Skip Navigation


International Immunology Advance Access originally published online on June 13, 2006
International Immunology 2006 18(7):1147-1157; doi:10.1093/intimm/dxl049
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/7/1147    most recent
dxl049v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Biswas, C.
Right arrow Articles by Argon, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Biswas, C.
Right arrow Articles by Argon, Y.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?


© The Japanese Society for Immunology. 2006. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

The N-terminal fragment of GRP94 is sufficient for peptide presentation via professional antigen-presenting cells

Chhanda Biswas1, Uma Sriram2, Bogoljub Ciric1, Olga Ostrovsky1, Stefania Gallucci2 and Yair Argon1

1 Department of Pathology and Laboratory Medicine, Children's Hospital of Philadelphia and University of Pennsylvania, Philadelphia, PA 19104, USA
2 Department of Pediatrics, Children's Hospital of Philadelphia and University of Pennsylvania, Philadelphia, PA 19104, USA

Correspondence to: Y. Argon; E-mail: yargon{at}mail.med.upenn.edu


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 References
 
The chaperone glucose-regulated protein 94 (GRP94) has long been used to augment peptide presentation to T cells. This chaperone binds antigenic peptides, binds to receptors on professional antigen-presenting cells (APCs), activates these cells and after internalization, transfers the peptides to MHC class I for activation of T cells. Here we show that all these activities reside within amino acids 1–355 of GRP94. This small fragment is sufficient to bind peptides, to bind and be taken up by the receptors CD91 and scavenger receptor type A on either dendritic cells or macrophages. The minimal construct can augment peptide presentation in culture and induce antigen-specific CTL in naive mice only because it loads APCs with the relevant peptide. Thus, the sequence 1–355 is the immunologically sufficient module of GRP94 and we propose that this ‘mini-chaperone’ can be used in immunotherapy of tumors and vaccine development.

Keywords: antigen presentation, cross presentation, dendritic cells, macrophages, molecular chaperones, peptide binding


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 References
 
The stress protein glucose-regulated protein 94 (GRP94), also known as gp96, has long been known for its ability to induce T cell immunity. It can induce MHC class I-restricted CTL responses against tumor antigens (1), viral antigens (2) and cellular proteins (3). This ability to boost CTL responses is generally [though not universally (4, 5)] attributable to the capacity of GRP94 to bind peptides and present them to T cells via a pathway involving professional antigen-presenting cells (APCs). GRP94 can carry a number of peptides (6), bind to at least two receptors on APCs, CD91 and scavenger receptor type A (SR-A) (7–9), and be taken up by receptor-mediated endocytosis (10), followed by transfer of the peptides to MHC class I proteins. In addition, and importantly, GRP94 can activate the APCs to express co-stimulatory molecules and secrete cytokines (11–13) responsible for stimulating both the innate and the adaptive arms of the immune system (14). These features of GRP94's action provide an explanation for its ability to mediate cross-presentation for priming CD8+ T cell responses.

Nonetheless, GRP94 is not unique in this activity: Binder and Srivastava (15) and others achieved cross-presentation with chaperones from the heat shock protein 90 (HSP90) family, to which GRP94 belongs, as well as with bacterial and mammalian HSP70 proteins (16), HSP60s (17) and calreticulin, a lectin-type chaperone (18, 19). Consistent with these observations, neither CD91 nor SR-A are specific receptors for GRP94 and can bind other chaperones as well (8, 9, 16). Nevertheless, GRP94 is considered one of the most potent in chaperoning peptides for cross-presentation (20, 21). Presumably, the potency of GRP94 in cross-presentation is due to the combination of its ability to chaperone peptides into APCs as well as to independently signal their activation.

Despite the body of work demonstrating its existence, the chaperone-mediated mechanism of cross-presentation has been controversial. In particular, the peptide dependence of the stimulating activity of GRP94 has been questioned. Baker-LePain et al. (22) reached the conclusion that the stimulation of tumor rejection by a construct of GRP94, and by implication, the activity of the full-length protein, is peptide non-specific. Li et al. (23) showed that a similar N-terminal construct of GRP94 has adjuvant activity when mixed with an hepatitis B virus (HBV) peptide, though the authors did not attempt to ascertain peptide binding to their construct. Furthermore, a recent mass spectrometry analysis of tissue-derived GRP94 showed it to be relatively poor in associated peptides (5).

Having identified a peptide-binding site in the N-terminal portion of GRP94 (24, 25), we sought to resolve whether this protein serves merely as an adjuvant or as peptide delivery vehicle, because future use of the chaperone in vaccination depends heavily on the answer to this question. Here we report that a truncated GRP94, containing as little as amino acids 1–355, is an immunologically active module that can account for the ability of the full-length chaperone to augment CTL responses, and that it does so only when loaded with the appropriate peptide.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 References
 
Mice
C57Bl/6 and C57Bl/6 RAG-2–/– (H-2b) mice were obtained from Jackson Laboratory (Bar Harbor, ME, USA). The mice were bred in the animal facility of The Children's Hospital of Philadelphia, an AAALAC-accredited facility. Animals were maintained and treated following the guidelines of our Institutional and Animal Care and Use Committee, which reviewed and approved this work.

Cell lines
RAW 264.7 macrophage-like cells and EL4 thymoma cells were obtained from the American Type Tissue Collection and were cultured in RPMI with 10% FCS. RAW 264.7/H-2Kb cells are a stably transfected line expressing the Kb cDNA (gift of H. Schreiber, The University of Chicago). They were selected with G418 and the expression of Kb was quantified by FACS. N1 are EL4 cells transfected with the envelope gene of vesicular stomatitis virus (VSV) (gift of D. Lyles, Wake Forest University). They were cultured in RPMI supplemented with 0.3 mg ml–1 of G418. The T cell hybridoma N15 was obtained from E. Reinherz (Dana-Farber Cancer Institute, Boston, MA, USA) and cultured in RPMI, 10% FCS, 0.1 mg ml–1 G418 and 0.2 mg ml–1 of hygromycin.

Generation of primary dendritic cells
The CD11c+ bone marrow-derived immature dendritic cells (BM–DCs) were propagated from bone marrow (BM) progenitor cells harvested from C57Bl/6 RAG–/– (2–4 months old), as described previously (26). The cells at a concentration of 1 x 106 cells per well were cultured in 24-well plate in Iscove's modified Dulbecco's medium supplemented with 10% FCS, glutamine, beta-mercaptoethanol (ME), 100 IU ml–1 penicillin/streptomycin and 50 µg ml–1 of gentamycin, in the presence of 1–3 ng ml–1 recombinant murine granulocyte monocyte colony-stimulating factor (BD PharMingen, CA, USA) and 2.5 ng ml–1 recombinant IL-4 (BD PharMingen) for 5–8 days renewing the media in every 2 days. Dendritic cells (DCs) derived from BM of C57Bl/6 RAG–/– mice behave identically to those from normal mice, and have the advantage of obviating the need for depletion procedures to remove contaminating T and B cell. DCs were used at day 6–7 of culture, a time when >80% of the cells are CD11c+.

Purification of proteins
GRP94 was purified from mouse liver by the method in (27), which entails affinity chromatography over concanavalin A–Sepharose (Sigma Chemicals, St Louis, MO, USA), followed by ion exchange chromatography over a diethylaminoethyl–Sepharose column (Amersham Biosciences).

The N-terminal portion of GRP94, amino acids 1–355, containing the nucleotide/peptide-binding domain and the first charged domain was purified as per Vogen et al. (24). In brief, the full-length GRP94 was truncated by introducing stop codon into codon 356 by PCR and was expressed in pFastBacHTB baculovirus expression vector (GIBCO/Invitrogen, CA, USA). The expression construct containing a His6 tag at its N-terminus followed by a tobacco etch virus protease cleavage site was expressed in SF9 cells. The soluble recombinant protein was purified on affinity column, Ni-NTA (Amersham Pharmacia Biotech) on AKTA FPLC. The protein was further purified by size exclusion chromatography on a Superdex-75 column (Amersham Pharmacia). Prior to purification, the columns were made LPS free by treating with 70% ethanol and by using endotoxin-free reagents and glasswares. The standard limulus assay (Sigma) for LPS indicated no contamination in the protein preparations and an alternative test, induction of tumor necrosis factor (TNF)-alpha secretion by the macrophage cell line RAW 264.7, also indicated minimal or no LPS in the preparations.

N34–355 was cloned into the pQE30Xa bacterial expression vector (Qiagen) and expressed in Escherichia coli strain M15 by isopropyl-L-thio-ß-galactoside induction. In addition to an N-terminal His6 tag, this construct has a C-terminal birA target sequence (28), which allows the bacterial biotin ligase to attach a biotin molecule in vivo specifically to Lys370. N34–355 was mostly soluble and was purified by Ni-NTA chromatography followed by mono-Q ion exchange chromatography (Amersham Pharmacia Biotech).

Peptides
The peptide VSV8, RGYVYQGL, the antigenic fragment of VSV N protein was synthesized at The University of Chicago facility, purified by HPLC and verified by mass spectrometry. VSV8 was iodinated using IodoBeads (Pierce) and unincorporated iodine was removed by passage over a Dowex AG1X8 beads (Supelco, Bellefonte, PA, USA).

Peptide-binding assay
Peptide binding was carried out according to Vogen et al. (24). Briefly, for routine peptide binding and to determine the binding stoichiometry, full-length GRP94 or N1–355 or N34–355 (3–4 µM each) and 800 µM of 125I-labeled VSV8 in 25 µl of buffer A (HEPES 20 mM, pH 7.2, 20 mM NaCl, 110 mM KOAc, 1 mM MgCl2 and 0.1 mM CaCl2) were heat shocked for 10 min at 50°C followed by incubation at room temperature for 30 min. The complex was separated from unbound peptide on mini spin column (Biorad P10) and resolved by SDS-PAGE. The gel was dried, exposed overnight to a phosphorimager screen and the image developed with a Typhoon imager (Amersham). The radioactive signal was quantified with the program ImageQuant. For binding to DCs, antigen presentation and immunization experiments, the complex was separated from the unbound peptide by size exclusion chromatography on a Superdex-75 column (Pharmacia). The removal of free peptide was essentially complete.

Binding of N1–355 and N1-355–VSV8 complex to APCs
Six- to seven-day old cultures of BM–DCs, or the macrophage cell line RAW 264.7, were incubated either with 10 µg ml–1 of N1–355 or N1-355–VSV8 complex or 10 ng ml–1 of VSV8 on ice for 30 min. The incubations were in the presence or absence of 80 µg ml–1 fucoidan (Sigma), activated alpha-2-macroglobulin ({alpha}2M), equimolar ratio of both or increasing concentrations of full-length GRP94. {alpha}2M (Boehringer Mannheim, Germany) was converted to the CD91-binding, thiol-ester-cleaved activated alpha-2M by incubation with 0.2 M NH4HCO3 (29). The cells were washed with cold PBS, lysed in SDS-reducing sample buffer and proteins resolved on SDS-PAGE and transblotted onto nitrocellulose membranes. The membranes were probed with anti-GRP94, which also recognized the endogenous GRP94 that served as a loading control.

Internalization of N1–355 by DCs
For receptor-mediated internalization experiments, cover slips with adherent DCs were placed in six-well plates on ice, washed 4x with ice-cold RPMI. The cells were incubated with N1-355–VSV8 complex (10 µg protein and 10 ng peptide ml–1), N1–355 or N34–355 (10 µg ml–1) alone or VSV8 (10 ng ml–1) alone, in the presence or absence of fucoidan or {alpha}2M for 2 h on ice. Subsequently, cells were washed three times in PBS, and warmed to 37°C for 30–120 min. Cells were then fixed in 2% PFA for 10 min, washed three times in PBS and permeabilized in 0.25% Triton X-100 for 10 min. Next, the cells were reacted with anti-biotin antibody directed to the biotinylated Lysine 370, and anti-LAMP1 primary antibody. After washes in PBS, the cells were reacted with an FITC-conjugated secondary antibody against the anti-biotin antibody and rhodamine-conjugated secondary antibody against the anti-LAMP1 (Jackson ImmunoResearch Laboratories, Inc.). The cells were washed three times in PBS, and the final wash contained 1 ng ml–1 diamidino-2-phenylindole to stain nuclei. The stained cells were viewed in a Zeiss Axiovert 200M (Germany) fluorescence microscope equipped with the 3I software package.

Peptide presentation
Peptide presentation assays were performed with N1-355–VSV8 complex separated from unbound VSV8 on Superdex-75 size exclusion chromatography. The DCs or RAW 264.7 cells transfected with Kb, each 0.5 x 105 in 100 µl of RPMI, were plated in 96-wells in triplicates. The cells were either untreated or pulsed for 4 h with N1–355, N1-355–VSV8 complex and N1–355 mixed with free peptide. After thorough washing, equal numbers of N15 cells, T cell hybridomas specific for VSV8 (30), were added to the APCs. The supernatants were collected after 24 h and IL-2 in them was determined by ELISA, using a kit from R&D Systems (Minneapolis, MN, USA), as a measure of T cell activation. The fraction from the Superdex-75 column corresponding to the migration of free VSV8 was also used as a control to determine that the complex was not contaminated with free peptide.

Activation of DCs
To determine the ability of N1–355 or the complex to activate DCs, immature DCs (1 x 106 cells per well in a 24-well plate) were incubated for 24 h with 20 µg ml–1 of N1–355, or the complex or 20 ng ml–1 VSV8, or 100 ng ml–1 LPS (E. coli 055:B5, Sigma). The activation was carried out both in the presence and absence of polymyxin B (50 µg ml–1). In some cases, N1–355, the complex and LPS were treated at 95°C for 1 h and then added to the cells.

DC staining and FACS analysis
At the end of the incubations, BM–DCs were harvested, incubated with rat anti-mouse CD16/CD32 (clone 2.4G2) antibody to block Fc{gamma}R and then stained for 30 min with the following mAbs (PharMingen): allophycocyanin-conjugated hamster anti-mouse CD11c and PE-conjugated rat anti-mouse CD80, CD86, and MHC class I and II and FITC-conjugated hamster anti-mouse CD40. In parallel, cells were stained with isotype-matched antibodies to correct for non-specific staining. Cells were analyzed on a FACSCalibur flow cytometer (BD Bioscience, CA, USA) and data analyzed with the CellQuest software. Cells were gated according to size and scatter to eliminate dead cells and debris.

Cytokine detection
The supernatants from the activation experiments (24 h stimulation) were assayed by ELISA for the presence of TNF{alpha} and IL-12p70 using detection kit from R&D Systems.

Immunization of mice
Female C57BL6 (H2b) mice (4–6 months old) were immunized at the base of the tail with either peptide alone (20 ng per mouse) or an emulsified mixture of peptide (20 ng–100 µg per mouse) with CFA or with N1-355–VSV8 complex, containing 20 µg N1–355 and 20 ng VSV8 per mouse. The injection volume was adjusted to 100–150 µl. Subsequently, the mice were given a booster injection 2 weeks later in a similar way only substituting CFA with incomplete Freund's adjuvant. Either 2 weeks or 2 months later, the mice were sacrificed and the spleens were removed. The splenocytes were obtained by dispersing the cells with a syringe plunger following their filtration through cell strainers. Erythrocytes were lysed with 0.83% ammonium chloride lysis solution. The splenocytes were washed and re-suspended at 10 million cells ml–1.

In vitro re-stimulation of the splenocytes
In order to increase the frequency of antigen-specific T cells, splenocytes from immunized mice were re-stimulated in vitro. In brief, five million splenocytes were added per well in 24-well plates layered with two million irradiated syngenic splenocytes as a source of more APCs and the cells were stimulated either with VSV8 (1 µg ml–1) or IL-2 (20 U ml–1) or together with VSV8 and IL-2 or concanavalin A (5 µg ml–1) for 5 days to be tested as effector cells in the cytotoxicity assay.

Measurement of cytotoxicity by 51Cr release
The stimulated splenocytes were washed, re-suspended in RPMI 1640 medium supplemented with antibiotic, L-glutamine plus 10% FCS and 2-ME and used as effector cells (E). The CTL assay of these cells was measured against the VSV8 transfected N1 cells, the target cells (T), which were labeled with 51Cr. The cells were washed three times to remove free 51Cr and subsequently added to the round-bottom 96-well plate at various E:T cell ratios (50:1, 25:1, 12.5:1 and 6:1) in a final volume of 200 µl. After 4 h of incubation, supernatants (100 µl) were transferred to Luma plate-96 (Perkin Elmer) and the released 51Cr was measured by a Top count NXT (Perkin Elmer). Spontaneous release was measured in wells containing target cells alone. Igepal (NP40) was used to lyse the target cells maximally. The percentage of specific lysis was measured by the following formula:

Formula
Lysis of EL4 cells not expressing the VSV8 peptide was negligible.

Statistical analyses
The confidence levels in the significance of data points were ascertained by a two-tailed, paired Student's t-test.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 References
 
Rationale and experimental approaches for assaying peptide-specific immune responses by N1–355 instead of full-length GRP94
When using purified chaperones to elicit immune responses, several aspects need to be considered carefully. First, there are issues of purity and in particular non-protein contaminants that can activate innate immunity pathways. Second, there are issues of multiple biological interactions that chaperones are designed to engage in, which can confound the interpretation of an observed immune response.

Having previously demonstrated that the chaperone GRP94 binds peptides via its N-terminal domain and that the N1–355 fragment accounted for the peptide-binding activity of the full-length protein (24, 25), we used this truncated protein for antigen cross-presentation, because it should be a better vehicle in terms of the above concerns. We expressed N1–355 in the insect cell line SF9 to avoid bacterial contaminants, especially LPS, and to avoid the need to refold the recombinant protein in vitro. This recombinant N1–355 was loaded with VSV8, a known antigenic octamer presented in the context of Kb (30). Because the quality of the protein and the complex with peptide are crucial for data interpretation, we validated its usage by a number of tests, as shown below, to ensure that N1–355 follows the cross-presentation pathway proposed for the full-length, tissue-derived chaperone.

Efficient loading of the protein with antigenic peptide
First, we ascertained and quantified the loading of peptide onto the complex used for the presentation assays. We employed iodinated VSV8 as a tracer, mixed with unlabeled VSV8 to a defined specific radioactivity and allowed it to bind in solution to heat-shocked recombinant N1–355. As others and we have already demonstrated (6, 24), this treatment accelerates the loading of peptide onto the binding site without changing the specificity. Peptide–N1-355 complexes were then resolved by reducing SDS-PAGE and the amount of radioactivity migrating with the protein was quantified to measure the binding (Fig. 1A). The data in Fig. 1 show that the majority of the recombinant protein can be loaded with VSV8, to a similar extent as compared with full-length GRP94 purified from mouse liver by the method in (27) and to an even shorter portion of GRP94, N34–355, expressed in bacteria. Binding of the peptide was inhibited in the presence of radicicol, the fungal metabolite that binds to the nucleotide-binding pocket of GRP94 and inhibits it (31), demonstrating specificity of binding and not merely adsorption of peptide (Fig. 1A and B). The binding stoichiometries of the three proteins, and the molar ratios of bound VSV8 to chaperone were comparable in the range of 0.5–0.7 mole peptide per mole of protein [Fig. 1B; see also (24)]. Given that some unknown fraction of bound peptide might dissociate during the electrophoresis, these ratios show that the majority of N1–355 molecules used in this work to stimulate responses are loaded with the relevant peptide.


Figure 1
View larger version (27K):
[in this window]
[in a new window]
  		Fig. 1 Peptide binding to GRP94 and N1–355. Mouse liver GRP94 and recombinant, insect cell-derived N1–355 or bacterial N34–355 (3 µM each) were incubated with 800 µM 125I-VSV8 in buffer A at 50°C for 10 min and then at room temperature for further 30 min, to form protein–peptide complexes in the presence of 300 µM radicicol (Rad) or the solvent dimethyl sulfoxide (DMSO) alone. Samples were resolved by reducing SDS gel electrophoresis and peptide binding was visualized by phosphorimaging (panel A). Coomassie blue staining of the gels shows equal loads of samples. (B) The radioactivity from the bound peptide was quantified by the program ImageQuant and the molar ratios of peptide and protein were calculated to compare the binding between the two proteins. Data are presented as averages ± SD from three separate experiments.

 
Specific binding of N1–355 to APCs
Having ascertained that N1–355 binds antigenic peptide efficiently, we further validated its use by characterizing the binding of N1–335 to professional APCs. It was previously shown by Baker-LePain et al. (22) that a similar N-terminal domain binds to DCs and activates them. We incubated N1–355 at 4°C with DCs or with a macrophage-like cell line, RAW 264.7, and the cell-associated protein was assayed after washing by western blotting with anti-GRP94. Binding was normalized to the endogenous contents of full-length GRP94 (the recombinant proteins migrate on gels as proteins of 40 kDa, and endogenous GRP94 migrates as an 100-kDa protein; Fig. 2A). Since binding was proportional to the concentration of N1–355 and reached apparent saturation at 50 µg ml–1 (data not shown), we chose to use 10 µg ml–1 of the protein or the protein–peptide complex, a concentration within the linear range. To ask if the binding of N1–355 uses the same receptor previously shown to bind full-length GRP94, CD91 (8) and SR-A (9), we used their respective competitive inhibitors. As shown in Fig. 2, both N-terminal fragments of GRP94 bound to DCs as well as to the macrophage line, and the binding was specific. In each case, binding of the truncated GRP94 versions was sensitive to the presence of activated {alpha}2M, which inhibits CD91 binding, and also to fucoidan, which inhibits SR-A. Either {alpha}2M or fucoidan inhibited the binding. Although {alpha}2M and fucoidan should be specific for each of the two receptors above, in our hands, the inhibition by either was almost complete in three independent experiments (as shown in Fig. 2A–C). In two other experiments (data not shown), the inhibition by each agent was partial and only the combination of {alpha}2M and fucoidan abolished binding of the truncated GRP94 constructs completely.


Figure 2
View larger version (23K):
[in this window]
[in a new window]
  		Fig. 2 N1–355 and N34–355 bind to the CD91 and SR-A receptors on APCs. (A) N1–355 (10 µg ml–1) was incubated with either RAW 264.7 cells (106 cells in 100 µl) (top two panels) or with primary cultured DCs (lower two panels). The cells were incubated either with the protein alone (N1–355) or with the protein loaded with the peptide VSV8 (complex), without inhibitors or in the presence of 8 µM activated {alpha}2M, 80 µg ml–1 fucoidan (Fu) or both ({alpha}2M + Fu). After washing with cold PBS, lysates (50 µg) from each condition were resolved by SDS-PAGE, transblotted and probed with anti-GRP94 (9G10). Since this antibody detects the exogenous N1–355 protein (upper panel of each pair) as well as the endogenous GRP94 (lower panel of each pair), it provides an internal control for the amount of protein loaded, allowing quantitative comparisons. (B) Quantitation of N1–355 binding to RAW 264.7 cells. The bands from the experiment shown in (A) were scanned and their intensities were plotted to quantify the binding in each condition. Two other experiments gave similar results. (C) Quantitation of N1–355 binding to DCs, performed as described in (B). Another experiment gave similar results. (D) Binding of N34–355 to DC is inhibited by full-length GRP94. DCs (106 cells in 100 µl) were incubated with N34–355 (125 nM), tagged with biotin specifically at Lys370 in the C-terminus, in the presence of increasing concentration of GRP94 (whose first 33 amino acids were deleted, similar to N34–355) on ice for 1 h. Binding was quantified as above, with the units of binding in the absence of competitor taken as 100%. A two-tailed t-test was used to calculate P values for the difference in binding in the presence and absence of competitor GRP94. **P < 0.01.

 
Similar results were obtained with the even shorter version of GRP94, N34–355. This version can be labeled during expression in E. coli with biotin in site-specific manner, because it has a birA target sequence grafted onto the C-terminus of N34–355. The expressed protein is, therefore, a substrate for the bacterial biotin ligase, which attached a biotin molecule to Lys370 (28), enabling sensitive detection with streptavidin. Binding of this protein construct to DC or to RAW cells was similar to that of N1–355, indicating that the receptor-binding sites are within the amino acids 34–355.

Finally, in the presence of increasing concentrations of recombinant GRP94, binding of N34–355 was progressively inhibited, confirming that they indeed compete for the same binding sites (Fig. 2D). The incomplete inhibition may reflect the quality of the recombinant full-length GRP94 or differential affinities to the cell-surface receptors. At the same time, the residual binding may represent binding of the N34–355 protein to additional receptors that do not bind full-length GRP94. In either case, we conclude that amino acids 34–355 harbor the binding sites for SR-A and CD91.

Interestingly, when recombinant N1–355 was loaded with VSV8, it bound to APCs consistently better than ‘empty’ N1–355. The increased binding activity was observed with both RAW 264.7 cells (Fig. 2A and B) as well as DCs (Fig. 2A and C). In addition, binding of the truncated GRP94 was greater following a heat shock treatment than without it (data not shown). These two observations indicate that the APC binding activity may be a property of a conformational state of the truncated chaperone protein.

Internalization of truncated GRP94 constructs by DCs
When DCs were allowed to bind either N34–355 or N1–355 at 4°C and then warmed to 37°C, the intracellular distribution as seen by immunofluorescence microscopy was punctate, showing that the proteins were internalized (Fig. 3A, only N34–355 is shown). Double labeling with anti-GRP94 and anti-LAMP-1, an endosomal/lysosomal marker, showed that after 15–60 min of uptake many of the endocytic structures were positive for both, suggesting receptor-mediated endocytosis into the endosomal system. The uptake of N34–355 was inhibited by a combination of {alpha}2M and fucoidan, and did not require prior loading with peptide, as empty protein was taken along the same pathway (data not shown). We conclude from the observations in Figs 2 and 3 that the sequence N34–355 contains the information responsible for the receptor-mediated endocytosis of full-length GRP94 into APCs.


Figure 3
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 3 Internalization of N34–355 by DCs. DCs grown on glass coverslips were incubated with biotinylated N34–355 (10 µg ml–1) for 30 min on ice and then for 30 min at 37°C. (A) The uptake of N34–355 was probed with anti-biotin antibody, followed by an FITC-conjugated mouse secondary antibody. Nuclei were counterstained with Hoechst. (B) The same cells were co-stained with monoclonal anti-LAMP-1 antibody followed by rhodamine-conjugated anti-mouse Ig. Note the correspondence of N34–355-positive structures with large endosomal vesicles. Bar, 10 µm.

 
Induction of DC maturation
Up-regulation of co-stimulatory molecules on APCs is important for their ability to activate T cells and binding of either full-length GRP94 or its N-terminal domain have been shown to stimulate the maturation of DCs (12, 13, 22, 32). Therefore, to validate our protein, we investigated the ability of N1–355 to induce similar functional maturation of DCs. Incubation with the N1-355–VSV8 complex activated DCs, as indicated by up-regulation of CD40, CD80, CD86, and to a lesser extent MHC class II (Fig. 4A). Both the mean fluorescence intensities and the percentages of cells expressing these activation-related surface proteins were increased (Fig. 4A and B). The activation was dependent on the presence of native chaperone (Fig. 4C), and free peptide, even at 1 µg ml–1 (50-fold higher than the amount of VSV8 in the complex), did not lead to increased expression of these activation markers (Fig. 4A and B). On the other hand, N1–355 itself, even without a deliberately bound peptide, was a potent activator of DCs (Fig. 4B), quite comparable to the effect of LPS at a concentration 100 ng ml–1 (Fig. 4C).


Figure 4
View larger version (47K):
[in this window]
[in a new window]
  		Fig. 4 Activation of DC by N1–355. (A) Increased expression of activation markers. BM–DCs were treated with VSV8 peptide (20 ng ml–1) or N1-355–VSV8 complex (20 µg ml–1 protein and 20 ng ml–1 peptide) on day 6–7 of culture (see Methods). After 24 h, cells were harvested and stained with anti-CD11c, -CD40, -CD80, -CD86, -class I and -class II fluorophore-tagged antibodies and analyzed in a BD FACSCalibur. The live CD11c+ DCs were gated and all the markers were analyzed among this population. Histograms show the fluorescence intensities of the cells stained with isotype-matched control antibody (shaded area), mock-treated cells (discontinuous line), cells treated with VSV8 peptide (thin line) and with the complex (thick line). The mean fluorescence intensity (MFI) values of the staining histograms are shown separately in the table. MFIs of control antibodies for peptide-pulsed DCs and for complex-pulsed DCs were almost identical. (B) Quantitation of cell-surface proteins expression. FACS analysis of the expression of DC markers after 24 h incubation with buffer alone (control), or free VSV8 (1 µg ml–1), N1–355 (20 µg ml–1), or the complex (20 µg ml–1 protein and 20 ng ml–1 peptide). DCs were stained for the indicated markers or with isotype-matched control antibodies. All FACS data were analyzed using Cell Quest software in the area gated for CD11c+ cells. The data are presented as percent of CD11c+ cells that are positive for the indicated markers, defined as cells stained more brightly than all but 1% of the staining with isotype control antibody. Bars represent the averages and SDs of data from three experiments, conducted with three independent BM–DC cultures. *P < 0.05, by a two-tailed t-test. **P < 0.01. (C) Similar FACS analysis as in (B), comparing the effects on DCs of LPS (100 ng ml–1), N1-355–VSV8 complex and boiled complex. The latter is to check that the activation is not due to contaminated LPS. (D and E) Cytokine secretion. The levels of the cytokines TNF{alpha} (D) and IL-12p70 (E) in the supernatants from each activation experiment was quantified by ELISA. The data shown are averages of duplicates from two independent mice.

 
Because of its ubiquitous presence, LPS contamination may account for the activity of the N1–355 preparations (11, 33). Three lines of evidence, however, show that the observed phenotypic maturation of DCs is due to the chaperone and not to LPS contamination. First, the protein, which was derived from LPS-free insect cells, and the peptide preparations were checked routinely by the standard Limulus assay, and found to contain <10 pg ml–1 LPS. Such dose could not mature DCs in our hands (data not shown). Only by using the much more sensitive assay of TNF-alpha secretion by exposed RAW 264.7 cells could we show any activity consistent with LPS (data not shown). Thus, if present, the level of LPS was two to three orders of magnitude lower than the level required by deliberate addition of LPS to induce the activation markers on DCs. Second, boiling N1–355 (data not shown) or the complex (Fig. 4C) for 60 min destroyed the stimulating activity, whereas LPS is resistant to this treatment. Third, polymyxin B, a neutralizer of LPS, reduced drastically DC activation by LPS (data not shown), but had only marginal influence on the activation mediated by N1–355 or its complex: indeed, the data in Fig. 4(B and C) were generated by stimulating DCs with complexes in the presence of 50 µg ml–1 of polymyxin B. The results of DC activation by complexes were similar when added in the absence of polymyxin B (data not shown). Thus, we conclude that the interaction of N1–355 with DCs activates them, and that LPS contamination does not play a role in this process.

In addition to expression of surface proteins, activation of DCs was also measured by induction of cytokine secretion. We quantified the levels of TNF{alpha} and IL-12p70 in the media after exposure of DCs to N1–355 (Fig. 4D and E). Secretion of both cytokines was induced by treatment with empty N1–355, and peptide-loaded N1–355 was slightly more effective. As with the surface markers, the extent of stimulation of cytokine secretion by the chaperones was comparable to that found in response to deliberate addition of LPS.

Presentation of antigenic peptide chaperoned by N1–355
To measure the ability of N1–355 to cross-present its bound peptide via MHC class I, either RAW 264.7 or DCs were pulsed for 6 h with N1-355–VSV8 complex. The complex was formed by the heat-shock method (see Methods), excess peptide was removed with a spin column and the complex was purified further by FPLC on Superdex-75 (Fig. 5A). The presentation of the peptide was measured by the ability of the pulsed APCs to activate T cell hybridomas, N15, specific for VSV8 in the context of H-2Kb (30). To ensure MHC compatibility, we used either RAW 264.7 cells stably transfected with Kb or DCs from C57Bl/6/RAG-2ko mice. The pulsed APCs were cultured with the N15 hybridoma for 18 h and stimulation of the hybridoma cells was measured by the level of IL-2 secreted. A significant amount of IL-2 was secreted in response to even a 20 ng ml–1 VSV8 when bound to N1–355, and the entire dose-response curve forVSV8 was shifted to lower peptide doses (N1-355–VSV8, Fig. 5B and C). Similar activation results were obtained using the RAW 264.7 line (Fig. 5B) and primary cultures of DCs (Fig. 5C). Free VSV8 was only effective at 500 ng ml–1, likely due to exchange of VSV8 onto MHC class I already expressed on the surface of APCs. Importantly, active loading of N1–355 with peptide was necessary, either by an overnight incubation with VSV8 or by heat shocking the protein, because simply mixing the peptide and protein did not increase the efficiency of presentation (Fig. 5B and C, mixture). We conclude that N1–355 is a sufficient portion of GRP94 to augment the presentation of an antigenic peptide in culture.


Figure 5
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 5 Presentation of peptide by APCs. (A) Purification of N1-355–VSV8 complex. N1–355 and VSV8 were mixed, heat shocked at 50°C for 10 min and then incubated at room temperature for 30 additional minutes to form a complex. The bulk of unbound peptide was separated from the complex by a P30 spin column, followed by size exclusion chromatography on a Superdex-75 column (top panel). Chromatography of protein itself and the free peptide are shown in the middle and bottom panels, respectively. The molar ratio of protein to peptide was 1:0.5. (B) RAW cells expressing Kb in 96-well plates (0.5 x 105 cells per well) were pulsed with increasing concentrations of N1–355, N1-355–VSV8 complex, free VSV8 or a mixture of VSV8 and N1–355 that was neither heat shocked nor pre-incubated. Cells were pulsed for 6 h and then washed and co-cultured with same number of N1 cells for 18–20 h. The presence of IL-2 in the supernatant of each well was measured by ELISA. To confirm that the activity of the complex in the presentation assay was not due to contamination of the fraction with free peptide, the fraction V0 (indicated by an arrow in A) obtained from chromatography of free peptide was tested for stimulation of the N1 hybridoma and was nagative (data not shown). N = 2. (C) A similar presentation assay to the one shown in (B), employing DCs and a broader range of peptide input. N = 2.

 
Immunization with N1-355–VSV8 complex generated peptide-specific cytotoxic T cells in vivo
Thus far, we have shown that N1–355 can deliver peptides efficiently to DCs and macrophages and can stimulate the up-regulation of co-stimulatory molecules and pro-inflammatory cytokines in primary resting DCs. Therefore, this fragment of GRP94 could arm innate immune cells to efficiently present antigens to naive T cells. To investigate if this is the case, we used N1-355–VSV8 complexes to prime an antigen-specific CTL response in vivo. Mice were immunized with N1-355–VSV8 complexes twice with a 10-day interval and CTL activity in their spleens was assayed either 2 weeks or 2 months after immunization. Splenocytes were re-stimulated in culture with VSV8 and IL-2 to expand the frequency of effector T cells, and their CTL activity was measured after 5 days on N1 cells as targets. Mice immunized with the complex developed VSV8-specific CTL while mice immunized with the same amount of free peptide showed no significant development of immunity (Fig. 6). The complex was even more effective in generating CTL than an emulsion of peptide in Freund's adjuvant: 100 µg VSV8 emulsified in Freund's adjuvant did not stimulate CTL to the same extent as 20 ng of peptide complexed with N1–355. Therefore, N1–355 is capable of augmenting CTL responses in vivo like the full-length chaperone.


Figure 6
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 6 In vivo generation of peptide-specific CTLs by the N1-355–VSV8 complex. Groups of C57Bl/6 mice were immunized at the base of the tail with N1-355–VSV8 complex (20 µg protein and 20 ng peptide per mouse), free VSV8 (20 ng per mouse) or VSV8 (100 µg per mouse) emulsified in Freund's adjuvant. The control group was injected with buffer alone. Mice were immunized twice with a 10-day interval. Either 2 weeks after the second injection (experiment 1) or 2 months later (experiment 2), splenocytes were harvested and re-stimulated in culture with VSV8 (1 µg ml–1) and mouse recombinant IL-2 (20 U ml–1). On the 5th day of re-stimulation, the cells were harvested and assayed for their CTL activity on 51Cr-labeled N1 cells, at different E:T ratios. Data are averages ± SD from two independent experiments and a total of six mice per group.

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
References
 
A critical and controversial issue in assessing the utility of the molecular chaperone GRP94 as a vaccine is whether it binds peptides and transfers them to APCs. While a number of papers demonstrate that exogenous provision of GRP94–peptide complexes stimulates specific T cell immune responses, others argue that GRP94 acts as a general adjuvant and not in peptide-specific fashion. This work provides evidence that GRP94 can be used as a peptide-specific vehicle. We show that amino acids 34–355 are sufficient not only to account for binding of antigenic peptides to the full-length protein but also for the binding to receptors on DCs and the subsequent endocytosis. Amino acids 1–355 are sufficient to reproduce these activities as well as the peptide-specific cross-presentation to CD8+ T cells previously observed with the full-length GRP94. We suspect that the 34–355 construct would also be able to cross-present peptides in vivo when produced in endotoxin-free form. Therefore, the ‘mini-chaperone’ is sufficient to activate class I-restricted T cell responses.

We previously reported that N34–355 contains the binding site for the octameric T cell epitope VSV8 and showed that monomeric recombinant protein binds this and other peptides with a 1:1 stoichiometry (24, 25). The binding of peptides is regulated by binding of ligands, such as inhibitors and adenine nucleotides, to a spatially distinct site on the protein (25). Therefore, we consider peptide binding an inherent activity of GRP94, which is not dependent on dimerization or oligomerization of the protein. It is interesting to note that, similar to peptide binding by MHC molecules, the on-rate of peptide binding to N1–355 or N34–355 is slow, but the stability of the complexes is extremely high (6, 24, 34). Another important property regarding antigen presentation is that GRP94 can be loaded with peptides as long as 40-mers (5, 35), within the size range that is translocated via the endoplasmic reticulum membrane transporter associated with antigen processing.

This peptide-binding activity of the N-terminal domain of GRP94 should be considered in re-evaluating the study of Baker-LePain et al. (22), which showed that an N-terminal domain similar to the one employed here is effective in tumor rejection, but ascribed this activity to peptide-independent effects. Their conclusion was based on the assumption that the only peptide-binding site of GRP94 is the one identified previously in the C-terminal domain (36, 37). While this was a reasonable supposition at the time, the newer data (24; T. Gidalevitz, B. B. Simen, O. Ostrovsky, S. M. Vogen, J. L. Dul and Y. Argon, in preparation) show that the construct used by Baker-LePain et al. (22) did in fact contain a peptide-binding site.

Another report has recently shown that the N-terminal domain of GRP94, but not a C-terminal domain, can stimulate CTL responses to an HBV peptide (23). Since that report only employed co-administration of a mixture of HBV peptide with the recombinant protein, without data on actual binding, the observed stimulation of APC could only be interpreted as an adjuvant effect, without presuming a peptide-specific action of the chaperone. Our work, employing a purified complex of peptide with protein with defined stoichiometry, resolves this issue and strongly argues that the N-terminal portion of GRP94 carries with it peptides that can be cross-presentated. In fact, we show here that antigen presentation by the mini-chaperone strictly requires active peptide loading and that simple mixing does not suffice.

In addition to peptide binding, the N-terminal portion of GRP94 also binds professional APCs in a manner similar to the full-length protein. This activity was initially shown by Baker-LePain et al. and our results extend its characterization. First, we show that the N-terminal portion binds to DCs and macrophages via the same CD91 and SR-A receptors previously reported to mediate binding of the full-length protein, suggesting that the receptor-binding site is within amino acids 34–355. Second, we show that peptide-loaded N1–355 binds more avidly than the presumably empty protein, suggesting that upon peptide binding, the chaperone undergoes a conformation change that increases the affinity for the APC receptors. Third, because excess GRP94 did not abolish the binding of N34–355 to APCs completely, it is still possible that the truncated protein can also use cell-surface receptors on these cells that GRP94 does not utilize.

Following binding to the cell-surface receptors, the mini-chaperone is endocytosed with properties similar to those by full-length GRP94 reported in (10): co-localization in endosomes with LAMP-1 and similar kinetics. Based on the reports that GRP94 binds LPS tightly and this facilitates binding and uptake of GRP94 via Toll-like receptors (TLRs) (13), we deliberately added increasing concentrations of LPS to the recombinant truncated chaperone, or performed the binding to APC in the presence of polymixin B, yet we did not find a significant difference in the amount or pattern of cell-associated N1–355 (data not shown). This suggests that the bulk of the mini-chaperone uptake does not involve the TLRs and perhaps the TLR-interacting domain is located elsewhere on GRP94. Thus, we conclude that the mini-chaperone 34–355 is sufficient to bind peptide, contain the SR-A and CD91 binding sites, and is sufficient to mediate the uptake of peptides into the endosomal compartment of APCs.

In addition to being internalized by DCs, GRP94 has been shown to stimulate and activate DCs (13, 20, 38), an activity which is key to its effectiveness in cross-presentation. Binding of full-length GRP94 to TLRs initiates signal transduction pathways that lead to maturation of DCs, as reflected in increased expression of surface markers and secretion of cytokines (4, 13, 20, 38). As we show here, N1–355 has the same activity as the full-length GRP94: it up-regulates the same co-stimulatory receptors and induces secretion of the same cytokines. Furthermore, the induction of IL-12p70 production by DCs, at levels comparable to those induced by a strong IL-12 inducer such as LPS, suggests that one of the mechanisms by which N1–355 is so efficient in eliciting antigen-specific cytotoxic T cells may be by ‘biasing’ the immune system toward a Th1 response.

Recently, Binder and Srivastava (15) have shown that full-length GRP94, present in cell lysates, is a necessary source of antigen for cross-presentation to CD8+ T cells in vivo. Since we use recombinant proteins, there are no contaminants from cellular lysates that could have adjuvant effects and synergize with the GRP94 fragment (39). Of note, the dose of peptide that elicits T cell response when bound to N1–355, 20 ng, is the same as the dose used with full-length GRP94 (15) and is much lower than the peptide dose commonly used to elicit CTL priming with the traditional adjuvant CFA (20–100 µg). Thus, our data confirm and expand the conclusion of Binder et al. and definitively demonstrate that amino acids 1–355 are a sufficient module that can be used instead of the 802 amino acid long GRP94 protein, and is better than many other adjuvants, to elicit a powerful immune response.

The use of constructs derived from GRP94 for immunization does not address the debate about the physiological role of chaperone–peptide transfer. Shen and Rock (40) argued that the physiological relevant form of antigen that is cross-presented in vivo is the whole cellular protein from which the antigenic peptide is derived. Similarly, Norbury et al. (41) showed that cross-presentation is based on the transfer of proteasome substrates rather than peptides. Thus, cross-priming of T cells in vivo may not be mediated by the mechanism described in this work. Nonetheless, even if chaperone–peptide complexes are not important for in vivo cross-presentation, if they can deliver immunogenic peptides to APC, they are a viable mode of deliberate immunization (42) that should be pursued. The availability of a short version of GRP94 is of major advantage for such vaccine application: it is expressed at higher yields and is easier to purify (C. Biswas, unpublished results); it is quite stable and superior to full-length GRP94 in its aggregation behavior. Furthermore, the absence of other domains of GRP94, which are designed to mediate a variety of protein–protein interactions, minimizes potentially irrelevant activities of the chaperone, focusing instead on the activities required for efficient antigen presentation. It is, therefore, likely that the mini-chaperone is a better reagent for immunization.


   Acknowledgements
 
We thank Hua Ding and Cynthia Spitalny for preparing the biotinylated protein, Steven Berardi for competent mouse husbandry, Noreen Ahmed and Steven Berardi for protein purification, Renell Morgan and Sean McCarny for help with microscopy and Julian Ostrovsky for help with image quantitation. Sherry Wanderling made the macrophage cell line expressing H2-Kb, using a construct that was a gift from H. Schreiber (The University of Chicago). We thank D. Lyles (Wake Forest University) and E. Reinherz (Dana-Farber Cancer Institute) for gifts of cell lines and J. Orange for help with the cytotoxicity assays. Finally, we thank a reviewer for insightful comments that improved the quality of the paper. This work was supported by a grant from the W.W. Smith Charitable Trust (to Y.A.), by National Institutes of Health grant AI-049892 (to S.G.) and by grants from the Stokes Institute and the Penn Center for AIDS research. O.O. was partly supported by a postdoctoral fellowship from the Juvenile Diabetes Research Foundation.


   Abbreviations
 
APC, antigen-presenting cell
{alpha}2M, alpha-2-macroglobulin
BM, bone marrow
BM–DC, bone marrow-derived dendritic cell
DC, dendritic cell
GRP94, glucose-regulated protein
HBV, hepatitis B virus
HSP, heat shock protein
ME, mercaptoethanol
SR-A, scavenger receptor type A
TLR, Toll-like receptor
TNF, tumor necrosis factor
VSV, vesicular stomatitis virus

   Notes
 
Transmitting editor: A. Singer

Received 7 February 2006, accepted 27 April 2006.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 

  1. Srivastava PK and Udono H. (1994) Heat shock protein-peptide complexes in cancer immunotherapy. Curr. Opin. Immunol. 6:728.[CrossRef][ISI][Medline]
  2. Suto R and Srivastava PK. (1995) A mechanism for the specific immunogenicity of heat shock protein-chaperoned peptides. Science 269:1585.[Abstract/Free Full Text]
  3. Arnold D, Faath S, Rammensee H, Schild H. (1995) Cross-priming of minor histocompatibility antigen-specific cytotoxic T cells upon immunization with the heat shock protein gp96. J. Exp. Med. 182:885.[Abstract/Free Full Text]
  4. Baker-LePain JC, Reed RC, Nicchitta CV. (2003) ISO: a critical evaluation of the role of peptides in heat shock/chaperone protein-mediated tumor rejection. Curr. Opin. Immunol. 15:89.[CrossRef][ISI][Medline]
  5. Demine R and Walden P. (2005) Testing the role of gp96 as peptide chaperone in antigen processing. J. Biol. Chem. 280:17573.[Abstract/Free Full Text]
  6. Blachere NE, Li Z, Chandawarkar RY, et al. (1997) Heat shock protein-peptide complexes, reconstituted in vitro, elicit peptide-specific cytotoxic T lymphocyte response and tumor immunity. J. Exp. Med. 186:1315.[Abstract/Free Full Text]
  7. Binder RJ, Han DK, Srivastava PK. (2000) CD91: a receptor for heat shock protein gp96. Nat. Immunol. 1:151.[Medline]
  8. Basu S, Binder RJ, Ramalingam T, Srivastava PK. (2001) CD91 is a common receptor for heat shock proteins gp96, hsp90, hsp70, and calreticulin. Immunity 14:303.[CrossRef][ISI][Medline]
  9. Berwin B, Hart JP, Rice S, et al. (2003) Scavenger receptor-A mediates gp96/GRP94 and calreticulin internalization by antigen-presenting cells. EMBO J. 22:6127.[CrossRef][ISI][Medline]
  10. Berwin B, Rosser MF, Brinker KG, Nicchitta CV. (2002) Transfer of GRP94(Gp96)-associated peptides onto endosomal MHC class I molecules. Traffic 3:358.[CrossRef][ISI][Medline]
  11. Reed RC, Berwin B, Baker JP, Nicchitta CV. (2003) GRP94/gp96 elicits ERK activation in murine macrophages: a role for endotoxin contamination in NF-kappaB activation and nitric oxide production. J. Biol. Chem. 278:31853.[Abstract/Free Full Text]
  12. Singh-Jasuja H, Scherer HU, Hilf N, et al. (2000) The heat shock protein gp96 induces maturation of dendritic cells and down-regulation of its receptor. Eur. J. Immunol. 30:2211.[ISI][Medline]
  13. Vabulas RM, Braedel S, Hilf N, et al. (2002) The endoplasmic reticulum-resident heat shock protein Gp96 activates dendritic cells via the Toll-like receptor 2/4 pathway. J. Biol. Chem. 277:20847.[Abstract/Free Full Text]
  14. Hilf N, Singh-Jasuja H, Schild H. (2002) The heat shock protein Gp96 links innate and specific immunity. Int. J. Hyperthermia 18:521.[CrossRef][ISI][Medline]
  15. Binder RJ and Srivastava PK. (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.[CrossRef][ISI][Medline]
  16. Tobian AA, Canaday DH, Boom WH, Harding CV. (2004) Bacterial heat shock proteins promote CD91-dependent class I MHC cross-presentation of chaperoned peptide to CD8+ T cells by cytosolic mechanisms in dendritic cells versus vacuolar mechanisms in macrophages. J. Immunol. 172:5277.[Abstract/Free Full Text]
  17. Suzue K, Zhou X, Eisen HN, Young RA. (1997) Heat shock fusion proteins as vehicles for antigen delivery into the major histocompatibility complex class I presentation pathway. Proc. Natl Acad. Sci. USA 94:13146.[Abstract/Free Full Text]
  18. Castellino F, Boucher PE, Eichelberg K, et al. (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.[CrossRef][ISI][Medline]
  19. Nair S, Wearsch PA, Mitchell DA, Wassenberg JJ, Gilboa E, Nicchitta CV. (1999) Calreticulin displays in vivo peptide-binding activity and can elicit CTL responses against bound peptides. J. Immunol. 162:6426.[Abstract/Free Full Text]
  20. Schild H and Rammensee HG. (2000) gp96—the immune system's Swiss army knife. Nat. Immunol. 1:100.[CrossRef][ISI][Medline]
  21. SenGupta D, Norris PJ, Suscovich TJ, et al. (2004) Heat shock protein-mediated cross-presentation of exogenous HIV antigen on HLA class I and class II. J. Immunol. 173:1987.[Abstract/Free Full Text]
  22. Baker-LePain JC, Sarzotti M, Fields TA, Li CY, Nicchitta CV. (2002) GRP94 (gp96) and GRP94 N-terminal geldanamycin binding domain elicit tissue nonrestricted tumor suppression. J. Exp. Med. 196:1447.[Abstract/Free Full Text]
  23. Li H, Zhou M, Han J, Zhu X, Dong T, Gao GF, Tien P. (2005) Generation of murine CTL by a hepatitis B virus-specific peptide and evaluation of the adjuvant effect of heat shock protein glycoprotein 96 and its terminal fragments. J. Immunol. 174:195.[Abstract/Free Full Text]
  24. Vogen SM, Gidalevitz T, Biswas C, et al. (2002) Radicicol-sensitive peptide binding to the N-terminal portion of GRP94. J. Biol. Chem. 277:40742.[Abstract/Free Full Text]
  25. Gidalevitz T, Biswas C, Ding H, et al. (2004) Identification of the N-terminal peptide binding site of glucose-regulated protein 94. J. Biol. Chem. 279:16543.[Abstract/Free Full Text]
  26. Gallucci S, Lolkema M, Matzinger P. (1999) Natural adjuvants: endogenous activators of dendritic cells. Nat. Med. 5:1249.[CrossRef][ISI][Medline]
  27. Zabrecky JR and Sawlivich W. (2004) Purification of the heat shock protein, gp96, from natural sources. Methods 32:3.[CrossRef][ISI][Medline]
  28. Scott CJ, Martin SL, Wallace A, Curran MD, Walker B. (2000) Characterization of the affinity of streptavidin toward a peptide sequence previously identified as a target substrate for biotinylation by the Escherichia coli biotin holoenzyme synthetase, BirA. Anal. Biochem. 284:416.[CrossRef][ISI][Medline]
  29. Wu SM, Boyer CM, Pizzo SV. (1997) The binding of receptor-recognized alpha2-macroglobulin to the low density lipoprotein receptor-related protein and the alpha2M signaling receptor is decoupled by oxidation. J. Biol. Chem. 272:20627.[Abstract/Free Full Text]
  30. Liu J, Tse AG, Chang HC, et al. (1996) Crystallization of a deglycosylated T cell receptor (TCR) complexed with an anti-TCR Fab fragment. J. Biol. Chem. 271:33639.[Abstract/Free Full Text]
  31. Schulte TW, Akinaga S, Murakata T, et al. (1999) Interaction of radicicol with members of the heat shock protein 90 family of molecular chaperones. Mol. Endocrinol. 13:1435.[Abstract/Free Full Text]
  32. Singh-Jasuja H, Toes RE, Spee P, et al. (2000) Cross-presentation of glycoprotein 96-associated antigens on major histocompatibility complex class I molecules requires receptor-mediated endocytosis. J. Exp. Med. 191:1965.[Medline]
  33. Wallin RP, Lundqvist A, More SH, von Bonin A, Kiessling R, Ljunggren HG. (2002) Heat-shock proteins as activators of the innate immune system. Trends Immunol. 23:130.[CrossRef][ISI][Medline]
  34. Wearsch PA and Nicchitta CV. (1997) Interaction of endoplasmic reticulum chaperone GRP94 with peptide substrates is adenine nucleotide-independent. J. Biol. Chem. 272:5152.[Abstract/Free Full Text]
  35. Spee P and Neefjes J. (1997) TAP-translocated peptides specifically bind proteins in the endoplasmic reticulum, including gp96, protein disulfide isomerase and calreticulin. Eur. J. Immunol. 27:2441.[ISI][Medline]
  36. Linderoth NA, Popowicz A, Sastry S. (2000) Identification of the peptide-binding site in the heat shock chaperone/tumor rejection antigen gp96 (Grp94). J. Biol. Chem. 275:5472.[Abstract/Free Full Text]
  37. Linderoth NA, Simon MN, Hainfeld JF, Sastry S. (2001) Binding of antigenic peptide to the endoplasmic reticulum-resident protein gp96/GRP94 heat shock chaperone occurs in higher order complexes. Essential role of some aromatic amino acid residues in the peptide-binding site. J. Biol. Chem. 276:11049.[Abstract/Free Full Text]
  38. Basu S, Binder RJ, Suto R, Anderson KM, Srivastava PK. (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.[Abstract/Free Full Text]
  39. Gallucci S and Matzinger P. (2001) Danger signals: SOS to the immune system. Curr. Opin. Immunol. 13:114.[CrossRef][ISI][Medline]
  40. Shen L and Rock KL. (2004) Cellular protein is the source of cross-priming antigen in vivo. Proc. Natl Acad. Sci. USA 101:3035.[Abstract/Free Full Text]
  41. Norbury CC, Basta S, Donohue KB, et al. (2004) CD8+ T cell cross-priming via transfer of proteasome substrates. Science 304:1318.[Abstract/Free Full Text]
  42. Belli F, Testori A, Rivoltini L, et al. (2002) Vaccination of metastatic melanoma patients with autologous tumor-derived heat shock protein gp96-peptide complexes: clinical and immunologic findings. J. Clin. Oncol. 20:4169.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/7/1147    most recent
dxl049v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)