Cells + Gp96, Hsp90, and Hsp70 with Cd11b Interaction of Heat Shock

Home , Hsp90

Saturation, Competition, and Specificity in Interaction of Heat Shock Proteins (hsp) gp96, hsp90, and hsp70 with CD11b+ Cells

This information is current as Robert J. Binder, Michelle L. Harris, Antoine Ménoret and of September 25, 2021. Pramod K. Srivastava J Immunol 2000; 165:2582-2587; ; doi: 10.4049/jimmunol.165.5.2582 http://www.jimmunol.org/content/165/5/2582 Downloaded from

References This article cites 26 articles, 15 of which you can access for free at: http://www.jimmunol.org/content/165/5/2582.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 25, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Saturation, Competition, and Speciﬁcity in Interaction of Heat Shock Proteins (hsp) gp96, hsp90, and hsp70 with CD11b؉ Cells1

Robert J. Binder, Michelle L. Harris, Antoine Me´noret, and Pramod K. Srivastava2

Heat shock proteins (hsp(s)) have been postulated to interact with APCs through specific receptors, although the receptors are yet to be identified. Specificity, saturation, and competition are the three defining attributes of a receptor-ligand interaction. We demonstrate here that the interaction of the heat shock proteins gp96 and hsp90 with CD11b؉ cells is specific and saturable and that gp96 can compete with itself in gp96-macrophage interaction. Interestingly, the phylogenetically related hsp90 also competes quite effectively with gp96 for binding to macrophages, whereas the unrelated hsp70 does so relatively poorly, although it binds CD11b؉ cells just as effectively. These data provide evidence that the heat shock proteins interact with APCs with specificity and Downloaded from for the existence of at least two distinct receptors, one for gp96 and hsp90 and the other for hsp70. The Journal of Immunology, 2000, 165: 2582–2587.

mmunization with the heat shock protein (hsp)3-peptide com- suggesting the presence of a shared receptors between hsp90 and plexes has been shown to elicit potent CD8ϩ cytotoxic T gp96 and a unique receptor for hsp70. lymphocyte response against the peptide chaperoned by the I http://www.jimmunol.org/ hsp (1). This phenomenon has been demonstrated extensively in a Materials and Methods number of models including cancers (2–9), virus-infected cells (1, Cells and mice 10, 11), and cells expressing minor histocompatibility (12) or Peritoneal exudate cells (PEC) were obtained from C57BL/6 mice (The model Ags (13). A remarkable attribute of this phenomenon is the Jackson Laboratory, Bar Harbor, ME) injected 5–15 days earlier with 0.5 extremely small quantities of peptide required for generation of ml pristane (2,6,10,14-tetramethylpentadecane). CD11bϩ cells were posi- CD8ϩ T cell responses (1). As little as a few hundred picograms tively selected from mouse peritoneal exudates by the magnetic bead sys- tem (Miltenyi Biotec, Auburn, CA). Purifications of Ͼ93% CD11bϩ (as of peptide, if chaperoned by hsps, are sufficient for eliciting pep- ϩ measured by FACScan analysis) were always achieved. These cells are tide-specific CD8 T cells. It had been demonstrated earlier that macrophages as defined by the following: MHC IIϩ, CD11cϪ, phagocytic.

