| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institutes of
*
Medical Microbiology, Immunology and Hygiene and
Virology, Technische Universität München, and
Clinical Cooperation Group Vaccinology, National Research Center for Environment and Health (GSF) and Technische Universität, Munich, Germany; and
The Walter and Eliza Hall Institute of Medical Research, Melbourne, Victoria, Australia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
+ dendritic cell (DC)4 (5, 6). Even though mechanistic details underlying cross-presentation are poorly understood, recent results imply that exogenous Ag gains access to the endogenous MHC class I Ag-presenting pathway at the phagoendosomal compartment of DCs, shown to fuse with the endoplasmic reticulum (ER) (7). During this fusion process, components of the ER-borne class I processing machinery reportedly access the phagoendosome, thereby donating competence for cross-presentation (8, 9, 10). Integration of this knowledge into a rational subunit Ag vaccine design suggests that subunit Ag ought to be targeted to the phagoendosome to achieve robust cross-presentation. Besides the necessity to deliver exogenous Ag into the MHC class I presentation pathway, the DC needs to mature for efficient "cross-priming" of CD8 T cells. TLRs are type I transmembrane proteins expressed by innate immune cells (reviewed in Refs. 11 and 12). They provide a repertoire to DCs for sensing pathogen-derived ligands (13). TLRs are expressed either at the cell membrane (TLR1, 2, 4, 5, and 6) or at the phagoendosome (TLR3, 7, 8, 9) (reviewed in Ref. 12), presumably upon translocation from the ER (14, 15). Specifically, TLR9 recognizes at the phagoendosome immunostimulatory CpG-DNA, a process known to lead to maturation of immature DCs into professional APCs (16). In DCs, the phagoendosome thus appears to function as a crossroad at which immunostimulatory CpG-DNA initiates its adjuvant effects and exogenous Ag becomes cross-presented. It follows that targeting both CpG-DNA (adjuvants) and exogenous Ag into the phagoendosome of DCs might be the key to improve the immunogenicity of exogenous Ag for MHC class I-restricted Ag-specific CD8 T cell responses.
Multiple studies have shown that CpG-DNA admixed to proteinaceous Ag promotes humoral and cellular Th1 responses (reviewed in Refs. 17 and 18). The pioneering work of Raz and colleagues (19, 20, 21, 22) unraveled that covalently conjugating CpG-DNA to Ag creates an even more potent immunogen, effectively driving Ag-specific Ab responses, a Th1 biased cytokine profile from CD4 T cells, and CD8 CTL responses that are largely CD4 T cell-independent (reviewed in Ref. 23). This finding is further supported by recent clinical trials in which Amb a1 linked to CpG-DNA improved symptom relief and immunomodulation of the allergic response in patients with allergic rhinitis (24, 25).
One of the hallmarks of CpG-Ag conjugates is reflected by the ability of CpG-DNA to act as a leader for sampling and presentation of CpG-DNA tagged Ag by DCs (26, 27), a phenomenon that was subsequently confirmed (28). Cellular uptake of CpG-DNA is mediated by a sequence nonspecific DNA receptor that drives an endocytosis process, which translocates CpG-DNA into the phagoendosome (29, 30). Ag-linked CpG-DNA thus acts as ligand for the DNA receptor and mediates robust internalization of Ag-CpG complex (27, 31). In this study, we tested the hypothesis that receptor-mediated endocytosis of CpG-Ag complex is key to trigger protective CD8 T cell immunity by translocating immunostimulatory CpG-DNA plus Ag into the phagoendosome of DCs known to express competence for cross-presentation and TLR9-mediated DC activation (8, 9, 14, 27, 30, 31, 32). To clarify aspects of these matters, we reanalyzed endosomal uptake of CpG-OVA complex by DCs in vitro, as well as subsequent CpG-DNA-driven DC maturation and cytokine production (23, 30, 31). We then compared the immunogenicity of CpG-OVA complex with that of live bacterial and viral vectors genetically engineered to produce OVA. Specifically we analyzed primary and secondary clonal CD8 T cell "burst size," the cell phenotype of clonally expanding CD8 T cells, as well as their protective effector functions. We describe that the immunogenicity of subunit vaccine CpG-OVA complex competes well with that of the live vaccine Listeria monocytogenes producing OVA (LM-OVA) and the viral vector-based OVA vaccines we analyzed. We conclude that endosomal translocation of CpG-DNA Ag complex is key for robust CD8 T cell-based immunoprotection by vaccines build on subunit Ag.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
OVA (chicken egg OVA) was purchased from Sigma-Aldrich. Phosphothioate-stabilized immunostimulatory CpG-DNA was purchased from TIB MolBiol. The immunostimulatory CpG sequence used as oligonucleotide (ODN) is ODN 1668: 5'-TCCATGACGTTCCTGATGCT-3' (16). For inhibition experiments, the ODN pG-1668 5'-TCCATGACGTTCCTGGGGGG-3' (28) was used. As non-CpG-related control ODN AP-1 with the sequence 5'-GCTTGATGACTCAGCCGGAA-3' and the GpC-ODN 1720 5'-TCCATGAGCTTCCTGATGCT-3' were used. CpG-DNA was covalently linked to OVA essentially as described (19, 28, 31) by using as a cross-link sulfo-maleimidobezoyl-N-hydroxy succinimide ester (Pierce). The batch used in this study contained 
2–3 CpG-DNA molecules per OVA molecule, with 10 µg of CpG-OVA complex containing 0.44 nM CpG.
CpG uptake measurement
To examine uptake and blocking of FITC-labeled CpG-OVA complex in vitro, bone marrow-derived DCs grown in GM-CSF conditioned medium (0.2 µg/ml rmGM-CSF; PeproTech) (31). The cells were exposed to FITC-labeled CpG-OVA complex (5 µg/ml), mixed with poly (G) CpG-DNA (50 µM), CpG-OVA FITC conjugates (5 µg/ml), OVA-FITC (5 µg/ml), or with medium (30 min at 37°C), washed twice with ice-cold 3% FCS/PBS, and stained with allophycocyanin-labeled anti-CD11c (clone HL3; BD Pharmingen).
Mice, bacteria, and virus
C57BL/6 mice were obtained from Harlan Winkelmann. Mice were kept under specific pathogen-free conditions and used at 6–12 wk of age. Infection experiments were performed by i.v. injection of LM-OVA (kindly provided by H. Shen, University of Pennsylvania Medical School, Philadelphia, PA). A dose of 2 x 104 i.v.-injected LM-OVA represents the LD50 for C57BL/6 mice. For primary infection and secondary expansion (detailed in Fig. 2) 2 x 103 CFU of bacteria and for lethal challenge 2 x 105 CFU of bacteria were injected. In some experiments LM wild-type strain 10403s was used as specificity control. For infection with viral vectors, 108 PFU of modified vaccinia virus Ankara (MVA)-OVA (MVA genetically engineered to produce OVA upon cellular infection) (33) or 4 x 106 PFU of rHSV-OVA (attenuated HSV genetically engineered to produce OVA; kindly provided by T. Brocker, Ludwig-Maximilians-Universität München, München, Germany) were injected i.v. (34).
|
Survival statistic
Survival curves were drawn by the Kaplan-Meier method, and differences between curves were analyzed using the log-rank test.
DC preparation and transfer
DCs from spleens of naive mice were purified as described (35). Briefly, spleen fragments were digested for 20 min at room temperature with collagenase DNase. All subsequent procedures were at 0–4°C in a Ca2+- and Mg2+-free medium. Light density cells were selected by centrifugation in a Nycodenz medium of 1.077 g/cm3 (Nycomed). Cells not of the DC lineage were then depleted by incubating the cells with previously optimized amounts of anti-CD3 (KT3), anti-Thy1 (T24/31.7), anti-CD19 (1D3), anti-Gr-1 (RB6-8C5), and anti-erythrocyte (TER-119) and then removing the Ab-binding cells with goat anti-rat Ig-coupled magnetic beads (BioMag; Qiagen). The preparation was used directly for immunofluorescent labeling with CD45RA (14.8), CD11c (HL3), CD4 (L3T3, RM4-5), and CD8 (53-6.7) (all BD Pharmingen). Sorting was performed using MoFlo Cell Sorter (DakoCytomation). We used propidium iodide (Molecular Probes) for live-death discrimination. Approximately 2 x 106 DCs were injected intravenously.
Surface staining and cytokine measurement
H2-Kb/SIINFEKL multimer reagents were generated as previously described (36). The 200 µl of blood from the tail vein or the total spleen were taken, and erythrocytes were lysed using ammonium chloride-Tris. Cells (1–6 x 106) were stained for each sample. For live-dead discrimination, cells were incubated with ethidium monazide (Molecular Probes), followed by MHC multimer and surface marker staining for 45 min at 4°C. The following mAbs were used: anti-CD8 (clone 3B5; Caltag Laboratories), anti-CD127 (clone A7R34; BD Biosciences), anti-CD62L (clone MEL-14; BD Pharmingen), and anti-CD44 (clone IM7; BD Pharmingen).
To analyze activation of APCs by CpG-OVA complex, bone marrow-derived DCs generated in GM-CSF conditioned medium (day 5) were incubated with 10 µg/ml CpG-OVA complex for 24 h, and thereafter the cells were washed twice followed by staining with anti-CD11c (clone HL3; BD Pharmingen), anti-MHC class II (clone M5/114.15.2; BD Pharmingen), anti-CD40 (clone 3/23; BD Pharmingen), and anti-CD86 (clone GL1; BD Pharmingen). Data were acquired on a FACSCalibur (BD Biosciences) and analyzed with FlowJo (Tree Star) software.
Supernatant of 24 h CpG-OVA complex incubated bone marrow-derived DC were collected for ELISA. Cytokine release (TNF-, IL-12p40, and IL-6; R&D Systems) was assayed as instructed by the manufacturer.
Confocal microscopy
Bone marrow-derived DCs were plated on eight-well chamber slides (BD Falcon) 1 day before analysis. Cells were incubated with either 4 µg/200 µl of OVA-FITC or ODN 1668-OVA-FITC for 45 min and washed with PBS. Fixation was performed with 1% formaldehyde/PBS and permeabilization in staining buffer (PBS, 0.5% saponin, 2% normal goat serum). Primary Ab (anti-CD107a, lysosomal-associated membrane protein 1 (LAMP-1); BD Pharmingen) and secondary Ab (Alexa 488-labeled anti-FITC and Alexa 546-labeled anti-rat IgG; Molecular Probes) incubation was performed for 1 h and for 30 min, respectively.
Cells were viewed with a Zeiss (Carl Zeiss) LSM 510 confocal microscope equipped with Ar/Kr (458 and 488 nm) and He/Ne (543 nm) lasers. The lens used was a Plan-Neofluar 40 1.3 oil lens. In general the confocal pinhole setting was chosen to provide a <1-µm section of the cells. Images recorded with the confocal microscope were exported as single-image files in the TIFF format with the LSM5 Image Browser (version 3.1.0.99) and processed with Adobe Photoshop CS software. OVA-FITC is shown in green and LAMP-1-positive staining in red (see Fig. 1); the overlay image is also presented.
|
Splenocytes were incubated for 5 h in the presence of the CD8 T cell epitope SIINFEKL (peptide) and for the last 3 h brefeldin A (GolgiPlug; BD Pharmingen) was added. Intracellular cytokine staining for IFN- and TNF- (FITC-labeled; BD Pharmingen) was performed using the Cytofix/Cytoperm kit (BD Pharmingen) according the manufacturer’s recommendations. For in vivo cytotoxicity assay, splenocytes from naive C57BL/6 mice were either loaded or not loaded with SIINFEKL stained with carboxyfluorescein diacetate succinimidyl ester (CFDA-SE; Molecular Probes) at concentrations of 5 µM (SIINFEKL-loaded cells) or 0.5 µM (unloaded cells). Equal numbers of either cells were injected i.v. into primed animals. After 5 and 21 h, blood was collected from the tail vein and analyzed for CFSE-positive cells.
For in vitro cytotoxicity experiments, spleen cell suspensions were cocultured with SIINFEKL-labeled (1 µM) irradiated (15 Gy) syngenic spleen cells plus 5 U/ml rIL-2. At day 7, cytolytic activity was measured by the 51Cr release assay essentially as described (31). Untreated EL4 cells (American Type Culture Collection) served as specificity control. Radioactivity was measured with a PerkinElmer gamma counter. Specific lysis was calculated according to the formula: percentage of specific lysis = [(cpmsample – cpmspontaneous release)/(cpmmaximum release – cpmspontaneous release)] x 100. Spontaneous release ranged from 5 to 15%.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Conjugation of OVA with immunostimulatory CpG-DNA creates potent immunogens (19, 23, 26, 31). In confirmation of prior work (31), Fig. 1a shows that cellular uptake by DCs of CpG-OVA complex is 10- to 50-fold enhanced compared with that of OVA, and can be competed for by third party ODN. Thus a yet ill-defined DNA receptor appears to drive endocytosis (26, 27, 28) shown previously to be unrelated to TLR9 (31). Upon CpG-OVA complex internalization, immature DCs cross-present the OVA-derived CD8 T cell epitope SIINFEKL and become activated in a TLR9-dependent fashion to up-regulate MHC class II, CD86, and CD40 (Fig. 1b). In parallel, the activated DCs produce IL-12p40, TNF-, and IL-6 (Fig. 1c). Fig. 1d visualizes that upon receptor-mediated endocytosis the CpG-OVA complex becomes translocated to LAMP-1-positive endosomal-lysosomal compartments, previously shown to express TLR9 (14, 30) and competence for cross-presentation (8, 9). In contrast, cellular uptake of OVA-FITC is poor and only minute amounts of OVA-FITC were found in vesicular structures some of which colocalized with LAMP-1-positive compartments (Fig. 1d and data not shown). We conclude that similar to CpG-DNA (14, 29, 30), CpG-OVA complex effectively translocates into vesicular LAMP-1-positive compartments, and thus we hypothesize that translocation of CpG-ODN complex to the endosome of immature DCs might enhance the ability of DCs for cross-presentation and cross-priming.
CpG-OVA conjugates induce clonal expansion and contraction of Ag-specific CD8 T cells
We analyzed the ability of CpG-OVA complex (subunit vaccine) to drive MHC class I-restricted CD8 T cell responses and compared it with the ability of live vaccine. As examples for a live vaccine we chose LM-OVA (36), and as example for viral vectors we studied attenuated MVA with an integrated OVA expression cassette (37) as well as a replication attenuated HSV with an integrated OVA expression cassette (34). As readout we used H-2Kb/SIINFEKL tetramers to visualize and to quantitate in peripheral blood the frequency of emanating Ag-specific CD8 T cells. Intravenous infection of C57BL/6 mice with LM-OVA (sublethal dose, 0.1 x LD50) caused expansion and contraction of the SIINFEKL-specific CD8 T cells typical for a living bacterium (Fig. 2a) and protected against a second otherwise lethal infection (see Fig. 5) (38, 39). Thus, low dose infection with LM-OVA can be seen as a gold standard for OVA vaccine. In comparison, a single s.c. challenge of C57BL/6 mice with the subunit vaccine CpG-OVA complex triggered primary as well as secondary clonal expansion and contraction of SIINFEKL-specific CD8 T cells similar in kinetic and magnitude as that observed with live vaccine LM-OVA (Fig. 2a). However, a single challenge with CpG-DNA (0.44 nM) admixed to OVA (10 µg) failed to trigger significant clonal expansion of CD8 T cells (Fig. 2a). As expected, nonstimulatory GpC-ODN linked to OVA fell short of triggering clonal CD8 T cell expansion (data not shown).
|
DCs display the ability to cross-present exogenous Ag (5). We tested whether FACS sorted splenic C57BL/6 CD8 DCs loaded ex vivo with CpG-OVA complex would trigger clonal expansion of SIINFEKL-specific CD8 T cells upon i.v. transfer into naive mice. As shown in Fig. 2b, this protocol yielded within 7 days in the spleen of the challenged mice 2% of Ag-specific CD8 T cells, whereas CD8–CD4– DC failed to induce significant clonal CD8 T cell expansion. Because T cells in blood (Fig. 1a) are migrating and thus might not mirror T cell frequencies in secondary lymphoid organs or in spleen, we also analyzed splenocytes of mice challenged with CpG-OVA complex, live LM-OVA, MVA-OVA, or rHSV-OVA (viral vectors). Depending on the experiment, 2–5% of splenic CD8 T cells were SIINFEKL-specific 7 days after a single s.c. challenge with subunit vaccine CpG-OVA complex (Fig. 2c). These splenic frequencies were slightly lower than those triggered by a sublethal infection with LM-OVA. Of note, the immunogenicity of CpG-OVA complex exceeded that of the viral OVA vectors analyzed (Fig. 2c). For the additional experiments we decided to compare CpG-OVA complex with the best live vaccine, i.e., low dose infection with LM-OVA.
CpG-OVA complex induces CD8 T cells with memory phenotype
MHC class I tetramers allow visualization and thus quantitation of Ag-specific CD8 T cells, yet provide no information on their functional status. In contrast to naive T cells, Ag-experienced T cells, or memory T cells differentially express certain cell surface markers including CD44, CD62L, and CD127 (39, 40, 41). Therefore, we phenotyped CpG-OVA complex (subunit vaccine) as well as LM-OVA (live vaccine) that triggered SIINFEKL-specific CD8 T cells (primary responses, day 7) to determine cell surface expression of these markers. As shown in Fig. 3a, the Ag-specific CD8 T cells that clonally expanded in LM-OVA-infected or in CpG-OVA complex-challenged mice were CD44-positive, which is a marker implying Ag experience (Fig. 3a). Most of the cells were CD62Llow, implying an activated phenotype. Interestingly, the subunit vaccine CpG-OVA complex induced almost equal percentages of CD62Lhigh Ag-specific CD8 T cells as did infection with LM-OVA. CD62Lhigh Ag-specific CD8 T cells are viewed as central memory T cells that gives rise to a long lasting proliferating pool of memory T cells (39, 40, 41). As for CD62Lhigh, almost equal percentages of Ag-specific CD8 T cells were CD127high (Fig. 3a), a memory marker characterizing long-lasting memory cells with immediate effector function known to be expressed already in a subset of clonally expanding primary CD8 T cells (39, 40, 41). We thus concluded that the difference in the percentage of SIINFEKL-specific CD8 T cells observed upon challenge with CpG-OVA complex and of the LM-OVA refers to a relative lack of short-living CD8 effector cells, whereas the memory T cell subsets appear to evolve equally well. This conclusion was further supported by phenotyping Ag-specific CD8 T cells 35 days after primary challenge with CpG-OVA complex or LM-OVA, which is a point in time where short-living effector (CD127low cells) cells have nearly completely disappeared. The percentage of remaining SIINFEKL-specific memory cells appeared at this point in time to be equal between LM-OVA-infected and CpG-OVA complex vaccinated animals (Fig. 3b). Again, no expansion of SIINFEKL-specific cells was seen in OVA-treated mice.
|
MHC class I tetramers or surface markers associated with effector or memory T cells provide only an indirect picture of the respective effector functions. To directly assess effector functions we performed intracellular cytokine staining and cytotoxicity assays at day 7. After short in vitro restimulation of splenocytes of CpG-OVA complex primed mice with the SIINFEKL peptide, 2% of splenic CD8 T cells stained positive for IFN-
or TNF- (Fig. 4a). Infection with the live vaccine LM-OVA led to 2-fold higher frequencies, a finding in line with the higher frequencies of Ag-specific effector cells (CD62Llow and CD127low; see Fig. 3). In vivo CTL assays revealed strong Ag-specific lytic activity in mice 7 days after challenge with CpG-OVA complex, with the lytic activity induced upon infection with LM-OVA being slightly higher (Fig. 4b).
|
CpG-OVA complex induces protective CD8 T cells
A stringent test for the efficacy of vaccination represents challenge with lethal doses of live bacterium expressing the Ag of interest. We compared CD8 T cell-mediated protection due to CpG-OVA complex subunit vaccination with that of LM-OVA or MVA-OVA by lethal challenge with LM-OVA or wild-type LM (specificity control). To this end, groups of C57BL/6 mice were first vaccinated with CpG-OVA complex or live LM-OVA or MVA-OVA and 35 days later challenged with 10 x LD50 of LM-OVA or wild-type LM. Thereafter survival was monitored. Kaplan-Mayer curves depicted in Fig. 5 demonstrate survival of 85% of CpG-OVA complex vaccinated mice (group of 15 mice) and 60% survival of MVA-OVA vaccinated mice, whereas all mice vaccinated with LM-OVA survived. All animals vaccinated with CpG-OVA complex and challenged with wild-type LM (specificity control) succumbed to infection, whereas all primarily LM-OVA-infected mice survived presumably due to immunity to other Listeria epitopes.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
By definition vaccines based on subunit Ag depend on cross-presentation and on stimuli triggering DC maturation required for cross-priming. We assume that poor access of soluble Ag to the MHC class I cross-presentation pathway and inadequate adjuvants explain the poor immunogenicity experienced with conventional subunit Ag-based vaccines. Uptake of soluble Ag by DCs appears to be an inefficient (46) process augmented by receptor-mediated endocytosis or by phagocytosis (reviewed in Ref. 47). As previously noted (23, 26, 27, 28, 31), the cellular uptake of immunostimulatory CpG-DNA is mediated by a yet ill-defined sequence nonspecific DNA receptor, which translocates CpG-DNA into an endosomal compartment (29, 30, 31). Notably, several groups recently reported in DCs the existence of a process, termed ER-mediated phagocytosis, in which during endosome biogenesis the endosome membranes fuse with ER membranes (7). This process provides access of the MHC class I processing machinery to the endosomal compartment (7, 8, 9). As a consequence, endosomes acquire competence for cross-presentation. In this process, a tyrosine-based targeting signal within its cytoplasmic domain controls routing of MHC class I molecules from ER to the endosome (32). Furthermore, transfer of sufficient Ag doses to the endosome appears key for robust cross-presentation because Ags that are degraded rapidly after synthesis are only weakly cross-presented, as in for example, CD8 epitopes derived from leader sequences (10).
Besides efficient endosomal translocation of Ag to be cross-presented, maturation of DC is indispensable for cross-priming. Clearly, the use of strong adjuvants such as the TLR9 ligand CpG-DNA allows priming of T cells when admixed with Ag (reviewed in Refs. 17 , 18). However, by linking CpG-DNA to Ag (19) both the adjuvant CpG-DNA and Ag translocate together to the endosomal-lysosomal compartment of DCs (Fig. 1). Members of the TLR9 subfamily (TLR7–TLR9) are expressed at the endosome (reviewed in Ref. 12), presumably upon translocation from ER during biogenesis of the phagoendosome (14, 15). It follows that translocation to the phagoendosome allows CpG-OVA complex to enter a crossroad organelle at which competence for cross-presentation meets competence for cross-priming.
This information in mind, we report our results of a systematic comparison of the immunogenicity of live bacterial and viral vectors genetically engineered to produce OVA with that of CpG-OVA complex. First we show that in vitro CpG-OVA complex efficiently translocates to LAMP-1-positive endosomal-lysosomal compartments of DCs. More importantly, single s.c. challenges with CpG-OVA complex triggered in vivo primary and secondary clonal expansion and contraction of Ag-specific CD8 T cells comparable in kinetics and magnitude to that of the live vector LM-OVA; the viral vectors tested were less immunogenic. Although long-lasting CD127high memory CD8 T cells were induced equally effective by CpG-OVA complex (subunit) vaccine and by LM-OVA (live vaccine), the latter were more effective in triggering primary CD8 T effector cells.
LM-OVA vaccination protected 100% of the animals against an otherwise lethal challenge with LM-OVA, whereas CpG-OVA complex vaccination protected 85% of vaccinated animals, and MVA-OVA primed mice appeared to be less protected. This indicates that the overall burst size of Ag-specific CD8 T cells may not be considered as valid marker to predict subsequent protection because infectious replicating bacteria induce more short-living effector cells. Evaluation of cell surface markers expressed by clonally expanding CD8 T cells revealed equally effective generation of long-living memory CD127high CD8 T cells (39, 40, 41). It is thus tempting to attribute resistance toward lethal LM-OVA infections of mice challenged with CpG-OVA complex 35 days earlier to the robust generation of long-living CD127high memory CD8 T cells. Of note, the dose of our secondary LM-OVA infection was drastically high with 10 x LD50. We explain the more efficient protection of mice vaccinated with low dose infection LM-OVA as compared with CpG-OVA complex vaccinated animals by specific responses against other epitopes of LM-OVA, as these animals also survive a secondary infection with LM wild type.
Several groups demonstrated robust immunostimulatory effects of CpG-Ag conjugates (CpG-Ag complex) in different model systems (19, 21, 28). For example, application of CpG-Ag complex diminished eosinophilia and shifted the Th1/Th2 balance toward Th1 both in allergic mice and in patients. Furthermore, cell-mediated responses were enhanced Ag specifically (reviewed in Ref. 23). In this study we extend these studies to demonstrate protection against replicating live LM bacteria. CD8+ DCs cross-presented efficiently and CpG-Ag complex triggered generation of CD8 memory subsets similar in magnitude to live LM-OVA. Future studies have to address whether, in addition to CD8 T cells, other lymphocyte subsets like CD4 T cells are activated by CpG-OVA complex and might contribute to protection toward live LM-OVA.
In conclusion we provide compelling evidence that targeting CpG-OVA complex to endosomes of murine DCs donates immunogenicity for protective CD8 T cell responses comparable in size and kinetics to that triggered by the live vector LM-OVA. These findings imply that effective levels of protective CD8 T cell memory do not depend solely on persistent infection and thus persistent T cell activation to imitate the Mackaness (42) paradigm of infection-immunity. Efficient translocation of CpG-DNA (adjuvants) together with proteinaceous Ag appears operationally to be sufficient. To test whether this vaccination protocol can be translated to humans, we analyze whether CpG-OVA complex also activates in human plasmacytoid DCs competence for cross-presentation and cross-priming.
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 This work was supported by Grants from Sonderforschungsbereich (SFB)456/TeilProjekt (TP)-B3 Hochschul-und Wissenschaftsprogramm (HWP)-1, SFB576/TP-A8, and the Deutsche Forschungsgemeinschaft (DFG) Gerhard Hess Program. 
2 H.W. and K.M.H. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Hermann Wagner, Institute of Medical Microbiology, Immunology and Hygiene, Trogerstrasse 9, 81675 Munich, Germany. E-mail address: h.wagner{at}lrz.tu-muenchen.de
4 Abbreviations used in this paper: DC, dendritic cell; LM-OVA, Listeria monocytogenes producing OVA; ER, endoplasmic reticulum; ODN, oligonucleotide; MVA, modified vaccinia virus Ankara; LAMP-1, lysosomal-associated membrane protein 1.
Received for publication October 27, 2004. Accepted for publication January 10, 2005.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
+ dendritic cells are generally involved in priming CTL immunity to viruses. J. Immunol. 172:1996.This article has been cited by other articles:
|
|
 |
|
  C. Berchtold, K. Panthel, S. Jellbauer, B. Kohn, E. Roider, M. Partilla, J. Heesemann, S. Endres, C. Bourquin, and H. Russmann Superior Protective Immunity against Murine Listeriosis by Combined Vaccination with CpG DNA and Recombinant Salmonella enterica Serovar Typhimurium Infect. Immun., December 1, 2009; 77(12): 5501 - 5508. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Kaiser-Schulz, A. Heit, L. Quintanilla-Martinez, F. Hammerschmidt, S. Hess, L. Jennen, H. Rezaei, H. Wagner, and H. M. Schatzl Polylactide-Coglycolide Microspheres CoEncapsulating Recombinant Tandem Prion Protein with CpG-Oligonucleotide Break Self-Tolerance to Prion Protein in Wild-Type Mice and Induce CD4 and CD8 T Cell Responses J. Immunol., September 1, 2007; 179(5): 2797 - 2807. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Khan, M. S. Bijker, J. J. Weterings, H. J. Tanke, G. J. Adema, T. van Hall, J. W. Drijfhout, C. J. M. Melief, H. S. Overkleeft, G. A. van der Marel, et al. Distinct Uptake Mechanisms but Similar Intracellular Processing of Two Different Toll-like Receptor Ligand-Peptide Conjugates in Dendritic Cells J. Biol. Chem., July 20, 2007; 282(29): 21145 - 21159. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hamm, A. Heit, M. Koffler, K. M. Huster, S. Akira, D. H. Busch, H. Wagner, and S. Bauer Immunostimulatory RNA is a potent inducer of antigen-specific cytotoxic and humoral immune response in vivo Int. Immunol., March 1, 2007; 19(3): 297 - 304. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Adamsson, M. Lindblad, A. Lundqvist, D. Kelly, J. Holmgren, and A. M. Harandi Novel immunostimulatory agent based on CpG oligodeoxynucleotide linked to the nontoxic B subunit of cholera toxin. J. Immunol., April 15, 2006; 176(8): 4902 - 4913. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), with LM-OVA (
), and with CpG (0.44 nM) (
) and OVA (10 µg;
). b, Dot plots show CD8+ splenocytes 7 days after i.v. injection of 2 x 106 CpG-OVA complex loaded DCs stained with CD62L (x-axis) and tetramer (y-axis). Subgroup of DC as indicated atop the graph. c, Dot plots show splenocytes stained with CD8 (x-axis) and H2-Kb-SIINFEKL tetramer (y-axis). Numbers represent percentage of SIINFEKL-specific cells per CD8 T cells. Treatment of mice 7 days before staining is indicated atop each dot panel. A bar chart (far right) with average frequencies (resulting from at least three mice per group) + SD is shown. Treatment information is listed (x-axis).
), or LM-OVA (
). Bars indicate percentage of IFN-