while its membrane is still intact.
To determine whether the ability of Shigella to generate
apoptosis is dependent on its virulence, we performed
infection experiments using the avirulent S. flexneri mutant
N1411, which lacks ipaBCDA. As shown in Fig. 1(b),
mutant N1411 did not induce rapid necrotic cell death in
any cell line used. However, the cleavage of PARP by
caspase-3/-7 was seen in infected U937IFN and U937UD
cells irrespective of pathogenicity (Fig. 3b). Furthermore,
U937IFN cells were incubated with N1411 at an m.o.i. of
Fig. 7. Infection of macrophage-like cells with Shigella results
in the formation of pores on the host cell membrane. U937RA
cells grown in a 12-well plate were pre-incubated with 20 mM
of several osmoprotectants and then infected with S. flexneri
YSH6000 (m.o.i. of 50). Cytotoxicity was measured 2 h postinfection by the LDH release assay. Data represent means±SD
from three separate experiments.
2522
Apoptosis (%)
50 for 5 h and caspase-3/-7 and caspase-1 activity in the
lysate was measured. Caspase-3/-7 but not caspase-1 activity
was detected as summarized in Table 2. Thus, avirulent
strains can also induce apoptosis in U937IFN cells. We then
examined whether S. flexneri killed by pre-treatment with
antibiotics or the E. coli strain JM109 could also induce
apoptosis in U937IFN cells. U937IFN cells were mixed with
the bacterial suspensions at the indicated m.o.i., incubated
for 5 h, and cytosolic fractions were prepared. Caspase-3/-7
activity was detected in the U937IFN line when it had been
exposed to killed wild-type Shigella and even E. coli JM109,
as summarized in Table 2. Caspase-1 activity was not
detected at all (data not shown).
Bacterial internalization is not required for
Shigella-induced U937IFN cell apoptosis
To examine whether phagocytosis of Shigella by host cells
is required for Shigella-induced U937IFN apoptosis, we
incubated U937IFN cells with cytochalasin D for 1 h
prior to incubation with wild-type Shigella or Gm-treated
bacteria at an m.o.i. of 50, and analysed the cleavage of
PARP in infected cells. In the absence of cytochalasin D,
the cleaved 85 kDa PARP fragment was detected 2 and 5 h
post-infection in U937IFN cells incubated with both the
wild-type and Gm-treated Shigella as shown in Fig. 8.
Moreover, the amount of cleaved PARP was enhanced in
the presence of cytochalasin D for both wild-type and
Gm-treated bacteria. TUNEL assays on U937IFN cells
infected with wild-type Shigella or N1411 (m.o.i. of 50)
showed that TUNEL-positive cells with nuclear condensation were detected in both the absence and the presence
of cytochalasin D (Fig. 9). Thus, the internalization of
Shigella is not required for the induction of U937IFN
apoptosis, indicating that extracellular Shigella can transmit the apoptosis signal into host cells through the cell
membrane in a manner that is not dependent on its
virulence.
DISCUSSION
While it has been reported that S. flexneri induces macrophage apoptosis (Zychlinsky et al., 1992; Chen et al., 1996;
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
Microbiology 149
Macrophage-like cell death by Shigella infection
Table 2. Activation of caspase-3/-7 by interaction with several bacteria
U937UD, U937IFN, U937RA and J774 cells were infected with wild-type YSH6000, avirulent N1411,
YSH6000 killed by pre-treatment with 100 mg Gm ml21, or E. coli strain JM109 at an m.o.i. of 50, or
treated with 0?5 mM stsp. After 5 h incubation, caspase-3/-7 activities were determined by using the
peptide substrate Ac-DEVD-MCA as described in Methods. Each value represents the mean±SD from
triplicate measurements.
