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


     
 

The Journal of Immunology, 2007, 179, 7254-7261
Copyright © 2007 by The American Association of Immunologists, Inc.
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 Binder, R. J.
Right arrow Articles by Srivastava, P. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Binder, R. J.
Right arrow Articles by Srivastava, P. K.

Specific Immunogenicity of Heat Shock Protein gp96 Derives from Chaperoned Antigenic Peptides and Not from Contaminating Proteins1

Robert J. Binder2,*, John B. Kelly, III{dagger}, Ralph E. Vatner{dagger} and Pramod K. Srivastava{dagger}

* Department of Immunology, University of Pittsburgh, Pittsburgh, PA 15261; and {dagger} Center for Immunotherapy of Cancer and Infectious Diseases, Department of Immunology, University of Connecticut School of Medicine, Farmington, CT 06030


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The peptide-binding property of MHC is central to adaptive immunological functions. A similar property of heat shock proteins (HSPs) hsp70 and hsp90 has been implicated in Ag presentation by MHC and in cross-priming. The peptide-binding pocket of hsp70 has been characterized structurally and functionally and a peptide-binding site in gp96 (of hsp90 family) has been defined. Nonetheless, questions persist whether the specific immunogenicity of HSP preparations derives from the peptides chaperoned by the HSPs or by proteins contaminating the HSP preparations. Because absolute purity of a protein preparation is a metaphysical concept, other approaches are necessary to address the question. In this study, we demonstrate that the specific immunogenicity of gp96 preparations isolated from cells expressing β-galactosidase derives from the MHC I epitope precursors associated with the gp96 and not from contaminating β-galactosidase protein nor unassociated fragments derived from it. Although the observations here are limited to a single HSP and antigenic peptides chaperoned by it, they can be extended broadly.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Apparently homogeneous preparations of selected heat shock proteins (HSPs)3 (gp96, hsp90, hsp70, calreticulin, hsp110, grp170) have been shown to elicit specific CD8 T cell response to Ags expressed by the cells from which the HSPs have been purified (for review, see Ref. 1). Such immunity has been shown to be protective against the cancers, viruses, or intracellular bacteria that express the cognate Ags. The same HSPs purified from normal tissues or from tissues expressing irrelevant Ags show no immunogenicity. When the observations were originally made, it was suggested that the genes encoding HSPs were hypervariable (2) and, thus, lent themselves to be sources of a tremendous range of antigenic variation. However, the sequencing of cDNAs of HSPs from normal tissues and tumors failed to show any evidence of the proposed hypervariability of the hsp genes. Since 1) the HSP preparations appeared, by all of the methods available at the time, to be homogenous and devoid of detectable contaminating proteins, 2) elution of HSP bands from polyacrylamide gels led to isolation of immunologically active material (3), and 3) glycosylation did not appear to play a role in immunogenicity, it was suggested in 1991 (4, 5) that the immunogenicity of HSP preparations was derived from antigenic peptides associated with the HSPs. That idea has derived significant support since then. Thus, defined antigenic peptides derived from viral Ags (6, 7, 8), tumor Ags (9), mycobacterial Ags (10), model Ags (11), or minor histocompatibility Ags (12) have been isolated from gp96, hsp90, or hsp70. Furthermore, immunological evidence has been reported for association of these same HSPs with well-characterized epitopes of viral Ags (7, 8), tumor Ags (13, 14, 15), or model Ags (11, 16). Peptide-binding sites of hsp70 (17, 18) and gp96 (19, 20) have been demonstrated by crystallography and modeling studies. Nonetheless, the notion that specific immunogenicity of HSP preparations derives from peptides noncovalently and physiologically associated with the HSP molecule has been questioned recently (21, 22, 23, 24). The questioning is based on the idea that HSP preparations are contaminated with trace amounts of other cellular proteins and that this trace contamination and not the HSP-chaperoned peptides is responsible for the specific immunogenicity of the HSP. This criticism has not been experimentally substantiated, but is a reasonable one and merits experimental scrutiny.

In this study, we provide such a scrutiny, using the HSP gp96 purified from murine P13.4 cells, which are P815 cells transfected with the gene encoding β-galactosidase (β-gal). The choice of the Ag was influenced by the fact that it is detectable structurally as well as enzymatically, such that logarithmically smaller quantities of β-gal can be detected than possible for other well-used model Ags. Furthermore, a range of immunologically relevant tools are available for this Ag. Using a range of sensitive, complementary, and redundant tools, our data clearly demonstrate that the specific immunogenicity of gp96 preparations isolated from β-gal-expressing cells derives from gp96-chaperoned peptides and not from contaminating β-gal.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice and reagents

