Autophagy Limits Listeria monocytogenes Intracellular Growth in the ...

[Autophagy 3:2, 117-125; March/April 2007]; ©2007 Landes Bioscience

Research Paper

Autophagy Limits Listeria monocytogenes Intracellular Growth in the Early Phase of Primary Infection Bénédicte F. Py† Marta M. Lipinski† Junying Yuan*

Abstract

Previously published online as an Autophagy E-Publication: http://www.landesbioscience.com/journals/autophagy/abstract.php?id=3618

Key words

Abbreviations

RIB

IEN

listeriolysin O mouse embryonic fibroblast phosphatidylinositol-specific phospholipase C broad-range phospholipase C multiplicity of infection microtubule-associated protein light chain 3

CE

.D

autophagy, Listeria monocytogenes, innate immunity, listeriolysin O, phospholipase C

UT E

.

Original manuscript submitted: 08/29/06 Manuscript accepted: 11/27/06

IST

*Correspondence to: Junying Yuan; Department of Cell Biology; Harvard Medical School; 240 Longwood Avenue; Boston, Massachusetts 02115 USA; Tel.: 617.432.4170; Fax: 617.432.4177; Email: [email protected]

OT D

†These authors contributed equally to this work.

ON

Department of Cell Biology; Harvard Medical School; Boston, Massachusetts USA

Introduction

SC

Listeria monocytogenes is a food‑borne pathogen capable of infecting humans and other animal species to cause listeriosis, a severe gastroenteritis with a possible central nervous system infection and up to 30% mortality.1‑3 On the cellular level, L. monocytogenes, a gram‑positive bacterium, is taken up by the host cell through pathogen‑induced endocytosis and enclosed in an endocytic vacuole (subsequently referred to as the vacuole). Listeriolysin O (LLO), a bacterial cholesterol‑dependent cytolysin, is required to perforate the vacuole membrane and gain access to the host cell cytosol. In addition to LLO, two bacterial phospholipases, a phosphatidylinositol (PI) specific phospholipase C (PI‑PLC) and a broad‑range phospholipase C (PC‑PLC), encoded respectively by the plcA and plcB genes, play a role in escape from the vacuole.2,4,5 PI‑PLC and PC‑PLC have been shown to hydrolyze host lipids to produce diacylglycerol and inositol phosphate and ceramide, respectively, and to participate in signaling events aimed at subverting cellular functions.2 Following escape from the vacuole, the bacteria grow and divide in the host cytosol. The product of the actA gene induces the polymerization of host actin which eventually forms a comet‑like tail propelling bacterial movement within the host cytosol and is necessary for spreading to neighboring cells. L. monocytogenes strains lacking the actA gene are able to infect mammalian cells and grow similarly to wild‑type but are defective in cell‑to‑cell spread.6 Autophagy is a cellular catabolic process mediating degradation of bulk cytoplasm, including organelles, through lysosomal proteolysis.7,8 Autophagy sequesters a portion of the cytoplasm within a double‑membrane vesicle named autophagosome, which then fuses with lysosomes, delivering its cargo for degradation by lysosomal proteases. Autophagy has been shown to be essential for cellular survival under stress conditions, including starvation,9,10 as well as to play a role in the survival of organisms during neonatal development11 and in the protection from neurodegeneration.12,13 Additionally, recent data

BIO

Acknowledgements

ES

LLO MEF PI-PLC PC-PLC MOI LC3

Autophagy has been recently proposed to be a component of the innate cellular immune response against several types of intracellular microorganisms. However, other intracellular bacteria including Listeria monocytogenes have been thought to evade the autophagic cellular surveillance. Here, we show that cellular infection by L. monocytogenes induces an autophagic response, which inhibits the growth of both the wild‑type and a DactA mutant strain, impaired in cell‑to‑cell spreading. The onset of early intracel‑ lular growth is accelerated in autophagy‑deficient cells, but the growth rate once bacteria begin to multiply in the cytosol does not change. Moreover, a significant fraction of the intra‑ cellular bacteria colocalize with autophagosomes at the early time‑points after infection. Thus, autophagy targets L. monocytogenes during primary infection by limiting the onset of early bacterial growth. The bacterial expression of listeriolysin O but not phospholipases is necessary for the induction of autophagy, suggesting a possible role for permeabiliza‑ tion of the vacuole in the induction of autophagy. Interestingly, the growth of a DplcA/B L. monocytogenes strain deficient for bacterial phospholipases is impaired in wild‑type cells, but restored in the absence of autophagy, suggesting that bacterial phospholipases may facilitate the escape of bacteria from autophagic degradation. We conclude that L. monocytogenes are targeted for degradation by autophagy during the primary infection, in the early phase of the intracellular cycle, following listeriolysin O‑dependent vacuole perforation but preceding active multiplication in the cytosol, and that expression of bacterial phospholipases is necessary for the evasion of autophagy.

©

20

06

LA

ND

We thank Noboru Mizushima for the gift of anti-LC3 antibody, GFP-LC3 expression vector and Atg5-/- and wild-type control MEFs, Darren E. Higgins for the gift of the bacteria strains, helpful suggestions during the course of this work and for critical reading of the manuscript, Laura S. Burrack for technical advice and Jennifer C. Waters and Lara Petrak of the Nikon Image Center at Harvard Medical School for help with fluorescent microscopy. This work was supported in part by R37AG12859 and PO1AG027916 from National Institute of Aging (to J.Y.) and Lavoisier fellowship from French Foreign Ministry (to B.P.).

