Elimination of Immunodominant Epitopes from Multispecific DNA-Based Vaccines Allows Induction of CD8 T Cells That Have a Striking Antiviral Potential This information is current as of September 29, 2021. Petra Riedl, Andreas Wieland, Kasper Lamberth, Soren Buus, Francois Lemonnier, Kurt Reifenberg, Jörg Reimann and Reinhold Schirmbeck J Immunol 2009; 183:370-380; ; doi: 10.4049/jimmunol.0900505 Downloaded from http://www.jimmunol.org/content/183/1/370

References This article cites 63 articles, 30 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/183/1/370.full#ref-list-1

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists by guest on September 29, 2021 • Fast Publication! 4 weeks from acceptance to publication

*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 © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Elimination of Immunodominant Epitopes from Multispecific DNA-Based Vaccines Allows Induction of CD8 T Cells That Have a Striking Antiviral Potential1

Petra Riedl,2* Andreas Wieland,2* Kasper Lamberth,† Soren Buus,† Francois Lemonnier,‡ Kurt Reifenberg,§ Jo¨rg Reimann,* and Reinhold Schirmbeck3*

Immunodominance limits the TCR diversity of specific antiviral CD8 T cell responses elicited by vaccination or . To prime multispecific T cell responses, we constructed DNA vaccines that coexpress chimeric, multidomain Ags (with CD8 T cell-defined epitopes of the (HBV) surface (S), core (C), and polymerase (Pol) proteins and/or the OVA Ag as stress protein- capturing fusion proteins. Priming of mono- or multispecific, HLA-A*0201- or Kb-restricted CD8 T cell responses by these DNA b b b vaccines differed. K /OVA257–264- and K /S190–197-specific CD8 T cell responses did not allow priming of a K /C93–100-specific CD8 Downloaded from T cell response in mice immunized with multidomain vaccines. Tolerance to the S- Ag in transgenic Alb/HBs mice (that express b large amounts of transgene-encoded S- Ag in the liver) facilitated priming of subdominant, K /C93–100-specific CD8 T cell im- b munity by multidomain Ags. The “weak” (i.e., easily suppressed) K /C93–100-specific CD8 T cell response was efficiently elicited by a HBV core Ag-encoding vector in 1.4HBV-Smut tg mice (that harbor a replicating HBV that produces HBV surface, b mut core, and precore Ag in the liver). K /C93–100-specific CD8 T cells accumulated in the liver of vaccinated 1.4HBV-S transgenic mice where they suppressed HBV replication. Subdominant epitopes in vaccines can hence prime specific CD8 T cell immunity http://www.jimmunol.org/ in a tolerogenic milieu that delivers specific antiviral effects to HBV-expressing hepatocytes. The Journal of Immunology, 2009, 183: 370–380.

nfections with hepatotropic , e.g., that high levels of circulating “free” HBV Ags and/or viral parti- (HBV)4 or virus, often persist in the liver (1). cles induce peripheral tolerance and/or suppress antiviral immu- I Factors of the pathogen (e.g., genotype) and the host (e.g., nity (11–13). Defects in anti-HBV immunity are reversible as age or immune status) critically influence if such an infection is “spontaneous” or (IFN or lamivudine) induced reconstitution of cleared or persists chronically (2). Patients who control infection antiviral immunity has been reported in chronically HBV-infected by guest on September 29, 2021 usually mount vigorous multispecific CD8 T cell responses to patients (14–16). HBV (3, 4). A small number of HLA-A*0201-restricted HBV Reconstitution of antiviral immunity by vaccination is an attrac- epitopes quantitatively dominated the CD8 T cell response to HBV tive option for the specific control of chronic HBV infection. Al- in patients with acute hepatitis (5–7), indicating that individual though many experimental, therapeutic HBV vaccines have been patients develop immunological hierarchies of HBV-specific T cell developed (17–21), these have failed so far to reproducibly clear responses (8). In contrast, patients who develop chronic HBV in- chronic HBV infection. The major problem for therapeutic HBV fection tend to show low and transient antiviral T cell responses vaccines is the delivery of Ags into the tolerogen milieu of chron- (8). Development of chronic HBV can also result from ically infected patients. Vaccine-induced cellular immune re- defects in the local delivery of antiviral effector functions by spe- sponses often show only a limited repertoire of specific CD8 T cell cific CD8 T cells (2, 9). Prominent among these are IFNs and TNF responses because they are suppressed by immunodominance that down-regulate HBV replication (10). It is generally assumed d mechanisms (22, 23). We reported that L /S28–39-restricted CD8 T cell responses efficiently suppress copriming to other HBV surface d b *Department of Internal Medicine I, University of Ulm, Germany; †Laboratory of Ag (HBsAg)-specific D - and K -restricted CD8 T cell responses Experimental Immunology, University of Copenhagen, Denmark; ‡Institute Pasteur, (24). This indicated that these HBV-specific CD8 T cell responses Unite dЈImmunite Cellulaire Antivirale, Paris, France; and §Central Laboratory An- imal Facility, University of Mainz, Germany are subjected to regulatory immunodominance control. Moreover, Received for publication February 13, 2009. Accepted for publication May 4, 2009. adenovirus-specific CD8 T cell responses down-modulate sub- d b d The costs of publication of this article were defrayed in part by the payment of page dominant (D - and K -restricted) but not dominant (L -restricted) charges. This article must therefore be hereby marked advertisement in accordance CD8 T cell responses to HBsAg (25). Dominant CD8 T cell re- with 18 U.S.C. Section 1734 solely to indicate this fact. sponses thus strikingly limit the diversity of HBsAg-specific CD8 1 This work was supported by grants from the Deutsche Forschungsgemeinschaft (Ri T cell immunity. Identification and reconstitution of subdominant 1297/1-1 to P.R. and Schi 505/2H4 to R.S.). CD8 T cell immunity is thus an attractive option in the design of 2 P.R. and A.W. contributed equally to this work. therapeutic HBV vaccines. 3 Address correspondence and reprint requests to Dr. Reinhold Schirmbeck, Depart- ment of Internal Medicine I, University of Ulm, Albert Einstein Allee 23, D-89081 Transgenic (tg) mouse models have been used to study chronic Ulm, Germany. E-mail address: [email protected] HBV infection of the liver and the establishment of local antiviral 4 Abbreviations used in this paper: HBV, hepatitis B virus; HBsAg or S, HBV surface T cell immunity in an immunologically well-defined model. The Ag; HBcAg, or C, HBV core Ag; HBeAg, HBV e Ag; Pol, polymerase; tg, transgenic; first generation models expressed subgenomic HBV fragments un- Hsp, heat shock protein. der heterologous, liver-specific control (26–29). The sec- Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 ond generation models expressed a full, replicating HBV genome www.jimmunol.org/cgi/doi/10.4049/jimmunol.0900505 The Journal of Immunology 371 Downloaded from

FIGURE 1. Expression of Hsp-Ag complexes. A, Schematic presentation of expression vectors encoding single (pCI/S, pCI/C, pCI/Pol, and pCI/OVA) or multidomain (pCI/HB-multi) Ags. The multidomain consists of the following Ag fragments: S2–50, Pol131–160, Pol 791–832, OVA246–353,C79–149,C10–50, and S140–226. The Strep tag, the Hsp-binding SV40-T77-derived DnaJ-like sequence, and the MHC class I-restricted epitopes (epitopes 1–8; for detail see Table I) are indicated. B, HEK293 cells were transiently transfected with pCI/HB-multi (lane 1), pCI/HB-multi⌬DnaJ (in which the Hsp-binding DnaJ is deleted; lane 2), pCI (lane 3), and pCI/HB-multi⌬OVA (in which the OVA sequence is deleted; lane 4). Cells were labeled with [35S]methionine, lysed, http://www.jimmunol.org/ and the respective Ags were immunopurified with Strep-Tactin-Sepharose followed by release of Ag complexes with SDS. Samples were analyzed by SDS-PAGE and fluorography of the gels. The positions of the HB-multi, HB-multi⌬DnaJ, and HB-multi⌬OVA fusion proteins and the cellular Hsp73 are indicated. in the liver of tg mice (30–32). These models allowed important of Ulm University (Ulm, Germany). Male and female mice were used at insight into viral and host factors involved in HBV pathogenesis 12–16 wk of age. We used 1.4HBV-Smut tg males for immunization studies and replication. Most striking was the observation that the immune because HBV replication was reproducible higher and shows less variabil- ity than in the liver of tg females (32). All animal experiments were per- by guest on September 29, 2021 response can noncytolytically clear HBV from hepatocytes (33). formed according to the guidelines of the local Animal Use and Care Com- mut We used the 1.4HBV-S tg mice (32) to test whether CD8 T cells mittee and the National Animal Welfare Law. deliver antiviral (replication-inhibiting) effector functions to the liver. 1.4HBV-Smut tg mice express a mutant, replicating HBV Vector constructs genome in the liver. An introduced T1438C mutation blocks ex- The construction of the pCI/S, pCI/C, pCI/Pol, pCI/OVA, pCI/T77-C79–149, pression of the small HBsAg (and hence secretion of HBsAg par- and pCI/T77-S140–226 vectors has been described previously (24, 34, 39, ⌬ ticles or virions from transgene-expressing hepatocytes) but allows 40). The pCI/HB-multi and its deletion variants pCI/HB-multi DnaJ and pCI/HB-multi⌬OVA were generated by standard PCR-based cloning HBV replication in hepatocytes, HBcAg expression by hepato- techniques. cytes, and HBeAg secretion into the serum (32). This allowed us to investigate which vaccine-induced CD8 T cells can locally sup- Detection of chimeric Ags press HBV replication in hepatocytes. HEK cells were transiently transfected with plasmids using the calcium Expression of Ags as heat shock protein (Hsp)-bound fusion phosphate method. Thirty-six hours after transfection, cells were labeled proteins is an attractive strategy to induce multispecific CD8 T cell with [35S]methionine. Subsequently, cells were lysed with lysis buffer (150 responses (34–37). Constitutively expressed Hsp73 binds mutant mM NaCl, 0.5% Nonidet P40, and 100 mM Tris-hydrochloride (pH 8.0)) supplemented with the protease inhibitors aprotinin and leupeptin. Extracts SV40 T-Ag in an ATP-dependent manner through a specific DnaJ- were cleared by centrifugation and precipitated with 100 ␮l of Strep- like domain present in the N-terminal 77-residue fragment of the Tactin-Sepharose (catalog no. 2-1201-025; IBA). Precipitates were washed

T-Ag (34). Fusion constructs containing the N-terminal T77 do- (100 mM NaCl and 150 mM Tris-hydrochloride (pH 8.0)), recovered from main of SV40 T-Ag and unrelated COOH-terminal domains effi- Strep-Tactin-Sepharose by pH 6.8 elution buffer (1.5% SDS, 5% 2-ME, and 7 mM Tris-hydrochloride) and processed for SDS-PAGE and fluorog- ciently bind Hsp73 and accumulate to high steady-state levels in raphy of the gels. eukaryotic transfectants (34, 37). The system is exceptionally po- tent to induce mono- or multispecific CD8 T cell responses when DNA immunization used in DNA- or protein-based vaccines. In this article, we con- Mice were injected with 50 ␮g of plasmid DNA into each tibialis muscle structed multispecific, Hsp-binding DNA vaccines encoding dif- (100 ␮g of plasmid DNA/mouse). ferent domains of HBV Ags to characterize immunodominant and subdominant epitopes. DNA isolation, Southern blot analysis, and detection of the integrated transgene Materials and Methods Tissues were frozen, homogenized in a grinder, and lysed with proteinase Mice K (100 ␮g/ml, catalog no. 83767023-58; Roche)-containing buffer. DNA was extracted by standard procedures and treated for3hat37°C with C57BL/6JBom (B6) H-2b mice, C57BL/6J-TgN(Alb1HBV)44Bri tg (Alb/ RNase (catalog no. 109169; Roche). Thirty micrograms of DNA was di- HBs) (27), 1.4HBV-Smut tg (32), and HLA-A*0201 (HHD) mice (38) were gested with HindIII and electrophoretically separated on an agarose gel. bred and kept under standard pathogen-free conditions in the animal colony DNA was transferred to a nylon membrane by capillar blotting. Plasmid 372 INDUCTION OF SUBDOMINANT CD8 T CELL RESPONSES

Table I. CD8 T cell epitopes

Binding Affinity Epitope Ag 1 Epitope Sequence Restriction (IC 50% nM) Ref.

1 HBsAg S20–28 FLLTRILTI HLA-A*0201 10 7, 41 2 HBsAg S185–194 GLSPTVWLSV HLA-A*0201 18 7, 41 a 3 HBsAg S208–216 ILSPFLPLL HLA-A*0201 nt 4 HBcAg C18–27 FLPSDFFPSV HLA-A*0201 3 41 5 HBV-Pol Pol803–811 SLYADSPSV HLA-A*0201 14 7, 41 b 6 OVA OVA257–264 SIINFEKL K 4 This study b 7 HBsAg S190–197 VWLSVIWM K 13 This study b 8* HBcAg C93–100 MGLKFRQL K 4 This study a 8* HBcAgadw C93–100 MGLKIRQL nt a nt, Not tested. * The difference in the C93–100 sequence is marked with bold underlining. pHBV-1 containing a full-length HBV genome (provided by F. V. paraffin sections using a polyclonal rabbit anti-HBcAg antiserum (catalog Chisari, The Scripps Research Institute, La Jolla) was digested with no. RP017; Diagnostic BioSystems) followed by peroxidase-conjugated EcoRI and PvuI. The 3.2-kb HBV DNA fragment was isolated and used goat anti-rabbit IgG (catalog no. 111-035-003; Dianova). The enzymes as template. [␣-32P]dCTP-labeled HBV DNA was generated using the were visualized using the chromogen 3,3Ј diaminobenzidine. The sections Downloaded from High Prime DNA Labeling Kit (catalog no. 11585584001; Roche) ac- were further stained with hemalaun. cording to the manufacturer’s instructions. The HBV-specific DNA bands were quantitatively analyzed by Aida Image Analyzer version Determination of specific CD8 T cell frequencies 3.52 software (raytest). Spleen cells and hepatic nonparenchymal cells were isolated. Cells (5 ϫ Determination of serum transaminase levels 105/100 ␮l) were incubated in RPMI 1640 medium at 37oC with 5 ␮g/ml

of the indicated peptides. After a 1-h incubation, brefeldin A (catalog no. http://www.jimmunol.org/ Serum alanine aminotransferase activity was determined using the Re- 15870; Sigma-Aldrich) was added to a final concentration of 5 ␮ flotron test (catalog no. 745138; Roche). g/ml and the cultures were incubated for another 4 h. Cells were harvested, washed, Histology and surface stained with PE-conjugated anti-CD8 mAb (catalog no. 553033; BD Biosciences). Surface-stained cells were fixed with 2% para- Thin slides of liver tissue (Ͻ3 mm) were fixed in 4% formalin (pH 7.0) for formaldehyde in PBS. Fixed cells were resuspended in permeabilization 24 h and embedded in paraffin. Paraffin sections, 2-␮m thick, were stained buffer (HBSS, 0.5% BSA, 0.5% saponin, and 0.05% sodium azide) and with H&E. Immunohistological detection of HBcAg was performed on incubated with FITC-conjugated anti-IFN-␥ mAb (catalog no. 554411; BD by guest on September 29, 2021

FIGURE 2. Immunogenicity of recombinant vaccines. HLA-A*0201 tg mice (four mice per group) were immunized with pCI/S (group 1), pCI/C (group 2), pCI/Pol (group 3), pCI/HB-multi (group 4), or with the noncoding pCI vector (group 5). Twelve days after injection, epitope-specific CD8 T cell responses were determined by 5-h ex vivo stimulation of splenic T cells with the A2/C18–27 (A), A2/Pol803–811 (B), A2/S20–28 (C), A2/S185–194 (D), or ␥ϩ ␥ϩ Ϯ A2/S208–216 (E) peptides followed by determination of IFN- CD8 T cell frequencies. Mean percentages of IFN- CD8 T cells ( SD) of a representative experiment (of three independent experiments performed) are shown. The statistical significance of differences between groups was determined by the .p Ͻ 0.05; ns, not significant ,ء .unpaired Student t test The Journal of Immunology 373 Downloaded from http://www.jimmunol.org/

FIGURE 3. Immunogenicity of recombinant vaccines. B6 mice (four mice per group) were immunized with pCI/S (group 1), pCI/C (group 2), pCI/OVA (group 3), pCI/HB-multi (group 4), pCI/T-S140–226 (group 5), or the pCI/T-C79–149 DNA. Twelve days after injection, epitope-specific CD8 b b b T cell responses were determined by K /OVA257–264 (A)-, K /S190–197 (B)-, or K /C93–100 (C)-specific tetramer staining of the splenic T cells or by b b b 5-h ex vivo stimulation of splenic T cells with the respective K /OVA257–264 (D), K /S190–197 (E), or K /C93–100 (F) peptides followed by deter- mination of IFN-␥ϩ CD8 T cell frequencies. Mean percentages of IFN-␥ϩ or tetramerϩ CD8 T cells (ϮSD) of a representative experiment (of two independent experiments performed) are shown. The statistical significance of differences between groups was determined by the unpaired Student p Ͻ 0.05; ns, not significant. by guest on September 29, 2021 ,ء .t test

Biosciences) for 30 min at room temperature and washed twice in perme- HBV domains (i.e., S2–50, Pol131–180, Pol791–832,C79–149,C10–50, abilization buffer. Stained cells were resuspended in PBS/0.3% (w/v) BSA and S140–226) and an OVA246–353 fragment used to characterize supplemented with 0.1% (w/v) sodium azide. We determined the frequen- b ϩ ϩ the well-defined K -restricted CD8 T cell response to OVA . cies of IFN-␥ CD8 T cells by flow cytometry analyses. 257–264 Alternatively, 106 freshly isolated spleen cells or liver (nonparen- The antigenic sequence was COOH-terminally fused in frame chymal cells) were stained with FITC-conjugated anti-CD8 mAb (cat- to the Hsp-binding T77 sequence and a Strep tag sequence (HB- alog no. 100706; BioLegend) and PE-conjugated C93–100-, S190–197-, or multi; Fig. 1A). This fusion protein was efficiently expressed as a OVA257–264-specific tetramers (Beckman Coulter) for 30 min at 4°C. complex with cytosolic, constitutively expressed Hsp73 in eukary- We determined the frequencies of tetramerϩ CD8 T cells by flow cy- tometry analyses. otic transfectants (Fig. 1B).

Synthetic peptides Induction of multispecific murine CD8 T cell responses The synthetic peptides used in this study were obtained from JPT Peptide We analyzed the CD8 T cell responses to five well-defined Technologies. Peptide purity was Ͼ70%. Before use, peptides were dissolved in DMSO at a concentration of 10 mg/ml and diluted with culture medium. HLA-A*0201-restricted HBV epitopes (7, 41, 42) (S20–28, S185–194,S208–216,C18–27, and Pol803–811; Table I) in HLA- Statistics A*0201 tg HHD mice (38). A single immunization with pCI/C Data were analyzed using PRISM software (GraphPad). The statistical sig- or pCI/Pol elicited CD8 T cell responses to either the C18–27 nificance of differences in the mean CD8 T cell frequencies between groups (group 2 in Fig. 2A) or the Pol803–811 epitope (group 3 in Fig. was determined by using the unpaired Student t test. A value of p Ͻ 0.05 2B). Immunization of mice with pCI/HB-multi induced CD8 T was considered significantly different. cell responses to both epitopes (group 4 in Fig. 2, A and B) but CD8 T cell frequencies were slightly lower than in mice vac- Results cinated with pCI/C or pCI/Pol (cf groups 2 and 4 in Fig. 2A and Design of multidomain, stress protein-capturing DNA vaccines groups 3 and 4 in Fig. 2B). pCI/S immunization elicited CD8 T A large amount of antigenic information can be incorporated into cell responses to all three S epitopes (group 1 in Fig. 2, C–E) the Hsp-binding fusion proteins, thus allowing expression of many but pCI/HB-multi immunization induced CD8 T cell responses

T cell-stimulating epitopes together with their flanking sequences to the S20–28 but not (or barely) to the S185–194 and S208–216

(that are potentially relevant for their processing) (34). We con- epitopes (group 4 in Fig. 2, C–E). S20–28-specific CD8 T cell structed a multispecific vaccine encoding different domains of the responses were comparable in pCI/S- and pCI/HB-multi-immu- HBV surface (HBsAg or S), core (HBcAg or C), and polymerase nized mice (cf groups 1 and 4 in Fig. 2C). Only a fraction of the (Pol) proteins (HB-multi; Fig. 1A). The fusion Ag encodes six HLA-A*0201-restricted epitope repertoire of HBsAg could thus 374 INDUCTION OF SUBDOMINANT CD8 T CELL RESPONSES Downloaded from http://www.jimmunol.org/

b FIGURE 4. A, Kinetics of K /C93–100-specific CD8 T cell responses primed in B6 mice by pCI/C or pCI/HB-multi DNA. Mice were immunized once ␮ b i.m. with 100 g of pCI/C or pCI/HB-multi DNA. At the indicated time points after vaccination, splenic K /C93–100-specific CD8 T cell numbers were determined by tetramer staining. The mean numbers of tetramerϩ CD8 T cells (percent) of three mice per time point per group are shown. B–D, Immunogenicity of recombinant vaccines. B6 mice (four mice per group) were injected with pCI (group 1), pCI/HB-multi⌬DnaJ (group 2), pCI/HB-multi (group 3), or pCI/HB-multi⌬OVA DNA (group 4). Twelve days after injection, epitope-specific CD8 T cell responses were determined by Kb/OVA 257–264 by guest on September 29, 2021 b b ϩ (B)-, K /S190–197 (C)-, or K /C93–100 (D)-specific tetramer staining of splenic T cells. The mean numbers of tetramer CD8 T cells (percent) of a representative experiment (of two independent experiments performed) are shown. The statistical significance of differences between groups was deter- .p Ͻ 0.05 ,ء .mined by the unpaired Student t test be specifically activated by the pCI/HB-multi vaccine while the sponses were not a result of different kinetics. CD8 T cell re- pCI/S vaccine supported efficient priming of a multispecific sponses to the HBsAg and HBcAg epitopes of HBV were thus response. Thus, the diversity of the CD8 T cell responses to suppressed in the presence of a CD8 T cell response to the Kb/

HBV Ags differed in mice vaccinated with either the pCI/HB- OVA257–264 epitope. This was further supported by immunization multi or the pCI/S, pCI/C, or pCI/Pol DNA vaccines. experiments using vaccines that encode the isolated, epitope-con- b The fusion Ag also contains two well-defined K -restricted taining S140–226 or C79–149 fragment. The pCI/T-S140–226 and pCI/ b HBV surface and core epitopes (34, 39) and one K -restricted T-C79–149 efficiently expressed Hsp-bound fusion Ags (34) and (in b OVA epitope (Table I). We immunized B6 mice with pCI/S, the absence of K /OVA257–264-specific immune responses) elicited b b pCI/C, pCI/OVA, or pCI/HB-multi DNA and determined spe- CD8 T cell responses to the K /S190–197 or K /C93–100 epitopes cific (tetramerϩ and IFN-␥ϩ) CD8 T cell frequencies in the (group 5 in Fig. 3, B and E; group 6 in Fig. 3, C and F). Further- b spleen 12 days after immunization. Although K /OVA257–264- more, a suppression of CD8 T cell responses to these epitopes was specific CD8 T cells were efficiently induced in B6 mice by also evident in mice immunized with the pCI/HB-multi⌬DnaJ vac- b b vaccination with pCI/HB-multi, neither K /S190–197- nor K / cine (group 2 in Fig. 4). This construct encodes the HB-multi se-

C93–100-specific CD8 T cell responses could be elicited in these quence without the Hsp-binding T77 domain, and the expression animals by immunization with this vaccine (group 4 in Fig. 3). level of the HB-multi⌬DnaJ Ag from this construct was signifi- b K /OVA257–264-specific CD8 T cell responses were comparable cantly lower than the expression level of its Hsp-bound form (cf in pCI/OVA- and pCI/HB-multi-immunized mice (cf groups 3 lanes 1 and 2 in Fig. 1B). Immunization with pCI/HB-multi⌬DnaJ b and 4 in Fig. 3, A and D). Mice immunized with pCI/S or pCI/C elicited lower K /OVA257–264-specific CD8 T cell responses than DNA vaccines efficiently generated CD8 T cell responses spe- immunization with the Hsp-bound HB-multi Ag (cf groups 2 and b b cific for the K /S190–197 or K /C93–100 epitope (groups 1 and 2 3 in Fig. 4B). This confirms our previous findings that stress pro- b in Fig. 3, B, C, E, and F). The pCI/C DNA-primed K /C93–100- tein binding enhances the expression and immunogenicity of a b b specific CD8 T cell response peaked at 11 days after vaccination construct (34, 35). Neither K /S190–197- nor K /C93–100-specific and declined to low levels in the following 10 days (Fig. 4A). In CD8 T cell responses could be elicited in pCI/HB-multi⌬DnaJ- contrast, the pCI/HB-multi DNA vaccine induced only low num- immunized mice (group 2 in Fig. 4, C and D). Thus, the T77 DnaJ- b Ͼ bers of K /C93–100-specific CD8 T cells during 20 days after like domain and/or Hsp capture did not change the immunodomi- vaccination (Fig. 4A), demonstrating that the different T cell re- nance hierarchy of this response. The Journal of Immunology 375 Downloaded from http://www.jimmunol.org/

b FIGURE 5. Induction of K /C93–100-specific CD8 T cell responses in HBsAg-expressing Alb-HBs tg mice. A, Schematic presentation of a Hsp73-binding construct encoding fragments of the HBV surface and core Ag (pCI/HB-SC). The SC domain consists of the following Ag fragments: S140–226,C79–149, b and C10–50. The Strep tag, the Hsp-binding, SV40-T77-derived DnaJ sequence, and the K epitopes (for detail see Table I) are indicated. B, HEK293 cells by guest on September 29, 2021 were transiently transfected with pCI (lane 1) or the pCI/HB-SC DNA (lane 2). Cells were lysed and the HB-SC Ag was immunopurified with Strep- Tactin-Sepharose followed by release of Ag complexes with SDS. Samples were analyzed by SDS-PAGE and fluorography of the gels. The positions of the HB-SC fusion protein and the cellular Hsp73 are indicated. C and D, B6 and Alb-HBs tg mice were immunized i.m. with pCI/S (groups 1), pCI/C b b (groups 2), or pCI/HB-SC (groups 3). Twelve days after injection, epitope-specific CD8 T cell responses were determined by K /S190–197 (C)- or K /C93–100 (D)-specific tetramer staining of splenic T cells. The mean percentages of tetramerϩ CD8 T cells of a representative experiment (of two independent ;p Ͻ 0.05 ,ء .experiments performed) are shown. The statistical significance of differences between groups was determined by the unpaired Student t test ns, not significant.

Characterization of immunodominance hierarchies primed by ple epitope competition model for Kb class I molecules cannot the HB-multi vaccine explain the observed immunodominance hierarchy. b b b To further characterize the K /S190–197-mediated suppression The K /S190–197- but not the K /C93–100-specific CD8 T cell response was efficiently induced in B6 mice immunized with mechanism, we used Alb/HBs tg mice (27). These mice overex- the HB-multi⌬OVA construct (Fig. 1) from which the OVA press the large surface Ag in the liver under heterologous albumin domain was deleted (group 4 in Fig. 4, B–D). Similarly, a promoter control. DNA immunization with HBsAg-encoding vec- b HB-SC vaccine (encoding only HBsAg and HBcAg domains; tors did not induce K /S190–197-specific CD8 T cells (group 1 in b Fig. 5C), indicating that Alb/HBs tg mice are tolerant to this Kb/ Fig. 5, A and B) efficiently induced K /S190–197- but only low b S -specific immune response. Similarly, the pCI/HB-SC K /C93–100-specific CD8 T cell responses in vaccinated B6 mice 190–197 b (group 3 in Fig. 5, C and D). Thus, distinct immunodominance DNA vaccine did not induce K /S190–197-specific CD8 T cell re- mechanisms were detected in the Kb-restricted CD8 T cell re- sponses in Alb/HBs tg mice (group 3 in Fig. 5C). However, in b contrast to the vaccination studies in non-tg B6 mice, the missing sponses to multidomain Ags: while the K /OVA257–264-specific b b Kb/S -specific response allowed priming of Kb/C -spe- CD8 T cell response suppressed K /S190–197- and K /C93–100- 190–197 93–100 b specific CD8 T cell responses, the K /S190–197-specific CD8 T cific CD8 T cells in Alb/HBs tg mice (group 3 in Fig. 5D). Thus, b b cell response suppressed the K /C93–100-specific CD8 T cell the absence of the K /S190–197-specific CD8 T cell response al- b b response. We conclude that the K /C93–100 epitope is subdomi- lowed priming of the subdominant K /C93–100-specific responses. b nant because it is readily suppressed by immune responses to Notably, comparable levels of K /C93–100-specific CD8 T cells b OVA or the HBsAg. This was unexpected since the K /C93–100 were primed in B6 and Alb/HBs tg mice by the pCI/C vaccine epitope contains optimal anchor motifs at positions F5 and L8 (group 2 in Fig. 5D), indicating that this epitope is not suppressed (43) and its binding affinity for Kb is comparable to that of the when delivered as intact core protein. Since Ag processing/pre- b b K /OVA257–264 and K /S190–197 epitopes (Table I). Thus, a sim- sentation are likely to be identical in pCI/HB-SC-transduced APC 376 INDUCTION OF SUBDOMINANT CD8 T CELL RESPONSES

b but the numbers of K /C93–100-specific CD8 T cells rapidly de- clined thereafter (Fig. 6B). A similar pattern of expansion and b contraction of the K /C93–100-specific CD8 T cell response was induced in wild-type B6 mice vaccinated with pCI/C DNA (see Fig. 4A). It is noteworthy that specific tetramerϩ CD8 T cells were readily detectable in the transgene-expressing liver (Fig. 6) but not in the spleen of 1.4HBV-Smut tg mice (data not shown). b We analyzed whether K /C93–100-specific CD8 T cells down- modulate HBV replication in the liver of tg mice. Southern blot analysis of liver DNA obtained from nonimmunized or pCI-in- jected 1.4HBV-Smut tg mice showed three prominent bands cor- responding to the expected sizes of the relaxed circular DNA, dou- ble-stranded linear DNA, and ssDNA species of HBV (group 1 in Fig. 7A) (32). This analysis did not reveal a cccDNA species. A significant reduction of HBV replication in the liver of tg mice was observed 12 days after immunization with the pCI/C vector (i.e., at b the peak of the K /C93–100-specific CD8 T cell response, Fig. 6B; group 3 in Fig. 7A) but not after immunization with the pCI/S

vaccine (group 2 in Fig. 7A). This reduction in HBV replication Downloaded from was transient because it rebounded between days 18 and 28 after immunization (group 1 in Fig. 7B), i.e., at a time point at which b K /C93–100-specific CD8 T cell levels had declined in the liver (Fig. 6B). A similar pattern of “transient” inhibition of HBV rep- lication was induced in 1.4HBV-Smut tg mice after a boost injec-

FIGURE 6. Characterization of antiviral CD8 T cell responses in vac- tion with pCI/C DNA (Fig. 7B). Maximal reduction of viral rep- http://www.jimmunol.org/ cinated 1.4HBV-Smut tg mice. A, 1.4HBV-Smut tg mice were immunized lication in the liver of 1.4HBV-Smut tg mice was observed at i.m. with pCI/S (group 1), pCI/C (group 2), or pCI/Cadw (group 3). Twelve approximately day 11 after boost immunization (group 3 in Fig. days after injection, epitope-specific CD8 T cell responses were determined 7B). pCI/C vaccination had no significant effect on viral RNA b b by K /S190–197-orK/C93–100-specific tetramer staining of liver T cells. levels in the liver of tg animals (data not shown), confirming pre- The mean percentages of tetramerϩ CD8 T cells of a representative ex- b viously published data from an adenovirus-based vaccination periment (of three independent experiments) are shown. B,K/C93–100 tet- ϩ mut study in another HBV tg mouse model (46). The HBcAg-specific, ramer CD8 T cells in the livers of immunized 1.4HBV-S tg mice were mut determined at the indicated time points after prime. antiviral immune response in the liver of 1.4HBV-S tg mice did not elicit a rise in serum transaminase levels (Fig. 7C). We repro- duciblely detected small centers of cell infiltrates in the liver of by guest on September 29, 2021 from B6 and Alb/HBs tg mice, we conclude that epitope compe- pCI/C-immunized (but not pCI/S or nonimmunized) 1.4HBV-Smut b b tition for K class I molecules is not involved in the K /S190–197- tg mice by immunohistology (Fig. 7D and data not shown). b mediated suppression of the K /C93–100-specific CD8 T cell re- HBcAg is expressed in the majority of hepatocytes (Fig. 7E) (32) sponse in B6 mice. and its expression significantly declined in pCI/C-vaccinated mice b b at the peak of the K /C93–100-specific CD8 T cell response (day 12 Vaccine-induced, K /C93–100-specific, subdominant CD8 T cells after immunization; Fig. 7E). Antiviral CD8 T cells can thus de- down-modulate HBV replication in the liver liver their virus-replication-inhibition effect without the destruc- b The above analysis identified the K /C93–100 epitope of HBV as tion of hepatocytes (32). b subdominant in the model systems used. Subdominant (weak) We next asked whether K /C93–100-specific CD8 T cells are the epitopes may play an essential role to break T cell tolerance (44, main players in the vaccine-induced inhibition of HBV replication. 45). This epitope may therefore be useful for therapeutic vaccines. In the above experiments, we used the pCI/C vaccine (that encodes We previously generated the 1.4HBV-Smut tg line that expresses a HBcAg of the HBV/ayw serotype (genotype D) that is identical the intracellular large surface but not the secreted small surface to HBcAg expressed by the 1.4HBV-Smut tg mouse). A natural HBsAg (due to a mutation introduced at the translation start codon HBcAg variant of the adw2 serotype (genotype A) contains four of the S- gene) (32). HBeAg is readily detected in the sera of amino acid exchanges and two additional amino acids at position mut 1.4HBV-S tg mice and HBcAg is efficiently expressed in the C153/154. A single amino acid exchange locates within the central b liver (32). In the subsequent experiments, we used DNA vaccines anchor motif of the K /C93–100 epitope (i.e., at position 5 from a F encoding single Ags of HBV because they are closely associated to an I; Table I). Comparable levels of HBcAg were expressed in with the natural HBV-specific Ag processing and epitope presen- cells transiently transfected with the pCI/C and pCI/Cadw (data not mut tation. We vaccinated 1.4HBV-S tg mice with either the pCI/S shown). A pCI/Cadw vaccine encoding this HBcAg variant did not b mut or the pCI/C vaccine and tested whether this vaccination protocol induce K /C93–100-specific CD8 T cells in B6 and 1.4HBV-S tg primes specific CD8 T cell responses in their target organ (i.e., the mice (group 3 in Fig. 6A) and did not reduce HBV replication in mut liver). Single or repeated injections of the pCI/S DNA vaccine into 1.4HBV-S tg mice (group 4 in Fig. 7A). Notably, pCI/Cadw- mut b 1.4HBV-S tg mice did not elicit K /S190–197-specific CD8 T primed CD8 T cell responses were neither detectable by specific in cell immunity in spleen (data not shown) or liver (group 1 in Fig. vitro stimulation with wild-type or mutant C93–100 peptides nor by 6A). In contrast, a single injection of pCI/C into 1.4HBV-Smut tg specific tetramer staining (data not shown). These findings thus b b mice elicited a strong K /C93–100-specific CD8 T cell immunity in show that the antigenicity of the K /C93–100 epitope is destroyed the livers of immunized tg mice (group 2 in Fig. 6A). Seven to 14% by the mutation and confirm that immune responses to the sub- b of the intrahepatic CD8 T cells displayed specific reactivity for this dominant K /C93–100 epitope play a central role in the inhibition of epitope at the peak of the response at days 11–12 after vaccination, HBV replication in this model. The Journal of Immunology 377 Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

b mut mut FIGURE 7. K /C93–100-specific CD8 T cells primed by pCI/C inhibit viral replication in the liver of 1.4HBV-S tg mice. A, 1.4HBV-S tg mice were immunized i.m. with “empty” pCI (group 1), pCI/S (group 2), pCI/C (group 3), or pCI/Cadw (group 4). At 12 days after vaccination, HBV replication in the livers of immunized mice was investigated by Southern blot analysis (upper panel). Two representative mice per group are shown. The integrated transgene (tg), relaxed circular (RC), double-stranded linear (DS), and single-stranded (SS) HBV DNAs are indicated. HBV replication (three to six mice per group) was quantitatively determined as a ratio of HBV replicates:transgene, i.e., bands for the integrated transgene are used for normalization of the amount of HBV-specific replicative intermediates (lower panel). B, 1.4HBV-Smut tg mice were preimmunized with pCI/C. At 28 days after vaccination, HBV replication in the livers of immunized mice was determined (group 1) and other groups of mice were boosted with pCI/C (groups 2–5). At day 8 (group 2), day 11 (group 3), day 21 (group 4), and day 28 (group 5) after boost, HBV replication was investigated by Southern blot analysis (upper panel) and quantitatively determined (three to four mice per group) as a ratio of HBV replicates:transgene (lower panel) as described above. C, Antiviral immune responses do not induce detectable liver injury in 1.4HBV-Smut tg mice. 1.4HBV-Smut tg mice were vaccinated i.m. pCI/C DNA and serum alanine aminotransferase (ALT) activity was determined in sera before (day 0) and up to 18 days after immunization. D, At 12 days after vaccination, livers of pCI- and pCI/C-immunized 1.4HBV-Smut tg mice were analyzed by histology (H&E staining). E, Immunohistochemical analysis of HBcAg in the livers of pCI- and pCI/C-immunized 1.4HBV-Smut tg mice. HBcAg expression was detected in the nuclei and cytosol with polyclonal rabbit anti-HBcAg antiserum and peroxidase-conjugated goat anti-rabbit IgG as previously described (32).

Discussion CD8 T cell responses investigated. These operate in immune re- We used a simple i.m. DNA vaccination approach to demonstrate sponses to different epitopes of the same Ag (e.g., HBsAg), be- under well-controlled experimental conditions (inbred mice, well- tween different Ags of the same virus (e.g., HBsAg and HBcAg), defined Ags and their MHC class I-restricted epitopes) that immu- or between unrelated Ags (e.g., OVA and HBsAg or HBcAg). nodominance hierarchies are readily apparent in all multispecific Deletion of an immunodominant, antigenic domain from a vaccine 378 INDUCTION OF SUBDOMINANT CD8 T CELL RESPONSES or immunization of a host selectively unresponsive to an immu- control of difficult viruses, evaluating immunodominance phenom- nodominant epitope present on a given vaccine reveals a repertoire ena will always be a major task but difficult to perform. This is of “subdominant” or “cryptic” epitopes that can mediate important especially challenging in clinical settings in which an outbred host biological effects. One of these is the potential usefulness of such confronts a virus with many different Ags. epitopes in attempts to reconstitute specific protective CD8 T cell We consider the identification of the potential therapeutic value immunity in chronic infection (in which the host failed its first of selective priming of CD8 T cells to subdominant viral epitopes attempt to establish protective, specific T cell immunity against (under conditions in which nonresponsiveness to immunodomi- immunodominant determinants of a pathogen). nant epitopes prevails) the key observation of our report. We dem- Interference between responses to individual epitopes presented onstrate that the weak (i.e., easily suppressed) CD8 T cell response by (the same or different) MHC class I molecules is the well- b to the K /C93–100 epitope, if allowed to develop, can at least tran- established phenomenon of “immunodominance” in multispecific siently and partially control HBV replication. This suggests that CD8 T cell responses that limits priming of responses with exten- efforts to induce protective antiviral T cell immunity should in- sive repertoire diversity. Immunodominance is an incompletely clude a consideration of the cryptic repertoire of subdominant understood phenomenon (22, 23). Little is known about how im- weak epitopes. munodominant and subdominant determinants are distinguished Aspects of the vaccination approach used to deliver the viral Ag by the CD8 T cell system. Immunodominance is observed between as well as the tg readout system used to detect the antiviral effect epitopes from viruses and model Ags (24, 25, 40, 47–54). Different are relevant to critically evaluate the described data. DNA vacci- factors have been proposed to contribute to immunodominance. nation of the mouse is a potent tool to elicit CD8 T cell responses. These include: competition between responding T cells (47) or T Unfortunately, the immunogenicity of DNA vaccines in is Downloaded from cells and APCs (55), inefficient Ag processing (51), or peptide b limited. We show that the K /S190–197 CD8 T cell response dom- binding to MHC class I molecules (41, 48, 56), only a few specific b inates the K /C93–100 CD8 T cell response but that this dominant precursors in the naive T cell pool for the subdominant epitopes or HBsAg-specific response does not prime an antiviral response in high T cell precursor frequencies specific for an immunodominant 1.4HBV-Smut tg mice. DNA vaccination studies in MHC class epitope (57, 58), competition between MHC class I alleles for cell II-deficient (A␣Ϫ/Ϫ) mice showed that priming an HBsAg-specific surface expression (49), the kinetics of expression CD8 T cell response by pCI/S is CD4 T help dependent (61), http://www.jimmunol.org/ (50), and regulatory T cells that selectively suppress certain re- whereas priming a HBcAg-specific CD8 T cell response by pCI/C sponses (59). We previously reported immunodominance phenom- is not (data not shown). CD4 T cell regulation of the dominant but ena between Ld- and Kb/Dd-restricted CD8 T cell responses (24). not subdominant response is thus possible. As expected, it is not Ld-mediated immunodominance can be overcome at least partially the native (particulate) structure of HBcAg that is important for by expressing multiepitope vaccines as chimeric Ags that endog- priming this CD8 T cell response because DNA vaccines encoding enously capture constitutively expressed Hsp73 (35). This may be fragments of the HBcAg were equally efficient in priming the related to specific effects of Hsp73, e.g., enhanced expression of response. Ags, induction of cross-priming (34, 60). Notably, expression of mut

We used the 1.4HBV-S tg line (32) to detect antiviral CD8 T by guest on September 29, 2021 Hsp73-bound Ags cannot overcome HLA-A*0201-restricted (Fig. cells. The 1.4HBV-Smut tg mouse is a variant of the well-estab- 2) (37) and Kb-restricted immunodominance hierarchies (Figs. 3 lished 1.3HBV model in which a replicating 1.3HBV genome is and 5). Our data make Ag processing and MHC class I-restricted transcribed from a transgene in hepatocytes (31). The 1.4HBV- epitope presentation unlikely candidates for the observed immu- mut nodominance because elimination of (or specific tolerance to) a S tg model was designed to allow the following investigations: single epitope revealed efficient responsiveness to the subdominant 1) The mice have a replicating HBV genome and express all an- epitope. Differences in the specific precursor frequency of naive tigenic information of HBV in hepatocytes which allows us to CD8 T cells specific for dominant vs subdominant epitopes are readout antiviral effects of T cells to all HBV Ags. 2) No intact also unlikely to explain the described immunodominance hierar- HBV virions are produced in the liver, similar to the situation in chy. Injection of a DNA vaccine encoding an immunodominant chronic HB patients treated with IFNs and/or antivirals. 3) HBsAg epitope into the left leg did not suppress CD8 T cell responses is produced and presented by hepatocytes but is not detectable in b primed at the same time to a subdominant epitope encoded by a serum or peripheral tissues. Hence, K /S190–197-specific immune DNA vaccine (in the absence of the dominant epitope) into the responses are focused to the liver and not “distracted” by an almost right leg (data not shown). These data suggest a “local regulator ubiquitous, alternative Ag presentations which allows us to study model” in which differential recruitment of CD8 T cells with dif- local delivery of antiviral effects to hepatocytes by HBsAg-specific b ferent recognition specificities, differential efficiencies in synapse CD8 T cells. Unexpectedly, it was the K /C93–100- but not the b formation with the APCs, different relative efficiencies in the com- K /S190–197-specific CD8 T cell response that could be elicited in mut mitment for the priming program, or the clonal expansion may 1.4HBV-S tg mice and induced a specific antiviral immune re- operate. sponse. Injection of the pCI/S DNA vaccine efficiently expanded b The data have practical implications for understanding antiviral K /S190–197-specific CD8 T cells in B6 but not in congenic Alb/ CD8 T cell responses and for the design of T cell-stimulating vac- HBs or 1.4HBV-Smut tg mice. We speculated that Alb/HBs mice cines. Escape mutations in an immunodominant epitope that elim- (that secrete large amounts of HBsAg particle into the circulation) inates its antigenicity do not necessarily lead to an inability to clear should be tolerant to HBsAg while 1.4HBV-Smut tg mice (that the infection but may pave the way for responses to a diverse express only low amounts of intracellular HBsAg in hepatocytes) repertoire of subdominant epitopes that can be functionally com- should allow peripheral priming of HBsAg-specific CD8 T cells petent. In designing polytope vaccines, more may not necessarily that subsequently find their Ag in the liver. This assumption was be better. Delivery of multiple polytope vaccines with a limited obviously wrong. High levels of free HBsAg in the circulation are number of epitopes to different sites at different times may be more not critical for establishing tolerance to HBV. Furthermore, toler- efficient than delivering many epitopes at the same time to the ance can be broken in the presence of large doses of circulating same site. Because the establishment of robust, multispecific CD8 HBeAg in the serum because the pCI/C DNA vaccine efficiently b mut T cell responses is a key priority in novel vaccine designs for the primed K /C93–100-specific CD8 T cells in 1.4HBV-S tg mice. The Journal of Immunology 379

These autoreactive T cells accumulated in the liver of 1.4HBV- and its role in response to therapeutic vaccination in humans. J. Immunol. 162: Smut tg mice and were down-regulated in this target organ by un- 3088–3095. 18. Pol, S., B. Nalpas, F. Driss, M. L. Michel, P. Tiollais, J. Denis, and C. Brecho. known mechanisms (62, 63). The data reported here show that 2001. Efficacy and limitations of a specific immunotherapy in chronic hepatitis B. vaccines can induce specific CD8 T cells in a system modeling J. Hepatol. 34: 917–921. chronic infection of the liver but that the rules that allow or sup- 19. Jung, M. C., N. Gruner, R. Zachoval, W. Schraut, T. Gerlach, H. Diepolder, C. A. Schirren, M. Page, J. Bailey, E. Birtles, E. Whitehead, J. Trojan, S. Zeuzem, press this re-emergence of specific antiviral immunity remain to be and G. R. Pape. 2002. Immunological monitoring during therapeutic vaccination defined. as a prerequisite for the design of new effective therapies: induction of a vaccine- specific CD4ϩ T-cell proliferative response in chronic hepatitis B carriers. Vac- cine 20: 3598–3612. Acknowledgments 20. Bienzle, U., M. Gunther, R. Neuhaus, P. Vandepapeliere, J. Vollmar, A. Lun, and We greatly appreciate the expert technical assistance of Katrin O¨ lberger, P. Neuhaus. 2003. Immunization with an adjuvant hepatitis B vaccine after liver transplantation for hepatitis B-related disease. Hepatology 38: 811–819. Claudia Heilig, and Sarah Fischer. We are very grateful for the detailed 21. Depla, E., A. Van der Aa, B. D. Livingston, C. Crimi, K. Allosery, help of Drs. H. Maier and F. V. Chisari (Division of Experimental Pathol- V. De Brabandere, J. Krakover, S. Murthy, M. Huang, S. Power, et al. 2008. ogy, The Scripps Research Institute, La Jolla, CA) to characterize HBV Rational design of a multiepitope vaccine encoding T-lymphocyte epitopes for replication in tg mice. We thank F. Leitha¨user (Department of Pathology, treatment of chronic hepatitis B virus infections. J. Virol. 82: 435–450. University of Ulm, Ulm, Germany) for support in histology. 22. Yewdell, J. W., and J. R. Bennink. 1999. Immunodominance in major histocom- patibility complex class I-restricted T lymphocyte responses. Annu. Rev. Immu- nol. 17: 51–88. Disclosures 23. Kedl, R. M., J. W. Kappler, and P. Marrack. 2003. Epitope dominance, compe- tition and T cell affinity maturation. Curr. Opin. Immunol. 15: 120–127. The authors have no financial conflict of interest. 24. Schirmbeck, R., D. Stober, S. El Kholy, P. Riedl, and J. Reimann. 2002. The

