The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Flechtner, J. B.
Right arrow Articles by Andjelic, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Flechtner, J. B.
Right arrow Articles by Andjelic, S.
The Journal of Immunology, 2006, 177: 1017-1027.
Copyright © 2006 by The American Association of Immunologists

High-Affinity Interactions between Peptides and Heat Shock Protein 70 Augment CD8+ T Lymphocyte Immune Responses

Jessica B. Flechtner, Kenya Prince Cohane, Sunil Mehta1, Paul Slusarewicz2, Alexis Kays Leonard3, Brian H. Barber4, Daniel L. Levey5 and Sofija Andjelic5

Antigenics Inc., Lexington, MA 02421


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Exogenously delivered antigenic peptides complexed to heat shock proteins (HSPs) are able to enter the endogenous Ag-processing pathway and prime CD8+ CTL. It was determined previously that a hybrid peptide containing a MHC class I-binding epitope and HSP70-binding sequence Javelin (J0) in complex with HSP70 could induce cytotoxic T cell responses in vivo that were more robust than those induced by the minimal epitope complexed with HSP70. The present study introduces a novel, higher-affinity HSP70-binding sequence (J1) that significantly enhances binding of various antigenic peptides to HSP70. A competition binding assay revealed a dissociation constant that was 15-fold lower for the H2-Kb OVA epitope SIINFEKL-J1 compared with SIINFEKL-J0, indicating a substantially higher affinity for HSP70. Further, modifying the orientation of the hybrid epitope and introducing a cleavable linker sequence between the Javelin and the epitope results in even greater immunogenicity, presumably by greater efficiency of epitope processing. The enhanced immunogenicity associated with Javelin J1 and the cleavable linker is consistently observed with multiple mouse and human epitopes. Thus, by creating a series of epitopes with uniform, high-affinity binding to HSP70, successful multiple epitope immunizations are possible, with equal delivery of each antigenic epitope to the immune system via HSP70. These modified epitopes have the potential for creating successful multivalent vaccines for immunotherapy of both infectious disease and cancer.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Heat shock proteins (HSPs),6 the most abundant intracellular proteins found in eukaryotic cells, are a family of evolutionarily conserved molecules with multiple functions. HSP70 proteins were first described as molecular chaperones that increased in concentration as a result of thermal stress and that facilitated the folding, assembly, or disassembly of other proteins. Later, more ubiquitous roles that occur under both stress and nonstress conditions were identified, including HSP-mediated membrane transport (1, 2).

More recently, HSPs have been shown to have a role in priming immune responses. They exert their effects by maturing dendritic cells (3, 4, 5, 6), inducing type-1 cytokines and CC chemokines (6, 7, 8), and by causing production of NO by APC (9). In addition to these Ag-independent interactions, HSP70 and other HSP family members have been shown to bind antigenic peptides endogenously. Such HSP-peptide complexes purified to homogeneity induce specific immunity to tumor, viral, and bacterial targets (10, 11, 12, 13, 14, 15, 16). In fact, recent data elegantly show that protein fragments chaperoned by HSPs to APC are a necessary source of Ag for priming CD8+ T cell responses (17). HSPs can also be reconstituted in vitro with defined Ags and elicit specific immunity in a large variety of systems (18, 19, 20). Stimulation of the adaptive immune response occurs by receptor-mediated uptake of HSP-peptide complexes (21) that then deliver the MHC class I epitopes via both cytosolic and endocytic routes of Ag processing for re-presentation on the cell surface. Several receptors have been identified that may bind HSP (22, 23, 24, 25, 26, 27, 28, 29, 30, 31), including CD40, CD91, TLR2, TLR4, CD14, Lox-1, scavenger receptor class A, and scavenger receptor expressed by endothelial cells, although only CD91 has been independently demonstrated to be involved in representation of HSP-chaperoned peptides (32).

One potential roadblock to successful immunization with in vitro-reconstituted HSP-peptide complexes is that only a small subset of antigenic peptides may bind HSP with high affinity (33). The ability of the HSP to chaperone peptides for stimulation of immune responses requires that the epitopes remain noncovalently complexed long enough to be delivered to APCs; therefore, some epitopes may not be of high enough affinity to be immunogenic in the context of an HSP-mediated vaccine. Moreover, because the peptide/HSP interaction is noncovalent multiple epitope vaccines may be hindered by high-affinity peptides competing with lower affinity peptides for HSP binding. To overcome these potential hindrances, hybrid peptides were designed that contain defined T cell epitopes colinearly synthesized with a short hydrophobic binding sequence, Javelin (J0), predicted to have a high affinity for HSP70. With the addition of a high-affinity Javelin sequence, epitopes with intrinsically low HSP70 affinity can now form immunizing HSP70:Javelin-hybrid peptide complexes and thus can more effectively prime CTL responses, compared with the unmodified epitope (34). The increased potency of immunization with HSP70:Javelin-hybrid peptide complexes can be at least partially attributed to the enhanced binding affinity for HSP70 allowing for a longer half-life of peptide-HSP70 interaction; in fact, data suggest that one of the putative receptors for HSP70, CD40, will only bind HSP70 strongly in the presence of the peptide substrate (31), which is facilitated by the higher affinity interaction between the peptide and HSP.

Evaluation of the binding kinetics of the original Javelin-hybrid epitope constructs revealed that while J0 substantially enhanced the affinity between the hybrid peptide and HSP70, there was still a high degree of variability in the binding affinities of Javelin-hybrid peptide constructs containing various antigenic epitopes. As a result, we sought to improve the hybrid peptide design to make the Javelin a more universal affinity-enhancing sequence. In this study, we report that by creating hybrid epitopes with a higher HSP70-binding affinity Javelin (J1), modifying the linker sequence to enhance Ag processing, and moving the Javelin-linker sequence to the N-terminal side of the epitope, we were able to dramatically improve the magnitude of immune responses to several mouse and human antigenic epitopes in the absence of conventional adjuvant. Moreover, the uniform, high-affinity interaction allows for equal delivery of multiple epitopes in a single formulation with the potential for highly successful therapeutic and prophylactic vaccines for treatment and prevention of infectious diseases and cancer.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice and cell lines

Six- to 12-wk-old female C57BL/6 mice were obtained from the National Cancer Institute Animal Resource Center (Frederick, MD) or The Jackson Laboratory. TAP–/– mice were obtained from The Jackson Laboratory. HHDII mice (35) were obtained from F. Lemonnier at the Institute Pasteur and bred at Charles River Laboratories. Mice were housed at New York Medical College or at Antigenics, and cared for following the guidelines of the Institutional Animal Care and Use Committee. The mouse thymoma cell line EL4 and its E.G7 derivative (EL-4 transfected with cDNA encoding OVA) were obtained from American Type Culture Collection and cultured according to the supplier recommendations. The B3Z T-T hybridoma (specific for the peptide SIINFEKL presented in the context of H2-Kb) was previously described (34).

Peptides and proteins

Peptides were purchased from New England Peptide or CS Bio and were >96% pure. Table I shows the panel of peptides that constructed for these experiments.


View this table:
[in this window]
[in a new window]
  		Table I. Peptides used in this study

 
BSA fraction V (BSA) was purchased from Invitrogen Life Technologies. Phosphorylase B was purchased from Sigma-Aldrich and reconstituted in 50 mM HEPES with 150 mM NaCl, then filtered through a 45-µm filter. Recombinant human cytosolic HSP70 was prepared as described previously (34).

Binding assay for hybrid peptide and HSP70

Eighty-microliter binding reactions were set up that spanned a dilution series of the competitor peptide of interest and controls. Binding reactions contained fluoresceinated peptide at a constant concentration, a varying range of unlabeled competitor peptide, and a constant amount of HSP70 in PBS plus 1 mM ADP (Crescent Chemical). The fluorescent peptide used in the binding assay was fluorescently labeled P2-L1-J0: ALFDIESKVGSGHWDFAWPW. Fluorescein was covalently attached to the N terminus, strategically located on the Ag sequence and away from the HSP70-binding sequence. Reactions were incubated for 1 h at 25°C. A total of 25 µl of each reaction mix was then centrifuged through Microspin G-50 columns (Amersham Biosciences). The column is designed to retain free peptide while allowing protein and presumed protein-peptide complexes to pass through. For every peptide tested, a control containing peptide in the absence of HSP70 was used to ensure that all peptides were quantitatively retained in the G-50 column. Upon centrifugation, material passing through the columns was examined by fluorometry and by Bradford assay and no fluorescence of or presence of protein was observed when HSP70 was not present in the binding solution, confirming that free peptides were fully retained on the column.

Material passing through the columns was diluted in 8 M urea and incubated at 95°C for 5 min to ensure dissociation of complexes. Samples were then cooled on ice and analyzed by Bradford assay and fluorometry. Fluorescence in 8 M urea was different from in buffer alone, so to correct for this, a standard curve of P2-L1-J0 was made in 8 M urea and the fluorescence from the binding reactions was measured against this standard curve. Data were analyzed by determining the Hill coefficient: the IC50 of the competitor ligand was determined by graphing log [Y/(1 – Y)] on the y-axis vs log [S] on the x-axis, where Y is the known ligand bound and S is the concentration of the competitor ligand. The IC50 was then used to extrapolate the Ki of the competitor ligand: Ki = IC50/[1 + (Y/Kd)].

The dissociation constant of P2-L1-J0 with HSP70 was determined by using a similar binding assay to that described for the competitor binding assay, except a range of P2-L1-J0 concentrations were incubated with excess HSP70 in the absence of competitor peptide. The Kd was determined by plotting the fraction of bound HSP70 over the concentration of free P2-L1-J0 vs the fraction of bound HSP70. The Kd is the negative inverse of the slope of the resulting line.

HSP70/Javelin-hybrid peptide complex preparation

For preparation of complexes of HSP70 coupled with hybrid peptides, the indicated amounts of HSP70 and peptide were mixed in PBS (pH 7.4; Zymed Laboratories) and incubated for 1 h at 25°C. After the 1-h incubation, 0.1 mM yeast-derived ADP (Crescent Chemical) was added, and the complexes incubated at 25°C for an additional 30 min. All control samples, including nonchaperone protein samples, were treated in the same manner as the complex samples. After the final incubation, complexes were immediately transferred to ice until used.

In vitro cross-presentation assay

Mice were injected i.p. with 1 cc 3% brewer thioglycolate yeast (Sigma-Aldrich). Five days later, mice were euthanized and peritoneal exudate cells were harvested by peritoneal lavage with ice-cold PBS (Invitrogen Life Technologies). Cells were plated at 2 x 106/ml in flat-bottom 96-well plates in serum-free medium (AIM-V; Invitrogen Life Technologies). Plates were incubated at 37°C, 5% CO2 for 1 h, and then nonadherent cells were washed off with 37°C medium. Complexes or control samples were added to triplicate wells, and then the B3Z T-T hybridoma was added at a final concentration of 5 x 105 cells/ml in 200 µl of AIM-V. Plates were incubated overnight at 37°C in 5% CO2, then cell-free supernatants were harvested. Supernatants were stored at –80°C until analyzed for IL-2 levels using the Opt-EIA IL-2 ELISA kit (BD Pharmingen), according to manufacturer’s instructions with the exception that all indicated volumes were halved.

Immunization, ELISPOT, and 51Cr-release assays

For mouse immunization, 50 µl of complex or control solutions was injected s.c. into the base of the tail. Seven days later, spleens were harvested and either put directly into the ex vivo ELISPOT assay (details below) or restimulated at 1–2 x 107 responder cells/flask in the presence of 1–2 x 107 peptide-pulsed, irradiated (3000 rad) normal syngeneic spleen cells in RPMI 1640 containing 10% FCS (cRPMI-10%; HyClone), supplemented with 2 mM L-glutamine, 100 µm of penicillin-100 µg of streptomycin, 0.1 mM MEM nonessential amino acids, 1 mM sodium pyruvate, and 50 µM 2-ME (Invitrogen Life Technologies). Peptide pulsing was performed by incubation of spleen cells with 10 µg/ml peptide for 30 min at room temperature. After 5 days of incubation, cytotoxic activity of restimulated cells was measured in a standard 4-h 51Cr-release assay. EL4 target cells were labeled with 100 µCi of sodium [51Cr]chromate and incubated with or without 1 µg/ml peptide for 1 h then extensively washed and used as target cells. Specific lysis was determined using the following formula: percent-specific release = 100 x (release by effector cells – spontaneous release)/(maximal release – spontaneous release). Spontaneous release was <30% maximum release in all experiments.

For the IFN-{gamma} ELISPOT assay, polyvinylidene difluoride membrane microtiter plates (Millipore) were coated with 10 µg/ml capture Ab (Mabtech) and incubated overnight at 4°C. The following day, wells were washed and blocked with cRPMI-10%. Spleens were pooled within groups and CD8+ T cells were enriched via the MidiMACS cell separation system (Miltenyi Biotec) following the manufacturer’s instructions. Briefly, RBC were lysed with ACK buffer, and the remaining cells were incubated with anti-CD8 microbeads for 20 min at 4°C. Cells were washed once then applied over the MidiMACS column attached to a magnet. Columns were washed four times, and the flow-through was discarded. The CD8+ T cells were plunged off the column with 6 ml of buffer, washed, and resuspended in complete medium. The enrichment success was routinely between 92 and 98% (data not shown). Between 2.5 and 4 x 105 cells were plated per well of the ELISPOT plate, as indicated in the results. Relevant or irrelevant peptide was added for a final concentration of 10 µg/ml (negative and positive control wells contained medium or 5 µg/ml Con A, respectively). In some experiments, peptide-pulsed naive splenocytes were used as APC at a concentration of 5 x 105 cells/well. Plates were wrapped in foil and incubated for 18 h in a 37°C, humidified chamber with 5% CO2. For spot development, wells were extensively washed, then incubated with biotinylated detection Ab (Mabtech) for 2 h at 37°C. The wells were then washed and Vectastain ABC peroxidase (Vector Laboratories) was added and plates incubated at room temperature °C for 1 h. Spots were visualized after incubation with AEC (Sigma-Aldrich) for 4 min, washed with tap water, then dried. Analysis was performed on the CTL Immunospot Reader (Cellular Technology) or by Zellnet Consulting.

Tumor challenge experiments

A total of 7.5 x 105 E.G7 cells was injected s.c. into the right flank of mice that had been immunized 14 and 7 days previously s.c. at the base of the tail. Tumor growth was monitored every 3–4 days, and the results were expressed as the mean volume (mm3) calculated from the longest and its perpendicular diameter of the tumor.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
The Javelin sequence confers the ability of antigenic epitopes to form a stable complex with HSP70

We propose that the immune response to peptides with high affinity for HSP70 will be greater in magnitude than the response to low-affinity peptides because higher affinity peptides are more likely to remain in complex with HSP70 long enough to be chaperoned into APCs. Thus, any epitope modified to contain a Javelin sequence should bind HSP70 with uniformly high affinity and be more immunogenic when injected in vivo compared with the unmodified epitope. To examine this likelihood, several hybrid peptide constructs containing antigenic epitopes colinearly synthesized with a flexible linker (GSG) and the Javelin J0 sequence (HWDFAWPW; Ref. 33) were created. These peptide constructs were first analyzed for their HSP70-binding kinetics in the presence of ADP, as described in Materials and Methods. All Javelin-hybrid peptides displayed higher affinities for HSP70 than their unmodified epitope counterpart; however, the relative affinities between different hybrid peptides were quite variable (Table II, "No Javelin" vs "+ J0"). Moreover, while the specific immunogenicity of the model hybrid Ag OVA-L1-J0 complexed with HSP70 was consistently greater than complexes of HSP70 and the minimal epitope OVA (OVA-L1-J0 and OVA tested at molar equivalents), studies demonstrated considerable variability in the magnitude of the responses to the hybrid Ag (data not shown).


View this table:
[in this window]
[in a new window]
  		Table II. Dissociation constants of peptide constructs

 
The variability in binding and immunogenicity that was characteristic of the hybrid peptides containing J0 led to the hypothesis that even though these peptides individually bind HSP with higher affinity than the unmodified epitopes, they may still not bind with a high enough affinity for uniformity of responses in vivo, and they may not all induce robust immune responses when immunized in an HSP-based polyvalent vaccine. The unevenness in the binding kinetics could result in one epitope being delivered with higher frequency to APCs than another, due to the longer interaction with HSP70. As a result, a new Javelin sequence was created that binds HSP70 with a higher affinity than the J0 sequence, and when synthesized in a string with antigenic epitopes, was predicted to standardize the affinity between the epitopes and HSP70. When the new Javelin construct, J1 (NLLRLTGW; Refs. 36, 37, 38, 39, 40), was analyzed in the kinetic assay, the data revealed a 131-fold greater affinity for HSP70 than J0 (Table II). To determine whether the new Javelin facilitated uniformly higher affinity interactions between HSP70 and epitope, hybrid peptides were synthesized that contained defined epitopes in sequence with the flexible linker -GSG- (L1) and J1, then tested in the binding assay. A Hill plot showing representative data from the mouse MHC class I Kb binding epitope, SINFEKL, from hen egg OVA, is shown in Fig. 1. The dissociation constant for OVA-L1-J0 was extrapolated to be 26 µM whereas the Kd for the OVA-L1-J1 construct was 0.8 µM, a 32-fold difference in affinity. Both affinities were substantially greater than the 236 µM dissociation constant for OVA. The additional results for several human HLA-A*0201-binding epitopes are also shown in Table II. The increase in the affinities of each of these hybrid-J1 peptides over their unmodified counterpart ranged from 37-fold for the CMV epitope to 5100-fold for the gp100 epitope. In fact, all of the sequences containing J1 exhibited low micromolar/high nanomolar dissociation constants for HSP70 signifying a substantial improvement in binding affinity. These data indicate that both murine and human epitopes can be modified with the -L1-J1 sequence and bind HSP70 with higher affinity than the minimal epitope. Synthesizing epitopes with the addition of the J1 sequence creates Javelin-hybrid peptide constructs that more uniformly bind HSP70 and as a result possibly provide stronger priming of immune responses, especially for the epitopes that are naturally poor HSP70 binders like gp100 (Tables I and II).


Figure 1
View larger version (13K):
[in this window]
[in a new window]
  		FIGURE 1. Hill plot for calculation of dissociation constants of OVA, OVA-L1-J0, and OVA-L1-J1. Various concentrations of SIINFKEL (OVA), OVA-L1-J0, or OVA-L1-J1 hybrid peptides were titrated into binding reactions containing constant amounts of both HSP70 and a labeled reporter peptide of known affinity for HSP70. The abilities of these peptides to compete out the binding of the reporter were analyzed using a Hill plot and the IC50 of each determined as the point where the plot intersected the y-axis. The Kd of each peptide was then calculated from its experimentally determined IC50. [S], Concentration of hybrid peptide; Y, fraction of labeled reporter peptide bound.

 
High-affinity HSP70:Javelin-hybrid peptide complexes are better at cross-presentation and induce stronger immune responses than unmodified peptide

The OVA-Javelin hybrid peptides containing the J0 or J1 sequence were tested in an in vitro cross-presentation assay to compare their processing and presentation by APCs. As shown in Fig. 2A, at the doses used, each of the Javelin-hybrid peptides alone, without exogenously added HSP70 can be processed and presented by APCs at very low levels; however, the amount of IL-2 detected was not significantly different from the nonspecific stimulation of the B3Z T-T hybridoma by APCs treated with HSP70 without peptide. When delivered in complex with HSP70, both Javelin-hybrid peptide constructs could be processed, and the antigenic epitope presented by activated murine macrophages. On average, APCs pulsed with 40 nM (85 ng/ml) OVA-L1-J1 in complex with 400 nM (28 µg/ml) HSP70 induced approximately twice as much IL-2 from the B3Z cell line than the molar equivalent of OVA-L1-J0 in complex with HSP70 (p = 0.024; Fig. 2A). Analysis of levels of intracellular beta-galactosidase showed the same pattern as IL-2 production (data not shown). These data indicate that the higher affinity J1 sequence improves the cross-presentation of Javelin-hybrid peptide, presumably by shifting the equilibrium of the interaction so that more Javelin-hybrid peptide is bound to HSP70 in a given formulation.


Figure 2
View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 2. Cross-presentation of the OVA epitope delivered as a hybrid peptide complex with HSP70. A, Adherent PEC from thioglycolate-induced mice were pulsed with 40 nM peptides, alone or in complex with 400 nM HSP70, then cocultured with the B3Z T-T hybridoma as described in Materials and Methods. Pooled data from three experiments are shown, represented as mean supernatant IL-2 quantities ± SE. The p value was calculated using the Student’s t test. B, Mice were immunized with 17.5 µM (2 µg) OVA-L1-J0 in complex with 3.6 µM (12.5 µg) HSP70 or 3.8 µM (0.4 µg) OVA-L1-J1 in complex with 0.74 µM (2.6 µg) HSP70 or controls in a 50-µl volume; 7 days later splenic CD8 T cell responses were evaluated by ELISPOT with peptide pulsed naive splenocytes as APC. Representative data (mean IFN-{gamma} spot number ± SD) for greater than four experiments is shown.

 
Next, the immunization capacity of the HSP70-Javelin hybrid peptide vaccine was evaluated in C57BL/6 mice. It was hypothesized that the greater cross-presentation that occurred with HSP70:OVA-L1-J1 complex compared with HSP70:OVA-L1-J0 complex would enable use of a lower dose of the former complex to achieve the same response as a higher dose of the latter. The difference in dosing requirements could be a direct result of the higher affinity interaction between the J1-containing hybrid peptide and HSP70, and hence a longer interaction time with more peptide bound at equilibrium. Indeed, even when mice were immunized with HSP70:OVA-L1-J1 at a 5-fold lower dose than HSP70:OVA-L1-J0, equivalent results were obtained in the ex vivo IFN-{gamma} ELISPOT (Fig. 2B). Mice immunized with the equivalent low dose of the HSP70:OVA-L1-J0 did not have a measurable response in the assay (data not shown). Additionally, the immunological activity of complexes of HSP70:OVA-L1-J1 tested over many experiments was much less variable than the J0 modified counterpart. Taken together, these data indicate that the higher affinity J1-hybrid peptide in complex with HSP70 consistently induces more vigorous immune responses and better cross-presentation than the previously used, lower affinity J0-hybrid peptide:HSP70 complex.

A novel cleavable linker sequence ensures optimal Ag processing leading to the amplification of immune responses

The epitope contained within the J1-hybrid peptide constructs needs to be processed by the intracellular machinery of APCs to be presented by MHC class I molecules on the cell surface. The original Javelin constructs contained a flexible -GSG- linker, but it became apparent that further improvements in immunogenicity might be achieved by creating a cleavable linker sequence between the epitope and J1 that will be accessible to intracellular enzymes. The linker sequence FFRK (L2) was designed to contain both cathepsin and proteasomal cleavage sites (41, 42, 43, 44, 45). In addition, a peptide was created in which the J1-linker sequence was reoriented to the N-terminal side of the epitope and linker, based on the published data that peptides with a C-terminal Javelin (BiP) appear to be processed via the proteasomal-processing pathway, but peptides with an N-terminal Javelin are processed via an endosomal-processing pathway (21). Controls, or the new L2 sequence-containing constructs, in complex with HSP70 were used to immunize mice, and the results compared with responses from mice immunized with complexes formed with the other hybrid peptides (peptide sequences shown in Table I). T cell responses from immunized animals were analyzed either by ex vivo IFN-{gamma} ELISPOT (Fig. 3A), or by a standard 51Cr-release assay as described in Materials and Methods (Fig. 3B). The ex vivo IFN-{gamma} ELISPOT assay revealed that the most remarkable immune response was detected in mice immunized with HSP70:J1-L2-OVA complexes. The numbers of IFN-{gamma}-secreting CD8+ T cells were increased by >5-fold compared with the number induced by the OVA epitope immunized in adjuvant (TiterMax, Fig. 3A). The other Javelin-hybrid peptide constructs complexed to HSP70 gave responses comparable to the OVA plus TiterMax immunization. This result shows that by including a cleavable linker and changing the orientation of the Javelin-hybrid peptide construct, the responder frequency increases to ~50 per 100,000 CD8+ T cells. It is important to note that in the ex vivo ELISPOT assay, HSP70 complexes containing the OVA-J1 sequence without the L1 linker exhibited the same level of response as the OVA-L1-J1 construct (Fig. 3A), thus confirming that L1 was not a readily cleavable linker and that it did not have a large influence on the hybrid peptide processing. The fact that the minimal OVA epitope in complex with HSP70 (HSP:OVA) elicits a T cell response of similar magnitude to the OVA-J1 and OVA-L1-J1 peptides complexed to HSP70 is likely attributable to cell surface loading of the epitope onto MHC class I molecules in the case of the former immunogen and the suboptimal attributes of the extended peptides in the case of the latter immunogens. The most relevant comparisons to assess in this experiment are among the J1-L2-OVA peptide and the OVA-L1-J1 or J1-L1-OVA peptides where in all cases MHC class I surface loading is unlikely and where the enhanced immunogenicity associated with the optimized Javelin (J1), linker (L2), and orientation (N terminus) is clear.


Figure 3
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 3. Evaluation of new linker/orientation epitopes in vivo. C57BL/6 mice were immunized s.c. at the base of the tail with 19 µM (~2 µg) hybrid peptide complexed with 1.3 µM (4.4 µg) HSP70 or the appropriate controls in 50 µl of saline containing 0.1 mM ADP, as described in Materials and Methods. Seven days later, mice were euthanized and the spleens harvested for analysis. A, CD8+ T cells were enriched from one-half a spleen of each immunized mouse and put into an ex vivo ELISPOT assay to measure epitope-specific IFN-{gamma} production. Inset, The Ags used to pulse naive splenic APC during in vitro stimulation and data are shown as the mean IFN-{gamma} spot number ± SD for three mice per group. B, Epitope-specific CTL responses to complexes formed with new Javelin-hybrid epitope constructs after one restimulation in vitro from the same mice shown in A. Inset, The constructs used for immunization. Cytotoxicity against SIINFEKL-pulsed EL4 cells is plotted, and killing of irrelevant peptide pulsed targets did not exceed 10%. Data are average of three mice per group from one representative experiment.

 
The corresponding CTL assay showed the highest cytotoxic activity was induced in mice immunized with HSP70:J1-L2-OVA complex, followed by HSP70:OVA-L1-J1 complex (Fig. 3B). The differences in rank in potency between the constructs as evaluated by the ex vivo ELISPOT and the 51Cr-release assay may reflect the change in frequency of reactive CD8 T cells that occur as a result of the in vitro expansion of CTL for the 51Cr-release assay. Regardless of the differences between the two assays, J1-L2-OVA was the most effective construct tested in both assays. Among the least potent immunogens in these experiments was HSP70:OVA, which is also intriguing because the epitope does not need to be processed before loading onto MHC class I molecules. Most likely, the low affinity binding between OVA and HSP70 resulted in fewer OVA peptides complexed with HSP70 at the given dose, which was insufficient when compared with the molar equivalent of the Javelin-hybrid peptides. Taken together, these data indicate that designing the cleavable linker to facilitate the processing of Javelin-hybrid peptide constructs and changing the orientation of the hybrid constructs created a more efficient HSP70 peptide complex vaccine that is able to induce vigorous immune responses in vivo.

J1-L2-hybrid peptide processing can occur independently of the proteasome

There has been much attention in the recent literature on the means by which exogenous Ags are delivered into the endogenous pathway of Ag processing and loaded onto MHC class I molecules. Evidence exists for endogenous Ags delivered to MHC class I molecules by both the proteasomal pathway of Ag processing, feeding peptides through the TAP transporter into the endoplasmic reticulum for loading onto MHC class molecules, and by the endosomal route, where MHC class I molecules are recycled through the endosomes to the cell surface (41). To determine whether the processing of the new J1-L2-OVA peptide required the proteasome or was dependent on the TAP transporter for loading onto MHC class I molecules, cross-presentation in the presence of the proteasome inhibitor lactacystin or by peritoneal exudate cells (PECs) from TAP–/– mice was evaluated. As shown in Fig. 4A, the HSP70:OVA-L1-J1 complex required the proteasome for processing, because cross-presentation was significantly inhibited in the presence of 20 µM lactacystin, verifying published observations with HSP70:OVA-L1-J0 (21). In contrast, there was no difference in the cross-presentation of HSP70:J1-L2-OVA between the untreated and lactacystin-treated samples, indicating that there was no requirement for passage through the proteasome. This result is not surprising, because the proteasome is required for proper C-terminal cleavage of epitopes, and the J1-L2-OVA peptide already has the correct C terminus. In addition, MALDI analysis of the J1-L2-OVA peptide treated in vitro with cathepsin B revealed that the peptide is cleaved as predicted in the L2 region, again supporting that the proteasome is not an absolute requirement for epitope cleavage from the J1-L2-hybrid peptides (data not shown). Interestingly, the TAP transporter was not absolutely required for presentation of OVA from either HSP70:OVA-L1-J1 or HSP70:J1-L2-OVA. There was some diminishment in cross-presentation with HSP70:OVA-L1-J1 complex, however, there was no variation between C57BL/6 and TAP–/– PECs in the ability to cross-present OVA derived from HSP70:J1-L2-OVA complex (Fig. 4B). In addition, lactacystin treated or untreated C57BL/6 PECs and TAP–/– PECs all exhibited similar abilities to present exogenously added OVA peptide (Fig. 4). These data indicate that while the new J1-L2-epitope peptides may enter the proteasomal pathway for processing and presentation, it is not an absolute requirement, indicative of the epitopes being loaded into the MHC class I peptide-binding groove by at least one other means.


Figure 4
View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 4. Neither the proteasome nor the TAP transporter are required for cross-presentation of epitopes from HSP70:J1-L2-OVA complexes. Adherent PEC from thioglycolate-induced mice were pretreated for 1 h with 20 µM lactacystin where applicable, pulsed with 40 nM peptide in complex with 400 nM HSP70, then cocultured with the B3Z T-T hybridoma as described in Materials and Methods. Supernatant IL-2 quantities were measured by ELISA after 18 h. A, C57BL/6 PECs untreated ({square}) or lactacystin treated ({blacksquare}); B, C57BL/6 ({square}) or TAP–/– PECs ({blacksquare}). IL-2 levels ± SD from triplicate wells evaluated in duplicate from one representative experiment of at least six similar experiments are shown.

 
The Javelin sequence is specific for HSP interaction and is necessary for enhanced immunogenicity in response to HSP70:Javelin hybrid peptide complexes

We next determined whether the optimized Javelin sequence (J1) was required for the enhanced immunogenicity of the new J1-L2-epitope constructs when injected in complex with HSP70. Mice were immunized with HSP70 complexed with either J1-L2-OVA or L2-OVA, and immune responses evaluated in the ex vivo ELISPOT assay. As shown in Fig. 5A, there were no responses to HSP70:L2-OVA above the negative control; however, there was a good ex vivo response induced in mice immunized with HSP70:J1-L2-OVA at an equimolar dose. There was likewise no response in mice immunized with HSP70 alone (data not shown). These results indicate that the linker alone does not confer the ability of hybrid peptide to elicit immunity.


Figure 5
View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 5. The Javelin sequence and HSP are required for the enhanced immunogenicity of HSP70:Javelin-peptide complexes. A, Mice were immunized with HSP70:J1-L2-OVA or HSP70:L2-OVA or L2-OVA alone (1.3 µM protein:19 µM peptide), then CD8-enriched splenocytes were tested in an ex vivo ELISPOT assay 7 days postimmunization. Solid bars, Effectors stimulated with an irrelevant epitope from vesicular stomatitis virus (VSV) in the absence of APC; open bars, effectors stimulated with OVA in the absence of APC. B, Mice were immunized with 19 µM (2.5 µg) J1-L2-OVA mixed with 1.3 µM of the indicated proteins, or the appropriate controls, and the immune responses measured by ex vivo ELISPOT in the absence of APC. Data are the mean IFN-{gamma} spot number ± SE from three experiments. PhB, phosphorylase B.

 
Although J1 improved immunogenicity with peptides in complex with HSP, it was important to determine whether the effect was specific to HSP or was a property of any large protein that can act as a potential carrier. Mice were immunized with peptides complexed with HSP70, or mixed with phosphorylase B or BSA under the same conditions in which HSP70:peptide complexes are formed, then evaluated for immunogenicity in the ex vivo ELISPOT. As shown in Fig. 5B, there was minimal effect of peptides mixed with BSA or phosphorylase B; however, there was a robust immune response when the J1-L2-OVA peptide was immunized in complex with HSP70. Taken together, these results show that peptides containing the Javelin sequence and the chaperone property associated with HSP70 are necessary for the induction of enhanced immune responses to defined epitopes.

Mice immunized with complexes of HSP70 and the new J1-L2-OVA hybrid peptide are protected from tumor challenge

The new J1-L2-OVA peptide construct was additionally evaluated for its ability to protect against tumor challenge in a prophylactic model of tumor rejection. Mice were immunized on days 0 and 7 with HSP70:J1-L2-OVA or the appropriate controls and, on day 14, challenged in the flank with a s.c. injection of E.G7 cells (the EL4 thymoma transfected with OVA cDNA). Tumor growth was monitored every 3–4 days throughout the duration of the experiment, as described in Materials and Methods. As shown in Fig. 6, the mice immunized with either J1-L2-OVA alone or HSP70:J1-L2-OVA complexes had significantly decreased tumor burden relative to the group immunized with diluent (p < 0.02 and p < 0.001, respectively). Moreover, there were no differences between the diluent group and any of the other immunized groups, including the HSP70:OVA complex-immunized mice, indicating the robustness of the J1-L2-OVA vaccine. These data indicate that the J1-L2- modification of epitopes not only increases their cross-presentation and ex vivo immunogenicity, but also renders the modified epitopes powerfully immunogenic in vivo in the presence and absence of the HSP70 chaperone.


Figure 6
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 6. Immunization with the complex of HSP70 and the improved J1-L2-OVA construct provides protection against subsequent tumor challenge. Ten mice per group were immunized s.c. base of the tail on days 0 and 7 then challenged with 7.5 x 105 EG.7 cells on day 14, as described in Materials and Methods. Mice were observed two to three times per week for the presence of tumors and data are shown as the average tumor volume ± SE for peptide only (16 µM; solid symbols) or HSP70:peptide complexes (1.1 µM protein with 16 µM peptide; open symbols). *, p value <0.02; ***, p value <0.001 by Dunnett’s method.

 
The J1-L2- sequence increases the affinity of several human and murine epitopes for HSP70 resulting in robust immune responses

To test whether the J1-L2-epitope hybrid peptide can be used to deliver other MHC class I-binding epitopes in addition to OVA, we synthesized constructs containing various defined murine and human MHC class I-binding epitopes, and tested their ability to induce immune responses in C57BL/6 and HLA-A*0201-transgenic HHDII mice, respectively. Table III shows the dissociation constant (Kd) for each of the epitopes with and without the addition of the Javelin linker. On average, there was a 68-fold increase in affinity with the J1-L2- modification to each of the epitopes, excluding J1-L2-IMD, which had a 976-fold increase in affinity for HSP70 compared with unmodified IMD (Table III). Mice were immunized with the HSP70:J1-L2-epitope complexes and evaluated for CD8+ Ag-specific T cell responses in the ex vivo IFN-{gamma} ELISPOT assay. Fig. 7A shows the Ag-specific immune responses to three different murine H2-Kb-binding antigenic epitopes, J1-L2-Bcas, J1-L2-SdV, and J1-L2-Vsv. Substantial IFN-{gamma} levels were secreted in response to all three epitopes upon immunization with HSP70:J1-L2-epitope complexes. Cells restimulated with irrelevant peptide had negligible spot numbers (data not shown). Fig. 7B illustrates representative data from HHDII-transgenic mice that were immunized with complexes of HSP70 and HLA-A*0201-presented melanoma hybrid epitopes J1-L2-YMD, J1-L2-IMD, J1-L2-Trp2, and J1-L2-MelA. In all cases there was a high level of IFN-{gamma} produced upon immunization with HSP70:J1-L2-epitope complexes. The background spot number in response to irrelevant peptides in these experiments was less than five spots per well (data not shown). Taken together, these data indicate that the J1-L2- sequence can be used for several human and murine antigenic epitopes to increase their affinity for HSP70, and form immunogenic complexes.


View this table:
[in this window]
[in a new window]
  		Table III. Affinities of multiple murine and human epitopes

 

Figure 7
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 7. The J1-L2-modification is functional for several antigenic epitopes. Mice were immunized s.c. base of the tail with peptides complexed with HSP70 at the optimal concentrations determined for each peptide. CD8-enriched splenocytes were evaluated 7 days later for peptide-specific immune responses in the ex vivo ELISPOT assay without the addition of naive APC. A, Murine hybrid epitopes immunized in complex with HSP70 elicit immune responses in C57BL/6 mice. All data are shown as the mean ± SE of between 3 and 18 observations for each epitope. There were no spots in irrelevant peptide wells (data not shown). B, Human hybrid epitopes (5–10 µg) immunized in complex with HSP70 (20–25 µg) elicit immune responses in HHDII HLA-A2.1-transgenic mice. Data are shown as the mean ± SD of four mice per group from one representative experiment per peptide (from at least three observations per epitope). These were negligible spots in unrelated peptide wells (data not shown). NT, Not tested.

 
Hybrid J1-L2-epitope peptides with uniform HSP70-binding affinities allow multiple-epitope complexes to be immunized within in a single vaccine

Creating hybrid peptides that shared similar affinities for HSP70 enabled us to immunize mice with more than one epitope in a single injection. In theory, the similar HSP70-binding affinities resulted in equimolar amounts of each peptide associated with HSP70 at equilibrium, thus increasing the likelihood of generating robust immune response to each peptide present in a multivalent vaccine. To test this hypothesis, we immunized C57BL/6 or HLA-A*0201 HHDII-transgenic mice with multiple-epitope vaccines. Fig. 8A shows IFN-{gamma} ELISPOT data from mice that were immunized with HSP70:J1-L2-OVA, HSP70:J1-L2-Vsv, or a complex formulated with both peptides. Complexes were prepared as described in Materials and Methods; where more than one peptide was included in a complex the peptides were mixed together when added to HSP70. The amount of HSP70 in each formulation was the same. As observed previously, there was a response to the J1-L2-OVA epitope immunized in the absence of HSP70. The increase in observed immunogenicity of the J1-L2-OVA peptide in Fig. 8A vs Fig. 5 is attributable to the increased dose of protein and peptide immunized in Fig. 8A. J1-L2-Vsv alone did not stimulate IFN-{gamma} production. The lack of response to this particular hybrid peptide in the absence of HSP70 is not clear.


Figure 8
View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 8. Ex vivo IFN-{gamma} ELISPOT responses to multiple epitopes in a single immunization. A, C57BL/6 mice were immunized with controls, single-epitope complexes, or two hybrid epitope complexes (in a single injection), then tested for epitope-specific immune responses in the ex vivo ELISPOT assay without the addition of naive APC. The HSP70 concentration in each complex was 3.2 µM. Open bars represent OVA-specific responses, dark gray bars correspond to Vsv-specific responses, Light gray bars show response to irrelevant peptide, and black bars are the medium control wells. B, HHDII mice were immunized with single-epitope complexes, a mixture of two single-epitope complexes (complex mix), or a complex formulated with both epitopes at the same time (single complex); 7 days later, CD8 T cells were analyzed for epitope-specific IFN-{gamma} production using peptide-pulsed naive splenocytes as APCs.

 
Also shown in Fig. 8A, there were substantial IFN-{gamma} responses from mice immunized with HSP70:J1-L2-OVA or HSP70:J1-L2-Vsv complexes; importantly, not only was there a robust response when the mice were immunized with the two hybrid epitopes in a single complex, there was also no diminishment in the responsiveness to either epitope when compared with the single complex immunizations. Likewise, as shown in Fig. 8B, when complexes containing human J1-L2-epitope melanoma Ags in association with HSP70 were injected into the HHDII-transgenic mice, the mice were able to generate responses to both epitopes. Coimmunization of complexes that were prepared individually then mixed just before immunization resulted in immune responses that mirrored the response to each epitope delivered in a single epitope complex, although the response to the IMD epitope was somewhat diminished relative to that peptide immunized in a single epitope complex. In contrast, when the two epitopes were mixed when complexed with HSP70, there was an "equalized" immune response such that the response to the YMD was augmented over the response to the same epitope immunized in a single epitope complex.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
The modification of antigenic epitopes by collinear synthesis with a high-affinity HSP-binding sequence enhances their immunogenicity (34). As shown here, optimization of the Javelin sequence further increases and equalizes the affinity of antigenic epitopes for HSP70. Additionally modifying the linker sequence to include putative cleavage sites and changing the orientation of the construct to the N terminus of the epitopes resulted in added enhancement of immunogenicity. Moreover, this peptide modification is universally applicable to both murine and human epitopes, enhancing the likelihood of efficacy of vaccines containing multivalent, HSP70:peptide complexes.

There is a strong body of literature that illustrates how HSP70 and other cell-derived chaperones can serve as adjuvants to deliver antigenic epitopes to the host immune system and induce cell-mediated immunity against tumor and viral targets (10, 11, 12, 13, 14, 15, 17, 18, 19, 20, 34, 46, 47). These are very crucial observations because historically, vaccines delivered with conventional adjuvants designed to elicit T cell responses have been poorly tolerated or plagued with side effects that have restricted their approval for human use (48). However, there are potential limitations to vaccines made by loading HSPs with diverse peptides in vitro. Individual peptide sequences have quite variable affinities for HSP (33, 34), and in equilibrium reactions in which more than one peptide is complexed with HSP70 in vitro, a higher affinity peptide will bind HSP at the expense of the lower affinity peptide, with the potential effect of inefficient priming of the immune system to the underrepresented peptide.

It was previously demonstrated that hybrid peptides containing an antigenic epitope collinearly synthesized with J0, a high-affinity HSP70-binding sequence, complexed with HSP70 could dramatically enhance the potency of immunization when compared with the unmodified epitope complexed with HSP70 (34). The increased efficacy of immunization was attributed to improved affinity of the peptides for the HSP70. However, evaluation of the binding kinetics of a variety of the J0-hybrid peptide constructs revealed that while the dissociation constants were lowered 5- to 30-fold relative to the minimal epitope alone, the epitopes still greatly influenced the binding affinity of the J0-hybrid peptides. The binding affinities of the J0-hybrid peptides reported here cover two orders of magnitude (4.2–180 µM; Table II). These differences could affect the ability of multiple epitopes to be delivered equally in a single vaccination.

Modifying the Javelin to an even higher affinity HSP-binding sequence (J1) normalizes the HSP70 affinity of all the J1-hybrid peptides tested such that their dissociation constants are within ~3 µM of one another, thus creating a more "universal" Javelin (Tables II and III; Fig. 1). The higher affinity interaction of any given peptide with HSP70 results in a better chance for the HSP-complexed peptide to be delivered to APCs with high efficiency resulting in robust immune responses. Moreover, the higher affinity sequence resulted in greater uniformity of responses between experiments. In addition, combining multiple peptides that have been modified with the J1 sequence with HSP70 makes it possible to deliver more than one epitope in a single injection with similar efficiencies (Fig. 8).

Evaluation of the new Javelin (J1)-hybrid peptide led to the conclusion that the higher affinity Javelin sequence was an improvement over the previously published Javelin (J0) constructs. Nevertheless, the frequency of IFN-{gamma}-secreting CD8+ T cells induced by immunization with HSP70:J1-hybrid peptide complexes was rather low (Fig. 2B, 1 cell/10,000), indicating less-than-optimal immune responses. This observation led to additional reevaluation of the J1-hybrid peptide in an attempt to further improve the immunogenicity of the HSP70:Javelin-peptide complexes.

Data from Castellino et al. (21) using J0-modified peptides revealed that both cytosolic and endocytic routes were responsible for MHC class I presentation of the HSP-associated peptides. The orientation of the Javelin-peptide sequence governed the observed effect, such that when the J0 was synthesized C-terminal to the epitope, the peptide was processed via the cytosolic route, but when the J0 was synthesized N-terminal to the epitope, the peptide was processed via the endocytic route. One of the explanations for the difference may be the affinity of the peptide for HSP70. OVA-L1-J0 has ~16-fold lower affinity for HSP70 than J0-L1-OVA (P. Slusarewicz and A. Kays Leonard, unpublished observations). The difference in affinity has at least three implications. First, the literature reveals that peptide binding changes the conformation and rigidity of HSP70 (49, 50), leaving the possibility that higher affinity peptides may have a more profound affect on the HSP70 conformation, potentially altering which receptor the HSP-peptide complex binds and as a result becomes incorporated into the APC. Second, a higher affinity interaction may affect how a peptide is processed within the cell by protecting the peptide from proteolytic digestion thereby increasing its half-life within the cell (51). Or third, the result may be that a greater proportion of HSP70 is bound by peptide, and at least for the CD40 receptor, the interaction with HSP70 is strongly increased when HSP70 is complexed with a peptide substrate (31). However, it seems the affinity is not the only explanation for route of processing, because OVA-L1-J1 and J1-L2-OVA have similar affinities for HSP70, yet OVA-L1-J1 is dependent on the proteasome for processing, but J1-L2-OVA is not (Fig. 4). Because the proteasome is responsible for correct C-terminal cleavage of epitopes (reviewed in Ref. 52), and the peptide J1-L2-OVA already has the correct C terminus but OVA-L1-J1 does not, it is possible that the difference observed is simply due to the requirement for processing. The data do not rule out that J1-L2-OVA can enter the cytosolic pathway of peptide processing. Rather, they suggest that the proteasome is not an absolute requirement for MHC class I presentation of this peptide.

Further evidence that peptide processing is also significant for increased immunogenicity is illustrated by comparing J1-L1-OVA and J1-L2-OVA. These two peptides share the Javelin-epitope orientation and the concomitantly low HSP70-binding affinities of the J1 epitope (1.63 vs 2.26 µM, respectively), yet J1-L2-OVA induces immune responses greater than predicted by the difference in the binding affinities (Fig. 3). Indeed, the increase in immune responses must be attributed to the linker modification, because all other components are the same. The L2 linker sequence created an optimal target for constitutively expressed proteolytic enzymes as well as a potential cleavage site for the proteasome itself, thereby increasing the potential for the correct epitope sequence to be generated regardless of peptide delivery into the cytosolic or endocytic routes of peptide processing.

The optimized Javelin-linker-epitope sequence is quite potent, and it is interesting to note that the Javelin-modified peptides have weak to moderate immunizing potential on their own (Figs. 6–8), repeating a phenomenon observed with the OVA-L1-J0 peptide (34). Experiments are currently underway to understand how the peptides can be immunogenic in the absence of HSP70 or adjuvant, but there are several working hypotheses. First, the peptide may be binding endogenous HSP70 that is released as a result of cellular damage that is incurred upon immunization or present in the serum of individuals. Indeed, the peptide induces better immune responses in the presence of ADP (J. B. Flechtner, unpublished observations), suggesting that this may be the case because ADP enhances HSP70-peptide interactions. Second, the hydrophobic nature of the peptide may render it a cell-penetrating peptide, such that it is inserted into the cytoplasmic membrane of the cell and internalized. Third, again due to the hydrophobic nature of the Javelin sequence, the peptide may form aggregates or other higher-order structures in the absence of HSP and therefore be taken up into APCs by phagocytosis. Regardless of the mechanism, immunization of mice with J1-L2-OVA in saline is quite effective at protecting them against a tumor challenge. It is surprising, therefore, that there can be such low numbers of CD8 T cells producing IFN-{gamma} in response to the J1-L2-OVA peptide immunized in the absence of HSP70 (Fig. 5B) but robust antitumor responses (Fig. 6). This may be a direct reflection of the relative insensitivity of the ELISPOT to predict effector T cell efficiency, or more likely, a reflection of increase in CTL frequency that occurs as a result of the prime-boost regimen used in the tumor rejection assay compared with the prime-only protocol used for the ex vivo ELISPOT analysis. It will be of interest to test whether under conditions of limited peptide (<16 µM in the experiment described in Fig. 6), the J1-L2 modification renders peptides delivered in complex with HSP more immunogenic than the corresponding amounts of unmodified peptide in complex with HSP.

Importantly, robust immune responses to peptides modified with the Javelin-linker sequence are not limited to a single epitope. Both human and mouse antigenic epitopes are amenable to modification with the Javelin sequence and remain potent stimulators of the immune system when delivered in complex with HSP70. Interestingly, the Trp2 Ag is a "self" epitope for both mice and humans that typically requires multiple immunizations to break tolerance to the Ag (53). However, in the HHDII mice, only one immunization was required to break tolerance and induce strong immune responses.

Finally, the Javelin modification allows more than one epitope to be delivered in a single immunization. Notably, there does not appear to be an immunodominance issue–multiple epitopes can be delivered in a single vaccine without great expense to the response to an individual epitope. Currently, work is in progress to determine whether a single Javelin-linker sequence can be used to modify a "string" of several epitopes to circumvent potential solubility issues with the relatively hydrophobic Javelin sequence.

In summary, modifying the sequence, linker, and orientation of Javelin-epitope constructs not only increases and normalizes their affinity for HSP70 but optimizes their ability to be delivered to APCs, processed and presented by the cell, and in turn induce robust immune responses. The Javelin-hybrid peptides can be mixed together and complexed with HSP70 resulting in multivalent immune responses that are as potent as immunizing separate HSP70/peptide complexes that are mixed just before immunization. Thus, with the new Javelin-linker modified peptides, the elegantly characterized, adjuvant-free, HSP-based vaccine is optimized to deliver multiple Ags with equal efficiency to induce potent immune responses to either infectious disease or tumor targets.


   Acknowledgments
 
We are grateful to the former crew of Mojave Therapeutics, especially Priscilla Calderon, Armin Lahiji, Kevin Wright, Adrienne Scott, and Cara Miller for technical assistance with immunological assays, as well as Nadine Soplop, Nicole Covino, George Angelos, Jeff Courter, and Jason Tenzer for technical assistance with biochemical assays. Thanks to Denise Ireland, Jennifer Burke, and the crew of the animal facility at Antigenics for experimental help and animal handling. We also thank Nilabh Shastri for the B3Z T-T hybridoma, François Lemmonier for supplying the HHDII-transgenic mice, and Roman Chicz and Robert Binder for critically reading this manuscript.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
J. B. Flechtner, K. P. Cohane, S. Mehta, P. Slusarewicz, B. H. Barber, and S. Andjelic have two pending patents, both titled "Improved heat shock protein-based vaccines and immunotherapies." Both patents were filed by Mojave Therapeutics, and the Intellectual Property was assigned to Antigenics Inc. D. L. Levey is a current, stockholding employee of Antigenics Inc.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 Current address: Centocor, 145 King of Prussia Road, Radnor, PA 19087. Back

2 Current address: DFB Pharmaceuticals, 318 McCullough, San Antonio, TX 78215. Back

3 Current address: Nastech Pharmaceutical, 3450 Monte Villa Parkway, Bothell, WA 98021. Back

4 Current address: University Health Network, 7-504, 610 University Avenue, Toronto, Ontario, Canada M5G 2M9. Back

5 Address correspondence and reprint requests to Dr. Daniel L. Levey, Antigenics, 3 Forbes Road, Lexington, MA 02421; E-mail address: daniel.levey{at}antigenics.com or Dr. Sofija Andjelic at the current address: Progenics Pharmaceuticals, 777 Old Saw Mill River Road, Tarrytown, NY 10591; E-mail address: sandjelic{at}progenics.com Back

6 Abbreviations used in this paper: HSP, heat shock protein; PEC, peritoneal exudate cell. Back

Received for publication July 21, 2005. Accepted for publication April 28, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Chappell, T. G., B. B. Konforti, S. L. Schmid, J. E. Rothman. 1987. The ATPase core of a clathrin uncoating protein. J. Biol. Chem. 262: 746-751. [Abstract/Free Full Text]
  2. Flynn, G. C., T. G. Chappell, J. E. Rothman. 1989. Peptide binding and release by proteins implicated as catalysts of protein assembly. Science 245: 385-390. [Abstract/Free Full Text]
  3. Basu, S., R. J. Binder, R. Suto, K. M. Anderson, P. K. Srivastava. 2000. Necrotic but not apoptotic cell death releases heat shock proteins, which deliver a partial maturation signal to dendritic cells and activate the NF-{kappa}B pathway. Int. Immunol. 12: 1539-1546. [Abstract/Free Full Text]
  4. Singh-Jasuja, H., H. U. Scherer, N. Hilf, D. Arnold-Schild, H. G. Rammensee, R. E. Toes, H. Schild. 2000. The heat shock protein gp96 induces maturation of dendritic cells and down-regulation of its receptor. Eur. J. Immunol. 30: 2211-2215. [Medline]
  5. Flohe, S. B., J. Bruggemann, S. Lendemans, M. Nikulina, G. Meierhoff, S. Flohe, H. Kolb. 2003. Human heat shock protein 60 induces maturation of dendritic cells versus a Th1-promoting phenotype. J. Immunol. 170: 2340-2348. [Abstract/Free Full Text]
  6. Wang, Y., T. Whittall, E. McGowan, J. Younson, C. Kelly, L. A. Bergmeier, M. Singh, T. Lehner. 2005. Identification of stimulating and inhibitory epitopes within the heat shock protein 70 molecule that modulate cytokine production and maturation of dendritic cells. J. Immunol. 174: 3306-3316. [Abstract/Free Full Text]
  7. Wang, Y., C. G. Kelly, M. Singh, E. G. McGowan, A. S. Carrara, L. A. Bergmeier, T. Lehner. 2002. Stimulation of Th1-polarizing cytokines, C-C chemokines, maturation of dendritic cells, and adjuvant function by the peptide binding fragment of heat shock protein 70. J. Immunol. 169: 2422-2429. [Abstract/Free Full Text]
  8. Wan, T., X. Zhou, G. Chen, H. An, T. Chen, W. Zhang, S. Liu, Y. Jiang, F. Yang, Y. Wu, X. Cao. 2003. Novel heat shock protein Hsp70L1 activates dendritic cells and acts as a Th1 polarizing adjuvant. Blood 103: 1747-1754. [Medline]
  9. Panjwani, N. N., L. Popova, P. K. Srivastava. 2002. Heat shock proteins gp96 and hsp70 activate the release of nitric oxide by APCs. J. Immunol. 168: 2997-3003. [Abstract/Free Full Text]
  10. Srivastava, P. K., H. Udono. 1994. Heat shock protein-peptide complexes in cancer immunotherapy. Curr. Opin. Immunol. 6: 728-732. [Medline]
  11. Srivastava, P. K., A. B. DeLeo, L. J. Old. 1986. Tumor rejection antigens of chemically induced sarcomas of inbred mice. Proc. Natl. Acad. Sci. USA 83: 3407-3411. [Abstract/Free Full Text]
  12. Udono, H., P. K. Srivastava. 1993. Heat shock protein 70-associated peptides elicit specific cancer immunity. J. Exp. Med. 178: 1391[Abstract/Free Full Text]
  13. Udono, H., D. L. Levey, P. K. Srivastava. 1994. Cellular requirements for tumor-specific immunity elicited by heat shock proteins: tumor rejection antigen gp96 primes CD8+ T cells in vivo. Proc. Natl. Acad. Sci. USA 91: 3077-3081. [Abstract/Free Full Text]
  14. Li, Z., P. K. Srivastava. 1993. Tumor rejection antigen gp96/grp94 is an ATPase: implications for protein folding and antigen presentation. EMBO J. 12: 3143-3151. [Medline]
  15. Sato, K., Y. Torimoto, Y. Tamura, M. Shindo, H. Shinzaki, K. Hirai, Y. Kohgo. 2001. Immunotherapy using heat-shock protein preparations of leukemia cells after syngeneic bone marrow transplantation in mice. Blood 98: 1852-1857. [Abstract/Free Full Text]
  16. Zugel, U., A. M. Sponaas, J. Neckermann, B. Schoel, S. H. Kaufmann. 2001. gp96-peptide vaccination of mice against intracellular bacteria. Infect. Immun. 69: 4164-4167. [Abstract/Free Full Text]
  17. Binder, R. J., P. K. Srivastava. 2005. Peptides chaperoned by heat-shock proteins are a necessary and sufficient source of antigen in the cross-priming of CD8+ T cells. Nat. Immunol. 6: 593-599. [Medline]
  18. Blachere, N. E., Z. Li, R. Y. Chandawarkar, R. Suto, N. S. Jaikaria, S. Basu, H. Udono, P. K. Srivastava. 1997. Heat shock protein-peptide complexes, reconstituted in vitro, elicit peptide-specific cytotoxic T lymphocyte response and tumor immunity. J. Exp. Med. 186: 1315-1322. [Abstract/Free Full Text]
  19. Ciupitu, A. M., M. Petersson, C. L. O’Donnell, K. Williams, S. Jindal, R. Kiessling, R. M. Welsh. 1998. Immunization with a lymphocytic choriomeningitis virus peptide mixed with heat shock protein 70 results in protective antiviral immunity and specific cytotoxic T lymphocytes. J. Exp. Med. 187: 685-691. [Abstract/Free Full Text]
  20. Navaratnam, M., M. S. Deshpande, M. J. Hariharan, D. S. Zatechka, Jr, S. Srikumaran. 2001. Heat shock protein-peptide complexes elicit cytotoxic T-lymphocyte and antibody responses specific for bovine herpesvirus 1. Vaccine 19: 1425-1434. [Medline]
  21. Castellino, F., P. E. Boucher, K. Eichelberg, M. Mayhew, J. E. Rothman, A. N. Houghton, R. N. Germain. 2000. Receptor-mediated uptake of antigen/heat shock protein complexes results in major histocompatibility complex class I antigen presentation via two distinct processing pathways. J. Exp. Med. 191: 1957-1964. [Medline]
  22. Sondermann, H., T. Becker, M. Mayhew, F. Wieland, F. U. Hartl. 2000. Characterization of a receptor for heat shock protein 70 on macrophages and monocytes. Biol. Chem. 381: 1165-1174. [Medline]
  23. Wang, Y., C. G. Kelly, J. T. Karttunen, T. Whittall, P. J. Lehner, L. Duncan, P. MacAry, J. S. Younson, M. Singh, W. Oehlmann, et al 2001. CD40 is a cellular receptor mediating mycobacterial heat shock protein 70 stimulation of CC-chemokines. Immunity 15: 971-983. [Medline]
  24. Basu, S., R. J. Binder, T. Ramalingam, P. K. Srivastava. 2001. CD91 is a common receptor for heat shock proteins gp96, hsp90, hsp70, and calreticulin. Immunity 14: 303-313. [Medline]
  25. Vabulas, R. M., P. Ahmad-Nejad, S. Ghose, C. J. Kirschning, R. D. Issels, H. Wagner. 2002. HSP70 as endogenous stimulus of the Toll/interleukin-1 receptor signal pathway. J. Biol. Chem. 277: 15107-15112. [Abstract/Free Full Text]
  26. Asea, A., S. K. Kraeft, E. A. Kurt-Jones, M. A. Stevenson, L. B. Chen, R. W. Finberg, G. C. Koo, S. K. Calderwood. 2000. HSP70 stimulates cytokine production through a CD14-dependant pathway, demonstrating its dual role as a chaperone and cytokine. Nat. Med. 6: 435-442. [Medline]
  27. Asea, A., M. Rehli, E. Kabingu, J. A. Boch, O. Bare, P. E. Auron, M. A. Stevenson, S. K. Calderwood. 2002. Novel signal transduction pathway utilized by extracellular HSP70: role of Toll-like receptor (TLR) 2 and TLR4. J. Biol. Chem. 277: 15028-15304. [Abstract/Free Full Text]
  28. Delneste, Y., G. Magistrelli, J. Gauchat, J. Haeuw, J. Aubry, K. Nakamura, N. Kawakami-Honda, L. Goetsch, T. Sawamura, J. Bonnefoy, P. Jeannin. 2002. Involvement of LOX-1 in dendritic cell-mediated antigen cross-presentation. Immunity 17: 353-362. [Medline]
  29. Berwin, B., J. P. Hart, S. Rice, C. Gass, S. V. Pizzo, S. R. Post, C. V. Nicchitta. 2003. Scavenger receptor-A mediates gp96/GRP94 and calreticulin internalization by antigen-presenting cells. EMBO J. 22: 6127-6136. [Medline]
  30. Berwin, B., Y. Delneste, R. V. Lovingood, S. R. Post, S. V. Pizzo. 2004. SREC-I, a type F scavenger receptor, is an endocytic receptor for calreticulin. J. Biol. Chem. 279: 51250-51257. [Abstract/Free Full Text]
  31. Becker, T., F. U. Hartl, F. Wieland. 2002. CD40, an extracellular receptor for binding and uptake of Hsp70-peptide complexes. J. Cell Biol. 158: 1277-1285. [Abstract/Free Full Text]
