Listeria Innocua Nonpathogenic P60-Specific CD4 and CD8 T Cell

Enhancement of the Listeria monocytogenes p60-Specific CD4 and CD8 T Cell Memory by Nonpathogenic Listeria innocua

This information is current as Gernot Geginat, Thomas Nichterlein, Marianne Kretschmar, of September 25, 2021. Simone Schenk, Herbert Hof, Mio Lalic-Mülthaler, Werner Goebel and Andreas Bubert J Immunol 1999; 162:4781-4789; ; http://www.jimmunol.org/content/162/8/4781 Downloaded from

References This article cites 56 articles, 30 of which you can access for free at: http://www.jimmunol.org/content/162/8/4781.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 September 25, 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. Enhancement of the Listeria monocytogenes p60-Speciﬁc CD4 and CD8 T Cell Memory by Nonpathogenic Listeria innocua1

Gernot Geginat,2* Thomas Nichterlein,* Marianne Kretschmar,* Simone Schenk,* Herbert Hof,* Mio Lalic-Mu¨lthaler,† Werner Goebel,† and Andreas Bubert3†

The contact of T cells to cross-reactive antigenic determinants expressed by nonpathogenic environmental micro-organisms may contribute to the induction or maintenance of T cell memory. This hypothesis was evaluated in the model of murine Listeria monocytogenes infection. The influence of nonpathogenic L. innocua on the L. monocytogenes p60-specific T cell response was analyzed. We show that some CD4 T cell clones raised against purified p60 from L. monocytogenes cross-react with p60 purified from L. innocua. The L. monocytogenes p60-specific CD4 T cell clone 1A recognized the corresponding L. innocua p60 peptide QAAKPAPAPSTN, which differs only in the first amino acid residue. In vitro experiments revealed that after L. monocytogenes infection of APCs, MHC class I-restricted presentation of p60 occurs, while MHC class II-restricted p60 presentation is inhibited. Downloaded from L. innocua-infected cells presented p60 more weakly but equally well in the context of both MHC class I and MHC class II. In contrast to these in vitro experiments the infection of mice with L. monocytogenes induced a strong p60-specific CD4 and CD8 T cell response, while L. innocua infection failed to induce p60-specific T cells. L. innocua booster infection, however, expanded p60-specific memory T cells induced by previous L. monocytogenes infection. In conclusion, these findings suggest that infection with a frequently occurring environmental bacterium such as L. innocua, which is nonpathogenic and not adapted to intracellular replication, can contribute to the maintenance of memory T cells specific for a related intracellular pathogen. The Journal of http://www.jimmunol.org/ Immunology, 1999, 162: 4781–4789.

isteria monocytogenes is a facultative intracellular bacte- which have been shown to play a role in the bacterial uptake by rium that can cause severe systemic infections in human various types of mammalian cells, do not seem to be present in Lgroups at risk, i.e., in immunosuppressed and elderly people, nonpathogenic Listeria species (10). as well as in neonates and pregnant women (1). It represents the The established model of the murine L. monocytogenes infection best studied member of the genus Listeria, which includes five is characterized by the development of a long-lasting T cell-de- additional species, L. ivanovii, L. innocua, L. seeligeri, L. welshi- pendent immunity (11). The murine L. monocytogenes infection by guest on September 25, 2021 meri, and L. grayi. In addition to L. monocytogenes the only other induces an initial induction and activation of Ag-specific effector T potentially pathogenic species of the genus is L. ivanovii (2). All cells followed by subsequent contraction by apoptosis and the es- other members are considered harmless environmental bacteria. L. tablishment of a T cell memory compartment (12, 13). It remains innocua shows the closest phylogenetic relationship to L. mono- uncertain whether the prolonged life span of memory T cells is an cytogenes (3) and is the most frequent listerial isolate from human intrinsic property of these cells or whether it reflects their inter- feces (4) and various food sources (5–8). Pathogenic and non- mittent stimulation by residual deposits of specific Ag, through pathogenic Listeria species differ in multiple pathogenicity-asso- cross-reactive contact with environmental Ags or through by- ciated gene loci. The gene products of the PrfA-dependent gene stander stimulation (14–16). The investigation of listerial target cluster, which are crucial for the release from the host cell phago- Ags recognized by CD4 and CD8 T cells identified several se- some, intracellular replication, intracellular movement, and the creted proteins of L. monocytogenes as potent T cell Ags. The cell-to-cell spread of pathogenic Listeria species (9), are lacking in listeriolysin O (LLO)4 and the p60 protein from L. monocytogenes L. innocua. In addition, genes coding for internalins, some of are targets for both CD4 and CD8 T cells (17–20). It has been shown that LLO-specific CD8 T cell clones and p60-specific CD4 or CD8 T cell clones mediate protective immunity against L. *Institut fu¨r Medizinische Mikrobiologie und Hygiene, Fakulta¨t fu¨r Klinische Medi- zin Mannheim der Universita¨t Heidelberg, Mannheim, Germany; and †Lehrstuhl fu¨r monocytogenes in vivo (20–22). In contrast to the PrfA-regulated Mikrobiologie, Theodor-Boveri-Institut fu¨r Biowissenschaften, Am Hubland, Wu¨rz- virulence factors of L. monocytogenes, the PrfA-independent p60 burg, Germany protein is essential for cell viability, and it acts basically as a Received for publication October 27, 1998. Accepted for publication January murein hydrolase required in a late step of cell division (23). It is 14, 1999. the major extracellular protein from pathogenic L. monocytogenes The costs of publication of this article were defrayed in part by the payment of page and contributes to the uptake of this pathogen into some mamma- charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. lian cell types such as fibroblasts and macrophages (23–26). Be- 1 This work was supported by the Forschungsfond der Fakulta¨t fu¨r Klinische Medizin cause of the essential function of p60 for bacterial cell division, it Mannheim der Universita¨t Heidelberg, Projekt 13/1998. M.L.-M. received a stipend is not surprising that a p60-related protein is produced by all mem- from the Graduiertenkolleg Infektiologie funded by the Deutsche Forschungsgemein- bers of the genus Listeria (25). The amino acid sequence compar- schaft. ison of the whole p60 protein family revealed that the amino- and 2 Address correspondence and reprint requests to Dr. G. Geginat, Institut fu¨r Mediz- inische Mikrobiologie und Hygiene, Fakulta¨t fu¨r Klinische Medizin Mannheim der Universita¨t Heidelberg, Theodor-Kutzer-Ufer 1-3, 68167 Mannheim, Germany. E- 4 Abbreviations used in this paper: LLO, listeriolysin O; LM-p60, p60 produced by mail address: [email protected] Listeria monocytogenes; LI-p60, p60 produced by Listeria innocua; rmIFN, recom- 3 Current address: Microbiological Analytics, Merck KGaA, Darmstadt, Germany. binant murine IFN; ELISPOT, enzyme-linked immunospot.

Copyright © 1999 by The American Association of Immunologists 0022-1767/99/$02.00 4782 NONPATHOGENIC LISTERIA AND T CELL MEMORY carboxyl-terminal portions of p60 are highly conserved in all Lis- Peptide libraries and synthetic peptides teria species, while the inner portions of p60 differ in a species- For mapping of the LI-p60 epitope recognized by the cross-reactive CD4 specific way (25). T cell clone 1A, soluble libraries of overlapping deca- and dodecapeptides The remarkable conservation of the p60 protein, which is a (offset ϭ one amino acid) covering the previously identified antigenic re- strong target Ag for protective L. monocytogenes-specific CD4 and gion of the p60 molecule from L. innocua were synthesized. The synthesis CD8 T cells, provides an excellent model to study the influence of of a soluble peptide library representing the LM-p60 was described previously (20). The individual peptide spots were excised and distributed into a related nonpathogenic species on the T cell-mediated immunity different 1.5-ml tubes for cleavage due to diketopiperazine formation at against a pathogenic bacterium. We show that infection with an neutral pH. The biological activity of individual peptides diluted 1/20 was environmental nonpathogenic bacterium such as L. innocua, which performed with a T cell activation assay as described below. Definitive by itself is unable to induce a primary p60-specific T cell response, amounts of peptides capable of inducing T cell proliferation were synthesized on a Zinsser (Frankfurt, Germany) Analytic SMPS 350 A peptide mediates the expansion of p60-specific memory T cells induced by synthesizer. previous L. monocytogenes infection. These findings suggest that contact with a frequently occurring harmless bacterium that is not Accessory cells adapted to an intracellular life can contribute to the maintenance of Spleen cells, B7-transfected P815 cells, or P388D1 cells were used as ac- memory T cells specific for an antigenically related intracellular cessory cells. Mitomycin C-inactivated spleen cells were prepared as de- pathogen. scribed previously (20) and were used as accessory cells in T cell proliferation assays and T cell activation assays. MC201A5, P815 transfected with the human B7.1 gene (P815/B7) (28), were obtained from M. Red- Materials and Methods dehase (University Mainz, Mainz, Germany) with the permission of the Bacterial strains DNAX Research Institute (Palo Alto, CA). Mitomycin C-inactivated Downloaded from P815/B7 were used as accessory cells for the restimulation of CD8 T cell L. monocytogenes serovar 1/2a EGD, L. innocua sv6b, and the actA dele- lines. P815/B7 cells were cultured in RPMI supplemented with 5% FCS, tion mutant L. monocytogenes ⌬actA were taken from the strain collection Ϫ 100 U/ml penicillin, 100 ␮g/ml streptomycin, 1 ϫ 10 5 M 2-ME, 2 mM of the Institute of Microbiology at the University of Wu¨rzburg (Wu¨rzburg, glutamine, and 167 ␮g/ml G418. For in vitro infection experiments Germany). The listeriolysin deletion mutant L. monocytogenes ⌬hly was P388D cells were used as APCs. obtained from D. Portnoy, University of Pennsylvania (Philadelphia, PA), 1 and Bacillus subtilis DB104 was provided by R. Doi, University of Cali- T cell clones and lines fornia (Davis, CA). All bacteria were grown in brain-heart infusion broth http://www.jimmunol.org/ (Difco, Augsburg, Germany). The bacterial concentration was estimated by The generation and culture of the p60-specific CD4 T cell clones 1A, 3D, determination of the OD600 and was confirmed by colony counts on sheep 5F, and 6G were described previously (20). CD8 T cell lines specific for blood agar plates. p60217–225 and p60449–457 were derived from L. monocytogenes-infected BALB/c mice. Mice were primed with 5 ϫ 103 CFU of L. monocytogenes Mice i.p. and boosted with 1 ϫ 106 CFU 14 days later. Spleens were removed 10 days later, and bulk cultures were set in 24-well flat-bottom plates with an Female BALB/cOlaHsd (H-2d) mice were purchased (Harlan-Winkel- initial concentration of 15 ϫ 106 spleen cells/well in 1.5 ml of cell culture mann, Borchen, Germany), kept under conventional conditions, and used at medium (␣ modification of Eagle’s medium; PAA, Wien, Austria) supple- 6–8 wk of age. For long term experiments mice were kept under specific mented with 10% FCS, 100 U/ml penicillin, 100 ␮g/ml streptomycin, 1 ϫ pathogen-free conditions in a laminar air flow container. Ϫ5 ␮ ␣

10 M 2-ME, 2 mM glutamine, and 0.5 g/ml amphothericin B ( MEM) by guest on September 25, 2021 Ϫ8 Infection of mice and in vivo protection assay and 10 M synthetic HPLC-purified p60217–225 or p60449–457 peptide, respectively. After an initial 5-day culture period T cells were grown in Mice were infected by i.v. injection of Listeria in 0.2 ml of PBS. Primary ␣MEM supplemented with 20 U of rmIL-2/ml (R&D Systems, Wiesbaden- L. monocytogenes infection was performed with 1 ϫ 103 CFU of L. mono- Nordenstadt, Germany). CD8 T cell lines were restimulated every 3–4 wk cytogenes. Secondary L. monocytogenes infection was performed with 1 ϫ with mitomycin C-inactivated P815/B7 cells as accessory cells. Per well of 106 CFU of L. monocytogenes. Priming with low dose L. monocytogenes a 24-well plate, 0.4 ϫ 106 T cells were seeded with 0.8 ϫ 106 P815/B7 in was performed with a dose of 1 ϫ 102 CFU. Priming with high dose L. ␣MEM containing 20 U of rmIL-2/ml and supplemented with 10Ϫ9 M ϫ 3 Ϫ9 monocytogenes was performed with 5 10 CFU. Primary and secondary p60217–225 or 10 M p60449–457 peptide. Cytotoxic CD8 T cell lines gen- L. innocua infections were performed with 2 ϫ 107 CFU. L. innocua erated in the presence of P815/B7 showed the same characteristics as CD8 booster infection of L. monocytogenes-primed mice was performed with T cell lines generated in the presence of inactivated spleen cells. 1 ϫ 106 CFU. For the in vivo protection assay immunized mice were challenged by i.v. injection of 1 ϫ 104 CFU of L. monocytogenes. Spleen T cell activation assays and liver were removed 72 h after infection and were homogenized in T cell activation in the presence of inactivated spleen accessory cells and Tenbroeck tissue grinders (Wheaton, Millville, NY) with 10 ml of sterile Ag was measured by [3H]thymidine incorporation into proliferating T cells H O. Homogenates were diluted serially, and aliquots of relevant dilutions 2 or by detection of IL-2 in culture supernatants as previously described (20). were plated on tryptose agar. Colonies were counted after 48 h of incuba- In brief, 5 ϫ 104 washed T cells were cultured with 2 ϫ 105 inactivated tion. Colony counts were corrected for dilution and averaged to yield CFU accessory cells and the appropriate Ag in 150 ␮lof␣MEM/well in round- per organ. The level of protection was calculated as the log difference of 10 bottom 96-well microtiter plates. Culture supernatants were harvested the bacterial count from immunized mice and naive control mice. Data are 18–24 h after initiation of cultures. IL-2 activity in supernatants was de- presented as the average of individual experiments with five mice per tected in a bioassay using IL-2-dependent HT-2 cells. Proliferation of T group. Each experiment was performed at least twice with similar results. cells and HT-2 cells was measured by [3H]thymidine incorporation. The statistical significance of results was tested as described below. T cell activation by P388D1 macrophage-like cells was measured by the ␥ Purification of p60 detection of IFN- in culture supernatants. P388D1 cells were precultured for 48 h without antibiotics and in the presence of 20 U of rmIFN-␥/ml The purification of p60 from the p60-overproducing strain L. monocyto- (R&D Systems) to increase MHC class II expression and were plated in a genes MR1 pGB363–1p60 (LM-p60) was described previously (20). Pu- final concentration of 1 ϫ 105 cells in flat-bottom 96-well microtiter plates. rification of p60 from L. innocua (LI-p60) was performed similarly. In After the removal of IFN-␥-supplemented medium cells were either in- brief, supernatants from L. innocua cultures were harvested at early sta- fected or loaded with synthetic peptides or purified p60. Bacterial infection tionary growth phase, and the proteins were precipitated with TCA, washed was performed with centrifugal enhancement (10 min at 200 ϫ g). After 30 with acetone, dissolved in sample buffer, heated at 95°C for 20 min, and min at 37°C infected cells were washed once with ␣MEM supplemented separated by SDS-PAGE. After staining, the 60-kDa band representing the with 50 ␮g/ml gentamicin, and culture medium supplemented with p60 protein was excised and transferred into a BIOTRAP BT 1000-cham- 10 ␮g/ml gentamicin was added. After 6-h incubation at 37°C cells were ber (Schleicher & Schuell, Dassel, Germany) for overnight elution. Purified fixed for 10 min with 1% paraformaldehyde in PBS (29). After washing proteins were used after dialysis against Tris-HCl buffer. To confirm purity, with culture medium without FCS (wash medium) paraformaldehyde was aliquots of the protein samples were reseparated by SDS-PAGE and either inactivated by addition of a 1/1 mixture of lysine buffer (36.54 g of lysine stained with Coomassie blue or transferred onto nitrocellulose for immu- in 500 ml of H2O) with wash medium for 20 min. Finally after four ad- noblot analysis with a rabbit anti-p60 antiserum (27). ditional washing steps 5 ϫ 104 T cells were added to each well in ␣MEM The Journal of Immunology 4783 supplemented with 10 ␮g/ml gentamicin. Culture supernatants were harvested after 12–18 h at 37°C, and the IFN-␥ concentration was subsequently determined with an IFN-␥-specific sandwich ELISA kit (R&D Systems), which binds and detects IFN-␥ with a pair of specific mAb. The ELISA assay was performed as suggested by the manufacturer. IFN-␥ production in individual samples was calculated from the titration of a sup- plied rmIFN-␥ standard. Results were corrected for the sample dilution to yield the sample concentration in picograms per milliliter. Enzyme-linked immunospot (ELISPOT) assay The frequency of p60-specific T lymphocytes was determined with the ELISPOT assay (12, 30). ELISPOT assays were performed in polyvinyli- dene fluoride-backed 96-well microtiter plates (Millipore, Eschborn, Ger- many) using a modified standard protocol provided by the manufacturer. Wells were coated overnight with bicarbonate coating buffer, pH 9.6, supplemented with 10 ␮g/ml of rat anti-mouse IFN-␥ mAb (RMMG-1; Bio- Source, Camarilla, CA) or rat anti-mouse IL-4 (clone BVD4-1D11; PharMingen, San Diego, CA) for the detection of IFN-␥- or IL-4-producing cells, respectively. Before the addition of cells all wells were washed ␣ four times with H2O and subsequently blocked with MEM for 30 min at 37°C. To determine the frequency of p60-specific cells in infected mice, ϫ 5 4 10 splenocytes were preincubated for 18 h in round-bottom 96-well FIGURE 1. Purification of the p60 protein from L. innocua. Purification Downloaded from microtiter plates in a final volume of 150 ␮lof␣MEM in the presence of ␮ ␮ Ϫ8 Ϫ8 of p60 from culture supernatants was performed by TCA precipitation 5 g/ml LM-p60, 5 g/ml LI-p60, 10 M p60217–225,10 M p60301–312, Ϫ8 followed by preparative SDS-gel chromatography. Aliquots of p60 prep- or 10 M p60449–457. Controls were performed without Ag or in the Ϫ8 arations were analyzed by SDS-PAGE. Proteins were visualized directly presence of 10 M LLO91–99. This preincubation step in round-bottom wells was required for the optimal activation of T cells. From these primary with Coomassie blue (A) or were blotted on nitrocellulose and analyzed cultures 1 ϫ 105 or 1 ϫ 104 cells were transferred to anti-IFN-␥- or anti- with an anti-p60 antiserum (B). Lane M, The m.w. standard; lane 1, TCA- IL-4-coated wells. After 12–16 h at 37°C wells were washed 10 times with precipitated supernatant from the pGB363–1p60-transformed, LM-p60- 200 ␮l PBS/0.25% Tween 20 (wash buffer). For the detection of bound overproducing strain L. monocytogenes EGD MR1; lane 2, purified LM- http://www.jimmunol.org/ ␥ ␮ IFN- or IL-4, wells were incubated for2hat37°C with 1 g/ml biotin- p60 protein from L. monocytogenes; lane 3, TCA-precipitated supernatant ␥ labeled rat anti-mouse IFN- mAb (clone XMG1.2, PharMingen) or rat from L. innocua; lane 4, purified LI-p60 protein from L. innocua. anti-mouse IL-4 mAb (clone BVD6-24G2; PharMingen), respectively. Subsequently, after washing five times with wash buffer 50 ␮l of horse- radish peroxidase-streptavidin conjugate (Dianova, Hamburg, Germany) diluted to a concentration of 4 ␮g/ml in wash buffer was added. After 2-h After purification, the LI-p60 preparation revealed a homogeneous incubation at room temperature wells were washed four times with wash molecular mass of 60 kDa without any visible contamination (Fig. buffer and developed with 50 ␮l/well aminoethylcarbazole solution. A 1A, lane 4). The identity of the 60-kDa protein was confirmed by 20-mg tablet of 3-amino-9-ethylcarbazole (Sigma, Deisenhofen, Germany) Western blot analysis with an LM-p60-specific antiserum that

was dissolved in 2.5 ml n,n-dimethylformamide (Sigma). After addition of by guest on September 25, 2021 ␮ cross-reacts with LI-p60 (25) (Fig. 1B). 47.5 ml of sodium acetate buffer, pH 5.0, and 25 lofH2O2 the solution was filtered through a 0.2-␮m syringe filter and used immediately. After Several LM-p60-specific CD4 T cell clones were described pre- 20–30 min at room temperature the wells were washed three times with viously (20). These LM-p60-specific T cell clones were now tested H2O and air-dried. The frequency of Ag-specific cells was calculated as the for cross-reactivity with LI-p60. T cells were cocultivated with number of spots per number of splenocytes seeded. The specificity and inactivated spleen cells in the presence of graded amounts of either sensitivity of the ELISPOT assay were controlled with IFN-␥-secreting CD4 T cell clones (20) and an IL-4-transfected cell line provided by A. purified LM-p60 or LI-p60. T cell activation was measured by Limmer (Zentrum fu¨r Molekulare Biologie, Heidelberg, Germany). The [3H]thymidine incorporation and expressed as the stimulation in- plating efficiency was ϳ80% for both cell types. dex. All clones revealed a dose-dependent LM-p60-specific proliferation (Fig. 2, open bars). In the presence of LI-p60 (Fig. 2, Statistical analysis solid bars) only clones 1A and 3D showed Ag-specific prolifera- Statistical analysis of the results of in vitro experiments was performed tion, while clones 5F and 6G were strictly LM-p60 specific. The with the Newman-Keuls multiple comparison test at the 0.05 significance level. The statistical significance of the results of in vivo experiments was checked using the nonparametric Tukey multiple comparison test at the 0.05 significance level. All tests were performed using the WINKS statistical analysis software (Texasoft, Cedar Hill, TX). Results Some CD4 T cell clones specific for L. monocytogenes p60 cross-react with L. innocua p60 L. monocytogenes infection induces a potent p60-specific Th1 immune response. We have shown previously that p60-specific Th1 clones mediate significant protection against L. monocytogenes infection (20). Because LM-p60 from pathogenic L. monocytogenes and LI-p60 from nonpathogenic L. innocua display about 90% homology (25), it was tested whether LM-p60-specific CD4 T cells FIGURE 2. Th1 clones 1A and 3D raised against LM-p60 cross-react cross-react with LI-p60. LI-p60 from culture supernatants of L. with LI-p60. T cell clones 1A, 3D, 5F, and 6G specific for LM-p60 were innocua was purified by preparative SDS-PAGE and subsequent incubated with graded amounts of either purified LM-p60 protein (open gel elution. LM-p60 was purified from the p60-overproducing bars) or LI-p60 (filled bars) in the presence of inactivated syngenic spleen strain L. monocytogenes EGD MR1 pGB363-1p60 (23). SDS- cells as accessory cells. The proliferation of T cell clones was measured by PAGE analyses of TCA-precipitated supernatants revealed that incorporation of [3H]thymidine. Results are expressed as the mean stimu- p60 is a major secreted protein from L. innocua (Fig. 1A, lane 3). lation index (SI) of triplicate determinations. 4784 NONPATHOGENIC LISTERIA AND T CELL MEMORY

Differences in p60 Ag presentation by macrophages infected with L. monocytogenes and L. innocua Induction of either Th cells or cytolytic T cells depends on the access of the Ag to the appropriate Ag presentation pathway (31). L. monocytogenes escapes into the cytosol and gains access to the cytosolic MHC class I presentation pathway mainly by the activity of LLO (32). To compare p60 Ag presentation by L. monocytogenes and L. innocua-infected cells, the recognition of Listeria-

infected macrophage-like P388D1 cells by LM-p60301–312-speciﬁc

CD4 (20) and LM-p60217–225 or LM-p60449–457-speciﬁc CD8 T cells (33) was analyzed. The sequences of both p60 CD8 T cell epitopes are completely conserved in LM-p60 and LI-p60 (25), and we have shown that LI-p60 is recognized by the CD4 T cell clone 1A (Fig. 2). T cell activation was measured by IFN-␥ se-

cretion into the culture supernatant. P388D1 cells loaded with

p60217–225 or with p60449–457 were recognized in a dose-dependent way by the corresponding peptide-speciﬁc CD8 T cells but not at all by p60-speciﬁc CD4 T cells (Fig. 4A). APC loaded either Downloaded from

with LM-p60 or LI-p60 were recognized by the p60301–312-specific CD4 T clone with similar sensitivity but were not recognized by p60-specific CD8 T cells (Fig. 4B), indicating that the MHC class I-presented p60 peptides were not generated from exogenous p60 protein. http://www.jimmunol.org/ FIGURE 3. Characterization of the LI-p60 epitope recognized by the The presentation of p60 by Listeria-infected cells was analyzed cross-reactive CD4 T cell clone 1A. A, Mapping of the cross-reactive antigenic under conditions that limit Ag processing to a defined period. T peptide recognized by clone 1A with a set of dodecamer peptides with one- cells were added after APC were fixed 6 h postinfection. L. mono- residue offset that maps the relevant region of LI-p60 (LI-p60 ). Recog- 289–318 cytogenes-infected P388D1 APC were recognized preferentially by nition of synthetic peptide by clone 1A was measured as IL-2 activity in cul- p60-specific CD8 T cells. These cells were recognized poorly by ture supernatants. B, The results of the peptide mapping analysis were p60-specific CD4 T cells, which showed only weak IFN-␥ pro- confirmed with synthetic HPLC-purified p60301–312 peptide. Ag-specific proliferation of clone 1A was measured in the presence of graded concentrations duction (Fig. 4C, upper left panel). In contrast to L. monocyto- of synthetic LI-p60299–310 (circles) or LM-p60301–312 (squares). Results are genes-infected cells, which were recognized by p60-specific CD8 Ϫ Ϫ expressed as the mean stimulation idex (SI) of triplicate cultures. T cells even after infection with 10 2-10 3 CFU/cell, CD8 T cell by guest on September 25, 2021 recognition of L. innocua-infected cells required infection with at sensitivity of clone 1A was almost equal for both p60 preparations least 10 bacteria/cell. Remarkably, L. innocua-infected cells pre- (Fig. 2, upper left panel). Clone 3D was ϳ5-fold more sensitive for sented p60 equally in the MHC class I and the MHC class II LM-p60 than LI-p60 (Fig. 2, upper right panel). context, as indicated by the activation of p60-specific CD4 and We have previously shown that the LM-p60-specific T cell CD8 T cells (Fig. 4C, lower right panel). To study the influence of clone 1A is specific for the 12-mer peptide LM-p60301–312 a possible cell-to-cell spread of L. monocytogenes on Ag presen- (EAAKPAPAPSTN) (20). The comparison of the known se- tation of p60 (29), APC were infected with the L. monocytogenes quences of LM-p60 and LI-p60 reveals that this antigenic region is ⌬actA deletion mutant, which lacks actA expression and is there- highly conserved between LM-p60 and LI-p60 (25). The LM-p60- fore unable to infect secondary cells by cell-to-cell spread (34). specific CD4 T cell clone 1A was screened against a decamer and Under the experimental conditions used, recognition of L. mono- dodecamer peptide library with one residue offset covering the cytogenes ⌬actA-infected cells by p60-specific CD4 and CD8 T corresponding LI-p60 region LI-p60289–318. All fractions were cells (Fig. 4C, upper right panel) was similarly efficient as recog- tested with clone 1A in the presence of syngenic inactivated spleen nition of L. monocytogenes wild-type-infected cells, indicating that cells, and the IL-2 activity in supernatants was screened with HT-2 no active cell-to-cell spread occurred during the first 6 h postin- cells. Only the peptide QAAKPAPAPSTN (LI-p60 )ofthe 299–310 fection. The relevance of active LLO secretion for the inhibition of dodecamer library with one residue offset between subsequent pep- the MHC class II-restricted p60 presentation by L. monocytogenes- tides was found to be biologically active (Fig. 3A). No activity was infected cells was analyzed with the LLO deletion mutant L. mono- obtained with any peptide of the decamer peptide library with one cytogenes ⌬hly. APC infected with this mutant, which is unable to residue offset (data not shown). The antigenicity of the LI-p60 escape from the host cell phagosome (9), were recognized by p60- epitope 299–310 was confirmed with newly synthesized, HPLC- purified LI-p60 peptide in a proliferation assay with clone specific CD8 and CD4 T cells with similar efficacy, indicating that 299–310 LLO secretion is responsible for the different p60 Ag presentation 1A. Remarkably, the recognition of LI-p60299–310 by clone 1A was 5- preferences of APC infected with L. monocytogenes or L. innocua, to 10-fold less sensitive than the recognition of LM-p60301–312 (Fig. 3B), although recognition of the whole purified p60 proteins of respectively (Fig. 4C, lower left panel). Thus, p60 Ag presentation both species was equally sensitive (Fig. 2, upper left panel). Thus, by L. monocytogenes and L. innocua-infected macrophages is sig- these results indicate that some LM-p60-specific CD4 T cell clones nificantly different. Whereas the strong p60 presentation by L. have the potential to cross-react with the LI-p60 derived from non- monocytogenes-infected macrophage-like cells is restricted pre- pathogenic L. innocua. Particularly, the epitope recognized by dominantly by MHC class I molecules, the p60 presentation by L. clone 1A is functionally conserved in the p60 proteins from L. innocua-infected cells is weaker and occurs in the context of both monocytogenes and L. innocua. MHC class I and MHC class II molecules. The Journal of Immunology 4785 Downloaded from

FIGURE 5. Frequency of Ag-speciﬁc T cells in the spleens of L. mono-

cytogenes and L. innocua-infected mice. The frequency of Ag-speciﬁc T http://www.jimmunol.org/ cells was determined with the ELISPOT assay as described in Materials

and Methods. The frequency of cells reactive with p60449–457, p60217–225, p60301–312, LM-p60, or LI-p60 or without Ag is shown as the number of reactive cells per 1 ϫ 105 splenocytes. The dotted line at 0.2 ϫ 10Ϫ5 indicates the detection limit of the ELISPOT assay. The frequency of IFN- ␥-secreting (filled bars) or IL-4-secreting (open bars) T cells was determined in spleens of BALB/c mice infected and boosted with L. monocytogenes or L. innocua or of mice that remained uninfected as indicated. The SD of triplicate cultures is indicated. Asterisks indicate values significantly Ͻ by guest on September 25, 2021 FIGURE 4. Ag presentation of p60 by P388D1 cells infected with L. (p 0.05) above background activity without Ag. monocytogenes or L. innocua. Cloned p60301–312-specific CD4 T cells (filled diamonds), p60217–225-specific CD8 T cells (open circles), or p60449–457-specific CD8 T cells (open squares) were incubated with cific IFN-␥-producing Th1 cells and IL-4-producing Th2 cells, re- P388D1 cells. APC were loaded with p60217–225 or p60449–457 (A), loaded spectively (35). Control experiments have shown that APC loaded with LM-p60 or LI-p60 (B), or infected with L. monocytogenes wild-type with purified p60 protein are efficiently recognized by p60-specific (LM wt), L. innocua (LI), or the deletion mutants L. monocytogenes ⌬actA CD4 T cells but not by CD8 T cell lines specific for p60 or (LM ⌬actA) and L. monocytogenes ⌬hly (LM ⌬hly; C). Ag processing by 217–225 APC was limited to a period of 6 h postinfection by fixation of APC with p60449–457 (Fig. 4B). Mice were primed with L. monocytogenes or 1% paraformaldehyde before addition of T cells. Activation of T cells was with L. innocua and boosted on day 14 postinfection. Control mice measured as the IFN-␥ concentration in culture supernatants. Results are were not infected. Spleens were removed 10 days after the booster expressed as the IFN-␥ concentration in picograms per milliliter, and the infection and used for the ELISPOT test. L. monocytogenes-in- SD of triplicate cultures is indicated. The dotted line at 20 pg/ml indicates fected mice showed a significant ( p Ͻ 0.05) increase in the num- ␥ ␥ ϫ the detection limit of the IFN- ELISA. ber of IFN- -producing cells reactive with p60217–225 (115 Ϫ5 ϫ Ϫ5 ϫ Ϫ5 10 ), p60449–457 (24 10 ), LM-p60 (101 10 ), LI-p60 ϫ Ϫ5 ϫ Ϫ5 (93 10 ), or p60301–312 (28 10 ) compared with uninfected Failure of L. innocua infection to induce a p60-specific T cell control mice (Fig. 5, upper panel). The background activity was response in vivo ϳ10 ϫ 10Ϫ5. No significant increase in the frequency of IL-4- Because p60 Ag presentation by L. monocytogenes- and L. in- producing T cells was found after L. monocytogenes infection, nocua-infected cells differed, we wondered to what extent this dis- indicating that the p60-specific Th cell response was predomi- tinct preference of Ag presentation in vitro is mirrored by the fre- nantly of the Th1 type. In contrast, infection with L. innocua did quency of p60-specific CD4 and CD8 T cells in Listeria-infected not induce a significant increase in the number of p60-reactive T mice. ELISPOT assays were used to determine the frequency of cells compared with that in uninfected mice (Fig. 5, middle panel). ␥ p60-specific T cells in vivo (12). The frequency of p60217–225- The frequency of IFN- -producing p60217–225-, p60449–457-, LM- specific or p60449–457-specific CD8 T cells was calculated as the p60-, LI-p60-, or p60301–312-reactive cells was not significantly number of IFN-␥ spots generated per 1 ϫ 105 spleen cells in the ( p Ͻ 0.05) different from that in controls without addition of Ag. presence of the corresponding synthetic peptide. The frequency of The frequency of IL-4-producing cells in the presence of p60 pro- p60-specific CD4 T cells was recorded as the number of spots teins or p60 peptides varied only insignificantly, indicating that no generated by stimulation with purified LM-p60 or LI-p60. Addi- prominent p60-specific Th2 cell response was induced by L. in- tionally, the frequency of p60301–312-specific CD4 T cells was de- nocua infection. No significant p60-dependent increase in the fre- termined. All assays were performed in parallel with plates coated quency of IFN-␥- or IL-4-producing cells was observed in unin- with anti-IFN-␥ or with anti-IL4 to differentiate between p60-spe- fected mice (Fig. 5, lower panel). In summary, these experiments 4786 NONPATHOGENIC LISTERIA AND T CELL MEMORY demonstrate that L. innocua infection is insufficient to induce a p60- specific CD4 or CD8 T cell response, while L. monocytogenes infection induces a strong p60-specific CD4 and CD8 T cell response. L. innocua infection enhances the p60-specific T cell memory established by previous L. monocytogenes infection Although LI-p60 was presented by L. innocua-infected macrophages in vitro, L. innocua infection did not induce a primary p60-specific immune response. Some experimental evidence exists that primary induction of naive T cells requires a stimulus different from that required for the activation of experienced T cells (36). Therefore, it was tested whether L. innocua infection is able to boost p60-specific memory T cells established by prior L. monocytogenes infection. For these long term experiments specific pathogen-free mice were used; they were kept in a laminar air flow compartment to prevent contact with environmental micro-organisms. The primary infection was performed i.v. with a dose of either 1 ϫ 102 CFU (low dose) or 5 ϫ 103 CFU (high dose) of L. monocytogenes. In the booster groups mice were subsequently (1, Downloaded from 2, and 3 mo after L. monocytogenes infection) boosted with 1 ϫ 106 CFU of L. innocua. In the control groups booster infections were not performed. The frequency of Ag-specific memory cells was determined 4 mo after primary L. monocytogenes infection or 1 mo after the last booster infection to obtain the frequency of p60-specific T cells in the early memory phase because from the http://www.jimmunol.org/ study of L. monocytogenes infection it is known that after initial expansion of the T cell pool upon Ag contact most T cells die within 3 wk (12, 13). The frequency of Ag-specific, IFN-␥-pro- ␥ ducing CD4 and CD8 T cells was determined with the IFN- FIGURE 6. L. innocua infection boosts a previously established LM- ELISPOT assay in the presence of LM-p60, LI-p60, p60301–312, p60-specific T cell response. The frequency of Ag-specific IFN-␥-produc- p60217–225,orp60449–457 or without Ag (Fig. 6). In all groups the ing cells was determined by ELISPOT either 4 mo after primary infection background frequency of spontaneously IFN-␥-producing cells with L. monocytogenes only (filled bars) or 1 mo after primary infection measured in the absence of Ag was Ͻ0.2 ϫ 10Ϫ5 cells. Remark- with L. monocytogenes and boosting with L. innocua 1, 2, and 3 mo after ably, repeated booster infections with L. innocua significantly primary infection (open bars). A, Primary infection with a high dose (5 ϫ by guest on September 25, 2021 3 ϫ ( p Ͻ 0.05) increased the frequency of p60-specific T cells. In the 10 CFU) of L. monocytogenes. B, Primary infection with a low dose (1 2 group immunized with high dose L. monocytogenes the frequency 10 CFU) of L. monocytogenes. The frequency of cells reactive with LLO , p60 , LM-p60, or LI-p60 or without Ag is shown as the of LM-p60 and LI-p60-specific Th1 cells increased 9- and 14-fold, 91–99 217–225 number of reactive cells per 1 ϫ 105 splenocytes. The SD of triplicate respectively, and the frequency of p60217–225-specific cells in- cultures is indicated. The dotted line at 0.2 ϫ 10Ϫ5 indicates the detection creased 8-fold (Fig. 6A). In the group primed with low dose L. limit of the ELISPOT assay. Asterisks indicate a significant (p Ͻ 0.05) monocytogenes (Fig. 6B) the increases in the frequency of LM- difference between the booster group and the control group with primary p60- and LI-p60-specific Th1 cells were 13- and 15-fold, respec- infection only. tively, and the increase in the frequency of p60217–225-specific cells was 11-fold. In both groups the frequency of LLO91–99-specific cells was not significantly altered after L. innocua booster with L. innocua as described in the previous section. Mice were infection, indicating that the expansion of p60-specific T cells ob- challenged with L. monocytogenes 4 mo after primary infection. In tained by L. innocua infection was Ag specific. If the primary the L. innocua booster groups this was 1 mo after the last booster induction of p60-specific T cells by primary L. monocytogenes infection. After primary immunization by high dose L. monocyto- infection was omitted, and primary immunization was instead per- genes infection excellent protection was observed in the spleen formed by L. innocua infection a p60- or LLO-specific T cell re- (protection, Ͼ6) and in the liver (protection, Ͼ6) upon L. mono- sponse was not observed (data not shown). Thus, although L. in- cytogenes challenge (Fig. 7). This protection was not further im- nocua infection is insufficient to induce a primary p60-specific T proved by L. innocua booster infections. After primary immuni- cell response in vivo, this nonpathogenic Listeria species is able to zation by low dose L. monocytogenes infection the protection upon expand p60-specific CD4 and CD8 memory T cells induced by L. monocytogenes challenge 4 mo postimmunization was less com- previous L. monocytogenes infection. plete in the spleen (protection, 3.8) and liver (protection, 2.9). L. innocua booster infections significantly ( p Ͻ 0.05) improved the Limited protective efficacy of L. innocua infection protection in the liver (protection, 3.9), while protection in the L. innocua is an environmental bacterium that is frequently iso- spleen (protection, 4.0) was not significantly influenced. No in- lated from food (5–8). Frequent encounters with L. innocua could creased protection was observed if mice were boosted with B. sub- stimulate L. monocytogenes cross-reactive memory T cells and tilis instead of L. innocua, indicating that the increased protection play a potential role in the maintenance of protective antilisterial observed in the liver after L. innocua booster infection was not the immunity. To test whether infection with L. innocua improves the result of an unspecific stimulation (Fig. 7). In contrast, primary L. protective immune response against L. monocytogenes infection, innocua infection boosted by subsequent L. innocua infection after mice were immunized by primary low dose or high dose L. mono- 1, 2, and 3 mo did not confer significant protection against chal- cytogenes infection and were subsequently boosted repeatedly lenge with L. monocytogenes. Together, these results demonstrate The Journal of Immunology 4787

unless accompanied by specific Ag (38–40). This observation has been challenged by the finding that after adoptive transfer of memory CD8 T cells these cells persist in the complete absence of specific Ag (41–44). Several models were proposed to explain the maintenance of T cell memory in the absence of specific Ag. It was proposed that chronic stimulation of memory T cells could reflect repeated contact with cross-reactive environmental Ags (37). This speculation was supported by the idea that memory T cells are hyper-reactive to Ag (43, 45). In one study memory CD8 T cells responded to a peptide at a 10- to 50-fold lower concentration than that required for the stimulation of naive CD8 T cells (45). Alter- native models, which are supported by experimental data mostly from viral infection systems, suggest that without specific TCR- mediated antigenic stimulation, memory T cells might participate as bystander cells from cytokines produced during the T cell response directed against an unrelated micro-organism (46, 47). In our experimental system the expansion of memory CD8 T cells was exclusively Ag specific. The frequency of CD8 T cells specific for the unrelated T cell Ag LLO was not altered by L. innocua Downloaded from booster infection. Thus, it has to be assumed that p60-specific memory T cells were specifically activated by contact with a related cross-reactive Ag. This observation corroborates the recent reports from viral infection models that most T cells that are activated during infection are strictly Ag specific and are not ex- FIGURE 7. Limited protective efficacy of L. innocua infection. Mice panded during infection with a heterologous virus (48, 49). http://www.jimmunol.org/ were immunized by primary infection with a low dose (LM-low dose) or The virulence-associated proteins of L. monocytogenes, such as a high dose (LM-high dose) inoculum of L. monocytogenes or by L. innocua (LI) infection. Subsequently, booster groups were repeatedly LLO and the metalloprotease, which are targets for L. monocyto- boosted, as indicated, by L. innocua or B. subtilis infection 1, 2, and 3 mo genes-specific T cells (17, 21, 50), are not shared with nonpatho- after primary infection. One month after the last booster infection mice genic Listeria (9). Because p60 is essential for cell viability, it is were challenged with L. monocytogenes. The bacterial loads of spleen and not surprising that a p60 homologue is also produced by nonpatho- liver were determined 72 h postinfection. Results are expressed as the mean genic Listeria species, including L. innocua (23). The LM-p60 Ϯ log10 CFU SD of groups of five mice. A significant increase in protec- from pathogenic L. monocytogenes and the LI-p60 from nonpatho- tion after booster infection is indicated by an asterisk. Primary infection genic L. innocua display about 90% homology (25). In addition to with L. monocytogenes and subsequent booster infection with L. innocua the MHC class II epitope, both known MHC class I epitopes from by guest on September 25, 2021 are indicated as LM-high dose/LI-boost if the primary inoculum was high LM-p60, p60 and p60 , are conserved in both species dose L. monocytogenes and as LM-low dose/LI-boost if the primary inoc- 217–225 449–457 (25) and are presented by L. innocua-infected macrophage-like ulum was low dose L. monocytogenes. Primary infection with low dose L. monocytogenes and booster infection with B. subtilis is indicated as LM- P388D1 cells. The MHC class II epitope p60301–312 recognized by low dose/BS-boost. Primary and booster infection with L. innocua is in- the CD4 T cell clone 1A is conserved in L. innocua, with the dicated as LI/LI-boost. Control mice were not infected. exception of position 1, which is a glutamine residue in the LI-p60 and a glutamic acid residue in LM-p60. The MHC class I-restricted Ag presentation of bacterial proteins by cells infected with L. innocua (51) or with LLO-negative L. monocytogenes strains (52) that repeated L. innocua infections fail to induce protective immu- has been reported previously. However, no direct comparison of nity against challenge with L. monocytogenes. However, L. in- the relative strength of MHC class I and MHC class II-restricted nocua booster infections induced additional protection in the liver presentation was performed. Although our experiments demon- upon challenge with L. monocytogenes in the situation of subop- strated MHC class I-restricted p60 presentation independent of timal antilisterial immunity. LLO, the comparison with LLO-secreting L. monocytogenes clearly corroborates the importance of LLO for efficient MHC Discussion class I-restricted Ag presentation. This observation strengthens the The present study investigated the influence of L. innocua, a non- results of a detailed study of the Ag presentation of LLO with pathogenic environmental Listeria species, on the induction and LLO-specific MHC class I and MHC class II-restricted T cell hy- maintenance of an L. monocytogenes-specific immune response. bridomas (29), which showed that MHC class I-restricted presen- We show that although L. innocua infection itself fails to induce a tation of LLO depends on the hemolytic activity of LLO. In con- primary L. monocytogenes-specific T cell response, infection with trast to this study we investigated the influence of secreted LLO on this nonpathogenic Listeria species has the potential to enhance a the Ag presentation of an independent bacterial protein. The com- previously established L. monocytogenes-specific T cell memory. parison of p60 presentation by cells infected with LLO-secreting L. This is the first experimental support for the hypothesis that cross- monocytogenes with that by cells infected with LLO-negative L. reactive nonpathogenic micro-organisms influence the mainte- monocytogenes or by L. innocua provided clear evidence that bac- nance of T cell memory (14, 15, 37). terial secretion of LLO strongly enhances the MHC class I-re- T cell memory is a hallmark of the immune system, and ever stricted presentation and inhibits the MHC class II-restricted pre- since its recognition there has been considerable interest in under- sentation of a listerial protein in infected cells. standing how it is maintained. The idea that maintenance of long- Although strong inhibition of MHC class II-restricted p60 Ag lived T cell memory requires chronic exposure to Ag came from presentation occurred in L. monocytogenes-infected cells in vitro, reports that memory T cells survive poorly on adoptive transfer a high frequency of p60-specific CD4 T cells was observed in vivo. 4788 NONPATHOGENIC LISTERIA AND T CELL MEMORY

The frequency of cells specific for the CD4 T cell epitope the stimulation of L. monocytogenes-specific CD4 and CD8 mem- p60301–312 was similar to the frequency of cells specific for the ory T cells. A general model of the role of environmental micro- CD8 T cell epitope p60449–457, which is subdominant relative to organisms for the maintenance of T cell memory would suggest the p60217–225 epitope (12). The response to p60301–312 repre- that specific T cells established by primary infection with a patho- sented Ͻ10% of the total p60-specific CD4 T cell response. The genic micro-organism can be boosted by contact with any anti- frequency of T cells specific for the whole p60 was similar to the genically related nonpathogenic species. frequency of T cells specific for the immunodominant epitope p60217–225. A possible explanation for the efficient MHC class II- Acknowledgments restricted in vivo presentation of p60 is the MHC class II-restricted presentation after secondary uptake of bacterial protein or killed We thank R. Holtappels (University of Mainz, Mainz, Germany) and J. Daniels (University of Wu¨rzburg, Wu¨rzburg, Germany) for critical review bacteria released from dead cells by neighboring noninfected cells. of the manuscript, and D. Palm (University of Wu¨rzburg) for his support This mechanism enables the MHC class II-restricted presentation with generation of the peptide library. of LLO in cultures of infected macrophages (29). In contrast to the strong p60-specific T cell response after L. monocytogenes infection, no p60-specific CD4 or CD8 T cell response was observed References after primary L. innocua infection, although we found that p60 is 1. Farber, J. M., and P. I. Peterkin. 1991. Listeria monocytogenes, a food-borne pathogen. Microbiol. Rev. 55:476. presented by L. innocua-infected macrophages. One plausible ex- 2. Seeliger, H. P. 1984. Modern taxonomy of the Listeria group relationship to its planation for this failure might be the rapid clearance of L. innocua pathogenicity. Clin. Invest. Med. 7:217. from spleen and liver after infection (53–55). Similarly, it has been 3. Sallen, B., A. Rajoharison, S. Desvarenne, F. Quinn, and C. Mabilat. 1996. Com- Downloaded from parative analysis of 16S and 23S rRNA sequences of Listeria species. Int. J. Syst. reported that antibiotic abridgement of L. monocytogenes infection Bacteriol. 46:669. by ampicillin treatment started as late as 5 days postinfection 4. Muller, H. E. 1990. Listeria isolations from feces of patients with diarrhea and causes a significant reduction of protection against a secondary from healthy food handlers. Infection 18:97. 5. Harvey, J., and A. Gilmour. 1993. Occurrence and characteristics of Listeria in challenge infection (56). However, the activation of experienced foods produced in Northern Ireland. Int. J. Food Microbiol. 19:193. p60-specific CD4 and CD8 memory T cells by L. innocua showed 6. Pinto, B., and D. Reali. 1996. Prevalence of Listeria monocytogenes and other that in vivo the p60 of this species was indeed presented by MHC listerias in Italian-made soft cheeses. Zentralbl. Hyg. Umweltmed. 199:60. http://www.jimmunol.org/ 7. Wilson, I. G. 1996. Occurrence of Listeria species in prepacked retail sand- class I and MHC class II molecules, as predicted from the in vitro wiches. Epidemiol. Infect. 117:89. Ag presentation experiments. Generally, this observation suggests 8. Wilson, I. G. 1995. Occurrence of Listeria species in ready to eat foods. Epide- that different stimuli are required for the induction of a primary vs miol. Infect. 115:519. 9. Portnoy, D. A., T. Chakraborty, W. Goebel, and P. Cossart. 1992. Molecular a recall T cell response similar to the situation observed with some determinants of Listeria monocytogenes pathogenesis. Infect. Immun. 60:1263. nonimmunogenic tumor cell lines, which are only able to induce a 10. Gaillard, J. L., P. Berche, C. Frehel, E. Gouin, and P. Cossart. 1991. Entry of L. recall T cell response in previously immunized mice (36). The monocytogenes into cells is mediated by internalin, a repeat protein reminiscent of surface antigens from Gram-positive cocci. Cell 65:1127. reason for the different immunogenicities of L. monocytogenes and 11. Kaufmann, S. H. E. 1993. Immunity to intracellular bacteria. Annu. Rev. Immu- L. innocua is unknown. Recently, it was reported that dead bacteria nol. 11:129. suppress IL-12 production in contrast to live bacteria, which in- 12. Vijh, S., and E. G. Pamer. 1997. Immunodominant and subdominant CTL re- by guest on September 25, 2021 sponses to Listeria monocytogenes infection. J. Immunol. 158:3366. duce IL-12 in vivo (57). Because IL-12 plays an important role in 13. Busch, D. H., I. M. Pilip, S. Vijh, and E. G. Pamer. 1998. Coordinate regulation the induction of a cellular immune response (58), the suppressive of complex T cell populations responding to bacterial infection. Immunity 8:353. effect of killed L. innocua could prevent the induction of a T cell 14. Sprent, J. 1997. Immunological memory. Curr. Opin. Immunol. 9:371. 15. Zinkernagel, R. M., M. F. Bachmann, T. M. Ku¨ndig, S. Oehen, H. Pirchet, and immune response. H. Hengartner. 1996. On immunological memory. Annu. Rev. Immunol. 14:333. Human infection by L. monocytogenes is a food-borne infection 16. Ahmed, R., and D. Gray. 1996. Immunological memory and protective immunity: (5–8). Using the ligated ileal loop model of the rat it was recently understanding their relation. Science 272:54. 17. Safley, S. A., C. W. Cluff, N. E. Marshall, and H. K. Ziegler. 1991. Role of shown that L. monocytogenes is taken up in vivo by epithelium listeriolysin-O (LLO) in the T lymphocyte response to infection with Listeria rather nonspecifically, i.e., no specific virulence factors of L. monocytogenes: identification of T cell epitopes of LLO. J. Immunol. 146:3604. 18. Pamer, E. G. 1994. Direct sequence identification and kinetic analysis of an MHC monocytogenes seem to be required for this step, and even non- class I-restricted Listeria monocytogenes CTL epitope. J. Immunol. 152:686. pathogenic L. innocua are translocated at a similar rate as the L. 19. Pamer, E. G., J. T. Harty, and M. J. Bevan. 1991. Precise prediction of a dominant monocytogenes wild-type strain (59). Remarkably, the screening class I MHC-restricted epitope of Listeria monocytogenes. Nature 353:852. 20. Geginat, G., M. Lalic, M. Kretschmar, W. Goebel, H. Hof, D. Palm, and of sera from healthy volunteers for the occurrence of p60-specific A. Bubert. 1998. Th1 cells specific for a secreted protein of Listeria monocyto- Abs revealed a high prevalence of exclusively LI-p60-specific genes are protective in vivo. J. Immunol. 160:6046. ϩ Abs, indicating that nonpathogenic L. innocua from food are fre- 21. Harty, J. T., and M. J. Bevan. 1992. CD8 T cells specific for a single nonamer epitope of Listeria monocytogenes are protective in vivo. J. Exp. Med. 175:1531. quently exposed to the immune system of normal hosts (M. Lalic 22. Harty, J. T., and E. G. Pamer. 1995. CD8 T lymphocytes specific for the secreted and A. Bubert, unpublished observations). In contrast to this nat- p60 antigen protect against Listeria monocytogenes infection. J. Immunol. 154: ural mode of infection we studied L. innocua in a systemic infec- 4642. 23. Wuenscher, M. D., S. Ko¨hler, A. Bubert, U. Gerike, and W. Goebel. 1993. The tion model, because this is an established model for the study of iap gene of Listeria monocytogenes is essential for cell viability, and its gene the antilisterial T cell response. In our protection assays it was product, p60, has bacteriolytic activity. J. Bacteriol. 175:3491. difficult to obtain additional protection after booster immunization 24. Kuhn, M., and W. Goebel. 1989. Identification of an extracellular protein of Listeria monocytogenes possibly involved in intracellular uptake by mammalian with nonpathogenic Listeria, because even primary immunization cells. Infect. Immun. 57:55. with low doses of L. monocytogenes induced strong protection 25. Bubert, A., M. Kuhn, W. Goebel, and S. Ko¨hler. 1992. Structural and functional properties of the p60 proteins from different Listeria species. J. Bacteriol. 174: against reinfection. Upon challenge with L. monocytogenes 4mo 8166. postimmunization, protection in the spleen was at least 10-fold 26. Hess, J., I. Gentschev, G. Szalay, C. Ladel, A. Bubert, W. Goebel, and better than that in the liver. Thus, it is not unexpected that L. S. H. E. Kaufmann. 1995. Listeria monocytogenes p60 supports host cell invasion by and in vivo survival of attenuated Salmonella typhimurium. Infect. Immun. innocua-mediated enhancement of protection against L. monocy- 63:2047. togenes was limited to the situation of weak antilisterial immunity, 27. Ko¨hler, S., M. Leimeister-Wa¨chter, T. Chakraborty, F. Lottspeich, and as observed in the liver after low dose primary immunization. In W. Goebel. 1990. The gene coding for protein p60 of Listeria monocytogenes and its use as a specific probe for Listeria monocytogenes. Infect. Immun. 58:1943. summary, we conclude that after L. innocua booster infection 28. Azuma, M., H. Yssel, J. H. Phillips, H. Spits, and L. L. Lanier. 1993. Functional weak Ag presentation of a cross-reactive protein is sufficient for expression of B7/BB1 on activated T lymphocytes. J. Exp. Med. 177:845. The Journal of Immunology 4789