Caspase-3/-7 activity [nmol AMC h”1 (mg protein)”1]
No treatment
Wild-type
DipaBCDA
Gm-treated wild-type
E. coli JM109
0?5 mM stsp
U937UD
U937IFN
U937RA
J774
0?8±0?1
2?3±0?6
2?9±0?7
2?3±0?3
1?8±0?6
37?5±2?1
1?2±0?6
11?0±0?4
8?5±1?1
10?6±0?9
6?5±0?4
34?3±1?3
0?6±0?4
1?5±0?9
0?6±0?4
1?8±0?6
1?0±0?6
31?3±2?4
0?5±0?2
0?8±0?4
0?5±0?2
0?7±0?3
0?8±0?3
32?9±1?9
Hilbi et al., 1998), this pathogen-induced cell death has
unusual features that distinguish it from typical apoptosis
in that activation of caspase-1 but not caspase-3/-7 occurs.
In general, during apoptosis, the executioner caspases,
which include caspase-3 and -7, are activated and these are
predominantly responsible for the limited proteolysis that
characterizes the apoptotic dismantling of the cell. Caspase1 is now thought not to be essential for the induction of
apoptosis (Kuida et al., 1995; Li et al., 1995, 1997). Rather, it
Cleaved PARP (a.u.)
600
500
400
300
200
100
0 2
_
5
0 2 5
+
Wild-type
0 2 5
_
0 2
5
+
Time (h)
Cyto D
Gm-treated wild-type
Fig. 8. Internalization of Shigella is not required for its induction of U937IFN apoptosis: PARP cleavage assay. U937IFN
cells were infected with wild-type Shigella strain YSH6000 or
YSH6000 killed by pre-treatment with 100 mg Gm ml”1 at an
m.o.i. of 50 in the absence (hatched bars) or presence (shaded
bars) of 5 mg ml”1 cytochalasin D. At 2 and 5 h post-infection,
infected cells were prepared for immunoblot analyses.
Approximately 56104 cells were separated by SDS-PAGE and
the gel was immunoblotted with a polyclonal antibody specific
for the amino-terminal region of the 85 kDa fragment of human
PARP. The immunostained band corresponding to the cleaved
PARP (85 kDa) was quantified using the software NIH image
1.60. Error bars represent SD.
http://mic.sgmjournals.org
appears to be associated with the production of cytokines
during inflammation.
Zychlinsky and co-workers previously showed that
Shigella induces IpaB-mediated apoptosis in macrophages
(Zychlinsky et al., 1992; Chen et al., 1996; Hilbi et al., 1998),
so we tried to reproduce their results using similar methods.
However, we detected only virulence-dependent cell death
of infected macrophage-like cells (Fig. 1), but not the
effective suppression of cell death by caspase-1 inhibitor
(Fig. 4), the binding of caspase-1 and IpaB, or the activation of caspase-1 (i.e. processing of pro-caspase-1) in
infected cells (data not shown). Recently, it has been
reported that the release of mature IL-1b appears to be
linked to the processing of precursor forms by caspase-1,
although functional caspase-1 has not been detected in
monocytes and macrophages (Ayala et al., 1994; Wewers
et al., 1997; Mehta et al., 2001). The researchers also
suggested that caspase-1 may have a role in release of
mature IL-1b that is separate from its function as a
protease. Furthermore, it has been also reported that
another protease(s) could be involved in the processing of
the pro-form of IL-1b (Schonbeck et al., 1998). These
findings raised the possibility that caspase-1 may not be
associated with induction of cell death by infection of
Shigella. Therefore, we tried to re-analyse cell death of
infected macrophages by different methods, including
direct assay of caspase activity using synthetic peptide
substrates, flow cytometric analysis using annexin V/PI
staining, and confocal laser microscopic analysis with
TUNEL staining. To examine if the activity of caspase-1
could be required for induction of cell death by Shigella
infection, we directly assayed the activity of caspase-1 with
a specific synthetic peptide substrate, Ac-WEHD-MCA.
However, no activity was detected in lysates from U937RA
and J774 cells infected with wild-type Shigella. Similarly,
the activity of caspase-3 was not detected at all. We also
observed that caspase-1 inhibitor, Ac-YVKD-CHO, had
only partial effects on cell death by Shigella early during
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
2523
Takashi Nonaka and others
Transmission
TUNEL
Anti-LPS
(a)
(b)
Fig. 9. Internalization of Shigella is not required for its induction of U937IFN apoptosis: immunostaining and TUNEL assay.