Mice were purchased from The Jackson Laboratory and maintained in the Center for Laboratory Animal Care facilities at the University of Connecticut Health Center. Female C57BL/6 mice were used at 6–8 wk of age. gp96 was purified from P13.4, P815, or PIK23 cells in identical procedures as previously described (3). Peptides were stripped off gp96 exactly as previously published (9). Untransfected P815 and β-gal- transfected P13.4 cells were maintained in complete medium (RPMI 1640 supplemented with 5% FCS and 1% each of nonessential amino acids, glutamine, and pyruvate) in addition to geneticin (Invitrogen Life Technologies) for P13.4 cells. Cells transfected with pNEBR-R1 were supplemented with geneticin (400 µg/ml); transfectants with pNEBR-X1Hygro 3C5 also required the addition of 400 µg/ml hygromycin B (Invitrogen Life Technologies). Purified recombinant β-gal was purchased from Calbiochem and the polyclonal antiserum against it from Oncogene. LPS was removed from β-gal by incubation of solubilized protein with endotoxin-removal Sepharose beads (Pierce) for 12 h, followed by fresh beads for an extra hour. Beads were removed by centrifugation and the LPS content was subsequently determined to be lower than can be detected (0.01 endotoxin units) by the Limulus amebocyte lysate assay (BioWhittaker). Anti-gp96 Ab used for immunoblotting was purchased from StressGen. 5-Bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-GAL) was used at a final concentration of 2.5 mM in the β-gal detection assay. Enzymatic digestion of X-GAL was monitored by measuring absorbance at 595 nm.

Immunization and measurement of CTL activity

Mice were immunized s.c. twice (1 wk apart) in the nape of the neck. One week later, spleens were removed and splenocytes were incubated in complete medium and relevant peptide as indicated for 1 wk. On day 5 after one restimulation, cytotoxicity was measured in a chromium release assay as described. Briefly, peptide-pulsed or unpulsed target cells were loaded with 51Cr and incubated with titrated numbers of T cells. Amount of 51Cr released was measured and percent specific release was calculated. Percent lysis of un-pulsed cells (<2% in all mice) was subtracted from percent lysis of peptide-pulsed cells.

Preparative gel electrophoresis

Proteins were applied to a preparative gel (Bio-Rad) under denaturing conditions. Proteins were resolved at 100 V and eluted into 3-ml fractions under conditions specified by the manufacturer.

Anion exchange chromatography

Proteins were applied to a HiPrep 16/10 DEAE FF column (Amersham Biosciences) attached to a System Gold HPLC (Beckman Coulter) and used according to the manufacturer’s recommendations. Proteins were eluted over increasing salt concentrations from 200 to 800 mM NaCl.

Plasmids

The plasmid pNEBR-R1 is available from New England Biolabs (NEB) as part of their Rheoswitch Mammalian Inducible Protein Expression System. The KOVAK plasmid was a gift from N. Shastri (25). The coding region for the KOVAK protein (beginning at the PvuII site) was amplified via PCR using a forward primer 5'-CACCATGGCTGCAGATCAAGCCAGAGA-3' and reverse primer 5'-AAAAGGGGAAACACATCTGCCAAAG-3' and inserted into the directional TOPO cloning vector pBAD202-DTOPO (Invitrogen Life Technologies). This construct was verified to produce the protein of interest, fused to the V5 epitope and 6x His tag via sequencing and expression studies in Escherichia coli. The KOVAK protein was subcloned from this vector using the forward primer 5'-CTCGAGGAATGGCTGCAGATCAAGCCAGAGAGCTC-3' and reverse primer 5'-GGATCCGGCGCGCCTCAATGGTGATGGTGATGATGACCGGT-3' and ligated into pAcGFP-C1 (BD Clontech) using the XhoI and BamHI unique restriction enzyme sites. The entire fusion coding sequence was digested directly out of this vector using the existing NheI and introduced AscI sites, and this was ligated into pNEBR-X1Hygro (New England Biolabs) using the same enzyme sites to create pNEBR-X1HGK 3C5.

Transfection and cloning

P815 cells growing exponentially were diluted to 2 x 105/ml in 5 ml and grown overnight. The next day cells were transfected with 200 ng/ml pNEBR-R1 in a 1 µg:3 µl liposomal context with FuGene 6 (Roche). After 6 h, the cells were washed and cloned by limiting dilution in selected antibiotic-containing medium. Effective, stably transfected clones were verified via transient transfection with a luciferase reporter construct (NEB). P815-R23 was a clone that exhibited good induction and titratability for luciferase and was diluted 1 day before secondary transfection as described above. P815-R23 cells were then transfected with pNEBR-XIHGK 3C5 (200 ng/ml, 1 µg:3 µl DNA:FuGene 6) and 6 h later they were washed and supplemented with 400 µg/ml hygromycin B. Stable transfectants were induced with 500 nM RSL1 overnight and sorted for high GFP-KOVAK expression using a FACSDiva (BD Biosciences). Cells exhibited good protein induction and dose-response kinetics to RSL1 and are subsequently referred to as PIK23.

Depletion of gp96 or β-gal

Protein G beads (Sigma-Aldrich) were blocked with albumin, saturated with the relevant Ab for the protein to be depleted, and washed to remove excess Ab. A rabbit anti-gp96 antiserum (courtesy of Antigenics) was used for depletion of gp96. The antiserum against β-gal is described in a previous section. Ab-coated protein G beads were placed in protein sample and incubated for 8 h and the beads were removed by low-speed centrifugation. The process was repeated with fresh Ab-coated beads and depletion was monitored by immunoblotting of the protein sample with the relevant Ab. Typically, three sequential incubations with Ab-coated beads were sufficient for complete depletion.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Immunogenicity of apparently homogenous gp96 preparations

gp96 was purified from the β-gal-expressing P13.4 and parental P815 cells and analyzed for purity and identity (Fig. 1A). The major 96-kDa band on SDS-PAGE from both cell types reacts with the anti-gp96 mAb 9G10. gp96 constitutes nearly all detectable protein on SDS-PAGE in the preparations from both cell types. However, protein bands constituting <50 ng, the limit of detection by silver staining in our hands, may remain undetected. Hence, gp96 preparations were immunoblotted with a polyclonal antiserum to β-gal to detect contaminating β-gal. The choice of a polyclonal antiserum was made so that the largest possible repertoire of β-gal and its derivative fragments could be detected. No β-gal was detected in either of the gp96 preparations, even though control β-gal was easily detected (Fig. 1A). (One nanogram of β-gal can be easily detected under these conditions (data not shown).) To obtain another measure of contaminating β-gal, the gp96 preparations were tested for enzymatic activity of β-gal as described in Materials and Methods. Titrated quantities of commercially obtained β-gal were assayed in parallel in the same experiment. Under experimental conditions where as little as 5 femtograms (fg) of β-gal could be detected enzymatically, no enzymatic activity was detected in gp96 preparations from P13.4 or from the parental P815 cell line (Fig 1B). These observations indicate that the gp96 preparation applied to the SDS-PAGE (5 µg absolute quantity) contained <50 ng of intact β-gal or a fragment thereof (based on silver staining), or <1 ng of intact β-gal or a fragment thereof (based on immunoblotting), or <5 fg of enzymatically active β-gal (based on enzymatic assay). Since immunogenic contaminants need not be intact, nor enzymatically active, and as any number of fragments below the level of detection may still be present, this apparently homogenous gp96 preparation may still potentially contain immunogenic β-gal fragments.


Figure 1
View larger version (35K):
[in this window]
[in a new window]

  		FIGURE 1. Immunogenic gp96 preparations did not contain intact β-gal or its large fragments. A, gp96 preparations from P13.4 or P815 (5 µg of each) were analyzed by SDS-PAGE followed by silver staining or immunoblotting. Purified β-gal (5 µg) was used as a control. B, Titrated quantities of β-gal were incubated with its substrate X-GAL for 8 h. The absorbance at 595 was measured. This assay showed a minimum sensitivity detected of 0.05 pg of β-gal. gp96 (50 µg) purified from β-gal-expressing P13.4 cells or parental P815 cells was incubated with X-GAL under the same conditions. C, Titrated quantities (1 or 10 µg) of P13.4- or P815-derived gp96 preparations (as in A) were used to immunize naive C57BL/6 mice. Control mice were immunized with purified β-gal (1 or 10 µg) or PBS. After 1 wk, spleen cells were cultured with DAPIYTNV peptide for 1 wk, restimulated, and tested for cytotoxicity 5 days later. Each line represents the specific (peptide-pulsed minus unpulsed) lysis of spleen cells from one mouse.

 
The gp96 preparation shown in Fig. 1A was tested for immunogenicity. C57BL/6 mice were immunized with PBS, or 1 or 10 µg of gp96, purified from P13.4 or P815 cells or with 1 or 10 µg of β-gal as a positive control. Mice immunized with 10 µg of P13.4-derived, but not P815-derived gp96 generated T cells specific for the Kb-restricted β-gal epitope DAPIYTNV as measured in a CTL assay (Fig. 1C). Mice immunized with 10 µg of β-gal also showed a CTL response. No CTL were detected in mice immunized with 1 µg of gp96 or β-gal. Intermediate doses were not tested. The requirement of ~10 µg of gp96 for specific immunogenicity of gp96 is consistent with previous studies (3, 26). With respect to β-gal, although the experiments in Fig. 1C do not establish the minimal quantity of free β-gal required for immunogenicity, they do establish that 1 µg is not sufficient, a point that is significant in future considerations (see Discussion).

Immunogenicity of gp96 preparations derived from 96-kDa proteins

To assess the contribution of enzymatically inactive intact β-gal and of β-gal fragments smaller than the intact β-gal, gp96 preparations from Fig. 1 were subjected to preparative SDS-PAGE. Previous studies have demonstrated that gp96, hsp90, and hsp70 preparations subjected to SDS-PAGE retain their immunogenicity (3, 19, 27). The SDS-PAGE fractions were tested for the presence of gp96 and β-gal. In parallel, purified β-gal preparations were also subjected to preparative SDS-PAGE and the fractions were tested for the presence of gp96 and β-gal. gp96 molecules eluted in fractions 45–51 and β-gal molecules eluted as a sharp peak (band) in fraction 114. The two preparations were analyzed by immunoblotting (Fig. 2A): 96-kDa bands in fractions 45–51 (of the gp96 preparation) were detected by the anti-gp96 Ab 9G10, but the polyclonal Ab to β-gal did not detect any bands in fraction 114 of the gp96 preparation. Conversely, fraction 114 of the β-gal preparation probed positively with the anti-β-gal antiserum, but fractions 45–51 of this preparation were not detected by the anti-gp96 Ab 9G10 (Fig. 2A).


Figure 2
View larger version (26K):
[in this window]
[in a new window]

  		FIGURE 2. The immunogenic entity in gp96 preparations was of the 96-kDa size. A, gp96 or β-gal was loaded individually onto preparative SDS-polyacrylamide gels and fractions were collected according to size. gp96 and β-gal eluted in fractions 45–51 and 114, respectively, as detected by immunoblotting. B, Fractions 45–51 and 114 from either preparation were used to immunize mice. One week after immunization, spleen cells were removed and cultured with peptide for 1 wk. Cytotoxicity was tested 5 days thereafter as in Fig. 1. Each line represents the specific (peptide-pulsed minus un-pulsed) lysis of spleen cells from one mouse.

 
Fractions 45–51 and, separately, fraction 114 of the gp96 and β-gal preparations were used to immunize C57BL/6 mice. Fractions 45–51 but not fraction 114 of the gp96 preparation elicited a clear CTL response to β-gal epitope DAPIYTNV (Fig. 2B), and the magnitude of the response was comparable to that seen with the corresponding amount of gp96 preparation before it underwent fractionation on preparative SDS-PAGE (Fig. 1C). As a positive control, fraction 114 from the β-gal preparation also elicited a comparable CTL response. These results show that the Kb/DAPIYTNV-specific immunogenicity of gp96 preparations purified from P13.4 cells derives from ~96-kDa proteins, including gp96 and theoretically ~96-kDa β-gal fragments, but not from contaminating intact β-gal nor from β-gal fragments very different in size from ~96 kDa.

In a separate approach to address the contribution of putative β-gal contaminants to the immunogenicity of gp96 preparations, the gp96 preparations were repurified over anion exchange columns. Proteins were eluted over increasing salt gradients (Fig. 3A). Elution profiles of purified gp96 and β-gal did not overlap. gp96 eluted in fraction 30 while β-gal eluted in fraction 24 as analyzed by SDS-PAGE and immunoblotting with anti-gp96 and anti-β-gal Abs (Fig. 3, A and B). Importantly, no β-gal was detected in any of the fractions from the gp96 preparation. C57BL/6 mice were then immunized with fractions 24 or 30 from the gp96 or β-gal preparations. CTL responses against Kb/DAPIYTNV were analyzed as in Figs. 1 and 2. Although mice immunized with fraction 30 of the gp96 preparation gave a robust CTL response, no response was detected in mice immunized with fraction 24 (where contaminating β-gal would have eluted) (Fig. 3C). Conversely, fraction 24 but not fraction 30 of the β-gal preparation provided a CTL response as expected. These results demonstrate that the immunogenic entity in gp96 preparations derived from gp96 or contaminating β-gal fragments with the same ionic charge as gp96. Along with results shown in Fig. 2, the immunogenic entity in P13.4-gp96 must be either gp96- or a β-gal-derived contaminant with the same size and charge as gp96.


Figure 3
View larger version (38K):
[in this window]
[in a new window]

  		FIGURE 3. Repurified gp96 preparations retained immunogenicity. A, Purified P13.4-gp96 or purified β-gal was repurified over an ion exchange column with an increasing salt gradient from 200 to 800 mM. gp96 eluted in fraction 30 distinct from β-gal in fraction 24. B, Elution peaks for gp96 or β-gal were analyzed by SDS-PAGE and immunoblotting. gp96 and β-gal did not cross-contaminate each other. C, C57BL/6 mice were immunized with fractions 30 or 24 from the final repurification step (as in A) of P13.4-gp96 or β-gal or with PBS. CTL activity was tested on spleen cells as in Fig. 1.

 
Immunodepletion of the active entity in gp96 preparations

To probe further the relative contributions of the gp96 band and undetectable but putative contaminating β-gal fragments of ~96-kDa size and the same charge as gp96, P13.4-derived gp96 preparation was immunodepleted of gp96 or of any β-gal-derived fragments of ~96-kDa size that may be reactive to the polyclonal antiserum to β-gal. It is to be noted that a 96-kDa fragment of β-gal shall constitute ~90% of the total length of β-gal and the polyclonal antiserum against β-gal used here recognizes a broad array of fragments of β-gal artificially generated by limited proteolysis (data not shown).

Depletion of the gp96 preparation with an antiserum to gp96 led to a complete absence of detectable gp96 in this preparation (Fig. 4A). Depletion of this same preparation by antiserum to β-gal did not cause depletion of gp96. The ability of the anti- β-gal antiserum to deplete a protein sample of β-gal was shown by a parallel experiment where a β-gal-containing sample was depleted with this antiserum and as a negative control with the anti-gp96 antiserum. The anti-β-gal antiserum depleted this sample of β-gal effectively, whereas the anti-gp96 antiserum did not. Thus, each antiserum depleted the cognate protein effectively and specifically (Fig. 4A). The purpose of this experiment was to create gp96 preparations that were depleted of gp96 or any contaminating β-gal fragments and to test their immunogenicity. If the Kb/DAPIYTNV-specific immunogenicity of gp96 preparations purified from P13.4 cells derives from β-gal fragments, effective depletion of gp96 preparations of such fragments should abrogate their immunogenicity. Additionally, depletion of gp96 should have no effect on the immunogenicity of the gp96 preparations. The converse arguments also apply.


Figure 4
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 4. Depletion of hypothetical contaminants did not deplete immunogenicity from gp96 preparations. A, Purified P13.4-gp96 preparations or β-gal were depleted with Abs to each protein. Depletions were monitored by SDS-PAGE and immunoblotting and were specific and complete. B, gp96 or β-gal, undepleted or depleted with the indicated Abs, were used to immunize mice. One week after immunization, spleen cells were removed and cultured with peptide for 1 wk. Cytotoxicity was tested 5 days thereafter as in Fig. 1. Each line represents the specific (peptide-pulsed minus unpulsed) lysis of spleen cells from one mouse.

 
Mice were immunized with preparations of gp96 depleted of gp96 or β-gal and the splenocytes of the immunized mice were tested for CTL activity as described in Figs. 1 and 2. gp96-depleted samples of the gp96 preparation completely lost their immunogenicity, while depletion of β-gal had no influence on the specific immunogenicity (Fig. 4B). The β-gal preparations depleted of β-gal completely lost their immunogenicity while the same preparations theoretically depleted of gp96 had no influence.

gp96 does not act as an adjuvant to proteins mixed with it

gp96 provides adjuvanticity to the peptides that it chaperones (27). An argument has been made, without any accompanying evidence, that gp96 is such an excellent adjuvant for any proteins present in the same solution as gp96 and that this adjuvanticity of gp96 is responsible for eliciting immune response to "contaminating proteins" (21, 22, 23, 24). The corresponding evidence with respect to peptides is clear; immunization with mixtures of gp96 and peptides does not elicit immunity specific for the peptide. The peptide must be noncovalently associated with it (27). Nonetheless, the premise that gp96 provides adjuvanticity to subimmunogenic quantities of intact proteins was tested experimentally. Mice were immunized with titrated quantities of β-gal, including subimmunogenic doses, mixed with gp96 purified from P815 cells. Splenocytes were tested for CTL activity as in Figs. 1–3. As shown in Fig. 1, 10 µg of β-gal is immunogenic while 1 µg is not. Mice were therefore immunized with 1 µg of β-gal mixed with 10 µg of gp96 as an experimental adjuvant. The quantity of gp96 was chosen as 10 µg because that quantity of gp96 purified from P13.4 cells was sufficient to elicit Kb/DAPIYTNV-specific immunogenicity. Mice were also immunized with 10 µg of β-gal as a positive control. The results show that although all positive and negative controls behave as expected, mixing 10 µg of gp96 with 1 µg of subimmunogenic doses of β-gal does not confer immunogenicity (Fig. 5).


Figure 5
View larger version (9K):
[in this window]
[in a new window]

  		FIGURE 5. gp96 does not augment immune responses of suboptimal quantities of whole β-gal. Mice were immunized with 10 µg of P815-derived gp96 alone or mixed with 1 µg of β-gal protein. Spleens from immunized mice were cultured and tested in cytotoxicity assays for anti-β-gal responses as in Fig. 1. Control mice were immunized with 10 µg of P13.4-derived gp96 or 10 µg of β-gal.

 
Detection of MHC class I (MHC I) precursor peptides chaperoned by gp96

Since immunogenicity of gp96 preparations was determined to be associated with the gp96 molecule itself and not with contaminating Ag (as shown above), the presence of gp96-associated peptides was tested. Peptides were stripped off P13.4-derived gp96 as described in Materials and Methods and analyzed by mass spectrometry. Although a single 19-mer peptide (EVASDTPHPARIGLNCQLA) was found and fragmented by tandem mass spectrometry, we were not sufficiently confident in assigning the sequence due to low accuracy of the precursor ion measurement (A. Tomlinson, unpublished data). Additionally, although major fragmentation ions were consistent with the 19-mer sequence, overall signal intensity was low and current efforts are underway to compare the sequence information collected for this peptide to synthetic analogs to increase confidence and verify this result.

Detection of MHC I epitope precursors (as opposed to precise epitopes) has traditionally been difficult because of the variability inherent in the precursors, and with rare exceptions (9) has not been made. Shastri and colleagues (25) devised an experimental system, precisely to address this question, using the OVA model. They modified the SIINFEKL epitope of OVA to SIINFEHL and flanked each end of this sequence by a lysine to generate trypsin and carboxypeptidase B cleavage sites on the termini of SIINFEHL. The OVA gene with these mutations was called KOVAK. A peptide pool obtained from KOVAK-expressing cells and treated with a combination of trypsin and carboxypeptidase B liberates the SIINFEHL epitope from all of the precursor peptides that might contain it (Fig. 6A). The SIINFEHL epitope, which is presented by H-2Kb, can be measured in a T cell-based assay.


Figure 6
View larger version (10K):
[in this window]
[in a new window]

  		FIGURE 6. MHC precursor peptides could be eluted from purified gp96 preparations. A, Schematic design for the KOVAK-GFP construct for detection of MHC I peptide precursor (based on Ref. 25 ). B, PIK23 cells were induced with RSL1 (solid line) and were observed expressing the KOVAK-GFP protein. The dashed line shows lack of KOVAK expression in uninduced cells. C, One milligram of gp96, purified from uninduced or RSL1-induced PIK23 cells, was stripped off associated peptides as described in Materials and Methods. Eluted peptides (<10 kDa) were treated with trypsin and carboxypeptidase to release SIINFEHL and pulsed onto EL-4 cells and used as targets in a CTL assay. Peptides from induced but not uninduced PIK23-derived gp96 were able to charge EL.4 cells. Synthetic SIINFEKL peptide (1 µM) was pulsed onto EL.4 cells as a positive control. (The T cells recognize equally SIINFEKL- or SIINFEHL-pulsed cells.) There was no lysis of unpulsed EL.4 cells.

 
We have created PIK23 cells (H-2d) that contain an RSL1-inducible KOVAK construct fused to GFP. Induction of KOVAK by RSL1 was monitored by GFP expression as described in Materials and Methods (Fig. 6B). PIK23 cells induced with RSL1 or uninduced were used as sources of purified gp96 of the same purities as used in Figs. 1–5. gp96 was purified to apparent homogeneity by the same procedure as used for P13.4 cells. gp96 preparations were immunoblotted with polyclonal anti-GFP Ab to detect GFP-KOVAK or its fragments. No Ag was detected (data not shown) in a manner analogous to the data shown in Fig. 2 for the β-gal system. Peptides were stripped off gp96 and treated with a combination of trypsin and carboxypeptidase B. The resulting peptides were pulsed onto Kb-expressing EL.4 cells and tested for lysis by a T cell line specific for SIINFE(H/K)L. Peptides thus obtained from RSL1-induced, PIK23-derived gp96 were able to charge target cells for lysis while peptides obtained from control uninduced, PIK23-derived gp96 did not (Fig. 6C). This result shows the association of SIINFEHL peptides with gp96 and that these peptides include the direct MHC I binding peptide as well as the precursor peptides.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Our results demonstrate that 1) apparently homogenous gp96 preparations from P13.4 cells elicit Kb/DAPIYTNV-specific CTL response, and this activity resides in proteins of ~96 kDa size; 2) if these preparations contain contaminating β-gal, its content is <5 pg/mg of gp96, and <50 fg per immunizing dose of gp96 preparations; 3) the specific immunogenicity of the gp96 preparation cannot be attributed to contaminating intact β-gal nor to any fragments of β-gal that may react with a polyclonal antiserum against β-gal; 4) depletion of gp96 preparations of gp96 species depletes the specific immunogenicity of the gp96 preparation; however, depletion of these same preparations of any theoretically contaminating (but undetectable) β-gal has no influence on the specific immunogenicity of the gp96 preparation; 5) gp96 does not confer detectable adjuvanticity to subimmunogenic doses of β-gal; and 6) MHC I-binding peptides (and their precursors) can be eluted off gp96 molecules. These observations demonstrate that the Kb/DAPIYTNV-specific immunogenicity of gp96 preparations from P13.4 cells does not derive from contaminating β-gal. Conversely, the hypothesis that gp96 (and other HSPs) copurify with trace but undetectable quantities of intact proteins and that these contaminants are responsible for the specific immunogenicity of HSP preparations (21, 22, 23, 24) cannot be substantiated.

The possibility that gp96 molecules are not associated with endogenous peptides was first raised from experiments that showed that truncated gp96 molecules lacking the carboxyl terminus (where the first putative peptide binding site was previously reported to be (20)) could still confer protective immunity (21). While this article was under publication, a second peptide binding site of gp96 was identified on the amino terminus of gp96 (19). The presence of this second, N-terminally located peptide binding site invalidated the conclusion that the immunogenicity of gp96 is peptide independent. Furthermore, the immunogenicity of gp96 shown in that study was extremely weak and comparable to the nonspecific innate immune activity of gp96 as published by us previously (28, 29). Thus, the single study that purported to show evidence of the peptide independence of immunogenicity of gp96 could not be upheld after the identification of this second peptide binding site.

An extensive biochemical analysis of this scale has not been performed with other Ags; nonetheless, results of solid studies in the OVA system are consistent with the results shown here in the β-gal system. Thus, Berwin et al. (30) showed that gp96 constituted the immunogenic component of virally lysed cells. These authors showed through careful quantitative analyses that OVA-depleted but gp96-enriched fractions of the lysates contained the major immunogenic activity; conversely, gp96-depleted lysates had little activity. Binder and Srivastava (16) showed that cell lysates replete with OVA Ag but depleted of HSPs were devoid of immunogenic activity; a previous study (31) that had come to an opposite conclusion was shown to have done so because of dose restriction of the immunogenicity of cell lysates (16). Studies of Shastri and colleagues (25) have provided a definitive and independent verification of the original idea (4, 5) that selected HSPs chaperone precursors of MHC I epitopes in vivo. Using the KOVAK system described in Results (25), they observed that the chaperones TriC (of the hsp60 family) and hsp90{alpha} associate with precursors of the Kb-binding SIINFEKL peptide, and that extinction of expression of these chaperones through genetic means, leads to generation of empty MHC I molecules (32, 33). Collectively, these results argue unequivocally in favor of the observations reported in the present study.

The HSPs are the only molecules other than the MHC, to be peptide-binding proteins of immunological significance. The increasingly clear demonstration that the HSP-peptide complexes play a critical and essential role in Ag presentation (9, 32, 33, 34, 35) and in cross-priming (7, 12, 13, 15, 16) have now begun to reveal that the two peptide-binding activities are in fact functionally related.


   Acknowledgments
 
We thank Drs. Adam Adler and Robert Clark for critically reading this manuscript. We thank Dr. A. Tomlinson (Antigenics) for his role in the identification of peptide epitopes eluted from gp96.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
P.K.S. has a significant financial interest in Antigenics.


   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 This work was supported by National Institutes of Health Grant CA84479-07 and a sponsored research agreement with Antigenics Incorporated. Back

2 Address correspondence and reprint requests to Dr. Robert J. Binder, BSTWR E1051, 200 Lothrop Street, University of Pittsburgh, Pittsburgh, PA 15261. E-mail address: rjb42{at}pitt.edu Back

3 Abbreviations used in this paper: HSP, heat shock protein; β-gal, β-galactosidase; X-GAL, 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside; fg, femtogram; MHC I, MHC class I. Back

Received for publication April 16, 2007. Accepted for publication September 17, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Srivastava, P.. 2002. Interaction of heat shock proteins with peptides and antigen presenting cells: chaperoning of the innate and adaptive immune responses. Annu. Rev. Immunol. 20: 395-425. [Medline]
  2. Srivastava, P. K., M. R. Das. 1984. The serologically unique cell surface antigen of Zajdela ascitic hepatoma is also its tumor-associated transplantation antigen. Int. J. Cancer. 33: 417-422. [Medline]
  3. 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]
  4. Srivastava, P. K., M. Heike. 1991. Tumor-specific immunogenicity of stress-induced proteins: convergence of two evolutionary pathways of antigen presentation?. Semin. Immunol. 3: 57-64. [Medline]
  5. Srivastava, P. K., R. G. Maki. 1991. Stress-induced proteins in immune response to cancer. Curr. Top. Microbiol. Immunol. 167: 109-123. [Medline]
  6. Nieland, T. J., M. C. Tan, M. Monne-van Muijen, F. Koning, A. M. Kruisbeek, G. M. van Bleek. 1996. Isolation of an immunodominant viral peptide that is endogenously bound to the stress protein GP96/GRP94. Proc. Natl. Acad. Sci. USA 93: 6135-6139. [Abstract/Free Full Text]
  7. Suto, R., P. K. Srivastava. 1995. A mechanism for the specific immunogenicity of heat shock protein-chaperoned peptides. Science 269: 1585-1588. [Abstract/Free Full Text]
  8. 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]
  9. Ishii, T., H. Udono, T. Yamano, H. Ohta, A. Uenaka, T. Ono, A. Hizuta, N. Tanaka, P. K. Srivastava, E. Nakayama. 1999. Isolation of MHC class I-restricted tumor antigen peptide and its precursors associated with heat shock proteins hsp70, hsp90, and gp96. J. Immunol. 162: 1303-1309. [Abstract/Free Full Text]
  10. Zugel, U., A. M. Sponaas, J. Neckermann, B. Schoel, S. H. Kaufmann. 2001. 96-peptide vaccination of mice against intracellular bacteria. Infect. Immun. 69: 4164-4167. [Abstract/Free Full Text]
  11. Breloer, M., T. Marti, B. Fleischer, A. von Bonin. 1998. Isolation of processed, H-2Kb-binding ovalbumin-derived peptides associated with the stress proteins HSP70 and gp96. Eur. J. Immunol. 28: 1016-1021. [Medline]
  12. 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]
  13. Castelli, C., A. M. Ciupitu, F. Rini, L. Rivoltini, A. Mazzocchi, R. Kiessling, G. Parmiani. 2001. Human heat shock protein 70 peptide complexes specifically activate antimelanoma T cells. Cancer Res. 61: 222-227. [Abstract/Free Full Text]
  14. Parmiani, G., A. De Filippo, L. Pilla, C. Castelli, L. Rivoltini, L. . 2006. Heat shock proteins gp96 as immunogens in cancer patients. Int. J. Hyperthermia 22: 223-227. [Medline]
  15. Noessner, E., R. Gastpar, V. Milani, A. Brandl, P. J. Hutzler, M. C. Kuppner, M. Roos, E. Kremmer, A. Asea, S. K. Calderwood, R. D. Issels. 2002. Tumor-derived heat shock protein 70 peptide complexes are cross-presented by human dendritic cells. J. Immunol. 169: 5424-5432. [Abstract/Free Full Text]
  16. 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: 593-999. [Medline]