immunization with hsp-peptide complexes is exquisitely depen- by guest on September 25, 2021 dent on the presence of functional APCs in the immunized host FITC labeling of proteins (14). Collectively, these two observations led to the suggestion that Labeling of proteins with FITC was performed using FITC-conjugation the unusually high immunogenicity of hsp-peptide complexes re- kits and according to the manufacturer’s recommendations. Briefly, puri- sults from the presence of hsp receptors on APCs such as macro- fied hsps or mouse serum albumin (SA, 1 mg/ml) were incubated with FITC in 0.1 M carbonate-bicarbonate buffer for2hatroom temperature. phage and dendritic cells (15). Indeed, a number of recent studies Free unconjugated FITC was removed by passing the mixture over gel have suggested the presence of hsp-binding moieties on the sur- filtration columns. The number of FITC molecules bound per protein mol- faces of APCs (16–18). These studies, although consistent with the ecule was estimated by measuring the optical density at 280, 495, and 490 idea of a receptor(s) for hsps, fall short of defining the two minimal nm as recommended. Estimates show that between 4 and 5 FITC molecules bind to each protein molecule. All conjugated proteins were analyzed by essential criteria for receptors, i.e., saturability and competibility SDS-PAGE and immunoblotting with the respective anti-hsp mAbs, anti- of binding. The studies reported here make that incremental ad- SA Ab or anti-FITC mAb. Abs used for the immunoblots were mouse mAb vance and show that the interaction of the cytosolic hsps, hsp70 SPA-820 (clone N27F3-4 specific for both constitutive hsp73 and inducible and hsp90, and endoplasmic reticulum resident hsp gp96, with hsp72 forms), rat mAb SPA-850 (clone 9G10 specific for gp96), all from APCs is saturable. They further show that whereas hsp90 and gp96 StressGen Biotechnologies (Victoria, Canada). Anti-hsp90 Ab was a rabbit mAb (NeoMarkers, Lab Vision, Fremont, CA) specific for hsp84 and compete with gp96 for binding to APCs, hsp70 does not, thus hsp86. Anti-SA and anti-FITC Abs were purchased from SIGMA (St. Louis, MO). All FITC-labeled proteins were centrifuged at 100,000 ϫ g before use to remove any particulate matter. Center for Immunotherapy of Cancer and Infectious Diseases, University of Con- Purification of hsps necticut School of Medicine, Farmington, CT 06030 hsp70 and gp96 were purified from liver cells of C57BL/6 mice according Received for publication April 25, 2000. Accepted for publication June 19, 2000. to previously described methods (19, 20). Supernatants of 100,000 ϫ g The costs of publication of this article were defrayed in part by the payment of page centrifugations of cell lysates were subjected to stepwise ammonium sul- charges. This article must therefore be hereby marked advertisement in accordance fate precipitations. Pellets of 50% precipitations were used for hsp70 pu- with 18 U.S.C. Section 1734 solely to indicate this fact. rification. Pellets of 80% precipitates were resuspended in PBS (with 2 mM 1 The work was supported by National Institutes of Health Grants CA64394 and MgCl2, 2 mM CaCl2, and 2 mM PMSF) and applied to preequilibrated Con CA44786, U.S. Department of Defense Grant BAA96024, and a research agreement A affinity columns. Con A eluents were exchanged into a phosphate buffer with Antigenics, in which P.K.S. has a significant financial interest. with PD-10 gel exclusion columns and applied to DEAE anion exchange 2 Address correspondence and reprint requests to Dr. Pramod K. Srivastava, Univer- columns. gp96 eluted from these columns as a purified protein and was sity of Connecticut School of Medicine, MC1601, Farmington, CT 06030-1920. E- identified by immunoblotting. Pellets from 50% ammonium sulfate pre- mail address: [email protected] cipitation were resuspended in buffer D (20 mM Tris with 20 mM NaCl, 3 3 Abbreviations used in this paper: hsp, heat shock proteins; PEC, peritoneal exudate mM MgCl, and 15 mM 2-ME) and applied to ADP-affinity columns as cells; SA, mouse serum albumin. previously described. hsp70 was eluted with ADP-buffer D and applied to

DEAE anion exchange columns in a phosphate buffer. hsp70 eluted off these columns as a purified protein shown by a single band on SDS-PAGE and identified in immunoblotting. hsp90 was purified according to the protocol of Denis (21) with minor modifications. Briefly, 100,000 ϫ g supernatants were obtained from liver cell lysates and applied to a Mono Q column (Mono Q HR 16/10, purchased from Pharmacia Biotech (Uppsala, Sweden) and attached to the BIOCAD Perseptives Biosystems, Cambridge, MA) in 200 mM sodium phosphate buffer and eluted with a gradient salt concentration to 600 mM. hsp90-positive fractions were collected and changed to Tris buffer, pH 8. Hsp90 was reapplied to the MonoQ column in the Tris buffer and eluted with a gradient salt concentration from 0 to 1 M NaCl. The Mono Q column was used according to the recommended conditions. hsp90 was eluted in a pure form as shown by SDS-PAGE and identified by immunoblotting.