www.landesbioscience.com

Autophagy

117

Autophagy Limits L. Monocytogenes Intracellular Growth

suggest that autophagy plays an essential role in both acquired14 and innate immune responses. In fact, several intracellular bacterial pathogens, including Group A Streptococcus,15 Salmonella enterica16 and Mycobacterium tuberculosis,17 have been shown to be ensnared and degraded by autophagy. However, some intracellular pathogens, for example Shigella flexneri, are able to escape the autophagic surveillance while others such as Coxiella burnetii take advantage of this cellular response.18,19 Under normal growth conditions L. monocytogenes has been thought to be able to evade the autophagic cellular surveillance.20 However, accumulative evidence suggest a possible role for autophagy in L. monocytogenes infection. L. monocytogenes can be trapped within multilayer membrane vesicles during impaired secondary infection,21 and cytoplasmic bacteria are engulfed in AVs when treated with chloramphenicol, a specific inhibitor of bacterial protein synthesis.6 In this study, we examined whether under normal growth conditions L. monocytogenes may also be a target for autophagy. Our data demonstrate that infection of mouse embryonic fibroblast cells (MEFs) with wild‑type L. monocytogenes leads to the induction of autophagy and the uptake of the intracellular bacteria by the autophagosomes. Autophagy plays a protective role during L. monocytogenes infection and can limit bacterial growth independently of secondary spreading. Induction of autophagy is dependent on the expression of LLO, suggesting that vacuole permeabilization is a prerequisite for autophagy. Autophagy also appears to be responsible for the growth defect of a mutant L. monocytogenes strain deficient for production of bacterial phospholipases, suggesting a role for the phospholipases in the evasion of autophagy.

Table 1

L. monocytogenes strains used in this study

Strain Genotype Reference 10403S

Wild‑type strain

DP‑L2161

10403S Dhly

35 36

DP‑L3078

10403S DactA

37

DP‑L1936

10403S DplcA/B

27

DH‑L1039

10403S GFP

38

15‑ml conical tubes containing 5 ml of sterile water. After vortexing, appropriate dilutions were plated onto LB agar plates. Light microscopy. The coverslips were stained with Diff‑Quik (DADE‑Behring) according the manufacturer’s protocol, and analyzed by light microscopy. Fluorescence microscopy. For the colocalization experiment of L. monocytogenes wthin GFP‑LC3 autophagosomes, a total of 1.0 x 106 H4 cells stably transfected with a vector encoding GFP‑LC3 (a kind gift from N. Mizushima) were seeded one day prior to infection in 60‑mm‑diameter culture dishes containing 12‑mm‑diameter round glass coverslips. Overnight cultures of 10403S wild‑type bacteria were used to infect monolayers of host cells at a MOI of 500:1 (bacterium/ host cell ratio). 1 h after infection the monolayers were washed two times in PBS buffer, and DMEM + 10% NCS medium containing gentamicin (10 mg/ml) was added. Two hours and 6 h after infection, the cells were fixed in 3.2% formaldehyde (Polysciences) in PBS for 15 min and permeabilized in 0.1% Triton X‑100 in PBS. The cells were then incubated successively in PBS‑BSA (PBS, BSA 0.2%) for 30 min, with a rabbit polyclonal anti‑Listeria antiserum (1:200, Materials and Methods Difco) in PBS‑BSA for 1 h, and with rhodamine red conjugated Bacterial and eukaryotic cell growth conditions. The bacterial goat anti‑rabbit (1:200) in PBS‑BSA for 30 min. The coverslips were strains used in the present study were kind gifts from D.E. Higgins washed and mounted on slides using vectashield mounting medium (Harvard Medical School, Boston, Massachusetts) and are listed containing DAPI (Vector laboratories). For the colocalization experiin Table 1. L. monocytogenes strains were grown in brain heart ment of L. monocytogenes and polymerized actin, a total of 0.8 x 106 infusion (BHI) medium (Difco). Atg5‑/‑ mouse embryonic fibro- wild‑type or Atg5‑/‑ MEFs were seeded 1 day prior to infection in blasts (MEFs) transformed with pEF321‑T and matched wild‑type 60‑mm‑diameter culture dishes containing 12‑mm‑diameter round counterparts were a kind gift from N. Mizushima (Tokyo Metropolitan glass coverslips. Overnight cultures of the 10403S‑GFP or DplcA/B Institute of Medical Science, Tokyo, Japan).11 MEFs were strains were used to infect monolayers of host cells at a MOI of maintained in Dulbecco’s modified Eagle’s Medium (DMEM) 500:1 (bacterium/host cell ratio). 1 h after infection monolayers were (Gibco) supplemented with 10% Newborn Calf Serum (NCS) and washed two times in PBS buffer, and DMEM + 10% NCS medium 1X penicillin/streptomycin (PS) (Gibco). H4 cells were transfected containing gentamicin (10 mg/ml) was added. 4 h after infection, with pEGFPC1‑LC3 encoding the GFP‑LC3 fusion protein (a kind cells were fixed as described above and permeabilized in TBS‑TX gift from N. Mizushima), followed by selection with 1.5 mg/ml (25 mM Tris‑HCl pH 8.0, 150 mM NaCl, 0.1% Triton X‑100). The G418 (Sigma). H4 cells were maintained in DMEM supplemented cells infected with 10403S‑GFP were then successively incubated with 10% NCS, 1X PS, 1X sodium pyruvate (Gibco) and 1.5 mg/ml in antibody buffer (TBS‑TX, 10% donkey serum) for 30 min and G418. with Texas Red‑X phalloidin 1/200 (Molecular Probes) in antibody Intracellular growth assay in MEF cells. A total of 0.8 x 106 buffer for 30 min. The cells infected with DplcA/B were incubated Atg5‑/‑ or control wild‑type MEFs were seeded 1 day prior to infection with anti‑Listeria antibody (1:200) in antibody buffer for 1 h, and in 60‑mm‑diameter culture dishes containing 14–12‑mm‑diameter with Alexa488 conjugated donkey anti‑rabbit (1:200, Molecular round glass coverslips. L. monocytogenes strains were grown overnight Probes) in antibody buffer for 30 min prior to phalloidin‑Texas Red. in 2 ml of BHI medium at 30˚C without shaking. L. monocytogenes Coverslips were then processed as already described. Fluorescence overnight cultures grown under these conditions contained ~2 x 109 microscopy was performed with a Nikon 90i Automated Upright bacteria/ml. Bacterial cultures were used to infect monolayers of host microscope. Colocalization was evaluated in at least 40 random fields cells at a multiplicity of infection of 50:1 (MOI, bacterium/host cell for each experiment. ratio). At 1 h after infection, monolayers were washed 2 times in PBS Western blot analysis. A total of 7 x 104/well Atg5‑/‑ or wild‑type buffer, and DMEM + 10% NCS medium containing gentamicin MEFs were seeded 1 day prior to infection in 24‑well‑plates. (10 mg/ml) was added to kill all extracellular bacteria. The numbers L. monocytogenes overnight cultures were used to infect monolayers of bacteria per coverslip was determined at the time points indicated of host cells at a MOI of 500:1 (bacterium/host cell ratio). 1 h after in each figure by separately placing coverslips, in triplicate, into infection, monolayers were washed two times in PBS buffer, and 118

Autophagy

2007; Vol. 3 Issue 2


DMEM + 10% NCS medium containing gentamicin (10 mg/ml) was added. Cells were collected at indicated time points, lysed in 2X Laemmli sample buffer, analyzed by SDS‑PAGE, transferred to PVDF membranes, blocked and probed according to standard procedures. The following antibodies and dilutions were used: anti‑LC3 (1:1000) (a kind gift from N. Mizushima), anti‑LC3 (1:200) (MBL), anti‑tubulin (1:5000) (Cell Signaling). Western blot quantification was performed with the NIH image software.

Results Listeria monocytogenes infection induces autophagy and autophagy limits bacterial intracellular growth. In order to determine whether infection with L. monocytogenes can stimulate autophagy in host cells, we infected mouse embryonic fibroblast cells (MEFs) with wild‑type L. monocytogenes (strain 10403S) and followed the induction of cellular autophagy by measuring the conversion of

Figure 1. L. monocytogenes infection induces autophagy and autophagy limits bacterial intracellular growth. (A) Wild‑type and Atg5‑/‑ MEFs were infected with the wild‑type 10403S L. monocytogenes strain. LC3‑I to LC3‑II conversion was analyzed by Western blot at different time points after infection as indicated. The LC3‑II/LC3‑I ratio was quantified. The presented result is representative of 3 independent experiments. (B) Wild‑type and Atg5‑/‑ MEFs were treated with 5 mM rapamycin for 24 h. LC3‑I to LC3‑II conversion was analyzed by Western blot and quantified. (C) Wild‑type MEFs were infected with the 10403S L. monocytogenes strain and 8 h later the LC3‑I to LC3‑II conversion was quantified by Western blot. Results are presented as the ratio between LC3‑II/LC3‑I in infected vs uninfected cells in a set of six independent experiments (T‑test, p < 0.001). (D) Atg5‑/‑ MEFs and their wild‑type counterparts were infected with 10403S L. monocytogenes. Bacterial intracellular growth was measured at different time points after infection as described in Materials and Methods. The data are representative of at least three independent experiments. (E) Atg5‑/‑ MEFs and their wild‑type counterparts were infected with wild‑type 10403S L. monocytogenes for 8 h and analyzed by light microscopy after Diff‑Quik staining (x100). (F) H4 cells stably expressing GFP‑LC3 were infected with 10403S L. monocytogenes. Two hours (a–d) or 6 h after infection, colocalization of the bacteria (a) and the autophagosome marker LC3 (b) was analyzed by fluorescent microscopy. An enlargement of (c) is represented in (d). The proportion of bacteria colocalizing with GFP‑LC3 was quantified for each time point.


Autophagy

119


the specific autophagy marker LC3 at different time points after infection. During autophagy, the cytosolic LC3‑I form is converted to the LC3‑II form, which is covalently linked to phospholipids and associated with autophagosomes.22 As this modification alters LC3 gel migration properties, the conversion of LC3 can be followed by Western blot. We detected a significant increase in the LC3‑II/LC3‑I ratio over the time of infection as compared to noninfected cells (Figs. 1A, 1C). Induction of autophagy was noticeable by 2 h after infection, which corresponds to an early stage of the bacteria intracellular cycle. As a control the same experiment was performed in Atg5‑/‑ MEFs as Atg5 is an essential mediator of LC3‑I to LC3‑II conversion and Atg5‑/‑ cells are unable to execute autophagy.23 As expected, no LC3 conversion was detected in infected Atg5‑/‑ MEFs. As a positive control, wild‑type and Atg5‑/‑ MEFs were treated with rapamycin (5 mM) for 24 h. As expected, rapamycin, an inhibitor of the Tor kinase, induced an increase in the LC3‑II/LC3‑I ratio in the wild‑type but not in the Atg5‑/‑ MEFs (Fig. 1B). Autophagy has been reported to be induced following infection by many intracellular bacterial pathogens, however, the outcome of this process on pathogen growth is highly variable.24 In order to determine whether autophagy limits or favors its intracellular growth, we compared the growth rate of L. monocytogenes in Atg5‑/‑ MEFs and in their wild‑type counterparts. L. monocytogenes infected both cell types with similar efficiency, and was able to multiply in both wild‑type and Atg5‑/‑ MEFs. However, intracellular growth was delayed by approximately 2 hours in the wild‑type MEFs as compared to the Atg5‑/‑ cells (Fig. 1D). Atg5‑/‑ MEFs are thus more permissive for L. monocytogenes growth than their wild‑type counterparts. After 8 h of infection, wild‑type and Atg5‑/‑ MEFs were analyzed by light microscopy. The proportion of infected cells was similar in both cell types. However, the number of bacteria per infected cell was much higher in Atg5‑/‑ MEFs as compared to wild‑type MEFs (Fig. 1E). Intracellular L. monocytogenes localize to the autophagosomes. In order to determine if autophagy limits bacterial growth by directly taking up intracellular L. monocytogenes and targeting it for degradation, we investigated whether the bacteria colocalize with autophagosomes at different time points following the infection. For this purpose we used human H4 cells expressing GFP‑LC3. Similarly to endogenous LC3, GFP‑LC3 shows a diffuse cytosolic localization in uninduced cell. Following stimulation of autophagy GFP‑LC3 translocates to the autophagosomes and serves as a specific marker of this compartment. The cells were infected with L. monocytogenes and intracellular localization of the bacteria was analyzed using a specific antibody (Fig. 1F). 2 h after infection, 35% of the bacteria were localized to GFP‑LC3 positive autophagosomes, whereas 6 h after infection only 1% of the bacteria were found to colocalize with GFP‑LC3. Interestingly, the autophagosomes surrounding intracellular bacteria were much larger that normal autophagosomes, which were also present in the infected cells (Fig. 1F, panel d). The kinetic suggests that intracellular L. monocytogenes are targeted by the autophagic pathway early on following infection. However, at later time‑points bacteria are largely able to evade the autophagic surveillance. This is consistent with our observation that the capacity for autophagy leads to a 2‑hour delay in intracellular growth, rather than a lower overall growth rate. Together, our data demonstrate a role for autophagy as a cellular defense mechanism against L. monocytogenes infection targeting bacteria for degradation during the early stage of the intracellular infection. Autophagy plays a role in limiting the primary infection by L. monocytogenes. Although our data indicate that autophagy counters 120

Figure 2. Intracellular growth of DactA L. monocytogenes is limited by autophagy. (A) Wild‑type MEFs were infected with DactA L. monocytogenes and the host cell lysate was analyzed by anti‑LC3 Western blot at different time points after infection as indicated. The cell lysates of uninfected cells cultured in the same conditions for 2 h or 8 h are shown as a control (*, non-specific band). The LC3‑II/LC3‑I ratio was quantified. The presented result is representative of 3 independent experiments. (B) Wild‑type MEFs were infected with the DactA L. monocytogenes strain and 8 h later the LC3‑I to LC3‑II conversion was quantified by Western blot. Results are presented as the ratio between LC3‑II/LC3‑I in infected vs uninfected cells and are the means of six independent experiments (T‑test, p < 0.001). (C) Atg5‑/‑ MEFs and their wild‑type counterparts were infected with DactA L. monocytogenes. Bacterial intracellular growth was measured at different time points after infection. The data are representative of three independent experiments.

intracellular growth in the early stages of primary cell infection, it has been previously suggested that autophagy may function in limiting infection of secondary cells following cell‑to‑cell spread.21 To further investigate this point, we used a DactA mutant of L. monocytogenes to assess whether ActA expression and cell‑to‑cell spread are required for L. monocytogenes to induce autophagy. This mutant strain can replicate normally within host cells, but is defective in triggering actin polymerization and as a consequence unable to spread from cell‑to‑cell.25 As soon as 2 h after infection by the DactA mutant wild‑type MEFs showed a significantly higher LC3‑II/LC3‑I ratio as compared to uninfected cells (Fig. 2A and B). Thus, actin‑based motility and spread to secondary cells is not necessary to induce

Autophagy



Figure 3. L. monocytogenes require LLO expression to induce autophagy. (A) Intracellular growth of Dhly L. monocytogenes in Atg5‑/‑ MEFs and their wild‑type counterparts. The data are representative of seven independent experiments. (B) Infection with the Dhly L. monocytogenes strain does not trigger autophagy. Atg5‑/‑ MEFs and their wild‑type counterparts were infected with Dhly L. monocytogenes. The LC3 conversion was analyzed by Western blot at different time points after infection as indicated (*, nonspecific band). The LC3‑II/LC3‑I and the LC3‑II/Tubulin ratios were quantified. The presented result is representative of three independent experiments. (C) Wild‑type MEFs were infected with the Dhly L. monocytogenes strain and 8 h later the LC3‑I to LC3‑II conversion was quantified by Western blot. Results are presented as the ratio between LC3‑II/LC3‑I in infected vs LC3‑II/ LC3‑I in uninfected cells and are the means of six independent experiments (t‑test, p > 0.5).

autophagy. We then compared the growth of the DactA mutant in Atg5‑/‑ MEFs and in their wild‑type counterparts. The DactA mutant strain of L. monocytogenes infected Atg5‑/‑ and wild‑type MEFs with similar efficiency. DactA bacteria grew in both cell types (Fig. 2C), but intracellular growth of the DactA strain was delayed in wild‑type MEFs as compared to Atg5‑/‑ MEFs. The extent of this delay was comparable to that observed for the wild‑type L. monocytogenes strain. We conclude that cell‑to‑cell spread is not necessary for the induction of autophagy and autophagy is able to limit bacterial growth of the DactA strain during the primary infection. L. monocytogenes requires listeriolysin O (LLO) expression to trigger autophagy. We next attempted to further narrow down the www.landesbioscience.com

stage of intracellular infection at which L. monocytogenes is targeted by autophagy. In order to determine whether L. monocytogenes could be targeted by autophagy before the lysis of host endocytic vacuoles and release of bacteria into the cytosol, we compared the growth of a Dhly mutant strain of L. monocytogenes in wild‑type and Atg5‑/‑ MEFs. This strain does not express the pore-forming protein LLO,26 which is essential for the escape of bacteria from host vacuoles formed upon the initial entry. Wild‑type bacteria rapidly escape from the vacuole and multiply in the host cytosol, while Dhly L. monocytogenes mutants remain trapped within the vacuole, and are unable to grow.1 Dhly L. monocytogenes infected both of the MEF cell lines with similar efficiency (Fig. 3A). As expected, Dhly bacteria did not grow in the wild‑type MEFs. In addition, there was no statistical difference in the growth of Dhly bacteria in the Atg5‑/‑ MEFs compared to wild‑type MEFs. It is interesting to note that we repeatedly observed a slight decrease in the number of intracellular bacteria in both wild‑type and Atg5‑/‑ MEFs over the course of this experiment. This suggests that irrespective of their genotype MEF cells might display minor bactericidal properties. We next assessed whether LLO is necessary for autophagy induction during L. monocytogenes infection. MEFs were infected with the Dhly L. monocytogenes strain, and the induction of autophagy was followed over time by anti‑LC3 Western blot. No significant difference in the LC3‑II/LC3‑I ratio was detected in wild‑type MEFs infected with the Dhly mutant as compared to uninfected cells (Fig. 3B, Fig. 3C). Because the level of LC3‑I was low, as an additional control, we measured the LC3‑II/Tubulin ratio to assess accumulation of LC3‑II in the cells. The LC3‑II/Tubulin ratio remained unchanged, and no accumulation of LC3‑II was detected in MEFs infected with the Dhly mutant as compared to uninfected cells. We conclude that expression of LLO is necessary for the induction of autophagy in L. monocytogenes‑infected host cells. Bacterial phospholipases play a role in allowing L. monocytogenes to escape host cell autophagy. In addition to LLO, L. monocytogenes secretes two phospholipases C (PLCs)—PI‑PLC and PC‑PLC, encoded by plcA and plcB, respectively.1 It has been shown that PI‑PLC and PC‑PLC act synergistically with LLO in lysing the endocytic vacuole to allow the invasion of the host cell cytosol.27 Release of DplcA/B bacteria from the host vacuole is delayed, but once released this double mutant strain remains able to grow within host cells. As the perforation of the vacuole by LLO is required for the induction of autophagy, we assessed whether the complete lysis of the vacuole and release of bacteria into the cytosol favored by the bacterial PLCs was also necessary for autophagy induction. Infection of wild‑type MEFs by the DplcA/B mutant increased the LC3‑II/ LC3‑I ratio as compared to uninfected cells with the same kinetics as infection with wild‑type bacteria (Fig. 4A and B). We conclude that bacterial PLCs are not necessary for autophagy induction by L. monocytogenes. It is noteworthy that infection by the DplcA/B strain induced a significantly higher increase in the LC3‑II/LC3‑I ratio (5.2 ± 3.6 fold) than infection by the wild‑type strain (3.6 ± 1.8 fold) (T‑test p = 0.016). As our data suggested that autophagy may counter bacterial growth in the early stages of intracellular infection, we hypothesized that the delayed release of DplcA/B bacteria from the host vacuole may favor the anti‑bacterial effects of autophagy. We compared the growth of DplcA/B L. monocytogenes in wild‑type and Atg5‑/‑ MEFs. DplcA/B L. monocytogenes initially infected wild‑type and Atg5‑/‑ MEFs with similar efficiency, and to the same extent as the wild‑type bacteria. However, the growth of the DplcA/B L. monocytogenes strain in wild‑type MEFs was almost completely inhibited, but restored to levels comparable to those observed with wild‑type

Autophagy

121


bacteria in Atg5‑/‑ MEFs (Fig. 4B). This suggests that in the absence of expression of the bacterial phospholipases, L. monocytogenes is especially vulnerable to the bactericidal effects of autophagy. Autophagy reduces the early appearance of cytoplasmic bacteria colocalizing with polymerized actin. As our results suggested that autophagy acts during an early stage of the L. monocytogenes intracellular cycle, we tested whether autophagy limits bacterial escape from the permeabilized vacuole. Atg5‑/‑ and wild‑type MEFs were infected with DplcA/B or with GFP‑expressing wild‑type L. monocytogenes. Four hours after infection, colocalization of bacteria with polymerized actin was assessed by fluorescence microscopy (Fig. 5A). Polymerized actin was stained with fluorescent phalloidin, and the DplcA/B bacteria were stained with anti‑Listeria antibody. L. monocytogenes trigger actin polymerization around them only once released in the cytosol, thus actin polymerization around the bacteria is a marker of vacuolar escape. Four hours after infection, 54% of the infected Atg5‑/‑ MEFs but only 44% of the infected wild‑type MEFs contained 10403S bacteria colocalizing with polymerized actin. More strikingly, in cells infected with the DplcA/B strain, 4 h after infection 53% of the Atg5‑/‑ MEFs, but only in 15% of the wild‑type MEFs contained bacteria colocalizing with polymerized actin (Fig. 5B). Thus, autophagy appears to limit the ability of L. monocytogenes to escape from the vacuole and/or to enter the actin polymerization phase. Together, our data suggest that the bacteria are targeted for degradation by autophagy early during the intracellular infection cycle, after vacuole permeabilization by LLO but before the recruitment and polymerization of actin.

Discussion Our results demonstrate for the first time that infection of mammalian cells with L. monocytogenes under normal growth conditions can trigger autophagy and that the autophagic process attenuates intracellular bacteria growth. Previous studies reported that metabolically arrested L. monocytogenes could be engulfed into autophagosomes with multi‑layer membranes.6,21 However, no data were available about the outcome of this engulfment on the ability of L. monocytogenes to grow intracellularly. Autophagy has been recently described as a component of the innate immune response against diverse intracellular microorganisms.15 However, this assumption should be verified on a case‑by‑case basis depending on the pathogen. Indeed, some pathogens like Coxiella burnetii are benefited by this cellular response and subvert autophagy resulting in more efficient bacterial replication.18 The capability of macrophages to eliminate L. monocytogenes through the fusion of the primary vacuole with lysosomes is well known. Here, we demonstrate that nonphagocytic cells have the potential to limit intracellular L. monocytogenes growth through autophagy, a pathway also leading to lysosomal degradation but using a different route. It has been reported that L. monocytogenes can become trapped within vesicles with multiple membrane layers during secondary infection of J774 cells in the absence of LLO expression.21 These results suggest a possible role for autophagy in limiting secondary infection, although the dependence of formation of multi‑lamellar structures on the presence of autophagy genes was not demonstrated in that study. Our results show that autophagy limits growth of the DactA bacterial mutant, which is defective in cell‑to‑cell spread. We thus demonstrate that besides the possible role for autophagy in secondary infection, autophagy also limits bacterial growth during primary infection. Moreover, our data show that (1) autophagy is 122

Figure 4. L. monocytogenes deficient for bacterial phospholipases are more susceptible to autophagy. (A) Wild‑type MEFs were infected with DplcA/B L. monocytogenes and the host cell lysate was analyzed by anti‑LC3 Western blot at different time points after infection as indicated. The cell lysates of uninfected cells cultured in the same conditions for 2 h or 8 h are shown as a control (*, non-specific band). The LC3‑II/LC3‑I ratio was quantified. The presented result is representative of 3 independent experiments. (B) Wild‑type MEFs were infected with the DplcA/B L. monocytogenes strain and 8 h later the LC3‑I to LC3‑II conversion was quantified by Western blot. Results are presented as the ratio between LC3‑II/LC3‑I in infected vs LC3‑II/LC3‑I in uninfected cells and are the means of 6 independent experiments (T‑test, p < 0.001). (C) DplcA/B L. monocytogenes intracellular growth in Atg5‑/‑ and wild‑type MEFs. The data are representative of three independent experiments.

induced between 1 h and 2 h after infection, (2) a high proportion of bacteria are engulfed by autophagosomes at 2 h but not at 6 h after infection, (3) the colocalization of the bacteria to the autophagosomes precedes significant induction of actin polymerization (apparent at 4 h but not 2 h), (4) impairment of autophagy accelerates initiation

Autophagy



bacteria can be engulfed by autophagy in the presence of chloramphenicol, this resistance may be dependent on the continued synthesis of one or more bacterial proteins. One possible explanation for the resistance of cytosolic L. monocytogenes to autophagy could be that the actin cloud recruited by bacteria following vacuolar escape and/or actin‑based motility could prevent detection and/or capture by autophagy. However, our results did not show any increase in the growth rate of DactA mutant bacteria at later time points within autophagy‑ deficient MEFs as compared to wild‑type MEFs. In addition infection by the DactA strain does not induce autophagy to a higher extent than infection by wild‑type L. monocytogenes. Thus, if L. monocytogenes has developed a strategy to overcome autophagic degradation, it is unlikely to involve ActA. Whereas autophagy has been implicated in the response to many cellular insults, including infection by multiple pathogens, little is known about the signaling pathFigure 5. Autophagy limits L. monocytogenes escape from the vacuole and actin polymerization. ways responsible for its induction. Here, we (A) Atg5‑/‑ MEFs (i–l) and their wild‑type counterparts (a‑h) were infected with the wild‑type 10403S‑GFP show that induction of autophagy during L. (a–d) or the DplcA/B strains (e–l). Colocalization of the bacteria (a, e and i) with polymerized actin monocytogenes infection requires the expres(b, f and j) was assessed by fluorescence microscopy 4 h after infection. (d, h and l) are enlargements sion of LLO, but PI‑PLC and PC‑PLC are of (c, g and k), respectively. (B) The percentage of infected MEFs containing bacteria colocalizing with dispensable. LLO is a member of a large polymerized actin 4 h after infection was quantified. family of pore-forming cytolysins. These of bacterial growth within the first 4 hours of intracellular infection proteins insert into eukaryotic cell membranes upon interaction rather than limits the global growth rate and (5) autophagy delays with cholesterol. Monomers polymerize and form pores of varying the appearance of bacteria free in the cytosol and able to polymerize size depending on the available concentration of monomers.28 It has actin. Altogether, these results indicate that autophagy plays a role been recently shown that LLO induces small membrane perforations during the early stages of the primary infection, including inter- within vacuoles that persist for several minutes and thus allow protons nalization of bacteria by host cells, perforation and lysis of the host and calcium leakage into the cytosol. These small perforations may vacuole, and release of bacteria into the cytosol.6 As we have not expand to allow the exchange of larger molecules,29 thus forming a observed any differences in the infection rates between wild‑type and transient stage where bacteria still remain enveloped in perforated Atg5‑/‑ cells, it is unlikely that autophagy affects internalization of vacuoles during the first 25 minutes post‑infection. Bacterial phosL. monocytogenes. In addition, the expression of LLO was necessary pholipases do not appear to influence these early perforations.29 As for the induction of autophagy, and the growth of Dhly bacteria was early perforation and autophagy both depend on LLO, but not on not significantly favored in the absence of autophagy. Therefore, PLCs, we hypothesize that the diffusion of vacuolar contents into autophagy does not target intact host vacuoles, but rather bacteria in the cytosol may be responsible for the induction of autophagy. The perforated vacuoles, or recently released and free in the cytosol but nature of the specific molecule(s) involved remains to be elucidated. not yet able to polymerize actin. Alternatively, LLO has been shown Given the low pH and increased calcium concentration within the to play a role in host cell signaling,2 which may be necessary for the bacterial vacuoles, acidification of the cytosol or elevation of cytosolic induction of autophagy. calcium concentration are possible candidates. Indeed, autophagy After 4 h of infection, impairment of autophagy triggers no or has already been described to be dependent on calcium signaling only a modest increase in the growth rate of wild‑type L. monocyto- in other circumstances.30 Alternatively, diffusion of a specific factor genes. Moreover, very few bacteria localized in the autophagosomes synthesized by the bacteria could also mediate the pro‑autophagic 6 h after infection. This suggests that autophagy does not efficiently signaling. Nevertheless, we cannot exclude that autophagy could be target bacteria and limit their growth once they are replicating triggered only following release of the bacteria into the cytosol by in the cytosol. Previous studies suggested that cytosolic L. mono- either secreted factors or proteins exposed at the bacterial surface. cytogenes were engulfed by autophagosomes 6 h after infection in In fact, Shigella flexneri, an intracellular pathogen related to L. J774 cells.6 However, these results were obtained in the presence monocytogenes has been shown to express a surface protein directly of chloramphenicol, a specific inhibitor of bacterial protein synthesis. recognized by the autophagic machinery.19 This protein, VirG, serves We hypothesize that under normal growth conditions, L. monocytogenes a function in actin polymerization similar to that performed by ActA that have successfully invaded the cytosol and have begun replicating in L. monocytogenes and is necessary for the actin‑mediated motility are no longer susceptible to autophagy. However, since cytosolic of S. flexneri. Despite their similar functions, ActA and VirG are not www.landesbioscience.com

Autophagy

123


related, but rather products of convergent evolution.31 Our data also demonstrate that unlike VirG, ActA is not essential for the recognition of L. monocytogenes by the autophagic machinery. Although the DplcA/B L. monocytogenes strain is able to grow efficiently in many cell lines, we show that the growth of this mutant strain is greatly impaired in MEFs. Since DplcA/B L. monocytogenes are able to proliferate at a rate comparable to that of the wild‑type strain in Atg5‑/‑ MEFs, our data indicate that the growth of this L. monocytogenes strain in wild‑type MEFs is impaired as a result of autophagy. Both bacterial PLCs have been implicated in facilitating vacuole lysis by LLO, and impaired expression of either or both bacterial PLCs is associated with a delay in the bacterial release from the vacuole into the cytosol. However, the respective roles of each of the PLCs seems to depend on the cell type as well as on whether the bacteria are involved in primary infection or secondary cell‑to‑cell spread.2,32‑34 Our data suggest that autophagy impairs growth of L. monocytogenes in an early phase of primary infection following vacuole perforation by LLO but preceding actin polymerization and active bacterial growth in the cytosol, suggesting a short period of time during which bacteria are vulnerable to autophagic degradation. The absence of PLCs expression in the DplcA/B strain increases the time between vacuole perforation and release of bacteria into the cytosol, and therefore may increase the period during which bacteria are vulnerable to autophagy allowing almost complete inhibition of DplcA/B bacterial growth. It is noteworthy that the pattern of intracellular growth observed with DplcA/B L. monocytogenes is reminiscent of that of DicsB S. flexneri.19The DicsB S. flexneri strain is fully invasive and capable of escaping from the vacuole, but ultimately loses its ability to multiply within the host cells. Intracellular growth of the DicsB mutant is restored to wild‑type levels in Atg5‑/‑ MEFs.19 IcsB has been shown to block bacterial engulfment by autophagosomes by camouflaging the bacterial surface protein VirG and inhibiting its interaction with Atg5. A similar role for the L. monocytogenes PLCs in inhibiting autophagy could be hypothesized. Supporting this hypothesis, infection by the DplcA/B strain induced autophagy to a higher extent than infection by the wild‑type L. monocytogenes. As secreted proteins, PLCs may directly bind and inhibit the recognition or function of pro‑autophagic factors in a manner similar to IcsB. However, it is also possible that PLCs may inhibit autophagy by either allowing a faster escape from the vacuole or by their effect on host signaling.2 Bacterial PLCs have been shown to trigger host PLC activity and to activate inositol triphosphate /calcium and diacylglycerol /protein kinase C signaling. Further studies will be needed to determine if any of these signaling pathways may play a role in inhibiting L. monocytogenes uptake by autophagosomes during replication in the host cell cytosol.

Conclusions Our findings demonstrate that under normal intracellular growth conditions L. monocytogenes is targeted by autophagy and that this process efficiently limits bacterial growth. Our data suggest that bacteria are vulnerable to autophagic degradation during a short window of time following vacuole perforation by LLO, preceding actin polymerization and active replication in the host cytosol and that bacterial phospholipases are necessary for escape of bacteria from the autophagic surveillance. Our study opens new perspectives in the further characterization of the signal(s) triggering autophagy during L. monocytogenes infection, as well as the putative role for the bacterial PLCs in inhibiting autophagy. The ability of L. monocytogenes 124

to invade the host cell cytosol and then spread intracellularly to adjacent cells was thought to protect bacteria from any host immune responses. Our study suggests that autophagy is one of a formidable battery of intracellular host cell defenses that bacteria must face immediately upon cell entry. References 1. Vazquez‑Boland JA, Kuhn M, Berche P, Chakraborty T, Dominguez‑Bernal G, Goebel W, Gonzalez‑Zorn B, Wehland J, Kreft J. Listeria pathogenesis and molecular virulence determinants. Clin Microbiol Rev 2001; 14:584‑640. 2. Goldfine H, Wadsworth SJ. Macrophage intracellular signaling induced by Listeria monocytogenes. Microbes Infect 2002; 4:1335‑43. 3. Pizarro‑Cerda J, Cossart P. Subversion of cellular functions by Listeria monocytogenes. J Pathol 2006; 208:215‑23. 4. Grundling A, Gonzalez MD, Higgins DE. Requirement of the Listeria monocytogenes broad‑range phospholipase PC‑PLC during infection of human epithelial cells. J Bacteriol 2003; 185:6295‑307. 5. Poussin MA, Goldfine H. Involvement of Listeria monocytogenes phosphatidylinositol‑specific phospholipase C and host protein kinase C in permeabilization of the macrophage phagosome. Infect Immun 2005; 73:4410‑3. 6. Rich KA, Burkett C, Webster P. Cytoplasmic bacteria can be targets for autophagy. Cell Microbiol 2003; 5:455‑68. 7. Levine B, Klionsky DJ. Development by self‑digestion: Molecular mechanisms and biological functions of autophagy. Dev Cell 2004; 6:463‑77. 8. Cuervo AM. Autophagy: In sickness and in health. Trends Cell Biol 2004; 14:70‑7. 9. Lum JJ, Bauer DE, Kong M, Harris MH, Li C, Lindsten T, Thompson CB. Growth factor regulation of autophagy and cell survival in the absence of apoptosis. Cell 2005; 120:237‑48. 10. Levine B. Eating oneself and uninvited guests: Autophagy‑related pathways in cellular defense. Cell 2005; 120:159‑62. 11. Kuma A, Hatano M, Matsui M, Yamamoto A, Nakaya H, Yoshimori T, Ohsumi Y, Tokuhisa T, Mizushima N. The role of autophagy during the early neonatal starvation period. Nature 2004; 432:1032‑6. 12. Komatsu M, Waguri S, Chiba T, Murata S, Iwata J, Tanida I, Ueno T, Koike M, Uchiyama Y, Kominami E, Tanaka K. Loss of autophagy in the central nervous system causes neurodegeneration in mice. Nature 2006; 441:880‑4. 13. Hara T, Nakamura K, Matsui M, Yamamoto A, Nakahara Y, Suzuki‑Migishima R, Yokoyama M, Mishima K, Saito I, Okano H, Mizushima N. Suppression of basal autophagy in neural cells causes neurodegenerative disease in mice. Nature 2006; 441:885‑9. 14. Paludan C, Schmid D, Landthaler M, Vockerodt M, Kube D, Tuschl T, Munz C. Endogenous MHC class II processing of a viral nuclear antigen after autophagy. Science 2005; 307:593‑6. 15. Nakagawa I, Amano A, Mizushima N, Yamamoto A, Yamaguchi H, Kamimoto T, Nara A, Funao J, Nakata M, Tsuda K, Hamada S, Yoshimori T. Autophagy defends cells against invading group A Streptococcus. Science 2004; 306:1037‑40. 16. Birmingham CL, Smith AC, Bakowski MA, Yoshimori T, Brumell JH. Autophagy controls Salmonella infection in response to damage to the Salmonella‑containing vacuole. J Biol Chem 2006; 281:11374‑83. 17. Gutierrez MG, Master SS, Singh SB, Taylor GA, Colombo MI, Deretic V. Autophagy is a defense mechanism inhibiting BCG and Mycobacterium tuberculosis survival in infected macrophages. Cell 2004; 119:753‑66. 18. Gutierrez MG, Vazquez CL, Munafo DB, Zoppino FC, Beron W, Rabinovitch M, Colombo MI. Autophagy induction favours the generation and maturation of the Coxiella‑replicative vacuoles. Cell Microbiol 2005; 7:981‑93. 19. Ogawa M, Yoshimori T, Suzuki T, Sagara H, Mizushima N, Sasakawa C. Escape of intracellular Shigella from autophagy. Science 2005; 307:727‑31. 20. Webster P. Cytoplasmic bacteria and the autophagic pathway. Autophagy 2006; 2:159‑61. 21. Gedde MM, Higgins DE, Tilney LG, Portnoy DA. Role of listeriolysin O in cell‑to‑cell spread of Listeria monocytogenes. Infect Immun 2000; 68:999‑1003. 22. Kabeya Y, Mizushima N, Ueno T, Yamamoto A, Kirisako T, Noda T, Kominami E, Ohsumi Y, Yoshimori T. LC3, a mammalian homologue of yeast Apg8p, is localized in autophagosome membranes after processing. Embo J 2000; 19:5720‑8. 23. Mizushima N, Yamamoto A, Hatano M, Kobayashi Y, Kabeya Y, Suzuki K, Tokuhisa T, Ohsumi Y, Yoshimori T. Dissection of autophagosome formation using Apg5‑deficient mouse embryonic stem cells. J Cell Biol. 2001; 152:657‑68. 24. Colombo MI, Gutierrez MG, Romano PS. The two faces of autophagy: Coxiella and mycobacterium. Autophagy 2006; 2:162‑4. 25. Brundage RA, Smith GA, Camilli A, Theriot JA, Portnoy DA. Expression and phosphorylation of the Listeria monocytogenes ActA protein in mammalian cells. Proc Natl Acad Sci USA 1993; 90:11890‑4. 26. Bayley H. Toxin structure: Part of a hole? Curr Biol 1997; 7:R763‑7. 27. Smith GA, Marquis H, Jones S, Johnston NC, Portnoy DA, Goldfine H. The two distinct phospholipases C of Listeria monocytogenes have overlapping roles in escape from a vacuole and cell‑to‑cell spread. Infect Immun 1995; 63:4231‑7. 28. Jacobs T, Darji A, Frahm N, Rohde M, Wehland J, Chakraborty T, Weiss S. Listeriolysin O: Cholesterol inhibits cytolysis but not binding to cellular membranes. Mol Microbiol 1998; 28:1081‑9.

Autophagy


Autophagy Limits L. Monocytogenes Intracellular Growth 29. Shaughnessy LM, Hoppe AD, Christensen KA, Swanson JA. Membrane perforations inhibit lysosome fusion by altering pH and calcium in Listeria monocytogenes vacuoles. Cell Microbiol 2006; 8:781‑92. 30. Gordon PB, Holen I, Fosse M, Rotnes JS, Seglen PO. Dependence of hepatocytic autophagy on intracellularly sequestered calcium. J Biol Chem 1993; 268:26107‑12. 31. Kocks C, Marchand JB, Gouin E, d’Hauteville H, Sansonetti PJ, Carlier MF, Cossart P. The unrelated surface proteins ActA of Listeria monocytogenes and IcsA of Shigella flexneri are sufficient to confer actin‑based motility on Listeria innocua and Escherichia coli respectively. Mol Microbiol 1995; 18:413‑23. 32. Camilli A, Tilney LG, Portnoy DA. Dual roles of plcA in Listeria monocytogenes pathogenesis. Mol Microbiol 1993; 8:143‑57. 33. Wadsworth SJ, Goldfine H. Listeria monocytogenes phospholipase C‑dependent calcium signaling modulates bacterial entry into J774 macrophage‑like cells. Infect Immun 1999; 67:1770‑8. 34. Bannam T, Goldfine H. Mutagenesis of active‑site histidines of Listeria monocytogenes phosphatidylinositol‑specific phospholipase C: Effects on enzyme activity and biological function. Infect Immun 1999; 67:182‑6. 35. Bishop DK, Hinrichs DJ. Adoptive transfer of immunity to Listeria monocytogenes: The influence of in vitro stimulation on lymphocyte subset requirements. J Immunol 1987; 139:2005‑9. 36. Jones S, Portnoy DA. Characterization of Listeria monocytogenes pathogenesis in a strain expressing perfringolysin O in place of listeriolysin O. Infect Immun 1994; 62:5608‑13. 37. Skoble J, Portnoy DA, Welch MD. Three regions within ActA promote Arp2/3 complex‑mediated actin nucleation and Listeria monocytogenes motility. J Cell Biol 2000; 150:527‑38. 38. Agaisse H, Burrack LS, Philips JA, Rubin EJ, Perrimon N, Higgins DE. Genome‑wide RNAi screen for host factors required for intracellular bacterial infection. Science 2005; 309:1248‑51.


Autophagy

125