j ourna l homepage: www.e lsev ie r .com/ locate /bbacan
Review
Immunogenic cell death, DAMPs and anticancer therapeutics:An emerging amalgamation
Abhishek D. Garg a, Dominika Nowis b, Jakub Golab b, Peter Vandenabeele c,d,Dmitri V. Krysko c,d,⁎, Patrizia Agostinis a,⁎a Department of Molecular Cell Biology, Catholic University of Leuven, Belgiumb Department of Immunology, Center of Biostructure, Medical University of Warsaw, Polandc Molecular Signalling and Cell Death Unit, Department for Molecular Biomedical Research, VIB, Belgiumd Department of Molecular Biology, Ghent University, Belgium
Abbreviations: APCs, Antigen-presenting Cells; ATP, ADCs, Dendritic cell(s); DT-EGF, Epidermal Growth FactoMobility Group Box-1; HSP, Heat Shock Protein(s); IFN, IComplex; NFκB, Nuclear Factor kappa-light-chain-enhanTherapy; PERK, PKR-like ER kinase; PKR, Protein kinaseFactor; TLR, Toll-like Receptor(s); TNF, Tumour Necrosi⁎ Corresponding authors. P. Agostinis is to be contacte
Immunogenic profile of certain cancer cell death mechanisms has been transmuted by research publishedover a period of last few years and this change has been so drastic that a new (sub)class of apoptotic cancercell death, redefined as ‘immunogenic apoptosis’ has started taking shape. In fact, it has been shown that thischemotherapeutic agent-specific immunogenic cancer cell death modality has the capabilities to induce‘anticancer vaccine effect’, in vivo. These new trends have given an opportunity to combine tumour cell killand antitumour immunity within a single paradigm, a sort of ‘holy grail’ of anticancer therapeutics. At themolecular level, it has been shown that the immunological silhouette of these cell death pathways is definedby a set of molecules called ‘damage-associated molecular patterns (DAMPs)’. Various intracellularmolecules like calreticulin (CRT), heat-shock proteins (HSPs), high-mobility group box-1 (HMGB1) protein,have been shown to be DAMPs exposed/secreted in a stress agent/factor-and cell death-specific manner.These discoveries have motivated further research into discovery of new DAMPs, new pathways for theirexposure/secretion, search for new agents capable of inducing immunogenic cell death and urge to solvecurrently present problems with this paradigm. We anticipate that this emerging amalgamation of DAMPs,immunogenic cell death and anticancer therapeutics may be the key towards squelching cancer-relatedmortalities, in near future.
Resistance to cell death and the ability to prevaricate immuno-logical surveillance are two of themostmalicious strategies within thedefence arsenal of tumour cells [8,9]. Thus, over a long period of time,the strategies that have been anticipated to be capable of inflictingmaximal damage upon the cancer/tumour cells have consisted ofeither increasing the cancer cell's susceptibility towards death orincreasing/refurbishing of the immunological recognition of poorlyimmunogenic cancer cells [10]. A further more attractive ploy thatholds the highest therapeutic value is the one that combines both ofthese above as it would not only ensure complete or extensiveobliteration of cancerous cells but also provide a potentially firmbarrier to cancer's recurrence habits. Paradigms based upon this ployhave the potential to become practically applicable if therapeuticstrategies capable of inducing immunogenic cell death are explored.
In recent times, the concept of immunogenic cell death has beenaddressed by many extensive studies [2,12–14] and the paradigmsbased upon this concept have come closer to applicability in real-timethan it was previously anticipated. These latest trends have not onlychanged ‘classical’ immunological profiles of certain cell deathpathways but have also revealed that a subset of DAMPs areresponsible for immunogenicity of these cell death modalities.These recent developments have made DAMPs an important set ofmolecules to focus upon as far as immunogenic cancer cell death isconcerned.
Box 1. Cells of the innate and adaptive immune system
Development of an effective antitumour immune responsedepends on coordinated interactions between various populationsof innate and adaptive immune cells [1] briefly described below.
1. Innate immunity mechanisms involved in tumourelimination are orchestrated by a vast array of immunecells, of which, natural killer (NK) cells, NKT cells, γδ-Tcells, macrophages, granulocytes and dendritic cells(DCs) seem to be the most important ones.
1.1. Antigen-presenting cells (APCs)—APCs like DCs, macro-phages and B cells play a crucial role in antigenpresentation [7]. Their maturation after encounter ofvarious “danger signals” initiates a series of processesleading to activation of antigen-specific T cell response.
1.2. NK cells—Activation of these cells is usually triggered bydisappearance of class I MHC molecules from the surfaceof tumour cells or by exposure to antigens such as MHCclass-I related molecules, MIC A and MIC B, which areNKG2D ligands. Their expression is triggered by DNAdamage in tumour cells [16].
1.2.1. NKTcells—These cells express an invariant T cell receptoralpha chain that recognizes glycolipid antigens (e.g.gangliosides) presented by CD1d molecules on thesurface of tumour cells following which, NKT cells exerttheir antitumour effects primarily through secretion ofinterferon γ and direct cytotoxicity [16–18].
1.3. γδ-T cells—These cells are considered to be the mostimportant early source of IFN-γ and they mainly recognizeheat-shock proteins (HSPs), MIC A and MIC B orphosphoantigens and kill tumour cells through directcytotoxicity [16,19].
1.4. Macrophages and neutrophils—These are the mostimportant subsets of phagocytes and might be activateddirectly by stress products of cancer cells through TLR(Toll-like receptor) signalling. Here, secretion of pro-inflammatory cytokines and extensive production ofreactive oxygen as well as nitrogen species, mightserve in the protective mechanisms.
2. Adaptive antitumour effector mechanisms include theactivity of several populations of T cells and antibodyproduction by B cells. Here, antitumour antibodies executecancer cell death either through direct interplay with vitalintracellular signalling pathways or through induction ofcomplement-dependent cytotoxicity (CDC) and antibody-dependent cellular cytotoxicity (ADCC). The latter (i.e.ADCC) involves cytotoxic activity of NK cells, macro-phages, neutrophils and eosinophils, triggered by activa-tion of receptors for the constant region of the antibody(FcR).
2.1. Cytotoxic T cells (CTLs) or CD8+ T cells—CTLs are verypotent professional killers; a single activated CTL caneliminate hundreds of target tumour cells. Once activatedby mature DCs-based antigen presentation, they recog-nize and eliminate tumour cells bearing the particularantigen via secretion of monomeric perforin (whenpolymerizes, forms tiny holes in the plasma membraneallowing free water and ions flow), granzyme B (a serineprotease that activates both intrinsic and extrinsic path-ways of apoptosis induction) and activation of membranedeath receptors (Fas/CD95, TNF-RI, DR4 and DR5 viaFasL, TNF and TRAIL, respectively).
2.2. Helper T cells or CD4+ T cells—They are the majorconductors and orchestrators of the adaptive immuneresponse. Through cytokine secretion they stimulateproliferation and enhance activation of CD8+ T cells andcontribute to DCs maturation. Moreover, CD4+ cells arealso indispensable for B cell activation and antibodies classswitching [9,19].
In the present review, the ‘classical’ as well as the ‘emerging’immunological silhouette of necrosis, immunogenic apoptosis and‘autophagic’ cell death will be discussed along with specific attentionpaid to certain particular DAMPs (confirmed or proposed to be)associated with them. Also, the various resistance tactics employed bycancer cells along with a discussion on envisaged trends in thedevelopment of anticancer therapy based on ‘immunogenic cancercell death’ concept, will be discussed.
1.1. Cell death pathways: ‘classical’ and ‘new’ immunological profiles
In ‘classical’ terms, the immunological profile of at least apoptoticand necrotic cell death mechanisms is considered to be largely‘straight-forward’.
55A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2010) 53–71
Apoptosis (i.e. type 1 cell death), which is a prominent‘physiological’ pathway of cell death is considered to be immunolog-ically ‘silent’ or even tolerogenic [10]. Apoptotic cells are usuallyrapidly recognized by phagocytic cells [25] owing to the release ofvarious ‘find me’ signals [27]. Some of the most prominent apoptotic‘findme’ signals are listed in Table 1. Once recognized by professional/non-professional phagocytes, the apoptotic cells are subsequentlycleaned up very silently via phagocytosis owing to surface exposureof various ‘eat me’ signals (e.g. phosphatidylserine (PS), opsonins,modified ICAM-3, modified low-density lipoproteins (LDL), throm-bospondin binding sites, phosphatidylethanolamine, phosphatidyli-nositol, lactoferrin, PTX3 binding sites and HRG1 binding sites [29–31]) and suppression of ‘don't eat me’ signals (like plasmamembrane exposed CD31 [32] and CD47/integrin-associated proteinmolecules [33]). This process of apoptotic cell clean-up is carried outwithout eliciting any major immunological response owing torelease of anti-inflammatory signals such as transforming growthfactor β1 (TGF-β1), prostaglandin (PG) E2 and platelet-activatingfactors [29]. TGF-β1, a central anti-inflammatory modulator duringclearance of the apoptotic cells [34], acts by inhibiting p38 MAPKphosphorylation and NF-κB activation, and subsequent cytokineproduction [35]. It has been well established under different in vitroconditions that macrophage interactions with apoptotic cellspotently attenuates the response to lipopolysaccharide (LPS) andsuppresses the secretion of pro-inflammatory mediators such astumour necrosis factor (TNF), IL-1 and IL-12 [36,37]. Moreover,apoptotic cells can protect mice from LPS-induced death by reducing
Table 1A list of prominent factors (or ‘find me’ signals) released by various cell death pathways.
Factors released Receptors Type of cell death Fu
Apoptotic Micro-blebsor Microvesicles
? Apoptosis Apr
Endothelial Monocyte-activatingPolypeptide II (EMAPII)
CXCR3? Apoptosis Agrpr
Covalent/Cross-linked dimer ofribosomal protein S19 (dRP S19)
CD88 Apoptosis Aan
Lysophosphatidylcholine (LPC) G2A Apoptosis Aas
Fragments of human tyrosyl tRNASynthetase (TyrRS)
? Apoptosis Typoacan
Thrombospondin 1 and itsHeparin-Binding Domain
αvβ3 integrin Apoptosis Awe
H202 ? Apoptosis ImAnnexin-I ? Apoptosis Sti
prHMGB1 TLR2, TLR4 Necrosis,
apoptosis,autophagy
Prmopr
S100/Calgranulin Protein FamilyMembers- S100A8, S100A9,S100A12/EN-RAGE
RAGE Necrosis Capr
Heat Shock Proteins- HSP70, HSP90,HSP60, HSP72 and GP96
CD91, TLR-2/4, SREC-1,FEEL-1/ CLEVER-1
Necrosis Ca
Hepatoma-derived GrowthFactor (HDGF)
? Necrosis Sh
Monosodium Urate (MSU) ? Necrosis Delea
ATP P2XR, P2YR Necrosis Carec
IL-6 IL-6R Necrosis PoRibonucleoproteins (snRNPs),mRNA and Genomic DNA
TLR-3 Necrosis Intor
IL-1α IL-1R Necrosis Po
Abbreviations: CD—Cluster of Differentiation; CXCR3—C-X-C Motif Receptor 3; FEEL-1/CLReceptor-1; G2A—G2 Accumulation; HMGB1—HighMobility Group Box 1; HSP—Heat Shock PAdvanced Glycation Endproducts; SREC-1—Scavenger Receptor Class F member 1; TLR—Tol
serum LPS levels [38]. Here, the authors demonstrated that LPS couldquickly bind to apoptotic cells and that LPS-coated apoptotic cellscan be recognized and cleared by macrophages. In addition to theabove-mentioned autocrine and paracrine effects of apoptotic cells, itis important to mention a recently emerging concept [39] thatapoptotic cells also exert their anti-inflammatory effects directly, bybinding to macrophages, independent of phagocytosis process orsoluble factors. This apparent immunological ‘tolerogenicity’ towardsapoptosis is considered to be a mechanism playing an important rolein host's protection [10].
On the other hand, necrosis (i.e. type 3 cell death), which is inmost cases a kind of ‘accidental’ cell death mechanism, can be highlyinflammatory, sometimes even to an extent of being harmful [40],owing to the sudden release of various intracellular factors [27], themost prominent of which are listed in Table 1. These componentstend to sensitise and attract various professional and non-profes-sional phagocytes which further secrete different pro-/terminal-inflammatory factors/cytokines, such as IL-8, macrophage-inflam-matory protein 2 and TNF [29]. Although necrosis has beenconsidered for a long time to be a kind of ‘accidental’ cell deathmechanism, recent studies indicate that certain forms of necrosisare molecularly regulated, and have the serine/threonine kinasereceptor-interacting protein 1 (RIP1) and reactive oxygen speciesat the core of their molecular machinery [20]. Thus, as apparent,there exists a large difference in the overall immunological pro-files of apoptosis (tolerogenic) as compared to necrosis (highlyinflammatory).
nction(s)/feature(s) References
‘find me’ signal, chemoattractant for monocytes and possesseso- and anti-inflammatory features.
[27,209,210]
‘find me’ signal, chemoattractant for polymorphonuclearanulocytes and monocytes and possesses pro-inflammatoryoperties.
[27,211,212]
‘find me’ signal, chemoattractant for monocytes/macrophagesd chemorepellent for neutrophils.
[27,213–215]
‘find me’ signal and chemoattractant for monocytes as wellmacrophages
[27,216]
rRS's N-terminal fragment acts as chemoattractant forlymorphonuclear leukocytes while the C-terminal fragmentts as chemoattractant for both polymorphonuclear leukocytesd mononuclear phagocytes. It is a ‘find me’ signal.
[27]
‘find me’ signal and putative chemoattractants for monocytes asll as neutrophils. Supposed to confer tolerogenicity.
[27,217]
portant role in cell death [218]mulation of engulfment and possesses anti-inflammatoryoperties.
[217]
ominent and vital necrotic ‘cytokine’. Capable of attractingnocytes and neutrophils as well as possesseso-inflammatory properties
[41,108,219,220]
pable of attracting monocytes and neutrophils. Possesseso-and anti-inflammatory properties.
[27,221]
pable of attracting monocytes and neutrophils. [27]
ares similarity with HMGB1 [222]
rived from uric acid. Capable of attracting neutrophils as well asding to DC maturation. Possesses pro-inflammatory properties.
[27,123]
pable of instigating microglia recruitment and DCruitment/maturation.
[27,43]
tent pro-inflammatory activity. [138]eracts with various receptors on innate immunity—chestrating cells. Possesses pro-inflammatory properties.
56 A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2009) 53–71
In contrast to apoptosis and necrosis though, there exists a bit ofdiscord regarding the immunogenic impact of cell death associatedwith macroautophagy (thereafter simply called autophagy) or‘autophagic’ cell death (i.e. type 2 cell death) [10], that is cell deathcaused by autophagy. Interestingly, over the period of last few years,the classical immunological profiles of various (cancer) cell deathmechanisms have been considerably rectified such that while thestatus of necrosis has remained largely unaltered a separate (sub)class of apoptotic cell death called ‘immunogenic’ apoptosis hasstarted taking shape [13]. Even ‘autophagic’ cell death has shownsigns of being capable of creating some immunological flutters[41].
Apart from the above-mentioned morphologically well definedcell death modalities, certain atypical cell death mechanisms likemitotic catastrophe, anoikis, excitotoxicity, Wallerian degeneration,paraptosis, pyroptosis, pyronecrosis and entosis have received someattention recently [42]. In terms of immunological impact, verylittle is known about these cell death pathways however theirmolecular nature raises the possibility that they might exhibitinflammatory potential. For example, pyroptosis has been shown tocause release of IL-1β (major fever-inducing cytokine) and IL-18.Although in vivo data on immunogenicity of pyroptosis is stillunavailable [43], these cytokines may trigger local and systemicinflammatory reactions [44,45]. Similarly, pyronecrosis has beenshown to be capable of releasing certain DAMPs (i.e. HMGB1, seeSections 2.3 and 3) and IL-1β [42,46]. However, since the true identityof these cell death modalities is still a matter of debate, theirimmunology remains largely a conundrum. Thus, further research isrequired to ascertain the identity of these pathways in both molecularas well as immunological terms.
Box 2. Four major cell death morphotypes
Type 1 cell death or apoptosis is morphologically characterizedby chromatin condensation, cleavage of chromosomal DNA intointernucleosomal fragments, cell shrinkage, membrane blebbingand formation of apoptotic bodies without plasma membranebreakdown. At the biochemical level, apoptosis entails theactivation of caspases, a highly conserved family of cysteine-dependent aspartate-specific proteases, of which 11 membershave been identified in humans [3].Type 2 cell death or autophagy is characterized by a massivevacuolization of the cytoplasm. Autophagic cytoplasmic degra-dation requires the formation of a double-membrane structurecalled the autophagosome, which sequesters cytoplasmic com-ponents as well as organelles and traffics them to the lyso-somes. Autophagosome–lysosome fusion results in thedegradation of the cytoplasmic components by the lysosomalhydrolases. In adult organisms, autophagy functions as a self-digestion pathway promoting cell survival in an adverse environ-ment and as a quality control mechanism by removing damagedorganelles, toxic metabolites or intracellular pathogens. How-ever, autophagy may also promote cell death through exces-sive self-digestion and degradation of essential cellularconstituents, or by instigating a still ill-defined caspase-inde-pendent cell death pathway [11].Type 3 cell death or necrosis is morphologically characterizedby vacuolization of the cytoplasm, swelling and breakdown ofthe plasma membrane resulting in the release of the intracellularcontent and pro-inflammatory molecules. The biochemistry ofnecrosis is characterizedmostly in negative terms by the absenceof caspase activation, cytochrome c release and DNA oligonu-cleosomal fragmentation. Although necrosis has long been des-cribed as a passive, unorganized way to die, recent evidencesuggests that necrotic cell death can be actively propagated aspart of a signal transduction pathway [20].
Secondary necrosis is a terminal process experienced by apop-totic cells when their clearance via professional/non-professionalphagocytes is absent (in vitro conditions) or fails (certain in vivoconditions). Secondary necrosis typically involves swelling ofcell/apoptotic bodies, mutilated cytoplasmic membrane, andself-hydrolytic enzymes' activation, finally followed by completecell disruption [26].
1.2. Defining immunogenicity of cell death pathways:damage-associated molecular patterns (DAMPs)
The immunogenicity of the cells dying via various cell deathpathways is usually considered to be mediated by a variety ofmolecules categorised as ‘damage-associated molecular patterns’(DAMPs) also known as alarmins [47]. DAMPs are intracellularmolecules normally hidden within live cells, which acquire immu-nostimulatory properties upon exposure or secretion by damaged/dying cells. Certain proteins categorised as DAMPs are also oftencalled leaderless secretory proteins (LSPs) since these are intracellularproteins lacking secretory leader peptide, having inflammatoryactivity and are secreted actively by non-classical molecular pathways[48,49]. These molecules have the ability to exert various effects onantigen-presenting cells (APCs), like dendritic cells (DCs), such asmaturation, activation and antigen processing/presentation. Bio-chemically speaking, it has been proposed that ‘hydrophobicity’ isprobably the most ‘ancient’ DAMP capable of activating the innateimmune system and hence most known DAMPs tend to contain manyhydrophobic regions such as the chaperones belonging to the heatshock proteins family [50].
Upon exposure or release from the dying cells DAMPs interact withmembrane-bound or vesicular pattern-recognition receptors (PRRs)such as Toll-like receptors (TLRs), the NOD-like receptors (NLRs) andRIG-I-like receptors (RLRs) [48]. In fact, receptors associated withDAMPs are often sharedwith those for pathogen-associatedmolecularpatterns (PAMPs) e.g. membrane-bound PRRs (TLR4 and TLR-2) andintracellular PRRs (TLR3, TLR7–9, NLRs and RLRs) [48,51]. A recentstudy has shown that cancer cells artificially made to express PAMPssuch as flagellin (from Listeria monocytogenes) or P40 protein (fromKlebsiella pneumoniae) may induce antitumour immunological re-sponse against themselves, thus providing strong evidence for the roleof PRRs-PAMPs/DAMPs interactions in regulating cancer cell'simmunogenicity [52].
Roughly speaking, all DAMPs might be sub-divided into three‘major’ sub-classes (on the basis of their stage and place of locali-zation/release) i.e. (1) DAMPs exposed on plasma membrane (e.g.calreticulin, HSP70 and HSP90), (2) DAMPs secreted extracellularly(e.g. HMGB1, uric acid, IL-1α and other pro-inflammatory cytokines)and (3) DAMPs produced as end-stage degradation products (e.g. ATP,DNA and RNA). Another (relatively less studied) ‘minor’ sub-class ofDAMPs includes those resulting from the extracellular matrix [53].These include compounds like hyaluronan, heparan sulphate anddegraded matrix constituents [53]. Thus, it might be considered thatthe actual diversity of DAMPsmaydependupon various factors such astype of cell death, cell-type as well as tissue injury [48]. However,while we have considerable knowledge regarding the diversity ofnecrosis-associated DAMPs and their immunological impact, we knowmuch less about the diversity of DAMPs associated with ‘immuno-genic’ apoptosis and as far as ‘autophagic’ cell death is concerned, weare only starting to scratch the surface.
2. Immunogenic apoptosis: an emerging concept incancer therapy
Immunogenic apoptosis is a relatively new concept whichemerged due to a series of important studies published back-to-
57A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2010) 53–71
back. Compared to the ‘classical’ apoptosis, immunogenic apo-ptosis has been found to have immune system eliciting pro-perties and capabilities of inducing DC-based anticancer vaccineeffect.
First instance of induction of effective antitumour immunity inresponse to chemotherapy (in vitro and in vivo) was observed intumour-bearing immunocompetent mice treated with anthracycline,accompanied by tumour regression [54]. Soon it was found that apartfrom anthracyclines, various other agents were capable of inducingimmunogenic cancer cell Death In another of these studies it wasfound that bortezomib-killed tumour cells were capable of activatingDCs and subsequently inducing a tumour specific T cell response, invitro [55, 56]. Further investigation showed that this antitumourresponse was dependent on cell-to-cell contact between dyingcancer cells and the DCs and was mediated by cell surface exposedHSP90 proteins since geldanamycin (an HSP90 inhibitor) blocked it[56].
Subsequently, the concept of immunogenic cell death wasfurthered by an array of studies showing that cancer cellsundergoing in vitro apoptosis induced by certain agents (e.g.anthracyclines, oxaliplatin and ionizing irradiation) were capableof mediating an ‘anticancer vaccine effect’, in absence of anyadjuvants or immunostimulatory substances, once subcutaneouslyimplanted into immunocompetent mice [2,13]. Thus the mostattractive property of immunogenic apoptosis is that, when inducedin the cancer cells it combines the ‘physiological’ cell deathmechanism with potent anticancer immunity [13]. Further research
Fig. 1. Ecto-CRT/ERp57 translocation pathway during immunogenic apoptosis. Cancer cellssignalling pathway which might be split into two major arms i.e. apoptotic arm (red arrows)activated via ER stress induced by the above-mentioned agents. Thereafter, the activated PEleaflet of the plasma membrane via SNARE-dependent exocytosis, transiting through Golgiapoptotically activated caspase 8, BAP31, Bax/Bak and ER-Ca2+ leakage, which leads to eximportant to note here that the translocation arm and anthracycline-induced cell death areinhibits the CRT/ERp57 translocation yet the cell death persists.
showed that the ‘eat me’ signal critical for execution of ‘immuno-genic’ apoptosis by these agents was calreticulin (CRT), present onthe surface of these apoptotic cells (see Fig. 1) [12,13]. Theserevelations helped in adding CRT to the relatively short list ofapoptotic DAMPs primarily consisting of the HSPs.
2.1. Calreticulin: a critical DAMP and ‘eat me’ signal forimmunogenic apoptosis
Calreticulin (CRT) is a highly conserved, 46 kDa Ca2+-bindingprotein, mainly located in the lumen of endoplasmic reticulum (ER)and serving various versatile functions like chaperone activity andregulation of Ca2+ homeostasis/signalling [57]. Apart from the ER,where it remains sequestered through its KDEL sequence, CRT hasalso been reported to be residing in the nucleus (specially, nuclearenvelope lumen) and the cytoplasm [57,58]. While CRT's cytoplasmicfunction is as of now unclear, nuclear CRT has been suggested toregulate nuclear protein transport (both import and export of pro-teins like NFAT3, MEF2C and glucocorticoid Receptor) and therebyinfluence signalling via nuclear steroid receptors and integrins[10,57]. Within the ER, CRT interacts with various other moleculeslike ERp57 and calnexin (CNX) in order to help in proper folding ofproteins [59]. Further, CRT also plays an important role in assemblyof the MHC class I molecule as well as in loading of the antigenpeptides onto the MHC-I molecule within the ER [60], therebyplaying an important role in proper functioning of the immunesystem.
stressed by anthracyclines, UVC light, oxaliplatin and γ-irradiation tend to activate aand translocation arm (blue arrows). Both these arms originate around PERK, which isRK phosphorylates eIF2α, which then assists in CRT/ERp57 translocation to the outernetwork. In parallel, the apoptotic arm is orchestrated via co-ordinated action of pre-ecution of apoptosis as well as surface exposure of legitimate apoptotic signals. It isindependent of each other such that while the depletion of PERK, caspase 8 or SNARE
58 A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2009) 53–71
Apart from the ER, nucleus and cytoplasm, a fraction of CRTwas also found on the plasma membrane of the viable cells (ecto-CRT) [33]. Newly synthesized CRT was preferentially transportedto the cell surface through a still elusive mechanism, whereinit possessed a half-life of around 12 h [61]. Once on the surfaceof the viable cells, ecto-CRT has been proposed to play variousroles such as regulation of focal adhesion, various cell-to-extra-cellular matrix (ECM) interactions and cell migration/locomotion[61].
Interestingly Gardai et al. have shown that cancer cells inducedto die apoptotically by UV, expose larger amounts of ecto-CRT,which is redistributed in the form of ‘patches’ [33] and pre-dominantly co-localized with PS [33]. Here, ecto-CRT was supposedto act as an ‘eat me’ signal for professional phagocytes, e.g.macrophages, so that they can engulf the apoptotic cells [33]. Interms of plasma membrane structuring, both PS and ecto-CRT co-localized with cholesterol-rich, GM-1 ganglioside-containing “rafts”[33]. However, apart from these differences, what made ecto-CRTexposure on (UV-induced) apoptotic cells more receptive toprofessional phagocytes compared to living cells was the apparent“suppression” of the ‘don't eat me’ signal—CD47, which inhibitsengulfment of living cells by interacting with the SIRPα (SHPS-1)protein on the surface of the professional phagocytes [62]. Thus,along with increased amounts, patched distribution and co-localization with PS, the suppression of the ‘don't eat me’ signalis crucial for ecto-CRT to serve as an ‘eat me’ signal [33]. In fact,when CD47 was absent in the viable (erythroid) cells (having ecto-CRT on their surface), they became receptive to phagocytosis[33,63]. Further, it has been found that once ecto-CRT is exposed onthe apoptotic cell surface, it interacts with various proteins such asthrombospondin [64], C1q, mannose binding lectin (MBL) [65] andmore importantly the internalization receptor, CD91 (also calledLDL-receptor-related protein or LRP) on the professional phago-cytes, which stimulates Rac-1 and drives engulfment of apoptoticcells [33].
Recently, a series of studies showed that ecto-CRT exposure,rather than being a general phenomenon associated with apoptosis,occurs in response to specific stressors/drugs capable of inducingapoptosis [12]. While screening various apoptosis-inducing che-motherapeutic agents or cellular stressors, Obeid and co-workersdelineated a subset of apoptosis inducers which were endowedwith the ability of inducing effective ecto-CRT exposure, such asanthracyclines, oxaliplatin, UVC and γ-radiation [2,10,13]. Apopto-tic cancer cells upon exposure of ecto-CRT induced by thesecytotoxic drugs became receptive for engulfment by DCs [13] andwhen implanted into immunocompetent mice, they were able toconvert the ‘classical’ silent nature of apoptosis into an ‘immuno-genically active’ form, fittingly termed as ‘immunogenic apoptosis’[2,12,13]. However, it must be noted that for ‘immunogenic’apoptosis to be induced, ecto-CRT was required to be presentalong with legitimate apoptotic signals, i.e. sole presence of ecto-CRT on any cancer cell was not sufficient to make those cells afavourite target for professional APCs and/or phagocytes [10,12].Though CRT has been observed previously to exhibit the ability tointeract with tumour-specific immunogenic peptides [66,67], its‘anticancer vaccine effect’ has been suggested to reside in thecapability of ecto-CRT to incite the uptake of tumour cells by DCs,rather than in its association with these antigenic peptides on thesurface of tumour cells [10,12].
Ecto-CRT exposure was found to be a pre-apoptotic event,preceding various distinct morphological signs of apoptosis such asPS exposure on the outer leaflet of the plasma membrane as well asmitochondrial depolarization and was always accompanied by co-translocation of ERp57 [2]. Recent reports into the molecularmechanism dictating this process, proposed that ER-stress, andmore specifically the PERK/eIF2α arm of the unfolded protein
response (UPR) pathway [68], plays an important role in ecto-CRT/ERp57 translocation (see Fig. 1) [2].
Recent reports have documented the molecular mechanismsgoverning ecto-CRT/ERp57 exposure in dying cells (see Fig. 1).These consisted of ERCa2+-leakage, accompanied by phosphory-lation of the ER-transmembranal kinase PERK (on threonine 980)that leads to phosphorylation of eIF2α (on serine 51), followed bya pre-apoptotic and partial activation of caspase-8 [2]. Thecleavage of the caspase-8 substrate BAP31, an ER-sessile protein,alongwith conformational activation of Bax and Bak proteins, wasrequired and accompanied by movement of CRT/ERp57 throughGolgi bodies. Later CRT/ERp57 translocation onto the plasmamembrane occurred via SNARE-dependent exocytosis possiblyinvolving both vesicle-associated SNAREs (like VAMP1) andmembrane-associated SNAREs (like SNAP23/25) (see Fig. 1) [2].Absence of ER-Ca2+ leakage [12], PERK, BAP31, Bax, Bak,SNARE or presence of mutated eIF2α (rendering it non-phosphorylatable) abolished ecto-CRT/ERp57 translocation andexposure [2]. This pre-apoptotic ecto-CRT translocation path-way has been found to be evolutionarily conserved and sharedbetween yeast and mammals such that under anthracycline-stress the yeast apoptotic cells were found to expose CNE-1 (ahomologue of mammalian CRT) onto their surface [22].
From the above discussion it is clear that ecto-CRT is not only animportant ‘eat me’ signal but also an important anticancer immunitymediator capable of supplying the host immune system with thetumour-associated antigen's cargo. However, it is worth mentioninghere that this is not the first instance of CRT's immunologicalimpact. In fact, CRT's importance in regulation of various immuno-logical functions has been known for nearly past 15 years [69].Immunological exploits of CRT first came to attention when itwas shown that CRT might be a part of cluster of intracellularautoantigens called Ro/SS-A complex recognized by autoanti-bodies typically found in sera of patients suffering from auto-immune diseases like primary Sjogren's syndrome and systemiclupus erythematosus (SLE) [69,70]. Since these autoantibodies weregenerated against a cluster of ‘strictly’ intracellular proteins like CRTit was suggested that these intracellular proteins might be releasedduring necrotic cell death [69]. CRT was later found to be not aspecific member of the Ro/SS-A complex [71] but an autonomousautoantigen [69].
What is further more intriguing is that ecto-CRT has beenobserved not only on the side of the (chemotherapeuticallyinduced) apoptotic cancer cells but also on the side of the immunesystem cells. For instance, ecto-CRT can be found on the surface ofhuman monocyte-derived macrophages wherein it acts as areceptor for C1q so as to opsonise apoptotic cell in order to carryout phagocytosis [72]. Similarly, it has been observed that ecto-CRTmight also be present on normal DC surface and can interact withtumour-associated antigens like NY-ESO-1, thereby playing animportant role in interfacing of tumours and host innate immunesystem [67]. Also, ecto-CRT might be found on the surface ofresting as well as activated T-cells, in association with β2-free MHCclass I molecules [73]. Whether the presence of ecto-CRT on bothsides of the tumour–immune cells interaction has something to dowith its ‘anticancer vaccine effect’ mediating properties is stillunclear but further research into this aspect might shed more lighton tumour–immune system interfacing and interactions. This couldalso potentially open up some exploitable lacunae against cancercells.
59A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2010) 53–71
2.2. Heat-shock proteins (HSPs): a group of seasoned apoptotic DAMPs
HSPs are a family of highly conserved chaperone proteins that playan important role in folding of newly synthesized proteins as well asrefolding of proteins affected due to various stress conditions [74].Mammalian HSPs have been mainly divided into two groups (basedon their sizes) i.e. high molecular weight (HMW) HSPs and smallmolecular weight (SMW) HSPs [74], all of which are most stronglyexpressed under stresses. HMW-HSPs are ATP-dependent chaperonesrequiring co-chaperones for their activity and consist of three majorfamilies—HSP90, HSP70 and HSP60 [74]. On the other hand, SMW-HSPs are ATP-independent chaperones consisting of members likeHSP27 [75]. Various kinds of cellular stresses, e.g. oxidative stress,irradiation, serum deprivation and chemotherapeutic drugs, mightlead to transcriptional and translational activation of HSPs [74]. Whileunder stress most of the HSPs are expressed within the cellularcytoplasm or organelles (i.e. mitochondria and ER) yet, certaininducible HSPs, such as HSP70 and HSP90, can translocate to theplasmamembrane [10]. Based on the presently available literature wecan safely say that whereas intracellular expression of HSPs has canceraugmenting or pro-survival properties, HSPs surface exposure haspossibly tumour suppressing properties due to their ability to attractthe attention of the innate immune system towards the cancerous/tumour cells [10].
Intracellularly overexpressed HSP70, HSP90, HSP60 or HSP27 havebeen shown to exhibit apoptosis-inhibitory and cytoprotectiveactivity whereas their depletion increases the cell's sensitivity toseveral stress signals utilizing mitochondria as central executioners ofapoptosis [74]. For detailed account on anti-apoptotic activity of theHSPs one might refer to the recent review by Lanneau et al. [74]. Notsurprisingly thus, tumours have been shown to overexpress HSPs,probably due to ‘stressful’ conditions that might exist in the tumourmicroenvironment [76].
When we shift their position from intracellular to plasmamembrane surface, HSPs can exhibit potent immunostimulatoryactivity as observed in surface HSP enriched tumour cells (inducedby hyperthermia) [55]. More specifically, HSP70 and HSP90 exhibitimmunostimulatory activity when exposed on the surface of stressedor dying cells [10]. HSP70 has been found on the plasma membranein tight association with PS [77], along with trafficking factors, duringanalysis of lipid rafts [78,79]. In fact the association between HSP70and PS has been found to accelerate apoptosis [77]. Apart from this, asphingolipid Gb3 has also been implicated in HSP70's tumour cellmembrane localization [80]. It has been postulated that, presence ofHSPs on outer leaflet of plasma membrane is probably attributableto their functions in transportation and primary chaperoning ofproteins [10].
Ecto-HSP70 and HSP90 (the latter to a lesser extent) act as DAMPsand determine the immunogenicity of stressed/dying cells [81]. Thisis due to their ability to interact with a number of APC surface-receptors [82] like CD91, LOX1 and CD40 [55] and to facilitate cross-presentation of antigens derived from tumour cells (i.e. tumour-associated antigens or TAAs), on MHC class I molecule, capable ofevoking (CD8+) T-cell response [83–85]. Such a HSP-TAA complexalong with MHC class I molecule can activate antigen processing andpresentation in DCs via a complex of TLR4 receptors and CD14 [86,87].TLR4-based activation can also lead to cytokine-mediated hostimmune system activation via activation of NF-κB signalling pathwayin DCs, accompanied by release of pro-inflammatory cytokines likeTNF, ILs (types-1β, -12, -6) and GM-CSF (granulocyte macrophage-colony stimulating factor) [86–89]. Further to this, HSP70 can alsoupregulate CD86 and CD40 thereby promoting DC maturation [90].Intriguingly, HSP proteins have also been found to be capable ofinteracting with the natural killer (NK) cells and play important rolein their activation [10]. NK cells are specialized immune cells whoseactivity is dependent upon a fine balance between activating and
inhibitory signals wherein the normal/healthy cells inhibit NK cellsvia higher amounts of surface MHC class I molecules and less or noamounts of activating NK ligands [91]. On the other hand, the harmfulcells like cancer cells lead to NK cell activation due to increasedamounts of activating surface NK ligands and reduced amounts ofMHC class I molecules [91]. Since the activity of the NK cells isregulated by various inhibitory/activating receptors, presence of HSPson the surface of stressed/dying cells leads to prevention of HLA-Epeptide complex recognition by CD94/NKG2A (inhibitory receptorcomplex on NK cells) thereby leading to destruction of the cell(s) bythe NK cells [92]. Here, surface HSP70 on cancer cells can interactwith CD94 receptor and activate the NK cells, via its C-terminaldomain [93,94].
As discussed previously, bortezomib-mediated surface expressionof HSP90 on the plasma membrane of myeloma cells was reported toinduce immunogenic cancer cell death by delivering activation signalsto the DCs [56,82]. This finding of HSPs being potential immunogenicsignals is supported by earlier studies [95]. For example, increasedimmunogenicity of the tumour cells exposing high amounts of surfaceHSPs due to heat shock treatment has also been observed in variousanimal models as well as human models, in vitro [96,97]. Similarly ithas been observed that apoptotic leukaemia cells might be induced tobehave in an immunogenic manner in vivo, by heat-stressing themprior to apoptosis induction [98]. All together these studies supportthe concept that surface exposure of HSPs on tumour cells might be acrucial event for initiating effective anticancer immune response bothon the level of APCs as well as NK cells (in certain cases), by increasingthe immunogenicity of tumour cells. This makes HSPs one of the mostvital groups of DAMPs due to their relative diversity and a well-documented surface-association during stress conditions.
However, the mechanisms behind surface translocation of HSPsremain elusive. Moreover, it is not known whether the surface heatshock protein fraction is derived from the ER, cytosol or both [55]. Thebest available premise might be that since bortezomib has the abilityto induce ER stress [99,100], we might use ER stress signallingpathways as the starting step for investigation of surface HSPexposure. Therapeutically speaking, it has also been observed thatvaccination with various HSPs/chaperones (enriched from tumourlysate) has the ability to induce a specific antitumour immuneresponse [98]. In fact it has been found that the DC-stimulatoryproperties of tumour cell derived-lysate rich in chaperones are farmore superior to individually purified HSPs [98]. Similarly it has beenobserved that stressed tissue-derived HSPs were more immunosti-mulatory than recombinant HSPs [101]. These observations haveessentially led to a theory that HSPs might actually be carriers ofDAMPs and that it is the complex of HSP-DAMP that might beimmunostimulatory rather than the HSPs alone [101]. This theoryapparently explains why there exists a large gap between theimmunogenicity of inducible HSPs (isolated from stressed cells)and constitutively expressed HSPs (isolated from un-stressed cells)[101–103]. Clearly, further research is required to be done in order toascertain the exact immune system activation properties of HSPs.
2.3. HMGB1: a late apoptotic DAMP?
The high mobility group box 1 protein (HMGB1), (also calledamphoterin) is a very abundant nucleus-localizing non-histonechromatin-binding protein that affects various nuclear functions liketranscription as well as assembly of nucleoprotein complexes andmight be secreted actively (from inflammatory cells) or passively(from necrotic cells, in soluble form) [10,104]. There exist three typesof HMGB proteins composed of two basic DNA-binding domains (i.e.HMG boxes A and B) and a COOH-terminal tail [104]. All three HMGBproteins havemore than 80% amino acid identity amongst themselveshowever, whereas HMGB1 is ubiquitously expressed [105], HMGB2and HMGB3 are expressed mostly during embryogenesis and
60 A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2009) 53–71
restrictively in adult-stage [106,107]. While the above-mentionedfeatures are typical for HMGB1 protein within viable cells, itsfunctionalities (and its effects) change significantly when it is presentextracellularly. Though HMGB1 might directly be secreted by (IL-1β,TNF or LPS) activated macrophages and monocytes, more impor-tantly, dying cells (usually necrotic, as discussed in Section 3) tend torelease HMGB1 protein which then acts as a ‘cytokine’ and exhibitspro-inflammatory properties [104]. Thus its release is deemed aspotential immune-response inducer against dying (necrotic) tumourcells. It is worthmentioning here that, the HMGB1 actively secreted bycertain immune cells is molecularly different from the one passivelyreleased by the necrotic cells since the active secretion requiresHMGB1 to be acetylated on several specific lysine residues [104].Though it has been long known that HMGB1 is released as a DAMP bynecrotic cells, it is now increasingly coming to light that other celldeath pathways like apoptosis [43] and ‘autophagic’ cell death [41]might also have something to dowith HMGB1 release at some point intheir respective execution phases.
Association of HMGB1 with apoptosis has been a matter of debate.As discussed below, HMGB1 is an indispensable necrotic DAMPhowever recently it has been reported that HMGB1 might beassociated with apoptosis, specifically during secondary necrosis[43,108,109]. This is because, nuclear DNA has been found to bereleased during apoptosis in a time-dependent fashion while on theother hand HMGB1–DNA binding has been found to tighten duringapoptotic process [10]. These observations led to a concept thatapoptotic cells might release DNA as well as HMGB1 during its laterstages of unfolding. In line with this hypothesis, recent studies haverevealed that apoptotic tumour cells release HMGB1 at least duringtheir later stages (e.g. during secondary necrosis), through a processthat was found to be blocked by z-VAD-fmk, a pan-caspase inhibitor
Fig. 2. Immunology of necrotic cell Death. Necrotic cells release intracellular organelles as wHDGF, SAP130 and S100 protein family members. Moreover, necrotic cells secrete IL-1α asignals attract the attention of phagocytic cells and interact with their surface receptorsinflammatory signals like IL-8 and TNF-α, depending upon the situation and receptor–ligand
and a secondary necrosis delayer [110,111]. Once released outside,HMGB1 can bind to various receptors like, TLR2/TLR4 and thereceptor for advanced glycosylation end products (RAGE), therebyinstigating inflammatory response [112,113].
3. Necrosis: cell death pathway with a diverseimmunological profile
Necrosis is well-known to have the capabilities of inducing aconsiderable inflammatory response. Due to its pro-inflammatorynature, cells undergoing necrosis exhibit the knack to attractattention of various immune cells, such as macrophages and otherphagocytes. These phagocytes usually take up necrotic cells viamacropinocytosis [114] and their uptake is delayed as well as lessefficient than uptake of apoptotic cells; and it occurs only after lossof membrane integrity [115]. This indicates that necrotic cells releasetheir intracellular contents (e.g. DAMPs) before they are cleared byphagocytes. One such DAMP is HMGB1 (see Fig. 2). Although it wasinitially thought that HMGB1 is released from the nucleus of primarynecrotic cells only, recent studies demonstrated that it could bereleased during secondary apoptosis as well [108] (as discussed inSections 2.3 and 6.4). Extracellular HMGB1 activates macrophages aswell as DCs [116,117] and promotes neutrophil recruitment [118].HMGB1 has been shown to bind several PRRs (e.g. TLR2, TLR4 andRAGE) [43]. However, since possible contamination of DAMPs withTLRs ligands is always possible, it is important to take great carewhen interpreting the role of TLRs in these studies [43].
Here, TLR4 controls antigen presentation by inhibiting the fusionof phagosomes with lysosomes [10] such that the tumour cell materialwithin the phagosome is then diverted to the ER for forming anti-genic peptides rather than being degraded by the lysosomes. The
ell as intracellular factors like uric acid/MSU, HSP70, HSP90, HMGB1, ATP, DNA, RNA,nd IL-6 in a passive and active manner, respectively. These pro-inflammatory factors//co-receptors. Following this, the phagocytes either remain neutral or secrete pro-interaction. The latter instigates an inflammatory response of considerable magnitude.
61A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2010) 53–71
significance of extracellularly released HMGB1′s binding to TLR4 isfurther elucidated by the fact that, absence of either of thesemolecules (e.g. due to antibody-based blockage or RNAi-basedknock-down), dampens the immune response initiation, both invitro as well as in vivo [10]. Under these conditions while the DC-based engulfment of tumour cargo remains intact, yet the phago-somes containing this tumour cargo fuse with lysosomes, therebypreventing assemblage of proper tumour-antigen peptides withantigen-presentation molecules like MHC (reversed by lysosomalinhibitors like chloroquine or bafilomycin A1) [10]. This observationhas also been ascertained in humans, wherein, patients bearing adefective TLR4 allele have experienced increased incidence oftumour metastasis and reduced capacity of DCs in presentingantigens from dying tumour cells [108]. Furthermore and apartfrom interaction with TLR2/TLR4, HMGB1 complexed with self-DNAexhibits an ability to interact with TLR9 and stimulate interferon-αproduction [119,120].
Necrotic cells also release heat shock proteins (gp96, HSP70,HSP90) [121,122] (see Fig. 2) [123], which have a pro-inflammatoryeffect through TLR2 and TLR4 [122,124]. In addition, researchers havedetected an increased level of uric acid (the end product of purinemetabolism in uricotelic mammals) during cell injury, probably due tothe augmented degradation of cellular RNA and DNA and subsequentmetabolization of the liberated purines [123]. Moreover, oncereleased uric acid enhances CD8+ T cell responses [125] as well asCD4+ T cell responses towards particulate antigens [126]. Similarly,when added to dying tumour cells or with whole protein antigen in atumour vaccination protocol, uric acid tends to increase antibodyimmunity and improved inhibition of tumour growth [127]. Uric acidhas been known for a long time as a pro-inflammatorymolecule whencrystallized in the joints in gout, and lately it was shown to activatethe inflammasome, which is involved in the maturation of theinflammatory cytokine IL-1β [128]. It is quite likely that upon loss ofmembrane integrity many other intracellular substances that arenormally confined within cells are released. It has been shown thatRNA is released from injured or necrotic neutrophils in vivo and that itcontributes to the pathogenesis of polymicrobial septic peritonitis andischemic gut injury in a TLR3-dependent manner [129]. Moreovergenomic DNA (double-stranded), nucleotides (ATP) and nucleosides(adenosine) can be released from necrotic cells exerting immunos-timulatory effects on macrophages and dendritic cells [130–134.Recently, another nuclear protein, a spliceosome-associated protein130 (SAP130), was identified as another DAMP molecule releasedfrom necrotic cells and specifically binding to macrophage-inducibleC-type lectin (MINCLE) on macrophages to induce inflammatoryresponse [135].
It has been shown that necrotic cells can also induce aninflammatory response by active or passive secretion of inflamma-tory cytokines (see Fig. 2). For example, necrotic cells injectedintraperitoneally induce an inflammatory response and neutrophilicinflammation [136]. This response requires the signalling proteinmyeloid differentiation primary response gene 88 (MyD88) andIL1R on non-bone-marrow-derived cells [136]. Eigenbrod et al.further explored this issue and found that IL-1α is passivelyreleased from necrotic cells and that it induces secretion of CXCL1by mesothelial cells of the peritoneum [137]. The authors showedthat this leads to neutrophilic inflammation in a MyD88 and IL1Rdependent manner. Moreover, it was found that necrotic cellsactively release the pro-inflammatory cytokine IL-6 due to upregu-lation of NF-κB and p38MAPK [138]. These observations are inagreement with the notion that necrotic cells retain their ability tosynthesize proteins until the very last moment of plasma membranepermeabilization [139].
All the above-mentioned data indicate that necrotic cells couldrelease an array of immune-modulatory factors or DAMPs, whichinteract with immune system and modulate its responses. Obviously,
information on the DAMPs released from necrotic cells is still limited,but many more interesting and challenging findings are expected.That knowledge will help us to understand the roles of DAMPs inmaintenance of homeostasis or induction of inflammation.
4. Immunogenic impact of ‘autophagic’ cell death: a moot case?
Autophagy is foremost a survival mechanism that is stimulated incells subjected to nutrient or growth factor deprivation.When cellularstress continues, cell death may ensue through aberrant autophagywhich might become associated with features of apoptotic or necroticcell death, depending on the stimulus and cell type [11]. While‘autophagic’ cell death has been a subject of intense attention inrecent times [11] the role of autophagy in immunology has also comeunder focus. There are a number of investigations which haverevealed that within APCs autophagy might play an important rolein MHC class II presentation of certain peptides [140,141]. Howeverthe role of autophagy in MHC class I-based antigen presentationremains speculative at best [142] and has been suggested by only onestudy [143]. Similarly, autophagy has also been shown to play animportant role within macrophages in TLR-mediated recognition andactivation of innate immunity [140]. In case of APCs, researchers havefound that autophagosome(s), an important autophagy-organelleexhibits the ability to deliver peptides for loading onto MHC class IImolecules for presenting to CD4+ T cells [141] but, the validity for thisbeing a major pathway in vivo, still requires further investigation.Moreover, autophagosomes have been proposed to be capable offerrying various tumour antigens including HSP-bound peptides anddefective ribosomal products, which has led to the idea thatautophagosomes-based vaccines might be used against tumours[140]. Thus, based upon the accumulating evidence, it might be safeto propose that autophagy could have vital role to play in case of bothinnate and adaptive immunity. Interestingly, it has been shown thatautophagy might also be involved in antigen cross-presentationwithin the antigen donor cells [144] and again, autophagosomesmight be playing the role of antigen carriers. These proposed roles ofautophagy in antigen presentation suggest that ‘autophagic’ cell deathplays a possible role in modifying the surface proteome of dyingtumour cells or causing exposure/release of DAMPs, a premise thathas recently been vindicated by various studies. It has been knownfor some time that autophagy is capable of modifying the surfaceproteome of cancer cells in vitro [145] however, it was only recentlythat ‘autophagic’ cell death or cell death associated/accompanied byautophagy (in most cases, macroautophagy) was discovered to havethe ability to increase the immunogenicity of the tumour cells (seeFig. 3). Here, cancer cells dying by autophagy or by anothermechanism associated with autophagy following tamoxifen treat-ment were engulfed by professional and non-professional phago-cytes through a process that can be inhibited by 3-MA (see Fig. 3),thereby implying that autophagy somehow modifies the surface ofthe dying cells predisposing them to phagocytic engulfment [146].Here, PS exposure was shown to mediate the recognition andengulfment of these cells [146]. Moreover, it was shown thatautophagy might in fact be essential for exposing the ‘eat-me’ signals(e.g. PS and possibly CRT) on apoptotic cells, which indicates thatautophagosome formation during apoptosis might be one of themechanisms used for exposing phagocytosis signals on the surface ofdying cells [147,148]. Even release of ‘find me’ signals such as LPC hasbeen attributed to autophagy [148]. Further it has been shown thatphagocytic uptake of these dying autophagic cells by macrophagesmight lead to a pro-inflammatory response characterized by theproduction of IL-6, TNF and IL-8 [149]. Furthermore in epidermalgrowth factor receptor-targeted diphtheria toxin (DT-EGF) induceddying cancer cells, selective release of HMGB1 protein occurredwithout showing signs of membrane lysis or classical necrosis andrequired autophagosomes (a fraction of them) formation (see Fig. 3),
Fig. 3. Immunology of ‘autophagic’ cell Death. Two independent studies have recently shed light on the ability of ‘autophagic’ cell death to create immunological flutters. In the first ofthese studies it was shown that cancer cells undergoing cell death associated with/accompanied by autophagy, induced by tamoxifen (red arrows), exposed ‘eat me’ signals likephosphatidylserine (PS) which attracted the attention of macrophages. Subsequently, it was shown that autophagy activated in dying cancer cells via epidermal growth factorreceptor-targeted diphtheria toxin (DT-EGF) led to autophagosome-mediated HMGB1 release (black arrows). This HMGB1 could then interact with the surface receptors of variousphagocytes such as DCs, however this has not been confirmed yet.
62 A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2009) 53–71
thus suggesting another link between autophagy and immuneresponse to cell death [41]. However, the effects of this autophagy-mediated HMGB1 protein release on innate immunity cells have notbeen investigated yet.
The above discussion points towards the fact that ‘autophagic’ celldeath probably has the capability of ‘painting’ the dying tumour cells,immunogenic. This notion is further strengthened by a very recentstudy which revealed that immunization with cells undergoingautophagy was superior to apoptotic cells in facilitating the cross-priming of antigen-specific CD8+ T cells [150]. Strikingly, silencing ofAtg5 (autophagy-related protein 5) expression inhibited priming[150]. However, there is only limited knowledge available on thereleased immuno-modulatory factors from the cells dying inassociation with autophagy; hence, additional in depth studies arerequired.
5. Cancer's escape from immunosurvelliance
Accumulating evidence supports to some extent, a long-time heldconcept of cancer immune surveillance, which states that theimmune system is a watchdog that finds as well as eliminates thepotentially dangerous cells (including neoplastically transformedones) and tumour development is probably a result of immuneresponse failure. Indeed, the incidence of spontaneous, geneticallyengineered (by deletion of tumour suppressor genes) or chemicallyinduced tumours, is increased in immunocompromised mice. Also, inhumans with impaired immune system functionality, both virus-induced and some non-virus-induced tumours (e.g. melanomas) aremore frequent [19,151]. Complex interactions between tumour cellsand the immune system are nowadays described as cancer immuno-editing. It consists of three phases—referred to as three “Es”, namely
elimination, equilibrium and escape [152]. The elimination phase,which is mainly parallel to cancer immune surveillance, involvessuccessful recognition and elimination of tumour cells by the immunesystem.
Cancer cells that manage to survive elimination, proceed to the“equilibrium” phase wherein, in a Darwinian way, the constantinteraction of tumour cells with the immune system over a very longperiod of time, promotes survival and selection (shaping or “editing”)of tumour cells resistant to various antitumour immune mechanisms[151,153]. Finally, the resultant “edited” cells enter an escape phasewhen tumours grow progressively and eventually metastasize,leading to patient's Death Proper induction of antitumour immuneresponse heavily depends upon the recognition of tumour antigens bythe immune system. These antigens are usually divided into two‘main’ groups: tumour-specific antigens (TSA)—molecules that areunique to tumour cells and tumour-associated antigens (TAA)—molecules present on both normal and transformed cells [1]. TSAinclude products of mutated genes (resulting from point mutations ortranslocations) or antigens of viral origin (e.g. EBNA—Epstein–Barrvirus nuclear antigen in Burkitt's lymphoma). TAA include differen-tiation antigens (e.g. melanocyte differentiation antigens or MART-1and tyrosinase in malignant melanoma), cancer-testis antigens (e.g.MAGE, GAGE and NY-ESO-1 in malignant melanoma and othertumours) that are characterized by ubiquitous expression in cancercells but restricted to adult germ cells in healthy individuals [153,154]and last but not the least, overexpressed or amplified antigens (e.g.HER2/neu in metastatic breast cancer). These tumour antigens arevital for activation of both innate as well as adaptive antitumourimmune responses.
Tumour cells that successfully undergo “equilibrium” phase (asdescribed in the previous section) develop a broad spectrum of
63A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2010) 53–71
mechanisms that not only prevent cellular destruction by immunecells but also effectively inhibit both innate and adaptive immuneresponses, providing a general immunosuppressive environment intumour tissues. Apparently, development of tolerance might befacilitated over immunity possibly because immune system isexposed to the tumour-derived antigens gradually, due to relativelyslow tumour growth. Also, the presentation of tumour antigens occursin the absence of “danger signals” that influence maturation of APCs(like DAMPs), a problem that is further augmented by the fact thatmany cytokines secreted within tumour microenvironment such asTGF-β1 and VEGF tend to suppress maturation of DCs and all thiscollectively leads to induction of tolerance rather than antitumourimmune response [9]. This immunosuppressive environment mightbe further strengthened due to excessive secretion of chemokines orarachidonic acidmetabolites (such as PGE2) that promote recruitmentof myeloid-derived suppressor cells (MDSCs), which might in turninhibit DC-maturation via secretion of arginase-1 and NOS2. More-over, tumour cells often express FasL whose interaction with Fasreceptor typically present on the surface of activated T cells, mightinduce T cell apoptosis [1].
Similarly, on the cellular level, the poor immunogenicity oftumour cells might be accounted for by multiple mechanisms. Firstof all, tumour cells are derived from normal cells and they differminimally in their antigenic repertoire from normal cells. Thisproblem is further compounded by the ability of tumour cells tovary their antigenic profile [155]. Also, T cells recognizing selfantigens usually have low affinity to the MHC-peptide complex dueto negative selection taking place in the thymus. Additionally, tumourgrowth is frequently associated with the loss of MHC class I moleculesexpression thereby efficiently preventing recognition of tumourantigens by cytotoxic T cells. Roughly 40–90% of human tumoursdisplay total or selective loss of certain MHC class I molecules[156,157]. Moreover, tumour cells also exhibit deregulation ofantigen processing and presentation pathways through downregulation of crucial participating proteins such as TAP1 (involvedin transportation of polypeptide-derived proteasomal degradationproducts from the cytoplasm to the ER for loading onto the MHC classI molecules) as well as LMP2 (low-molecular-mass protein 2) andLMP7 (both are components of the so called immunoproteasome—anIFN-γ-induced proteasome with increased peptide-processingcapacity). Furthermore, to make matters worse, tumour cells mightoverexpress self-antigens thereby leading to stimulation of theregulatory CD4+CD25+FoxP3+ T cells (Tregs) [19]. Here, Tregshave the potential to suppress DC-maturation (due to highexpression of inhibitory molecule CTLA-4, a ligand for CD80/CD86co-stimulatory molecules) and stimulate them to secrete thetryptophan catabolizing enzyme, indoleamine 2,3-dioxygenase(IDO) [9,16]. Since, lymphocytes cannot restore tryptophan (a crucialenergy source) and are very sensitive to toxic effects of tryptophandegradation products (like kynurenines); IDO expression efficientlyblocks proliferation of CD8+ T cells and promotes apoptosis of CD4+
T cells [9]. T cell function might be further inhibited by nitric oxideand arginase-1, both secreted by plethora of tumour cells [158]. Thus,not surprisingly, tumour cells commonly secrete chemokines (likeCCL22) that attract Tregs to tumour tissue [159]. In fact, highexpression of Tregs in tumour samples taken from patients withmetastatic ovarian cancer negatively correlates with overall survivalrate [159]. Cancer cells may also develop strategies preventing cell-mediated lysis by secreting various molecules such as, TRAIL decoyreceptors, proteins that inhibit caspase activity (e.g. FLIP—caspase-8 (FLICE)-like inhibitory protein) and serine-protease inhibitor PI9which can efficiently inhibit granzyme B-mediated apoptosis-induc-tion pathway as well as down-regulate or mutate death receptors andcaspase-8 [9,160,161]. Genetic instability and frequent p53 mutationsin tumour cells have also been considered to be possible reasonsbehind cancer cell's resistance to CTLs-mediated lyses however, the
exact mechanisms of this interaction needs to be elucidated [162].Moreover, cancer cells might also exhibit a dodging strategy ofshedding tumour antigens from the plasma membrane, which canhelp in blocking tumour-specific antibodies and TCRs while still incirculation before they reach the tumour itself [16]. Altogether, itseems that tumour may serve as a “false” immune organ, wheretolerogenic DCs, tumour-associated macrophages (TAMs), myeloid-derived suppressor cells (MSDCs), regulatory T cells and variousmediators/factors secreted by it could create a highly unfriendlyenvironment for the development of effective and successfulantitumour immune response. The malevolent immunosurveillanceevading capabilities of tumours, which have been described above,have been formidable barriers to the clinical success of cancerimmunotherapy [155,163], a predicament which is further ebbed bypatient-to-patient tumour microenvironment variability [163]. Thus,not surprisingly, the field of cancer immunotherapy has experienceda number of failures [155]. Moreover, cancer immunotherapy isconsidered to be almost completely ineffective in patients with solidtumours [164]. However, although one might argue that, in recenttimes, cancer immunotherapy has achieved some degree of success asan alternative therapy (thanks largely to the exploits of T cell-basedimmunotherapy instigated via IL-2 administration) yet the promiseof a viable cancer vaccine has remained unfulfilled [163]. Due to thesereasons, tumour-killing therapies like chemotherapy have beenpreferred over immunotherapy, in recent past [165]. However, it isincreasingly becoming evident that there exist considerable cases ofcancer resistance to these cell death-inducing therapies as well (asdiscussed later in Section 6) and since they mostly induce apoptosis,the resultant host immune response is relatively flimsy [165]. Thus, ininterest of more long-term gains, researchers have been in pursuit ofa method to combine cell death-inducing therapies and immu-notherapy; and hence the concept of ‘immunogenic cancer cell death’is being hailed as a vital premise.
The immune system, however, is a double edged sword. There isstrong evidence indicating that even chronic inflammation couldincrease the risk of malignant transformation in certain tissues[16,166]. This “smouldering” inflammation is mainly executed byactivated macrophages, mast cells and neutrophils, which mightpromote DNA damage in cancer cells through production of reactiveoxygen species. They might also further enhance the favourability fortumour progression via various processes like secretion of pro-inflammatory/pro-angiogenic cytokines (TNF, IL-1, IL-6 or VEGF),various metalloproteinases (such as MMP-9), prostaglandins and anumber of other mediator factors/molecules.
Box 4.Cell death resistance to tumour therapeutics
Resistance of tumour cells to cytotoxic therapies is amongst themost important reasons for treatment failure. Multiple mechan-isms linked to the tumour cells and their microenvironmentaccount for this.
• Tumour cells: One of the common drug resistance mecha-nisms involves increased efflux of hydrophobic cytotoxic che-motherapeutics mediated by a large family of ATP-bindingcassette (ABC) transporters,which includePglycoprotein (Pgp).ABC transporters genes are frequently overexpressed by drug-experienced tumour cells. Several chemotherapeutics are notso good substrates for ABC transporters per se, but withincells, their covalent interaction with low-molecular weightthiol-containing peptides (such as glutathione) leads to theireffective removal from the cells [4]. Resistance to some drugssuch as cisplatin may also result from decreased drug uptakerather than efflux. Effective drug resistance mechanism isalso based on increased DNA-repair capacity of tumour cells[5,6].
64 A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2009) 53–71
• Resistance to tumour cell death: Tumour cell death inresponse to cytotoxic treatments is usually a direct resultof drug-induced or radiation-induced damage. Excessivedemolition by necrosis is rarely encountered in the clinicalsetting as it would require drug doses causing unaccept-able toxicity. At lower drug doses tumour cells try to copewith the damage by triggering cytoprotective mechanisms.In case of failure, cells die via apoptotic or apoptotic-likemechanisms, including mitotic catastrophe, or enter cellularsenescence. Although autophagy is primarily a cytopro-tective mechanism, it can cause cell death when activated inexcess and its role following anticancer treatment is currentlyunder investigation. Cancer resistance mechanism often in-volves deregulation in the p53 pathway which associates withimpaired induction of apoptosis and increased radio- andchemo-resistance. Impairment of apoptosis signalling cas-cades (e.g. on the level of the Bcl-2 family members, inhi-bitor of apoptosis proteins etc.) can also contribute totumour resistance to therapy [15].
• Tumour microenvironment: Abnormalities of the tumourvascular network confer blood flow resistance and impairblood supply causing formation of hypoxic regions withintumours, which severely affect the distribution and activityof the anticancer drugs. Impaired nutrients delivery drivescell cycle arrest in tumour cells thereby making them lesssusceptible to many chemotherapeutics that target activelyproliferating cells [21]. Tumour cell metabolic adaptations inthe face of the hostile tumour microenvironment, such asswitch to anaerobic glycolysis are associated with forma-tion of lactic acid that lowers extracellular pH. Anticanceragents such as doxorubicin, vinblastine or vincristine areprotonated at low pH and cannot penetrate plasma membraneof tumour cells [23,24]. Doxorubicin exerts some of its cyto-toxic effects through generation of free radicals that damageDNA. At lowO2 concentration its capacity to reduce oxygen tosuperoxide anion is decreased [28].
6. Anticancer therapeutics and immunogenic cell death: envisagedfuture trends
As apparent from the above discussion, during anticancertreatment regimes, the refurbishing of anticancer immunity couldbe just as important as tumour cell death and immunogenicapoptosis/cell death seems to be capable of providing both theseoptions [10,12,14].
Antitumour immunity might be mediated by different DAMPs incase of different chemotherapeutic agents however the overall resultmight still be dependent on, (i) induction of immunogenic cancer celldeath and (ii) acquiring of tumour antigens (possessed by dyingtumour cells) by antigen presenting cells like DCs (followed byeliciting tumour specific killer T cell response) [55].
Although the concept of DAMP-defined immunogenicity of celldeath is relatively new, the fact that certain types of chemothera-peutic drugs could have immuno-modulatory effects has beenknown for some time. And it would not be wrong to say that theprior knowledge of such immuno-modulatory activity finally led toderivation of the immunogenic cell death concept. This argument isvindicated by the fact that doxorubicin [167] was already knownto be capable of modulating the immune response long before itwas shown to be inducing surface exposure of DAMPs [13,108].Subsequently, various therapeutic agents like oxaliplatin, γ-irradia-tion, bortezomib and mitoxantrone have joined the list of immuno-genic cell death inducers thereby supporting the premise that manytherapeutic strategies which have known immuno-modulatory orimmuno-stimulatory effects ought be investigated further to ascer-tain whether they are associated with exposure or release of known/unknown DAMPs.
6.1. Current inducers of immunogenic cell death: glitches and side effects
The requirement for searching new therapeutic agents capable ofinducing immunogenic cell death is further pushed by the fact thatvarious existing inducers of immunogenic cell death suffer fromdifferent disadvantages.
For instance, anthracyclines like doxorubicin have been reportedto have various side effects [168,169] as well as multi-factorial cancerresistance problems [170]. Doxorubicin in particular has manyserious, clinical use-limiting side effects such as bone marrowdepression [171], liver and kidney injury [172,173], and cardiactoxicity [174]. In fact, cardiotoxicity is considered to be doxorubicin'smost serious side effect [175] since it has been reported to be fatal incertain cases [176]. Though the doxorubicin-associated cardio-myopathy/congestive heart failure is dose-dependent [176] yetmyocardial necrosis and acute myocardial dysfunction have beenreported following doxorubicin administration in a cancer patienteven though the dose used was well below the one reported to havesevere cardiotoxic effects [177]. These problems have led toimposition of stringent limits on overall chemotherapeutic usage ofdoxorubicin. Here, the cardiotoxicity induced by doxorubicin has beenattributed to its reactive oxygen species (ROS) producing ability sincecardiomyocytes are very susceptible to ROS-induced apoptosis [178].Ironically, it is this ROS-producing ability of doxorubicin, which hasbeen hypothesized to be behind its ER-stressor capabilities ultimatelyleading to surface-CRT exposure [2]. Interestingly, it has beenobservedthat TLR-2 and TLR-4 knockout mice are resistant to doxorubicininduced cardiotoxicity [179,180] and demonstrate reduced oxidativeand inflammatory stress responses, including reduced cardiacapoptosis. In this regard it might be conceivable that dying cardio-myocytes expose and/or release DAMPs which in turn contribute toinflammation and damage of cardiomyocytes. Similarly, mitoxantronehas also been shown to have cumulative cardiotoxicity as one of itsside effects besides ability to induce secondary malignancy (acuteleukaemia), nausea, alopecia and infections [181]. Further, oxaliplatinhas also been shown to have its own distinctive spectra of side effectssuch as gastrointestinal problems, ototoxicity, myelosuppression andacute neurotoxicity [169] wherein, neurotoxicity has dose-limitingside effects [182]. Also, ionizing radiations have been known to exhibitside effects like bone marrow toxicity, neutropaenia and myelo-suppression [183]. Moreover, even bortezomib has been shown toexhibit side effects such as peripheral neuropathy (PN), an importantdose-limiting toxicity [184] and certain skin-associated side effectssuch as rash and lupus erythematosus tumidus, a single case of whichwas reported recently [185].
As apparent from the above discussion, if the idea of immunogeniccancer cell death is to be further strengthened then it demandsscreening and selection of newer chemotherapeutic agents (previ-ously known or newly discovered), capable of inducing it. In thefollowing sections we will describe promising anticancer therapiesknown to be associated with a pronounced immuno-stimulatoryresponse.
6.2. Photodynamic therapy: illuminating the road towards immunogeniccell death
Photodynamic therapy (PDT) is a very promising therapeuticmodality for treatment of various cancers with several advantagesover ‘classical’ anticancer regimes [186,187]. PDT is basically a two-step procedure involving administration of (tumour-localizing)photosensitiser followed by its activation with a light of specificwavelength absorbed by it, which ultimately leads to production ofROS, thereby causing oxidative stress [188]. The best feature aboutPDT is that, this oxidative stress could be directed at a particularsub-cellular organelle within the cancer cells based upon thehydrophilicity/hydrophobicity based sub-cellular localization of
65A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2010) 53–71
the photosensitiser (extensively reviewed by Buytaert et al. [188])e.g. a photosensitiser localizing in mitochondria would eventuallylead to mitochondrial oxidative stress [189,190]. Fortunately, thereexist a variety of photosensitisers with different sub-cellularlocalizations, which could be used for further exploring theimmunogenic cell death concept; for example, since immunogenicapoptosis has been linked to ER-stress, we envisage that furtherinvestigation into PDT treatment with ER-localizing photosensitisers[188] might lead to interesting results in terms of immunogenic celldeath-induction. Importantly, the ability of PDT to induce activationof the innate immune system and specific ‘antitumour immunity’ isan established phenomenon [191].
While preclinical studies analyzing antitumour immune responsein PDT treated murine tumours have been carried out for sometime[183,192], it was only recently that a ‘pioneering’ study showed forthe first time that PDT in clinical settings could enhance immuno-logical recognition of TAAs [192]. More specifically it was shown that5-aminolevulinic acid (5-ALA)- and Porfimer sodium (Photofrin)-based PDT could enhance the systemic immune response againstbasal cell carcinoma-associated tumour antigen, Hip-1 (hedgehog-interacting protein 1, a transmembrane protein) [192]. It was alsoobserved that, PDT-based enhancement of antitumour immuneresponse was inversely correlated with the surface area treated(superficial vs. nodular lesions) and the light dosage (fluence andfluence rates) used. Here, according to the authors, higher lightdosage probably lead to vascular shut-down thereby hamperingantitumour response by reducing the infiltration of immune cellsinto the treated area [192]. We also suspect that, the lattercorrelation might exist since higher light dosage may result inmore tumour cell membrane permeabilization or necrosis, whichmight lead to a relative ‘slack’ in antigen presentation andprocessing, a premise driven by the fact that clearance of necroticcells by phagocytes is less efficient than that for apoptotic ones[115]. And last but not the least, the authors also observed thatcompared to surgery, PDT resulted in significantly higher immunereactivity [192].
Likewise, lysates generated by PDT have the ability to induce DCmaturation and IL-12 expression [191]. Even PDT-based vaccines havebeen shown to have the ability to activate both innate as well asadaptive immune system [191]. In terms of DAMPs it has beenobserved that various PDT treatments usually associate with extra-cellular release of HSP70 [183]. In fact, a recent report observed thatsignals such as HSP70 released from PDT-treated cancer cells couldillicit the tumour-associated macrophages (TAMs) to producecomplement proteins, a process mediated by the communicationinvolving TLR 2 and 4 receptors along with NFκB activation [193].Besides, at certain doses, PDT is often accompanied by release of
Table 2Anticancer immuno-stimulatory activity of certain chemotherapeutic agents or treatment m
Chemotherapeutic agents ortreatment modalities
Comments on anticancer immun
Gemcitabine and Anti-CD40 antibodies It has been shown that antigenscross-presentation by DCs [224].gemcitabine with anti-CD40 antiimmunity [225].
Cyclophosphamide and Anti-OX40 antibodies Combined administration of cyclco-stimulatory signals to T and Na potential DAMP [227]. Also, cyctheir immunosuppressive functio
Combination Immunotherapies ‘Combination Immunotherapies’released by activated DCs) can in
Irinotecan (CPT-11) Irinotecan (CPT-11), a chemothesurface of tumour cells, which ca
various inflammatory cytokines/chemokines and it also has theability to activate the immune cells like macrophages [183].Furthermore, it has been shown that loading of dendritic cells withirradiated or PDT-treated apoptotic tumour cells could induce potentantitumour immune response [194,195].
6.3. Exploration towards potential inducers of immunogenic cell death:a snap-shot
Apart from PDT, another ‘classical’ methodology potentiallycapable of inducing immunogenic cancer cell death is the one thatutilizes ‘exo-/endo-toxins’. Ability of attenuated microbes to induceanticancer immune response is a well-established phenomenon,clinically practiced pre-dominantly in the form of treatment withBacillus Calmette-Guerin (BCG) [196]. However, the studies elucidat-ing the ability of exotoxins to induce release of DAMPs are very recente.g. the ability of epidermal growth factor receptor-targeted diphthe-ria toxin (DT-EGF) to induce HMGB1 release (as discussed previous-ly). Though endotoxins have not been shown to associate with aparticular DAMP per se, an endotoxin in particular, called lipopoly-saccharide (LPS) has been shown to be capable of inducing ‘ER failure’,a phenomenon accompanied by persistent ER stress, increase of UPRmarkers, down-regulation of functional ER proteins and absence of ERrecovery or apoptosis [197]. Whether LPS or other endo-/exo-toxinscould have an untapped potential in terms of DAMP release is aninteresting question, which needs to be further investigated, alsobecause of relatedness of these molecules with PAMPs. Otherchemotherapeutic drugs or treatment modalities found to beexhibiting anticancer immuno-stimulatory activity have been brieflydescribed in Table 2.
On a more generalized level, we may propose that agents havingthe ability to induce ER stress or probably eIF2α (hyper)phosphor-ylation might be the best initial stepping stones towards creating a‘library’ of immunogenic cell death inducing compounds/drugs.Moreover, eIF2α might be phosphorylated by kinases other thanPERK, such as GCN2 and PKR [100] under ER stress or through non-canonical pathways involving HspB8 and Bag3 independent of ERstress [201]. Although the relationship with immunogenic cell deathwas not explored in the latter reports, these findings suggest thatthere may be non-overlapping and perhaps stress-specific mechan-isms regulating the phosphorylation status of eIF2α and its down-stream targets. Further investigation into these ‘non-canonical’pathways of eIF2α phosphorylation might assist us in uncoveringnewer and unique mechanics of ecto-DAMP exposure, which canopen up new possibilities for novel ‘alternative’ drug targets. We alsoneed to investigate how heavily induction of immunogenic cell deathdepends upon the ER stress signalling pathways.
odalities.
o-stimulatory activity
from gemcitabine-treated tumour cells can reach local lymph nodes and undergoAlthough, this effect alone is insufficient to trigger immunity yet combination ofbodies (that activate DCs) results in development of an effective antitumour
ophosphamide and anti-OX40 antibodies has been found to be capable of deliveringK cells [226]. Moreover, cyclophosphamide could induce high levels of uric acid,lophosphamide can selectively decrease the percentages of Tregs and also inhibitn [228].based on administration of chemotherapeutic drugs together with IL-12 (a cytokineduce long-term antitumour immunity in animals [229].ociated with the release of proinflammatory cytokines such as IL-6 or IL-8 [230].ng agents might be useful in antitumour immuno-stimulation.rapeutic agent, has the ability to induce expression of LY6D/E48 antigen on then in turn be targeted by monoclonal antibodies [231].
66 A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2009) 53–71
6.4. Immunological conundrums and therapeutic restrictions of DAMPs
Apart from the exploration related to the therapeutic agents/modalities, it is also very important to investigate a few ‘problems’that might be associated with certain DAMPs and could either lead toreduction of their immuno-stimulatory activity or increase in sideeffects (e.g. autoimmune diseases). For instance, it is noteworthythat the same molecules released from apoptotic and necrotic cellscould have different effect on immune response. Here, the cellularcontent of apoptotic cells, which had been exposed to activatedcaspases, seems to differ from that of necrotic cells. Activatedcaspases could also modify crucial cellular molecules therebysilencing them or granting them with anti-inflammatory properties.For example, it has been shown that caspase activation could targetthe mitochondria to produce ROS, which could prove to be critical ininducing tolerance to apoptotic cells [31] because these ROS couldoxidise the potential danger signal HMGB1 released from dying cellsand thereby neutralize its stimulatory activity [31,198]. Whether thisoxidation based modification is also capable of diminishing theimmunological properties of other DAMPs is as of now unclear. Onthe other hand, therapeutic strategies involving direct usage ofknown DAMPs also need further investigation due to the potentialside effects associated with them. For example, although directadministration of recombinant CRT (rCRT) could be one of thestrategies for the future therapy, it has a few side effects-relatedconcerns [66]. For instance, while it has been shown that the N-terminal fragment of CRT (vasostatin) has potent antiangiogeniceffects yet there is a risk that topical application of full-length rCRTcould promote tumour angiogenesis [66]. Also, CRT might preventperforin- and complement-mediated cytolysis of tumour targets.Moreover, CRT profoundly affects wound healing by recruiting cellsessential for repair, thereby stimulating cell growth and increasingextracellular matrix production [66].
And last but not the least, it is important to question the overallanticancer therapeutic relevance (as well as reliability) of variousDAMPs released during secondary necrosis (like HMGB1 or probablySAP130 [43]), partly due to the very nature of secondary necrosis andpartly due to the autoimmune response concerns. While secondarynecrosis (which follows apoptosis in a cell) is predominantly an invitro phenomenon (primarily due to the absence of proper clearancemechanisms) yet it has been reported that secondary necrosis mightalso exist in in vivo conditions [26]. It has been reported that thereare conditions in vivo where apoptotic scavenging/clearance mightfail thereby paving way for elimination via secondary necrosis[199,200]. For example, apoptosis in cartilage chondrocytes usuallyends up as secondary necrosis since lack of direct cell-to-cell contactsand the avascular nature of articular cartilage restricts access andremoval of dying cells by scavenger/phagocytic cells [201,202]. Onthe other hand, secondary necrosis might also be induced in vivo inhyper-inflammation conditions where excessively induced apoptosisoverwhelms the scavenging or phagocytic capacity of the host'simmune system or in conditions where normal scavenging orphagocytic functions of the immune cells might be compromiseddue to certain agents/factors [26]. However in terms of anticancertherapeutics we probably cannot completely rely on secondarynecrosis due to the uncertainty surrounding the inflammation causedby it [203]. And even if such inflammation might be controlledsomehow, the problem of autoimmune disease still looms large.More specifically it has been proposed that, the debris released bycells undergoing secondary necrosis might be taken up by antigenpresenting cells like DCs probably attracted and/or sensitised byDAMPs (like HMGB1 and other end-stage degradation products)released by secondary necrotic cells [26]. This debris might becomposed of autoantigens which can ultimately lead to autoimmuneresponse [26]. In fact HMGB1 (released during secondary necrosis)has been considered to participate in the pathogenesis of autoim-
mune diseases [109,204]. For instance, HMGB1-nucleosome com-plexes have been shown to be capable of activating APCs and therebycontributing towards pathogenesis of systemic lupus erythematosus(SLE) [205]. Similarly, a study of certain autoantigens has showedthat, once their cleaved forms (cleaved during apoptotic executionphase) are released, they are further cleaved (in a caspase-independent fashion) and the resultant products might induceinflammation and stimulate adverse autoimmune responses[206,207]. Therefore, one may wonder how reliable or rather howadvisable is it, to consider DAMPs like HMGB1 released duringsecondary necrosis to be capable of instigating a specific anticancerresponse based on processing of TAAs. From the above discussion itis unambiguous that one needs to be careful in deeming DAMPs likeHMGB1 released during secondary necrosis as therapeuticallyrelevant.
The situation might however be different if HMGB1 is found to bereleased in more early or intermediate apoptotic phases or if it isascertained that DAMPs released during secondary necrosis mightsomehow be utilized without harming the patient in the process.Clearly, more in vivo research is required to be done in order toascertain as to how DAMPs like HMGB1 might be considered usefulwith respect to apoptosis. One recent research which has supportedthe vitality of HMGB1 for immunogenic cell death showed thatabsence of HMGB1 in tumour cells undergoing immunogenicapoptosis (induced via anthracyclines, oxaliplatin or ionizing radia-tion) ultimately abolished their ability to induce T cell basedanticancer immune response [43]. On a different note though,interestingly it has been found that several patients suffering fromvarious types of cancers have autoantibodies against certain notoriousDAMPs (probably acting as or surrounding the tumour-associatedantigens), known examples of which are: HSP70 (esophageal cancer,hepatocellular carcinoma and leukaemia), HSP60 (colon cancer), CRT(pancreatic cancer and hepatocellular carcinoma) and S100A7(ovarian cancer) [208]. It has been proposed that the presence ofthese autoantibodies, generated against corresponding autoantigens,might actually be used as potential prognostic and diagnosticbiomarker for the manifestation of cancer [208]. This ‘autoantigen’status of certain DAMPs might have some implications for immuno-genic cancer cell death research, and it is imperative that thisconnection be further investigated.
7. Conclusions
Mounting evidence from recent studies in cancer research andtumour immunology argue that the most effective therapeuticapproach will be the one dealing concurrently with two of the mostdreadful cancer features; the ability to overrule (apoptotic) cell deathmechanisms as well as immunosurveillance. Hence, there is an urgentneed to optimize current treatment modalities and/or chemothera-peutic agents or develop new ones, so as to combine these properties;i.e. induce apoptotic cell death and concomitantly have immuno-stimulatory effects, in association with certain DAMPs. Thesemodalities/agents, if further investigated, might prove to be potentialinducers of ‘immunogenic cancer cell death’ and have potential asantitumour vaccination strategies. A major roadblock in the way frombench-to-bed side that will be needed to overcome is the dosagedifference between in vitro and in vivo conditions. For instance, thedoses of drugs used to kill tumour cells in vitro are at least 10-foldhigher than maximally tolerated by patients. Therefore readytranslation of the in vitro findings to a clinical setting is incrediblytough. Moreover, restrained diffusion of chemotherapeutics withintumours reduces their cytotoxicity and definitely does not help ininduction of immunogenic cell Death Also a critical issue for furtherdevelopment of the immunogenic cell death concept will be thecharacterization of more potential DAMPs and their secretion/exposure pathways. This might also open up new therapeutic targets
67A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2010) 53–71
for targeted chemotherapeutics. It will also be vital to furtherdelineate the exact mechanisms of interaction between dying tumourcells and innate immune system, our current understanding of whichcould well be a tip of the iceberg.
Acknowledgements
The work from the author's laboratories was supported by thegrants from the K.U.Leuven (OT49/06), F.W.O Flanderen (G.0661.09)to P.A. This paper presents research results of the IAP6/18, funded bythe Interuniversity Attraction Poles Programme, initiated by theBelgian State, Science Policy Office. D.V.K. is a postdoctoral researchfellow of the Fund for Scientific Research Flanders (FWO-Vlaanderen),Belgium. J.G. receives funding from the Foundation for Polish ScienceTeam Programme co-financed by the EU European Regional Devel-opment Fund and is a recipient of the Mistrz Award from theFoundation for Polish Science. Lastly, all figures were produced usingServier Medical Art (www.servier.com) for which the authors wouldlike to acknowledge Servier.
References
[1] O.J. Finn, Cancer immunology, N. Engl. J. Med. 358 (2008) 2704–2715.[2] T. Panaretakis, O. Kepp, U. Brockmeier, A. Tesniere, A.C. Bjorklund, D.C.
Chapman, M. Durchschlag, N. Joza, G. Pierron, P. van Endert, J. Yuan, L. Zitvogel,F. Madeo, D.B. Williams, G. Kroemer, Mechanisms of pre-apoptotic calreticulinexposure in immunogenic cell death, EMBO J. 28 (2009) 578–590.
[3] C. Garrido, G. Kroemer, Life's smile, death's grin: vital functions of apoptosis-executing proteins, Curr. Opin. Cell Biol. 16 (2004) 639–646.
[4] T. Ishikawa, The ATP-dependent glutathione S-conjugate export pump, TrendsBiochem. Sci. 17 (1992) 463–468.
[5] L. Kelland, The resurgence of platinum-based cancer chemotherapy, Nat. Rev.Cancer 7 (2007) 573–584.
[6] L.P. Martin, T.C. Hamilton, R.J. Schilder, Platinum resistance: the role of DNArepair pathways, Clin. Cancer Res. 14 (2008) 1291–1295.
[7] O. Joffre, M.A. Nolte, R. Sporri, C. Reis e Sousa, Inflammatory signals in dendriticcell activation and the induction of adaptive immunity, Immunol. Rev. 227(2009) 234–247.
[8] D. Hanahan, R.A. Weinberg, The hallmarks of cancer, Cell 100 (2000) 57–70.[9] L. Zitvogel, A. Tesniere, G. Kroemer, Cancer despite immunosurveillance:
immunoselection and immunosubversion, Nat. Rev. Immunol. 6 (2006) 715–727.[10] A. Tesniere, T. Panaretakis, O. Kepp, L. Apetoh, F. Ghiringhelli, L. Zitvogel, G.
Kroemer, Molecular characteristics of immunogenic cancer cell death, Cell DeathDiffer. 15 (2008) 3–12.
[11] M.C. Maiuri, E. Zalckvar, A. Kimchi, G. Kroemer, Self-eating and self-killing:crosstalk between autophagy and apoptosis, Nat. Rev. Mol. Cell. Biol. 8 (2007)741–752.
[12] O. Kepp, A. Tesniere, L. Zitvogel, G. Kroemer, The immunogenicity of tumor celldeath, Curr. Opin. Oncol. 21 (2009) 71–76.
[13] M. Obeid, A. Tesniere, F. Ghiringhelli, G.M. Fimia, L. Apetoh, J.L. Perfettini, M.Castedo, G. Mignot, T. Panaretakis, N. Casares, D. Metivier, N. Larochette, P. vanEndert, F. Ciccosanti, M. Piacentini, L. Zitvogel, G. Kroemer, Calreticulin expo-sure dictates the immunogenicity of cancer cell death, Nat. Med. 13 (2007)54–61.
[14] L. Zitvogel, L. Apetoh, F. Ghiringhelli, F. Andre, A. Tesniere, G. Kroemer, Theanticancer immune response: indispensable for therapeutic success? J. Clin.Invest. 118 (2008) 1991–2001.
[15] F.H. Igney, P.H. Krammer, Death and anti-death: tumour resistance to apoptosis,Nat. Rev. Cancer 2 (2002) 277–288.
[16] M. Jakobisiak, W. Lasek, J. Golab, Natural mechanisms protecting against cancer,Immunol. Lett. 90 (2003) 103–122.
[17] J. Cui, T. Shin, T. Kawano, H. Sato, E. Kondo, I. Toura, Y. Kaneko, H. Koseki, M.Kanno, M. Taniguchi, Requirement for Valpha14 NKT cells in IL-12-mediatedrejection of tumors, Science (New York, NY) 278 (1997) 1623–1626.
[18] S. Gillessen, Y.N. Naumov, E.E. Nieuwenhuis, M.A. Exley, F.S. Lee, N. Mach, A.D.Luster, R.S. Blumberg, M. Taniguchi, S.P. Balk, J.L. Strominger, G. Dranoff, S.B.Wilson, CD1d-restrictedT , cells regulate dendritic cell function and antitumorimmunity in a granulocyte-macrophage colony-stimulating factor-dependentfashion, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 8874–8879.
[19] M. Dougan, G. Dranoff, The immune response to tumors, Curr. Protoc. Immunol.Chapter 20 (2009) 11 Unit 20.
[20] N. Vanlangenakker, T.V. Berghe, D.V. Krysko, N. Festjens, P. Vandenabeele,Molecular mechanisms and pathophysiology of necrotic cell death, Curr. Mol.Med. 8 (2008) 207–220.
[21] I. Tannock, Cell kinetics and chemotherapy: a critical review, Cancer Treatmentreports 62 (1978) 1117–1133.
[22] F. Madeo, M. Durchschlag, O. Kepp, T. Panaretakis, L. Zitvogel, K.U. Frohlich, G.Kroemer, Phylogenetic conservation of the preapoptotic calreticulin exposurepathway from yeast to mammals, Cell Cycle 8 (2009) 639–642.
[23] I.F. Tannock, D. Rotin, Acid pH in tumors and its potential for therapeuticexploitation, Cancer Res. 49 (1989) 4373–4384.
[24] L.E. Gerweck, S. Vijayappa, S. Kozin, Tumor pH controls the in vivo efficacy ofweak acid and base chemotherapeutics, Mol. Cancer Ther. 5 (2006)1275–1279.
[25] J. Savill, I. Dransfield, C. Gregory, C. Haslett, A blast from the past: clearanceof apoptotic cells regulates immune responses, Nat. Rev. Immunol. 2 (2002)965–975.
[26] M.T. Silva, A. do Vale, N.M. dos Santos, Secondary necrosis in multicellularanimals: an outcome of apoptosis with pathogenic implications, Apoptosis 13(2008) 463–482.
[27] C. Peter, S. Wesselborg, K. Lauber, Role of attraction and danger signals in theuptake of apoptotic and necrotic cells and its immunological outcome, in: D.V.Krysko, P. Vandenabeele (Eds.), Phagocytosis of Dying Cells, Springer Science +Business Media B.V., Berlin, 2009, pp. 63–101.
[28] K.A. Kennedy, Hypoxic cells as specific drug targets for chemotherapy, Anti-Cancer Drug. Design. 2 (1987) 181–194.
[29] M. Napirei, H.G. Mannherz, Molecules involved in recognition and clearance ofapoptotic/necrotic cells and cell debris, in: D.V. Krysko, P. Vandenabeele (Eds.),Phagocytosis of Dying Cells, Springer Science + Business Media B.V., Berlin,2009, pp. 103–145.
[30] D.V. Krysko, K. D'Herde, P. Vandenabeele, Clearance of apoptotic and necroticcells and its immunological consequences, Apoptosis 11 (2006) 1709–1726.
[31] I. Bournazou, J.D. Pound, R. Duffin, S. Bournazos, L.A. Melville, S.B. Brown, A.G.Rossi, C.D. Gregory, Apoptotic human cells inhibit migration of granulocytes viarelease of lactoferrin, J. Clin. Invest. 119 (2009) 20–32.
[32] S. Brown, I. Heinisch, E. Ross, K. Shaw, C.D. Buckley, J. Savill, Apoptosis disablesCD31-mediated cell detachment from phagocytes promoting binding andengulfment, Nature 418 (2002) 200–203.
[33] S.J. Gardai, 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, Cell-surfacecalreticulin initiates clearance of viable or apoptotic cells through trans-activation of LRP on the phagocyte, Cell 123 (2005) 321–334.
[34] V.A. Fadok, D.L. Bratton, A. Konowal, P.W. Freed, J.Y. Westcott, P.M. Henson,Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatorycytokine production through autocrine/paracrine mechanisms involving TGF-beta, PGE2, and PAF, J. Clin. Invest. 101 (1998) 890–898.
[35] Y.Q. Xiao, K. Malcolm, G.S. Worthen, S. Gardai, W.P. Schiemann, V.A. Fadok, D.L.Bratton, P.M. Henson, Cross-talk betweenERK , p38MAPK , mediates selectivesuppression of pro-inflammatory cytokines by transforming growth factor-beta,J. Biol. Chem. 277 (2002) 14884–14893.
[36] R.E. Cocco, D.S. Ucker, Distinct modes of macrophage recognition for apoptoticand necrotic cells are not specified exclusively by phosphatidylserine exposure,Mol. Biol. Cell. 12 (2001) 919–930.
[37] R.E. Voll, M. Herrmann, E.A. Roth, C. Stach, J.R. Kalden, I. Girkontaite,Immunosuppressive effects of apoptotic cells, Nature 390 (1997) 350–351.
[38] Y. Ren, Y. Xie, G. Jiang, J. Fan, J. Yeung, W. Li, P.K. Tam, J. Savill, Apoptotic cellsprotect mice against lipopolysaccharide-induced shock, J. Immunol. 180 (2008)4978–4985.
[39] R.B. Birge, D.S. Ucker, Innate apoptotic immunity: the calming touch of death,Cell Death Differ. 15 (2008) 1096–1102.
[40] J. Vakkila, M.T. Lotze, Inflammation and necrosis promote tumour growth, Nat.Rev. Immunol. 4 (2004) 641–648.
[41] J. Thorburn, H. Horita, J. Redzic, K. Hansen, A.E. Frankel, A. Thorburn, Autophagyregulates selective HMGB1 release in tumor cells that are destined to die, CellDeath Differ. 16 (2009) 175–183.
[42] G. Kroemer, L. Galluzzi, P. Vandenabeele, J. Abrams, E.S. Alnemri, E.H. Baehrecke,et al., Classifications of cell death: recommendations of the NomenclatureCommittee on Cell Death 2009, Cell Death Differ. 16 (2009) 3–11.
[43] D.R. Green, T. Ferguson, L. Zitvogel, G. Kroemer, Immunogenic and tolerogeniccell death, Nat. Rev. Immunol. 9 (2009) 353–363.
[44] S.L. Fink, B.T. Cookson, Pyroptosis and host cell death responses duringSalmonella infection, Cell. Microbiol. 9 (2007) 2562–2570.
[45] F. Martinon, J. Tschopp, Inflammatory caspases and inflammasomes: masterswitches of inflammation, Cell Death Differ. 14 (2007) 10–22.
[46] S.B. Willingham, D.T. Bergstralh, W. O'Connor, A.C. Morrison, D.J. Taxman, J.A.Duncan, S. Barnoy, M.M. Venkatesan, R.A. Flavell, M. Deshmukh, H.M. Hoffman,J.P. Ting, Microbial pathogen-induced necrotic cell death mediated by theinflammasome components CIAS1/cryopyrin/NLRP3 and ASC, Cell Host Microbe.2 (2007) 147–159.
[47] M.E. Bianchi, DAMPs, PAMPs and alarmins: all we need to know about danger,J. Leukoc. Biol. 81 (2007) 1–5.
[48] A. Rubartelli, M.T. Lotze, Inside, outside, upside down: damage-associatedmolecular-pattern molecules (DAMPs) and redox, Trends Immunol. 28 (2007)429–436.
[49] W. Nickel, Themystery of nonclassical protein secretion. A current view on cargoproteins and potential export routes, Eur. J. Biochem. 270 (2003) 2109–2119.
[50] S.Y. Seong, P. Matzinger, Hydrophobicity: an ancient damage-associatedmolecular pattern that initiates innate immune responses, Nat. Rev. Immunol.4 (2004) 469–478.
[51] M.T. Lotze, H.J. Zeh, A. Rubartelli, L.J. Sparvero, A.A. Amoscato, N.R. Washburn,M.E. Devera, X. Liang, M. Tor, T. Billiar, The grateful dead: damage-associatedmolecular pattern molecules and reduction/oxidation regulate immunity,Immunol. Rev. 220 (2007) 60–81.
[52] C. Tosch, M. Geist, C. Ledoux, C. Ziller-Remi, S. Paul, P. Erbs, N. Corvaia, P. VonHoegen, J.M. Balloul, H. Haegel, Adenovirus-mediated gene transfer of pathogen-
68 A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2009) 53–71
associated molecular patterns for cancer immunotherapy, Cancer Gene. Ther. 16(2009) 310–319.
[53] K.A. Scheibner, M.A. Lutz, S. Boodoo, M.J. Fenton, J.D. Powell, M.R. Horton,Hyaluronan fragments act as an endogenous danger signal by engaging TLR2,J. Immunol. 177 (2006) 1272–1281.
[54] N. Casares, M.O. Pequignot, A. Tesniere, F. Ghiringhelli, S. Roux, N. Chaput, E.Schmitt, A. Hamai, S. Hervas-Stubbs, M. Obeid, F. Coutant, D. Metivier, E. Pichard,P. Aucouturier, G. Pierron, C. Garrido, L. Zitvogel, G. Kroemer, Caspase-dependentimmunogenicity of doxorubicin-induced tumor cell death, J. Exp. Med. 202(2005) 1691–1701.
[55] R. Spisek, M.V. Dhodapkar, Towards a better way to die with chemotherapy: roleof heat shock protein exposure on dying tumor cells, Cell Cycle 6 (2007)1962–1965.
[56] R. Spisek, A. Charalambous, A. Mazumder, D.H. Vesole, S. Jagannath, M.V.Dhodapkar, Bortezomib enhances dendritic cell (DC)-mediated induction ofimmunity to human myeloma via exposure of cell surface heat shock protein90 on dying tumor cells: therapeutic implications, Blood 109 (2007)4839–4845.
[57] N. Mesaeli, C. Phillipson, Impaired p53 expression, function, and nuclearlocalization in calreticulin-deficient cells, Mol. Biol. Cell. 15 (2004)1862–1870.
[58] K.L. Shaffer, A. Sharma, E.L. Snapp, R.S. Hegde, Regulation of proteincompartmentalization expands the diversity of protein function, Dev. Cell. 9(2005) 545–554.
[59] L. Ellgaard, E.M. Frickel, Calnexin, calreticulin, and ERp57: teammates inglycoprotein folding, Cell. Biochem. Biophys. 39 (2003) 223–247.
[60] J.C. Solheim, Class I MHC molecules: assembly and antigen presentation,Immunol. Rev. 172 (1999) 11–19.
[61] S.M. Goicoechea, J.E. Murphy-Ullrich, Cell surface calreticulin: role in signalingthrombospondin anti-adhesive activity, in: P. Eggleton, M. Michalak (Eds.),Calreticulin, 2nd Ed, Springer, Berlin, Molecular Biology Intelligence Unit, 2003,pp. 193–204.
[62] M.L. Johansen, E.J. Brown, Dual regulation of SIRPalpha phosphorylation byintegrins and CD47, J. Biol. Chem. 282 (2007) 24219–24230.
[64] A.W. Orr, C.A. Elzie, D.F. Kucik, J.E. Murphy-Ullrich, Thrombospondin signalingthrough the calreticulin/LDL receptor-related protein co-complex stimulatesrandom and directed cell migration, J. Cell. Sci. 116 (2003) 2917–2927.
[65] C.A. Ogden, A. deCathelineau, P.R. Hoffmann, D. Bratton, B. Ghebrehiwet, V.A.Fadok, P.M. Henson, C1q and mannose binding lectin engagement of cell surfacecalreticulin and CD91 initiates macropinocytosis and uptake of apoptotic cells,J. Exp. Med. 194 (2001) 781–795.
[66] W.J. Storkus, L.D. Falo Jr, A ‘good death’ for tumor immunology, Nat. Med. 13(2007) 28–30.
[67] G. Zeng, M.E. Aldridge, X. Tian, D. Seiler, X. Zhang, Y. Jin, J. Rao,W. Li, D. Chen, M.P.Langford, C. Duggan, A.S. Belldegrun, S.M. Dubinett, Dendritic cell surfacecalreticulin is a receptor for NY-ESO-1: direct interactions between tumor-associated antigen and the innate immune system, J. Immunol. 177 (2006)3582–3589.
[68] M. Schroder, R.J. Kaufman, The mammalian unfolded protein response, Annu.Rev. Biochem. 74 (2005) 739–789.
[69] P. Eggleton, M.Michalak, Introduction to calreticulin, in: P. Eggleton, M.Michalak(Eds.), Calreticulin, 2nd Ed, Springer, Berlin, Molecular Biology Intelligence Unit,2003, pp. 1–8.
[70] D.P. McCauliffe, F.A. Lux, T.S. Lieu, I. Sanz, J. Hanke, M.M. Newkirk, L.L. Bachinski,Y. Itoh, M.J. Siciliano, M. Reichlin, et al., Molecular cloning, expression, andchromosome 19 localization of a human Ro/SS-A autoantigen, J. Clin. Invest. 85(1990) 1379–1391.
[71] L.A. Rokeach, J.A. Haselby, J.F. Meilof, R.J. Smeenk, T.R. Unnasch, B.M. Greene, S.O.Hoch, Characterization of the autoantigen calreticulin, J. Immunol. 147 (1991)3031–3039.
[72] S. Papp, M. Opas, Sub-cellular distribution of calreticulin, in: P. Eggleton, M.Michalak (Eds.), Calreticulin, 2nd Ed, Springer, Berlin, Molecular BiologyIntelligence Unit, 2003, pp. 38–48.
[73] F.A. Arosa, O. de Jesus, G. Porto, A.M. Carmo, M. de Sousa, Calreticulin isexpressed on the cell surface of activated human peripheral blood T lymphocytesin association with major histocompatibility complex class I molecules, J. Biol.Chem. 274 (1999) 16917–16922.
[74] D. Lanneau, M. Brunet, E. Frisan, E. Solary, M. Fontenay, C. Garrido, Heat shockproteins: essential proteins for apoptosis regulation, J. Cell. Mol. Med. 12 (2008)743–761.
[75] A. Parcellier, S. Gurbuxani, E. Schmitt, E. Solary, C. Garrido, Heat shock proteins,cellular chaperones that modulate mitochondrial cell death pathways, Biochem.Biophys. Res. Commun. 304 (2003) 505–512.
[76] C. Garrido, M. Brunet, C. Didelot, Y. Zermati, E. Schmitt, G. Kroemer, Heat shockproteins 27 and 70: anti-apoptotic proteins with tumorigenic properties, CellCycle 5 (2006) 2592–2601.
[77] N. Arispe, M. Doh, O. Simakova, B. Kurganov, A. De Maio, Hsc70 and Hsp70interact with phosphatidylserine on the surface of PC12 cells resulting in adecrease of viability, FASEB. J. 18 (2004) 1636–1645.
[78] L.J. Foster, C.L. De Hoog, M. Mann, Unbiased quantitative proteomics of lipid raftsreveals high specificity for signaling factors, Proc. Natl. Acad. Sci. U. S. A. 100(2003) 5813–5818.
[79] N. Li, A.R. Shaw, N. Zhang, A. Mak, L. Li, Lipid raft proteomics: analysis of in-solution digest of sodium dodecyl sulfate-solubilized lipid raft proteins by liquid
chromatography-matrix-assisted laser desorption/ionization tandem massspectrometry, Proteomics 4 (2004) 3156–3166.
[80] M. Gehrmann, G. Liebisch, G. Schmitz, R. Anderson, C. Steinem, A. De Maio, G.Pockley, G. Multhoff, Tumor-specific Hsp70 plasma membrane localization isenabled by the glycosphingolipid Gb3, PLoS ONE 3 (2008) e1925.
[81] H. Udono, P.K. Srivastava, Comparison of tumor-specific immunogenicities ofstress-induced proteins gp96, hsp90, and hsp70, J. Immunol. 152 (1994)5398–5403.
[82] J. Bartunkova, R. Spisek, Impact of tumour cell death on the activation of anti-tumour immune response, in: D.V. Krysko, P. Vandenabeele (Eds.), Phagocytosisof Dying Cells, Springer Science + Business Media B.V., Berlin, 2009,pp. 347–370.
[83] A.D. Doody, J.T. Kovalchin, M.A. Mihalyo, A.T. Hagymasi, C.G. Drake, A.J. Adler,Glycoprotein 96 can chaperone both MHC class I-and class II-restricted epitopesfor in vivo presentation, but selectively primes CD8+ T cell effector function,J. Immunol. 172 (2004) 6087–6092.
[84] H. Schild, D. Arnold-Schild, E. Lammert, H.G. Rammensee, Stress proteins andimmunity mediated by cytotoxic T lymphocytes, Curr. Opin. Immunol. 11 (1999)109–113.
[85] R.J. Binder, N.E. Blachere, P.K. Srivastava, Heat shock protein-chaperonedpeptides but not free peptides introduced into the cytosol are presentedefficiently by major histocompatibility complex I molecules, J. Biol. Chem. 276(2001) 17163–17171.
[86] A. Asea, M. Rehli, E. Kabingu, J.A. Boch, O. Bare, P.E. Auron, M.A. Stevenson, S.K.Calderwood, Novel signal transduction pathway utilized by extracellularHSP70: role of toll-like receptor (TLR) 2 and TLR4, J. Biol. Chem. 277 (2002)15028–15034.
[87] A. Asea, S.K. Kraeft, E.A. Kurt-Jones, M.A. Stevenson, L.B. Chen, R.W. Finberg, G.C.Koo, S.K. Calderwood, HSP70 stimulates cytokine production through a CD14-dependant pathway, demonstrating its dual role as a chaperone and cytokine,Nat. Med. 6 (2000) 435–442.
[88] Y. Wang, C.G. Kelly, M. Singh, E.G. McGowan, A.S. Carrara, L.A. Bergmeier, T.Lehner, Stimulation of Th1-polarizing cytokines, C-C chemokines, maturation ofdendritic cells, and adjuvant function by the peptide binding fragment of heatshock protein 70, J. Immunol. 169 (2002) 2422–2429.
[89] T. Lehner, Y. Wang, T. Whittall, E. McGowan, C.G. Kelly, M. Singh, Functionaldomains of HSP70 stimulate generation of cytokines and chemokines,maturation of dendritic cells and adjuvanticity, Biochem. Soc. Trans. 32 (2004)629–632.
[90] H. Singh-Jasuja, R.E. Toes, P. Spee, C. Munz, N. Hilf, S.P. Schoenberger, P. Ricciardi-Castagnoli, J. Neefjes, H.G. Rammensee, D. Arnold-Schild, H. Schild, Cross-presentation of glycoprotein 96-associated antigens on major histocompatibilitycomplex class I molecules requires receptor-mediated endocytosis, J. Exp. Med.191 (2000) 1965–1974.
[91] I. Waldhauer, A. Steinle, NK cells and cancer immunosurveillance, Oncogene 27(2008) 5932–5943.
[92] H.D. Hickman-Miller, W.H. Hildebrand, The immune response under stress: therole of HSP-derived peptides, Trends Immunol. 25 (2004) 427–433.
[93] C. Botzler, G. Li, R.D. Issels, G. Multhoff, Definition of extracellular localizedepitopes of Hsp70 involved in an NK immune response, Cell Stress Chaperones 3(1998) 6–11.
[94] G. Multhoff, L. Mizzen, C.C. Winchester, C.M. Milner, S. Wenk, G. Eissner, H.H.Kampinga, B. Laumbacher, J. Johnson, Heat shock protein 70 (Hsp70) stimulatesproliferation and cytolytic activity of natural killer cells, Exp. Hematol. 27 (1999)1627–1636.
[95] J. Dai, B. Liu, M.M. Caudill, H. Zheng, Y. Qiao, E.R. Podack, Z. Li, Cell surfaceexpression of heat shock protein gp96 enhances cross-presentation of cellularantigens and the generation of tumor-specific T cell memory, Cancer Immun. 3(2003) 1.
[96] D. Masse, F. Ebstein, G. Bougras, J. Harb, K. Meflah, M. Gregoire, Increasedexpression of inducible HSP70 in apoptotic cells is correlated with their efficacyfor antitumor vaccine therapy, Int. J. Cancer 111 (2004) 575–583.
[97] H. Shi, T. Cao, J.E. Connolly, L. Monnet, L. Bennett, S. Chapel, C. Bagnis, P. Mannoni,J. Davoust, A.K. Palucka, J. Banchereau, Hyperthermia enhances CTL cross-priming, J. Immunol. 176 (2006) 2134–2141.
[98] H. Feng, Y. Zeng, M.W. Graner, A. Likhacheva, E. Katsanis, Exogenous stressproteins enhance the immunogenicity of apoptotic tumor cells and stimulateantitumor immunity, Blood 101 (2003) 245–252.
[99] M. Milani, T. Rzymski, H.R. Mellor, L. Pike, A. Bottini, D. Generali, A.L. Harris, Therole of ATF4 stabilization and autophagy in resistance of breast cancer cellstreated with bortezomib, Cancer Res. 69 (2009) 4415–4423.
[100] D.M. Schewe, J.A. Aguirre-Ghiso, Inhibition of eIF2alpha dephosphorylationmaximizes bortezomib efficiency and eliminates quiescent multiple myelomacells surviving proteasome inhibitor therapy, Cancer Res. 69 (2009) 1545–1552.
[101] M.V. Dhodapkar, K.M. Dhodapkar, Z. Li, Role of chaperones and FcgammaR inimmunogenic death, Curr. Opin. Immunol. 20 (2008) 512–517.
[102] A. Menoret, Y. Patry, C. Burg, J. Le Pendu, Co-segregation of tumor immunoge-nicity with expression of inducible but not constitutive hsp70 in rat coloncarcinomas, J. Immunol. 155 (1995) 740–747.
[103] A. Melcher, S. Todryk, N. Hardwick, M. Ford, M. Jacobson, R.G. Vile, Tumorimmunogenicity is determined by the mechanism of cell death via induction ofheat shock protein expression, Nat. Med. 4 (1998) 581–587.
[104] R. Palumbo, M. Sampaolesi, F. De Marchis, R. Tonlorenzi, S. Colombetti, A.Mondino, G. Cossu, M.E. Bianchi, Extracellular HMGB1, a signal of tissue damage,induces mesoangioblast migration and proliferation, J. Cell. Biol. 164 (2004)441–449.
69A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2010) 53–71
[105] S. Guazzi, A. Strangio, A.T. Franzi, M.E. Bianchi, HMGB1, an architecturalchromatin protein and extracellular signalling factor, has a spatially andtemporally restricted expression pattern in mouse brain, Gene. Expr. Patterns3 (2003) 29–33.
[106] T. Vaccari, M. Beltrame, S. Ferrari, M.E. Bianchi, Hmg4, a new member of theHmg1/2 gene family, Genomics 49 (1998) 247–252.
[107] L. Ronfani, M. Ferraguti, L. Croci, C.E. Ovitt, H.R. Scholer, G.G. Consalez, M.E.Bianchi, Reduced fertility and spermatogenesis defects in mice lackingchromosomal protein Hmgb2, Development 128 (2001) 1265–1273.
[108] L. Apetoh, F. Ghiringhelli, A. Tesniere, M. Obeid, C. Ortiz, A. Criollo, G. Mignot,M.C. Maiuri, E. Ullrich, P. Saulnier, H. Yang, S. Amigorena, B. Ryffel, F.J. Barrat, P.Saftig, F. Levi, R. Lidereau, C. Nogues, J.P. Mira, A. Chompret, V. Joulin, F. Clavel-Chapelon, J. Bourhis, F. Andre, S. Delaloge, T. Tursz, G. Kroemer, L. Zitvogel,Toll-like receptor 4-dependent contribution of the immune system toanticancer chemotherapy and radiotherapy, Nat. Med. 13 (2007) 1050–1059.
[109] C.W. Bell, W. Jiang, C.F. Reich, 3rdD.S. , Pisetsky, The extracellular release ofHMGB1 during apoptotic cell death, Am. J. Physiol. Cell. Physiol. 291 (2006)C1318–C1325.
[110] J. Tian, A. Avalos, S. Mao, B. Chen, K. Senthil, H Wu, et al., Toll-like receptor 9-dependent activation by DNA-containing immune complexes is mediated byHMGB1 and RAGE, Nat. Immunol. 8 (2007) 487–496.
[111] C. Bell, W. Jiang, C.R. III, D. Pisetsky, The extracellular release of HMGB1 duringapoptotic cell death, Am. J. Physiol. Cell. Physiol. 291 (2006) C1318–C1325.
[112] J.S. Park, F. Gamboni-Robertson, Q. He, D. Svetkauskaite, J.Y. Kim, D. Strassheim,J.W. Sohn, S. Yamada, I. Maruyama, A. Banerjee, A. Ishizaka, E. Abraham, Highmobility group box 1 protein interacts with multiple Toll-like receptors, Am. J.Physiol. Cell. Physiol. 290 (2006) C917–C924.
[113] P. Rovere-Querini, A. Capobianco, P. Scaffidi, B. Valentinis, F. Catalanotti, M.Giazzon, I.E. Dumitriu, S. Muller, M. Iannacone, C. Traversari, M.E. Bianchi, A.A.Manfredi, HMGB1 is an endogenous immune adjuvant released by necrotic cells,EMBO Rep. 5 (2004) 825–830.
[114] D.V. Krysko, G. Denecker, N. Festjens, S. Gabriels, E. Parthoens, K. D'Herde, P.Vandenabeele, Macrophages use different internalization mechanisms to clearapoptotic and necrotic cells, Cell Death Differ. 13 (2006) 2011–2022.
[115] G. Brouckaert, M. Kalai, D.V. Krysko, X. Saelens, D. Vercammen, M. Ndlovu, G.Haegeman, K. D'Herde, P. Vandenabeele, Phagocytosis of necrotic cells bymacrophages is phosphatidylserine dependent and does not induce inflamma-tory cytokine production, Mol. Biol. Cell. 15 (2004) 1089–1100.
[116] J.S. Park, D. Svetkauskaite, Q. He, J.Y. Kim, D. Strassheim, A. Ishizaka, E. Abraham,Involvement of toll-like receptors 2 and 4 in cellular activation by high mobilitygroup box 1 protein, J. Biol. Chem. 279 (2004) 7370–7377.
[117] D. Yang, Q. Chen, H. Yang, K.J. Tracey, M. Bustin, J.J. Oppenheim, High mobilitygroup box-1 protein induces the migration and activation of human dendriticcells and acts as an alarmin, J. Leukoc. Biol. 81 (2007) 59–66.
[118] V.V. Orlova, E.Y. Choi, C. Xie, E. Chavakis, A. Bierhaus, E. Ihanus, C.M. Ballantyne,C.G. Gahmberg, M.E. Bianchi, P.P. Nawroth, T. Chavakis, A novel pathway ofHMGB1-mediated inflammatory cell recruitment that requires Mac-1-integrin,EMBO J. 26 (2007) 1129–1139.
[119] S. Ivanov, A.M. Dragoi, X. Wang, C. Dallacosta, J. Louten, G. Musco, G. Sitia, G.S.Yap, Y.Wan, C.A. Biron, M.E. Bianchi, H.Wang,W.M. Chu, A novel role for HMGB1in TLR9-mediated inflammatory responses to CpG-DNA, Blood 110 (2007)1970–1981.
[120] J. Tian, A.M. Avalos, S.Y. Mao, B. Chen, K. Senthil, H. Wu, P. Parroche, S. Drabic, D.Golenbock, C. Sirois, J. Hua, L.L. An, L. Audoly, G. La Rosa, A. Bierhaus, P.Naworth, A. Marshak-Rothstein, M.K. Crow, K.A. Fitzgerald, E. Latz, P.A. Kiener,A.J. Coyle, Toll-like receptor 9-dependent activation by DNA-containingimmune complexes is mediated by HMGB1 and RAGE, Nat. Immunol.8 (2007) 487–496.
[121] S. Basu, R.J. Binder, R. Suto, K.M. Anderson, P.K. Srivastava, Necrotic but notapoptotic cell death releases heat shock proteins, which deliver a partialmaturation signal to dendritic cells and activate the NF-kappa B pathway, Int.Immunol. 12 (2000) 1539–1546.
[122] R.J. Binder, R. Vatner, P. Srivastava, The heat-shock protein receptors: someanswers and more questions, Tissue Antigens 64 (2004) 442–451.
[123] Y. Shi, J.E. Evans, K.L. Rock, Molecular identification of a danger signal that alertsthe immune system to dying cells, Nature 425 (2003) 516–521.
[124] M.F. Tsan, Toll-like receptors, inflammation and cancer, Semin. Cancer Biol. 16(2006) 32–37.
[125] Y. Shi, W. Zheng, K.L. Rock, Cell injury releases endogenous adjuvants thatstimulate cytotoxic T cell responses, Proc. Natl. Acad. Sci. U. S. A. 97 (2000)14590–14595.
[126] Y. Shi, K.L. Rock, Cell death releases endogenous adjuvants that selectivelyenhance immune surveillance of particulate antigens, Eur. J. Immunol. 32 (2002)155–162.
[128] F. Martinon, V. Petrilli, A. Mayor, A. Tardivel, J. Tschopp, Gout-associated uric acidcrystals activate the NALP3 inflammasome, Nature 440 (2006) 237–241.
[129] K.A. Cavassani, M. Ishii, H. Wen, M.A. Schaller, P.M. Lincoln, N.W. Lukacs, C.M.Hogaboam, S.L. Kunkel, TLR3 is an endogenous sensor of tissue necrosis duringacute inflammatory events, J. Exp. Med. 205 (2008) 2609–2621.
[130] K.J. Ishii, K. Suzuki, C. Coban, F. Takeshita, Y. Itoh, H. Matoba, L.D. Kohn, D.M.Klinman, GenomicDNA , released by dying cells induces the maturation of APCs,J. Immunol. 167 (2001) 2602–2607.
[131] D.V. Krysko, L. Leybaert, P. Vandenabeele, K. D'Herde, Gap junctions and thepropagation of cell survival and cell death signals, Apoptosis 10 (2005)459–469.
[132] E. Decrock, M. Vinken, E. De Vuyst, D.V. Krysko, K. D'Herde, T. Vanhaecke, P.Vandenabeele, V. Rogiers, L. Leybaert, Connexin-related signaling in cell death:to live or let die? Cell Death Differ. 16 (2009) 524–536.
[133] M. Skoberne, A.S. Beignon, N. Bhardwaj, Danger signals: a time and spacecontinuum, Trends Mol. Med. 10 (2004) 251–257.
[134] F. Wilkin, X. Duhant, C. Bruyns, N. Suarez-Huerta, J.M. Boeynaems, B. Robaye, TheP2Y11 receptor mediates the ATP-induced maturation of human monocyte-derived dendritic cells, J. Immunol. 166 (2001) 7172–7177.
[135] S. Yamasaki, M. Matsumoto, O. Takeuchi, T. Matsuzawa, E. Ishikawa, M. Sakuma,H. Tateno, J. Uno, J. Hirabayashi, Y. Mikami, K. Takeda, S. Akira, T. Saito, C-typelectin Mincle is an activating receptor for pathogenic fungus, Malassezia, Proc.Natl. Acad. Sci. U. S. A. 106 (2009) 1897–1902.
[136] C.J. Chen, H. Kono, D. Golenbock, G. Reed, S. Akira, K.L. Rock, Identification of akey pathway required for the sterile inflammatory response triggered by dyingcells, Nat. Med. 13 (2007) 851–856.
[137] T. Eigenbrod, J.H. Park, J. Harder, Y. Iwakura, G. Nunez, Cutting edge: criticalrole for mesothelial cells in necrosis-induced inflammation through therecognition of IL-1 alpha released from dying cells, J. Immunol. 181 (2008)8194–8198.
[138] T. Vanden Berghe, M. Kalai, G. Denecker, A. Meeus, X. Saelens, P. Vandenabeele,Necrosis is associated with IL-6 production but apoptosis is not, Cell. Signal. 18(2006) 328–335.
[139] X. Saelens, N. Festjens, E. Parthoens, I. Vanoverberghe, M. Kalai, F. van Kuppeveld,P. Vandenabeele, Protein synthesis persists during necrotic cell death, J. Cell. Biol.168 (2005) 545–551.
[140] Y. Li, L.X. Wang, P. Pang, C. Twitty, B.A. Fox, S. Aung, W.J. Urba, H.M. Hu, Cross-presentation of tumor associated antigens through tumor-derived autophago-somes, Autophagy 5 (2009) 576–577.
[141] H.W. Virgin, B. Levine, Autophagy genes in immunity, Nat. Immunol. 10 (2009)461–470.
[142] J.M. Vyas, A.G. Van der Veen, H.L. Ploegh, The known unknowns of antigenprocessing and presentation, Nat. Rev. Immunol. 8 (2008) 607–618.
[143] L. English, M. Chemali, J. Duron, C. Rondeau, A. Laplante, D. Gingras, D. Alexander,D. Leib, C. Norbury, R. Lippe, M. Desjardins, Autophagy enhances thepresentation of endogenous viral antigens on MHC class I molecules duringHSV-1 infection, Nat. Immunol. 10 (2009) 480–487.
[144] Y. Li, L.X. Wang, G. Yang, F. Hao, W.J. Urba, H.M. Hu, Efficient cross-presentationdepends on autophagy in tumor cells, Cancer Res. 68 (2008) 6889–6895.
[145] L. Baricault, J.A. Fransen, M. Garcia, C. Sapin, P. Codogno, L.A. Ginsel, G. Trugnan,Rapid sequestration of DPP IV/CD26 and other cell surface proteins in anautophagic-like compartment in Caco-2 cells treated with forskolin, J. Cell. Sci.108 (Pt 5) (1995) 2109–2121.
[146] G. Petrovski, G. Zahuczky, K. Katona, G. Vereb, W. Martinet, Z. Nemes, W. Bursch,L. Fesus, Clearance of dying autophagic cells of different origin by professionaland non-professional phagocytes, Cell Death Differ. 14 (2007) 1117–1128.
[147] M.A. Mellen, E.J. de la Rosa, P. Boya, The autophagic machinery is necessary forremoval of cell corpses from the developing retinal neuroepithelium, Cell DeathDiffer. 15 (2008) 1279–1290.
[148] X. Qu, Z. Zou, Q. Sun, K. Luby-Phelps, P. Cheng, R.N. Hogan, C. Gilpin, B. Levine,Autophagy gene-dependent clearance of apoptotic cells during embryonicdevelopment, Cell 128 (2007) 931–946.
[149] G. Petrovski, G. Zahuczky, G. Majai, L. Fesus, Phagocytosis of cells dying throughautophagy evokes a pro-inflammatory response in macrophages, Autophagy 3(2007) 509–511.
[150] M. Uhl, O. Kepp, H. Jusforgues-Saklani, J.M. Vicencio, G. Kroemer, M.L. Albert,Autophagy within the antigen donor cell facilitates efficient antigen cross-priming of virus-specific CD8(+) T cells, Cell Death Differ. 16 (2009) 991–1005.
[151] G.P. Dunn, A.T. Bruce, H. Ikeda, L.J. Old, R.D. Schreiber, Cancer immunoediting:from immunosurveillance to tumor escape, Nat. Immunol. 3 (2002) 991–998.
[152] M.W. Teng, J.B. Swann, C.M. Koebel, R.D. Schreiber, M.J. Smyth, Immune-mediated dormancy: an equilibrium with cancer, J. Leukoc. Biol. 84 (2008)988–993.
[153] G.P. Dunn, L.J. Old, R.D. Schreiber, The three Es of cancer immunoediting, Annu.Rev. Immunol. 22 (2004) 329–360.
[154] S. Lucas, P.G. Coulie, About human tumor antigens to be used in immunotherapy,Semin. Immunol. 20 (2008) 301–307.
[155] J. Copier, A.G. Dalgleish, C.M. Britten, L.H. Finke, G. Gaudernack, S. Gnjatic, K.Kallen, R. Kiessling, M. Schuessler-Lenz, H. Singh, J. Talmadge, H. Zwierzina, L.Hakansson, Improving the efficacy of cancer immunotherapy, Eur. J. Cancer 45(2009) 1424–1431.
[156] F.M. Marincola, E.M. Jaffee, D.J. Hicklin, S. Ferrone, Escape of human solid tumorsfrom T-cell recognition: molecular mechanisms and functional significance, Adv.Immunol. 74 (2000) 181–273.
[157] I. Algarra, T. Cabrera, F. Garrido, The HLA crossroad in tumor immunology, Hum.Immunol. 61 (2000) 65–73.
[158] V. Bronte, P. Zanovello, Regulation of immune responses by L-argininemetabolism, Nat. Rev. Immunol. 5 (2005) 641–654.
[159] R.A. Weinberg, Crowd Control: Tumor Immunology and Immunotherapy, TheBiology of Cancer, Garland Science, Taylor and Francis Group, LLC, New York,2007, pp. 655–724.
[160] J.P. Medema, J. de Jong, L.T. Peltenburg, E.M. Verdegaal, A. Gorter, S.A. Bres, K.L.Franken, M. Hahne, J.P. Albar, C.J. Melief, R. Offringa, Blockade of the granzyme B/perforin pathway through overexpression of the serine protease inhibitor PI-9/
70 A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2009) 53–71
SPI-6 constitutes a mechanism for immune escape by tumors, Proc. Natl. Acad.Sci. U. S. A. 98 (2001) 11515–11520.
[162] S. Chouaib, F. Meslin, J. Thiery, F. Mami-Chouaib, Tumor resistance to specificlysis: amajor hurdle for successful immunotherapy of cancer, Clin. Immunol. 130(2009) 34–40.
[163] G. Lizee, M.A. Cantu, P. Hwu, Less yin, more yang: confronting the barriers tocancer immunotherapy, Clin. Cancer Res. 13 (2007) 5250–5255.
[164] D.I. Gabrilovich, Combination of chemotherapy and immunotherapy for cancer:a paradigm revisited, Lancet Oncol. 8 (2007) 2–3.
[165] R.A. Lake, B.W. Robinson, Immunotherapy and chemotherapy—a practicalpartnership, Nat. Rev. Cancer 5 (2005) 397–405.
[166] S.P. Hussain, C.C. Harris, Inflammation and cancer: an ancient link with novelpotentials, Int. J. Cancer 121 (2007) 2373–2380.
[167] H.S. Schwartz, G.B. Grindey, Adriamycin and daunorubicin: a comparison ofantitumor activities and tissue uptake in mice following immunosuppression,Cancer Res. 33 (1973) 1837–1844.
[168] L. Lothstein, M. Israel, T.W. Sweatman, Anthracycline drug targeting: cytoplas-mic versus nuclear—a fork in the road, Drug Resist. Updat. 4 (2001) 169–177.
[169] D.E. Thurston, Chemistry and Pharmacology of Anticancer Drugs, CRC Press,New York, 2007.
[170] D. Nielsen, C. Maare, T. Skovsgaard, Cellular resistance to anthracyclines,Gen. Pharmacol. 27 (1996) 251–255.
[171] M. Igawa, H. Kadena, M. Ueda, T. Usui, Association between patient character-istics and treatment history, and toxicity associated with methotrexate,vinblastine, adriamycin and cisplatin (M-VAC) for advanced urothelial cancer,Br. J. Urol. 73 (1994) 263–267.
[172] Q. Zeidan, M. Strauss, N. Porras, G. Anselmi, Differential long-term subcellularresponses in heart and liver to adriamycin stress. Exogenous L-carnitine cardiacand hepatic protection, J. Submicrosc. Cytol. Pathol. 34 (2002) 315–321.
[173] A. Di Donato, G.M. Ghiggeri, M. Di Duca, E. Jivotenko, R. Acinni, J. Campolo, F.Ginevri, R. Gusmano, Lysyl oxidase expression and collagen cross-linking duringchronic adriamycin nephropathy, Nephron 76 (1997) 192–200.
[174] P.K. Singal, C.M. Deally, L.E. Weinberg, Subcellular effects of adriamycin in theheart: a concise review, J. Mol. Cell. Cardiol. 19 (1987) 817–828.
[175] E.A. Lefrak, J. Pitha, S. Rosenheim, J.A. Gottlieb, A clinicopathologic analysis ofadriamycin cardiotoxicity, Cancer 32 (1973) 302–314.
[176] P.K. Singal, N. Iliskovic, Doxorubicin-induced cardiomyopathy, N. Engl. J. Med.339 (1998) 900–905.
[177] E.R. Hayek, E. Speakman, E. Rehmus, Acute doxorubicin cardiotoxicity, N. Engl. J.Med. 352 (2005) 2456–2457.
[178] C.D. Venkatakrishnan, A.K. Tewari, L. Moldovan, A.J. Cardounel, J.L. Zweier, P.Kuppusamy, G. Ilangovan, Heat shock protects cardiac cells from doxorubicin-induced toxicity by activating p38 MAPK and phosphorylation of small heatshock protein 27, Am. J. Physiol. Heart Circ. Physiol. 291 (2006) H2680–2691.
[179] N. Nozaki, T. Shishido, Y. Takeishi, I. Kubota, Modulation of doxorubicin-inducedcardiac dysfunction in toll-like receptor-2-knockout mice, Circulation 110(2004) 2869–2874.
[180] A. Riad, S. Bien, M. Gratz, F. Escher, D.Westermann, M.M. Heimesaat, S. Bereswill,T. Krieg, S.B. Felix, H.P. Schultheiss, H.K. Kroemer, C. Tschope, Toll-like receptor-4deficiency attenuates doxorubicin-induced cardiomyopathy inmice, Eur. J. HeartFail. 10 (2008) 233–243.
[181] A. Chaudhuri, P.O. Behan, Mitoxantrone trial in multiple sclerosis, Lancet 361(2003) 1133–1134 author reply 1134.
[182] A. Broomand, E. Jerremalm, J. Yachnin, H. Ehrsson, F. Elinder, Oxaliplatinneurotoxicity—no general ion channel surface-charge effect, J. Negat. ResultsBiomed. 8 (2009) 2.
[183] A.P. Castano, P. Mroz, M.R. Hamblin, Photodynamic therapy and anti-tumourimmunity, Nat. Rev. Cancer 6 (2006) 535–545.
[184] A.A. Argyriou, G. Iconomou, H.P. Kalofonos, Bortezomib-induced peripheralneuropathy in multiple myeloma: a comprehensive review of the literature,Blood 112 (2008) 1593–1599.
[185] B.C. Bockle, M. Baltaci, W. Weyrer, N.T. Sepp, Bortezomib-induced lupuserythematosus tumidus, Oncologist 14 (2009) 637–639.
[186] P. Agostinis, E. Buytaert, H. Breyssens, N. Hendrickx, Regulatory pathways inphotodynamic therapy induced apoptosis, Photochem. Photobiol. Sci. 3 (2004)721–729.
[187] P. Agostinis, A. Vantieghem, W. Merlevede, P.A. de Witte, Hypericin in cancertreatment: more light on the way, Int. J. Biochem. Cell. Biol. 34 (2002) 221–241.
[188] E. Buytaert, M. Dewaele, P. Agostinis, Molecular effectors of multiple cell deathpathways initiated by photodynamic therapy, Biochim. Biophys. Acta 1776(2007) 86–107.
[189] A.C.E. Moor, B. Ortel, T. Hasan, Mechanisms of photodynamic therapy, in: T.Patrice (Ed.), Photodynamic Therapy, The Royal Society of Chemistry,Cambridge, 2003, pp. 19–57.
[190] N. Rousset, L. Bourre, S. Thibaud, Sensitisers in photodynamic therapy, in: T.Patrice (Ed.), Photodynamic Therapy, The Royal Society of Chemistry,Cambridge, 2003, pp. 19–57.
[191] M.R. Hamblin, A.P. Castano, P. Mroz, Combination immunotherapy andphotodynamic therapy for cancer, in: R. Waynant, D.B. Tata (Eds.), Proceedingsof Light-Activated Tissue Regeneration and Therapy Conference, SpringerScience + Business Media, Berlin, 2008, pp. 99–113.
[192] E. Kabingu, A.R. Oseroff, G.E. Wilding, S.O. Gollnick, Enhanced systemic immunereactivity to a basal cell carcinoma associated antigen following photodynamictherapy, Clin. Cancer Res. 15 (2009) 4460–4466.
[193] M. Korbelik, Complement upregulation in photodynamic therapy-treatedtumors: role of Toll-like receptor pathway and NFkappaB, Cancer Lett. 281(2009) 232–238.
[194] T.J. Fry, J.L. Shand, M. Milliron, S.K. Tasian, C.L. Mackall, Antigen loading of DCswith irradiated apoptotic tumor cells induces improved anti-tumor immunitycompared to other approaches, Cancer Immunol. Immunother. 58 (2009)1257–1264.
[195] S.O. Gollnick, L. Vaughan, B.W. Henderson, Generation of effective antitumorvaccines using photodynamic therapy, Cancer Res. 62 (2002) 1604–1608.
[196] B.K. Al-Ramadi, M.J. Fernandez-Cabezudo, H. El-Hasasna, S. Al-Salam, S. Attoub,D. Xu, S. Chouaib, Attenuated bacteria as effectors in cancer immunotherapy,Ann. N.Y. Acad. Sci. 1138 (2008) 351–357.
[197] A.V. Kozlov, J.C. Duvigneau, I. Miller, S. Nurnberger, B. Gesslbauer, A. Kungl, W.Ohlinger, R.T. Hartl, L. Gille, K. Staniek, W. Gregor, S. Haindl, H. Redl, Endotoxincauses functional endoplasmic reticulum failure, possibly mediated by mito-chondria, Biochim. Biophys. Acta 1792 (2009) 521–530.
[198] H. Kazama, J.E. Ricci, J.M. Herndon, G. Hoppe, D.R. Green, T.A. Ferguson, Inductionof immunological tolerance by apoptotic cells requires caspase-dependentoxidation of high-mobility group box-1 protein, Immunity 29 (2008) 21–32.
[199] J.F. Kerr, J. Searle, Deletion of cells by apoptosis during castration-inducedinvolution of the rat prostate, Virchows Arch B Cell Pathol. 13 (1973) 87–102.
[200] J. Searle, T.A. Lawson, P.J. Abbott, B. Harmon, J.F. Kerr, An electron-microscopestudy of the mode of cell death induced by cancer-chemotherapeutic agents inpopulations of proliferating normal and neoplastic cells, J. Pathol. 116 (1975)129–138.
[201] K. Polzer, G. Schett, J. Zwerina, The lonely death: chondrocyte apoptosis in TNF-induced arthritis, Autoimmunity 40 (2007) 333–336.
[202] S.N. Redman, I.M. Khan, S.R. Tew, C.W. Archer, In situ detection of cell death inarticular cartilage, Methods Mol. Med. 135 (2007) 183–199.
[203] V.A. Patel, A. Longacre, K. Hsiao, H. Fan, F. Meng, J.E. Mitchell, J. Rauch, D.S. Ucker,J.S. Levine, Apoptotic cells, at all stages of the death process, trigger characteristicsignaling events that are divergent from and dominant over those triggered bynecrotic cells: implications for the delayed clearance model of autoimmunity,J. Biol. Chem. 281 (2006) 4663–4670.
[204] W. Jiang, C.W. Bell, D.S. Pisetsky, The relationship between apoptosis and high-mobility group protein 1 release from murine macrophages stimulated withlipopolysaccharide or polyinosinic-polycytidylic acid, J. Immunol. 178 (2007)6495–6503.
[205] V. Urbonaviciute, B.G. Furnrohr, S. Meister, L. Munoz, P. Heyder, F. De Marchis,M.E. Bianchi, C. Kirschning, H. Wagner, A.A. Manfredi, J.R. Kalden, G. Schett, P.Rovere-Querini, M. Herrmann, R.E. Voll, Induction of inflammatory andimmune responses by HMGB1-nucleosome complexes: implications for thepathogenesis of SLE, J. Exp. Med. 205 (2008) 3007–3018.
[206] J.C. Hall, L. Casciola-Rosen, A. Rosen, Altered structure of autoantigens duringapoptosis, Rheum. Dis. Clin. North Am. 30 (2004) 455–471 vii.
[207] X. Wu, C. Molinaro, N. Johnson, C.A. Casiano, Secondary necrosis is a source ofproteolytically modified forms of specific intracellular autoantigens: implica-tions for systemic autoimmunity, Arthritis Rheum. 44 (2001) 2642–2652.
[208] Z. Mou, Y. He, Y. Wu, Immunoproteomics to identify tumor-associated antigenseliciting humoral response, Cancer Lett. 278 (2009) 123–129.
[209] C. Segundo, F. Medina, C. Rodriguez, R. Martinez-Palencia, F. Leyva-Cobian, J.A.Brieva, Surface molecule loss and bleb formation by human germinal center Bcells undergoing apoptosis: role of apoptotic blebs in monocyte chemotaxis,Blood 94 (1999) 1012–1020.
[210] J.H. Distler, L.C. Huber, S. Gay, O. Distler, D.S. Pisetsky, Microparticles as mediatorsof cellular cross-talk in inflammatory disease, Autoimmunity 39 (2006) 683–690.
[211] J. Kao, K. Houck, Y. Fan, I. Haehnel, S.K. Libutti, M.L. Kayton, T. Grikscheit, J.Chabot, R. Nowygrod, S. Greenberg, et al., Characterization of a novel tumor-derived cytokine. Endothelial-monocyte activating polypeptide II, J. Biol. Chem.269 (1994) 25106–25119.
[212] U.E. Knies, H.A. Behrensdorf, C.A. Mitchell, U. Deutsch, W. Risau, H.C. Drexler, M.Clauss, Regulation of endothelial monocyte-activating polypeptide II release byapoptosis, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 12322–12327.
[213] K. Horino, H. Nishiura, T. Ohsako, Y. Shibuya, T. Hiraoka, N. Kitamura, T.Yamamoto, A monocyte chemotactic factor, S19 ribosomal protein dimer, inphagocytic clearance of apoptotic cells, Lab. Invest. 78 (1998) 603–617.
[214] T. Nishimura, K. Horino, H. Nishiura, Y. Shibuya, T. Hiraoka, S. Tanase, T.Yamamoto, Apoptotic cells of an epithelial cell line, AsPC-1, release monocytechemotactic S19 ribosomal protein dimer, J. Biochem. 129 (2001) 445–454.
[215] T. Yamamoto, Roles of the ribosomal protein S19 dimer and the C5a receptor inpathophysiological functions of phagocytic leukocytes, Pathol. Int. 57 (2007)1–11.
[216] K. Lauber, E. Bohn, S.M. Krober, Y.J. Xiao, S.G. Blumenthal, R.K. Lindemann, P.Marini, C. Wiedig, A. Zobywalski, S. Baksh, Y. Xu, I.B. Autenrieth, K. Schulze-Osthoff, C. Belka, G. Stuhler, S. Wesselborg, Apoptotic cells induce migration ofphagocytes via caspase-3-mediated release of a lipid attraction signal, Cell 113(2003) 717–730.
[217] A. Krispin, Y. Bledi, M. Atallah, U. Trahtemberg, I. Verbovetski, E. Nahari, O. Zelig,M. Linial, D. Mevorach, Apoptotic cell thrombospondin-1 and heparin-bindingdomain lead to dendritic-cell phagocytic and tolerizing states, Blood 108 (2006)3580–3589.
[218] O.Y. Pletjushkina, E.K. Fetisova, K.G. Lyamzaev, O.Y. Ivanova, L.V. Domnina, M.Y.Vyssokikh, A.V. Pustovidko, J.M. Vasiliev, M.P. Murphy, B.V. Chernyak, V.P.Skulachev, Long-distance apoptotic killing of cells is mediated by hydrogenperoxide in amitochondrial ROS-dependent fashion, Cell Death Differ. 12 (2005)1442–1444.
71A.D. Garg et al. / Biochimica et Biophysica Acta 1805 (2010) 53–71
[219] C. Semino, G. Angelini, A. Poggi, A. Rubartelli, NK/iDC interaction results in IL-18secretion by DCs at the synaptic cleft followed by NK cell activation and releaseof the DC maturation factor HMGB1, Blood 106 (2005) 609–616.
[220] P. Scaffidi, T. Misteli, M.E. Bianchi, Release of chromatin protein HMGB1 bynecrotic cells triggers inflammation, Nature 418 (2002) 191–195.
[221] R. Donato, RAGE: a single receptor for several ligands and different cellularresponses: the case of certain S100 proteins, Curr. Mol. Med. 7 (2007) 711–724.
[222] Z. Zhou, Y. Yamamoto, F. Sugai, K. Yoshida, Y. Kishima, H. Sumi, H. Nakamura, S.Sakoda, Hepatoma-derived growth factor is a neurotrophic factor harbored inthe nucleus, J. Biol. Chem. 279 (2004) 27320–27326.
[223] K. Kariko, H. Ni, J. Capodici, M. Lamphier, D. Weissman, mRNA is an endogenousligand for Toll-like receptor 3, J. Biol. Chem. 279 (2004) 12542–12550.
[224] A.K. Nowak, R.A. Lake, A.L. Marzo, B. Scott, W.R. Heath, E.J. Collins, J.A. Frelinger,B.W. Robinson, Induction of tumor cell apoptosis in vivo increases tumor antigencross-presentation, cross-priming rather than cross-tolerizing host tumor-specific CD8 T cells, J. Immunol. 170 (2003) 4905–4913.
[225] A.K. Nowak, B.W. Robinson, R.A. Lake, Synergy between chemotherapy andimmunotherapy in the treatment of established murine solid tumors, CancerRes. 63 (2003) 4490–4496.
[226] D. Hirschhorn-Cymerman, G.A. Rizzuto, T. Merghoub, A.D. Cohen, F. Avogadri,A.M. Lesokhin, A.D. Weinberg, J.D. Wolchok, A.N. Houghton, OX40 engagementand chemotherapy combination provides potent antitumor immunity withconcomitant regulatory T cell apoptosis, J. Exp. Med. 206 (2009) 1103–1116.
[228] C. Menard, F. Martin, L. Apetoh, F. Bouyer, F. Ghiringhelli, Cancer chemotherapy:not only a direct cytotoxic effect, but also an adjuvant for antitumor immunity,Cancer Immunol. Immunother. 57 (2008) 1579–1587.
[229] J. Gol b, R. Zagozdzon, R. Kaminski, K. Kozar, K. Gryska, D. Izycki, A. Mackiewicz, T.Stoklosa, A. Giermasz, W. Lasek, M. Jakobisiak, Potentiated antitumor effective-ness of combined chemo-immunotherapy with interleukin-12 and 5-fluoroura-cil of L1210 leukemia in vivo, Leukemia 15 (2001) 613–620.
[230] T. Kuilman, C. Michaloglou, L.C. Vredeveld, S. Douma, R. van Doorn, C.J. Desmet,L.A. Aarden, W.J. Mooi, D.S. Peeper, Oncogene-induced senescence relayed byan interleukin-dependent inflammatory network, Cell 133 (2008) 1019–1031.
[231] B. Rubinfeld, A. Upadhyay, S.L. Clark, S.E. Fong, V. Smith, H. Koeppen, S. Ross, P.Polakis, Identification and immunotherapeutic targeting of antigens induced bychemotherapy, Nat. Biotechnol. 24 (2006) 205–209.