immunodominant. Ld-restricted T cell response to hepatitis B surface Downloaded from References (HBsAg) efficiently suppresses T cell priming to multiple Dd-, Kd-, and Kb- restricted HBsAg epitopes. J. Immunol. 168: 6253–6262. 1. Rehermann, B., and M. Nascimbeni. 2005. Immunology of hepatitis B virus and 25. Schirmbeck, R., J. Reimann, S. Kochanek, and F. Kreppel. 2008. The immuno- infection. Nat. Rev. Immunol. 5: 215–229. genicity of adenovirus vectors limits the multispecificity of CD8 T-cell responses 2. Wieland, S. F., and F. V. Chisari. 2005. Stealth and cunning: hepatitis B and to vector-encoded transgenic . Mol. Ther. 16: 1609–1616. hepatitis C viruses. J. Virol. 79: 9369–9380. 3. Maini, M. K., C. Boni, G. S. Ogg, A. S. King, S. Reignat, C. K. Lee, 26. Chisari, F. V., C. A. Pinkert, D. R. Milich, P. Filippi, A. McLachlan, J. R. Larrubia, G. J. Webster, A. J. McMichael, C. Ferrari, et al. 1999. Direct ex R. D. Palmiter, and R. L. Brinster. 1985. A transgenic mouse model of the vivo analysis of hepatitis B virus-specific CD8ϩ T cells associated with the con- chronic hepatitis B surface antigen carrier state. Science 230: 1157–1160. http://www.jimmunol.org/ trol of infection. Gastroenterology 117: 1386–1396. 27. Chisari, F. V., K. Klopchin, T. Moriyama, C. Pasquinelli, H. A. Dunsford, S. Sell, 4. Maini, M. K., C. Boni, C. K. Lee, J. R. Larrubia, S. Reignat, G. S. Ogg, C. A. Pinkert, R. L. Brinster, and R. D. Palmiter. 1989. Molecular pathogenesis A. S. King, J. Herberg, R. Gilson, A. Alisa, et al. 2000. The role of virus-specific of in hepatitis B virus transgenic mice. Cell 59: CD8ϩ cells in liver damage and viral control during persistent hepatitis B virus 1145–1156. infection. J. Exp. Med. 191: 1269–1280. 28. Moriyama, T., S. Guilhot, K. Klopchin, B. Moss, C. A. Pinkert, R. D. Palmiter, 5. Bertoni, R., J. Sidney, P. Fowler, R. W. Chesnut, F. V. Chisari, and A. Sette. R. L. Brinster, O. Kanagawa, and F. V. Chisari. 1990. Immunobiology and patho- 1997. histocompatibility leukocyte antigen-binding supermotifs predict genesis of hepatocellular injury in hepatitis B virus transgenic mice. Science 248: broadly cross-reactive cytotoxic T lymphocyte responses in patients with acute 361–364. hepatitis. J. Clin. Invest. 100: 503–513. 29. Reifenberg, K., J. Lohler, H. P. Pudollek, E. Schmitteckert, G. Spindler, J. Kock, 6. Webster, G. J., S. Reignat, M. K. Maini, S. A. Whalley, G. S. Ogg, A. King, and H. J. Schlicht. 1997. Long-term expression of the hepatitis B virus core-e- D. Brown, P. L. Amlot, R. Williams, D. Vergani, et al. 2000. Incubation phase of and X-proteins does not cause pathologic changes in transgenic mice. J. Hepatol. acute hepatitis B in man: dynamic of cellular immune mechanisms. Hepatology 26: 119–130. by guest on September 29, 2021 32: 1117–1124. 30. Guidotti, L. G., K. Ando, M. V. Hobbs, T. Ishikawa, L. Runkel, R. D. Schreiber, 7. Webster, G. J., S. Reignat, D. Brown, G. S. Ogg, L. Jones, S. L. Seneviratne, and F. V. Chisari. 1994. Cytotoxic T lymphocytes inhibit hepatitis B virus gene R. Williams, G. Dusheiko, and A. Bertoletti. 2004. Longitudinal analysis of expression by a noncytolytic mechanism in transgenic mice. Proc. Natl. Acad. CD8ϩ T cells specific for structural and nonstructural hepatitis B virus proteins Sci. USA 91: 3764–3768. in patients with chronic hepatitis B: implications for immunotherapy. J. Virol. 78: 31. Guidotti, L. G., B. Matzke, H. Schaller, and F. V. Chisari. 1995. High-level 5707–5719. hepatitis B virus replication in transgenic mice. J. Virol. 69: 6158–6169. 8. Bertoletti, A., and A. J. Gehring. 2006. The immune response during hepatitis B 32. Halverscheid, L., N. K. Mannes, R. Weth, M. Kleinschmidt, U. Schultz, virus infection. J. Gen. Virol. 87: 1439–1449. K. Reifenberg, R. Schirmbeck, M. Nassal, H. E. Blum, J. Reimann, and 9. Chisari, F. V., and C. Ferrari. 1995. Hepatitis B virus immunopathogenesis. M. Geissler. 2008. Transgenic mice replicating hepatitis B virus but lacking ex- Annu. Rev. Immunol. 13: 29–60. pression of the major HBsAg. J. Med. Virol. 80: 583–590. 10. Wieland, S. F., R. G. Vega, R. Muller, C. F. Evans, B. Hilbush, L. G. Guidotti, 33. Guidotti, L. G., and F. V. Chisari. 1996. To kill or to cure: options in host defense J. G. Sutcliffe, P. G. Schultz, and F. V. Chisari. 2003. Searching for interferon- against viral infection. Curr. Opin. Immunol. 8: 478–483. induced genes that inhibit hepatitis B virus replication in transgenic mouse hepa- 34. Schirmbeck, R., M. Kwissa, N. Fissolo, S. Elkholy, P. Riedl, and J. Reimann. tocytes. J. Virol. 77: 1227–1236. 2002. Priming polyvalent immunity by DNA vaccines expressing chimeric anti- 11. Schlaak, J. F., G. Tully, H. F. Lohr, G. Gerken, and K. H. Meyer Zum Buschen- gens with a stress protein-capturing, viral J-domain. FASEB J. 16: 1108–1110. felde. 1999. The presence of high amounts of HBV-DNA in serum is associated 35. Schirmbeck, R., N. Fissolo, P. Chaplin, and J. Reimann. 2003. Enhanced priming with suppressed costimulatory effects of interleukin 12 on HBV-induced immune ϩ of multispecific, murine CD8 T cell responses by DNA vaccines expressing response. J. Hepatol. 30: 353–358. stress protein-binding polytope peptides. J. Immunol. 171: 1240–1246. 12. Chen, M., M. Sallberg, J. Hughes, J. Jones, L. G. Guidotti, F. V. Chisari, 36. Schirmbeck, R., P. Riedl, M. Kupferschmitt, U. Wegenka, H. Hauser, J. Rice, J. N. Billaud, and D. R. Milich. 2005. Immune tolerance split between . Kroger, and J. Reimann. 2006. Priming protective CD8 T cell immunity by B virus precore and core proteins. J. Virol. 79: 3016–3027. DNA vaccines encoding chimeric, stress protein-capturing tumor-associated an- 13. Frelin, L., T. Wahlstrom, A. E. Tucker, J. Jones, J. Hughes, B. O. Lee, tigen. J. Immunol. 177: 1534–1542. J. N. Billaud, C. Peters, D. Whitacre, D. Peterson, and D. R. Milich. 2009. A mechanism to explain the selection of the antigen-negative mutant 37. Wieland, A., M. Denzel, E. Schmidt, S. Kochanek, F. Kreppel, J. Reimann, and during chronic hepatitis B virus infection. J. Virol. 83: 1379–1392. R. Schirmbeck. 2008. Recombinant complexes of antigen with stress proteins are 14. Boni, C., A. Bertoletti, A. Penna, A. Cavalli, M. Pilli, S. Urbani, P. Scognamiglio, potent CD8 T-cell-stimulating immunogens. J. Mol. Med. 86: 1067–1079. 38. Pascolo, S., N. Bervas, J. M. Ure, A. G. Smith, F. A. Lemonnier, and B. Perarnau. R. Boehme, R. Panebianco, F. Fiaccadori, and C. Ferrari. 1998. Lamivudine ϩ treatment can restore T cell responsiveness in chronic hepatitis B. J. Clin. Invest. 1997. HLA-A2.1-restricted education and cytolytic activity of CD8 T lympho- ␤ ␤ b ␤ 102: 968–975. cytes from 2 microglobulin ( 2m) HLA-A2.1 monochain transgenic H-2D 2m 15. Boni, C., A. Penna, G. S. Ogg, A. Bertoletti, M. Pilli, C. Cavallo, A. Cavalli, double knockout mice. J. Exp. Med. 185: 2043–2051. S. Urbani, R. Boehme, R. Panebianco, F. Fiaccadori, and C. Ferrari. 2001. Lami- 39. Kuhober, A., H. P. Pudollek, K. Reifenberg, F. V. Chisari, H. J. Schlicht, vudine treatment can overcome cytotoxic T-cell hyporesponsiveness in chronic J. Reimann, and R. Schirmbeck. 1996. DNA immunization induces antibody and b hepatitis B: new perspectives for immune therapy. Hepatology 33: 963–971. cytotoxic T cell responses to hepatitis B core antigen in H-2 mice. J. Immunol. 16. Tsai, S. L., P. J. Chen, M. Y. Lai, P. M. Yang, J. L. Sung, J. H. Huang, 156: 3687–3695. L. H. Hwang, T. H. Chang, and D. S. Chen. 1992. Acute exacerbations of chronic 40. Fissolo, N., P. Riedl, J. Reimann, and R. Schirmbeck. 2005. DNA vaccines prime ϩ type B hepatitis are accompanied by increased T cell responses to hepatitis B core CD8 T cell responses to epitopes of viral antigens produced from overlapping and e antigens: implications for hepatitis B e antigen seroconversion. J. Clin. reading frames of a single coding sequence. Eur. J. Immunol. 35: 117–127. Invest. 89: 87–96. 41. Sette, A., A. Vitiello, B. Reherman, P. Fowler, R. Nayersina, W. M. Kast, 17. Livingston, B. D., J. Alexander, C. Crimi, C. Oseroff, E. Celis, K. Daly, C. J. Melief, C. Oseroff, L. Yuan, J. Ruppert, et al. 1994. The relationship be- L. G. Guidotti, F. V. Chisari, J. Fikes, R. W. Chesnut, and A. Sette. 1999. Altered tween class I binding affinity and immunogenicity of potential cytotoxic T cell helper T lymphocyte function associated with chronic hepatitis B virus infection epitopes. J. Immunol. 153: 5586–5592. 380 INDUCTION OF SUBDOMINANT CD8 T CELL RESPONSES

42. Rehermann, B., P. Fowler, J. Sidney, J. Person, A. Redeker, M. Brown, B. Moss, nodominance hierarchy determined by antigen dose and precursor frequencies. A. Sette, and F. V. Chisari. 1995. The cytotoxic T lymphocyte response to mul- Proc. Natl. Acad. Sci. USA 103: 994–999. tiple hepatitis B virus polymerase epitopes during and after acute . 54. Rutigliano, J. A., T. J. Ruckwardt, J. E. Martin, and B. S. Graham. 2007. Relative ϩ J. Exp. Med. 181: 1047–1058. dominance of epitope-specific CD8 T cell responses in an F1 hybrid mouse 43. Rammensee, H. G., T. Friede, and S. Stevanoviic. 1995. MHC ligands and pep- model of respiratory syncytial virus infection. Virology 362: 314–319. tide motifs: first listing. Immunogenetics 41: 178–228. 55. Willis, R. A., J. W. Kappler, and P. C. Marrack. 2006. CD8 T cell competition 44. Moudgil, K. D., and E. E. Sercarz. 2005. Understanding crypticity is the key to for dendritic cells in vivo is an early event in activation. Proc. Natl. Acad. Sci. revealing the pathogenesis of autoimmunity. Trends Immunol. 26: 355–359. USA 103: 12063–12068. 45. Zehn, D., and M. J. Bevan. 2006. T cells with low avidity for a tissue-restricted 56. Price, D. A., J. M. Brenchley, L. E. Ruff, M. R. Betts, B. J. Hill, M. Roederer, R. A. Koup, S. A. Migueles, E. Gostick, L. Wooldridge, et al. 2005. Avidity for antigen routinely evade central and peripheral tolerance and cause autoimmunity. ϩ Immunity 25: 261–270. antigen shapes clonal dominance in CD8 T cell populations specific for persis- 46. Dumortier, J., K. Schonig, H. Oberwinkler, R. Low, T. Giese, H. Bujard, tent DNA viruses. J. Exp. Med. 202: 1349–1361. P. Schirmacher, and U. Protzer. 2005. Liver-specific expression of interferon ␥ 57. Choi, E. Y., G. J. Christianson, Y. Yoshimura, T. J. Sproule, N. Jung, S. Joyce, following adenoviral gene transfer controls hepatitis B virus replication in mice. and D. C. Roopenian. 2002. Immunodominance of H60 is caused by an abnor- Gene Ther. 12: 668–677. mally high precursor T cell pool directed against its unique minor histocompat- ibility antigen peptide. Immunity 17: 593–603. 47. Sandberg, J. K., P. Grufman, E. Z. Wolpert, L. Franksson, B. J. Chambers, and 58. Facchinetti, A., S. S. Dalla, S. Mezzalira, A. Rosato, and G. Biasi. 2005. A large K. Karre. 1998. Superdominance among immunodominant H-2Kb-restricted number of T lymphocytes recognize Moloney-murine leukemia virus-induced epitopes and reversal by dendritic cell-mediated antigen delivery. J. Immunol. antigens, but a few mediate long-lasting tumor immunosurveillance. J. Immunol. 160: 3163–3169. 174: 5398–5406. 48. Chen, W., L. C. Anton, J. R. Bennink, and J. W. Yewdell. 2000. Dissecting the 59. Haeryfar, S. M., R. J. DiPaolo, D. C. Tscharke, J. R. Bennink, and J. W. Yewdell. multifactorial causes of immunodominance in class I-restricted T cell responses ϩ 2005. Regulatory T cells suppress CD8 T cell responses induced by direct to viruses. Immunity 12: 83–93. priming and cross-priming and moderate immunodominance disparities. J. Im- 49. Tourdot, S., and K. G. Gould. 2002. Competition between MHC class I alleles for munol. 174: 3344–3351. cell surface expression alters CTL responses to influenza A virus. J. Immunol. 60. Kammerer, R., D. Stober, P. Riedl, C. Oehninger, R. Schirmbeck, and 169: 5615–5621. J. Reimann. 2002. Noncovalent association with stress protein facilitates cross- Downloaded from 50. Probst, H. C., K. Tschannen, A. Gallimore, M. Martinic, M. Basler, T. Dumrese, priming of CD8ϩ T cells to tumor cell antigens by dendritic cells. J. Immunol. E. Jones, and M. F. van den Broek. 2003. Immunodominance of an antiviral 168: 108–117. cytotoxic T cell response is shaped by the kinetics of viral protein expression. 61. Wild, J., M. J. Grusby, R. Schirmbeck, and J. Reimann. 1999. Priming MHC-I- J. Immunol. 171: 5415–5422. restricted cytotoxic T lymphocyte responses to exogenous hepatitis B surface 51. Basler, M., N. Youhnovski, B. M. Van Den, M. Przybylski, and M. Groettrup. antigen is CD4ϩ T cell dependent. J. Immunol. 163: 1880–1887. 2004. Immunoproteasomes down-regulate presentation of a subdominant T cell 62. Maier, H., M. Isogawa, G. J. Freeman, and F. V. Chisari. 2007. PD-1:PD-L1 epitope from lymphocytic choriomeningitis virus. J. Immunol. 173: 3925–3934. interactions contribute to the functional suppression of virus-specific CD8ϩ T ϩ

52. Yewdell, J. W., and M. Del Val. 2004. Immunodominance in T CD8 responses lymphocytes in the liver. J. Immunol. 178: 2714–2720. http://www.jimmunol.org/ to viruses: cell biology, cellular immunology, and mathematical models. Immu- 63. Blackburn, S. D., H. Shin, W. N. Haining, T. Zou, C. J. Workman, A. Polley, nity 21: 149–153. M. R. Betts, G. J. Freeman, D. A. Vignali, and E. J. Wherry. 2009. Coregulation 53. La Gruta, N. L., K. Kedzierska, K. Pang, R. Webby, M. Davenport, W. Chen, of CD8ϩ T cell exhaustion by multiple inhibitory receptors during chronic viral S. J. Turner, and P. C. Doherty. 2006. A virus-specific CD8ϩ T cell immu- infection. Nat. Immunol. 10: 29–37. by guest on September 29, 2021