Immunization with hsps Mice were immunized with each of the three hsps as previously described (1). Visualization of hsp receptors Incubations of indicated amounts of FITC-labeled proteins and cells were done in the presence of Carnation 1% nonfat dry milk (Nestle´, Glendale, Downloaded from CA) in PBS for 20 min at 4°C. After repeated washing, cells were analyzed by ﬂow cytometry (Becton Dickinson, La Jolla, CA). Cells were also labeled with propidium iodide just before FACScan analysis. Cells staining positive for propidium iodide were gated out of the events. No differences were observed in the binding of hsps to CD11bϩ cells from pristaned or nonpristaned mice. For the saturation studies, mean ﬂuorescent intensities

were obtained from histogram plots. Competition studies were performed http://www.jimmunol.org/ by mixing labeled and unlabeled competitor proteins together before incubation with cells. Excess protein was removed, and mean fluorescence intensities were measured by FACS analysis. Confocal imaging Paraformaldehyde-fixed or unfixed cells were labeled with FITC-labeled FIGURE 1. Characterization of the hsp preparations used. A, Homoge- hsp as above. Labeled cells were visualized using a Zeiss LSM confocal neity and identity of gp96, hsp90, and hsp70 preparations used in the study. microscope. For internalization studies, unfixed cells were used, and im- gp96, hsp90, and hsp70 were purified as described in Materials and Meth- ages were recorded at various intervals at the exact same focal plane ods. hsp preparations were analyzed by SDS-PAGE and silver staining

and view. (left) and by immunoblotting with anti-gp96, anti-hsp 90, and anti- by guest on September 25, 2021 hsp72/73 mAbs (right). B, hsps and SA are complexed to FITC. Homo- Results and Discussion geneous preparations of hsps, gp96, hsp90, hsp70, and SA were complexed FITC-labeled hsps remain immunogenic covalently to FITC. FITC-labeled proteins were analyzed by SDS-PAGE and silver staining (B) and by immunoblotting with anti-FITC or anti-gp96 Homogeneous preparation of hsps gp96, hsp90, and hsp70 or se- mAbs (C). rum albumin (Fig. 1A) were labeled with FITC as described in Materials and Methods. The labeling was confirmed by SDS- PAGE and immunoblotting with anti-hsp and anti-FITC Abs (Fig. 1, B and C). Each hsp or albumin molecule bound to an average of five FITC molecules. The resulting 2-kDa increase in the size of the proteins was detectable by SDS-PAGE (Fig. 1B). By immunoblotting with an anti-FITC mAb, conjugation of gp96 to FITC was confirmed (Fig. 1C). Similar results were obtained with other hsps (not shown). Because modification of proteins by conjugation to FITC could result in the loss of binding of hsp to their putative receptors, we tested whether the FITC-labeled hsps remained immunogenic. We used the unique ability of hsp-peptide complexes to generate peptide-specific CTLs as the functional assay (1). FITC-labeled gp96 was complexed in vitro to the Kb-binding OVA epitope SIINFEKL and used to immunize C57BL/6 mice. Spleens were removed from the immunized mice 1 wk later, stimulated in vitro with the OVA-expressing cell line E.G7, and tested for cy- totoxicity of E.G7 or EL4 cells. It was observed that FITC-labeled gp96 complexed to SIINFEKL was able to elicit SIINFEKL-spe- FIGURE 2. FITC-labeled hsps are functionally immunogenic. C57BL/6 ␮ cific CTLs as effectively as unlabeled gp96-SIINFEKL and thus mice were immunized twice, 1 wk apart, with 50 g FITC-gp96-SIIN- FEKL, gp96-SIINFEKL, or FITC-gp96 complexes i.p., as described in remained immunogenic (Fig. 2). Control preparations of FITC- Ref. 1. Spleen cells were harvested from the mice, restimulated once with labeled gp96 uncomplexed to peptide were not able to generate SIINFEKL-pulsed APCs, and tested for the presence of anti-SIINFEKL SIINFEKL specific CTLs. Identical experiments were conducted CTLs in a 51Cr release assay using the SIINFEKL-expressing cell line with FITC-labeled hsp70 and hsp90 with similar results (data not E.G7 or the control EL4 cells as targets. Two mice per group were immu- shown). The functionally active FITC-labeled hsp preparations nized; each mouse is represented by a different symbol (E, F, mouse 1; Ⅺ, were therefore used for further studies. f, mouse 2). 2584 hsp RECEPTORS ON APC Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 3. hsps but not SA bind specifically to CD11bϩ cells but not control cells. Paraformaldehyde-fixed cells were incubated with various doses of FITC-labeled hsp or SA as indicated. Cells were extensively washed to remove excess protein and analyzed by flow cytometry as described in Materials and Methods. hsps bind CD11bϩ cells specifically and saturably inability of FITC-labeled SA to label cells also shows that binding Peritoneum-derived CD11bϩ cells can represent gp96-chaperoned of hsps was not an artifact of the FITC label. FITC-labeled hsps and SA were also incubated with Meth A tumor cells (represen- antigenic peptides, suggesting that they express receptors capable Ϫ Ϫ of internalizing hsp-peptide complexes (10). CD11bϩ cells were tative of CD11b cells). No binding of hsps or SA to CD11b purified from peritoneal exudates of mice and were fixed in 2% cells was observed. paraformaldehyde so as to render them incapable of phagocytosis. Saturability of hsp-APC interaction was tested. Paraformalde- Fixed cells were incubated with FITC-labeled gp96, hsp70, hsp90, hyde-fixed cells were used as in the previous experiment to elim- or SA. Excess or unbound protein was removed by extensive inate phagocytosis, turnover, internalization, and other vital phe- washing, and cells were analyzed by FACS analysis. As shown in nomena. Increasing quantities of FITC-labeled gp96, hsp90, Fig. 3, FITC-labeled hsps could effectively label CD11bϩ cells. hsp70, or SA (as a negative control), between 0 and 200 ␮g/ml, ϩ Increasing quantities of hsps used for the incubations labeled an were incubated with a constant number of fixed CD11b cells, as increasing percentage of cells. The labeling was specific to hsps indicated earlier. The mean fluorescent intensity was measured by because SA even at the highest concentration of protein used did FACS analysis of cells. Increasing and saturable binding of gp96 not demonstrate significant labeling to CD11bϩ cells (Fig. 3). The and hsp90 to APCs was observed with higher concentrations. The Journal of Immunology 2585

