Salmonella Macrophages with Jun N-Terminal Kinase During Infection

Distinct Mechanisms Target Stress and Extracellular Signal-Activated Kinase 1 and Jun N-Terminal Kinase During Infection of Macrophages with Salmonella This information is current as of October 1, 2021. Katarzyna J. Procyk, Maria Rita Rippo, Roberto Testi, Fred Hoffmann, Peter J. Parker and Manuela Baccarini J Immunol 1999; 163:4924-4930; ; http://www.jimmunol.org/content/163/9/4924 Downloaded from

References This article cites 58 articles, 39 of which you can access for free at: http://www.jimmunol.org/content/163/9/4924.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on October 1, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 1999 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Distinct Mechanisms Target Stress and Extracellular Signal-Activated Kinase 1 and Jun N-Terminal Kinase During Infection of Macrophages with Salmonella1

Katarzyna J. Procyk,2* Maria Rita Rippo,† Roberto Testi,† Fred Hoffmann,‡ Peter J. Parker,§ and Manuela Baccarini3*

The interaction between bacteria and macrophages is central to the outcome of Salmonella infections. Salmonella can escape killing by these phagocytes and survive and multiply within them, giving rise to chronic infections. Cytokines produced by infected macrophages are involved in the early gastrointestinal pathology of the infection as well as in the induction and maintenance of the immune response against the invaders. Jun N-terminal kinases (JNK) are activated by inflammatory stimuli and play a role in cytokine production. We have investigated the signaling routes leading to JNK activation in Salmonella-infected macrophages and have discovered that they differ radically from the mechanisms operating in epithelial cells. In particular, activation of the Downloaded from JNK kinase stress and extracellular-activated kinase 1 (SEK1) and of JNK in macrophages occurs independently of actin rearrangements and of the GTPases Cdc42 and Rac, essential mediators in other cells. Activation of JNK is effected by a novel pathway comprising tyrosine kinase(s), phosphoinositide 3-kinase and, likely, atypical protein kinase C ␨. SEK1 is stimulated by a distinct mechanism involving phosphatidylcholine-phospholipase C and acidic sphingomyelinase. Dominant-negative SEK1 can block JNK activation by LPS, but not by Salmonella. These data demonstrate that SEK1 and JNK are activated independently in Salmonella- infected macrophages and offer experimental support for the concept that incoming signals can direct the selective coupling of http://www.jimmunol.org/ downstream pathways to elicit highly specific responses. Inhibitors of stress kinase pathways are receiving increasing attention as potential anti-inflammatory drugs. The precise reconstruction of stimulus-specific pathways will be instrumental in predicting/ evaluating the effects of the inhibitors on a given pathological condition. The Journal of Immunology, 1999, 163: 4924–4930.

almonellae are facultative intracellular pathogens that cause pathways (11). The bacterial product SopE, for instance, activates a variety of illnesses from localized gastroenteritis to more the small GTPases Cdc42 and Rac-1 (7) and by so doing induces S overt, host-specific ones, such as typhoid fever. In the host’s the cytoskeletal changes required for membrane ruffling and for the intestine, Salmonella adheres to specialized epithelial cells (M macropinocytosis of Salmonella (12). A functional type III secre- by guest on October 1, 2021 cells) and cause cytoskeletal and membrane rearrangements that tion system is also a prerequisite for the activation of the mitogen- result in its uptake (1). By destroying the infected M cells (2), the activated protein kinase (MAPK)4 subgroups extracellular-regu- bacteria gain access to the mesenteric lymph follicles, where they lated kinase (ERK), Jun N-terminal kinase (JNK), and p38 and for encounter and infect macrophages. Recently, the induction of mac- the production of proinflammatory cytokines by epithelial cells rophage apoptosis by invasive Salmonella has been documented in infected with Salmonella (13). vitro (3–5, 60) and in vivo (6). Both epithelial cells invasion (7) For Salmonella as for many other facultative intracellular patho- and the induction of macrophage apoptosis (8) depend on bacterial gens the key to a successful infection lies in the outcome of their virulence determinants translocated into the eukaryotic cell by a encounter with the host’s macrophages. Besides playing a crucial specialized, host-dependent secretion system (9, 10). role in the immune response against the bacteria, the pro-inflam- The analysis of the biochemical cross-talk between Salmonella matory cytokines produced by these cells mediate the early gas- and epithelial cells has shown that the proteins secreted by the type trointestinal pathology of the infection (14). Still, much less is III secretion system have the capacity to trigger host cell signaling known about the molecular mechanisms operating during the interaction of Salmonella typhimurium with macrophages than about

*Department of Cell and Microbiology, Institute of Microbiology and Genetics, Vi- the signaling events taking place during epithelial cell invasion. enna Biocenter, Vienna, Austria; †Department of Experimental Medicine and Bio- We have previously addressed the question of Salmonella-medi- chemical Sciences, University of Rome “Tor Vergata,” Rome, Italy; ‡Institut fu¨r ated ERK activation, showing that LPS is the major determinant Pharmakologie und Toxikologie, Freiburg, Germany; and §Protein Phosphorylation Lab, Imperial Cancer Research Fund, London United Kingdom responsible for ERK stimulation by this pathogen (15). In the Received for publication June 10, 1999. Accepted for publication August 13, 1999. present study, we focus on reconstructing the mechanism of JNK activation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance JNK can be stimulated by cellular stress signals like irradiation, with 18 U.S.C. Section 1734 solely to indicate this fact. heat shock, osmotic stress, and protein synthesis inhibitors (16), 1 This work was supported by Grants P10766-MED and P13252-MOB of the Austrian Research Fund (to M.B.) and by a grant from the Associazione Italiana per la Ricerca sul Cancro (to R.T.). 4 Abbreviations used in this paper: MAPK, mitogen-activated protein kinase; AS- Mase, acidic sphingomyelinase; BIM, bisindoleylmaleimide; DAG, diacylglycerol; 2 Current address: Protein Phosphorylation Lab, Imperial Cancer Research Fund, ERK, extracellular-regulated kinases; JNK, Jun N-terminal kinases; PC-PLC, phos- London, United Kingdom. phatidylcholine phospholipase C; PDK1, phosphoinositide-dependent kinase 1; PI, 3 Address correspondence and reprint requests to Dr. Manuela Baccarini, Institute of phosphoinositides; PI 3-K, phosphoinositide 3-kinase; PKC, protein kinase C; SEK1, Microbiology and Genetics, Dr. Bohrgasse 9, A-1030, Vienna, Austria. E-mail ad- stress and extracellular signal-activated kinase 1; moi, multiplicity of infection; PLD, dress: [email protected] phospholipase D; wt, wild type; KR, kinase-defective; HA, hemagglutinin.