U937IFN cells were incubated with the virulent Shigella strain YSH6000 (a) or the avirulent mutant N1411, which lacks the
ipaBCDA genes (b), at an m.o.i. of 50 in the absence (upper panels) or presence (lower panels) of 5 mg cytochalasin D ml”1.
At 5 h post-infection, infected cells were stained with Cy5-labelled anti-S. flexneri 2a LPS and by the TUNEL method, which
detects fragmented DNA. The cells were observed by confocal laser microscopy (606 objective). Results are representative
of at least three experiments.
infection, correlating with the results of the caspase-1
activity assay. We conclude that caspase-1 may have only
partial effects on Shigella-induced cell death early during
2524
infection under our experimental conditions, but caspase3/-7 do not contribute to this cell death. Also, we can not
rule out the possibility that caspase(s) other than caspase-1,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
Microbiology 149
Macrophage-like cell death by Shigella infection
-3 and -7 are associated with induction of Shigella-induced
cell death, because Z-VAD-fmk significantly suppressed
cell death by Shigella at 2 h post-infection. Furthermore,
to distinguish clearly necrosis from apoptosis in infected
cells, we examined infected cells with annexin V/PI staining
or the TUNEL method. It has been reported that the
cleavage of chromatin DNA is caused not only in apoptotic
cells but also in necrotic cells (Dong et al., 1997; de Torres
et al., 1997). In apoptotic cells, the cleavage of chromatin
DNA is accompanied by nuclear condensation. On the
other hand, necrotic cells showed DNA fragmentation
without nuclear condensation. Our results clearly revealed
that macrophage-like cell death by infection of Shigella is
necrosis rather than classical apoptosis.
A possible mechanism by which Shigella kills macrophagelike cells is that it inserts a pore into the plasma membrane
that causes osmotic lysis. This is suggested by the work of
Blocker et al. (1999), who showed that Shigella can lyse
RBCs by inserting a 2?5 nm pore into the cell membrane.
However, it was initially not clear if the macrophage-like
cell death induced by wild-type Shigella occurs through a
similar mechanism, because Blocker et al. (1999) examined
the causes of Shigella-induced cytotoxicity only with RBCs,
not with living macrophage cells. Furthermore, the Shigellainduced haemolysis requires centrifugation (Blocker et al.,
1999), indicating that Shigella can lyse the plasma
membrane by contact from the outside of the RBCs. In
contrast, we have shown here that the Shigella-induced
macrophage-like cell death can be completely inhibited
by the pre-treatment of the host cells with cytochalasin
D. Thus, phagocytosis is a prerequisite for Shigellainduced macrophage-like cell death, which indicates that
only internalized bacteria can lyse the plasma membrane
of macrophage-like cells. Nevertheless, we have shown
that Shigella kills macrophage-like cells in a manner that
resembles haemolysis, at least superficially, in that it
forms a pore that could result in osmotic lysis. Although
the pore size we measured (2?87±0?4 nm) on macrophagelike cells was very similar to that (2?5 nm) on RBCs (Blocker
et al., 1999), it is not yet clear if the same pore is involved
in both phenomena. Supporting the possibility that the
same pore is involved is that Blocker et al. (1999) have
found that pore formation in haemolysis is dependent on
both IpaB and IpaC. Similarly, macrophage-induced cell
death is dependent on the ipaBCDA genes, because Shigellainduced killing does not occur upon infection with the
avirulent mutant of the YSH6000 strain, N1411, which
lacks ipaBCDA.
In intestinal mucosa, phagocytic cells such as macrophages
have a variety of phenotypes. It is known that CD11b is one
of the representative surface markers of macrophages, and
the expression level is not constant between cell types (i.e.
source organs or differentiation stages) (Rogler et al., 1998).
Therefore, it has been suggested that phagocytic cells with
different levels of CD11b could be a relevant target for
Shigella. The U937 cell line can be induced to differentiate
http://mic.sgmjournals.org
into macrophage-like cell lines by treatment with RA or
IFN, generating the cell lines we have denoted as U937RA
and U937IFN. These two lines differ in that U937RA
appears to be more macrophage-like since the cells generate
more superoxide and express higher levels of CD11b
(Kikuchi et al., 1994, 1996). In this study, we used
U937IFN and U937RA as reproducibly available cell lines
representing the different cell types in the intestinal
mucosa, not the partially differentiated myeloid cells. We
have shown that Shigella induces necrosis in macrophagelike cells with higher expression of CD11b and apoptosis
in cells with lower expression levels. Surprisingly, the
inhibition of phagocytosis by cytochalasin D did not
suppress Shigella-induced U937IFN apoptosis; rather, it
appeared to enhance it. In addition, not only the wild-type
YSH6000 strain but also the avirulent N1411 mutant,
Gm-killed YSH6000, and the E. coli strain JM109 could
induce apoptosis in U937IFN. These observations suggest
that extracellular bacteria can transmit apoptotic signals
in a manner that is independent of their pathogenicity.
Susceptibility to such apoptotic signals was not observed
for U937RA and J774 cells. One possible molecule involved
in this process could be the Toll-like receptor (TLR)-2
(Kirschning et al., 1998). Aliprantis et al. (1999) have
reported that bacterial lipoproteins, which are expressed by
all bacteria, are potent activators of TLR-2 and that TLR-2
transmits a pro-apoptotic signal. Recently, Y. Niikura and
co-workers (personal communication) showed that Fas
is up-regulated in U937IFN cells, while the receptor is
down-regulated in U937RA cells. U937IFN is more sensitive
to apoptotic stimuli through Fas than U937RA. Similar
regulation may be seen in the case of other receptor(s) e.g.
TLR-2 and porimin (Ma et al., 2001), which is thought to
be a receptor associated with induction of oncosis. These
findings suggest that the expression level of such receptors
involved in induction of cell death could be associated with
Shigella’s ability to induce necrosis or apoptosis.
In summary, we have found that infection of macrophagelike cell lines with S. flexneri strain YSH6000 results in death
that is due mainly to the rapid induction of necrosis. The
mechanism may involve the formation of a pore in the
plasma membrane of infected cells, which is dependent on
the ipaBCDA virulence genes. Shigella can also induce death
in other cells, without even being taken up, through a
mechanism that does not involve the ipaBCDA genes and
results in typical apoptosis. These observations point to the
ability of Shigella to control the way it kills host cells, as it
clearly commands mechanisms that lead to either apoptosis
or necrosis. This ability most likely plays a crucial role in
its initiation of infection, survival, and escape from host
immune responses.
ACKNOWLEDGEMENTS
We thank Chihiro Sasakawa and Toshihiko Suzuki for an anti-LPS
antibody and helpful discussions, Dai Ishikawa and Takayasu Noguchi
for helpful advice in the osmoprotectant experiments, and Shionogi
Pharmaceuticals for IFN.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
2525
Takashi Nonaka and others
This work was supported in part by grants from the Ministry of
Education, Science, Sports and Culture of Japan and from Research
Fellowships of the Japan Society for the Promotion of Science for
Young Scientists.
human macrophages prevents the mycobacteria from spreading and
induces mycobacterial growth inhibition by freshly added, uninfected
macrophages. J Immunol 158, 4320–4327.
Gao, L. Y. & Abu Kwaik, Y. (2000). Hijacking of apoptotic pathways
by bacterial pathogens. Microb Infect 2, 1705–1719.
Hakansson, S., Schesser, K., Persson, C., Galyov, E. E.,
Rosqvist, R., Homble, F. & Wolf-Watz, H. (1996). The YopB protein
REFERENCES
Aliprantis, A. O., Yang, R. B., Mark, M. R., Suggett, S., Devaux,
B., Radolf, J. D., Kimpel, G. R., Godowski, P. & Zychlinsky, A. (1999).
Cell activation and apoptosis by bacterial lipoproteins through
Toll-like receptor-2. Science 285, 736–739.
Ayala, J. M., Yamin, T. T., Egger, L. A., Chin, J., Kostura, M. J. &
Miller, D. K. (1994). IL-1 beta-converting enzyme is present in
monocytic cells as an inactive 45-kDa precursor. J Immunol 153,
2592–2597.
of Yersinia pseudotuberculosis is essential for the translocation of
Yop effector proteins across the target cell plasma membrane and
displays a contact-dependent membrane disrupting activity. EMBO J
15, 5812–5823.
Hersh, D., Monack, D. M., Smith, M. R., Ghori, N., Falkow, S. &
Zychlinsky, A. (1999). The Salmonella invasin SipB induces
macrophage apoptosis by binding to caspase-1. Proc Natl Acad Sci
U S A 96, 2396–2401.
Blocker, A., Gounon, P., Larquet, E., Niebuhr, K., Cabiaux, V.,
Parsot, C. & Sansonetti, P. (1999). The tripartite type III secreton of
Hilbi, H., Moss, J. E., Hersh, D. & 7 other authors (1998). Shigellainduced apoptosis is dependent on caspase-1 which binds to IpaB.
J Biol Chem 273, 32895–32900.
Shigella flexneri inserts IpaB and IpaC into host membranes. J Cell
Biol 147, 683–693.
Kato, M., Nonaka, T., Maki, M., Kikuchi, H. & Imajoh-Ohmi, S.
(2000). Caspases cleave the amino-terminal calpain inhibitory unit
Boise, L. H. & Collins, C. M. (2001). Salmonella-induced cell death:
apoptosis, necrosis or programmed cell death? Trends Microbiol 9,
64–67.
of calpastatin during apoptosis in human Jurkat T cells. J Biochem
127, 297–305.
Bradford, M. M. (1976). A rapid and sensitive method for the
Kikuchi, H., Fujinawa, T., Kuribayashi, F., Nakanishi, A., ImajohOhmi, S. & Kanegasaki, S. (1994). Induction of essential
quantitation of microgram quantities of protein utilizing the
principle of protein-dye binding. Anal Biochem 72, 248–254.
components of the superoxide generating system in human
monoblastic leukemia U937 cells. J Biochem 116, 742–746.
Brennan, M. A. & Cookson, B. T. (2000). Salmonella induces
Kikuchi, H., Iizuka, R., Sugiyama, S., Gon, G., Mori, H., Arai, M.,
Mizumoto, K. & Imajoh-Ohmi, S. (1996). Monocytic differentiation
macrophage death by caspase-1-dependent necrosis. Mol Microbiol
38, 31–40.
Chen, Y., Smith, M. R., Thirumalai, K. & Zychlinsky, A. (1996). A
bacterial invasin induces macrophage apoptosis by binding directly
to ICE. EMBO J 15, 3853–3860.
Clifton, D. R., Goss, R. A., Sahni, S. K., Antwerp, D. V., Baggs, R. B.,
Marder, V. J., Silverman, D. J. & Sporn, L. A. (1998). NF-kappa
modulates apoptotic response to cytotoxic anti-Fas antibody and
tumor necrosis factor alpha in human monoblast U937 cells. J Leukoc
Biol 60, 778–783.
Kirby, J. E., Vogel, J. P., Andrews, H. L. & Isberg, R. R. (1998).
Evidence for pore-forming ability by Legionella pneumophila. Mol
Microbiol 27, 323–336.
B-dependent inhibition of apoptosis is essential for host cell survival
during Rickettsia rickettsii infection. Proc Natl Acad Sci U S A 95,
4646–4651.
Kirschning, C. J., Wesche, H., Ayres, T. M. & Rothe, M. (1998).
Dacheux, D., Goure, J., Cabert, J., Usson, Y. & Attree, I. (2001).
Kuida, K., Lippke, J. A., Ku, G., Harding, M. W., Livingston, D. J., Su,
M. S. & Flavell, R. A. (1995). Altered cytokine export and apoptosis
Pore-forming activity of type III system-secreted proteins leads to
oncosis of Pseudomonas aeruginosa-infected macrophages. Mol
Microbiol 40, 76–85.
de Torres, C., Munell, F., Ferrer, I., Reventos, J. & Macaya, A. (1997).
Identification of necrotic cell death by the TUNEL assay in the
hypoxic-ischemic neonatal rat brain. Neurosci Lett 230, 1–4.
Dong, Z., Saikumar, P., Weinberg, J. M. & Venkatachalam, M. A.
(1997). Internucleosomal DNA cleavage triggered by plasma
membrane damage during necrotic cell death. Involvement of
serine but not cysteine proteases. Am J Pathol 151, 1205–1213.
Fan, T., Lu, H., Hu, H., Shi, L., McClarty, G. A., Nance, D. M.,
Greenberg, A. H. & Zhong, G. (1998). Inhibition of apoptosis in
chlamydia-infected cells: blockade of mitochondrial cytochrome c
release and caspase activation. J Exp Med 187, 487–496.
Fernandez-Prada, C. M., Hoover, D. L., Tall, B. D. & Venkatesan,
M. M. (1997). Human monocyte-derived macrophages infected with
virulent Shigella flexneri in vitro undergo a rapid cytolytic event
similar to oncosis but not apoptosis. Infect Immun 65, 1486–1496.
Fernandez-Prada, C. M., Hoover, D. L., Tall, B. D., Hartman, A. B.,
Kopelowitz, J. & Venkatesan, M. M. (2000). Shigella flexneri IpaH7?8
facilitates escape of virulent bacteria from the endocytic vacuoles of
mouse and human macrophages. Infect Immun 68, 3608–3619.
Fratazzi, C., Arbeit, R. D., Carini, C. & Remold, H. G. (1997).
Programmed cell death of Mycobacterium avium serovar 4-infected
2526
Human Toll-like receptor 2 confers responsiveness to bacterial
lipopolysaccharide. J Exp Med 188, 2091–2097.
in mice deficient in interleukin-1 beta converting enzyme. Science
267, 2000–2003.
Lazebnik, Y. A., Kaufmann, S. H., Desnoyers, S., Poirier, G. G. &
Earnshaw, W. C. (1994). Cleavage of poly(ADP-ribose) polymerase
by a proteinase with properties like ICE. Nature 371, 346–347.
Li, P., Allen, H., Banerjee, S. & 7 other authors (1995). Mice
deficient in IL-1 beta-converting enzyme are defective in production of mature IL-1 beta and resistant to endotoxic shock. Cell
80, 401–411.
Li, P., Allen, H., Banerjee, S. & Seshadri, T. (1997). Characterization
of mice deficient in interleukin-1 beta converting enzyme. J Cell
Biochem 64, 27–32.
Ma, F., Zhang, C., Prasad, K. V., Freeman, G. J. & Schlossman,
S. F. (2001). Molecular cloning of Porimin, a novel cell surface
receptor mediating oncotic cell death. Proc Natl Acad Sci U S A 98,
9778–9783.
Mehta, V. B., Hart, J. & Wewers, M. D. (2001). ATP-stimulated release
of interleukin (IL)-1b and IL-18 requires priming by lipopolysaccharide and is independent of caspase-1 cleavage. J Biol Chem
276, 3820–3826.
Monack, D. M., Raupach, B., Hromocky, A. E. & Falkow, S. (1996).
Salmonella typhimurium invasion induces apoptosis in infected
macrophages. Proc Natl Acad Sci U S A 93, 9833–9838.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
Microbiology 149
Macrophage-like cell death by Shigella infection
Moran, O., Zegarra-Moran, O., Virginio, C., Gusmani, L. &
Rottini, G. D. (1992). Physical characterization of the pore forming
Scherrer, R. & Gerhardt, P. (1971). Molecular sieving by the Bacillus
megaterium cell wall and protoplast. J Bacteriol 107, 718–735.
cytolysine from Gardnerella vaginalis. FEMS Microbiol Immunol
5, 63–69.
Schesser, K., Spiik, A. K., Dukuzumuremyi, J. M., Neurath, M. F.,
Pettersson, S. & Wolf-Watz, H. (1998). The yopJ locus is required for
Nonaka, T., Kuwae, A., Sasakawa, C. & Imajoh-Ohmi, S. (1999).
Yersinia-mediated inhibition of NF-kappaB activation and cytokine
expression: YopJ contains a eukaryotic SH2-like domain that is
essential for its repressive activity. Mol Microbiol 28, 1067–1079.
Shigella flexneri YSH6000 induces two types of cell death, apoptosis
and oncosis, in the differentiated human monoblastic cell line
U937. FEMS Microbiol Lett 174, 89–95.
Orth, K., Palmer, L. E., Bao, Z. Q., Stewart, S., Rudolph, A. E., Bliska,
J. B. & Dixon, J. E. (1999). Inhibition of the mitogen-activated
protein kinase kinase superfamily by a Yersinia effector. Science 285,
1920–1923.
Rogler, G., Hausmann, M., Vogl, D., Aschenbrenner, E., Andus, T.,
Falk, W., Andreesen, R., Scholmerich, J. & Gross, V. (1998).
Isolation and phenotypic characterization of colonic macrophages.
Clin Exp Immunol 112, 205–215.
Ruckdeschel, K., Roggenkamp, A., Lafont, V., Mangeat, P.,
Hessemann, J. & Rouot, B. (1997). Interaction of Yersinia
enterocolitica with macrophages leads to macrophage cell death
through apoptosis. Infect Immun 65, 4813–4821.
Sasakawa, C., Kamata, K., Sakai, T., Murayama, S. Y., Makino, S. &
Yoshikawa, M. (1986). Molecular alteration of the 140-megadalton
plasmid associated with loss of virulence and Congo red binding
activity in Shigella flexneri. Infect Immun 51, 470–475.
Schonbeck, U., Mach, F. & Libby, P. (1998). Generation of
biologically active IL-1 beta by matrix metalloproteinases: a novel
caspase-1-independent pathway of IL-1 beta processing. J Immunol
161, 3340–3346.
Thornberry, N. A., Rano, T. A., Peterson, E. P. & 9 other authors
(1997). A combinatorial approach defines specificities of members of
the caspase family and granzyme B. J Biol Chem 272, 17907–17911.
Watarai, M., Kamata, Y., Kozaki, S. & Sasakawa, C. (1997). rho, a
small GTP-binding protein, is essential for Shigella invasion of
epithelial cells. J Exp Med 185, 281–292.
Watson, P. R., Gautier, A. V., Paulin, S. M., Bland, A. P., Jones, P. W.
& Wallis, T. S. (2000). Salmonella enterica serovars Typhimurium
and Dublin can lyse macrophages by a mechanism distinct from
apoptosis. Infect Immun 68, 3744–3747.
Wewers, M. D., Dare, H. A., Winnard, A. V., Parker, J. M. & Miller,
D. K. (1997). IL-1 beta-converting enzyme (ICE) is present and
functional in human alveolar macrophages: macrophage IL-1 beta
release limitation is ICE independent. J Immunol 159, 5964–5972.
Sasakawa, C., Kamata, K., Sakai, T., Makino, S., Yamada, M.,
Okada, N. & Yoshikawa, M. (1988). Virulence-associated genetic
Zychlinsky, A., Prevost, M. C. & Sansonetti, P. J. (1992). Shigella
regions comprising 31 kilobases of the 230-kilobase plasmid in
Shigella flexneri 2a. J Bacteriol 170, 2480–2484.
flexneri induces apoptosis in infected macrophages. Nature 358,
167–169.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 01:17:13
2527