FIGURE 4. Binding of hsps to CD11bϩ cells is saturable. Paraformal- FIGURE 5. Gp96 and hsp90 compete with FITC-gp96 for binding. dehyde-fixed CD1bϩ cells were incubated with increasing doses of FITC- Paraformaldehyde-fixed cells were incubated with gp96/competitor mix- labeled gp96, hsp90, hsp70, or SA as indicated. Cells were extensively tures. Gp96 was used at a final and fixed concentration of 100 ␮g/ml. washed to remove excess protein and analyzed by flow cytometry. Mean Competitor proteins were used at increasing doses, as indicated. Cells were fluorescent intensities of labeled stains are plotted against the concentra- Downloaded from extensively washed to remove excess protein and analyzed by flow cytom- tions of proteins. etry. Mean fluorescent intensities are plotted as a function of increasing quantities of competitor proteins. Interestingly, both hsp90 and gp96 reached saturable binding at 100 ␮g/ml, although the APCs bound nearly 5 times as much hsp90 as gp96 (Fig. 4). hsp70 did not reach saturable binding even petitor (unlabeled) gp96, hsp90, hsp70, or SA, followed by incu- ␮ ϩ at 200 g/ml. We are not certain why hsp70 does not reach satu- bation of the mixtures with fixed CD11b cells. As shown in Fig. http://www.jimmunol.org/ rability; we presume that this is because it binds a receptor distinct 5, increasing quantities of unlabeled gp96 were able to increas- from hsp90/gp96, one with a different binding capacity, or that if ingly inhibit binding of FITC-gp96 to CD11bϩ cells such that at it is the same receptor, it has different affinities for binding the the highest concentration tested (unlabeled:labeled ratio of 10:1), various hsps. As in the previous experiment, no detectable binding unlabeled gp96 could inhibit Ͼ50% of the binding of FITC-gp96. of SA with APCs was observed even at the highest concentration Unlabeled SA showed insignificant and untitratable competition of SA used (200 ␮g/ml). with FITC-gp96 under the circumstances ( p Ͻ 0.005). Interest- ingly, unlabeled hsp90 was also able to inhibit in a titratable man- gp96 and hsp90 but not hsp70 compete for binding of gp96 to ϩ ϩ ner, the binding of FITC-gp96 to CD11b cells quite effectively

CD11b cells such that, at the highest concentration of hsp90 tested (unlabeled by guest on September 25, 2021 FITC-gp96 (100 ␮g/ml ﬁnal concentration) was mixed with in- hsp90:labeled gp96 ratio of 10:1), it could reduce the binding of creasing quantities (100–1000 ␮g/ml ﬁnal concentration) of com- gp96 by nearly 90%. This observation suggests that gp96 and

FIGURE 6. Confocal microscopy of CD11bϩ cells stained with FITC-hsps. A, PEC (CD11bϩ) or P815 (CD11bϪ) cells were fixed with paraformaldehyde and incubated with FITC-labeled gp96 or FITC-labeled SA. Labeled cells were extensively washed and analyzed for transmission (right) and fluorescence (left) by confocal microscopy using a Zeiss LSM confocal microscope at ϫ100 magnification. B, Similar fluorescence analysis was conducted using hsp90 and hsp70. 2586 hsp RECEPTORS ON APC Downloaded from http://www.jimmunol.org/

FIGURE 7. Binding of gp96 to CD11bϩ cells is followed by rapid internalization. Live, nonﬁxed CD11bϩ cells were allowed to bind FITC-gp96 at 4°C and were then incubated at 25°C for the indicated periods of time. Cells were washed extensively and visualized by confocal microscopy as described.

hsp90 share a common receptor. Unlabeled hsp70 did not inhibit 5 min), then cluster at focal points on the cell surface (10 min), binding of FITC-gp96 to CD11bϩ cells at the scale that gp96 or followed by internalization of FITC-gp96 molecules into the cells hsp90 did. At the highest hsp70 concentration tested (unlabeled (20 min or later) (Fig. 7). Such internalization was not observed to by guest on September 25, 2021 hsp70:labeled gp96 ratio of 10:1), the degree of inhibition of bind- occur with fixed cells (data not shown), arguing against a passive ing of FITC-gp96 to CD11bϩ cells was almost identical with the mechanism of uptake such as diffusion. inhibition observed by SA. Nonetheless, the inhibition by unla- The observations reported here allow a formal claim for the beled hsp70 was titratable, whereas inhibition by unlabeled SA existence of hsp receptors on CD11bϩ cells. Specificity of binding, was not. This leads us to suggest ambiguously that although hsp70 saturability of interaction, and the ability of the ligand to compete may interact with C11bϩ cells through a distinct receptor, it may with itself are essential attributes of ligand-receptor interaction and at high concentrations interact with the gp96/hsp90 receptor as distinguish it from nonspecific aggregation, phagocytosis, adven- well. There may thus exist a mechanism of cross-talk among the titious adsorption, etc. Each of these characteristics has been ex- various hsps and their receptors. amined. Specificity of binding of gp96, hsp90, and hsp70 to APCs is shown; saturability of binding is shown only for gp96 and hsp90. Binding of hsps occurs on the surface of APCs followed by hsp70-APC interaction did not arrive at saturability within the internalization of hsps range of concentrations tested. We are not certain why hsp70 does To visualize the binding of hsps to APCs and the early downstream not reach saturability; we presume that this is because the receptor consequences of such binding, paraformaldehye-fixed CD11bϩ that hsp70 binds to is distinct from the receptor for hsp90/gp96, cells incubated with FITC-gp96 were analyzed by confocal mi- one with a different binding capacity. If the receptor for hsp70 and croscopy. Peripheral staining of CD11bϩ cells with FITC-gp96 gp96/hsp90 is the same, then we assume it has different affinities was observed (Fig. 6A). All CD11bϩ cells were found to stain; i.e., for binding the various hsps. The ability of a ligand to compete no heterogeneity in the population was observed. No staining was with itself was shown directly in case of gp96-APC interaction; observed when FITC-gp96 was incubated with P815 cells, an ob- hsp90 and hsp70 were tested for competition with gp96 but not servation consistent with the FACS data in Fig. 3, that CD11bϪ with themselves. Thus, the data speak most completely for gp96 cells do not interact with hsp. Also, no staining of CD11bϩ cells (and by implication, hsp90) and its receptor(s) on APCs. The stud- was observed with FITC-SA. The specificity of the interaction of ies reported by Fujihara and Nadler (22), Basu et al. (23), and Asea hsps with APC observed by FACS analysis was thus confirmed by et al. (24) have provided preliminary evidence for the existence of confocal analysis. Similar to gp96, incubation of FITC-hsp90 or hsp70 receptors on APCs. FITC-hsp70 led to peripheral staining of CD11bϩ cells but not to In addition to the above, the experiments shown here hint to CD11bϪ P815 cells (Fig. 6B). some other novel aspects of hsp-APC interaction. They suggest To further characterize the uptake, live unfixed CD11bϩ cells that gp96 and hsp90 share a receptor and that hsp70 may have a were incubated with FITC-gp96 and were incubated at 25°C for 0, distinct receptor, but it may also interact with the gp96/hsp90 re- 5, 10, 15, 20, or 30 min. FITC-gp96 was visualized by confocal ceptor at relatively higher concentrations. This result is consistent microscopy to first bind to the cell surface rather uniformly (0 and with the close phylogenetic relationship between gp96 and hsp90 The Journal of Immunology 2587 and with the lack of homology between them and hsp70 (9, 25, 9. Nicchitta, C.V. 1998. Biochemical, cell biological and immunological issues sur- 26). Conserved regions in gp96 and hsp90 are most likely to be rounding the endoplasmic reticulum chaperone GRP94/gp96. Curr. Opin. Immu- nol. 10:103. responsible for binding to this receptor. Secondly, our present re- 10. Suto, R., and P. K. Srivastava. 1995. A mechanism for the specific immunoge- sults suggest that there is no obvious heterogeneity in the CD11bϩ nicity of heat shock protein-chaperoned peptides. Science 269:1585. population with respect to binding any of the three hsps. This is in 11. Ciupitu, A. M. T., M. Petersson, C. L. O’Donnall, K. Williams, S. Jindal, R. Kiessling, and R. M. Welsh. 1998. Immunization with a lymphocytic chorio- contrast to the suggestion from our previous study in which only a meningitis virus peptide mixed with heat shock protein 70 results in protective small subpopulation of CD11bϩ cells were able to re-present antiviral immunity and specific cytotoxic T lymphocytes. J. Exp. Med. 187:685. 12. Arnold, D., S. Faath, H. G. Rammensee, and H. Schild. 1995. Cross-priming of gp96-chaperoned peptides (10). The differences between the two minor histocompatibility antigen-specific cytotoxic T cells upon immunization studies may lie in the events further downstream of binding such with the heat shock protein gp96. J. Exp. Med. 182:885. that binding of hsps to the cells may be necessary but not sufficient 13. Arnold, D., C. Wahl, S. Faath, H. G. Rammensee, and H. Schild. 1997. Influences of transporter associated with antigen processing (TAP) on the repertoire of pep- for re-presentation of hsp-chaperoned peptides. This remains to be tides associated with the endoplasmic reticulum-resident stress protein gp96. examined. Structural characterization of the hsp receptor(s) is the J. Exp. Med. 186:461. obvious next step, which will provide a degree of finality to these 14. Udono, H., D. L. Levey, and P. K. Srivastava. 1994. Cellular requirements for tumor-specific immunity elicited by heat shock proteins: tumor rejection antigen suggestions and, hopefully, open a new chapter in our understand- gp96 primes CD8ϩ T cells in vivo. Proc. Natl. Acad. Sci. USA 91:3077. ing of the role of APCs in priming of specific T cell responses. 15. Srivastava, P. K., H. Udono, N. E. Blachere, and Z. Li. 1994. Heat shock proteins transfer peptides during antigen processing and CTL priming. Immunogenetics Note added in proof. A receptor for heat shock protein gp96 has re- 39:93. cently been identified as CD91 (27). 16. Binder, R. J., A. Menoret, and P. K. Srivastava. 1998. Events involved in re- presentation of heat shock protein-chaperoned peptides. Cell Stress Chaperones

3:2. Downloaded from Acknowledgments 17. Wassenberg, J. J., C. Dezfulian, and C. V. Nicchitta. 1999. Receptor mediated We thank Sreyashi Basu, Kirsten Anderson, Toyoshi Matsutake, and Thiru and fluid phase pathways for internalization of the ER Hsp90 chaperone GRP94 Ramalingam (all of our laboratory) for reading the manuscript critically in murine macrophages. J Cell Sci. 112:2167. 18. Arnold-Schild D, D. Hanau, D. Spehner, C. Schmid, HG. Rammensee, and and Caroline Goupille for her input on the FACScan data. H. de la Salle, and H. Schild. 1999. Receptor-mediated endocytosis of heat shock proteins by professional antigen-presenting cells. J. Immunol. 162:3757. References 19. Peng, P., A. Menoret, and P. K. Srivastava. 1997. Purification of immunogenic heat shock protein 70-peptide complexes byADP-affinity chromatography. J. Im-

1. Blachere, N. E., Z. Li, R. Y. Chandawarkar, R. Suto, N. S. Jaikaria, S. Basu, http://www.jimmunol.org/ munol. Methods 204:13. H. Udono, and P. K. Srivastava. 1997. Heat shock protein-peptide complexes, reconstituted in vitro, elicit peptide-specific cytotoxic T lymphocyte response and 20. Udono, H., and P. K. Srivastava. 1994. Comparison of tumor-specific immuno- tumor immunity. J. Exp. Med. 186:1315. genicities of stress-induced proteins gp96, hsp90, and hsp70. J. Immunol. 152: 2. Srivastava, P. K., and M. R. Das. 1984. Serologically unique surface antigen of 5398. a rat hepatoma is also its tumor-associated transplantation antigen. Int. J. Cancer 21. Denis, M. 1988. Two-step purification and N-terminal amino acid sequence anal- 33: 417. ysis of the rat Mr 90,000 heat shock protein. Anal. Biochem. 173:405. 3. Srivastava, P. K., A. B. DeLeo, and L. J. Old. 1986. Tumor rejection antigens of 22. Fujihara, S. M., and S. G. Nadler. 1999. Intranuclear targeted delivery of func- chemically induced sarcomas of inbred mice. Proc. Natl. Acad. Sci. USA 83: tional NF-kB by 70 kDa heat shock protein. EMBO J. 18:411. 3407. 23. Basu S., R. J. Binder, S. Suto, and P. K. Srivastava. 2000. Mammalian heat shock 4. Ullrich, S. J., E. A. Robinson, L. W. Law, M. Willingham, and E. A. Appella. proteins signal cell death through activation of the NF-kB pathway in antigen 1986. Mouse tumor-specific transplantation antigen is a heat shock-related pro- presenting cells. FASEB J. 14:A945.

tein. Proc. Natl. Acad. Sci. USA 83:3121. 24. Asea, A., S.-K. Kraeft, E. A. Kurt-Jones, M. A. Stevenson, L. B. Chen, by guest on September 25, 2021 5. Palladino, M.A., P. K. Srivastava, H. F. Oettgen, and A. B. DeLeo. 1987. Ex- R. W. Finberg, G. C. Koo, and S. K. Calderwood. 2000. hsp70 stimulates cyto- pression of a share tumor-specific antigen by two chemically induced BALB/c kine production through a CD14-dependent pathway, demonstrating its dual role sarcomas. I. Detection by a cloned cytotoxic T cell line. Cancer Res. 47:5074. as a chaperone and cytokine. Nat. Med. 6:435. 6. Udono, H., and P. K. Srivastava. 1993. Heat shock protein 70-associated peptides 25. Csermely, P., T. Scnaider, C. Soti, Z. Prohaszka, and G. Nardai. 1998. The 90- elicit specific cancer immunity. J. Exp. Med. 178:1391. kDa molecular chaperone family: structure, function, and clinical applications: a 7. Tamura, Y., P. Peng, K. Liu, M. Daou, and P. K. Srivastava. 1997. Immunother- comprehensive review. Pharmacol. Ther. 79:129. apy of tumors with autologous tumor-derived heat shock protein preparations. 26. Gupta, R. S. 1995. Phylogenetic analysis of the 90kD heat shock family of protein Science 278:117. sequences and an examination of the relationship among animals, plants, and 8. Janetzki, S., N. E. Blachere, and P. K. Srivastava. 1998. Generation of tumor- fungi species. Mol. Biol. Evol. 12:1063. specific cytotoxic T lymphocytes and memory T cells by immunization with 27. Binder, R. J., D. K. Han, and P. K. Srivastava. 2000. CD91: a receptor for tumor-derived heat shock protein gp96. J. Immunother. 21:269. heat-shock protein gp96. Nat. Immunology. In press.