  17. Zhu, X., X. Zhao, W. F. Burkholder, A. Gragerov, C. M. Ogata, M. E. Gottesman, W. A. Hendrickson. 1996. Structural analysis of substrate binding by the molecular chaperone DnaK. Science 272: 1606-1614. [Abstract]
  18. 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]
  19. Vogen, S., T. Gidalevitz, C. Biswas, B. B. Simen, E. Stein, F. Gulmen, Y. Argon. 2002. Radicicol-sensitive peptide binding to the N-terminal portion of GRP94. J. Biol. Chem. 277: 40742-40750. [Abstract/Free Full Text]
  20. Linderoth, N. A., A. Popowicz, S. Sastry. 2000. Identification of the peptide-binding site in the heat shock chaperone/tumor rejection antigen gp96 (Grp94). J. Biol. Chem. 275: 5472-5477. [Abstract/Free Full Text]
  21. Baker-LePain, J. C., M. Sarzotti, T. A. Fields, C. Y. Li, C. V. Nicchitta. 2002. GRP94 (gp96) and GRP94 N-terminal geldanamycin binding domain elicit tissue nonrestricted tumor suppression. J. Exp. Med. 196: 1447-1459. [Abstract/Free Full Text]
  22. Baker-LePain, J. C., R. C. Reed, C. V. Nicchitta. 2003. ISO: a critical evaluation of the role of peptides in heat shock/chaperone protein-mediated tumor rejection. Curr. Opin. Immunol. 15: 89-94. [Medline]
  23. Nicchitta, C. V., D. M. Carrick, J. C. Baker-Lepain. 2004. The messenger and the message: gp96 (GRP94)-peptide interactions in cellular immunity. Cell Stress Chaperones 9: 325-331. [Medline]
  24. Yewdell, J. W.. 2005. The seven dirty little secrets of major histocompatibility complex class I antigen processing. Immunol. Rev. 207: 8-18. [Medline]
  25. Campbell, D. J., T. Serwold, N. Shastri. 2000. Bacterial proteins can be processed by macrophages in a transporter associated with antigen processing-independent, cysteine protease-dependent manner for presentation by MHC class I molecules. J. Immunol. 164: 168-175. [Abstract/Free Full Text]
  26. Chandawarkar, R. Y., M. S. Wagh, P. K. Srivastava. 1999. The dual nature of specific immunological activity of tumor-derived gp96 preparations. J. Exp. Med. 189: 1437-1442. [Abstract/Free Full Text]
  27. 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]
  28. Udono, H., P. K. Srivastava. 1994. Comparison of tumor-specific immunogenicities of stress-induced proteins gp96, hsp90, and hsp70. J. Immunol. 152: 5398-5403. [Abstract]
  29. Tamura, Y., P. Peng, K. Liu, M. Daou, P. K. Srivastava. 1997. Immunotherapy of tumors with autologous tumor-derived heat shock protein preparations. Science 278: 117-120. [Abstract/Free Full Text]
  30. Berwin, B., R. C. Reed, C. V. Nicchitta. 2001. Virally induced lytic cell death elicits the release of immunogenic GRP94/gp96. J. Biol. Chem. 276: 21083-21088. [Abstract/Free Full Text]
  31. Shen, L., K. L. Rock. 2004. Cellular protein is the source of cross-priming antigen in vivo. Proc. Natl. Acad. Sci. USA 101: 3035-3040. [Abstract/Free Full Text]
  32. Kunisawa, J., N. Shastri. 2003. The group II chaperonin TRiC protects proteolytic intermediates from degradation in the MHC class I antigen processing pathway. Mol. Cell 12: 565-576. [Medline]
  33. Kunisawa, J., N. Shastri. 2006. Hsp90{alpha} chaperones large C-terminally extended proteolytic intermediates in the MHC class I antigen processing pathway. Immunity 24: 523-534. [Medline]
  34. Srivastava, P. K., H. Udono, N. E. Blachere, Z. Li. 1994. Heat shock proteins transfer peptides during antigen processing and CTL priming. Immunogenetics 39: 93-98. [Medline]
  35. Binder, R. J., N. E. Blachere., P. K. Srivastava. 2001. Heat shock protein-chaperoned peptides but not free peptides introduced into the cytosol are presented efficiently by major histocompatibility complex I molecules. J. Biol. Chem. 276: 17163-17171. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Leukoc. Biol.Home page
M.-F. Tsan and B. Gao
Heat shock proteins and immune system
J. Leukoc. Biol., June 1, 2009; 85(6): 905 - 910.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
M. K. Callahan, M. Garg, and P. K. Srivastava
Heat-shock protein 90 associates with N-terminal extended peptides and is required for direct and indirect antigen presentation
PNAS, February 5, 2008; 105(5): 1662 - 1667.
[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 Binder, R. J.
Right arrow Articles by Srivastava, P. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Binder, R. J.
Right arrow Articles by Srivastava, P. K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS