Immunogenic cancer cell death: a key-lock paradigm

Available online at www.sciencedirect.com

Immunogenic cancer cell death: a key-lock paradigm Antoine Tesniere1,2,3, Lionel Apetoh2,3,4, Francois Ghiringhelli2,3,4, Nicholas Joza1,2,3, Theocharis Panaretakis1,2,3, Oliver Kepp1,2,3, Frederic Schlemmer1,2,3, Laurence Zitvogel2,3,4 and Guido Kroemer1,2,3 Physiological cell death, which occurs as a continuous byproduct of cellular turnover, is non-immunogenic or even tolerogenic, thereby avoiding autoimmunity. By contrast, cancer cell death elicited by radiotherapy and some chemotherapeutic agents such as anthracyclines is immunogenic. Recent data suggest that innate and cognate immune responses elicited by such anti-cancer agents are required for an optimal therapeutic outcome, underscoring the clinical relevance of immunogenic cell death. Here we discuss the concept that immunogenic death involves changes in the composition of the cell surface, as well as the release of soluble immunogenic signals that occur in a defined temporal sequence. This ‘key’ then operates on a series of receptors expressed by dendritic cells (DC, the ‘lock’) to allow for the presentation of tumor antigens to T cells and for the initiation of a productive immune response. Immunogenic cell death is characterized by the early cell surface exposure of chaperones including calreticulin and/or heat shock proteins, which determine the uptake of tumor antigens and/or affect DC maturation. Moreover, the late release of High mobility group box 1 (HMGB1), which acts on toll-like receptor 4 (TLR4), is required for optimal presentation of antigens from dying tumor cells. Nonetheless, numerous details on the molecular events that define immunogenicity remain to be defined, both at the level of the dying cancer cells and at the level of the responding innate effectors.

tiation is an irreversible process linked to proliferative arrest and apoptotic removal), proliferative arrest (which in the case of senescence is indefinite) or cell death. Cancer cell demise often occurs through apoptosis, a stereotyped pattern of morphological changes that involves chromatin condensation (pyknosis), nuclear fragmentation (karyorhexis), shrinkage of the cytoplasm, blebbing of the plasma membrane, and final disintegration of the cell into membrane-surrounded apoptotic bodies [1,2]. These changes are readily observed in vitro but are rarely seen in vivo, because dying cells are efficiently recognized, engulfed and degraded by neighboring cells before they enter the late stages of the apoptotic process [3,4]. In some cases, cancer cells succumb to non-apoptotic cell death. Thus, tumor cells can die by necrosis (characterized by swelling of the cell and the cytoplasmic organelles before the plasma membrane ruptures and the cellular content is spilled into the intercellular space), by manifesting massive autophagy (sequestration of large parts of the cytoplasm in autophagic vacuoles, often before the cells undergo apoptosis), or as a result of mitotic catastrophe (morphologically discernible by multi- or micronucleation and/or prolonged mitotic arrest before death) [1].

Addresses 1 INSERM, U848, F-94805 Villejuif, France 2 Institut Gustave Roussy, Pavillon de Recherche 1, 39 rue Camille-Desmoulins, F-94805 Villejuif, France 3 Universite´ Paris-Sud, F-94805 Villejuif, France 4 INSERM, U805, F-94805 Villejuif, France

There are teleological arguments to the notion that apoptosis must be non-immunogenic or even tolerogenic (because physiological cell death occurs through apoptosis, but does not lead to autoimmunity), while necrosis (which is often pathological) has been linked to inflammatory processes. However, the theoretical assumption that apoptosis equals non-immunogenic and necrosis equals immunogenic cell death does not withstand rigorous experimental verification, at least in models of tumor vaccination [5]. Rather, it seems that apoptosis is non-uniform in biochemical terms, meaning that distinct molecular pathways can lead to cell death and induce stimulus-specific alterations, for example, in the composition of the proteome that is associated with the plasma membrane, before the cells adopt the apparently uniform morphological characteristics of apoptosis. Cellular stress responses – which precede cell death – are highly diversified, meaning that the ‘history’ of the pre-apoptotic events conditions the composition and even the surface characteristics of apoptotic bodies. Moreover, the apoptotic execution phase itself can involve the variable contribution of distinct catabolic hydrolases including caspases and

Corresponding author: Kroemer, Guido ([email protected])

Current Opinion in Immunology 2008, 20:504–511 This review comes from a themed issue on Immune Response against Dying Cells Edited by Guido Kroemer and Laurence Zitvogel Available online 23rd June 2008 0952-7915/$ – see front matter # 2008 Elsevier Ltd. All rights reserved. DOI 10.1016/j.coi.2008.05.007

Introduction Malignant cells can respond to chemotherapy either by differentiation (which in the case of terminal differenCurrent Opinion in Immunology 2008, 20:504–511

www.sciencedirect.com

Immunogenic cell death Tesniere et al. 505

caspase-independent death effectors [6], implying that similar morphologies may have been acquired through distinct biochemical routes. In the healthy human adult, each second several million cells succumb to programmed cell death, a byproduct of tissue renewal and cell turnover. This continuous cell turnover does not lead to any adverse inflammatory or autoimmune reactions. By contrast, the first-line defense against viral invasion consists in the apoptotic elimination of infected cells. When cells die as a result of microbial infection, they must induce an immune response. Moreover, as discussed below, tumor cells succumbing to a defined subset of cytotoxic chemotherapeutics also trigger highly specific immune responses. The question is then how the immune system distinguishes between distinct immunogenic and nonimmunogenic cell death modalities. Here, we suggest that a defined sequence of molecular events act as a ‘key’ to open the ‘lock’ that triggers the immune response (Figure 1). In this review, we will discuss which distinctive biochemical features define the key, the immunogenic cancer cell death event, and which recognition mechanisms compose the lock at the level of dendritic cells (DC), the first-line decision-makers of the innate immune system.

Distinctive properties of immunogenic cell death When mouse cancer cells are treated in vitro to induce apoptosis and then are injected subcutaneously in the absence of any adjuvant, they elicit a specific immune response that precludes the growth of the same cancer cell line when it is injected at a distinct site. Thus, for instance anthracyclines and ionizing irradiation are highly efficient in eliciting immunogenic cell death, while most other agents (such as etoposide and mitomycin C) induce non-immunogenic cell death [7]. Calreticulin (CRT) exposure

CT26 colon cancer cells and MCA205 fibrosarcoma cells treated with anthracyclines or ionizing irradiation have been found to translocate CRT from the lumen of the endoplasmic reticulum (ER) to the cell surface [8] (Figure 2a). This conversion of endo-CRT to ectoCRT affects only a portion (5–10%) of the entire cellular pool of CRT. Blockade or knockdown of CRT suppressed the phagocytosis of anthracycline-treated tumor cells by dendritic cells and abolished their immunogenicity in mice. Conversely, adsorption of recombinant CRT onto cells treated with etoposide and mitomycin C conferred immunogenicity of cell death and enhanced antitumor effects in vivo [8].

Figure 1

Key events required to unlock dendritic cell activation by dying tumor cells. Immunogenic tumor cell death is characterized by a temporal sequence of events including early translocation of calreticulin to the cell surface, release and exposure of heat shock proteins, and late release of HMGB1, allowing to trigger DC-mediated specific anti tumor immune response. Components released from or exposed at the surface of dying tumor cells act at several levels of the immune response: uptake of apoptotic bodies (calreticulin), activation of dendritic cells (heat shock proteins), antigen processing (HMGB1), maturation of dendritic cells, and activation of T lymphocytes. www.sciencedirect.com

Current Opinion in Immunology 2008, 20:504–511

506 Immune Response against Dying Cells

Figure 2

governed by a few rules. First, it is suppressed when caspases are inhibited [8]. Second, phosphorylation of the eukaryotic initiation factor 2a (eIF2a), which is one of the hallmarks of ER stress, can be induced using inhibitors of the protein phosphatase PP1, and is both required and sufficient for CRT exposure. Moreover an increase in Ca2+ leakage from the ER induces CRT exposure [8,14]. Third, CRT exposure depends on the interaction of CRT with ERp57 as demonstrated by the observation that cells in which ERp57 has been removed by knockout or knockdown lose the capacity of CRT exposure [11]. These observations point to a complex mechanism of CRT exposure that remains to be elucidated in its molecular details. For instance, it is not known to which precise plasma membrane structures CRT binds, although there is clearly a saturable CRT binding site on the cell surface (our unpublished results). Moreover it is not know whether the CRT2 isoform, a novel cancer-testis antigen [15], can mediate similar immunogenic effects as the dominant, ubiquitous CRT isoform. Heat shock protein exposure

Molecular properties of immunogenic cancer cell death. Immunogenic cell death is associated with specific molecular events. (a) The rapid preapoptotic translocation of calreticulin from the endoplasmic reticulum lumen to the cell surface is mediated by eiF2a phosphorylation and caspase activation. (b) Heat shock proteins can be exposed at the membrane surface or released during tumor cell death. (c) HMGB1 is released from nucleus at late stage apoptosis in the extracellular medium.

In response to anthracyclines and ionizing irradiation, CRT exposure occurs at an early, pre-apoptotic stage, several hours before the cells expose phosphatidylserine on the outer leaflet of the plasma membrane [8–10]. Preapoptotic CRT exposure is accompanied by exposure of ERp57 but not of other ER proteins such as calnexin or the KDEL receptor [11]. By contrast, in response to other cell death inducers, CRT exposure occurs concomitantly with PS exposure, at a late apoptotic stage when multiple ER proteins are found simultaneously at the cell surface [12,13]. Pre-apoptotic CRT exposure is Current Opinion in Immunology 2008, 20:504–511

Cellular stress can elicit a number of adaptive mechanisms including the expression of heat shock protein (HSPs), which can be released from dead cells after primary or secondary necrosis [16]. One of the distinctive immunogenic signals may be the expression of heat shock proteins on the surface of dying tumor cells (Figure 2b). Bortezomib, a specific inhibitor of the 26S proteasome and inducer of HSPs, has shown clinical activity in several human tumors, including myeloma. Dhodapkar and coworkers [17] demonstrated that the uptake of human myeloma cells by DC after tumor cell death by bortezomib, but not gamma irradiation or steroids, leads to the induction of antitumor immunity, including against primary tumor cells, without the need for any additional adjuvants. Bortezomib induces the expression of heat shock protein 90 (hsp90) on the surface of dying human myeloma tumor cells, and it is the geldanamycin-inhibitable recognition of such tumor cells by DC that leads to the generation of anti-tumor T cells [17]. Similarly, in preclinical experiments, manipulation designed to increase the expression of HSPs such as local hyperthermia applied to tumors [18], infection with HSP70-encoding oncolytic viruses or transfection with HSP70 followed by irradiation [19] can increase immunogenicity. Extracellular targeting of the ER chaperone glucose-regulated protein 170 (GRP170) also enhances the immunogenicity of tumor cells [20]. There are numbers of mechanisms through which HSPs can enhance immunogenicity. On the one hand they may improve the recognition and uptake of dying cells by DC when they are present on the cell surface. On the other hand, tumor-derived antigenic peptides may bind to HSPs and may be recycled for antigenic presentation in a particularly efficient fashion [21]. www.sciencedirect.com

Immunogenic cell death Tesniere et al. 507

HMGB1 release

Nearly one decade ago, Bianchi et al. reported the seminal finding that necrotic cell death results in the release of the pro-inflammatory factor high mobility group box 1 (HMGB1) while apoptotic cell death leads to the retention of HMGB1 in the nucleus throughout the process, meaning that HMGB1 remains attached to the nucleus even when the plasma membrane bursts during secondary necrosis [22]. In the meantime, it has become clear that this distinction is less obvious than it has been thought initially and that apoptotic cells can in fact release HMGB1 [23]. Thus, genotoxic agents that activate polyadenosylribosyl polymerase (PARP), including alkylating agents, can stimulate the release of HMGB1 from its association with chromatin, presumably as a result of the direct PARP-mediated polyadenosylribosylation of HMGB1 [24]. Moreover, immunogenic cell death inducers such as anthracyclines stimulate the release of HMGB1, which can be detected in the supernatant of cultured cells when they undergo secondary necrosis [25] (Figure 2c). Knockdown of HMGB1 from tumor cells by transfection with small interfering RNAs decreased their immunogenicity; these cells hence become unable to mediate cancer vaccination after anthracycline treatment. Similarly, coinjection of a neutralizing antibody specific for HMGB1 abolishes the immunogenicity of anthracycline-treated CT26 or EG7 cells in vivo [25].

How do DC decode the immunogenicity of cell death? When mice carrying the diphtheria toxin receptor (DTR) under the control of the DC-specific CD11c (CD11c) promoter are injected with diphtheria toxin, DC are transiently eliminated from the immune system. In these conditions of DC depletion, mice lose their capacity to mount an immune response against dying tumor cells [7,25]. Immature bone marrow-derived DC (BM-DC) can phagocytose dying tumor cells in vitro and present class I and class II-restricted tumor-derived antigenic peptides to specific T cells in vitro or in vivo, after re-injecting them into naı¨ve recipients [25]. These results indicate that DC are the decisive effectors of the innate immune system that decode the immunogenicity of cell death. The question remains through which mechanisms cells that succumb to immunogenic cell death stimulate DC to present tumor antigens in an optimal fashion. The elusive CRT receptor on DC

Henson and colleagues have been the first to describe that CRT exposure, which in their system occurred concomitant with PS exposure, would facilitate the phagocytosis of dying cells by J774 macrophages [12]. In this particular scenario, blockade of CD91 (also called LDL-receptorrelated protein, LRP) on the surface of J774 macrophages, either by an antibody or by addition of recombinant receptor-associated protein (RAP), blocked phagocytosis [12]. However, CD91 is a multifunctional receptors that www.sciencedirect.com

recognizes some 30 different ligands, implying that the before mentioned results do not prove that CD91 is the (sole) CRT receptor. It is a well-established paradigm that different kinds of phagocytes use distinct receptors for the recognition of apoptotic corpses [26]. Hence, it remains to be determined whether DC, which can engulf cells that express CRT on their surface in the absence of PS exposure [8] use CD91 as the principal CRT receptor. CRT is a ‘sticky’ protein and it reportedly interacts with the soluble serum proteins thrombospondin as well as the complement factor C1q, both of which may serve as ‘bridges’ to facilitate the recognition of CRT by phagocytes [27]. On theoretical grounds, the receptor(s) of CRT on DC might be CD91 [12], scavenger receptor A (SR-A), scavenger receptor expressed by endothelial cell-I (SREC-I) [28], or CD40 ligand, TRAIL or Fas ligand [29], which all have been shown to interact with CRT and/or to mediate the phagocytic recognition of CRTexposing cells (Figure 3). Systematic knockout studies should evaluate which among these receptors is required for mounting a productive immune response against dying cells. Thus far, it appears that SR-A functions as a negative regulator of antigen presentation by DC [30], but the impact on this and other putative CRT receptors on the presentation of antigen from dying tumor cells has not been specifically addressed. At present, it is unclear whether cells that manifest early CRT exposure (without concomitant PS exposure) and cells that manifest late CRT exposure (with general mixing of the ER and plasma membranes) are recognized by the same population of phagocytic cells in vivo. It might be predicted that immunogenic events would destine stressed and dying tumor cells to be recognized preferentially by DC while corpses resulting from nonimmunogenic cell death should be handled by neighboring cells (that are not antigen presenting cells) or tissue macrophages. This conjecture awaits urgent experimental verification. TLR4 as a functionally important HMGB1 receptor

HMGB1 binds promiscuously to multiple proteins and reportedly can bind to at least three distinct surface receptors, namely the receptor for advanced glycosylation products (RAGE) [31,32], toll-like receptor 2 (TLR2) [33] and toll-like receptor 4 (TLR4) [34], all of which can be present on the surface of DC. However, the knockout of TLR4 (but not that of TLR2) was able to abolish the capacity of BM-DC to present antigen from dying tumor cells to CD4+ or CD8+ T cells in vitro, and the knockout of TLR4 (but not that of TLR2 or any other TLR) was able to abolish T cell priming by dying tumor cells in vivo. The TLR4 from mouse DC exposed to supernatants from dying tumor cells was found to bind HMGB1, as determined by co-immunoprecipitation experiments, Current Opinion in Immunology 2008, 20:504–511

508 Immune Response against Dying Cells

Figure 3

DC receptors involved in the response to immunogenic cancer cell death. HMGB1 is able to bind to the TLR4 receptor on DC. TLR4 then inhibits the fusion of phagosomes and lysosomes, thereby allowing tumor derived antigens to be processed and presented along with MHC and costimulatory molecules on the surface of DC. CRT interacts with multiple receptors on DC and mediate apoptotic bodies phagocytosis by DC. These potential targets include Scavenger receptor A, CD91, SREC1, C1q, and CD40. CRT.

indicating that the two molecules can indeed interact directly in this setting [25]. Importantly, recombinant HMGB1 added to HeLa cells (which express TLR4) could be co-immunoprecipitated with TLR4, and this interaction was increased upon transfection of HeLa cells with human wild type TLR4, yet was decreased upon transfection with a mutant TLR4 allele carrying the Asp299Gly mutation [35]. Thus, the Asp299Gly mutation has a dominant-negative effect on the TLR4/HMGB1 interaction.

previously been shown for macrophages, in a pioneering study by Nakanishi and coworkers [36]. Accordingly, inhibition of the fusion between lysosomes and phagosomes by addition of chloroquine restored antigen presentation by tlr4 / BM-DC from mice. Very similar results were obtained with BM-DC from patients carrying the TLR4 Asp299Gly mutation. Such BM-DC were relatively inefficient in presenting class I-restricted antigen from dying melanoma cells, and this defect could be restored by addition of chloroquine [25].

In contrast to wild-type (WT) BM-DC, tlr4 / BM-DC were defective in their capacity to present the OVAderived SIINFEKL peptide from OVA-expressing EG7 cells that were killed by X-irradiation, an immunogenic cell death inducer. This correlated with a more rapid lysosomal degradation of the phagocytic cargo in tlr4 / BM-DC as compared to WT BM-DC [25]. Thus, TLR4 acts as an endogenous inhibitor of the formation of phagolysosomes in DC (Figure 3), exactly as it has

The parallelism between the TLR4 defect (in mice) and the TLR4 Asp299Gly mutation (in humans) goes further. Syngenic tumor implanted in tlr4 / mice responded poorly to local radiotherapy or systemic chemotherapy as compared to similar tumors implanted in WT mice, indicating that a TLR4-dependent immune response contributes to the efficacy of the therapeutic regimen. Similarly, breast cancer patients that bear the tlr4 Asp299Gly allele (which in Caucasians affects 10–12%

Current Opinion in Immunology 2008, 20:504–511

www.sciencedirect.com

Immunogenic cell death Tesniere et al. 509

of the population) and that received local radiotherapy and systemic anthracycline therapy developed metastases more rapidly than patients bearing the normal TLR4 allele [25]. These results emphasize the importance of TLR4 (and by extension the TLR4/HMGB1 interaction) in the immune response against stressed and dying tumor cells. However, numerous important questions remain to be answered. For instance, it is not known what are the structural requirements that govern the interactions between HMGB1 and TLR4. Do they involve CD14 and MD2, both of which are required for the interaction between TLR4 and bacterial lipopolysaccharide? Similarly, it is not known which downstream signals link TLR4 to the suppression of phagolysosome formation. As it stands, myeloid differentiation primary response protein 88 (MyD88) (but not TRIF) is known to be required for the immune response against dying tumor cells [25]. However, the detailed mechanisms of TLR4mediated phagolysosome repression are entirely unknown. Elusive maturation signals

Antigen presentation by DC is a multi-step process in which the antigen is first taken up by immature DC, followed by a maturation step in which DC acquire the capacity for optimal antigen presentation and co-stimulation (Figure 3). The addition of recombinant CRT fails to induce DC maturation [37]. Moreover, live tumor cells to which CRT is adsorbed fail to induce DC maturation. By contrast, only tumor cells that undergo immunogenic cell death are able to stimulate DC maturation in vitro, suggesting that CRT, though necessary for optimal uptake of dying tumor cells by immature DC, is not a maturation signal for DC and that another property of cell death must account for DC maturation. One of these factors might be HMGB1, which can induce the phenotypic maturation of DC, as evidenced by increased CD83, CD54, CD80, CD40, CD58, and MHC class II expression [38]. However, Coyle and co-workers found in another study that HMGB1 itself was not sufficient to induce DC maturation unless it was complexed to DNA and in particular to class A CpG oligodeoxynucleotides [39]. In our hands, the presence of TLR4 was not required for HMGB1 to induce DC maturation [25], suggesting that another HMGB1 receptor such as RAGE might mediate DC maturation [39,40]. Other signals, different from TLR4 and HMGB1, are also potential candidates to stimulate DC maturation. Indeed, several chaperones including HSP70 [41] and GRP94/gp96 [42] can stimulate DC maturation in vitro. Nonetheless, the search for the DC maturation signal emanating from dying tumor cells is far from over. For instance, uric acid, an endogenous danger signaling molecule that can be released from dying cells, has been shown to stimulate DC to produce interferon-g (IFN-g) and interleukin-12 (IL-12) [43]. www.sciencedirect.com

Open questions and perspectives As outlined above, the overall shape of the key (immunogenic cell death) and lock (at the level of DC) that usually blocks immune responses is being characterized at the molecular level. However, the therapeutic induction of anti-cancer immune responses will require a detailed knowledge of all the peculiarities of this key. Which stimuli have to be delivered with which intensity in which exact spatiotemporary sequence to render cancer cell death immunogenic? Is there one key or are there several different cell death modalities that use only partially overlapping motifs for the induction of immune responses? That there may be different immunogenic events is indicated by the finding that DNA damage responses and oncogenic transformation can stimulate the expression of NKG2D ligands on the surface of tumor cells [44,45]. Moreover, p53induced cellular senescence, which in principle is different from apoptosis, can stimulate an innate immune response [46]. So, can we design one master key that opens different locks? Or is the immune response simultaneously secure behind several locks that all have to be unlocked, one after the other, to achieve a therapeutic immune response? The answers to these questions are not simply theoretical points; they will guide our efforts to improve conventional anti-cancer therapies and to design novel therapeutic approaches. As it stands, it appears clear that the plain success of current, clinically used radiotherapies and chemotherapies critically depends on the synergic interaction with the immune system, activated by dying cancer cells [25,47]. We anticipate that a more thorough understanding of the cellular and molecular processes involved in the immune response triggered by immunogenic tumor cell death will have a major impact on future oncology practice.

Acknowledgements The authors are supported by grants from the Ligue Nationale contre le Cancer, the European Union, Cance´ropoˆle Ile-de-France, Institut National du Cancer, and Association for International Cancer Research. AT is supported by INSERM, LA receives grants from Fondation pour la recherche me´dicale, TP is supported by a grant from the Swedish Research Council, OK receives fellowship from EMBO and GK is supported by Agence Nationale pour la Recherche. The authors declare no competing financial interests.

References 1.

Galluzzi L, Maiuri MC, Vitale I, Zischka H, Castedo M, Zitvogel L, Kroemer G: Cell death modalities: classification and pathophysiological implications. Cell Death Differ 2007, 14:1237-1243.

2.

Kroemer G, El-Deiry WS, Golstein P, Peter ME, Vaux D, Vandenabeele P, Zhivotovsky B, Blagosklonny MV, Malorni W, Knight RA et al.: Classification of cell death: recommendations of the Nomenclature Committee on Cell Death. Cell Death Differ 2005, 12(Suppl. 2):1463-1467.

3.

Savill J, Dransfield I, Gregory C, Haslett C: A blast from the past: clearance of apoptotic cells regulates immune responses. Nat Rev Immunol 2002, 2:965-975.

4.

Hengartner MO: The biochemistry of apoptosis. Nature 2000, 407:770-776.

5.

Zitvogel L, Casares N, Pequignot MO, Chaput N, Albert ML, Kroemer G: Immune response against dying tumor cells. Adv Immunol 2004, 84:131-179. Current Opinion in Immunology 2008, 20:504–511

510 Immune Response against Dying Cells

6.

Kroemer G, Martin SJ: Caspase-independent cell death. Nat Med 2005, 11:725-730.

7.

Casares N, Pequignot MO, Tesniere A, Ghiringhelli F, Roux S, Chaput N, Schmitt E, Hamai A, Hervas-Stubbs S, Obeid M et al.: Caspase-dependent immunogenicity of doxorubicin-induced tumor cell death. J Exp Med 2005, 202:1691-1701.

8.

9.

Obeid M, Tesniere A, Ghiringhelli F, Fimia GM, Apetoh L, Perfettini JL, Castedo M, Mignot G, Panaretakis T, Casares N et al.: Calreticulin exposure dictates the immunogenicity of cancer cell death. Nat Med 2007, 13:54-61. Obeid M, Panaretakis T, Joza N, Tufi R, Tesniere A, van Endert P, Zitvogel L, Kroemer G: Calreticulin exposure is required for the immunogenicity of gamma-irradiation and UVC light-induced apoptosis. Cell Death Differ 2007, 14:1848-1850.

10. Obeid M, Tesniere A, Panaretakis T, Tufi R, Joza N, van Endert P, Ghiringhelli F, Apetoh L, Chaput N, Flament C et al.: Ectocalreticulin in immunogenic chemotherapy. Immunol Rev 2007, 220:22-34. 11. Panaretakis T, Joza N, Modjtahedi N, Tesniere A, Vitale I, Durchschlag M, Fimia GM, Kepp O, Piacentini M, Froehlich KU, van Endert P, Zitvogel L, Madeo F, Kroemer G: The cotranslocation of ERp57 and calreticulin determines the immunogenicity of cell death. Cell Death Differ 2008. 12. Gardai SJ, McPhillips KA, Frasch SC, Janssen WJ, Starefeldt A, Murphy-Ullrich JE, Bratton DL, Oldenborg PA, Michalak M, Henson PM: Cell-surface calreticulin initiates clearance of viable or apoptotic cells through trans-activation of LRP on the phagocyte. Cell 2005, 123:321-334. 13. Franz S, Herrmann K, Fuhrnrohr B, Sheriff A, Frey B, Gaipl US, Voll RE, Kalden JR, Jack HM, Herrmann M: After shrinkage apoptotic cells expose internal membrane-derived epitopes on their plasma membranes. Cell Death Differ 2007, 14:733-742. 14. Tufi R, Panaretakis T, Bianchi K, Criollo A, Fazi B, Di Sano F, Tesniere A, Kepp O, Paterlini-Brechot P, Zitvogel L et al.: Reduction of endoplasmic reticulum Ca(2+) levels favors plasma membrane surface exposure of calreticulin. Cell Death Differ 2008, 15:274-282. 15. Hayashi E, Matsuzaki Y, Hasegawa G, Yaguchi T, Kurihara S, Fujita T, Kageshita T, Sano M, Kawakami Y: Identification of a novel cancer-testis antigen CRT2 frequently expressed in various cancers using representational differential analysis. Clin Cancer Res 2007, 13:6267-6274. 16. Saito K, Dai Y, Ohtsuka K: Enhanced expression of heat shock proteins in gradually dying cells and their release from necrotically dead cells. Exp Cell Res 2005, 310:229-236. 17. Spisek R, Charalambous A, Mazumder A, Vesole DH, Jagannath S, Dhodapkar MV: Bortezomib enhances dendritic cell (DC)mediated induction of immunity to human myeloma via exposure of cell surface heat shock protein 90 on dying tumor cells: therapeutic implications. Blood 2007, 109:4839-4845. 18. Mukhopadhaya A, Mendecki J, Dong X, Liu L, Kalnicki S, Garg M, Alfieri A, Guha C: Localized hyperthermia combined with intratumoral dendritic cells induces systemic antitumor immunity. Cancer Res 2007, 67:7798-7806. 19. Chan T, Chen Z, Hao S, Xu S, Yuan J, Saxena A, Qureshi M, Zheng C, Xiang J: Enhanced T-cell immunity induced by dendritic cells with phagocytosis of heat shock protein 70 gene-transfected tumor cells in early phase of apoptosis. Cancer Gene Ther 2007, 14:409-420. 20. Wang XY, Arnouk H, Chen X, Kazim L, Repasky EA, Subjeck JR: Extracellular targeting of endoplasmic reticulum chaperone glucose-regulated protein 170 enhances tumor immunity to a poorly immunogenic melanoma. J Immunol 2006, 177:1543-1551.

23. Bell CW, Jiang W, Reich CF 3rd, Pisetsky DS: The extracellular release of HMGB1 during apoptotic cell death. Am J Physiol Cell Physiol 2006, 291:C1318-C1325. 24. Ditsworth D, Zong WX, Thompson CB: Activation of poly(ADP)ribose polymerase (PARP-1) induces release of the proinflammatory mediator HMGB1 from the nucleus. J Biol Chem 2007, 282:17845-17854. 25. Apetoh L, Ghiringhelli F, Tesniere A, Obeid M, Ortiz C, Criollo A, Mignot G, Maiuri MC, Ullrich E, Saulnier P et al.: Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy. Nat Med 2007, 13:1050-1059. 26. Savill J, Fadok V: Corpse clearance defines the meaning of cell death. Nature 2000, 407:784-788. 27. Ogden CA, deCathelineau A, Hoffmann PR, Bratton D, Ghebrehiwet B, Fadok VA, Henson PM: C1q and mannose binding lectin engagement of cell surface calreticulin and CD91 initiates macropinocytosis and uptake of apoptotic cells. J Exp Med 2001, 194:781-795. 28. Berwin B, Delneste Y, Lovingood RV, Post SR, Pizzo SV: SREC-I, a type F scavenger receptor, is an endocytic receptor for calreticulin. J Biol Chem 2004, 279:51250-51257. 29. Duus K, Pagh RT, Holmskov U, Hojrup P, Skov S, Houen G: Interaction of calreticulin with CD40 ligand, TRAIL and Fas ligand. Scand J Immunol 2007, 66:501-507. 30. Wang XY, Facciponte J, Chen X, Subjeck JR, Repasky EA: Scavenger receptor-A negatively regulates antitumor immunity. Cancer Res 2007, 67:4996-5002. 31. Dumitriu IE, Baruah P, Bianchi ME, Manfredi AA, Rovere-Querini P: Requirement of HMGB1 and RAGE for the maturation of human plasmacytoid dendritic cells. Eur J Immunol 2005, 35:2184-2190. 32. Kokkola R, Andersson A, Mullins G, Ostberg T, Treutiger CJ, Arnold B, Nawroth P, Andersson U, Harris RA, Harris HE: RAGE is the major receptor for the proinflammatory activity of HMGB1 in rodent macrophages. Scand J Immunol 2005, 61:1-9. 33. Chen CJ, Kono H, Golenbock D, Reed G, Akira S, Rock KL: Identification of a key pathway required for the sterile inflammatory response triggered by dying cells. Nat Med 2007, 13:851-856. 34. Park JS, Gamboni-Robertson F, He Q, Svetkauskaite D, Kim JY, Strassheim D, Sohn JW, Yamada S, Maruyama I, Banerjee A et al.: High mobility group box 1 protein interacts with multiple Toll-like receptors. Am J Physiol Cell Physiol 2006, 290:C917-C924. 35. Apetoh L, Ghiringhelli F, Tesniere A, Criollo A, Ortiz C, Lidereau R, Mariette C, Chaput N, Mira JP, Delaloge S et al.: The interaction between HMGB1 and TLR4 dictates the outcome of anticancer chemotherapy and radiotherapy. Immunol Rev 2007, 220:47-59. 36. Shiratsuchi A, Watanabe I, Takeuchi O, Akira S, Nakanishi Y: Inhibitory effect of Toll-like receptor 4 on fusion between phagosomes and endosomes/lysosomes in macrophages. J Immunol 2004, 172:2039-2047. 37. Bak SP, Amiel E, Walters JJ, Berwin B: Calreticulin requires an ancillary adjuvant for the induction of efficient cytotoxic T cell responses. Mol Immunol 2008, 45:1414-1423. 38. Messmer D, Yang H, Telusma G, Knoll F, Li J, Messmer B, Tracey KJ, Chiorazzi N: High mobility group box protein 1: an endogenous signal for dendritic cell maturation and Th1 polarization. J Immunol 2004, 173:307-313.

21. Srivastava PK: Immunotherapy for human cancer using heat shock protein–peptide complexes. Curr Oncol Rep 2005, 7:104-108.

39. Tian J, Avalos AM, Mao SY, Chen B, Senthil K, Wu H, Parroche P, Drabic S, Golenbock D, Sirois C et al.: Toll-like receptor 9-dependent activation by DNA-containing immune complexes is mediated by HMGB1 and RAGE. Nat Immunol 2007, 8:487-496.

22. Scaffidi P, Misteli T, Bianchi ME: Release of chromatin protein HMGB1 by necrotic cells triggers inflammation. Nature 2002, 418:191-195.

40. Dumitriu IE, Baruah P, Valentinis B, Voll RE, Herrmann M, Nawroth PP, Arnold B, Bianchi ME, Manfredi AA, RovereQuerini P: Release of high mobility group box 1 by dendritic

Current Opinion in Immunology 2008, 20:504–511

www.sciencedirect.com

Immunogenic cell death Tesniere et al. 511

cells controls T cell activation via the receptor for advanced glycation end products. J Immunol 2005, 174:7506-7515. 41. Massa C, Melani C, Colombo MP: Chaperon and adjuvant activity of hsp70: different natural killer requirement for crosspriming of chaperoned and bystander antigens. Cancer Res 2005, 65:7942-7949. 42. Baker-LePain JC, Sarzotti M, Nicchitta CV: Glucose-regulated protein 94/glycoprotein 96 elicits bystander activation of CD4+ T cell Th1 cytokine production in vivo. J Immunol 2004, 172:4195-4203. 43. Endo Y, Sakai R, Ouchi M, Onimatsu H, Hioki M, Kagawa S, Uno F, Watanabe Y, Urata Y, Tanaka N et al.: Virus-mediated oncolysis induces danger signal and stimulates cytotoxic T-lymphocyte activity via proteasome activator upregulation. Oncogene 2007.

www.sciencedirect.com

44. Gasser S, Orsulic S, Brown EJ, Raulet DH: The DNA damage pathway regulates innate immune system ligands of the NKG2D receptor. Nature 2005, 436:1186-1190. 45. Unni AM, Bondar T, Medzhitov R: Intrinsic sensor of oncogenic transformation induces a signal for innate immunosurveillance. Proc Natl Acad Sci U S A 2008, 105:1686-1691. 46. Xue W, Zender L, Miething C, Dickins RA, Hernando E, Krizhanovsky V, Cordon-Cardo C, Lowe SW: Senescence and tumour clearance is triggered by p53 restoration in murine liver carcinomas. Nature 2007, 445:656-660. 47. Zitvogel L, Apetoh L, Ghiringhelli F, Andre´ F, Tesniere A, Kroemer G: The anticancer immune response: indispensable for therapeutic success? J Clin Invest 2008, 118(6):1991-2001.

Current Opinion in Immunology 2008, 20:504–511