SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Vaccines for the prophylactic and/or therapeutic immunization against hepatotropic pathogens (e.g., hepatitis B and hepatitis C virus) should establish long-lasting, specific antiviral effector/memory CD8+ T cell immunity in the liver. We describe a novel peptide-based vaccine in which antigenic major histocompatibility complex Class I–binding peptides are fused to a cationic (e.g., human immunodeficiency virus tat-derived) domain and complexed to immune-stimulating oligonucleotides. This vaccine formulation efficiently primes liver-homing, Class I–restricted CD8+ effector/memory T cell responses. In different antigen systems, this formulation was more potent in priming liver-homing CD8+ T cell responses than DNA-based vaccines delivering the same epitopes. CD8+ T cell priming was independent of CD4+ T cell “help” but submitted to regulatory control by CD25+ CD4+ T cells. The vaccine efficiently primed memory/effector CD8+ T cells detectable in the liver for more than 3 months after a single injection. With increasing time after priming, the phenotype of these specific memory CD8+ T cells shifted from an effector memory to a central memory type. The vaccine could override T cell tolerance in mice expressing the relevant antigen from a transgene in the liver. The CD8+ T cell immunity in the liver primed by this peptide formulation could be boosted by challenge injections. In conclusion, we describe a simple and potent vaccine formulation that has the potential to generate or reconstitute specific CD8+ T cell immunity to hepatotropic pathogens in the liver. (HEPATOLOGY 2004;40:300–309.)

Innovative vaccines against human immunodeficiency virus (HIV), hepatitis C virus (HCV), or hepatitis B virus (HBV) have shown that the cellular immune responses they elicit can attenuate the clinical disease associated with the respective chronic virus infection.1 Consequently, a key current objective in vaccinology is the priming of antiviral CD8+ T cell immunity by novel antigen delivery strategies. Four types of CD8+ T cell–stimulating vaccines are currently under investigation: recombinant viruses, DNA vaccines, virus-like particles, and peptides. Genetic vaccination is considered one of the most potent techniques to prime Class I–restricted T cell responses in the mouse.2, 3 We herein describe a peptide-based vaccine and compare its relative efficacy in CD8+ T cell priming to that of DNA vaccines.

The peptide vaccine was developed as a result of the observation that injection of a low dose (2 μg/mouse) of recombinant, native HBV core particles (without adjuvants) into a mouse primes a T helper 1 immune response, while injection of mutant core particles with a deletion of the C-terminal 20–30 cationic residues primes a T helper 2 response.4 When the C-terminal, cationic peptide was synthesized and complexed to low doses of oligonucleotides (ODNs), it provided a T helper 1 adjuvant effect to codelivered antigen.5 The most potent immunogen was generated when an antigenic peptide was fused to the cationic peptide and loaded with ODNs or polynucleotides.6 Priming mice with a low dose of this peptide vaccine elicited long-lasting CD8+ T cell immunity. Enhanced clonal expansion in vivo after priming and the longevity of the elevated numbers of specific CD8+ T cells in the spleen indicated that this vaccine formulation was superior to DNA vaccination in different antigen systems.

Priming CD8+ T cell response usually requires “help” from T cells recognizing different epitopes of the antigen, often in the context of major histocompatibility complex Class II molecules. Codelivery of either immune-stimulating ODNs7 or activation signals to the innate immune system8 can bypass the requirement for help in CD8+ T cell priming. The advantage of priming monospecific CD8+ T cell responses with single epitope vaccines is that competing T cell epitopes9 or epitopes stimulating “suppressive” regulatory T cells are absent.10–12 It is therefore of interest to elucidate the CD4+ T cell help requirements as well as the control by regulatory T (TR) cells of CD8+ T cell responses primed by such experimental peptide vaccines.

The objective of vaccination is the establishment of stable, long-lasting, specific memory. The longevity of CD8+ T cell memory is an indicator for the relative efficacy of a vaccine formulation in priming T cell immunity.13, 14 We therefore followed the establishment and decline of specific CD8+ T cell immunity after priming and tested its response to boost vaccinations.

Vaccine-primed T cells have to specifically deliver their effector functions to the relevant target organ. In the case of HCV or HBV, this is the liver. The liver is a difficult organ immunologically because it is as good at tolerating T cells as it is at supporting their specific activation.15, 16 The focus of this report is the establishment of CD8+ T cell immunity specific for the hepatotropic, human pathogen HBV in the liver via a novel peptide vaccination approach.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Mice.

C57BL/6JBom (B6) mice (H-2b), and CD4+ T cell–deficient Aα−/− or Aβ−/− B6 mice17, 18 were bred and kept under standard pathogen-free conditions in the animal colony of Ulm University (Ulm, Germany). C57BL/6J-TgN (Alb1HBV) 44Bri (hepatitis B surface antigen–expressing transgenic [HBs-tg]) mice from the Jackson Laboratory (Bar Harbor, ME) show albumin promoter-driven hepatitis B surface antigen (HBsAg)ayw expression from a transgene in the liver. All animal experiments were conducted according to the guidelines of the local Animal Use and Care Committees and were executed according to the National Animal Welfare Law.

Cationic Peptide/ODN Vaccines and Immunization of Mice.

CpG-containing ODN-1826 (TCCATGACGTTTCCTGACGTT; ISS-1826)19 obtained from MWG (Ebersberg, Germany) was dissolved in H2O (10 mg/mL stock solution). The synthetic peptides (Jerini BioTools, Berlin, Germany) used included the following: Kb-restricted T4404–412 VVYDFLKCM from the simian virus 40 (SV40) large tumor antigen (T-Ag); Kb-restricted S1208–215 ILSPFLPL from HBsAgayw, and its natural variant IVSPFIPL from HBsAgadw2 that binds Kb with higher affinity20, 21; Kb-restricted S2190–197 VWLSVIWM from HBsAgayw that binds Kb with high affinity20, 21; and Kb-restricted SIINFEKL from ovalbumin (OVA). Peptides were used either alone or as a fusion peptide with the cationic domain of the HIV-tat50–57 KKRRQRRR peptide. Peptides were dissolved in dimethyl sulfoxide at a concentration of 10 mg/mL. To generate peptide/ODN complexes, ODNs were incubated for 30 minutes with peptides in phosphate-buffered saline (pH 7.4) as described previously.6 Mice were immunized via a single intramuscular injection of 50 μg peptide and 30 μg ODN dissolved in phosphate-buffered saline. We injected 50 μL into each tibialis anterior muscle.22 In some groups, mice were immunized via an intramuscular injection of either 100 μg plasmid-DNA (pCI/T encoding T-Ag, pCI/S encoding HBsAg) or 10 μg/mouse HBsAg particles (mixed with 20 μg/mouse ODN). HBsAg particles produced in the Hansenula polymorpha host strain RB10 were kindly provided by Dr. K. Melber (Rhein Biotech GmbH, Düsseldorf, Germany).

CD25+ CD4+ T Cell Depletion By Treatment With α-CD25 mAb PC61.

B6 mice were injected intraperitoneally three times with 1 mg monoclonal antibody (mAb) PC61 (or an isotype-matched control mAb) at 2-day intervals.23 Depletion of CD25+ CD4+ cells was confirmed via flow cytometry. Mice were vaccinated 3 days after the last antibody injection.

Flow Cytometry Analysis.

Isolation of spleen cells and hepatic nonparenchymal cells has been described.6, 24 Cells were washed twice in phosphate-buffered saline/0.3% w/v bovine serum albumin supplemented with 0.1% w/v sodium azide. Nonspecific binding of antibodies to the Fc receptor was blocked by preincubating cells with mAb 2.4G2 (cat.no.01241D; BD Biosciences, Heidelberg, Germany) directed against the FcγRIII/II CD16/CD32 (0.5 μg mAb/106 cells/ 100 μL). Cells were washed and incubated with 0.5 μg/106 cells of the relevant mAb for 30 minutes at 4°C and washed again twice. In most experiments, cells were subsequently incubated with a second-step reagent for 10 minutes at 4°C. Four-color flow cytometry analyses were performed with FACScan (Becton Dickinson, Mountain View, CA). The forward narrow angle light scatter was used as an additional parameter to facilitate exclusion of dead cells and aggregated cell clumps. Data were analyzed using WinMDI software (The Scripps Research Institute, La Jolla, CA). The following reagents and mAb were obtained from BD Biosciences: fluorescein-isothiocyanate (FITC)- and phycoerythrin (PE)-conjugated anti-NK1.1 mAb PK136 (cat.no. 553165, 553165), PE- and allophycocyanin-conjugated anti-CD8α mAb 53-6.7 (cat.no.01045B and 553035), biotinylated and FITC-conjugated anti-CD44 (Pgp-1) mAb IM7 (cat.no.01222D and 01224D), biotinylated anti-CD28 mAb 37.51 (cat.no.01672D), biotinylated anti-CD69 mAb H1.2F3 (cat.no.01502D), biotinylated anti-B220 mAb (cat.no.553086), biotinylated anti-CD27 mAb (cat.no.558733), biotinylated anti-Annexin V mAb (cat.no. 556417), biotinylated anti-CD45RB mAb (cat.no.01532D), FITC-conjugated anti-CD62L mAb (cat.no.01265B), and FITC-conjugated anti-CCR5 mAb (cat.no.559923). SA-Red 670 (cat.no.19543-024) was obtained from Gibco-BRL (Berlin, Germany).

Determination of Splenic and Hepatic CD8+ T Cell Frequencies.

Determination of frequencies of CD8+ T cells specifically inducible to interferon γ (IFNγ) expression has been previously described.6, 24 Alternatively, cells from spleen or liver freshly isolated from nonimmune or immune mice (but not restimulated in vitro with peptide) were stained with FITC-conjugated α-CD8 mAb, PE-conjugated Kb/S2VWLSVIWM tetramer (ProImmune, Oxford, UK; 0.5 μg/106 cells/100 μL), and additional conjugated mAb's (binding markers of interest) for 30 minutes at 4°C. The cells were then washed twice and analyzed using multicolor flow cytometry.

Statistical Analyses.

Data were analyzed using GraphPAD prism software, version 3.0 (GraphPad Software, San Diego, CA). Values are presented as mean ± SEM.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Priming Specific CD8+ T Cell Immunity in the Liver With an Antigenic Peptide Containing a Cationic Domain With Complexed ODN.

We attempted to determine if an antigenic peptide fused to a cationic peptide and bound to ODN primes liver-homing CD8+ T cells using the Kb-restricted response to the T4 epitope of the SV40 T-Ag. Neither a single injection of the antigenic peptide (50 μg/mouse) alone (Fig. 1, group 1) nor the antigenic peptide mixed with ODN (group 2) established specific CD8+ T cell immunity. Similarly, injection of the T4-tat50–57 fusion peptide with antigenic and cationic sequences was not immunogenic for CD8+ T cell precursors (group 3). Multiple injections of different doses of these peptides and readouts at different time points after vaccination did not reveal evidence for specific CD8+ T cell priming (data not shown). In contrast, when the T4-tat50–57 fusion peptide was complexed with ODN, a single injection of this formulation efficiently primed CD8+ T cells (group 4). Even a single injection of 2 or 10 μg peptide (plus ODN) per mouse elicited a T cell response, stressing the exceptional potency of the vaccine formulation (data not shown). Most importantly, specific CD8+ T cells were found not only in the spleen but also in high numbers in the liver (see Fig. 1). We therefore tested if this vaccine formulation can establish stable CD8+ T cell immunity in this organ.

thumbnail image

Figure 1. The Kb-binding T4404–412 peptide of the SV40 T-Ag fused to a cationic HIV-tat50–57 domain and bound to oligonucleotides efficiently primes specific CD8+ T cell immunity. B6 mice were injected intramuscularly with either the T4404–412 VVYDFLKCM peptide or a fusion of the T4 peptide with the cationic tat KKRRQRRR peptide (T4-tat50–57). Peptides (50 μg/mouse) were injected with or without ODN (30 μg/mouse). Twelve days postvaccination, the frequencies of specific splenic or hepatic CD8+ IFNγ+ T cells were determined after a 4-hour ex vivo restimulation with T4 peptide (1 μg/mL) in the presence of brefeldin A (5 μg/mL). Mean numbers of specific CD8+ IFNγ+ T cells/105 CD8+ T cells from 3 individual mice (± SEM) from a representative experiment out of 3 independent experiments are shown. ODN, oligonucleotides; IFNγ, interferon γ.

Download figure to PowerPoint

ODN-Complexed Antigenic/Cationic Peptides Elicit Long-lasting, Specific CD8+ T Cell Immunity in the Liver.

We immunized B6 mice with a single dose of the ODN-complexed antigenic/cationic peptide vaccine and determined the number of specific CD8+ T cells in the liver at different time points of a 90-day period postvaccination (Fig. 2). The control plasmid DNA vaccine that encodes the complete SV40 T-Ag has been shown to be exceptionally potent in priming cytotoxic and tumor-rejecting CD8+ T cell responses.25 High numbers of specific CD8+ T cells were detected in the spleens and livers of mice vaccinated with the ODN-complexed antigenic/cationic peptide vaccine, even at early time points (day 7) postvaccination. Twelve days after a single injection of the T4-tat50–57 fusion peptide bound to ODN, up to 3% of the CD8+ T cells in the liver showed this specific reactivity. The frequency of specific CD8+ T cells within the total CD8+ T cell population was higher in the liver than in the spleen. The number of specific CD8+ T cells in the liver declined in the 3 months following a single injection. Compared with DNA vaccination, higher numbers of specific CD8+ T cells were found in the spleen and liver throughout the 90-day observation period after immunization with the T4-tat/ODN vaccine (see Fig. 2). Although antigen encoded by a DNA vaccine usually elicits potent CD8+ T cell responses in mice, the antigenic/cationic peptide complexed to ODN clearly displays superior immunogenicity. CD8+ T cell priming by most DNA vaccines is CD4+ T help–dependent.7 We did not incorporate a “helper” epitope into the antigenic/cationic peptide used. We immunized CD4+ T cell–deficient mice to determine if they can generate specific CD8+ T cell immunity in response to the antigenic/cationic peptide complexed to ODN.

thumbnail image

Figure 2. T-Ag–specific CD8+ T cells in liver and spleen after immunization with plasmid DNA or antigenic/cationic fusion peptide. B6 mice were immunized with the T4-tat50–57 peptide (as described in Fig. 1) or the plasmid DNA pCI/T encoding the T-Ag (100 μg/mouse). At the indicated time points liver and spleen cells were isolated and restimulated in vitro with the T4 peptide (1 μg/mL) for 4 hours in the presence of brefeldin A (5 μg/mL). Mean numbers of specific CD8+ IFNγ+ T cells/105 CD8+ T cells in 3 individual mice per time point (± SEM) are shown. IFNγ, interferon γ; ODN, oligonucleotides.

Download figure to PowerPoint

Priming Liver-Homing CD8+ T Cells By Antigenic Peptides With a Cationic Tail Complexed to ODN Is CD4+ T Cell Help-Independent.

Normal B6 mice and CD4+ T cell–deficient B6 mice (that lack either the Aα or Aβ chain of major histocompatibility complex Class II) were immunized with antigenic peptides containing a cationic tail complexed to ODN. The vaccine formulation efficiently primed CD8+ T cells in the presence and absence of CD4+ T cell help, as was evident by the readily detectable numbers of specific CD8+ T cells in the spleens and livers of vaccinated Aα−/− or Aβ−/− B6 mice (Fig. 3 and additional data not shown). The frequencies and the absolute numbers of specific CD8+ T cells were higher in spleens and livers of CD4+ T cell–deficient than in CD4+ T cell–competent mice (see Fig. 3). Because higher numbers of specific CD8+ T cells were found in CD4+ T cell–deficient mice, we asked if CD4+ T cells limit priming, clonal expansion, and/or survival of specific CD8+ T cells in spleen and liver.

thumbnail image

Figure 3. CD8+ T cell priming by cationic fusion peptide/ODN is CD4+ T cell “help”-independent. Normal B6 and congenic CD4+ T cell–deficient Aβ−/− B6 mice were immunized with T4-tat50–57/ODN intramuscularly. At the indicated time points postimmunization, liver and spleen mononuclear cells were isolated and restimulated in vitro with the Kb-binding T4 peptide (1 μg/mL) for 4 hours in the presence of brefeldin A (5 μg/mL). Cells were surface-stained with anti–CD8 mAb, washed, fixed, and stained for intracellular IFNγ.-4050μ Mean numbers of specific IFNγ+ CD8+ T cells/105 CD8+ T cells of 3 individual mice per group per time point (± SEM) are shown. IFNγ, interferon γ; B6, C57BL/6; wt, wild-type.

Download figure to PowerPoint

Priming of Liver-Homing CD8+ T Cells By Antigenic/Cationic Peptides Is Submitted to Negative Control of CD25+ CD4+ TR Cells.

CD8+ T cell activation and CD8+ T cell memory is controlled by CD25+ CD4+ TR cells.10, 11, 26, 27 We analyzed whether or not CD25+ TR cells limit the specific CD8+ T cell response primed by the injection of peptide/ODN complexes. Three injections of αCD25 mAb PC61 (but not isotype-matched control mAb) within a week depleted the CD25+ CD4+ T cell population (Fig. 4A). In mice with a suppressed CD25+ TR cell population, two- to threefold higher numbers of specific CD8+ T cells were found in spleen and liver at day 12 postvaccination (Fig. 4B). Higher numbers of specific CD8+ T cells were also found in spleen and liver 6 days after a specific challenge when B6 mice primed in the presence of CD25+ CD4+ T cells were depleted of CD25+ CD4+ T cells before the boost injection (Fig. 4C). Thus, priming of the CD8+ T cell response as well as the specific restimulation of its memory response by the vaccine formulation studied are negatively controlled by CD25+ CD4+ TR cells. It is unclear which epitopes of the vaccine are recognized by CD25+ TR cells, but the 16mer peptide used limits the available antigenic information.

thumbnail image

Figure 4. Priming of CD8+ T cells by the cationic peptide/ODN vaccines is negatively regulated by CD25+ CD4+ TR cells. Prior to vaccination, mice were injected repeatedly with either the α-CD25 mAb PC61, or an isotype-matched control mAb (1 mg antibody/injection). (A) Hepatic and splenic CD25+ CD4+ TR cells were depleted via antibody treatment at the time of vaccination. (B) B6 mice were immunized intramuscularly with the T4-tat50–57 or S2-tat50–57 fusion peptides complexed with ODN. Twelve days after priming, liver and spleen cells were isolated and restimulated in vitro (1 μg/mL for 4 hours) with either the T4 peptide VVYDFLKCM of T-Ag, the S2 peptide VWLSVIWM of HBsAg, or the control peptide SIINFEKL of ovalbumin. The mean numbers of specific IFNγ+ CD8+ T cells/105 CD8+ T cells in 3 individual mice per group (± SEM) from a representative out of 2 independent experiments are shown. (C) Three months after priming, T4-tat50–57–immune mice were treated with either the anti-CD25 or control mAb and boosted. Six days after the boost injection, the numbers of T4-specific CD8+ T cells were determined in spleen and liver. The mean numbers of specific IFNγ+ CD8+ T cells/105 CD8+ T cells in 5 individual mice per group (± SEM) are shown. mAb, monoclonal antibody; FITC, fluorescein-isothiocyanate; PE, phycoerythrin; OVA, ovalbumin; ODN, oligonucleotide; IFNγ, interferon γ.

Download figure to PowerPoint

Antigenic and Cationic Peptide/ODN Complexes Derived From HBsAg Efficiently Prime Liver-Homing CD8+ T Cell Responses.

We selected the two Kb-binding epitopes S1208–215 IVSPFIPL and S2190–197 VWLSVIWM of the HBsAg20, 21, 28 for the following vaccination experiments. These epitopes were synthesized as fusion peptides with a C-terminal, cationic tat50–57 KKRRQRRR tail. The fusion peptides complexed with ODN were injected into B6 mice. Control mice were immunized either with the pCI/S DNA vaccine that encodes the small HBsAgayw, or HBsAgayw particles with ODN.29 Both fusion peptides efficiently primed specific CD8+ T cells that were found in spleen and liver 12 days postvaccination (Fig. 5A). The numbers of specific CD8+ T cells present in both organs after cationic peptide/ODN vaccination were usually higher than those found after vaccination with plasmid DNA or recombinant HBsAg particles. The responses were specific for the immunizing epitope, because no cross-reactivities between the S1 and the S2 reactivity—or to the control (Kb-binding ovalbumin) peptide—were detectable.

thumbnail image

Figure 5. HBsAg-specific CD8+ T cells primed by cationic, antigenic eptide/ODN. (A) B6 mice were immunized with: the Kb-binding S1208–215 IVSPFLPL or S2190–197 VWLSVIWM peptide from HBsAg fused to tat50–57 KKRRQRRR domain; pCI/S plasmid DNA encoding HBsAg (100 μg/mouse); or HBsAg particles (10 μg/mouse) with ODN (20 μg/mouse). Twelve days postimmunization, the mean numbers of specific IFNγ+ CD8+ T cells/105 CD8+ T cells in liver and spleen of 4 individual mice per group (± SEM) was determined. (B) Normal (CD4+ T cell–competent) B6 mice and CD4+ T cell–deficient (Aα−/−) B6 mice were immunized with the Kb-binding S2190–197 VWLSVIWM peptide fused to the cationic tat50–57 domain and complexed to ODN. At the indicated time points postimmunization, liver and spleen cells were isolated and restimulated in vitro with the S2 peptide (1 μg/mL) in the presence of brefeldin A (5 μg/mL) before staining surface CD8 and intracellular IFNγ. In parallel experiments, freshly isolated liver and spleen mononuclear cells were directly stained with anti–CD8 mAb and Kb/S2-tetrameter (tet) (without in vitro restimulation) to assess the frequencies of tet+ CD8+ T cells. Mean numbers of specific IFNγ+ CD8+ T cells/105 CD8+ T cells or Kb/S2 tet+ CD8+ T cells/105 CD8+ T cells from 4 individual mice per group per time point (± SEM) are shown. ODN, oligonucleotide; OVA, ovalbumin; IFNγ, interferon γ; HBsAg, hepatitis B surface antigen; wt, wild-type; tet, tetramer.

Download figure to PowerPoint

In addition to specifically inducible IFNγ expression, we used Kb/S2 tetramers (tet) to determine the number of HBsAg-specific tet+ CD8+ T cells primed by the cationic S2 peptide complexed to ODN (Fig. 5B). Tet+ CD8+ T cell numbers were higher than those detected by specifically inducible IFNγ expression. Mean numbers and frequencies of specific tet+ or IFNγ+ CD8+ T cells in spleen or liver of vaccinated normal, CD4+ T cell–deficient, and CD25+ CD4+ TR cell–deficient B6 mice are summarized in Table 1. The data indicate that approximately half of the tet+ CD8+ T cells are specifically inducible to IFNγ expression within 4 hours. When specific CD8+ T cell numbers were followed for 35 days postvaccination in spleen and liver, a higher frequency of specific CD8+ T cells was found in the liver than in the spleen, and 30%–40% of the peak numbers of the specific CD8+ T cells were still detected in the liver 5 weeks after a single injection of a low dose of the vaccine (see Fig. 5B). The data confirm in an unrelated antigen system that higher numbers of specific CD8+ T cells are primed by the described peptide vaccine in normal and CD4+ T cell–deficient (Aα−/−) B6 mice stressing the exceptional immunogenicity of the vaccine formulation for CD8+ T cell precursors.

Table 1. Frequencies and Absolute Numbers of Antigen-Specific IFNγ+ and Tetramer (Kb/S2)+ CD8+ T Cells in Liver and Spleen After Immunization With S2-tat50–57/ODN
 Normal B6 MiceCD4+ T Cell–Deficient (Aα−/−) B6 MiceCD25+ TR Cell–Deficient B6 Mice
LiverSpleenLiverSpleenLiverSpleen
  • *

    Absolute numbers and frequencies of Kb/S2 tet+ CD8+ T cells 12 days after immunization with S2-tat50–57/ODN. Freshly isolated mononuclear cells from liver and spleen were surface-stained with anti–CD8 mAb and Kb/S2 tetramer without in vitro restimulation with peptide.

  • Absolute numbers and frequencies of IFNγ+ CD8+ T cells 12 days after immunization with S2-tat50–57/ODN. Freshly isolated mononuclear cells from liver and spleen were restimulated in vitro with the Kb-restricted S2 peptide (1 μg/mL) in the presence of brefeldin A (5 μg/mL) for 4 hours. Cells were subsequently surface-stained with anti–CD8 mAb, fixed, and stained for intracellular IFN-γ.

  • Percentage of intracellular IFNγ expressing cells within the CD8+ Kb/S2 tet+ T cell population. From freshly isolated, labeled cells, Kb/S2 tetα+ CD8+ T cells were electronically sorted and restimulated in vitro for 4 hours with RAG−/− B6 mouse spleen cells pulsed with the Kb-binding S2 peptide (1 μg/mL) in the presence of brefeldin A (5 μg/mL). Cells were harvested, washed, and stained for intracellular IFNγ.

Number of tet+ CD8+ T cells × 105/organ (± SEM)*3.8 ± 0.623 ± 1.48.1 ± 1.842.4 ± 2.37.1 ± 0.938.1 ± 4.1
Frequencies of tet+ CD8+ T cells/105 CD8+ T cells (± SEM)*5100 ± 3202900 ± 3509000 ± 4106100 ± 9107800 ± 2806300 ± 560
Number of IFN-γ+ CD8+ T cells × 105/organ (± SEM)2.4 ± 0.312 ± 1.14.3 ± 0.523 ± 2.14.7 ± 0.829 ± 1.1
Frequencies of IFN-γ+ CD8+ T cells/105 CD8+ T cells (± SEM)2700 ± 2401290 ± 1504700 ± 3403100 ± 3605300 ± 8903280 ± 230
Percentage of IFN-γ+ T cells in the specific tet+ CD8+ T cell population‡52 ± 2.244 ± 1.552 ± 3.150 ± 2.367 ± 2.952 ± 1.9

Surface Phenotype of Primed, Specific CD8+ T Cells in the Liver.

Two subsets of memory CD8+ T cell subpopulations have been identified: CD44hi CD45RBlo CD62Llo CD122lo effector memory CD8+ T cells and CD44hi CD45RBhi CD62Llo/hi CD122hi central memory CD8+ T cells.30 Effector memory CD8+ T cells rapidly respond within 4–6 hours with IFNγ expression to antigen challenge while central memory CD8+ T cells respond after a delay of more than 24 hours. In the first series of experiments, we tested the surface phenotype of CD8+ T cells with rapidly inducible, specific IFNγ expression in the liver. In a second series of experiments, we determined the surface phenotype of intrahepatic Kb/S2 tet+ CD8+ T cells at different time points after vaccination.

At day 12 postimmunization with the T4-tat/ODN vaccine, nearly all hepatic CD8+ T cells from immunized (CD4+ T cell–competent or CD4+ T cell–deficient) B6 mice showing specifically inducible IFNγ expression were CD44+ (Fig. 6). They showed evidence of activation (up-regulated CD69 and NK1 expression) and effector cell differentiation (down-regulated CD27 and CD28 expression) but no evidence of apoptosis (no change in B220 and Annexin V expression). A similar surface phenotype was found in hepatic CD8+ T cells from primed CD4+ T cell–competent B6 mice and CD4+ T cell–deficient B6 Aβ−/− mice (see Fig. 6). These (subset B) CD8+ T cells therefore express the phenotype of recently primed effector cells.

thumbnail image

Figure 6. Surface phenotype of specific CD8+ IFNγ+ T cells in the liver recently primed by cationic T4-tat50–57 fusion peptide/ODN. CD4+ T cell–competent and –deficient (Aβ−/−) B6 mice were immunized with the T4-tat50–57 fusion peptide/ODN. Twelve days after immunization, liver nonparenchymal cells were isolated and restimulated in vitro with the T4 peptide (1 μg/mL). Cells were surface-stained for CD8 and one of the additional surface markers depicted in the table and costained for intracellular IFNγ. In flow cytometry, analysis gates were set on either the IFNγ CD8+ (population A) or IFNγ+ CD8+ (population B) cells. The mean percentages of CD8+ T cells within the A and B subsets that express the respective markers in 3 individual mice per group (± SEM) from a representative out of 3 independent experiments are shown. B6, C57BL/6; IFNγ, interferon γ; PE, phycoerythrin; FITC, fluorescein-isothiocyanate.

Download figure to PowerPoint

We further analyzed the surface phenotype of hepatic Kb/S2 tet+ CD8+ in B6 mice 14 days after immunization with the S2-tat/ODN vaccine (Fig. 7, subset B1). Freshly isolated tet+ CD8+ T cells (that were not restimulated in vitro with peptide) showed the CD44+ CD45RBlo CD122lo CD27lo and CD62Llo effector memory surface phenotype.30 A small subset of the specific CD8+ T cells up-regulated the β-chemokine receptor CCR5 (CD195) that binds the chemokines RANTES or MIP-1α/β. Hepatic tet+ CD8+ T cells isolated 5 weeks postvaccination expressed a CD44+ CD45RBhi CD122hi/lo CD27hi and CD62Lhi surface phenotype (see Fig. 7, subset B2). Up-regulated surface expression of CD45RB, CD27, CD122, and CD62L in a fraction of specific CD8+ memory T cells indicates that central CD8+ T cell memory develops in the liver postvaccination. The vaccination protocol can thus establish central CD8+ T cell memory in the liver.

thumbnail image

Figure 7. Surface phenotype of specific liver CD8+ T cells primed by cationic S2190–197-tat50–57 fusion peptide/ODN. B6 mice were immunized with the S2190–197-tat50–57 fusion peptide/ODN vaccine. At the indicated time points postimmunization, freshly isolated liver cells were stained with allophycocyanin-labeled α–CD8 mAb, PE-labeled tet (Kb/S2 tet), and mAb binding the indicated surface markers. Cells were analyzed with four-color flow cytometry. The gates were set on the Kb/S2 tet (A) or the Kb/S2 tet+ (B1 and B2) CD8+ T cell subsets. Data from a representative, individual mouse out of 3 independent mice analyzed are shown in the dot blots. The mean percentage of T cells within the A and B subsets that express the respective surface marker in 3 individual mice analyzed (± SEM) is shown. tet, tetramer.

Download figure to PowerPoint

The Specific CD8+ T Cell Response Primed By Cationic Peptide/ODN Vaccines Can Be Boosted in a CD4+ T Cell–Independent Way.

Immunized normal (CD4+ T cell–competent) or Aβ−/− (CD4+ T cell–deficient) B6 mice were boosted 13 weeks postpriming. At this time point, the number of specific CD8+ T cells in spleen and liver was low (Fig. 8 and additional data not shown). An injection of the same dose of the T4-tat/ODN vaccine stimulated the reappearance of specific CD8+ T cells in spleen and liver. The numbers of specific CD8+ T cells were similar to those observed in the second week postpriming. This was seen after vaccination with different cationic peptide/ODN vaccines, and using inducible IFNγ or tetramer staining as the readout. Hence vaccination with antigenic peptide fused to a cationic domain and loaded with ODN efficiently primes liver-homing, antiviral CD8+ T cell responses with a central memory phenotype that can be boosted by challenge injections.

thumbnail image

Figure 8. Specific CD8+ T cell responses primed by T4-tat50–57/ODN vaccines can be boosted. Normal and CD4+ T cell–deficient Aβ−/− B6 mice were primed with T4-tat50–57 fusion peptide/ODN. Ninety days after priming, mice were boosted by the same vaccine. Liver nonparenchymal cells were isolated 12 and 40 days after the boost and restimulated with T4 peptide (1 μg/mL). (A) Frequencies of CD8+ IFNγ+ T cells/105 CD8+ T cells were determined in B6 mice immunized (prime/boost) with either T4-tat50–57/ODN or pCI/T. (B) Frequencies of CD8+ IFNγ+ T cells/105 CD8+ T cells were determined in CD4+ T cell–deficient Aβ−/− B6 mice after prime/boost immunization with the T4-tat50–57/ODN vaccine. Mean numbers of specific IFNγ+ CD8+ T cells/105 CD8+ T cells of 5 individual mice per group per time point (± SEM) are shown. ODN, oligonucleotide; IFNγ, interferon γ; B6, C57BL/6.

Download figure to PowerPoint

Cationic Peptide/ODN Vaccines Can Break CD8+ T Cell Tolerance to Epitopes of HBsAg Produced in the Liver.

Transgene-driven expression of HBsAgayw in the liver establishes tolerance to this viral antigen.31 We immunized HBs-tg B6 mice with cationic/antigenic peptides (plus ODN) containing Kb-binding HBsAg epitopes to test if this nonresponsiveness can be overcome.

In the first series of experiments, we immunized HBs-tg mice with two natural variants of the HBsAg epitope S1: S1 from transgene-encoded HBsAgayw binds Kb with low affinity, while a mutant S1 variant from HBsAgadw2 binds Kb with higher affinity.20, 21 A single vaccination with cationic/antigenic peptides (complexed to ODN) containing the high but not the low affinity S1 variant could break tolerance in HBs-tg mice (Fig. 9), because S1-specific CD8+ T cells were found in the spleens and livers of vaccinated transgenic mice. In the second series of experiments, we immunized HBs-tg mice with the S2 epitope of HBsAgayw that binds Kb with high affinity.20, 21 A single vaccination with cationic/antigenic peptides (complexed to ODN) containing the S2 epitope generated specific CD8+ T cells in HBs-tg mice readily detectable in spleen and liver (see Fig. 9). Hence tolerance to a liver-derived viral antigen can be overcome if a high affinity epitope is incorporated into the cationic peptide vaccine. This indicates that the described vaccine formulation can override nonresponsiveness.

thumbnail image

Figure 9. Breaking CD8+ T cell tolerance to HBsAg by cationic peptide/ODN vaccines. B6 mice and syngeneic HBs-tg B6 mice that express HBsAgayw in the liver were immunized once with cationic (tat50–57)/antigenic peptides (complexed to ODN) containing either the ayw or adw2 variant of S1208–215 or the ayw variant of S2190–197. Twelve days after immunization, liver nonparenchymal cells and spleen cells were isolated and restimulated with the peptides S1(adw2), S2(ayw), or ovalbumin (1 μg/mL for 4 hours). Cells were labeled with PE-conjugated anti–CD8 mAb and intracellularly stained with FITC-conjugated anti–IFNγ mAb. Mean numbers of IFNγ+ CD8+ T cells/105 CD8+ T cells in 4 mice per group (± SEM) are shown. wt, wild-type; B6, C57BL/6; HBs-tg, hepatitis B surface antigen–expressing transgenic; OVA, ovalbumin; IFNγ, interferon γ.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We describe a peptide-based vaccine formulation that efficiently primes Class I–restricted, liver-homing CD8+ T cell responses. In the antigen systems tested, the peptide-based formulation showed superior immunogenicity to DNA-based vaccines in priming CD8+ T cell responses. T cell priming was independent of CD4+ T cell help but was restricted by regulatory control of CD25+ CD4+ TR cells. The vaccine efficiently established memory/effector CD8+ T cell populations in the liver. At later time points postpriming, the central memory phenotype predominated in the specific CD8+ T cell population in the liver. The CD8+ T cell response could be boosted by challenge injections. The vaccine formulation described is thus an attractive candidate to generate specific CD8+ T cell immunity to hepatotropic pathogens in the liver.

CD8+ T cells are specific effector cells that control intracellular pathogens. The readout we used was either specifically inducible IFNγ expression or tet staining. All IFNγ+ CD8+ T cells were tet+. The tet stainings always detected two- to threefold more specific CD8+ T cells, suggesting that not all primed CD8+ T cells can be induced to IFNγ expression within a specific and restricted 4-hour restimulation ex vivo. The numbers we show may still be underestimates, because CD8+ T cells with altered tet binding have been previously described.32–35 Especially in the liver, this phenotype may be expressed by CD8+ T cells, though it is difficult to demonstrate this experimentally.

The T cell biology of the liver has different facets, which have been expertly reviewed recently.16 Local presentation of antigen causes T cell inactivation, tolerance, and apoptosis that may result from the need to maintain immunological silence to harmless antigenic material in food. In the absence of antigen in the liver, T cells home to and migrate through this organ,36 and a substantial pool of specific memory CD8+ T cells can be found in the liver for weeks postpriming. The specific CD8+ T cells we found in the liver were B220 and Annexin V, and they did not express CD95L. These data confirm our previous report24 that recently primed CD8+ T cells found in the liver are viable and functional, do not show evidence of apoptosis, and do not engage in CD95L-mediated fratricide. These data are difficult to reconcile because the liver is exceptionally efficient in eliminating CD8+ T blasts that enter it.37, 38 As has been correctly stated previously,39 we cannot exclude that a subset of the specific CD8+ T cell population that enters the liver is rapidly cleared by activation-induced cell death. The situation seems to change strikingly when the liver bears (and presents) the antigens, as is evident from liver transplantation experiments40–44 or chronic HBV/HCV infection.45–47 But our data in HBs-tg mice indicate that even in the presence of antigen-presenting hepatocytes, primed CD8+ T cells can be maintained in the liver for extended periods. We do not know if the vaccine-primed memory CD8+ T cells we find in the liver represent a resident population or result from continuous traffic of specific CD8+ T cells through this organ.

Specific priming, clonal expansion, differentiation, tissue-specific homing, and/or effector function delivery of CD8+ T cells are regulated by CD4+ T cells. Codelivery of ODN with the peptide vaccine has bypassed “help” requirements in our system. In addition to priming, CD8+ T cell memory generation and maintenance seems to depend on help.13, 48–50 In our system, the longevity of specific CD8+ T cell memory in spleen and liver and the ability to expand this memory population by specific challenge were CD4+ T cell–independent. This suggests that the vaccine formulation has a determining influence on the memory development of a T cell response at the stage of priming, because the synthetic peptide is expected to be cleared from the animal within days. This help-independent T cell memory generation makes the vaccine formulation an attractive choice for specific immunotherapy of immunodeficient hosts (e.g., HIV-infected patients). We used the HBs-tg mouse model to test if T cell tolerance to HBsAg can be broken by the vaccine formulation we describe. HBsAg-specific CD8+ T cell immunity could be established in transgenic mice by the novel peptide vaccine formulation. In these mice, the liver-specific expression of high levels of HBsAg under the control of the albumin promoter leads to variable but severe liver histopathology, antigenemia, and elevated serum transaminase levels. It is difficult to distinguish vaccine-induced from “spontaneous” immunopathologies in immune mice in this model. These data are thus not indicative of the potential therapeutic efficacy of the vaccine formulation in chronic HBV infection, but rather support the notion of an enhanced immunogenicity of the vaccine formulation (that can overcome nonresponsiveness).

CD25+ CD4+ TR cells have been shown to negatively regulate these T cell responses.10–12, 26, 27 Transient, acute TR cell depletion may facilitate therapeutic vaccination that targets specific CD8+ T cell activation against viruses or tumor-associated antigens.51, 52 Acute depletion of CD25+ T cells enhanced the number of specific, functional CD8+ T cells primed by the vaccine twofold. These TR cells are major histocompatibility complex Class II–restricted.53 It is not known if they recognize a determinant on the 12mer–20mer fusion peptide used for vaccination or if they respond to a self-determinant activated by the vaccine formulation as an endogenous “danger” signal.

In conclusion, our data show that small, synthetic peptide vaccines formulated in a simple and controlled way are potent activators of CD8+ T cells. Basically, the system mimics the well-known immunogenicity of many viral core (or nucleocapsid) particles, but in a drastically simplified way. As stated above, the system can be modified to incorporate larger protein antigens or DNA vaccines. It is thus an interesting experimental tool for the study of vaccine designs that target the induction of cellular immunity.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We greatly appreciate the expert technical assistance of Tanja Güntert, Daniela Schey, und Katrin Ölberger.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References