29. Hiltbold, E. M., S. A. Safley, and H. K. Ziegler. 1996. The presentation of class 45. Pihlgren, M., P. M. Dubois, M. Tomkowiak, T. Sjogren, and J. Marvel. 1996. I and class II epitopes of listeriolysin O is regulated by intracellular localization Resting memory CD8ϩ T cells are hyperreactive to antigenic challenge in vitro. and by intercellular spread of Listeria monocytogenes. J. Immunol. 157:1163. J. Exp. Med. 184:2141. 30. Miyahira, Y., K. Murata, D. Rodriguez, J. R. Rodriguez, M. Esteban, 46. Ehl, S., J. Hombach, P. Aichele, H. Hengartner, and R. M. Zinkernagel. 1997. ϩ M. M. Rodrigues, and F. Zavala. 1995. Quantification of antigen specific CD8 Bystander activation of cytotoxic T cells: studies on the mechanism and evalu- T cells using an ELISPOT assay. J. Immunol. Methods 181:45. ation of in vivo significance in a transgenic mouse model. J. Exp. Med. 185:1241. 31. Germain, R. N. 1994. MHC-dependent antigen processing and peptide presenta- 47. Tough, D. F., P. Borrow, and J. Sprent. 1996. Induction of bystander T cell tion: providing ligands for T cell activation. Cell 76:287. proliferation by viruses and type I interferon in vivo. Science 272:1947. 32. Portnoy, D. A. 1994. Cellular biology of Listeria monocytogenes infection. In 48. Murali-Krishna, K., J. D. Altman, M. Suresh, D. J. Sourdive, A. J. Zajac, Molecular Genetics of Bacterial Pathogenesis. V. L. Miller, J. B. Kaper, J. D. Miller, J. Slansky, and R. Ahmed. 1998. Counting antigen-specific CD8 T D. A. Portnoy, and R. R. Isberg, eds. ASM, Washington, D.C., p. 279. cells: a reevaluation of bystander activation during viral infection. Immunity 33. Sijts, A. J., A. Neisig, J. Neefjes, and E. G. Pamer. 1996. Two Listeria mono- 8:177. cytogenes CTL epitopes are processed from the same antigen with different ef- 49. Butz, E. A., and M. J. Bevan. 1998. Massive expansion of antigen-specific CD8ϩ ficiencies. J. Immunol. 156:683. T cells during an acute virus infection. Immunity 8:167. 34. Domann, E., J. Wehland, M. Rohde, S. Pistor, M. Hartl, W. Goebel, M. Lei- 50. Busch, D. H., H. G. Bouwer, D. Hinrichs, and E. G. Pamer. 1997. A nonamer meister-Wa¨chter, M. Wuenscher, and T. Chakraborty. 1992. A novel bacterial peptide derived from Listeria monocytogenes metalloprotease is presented to cy- virulence gene in Listeria monocytogenes required for host cell microfilament tolytic T lymphocytes. Infect. Immun. 65:5326. interaction with homology to the proline-rich region of vinculin. EMBO J. 11: 51. Zwickey, H. L., and T. A. Potter. 1996. Peptide epitopes from noncytosolic Lis- 1981. teria monocytogenes can be presented by major histocompatibility complex class 35. Mosmann, T. R., and R. L. Coffman. 1989. Th1 and Th2 cells: different patterns I molecules. Infect. Immun. 64:1870. of lymphokine secretion lead to different functional properties. Annu. Rev. Im- 52. Szalay, G., J. Hess, and S. H. E. Kaufmann. 1994. Presentation of Listeria mono- munol. 7:145. cytogenes antigens by major histocompatibility complex class I molecules to CD8 36. Ku¨ndig, T. M., M. F. Bachmann, L. Lefrancois, L. Puddington, H. Hengartner, cytotoxic T lymphocytes independent of listeriolysin secretion and virulence. and R. M. Zinkernagel. 1993. Nonimmunogenic tumor cells may efficiently re- Eur. J. Immunol. 24:1471. stimulate tumor antigen- specific cytotoxic T cells. J. Immunol. 150:4450. 37. Beverley, P. C. 1990. Is T-cell memory maintained by crossreactive stimulation? 53. Hof, H., and P. Hefner. 1988. Pathogenicity of Listeria monocytogenes in com- Downloaded from Immunol. Today 11:203. parison to other Listeria species. Infection 16(Suppl. 2):S141. 38. Gray, D., and P. Matzinger. 1991. T cell memory is short-lived in the absence of 54. von Koenig, C. H., B. Heymer, H. Hof, and H. Finger. 1983. Course of infection antigen. J. Exp. Med. 174:969. and development of immunity in experimental infection of mice with Listeria 39. Oehen, S., H. Waldner, T. M. Ku¨ndig, H. Hengartner, and R. M. Zinkernagel. serotypes. Infect. Immun. 40:1170. 1992. Antivirally protective cytotoxic T cell memory to lymphocytic choriomen- 55. Barbour, A. H., A. Rampling, and C. E. Hormaeche. 1996. Comparison of the ingitis virus is governed by persisting antigen. J. Exp. Med. 176:1273. infectivity of isolates of Listeria monocytogenes following intragastric and intra- 40. Ku¨ndig, T. M., M. F. Bachmann, S. Oehen, U. W. Hoffmann, J. J. Simard, venous inoculation in mice. Microb. Pathog. 20:247.

C. P. Kalberer, H. Pircher, P. S. Ohashi, H. Hengartner, and R. M. Zinkernagel. 56. North, R. J., P. A. Berche, and M. F. Newborg. 1981. Immunologic consequences http://www.jimmunol.org/ 1996. On the role of antigen in maintaining cytotoxic T-cell memory. Proc. Natl. of antibiotic-induced abridgement of bacterial infection: effect on generation and Acad. Sci. USA 93:9716. loss of protective T cells and level of immunologic memory. J. Immunol. 127: 41. Lau, L. L., B. D. Jamieson, T. Somasundaram, and R. Ahmed. 1994. Cytotoxic 342. T-cell memory without antigen. Nature 369:648. 57. Zhan, Y., and C. Cheers. 1998. Control of IL-12 and IFN-␥ production in re- 42. Mullbacher, A. 1994. The long-term maintenance of cytotoxic T cell memory sponse to life or dead bacteria by TNF and other factors. J. Immunol. 161:1447. does not require persistence of antigen. J. Exp. Med. 179:317. 58. O’Garra, A. 1998. Cytokines induce the development of functionally heteroge- 43. Bruno, L., J. Kirberg, and H. von Boehmer. 1995. On the cellular basis of im- neous T helper cell subsets. Immunity 8:275. munological T cell memory. Immunity 2:37. 59. Pron, B., C. Boumaila, F. Jaubert, S. Sarnacki, J. P. Monnet, P. Berche, and 44. Hou, S., L. Hyland, K. W. Ryan, A. Portner, and P. C. Doherty. 1994. Virus- J. L. Gaillard. 1998. Comprehensive study of the intestinal stage of listeriosis in speciﬁc CD8ϩ T-cell memory determined by clonal burst size. Nature 369:652. a rat ligated ileal loop system. Infect. Immun. 66:747. by guest on September 25, 2021