and also by growth factors (17, 18). Relevant to our study, this plemented with 0.5% sodium deoxycholate. Serial dilutions of the lysates pathway is activated by inflammatory stimuli (19–21) and upon were prepared in PBS and plated onto the Luria-Bertani agar plates. Col- infection of cultured cells by pathogens, including Gram-negative onies were allowed to develop for 18 h before counting. Assays were conducted in triplicates. (13, 22) and Gram-positive bacteria (23). Targeted distruption of the JNK kinase stress and extracellular signal-activated kinase 1 Cell lysis, immunoprecipitation and western blotting (SEK1) causes defects in the activity of AP-1 (24), a transcription Cells were lysed in solubilization buffer (10 mM Tris-base, 50 mM sodium factor implicated in the regulation of cytokine genes (25). Recent chloride, 30 mM sodium pyrophosphate, 50 mM sodium fluoride, and 1% data also implicate JNK in the stabilization (26, 27) and translation Triton X-100, pH 7.0) supplemented with 1 mM PMSF, 100 ␮M sodium (28, 29) of cytokine mRNAs. Consistently, T cell differentiation is vanadate, 1 mM DTT, and protease inhibitors (aprotinin (3 ␮g/ml), pep- ␮ ␮ defective in Jnk-1- (30) and Jnk-2-deficient mice (31). statin and leupeptin (each at 0.5 g/ml)). For immunoblotting, 30–40 g of whole cell extracts were separated by 10% SDS-PAGE and transferred In this study we show that SEK1 and JNK are activated in a onto nitrocellulose membranes. For immunoprecipitation, 500–600 ␮gof phagocytosis-independent manner upon infection of macrophages whole cell extracts were incubated in the presence of protein A beads by Salmonella. Stimulation of JNK, but not of SEK1, involves the (Amersham, Arlington, Heights, IL) with anti-PKC␨ (Santa Cruz Biotech- activation of tyrosine kinases, phosphoinositide 3-kinase (PI 3-K), nology, Santa Cruz, CA) or anti-HA Abs for 16–18 h at ϩ4°C. The beads and, likely, protein kinase C ␨ (PKC␨); on the other hand, SEK1 were collected and washed three times with lysis buffer before elution of the immunocomplexes by boiling in SDS sample buffer. Membranes were activation depends on the stimulation of phosphatidylcholine phos- blocked for 8–16 h at 4°C in TTBS (10 mM Tris-HCl (pH 8.0), 150 mM pholipase C (PC-PLC) and acidic sphingomyelinase (ASMase). NaCl, and 0.1% Tween 20) supplemented with 4% BSA (fraction V; Sig- Cdc42 and Rac, which mediate JNK activation by a variety of ma), and probed with the appropriate primary Abs in 1% BSA in TTBS stimuli in many cell types, are not required for SEK1 or JNK before incubation with peroxidase-conjugated secondary Abs and detection

by the enhanced chemiluminescence (ECL) system (Pierce, Rockford, IL). Downloaded from stimulation in macrophages. Thus, Salmonella activates the JNK The primary Abs used in this study recognize selectively phosphorylated pathway in macrophages by a mechanism that bypasses the ca- JNK1/2 (anti-phJNK, Thr183/Tyr185), SEK1 (anti-phSEK, Thr223; all from nonical activators Cdc42/Rac and SEK1. New England BioLabs, Schwalbach, Germany) and PKC␨ (36) or their unmodiﬁed forms (Santa Cruz Biotechnology). Materials and Methods PC-PLC and ASMase activity assays Bacteria

PC-PLC and ASMase activity of whole cell extracts was determined as http://www.jimmunol.org/ S. typhimurium strain LT2 (virulent, wild type (wt)) and SB111 (32) were previously described (37). Briefly, cells (2.5 ϫ 106) were scraped in 2 ml grown in Luria-Bertani (LB) broth (1% bactotryptone, 0.5% yeast extract, ice-cold PBS and centrifuged for 10 min at 400 rpm at 4°C. A total of 300 and 1% sodium chloride) at 37°C overnight under agitation (poorly inva- ␮l of Triton X 100 (0.01% for PC-PLC, 0.2% for ASMase activity mea- sive). To obtain highly invasive bacteria, bacteria from overnight cultures surements) were added to the pellet, and the samples were incubated on ice were diluted to an OD600 of 0.02 in 50 ml fresh LB and incubated for 5 h for 10 min before sonication. A total of 15 ␮g of lysate were incubated for under agitation (60). 2 h at 37°C either in PC-PLC buffer (50 mM Tris-HCl (pH 7.3), 6.3 mM Cell culture, stimulation, and pretreatment CaCl2, 150 mM ammonium sulfate, plus 50 nCi L-3-phosphatidyl[N-methyl-14C]choline ([14C]PC); 80 ␮l total volume) or in ASMase buffer (250 BAC-1.2F5 cells (33) were cultured in DMEM supplemented with 10% mM sodium acetate (pH 5.0), 0.2% Triton X 100, plus 50 nCi [methyl- FCS and 20% L cell conditioned medium as a source of CSF-1. Confluent 14C]sphingomyelin; 50 ␮l total volume). Labeled lipids were from Amer- cells (about 5 ϫ 106 cells/100-mm-diameter tissue culture dish) were cul- sham. The PC-PLC assay was terminated by extracting the lipids with by guest on October 1, 2021 ␮ ␮ ␮ tured for 16 h in medium without CSF-1, and then stimulated with 1.5 CHCl3:CH3OH (1:2 v/v, 180 l), 0.9% NaCl (60 l), and CHCl3 (60 l). ␮ g/ml bacterial LPS (from S. typhimurium; Sigma, St. Louis, MO) or in- The ASMase assay was terminated by extracting the lipids with CHCl3: ␮ ␮ fected with bacterial cultures as previously described (15). A multiplicity CH3OH (1:1 v/v, 400 l) and water (180 l). The aqueous and organic of infection (moi; bacteria per macrophage) of 25 was used. phases were separated and quantitated by liquid scintillation. The amount Actin polymerization was blocked by pretreatment with 10 ␮M cy- of substrate hydrolyzed was quantitated by liquid scintillation counting. tochalasin B (30 min; Sigma). Tyrosine kinases were inhibited by pretreat- PC-PLC and ASMase activity were expressed as percentage of control. ment with herbimycin A (4 ␮g/ml for 4 h; Sigma). Activation of PI 3-K was blocked by pretreatment with wortmannin (100 nM, 20 min; Sigma). Results Rho-family small GTPases (RhoA, Rac1, and Cdc42) were inhibited by a 60-min preincubation with toxin B (34) at a final concentration of 10 or 100 Salmonella induces SEK and JNK activation independently of ng/ml. Inhibition of PC-PLC activity was performed by preincubating the phagocytosis and of the function of Rho family GTPases ␮ cells for 60 min with 10 M xanthogenate tricyclodecan-9-yl (D609, Quiescent BAC-1.2F5 cells were infected with Salmonella for Alexis Biochemicals, Laufelfingen, Switzerland). PKC was inhibited by treating the cells with 10 ␮M bisindoleylmaleimide (BIM; Calbiochem, La different time periods. The activation state of the kinases was Jolla, CA) for 60 min before stimulation (35). Diacylglycerol (DAG)-de- assessed in whole cell lysates by immunoblotting with Abs that pendent PKC isoforms were down-regulated by a 24-h treatment with 5 specifically recognize the phosphorylated, activated form of each ␮ M PMA (Sigma). enzyme (Fig. 1A). All kinases were activated by Salmonella in- Transient transfection fection of BAC-1.2F5 cells with fast activation/inactivation kinetics. Peak activation occurred after 25 min and then decayed. In- The SuperFect reagent (Qiagen, Basel, Switzerland) was used according to the manufacturer’s instructions. Macrophages were transfected with an activation was complete by 1 h, and no further changes were epitope-tagged JNK1 (HA-JNK1, 2 ␮g/100-mm-diameter culture dish) to- observed over a period of 4 h (data not shown). These kinetics of gether either with an epitope-tagged (GST-) SEK wt, or with GST-SEK KR activation resembled those of the other MAPK subfamily, ERK (kinase defective mutant, in which the nucleophilic lysine was substituted (15). Highly invasive Salmonella, which causes apoptosis in mac- for by an arginine), or with pEBG vector as a control (each 8 ␮g/100-mm- diameter culture dish). Twenty-four hours after transfection, the cells were rophages, activates JNK less efficiently than the poorly invasive infected with Salmonella as described above. form (Fig. 1A) and than an invasion-defective mutant (Fig. 1B). The time frame investigated (5–45 min) allowed phagocytosis Phagocytosis assay of Salmonella by the macrophages. We therefore examined Colony counting assays were performed to assess phagocytosis of S. ty- whether phagocytosis was involved in JNK activation. Phagocy- 6 phimurium. Briefly, cells (0.05 ϫ 10 ) were seeded in 96-well plates and tosis of latex beads of the approximate size of the bacteria did not infected (moi of 25). Cells were allowed to phagocytose for 30 min and then washed three times with PBS. Fresh medium containing 50 ␮g/ml stimulate any of the kinases at study (data not shown). However, gentamicin was added and the cells were incubated for further 60 min to the mechanisms involved in the phagocytosis of bacteria and inert kill residual extracellular bacteria. Thereafter, cells were lysed in PBS sup- particles likely differ. We therefore analyzed the effect of inhibitors 4926 STRESS KINASE ACTIVATION IN SALMONELLA-INFECTED MACROPHAGES

FIGURE 1. Activation of SEK1 and JNK is an early event in Salmo- nella-induced signal transduction. A, Quiescent BAC-1.2F5 cells were infected with wt Salmonella (S.t., moi of 25) at 37°C for different times before solubilization. The presence of the phosphorylated, active forms of SEK1 (phSEK1) and JNK (phJNK) was detected by immunoblotting with Downloaded from the corresponding Abs. An anti-JNK1 immunoblot is shown as a loading control. In B, cells were infected with highly invasive Salmonella (wt) and with an invAϪ mutant. of specific signal transduction pathways on Salmonella phagocytosis and JNK activation (Fig. 2A). As a positive control we used http://www.jimmunol.org/ cytochalasin B, which destroys the actin cytoskeleton and totally abrogated phagocytosis. The PI 3-K inhibitor wortmannin, which blocks epithelial cell invasion by Listeria monocytogenes but not by Salmonella (24), fully blocked phagocytosis, as did toxin B (1), which glucosylates and inactivates the small GTPases Cdc42, Rac and Rho (Ref. 34; and data not shown). Cytochalasin B (Fig. 2B) did not suppress Salmonella-mediated activation of SEK1 or JNK. Wortmannin, on the other hand, decreased Salmonella-stimulated activation of JNK, but not SEK1 (Fig. 2C). Identical results were FIGURE 2. Salmonella-induced kinase activation is independent of by guest on October 1, 2021 obtained using a second PI 3-K inhibitor, LY294002 (data not phagocytosis. Cells were treated with cytochalasin B (cyto B, 10 ␮M for 30 shown). Thus, phagocytosis and kinase activation by Salmonella min; A and B), with wortmannin (WM, 100 nM for 20 min; A and C), or are independent events. Furthermore, the involvement of PI 3-K in with toxin B (tox B, 100 ng/ml for 60 min; A and D) before infection with Salmonella (S.t., moi of 25) for 15 min. A, The phagocytosis assay was JNK, but not SEK1 activation, suggests that these kinases are reg- conducted as described in Materials and Methods. The plot represents the ulated independently during infection of macrophages by mean of three independent experiments, and vertical bars indicate the SEs Salmonella. of the mean. B–D, The presence of phosphorylated forms of SEK1 and Neither SEK1 nor JNK activation were blocked by toxin B pre- JNK, as well as of JNK1 as a loading control, was detected by immuno- treatment (Fig. 2D). Thus, a novel pathway independent of func- blotting with the corresponding Abs. tional Cdc42/Rac operates in macrophages infected with Salmo- nella. This was all the more unexpected because PI 3-K-mediated JNK activation has been shown to be dependent on Rac (18, 38). can be inhibited by BIM (41). Activation of JNK was completely suppressed after pretreatment with BIM (Fig. 4A). In agreement Herbimycin A decreases Salmonella-mediated JNK, but not SEK with the hypothesis that distinct pathways cause SEK1 and JNK activation stimulation by Salmonella, SEK activation was insensitive to BIM treatment. Sustained treatment (up to 24 h) with 5 ␮M PMA, Herbimycin-dependent kinases have previously been implicated in the activation of PI 3-K (39) and JNK (19, 21) by LPS in monocytes. Salmonella-mediated JNK activation was also efficiently reduced by herbimycin A (Fig. 3). However, SEK1 activation was herbimycin-insensitive, confirming that the pathways targeting SEK1 and JNK differ in Salmonella-infected macrophages.

Atypical PKC is involved in Salmonella-mediated activation of JNK, but not of SEK The results described above suggest that a pathway comprising FIGURE 3. Herbimycin A treatment decreases Salmonella-induced ac- herbimycin-sensitive tyrosine kinases and PI 3-K, but independent tivation of JNK, but not SEK1. Quiescent BAC-1.2F5 cells were treated of Cdc42/Rac, targets JNK during Salmonella infection. PKC iso- with herbimycin A (1 ␮g/ml for 4 h) before infection with wt Salmonella forms can act as PI 3-K downstream effectors both in vitro and in (S.t., moi of 25 for 15 min). The presence of phosphorylated forms of vivo (36, 40). BAC-1.2F5 cells express the novel DAG-dependent SEK1 and JNK, as well as of JNK1 as a loading control, was detected by PKC isoforms ␦ and ⑀, and the atypical PKC␨ (35), all of which immunoblotting with the corresponding Abs. The Journal of Immunology 4927

ysis of whole cell extracts showed that the phosphorylation reached a maximum 10 min after infection and then slowly decayed (Fig. 4D). We conﬁrmed the identity of the PKC isoform activated by analyzing the phosphorylation status of immunoprecipitated PKC␨ (Fig. 4E). Taken together, the data identify PKC␨ as a target of Salmonella downstream of PI 3-K and PDK1. Be- cause the only other known kinase inhibited by BIM (albeit at higher concentrations) is protein kinase A (PKA), and this enzyme has never been connected with JNK activation, the data implicate PKC␨ in relaying the Salmonella signal to the JNK module.

Involvement of PC-PLC and ASMase in signaling to SEK1 High concentrations of the PC-PLC inhibitor D609 decrease LPS- mediated stimulation of Raf, MEK, and ERK (35). To investigate whether phospholipase activation was important for the stimulation of SEK1 or JNK by Salmonella, we treated quiescent BAC- 1.2F5 cells with low concentrations of D609 (10 ␮M) before infection. The inhibitor severely blunted SEK1 activation, but had only a minor impact on JNK stimulation (Fig. 5A). The concen- Downloaded from tration of D609 used is reportedly specific for PC-PLC, and does not affect phospholipase D (PLD) (43). Furthermore, Salmonella- induced SEK1 activation was not affected by the presence of 1% 1-butanol (competitive inhibitor of PLD; Ref. 44 and data not shown). This finding indicated that PC-PLC, and not PLD, mediated SEK1 activation during infection. To strengthen this conclusion, we measured PC-PLC activity in http://www.jimmunol.org/ extracts of untreated and infected cells. PC-PLC was strongly activated by Salmonella with kinetics consistent with the ones observed for SEK1/JNK activation (Fig. 5B). The activity of AS- Mase, the PC-PLC downstream target responsible for ceramide generation (45), increased steadily during the first 30 min of infection. Like SEK1 stimulation, ASMase activation was abrogated by pretreatment with the PC-PLC inhibitor D609 (Fig. 5C). Taken

together, the data strongly suggest that infection of macrophages by guest on October 1, 2021 generates ceramide via a PC-PLC/ASMase pathway and that in this specific setting the lipid second messenger is important for SEK1, but not for JNK activation. FIGURE 4. Involvement of PKC␨ in Salmonella-stimulated SEK1 and JNK activation. Quiescent BAC-1.2F5 cells were treated with the PKC Dominant negative SEK1 does not prevent Salmonella-mediated inhibitor BIM (10 ␮M for 60 min) before infection with wt Salmonella activation of the JNK (S.t., moi of 25 for 15 min; A). Alternatively, DAG-dependent PKC iso- The experiments reported above argue that Salmonella-mediated forms were down-regulated by prolonged PMA treatment (5 ␮M; B and C). activation of SEK1 and of JNK are regulated independently. To The presence of phosphorylated SEK1/JNK, of unmodified JNK1 (as a loading control), and of PKC␦ and PKC⑀ was detected by immunoblotting address this question directly, we cotransfected cells with either wt with the corresponding Abs. The presence of phosphorylated PKC␨ was or KR GST-SEK1 and HA-tagged JNK1. Twenty-four hours after assessed by immunoblot analysis of either whole cell extracts (D)orof transfection cells were either left untreated, treated with LPS, or anti-PKC␨ immunoprecipitates (E). Anti-PKC␨ blots of whole cell extracts infected with Salmonella. HA-JNK1 was immunoprecipitated and (D) or immunoprecipitates (E) are shown as a loading control. its activation state was assessed by Western blotting with anti-phJNK Abs. HA-JNK1 was activated by LPS or Salmonella in cells cotrans- which causes down-regulation of DAG-dependent PKC isoforms, fected with wt GST-SEK1. Expression of GST-SEK1 KR signif- did not affect JNK activation (Fig. 4B). This finding indicates that icantly reduced JNK stimulation by LPS but had no effect on Sal- DAG-dependent isoforms ␦ and ⑀, which were efficiently degraded monella-induced JNK activation (Fig. 6, top panel). Equal under these conditions (Fig. 4C), are not involved in Salmonella amounts of HA-JNK1 were present in the immunoprecipitates stimulation of JNK. Therefore, the effect of BIM on JNK activa- (Fig. 6, middle panel). Thus, Salmonella stimulated JNK by a tion must be due to the inhibition of an atypical PKC. SEK1-independent mechanism. Atypical PKCs would be natural targets of the phosphoinositides (PI) generated by PI 3-K. PKC␭ expression cannot be detected in BAC-1.2F5 macrophages. We monitored the stimulation of the Discussion only other known atypical PKC, PKC␨,byS. typhimurium. PKC␨ The interaction of Salmonella with the host’s macrophages is a key can be activated by PI(3,4,5)P3, PI(4,5)P2, and PI(3,4)P2 directly event in the early phases of infection. Still, the signaling steps (42) as well as indirectly via phosphorylation by PI-dependent ki- taking place during this interaction are largely unknown. nase 1 (36, 40). Activation was monitored by immunoblotting with In this paper we describe the mechanisms that lead to the acti- Abs that specifically recognize phosphorylation of the PDK1 site vation of stress-induced kinases after infection of macrophages Thr410 in the C-loop of the PKC␨ isoform (36). Western blot anal- with Salmonella. We find that a pathway involving the sequential 4928 STRESS KINASE ACTIVATION IN SALMONELLA-INFECTED MACROPHAGES

FIGURE 6. Dominant-negative SEK1 blocks LPS-induced, but not Sal- monella-induced, JNK activation. BAC-1.2F5 cells were transfected with a plasmid encoding HA-tagged JNK1 together with GST-tagged wt or KR SEK1. HA-tagged JNK1 was immunoprecipitated from untreated cell and from cells stimulated with either LPS (1 ␮g/ml for 20 min) or Salmonella (20 min). The presence of phosphorylated (top panel) and unmodiﬁed HA- JNK1 (loading control, middle panel) as well the expression of GST-SEK

wt and KR (bottom panel) was detected by immunoblotting with the cor- Downloaded from responding Abs.

JNK by Salmonella. At the same time, the ability of the macrophages to phagocytose and, on longer incubations, their adherence to the substrate (data not shown) were blocked efﬁciently, demon-

strating that the inhibitor functions to prevent Rho family-directed http://www.jimmunol.org/ cytoskeletal rearrangements in macrophages. Our results were all the more surprising in view of the fact that both cytoskeletal re- organization and JNK activation during invasion of epithelial cells with Salmonella depend on the function of Cdc42 and Rac-1 (7, FIGURE 5. Involvement of PC PLC and ASMase in Salmonella-in- 12). The most likely explanation for this discrepancy is that JNK duced kinase activity. Quiescent BAC-1.2F5 cells were pretreated with the PC-PLC inhibitor D609 (10 ␮M for 60 min) before infection wt Salmo- activation is implemented by the bacteria in a cell-type-specific nella (S.t., moi of 25; 15 min). To assess the contribution of PLD to kinase manner. In favor of this hypothesis, noninvasive Salmonella mu- activation, infection was conducted in the presence of PLD inhibitor 1-bu- tants are incapable of initiating JNK activation in epithelial cells by guest on October 1, 2021 tanol (1%). The presence of phosphorylated SEK1 and JNK was detected (13), whereas they do so in macrophages. Furthermore, highly in- by immunoblotting with the corresponding Abs (A). Neither of the inhib- vasive Salmonella, which induces apoptosis in macrophages, ac- itors had any effect on the basal level of kinase activation. An anti-JNK1 tivates JNK less efficiently than poorly invasive bacteria (Fig. 1B). blot is shown as loading control. PC-PLC (B) and ASMase (C) activity This phenomenon is reminiscent of the results obtained with Yer- were determined in whole cell extracts as described in Materials and Meth- sinia spp., in which invasive, virulent bacteria caused both mac- ␮ ods.InC, cells were pretreated with D609 (10 M for 60 min) to block rophage apoptosis and the suppression of MAPK activity (46, 47). PC-PLC. The results are expressed as percentage increase with respect to It is possible that Salmonella, like Yersinia (48), produces an in- the control values. The plot in B represents the mean of three independent experiments, and vertical bars represent the SEs of the mean. The 100% hibitor of eukaryotic MAPK and translocates it into the host cell value represents 455.33 cpm (Ϯ18.98) and corresponds to 0.47% (Ϯ0.023) via the type III secretion system. substrate hydrolized. The plot in C represents the mean of two independent Conventional, novel, and atypical isoforms of PKC have been experiments, and vertical bars represent the range of the samples. The previously implicated in MAPK stimulation (49–52), and are 100% value represents 45,023.5 cpm (range 39,390–50,657) and corre- downstream targets of PI 3-K (36, 40). During infection of mac- sponds to 44.25% (range 32.6–55.9) substrate hydrolized. rophages with Salmonella, DAG-dependent PKC isoforms are in fact activated as a result of PI 3-K-mediated stimulation of PLD and act as intermediates in ERK activation (15). Salmonella also activation of PI 3-K/PDK1 and PKC␨ is responsible for JNK ac- activates the atypical PKC␨ (Fig. 4, D and E) via phosphorylation tivation. PC-PLC and ASMase, also activated upon infection, sup- of Thr410 in its activation loop by PDK1, a phosphoinositide-de- port SEK1 but not JNK activation. Unexpectedly, stimulation of pendent kinase acting downstream of PI-3K (36, 40). Activation of both SEK1 and JNK is independent of the function of Cdc42 PKC␨ has mostly been connected with the stimulation of the ERK and Rac. pathway (49, 50, 52, 53), although this enzyme has been shown to modulate the JNK pathway target AP-1 (54, 55). Our data repre- PI 3-K and atypical PKC are novel intermediates in Rho family- sent the first demonstration that PKC␨ is activated downstream of independent JNK activation by Salmonella PI 3-K/PDK1 during infection of macrophages by Salmonella. The Herbimycin A and PI 3-K inhibitors severely blunted Salmonella- participation of this enzyme in the activation of JNK is supported mediated activation of JNK. PI 3-K has been reported previously by the inhibitory effect of BIM on JNK activation. to mediate JNK activation by tyrosine kinase and G-protein coupled receptors (17, 18, 38). However, in all cases in which this has PC-PLC and ASMase participate in SEK1 activation by been investigated, dominant-negative forms of Cdc42/Rac blocked Salmonella PI 3-K-dependent JNK activation. In contrast, toxin B, a bacterial SEK1 activation by Salmonella was resistant to all inhibitors that inhibitor of these GTPases, does not prevent the stimulation of efficiently prevented JNK stimulation by the bacterium. Besides The Journal of Immunology 4929 supporting the specificity of these substances, this indicated that cellularly inside macrophages and exerts a cytotoxic effect on phagocytes in vivo. distinct mechanisms target SEK1 and JNK during infection. PC- J. Exp. Med. 186:569. 7. Hardt, W. D., L. M. Chen, K. E. Schuebel, X. R. Bustelo, and J. E. Galan. 1998. PLC and ASMase were activated by Salmonella, and treatment S. typhimurium encodes an activator of Rho GTPases that induces membrane with the PC-PLC inhibitor D609 abolished both ASMase and ruffling and nuclear responses in host cells. Cell 93:815. 8. Hersh, D., D. M. Monack, M. R. Smith, N. Ghori, S. Falkow, and A. Zychlinsky. SEK1 activation, leaving JNK stimulation undisturbed. We pro- 1999. The Salmonella invasin SipB induces macrophage apoptosis by binding to pose that, as shown in Neisseria gonorrhoeae-infected epithelial caspase-1. Proc. Natl. Acad. Sci. USA 96:2396. cells (56) and in LPS-stimulated macrophages (57), the activation 9. Collazo, C. M., and J. E. Galan. 1997. The invasion-associated type-III protein secretion system in Salmonella: a review. Gene 192:51. of the PC-PLC/ASMase pathway by Salmonella leads to the gen- 10. Kubori, T., Y. Matsushima, D. Nakamura, J. Uralil, M. Lara-Tejero, A. Sukhan, eration of ceramide and thereby to SEK1 stimulation. The actual J. E. Galan, and S. I. Aizawa. 1998. Supramolecular structure of the Salmonella link between ceramide generation and SEK1 activation remains typhimurium type III protein secretion system. Science 280:602. 11. Galan, J. E. 1999. Interaction of Salmonella with host cells through the centisome unidentified. The TAK-1 kinase, a SEK1 activator stimulated by 63 type III secretion system. Curr. Opin. Microbiol. 2:46. endogenous and exogenous ceramides (58), would be a suitable 12. Chen, L. M., S. Hobbie, and J. E. Galan. 1996. Requirement of Cdc42 for Sal- candidate. monella-induced cytoskeletal and nuclear responses. Science 274:2115. 13. Hobbie, S., L. M. Chen, R. J. Davis, and J. E. Galan. 1997. Involvement of mitogen-activated protein kinase pathways in the nuclear responses and cytokine SEK1-independent JNK activation by Salmonella in production induced by Salmonella typhimurium in cultured intestinal epithelial macrophages cells. J. Immunol. 159:5550. 14. Arnold, J. W., D. W. Niesel, C. R. Annable, C. B. Hess, M. Asuncion, Y. J. Cho, The data discussed above demonstrate that distinct mechanisms J. W. Peterson, and G. R. Klimpel. 1993. Tumor necrosis factor-␣ mediates the implement SEK1 and JNK activation in Salmonella-infected mac- early pathology in Salmonella infection of the gastrointestinal tract. Microb. Pathog. 14:217. rophages. Cross-talk between these two signal transduction path- 15. Procyk, K. J., P. Kovarik, A. von Gabain, and M. Baccarini. 1999. Salmonella Downloaded from ways was not observed, and Salmonella-induced JNK activation typhimurium and lipopolysaccharide stimulate extracellularly regulated kinase was not blocked by a dominant-negative SEK1 mutant. In contrast, activation in macrophages by a mechanism involving phosphatidylinositol 3-kinase and phospholipase D as novel intermediates. Infect. Immun. 67:1011. the SEK1 dominant-negative mutant inhibited the bulk of JNK 16. Kyriakis, J. M., and J. Avruch. 1996. Protein kinase cascades activated by stress activation by LPS in macrophages (Fig. 6). These data demonstrate and inflammatory cytokines. BioEssays 18:567. 17. Logan, S. K., M. Falasca, P. Hu, and J. Schlessinger. 1997. Phosphatidylinositol the ability of extracellular signals to choose the signaling route 3-kinase mediates epidermal growth factor-induced activation of the c-Jun N- which leads to MAPK activation in distinct situations. It is tempt- terminal kinase signaling pathway. Mol. Cell. Biol. 17:5784. ing to speculate that they might do so by modulating the compo- 18. Timokhina, I., H. Kissel, G. Stella, and P. Besmer. 1998. Kit signaling through http://www.jimmunol.org/ PI 3-kinase and Src kinase pathways: an essential role for Rac1 and JNK acti- sition of MAPK modules comprising the dual specificity kinase, its vation in mast cell proliferation. EMBO J. 17:6250. activator, and its substrate (59). This hypothesis could help explain 19. Hambleton, J., S. L. Weinstein, L. Lem, and A. L. DeFranco. 1996. Activation of why many extracellular signals elicit highly specific responses de- c-Jun N-terminal kinase in bacterial lipopolysaccharide- stimulated macrophages. Proc. Natl. Acad. Sci. USA 93:2774. spite their apparent use of similar intracellular pathways. 20. Reimann, T., D. Buscher, R. A. Hipskind, S. Krautwald, M. L. Lohmann-Matthes, The nature of the JNK activator and of the SEK1 substrate(s) in and M. Baccarini. 1994. Lipopolysaccharide induces activation of the Raf-1/MAP ␤ ␤ Salmonella-infected macrophages is at present under investigation. kinase pathway: a putative role for Raf-1 in the induction of the IL-1 and the TNF- genes. J. Immunol. 153:5740. The data reported here extend our understanding of the bio- 21. Sanghera, J. S., S. L. Weinstein, M. Aluwalia, J. Girn, and S. L. Pelech. 1996. chemical events induced by the infection of macrophages with Activation of multiple proline-directed kinases by bacterial lipopolysaccharide in murine macrophages. J. Immunol. 156:4457. by guest on October 1, 2021 Salmonella, and show for the first time that a pathway comprising 22. Naumann, M., T. Rudel, B. Wieland, C. Bartsch, and T. F. Meyer. 1998. Coor- PI 3-K and PKC␨ leads to JNK activation independently of Cdc42/ dinate activation of activator protein 1 and inflammatory cytokines in response to Rac and SEK1. Stress kinase pathways involved in cytokine pro- Neisseria gonorrhoeae epithelial cell contact involves stress response kinases. J. Exp. Med. 188:1277. duction are emerging targets for the therapy of a variety of inflam- 23. Tang, P., C. L. Sutherland, M. R. Gold, and B. B. Finlay. 1998. Listeria mono- matory conditions. Understanding the alignment of stimulus- cytogenes invasion of epithelial cells requires the MEK-1/ERK-2 mitogen-acti- specific pathways will be important for selecting the appropriate vated protein kinase pathway. Infect. Immun. 66:1106. 24. Yang, D., C. Tournier, M. Wysk, H. T. Lu, J. Xu, R. J. Davis, and R. A. Flavell. kinase inhibitor(s) and for the evaluation of their effects in the 1997. Targeted disruption of the MKK4 gene causes embryonic death, inhibition context of a given pathological situation. of c-Jun NH2-terminal kinase activation, and defects in AP-1 transcriptional activity. Proc. Natl. Acad. Sci. USA 94:3004. 25. Foletta, V. C., D. H. Segal, and D. R. Cohen. 1998. Transcriptional regulation in Acknowledgments the immune system: all roads lead to AP-1. J. Leukocyte Biol. 63:139. 26. Chen, C. Y., F. Del Gatto-Konczak, Z. Wu, and M. Karin. 1998. Stabilization of We thank Dr. Silvo Gutkind (National Institute on Dental Research, National interleukin-2 mRNA by the c-Jun NH2-terminal kinase pathway. Science 280: Institutes of Health, Bethesda, MD) for the gift of the plasmids encoding HA- 1945. JNK1, GST-SEK wt, and GST-SEK KR. We also thank Pavel Kovarik, 27. Ming, X. F., M. Kaiser, and C. Moroni. 1998. c-Jun N-terminal kinase is involved Thomas Decker, and Alexander von Gabain (Vienna Biocenter) for construc- in AUUUA-mediated interleukin-3 mRNA turnover in mast cells. EMBO J. 17: 6039. tive discussions and continuous support. 28. Swantek, J. L., M. H. Cobb, and T. D. Geppert. 1997. Jun N-terminal kinase/ stress-activated protein kinase (JNK/SAPK) is required for lipopolysaccharide References stimulation of tumor necrosis factor ␣ (TNF-␣) translation: glucocorticoids inhibit TNF-␣ translation by blocking JNK/SAPK. Mol. Cell. Biol. 17:6274. 1. Francis, C. L., M. N. Starnbach, and S. Falkow. 1992. Morphological and cy- 29. Kontoyiannis, D., M. Pasparakis, T. T. Pizarro, F. Cominelli, and G. Kollias. toskeletal changes in epithelial cells occur immediately upon interaction with 1999. Impaired on/off regulation of TNF biosynthesis in mice lacking TNF AU- Salmonella typhimurium grown under low-oxygen conditions. Mol. Microbiol. rich elements: implications for joint and gut-associated immunopathologies. Im- 6:3077. munity 10:387. 2. Penheiter, K. L., N. Mathur, D. Giles, T. Fahlen, and B. D. Jones. 1997. Non- 30. Dong, C., D. D. Yang, M. Wysk, A. J. Whitmarsh, R. J. Davis, and R. A. Flavell. invasive Salmonella typhimurium mutants are avirulent because of an inability to 1998. Defective T cell differentiation in the absence of Jnk1. Science 282:2092. enter and destroy M cells of ileal Peyer’s patches. Mol. Microbiol. 24:697. 31. Yang, D. D., D. Conze, A. J. Whitmarsh, T. Barrett, R. J. Davis, M. Rincon, and ϩ 3. Chen, L. M., K. Kaniga, and J. E. Galan. 1996. Salmonella spp. are cytotoxic for R. A. Flavell. 1998. Differentiation of CD4 T cells to Th1 cells requires MAP cultured macrophages. Mol. Microbiol. 21:1101. kinase JNK2. Immunity 9:575. 4. Lindgren, S. W., I. Stojiljkovic, and F. Heffron. 1996. Macrophage killing is an 32. Galan, J. E., C. Ginocchio, and P. Costeas. 1992. Molecular and functional char- essential virulence mechanism of Salmonella typhimurium. Proc. Natl. Acad. Sci. acterization of the Salmonella invasion gene invA: homology of InvA to members USA 93:4197. of a new protein family. J. Bacteriol. 174:4338. 5. Monack, D. M., B. Raupach, A. E. Hromockyj, and S. Falkow. 1996. Salmonella 33. Morgan, C., J. W. Pollard, and E. R. Stanley. 1987. Isolation and characterization typhimurium invasion induces apoptosis in infected macrophages. Proc. Natl. of a cloned growth factor dependent macrophage cell line, BAC1.2F5. J. Cell. Acad. Sci. USA 93:9833. Physiol. 130:420. 6. Richter-Dahlfors, A., A. M. J. Buchan, and B. B. Finlay. 1997. Murine salmo- 34. Aktories, K. 1997. Rho proteins: targets for bacterial toxins. Trends Microbiol. nellosis studied by confocal microscopy: Salmonella typhimurium resides intra- 5:282. 4930 STRESS KINASE ACTIVATION IN SALMONELLA-INFECTED MACROPHAGES

35. Buscher, D., R. A. Hipskind, S. Krautwald, T. Reimann, and M. Baccarini. 1995. 48. Palmer, L. E., A. R. Pancetti, S. Greenberg, and J. B. Bliska. 1999. YopJ of Ras-dependent and -independent pathways target the mitogen-activated protein yersinia spp. is sufficient To cause downregulation of multiple mitogen-activated kinase network in macrophages. Mol. Cell. Biol. 15:466. protein kinases in eukaryotic cells. Infect. Immun. 67:708. 36. Le Good, J. A., W. H. Ziegler, D. B. Parekh, D. R. Alessi, P. Cohen, and 49. Berra, E., M. T. Diaz-Meco, J. Lozano, S. Frutos, M. M. Municio, P. Sanchez, P. J. Parker. 1998. Protein kinase C isotypes controlled by phosphoinositide 3-ki- L. Sanz, and J. Moscat. 1995. Evidence for a role of MEK and MAPK during nase through the protein kinase PDK1. Science 281:2042. signal transduction by protein kinase C ␨. EMBO J. 14:6157. 37. De Maria, R., M. R. Rippo, E. H. Schuchman, and R. Testi. 1998. Acidic sphin- 50. Schonwasser, D. C., R. M. Marais, C. J. Marshall, and P. J. Parker. 1998. Acti- gomyelinase (ASM) is necessary for Fas-induced GD3 ganglioside accumulation vation of the mitogen-activated protein kinase/extracellular signal-regulated ki- and efficient apoptosis of lymphoid cells. J. Exp. Med. 187:897. nase pathway by conventional, novel, and atypical protein kinase C isotypes. Mol. 38. Lopez-Ilasaca, M., J. S. Gutkind, and R. Wetzker. 1998. Phosphoinositide 3-ki- Cell. Biol. 18:790. nase ␥ is a mediator of ␤␥-dependent Jun kinase activation. J. Biol. Chem. 273: 2505. 51. Soh, J. W., E. H. Lee, R. Prywes, and I. B. Weinstein. 1999. Novel roles of 39. Herrera-Velit, P., and N. E. Reiner. 1996. Bacterial lipopolysaccharide induces specific isoforms of protein kinase C in activation of the c-fos serum response the association and coordinate activation of p53/56lyn and phosphatidylinositol element. Mol. Cell. Biol. 19:1313. 3-kinase in human monocytes. J. Immunol. 156:1157. 52. Takeda, H., T. Matozaki, T. Takada, T. Noguchi, T. Yamao, M. Tsuda, F. Ochi, 40. Chou, M. M., W. Hou, J. Johnson, L. K. Graham, M. H. Lee, C. S. Chen, K. Fukunaga, K. Inagaki, and M. Kasuga. 1999. PI 3-kinase ␥ and protein kinase ␨ A. C. Newton, B. S. Schaffhausen, and A. Toker. 1998. Regulation of protein C- mediate RAS-independent activation of MAP kinase by a Gi protein-coupled kinase C ␨ by PI 3-kinase and PDK-1. Curr. Biol. 8:1069. receptor. EMBO J. 18:386. 41. Martiny-Baron, G., M. G. Kazanietz, H. Mischak, P. M. Blumberg, G. Kochs, 53. Diaz-Meco, M. T., I. Dominguez, L. Sanz, P. Dent, J. Lozano, M. M. Municio, H. Hug, D. Marme, and C. Schachtele. 1993. Selective inhibition of protein E. Berra, R. T. Hay, T. W. Sturgill, and J. Moscat. 1994. ␨ PKC induces phos- kinase C isozymes by the indolocarbazole Go 6976. J. Biol. Chem. 268:9194. phorylation and inactivation of I␬B-␣ in vitro. EMBO J. 13:2842. 42. Nakanishi, H., K. A. Brewer, and J. H. Exton. 1993. Activation of the ␨ isozyme 54. Rooney, J. W., M. R. Hodge, P. G. McCaffrey, A. Rao, and L. H. Glimcher. 1994. of protein kinase C by phosphatidylinositol 3,4,5-trisphosphate. J. Biol. Chem. A common factor regulates both Th1- and Th2-specific cytokine gene expression. 268:13. EMBO J. 13:625. 43. Gratas, C., and G. Powis. 1993. Inhibition of phospholipase D by agents that inhibit cell growth. Anticancer Res. 13:1239. 55. Ways, D. K., K. Posekany, J. deVente, T. Garris, J. Chen, J. Hooker, W. Qin,

␨ Downloaded from 44. Chen, Y. G., A. Siddhanta, C. D. Austin, S. M. Hammond, T. C. Sung, P. Cook, D. Fletcher, and P. Parker. 1994. Overexpression of protein kinase C- M. A. Frohman, A. J. Morris, and D. Shields. 1997. Phospholipase D stimulates stimulates leukemic cell differentiation. Cell Growth Differ. 5:1195. release of nascent secretory vesicles from the trans-Golgi network. J. Cell Biol. 56. Grassme, H., E. Gulbins, B. Brenner, K. Ferlinz, K. Sandhoff, K. Harzer, F. Lang, 138:495. and T. F. Meyer. 1997. Acidic sphingomyelinase mediates entry of N. gonor- 45. Schutze, S., K. Potthoff, T. Machleidt, D. Berkovic, K. Wiegmann, and rhoeae into nonphagocytic cells. Cell 91:605. ␬ M. Kronke. 1992. TNF activates NF- B by phosphatidylcholine-speciﬁc phos- 57. MacKichan, M. L., and A. L. DeFranco. 1999. Role of ceramide in lipopolysac- pholipase C-induced “acidic” sphingomyelin breakdown. Cell 71:765. charide (LPS)-induced signaling: LPS increases ceramide rather than acting as a 46. Palmer, L. E., S. Hobbie, J. E. Galan, and J. B. Bliska. 1998. YopJ of Yersinia structural homolog. J. Biol. Chem. 274:1767.

pseudotuberculosis is required for the inhibition of macrophage TNF-␣ produc- http://www.jimmunol.org/ 58. Shirakabe, K., K. Yamaguchi, H. Shibuya, K. Irie, S. Matsuda, T. Moriguchi, tion and downregulation of the MAP kinases p38 and JNK. Mol. Microbiol. Y. Gotoh, K. Matsumoto, and E. Nishida. 1997. TAK1 mediates the ceramide 27:953. signaling to stress-activated protein kinase/c-Jun N-terminal kinase. J. Biol. 47. Ruckdeschel, K., J. Machold, A. Roggenkamp, S. Schubert, J. Pierre, R. Zumbihl, Chem. 272:8141. J. P. Liautard, J. Heesemann, and B. Rouot. 1997. Yersinia enterocolitica pro- motes deactivation of macrophage mitogen-activated protein kinases extracellular 59. Elion, E. A. 1998. Routing MAP kinase cascades. Science 281:1625. signal-regulated kinase-1/2, p38, and c-Jun NH2-terminal kinase: correlation with 60. Lundberg, V., V. Vinatzer, D. Berdnik, A. Gabain, and M. Baccarini. 1999. its inhibitory effect on tumor necrosis factor-␣ production. J. Biol. Chem. 272: Growth phase-regulated induction of Salmonella-induced macrophage apoptosis 15920. correlates with transient expression of SPI-1 genes. J. Bacteriol. 181:3433. by guest on October 1, 2021