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


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Robert, J.
Right arrow Articles by Cohen, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Robert, J.
Right arrow Articles by Cohen, N.
The Journal of Immunology, 1999, 163: 4133-4139.
Copyright © 1999 by The American Association of Immunologists

Cell Surface Expression of the Endoplasmic Reticular Heat Shock Protein gp96 Is Phylogenetically Conserved1

Jacques Robert2,*, Antoine Ménoret{dagger} and Nicholas Cohen*

* University of Rochester Medical Center, Rochester, NY 14642; and {dagger} Center for Immunotherapy of Cancer and Infectious Diseases, University of Connecticut School of Medicine, Farmington, CT 06030


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
In mammals, the heat shock protein gp96 complexed to antigenic peptides elicits T cell adaptive immunity. By itself, however, gp96 can evoke responses that are characteristic of innate immunity. Interestingly, this protein, which resides in the endoplasmic reticulum, is expressed on the surface of certain mouse tumor cells. Given that heat shock proteins are highly conserved, we investigated whether the cell surface expression of gp96 is also evolutionarily conserved. Our data reveal that gp96, most likely containing the endoplasmic reticulum retention motif (KDEL), is expressed on the surface of three different Xenopus lymphoid tumor cell lines, each derived from a different spontaneously arising thymic tumor. Levels of expression differ among the tumor lines tested, with more immunogenic tumors expressing greater amounts of surface gp96. Moreover, a high level of gp96 surface expression is detectable on a subset of Xenopus normal nontransformed splenic lymphocytes (mainly surface IgM+ B cells) but not on other normal cells, including macrophages and nucleated erythrocytes. Surface expression of a gp96 protein homologue occurs also on some, but not all, T and B cell clones derived from peripheral blood cells of the channel catfish, as well as on lymphocyte-like cells, but not on erythrocytes from the hagfish, a primitive agnathan vertebrate lacking markers of an adaptive immune system. gp96 is actively directed to and retained on the plasma membrane of Xenopus lymphocytes and tumor cells and hagfish lymphocyte-like cells by a process that requires vesicular transport. This selective surface expression of gp96 on some immune cells from different vertebrate classes is consistent with an ancestral immunological role of gp96 as danger-signaling molecule.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Glycoprotein 96 (gp96)3 (glucose-regulated protein 94 (grp94)/grp100/endoplasmic reticulum protein 99 (Erp99)/endoplasmin) is member of the heat shock protein (hsp) family, which consists of numerous sets of highly conserved intracellular proteins (1). Each hsp is localized in a restricted compartment: the cytosol and nucleus for hsp28, hsp70, and hsp90; the mitochondria for hsp60 and a subset of hsp70; and the endoplasmic reticulum (ER) for binding Ig protein (BiP) (grp78) and gp96 (grp94). Because hsps do not contain a transmembrane domain, they are, from a genetic and biochemical point of view, considered intracellular soluble proteins rather than integral surface membrane proteins. Like other proteins residing in the ER, gp96 contains a carboxy terminal KDEL sequence (Lys-Asp-Glu-Leu) that constitutes a feedback signal from the Golgi to the ER (2). Despite this KDEL sequence, Altmeyer et al. (3) have demonstrated surface expression of gp96 on mouse Meth-A sarcoma cells but not on normal embryonic mouse fibroblasts. Other hsps, such as hsp70 (4, 5, 6), BiP (7), and hsp90 (4, 8), are similarly expressed at the cell surface of a variety of human and murine tumor cells, but not on the limited number of transformed or normal cells that have been examined (human EBV-transformed B cells, peripheral blood leukocytes, and fibroblasts (4, 9)). Recently, however, a report of surface expression of several ER-resident molecular chaperones, including gp96, on a subset of murine immature thymocytes (10) has revealed that this phenomenon is not restricted to tumor cells.

Considering the proposed dual involvement of gp96 both in innate and adaptive immunity (11), the selective expression of hsps such as gp96 on the surface of some cells may be of immunological relevance. In mice, gp96 elicits MHC class I-restricted CD8+ cytotoxic T cell adaptive immune responses against the antigenic peptides it chaperones (12, 13, 14). In the absence of chaperoned peptides, however, gp96 can activate macrophages (13, 15) and T cells (16) to secrete cytokines such as IFN-{gamma} and TNF-{alpha}; in this sense, gp96 per se may be considered a "danger molecule" (17) that is relevant for innate rather than adaptive immunity. Modulated access of intracellular hsps such as gp96 on the cell surface may play a role in such signaling.

To begin to explore the biological relevance of this cell surface expression of gp96 from a phylogenetic perspective, we examined different cell types from several vertebrate species for their surface expression of gp96. This study reveals that in addition to being actively expressed at the surface of some normal amphibian (Xenopus) and teleost (catfish) lymphocytes, gp96 is also localized on the surface of lymphoid-like cells from the hagfish, a member of the only vertebrate class (Agnatha) that appears to lack an adaptive immune system.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals and cell lines

Peripheral blood from euthanized Pacific hagfish (Eptatretus stoutii) obtained from Marinus (Long Beach, CA) was collected from the caudal sinus (18). Hagfish hemopoietic pronephric tissue was treated mildly with 2% type IA collagenase (C-2674, Sigma, St. Louis, MO), and dissociated cells were collected in physiological saline. Kidney mononuclear leukocytes (5–10 x 106 cells/ml) that microscopically resembled lymphocytes from higher vertebrates were separated from erythrocytes by centrifugation for 30 min at 1500 x g at 4°C on a 28% Percoll cushion (hagfish osmolarity). {alpha}ßT cell lines and surface IgM+ (sIgM+) B cell lines from channel catfish (Ictalurus punctatus) were kindly provided by Dr. Norman Miller (University of Mississippi Medical Center, Jackson, MS) (19). Freshly harvested Xenopus spleens were dissociated, and lymphocytes were purified with Histopaque 1.083 (Sigma); the three lymphoid tumor cell lines (B3B7, ff-2, and 15/0), which were derived from spontaneously arising thymic tumors in different strains or clones of adult Xenopus, were maintained in vitro (20, 21) before analysis. Isomolar PBS for each species was prepared from mammalian PBS.

Monoclonal Abs

The following Xenopus-specific mouse mAbs were used: 10A9 directed against IgM (22); X71, specific for CTX (a cortical thymocyte-specific Xenopus cell surface receptor that is also present on the tumor cells used in this study (23)); AM22, specific for CD8-equivalent Ag (24); and TB17, an anti-MHC class I mAb (25). The anti-actin mouse mAb AC-40 was obtained from Sigma, the mouse mAb 10C3 specific for the KDEL C-terminal ER retention signal was purchased from StressGen (Biotechnologies, Victoria, British Columbia, Canada), and the rat mAb SPA-850 specific for gp96 (clone 9G10) was purchased from Neomarkers (Fremont, CA). This Ab recognizes a Xenopus glycoprotein of 98 kDa that can be purified using the same protocol for mouse gp96 (12, 14); four of the five carboxy-terminal amino acid residues of this 98-kDa glycoprotein are identical with those of mouse gp96 (our unpublished observations).

Flow cytometry

Samples of 105 cells were incubated with hybridoma supernatants or purified Abs followed by fluorescein-labeled goat anti-mouse or rabbit anti-rat IgG F(ab')2 (Southern Biotechnology Associates, Birmingham, AL), preadsorbed twice on Xenopus erythrocytes, and analyzed on an Elite flow cytometer (Beckman Coulter, Fullerton, CA) (23). Dead cells, detected by propidium iodide, were gated out. For two-color flow cytometry, spleen cells were first stained with rat anti-gp96 mAb followed by FITC-conjugated goat anti-rat (F(ab')2) Ab (Southern Biotechnology Associates) that had been preadsorbed on mouse and Xenopus cells. Cells were then stained with Xenopus anti-IgM 10A9 mAb followed by PE-conjugated goat anti-mouse (F(ab')2) Ab preadsorbed on rat and Xenopus cells. The two secondary Abs are not cross-reactive as certified by the supplier (Southern Biotechnology Associates); this was confirmed using isotype-matched rat and mouse control Abs.

Cell surface labeling and immunoprecipitation

Procedures for cell surface biotinylation, lysis in Nonidet P-40, and immunoprecipitation with protein G have been detailed elsewhere (21). Before and after labeling, cells were extensively washed three times in PBS (at the appropriate amphibian or fish osmolarity) that contained 1% BSA. Cell death, determined before lysis by trypan blue dye exclusion, was never >5%. Biotinylated cell surface lysates (corresponding to ~5 x 107 cells) were preincubated for 1 h at 4°C with 30 µl/ml of protein G. A total of 100 µl of such precleared lysates (corresponding to ~5 x 106 cells) were incubated overnight at 4°C either with 100 µl of mAb supernatants and 30 µl of protein G or with a mixture of 3 µl of anti-gp96 mAb, 3 µl of rabbit anti-rat Ab (Sigma), and 30 µl of protein G. Immunoprecipitates were separated on 7.5% SDS-PAGE gels under reducing conditions and transferred onto polyvinylidene difluoride (PVDF) membranes (Bio-Rad, Hercules, CA). Biotinylated proteins were revealed using HRP-conjugated streptavidin and chemiluminescence reagents from Amersham (Arlington Heights, IL). Nonbiotinylated proteins were detected after reprobing the membrane with specific Ab followed by a secondary rabbit anti-rat HRP-conjugated Ab. Enrichment of IgM+ splenic lymphocytes after surface biotinylation was performed using sheep anti-mouse magnetic beads adsorbed with Xenopus anti-IgM 10A9 mAb according to the manufacturer’s suggested protocol (Dynal, Fort Lee, NJ).

Cell surface re-expression assay

The protocol described by Wiest et al. (10) was used. Briefly, cells were incubated with pronase (0.4 mg/ml final concentration, P-6911, Sigma) for 45 min at 26°C with occasional agitation; digestion was quenched for 10 min on ice with 2.5% BSA (final concentration) and 10 mg/ml of DNase (final concentration). Cells were washed once with 5% BSA in PBS and then incubated in 1% BSA in PBS with 0.1 mM PMSF and 0.05 mM TLCK N{alpha}-p-tosyl-L-ysine chloromethyl ketone (both protein inhibitors) for 10 min on ice. After an additional wash, cells were put back in culture for 4 h at 26°C in either medium alone (for hagfish, Iscove-derived mammalian medium supplemented with 1.2% NaCl, 5% FBS, and 5% hagfish serum) or medium with 1 µg/ml of brefeldin A (BFA).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Surface expression of gp96 on tumor cells

Our previous characterization of the 15/0, ff-2, and B3B7 Xenopus tumors (20, 26, 27) and the present studies suggest a correlation between gp96 surface expression and immunogenicity (see Table IGo). The level of gp96 surface expression of the different Xenopus lymphoid tumor cell lines derived from independent spontaneously arising thymic tumors was analyzed by flow cytometry. Each of the three lymphoid tumor cell lines showed a reproducible but different level of gp96 surface expression (Fig. 1GoA). The least immunogenic tumor, 15/0, expresses the lowest level of surface gp96, whereas the highly immunogenic B3B7 tumor line expresses the most. The ff-2 cell line, which has an intermediate level of gp96 surface expression, is tumorigenic in larval (or less immunogenic) hosts but not in fully grown adult MHC identical hosts. Furthermore, tumor rejection of ff-2 cells is mediated by a thymus-dependent immune response against tumor-specific Ags that develop during metamorphosis (29). Another tumor clone (15/40, clone A1) that recently lost its tumorigenicity displays levels of surface gp96 that are as high as the immunogenic B3B7 tumor (our unpublished observations). It is noteworthy that of all the tumor lines examined, only ff-2 cells express surface MHC class I Ags, albeit at low levels. In addition, erythrocytes and macrophages that both express surface MHC class I did not stain with anti-gp96 mAb (Fig. 1GoB).


View this table:
[in this window]
[in a new window]
  		Table I. Characteristics of three different Xenopus thymic tumor cell lines

 


View larger version (28K):
[in this window]
[in a new window]
  		FIGURE 1. Flow cytometry analysis of Xenopus tumoral and normal cell lines. 15/0, ff-2, and B3B7 tumor cells (top panel) as well as normal erythrocytes, peritoneal macrophages, and splenic lymphocytes (bottom panel) were stained with anti-gp96 mAb, anti-Xenopus MHC class I mAb, and anti-Xenopus CD8 mAb. The negative control for gp96 was obtained by replacing anti-gp96 mAb with the same amount of control rat IgG2a isotype, and for the two other mAbs by using nonreactive mouse mAb (23 ). FITC-conjugated goat anti-rat or goat anti-mouse (F(ab')2) Abs preadsorbed twice on Xenopus blood cells were used as secondary Abs. Dead cells stained by propidium iodine were gated out.

 
Surface gp96 was also detectable by immunoprecipitation of surface-biotinylated 15/0 tumor cells. A major protein of 98–99 kDa was specifically detected in the sample immunoprecipitated with a mAb specific for gp96, but not in control samples immunoprecipitated with purified rat IgG or rat serum (Fig. 2GoA). The possibility that the biotinylation may have artifactually labeled intracellular proteins released from a minor fraction of dead cells that are ubiquitous in cultures is unlikely, because an anti-actin mAb did not precipitate any biotinylated material (data not shown). Because gp96 is normally a resident protein of the ER and contains the retention signal KDEL, we wondered whether surface expression of gp96 could have resulted from the absence of this signal caused by C-terminal degradation of a fraction of the gp96 protein. Immunoprecipitation of surface-biotinylated 15/0 tumor lysates with anti-KDEL mAb revealed weak but reproducible surface expression of intact gp96 (Fig. 2GoB) and two other KDEL-expressing proteins of 78 kDa and 160–160 kDa that might be the two hsps, BiP (grp78) and grp170, respectively (30). Interestingly, immunoblotting of 15/0 total cell lysates with the anti-KDEL Ab revealed at least four other proteins of 160–180, 60, 50, and 40 kDa. It appears that in addition to gp96, some, but not all, ER-resident proteins end up at the cell surface. Results similar to those obtained with 15/0 cells were also recorded for the B3B7 and ff-2 tumor cell clones (data not shown). The extensive washes before and after biotinylation were designed to eliminate any adventitious deposition of gp96 released in the medium by dead cells during the in vitro culture. To completely eliminate this possibility, 15/0 cells were incubated with an excess of biotinylated cell lysate (obtained by successive freeze/thaw cycles to avoid the use of detergent). No signal was detectable for cells incubated with biotinylated lysates after immunoprecipitation, whereas large amounts of biotinylated gp96 were detectable from the lysate (Fig. 2GoC).



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 2. Immunoprecipitation of surface-biotinylated Xenopus cell lysate. A, Surface-biotinylated lysate of 5 x 106 15/0 cells was precipitated with the same amount of anti-gp96 mAb, purified rat IgG, normal rat serum, and anti-Xenopus CTX mAb. Cell death was <5% as determined by trypan blue exclusion. Immunoprecipitates were separated on 7.5% SDS-PAGE gels under reducing conditions and transferred onto PVDF membranes. Biotinylated surface proteins were revealed using HRP-conjugated streptavidin and chemiluminescence reagents. B, Immunoprecipitation of surface-biotinylated lysate with anti-KDEL mAb specific for the ER retention signal; the general pattern of proteins with a carboxy-terminal KDEL obtained by immunoblotting total cell lysate is shown for comparison. C, The total biotinylated protein extract of lysate of 5 x 106 15/0 cells from which biotinylated gp96 can be precipitated (filled arrow) was incubated with 5 x 106 unlabeled 15/0 cells for 2 h at 26°C. Biotinylated gp96 could not be detected in the lysate from these cells after the usual washing protocol (*), indicating no adventitious deposition. Unbiotinylated gp96 was detectable after washing the same membrane and reprobing with anti-gp96 mAb and HRP-conjugated rabbit anti-rat Ab (open arrow).

 
Because passive loading from free extracellular gp96 can be ruled out as a source of surface gp96, we subsequently examined whether gp96 surface expression requires cell integrity by monitoring gp96 reappearance on the cell surface after complete digestion of all surface proteins with pronase (Fig. 3Go). After such a treatment (that also eliminates dead cells), surface gp96 was undetectable (Fig. 3Go); after 4 h, however, it could again be detected. Similar re-expression at 4 h after extracellular digestion was monitored by flow cytometry (data not shown) of cells stained for CTX, a well-characterized Xenopus transmembrane protein (23). In addition, surface re-expression of gp96 was abrogated in the presence of BFA (Fig. 3Go), which indicates that the presence of gp96 on the cell surface results from an active process that is dependent upon the translocation of protein from the ER to the Golgi. Note that treatment with pronase and BFA does not affect the overall abundance of gp96, because similar amounts of gp96 could be detected in lysates by Western blotting with anti-gp96. Similar BFA-sensitive re-expression after pronase treatment was obtained with the B3B7 tumor line (data not shown).



View larger version (39K):
[in this window]
[in a new window]
  		FIGURE 3. In vitro monitoring of gp96 re-expression after complete digestion of surface proteins with pronase. Left panel, Immunoprecipitation of lysate from class I-negative 15/0 lymphoid tumor cells: surface-biotinylated after pronase treatment; mock treatment as positive control; pronase treatment followed by 4 h of culture in the presence of 1 µg/ml BFA; or pronase treatment followed by 4 h of culture in medium alone. Right panel, Immunoprecipitation of Xenopus normal total spleen lymphocytes: surface-biotinylated after pronase treatment, mock treatment; pronase treatment followed by 4 h of culture in medium alone; or pronase treatment followed by 4 h of culture in medium containing 1 µg/ml BFA. Note that treatment with pronase and BFA does not affect the overall abundance of gp96, because similar amounts of gp96 could be detected in lysates by Western blotting with anti-gp96.

 
A final argument that the precipitation of surface gp96 from 15/0 cells is not an artifact of biotinylation or cross-reaction but rather is a selective phenomenon restricted to some type and/or differentiation step of cells is that no such signal could be obtained from the surface labeling of normal Xenopus erythrocytes, peritoneal macrophages (Figs. 1Go and 4Go), and most splenic lymphocytes (Fig. 1Go). Indeed, Xenopus erythrocytes that are nucleated and express surface MHC class I Ags also do not express surface gp96 (Figs. 1Go and 4Go). In immunoprecipitation experiments, reprobing membranes with the anti-gp96 Ab revealed a significant amount of precipitated intracellular unbiotinylated gp96 (Fig. 4Go, bottom panel).



View larger version (32K):
[in this window]
[in a new window]
  		FIGURE 4. A total of 5 x 106 surface-biotinylated cells from adult outbred Xenopus were immunoprecipitated using purified rat IgG, anti-gp96 mAb, and anti-Xenopus class I mAb (25 ). For erythrocytes and macrophages, the same membrane was washed and reprobed with anti-gp96 mAb and HRP-conjugated rabbit anti-rat Ab.

 
Surface expression of gp96 on a subpopulation of immune cells

Interestingly, surface gp96 was detected by immunoprecipitation of surface-biotinylated splenocytes. This signal was weaker than that of the tumor cells, but it has been obtained several times and has been confirmed by further analysis. Furthermore, in vitro monitoring of surface re-expression after complete digestion of surface protein by pronase (Fig. 3Go) revealed that as in the case of tumor cells, gp96 rapidly reappeared at the lymphocyte cell surface by a BFA-sensitive mechanism. The absence of signal in cells cultured in the presence of BFA further rules out a possible contribution of dead cells.

Flow cytometric analysis of splenocytes revealed that although the majority of lymphocytes do not stain with anti-gp96 mAb, a minor (5–9%) subset was strongly positive (about a two-log difference from the rat IgG control; Fig. 1Go). When viewed microscopically, these viable gp96+ small lymphocytes exhibited typical punctuate surface staining. Further analysis by two-color flow cytometry revealed that the majority (>90%) of these gp96+ small lymphocytes were also sIgM+ (Fig. 5GoA). No significant staining of this gp96+ IgM+ cell subset was obtained with either Xenopus anti-CD8 mAb or anti-CD5 mAb, both of which exclusively detect Xenopus T cells. To test whether this positive surface staining was the result of artifactual deposition of cytoplasmic gp96 and/or IgM released by dead cells, spleen cell suspensions were kept on ice for 2 h before the washing and staining steps. Although this treatment increased the numbers of dead cells, it did not increase the fraction of surface gp96+ splenocytes. Instead, it resulted in a decrease of ~25% of surface gp96+ cells (data not shown). Because spleens also contain a large number of macrophages, we purified them by adherence on a glass microscope coverslide and stained them with either anti-gp96, anti-class I, or class II mAbs followed by an FITC-conjugated secondary Ab before fixation with 2% paraformaldehyde. No fluorescence signal was detected with anti-gp96, whereas a bright surface staining pattern was obtained with both anti-MHC mAbs (data not shown). To further substantiate the two-color flow cytometric data, IgM+ spleen lymphocytes were positively selected with anti-IgM-coated magnetic beads after surface biotinylation. The amount of surface gp96 precipitated from this enriched IgM+ subset was markedly increased relative to the total or partially depleted splenocyte populations (Fig. 5GoB), whereas similar signals in all samples were detected by immunoprecipitation with Xenopus-specific anti-class I mAb.



View larger version (32K):
[in this window]
[in a new window]
  		FIGURE 5. Expression of gp96 at the surface of IgM+ B spleen cells in Xenopus. A, Two-color flow cytometry of spleen cells stained with rat anti-gp96 mAb followed by FITC-conjugated goat anti-rat (F(ab')2) Ab (preadsorbed on mice and Xenopus cells) and with Xenopus anti-IgM 10A9 mAb followed by PE-conjugated goat anti-mouse (F(ab')2) Ab (preadsorbed on rat and Xenopus cells). Note that the minor fraction (usually <1%) that was single positive for gp96 was also detected using rat IgG2a isotype as a negative control; therefore, this fraction is likely to be nonspecific or autofluorescent. B, Immunoprecipitation of surface-biotinylated gp96, MHC class I, and IgM was performed on total spleen cells, spleen cells depleted of sIgM-expressing cells, and spleen cells enriched for sIgM-expressing B cells as described in Fig. 3Go. Each lane represents the material precipitated from 7 x 106 cells.

 
Surface expression of gp96 on immune cells from other classes of vertebrate

If surface expression of gp96 on normal Xenopus lymphoid cells is biologically relevant, then we hypothesized that this hsp should be present on the surface of normal cells from representatives of phylogenetically more primitive vertebrates. To explore this possibility, we took advantage of the high degree of conservation of the epitope recognized by the anti-gp96 mAb (i.e., it reacts with a single monomeric protein of 96 kDa on Western blots in all vertebrate species tested, including agnathans (31)).4 Also, the same 96-kDa monomer can be detected by Western blotting with an anti-KDEL mAb. Immunoprecipitation of surface-biotinylated lysate with anti-gp96 mAb revealed surface expression of a gp96 homologue on the surface of some, but not all, T and B cell clones from the channel catfish (Fig. 6Go). In several experiments, this pattern of expression was reproducibly detected (i.e., surface positive and negative clones always expressed the same phenotype). This finding clearly indicates that gp96 surface expression is not merely the result of in vitro culture conditions but rather is restricted to certain cell types or stages of differentiation.



View larger version (28K):
[in this window]
[in a new window]
  		FIGURE 6. Expression of gp96 on the surface of some, but not all, catfish T and B cell clones. Surface-biotinylated lysates were immunoprecipitated from two different sIgM+ B cell clones (1G8, 3B11) and two {alpha}ßT cell clones (28S1, 43TA) from the channel catfish, I. punctatus. Results are representative of two different experiments

 
Agnathans, represented today by hagfish and lampreys, are the only known vertebrates for which components of the adaptive immune system (e.g., MHC, rearranging TCR, Ig) have not been identified despite repeated molecular approaches to identifying them (32). Because gp96 protein homologues can be detected with our anti-gp96 mAb, we analyzed the gp96 expression pattern of lymphocyte-like cells in the peripheral blood and the pronephros of the Pacific hagfish. Interestingly, a gp96 homologue containing the KDEL motif was clearly detectable on the surface of these cells (Fig. 7Go) but not on erythrocytes that are nucleated as they are in amphibians. Moreover, as in Xenopus, the gp96 hagfish homologue is loaded at the cell surface by an active mechanism that is sensitive to BFA (Fig. 7Go).



View larger version (29K):
[in this window]
[in a new window]
  		FIGURE 7. Expression of gp96 on the surface of hagfish lymphocyte-like cells but not erythrocytes. A, Immunoprecipitation of surface-biotinylated lysate from hagfish lymphocytes treated with streptavidin. (1) Mock-treated control; (2) pronase-treated; (3) pronase-treated followed by 4 h of culture in medium alone; or (4) pronase-treated followed by 4 h of culture in medium in the presence of BFA (1 µg/ml). Immunoprecipitation of surface-biotinylated lysate from hagfish lymphocyte-like cells and nucleated erythrocytes shows that gp96 is present only on the surface of lymphocytes. Immunoprecipitates were separated on 10% SDS-PAGE gels under nonreducing conditions and transferred onto PVDF membranes. Biotinylated surface proteins were revealed using HRP-conjugated streptavidin and chemiluminescence reagents. B, Same membrane as in A, reprobed with anti-gp96 mAb and rabbit anti-rat HRP-conjugated Ab to show that although there was no biotinylated gp96 (of surface origin) present in certain fractions, nonbiotinylated gp96 (of intracellular origin) was present in all fractions. C, Same membrane as in A, reprobed with anti-KDEL Ab, indicating the presence of the retention motif found on all ER-resident proteins.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The hsp gp96 has been described as a classical ER-resident protein whose expression on the surface of some mouse tumor cells has been considered atypical (3). Here, we show surface expression of gp96 on tumor cells of the frog, Xenopus. As in mice, the gp96 localized on the surface of Xenopus tumor cells is derived from the pool of endogenous gp96 synthesized by the cells, rather than from an adventitious deposition. The mechanism by which gp96 is directed and retained on the plasma membrane is still unknown. Detection of surface gp96 with a mAb specific for the C-terminal ER retention signal (KDEL) suggests that escape of C-terminal degraded gp96 is unlikely. The two other proteins of 78 kDa and 160–180 kDa are likely to be BiP (grp78) and grp170 (a newly identified ER protein), respectively. grp170 belongs to the hsp family, shows homology with both hsp70 and hsp110, binds ATP, and possesses a carboxyl terminal retention sequence (30).

The process by which gp96 is expressed at the cell surface was further characterized using freshly harvested Xenopus splenic lymphocytes as well as MHC class I-negative tumor lymphoid cell lines (B3B7, 15/0). Although digestion of surface proteins by pronase initially resulted in the loss of cell surface gp96, re-expression of this hsp at 26°C was detectable as early as 4 h after pronase treatment for both tumor cells and nontransformed splenocytes. Furthermore, this reappearance was abrogated by inhibiting protein translocation from the ER to the Golgi with BFA. These experiments rule out any contribution to surface expression by dead cells and suggest that surface expression results from an active process involving vesicular transport. It is noteworthy that this inhibitor interferes with re-expression of surface proteins (10) but not with internal labeling of cytosolic proteins in dying cells (33).

Our previous characterization of the 15/0, ff-2, and B3B7 Xenopus tumors and our present studies suggest a positive correlation between gp96 surface expression and immunogenicity (see Table IGo). No correlation between the immunogenicity of these three tumors and the level of surface MHC class I and class II molecules or other molecules was found. However, 15/0, the least immunogenic tumor, expresses the lowest level of surface gp96, whereas the highly immunogenic B3B7 tumor line expresses the most. The ff-2 cell line, which has an intermediate level of gp96 surface expression, is tumorigenic in larval but not in fully grown adult histocompatible hosts. Furthermore, tumor rejection of ff-2 cells is mediated by a thymus-dependent immune response against tumor-specific Ags that develop during metamorphosis (29). Another tumor clone (15/40.A1) that has recently lost its tumorigenicity expresses levels of surface gp96 that are comparable with the highly immunogenic B3B7 tumor (our unpublished observations).

Until recently, surface hsp expression has been reported only on transformed or virally infected mammalian cells (3, 4, 5). More recently, however, it has been shown that different ER-resident molecular chaperones are, in fact, expressed on the surface of immature murine thymocytes (10). Thus, this phenomenon does not appear to be restricted to transformed cells. We have used a comparative approach to further investigate this issue. Among the various normal Xenopus cell types examined by both immunoprecipitation and flow cytometry (i.e., nucleated erythrocytes, peritoneal macrophages, splenic macrophages, and splenic lymphocytes), surface gp96 was reproducibly detected only on a small fraction of splenic lymphocytes, which appear to be mainly sIgM+ lymphocytes. However, these results do not exclude other types of lymphocytes expressing surface gp96. The level of surface gp96 expression by splenic lymphocytes is 50–100 times greater than that on the different lymphoid tumor cells tested. Intuitively, expression of a high density of surface gp96 on lymphocytes argues for an immunological role of these molecules. This possibility is supported by our finding of a similar surface expression of gp96 protein homologues on cloned fish T and B cell clones derived from normal (i.e., nontransformed) lymphocytes (19) as well as on freshly harvested lymphocyte-like cells in the peripheral blood and pronephros of the agnathan hagfish. As in Xenopus, not all hagfish cells express surface gp96, indicating that this phenomenon is restricted to some cell types and/or to some stages of cell differentiation.

gp96, as well as other nonpolymorphic and highly conserved hsps of the ER, may chaperone antigenic peptides and present them to cells of the immune system (11, 13, 14). Indeed, soluble gp96-peptide complexes purified from murine tumors can specifically immunize mice against tumor challenges (12). We have recently reported the conserved ability of Xenopus gp96 to stably complex peptides noncovalently and the capacity of gp96, purified from Xenopus tumor but not normal tissue, to generate specific non-Ab-mediated MHC-unrestricted Xenopus antitumor immunity (31).4 Based on such data, gp96 has been proposed to play a role in Ag presentation and in the generation of classical adaptive immune responses (11, 34, 35). However, recent data suggest that in addition to this antigenic peptide-specific process, hsps such as gp96 per se can stimulate macrophages (13, 15) and thereby generate nonspecific defense responses that are mediated by enhanced cytokine production and nonspecific killing (16, 34, 35). gp96 expression can also be up-regulated by IFN-{gamma} (36) and "stressful" conditions (e.g., glucose deprivation). In this context, and given their ancient origin, gp96 and some other hsps have been proposed to play a role in innate immunity as molecular messengers of cell death (11, 37) or "danger" (17).

The activation of defense responses by hsp per se and/or hsp complexed to peptides is thought to be triggered by the release of hsp by dead or apoptotic cells. However, it is possible that modulated surface expression of such molecules may also play a role in this activation. It is with this in mind that we propose, and are currently testing, the hypothesis that gp96 surface expression represents a vestige of an ancestral system of Ag presentation and/or immune surveillance.

In summary, our results strongly suggest that cell surface expression of the ER-resident molecular chaperone gp96 is not unique to some mammalian tumors but rather, is reflective of a more general biological phenomenon that occurs on populations of normal lymphoid cell populations from phylogenetically diverse vertebrates, including the Agnatha, which lack an adaptive immune system.


   Acknowledgments
 
We thank Deborah M. Brown, Kara Jursak, Erika Torjusen, and Michael Sung for their contributions to this work. We also thank Nathalie Blachère, Daniel Levey, and Rajiv Chandawarkar for their collective criticisms and helpful discussions.


   Footnotes
 
1 Research at the University of Rochester was supported by National Institutes of Health Grants R37 HD-07901 and R01 CA-76312. Research at the University of Connecticut was supported by National Institutes of Health Grants RO1 CA-64394 and RO1 CA-44786 and by a research agreement with Antigenics, Inc. Back

2 Address correspondence and reprint requests to Dr. Jacques Robert, University of Rochester Medical Center, Department of Microbiology and Immunology, Box 672, 601 Elmwood Avenue, Rochester, NY 14642. E-mail address: Back

3 Abbreviations used in this paper: gp, glycoprotein; hsp, heat shock protein; ER, endoplasmic reticulum; PVDF, polyvinylidene difluoride; BFA, brefeldin A; sIgM, surface IgM; grp, glucose-regulated protein; BiP, binding IG protein. Back

4 J. Robert, A. Ménoret, S. Basu, N. Cohen, and P. K. Srivasta. The immunological properties of heat shock protein gp96 and hsp70 are phytogenetically conserved across distant species. Submitted for publication. Back

Received for publication May 18, 1999. Accepted for publication July 28, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Young, R. A.. 1990. Stress proteins and immunology. Annu. Rev. Immunol. 8:401.[Medline]
  2. Kock, G., M. Smith, D. Macer, P. Webster, R. Mortara. 1986. Endoplasmic reticulum contains a common, abundant calcium-binding glycoprotein, endoplasmin. J. Cell Sci. 86:217.[Abstract/Free Full Text]
  3. Altmeyer, A., R. Maki, A. M. Feldweg, M. Heike, V. Protopov, S. K. Masur, P. K. Srivastava. 1996. Tumor-specific cell surface expression of KDEL-containing, endoplasmic reticular heat shock protein gp96. Int. J. Cancer 69:340.[Medline]
  4. Ferrarini, M., S. Heltai, M. R. Xocchi, C. Rugarli. 1992. Unusual expression and localization of heat-shock proteins in human tumor cells. Int. J. Cancer 51:613.[Medline]
  5. Multhoff, G., C. Botzler, M. Wiesnet, E. Müller, T. Meier, W. Wilmann, R. D. Issels. 1995. A stress-inducible 72-kDa heat-shock protein (hsp72) is expressed on the surface of human tumor cells, but not on normal cells. Int. J. Cancer 61:272.[Medline]
  6. Vanbuskirk, A. M., B. L. Crump, E. Margoliash, S. K. Pierce. 1989. A peptide-binding protein having a role in antigen presentation is a member of the hsp70 family. J. Exp. Med. 170:1799.[Abstract/Free Full Text]
  7. Takemoto, H., T. Yoshimori, A. Yamamoto, Y. Myiata, I. Yahara, K. Inoue, Y. Tashiro. 1992. Heavy chain binding protein (BiP/GRP78) and endoplasmin are exported from the endoplasmic reticulum in rat exocrine pancreatic cells, similar to protein disulfide-isomerase. Arch. Biochem. Biophys. 296:129.[Medline]
  8. Ullrich, S. J., E. A. Robinson, L. W. Law, M. Willingham, E. Appella. 1986. A mouse tumor-specific transplantation antigen is a heat shock-related protein. Proc. Natl. Acad. Sci. USA 83:3121.[Abstract/Free Full Text]
  9. Multhoff, G., C. Botzler, M. Wiesnet, G. Eissner, R. Issels. 1995. CD3-large granular lymphocytes recognize a heat-inducible immunogenic determinant associated with the 72-kDa heat shock protein of human sarcoma cells. Blood 86:1374.[Abstract/Free Full Text]
  10. Wiest, D., A. Bhandoola, J. Punt, G. Kreibich, D. McKean, A. Singer. 1997. Incomplete endoplasmic reticulum (ER) retention in immature thymocytes as revealed by surface expression of "ER-resident" molecular chaperones. Proc. Natl. Acad. Sci. USA 94:1884.[Abstract/Free Full Text]
  11. Srivastava, P. K., A. Menoret, S. Basu, R. J. Binder, K. L. McQuade. 1998. Heat shock proteins come of age: primitive functions acquire new roles in an adaptive world. Immunity 8:657.[Medline]
  12. Udono, H., D. L. Levey, P. K. Srivastava. 1994. Cellular requirements for tumor-specific immunity elicited by heat shock proteins: tumor rejection antigen gp96 primes CD8+ T cells in vivo. Proc. Natl. Acad. Sci. USA 91:3077.[Abstract/Free Full Text]
  13. Suto, R., P. K. Srivastava. 1995. A mechanism for the specific immunogenicity of heat shock protein-chaperoned peptides. Science 269:1585.[Abstract/Free Full Text]
  14. Blachere, N. E., R. Y. Chandawarkar, R. Suto, N. S. Jaikaria, S. Basu, H. Udono, P. K. Srivastava. 1997. Heat shock protein-peptide complexes, reconstituted in vitro, elicit peptide-specific cytotoxic T lymphocyte responses and tumor immunity. J. Exp. Med. 186:1315.[Abstract/Free Full Text]
  15. Basu, S., R. Suto, R. J. Binder, P. K. Srivastava. 1998. Heat-shock proteins as novel mediators of cytokine secretion by macrophages. Cell Stress Chaperones 3:11.
  16. Breloer, M., B. Fleischer, A. Von Bonin. 1999. In vivo and in vitro activation of T cells after administration of Ag-negative heat shock proteins. J. Immunol. 162:3141.[Abstract/Free Full Text]
  17. Matzinger, P.. 1994. Tolerance, danger, and the extended family. Annu. Rev. Immunol. 12:991.[Medline]
  18. Newton, R. A., D. A. Raftos, R. L. Raison, C. L. Geczy. 1994. Chemotactic responses of hagfish (Vertebrata, Agnatha) leukocytes. Dev. Comp. Immunol. 18:295.[Medline]
  19. Clem, L. W., J. E. Bly, M. Wilson, V. G. Chinchar, T. Stuge, K. Barker, C. Luft, M. Rycyzyn, R. J. Hogan, T. van Lopik, et al 1996. Fish immunology: the utility of immortalized lymphoid cells–a mini review. Vet. Immunol. Immunopathol. 54:137.[Medline]
  20. Du Pasquier, L., J. Robert. 1992. In vitro growth of thymic tumor cell lines from Xenopus. Dev. Immunol. 2:295.[Medline]
  21. Robert, J., L. Du Pasquier. 1996. Xenopus lymphoid tumor cell lines. ed. Manual of Immunological Methods 2367. Academic Press, London.
  22. Hsu, E., L. Du Pasquier. 1984. Studies in Xenopus immunoglobulins using monoclonal antibodies. Mol. Immunol. 21:257.[Medline]
  23. Robert, J., D. Brown, L. Pasquier, N. Cohen. 1997. Antibody cross-linking of the thymocyte-specific cell surface molecule, CTX, causes abnormal mitosis and multinucleation of tumor cells. Exp. Cell Res. 235:227.[Medline]
  24. Du Pasquier, L., M. F. Flajnik. 1990. Expression of MHC class II antigens during Xenopus development. Dev. Immunol. 1:85.[Medline]
  25. Flajnik, M. F., E. Taylor, C. Canel., D. Grossberger, L. Du Pasquier. 1991. Reagents specific for MHC I antigens of Xenopus. Am. Zool. 31:580.
  26. Robert, J., C. Guiet, L. Du Pasquier. 1994. Lymphoid tumors of Xenopus laevis with different capacities for growth in larvae and adults. Dev. Immunol. 3:297.[Medline]
  27. Robert, J., C. Guiet, L. Du Pasquier. 1995. Ontogeny of the alloimmune response against transplanted tumor in Xenopus laevis. Differentiation 59:135.[Medline]
  28. Du Pasquier, L., J. Schwager, M. F. Flajnik. 1989. The immune system of Xenopus. Annu. Rev. Immunol. 7:251.[Medline]
  29. Robert, J., C. Guiet, N. Cohen, L. Du Pasquier. 1997. Effects of thymectomy and tolerance induction on tumor immunity in adult Xenopus laevis. Int. J. Cancer 70:330.[Medline]
  30. Chen, X., D. Easton, H. D. Lee-Yoon, X. Liu, J. Sujeck. 1996. The 170-kDa glucose stress-regulated protein is a large HSP70-HSP110-like protein of the endoplasmic reticulum. FEBS Lett. 380:68.[Medline]
  31. Robert, J., A. Menoret, K. Jursak, E. Torjusen, C. Smith, P. K. Srivastava, N. Cohen. 1998. Evolution of heat shock protein and vertebrate immunity. Cell Stress Chaperones 3:9.
  32. Flajnik, M.. 1998. Churchill and the immune system of ectothermic vertebrates. Immunol. Rev. 166:5.[Medline]
  33. Lippincott-Schwartz, J., L. C. Yuan, J. S. Bonifacino, R. D. Klausner. 1989. Rapid redistribution of Golgi proteins into the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER. Cell 56:801.[Medline]
  34. Srivastava, P. K., H. Udono, N. E. Blachere, Z. Li. 1994. Heat shock proteins transfer peptides during antigen processing and CTL priming. Immunogenetics 39:93.[Medline]
  35. Tamura, Y., P. Peng, K. Liu, M. Daou, P. K. Srivastava. 1997. Immunotherapy of tumors with autologous tumor-derived heat shock protein preparations. Science 278:117.[Abstract/Free Full Text]
  36. Anderson, S., T. Shen, J. Lou, L. Xing, N. Blachere, P. K. Srivastava, B. Y. Rubin. 1994. The endoplasmic reticular heat shock protein gp96 is transcriptionally upregulated in interferon-treated cells. J. Exp. Med. 180:1565.[Abstract/Free Full Text]
  37. Melcher, A., A. Todryk, N. Hardwick, M. Ford, M. Jacobson, R. G. Vile. 1998. Tumor immunogenicity is determined by mechanism of cell death via induction of heat shock protein expression. Nat. Med. 4:581.[Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
R. M. Srivastava, C. Varalakshmi, and A. Khar
The Ischemia-Responsive Protein 94 (Irp94) Activates Dendritic Cells through NK Cell Receptor Protein-2/NK Group 2 Member D (NKR-P2/NKG2D) Leading to Their Maturation
J. Immunol., January 15, 2008; 180(2): 1117 - 1130.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Cell Mol. Bio.Home page
Q. Lu, E. O. Harrington, J. Newton, M. Jankowich, and S. Rounds
Inhibition of ICMT Induces Endothelial Cell Apoptosis through GRP94
Am. J. Respir. Cell Mol. Biol., July 1, 2007; 37(1): 20 - 30.
[Abstract] [Full Text] [PDF]

Home page
 Infect. Immun.Home page
N. V. Prasadarao, P. K. Srivastava, R. S. Rudrabhatla, K. S. Kim, S.-h. Huang, and S. K. Sukumaran
Cloning and Expression of the Escherichia coli K1 Outer Membrane Protein A Receptor, a gp96 Homologue
Infect. Immun., April 1, 2003; 71(4): 1680 - 1688.
[Abstract] [Full Text]

Home page
 J. Immunol.Home page
P. P. Banerjee, D. S. Vinay, A. Mathew, M. Raje, V. Parekh, D. V. R. Prasad, A. Kumar, D. Mitra, and G. C. Mishra
Evidence That Glycoprotein 96 (B2), a Stress Protein, Functions as a Th2-Specific Costimulatory Molecule
J. Immunol., October 1, 2002; 169(7): 3507 - 3518.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
B. Liu, A. M. DeFilippo, and Z. Li
Overcoming Immune Tolerance to Cancer by Heat Shock Protein Vaccines
Mol. Cancer Ther., October 1, 2002; 1(12): 1147 - 1151.
[Abstract] [Full Text] [PDF]

Home page
 Infect. Immun.Home page
N. V. Prasadarao
Identification of Escherichia coli Outer Membrane Protein A Receptor on Human Brain Microvascular Endothelial Cells
Infect. Immun., August 1, 2002; 70(8): 4556 - 4563.
[Abstract] [Full Text] [PDF]

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

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS