Epitopes in Vivo -Derived CD8 T Cell Listeria Monocytogenes Dynamic

Dynamic Antigen Presentation Patterns of Listeria monocytogenes-Derived CD8 T Cell Epitopes In Vivo

This information is current as Mojca Skoberne, Rafaela Holtappels, Herbert Hof and of September 28, 2021. Gernot Geginat J Immunol 2001; 167:2209-2218; ; doi: 10.4049/jimmunol.167.4.2209 http://www.jimmunol.org/content/167/4/2209 Downloaded from

References This article cites 41 articles, 25 of which you can access for free at: http://www.jimmunol.org/content/167/4/2209.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 28, 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 © 2001 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Dynamic Antigen Presentation Patterns of Listeria monocytogenes-Derived CD8 T Cell Epitopes In Vivo1

Mojca Sùkoberne,* Rafaela Holtappels,† Herbert Hof,* and Gernot Geginat2*

Little information exists regarding the presentation of antigenic peptides in infected tissues. In this study the in vivo presentation of four different CD8 T cell epitopes of Listeria monocytogenes was monitored. Peptide presentation was measured by a new, highly sensitive, ex vivo Ag presentation assay that was based on the testing of freshly isolated cells from infected spleens with peptide- specific CD8 T cell lines in an IFN-␥-specific ELISPOT assay. Remarkably, the peptide presentation pattern of splenocytes and that of macrophages purified from spleens of L. monocytogenes-infected mice were different from those of in vitro infected macrophage-like cell lines. The in vivo Ag presentation pattern of splenocytes also exhibited dynamic changes during the first 48 h of infection. In vivo peptide presentation at later time points postinfection was biased toward immunodominant CD8 T cell epitopes, while at an early time point, 6 h postinfection, subdominant and dominant CD8 T cell epitopes were presented with Downloaded from similar strength. In summary, our studies show that Ag presentation during an infection is a highly dynamic process that only can be fully appreciated by the study of cells infected in their physiological environment. The Journal of Immunology, 2001, 167: 2209Ð2218.

he host response against infection with a complex micro- reticulum (6) or the modulation of the immunogenicity of T cell

organism comprises T cells specific for a multitude of Ag by variations in the sequences flanking a T cell epitope (7, 8). http://www.jimmunol.org/ T different antigenic peptides. Generally, the magnitude of The murine infection with Listeria monocytogenes is one of the the T cell response against different antigenic peptides exhibits a infection models where the mechanisms governing CD8 T cell remarkably constant hierarchy, and the majority of the responding induction and expansion were studied in detail. Mice infected with T cells are directed against few immunodominant T cell epitopes L. monocytogenes mount MHC class I Kd-restricted CD8 T cell (1). Numerous studies in different infectious disease model sys- responses against peptides encompassing aa 91Ð99 of listeriolysin tems revealed that the strength of the CD8 T cell response against O (LLO)3 (4); aa 217Ð225 (9), 449Ð457 (10), and 476Ð484 (11) a peptide is the result of the complex interplay of three major of the p60 protein; and aa 84Ð92 of the listerial metalloprotease factors: the quantity and stability of peptide MHC class I/com- (12), respectively. In vivo the majority of CD8 T cells are specific by guest on September 28, 2021 plexes expressed on APC, the TCR repertoire of the responding T for the immunodominant epitopes LLO91Ð99 and p60217Ð225, while cell population, and the suppression of T cells specific for sub- relatively few T cells are directed against the subdominant dominant epitopes by T cells specific for immunodominant epitopes p60449Ð457 and Mpl84Ð92 (13). The frequency of epitopes (reviewed in Ref. 2). Ag presentation is certainly required p60476Ð484-specific CD8 T cells is intermediate between the fre- for the induction and expansion of CD8 T cells. However, only a quency of p60217Ð225 and that of p60449Ð457-specific T cells (11). few studies exist about the processing and presentation of anti- Remarkably, among these four L. monocytogenes-derived peptides genic peptides in vivo. In principle, the extraction and quantifica- the immunodominant LLO91Ð99 is the least abundant endog- tion of naturally processed antigenic peptides allow the direct anal- enously processed peptide in infected cell lines, while the sub- ysis of Ag processing in tissues (3). Due to the relatively large dominant p60449Ð457 is the most abundant antigenic peptide in number of infected cells required for this method the extraction of infected cell lines (14). Thus, a paradoxical inverse correlation antigenic peptides from organs was only successful in a few model exists between the abundance of naturally processed antigenic pep- infections (4, 5). More indirectly, bacteria and viruses that express tides in infected cells and the frequency of peptide-specific CD8 T genetically engineered proteins were used to analyze the effects of cells in vivo. variations in Ag presentation on the strength of the CD8 T cell It must be kept in mind that the quantitative analysis of peptide response in vivo. Examples are the enhanced immunogenicity of processing is based on in vitro infected cells and that it is not preprocessed Ags that are directly targeted into the endoplasmic known to what extent this in vitro model represents the in vivo situation. Therefore, in the current study the presentation of L.

*Institut fu¬r Medizinische Mikrobiologie und Hygiene, Fakulta¬t fu¬r Klinische Medi- monocytogenes-derived antigenic peptides was monitored in vivo. zin Mannheim der Universita¬t Heidelberg, Mannheim, Germany; and †Institut fu¬r We used a novel approach for the direct measurement of Ag Virologie, Universita¬t Mainz, Mainz, Germany presentation in tissues that is based on the testing of in vivo-in- Received for publication February 15, 2001. Accepted for publication June 5, 2001. fected cells with peptide-specific CD8 T cell lines in a sensitive The costs of publication of this article were defrayed in part by the payment of page ELISPOT assay. Remarkably, the peptide presentation pattern of charges. This article must therefore be hereby marked advertisement in accordance splenocytes infected with L. monocytogenes in vivo exhibited dy- with 18 U.S.C. Section 1734 solely to indicate this fact. namic changes during the first 48 h of infection. In light of this new 1 This work was supported by Deutsche Forschungsgemeinschaft Grant GE 1081/1-1. M.S. was supported by the Slovenian Ministry of Science and the Medical Faculty of finding the possible correlation between peptide presentation and the University of Ljubljana. in vivo CD8 T cell expansion and function was reevaluated. 2 Address correspondence and reprint requests to Dr. Gernot Geginat, Institut fu¬r Medizinische Mikrobiologie und Hygiene, Fakulta¬t fu¬r Klinische Medizin Mannheim der Universita¬t Heidelberg, Theodor Kutzer Ufer 1-3, 68167 Mannheim, Germany. 3 Abbreviations used in this paper: LLO, listeriolysin O; AcN, acetonitrile; TFA, E-mail address: [email protected] trifluoroacetic acid; p.i., postinfection.

