Several pieces of experimental evidence indicate the following: 1) the most efficient antitumor treatments (this principle applies on both chemotherapy and radiotherapy) are those that induce immunogenic cell death and are able to trigger a specific antitumor immune response; and 2) the immunogenicity of cell death depends very closely on the plasma membrane quantity of calreticulin (CRT), an endoplasmic reticulum (ER) stress protein exposed to the cell membrane after immunogenic treatment. Nevertheless, the mechanisms implicated in CRT translocation are unknown. CRT is known to interact in the ER with ERP57, another ER stress protein. I sought to determine whether ERP57 would have any role in tumor immunogenicity. In this article I report that CRT exposure is controlled by ERP57 exposure. CRT and ERP57 are translocated together in the same molecular complex. ERP57 knockdown suppressed CRT exposure as well as phagocytosis by dendritic cells and abolished the immunogenicity in vivo. Knockdown or the absence of CRT abolishes ERP57 exposure. Administration of recombinant ERP57, unlike the administration of recombinant CRT, did not restore the immunogenicity of CRT or ERP57 small interfering RNA-transfected tumor cells. Together, these studies identify ERP57 as a key protein that controls immunogenicity by controlling CRT exposure and illustrate the ability of ERP57 to serve as a new molecular marker of immunogenicity.

We have recently shown that the exposure of calreticulin (CRT)3 dictates the immunogenicity of cancer cell death induced by chemotherapy compounds (1, 2), gamma irradiation (3), or UVC (3). The cell surface (ecto)CRT will act as an “eat me” signal and mediate the phagocytosis of tumor cell death by dendritic cells (DCs). Thus, the principal consequences of this uptake will be the following: 1) an increase in Ag cross-presentation; 2) an increase in the number of specific CTLs; 3) an increase in the production of IFN-γ by T cells; and 4) the eradication of implanted tumors (1, 2).

CRT is a 60-kDa chaperone Ca2+-binding protein and is mainly ubiquitous in the endoplasmic reticulum (ER). CRT has also been recognized as a multifunctional protein involved in a wide variety of cellular processes including modulation of cell adhesion (4), the folding of newly synthesized glycoproteins (5, 6, 7), modulation of gene expression (5), the regulation of Ca2+ homeostasis and Ca2+-dependent pathways (8), and lectin-like chaperone activity (7). Intracellular (endo)CRT can interact with many ER resident proteins, in particular with the family of protein disulfide isomerases and especially with the ERP57 protein (9).

CRT has been reported to be secreted from cells (10, 11) and is present at low levels in human plasma (12).

Nevertheless, CRT has been found to be localized in various subcellular compartments, including the cytosol, the nucleus, and the cell surface membrane. The role of CRT at the cell surface is more and more clear. It has been reported that ectoCRT orchestrates a number of cellular events including the modulation of cell adhesion and migration through interaction with integrins (13, 14) and with the extracellular matrix proteins fibrinogen (15) and laminin (16). We have shown lately that ectoCRT on dying or dead tumor cells represent an “eat me signal” for DCs as well as a major molecular determinant that makes the difference between immunogenic cell death and nonimmunogenic cell death (1, 3, 17).

However, the mechanisms by which CRT is translocated to the surface is still unknown. Because endoCRT can interact in the ER lumen with ERP57 and because ERP57 membrane translocation was induced by anthracyclines, I examined whether CRT and ERP57 are translocated together to the cell surface in the same molecular complex and whether ERP57 has any role in the inducing of immunogenic cell death. At this point, I identified another particular biochemical alteration in the plasma membrane of immunogenic dying cells, the surface exposure of ERP57, as a concomitant and determining event to CRT exposure that only occurs in immunogenic tumor cell death. Accordingly, I show that the ERP57 protein does not itself possess any immunogenic proprieties but could dictate indirectly the immunogenicity of tumor cell death by controlling the translocation of CRT.

The concentrations and times used in vitro for the treatment of CT26 cells with the different substances were established in preliminary dose-response experiments designed to determine the LC75, the dose that kills ∼75% of the cells. Therefore, CT26 cells were cultured at 37°C under 5% CO2 in RPMI 1640 medium supplemented with 10% FCS, penicillin, streptomycin, 1 mM pyruvate, and 10 mM HEPES in the presence of various substances to establish a dose-response analysis and a time kinetic. After that, I determined the LC75. The substances, concentrations, and culture periods used are as follows: doxorubicin at 25 μM for 24 h (Sigma-Aldrich); mitoxantrone at 1 μM for 24 h (Sigma-Aldrich); idarubicin at 1 μM for 24 h (Aventis); mitomycin C at 30 μM for 48 h (Sanofi-Synthelabo) and/or N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone (z-VAD-fmk) at 50 μM for 24 h (Bachem); tunicamycin at 65 μM for 24 h; thapsigargin at 30 μM for 24 h; brefeldin A at 50 μM for 24 h; etoposide at 25 μM for 48 h; MG132 at 10 μM for 48 h; N-acetyl-leucinyl-leucinyl-norleucinal (ALLN) at 45 μM for 48 h); betulinic acid at 10 μM for 24 h; Hoechst 33342 at 0.2 μM for 24 h; camptothecin for 15 μM for 24 h; lactacystin at 60 μM for 48 h; BAY 11-8072 at 30 μM for 24 h; staurosporine at 1.5 μM for 24 h; bafilomycin A1 at 300 μM for 48 h; arsenic trioxide at 30 μM for 24 h; C2-ceramide at 60 μM for 24 h; calyculin A at 30 nM for 48 h or tautomycin at 150 nM for 48 h (Sigma) and/or salubrinal at 20 μM for 48 h (Calbiochem). Cells were analyzed on a FACSVantage cell sorter after staining with 4,6-diamino-2-phenylindole at 2.5 μM for 10 min (Molecular Probes) for determination of cell viability and with annexin V conjugated with FITC (Bender MedSystems) for the assessment of phosphatidylserine exposure. Wild-type (K41) and calreticulin-deficient (K42) mouse embryonic fibroblasts were used in this study (18).

All animals were bred and maintained according the guidelines of both the Federation of European Laboratory Animal Science Associations (Tamworth, U.K.) and the Animal Experimental Ethics Committee (Paris, France). Animals were used at between 6 and 20 wk of age. Treated CT26 cells (3 × 106) were inoculated s.c. in 200 ml of PBS into the lower flank of BALB/c 6-wk-old female mice (Janvier), while 5 × 105 untreated control cells were inoculated into the opposite flank. For the tumorigenicity assay, 3 × 106 treated or untreated CT26 cells were injected s.c. into nu/nu mice (Charles River). To assess the specificity of the immune response against CT26, I injected either 5 × 105 or 5 × 106 CT26 cells (for the mice immunized in a standard protocol or vaccination protocol, respectively). Tumors were evaluated weekly using a caliper. In a series of experiments, BALB/c (wild type or nu/nu) carrying palpable CT26 tumors (implanted 14 days before for wild-type or 7 days before for nu/nu mice by injection of 106 tumor cells) received a single intratumoral injection of 100 μM PBS containing the same concentration of anticancer agents and PP1/GADD34 inhibitors as those used in vitro, as well as recombinant (r)CTR (45 μg) or rERP57 (45 μg). None of these treatments caused macroscopic necrosis.

Small interfering RNA (siRNA) heteroduplexes specific for CRT (sense strand: 5′-CCGCUGGGUCGAAUCCAAATT-3′), ERP57 (as described above), or an unrelated control (5′-GCCGGUAUGCCGGUUAAGUTT-3′) were synthesized by Sigma-Proligo. CT26 cells were transfected by siRNA at a final concentration of 100 nM using HiPerFect (Qiagen). Thirty-six hours posttransfection, CT26 cells were assessed for total CRT content by immunoblotting. To restore ectoCRT or ectoERP57 expression, cells were exposed to rCRT produced in insect cells (19) or rERP57 at 3 μg/106 cells in PBS on ice for 30 min, followed by three washes.

BALB/c animals were injected with 5 × 105 CT26 cells into the footpad. Five days later, popliteal lymph node cells were recovered by homogenizing and filtering the organ through a sterile cell strainer (70 μm; BD Biosciences). Lymph node cells (1 × 105) were cultured in 200 μl of complete culture medium in the presence or absence of CT26 cell lysates killed by a freeze-thaw cycle in round-bottom 96-well plates. Three days later, the supernatants were harvested and IFN-γ secretion was determined by ELISA (BD Biosciences).

Cells were washed with cold PBS at 4°C and lysed in a buffer containing 50 mM Tris-HCl (pH 6.8), 10% glycerol, and 2% SDS. Primary Abs detecting CRT (dilution: 1/2000; StressGen Biotechnologies), CD47 (dilution: 1/500; BD Biosciences), ERP57 (dilution: 1/1000, Abcam), tapasin (dilution: 1/1000; Abcam), GRP78 (dilution: 1 μg/ml; Abcam), and GRP94 (dilution: 1/1000; Abcam) were revealed with the appropriate HRP-labeled secondary Ab (Southern Biotechnologies Associates) and detected by ECL (Pierce), and the immunoreactive bands were quantified by densitometry. Anti-actin or anti-GAPDH Ab (Chemicon) was used to control equal loading.

CT26 cells were lysed in lysis buffer (20 mM Tris · HCl (pH 7.5), 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, and 1 mM Na3VO4 with 1% Triton X-100) containing a protease inhibitor mixture (Roche). Whole cell extracts were centrifuged at 14,000 rpm for 20 min to remove the debris. Immunoprecipitations were performed by incubating whole cell extracts with the indicated Ab preincubated with recombinant protein G agarose (Invitrogen) while rocking at 4°C overnight. Immunoprecipitates were washed three times with lysis buffer, resuspended in 50 μl of 1× Laemmli sample buffer, and then resolved by electrophoresis with 4–15% polyacrylamide Tris · HCl gel. In some experiments, I performed an immunoprecipitation on a plasma membrane protein isolated by biotinylation on CT26 cell surface membrane proteins.

Vector-based siRNA against Grp58 RNA was constructed as described in Ref. 20 . The human U6 promoter sequence was amplified from genomic DNA by nested PCR using the primers 5′-CCCGAGTCCAACACCCGTGG-3′ and 5′-GGTGTTTCGTCCTTTCCACA-3′ for the first PCR and then reamplified with 5′-ATAGAATTCCCGAGTCCAACACCCGTGGG-3′ and 5′ ATAGAATTCGGTGTTTCGTCCTTTCCACAAG-3′. The second amplicon was cloned using EcoRI (underlined) into a shuttle vector containing a hygromycin resistance. Two different siRNAs were cloned using a forward primer specific for the vector 5′-CCGATT TCGGCCTATTGGT-3′ and a reverse primer containing a portion that anneals with the 3′ end of the U6 promoter and the specific hairpin sequence (Grp58-1, 5′-ATAGCGCCGCAAAAAATAGTCCCATTAGCAAAGGG AAGCTTGAACCTTTGCTAATGGGACTATCGGTGTTTCGTCCTTTCCACA-3′; and Grp58-5, 5′-ATAGCGGCCGCAAAAAGAGATACCTGAAGTCTGGAAGCTTGAACAGACTTCAGGTATCTCGGTGTTTCGTCCTTTCCACA-3′). The PCR fragments obtained were digested using BglII and NotI and cloned into the shuttle vector predigested with the same enzymes. All PCRs were performed using Pfx polymerase (Invitrogen) following the instructions of the manufacturer. As a negative control, a complete U6 promoter fused to a degenerated hairpin was used. The selected sequences of both constructs (underlined) were “blasted” against the National Center for Biotechnology Information expressed sequence tag database (www.ncbi.nlm.nih.gov/BLAST/) and were specific for both human and mouse Grp58. The siRNA constructs were totally sequenced in both directions. Stably expressing CT26 cells were produced by transfection using the SuperFect kit (Qiagen), following the instructions of the manufacturer. After 48 h of transfection, cells were selected using G418 (geneticin; 1.3 mg/ml) for 5 days and individual clones were obtained by limiting dilution.

Bone marrow (BM) cells were flushed from the tibias and femurs of BALB/c mice with culture medium composed of RPMI 1640 medium supplemented with 10% heat-inactivated FCS (Invitrogen), sodium pyruvate, 50 μM 2-ME (Sigma-Aldrich), 10 mM HEPES (pH 7.4), and penicillin/streptomycin (Invitrogen). After one centrifugation, BM cells were resuspended in Tris-ammonium chloride for 2 min to lyse RBC. After one more centrifugation, BM cells (1 × 106 cells/ml) were cultured in medium supplemented with 100 ng/ml recombinant mouse FLT3 ligand (R&D systems) in 6-well plates (Costar Corning). After 7 days, the nonadherent and loosely adherent cells were harvested with Versene, washed, and transferred into 12-well plates (1.5 × 106 cells/plate) for cocultures with tumor cells.

In 12-well plates, 25 × 106 adherent CT26 cells were labeled with CellTracker Green or CellTracker Orange (Calbiochem) and then incubated with drugs. In some experiments viable CT26 cells were coated with chicken anti-CRT Ab (ABR Affinity Bioreagents), rabbit anti-ERP57 (Abcam), or an isotype control at 2 μl/106 cells for 30 min before washing and feeding to DCs. Alternatively, CT26 cells were coated with rCTR or of rERP57 protein 2 μg/106 cells on ice for 30 min and washed twice before addition to DCs. Cells were then harvested, labeled with CMTMR (Molecular Probes), washed three times with medium supplemented with FBS, and cocultured with immature DCs for 2 h at ratios of 1:1 and 1:5. At the end of the incubation, cells were harvested with versene, pooled with nonadherent cells present in the supernatant, and washed and stained with CD11c-FITC Ab. Phagocytosis was assessed by FACS analysis of double positive cells. Phagocytic indexes refer to the ratio between values obtained at 4°C and values obtained at 37°C from coincubation between DC and tumor cells.

CT26 cells (on a glass slide or in 12-well plates) were first washed with FACS buffer (1× PBS, 5% fetal bovine serum, and 0.1% sodium azide) and then incubated with rabbit anti-mouse CRT Ab (1/100; StressGen) or rabbit anti-mouse ERP57 Ab (1/200, Abcam) in FACS buffer at 4°C for 30 min. Cells reacted with anti-rabbit IgG (H+L) Alexa Fluor 488 conjugates (1/500) in FACS buffer at 4°C for 30 min. After washing three times with FACS buffer, ectoCRT or ectoERP57 was detected by cytofluorimetric analysis on a FACSVantage. In some experiments, cells were fixed with 4% paraformaldehyde and counterstained with Hoechst (2 μM; Sigma-Aldrich), followed by fluorescence microscopic assessment with a Leica IRE2 microscope equipped with a DC300F camera.

Biotinylation and recovery of cell surface proteins were performed with a method adapted from Gottardi et al. (21) and Hanwell et al. (22) that was described in detail in our previous work (2).

These protocols have been described in detail in our previous work (2).

Experimental results are expressed as mean ± SEM of triplicate plates. Statistical significance was determined by Student’s t test. For all tests, the statistical significance level was set at p < 0.001.

We have recently shown that CRT is exposed strictly on the surfaces of cells undergoing immunogenic tumor cell death (2). The membrane translocation of CRT was induced by immunogenic treatment such as the use of anthracycline and gamma irradiation contrarily to the use of agents that target the ER (thapsigargin, tunicamycin, and brefeldin), mitochondria (arsenite, betulinic acid, and C2-ceramide), or DNA (Hoechst 33342, camptothecin, etoposide, and mitomycin C) (2, 3). As reported in Ref. 2 , we conducted a proteomic study by using the technique of two-dimensional electrophoresis followed by mass spectroscopic analyses. In this study I have only shown a part of the gel two-dimensionally where I identified, after anthracycline treatment, an increase in ectoCRT. In another part of the same two-dimensional gel (not shown in the study reported in Ref. 2) I identified other increased spots after anthracycline treatment that correspond to an ER protein known as ERP57 (Fig. 1,a and data not shown). I identified four spots (spots 1, 2, 3, and 4) corresponding to a different phosphorylation status of ERP57. Thus, spots 1, 2, 3, and 4 were strongly induced (by factors of 4.3, 3.4, 8, and 8.1, respectively) by doxorubicin and (by factors of 2.7, 1.2, 1.2, and 1.7, respectively) by mitoxantrone, but less so (by factors of 2.2, 1.7, 1.2, and 1.5, respectively) by doxorubicin combined with z-VAD-fmk (Fig. 1,a). The surface exposure of ERP57 after treatment with anthracyclines was confirmed by Western blot analysis of purified plasma membrane surface proteins (Fig. 1,b) where the protein loading control was assured by the plasma membrane protein CD47, which has stable membrane expression after anthracycline treatment compared with untreated cells (2), and by immunoblot analysis of two-dimensional gels (data not shown). I verified the absence of contamination by an intracellular protein in the preparation of the plasma membrane protein by checking the expression of some intracellular proteins, e.g., GAPDH (Fig. 1,b). This ERP57 surface exposure was also detectable by both immunofluorescence staining of anthracycline-treated live cells or γ-irradiated live cells (data not shown) and cytofluorimetric analysis (Fig. 1,c) and was not accompanied by a general increase in the abundance of endoERP57 (Fig. 1,b and data not shown). The expression of intracellular CD47 was stable and not affected by anthracycline treatment (Fig. 1,b). Additionally, The increase in the level of ectoERP57 induced by anthracyclines was a quick process, detectable 1 h after treatment (Fig. 1, c and f), and hence preceded the apoptosis-associated phosphatidylserine exposure (Fig. 1, b and e). We showed previously, after treatment with anthracyclines, gamma irradiation, or UVC light, that CRT exposure in tumor cell death was an earlier process detectable as soon as 1 h after treatment and well before the manifestation of phosphatidylserine exposure (2, 3). Accordingly, I can report here that the kinetics of ERP57 exposure induced by anthracyclines was very similar to CRT exposure induced by anthracyclines. Moreover, there was a strong, positive, linear correlation (p < 0.001) between the appearance of ectoERP57 (measured at 24 h among viable cells) and the immunogenicity elicited by the 15 distinct apoptosis inducers (Fig. 1,d), similar to the correlation observed in previous work (2) between ectoCRT and the immunogenicity elicited by the same 15 distinct apoptosis inducers. Interestingly, there was a very strong correlation between ectoCRT and ecto-ERP57 (Fig. 1 g). These results suggest that ERP57 exposure is a second principal molecular alteration (besides ectoCRT) induced by immunogenic treatment on the plasma membrane of dying tumor cells.

FIGURE 1.

ERP57 surface exposure occurred specifically in immunogenic cell death and correlated with CRT surface exposure. a and c, Identification of ERP57 as a surface-exposed molecule elicited by anthracyclines. Cells were treated for 4 h with doxorubicin (DX) alone or in combination with Z-VAD-fmk (DXZ) or with mitoxantrone alone (MITOX). After this, I performed purification of the biotinylated plasma membrane proteins followed by two-dimensional gel electrophoresis (a) or immunoblot detection of ERP57 and CD47 in the plasma membrane protein fraction or in the total cell lysate (b). The circles from 1 to 4 indicate spots with different phosphorylation statuses of ERP57. CO, Control. c, Kinetics of ERP57 exposure. CT26 cells were treated with mitoxantrone (Mitox) for the indicated period, followed by immunofluorescence staining with an ERP57-specific Ab and cytofluorimetric analysis. Representative pictograms are shown. d, Correlation between ERP57 exposure and immunogenicity. The surface exposure of ERP57 was determined by immunofluorescence cytometry among viable (propidium iodine-negative) cells 24 h after treatment with the appropriate concentration and was correlated with the immunogenicity of cell death (as determined in Fig. 1 of Ref. 2 ). e and f, Preapoptotic exposure of ectoERP57. e, Kinetics of phosphatidylserine exposure and cell death. CT26 cells were treated with mitoxantrone for the indicated period followed by staining with annexin V (AnnV; which recognizes phosphatidylserine on the surface of dying cells, plus 4,6-diamino-2-phenylindole (DAPI; which stains dead cells) followed by FACS analysis. f, Cells were cultured as in e for the indicated periods followed by immunofluorescence staining with an ERP57 Ab and cytofluorimetric analysis. Quantitative data of the kinetics ERP57 exposure are reported. g, Correlation between ERP57 exposure and CRT exposure. The surface exposure of ERP57 and CRT was determined by immunofluorescence cytometry among viable (propidium iodine-negative) cells 24 h after treatment with different cell-death inducers. Note: idarubicin and doxorubicin (not included in d and g) emit strong autofluorescence, rendering FACS determinations of ERP57 exposure inaccurate.

FIGURE 1.

ERP57 surface exposure occurred specifically in immunogenic cell death and correlated with CRT surface exposure. a and c, Identification of ERP57 as a surface-exposed molecule elicited by anthracyclines. Cells were treated for 4 h with doxorubicin (DX) alone or in combination with Z-VAD-fmk (DXZ) or with mitoxantrone alone (MITOX). After this, I performed purification of the biotinylated plasma membrane proteins followed by two-dimensional gel electrophoresis (a) or immunoblot detection of ERP57 and CD47 in the plasma membrane protein fraction or in the total cell lysate (b). The circles from 1 to 4 indicate spots with different phosphorylation statuses of ERP57. CO, Control. c, Kinetics of ERP57 exposure. CT26 cells were treated with mitoxantrone (Mitox) for the indicated period, followed by immunofluorescence staining with an ERP57-specific Ab and cytofluorimetric analysis. Representative pictograms are shown. d, Correlation between ERP57 exposure and immunogenicity. The surface exposure of ERP57 was determined by immunofluorescence cytometry among viable (propidium iodine-negative) cells 24 h after treatment with the appropriate concentration and was correlated with the immunogenicity of cell death (as determined in Fig. 1 of Ref. 2 ). e and f, Preapoptotic exposure of ectoERP57. e, Kinetics of phosphatidylserine exposure and cell death. CT26 cells were treated with mitoxantrone for the indicated period followed by staining with annexin V (AnnV; which recognizes phosphatidylserine on the surface of dying cells, plus 4,6-diamino-2-phenylindole (DAPI; which stains dead cells) followed by FACS analysis. f, Cells were cultured as in e for the indicated periods followed by immunofluorescence staining with an ERP57 Ab and cytofluorimetric analysis. Quantitative data of the kinetics ERP57 exposure are reported. g, Correlation between ERP57 exposure and CRT exposure. The surface exposure of ERP57 and CRT was determined by immunofluorescence cytometry among viable (propidium iodine-negative) cells 24 h after treatment with different cell-death inducers. Note: idarubicin and doxorubicin (not included in d and g) emit strong autofluorescence, rendering FACS determinations of ERP57 exposure inaccurate.

Close modal

We have recently reported that anthracycline-, UVC-, or gamma irradiation-treated tumor cells were phagocytosed by DC quickly, well before the manifestation of apoptotic changes. This correlated with the rapid induction of CRT and the acquisition of immunogenicity (2, 3). In addition, they was a strong correlation between the surface exposure of ERP57 on the tumor cells and the phagocytosis mediated by DC (Fig. 2 a). The surface exposure of ERP57 was strong after immunogenic treatment and very weak after nonimmunogenic treatment.

FIGURE 2.

EctoCRT and not ectoERP57 mediate phagocytosis of tumor cells by DCs. a, Correlation between ERP57 cell surface exposure and phagocytosis of tumor cells by DCs. Tumor cells labeled with CellTracker Orange after the treatment for 4 h with different cell-death inducers were cultured for 1.5 h with DC-expressed CD11c. The percentage of DC taking up tumor cells was determined by gating on double positive DC for CellTracker Orange and for CD11c-FITC and correlated with ERP57 surface exposure (a), measured as in Fig. 1, d and g. b, Neutralizing of ectoERP57 did not affect the phagocytosis mediated by DCs, whereas neutralizing ectoCRT inhibited DC-mediated phagocytosis. Mitoxantrone-treated (Mitox) or control cells (CO) were incubated with a chicken blocking CRT Ab or with a rabbit blocking ERP57 Ab before their coculture with DCs. The percentage of phagocytosis was determined as in the Fig. 3 a. c–e, Knockdown of either CRT or ERP57 inhibits DC-mediated phagocytosis; otherwise, only rCRT and not rERP57 restores phagocytosis in vitro. CT26 transfected with siRNA ERP57 was followed by immunoblotting for ERP57 (c) and detection of surface ERP57 prior to the addition (+rERP57) of not (−) of rERP57 (d). In CT26 transfected with either CRT siRNA or ERP57 siRNA and optionally treated with rCRT or rERP57, I measured the phagocytosis of tumor cells by DC (e). The phagocytotic index consists in the ratio between values obtained with control cells and those obtained after mitoxantrone treatment (with or without siRNA CRT or siRNA ERP57 and with or without rCRT or rERP57). Results are triplicates from one experiment (mean ± SD) and representative of three independent experiments. ∗, p < 0.001 (Student’s t test).

FIGURE 2.

EctoCRT and not ectoERP57 mediate phagocytosis of tumor cells by DCs. a, Correlation between ERP57 cell surface exposure and phagocytosis of tumor cells by DCs. Tumor cells labeled with CellTracker Orange after the treatment for 4 h with different cell-death inducers were cultured for 1.5 h with DC-expressed CD11c. The percentage of DC taking up tumor cells was determined by gating on double positive DC for CellTracker Orange and for CD11c-FITC and correlated with ERP57 surface exposure (a), measured as in Fig. 1, d and g. b, Neutralizing of ectoERP57 did not affect the phagocytosis mediated by DCs, whereas neutralizing ectoCRT inhibited DC-mediated phagocytosis. Mitoxantrone-treated (Mitox) or control cells (CO) were incubated with a chicken blocking CRT Ab or with a rabbit blocking ERP57 Ab before their coculture with DCs. The percentage of phagocytosis was determined as in the Fig. 3 a. c–e, Knockdown of either CRT or ERP57 inhibits DC-mediated phagocytosis; otherwise, only rCRT and not rERP57 restores phagocytosis in vitro. CT26 transfected with siRNA ERP57 was followed by immunoblotting for ERP57 (c) and detection of surface ERP57 prior to the addition (+rERP57) of not (−) of rERP57 (d). In CT26 transfected with either CRT siRNA or ERP57 siRNA and optionally treated with rCRT or rERP57, I measured the phagocytosis of tumor cells by DC (e). The phagocytotic index consists in the ratio between values obtained with control cells and those obtained after mitoxantrone treatment (with or without siRNA CRT or siRNA ERP57 and with or without rCRT or rERP57). Results are triplicates from one experiment (mean ± SD) and representative of three independent experiments. ∗, p < 0.001 (Student’s t test).

Close modal

Having observed that ectoERP57 correlates strongly with the phagocytosis of tumor cells treated with anthracycline mediated by DC, I aimed to evaluate the exact role of ectoERP57 in phagocytosis. First, I proceeded to block ectoERP57 or ectoCRT with specific neutralizing Abs of avian origin for CRT and rabbit origin for ERP57 (both Abs cannot interact with mouse Fc receptors) respectively. The blocking of ectoERP57 on mitoxantrone-treated cancer cells has no effect on the phagocytosis of anthracycline-treated tumor cells by DC in vitro (Fig. 2,b). Otherwise, the blocking of ectoCRT inhibited their phagocytosis by DC (Fig. 2,b). In contrast, knockdown of ERP57 with ERP57 siRNA (Fig. 2,c) suppressed both ectoERP57 (Fig. 2,d) and phagocytosis of anthracycline-treated tumor cells by DC in vitro (Fig. 2,e) similarly as the effect observed by CRT siRNA (Fig. 2,e). Moreover, the addition of the rCRT protein, which binds to the surface of the tumor cells (1, 2, 3), reversed the defect induced by CRT or ERP57 siRNA as regards both CRT surface exposure (Ref. 2 and data not shown) and phagocytosis by DC in vitro (Fig. 2,e), whereas my restoration of the plasma membrane ERP57 through the addition of the rERP57 protein (Fig. 2,d) did not reverse the defect induced by CRT or ERP57 siRNA on the phagocytosis by DC in vitro (Fig. 2 e). Otherwise, neither rCRT (2) or rERP57 alone could induce DC maturation ex vivo over a large range of concentrations (10 ng/ml to 100 mg/ml; Ref. 2 and data not shown). Therefore, ectoCRT and not ectoERP57 specifically elicits phagocytosis by DC.

Next, I evaluated whether ectoERP57 would participate in the induction of immunogenicity in vivo. First, after the depletion of CRT by CRT siRNA, I abolished the immunogenicity of mitoxantrone-treated CT26 cells (Fig. 3, a and b). This immunogenicity was restored when rCRT was used to complement the ectoCRT defect induced by CRT siRNA, whereas the addition of rERP57 did not restore this defect (Fig. 3, a and b). Similarly, the knockdown of ectoERP57 abolished the immunogenicity of mitoxantrone-treated CT26 cells, and this defect was restored when rCRT was used to complement the ectoCRT defect induced by the ERP57 siRNA, whereas the addition of rERP57 did not restore this defect (Fig. 3, a and b). Second, I evaluated the role of ectoCRT and ectoERP57 in the production of IFN-γ by popliteal lymph nodes. The depletion of ectoCRT by CRT siRNA abolished the production of IFN-γ in a similar manner as the depletion of ectoERP57 by ERP57 siRNA (Fig. 3, d and e). In addition, only rCRT had the potential to restore IFN-γ production in either the siRNA CRT or the siRNA ERP57 system (Fig. 3, d and e). In contrast, the addition of rERP57 caused no significant restoration of the production of IFN-γ depleted by siRNA CRT or siRNA ERP57 (Fig. 3, d and e). However, the addition of rCRT and not rERP57 restored both IFN-γ production (Fig. 3, d and e) and the immunogenicity (Fig. 3 c) of nonimmunogenic treatment such as that of mitomycin C. Thus, ectoCRT and not ectoERP57 determines the immunogenicity of cancer cell death.

FIGURE 3.

EctoCRT and not ectoERP57 is required for the immune response against dying tumor cells and consequently for a successful antitumor vaccination. a–c, Antitumor vaccination depends on ectoCRT and not on ectoERP57. Untransfected CT26 colon cancer cells or cells transfected with the indicated CRT siRNA or ERP57 siRNA were treated for 24 h with medium alone (CO, Control), mitoxantrone (Mitox), or mitomycin C (MC) and then incubated (+) or not (−) with rCRT or rERP57. The antitumor vaccination was measured in BALB/c mice immunized s.c. with PBS or with dying tumor-treated cells in one flank followed 1 wk later with live tumor cells in the opposite flank. Only the supply of rCRT and not that of rERP57 was able to restore the immunogenicity of anthracyclines (b) and transform the nonimmunogenic death induced by mitomycin C into immunogenic death (c) (mean ± SEM) ∗, p < 0.001 (Student’s t test). d and e, EctoCRT and ectoERP57 dictate the priming of T cells. d, Untransfected CT26 cells or those transfected with the indicate siRNAs were treated for 24 h with medium alone (PBS) or mitoxantrone and then injected into the footpads of BALB/c mice to determine the capacity of draining lymph nodes to produce IFN-γ in response to dying CT26. e, Only the exogenous supply of rCRT and not that of rERP57 was able to restore the priming of T cells. CT26 cells lacking ectoCRT or ectoERP57 after depletion with the indicated siRNAs were coated with either rCRT or rERP57 and then inject into the footpads to assess to the production of IFN-γ draining lymph nodes. All experiments were repeated three times. For a–c each group includes 12 mice per experiment; for d and e each group includes three mice per experiment. ∗, p < 0.001 (Student’s t test.)

FIGURE 3.

EctoCRT and not ectoERP57 is required for the immune response against dying tumor cells and consequently for a successful antitumor vaccination. a–c, Antitumor vaccination depends on ectoCRT and not on ectoERP57. Untransfected CT26 colon cancer cells or cells transfected with the indicated CRT siRNA or ERP57 siRNA were treated for 24 h with medium alone (CO, Control), mitoxantrone (Mitox), or mitomycin C (MC) and then incubated (+) or not (−) with rCRT or rERP57. The antitumor vaccination was measured in BALB/c mice immunized s.c. with PBS or with dying tumor-treated cells in one flank followed 1 wk later with live tumor cells in the opposite flank. Only the supply of rCRT and not that of rERP57 was able to restore the immunogenicity of anthracyclines (b) and transform the nonimmunogenic death induced by mitomycin C into immunogenic death (c) (mean ± SEM) ∗, p < 0.001 (Student’s t test). d and e, EctoCRT and ectoERP57 dictate the priming of T cells. d, Untransfected CT26 cells or those transfected with the indicate siRNAs were treated for 24 h with medium alone (PBS) or mitoxantrone and then injected into the footpads of BALB/c mice to determine the capacity of draining lymph nodes to produce IFN-γ in response to dying CT26. e, Only the exogenous supply of rCRT and not that of rERP57 was able to restore the priming of T cells. CT26 cells lacking ectoCRT or ectoERP57 after depletion with the indicated siRNAs were coated with either rCRT or rERP57 and then inject into the footpads to assess to the production of IFN-γ draining lymph nodes. All experiments were repeated three times. For a–c each group includes 12 mice per experiment; for d and e each group includes three mice per experiment. ∗, p < 0.001 (Student’s t test.)

Close modal

Having shown that both phagocytosis and the efficiency of antitumor vaccination are dictated by ectoCRT and not by ectoERP57 and to confirm the absence of immunogenic properties in ERP57 protein, I established s.c. CT26 colon cancer in BALB/c mice followed by a single intratumoral injection of each treatment. A single injection of PBS, mitomycin C, rCRT, or rERP57 in such tumors fails to induce tumor regression (Fig. 4,a). Otherwise, the combined injection of a cell death inducer (mitomycin C) plus rCRT causes tumor regression in immunocompetent mice but not in immunodeficient nu/nu (data not shown) nude mice in contrast to the intratumoral injection of mitomycin C plus rERP57 (Fig. 4,a). To clarify this relationship between ectoERP57 and immunogenicity, CT26 cells were stably transfected with vector-based siRNA targeted against ERP57 mRNA. Several cell lines were obtained in which ERP57 expression was completely abolished and I selected one of them, clone ERP57 siRNA-1 (Fig. 4,b). No alteration in the expression levels of calreticulin, Grp78, or Grp94 was observed in these cells, suggesting that targeting ERP57 RNA was specific (Fig. 4,b). In Fig. 3 I reported a relationship between the depletion of ERP57 and the suppression of antitumor vaccination. Indeed, I noted that the exposure of CRT induced by mitoxantrone or mitomycin C plus tautomycin was inhibited in the stable transfected CT26 clone ERP57 siRNA-1 (Fig. 4,c). Therefore, I decided to determine the effect of this inhibition on the immunogenicity of the stable transfected CT26 clone. Indeed, I obtained the same result as that characterized in Fig. 3 by a suppression of the antitumor vaccination after mitoxantrone treatment or treatment with mitomycin C plus tautomycin (Fig. 4,d). The defect in this stable clone treated with mitoxantrone or with mitomycin C plus tautomycin was restored only after the addition of rCRT and not with the addition of rERP57 (Fig. 4,d). No defect of immunogenicity after mitoxantrone treatment or treatment with mitomycin C plus tautomycin was observed in untransfected CT26 cells or CT26 cells transfected with the control vector (Mock; Fig. 4,d). Additionally, I evaluated this stable clone by intratumoral injection into established tumors during immunogenic treatment of BALB/c mice. The immunogenic treatment efficiency of combined injection consists in the cell death inducers (mitomycin C) plus the inhibitors of protein phosphatase 1 (PP1)/growth arrest and DNA damage-inducible protein 34 (GADD34) (tautomycin); or only mitoxantrone was strongly abolished (Fig. 4,e). This defect observed in the ERP57 siRNA stable clone after a treatment with mitoxantrone or mitomycin C plus tautomycin was restored after the addition of rCRT and not rERP57 (Fig. 4 e). Altogether, these data indicate the following: 1) the efficiency of immunogenic chemotherapy is determined by ectoCRT and not ectoERP57; and 2) ectoERP57 does not possess any immunogenic properties.

FIGURE 4.

EctoCRT and not ectoERP57 dictates the efficacy of antitumor chemotherapy and immunogenicity. a, Impact of rCRT and rERP57 on the efficacy of intratumoral injection. CT26 colon cancer was established in BALB/c mice and when the tumors reached the size of 250 mm2 the mice were left untreated or injected locally with the indicated products of mitomycin C, rCRT, rERP57, or a combination of these different products. Each curve represents one mouse. The letter “n” of each graph indicates the number of mice that manifested complete tumor regression at day 60 and did not develop tumors upon rechallenge with live tumor cells at day 60 for another 45 days. The graphs depict one representative experiment of three, comprising five mice per group. b, Different CT26 subclones were obtained that were stably transfected with a siRNA-based vector against ERP57 RNA (siRNA; for RNA interference), an irrelevant siRNA (Mock), or an ERP57 expression plasmid (OE). The expression levels of endogenous ERP57, CRT, GRP78, GRP94, and GAPDH were analyzed in several different clones by Western blotting. c, EctoCRT depends on endoERP57. The exposure of CRT was determined in wild-type CT26 (CT26 Wt), CT26 ERP57 siRNA-1 (clone with stable expression of ERP57 siRNA vector), and mock CT26 (CT26 Mock) after treatment with the medium (CO, Control), mitoxantrone (Mitox), or tautomycin plus mitomycin C (TA + MC). d, The immunogenicity of mitoxantrone and inhibitor of PP1/GADD34 (tautomycin) was abolished in the CT26 ERP57 siRNA-1 clone, where both ectoCRT and ectoERP57 were abolished and restored only with rCRT as opposed to rERP57. The antitumor vaccination was monitored by challenging BALB/c mice with dying cells treated with mitoxantrone or mitomycin C plus tautomycin and, where optionally supplied, rCRT or rERP57 in one flank and live tumor cells in the opposite flank. e, The immunogenicity of an intratumoral injection depends on ectoCRT. The different CT26 colon cancer clones (Mock and stable) and wild-type CT26 were established in BALB/c mice and, when the tumors reached the size of 250 mm2, the mice were left untreated or injected locally with the indicate products of tautomycin, mitoxantrone, mitomycin C, rCRT, rERP57, or a combination of these different products. A cured mouse is a one that manifested complete tumor regression at day 60 and did not develop tumors upon rechallenge with live tumor cells at day 60 for another 45 days. For a and e each treatment group included five mice and experiments were repeated three times; for d, each treatment group included 12 mice and experiments were repeated three times. p < 0.001 (Student’s t test).

FIGURE 4.

EctoCRT and not ectoERP57 dictates the efficacy of antitumor chemotherapy and immunogenicity. a, Impact of rCRT and rERP57 on the efficacy of intratumoral injection. CT26 colon cancer was established in BALB/c mice and when the tumors reached the size of 250 mm2 the mice were left untreated or injected locally with the indicated products of mitomycin C, rCRT, rERP57, or a combination of these different products. Each curve represents one mouse. The letter “n” of each graph indicates the number of mice that manifested complete tumor regression at day 60 and did not develop tumors upon rechallenge with live tumor cells at day 60 for another 45 days. The graphs depict one representative experiment of three, comprising five mice per group. b, Different CT26 subclones were obtained that were stably transfected with a siRNA-based vector against ERP57 RNA (siRNA; for RNA interference), an irrelevant siRNA (Mock), or an ERP57 expression plasmid (OE). The expression levels of endogenous ERP57, CRT, GRP78, GRP94, and GAPDH were analyzed in several different clones by Western blotting. c, EctoCRT depends on endoERP57. The exposure of CRT was determined in wild-type CT26 (CT26 Wt), CT26 ERP57 siRNA-1 (clone with stable expression of ERP57 siRNA vector), and mock CT26 (CT26 Mock) after treatment with the medium (CO, Control), mitoxantrone (Mitox), or tautomycin plus mitomycin C (TA + MC). d, The immunogenicity of mitoxantrone and inhibitor of PP1/GADD34 (tautomycin) was abolished in the CT26 ERP57 siRNA-1 clone, where both ectoCRT and ectoERP57 were abolished and restored only with rCRT as opposed to rERP57. The antitumor vaccination was monitored by challenging BALB/c mice with dying cells treated with mitoxantrone or mitomycin C plus tautomycin and, where optionally supplied, rCRT or rERP57 in one flank and live tumor cells in the opposite flank. e, The immunogenicity of an intratumoral injection depends on ectoCRT. The different CT26 colon cancer clones (Mock and stable) and wild-type CT26 were established in BALB/c mice and, when the tumors reached the size of 250 mm2, the mice were left untreated or injected locally with the indicate products of tautomycin, mitoxantrone, mitomycin C, rCRT, rERP57, or a combination of these different products. A cured mouse is a one that manifested complete tumor regression at day 60 and did not develop tumors upon rechallenge with live tumor cells at day 60 for another 45 days. For a and e each treatment group included five mice and experiments were repeated three times; for d, each treatment group included 12 mice and experiments were repeated three times. p < 0.001 (Student’s t test).

Close modal

Because I found that both endoCRT and endoERP57 are crucial for immunogenicity, that only ectoCRT and not ectoERP57 possessed immunogenic properties, and that the membrane translocation of CRT depends on endoERP57, I suggest that endoERP57 would be implicated in the membrane translocation of CRT and not in the process of immunogenicity itself. To evaluate this hypothesis, ERP57 siRNA was tested to check its effect on the exposure of CRT. Consistently, markedly reduced CRT Ab staining was observed after the depletion of endoERP57, and no inhibition was observed with control siRNA (Fig. 5,a). Moreover, in a reverse experiment I depleted endoCRT with CRT siRNA, checked the consequence on ecotERP57, and observed a radical inhibition of ectoERP57 whereas no effect was detected with the control siRNA (Fig. 5,b). To confirm the relationship between endoCRT and endoERP57, I proceeded to test the translocation of ERP57 in K42-calreticulin embryonic fibroblast-deficient cells. After mitoxantrone or tautomycin treatment, I observed a strong inhibition of ectoERP57 in K42 cells compared with K41-wild-type mouse embryonic fibroblasts treated with mitoxantrone or tautomycin (Fig. 5,c). The membrane translocation of K42-deficient calreticulin cells was restored after the reconstitution of CRT level by a calreticulin-expressing vector (data not shown). Elsewhere, we have recently shown (1, 2) that CRT exposure is mediated by the inhibition of the PP1/GADD34 complex by a variety of chemical inhibitors (e.g., tautomycin, calyculin A, and salubrinal) as well as by PP1/GADD34 siRNA. In this study, I show that ectoERP57 is induced by the same inhibitors of PP1/GADD34 (e.g., tautomycin, calyculin A, and salubrinal) with the same kinetics that induce ectoCRT (Fig. 5,d). Besides, in Ref. 2 we reported that the membrane translocation of CRT is inhibited by latrunculin A, an inhibitor of the actin cytoskeleton and exocytosis and was not affected by the inhibitors of transcription (actinomycin D), translation (cycloheximide), or microtubule polymerization (nocodazole). Thereafter, after having tested the same inhibitors on the exposure of ERP57, I have observed the same profile of inhibition of ectoERP57 by latrunculin A and not by actinomycin D, cyclohexamide, or nocodazole. In addition, after treatment with the PP1/GADD34 inhibitor the total cellular expression of CRT, ERP57, and tapasin have not been modified compared with untreated cells (Fig. 5 f). According to the following facts, first, that the exposure of CRT depends on the exposure of ERP57 and vice versa; second, that CRT exposure and ERP57 exposure are induced by the same inducers (anthracyclines and not mitomycin C) and the same mediator (inhibitors of PP1/GADD34 complex); third, that they are inhibited by the same inhibitors; and fourth, that the total cellular expressions of CRT and ERP57 have not been affected by the treatment, I suggest that perhaps CRT and ERP57 are translocated together to the cell surface.

FIGURE 5.

ERP57 controls the membrane translocation of CRT and vice versa. a, The depletion of ERP57 abolishes CRT surface exposure. Cell surface CRT was measured in ERP57 siRNA-transfected CT26 and then treated with mitoxantrone (Mitox) for 4 h. b and c, The depletion of CRT abolishes ERP57 surface exposure. After mitoxantrone or tautomycin (TA) treatment for 4 h, cell surface ERP57 was measured in CRT siRNA-transfected CT26 cells (b) or in K42-calreticulin embryonic fibroblast-deficient cells (c). CO, Control. d, Kinetics of ERP57 exposure were determined by FACS analysis after incubation of cells with the indicated agents, mitoxantrone or inhibitors of PP1/GADD34. e, After preincubation for 1 h with the indicated inhibitors of protein synthesis (cycloheximide), RNA synthesis (actinomycin D), or the actin cytoskeleton (latrunculin A). Then, ERP57 expression was determined by immunocytofluorometry. Results are triplicates from one experiment (mean ± SD) and are representative of three independent experiments. f, Immunoblot staining of total lysate CRT, ERP57, and tapasin before and after treatment with PP1/GADD34 inhibitor.

FIGURE 5.

ERP57 controls the membrane translocation of CRT and vice versa. a, The depletion of ERP57 abolishes CRT surface exposure. Cell surface CRT was measured in ERP57 siRNA-transfected CT26 and then treated with mitoxantrone (Mitox) for 4 h. b and c, The depletion of CRT abolishes ERP57 surface exposure. After mitoxantrone or tautomycin (TA) treatment for 4 h, cell surface ERP57 was measured in CRT siRNA-transfected CT26 cells (b) or in K42-calreticulin embryonic fibroblast-deficient cells (c). CO, Control. d, Kinetics of ERP57 exposure were determined by FACS analysis after incubation of cells with the indicated agents, mitoxantrone or inhibitors of PP1/GADD34. e, After preincubation for 1 h with the indicated inhibitors of protein synthesis (cycloheximide), RNA synthesis (actinomycin D), or the actin cytoskeleton (latrunculin A). Then, ERP57 expression was determined by immunocytofluorometry. Results are triplicates from one experiment (mean ± SD) and are representative of three independent experiments. f, Immunoblot staining of total lysate CRT, ERP57, and tapasin before and after treatment with PP1/GADD34 inhibitor.

Close modal

To confirm our hypothesis that CRT and ERP57 are transported together to the cell surface, I performed coimmunoprecipitation experiments on the cell surface protein. Immunoprecipitation of endogenous ERP57 confirms that endogenous CRT interacts with endogenous ERP57 (Fig. 6,a) and that this interaction was not significantly modified after mitoxantrone treatment. In addition, a low amount of ectoCRT was detected in interaction with ectoERP57 in nontreated cells, but in mitoxantrone-treated cells the immunoprecipitation of ectoERP57 revealed large amount of ectoCRT in association with ectoERP57 (Fig. 6 a). The densitometry analysis of the ERP57/CRT ratio confirmed that after mitoxantrone treatment the quantity of CRT exposed to cell surface is nearly equal to the quantity of ERP57 exposed to cell surface. These data demonstrated clearly that ectoCRT and ectoERP57 are transported together to cell surface.

FIGURE 6.

EndoCRT interacts with endoERP57 and ectoCRT interacts with ectoERP57. CT26 cells untreated or treated for 4 h with mitoxantrone (Mitox) were lysed and immunoprecipitated (IP) with the anti-ERP57 mAb, or the immunoprecipitation was performed on the plasma membrane proteins purified by biotinylation from CT26 cells left untreated (−) or treated (+) for 4 h with mitoxantrone. The levels of CRT and ERP57 were analyzed by Western blotting (a). b, Arbitrary densitometry analysis for the ERP57/CRT ratio on the total extract and plasma membrane extract before and after mitoxantrone treatment. Results are triplicates from one experiment (mean ± SD) and are representative of three independent experiments. ∗, p < 0.001 (Student’s t test).

FIGURE 6.

EndoCRT interacts with endoERP57 and ectoCRT interacts with ectoERP57. CT26 cells untreated or treated for 4 h with mitoxantrone (Mitox) were lysed and immunoprecipitated (IP) with the anti-ERP57 mAb, or the immunoprecipitation was performed on the plasma membrane proteins purified by biotinylation from CT26 cells left untreated (−) or treated (+) for 4 h with mitoxantrone. The levels of CRT and ERP57 were analyzed by Western blotting (a). b, Arbitrary densitometry analysis for the ERP57/CRT ratio on the total extract and plasma membrane extract before and after mitoxantrone treatment. Results are triplicates from one experiment (mean ± SD) and are representative of three independent experiments. ∗, p < 0.001 (Student’s t test).

Close modal

The discovery of new molecular determinants that characterize the immunogenicity of cell death and are capable of triggering an anticancer immune response is a critical step for the design of more efficient antitumor therapies. Profiling of the cell surface and plasma membrane proteomes will likely provide novel insights and uncover disease-related alterations. Using this approach, a better molecular characterization of the difference between immunogenic and nonimmunogenic cell death has been crossed recently. Several lines of evidence (1, 2, 3, 23, 24, 25) have recently emerged to suggest that the immunogenicity of cancer cell death is dictated by the surface exposure of CRT, which makes the difference between immunogenic cell death and nonimmunogenic cell death. This exposure was conditioned by the inhibition of the PP1/GADD34 complex. Indeed, the injection of rCRT or inhibitors of PP1/GADD34 in combination with nonimmunogenic cell death inducers (e.g., etoposide and mitomycin C) allows us to transform nonimmunogenic death into immunogenic death capable of triggering an anticancer immune response. The surface exposure of CRT is a very rapid process that occurs upstream of apoptosis or necrosis and plays the role of an “eat me signal” for DCs as part of a specific danger-signaling system. In contrast, the molecular mechanisms governing the membrane translocation of CRT are still missing. Because if the possible interaction between CRT and many others ER resident proteins (such as ERP57/GRP58, CMHI, calnexin, protein disulfide, isomerase, etc.) (26) and because just the surface exposure of ERP57 and not that of the others ER resident proteins was increased after anthracycline treatment, I evaluated the possibility that ERP57 would be implicated in the process of immunogenicity.

In this study I checked the spectrum of ERP57 surface exposure compared with the one of CRT surface exposure, which was very similar. First, ERP57 and CRT exposure have been induced similarly by the same inducers such as anthracyclines (doxorubicin, mitoxantrone), irradiation-γ, UVC-light and inhibitors of PP1/GADD34 complex (such as tautomycin, calyculin A, salubrinal) and absent by many others non immunogenic treatment such as agents that target the endoplasmic reticulum (ER) (thapsigargin, tunicamycin and brefeldin A), mitochondria (arsenite, betulinic acid and C2 ceramide) or DNA (Hoechst 33342, camptothecin, etoposide and mitomycin C) (Table I). In addition, CRT and ERP57 exposure was occurred with the same exposure kinetics, detectable as soon as 1h after immunogenic treatment and with very strong correlation measured at same time of treatment. Moreover, they was very strong correlation between ectoERP57 and immunogenicity or between ectoCRT and immunogenicity (2) as well as between ectoERP57 and phagocytosis mediated by DCs or between ectoCRT and phagocytosis mediated by DCs (2).

Table I.

Comparative analysis of different substances on calreticulin CRT and ERP57 cell surface exposurea

Substances UsedPrincipal TargetEffect on CRTEffect on ERP57
Mitoxantrone DNA ++++ ++++ 
Doxorubicin DNA ++++ ++++ 
Idarubicin DNA ++++ ++++ 
Irradiation DNA ++++ ++++ 
UVC DNA ++++ ++++ 
Mitomycin C DNA 
Etoposide DNA ++ ++ 
Camptothecin DNA − − 
Hoechst 33342 DNA − − 
Tautomycin PP1/GADD34 ++++ ++++ 
Calyculin A PP1/GADD34 ++++ ++++ 
Salubrinal PP1/GADD34 ++++ ++++ 
Arsenic trioxide Mitochondria − − 
Betulinic acid Mitochondria 
C2-ceramide Mitochondria − − 
Thapsigargin ER − − 
Tunicamycin ER − − 
Brefeldin A ER − − 
MG132 Proteasome − − 
Lactacystin Proteasome − − 
ALLNb Proteasome − − 
Bafilomycin A1 Lysosome − − 
Bay 11-7082 NF-κβ 
Staurosporine Tyrosine Kinase 
Substances UsedPrincipal TargetEffect on CRTEffect on ERP57
Mitoxantrone DNA ++++ ++++ 
Doxorubicin DNA ++++ ++++ 
Idarubicin DNA ++++ ++++ 
Irradiation DNA ++++ ++++ 
UVC DNA ++++ ++++ 
Mitomycin C DNA 
Etoposide DNA ++ ++ 
Camptothecin DNA − − 
Hoechst 33342 DNA − − 
Tautomycin PP1/GADD34 ++++ ++++ 
Calyculin A PP1/GADD34 ++++ ++++ 
Salubrinal PP1/GADD34 ++++ ++++ 
Arsenic trioxide Mitochondria − − 
Betulinic acid Mitochondria 
C2-ceramide Mitochondria − − 
Thapsigargin ER − − 
Tunicamycin ER − − 
Brefeldin A ER − − 
MG132 Proteasome − − 
Lactacystin Proteasome − − 
ALLNb Proteasome − − 
Bafilomycin A1 Lysosome − − 
Bay 11-7082 NF-κβ 
Staurosporine Tyrosine Kinase 
a

Note: ++++, strong exposure; ++, middle exposure; +, weak exposure; −, no exposure. Some data are not presented in this article.

b

ALLN, N-acetyl-leucinyl-leucinyl-norleucinal.

Additionally, this concomitant exposure of ectoCRT and ectoERP57, which is induced by the same inducers, was: 1) abolished by latrunculin A, an inhibitor of the actin cytoskeleton and exocytosis, and was not affected by the inhibition of mRNA synthesis (by the addition of actinomycin D or by enucleation; data not shown) or the inhibition of transcription (actinomycin D), translation (cycloheximide), or microtubule polymerization (nocodazole); 2) colocalized by cell surface immunohistochemical staining; and 3) mechanistically and molecularly related because the depletion of CRT (by siRNA or by knockout) provokes inhibition of ectoERP57 and vice versa. These findings permit me to propose the presence of a commune translocation with two partners (CRT and ERP57). To validate this suggestion, I have confirmed by immunoprecipitation the intracellular interaction between ERP57 and CRT. This intracellular-intercellular interaction was not influenced by anthracycline treatment while counting the fact that the amount of endoCRT interacting with endoERP57 was not modified further to the treatment. In contrast, the amount of ectoCRT interacting with ectoERP57 was increased further during the treatment. In addition, the amount of colocalized cell surface ectoCRT and ectoERP57 detected by immunohistochemical staining was grater after mitoxantrone treatment (data not shown). Altogether, these results prove that CRT and ERP57 are translocated together to the cell surface.

Moreover, to establish that ERP57 is important for the translocation of CRT and does not possess any immunogenic proprieties, I have evaluated the “eat me signal” capacities of ectoERP57. The depletion of both ectoCRT and ectoERP57 by siRNA was enough to suppress the phagocytosis of mitoxantrone-treated cells by DCs, whereas this defect was restored after the reconstitution of the level of ectoCRT and not the level of ectoERP57. Similarly, the secretion of IFN-γ by popliteal lymph nodes was inhibited after the injection of mitoxantrone-treated cells further along with the suppression of both ectoCRT and ectoERP57 by siRNA and restored only after the reconstitution of the level of ectoCRT by rCRT and not by rERP57. In addition, the immunogenicity of tumor vaccination with mitoxantrone-treated cells or mitomycin C plus tautomycin- treated cells was abrogated by depletion of both ectoCRT and ectoERP57, but only the reconstitution of the level of ectoCRT by rCRT has the potential to restore the level of antitumor vaccination in contrast to rERP57. In a similar way, the co-intratumoral injection of cell death inducers (mitomycin C) plus rCRT can cause tumor regression in CT26 tumor-established mice in total contrast with the coinjection of mitomycin C plus rERP57. In a complementary system, the immunogenicity of the intratumoral coadministration of mitoxantrone or mitomycin C plus tautomycin was abrogated in CT26 ERP57 siRNA-1 tumor-established mice where the cotranslocation of CRT and ERP57 has been inhibited by the ERP57 siRNA vector system; in contrast, the immunogenicity of mitoxantrone or mitomycin C plus tautomycin was restored only by the coinjection of rCRT and not that of rERP57.

Altogether, these data establish first that immunogenic properties are illustrated only by ectoCRT and second that ERP57, by interacting with CRT, is crucial for the membrane translocation of CRT.

In summary, the molecular characterization of the different biochemical alterations occurring on the surface of dying cells after an immunogenic stimulus is progressing. Indeed, it is possible to use the surface exposure of ERP57 as a new molecular alteration that characterizes immunogenicity to enable the development of a second marker, in addition to ectoCRT, for the screening of new immunogenic antitumor compounds.

I thank Gian-Maria Fimia for precious help in the performance two-dimensional gel electrophoresis, Peter van Endert for providing the recombinant CRT, and David Levy for help in the preparation of DCs. I also thank Prof. Eliane Gluckman, Prof. Jean-Marc Egly, and Dr. Jean Masson for expert advice and help.

The author is a founder, employee, shareholder, and a member of the scientific advisory board of GenCal.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported in part by a grant from GenCal and by a fellowship to M.O. from Association pour la Recherche sur le Cancer (ARC).

M.O. performed the in vivo and in vitro experiments, conducted the data analysis, conceived and designed the study and wrote the manuscript.

3

Abbreviations used in this paper: CRT, calreticulin; BM, bone marrow; DC, dendritic cell; ecto, cell surface; endo, intracellular; ER, endoplasmic reticulum; GADD34, growth arrest and DNA damage inducible protein 34; PP1, protein phosphatase 1; siRNA, small interfering RNA; z-VAD-fmk, N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone.

1
Obeid, M., A. Tesniere, T. Panaretakis, R. Tufi, N. Joza, P. van Endert, F. Ghiringhelli, L. Apetoh, N. Chaput, C. Flament, et al
2007
. Ecto-calreticulin in immunogenic chemotherapy.
Immunol. Rev.
220
:
22
-34.
2
Obeid, M., A. Tesniere, F. Ghiringhelli, G. M. Fimia, L. Apetoh, J. L. Perfettini, M. Castedo, G. Mignot, T. Panaretakis, N. Casares, et al
2007
. Calreticulin exposure dictates the immunogenicity of cancer cell death.
Nat. Med.
13
:
54
-61.
3
Obeid, M., T. Panaretakis, N. Joza, R. Tufi, A. Tesniere, P. van Endert, L. Zitvogel, G. Kroemer.
2007
. Calreticulin exposure is required for the immunogenicity of gamma-irradiation and UVC light-induced apoptosis.
Cell Death Differ.
14
:
1848
-1850.
4
Johnson, S., M. Michalak, M. Opas, P. Eggleton.
2001
. The ins and outs of calreticulin: from the ER lumen to the extracellular space.
Trends Cell Biol.
11
:
122
-129.
5
Michalak, M., E. F. Corbett, N. Mesaeli, K. Nakamura, M. Opas.
1999
. Calreticulin: one protein, one gene, many functions.
Biochem. J.
344
:
281
-292.
6
Trombetta, E. S., A. J. Parodi.
2003
. Quality control and protein folding in the secretory pathway.
Annu. Rev. Cell Dev. Biol.
19
:
649
-676.
7
Trombetta, E. S..
2003
. The contribution of N-glycans and their processing in the endoplasmic reticulum to glycoprotein biosynthesis.
Glycobiology
13
:
77R
-91R.
8
Michalak, M., J. M. Robert Parker, M. Opas.
2002
. Ca2+ signaling and calcium binding chaperones of the endoplasmic reticulum.
Cell Calcium
32
:
269
-278.
9
Frickel, E. M., R. Riek, I. Jelesarov, A. Helenius, K. Wuthrich, L. Ellgaard.
2002
. TROSY-NMR reveals interaction between ERp57 and the tip of the calreticulin P-domain.
Proc. Natl. Acad. Sci. USA
99
:
1954
-1959.
10
Booth, C., G. L. Koch.
1989
. Perturbation of cellular calcium induces secretion of luminal ER proteins.
Cell
59
:
729
-737.
11
Patton, W. F., H. Erdjument-Bromage, A. R. Marks, P. Tempst, M. B. Taubman.
1995
. Components of the protein synthesis and folding machinery are induced in vascular smooth muscle cells by hypertrophic and hyperplastic agents. Identification by comparative protein phenotyping and microsequencing.
J. Biol. Chem.
270
:
21404
-21410.
12
Sueyoshi, T., B. A. McMullen, L. L. Marnell, T. W. Du Clos, W. Kisiel.
1991
. A new procedure for the separation of protein Z, prothrombin fragment 1.2 and calreticulin from human plasma.
Thromb. Res.
63
:
569
-575.
13
Zhu, Q., P. Zelinka, T. White, M. L. Tanzer.
1997
. Calreticulin-integrin bidirectional signaling complex.
Biochem. Biophys. Res. Commun.
232
:
354
-358.
14
Kwon, M. S., C. S. Park, K. Choi, J. Ahnn, J. I. Kim, S. H. Eom, S. J. Kaufman, W. K. Song.
2000
. Calreticulin couples calcium release and calcium influx in integrin-mediated calcium signaling.
Mol. Biol. Cell
11
:
1433
-1443.
15
Gray, A. J., P. W. Park, T. J. Broekelmann, G. J. Laurent, J. T. Reeves, K. R. Stenmark, R. P. Mecham.
1995
. The mitogenic effects of the Bβ chain of fibrinogen are mediated through cell surface calreticulin.
J Biol. Chem.
270
:
26602
-26606.
16
White, T. K., Q. Zhu, M. L. Tanzer.
1995
. Cell surface calreticulin is a putative mannoside lectin which triggers mouse melanoma cell spreading.
J. Biol. Chem.
270
:
15926
-15929.
17
Obeid, M., T. Panaretakis, A. Tesniere, N. Joza, R. Tufi, L. Apetoh, F. Ghiringhelli, L. Zitvogel, G. Kroemer.
2007
. Leveraging the immune system during chemotherapy: moving calreticulin to the cell surface converts apoptotic death from “silent” to immunogenic.
Cancer Res.
67
:
7941
-7944.
18
Nakamura, K., A. Zuppini, S. Arnaudeau, J. Lynch, I. Ahsan, R. Krause, S. Papp, H. De Smedt, J. B. Parys, W. Muller-Esterl, et al
2001
. Functional specialization of calreticulin domains.
J. Cell Biol.
154
:
961
-972.
19
Culina, S., G. Lauvau, B. Gubler, P. M. van Endert.
2004
. Calreticulin promotes folding of functional human leukocyte antigen class I molecules in vitro.
J. Biol. Chem.
279
:
54210
-54215.
20
Hetz, C., M. Russelakis-Carneiro, S. Walchli, S. Carboni, E. Vial-Knecht, K. Maundrell, J. Castilla, C. Soto.
2005
. The disulfide isomerase Grp58 is a protective factor against prion neurotoxicity.
J. Neurosci.
25
:
2793
-2802.
21
Gottardi, C. J., L. A. Dunbar, M. J. Caplan.
1995
. Biotinylation and assessment of membrane polarity: caveats and methodological concerns.
Am. J. Physiol.
268
:
F285
-F295.
22
Hanwell, D., T. Ishikawa, R. Saleki, D. Rotin.
2002
. Trafficking and cell surface stability of the epithelial Na+ channel expressed in epithelial Madin-Darby canine kidney cells.
J. Biol. Chem.
277
:
9772
-9779.
23
Gardai, S. J., K. A. McPhillips, S. C. Frasch, W. J. Janssen, A. Starefeldt, J. E. Murphy-Ullrich, D. L. Bratton, P. A. Oldenborg, M. Michalak, P. M. Henson.
2005
. Cell-surface calreticulin initiates clearance of viable or apoptotic cells through trans-activation of LRP on the phagocyte.
Cell
123
:
321
-334.
24
Storkus, W. J., L. D. Falo, Jr.
2007
. A ‘good death’ for tumor immunology.
Nat. Med.
13
:
28
-30.
25
Clarke, C., M. J. Smyth.
2007
. Calreticulin exposure increases cancer immunogenicity.
Nat. Biotechnol.
25
:
192
-193.
26
Ellgaard, L., A. Helenius.
2001
. ER quality control: towards an understanding at the molecular level.
Curr. Opin. Cell Biol.
13
:
431
-437.