Chaperoning Function of Stress Protein grp170, a Member of the hsp70 Superfamily, Is Responsible for its Immunoadjuvant Activity Jun-Eui Park, 1 John Facciponte, 2 Xing Chen, 1 Ian MacDonald, 1 Elizabeth A. Repasky, 2 Masoud H. Manjili, 4 Xiang-Yang Wang, 1,3 and John R. Subjeck 1 Departments of 1 Cell Stress Biology, 2 Immunology, and 3 Urologic Oncology, Roswell Park Cancer Institute, Buffalo, New York; and 4 Department of Microbiology and Immunology, School of Medicine, Virginia Commonwealth University, Massey Cancer Center, Richmond, Virginia Abstract When used as vaccines, tumor-derived stress proteins can elicit antitumor immune responses. For members of the hsp70 superfamily, like grp170, this seems to be due to ( a ) the chaperoning of antigenic peptide by the stress protein and (b) the binding of the stress protein to receptor(s) on antigen- presenting cells (APC) and subsequent antigen presentation. This suggests that domains exist on the stress protein for each function. In this study, we determine the ability of grp170 and its structural domains to ( a ) bind to and present melanoma-associated antigen gp100 to the immune system and (b) to bind to receptors on APCs. A direct correlation between chaperone function, binding to APCs in a receptor- like manner, and antitumor immunity was observed. Two mutants that share no common sequence, yet are both effec- tive in their antitumor activities, compete with one another for APC binding. Studies of other members of the hsp70 super- family, hsp110 and hsp70, or their domain deletion mutants, further confirmed that APC binding segregates with chaper- oning function and not sequence. Therefore, these studies suggest that molecular chaperoning is involved in stress pro- tein interactions with APCs, antigen binding, and in eliciting antitumor immunity, thus bridging this ancient function of stress proteins in prokaryotes to their ability to elicit immu- nity in higher organisms. (Cancer Res 2006; 66(2): 1161-8) Introduction Several studies have examined stress proteins/heat shock proteins (HSP) as tumor rejection antigens, most notably, hsc70 and grp94/gp96 (1–3). In some instances, exogenous stress proteins seem to act as vehicles for the delivery of antigens to professional antigen-presenting cells (APC) resulting in cross-priming. It has been shown that hsp70 and hsp90 family members can interact with various receptors on APCs leading to HSP-peptide uptake and antigen cross-priming (4–6), secretion of pro-inflammatory cyto- kines (7, 8), and maturation of dendritic cells (9). Thus, the adjuvant activity of some stress proteins seems to be several-fold in that they induce both innate and adaptive immunity. The fact that stress proteins of entirely different sequence each possess similar immune functions suggests that a fundamental underlying property of the HSP is involved. Stress proteins are molecular chaperones, and during stress (e.g., heat shock), act to inhibit the aggregation of other damaged proteins and, in concert with other chaperones, can often refold and reactivate damaged proteins. Molecular chaperones also participate in numerous normal cellular processes such as protein folding, transport, and peptide processing and trafficking (10, 11). The cellular functions of chaperones are essential to all living organisms from prokaryotes to man (12, 13). Grp170 is a major stress protein/molecular chaperone resident in the endoplasmic reticulum (ER; refs. 14–17) that is induced by stress conditions such as hypoxia, ischemia, and interference in calcium homeostasis (18). Studies have shown that grp170 is associated with the folding/ processing of secretory proteins such as thyroglobulin and immunoglobulin chains (14, 19), suggesting that it may be involved in protein/peptide import into the ER (20–23). We have recently described a novel approach to HSP vaccine formulation that uses the potent chaperoning property of hsp110 to form natural chaperone complexes with denatured protein antigens by heat shock (24, 25). Although heat is used as the denaturant of the mature protein in these studies, such chaperone complexes reflect folding and transport intermediates with nascent proteins characteristic of the natural functions of some molecular chaperones (26). In a recent study, a chaperone complex of hsp110 and the melanoma differentiation antigen gp100 was shown to activate antigen-specific T cell responses, leading to growth inhib- ition of B16 melanoma tumors (25). Importantly, these studies showed that it is the heat-induced chaperone complex itself that is required for immunologic activity and that other preparations containing hsp110 and gp100 (e.g., mixing without heat) show no antitumor activity. In this chaperone complex with gp100, mouse hsp110 (in mouse) was more effective as an adjuvant than was complete Freund’s adjuvant (CFA), suggesting potentially signifi- cant clinical applications. When purified from tumors, grp170 has been identified as a stress protein that can elicit antitumor immune responses. Immunization with tumor-derived grp170 can elicit tumor-specific CD8 + T cell responses and also significantly reduce pulmonary metastatic disease (27, 28). In order to gain insight into the mechanisms of grp170 immunogenicity, we now characterize complexes of gp100 with full-length grp170 and with mutants of grp170 that lack one or more of its structural domains (29). We continue to use melanoma-associated antigen gp100 as the substrate protein for chaperoning, paralleling our previous studies with hsp110, due to the potential clinical significance of this antigen (30, 31). We show here that the full-length grp170-gp100 chaperone complex generates an effective antigen-specific antitumor immu- nity in vivo . In addition, domain deletion mutants of grp170 in a complex with gp100 also elicit a potent antitumor immune Requests for reprints: John Subjeck or Xiang-Yang Wang, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263. Phone: 716-845-3147 or 716-845-2375; Fax: 716-845-8899; E-mail: john.subjeck@roswellpark.org or xiang-yang. wang@roswellpark.org. I2006 American Association for Cancer Research. doi:10.1158/0008-5472.CAN-05-2609 www.aacrjournals.org 1161 Cancer Res 2006; 66: (2). January 15, 2006 Research Article