Materials and Methods was 0Ð5 min of 0% B, 5Ð55 min linear increase to 50% B, 55Ð63 min Mice and infection of mice linear increase to 100% B, 63Ð66 min of 100% B, and 66Ð74 min linear decrease to 0% B. One-minute fractions were collected and stored at Female BALB/cOlaHsd (H-2d) mice were purchased (Harlan-Winkel- Ϫ70¡C. mann, Borchen, Germany), kept under conventional conditions and used at Isolation of naturally processed peptides from infected cell lines was 8Ð10 wk of age. Mice were infected with L. monocytogenes serovar 1/2a performed similarly. Six hours postinfection (p.i.) ϳ1 ϫ 108 adherent L. EGD in 0.2 ml PBS either i.v. or i.p. as indicated. Infectious doses were monocytogenes-infected P388 or J774 cells were washed twice with ice- 1 ϫ 106 and 1 ϫ 103 CFU i.v. for ex vivo peptide presentation experiments cold PBS and harvested with a cell scraper. Cell pellets were disrupted, and T cell induction studies, respectively. Bacteria used for infection were sonicated, and lysed for 30 min in 2 ml 0.5% TFA supplemented with in the logarithmic growth phase. The bacterial concentration was estimated proteinase inhibitors as described above. Subsequently, cell lysates were from the OD at 600 nm. centrifuged 30 min at 20,000 ϫ g, and supernatants were further purified by ultrafiltration using MICROSEP (Pall-Gelman, Dreieich, Germany) ul- APC, in vitro infection of APC, and CD8 T cell lines trafiltration units with a 10-kDa cutoff. The low Mr fraction was further P815 mastocytoma cells were used as targets in the cell-mediated cytotox- fractionated by HPLC as described above.

icity assay. For in vitro infection experiments macrophage-like J774A1 ϫ 8 (J774) and P388D1 (P388) cells were used as APC. Approximately 1 10 P388 or J774 cells were infected with L. monocytogenes at a multiplicity Quantification of endogenously processed peptides of infection of 10. After1hat37¡C, cells were washed once, and the For the quantification of endogenously processed antigenic peptides HPLC ␮ medium was exchanged with medium containing 5 g/ml gentamicin. Af- fractions were dried by vacuum concentration and resolved in 1 ml cell ter an additional 5-h incubation at 37¡C, cells were either detached by culture medium. The precise amount of antigenic peptides in HPLC frac- trypsin treatment and tested in the ELISPOT-based Ag presentation assay tions was determined as described previously (5). Fractions were tested in 51 or, alternatively for peptide extraction experiments, were harvested with a a standard chromium release assay with Cr-labeled P815 cells as APC. Downloaded from cell scraper without addition of trypsin. The peptide concentration of the fractions was calculated by linear inter- CD8 T cell lines specific for p60217Ð225, p60449Ð457, p60476Ð484, and polation from lysis data obtained with a synthetic peptide standard. The LLO were derived from the spleens of L. monocytogenes-infected 91Ð99 recoveries of p60217Ð225, p60449Ð457, p60476Ð484, and LLO91Ð99 were de- BALB/c mice. CD8 T cell lines were propagated by repeated restimulation termined in a preliminary experiment after admixture of synthetic peptides with P815 cells transfected with the human B7.1 gene (P815/B7) (15) in to mock extracts of cell lines or spleens. Recovery from spleen extracts was the presence of the appropriate synthetic peptide in medium supplemented ϳ 20% for p60217Ð225 and p60476Ð484, 10% for p60449Ð457, and 60% for with IL-2 as described previously (16). Synthetic peptides were purchased ϳ LLO91Ð99. Recovery from cell extracts was 20% for all three p60 pep- (Jerini Biotools, Berlin, Germany). The peptide concentration used for re- tides and 70% for LLO . In the calculation of the total number of http://www.jimmunol.org/ Ϫ10 91Ð99 stimulation was 10 M for all CD8 T cell lines. The detection limit of the peptides per organ or per cell the different recovery rates were counted. CD8 T cell lines used was between 10Ϫ11 and 10Ϫ12 M peptide as measured using a standard chromium release assay. Immunomagnetic isolation of macrophages from infected mice ELISPOT-based Ag presentation assay Macrophages from spleens of L. monocytogenes-infected mice were iso- Ag presentation by in vivo-infected splenocytes was assessed with an lated by immunomagnetic cell sorting using paramagnetic microbeads con- ELISPOT-based Ag presentation assay. This assay applies the basic prin- jugated to monoclonal hamster anti-mouse CD11b (clone M1/70.15.11.5) ciple of the ELISPOT assay for the detection of Ag presentation by target Abs (Miltenyi, Bergisch Gladbach, Germany). Spleens were removed 48 h cells infected in vitro or in vivo. Spleens were removed between 6 and 48 h 5 after infection of mice. Splenocytes were used as APC after passing after i.v. infection of mice with 1 ϫ 10 CFU L. monocytogenes. Spleens by guest on September 28, 2021 were injected with 500 ␮l of a 1 mg/ml solution of collagenase D (Roche through nylon gaze (80 mesh) and RBC lysis. Alternatively, in some ex- Diagnostics, Mannheim, Germany) in HBSS. Subsequently, spleens were periments macrophages isolated from infected spleens or in vitro infected P388 and J774 cells were also used as APC. P388 and J774 cells were cut in small pieces and incubated for 30 min at 37¡Cin5%CO2 in the collagenase D buffer. Cells were collected by centrifugation and subse- infected as described above. The setting of the assay was similar for these ϩ different APC types. All APC were tested in the presence of 10 ␮g/ml quently separated twice on MS selection columns (Miltenyi) following ␮ the standard positive selection protocol provided by the manufacturer. For gentamicin and 20 g/ml tetracycline. In round-bottom 96-well microtiter plates, 3 ϫ 104 peptide-specific CD8 T cells/well were added to graded each experiment spleens from three mice were pooled. At the end of the ␮ positive selection procedure, between 0.5 ϫ 106 and 1 ϫ 106 cells were numbers of APC in a final volume of 150 l. Plates were subsequently obtained per 1 ϫ 108 spleen cells. Aliquots of the selected cells were incubated for5hat37¡Cin5%CO2. This preincubation step in round- stained with FITC-labeled rat anti-mouse F4/80 IgG2b (clone CI:A3-1; bottom plates was required for optimal contact of APC and responder CD8 Serotec, Eching, Germany), rat anti-mouse CD11b IgG2b mAb (clone, T cells and resulted in a significantly better T cell activation compared with 5C6; Roche Diagnostics), rat Ig2b isotype control mAb (Serotec), and tests that were performed directly in flat-bottom ELISPOT plates (data not shown). After the preincubation cells were resuspended, and 100 ␮l cell hamster anti-mouse CD11c IgG mAb (clone HL3; PharMingen, San Diego, ␥ CA), respectively, and subjected to FACS analysis. The remaining cells suspension were transferred to rat anti-mouse IFN- mAb-coated were tested in the ELISPOT-based Ag presentation assay described below. (RMMG-1; BioSource International, Camarillo, CA) nylon membrane- backed 96-well microtiter plates (Nunc, Wiesbaden, Germany) and incu- Isolation of endogenously processed peptides bated overnight. During the primary incubation step Ͻ20% of infected P388 and J774 cells attached; thus, the majority of APC were transferred Peptide extraction from infected cells and organs was performed as de- to the ELISPOT plates together with CD8 T cells. ELISPOT plates were scribed previously with minor modifications (5). Spleens were removed developed with biotin-labeled rat anti-mouse IFN-␥ mAb (clone XMG1.2; ϫ 6 48 h after i.v. infection with 1 10 CFU L. monocytogenes. Organs from BD Biosciences, Heidelberg, Germany), HRP-streptavidin conjugate (Di- five mice were pooled and passed through a steel mesh. After removal of anova, Hamburg, Germany), and aminoethylcarbazole dye solution. an aliquot for plating of bacteria, trifluoroacetic acid (TFA) was added to achieve a pH of 2.0. The lysis solution was supplemented with COM- PLETE proteinase inhibitor (final concentration, 1 tablet/50 ml lysis buffer) Ex vivo enumeration of peptide-specific CD8 T cells and 1 ␮g/ml pepstatin (both from Roche Diagnostics). After homogeniza- tion extracts were sonicated, left for 30 min on ice, and centrifuged for 1 h The frequency of peptide-specific CD8 T lymphocytes was determined in at 50,000 ϫ g. Supernatants were removed and passed through a Sephadex an IFN-␥-specific ELISPOT assay 10 days after i.v. infection of mice with G-25 (Amersham Pharmacia, Freiburg, Germany) column with an isocratic 1 ϫ 103 CFU L. monocytogenes as described previously (11). Unseparated flow of 1 ml 0.1% TFA/min. Low m.w. fractions were collected and passed splenocytes (6 ϫ 105/well) were stimulated for6hinround-bottom 96- Ϫ7 through a Sep-Pak C18 reverse phase, solid phase extraction unit (Waters, well microtiter plates in the presence of 10 M peptide. Subsequently, Eschborn, Germany). After washing with 5 ml 10% acetonitrile (AcN), activated cells (4 ϫ 105 or 4 ϫ 104/well) were transferred to anti IFN-␥- bound material was eluted with 1.5 ml 50% AcN and 1.5 ml 100% AcN, coated ELISPOT plates that were developed as described above. The pooled, concentrated to a final volume of 0.5 ml by vacuum centrifugation, frequency of Ag-specific cells was calculated as the number of spots per ⌬ and further fractionated by HPLC on a reverse phase C18 column ( Pak splenocytes seeded. The specificity and sensitivity of the ELISPOT as- C18Ð300A, 3.9 ϫ 300 mm; Waters): 1 ml peptide extract was loaded and say were controlled with IFN-␥-secreting CD8 T cell lines specific for

eluted with a ﬂow rate of 1 ml/min on a linear AcN gradient. Solution A p60217Ð225, p60449Ð457, p60476Ð484, and LLO91Ð99. The recovery of seeded was 0.1% TFA; solution B was 70% AcN and 0.09% TFA. The gradient CD8 T cells was Ͼ90% for all T cell lines. The Journal of Immunology 2211

Adoptive transfer traction approach is that peptides presented on the cell surface Mice were infected i.v. with 1 ϫ 103 CFU L. monocytogenes and subse- cannot be measured selectively. To overcome these restrictions we quently received 5 ϫ 106 peptide-specific CD8 T cells i.v. in PBS. The applied the principle of the ELISPOT assay for analysis of in vivo- number of CFU in the organ homogenates was determined 72 h later as infected APC. The ELISPOT assay is generally used for the described previously (16). The statistical significance of results was detection of cells reactive against a defined Ag (13). For the checked with the Newman-Keuls multiple comparison test at the 0.05 significance level. All tests were performed using WINKS statistical analysis ELISPOT-based Ag presentation assay the test principle was re- software (Texasoft, Cedar Hill, TX). All experiments were repeated at least versed, and the assay was used to detect the activation of peptide- twice, with similar results. specific CD8 T cells by infected APC. To detect Ag presentation in vivo, mice were infected i.p. with Results a high dose (1 ϫ 107 or 1 ϫ 106 CFU) of L. monocytogenes. Applying the ELISPOT assay for measuring Ag presentation Spleens were removed 48 h p.i., and splenocytes were tested in the ex vivo ELISPOT-based Ag presentation assay (Fig. 1A). To quantify the Ag presentation studies are generally performed with permanent strength of Ag presentation the number of spots per well was cell lines that are cultured and infected in vitro. Ex vivo analysis counted (Fig. 1B). In vivo infected splenocytes showed a distinct of antigenic peptides from infected organs was only reported in a reaction pattern with CD8 T cells specific for p60217Ð225, few instances (4, 5, 17). Acidic extraction of naturally processed p60449Ð457, p60476Ð484, and LLO91Ð99. The strongest response was peptides from tissues is a time-consuming task that involves mul- obtained with CD8 T cells specific for LLO91Ð99 and p60217Ð225. tiple separation steps and generally requires a relatively large input CD8 T cells specific for p60 yielded an intermediate re-

476Ð484 Downloaded from of material. Another disadvantage of the quantitative peptide ex- sponse, and p60449Ð457-speciﬁc CD8 T cells yielded the weakest http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 1. Applying the ELISPOT assay for measuring Ag presentation ex vivo. Mice were infected i.p. with the indicated dose of L. monocytogenes. Ag presentation by splenocytes was tested 48 h p.i. with a panel of CD8 T cell lines in the ELISPOT-based Ag presentation assay as described in Materials ϫ 5 ϫ 5 and Methods. CD8 T cells specific for p60217Ð225, p60449Ð457, p60476Ð484,orLLO91Ð99 were added to either 5 10 or 1 10 splenocytes as indicated. Scans of individual wells (A) and the total number of spots per 2 ϫ 104 CD8 T cells (B) are shown. The peptide sensitivity of the used CD8 T cells was tested in an ELISPOT assay with splenocytes as APC in the presence of the indicated peptide concentrations (C). Shown is the total number of spots per 2 ϫ 104 CD8 T cells of the indicated specificity. 2212 PRESENTATION OF CD8 T CELL EPITOPES IN VIVO response. The reactivity of the CD8 T cell lines was dependent on the dose used for infection of mice (Fig. 1B). Spleens of mice infected with 1 ϫ 107 CFU compared with mice infected with 1 ϫ 106 CFU revealed an ϳ5-fold increased load with L. monocytogenes (56 ϫ 106 and 11 ϫ 106 CFU/spleen, respectively). Al- though the absolute strength of the CD8 T cell response against infected splenocytes varied in both groups of mice, the principal recognition pattern remained unchanged. The signal strength was also dependent on the number of APC added per well. A 5-fold reduction of the number of APC added per well (1 ϫ 105 vs 5 ϫ 105/well) also reduced the absolute number of spots per well, but did not alter the principal recognition pattern obtained with the panel of CD8 T cell lines tested (Fig. 1B). If CD8 T cell lines were cocultivated with noninfected spleen cells or if spleen cells were cultivated in the absence of CD8 T cells, a background activity between one and four spots per well was observed (data not shown). To enable quantitative comparison of the presentation of different peptides, CD8 T cell lines were selected that showed compa- Downloaded from rable peptide sensitivities (Fig. 1C). The peptide sensitivity of CD8 T cell lines was measured in the ELISPOT-based Ag presentation assay. Noninfected splenocytes were incubated with graded doses of synthetic p60217Ð225, p60449Ð457, p60476Ð484, and LLO91Ð99 peptide, respectively, and were tested with the CD8 T cell line of corresponding specificity. As shown in Fig. 1C all CD8 T cell lines FIGURE 2. Ex vivo Ag presentation analysis is independent of bacterial http://www.jimmunol.org/ required a similar minimal peptide concentration for a significant replication. To test the influence of ongoing bacterial replication and pro- response. tein secretion on the ex vivo peptide presentation pattern the ELISPOT- based Ag presentation assay was performed either in the absence of anti- Naturally processed antigenic peptides were detected after in- biotics (upper panel) or in the presence of 10 ␮g/ml gentamicin and 20 fection with a high dose of L. monocytogenes. When spleens were ␮g/ml tetracycline (lower panel). Graded numbers of splenocytes were removed 48 h p.i., a significant bacterial load was present in the tested with peptide-specific CD8 T cell lines. Shown is the mean total organ. Thus, the possibility exists that further replication of bac- number of spots per 2 ϫ 104 CD8 T cells of the indicated specificity. Error teria occurs in the in vitro phase of the assay. To address the bars indicate the SD of triplicate determinations. possible effect of ongoing intracellular bacterial replication on the Ag presentation assay control experiments were performed. by guest on September 28, 2021 Graded numbers of splenocytes were tested in the ELISPOT-based Ag presentation assay either in the absence of antibiotics (Fig. 2, panel) or J774 (Fig. 3A, lower panel) cell lines. Remarkably, in upper panel) or in the presence of gentamicin and tetracycline comparison to L. monocytogenes-infected P388 or J774 cells,

(Fig. 2, lower panel). As shown in Fig. 2 the presence of antibi- LLO91Ð99 was clearly presented stronger by infected splenocytes. otics did not inﬂuence the principle peptide presentation pattern of In contrast, the peptide presented strongest by P388 and J774 cells

L. monocytogenes-infected splenocytes. Thus, the peptide presen- was p60449Ð457, which was barely detectable on infected spleno- tation pattern of L. monocytogenes-infected splenocytes was not cytes. As it is not known whether the APC type principally influ- influenced by the inhibition of bacterial protein synthesis. To ences peptide recognition by CD8 T cells control experiments exclude any possible interference of viable bacteria in the Ag were performed with peptide-loaded APC. Splenocytes, P388, or Ϫ presentation assay, all tests were performed in the presence of J774 cells were loaded for2hinthepresence of 10 9 M synthetic gentamicin and tetracycline. peptides, washed, and tested with peptide-specific CD8 T cells of Taken together, these results demonstrate that the ELISPOT- the corresponding specificity. Fig. 3B shows the number of spots based Ag presentation assay allows the ex vivo detection of anti- per 2 ϫ 104 responder CD8 T cells. To directly compare the Ag genic peptides presented on the surface of L. monocytogenes-in- presentation efficacy of different cell types, the number of APC fected splenocytes. was adjusted to 1000 APC/well. As shown in Fig. 3B splenocytes were weaker APC than P388 or J774 cells. However, different L. monocytogenes-infected cells exhibit distinct Ag presentation APC types did not exert a selective influence on the recognition of patterns in vivo and in vitro any of the four antigenic peptides tested. In summary, these results Long term in vitro propagated cell lines poorly represent the char- show that ex vivo isolated L. monocytogenes-infected splenocytes acteristics of natural APC, which, e.g., are always exposed to a and in vitro infected cell lines exhibit different Ag presentation specific local cytokine environment. Therefore, we used the patterns. ELISPOT-based Ag presentation assay to compare in vitro and in vivo Ag presentation. Graded numbers of APC infected in vivo or Ag presentation pattern of in vivo-infected macrophages in vitro were added to CD8 T cells specific for p60217Ð225, Splenocytes are not a homogenous cell population. To test whether p60449Ð457, p60476Ð484,orLLO91Ð99. Fig. 3A shows the average the disparate Ag presentation patterns of whole spleen cells and in number of spots/well obtained in the presence of 2 ϫ 104 spleno- vitro infected macrophage-like cell lines are the result of a mixture cytes, 103 J774, or 103 P388 cells/well, respectively. The reactivity of different APC types in the spleen, the Ag presentation pattern of pattern of CD8 T cell lines with in vivo-infected splenocytes in vivo-infected macrophages was analyzed. CD11bϩ cells, which (Fig. 3A, upper panel) differed consistently from the reaction are mostly macrophages, were separated from spleens 48 h after pattern obtained with in vitro infected P388 (Fig. 3A, middle i.v. infection of mice with 1 ϫ 105 CFU L. monocytogenes. The The Journal of Immunology 2213 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 4. Ag presentation pattern of in vivo-infected macrophages. Macrophages were isolated by immunomagnetic separation from spleens 48 h after i.v. infection with 1 ϫ 105 CFU L. monocytogenes. A, Cells were stained as indicated with anti-F4/80- and anti-CD11c-mAb or with an rat IgG2b isotype control. The cytoﬂuorometric analysis of ex vivo isolated macrophages is shown. B, Isolated macrophages or unseparated splenocytes were tested in the ELISPOT-based Ag presentation assay. The number of spots per 2 ϫ 104 CD8 T cells and the indicated number of APC per well is shown.

FIGURE 3. L. monocytogenes-infected cells exhibit different Ag presentation patterns in vivo and in vitro. Ag presentation by in vivo-infected tion pattern of whole spleen cells, which is characterized by strong splenocytes (upper panel) and in vitro infected P388 (middle panel) and p60217Ð225 and LLO91Ð99 presentation. However, the direct com- J774 (lower panel) cells was compared (A). The peptide presentation was parison of the T cell activation by isolated macrophages and un- tested in the ELISPOT-based Ag presentation assay with CD8 T cell lines separated spleen cells showed that the CD11b-selected cells are of the indicated specificity. Shown is the mean total number of spots per much stronger Ag presenters (Fig. 4B). 2 ϫ 104 CD8 T cells Ϯ SD of triplicate determinations. The possible cell Thus, isolated, in vivo-infected macrophages demonstrate the line dependency of the peptide recognition pattern was analyzed with pep- same Ag presentation pattern as unseparated spleen cells, which is tide-loaded APC (B). J774, P388, and splenocytes were loaded 2 h with clearly different form the peptide presentation pattern of in vitro 10Ϫ9 M peptide and were tested with CD8 T cells of the corresponding infected macrophage-like cell lines. specificity as indicated. Shown is the number of spots per 2 ϫ 104 CD8 T cells and 1000 peptide-loaded APC per well. Acidic extraction of naturally processed antigenic peptides from infected organs immunomagnetic selection of CD11bϩ cells yielded a highly en- The ELISPOT-mediated Ag presentation assay measures selec- riched macrophage population. Approximately 90% of cells tively cell surface presentation of peptides. To determine whether stained positively for F4/80, and all cells were negative for the the Ag presentation pattern reflects the total peptide composition dendritic cell marker CD11c (Fig. 4A). These CD11bϩ cells were of infected cells, peptides were also quantified in whole cell ex- tested with peptide-specific CD8 T cell lines in the ELISPOT- tracts. Mice were infected i.v. with 1 ϫ 106 CFU L. monocyto- based Ag presentation assay (Fig. 4B). The Ag presentation pattern genes. Spleens were removed 48 h p.i., and peptides were ex- of isolated macrophages repeated the principle peptide presenta- tracted from a pool of five organs. Naturally processed antigenic 2214 PRESENTATION OF CD8 T CELL EPITOPES IN VIVO

peptides were also extracted from L. monocytogenes-infected P388 peptides (Fig. 5B, lower panel). The p60476Ð484 peptide was not cells. After HPLC separation, fractions containing antigenic activ- detected in cell extracts, indicating that less than five p60476Ð484 ity were identified with peptide-specific CD8 T cells in a 51Cr peptides were presented per cell. Thus, if the peptide presentation release assay. Fig. 5A shows the results of a representative CTL patterns of P388 cells and spleens are compared, it is clear that test of HPLC fractions from a peptide extract of infected spleens. p60449Ð457 was significantly less abundant than p60217Ð225 in in- The total peptide content of positive fractions was finally deter- fected spleens. As in infected spleens the amount of p60217Ð225 mined by linear interpolation from a synthetic peptide standard peptides was ϳ10-fold greater than the detection limit of

(data not shown). In L. monocytogenes-infected spleens only the p60449Ð457-specific CD8 cells, it is unlikely that the presence of an peptides LLO91Ð99 and p60217Ð225 were detectable (Fig. 5B, upper equal amount of p60449Ð457 was just overlooked. Taken together, panel). Both peptides were similarly abundant, in the range be- the quantitative peptide extraction procedure confirmed the results tween 3 ϫ 109 and 5 ϫ 109 peptides/spleen. The peptides obtained with the ELISPOT-based Ag presentation assay (Fig. p60476Ð484 and p60449Ð457 both ranged below the detection limit 3A). Therefore, we conclude that the Ag presentation pattern dis- of the assay, which was ϳ5 ϫ 108 peptides/spleen for both pep- played by infected splenocytes and P388 cells mirrors the total tides. In contrast to infected splenocytes, the most abundant pep- peptide composition of the infected cells. tides in P388 cells were p60449Ð457 and p60217Ð225, while LLO was significantly less abundant than these p60-derived 91Ð99 Kinetics of Ag presentation in vivo After sublethal L. monocytogenes infection of mice the bacterial load in the spleen peaks around day 3 p.i., and specific CD8 T cells Downloaded from can be detected 5 days p.i. and peak around day 7 p.i. (18). To study the kinetics of Ag presentation after L. monocytogenes infection the peptide presentation pattern of splenocytes was tested at different time points postinfection. Two different experimental approaches were chosen. To achieve a similar bacterial load in the

spleen mice were infected with different doses of L. monocyto- http://www.jimmunol.org/ genes. Spleens were tested 6, 24, and 48 h after i.v. infection with 1 ϫ 108,1ϫ 107, and 1 ϫ 106 CFU/mouse, respectively (Fig. 6A). Alternatively, all mice were infected with the same dose of 1 ϫ 105 CFU L. monocytogenes i.v. and tested 6, 24, and 48 after infection (Fig. 6B). The number of CFU per spleen of the different experimental groups is shown in Fig. 6C. Later time points p.i. were not included due to the increasing background activity of ex vivo isolated splenocytes. Spleen cells of all mice were tested in the ELISPOT-based Ag presentation assay with a set of CD8 T cell by guest on September 28, 2021 lines. Remarkably, the relative abundance of naturally processed antigenic peptides in vivo changed dramatically during the ﬁrst

48hofL. monocytogenes infection. Around 6 h p.i., p60217Ð225, p60449Ð457, p60476Ð484, and LLO91Ð99 were all presented with comparable strength. Later, between 24 and 48 h p.i., this relation

changed. At 48 h p.i. LLO91Ð99 and p60217Ð225 were presented much more strongly than p60449Ð457 and p60476Ð484 in both experimental groups. In summary, these data show that the Ag presentation pattern of L. monocytogenes-infected cells in vivo is highly dynamic, with rapid peptide-speciﬁc changes during the ﬁrst 48 h p.i.

Frequency and protective potential of peptide-specific CD8 T cells in vivo In previous reports a paradoxical inverse correlation between the FIGURE 5. Acidic extraction of naturally processed antigenic peptides abundance of naturally processed antigenic peptides extracted from L. monocytogenes-infected cell lines and organs. Naturally processed from in vitro infected cells and the frequency of peptide-specific antigenic peptides were extracted and quantified from the spleen of L. CD8 T cells in vivo was noted (13, 14). We found that the Ag monocytogenes-infected mice and from infected P388 cells. After HPLC presentation pattern of infected cell lines differed from the pattern separation, fractions containing antigenic activity were identified with pep- displayed by ex vivo isolated splenocytes. Furthermore, we found 51 tide-specific CD8 T cell lines in a Cr release assay. Peptides p60217Ð225, that in vivo the peptide presentation pattern of L. monocytogenes- p60449Ð457, p60476Ð484, and LLO91Ð99 were detected with CD8 T cells of infected cells changed during the course of infection. As Ag pre- corresponding specificity as indicated. A representative CTL test of HPLC sentation is principally required for CD8 T cell induction and also fractions prepared from a peptide extract of infected spleens 48 h p.i. is for the recognition of infected target cells, we analyzed the pos- shown (A). Arrows indicate the HPLC fraction where the indicated peptides elute. The total peptide content of positive fractions was determined sible correlation between in vivo peptide presentation and the fre- by linear interpolation from a synthetic peptide standard (B). The total quency and effector function of peptide-specific CD8 T cells. As number of peptides per L. monocytogenes-infected spleen is shown (upper shown in Fig. 7A the frequency of peptide-specific CD8 T cells in panel). For infected P388 cells the number of peptides per cell is indicated primarily infected BALB/c mice on day 10 p.i. exhibited a distinct (lower panel). Dotted lines indicate the detection limit of the assay. hierarchy. This hierarchy of L. monocytogenes-specific CD8 T The Journal of Immunology 2215 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 7. Frequency and protective potential of peptide-specific CD8 T cells in vivo. The frequency of peptide-specific CD8 T cells in vivo 10 days p.i. was determined in an ELISPOT assay in the presence of 10Ϫ7 M peptide (A). The mean number of IFN-␥-positive cells per 4 ϫ 105 splenocytes is shown. Error bars indicate the SD of three individual mice. The FIGURE 6. Kinetics of Ag presentation in vivo. The ex vivo Ag pre- protective potential of peptide-specific CD8 T cells was tested in adoptive sentation pattern was determined 6, 24, and 48 h after L. monocytogenes transfer experiments (B and C). Mice were injected i.v. with 1 ϫ 103 CFU infection of mice. Freshly isolated splenocytes from infected mice were L. monocytogenes and subsequently received 5 ϫ 106 peptide-specific CD8 tested in the ELISPOT-based Ag presentation assay using peptide-specific T cells of the indicated specificity. Results show the number of CFU per CD8 T cell lines. The strength of Ag presentation is shown as the mean spleen Ϯ SD of groups of five mice. Statistical analysis of results was total number of spots per 2 ϫ 104 CD8 T cells of the indicated specificity. performed with the Newman-Keuls multiple comparison test (p Ͻ 0.05). T Error bars indicate the SD of triplicate determinations. A, Mice were tested cell lines were ranked according to the protection mediated in the spleen. 6 h after infection with 1 ϫ 108 CFU, 24 h after infection with 107 CFU, The graphical representation of the Newman-Keuls test (C) indicates the and 48 h after infection with 106 CFU L. monocytogenes, respectively. B, rank and group of each T cell line. The means of any two lines belonging Mice were tested 6, 24, and 48 h after infection with 105 CFU L. mono- to the same group (underscored by the same line) are not significantly cytogenes. The different time points of each of the two experimental groups different. (A and B) were tested in the same experiment. C, Shown are the bacterial loads of spleens tested in A and B. To test the possible correlation of the protective potential of peptide-specific CD8 T cells to the peptide presentation pattern in vivo, adoptive transfer experiments were performed. The CD8 T Ͼ Ͼ Ͼ cells (LLO91Ð99 p60217Ð225 p60476Ð484 p60449Ð457) cor- cell lines were matched for comparable peptide sensitivity and roborates results published previously by Sijts et al. (13). The re- lytic potential in vitro. Mice received 5 ϫ 106 CD8 T cells im- cently described p60476Ð484 epitope exhibited an intermediate mediately after L. monocytogenes infection, and the bacterial load strength among the three known CD8 T cell epitopes of p60, con- of the spleen was determined 3 days later (Fig. 7B). All CD8 T cell firming previous results obtained after secondary L. monocyto- lines conferred significant ( p Ͻ 0.05) protection against L. mono- genes infection (11). Remarkably, the immunodominant peptides cytogenes. Generally, LLO91Ð99-specific CD8 T cells conferred the

LLO91Ð99 and p60217Ð115 were also the peptides that revealed the strongest protection, but this difference was not statistically sig- strongest Ag presentation 48 h p.i. (Fig. 6, A and B). niﬁcant (Fig. 7C). Corroborating earlier results by Dunn and North 2216 PRESENTATION OF CD8 T CELL EPITOPES IN VIVO

(19) CD8 T cells transferred 24 h p.i. conferred no significant determinant, while in other instances the limitation is located on protection against L. monocytogenes infection (data not shown). the side of the T cell (22). Taken together, from these experiments Compared with the quantitative hierarchy of CD8 T cells, the no obvious general correlation between the frequency of peptide- protection assay revealed a different hierarchy of the protective specific CD8 T cells in vivo and the abundance of antigenic pep- potential of transferred CD8 T cells. Remarkably, the protection tides in infected cells is evident. However, it should be noted that mediated by CD8 T cells did not correlate with the Ag presentation those studies generally analyzed Ag presentation in vitro. When pattern at 48 h p.i., when only poor presentation of p60449Ð457 and long-term in vitro propagated cell lines are studied, the most ob- p60476Ð484 was detected. However, the protection pattern of CD8 vious differences to the in vivo situation include the cell type and T cells correlated well with the Ag presentation pattern 6 h p.i., the absence of the physiological cytokine environment. Addition- when all peptides were presented with similar strength (Fig. 6, A ally, it has to be considered that compared with an in vitro cell and B). culture model, microorganisms in vivo generally exhibit different In summary, these data show that the protective capacity of growth kinetics and thus also the rate and kinetics of protein ex- adoptively transferred L. monocytogenes-specific CD8 T cells does pression might not be identical. Analysis of the Ag processing of not correlate to the immunodominance of these CD8 T cells in pp89168Ð176, an antigenic peptide derived from the murine CMV, vivo. Further it is shown that the expansion and the effector func- has shown that in vivo Ag processing is strongly influenced by tion of CD8 T cells correlate with the in vivo Ag presentation IFN-␥ (5) and also by the infected cell type (23). The important pattern at different time points p.i. influence of the cell type on Ag presentation has been shown in a study of virus-infected dendritic and fibroblast cell lines (24). A Discussion possible explanation for these observations is that cell type- Downloaded from We used a novel approach for the ex vivo measurement of Ag specific or IFN-␥-mediated differences in the composition of the presentation that is based on the testing of in vivo-infected APC proteasome (reviewed in Ref. 25) alter Ag presentation in vivo. with peptide-specific CD8 T cells in a sensitive ELISPOT assay. In the current study also the peptide presentation pattern of in The high sensitivity of this assay enabled the measurement of pep- vivo-infected macrophages was analyzed. On a per cell basis a tide presentation without the technically difficult classical peptide significantly stronger Ag presentation by ex vivo isolated macro- extraction method that is generally used for the quantification of phages was observed, while the peptide presentation pattern was http://www.jimmunol.org/ naturally processed peptides in tissues (3Ð5, 17). Remarkably, the identical with unseparated splenocytes. As it is known that in the comparison of in vivo-infected splenic macrophages and in vitro- spleen the majority of Listeriae resides in macrophages this result infected macrophage-like cell lines revealed distinct peptide pre- could be expected (26, 27). Thus, either the peptide presentation of sentation patterns for in vivo- and in vitro-infected APC. As dif- other infected cell types in the spleen, e.g., dendritic cells or si- ferent APC types and also APC harvested at different time points nusoidal lining cells is similar to macrophages or the quantitative p.i. were always directly compared in the same experiment with contribution of these cell types to the overall Ag presentation pat- the same set of CD8 T cell lines, the observed differences in the tern of unseparated splenocytes is to low to be detected. d peptide presentation patterns cannot be attributed to different sen- LLO91Ð99 and p60217Ð225 form relatively stable peptide/K com- ϳ d by guest on September 28, 2021 sitivities of the CD8 T cell lines used. In contrast to acidic peptide plexes with a half-life of 6 h, while p60449Ð457/K complexes extraction the Ag presentation assay detects selectively cell surface have a half-life of Ͻ1 h (28, 29). The relative strength of the CD8 peptide presentation. This is an important advantage if the rele- T cell response against these peptides correlates with the stability vance of Ag presentation for the induction of T cells shall be of the corresponding MHC class I/peptide complexes (29). Similar studied because some microbial immune escape mechanisms se- results were obtained with EBV-specific CD8 T cells (30) and in lectively prevent cell surface presentation of antigenic peptides a CD8 T cell immunization study with a large number of synthetic (20). However, after infection of mice with L. monocytogenes, the peptides (31). In the context of these data it is remarkable that in peptide presentation patterns of splenocytes and infected cell L. monocytogenes-infected spleens 6 h p.i. all peptides were pre- lines were principally confirmed by peptide extraction of whole sented with similar strength, while at later time points p.i. the organs and cells, respectively. The Ag presentation pattern of in presentation of peptides that form stable MHC/peptide complexes vitro infected P388 or J774 cells confirmed data reported by was significantly stronger than the presentation of peptides that Pamer’s group (14). Compared with acidic peptide extraction the form unstable complexes. The remarkable correlation between the ELISPOT-based Ag presentation assay requires less material and stability of MHC class I/peptide complexes, prolonged peptide is more sensitive. The ELISPOT-based assay also enables mea- presentation, and the frequency of peptide-specific CD8 T cells in surement of the Ag presentation pattern of defined, ex vivo isolated vivo suggests a model for the observed changes of the in vivo Ag cell populations. This gives the assay a broad range of possible presentation. In infected cells p60 protein secretion is limiting for applications for the study of animal and probably also human dis- the generation of p60-derived epitopes (32). Thus, over an ex- ease states. tended period of time it has to be expected that peptides that form The quantitative aspects of Ag processing are generally ana- stable MHC/peptide complexes outnumber peptides that form less lyzed in long term cell cultures. A number of previous studies stable complexes. Finally, this could result in the preferential stim- compared the abundance of naturally processed peptides in in- ulation and expansion of T cells directed against the more stable fected cell lines with the frequency of peptide-specific CD8 T cells peptide/MHC complexes. The importance of peptide stability for in vivo. In the murine L. monocytogenes infection model an in- the Ag presentation pattern is obvious when protein secretion is verse correlation between the abundance of antigenic peptides in inhibited. Sijts et al. have shown that after inhibition of p60 bio- infected cell lines and the frequency of peptide-specific CD8 T synthesis by tetracycline treatment p60449Ð457 that forms unstable cells has been noted (13, 14). Similarly, immunodominance among MHC class I/peptide complexes diminishes quickly, while

EBV-derived MHC class I-restricted epitopes does not correlate p60217Ð225 that forms stable complexes persists over an extended with the abundance of antigenic peptides in EBV-transformed cell time period (28). Accumulation of stable peptides over time could lines (21). Also, a detailed study of the CD8 T cell response also at least in part explain the observed differences of the Ag against influenza virus has shown that inefficient Ag processing presentation patterns of in vitro- and in vivo-infected APC. For accounts only for the poor immunogenicity of one subdominant peptide extraction macrophage-like cells were harvested 6 h p,i,, The Journal of Immunology 2217 while spleens were removed 48 h p.i.. Thus, the accumulation of the quantitative, cell type-specific, and regulatory aspects of Ag stable peptides in the spleen 48 h p.i. is clearly plausible. However, presentation in vivo. The answer to these questions should help to the peptide presentation patterns of splenocytes and macrophage- facilitate the design of T cell vaccines and the therapy of T cell- like cells differed also at an early time point 6 h after infection. At mediated autoimmune diseases. this early time point after infection, for example, LLO91Ð99 was presented clearly stronger on in vivo-infected APC than on in Acknowledgments vitro-infected APC. We thank S. Schenk for excellent technical assistance, and W. Meister for The relevance of prolonged Ag presentation for the induction of his valuable help while preparing the figures. protective anti-listerial immunity has been demonstrated by the antibiotic abridgement of bacterial replication in vivo. Abridge- References ment by ampicillin treatment during the first 5 days p.i. results in 1. Sercarz, E. E., P. V. Lehmann, A. Ametani, G. Benichou, A. Miller, and a diminished protective T cell response (33). Recently, Mercado et K. Moudgil. 1993. Dominance and crypticity of T cell antigenic determinants. Annu. Rev. Immunol. 11:729. al. have shown that the magnitude and the kinetics of the 2. Yewdell, J. W., and J. R. Bennink. 1999. Immunodominance in major histocompatibility complex class I-restricted T lymphocyte responses. Annu. Rev. Immu- p60217Ð225 and LLO91Ð99-specific T cell response against L. mono- nol. 17:51. cytogenes are determined during the first 24 h of bacterial infection 3. Griem, P., H. J. Wallny, K. Falk, O. Ro¬tzschke, B. Arnold, G. Scho¬nrich, G. independently from the infectious dose (34). Additionally they Ha¬mmerling, and H. G. Rammensee. 1991. Uneven tissue distribution of minor showed that the transfer of naive, p60 -specific CD8 T cells histocompatibility proteins versus peptides is caused by MHC expression. Cell 449Ð457 65:633. results in a strongly enhanced CD8 T cell response after subse- 4. Pamer, E. G., J. T. Harty, and M. J. Bevan. 1991. Precise prediction of a dominant quent L. monocytogenes infection, suggesting that Ag is also not class I MHC-restricted epitope of Listeria monocytogenes. Nature 353:852. Downloaded from limiting for the expansion of p60 -specific CD8 T cells in 5. Geginat, G., T. Ruppert, H. Hengel, R. Holtappels, and U. H. Koszinowski. 1997. 449Ð457 IFN-␥ is a prerequisite for optimal antigen processing of viral peptides in vivo. vivo. Therefore, and also because dendritic cells that possibly con- J. Immunol. 158:3303. tribute only very few to the overall Ag presentation pattern of 6. Restifo, N. P., I. Bacik, K. R. Irvine, J. W. Yewdell, B. J. McCabe, R. W. Anderson, L. C. Eisenlohr, S. A. Rosenberg, and J. R. Bennink. 1995. unseparated splenocytes play an important role in the primary Antigen processing in vivo and the elicitation of primary CTL responses. J. Im- stimulation of naive CD8 T cells, caution has to be exercised in the munol. 154:4414. interpretation of the observed correlation between the Ag presen- 7. del Val, M., K. Munch, M. J. Reddehase, and U. H. Koszinowski. 1989. Presen- http://www.jimmunol.org/ tation of CMV immediate-early antigen to cytolytic T lymphocytes is selectively tation pattern and the frequency of L. monocytogenes-specific CD8 prevented by viral genes expressed in the early phase. Cell 58:305. T cells in vivo. However, the timing requirements described by 8. Chen, W., S. Khilko, J. Fecondo, D. H. Margulies, and J. McCluskey. 1995. Mercado et al. are not strikingly different from the time frame of Determinant selection of major histocompatibility complex class I-restricted peptides is explained by class I-peptide affinity and is strongly influenced by non- the changing Ag presentation pattern in L. monocytogenes-infected dominant anchor residues. J. Exp. Med. 180:1471. mice, as by 24 h p.i. the presentation of the subdominant epitopes 9. Pamer, E. G. 1994. Direct sequence identification and kinetic analysis of an MHC class I-restricted Listeria monocytogenes CTL epitope. J. Immunol. 152:686. p60449Ð457 and p60476Ð484 started to decrease in relation to that of 10. Sijts, A. J., A. Neisig, J. Neefjes, and E. G. Pamer. 1996. Two Listeria mono- the dominant CD8 T cell epitopes. cytogenes CTL epitopes are processed from the same antigen with different ef- In vivo, CD8 T cells mediate protection against L. monocyto- ficiencies. J. Immunol. 156:683.

11. Geginat, G., S. Schenk, M. Skoberne, W. Goebel, and H. Hof. 2001. A novel by guest on September 28, 2021 genes (35, 36). To exert their protective function CD8 T cells must approach of direct ex vivo epitope mapping identifies dominant and subdominant recognize infected target cells. A number of observations suggest CD4 and CD8 T cell epitopes from Listeria monocytogenes. J. Immunol. 166: that the Ag presentation requirements for the primary stimulation 1877. 12. Busch, D. H., H. G. Bouwer, D. Hinrichs, and E. G. Pamer. 1997. A nonamer of naive T cells differ from the antigenic stimulus that is necessary peptide derived from Listeria monocytogenes metalloprotease is presented to cy- to stimulate the effector function of experienced T cells. Shen et al. tolytic T lymphocytes. Infect. Immun. 65:5326. 13. Vijh, S., and E. G. Pamer. 1997. Immunodominant and subdominant CTL re- have analyzed the CD8 T cell response against a model T cell Ag sponses to Listeria monocytogenes infection. J. Immunol. 158:3366. expressed by recombinant L. monocytogenes either as secreted or 14. Pamer, E. G., A. J. Sijts, M. S. Villanueva, D. H. Busch, and S. Vijh. 1997. MHC nonsecreted fusion protein (37). Remarkably, they have found that class I antigen processing of Listeria monocytogenes proteins: implications for dominant and subdominant CTL responses. Immunol. Rev. 158:129. the compartmentalization of bacterial Ags has differential effects 15. Azuma, M., H. Yssel, J. H. Phillips, H. Spits, and L. L. Lanier. 1993. Functional on priming of CD8 T cells and protective immunity. Dichotomous expression of B7/BB1 on activated T lymphocytes. J. Exp. Med. 177:845. requirements for CD8 T cell effector cell function and expansion 16. Geginat, G., M. Kretschmar, S. Walter, D. Junker, H. Hof, and T. Nichterlein. 1999. Suppression of acquired immunity against Listeria monocytogenes by am- are also implicated by the observation that subdominant CD8 T photericin B-mediated inhibition of CD8 T cell function. J. Infect. Dis. 180:1186. cell populations can have an important contribution to protective 17. Vijh, S., I. M. Pilip, and E. G. Pamer. 1998. Effect of antigen-processing efficiency on in vivo T cell response magnitudes. J. Immunol. 160:3971. immunity against diverse intracellular microorganisms (38Ð42). 18. Busch, D. H., I. M. Pilip, S. Vijh, and E. G. Pamer. 1998. Coordinate regulation The good protection obtained after adoptive transfer of of complex T cell populations responding to bacterial infection. Immunity 8:353. p60 -specific CD8 T cells extents these results to the infec- 19. Dunn, P. L., and R. J. North. 1991. Resolution of primary murine listeriosis and 449Ð457 acquired resistance to lethal secondary infection can be mediated predominantly tion with L. monocytogenes. The observed kinetic changes in the in by Thy-1ϩCD4ϪCD8Ϫ cells. J. Infect. Dis. 164:869. vivo peptide presentation pattern during L. monocytogenes infec- 20. Fru¬h, K., A. Gruhler, R. M. Krishna, and G. J. Schoenhals. 1999. A comparison tion further suggest an explanation for this dichotomy between the of viral immune escape strategies targeting the MHC class I assembly pathway. Immunol. Rev. 168:157. quantitative and protective hierarchy of CD8 T cells. The equal 21. Crotzer, V. L., R. E. Christian, J. M. Brooks, J. Shabanowitz, R. E. Settlage, adoptive protection mediated by different peptide-specific CD8 T J. A. Marto, F. M. White, A. B. Rickinson, D. F. Hunt, and V. H. Engelhard. 2000. Immunodominance among EBV-derived epitopes restricted by HLA-B27 cell lines correlated with the equal peptide presentation of these does not correlate with epitope abundance in EBV-transformed B-lymphoblastoid peptides 6 h p.i.. In contrast to the protective function, the fre- cell lines. J. Immunol. 164:6120. quency of L. monocytogenes-specific CD8 T cells correlated with 22. Chen, W., L. C. Anton, J. R. Bennink, and J. W. Yewdell. 2000. Dissecting the multifactorial causes of immunodominance in class I-restricted T cell responses the Ag presentation pattern 48 h p.i.. Taken together these corre- to viruses. Immunity 12:83. lations suggest distinct Ag presentation requirements for CD8 T 23. Hengel, H., U. Reusch, G. Geginat, R. Holtappels, T. Ruppert, E. Hellebrand, and cell expansion and CD8 T cell effector function. U. H. Koszinowski. 2000. Macrophages escape inhibition of major histocompatibility complex class I-dependent antigen presentation by cytomegalovirus. J. Vi- The current study represents the first study of the dynamics of rol. 74:7861. Ag presentation in vivo. The obtained results were unexpected and 24. Butz, E. A., and M. J. Bevan. 1998. Differential presentation of the same MHC class I epitopes by fibroblasts and dendritic cells. J. Immunol. 160:2139. highlight the limitations of in vitro experiments to predict Ag pre- 25. Fru¬h, K., and Y. Yang. 1999. Antigen presentation by MHC class I and its sentation in vivo. Therefore, it will be important to better define regulation by interferon ␥. Curr. Opin. Immunol. 11:76. 2218 PRESENTATION OF CD8 T CELL EPITOPES IN VIVO

26. Conlan, J. W., and R. J. North. 1994. Neutrophils are essential for early anti- 35. Harty, J. T., and M. J. Bevan. 1992. CD8ϩ T cells specific for a single nonamer Listeria defense in the liver, but not in the spleen or peritoneal cavity, as revealed epitope of Listeria monocytogenes are protective in vivo. J. Exp. Med. 175:1531. by a granulocyte-depleting monoclonal antibody. J. Exp. Med. 179:259. 36. Harty, J. T., and E. G. Pamer. 1995. CD8 T lymphocytes specific for the secreted 27. Conlan, J. W. 1996. Early pathogenesis of Listeria monocytogenes infection in p60 antigen protect against Listeria monocytogenes infection. J. Immunol. 154: the mouse spleen. J. Med. Microbiol. 44:295. 4642. 28. Sijts, A. J., and E. G. Pamer. 1997. Enhanced intracellular dissociation of major 37. Shen, H., J. F. Miller, X. Fan, D. Kolwyck, R. Ahmed, and J. T. Harty. 1998. histocompatibility complex class I-associated peptides: a mechanism for opti- Compartmentalization of bacterial antigens: differential effects on priming of mizing the spectrum of cell surface-presented cytotoxic T lymphocyte epitopes. J. Exp. Med. 185:1403. CD8 T cells and protective immunity. Cell 92:535. 29. Busch, D. H., and E. G. Pamer. 1998. MHC class I/peptide stability: implications 38. Gallimore, A., T. Dumrese, H. Hengartner, R. M. Zinkernagel, and for immunodominance, in vitro proliferation, and diversity of responding CTL. H. G. Rammensee. 1998. Protective immunity does not correlate with the hier- J. Immunol. 160:4441. archy of virus-specific cytotoxic T cell responses to naturally processed peptides. 30. Levitsky, V., Q. J. Zhang, J. Levitskaya, and M. G. Masucci. 1996. The life span J. Exp. Med. 187:1647. of major histocompatibility complex-peptide complexes influences the efficiency 39. van der Most, R. G., K. Murali-Krishna, J. L. Whitton, C. Oseroff, J. Alexander, of presentation and immunogenicity of two class I-restricted cytotoxic T lym- S. Southwood, J. Sidney, R. W. Chesnut, A. Sette, and R. Ahmed. 1998. Iden- phocyte epitopes in the Epstein-Barr virus nuclear antigen 4. J. Exp. Med. 183: tification of Db- and Kb-restricted subdominant cytotoxic T-cell responses in 915. lymphocytic choriomeningitis virus-infected mice. Virology 240:158. 31. van der Burg, S. H., M. J. Visseren, R. M. Brandt, W. M. Kast, and C. J. Melief. 40. Olsen, A. W., P. R. Hansen, A. Holm, and P. Andersen. 2000. Efficient protection 1996. Immunogenicity of peptides bound to MHC class I molecules depends on against Mycobacterium tuberculosis by vaccination with a single subdominant the MHC-peptide complex stability. J. Immunol. 156:3308. epitope from the ESAT-6 antigen. Eur. J. Immunol. 30:1724. 32. Villanueva, M. S., P. Fischer, K. Feen, and E. G. Pamer. 1994. Efficiency of MHC 41. Franke, E. D., A. Sette, J. Sacci, Jr., S. Southwood, G. Corradin, and class I antigen processing: a quantitative analysis. Immunity 1:479. ϩ 33. North, R. J., P. A. Berche, and M. F. Newborg. 1981. Immunologic consequences S. L. Hoffman. 2000. A subdominant CD8 cytotoxic T lymphocyte (CTL) of antibiotic-induced abridgement of bacterial infection: effect on generation and epitope from the Plasmodium yoelii circumsporozoite protein induces CTLs that loss of protective T cells and level of immunologic memory. J. Immunol. 127: eliminate infected hepatocytes from culture. Infect. Immun. 68:3403. 342. 42. Holtappels, R., D. Thomas, J. Podlech, G. Geginat, H. P. Steffens, and Downloaded from 34. Mercado, R., S. Vijh, S. E. Allen, K. Kerksiek, I. M. Pilip, and E. G. Pamer. 2000. M. J. Reddehase. 2000. The putative natural killer decoy early gene m04 (gp34) Early programming of T cell populations responding to bacterial infection. J. Im- of murine cytomegalovirus encodes an antigenic peptide recognized by protective munol. 165:6833. antiviral CD8 T cells. J. Virol. 74:1871. http://www.jimmunol.org/ by guest on September 28, 2021