| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Division of Immunology, Department of Biology, and
Laboratory of Analytical Chemistry, Department of Chemistry, University of Constance, Konstanz, Germany; and
Institute of Experimental Immunology, Department of Pathology, University Hospital Zürich, Zürich, Switzerland
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
enhanced the presentation of gp33–41, whereas presentation of the gp276–286 epitope from the same glycoprotein was markedly reduced. Because proteasomes are crucially involved in epitope generation and because IFN- treatment in vitro and lymphocytic choriomeningitis virus infection in vivo lead to a gradual replacement of constitutive proteasomes by immunoproteasomes, we investigated the role of proteasome composition on epitope hierarchy. Overexpression of the active site subunits of immunoproteasomes LMP2, LMP7, and MECL-1 as well as overexpression of LMP2 alone suppressed the presentation of the gp276–286 epitope. The ability to generate gp276–286-specific CTLs was enhanced in LMP2- and LMP7-deficient mice, and macrophages from these mice showed an elevated presentation of this epitope. In vitro digests demonstrated that fragmentation by immunoproteasomes, but not constitutive proteasomes led to a preferential destruction of the gp276 epitope. Taken together, we show that LMP2 and LMP7 can at least in part determine subdominance and shape the epitope hierarchy of CTL responses in vivo. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The infection of the mouse with lymphocytic choriomeningitis virus (LCMV)3 is a frequently used model of viral infection, and also, the phenomenon of immunodominance has been thoroughly investigated (2, 8, 9, 10, 11, 12). In C57BL/6 mice, this response is dominated by CTLs specific for the H-2Db-restricted epitopes gp33–41, nucleoprotein 396–404 (NP396–404), and the subdominant epitope gp276–286. The reasons for gp276 subdominance have been thoroughly investigated in the past (11). One contributing factor to the subdominance of gp276 could be that the amount of gp276 epitopes that was eluted from 109 LCMV-infected MC57 fibroblasts (0.16 ng) was about twice as low as that of NP396 (0.3 ng) and 12-fold lower than gp33 (2.0 ng). This amount corresponded to 92 H-2Db-bound gp276 epitopes per MC57 cell, which is in a range in which it could become limiting for the recognition by gp276-specific CTLs (13, 14). It hence appears that the intracellular processing of gp276 is less efficient than processing of the gp33 epitope derived from the same glycoprotein. Because the proteasome is involved in the processing of both epitopes (15), we decided to further investigate its impact on the establishment of the epitope hierarchy.
The proteasome is the main protease in the cytoplasm and the nucleus that generates the C termini of most peptide ligands of MHC class I molecules (16, 17). The proteolytic core complex of the proteasome system is the 20S proteasome, which is constructed like a cylinder of four stacked rings. The outer two rings consist of seven different 
-type subunits that bind to regulatory complexes of the 20S core particle, whereas the two inner rings are made up of seven different subunits of the 
-type. Three of the -subunits, designated 
(1), MB1 (5), and MC14 (Z, 2), bear the active centers of the 20S proteasome. Upon stimulation of cells with the inflammatory cytokine IFN-, these constitutively expressed subunits are replaced by inducible subunits named LMP2 (1i), LMP7 (5i), and MECL-1 (2i) during the de novo assembly of 20S proteasomes. This subunit exchange alters the cleavage pattern of the proteasome (16), which can lead to an enhancement of Ag presentation (18, 19, 20, 21). However, for a few epitopes from human tumors, the induction of immunoproteasomes was found to negatively affect their presentation (22). Although gene-targeted mice deficient for LMP2 (23) and LMP7 (24) have been generated almost a decade ago, the impact of these two subunits on the hierarchy of epitopes has been barely investigated, except for a recent study by Chen et al. (25), which shows that in LMP2–/– mice the CTL response to influenza virus follows a different hierarchy than in wild-type mice. This effect was due both to differences in the CTL precursor frequency as well as to changes in epitope presentation.
Another IFN--inducible complex, which affects epitope generation and therefore has the potential to shape epitope hierarchies, is the 11S regulator of the proteasome (26) (also called PA28 (27)). PA28 consists of two different subunits, and , which form a heptameric ring that binds to the 20S or 26S proteasomes and affects their peptidolytic properties (28). Overexpression of PA28 has been shown to enhance the presentation of a number of epitopes (29, 30, 31, 32), and, conversely, PA28 deficiency in PA28–/– or PA28–/– mice interferes with epitope presentation in several cases (33, 34).
In this study, we investigated whether the composition of proteasome active site subunits and PA28 could be involved in determining immunodominance and subdominance in the LCMV system. The finding that the IFN- treatment of MC57 fibroblasts enhanced gp33 presentation, but down-regulated the presentation of gp276 inspired our work. With the help of transfectants and knockout mice, we could indeed show that immunoproteasomes negatively affect gp276 processing and presentation in vitro and the generation of gp276-specific CTLs in vivo. Our data are discussed with respect to so far elusive changes of epitope hierarchy in acute vs persistent LCMV infection and in relation to different Ag presentation by fibroblasts as opposed to dendritic cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 mice (H-2b) were purchased from the animal facility of University of Constance. LMP2–/– (23) and LMP7–/– (24) gene-targeted mice were provided by J. Monaco (Department of Molecular Genetics, Cincinnati Medical Center, Cincinnati, OH), while PA28 mice were kindly contributed by T. Chiba (Department of Molecular Oncology, Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan) (34). All knockout mice were backcrossed onto the C57BL/6 background for at least 10 generations. Mice were kept in a specific pathogen-free facility and used at 6–10 wk of age. LCMV-WE was originally obtained from F. Lehmann-Grube (Heinrich Pette Institut, Universität Hamburg, Hamburg, Germany) and propagated on the fibroblast line L929. Recombinant vaccinia virus (rVV) encoding the LCMV glycoprotein (rVVG2) was obtained from D. Bishop (Institute of Virology, Oxford, U.K.) and was propagated on BSC40 cells. Mice were infected with 200 PFU of LCMV-WE i.v. or with 2 x 106 PFU of VVG2 i.p., and the specific CTL response was analyzed at day 8 or 6 after infection, respectively.
Cell lines
MC57 (H-2b) is a C57BL/6-derived methylcholanthrene-induced fibrosarcoma cell line (35). MCGP (H-2b) is a MC57-derived transfectant expressing the LCMV glycoprotein. B8 is a BALB/c-derived fibroblast line (H-2d) obtained by SV40 infection in vitro (28). B27M6 and B27M2 are triple transfectants of B8 cells expressing murine LMP2, LMP7, and MECL-1; BP13 is a double transfectant of B8 cells expressing murine PA28 and PA28 (19); BC2P6 is a transfectant of B8 cells expressing murine LMP2 (28); B7H6 is a transfectant of B8 cells expressing murine LMP7 (28). Hyb33 and Hyb276 are T cell hybridomas specific for gp33–41/H-2Kb or gp276–286/H-2Db, respectively (15). All cells were grown in IMDM supplemented with 2 mM glutamine, 10% FCS, and 100 U/ml penicillin/streptomycin. Selection drugs were required for MCGP (0.8 mg/ml G418), B27M6Db (0.5 mg/ml G418, 3 µg/ml puromycin, 0.4 mg/ml hygromycin B, 5 µg/ml blasticidin), BP13Db (0.5 mg/ml G418, 3 µg/ml puromycin, 0.4 mg/ml hygromycin B, 5 µg/ml blasticidin), B8Db (5 µg/ml blasticidin), B7H6Db and BC2P6Db (3 µg/ml puromycin, 0.4 mg/ml hygromycin B), and Hyb33 and Hyb276 (1x hypoxanthine thymidine, 0.5 mg/ml hygromycin B).
Synthetic peptides
The synthetic peptides gp33–41 (KAVYNFATC) and gp276–286 (SGVENPGGYCL) were obtained from Echaz Microcollections (Tubingen, Germany). The 25-mer peptide used for proteasome digestion encompassing LCMV-glycoprotein residues 271–295 (TLSSDSSGVEDPGGYCLTKWMILAAE) was synthesized by solid-phase peptide synthesis on a NovaSyn TGA resin (0.21 mmol/g) by Fmoc/tBu chemistry, using a semiautomated Economy Peptide Synthesizer EPS-221 (Abimed, Germany). The crude peptide was purified by reverse-phase HPLC on a C18 column (GROM-SIL 120 ODS-4 HE, 10 µm, 250 x 20 mm; Grom, Herrenberg-Kayh, Germany) using as mobile phases: eluent A (0.1% trifluoroacetic acid in water) and eluent B (80% AcCN, 0.1% trifluoroacetic acid in water). The following gradient was applied: 0 min, 10% eluent B; 45 min, 90% eluent B.
Antibodies
KL 25 is a mouse mAb reactive with the LCMV glycoprotein (36). The mAb KH95 (BD Pharmingen, San Diego, CA) reacts with the H-2Db MHC class I molecule.
Flow cytometry
A number of 5 x 105 infected and noninfected B8Db, B27M6Db, B27M2Db, B7H6Db, BC2P6Db, and BP13Db cells in 100 µl of PBS + 2% FCS were incubated in a round-bottom 96-well plate on ice for 20 min with 1 µg of mAb KL25, washed twice, and subsequently stained by a FITC-conjugated sheep anti-mouse Ig (Silenus, Victoria, Australia) for another 20 min on ice. Samples were washed twice and analyzed on a FACScan flow cytometer (BD Biosciences, Mountain View, CA). To check transfected B8Db, B27M6Db, B27M2Db, B7H6Db, BC2P6Db, and BP13Db for H-2Db expression, the staining was performed, as described above, with the H-2Db-specific Ab KH95.
For V staining splenocytes from uninfected or LCMV (8 days postinfection with 200 PFU of LCMV-WE i.v.)-infected C57BL/6, LMP2–/–, and LMP7–/– mice were treated with 1.66% NH4Cl (w/v), washed twice, and incubated for 30 min with biotin-conjugated anti-V8.1/8.2, anti-V9, or anti-V10b (BD Pharmingen) Abs on ice. Samples were washed twice and incubated for another 30 min with streptavidin-conjugated FITC and Cy5-conjugated mouse anti-CD8 (BD Pharmingen). After two washes, cells were acquired with the use of FACScan flow cytometer (BD Biosciences, Mountain View, CA) and analyzed with the FlowJo software (Tree Star, San Carlos, CA). Differences between groups were assessed by unpaired t test (www.graphpad.com). Values of p <0.05 are considered to be statistically significant.
Transfections
B8, B27M6, and BP13 were transfected with an expression plasmid encoding H-2Db (a kind gift from F. Momburg, Heidelberg, Germany). Cells were plated to 80% confluence and were transfected by the standard calcium phosphate coprecipitation method with 10 µg of H-2Db plasmid and 2 µg of blasticidin resistance vector pcDNA6/TR (Invitrogen Life Technologies, Karlsruhe, Germany). Two days after transfection, cells were plated in 96-well plates under cloning conditions and selected with 5 µg/ml blasticidin (Invitrogen Life Technologies). Because of instability of blasticidin, the selection medium was replaced every 4 days. Blasticidin-resistant cells were tested for H-2Db expression by FACS analysis, and positive cells were subcloned to obtain monoclonal cells.
B27M2, BC2P6, and B7H6 cells were stably transfected with H-2Db plasmid and either hygromycin or puromycin resistance plasmids, according to the manufacturer's protocol (FuGENE 6; Roche, Basel, Switzerland). Clonal and selection drug-resistant cells were tested for H-2Db expression by flow cytometry.
LacZ assay
For the lacZ assay, 5 x 104 LCMVgp33–41/H-2Kb- or LCMVgp276–286/H-2Db-specific T cell hybridomas (15) were cocultured overnight with 2.5 x 104 stimulator cells in 96-well plates. As stimulator cells, MCGP, B8Db, B27M6Db, B27M2Db, BC2P6Db, B7H6Db, and PB13Db, or thioglycolate-elicited peritoneal macrophages from C57BL/6, LMP2–/–, or LMP7–/– mice were used. MCGP cells were treated for 3 days with 60 U/ml murine rIFN- (Roche) to allow a complete replacement of constitutive proteasomes by immunoproteasomes (19). The lacZ-based color reaction was performed and measured, as detailed elsewhere (15).
Restimulation and cytolytic assay
B6, LMP2–/–, and LMP7–/– mice were infected with 2 x 106 VVG2 i.p. Six days later, 4 x 106 splenocytes were cocultured with 2 x 106 peptide-loaded (gp33–41 or gp276–286) irradiated (2000 rad) syngeneic splenocytes in 2 ml of IMDM supplemented with 10% FCS, penicillin/streptomycin, 2-ME, and 10% Con A supernatant (Con A-induced rat spleen culture supernatant) in 24-well plates. After 6 days of culture, a standard 51Cr release assay with peptide (gp33–41 or gp276–286)-loaded (10–6 M) or unloaded MC57 cells was performed. In brief, target cells were incubated with Na251CrO4 ± peptide for 90 min. Three-fold dilutions of the restimulated cultures were tested for cytotoxic activity using 104 51Cr-labeled MC57 as targets in a 5-h chromium release assay. The percentage of specific lysis was calculated as follows: percentage of specific release = (experimental release –spontaneous release) ÷ (maximal release –spontaneous release) x 100%. The spontaneous lysis was below 15%; the lysis of unloaded targets was below 10%.
Intracellular staining for IFN-
A total of 2 x 106 splenocytes was incubated in round-bottom 96-well plates with 10–7 M of the specific peptide in 100 µl of 10% IMDM for 2 h at 37°C. Then brefeldin A (10 µg/ml) was added, and the incubation was continued for another 4 h. Cells were incubated for 20 min at 4°C with Cy5-conjugated mouse anti-CD8 (clone 53-6.7; BD Pharmingen). Following fixation with 4% paraformaldehyde at 4°C for 5 min, the cells were incubated overnight with fluorescein-conjugated mouse anti-IFN- (clone XMG1.2; BD Pharmingen) in PBS containing 2% FCS and 0.1% (w/v) saponin (Sigma-Aldrich, St. Louis, MO). Samples were washed twice, acquired with the use of FACScan flow cytometer (BD Biosciences), and analyzed by the FlowJo software (Tree Star).
Metabolic labeling, immunoprecipitation, and two-dimensional gel electrophoresis
A total of 107 thioglycolate-elicited peritoneal macrophages from BALB/c mice was starved in cysteine/methionine-free RPMI 1640, 10% dialyzed FCS for 1 h at 37°C and labeled with 0.2 mCi/ml Met-[35S]-label (Hartmann Analytic, Braunschweig, Germany) for 8 h to allow full maturation of the proteasome. Cells were washed with PBS, harvested, and lysed for 30 min on ice in 20 mM Tris/HCl, pH 7.5, 150 mM NaCl, 1 mM MgCl2, and 2% Triton X-100. The lysate was precleared for 1 h with protein A-Sepharose CL-4B (Amersham Biosciences, Uppsala, Sweden), followed by overnight immunoprecipitation with an antiproteasome serum bound to protein A-Sepharose at 4°C. The precipitates were washed four times with NET-TON (650 mM NaCl, 5 mM EDTA, 50 mM Tris-HCl, 0.5% Triton-X-100, 0.05% NaN3, 1 mg/ml OVA) and twice with NET-T (150 mM NaCl, 5 mM EDTA, 50 mM Tris-HCl, 0.5% Triton X-100, 0.05% NaN3), and separated by nonequilibrium pH-gradient gel electrophoresis/SDS-PAGE, as previously described (28), and visualized by autoradiography on a Fuji BAS1500 radioimager.
Purification of 20S proteasome
The lysis of organ tissue, the purificaton of 20S proteasomes from liver, and the quantification of the 20S proteasome from uninfected and LCMV-infected (8 days postinfection with 200 PFU of LCMV-WE i.v.) C57BL/6, LMP2–/–, and LMP7–/– mouse livers were performed, as previously described (28).
Proteasomal fragmentation of polypeptide and mass spectrometric analysis of peptide products
Digestions of the 25-mer polypeptide spanning LCMV-WE glycoprotein residues 271–295 with purified 20S proteasomes were performed for indicated time periods exactly as previously described (19). HPLC-electrospray ionization mass spectrometry (ESI-MS) analysis was performed on a Q-Trap mass spectrometer (Applied Biosystems, Concord, Canada) equipped with Turbospray Ion Source (Applied Biosystems) and Agilent (Palo Alto, CA) 1100 HPLC system (Autosampler G1313A, Binary Pump G1312A, on-line vacuum degasser G1322A), ACCURATE flow splitter AC-100-VAR with 0.3 mm I.D. splitter capillary CAL-100-0.3 (LC Packings, Amsterdam, Holland), and control software: Analyst 1.3.2 (Applied Biosystems) under Windows 2000 operating system. HPLC conditions: mobile phase solvent A 0.1% aqueous solution of HCOOH, solvent B 0.1% solution of HCOOH in acetonitrile with following gradient: 0 min, 0% B; 4 min, 0% B; 44 min, 80% B. Binary pump flow 100 µl/min and after splitting (on column) 4–10 µl/min. The HPLC column was PepMap C18, 3 µm, 100 Å, 300 µm I.D. x 5 cm (LC Packings). MS conditions: scan type: enhanced MS; positive mode; mass range 340-1400 amu; declustering potential 30 V; entrance potential 10 V; curtain gas 20 PSI; ion source gas 1: 15 PSI; ion spray voltage 5000 V; source temperature 200°C. Sample injection volume 1 µl.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
on the presentation of gp33–41 and gp276–286
The stimulation of cells with IFN- strongly induces many proteins along the MHC class I presentation pathway, thus leading to a marked up-regulation of class I cell surface expression (16). When we infected the C57BL/6-derived mouse fibroblast line MC57 with the LCMV strain WE and treated the cells with or without murine IFN- for 3 days, we observed that IFN- caused a 2- to 3-fold up-regulation in the presentation of the immunodominant epitope gp33 as determined with a gp33-specific T cell hybridoma in a lacZ-based colorigenic assay (37) (Fig. 1A). Despite a 
10-fold increase of bulk H-2Db and H-2Kb class I cell surface expression (data not shown), the presentation of the subdominant gp276 epitope was consistently down-regulated by IFN- treatment by a factor of 2 in the same assay. Although we confirmed that the cells were equally infected with LCMV by determining the cell surface expression of the LCMV glycoprotein by flow cytometry (data not shown), we wanted to rule out that the obtained result is linked to IFN--dependent changes in viral replication and Ag production. We hence repeated this experiment with the MCGP line, which is a stable MC57 transfectant expressing the LCMV glycoprotein. Also in this virus-independent model, we observed an IFN--dependent up-regulation of gp33 presentation, whereas presentation of the gp276 epitope was markedly reduced (Fig. 1B).
|
on the presentation of gp276–286
Numerous gene products are induced with IFN-, which could account for an enhancement of gp33 presentation, but only few candidates are known that could be responsible for the down-regulation of gp276 presentation. Given that immunoproteasomes have been shown to interfere with the presentation of some tumor epitopes (22), we examined whether the IFN--inducible proteasome subunits LMP2, LMP7, and MECL-1, or the IFN--inducible proteasome activator PA28 could mediate the observed down-regulation of gp276 presentation. Previously, we have generated and carefully characterized stable transfectants of the mouse fibroblast line B8, which coexpress either all three active site subunits of immunoproteasomes (designated B27M6 or B27M2) or the - and -subunits of the proteasome regulator PA28 (designated BP13) to the same degree as B8 fibroblasts after 3 days of IFN- treatment (19). Because the recipient line B8 is derived from a BALB/c mouse (H-2d), we supertransfected the recipient line B8 as well as B27M6 and BP13 cells with an expression construct encoding the H-2Db restriction element. Clones named B8Db, B27M6Db, B27M2Db, and BP13Db, respectively, were selected that, according to flow cytometry, expressed similar levels of H-2Db class I molecules on the cell surface (Fig. 2B). Interestingly, when these clones were infected for 1.5 days with LCMV-WE and examined for gp276 presentation with our gp276/H-2Db-specific hybridoma, we found that the combined overexpression of LMP2, LMP7, and MECL-1 in B27M6Db and B27M2Db cells reduced gp276 presentation 2- to 3-fold compared with B8Db cells, whereas PA28 overexpression had only a minor effect (Fig. 2, A and C). This result, which was confirmed with independent clones, strongly suggested that the induction of immunoproteasomes is at least in part responsible for the IFN--mediated down-regulation of gp276.
|
4-fold, which indicates that the reduced presentation of gp276 in LMP2/LMP7/MECL-1 cells (Fig. 2, A and C) is mainly due to LMP2. Comparison of gp33 and gp276 presentation by LCMV-infected macrophages from C57BL/6, LMP2–/–, and LMP7–/– mice
To confirm this effect of immunoproteasomes in an independent system and to discriminate whether LMP2 or LMP7 contributes to down-regulation of gp276 presentation, we examined Ag presentation by LCMV-infected thioglycolate-elicited peritoneal macrophages from LMP2–/– and LMP7–/– gene-targeted as well as C57BL/6 control mice. The peritoneal macrophages were infected in vitro for 20 h with LCMV-WE and analyzed for gp33 and gp276 presentation with the respective T cell hybridomas (Fig. 3) using peptide-pulsed MC57 cells as specificity controls (lanes 1 and 2) and uninfected peritoneal macrophages from C57BL/6, LMP2–/–, and LMP7–/– mice as negative controls (lanes 3–5). Although no difference in gp33 presentation was found between C57BL/6 and LMP2–/– macrophages, gp33 presentation was reduced by 50% in LCMV-WE-infected macrophages from LMP7–/– mice. This result suggests that the induction of LMP7, but not of LMP2, contributes to the IFN--dependent enhancement of gp33 presentation (Fig. 1A). The presentation of gp276, in contrast, seems to benefit from the deficiency of LMP7 and to a minor extent from LMP2 as macrophages from the respective knockout mice stimulated gp276-specific hybridomas better than macrophages from the wild-type control. This result is consistent with the aforementioned reduction in gp276 presentation caused by immunoproteasome overexpression (Fig. 2B).
|
(containing 11 Cys/Met) (Fig. 3B). This might explain the only minor effect of LMP2 deficiency on gp276 presentation by peritoneal macrophages. The generation of gp276-specific CTLs is improved in LMP2–/– and LMP7–/– mice
To investigate how LMP2 and LMP7 would affect the CTL response to gp33, gp276, and NP396 in vivo, we first infected LMP2–/–, LMP7–/–, and PA28–/– knockout as well as C57BL/6 control mice with LCMV. However, standard 51Cr release assays performed on day 8 postinfection with peptide-loaded target cells did not detect significant differences between these mice strains irrespective of whether we infected with LCMV-WE or LCMV-Armstrong (data not shown). Moreover, virus titers in the spleen obtained on day 4 postinfection were not different in control and knockout mice (Table I). These results are not unexpected given that LCMV replicates very fast in numerous mouse tissues to high titers and overwhelms the organism with viral Ags. Hence, we turned to infection of these mice with rVVG2. VV replicates much slower in mice, and therefore produces a lower amount of Ag. Splenocytes from rVVG2-infected mice were harvested on day 6 after infection, and the CTLs were restimulated in vitro for another 6 days before they were used as effectors in a chromium release assay. As shown in Fig. 4A, the lysis of targets loaded with the gp276 peptide by LMP2–/– and LMP7–/– effectors was much higher (40–75%) as compared with CTLs from C57BL/6 mice (15–25%). This indicates that, in agreement with our in vitro data, the gp276 epitope became immunodominant in the absence of either LMP2 or LMP7. The lysis of gp33-loaded targets varied between 20 and 40% for all mice, thus indicating that the replacement of constitutive proteasomes by immunoproteasomes did not significantly affect the generation of gp33-specific CTLs. Even with this more sensitive in vivo system, we failed to detect any differences between PA28–/– and C57BL/6 wild-type mice with respect to the generation of either gp33- or gp276-specific CTLs (data not shown).
|
|
(intracellular cytokine staining) on day 7 after infection with rVVG2. Also with this ex vivo assay, we observed that the generation of gp276-specific precursors was clearly enhanced in LMP2–/– and LMP7–/– mice (Fig. 4, B and C). In contrast, there was no difference in thegeneration of gp33-specific CTLs in LMP2- and LMP7-deficient mice, compared with wild-type C57BL/6 mice (Fig. 4B). Taken together, it appears that the down-regulation of gp276 epitope presentation through LMP2 and LMP7 in vitro nicely correlates with an enhanced CTL elicitation in the respective gene-targeted mice.
The generation of V10b-specific CTLs is impaired in LCMV-infected LMP2- and LMP7-deficient mice
To investigate whether the improved generation of gp276-specific CTLs in LMP2- and LMP7-deficient mice is due to an altered CTL repertoire, splenocytes from naive and LCMV-WE-infected (8 days postinfection with 200 PFU of LCMV-WE) C57BL/6, LMP2–/–, and LMP7–/– mice were stained with different TCR-V-specific Abs (Fig. 5). There was a significant difference (2% less) of V8.1/8.2-specific CD8-positive cells in naive LMP2-deficient mice compared with C57BL/6 and LMP7-deficient mice. After LCMV infection, no difference could be detected for V8. Naive LMP7–/– mice showed a slight, but significant difference for V9-specific CTLs, which was abolished after LCMV infection. It has been shown that T cell lines specific for gp276 were using exclusively the V10 variable segment for their TCRs (38). Naive LMP2–/– and LMP7–/– mice showed no difference in V10 usage compared with C57BL/6. In contrast, after LCMV infection, the extent of CTLs using V10 was significantly increased in C57BL/6 mice compared with LMP2- and LMP7-deficient mice.
|
Given that the generation of the gp276 epitope is dependent on proteasome activity (19), we hypothesized that gp276 presentation is adversely affected by LMP2 and LMP7 because constitutive proteasomes and immunoproteasomes fragment precursor polypeptides of gp276 in a different way. To test this hypothesis, we investigated how immunoproteasomes and constitutive 20S proteasomes fragment the 25-mer precursor polypeptide covering residues 271–295 of the LCMV-WE glycoprotein (NH2-TLSDSSGVEDPGGYCLTKWMILAAE-COOH), which contains the underlined 11-meric gp276–286 epitope bearing an aspartate in position 5, which resulted from rapid deamidation of asparagine after systhesis (Fig. 6A). We isolated 20S proteasomes from the liver of uninfected mice as a source of constitutive proteasomes. Immunoproteasomes were isolated from the liver of mice on day 8 after LCMV infection. The subunit composition of the two proteasome populations on two-dimensional gels confirmed our previous finding that LCMV infection results in a virtually complete replacement of constitutive proteasomes by immunoproteasomes in vivo (39) (data not shown). The separation of the produced fragments after 8 h of in vitro digest by HPLC shows that constitutive proteasomes and immunoproteasomes fragment the 25-mer polypeptide in a different manner, and these differences in the HPLC profiles were confirmed in several independent experiments (Fig. 6B). The fragmentation was not observed in the presence of the proteasome inhibitor lactacystin, suggesting that 20S proteasomes were not contaminated by other proteases (data not shown). The time period of 8 h was chosen because the 25-mer substrate eluting late in the gradient is still by far the predominant peptide and the relative intensities of emerging peaks remained the same over a digestion period of up to 8 h. A further fragmentation of primary fragments was therefore unlikely to occur.
|
To address the effect of LMP2 only and LMP7 only, 20S proteasome was isolated from liver of LCMV-infected LMP2- and LMP7-deficient mice. Two-dimensional gels confirmed that the two other immunoproteasome subunits were induced (data not shown). Analysis of how the 25 mer was fragmented was obtained exactly as in Fig. 6C. The LCMV-derived epitope gp276–286 and its putative precursor gp275–286 were produced in larger amounts by LMP2- and LMP7-deficient immunoproteasome compared with normal immunoproteasome (Fig. 6D), which is in accordance with the results obtained in Fig. 4, in which LMP2- and LMP7-deficient mice elicited a stronger CTL response against gp276 than wild-type mice.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
10-fold lower concentration of gp276 was required to achieve optimal lysis (10, 11). Also, the recognition and elimination of LCMV-infected target cells by gp276-specific CTLs seem to be more efficacious compared with gp33-specific CTLs, as evidenced by adoptive transfer experiments, thus indicating that the binding of the gp276-specific TCRs to H-2Db/gp276 complexes on the surface of LCMV-infected cells is not a limitation (11). A factor that was shown to be a determinant of immunodominance at least in influenza infection is the availability of specific CTLs in the repertoire of peripheral T cell specificities (3, 25). T cell lines specific for gp276 were strongly biased for the usage of V4 and V10 variable segements for their TCRs (38), but it has not been investigated whether this bias imposes a limit on the availability of gp276-specific T cells in the repertoire.
Finally, gp33, gp276, and NP396 epitopes were eluted from H-2Db proteins of LCMV-infected MC57 fibroblast cells, and the approximate number of epitopes per cell was determined. The calculated numbers were 1080 for gp33, 92 for gp276, and 162 for NP396, which implies that the copies of gp276 and NP396 epitopes generated in MC57 cells and presented on the cell surface are in a range, in which recognition by CTLs could become limiting (13, 14). However, it must be pointed out that MC57 cells infected with LCMV in vitro may not be representative for the physiological situation because infected cells in vivo are confronted with IFN- produced by NK cells and Th1 cells, which has dramatic effects on the class I processing and presentation pathway. In particular, we have shown that the infection of mice with LCMV, with mouse CMV, or with Listeria monocytogenes results in an IFN--dependent and almost complete replacement of constitutive proteasomes by immunoproteasomes in the liver as well as other organs on day 7 or 8 after infection (39) (our unpublished data). We found that MC57 fibroblasts, even when they were infected with LCMV in vitro and hence produced type I IFNs, expressed mostly constitutive proteasomes (39). Moreover, not fibroblasts, but dendritic cells are the crucial APCs in the priming phase of the immune response, and these were shown to express high levels of immunoproteasomes (41, 42).
Hence, we set out to investigate immunoproteasomes as determinants of epitope hierarchy in the LCMV system. Our finding that IFN- treatment of MC57 cells resulted in enhanced gp33 presentation, whereas gp276 presentation was reduced (Fig. 1), inspired us to perform the experiments described in this work. An in vivo correlate to this finding has recently been reported by Rodriguez et al. (43), who showed that gp276-specific CTLs are much more prominent in IFN--deficient mice, but the gp33-specific CTL response was improved in these mice as well. Another interesting phenomenon was reported by Butz and Bevan (44). They noted that when CTLs from LCMV-infected C57BL/6 mice were weekly restimulated in vitro by LCMV-infected MC57 cells, gp276-specific CTLs outgrew gp33- and NP396-specific cells within 3 wk. However, when the restimulation was performed with the dendritic cell line JawsII, gp276-specific CTLs waned, and gp33- as well as NP396-specific CTLs predominated. Given that IFN- stimulation leads to an almost complete replacement of constitutive proteasomes by immunoproteasomes in mouse fibroblasts within 3 days and that dendritic cells constitutively express higher amounts of immunoproteasomes than fibroblasts, we hypothesized that immunoproteasomes may be responsible for both phenomena by down-regulating gp276 peptide presentation. This hypothesis turned out to be correct, as we have shown by two independent approaches. First, we demonstrated that the overexpression of the active site subunits of immunoproteasomes LMP2, LMP7, and MECL-1 in triple transfectants caused a 2- to 4-fold down-regulation of gp276 presentation in T cell hybridoma lacZ assays (Fig. 2, A and C). Second, we showed that LCMV-infected peritoneal macrophages from LMP7–/– mice and to a minor extent from LMP2–/– were better stimulators of gp276-specific T cell hybridomas than C57BL/6 wild-type macrophages. LMP2- and LMP7-overexpressing cells indicated that mostly LMP2 is responsible for the down-regulation of gp276 in LMP2, LMP7, and MECL-1 triple transfectant (Fig. 2C). Comparing the fragmentation of a gp276-containing polypeptide by constitutive proteasomes and immunoproteasomes in vitro suggests that immunoproteasomes produce less gp276 precursors and destroy the gp276 epitope more frequently than constitutive proteasomes through cleavages within the epitope (Fig. 6).
Taken together, it appears that the expression of LMP2 and LMP7 negatively affects the generation of the gp276 epitope. It is, however, not possible to assign this effect to one of the subunits exclusively because their incorporation into the proteasome is to some extent interdependent. We have analyzed the subunit composition of 20S proteasomes purified from livers of LCMV-infected LMP2–/– and LMP7–/– as well as C57BL/6 wild-type mice on Coomassie-stained two-dimensional gels and found that LMP7 incorporation fully occurs in LMP2–/– mice in accordance with previous data (45, 46, 47). However, because the incorporation of the subunit MECL-1 barely occurs when LMP2 is missing, the effects observed in LMP2–/– mice could also be caused by a lack of MECL-1. In contrast, we found that the incorporation of LMP2 and MECL-1 in livers of LCMV-infected LMP7–/– mice occurs only to an extent of 50%, which means that the more prominent enhancement of gp276 presentation in LMP7-deficient macrophages may at least in part be attributed to a reduction in LMP2 and MECL-1 incorporation (data not shown).
One consequence of our finding is that the number of 92 gp276 epitopes that Gallimore et al. (11) calculated to be presented by a single LCMV-infected MC57 cell will probably be much lower in cells containing immunoproteasomes, which almost completely replace constitutive proteasomes during LCMV infection in vivo. It is therefore not unexpected that the further down-regulation of gp276 epitopes by immunoproteasomes reduces gp276 epitope generation to an extent that puts a limit on the generation of gp276-specific CTLs in LCMV-infected mice. Nevertheless, the effect of immunoproteasomes on the generation of gp276-specific CTLs is not as prominent as we expected. It was not observed when LMP2–/– and LMP7–/– mice were infected with either the faster replicating WE strain or the slower replicating Armstrong strain of LCMV. However, when the mice were infected with rVVG2, which produces lower amounts of the LCMV glycoprotein, the impact of immunoproteasomes was nicely detectable (Fig. 4). We cannot rule out, however, that in addition to the effect on gp276 presentation, the greater number of gp276-specific CTLs in LMP2–/– and LMP7–/– knockout mice is due to a difference in the repertoire of peripheral T cells in these mice, as has been demonstrated for the influenza epitope NP366–374 in LMP2–/– mice (25). Differences in V10 usage of LCMV-infected LMP2- and LMP7-deficient mice might indicate that an altered T cell repertoire of gp276-specific T cells exists in these mice (Fig. 5).
Still another interesting phenomenon observed in mice persistently infected with LCMV may be linked to our results. In two different models of chronic LCMV infection, it was recently found that the gp276 epitope that is subdominant in acute infection becomes the immunodominant epitope in chronically infected mice, whereas gp33 and NP396 drop deeply in epitope hierarchy (48, 49). Both research groups report that in chronic infection, the LCMV-specific CTLs stop to produce TNF- and IFN-, which are the cardinal inducers of immunoproteasomes. We propose that this reversion in epitope hierarchy is at least in part due to a drop in the cellular content of immunoproteasomes.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 This work was supported by grants of the Swiss National Science Foundation (31-52284.97/1) and the University of Constance to M.G., and by a grant from the Cloëtta Foundation, Zürich, to M.v.d.B. 
2 Address correspondence and reprint requests to Dr. Marcus Groettrup, Department of Biology, Division of Immunology, University of Constance, P1101 Universitätsstrasse 10, D-78457 Konstanz, Germany. E-mail address: Marcus.Groettrup{at}uni-konstanz.de
3 Abbreviations used in this paper: LCMV, lymphocytic choriomeningitis virus; ESI-MS, electrospray ionization mass spectrometry; NP, nucleoprotein; VV, vaccinia virus.
Received for publication October 30, 2003. Accepted for publication July 2, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/ enhances the presentation of an immunodominant lymphocytic choriomeningitis virus T cell epitope. J. Immunol. 165:768.-inducible 11S regulator (PA28) and the LMP2/LMP7 subunits govern the peptide production by the 20S proteasome in vitro. J. Biol. Chem. 270:23808. in antigen presentation. Nature 381:166.[Medline]
/ can enhance antigen presentation without affecting 20S proteasome subunit composition. Eur. J. Immunol. 30:3672.[Medline]
and PA28. EMBO J. 20:5898.[Medline]
4 and V10 T cell receptor genes by lymphocytic choriomeningitis virus glycoprotein-specific H-2Db-restricted cytotoxic T cells. Eur. J. Immunol. 20:523.[Medline]
/ proteasome regulator but little increase in the steady state content of immunoproteasomes during dendritic cell maturation. Eur. J. Immunol. 31:3271.[Medline]
interferon. J. Virol. 76:4251. (IFN-)-inducible subunits. J. Exp. Med. 187:97.2 subunit propeptides influence cooperative proteasome assembly. J. Biol. Chem. 278:6153.This article has been cited by other articles:
|
|
 |
|
  N. J. Steers, K. K. Peachman, S. R. McClain, C. R. Alving, and M. Rao Human Immunodeficiency Virus Type 1 Gag p24 Alters the Composition of Immunoproteasomes and Affects Antigen Presentation J. Virol., July 15, 2009; 83(14): 7049 - 7061. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Riedl, A. Wieland, K. Lamberth, S. Buus, F. Lemonnier, K. Reifenberg, J. Reimann, and R. Schirmbeck Elimination of Immunodominant Epitopes from Multispecific DNA-Based Vaccines Allows Induction of CD8 T Cells That Have a Striking Antiviral Potential J. Immunol., July 1, 2009; 183(1): 370 - 380. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Strehl, K. Textoris-Taube, S. Jakel, A. Voigt, P. Henklein, U. Steinhoff, P.-M. Kloetzel, and U. Kuckelkorn Antitopes Define Preferential Proteasomal Cleavage Site Usage J. Biol. Chem., June 27, 2008; 283(26): 17891 - 17897. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hallermalm, K. Seki, A. De Geer, B. Motyka, R. C. Bleackley, M. J. Jager, C. J. Froelich, R. Kiessling, V. Levitsky, and J. Levitskaya Modulation of the Tumor Cell Phenotype by IFN-{gamma} Results in Resistance of Uveal Melanoma Cells to Granule-Mediated Lysis by Cytotoxic Lymphocytes J. Immunol., March 15, 2008; 180(6): 3766 - 3774. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. F. Kotturi, B. Peters, F. Buendia-Laysa Jr., J. Sidney, C. Oseroff, J. Botten, H. Grey, M. J. Buchmeier, and A. Sette The CD8+ T-Cell Response to Lymphocytic Choriomeningitis Virus Involves the L Antigen: Uncovering New Tricks for an Old Virus J. Virol., May 15, 2007; 81(10): 4928 - 4940. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. Robek, M. L. Garcia, B. S. Boyd, and F. V. Chisari Role of Immunoproteasome Catalytic Subunits in the Immune Response to Hepatitis B Virus J. Virol., January 15, 2007; 81(2): 483 - 491. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. O. Weinzierl, C. Lemmel, O. Schoor, M. Muller, T. Kruger, D. Wernet, J. Hennenlotter, A. Stenzl, K. Klingel, H.-G. Rammensee, et al. Distorted Relation between mRNA Copy Number and Corresponding Major Histocompatibility Complex Ligand Density on the Cell Surface Mol. Cell. Proteomics, January 1, 2007; 6(1): 102 - 113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. C. Pang, M. T. Sanders, J. J. Monaco, P. C. Doherty, S. J. Turner, and W. Chen Immunoproteasome Subunit Deficiencies Impact Differentially on Two Immunodominant Influenza Virus-Specific CD8+ T Cell Responses J. Immunol., December 1, 2006; 177(11): 7680 - 7688. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Strehl, T. Joeris, M. Rieger, A. Visekruna, K. Textoris-Taube, S. H. E. Kaufmann, P.-M. Kloetzel, U. Kuckelkorn, and U. Steinhoff Immunoproteasomes Are Essential for Clearance of Listeria monocytogenes in Nonlymphoid Tissues but Not for Induction of Bacteria-Specific CD8+ T Cells J. Immunol., November 1, 2006; 177(9): 6238 - 6244. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. A. York, M. A. Brehm, S. Zendzian, C. F. Towne, and K. L. Rock Endoplasmic reticulum aminopeptidase 1 (ERAP1) trims MHC class I-presented peptides in vivo and plays an important role in immunodominance PNAS, June 13, 2006; 103(24): 9202 - 9207. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Basler, J. Moebius, L. Elenich, M. Groettrup, and J. J. Monaco An Altered T Cell Repertoire in MECL-1-Deficient Mice. J. Immunol., June 1, 2006; 176(11): 6665 - 6672. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Chapatte, M. Ayyoub, S. Morel, A.-L. Peitrequin, N. Levy, C. Servis, B. J. Van den Eynde, D. Valmori, and F. Levy Processing of Tumor-Associated Antigen by the Proteasomes of Dendritic Cells Controls In vivo T-Cell Responses. Cancer Res., May 15, 2006; 66(10): 5461 - 5468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Szalay, S. Meiners, A. Voigt, J. Lauber, C. Spieth, N. Speer, M. Sauter, U. Kuckelkorn, A. Zell, K. Klingel, et al. Ongoing Coxsackievirus Myocarditis Is Associated with Increased Formation and Activity of Myocardial Immunoproteasomes Am. J. Pathol., May 1, 2006; 168(5): 1542 - 1552. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Osterloh, K. Linkemann, S. Tenzer, H.-G. Rammensee, M. P. Radsak, D. H. Busch, and H. Schild Proteasomes shape the repertoire of T cells participating in antigen-specific immune responses PNAS, March 28, 2006; 103(13): 5042 - 5047. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Ito, E. Kondo, A. Demachi-Okamura, Y. Akatsuka, K. Tsujimura, M. Tanimoto, Y. Morishima, T. Takahashi, and K. Kuzushima Three Immunoproteasome-Associated Subunits Cooperatively Generate a Cytotoxic T-Lymphocyte Epitope of Epstein-Barr Virus LMP2A by Overcoming Specific Structures Resistant to Epitope Liberation J. Virol., January 15, 2006; 80(2): 883 - 890. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Chapiro, S. Claverol, F. Piette, W. Ma, V. Stroobant, B. Guillaume, J.-E. Gairin, S. Morel, O. Burlet-Schiltz, B. Monsarrat, et al. Destructive Cleavage of Antigenic Peptides Either by the Immunoproteasome or by the Standard Proteasome Results in Differential Antigen Presentation J. Immunol., January 15, 2006; 176(2): 1053 - 1061. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Bruder, A. K. Nussbaum, D. M. Gakamsky, M. Schirle, S. Stevanovic, H. Singh-Jasuja, A. Darji, T. Chakraborty, H. Schild, I. Pecht, et al. Multiple synergizing factors contribute to the strength of the CD8+ T cell response against listeriolysin O Int. Immunol., January 1, 2006; 18(1): 89 - 100. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K. Nussbaum, M. P. Rodriguez-Carreno, N. Benning, J. Botten, and J. L. Whitton Immunoproteasome-Deficient Mice Mount Largely Normal CD8+ T Cell Responses to Lymphocytic Choriomeningitis Virus Infection and DNA Vaccination J. Immunol., July 15, 2005; 175(2): 1153 - 1160. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, Represents a VVG2-infected mouse without in vitro peptide stimulation. The y-axis shows intracellular IFN-