  32. Binder, R. J., R. Vatner, P. Srivastava. 2004. The heat-shock protein receptors: some answers and more questions. Tissue Antigens 64: 442-451. [Medline]
  33. Flynn, G. C., J. Pohl, M. T. Flocco, J. E. Rothman. 1991. Peptide-binding specificity of the molecular chaperone BiP. Nature 353: 726-730. [Medline]
  34. Moroi, Y., M. Mayhew, J. Trcka, M. H. Hoe, Y. Takechi, F. U. Hartl, J. E. Rothman, A. N. Houghton. 2000. Induction of cellular immunity by immunization with novel hybrid peptides complexed to heat shock protein 70. Proc. Natl. Acad. Sci. USA 97: 3485-3490. [Abstract/Free Full Text]
  35. Pascolo, S., N. Bervas, J. M. Ure, A. G. Smith, F. A. Lemonnier, B. Perarnau. 1997. HLA-A2.1-restricted education and cytolytic activity of CD8+ T lymphocytes from beta2 microglobulin (beta2m) HLA-A2.1 monochain transgenic H-2Db beta2m double knockout mice. J. Exp. Med. 185: 2043-2051. [Abstract/Free Full Text]
  36. Blond-Elguindi, S., S. E. Cwirla, W. J. Dower, R. J. Lipshutz, S. R. Sprang, J. F. Sambrook, M. J. Gething. 1993. Affinity panning of a library of peptides displayed on bacteriophages reveals the binding specificity of BiP. Cell 75: 717-728. [Medline]
  37. Gragerov, A., M. E. Gottesman. 1994. Different peptide binding specificities of hsp70 family members. J. Mol. Biol. 241: 133-135. [Medline]
  38. Gragerov, A., L. Zeng, X. Zhao, W. Burkholder, M. E. Gottesman. 1994. Specificity of DnaK-peptide binding. J. Mol. Biol. 235: 848-854. [Medline]
  39. Takenaka, I. M., S. M. Leung, S. J. McAndrew, J. P. Brown, L. E. Hightower. 1995. Hsc70-binding peptides selected from a phage display peptide library that resemble organellar targeting sequences. J. Biol. Chem. 270: 19839-19844. [Abstract/Free Full Text]
  40. MacAry, P. A., B. Javid, R. A. Floto, K. G. Smith, W. Oehlmann, M. Singh, P. J. Lehner. 2004. HSP70 peptide binding mutants separate antigen delivery from dendritic cell stimulation. Immunity 20: 95-106. [Medline]
  41. Ackerman, A. L., P. Cresswell. 2004. Cellular mechanisms governing cross-presentation of exogenous antigens. Nat. Immunol. 5: 678-684. [Medline]
  42. Kamboj, R. C., S. Pal, N. Raghav, H. Singh. 1993. A selective colorimetric assay for cathepsin L using Z-Phe-Arg-4-methoxy-beta-naphthylamide. Biochimie 75: 873-878. [Medline]
  43. Higaki, J., R. Catalano, A. W. Guzzetta, D. Quon, J. F. Nave, C. Tarnus, H. D’Orchymont, B. Cordell. 1996. Processing of beta-amyloid precursor protein by cathepsin D. J. Biol. Chem. 271: 31885-31893. [Abstract/Free Full Text]
  44. Kuttler, C., A. K. Nussbaum, T. P. Dick, H. G. Rammensee, H. Schild, K. P. Hadeler. 2000. An algorithm for the prediction of proteasomal cleavages. J. Mol. Biol. 298: 417-429. [Medline]
  45. Schwarz, G., W. H. Boehncke, M. Braun, C. J. Schroter, T. Burster, T. Flad, D. Dressel, E. Weber, H. Schmid, H. Kalbacher. 2002. Cathepsin S activity is detectable in human keratinocytes and is selectively upregulated upon stimulation with interferon-{gamma}. J. Invest. Dermatol. 119: 44-49. [Medline]
  46. Arnold, D., S. Faath, H. Rammensee, H. Schild. 1995. Cross-priming of minor histocompatibility antigen-specific cytotoxic T cells upon immunization with the heat shock protein gp96. J. Exp. Med. 182: 885-889. [Abstract/Free Full Text]
  47. Brenner, B. G., M. A. Wainberg. 1999. Heat shock protein-based therapeutic strategies against human immunodeficiency virus type 1 infection. Infect. Dis. Obstet. Gynecol. 7: 80-90. [Medline]
  48. Gupta, R. K., E. H. Relyveld, E. B. Lindblad, B. Bizzini, S. Ben-Efraim, C. K. Gupta. 1993. Adjuvants–a balance between toxicity and adjuvanticity. Vaccine 11: 293-306. [Medline]
  49. Slepenkov, S. V., S. N. Witt. 2003. Detection of a concerted conformational change in the ATPase domain of DnaK triggered by peptide binding. FEBS Lett. 539: 100-104. [Medline]
  50. Stevens, S. Y., S. Cai, M. Pellecchia, E. R. Zuiderweg. 2003. The solution structure of the bacterial HSP70 chaperone protein domain DnaK(393–507) in complex with the peptide NRLLLTG. Protein Sci. 12: 2588-2596. [Medline]
  51. Reits, E., A. Griekspoor, J. Neijssen, T. Groothuis, K. Jalink, P. van Veelen, H. Janssen, J. Calafat, J. W. Drijfhout, J. Neefjes. 2003. Peptide diffusion, protection, and degradation in nuclear and cytoplasmic compartments before antigen presentation by MHC class I. Immunity 18: 97-108. [Medline]
  52. York, I. A., A. L. Goldberg, X. Y. Mo, K. L. Rock. 1999. Proteolysis and class I major histocompatibility complex antigen presentation. Immunol. Rev. 172: 49-66. [Medline]
  53. Wang, R.-F., E. Appella, Y. Kawakami, X. Kang, S. A. Rosenberg. 1996. Identification of TRP-2 as a human tumor antigen recognized by cytotoxic T lymphocytes. J. Exp. Med. 184: 2207-2216. [Abstract/Free Full Text]
  54. Totsuka, M., M. Kakehi, M. Kohyama, S. Hachimura, T. Hisatsune, S. Kaminogawa. 1998. Enhancement of antigen-specific IFN-{gamma} production from CD8+ T cells by a single amino acid-substituted peptide derived from bovine {alpha}s1-casein. Clin. Immunol. Immunopathol. 88: 277-286. [Medline]
  55. Kast, W. M., L. Roux, J. Curren, H. J. Blom, A. C. Voordouw, R. H. Meloen, D. Kolakofsky, C. J. Melief. 1991. Protection against lethal Sendai virus infection by in vivo priming of virus-specific cytotoxic T lymphocytes with a free synthetic peptide. Proc. Natl. Acad. Sci. USA 88: 2283-2287. [Abstract/Free Full Text]
  56. Van Bleek, G. M., S. G. Nathenson. 1990. Isolation of an endogenously processed immunodominant viral peptide from the class I H-2Kb molecule. Nature 348: 213-216. [Medline]
  57. Skipper, J. C., R. C. Hendrickson, P. H. Gulden, V. Brichard, A. Van Pel, Y. Chen, J. Shabanowitz, T. Wolfel, C. L. Slingluff, Jr, T. Boon, et al 1996. An HLA-A2-restricted tyrosinase antigen on melanoma cells results from posttranslational modification and suggests a novel pathway for processing of membrane proteins. J. Exp. Med. 183: 527-534. [Abstract/Free Full Text]
  58. Parkhurst, M. R., M. L. Salgaller, S. Southwood, P. F. Robbins, A. Sette, S. A. Rosenberg, Y. Kawakami. 1996. Improved induction of melanoma-reactive CTL with peptides from the melanoma antigen gp100 modified at HLA-A*0201-binding residues. J. Immunol. 157: 2539-2548. [Abstract]
  59. Parkhurst, M. R., E. B. Fitzgerald, S. Southwood, A. Sette, S. A. Rosenberg, Y. Kawakami. 1998. Identification of a shared HLA-A*0201-restricted T-cell epitope from the melanoma antigen tyrosinase-related protein 2 (TRP2). Cancer Res. 58: 4895-4901. [Abstract/Free Full Text]
  60. Kawakami, Y., S. Eliyahu, K. Sakaguchi, P. F. Robbins, L. Rivoltini, J. R. Yannelli, E. Appella, S. A. Rosenberg. 1994. Identification of the immunodominant peptides of the MART-1 human melanoma antigen recognized by the majority of HLA-A2-restricted tumor infiltrating lymphocytes. J. Exp. Med. 180: 347-352. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
H. Bendz, S. C. Ruhland, M. J. Pandya, O. Hainzl, S. Riegelsberger, C. Brauchle, M. P. Mayer, J. Buchner, R. D. Issels, and E. Noessner
Human Heat Shock Protein 70 Enhances Tumor Antigen Presentation through Complex Formation and Intracellular Antigen Delivery without Innate Immune Signaling
J. Biol. Chem., October 26, 2007; 282(43): 31688 - 31702.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
B. Javid, P. A. MacAry, and P. J. Lehner
Structure and Function: Heat Shock Proteins and Adaptive Immunity
J. Immunol., August 15, 2007; 179(4): 2035 - 2040.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Flechtner, J. B.
Right arrow Articles by Andjelic, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Flechtner, J. B.
Right arrow Articles by Andjelic, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS