4 6 6 | K A N G E T A L . | M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4

INTRODUCTIONAcute pancreatitis (AP) is inflamma-

tion of the pancreas that can become afatal disease or lead to severe complica-tions (1). It is characterized clinically byabdominal pain and by increased pancre-atic enzyme levels in the blood or urine.Gallstone migration and alcohol abuseare the two major risk factors for AP inhumans (2,3). According to the updatedAtlanta classification, AP is generally di-vided into mild, moderate or severe pan-creatitis according to the presence or ab-sence of multiple organ failure (MOF) orlocal or systemic complications (4). Mildpancreatitis has a good prognosis withrapid recovery. The late consequences ofAP include impaired pancreatic exocrinefunction and glucose tolerance, diabetesand development of chronic pancreatitis(5). Moderately severe AP is character-ized by the presence of transient organfailure, local complications or exacerba-

tion of comorbid disease (4). About one-third of patients with AP develop severenecrotizing pancreatitis with persistentMOF and a high mortality rate. The maingoals in the clinical management of APare adequate fluid resuscitation and theprevention of MOF (6,7). Both geneticand environmental factors affect the de-velopment and severity of pancreatitis(8). Although the pathogenic mecha-nisms remain largely unknown, increas-ing evidence suggests that damage- associated molecular pattern molecules(DAMPs) play a central role in the patho-genesis of AP. DAMPs link local tissuedamage to systemic inflammation re-sponse syndrome (SIRS), which, if severeor sustained, can lead to subsequentMOF and even death (9,11) (Figure 1).Most DAMPs are recognized by mem-brane-bound and cytosolic pattern recog-nition receptors (PRRs) expressed byboth immune and nonimmune cell types.

This triggers downstream signaling andmanifests as sterile inflammation(9,12,14). In this review, we outline thepathogenesis of AP, discuss the relation-ship between cell death and AP, andhighlight the latest exciting advances inthe pathogenic role of DAMPs in AP.

PATHOGENESIS OF ACUTEPANCREATITIS

The pancreas is a dual-purpose glandwith exocrine and endocrine functions.The exocrine pancreas produces digestiveproteases in inactive proenzyme form,namely zymogens. The pancreas is nor-mally able to protect itself from zymogenactivation by synthesis of protease in-hibitors such as pancreatic secretorytrypsin inhibitor and serine protease in-hibitor (for example, SPINK1/Spink3). Bycontrast, pancreatic autodigestion andsubsequent AP is initiated once these de-fenses are impaired. In studies of rodentmodels (for example, repetitive ceruleinor L-arginine injections, choline-deficientethionine supplementation [CDE] diet,perfusion of taurocholate into the biliaryor pancreatic duct and surgical ligation orinfusion of pancreatic duct models), someessential pathogenic AP events have beenidentified (see Figure 1) (15,16).

The development of AP involves acomplex cascade of events (17), whichstart with injury or disruption of the pan-

Cell Death and DAMPs in Acute Pancreatitis

Rui Kang, Michael T Lotze, Herbert J Zeh, Timothy R Billiar, and Daolin Tang

Department of Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America

Cell death and inflammation are key pathologic responses of acute pancreatitis (AP), the leading cause of hospital admissionsfor gastrointestinal disorders. It is becoming increasingly clear that damage-associated molecular pattern molecules (DAMPs)play an important role in the pathogenesis of AP by linking local tissue damage to systemic inflammation syndrome. EndogenousDAMPs released from dead, dying or injured cells initiate and extend sterile inflammation via specific pattern recognition recep-tors. Inhibition of the release and activity of DAMPs (for example, high mobility group box 1, DNA, histones and adenosine triphos-phate) provides significant protection against experimental AP. Moreover, increased serum levels of DAMPs in patients with AP cor-relate with disease severity. These findings provide novel insight into the mechanism, diagnosis and management of AP. DAMPsmight be an attractive therapeutic target in AP.Online address: http://www.molmed.orgdoi: 10.2119/molmed.2014.00117

Address correspondence to Rui Kang or Daolin Tang, Department of Surgery, University of

Pittsburgh, 5117 Centre Ave, Hillman Cancer Center, Pittsburgh, Pennsylvania, 15213. Phone:

412-623-1211; Fax: 412-623-1212: E-mail: kangr@upmc.edu or tangd2@upmc.edu.

Submitted June 11, 2014; Accepted for publication August 4, 2014; Epub

(www.molmed.org) ahead of print August 5, 2014.

R E V I E W A R T I C L E

M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4 | K A N G E T A L . | 4 6 7

creatic acini, which then permits the leak-age of active pancreatic enzymes includ-ing amylolytic, lipolytic and proteolyticenzymes that destroy local tissues. Thisresults in edema, vascular damage, hem-orrhage and cell death (18,19). In additionto oxidative stress (20) and calcium over-load (21), hypotension (22) and low aci-nar pH (23) contribute to these initiationprocesses. After initial production of ac-tive pancreatic enzymes, local cell deathand systemic inflammation ensue. Mito-chondria, the energy factories of cells,regulate pancreatic cell death throughcontrol of the production of adenosinetriphosphate (ATP) and reactive oxygenspecies (ROS), as well as calcium (24).Dysfunction of mitochondrial calciumuptake and efflux, including elevation ofcytosolic calcium from the endoplasmic

reticulum, can cause mitochondrial cal-cium overload, which leads to enhancedgeneration of mitochondrial ROS and mi-tochondrial membrane permeabilization.Mitochondria dysfunction-mediated ox-idative injury results in endoplasmicreticulum stress, lysosomal damage andthe release of proteases (for example,cathepsin and trypsin) to degrade cytoso-lic proteins that cause pancreatic acinarcell death (25,26). Dead, dying and in-jured pancreatic acinar cells release intra-cellular contents, including DAMPs (forexample, high mobility group box 1[HMGB1], DNA, histones and ATP),which in turn promote infiltration of vari-ous immune cells (for example, neu-trophils, monocytes and macrophages)and activation of inflammatory signalingpathways (for example, nuclear factor-κB

[NF-κB], mitogen-activated protein ki-nase [MAPK], signal transducer and acti-vator of transcription 3 [STAT3] and in-flammasome). Of these, NF-κB appears tothe main inflammatory pathway in pan-creatitis, although conflicting reportshave been published (27–30). Recruitmentand activated innate immune cells at thesite of injury lead to further acinar cell in-jury and increase the levels of circulatingDAMPs. It is important to note that neu-trophil depletion markedly reduces tissuedamage severity in AP, suggesting a criti-cal role for tissue infiltration of neu-trophils in AP (31). DAMPs ultimately arethe mediators of the systemic inflamma-tory response and cause further pancre-atic damage, as well as MOF. In particu-lar, acute lung injury and adultrespiratory distress syndrome are theleading causes of death in patients withAP (6). More recently, myeloid NF-κB ac-tivation was found to contribute to IL-6synthesis and IL-6 transsignaling, whichpromotes pancreatitis-associated lung in-jury and lethality (32). A better under-standing of the relationship between localtissue injury and systemic inflammationsyndrome in the development of AP mayprovide more targeted therapeutic op-tions (see Figure 1).

CELL DEATH AND ACUTE PANCREATITISThe severity of experimental AP corre-

lates with the extent and type of cell in-jury and death. Although multiple formsof cell death exist in physiological andpathological conditions (33), necrosis andapoptosis are the most widely studiedtypes in both clinical and experimentalAP (34,35). Necrotic cells are capable ofactivating proinflammatory and im-munostimulatory responses by releasingDAMPs and other molecules, whereasapoptosis is usually considered immuno-logically silent because the cytoplasmiccontent is packaged in apoptotic bodiesand these membrane-bound cell frag-ments are rapidly taken up and de-graded by phagocytes or autophagy (36).However, excessive apoptotic cells canactivate the inflammatory and immuneresponses by releasing nuclear DAMPs

Figure 1. Emerging role of DAMPs in acute pancreatitis. A number of rodent models haveproven useful for studying the pathogenic mechanisms of acute pancreatitis. Increasingevidence indicates that DAMPs such as HMGB1, DNA, HSPs, histones and ATP play a cen-tral role in the pathogenesis of acute pancreatitis because DAMPs link local tissue dam-age and death to systemic inflammation response syndrome, which leads to subsequentMOF and even death.

4 6 8 | K A N G E T A L . | M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4

D A M P s A N D A C U T E P A N C R E A T I T I S

(nDAMPs) and mitochondrial DAMPs(mitDAMPs) (10,37–39). In addition tonecrosis and apoptosis, additional path-ways (for example, necroptosis, pyropto-sis and autophagy) regulate DAMP re-lease and affect AP progression (Figure 2).Collectively, these findings contributegreatly to our understanding of the im-munologic choices made during celldeath (12,14). The roles of different typesof cell death in AP are discussed below.

NecrosisNecrosis is a process of cell self-

destruction from external or internal

stimuli, whereas autolysis is caused bycell self-digestion from its own enzymes(40). Morphological features of necroticcells include swelling of the cell, cellmembrane rupture and the release of in-tracellular contents, including proteinsand nonproteins (for example, DNA andRNA). Nuclear morphological changesresulting from breakdown of DNA showthree types: pyknosis (nucleus shrinksand the chromatin condenses into a solidbasophilic mass), karyorrhexis (nucleusshrinks further) and karyolysis (DNA iscompletely digested and none of the nu-cleus is visible). Karyolysis usually is ob-

served in necrosis, whereas pyknosis andkaryorrhexis occur in apoptotic cells.

A number of observations have indi-cated that necrosis is the major type ofpancreatic acinar cell death (34,41). Al-though some data clearly implicates thatthe release of DAMPs from necrosis isresponsible for the inflammatory re-sponse, much less is known about thesignaling mechanisms responsible forthis function. A good example of such amechanism is derived from the study ofpoly (ADP-ribose) polymerase (PARP) innecrosis. PARP-1 activation during DNAdamage induces energy failure by de-pleting NAD+; the cell consumes ATP toreplete the NAD+ level, ultimately re-sulting in energy failure and DAMP re-lease in necrosis (42). Genetic or phar-macologic blockade of PARP inhibitsnecrosis and demonstrates a reducedpancreatitis response (43). In addition tothe passive ATP depletion-mediatedmechanisms, several active mechanisms(for example, oxidative stress, calciumoverload, mitochondrial permeabilitytransition pore opening and cathepsinrelease) also participate in the regulationof the necrotic process in AP (34,44,45).The actual mechanisms of necrosis in-volved in the development of sequentialcellular and organ response in AP needfurther investigation.

ApoptosisApoptosis is a process of programmed

cell death that generally includes a celldeath receptor-dependent (extrinsic)pathway and a mitochondria-dependent(intrinsic) pathway (46). Morphologicalfeatures of apoptotic cells include cellshrinkage, membrane blebbing, chro-matin condensing, DNA fragmentationand apoptotic body formation. The cas-pases are central initiators and effectorsof apoptosis, although the existence of acaspase-independent pathway promotesapoptosis (47). Caspase-8 mediates thecell death receptor-dependent pathway,whereas caspase-9 triggers the mitochon-dria-dependent apoptotic pathway. Bothcaspase-8 and -9 can activate caspase-3 tocleave specific groups of substrate pro-

Figure 2. Release and activity of DAMPs in acute pancreatitis. Various pathologic insultscause mitochondrial dysfunction and endoplasmic reticulum stress in pancreatic acinarcells, which leads to activation of multiple pathologic signals such as oxidative stress andcalcium overload. These signals are common triggers of cell death (for example, necrosis,apoptosis, necroptosis, pyroptosis and autophagic cell death) through different regulatoror effector proteins as indicated. In contrast to the intracellular physiological role ofDAMPs, extracellular DAMPs from cell death are important mediators of local and sys-temic inflammatory responses through different receptors (for example, TLRs, NLRs, RAGEand P2X7).

R E V I E W A R T I C L E

M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4 | K A N G E T A L . | 4 6 9

teins to induce apoptosis. The Bcl-2 fam-ily is divided into antiapoptosis (for ex-ample, Bcl-2, Bcl-XL, Bcl-w and MCL-1)and proapoptosis (for example, Bax, Bak,Box, Bad, Bim, Bid, PUMA and NOXA)proteins that are majorly responsible forregulating the intrinsic pathway by con-trolling mitochondrial outer-membranepermeabilization (MOMP) (48). WhenMOMP occurs, several mitochondrialproteins (for example, cytochrome c andsecond mitochondria-derived activator ofcaspases [Smac/DIABLO]) are releasedinto the cytosol to induce the formationof apoptosome (49,50).

Cerulein, L-arginine, ethanol,lipopolysaccharide (LPS), platelet-acti-vating factor, and cytokines (for exam-ple, tumor necrosis factor [TNF]-α andinterleukin [IL]-1β) have been shown toinduce apoptosis in AP. Besides caspaseand Bcl-2 family proteins, other compo-nents of the death machinery (for exam-ple, X-linked inhibitor of apoptosis andp53) are involved in the regulation of ex-trinsic or intrinsic apoptosis in AP.Given that apoptosis generally limits theinflammatory cascade, the apoptosislevel of acinar cells is inversely relatedto AP severity (35,51). Thus, induction ofapoptosis in pancreatic acinar cells ex-erts a protective effect (52,53), whereassuppression of apoptosis by caspase in-hibitors increases the severity of AP (54).Caspases not only mediate apoptosis butalso protect cells against necrosis bycleavage and inactivation of PARP ortrypsin (55). Perhaps paradoxically, in-tracellular heat shock proteins (HSPs)protect pancreatic acinar cells partly byinhibition of apoptosis (56,57), suggest-ing a pathogenic role of apoptosis in AP(58). Subsequently, researchers also ob-served that pancreatic acinar cells un-dergoing apoptosis can release histones,DNA and HMGB1, which facilitate pan-creatic injury and the inflammatory re-sponse (10,11). Whether induction or in-hibition of apoptosis would be beneficialin a clinical setting remains unproven. Inaddition, the current pharmacologicagents that target apoptosis largely lackspecificity.

AutophagyAutophagy, an evolutionarily con-

served process, can deliver endogenouscytoplasmic constituents (for example,damaged organelles and unused pro-teins) and exogenous pathogens to lyso-somes for degradation (59). Autophagyhas three different forms: microau-tophagy, chaperone-mediated autophagyand macroautophagy. Macroautophagy(hereafter called autophagy) is the well-studied dynamic membrane process thatis regulated by the family of autophagy-related proteins (ATGs). The major mor-phological features of autophagy involvethe formation and maturation of au-tophagic vacuoles, includingphagophores, autophagosomes and au-tolysosomes. Hydrolases (cathepsins Land B) from lysosomes contribute to thedegradation of substrates in the au-tolysosome, which produce amino acidsand ATP for protein synthesis and en-ergy metabolism. Autophagy generallyexerts a prosurvival function duringstress, whereas excessive autophagy maycontribute to cell death and DAMP re-lease (60,61). DAMP-induced autophagyis involved in the regulation of inflam-mation and the immune response(14,62,63), while autophagy participatesin the secretion, release and degradationof DAMPs in a context-dependent man-ner (64).

Autophagy dysfunction has been im-plicated in the pathogenesis of humandiseases, including AP (65,66). One earlyfeature of AP is the accumulation of largeautophagic vacuoles in pancreatic acinarcells (67,68). Autophagic flux is impairedin AP, which mediates acinar cell vacuoleformation, trypsinogen activation, celldeath and the inflammatory response(for example, inflammasome activation)(67,69,70), suggesting a protective role ofautophagy in AP. In contrast to the aboveconcept, ATG5-mediated autophagosomeformation in acinar cells is not blocked inpancreatitis, and inhibition of autophagyfrom ATG5-deficient mice improvescerulein-induced AP with reducedtrypsinogen activation, suggesting thatautophagy is a harmful response (67,68).

Perhaps more importantly, the expres-sion of vacuole membrane protein 1(VMP1) is increased in AP and mediates“zymophagy,” a selectively autophagicpathway to remove zymogen granules,and prevents pancreatic acinar cell deathby interaction with Beclin1, an essentialregulator of autophagy and apoptosis(71–74). Mitochondrial injury and lipidabnormalities have been associated withpancreatitis (66). The functional interac-tion of zymophagy and other selectiveautophagic pathways including mi-tophagy and lipophagy in AP have beeninsufficiently explored. Lysosomes canbe involved in various cellular transportprocesses, including autophagy. Lysoso-mal dysfunction from deficiency ofGnptab enzyme and lysosomal mem-brane protein (LAMP)-2 can lead to au-tophagic flux impairment in AP(67,69,70). Loss of LAMP-2 impairs au-tophagosome-lysosome fusion and pro-motes necrosis and inflammation, whichis associated with HMGB1 release in AP(70). Interestingly, autophagosomes areincreased significantly in trypsin in-hibitor SPINK1/Spink3-deficient pancre-atic acinar cells (75). However, it is un-clear how SPINK1 regulates autophagyflux and zymophagy. Collectively, thesestudies suggest that autophagy has bothpositive and negative impacts on pancre-atitis (Figure 3).

NecroptosisNecroptosis is a process of programmed

necrotic cell death (76). Morphologicalfeatures of necroptosis are similar to thatof necrosis and various types of au-tophagic vesicles are frequently observed(76). Stimulation of death receptors uponligation by TNF-α, FasL and TNF-relatedapoptosis-inducing ligand (TRAIL)under apoptosis-deficient conditions, inparticular with pan-caspase inhibitor Z-VAD-FMK or inhibition of caspase-8,may induce necroptosis. Receptor-inter-acting protein (RIP) kinases play a cen-tral role in the regulation of necroptosis.The RIP kinases contain seven membersthat facilitate NF-κB activation anddeath-inducing processes. As caspase-8

4 7 0 | K A N G E T A L . | M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4

D A M P s A N D A C U T E P A N C R E A T I T I S

substrates, binding of RIP1/RIPK-1 toRIP3/RIPK-3 forms a phosphorylationcomplex to assemble the necrosome (77).Cellular FLICE inhibitory protein (cFLIP)is a homolog of caspase-8. In the pres-ence of the long isoform of cFLIP(namely cFLIPL), caspase-8 fails to medi-ate apoptosis, but can cause cleavage ofboth RIP1 and RIP3, which inhibitsnecroptosis. By contrast, inhibition ofcaspase-8 expression or activity byknockout of caspase-8 or its interactorFAS-associated death domain protein(FADD) results in increased RIP1 stabil-ity that induces necroptosis. The mixedlineage kinase domain-like protein(MLKL) is the interacting target of RIP3and a component of necrosome to in-duce necroptosis (78). The discovery ofnecrostatin-1 as a RIP1-specific kinase

inhibitor enables selective inhibition ofnecroptosis and underscores its potentialcontribution to diseases (79).

The expression of RIP3 protein is in-creased in AP. Notably, RIP3 knockoutmice exhibit significantly decreasednecroptosis and MOF in experimental AP(80,81). These observations suggests thatnecroptosis mediates AP development.Inhibition of necroptosis by gene deletionof RIP3 or using necrostatin-1 diminishesDAMP release and protects mice fromlethal sepsis (82). RIP- mediated necroptosisalso is implicated in sterile inflammationfrom trauma and ischemia-reperfusion in-jury (79), which is accompanied by highlevels of DAMP release. These studiessuggest that inhibition of necroptosisblocks DAMP release and prevents sterileinflammation (83). Intriguingly, blocking

both apoptosis and necroptosis by com-bined Z-VAD-FMK and necrostatin-1treatment exacerbates cerulein-inducedAP (84). These data highlight the impor-tance of regulatory switches between dif-ferent forms of cell death that influencethe development of AP.

PyroptosisThe term pyroptosis was originally

coined to describe peculiar bacterial in-fection-induced death in macrophages(85) and it is now a form of cell death inimmune cells mediated by inflamma-some in response to immune stimuli (forexample, PAMPs and DAMPs) (86).Apart from extracellular ATP, HMGB1and histone also can induce pyroptosis inimmune cells and further augment theinfection or sterile inflammation (87,88).For example, RAGE-dependent uptake ofHMGB1 by macrophages leads to cas-pase-1-mediated pyroptosis in the devel-opment of inflammation, whereas TLR9is required for histone-mediated pyrop-tosis during liver injury (87–89). Pyropto-sis shares morphological features withboth apoptosis and necroptosis, includ-ing cytoplasmic swelling, DNA fragmen-tation and pore formation. The formationof pyroptotic DNA fragmentation is in-dependent of caspase-activated DNase.In contrast to apoptosis, pyroptosis has aproinflammatory nature because of therelease of cell contents after rapid plasmamembrane permeabilization or the secre-tion of proinflammatory cytokines (forexample, IL-1β, IL-18, IL-33 and HMGB1)after caspase-1 activation (86,90). Inflam-masome is a multiprotein oligomer thatis assembled in the cytoplasm by cytoso-lic NOD-like receptors (NLRs). A well-studied NLR inflammasome is theNLRP3 inflammasome, which recognizesvarious agonists. The NLRP3 inflamma-some activation requires the primingstep, in which DAMPs such as HMGB1or HMGB1/DNA complex might inducethe expression of NLRP3 by signalingthrough TLR4 and TLR9, respectively(91,92). In addition to NLRs, non-NLRinflammasomes such as absent in mela-noma 2 (AIM2) inflammasome have the

Figure 3. The yin and yang of autophagy in acute pancreatitis. On one hand, VMP1- Beclin1 complex-mediated zymophagy, a selective autophagy, prevents pancreatic aci-nar cell death through degradation of zymogen. On the other hand, deficiency ofLAMP2, Gnptab and SPINK1 cause impaired autophagy flux and increased autophago-some accumulation, which mediates vacuole formation in pancreatic acinar cells, intra-cellular activation of trypsinogen, cell death and the inflammatory response. By contrast,deficiency of ATG5 decreases autophagosome formation, which reduces intracellular ac-tivation of trypsinogen and pancreatic damage in AP. Thus, autophagy plays a dual rolein the regulation of AP.

R E V I E W A R T I C L E

M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4 | K A N G E T A L . | 4 7 1

ability to detect foreign dsDNA. RAGE isrequired for both HMGB1/DNA com-plex-induced AIM2 inflammasome activ-ity and autophagy, although upregulatedautophagy subsequently limits inflam-masome activity (93). The regulatorymechanisms of inflammasome activationare extremely complex (94). Activation ofdouble-stranded RNA-dependent pro-tein kinase (PKR) is implicated in thecrosstalk between inflammasomes,DAMP release and cell death (95).

TLR9 plays a fundamental role inDNA recognition and activation of innateimmunity. Both activation of TLR9 andthe NLRP3 inflammasome contribute topancreatic acinar cell death and sterileinflammation in AP (10). Thus, geneticdeletion of NLRP3, caspase-1 and TLR9in mice protects against cerulein-inducedpancreatitis (10). The potential mecha-nism responsible for these phenomena isinvolved in increased mitochondrialDNA (mitDNA) and nuclear DNA(nDNA) release during apoptosis, whichactivates TLR9 as well as NLRP3 inflam-masome pathways (10,38,39). Taken to-gether, interplay between pyroptosis andDNA-sensing pathway is involved in thesterile inflammatory response.

DAMPs AND THEIR RECEPTORSWhy is the immune system so con-

cerned with cell death? The current no-tion is that DAMPs released or exposedfrom dying or dead cells contribute to in-flammatory and immune responses to re-move dead cells and initiate tissue healing(12,13). Failure of this control mechanismcan lead to uncontrolled inflammationand serious diseases such as sepsis, arthri-tis, atherosclerosis, lupus and cancer. Bycontrast, pathogen-associated molecularpatterns (PAMPs) are foreign danger sig-nals and can induce immune cells to ac-tively secrete DAMPs by various nonclas-sical pathways, which indicates cross-talkbetween DAMPs and PAMPs in the regu-lation of innate immune responses (14).

A growing number of endogenoussubstances from multiple subcellularcompartments are being identified as po-tential DAMPs. The most prominent

DAMPs are nDAMPs, includingHMGB1, histones and nDNA. OtherPAMPs include S100 proteins, HSPs,complement, ATP and uric acid. Mito-chondria are now recognized not only ascentral players in cell death but also asan important source of DAMPs. mit-DAMPs, including mitDNA, N-formylpeptides, transcription factor A (TFAM, amitochondrial HMGB1 homologue) andROS, play emerging roles in inflamma-tion by the activation of neutrophils,monocytes and macrophages (96).DAMP sensing can be further classifiedas extracellular or intracellular depend-ing on whether the DAMP is releasedfrom the stressed cell. IntracellularDAMP sensing involves detection by in-tracellular receptors in the same cell thatproduced the DAMP (14).

A number of receptors (for example,TLRs, NLRs, RIG-I-like receptors (RLRs),the receptor for advanced glycation endproducts [RAGE], and P2X7) have beenreported to mediate DAMP activity indifferent cell types, including immuneand nonimmune cells. TLRs play a cen-tral role in innate immunity. TLR-1, -2, -4, -5 and -6 are located on the cell sur-face, whereas TLR-3, -7, -8 and -9 are ex-pressed on the endosomal and lysosomalcompartment. NLRs and RLRs act as in-tracellular surveillance molecules andRAGE is a member of the immunoglobu-lin gene superfamily. RAGE also is pres-ent in the mitochondria and regulatesmitochondrial respiration (97). Posttrans-lational modification (for example, oxi-dation and proteolysis) of DAMPs affectstheir receptor binding ability and activ-ity. In addition to receptors, endocytosisalso mediates DAMP activity in immuneand cancer cells (88,97). A fine-tunedmechanism of these actions in diseaseneeds further exploration.

Once released, extracellular DAMPsinduce the activation of multiple inflam-matory pathways that lead to the pro-duction and release of inflammatory cy-tokines, interferons, chemokines and celladhesion molecules. Apart from theirmain effects on immunity, DAMPs andtheir receptors also have been demon-

strated to regulate cell survival, prolifer-ation, differentiation and death in bothimmune and nonimmune cells. The pre-cise structures and affinities of these lig-and-receptor interactions remain to befully characterized.

HMGB1 AND ACUTE PANCREATITISHMGB1 is a highly conserved protein

with location-dependent functions. In thenucleus, HMGB1 as a DNA chaperoneregulates nucleosome structure, genomicstability and a number of DNA-associatedevents. Loss of HMGB1 in cells increasesDNA damage and apoptosis and inhibitsautophagy during stress (98,99). Once re-leased during death, HMGB1 acts as aDAMP, regulating the inflammatory andimmune responses (100,101). Ample evi-dence also shows that HMGB1 can act asan intracellular DAMP and drive activa-tion of intracellular receptors through nu-cleic acid recognition. Caspase-3/7 andPARP-1 are required for HMGB1 releasein apoptosis and necrosis, (42,102 respec-tively), whereas ATG5 and PKR are re-quired for HMGB1 release in autophagyand pyroptosis, (103,104 respectively).RAGE and TLRs are positive receptorsmediating the proinflammatory activityof HMGB1, whereas CD24 and TIM3 arenegative receptors limiting HMGB1 activ-ity in inflammation and tumor immunity(105). The redox status of HMGB1 alsoregulates its activity (106). In particular,HMGB1 in reduced all-thiol form pro-motes cell migration, whereas HMGB1 indisulfide form induces cytokine produc-tion (106). By contrast, oxidized HMGB1loses the above immune activities. More-over, reduced HMGB1 induces au-tophagy, whereas oxidized HMGB1 in-duces caspase-dependent apoptosis (107).However, the redox status within tissuein disease is a dynamic process and theextracellular activity of HMGB1 maychange accordingly (108).

HMGB1 play an important role inhuman health and disease (109). Theserum levels of HMGB1 are significantlyelevated and correlate with the severityof AP (110). HMGB1 seems to act as animportant cytokine mediator in the

4 7 2 | K A N G E T A L . | M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4

D A M P s A N D A C U T E P A N C R E A T I T I S

pathogenesis of severe AP and has awider therapeutic window. Early block-ade or delayed therapeutic delivery (forexample, ethyl pyruvate [111], A box[112] and anti-HMGB1 antibody [113])targeting HMGB1 limits developmentand associated MOF in experimental AP.Antioxidants (for example, pyrrolidinedithiocarbamate 114) and anticoagulants(for example, antithrombin III [115] anddanaparoid sodium [116]) also inhibitHMGB1 release and therefore protectagainst injury in AP. By contrast, intracel-lular HMGB1 protects against AP partlythrough limiting nDAMP (histones andDNA) release and subsequent inflamma-tory cell recruitment and activation (11).These observations indicate the dual roleof HMGB1 in the regulation of AP.HMGB1 function is not only complicatedin the pancreas, but also in other tis-sues/cells according to studies ofHMGB1 knockout or knockin mice (117).

HSPS AND ACUTE PANCREATITISHSPs are conserved stress proteins and

their expression is upregulated in acti-vated cells responding to heat shock, in-fection, hypoxia and other stressors. Themajor function of HSP is assisting withthe folding and unfolding of newlytranslated proteins during stress. Accord-ing to their molecular size, mammalianHSPs are categorized into the followinggroups: HSP100, HSP90, HSP70, HSP60,HSP40 and the small HSPs (for example,HSP25/27, αB-crystallin and HSP22).Overexpression of intracellular HSPs in-hibits apoptosis and inflammatory cy-tokine production (for example, TNF-α,IL-1β and HMGB1) following cellularstresses. HSP70 is directly involved inchaperone-mediated autophagy (118).HSP70 and HSP27 confer protection incellular stress by enhancing autophagyand mitophagy (119,120). Some HSPssuch as HSP70 and HSP60 are rapidly re-leased following nonprogrammed celldeath (12,13). Extracellular HSPs induceimmune cell activation and the proin-flammatory response through binding toPRRs, including TLR2, TLR4 and CD91.Due to their molecular chaperone prop-

erties, HSPs easily complex with othermolecules including PAMPs; severalstudies demonstrate that the previouslyreported cytokine activity of HSP wasdue to contaminating bacterial productsincluding LPS and flagellin (121,122).This supports the argument against therole of HSPs as DAMPs.

The expression of inducible HSPs (forexample, HSP27, HSP60, HSP70 andHSP90) is upregulated in experimentalAP (123,124). Preconditioning of animalswith either a thermal or chemical stressorinduces HSP expression, which in turnprotects against AP (56,57). Knockout ofheat shock factor protein 1 in mice in-hibits HSP synthesis and develops moresevere AP induced by cerulein (125).HSP27, HSP60 and HSP70 are direct ef-fectors in mediating this protective effectagainst AP by regulation of NF-κB sig-naling (30), intrapancreatic trypsinogenactivation (56,126), calcium overload(125), actin cytoskeleton (127), apoptosis(128) and autophagy (129). Althoughstudies of the role of extracellular HSP inAP are limited, one study shows that ad-ministration of recombinant HSP70 inmice aggravates cerulein-induced AP ina TLR4-dependent manner (130). SerumHSP27 is increased in patients with pan-creatitis and pancreatic cancer (131),highlighting its role as a potential bio-marker in these diseases.

DNA AND ACUTE PANCREATITISDNA is the hereditary material that

carries genetic information. According tothe endosymbiotic theory, nDNA andmtDNA have separate evolutionary ori-gins. mtDNA evolved from the circulargenomes of bacteria when an early eu-karyotic cell engulfed a prokaryotic cell(132). DNA is a potent activator of innateimmunity by activation of type I inter-feron, inflammasome, and NF-κB path-ways (133). Monitoring levels of circulat-ing DNA has been proposed as a usefultool in the noninvasive diagnosis ofhuman diseases. Serum DNA level is ele-vated significantly in patients with se-vere AP and may be an early predictor ofseverity in AP (134–137). Apoptotic or

necrotic cells are the primary source ofcirculating DNA during AP.

Different types of nucleic acid recep-tors and sensing molecules have beenidentified (133). TLR9 is critical for therecognition of CpG DNA. The expressionof TLR9 in the pancreas is increased inAP (138) and genetic deletion of TLR9protects mice from AP (10). TLR9 alsomediates the innate immune response tonDNA, mitDNA, histone and HMGB1 inAP and other sterile injuries (10,89). Theinteraction between RAGE and TLR9 isrequired for HMGB1-nDNA complexand TFAM-mitDNA complex-mediatedimmune responses (139,140). RAGE alsohas the ability to directly recognize DNA(141). A study indicates that besidesRAGE and TLR9, TLR2 is required forthe HMGB1-nucleosome complex- mediated immune response in macro -phages and dendritic cells (142). Thesefindings improve our understanding ofhow DNA-protein complexes trigger in-nate immune responses in human dis-eases, including AP.

HISTONE AND ACUTE PANCREATITISHistones are basic components of

chromosome structural units, namelynucleosomes. They are divided into corehistones (H2A, H2B, H3 and H4) andlink histones (H1/H5). In the nucleus,histones and their posttranslationalmodifications regulate chromosomestructure and function (143). In additionto their nuclear function, emergingstudies indicate that histones as well asnucleosomes can be released into the ex-tracellular space upon infection (for ex-ample, sepsis) (144), sterile inflamma-tion (for example, trauma,ischemia-reperfusion injury and pancre-atitis) (11,89,145) and death (for exam-ple, apoptosis, necrosis and NETosis)(146). NETosis is a unique form of celldeath that is first observed in neu-trophils upon microbial infection (147).Neutrophils have the ability to releasechromosomal components including hi-stones to capture and kill microbes(147). Once released, histones exhibitDAMP activity with significant induc-

R E V I E W A R T I C L E

M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4 | K A N G E T A L . | 4 7 3

tion of proinflammatory and toxic re-sponses in vivo and in vitro (144,145).Activation of TLRs (for example, TLR2,TLR4 and TLR9) and NLRP3 inflamma-some mediate the activity of histones(87,89,145,148). Dynamic changes in cir-culating levels of histones as well as nu-cleosomes serve as potential biomarkersand novel therapeutic targets in humandiseases (149,150).

Our recent study indicates that oxida-tive stress-mediated DNA damage is re-sponsible for histone release in AP (11).This process is controlled by intracellularHMGB1 (Figure 4). Loss of intracellularHMGB1 in pancreatic tissue leads tolocal oxidative injury, which facilitatesnuclear catastrophe and proinflamma-tory nucleosomal (histone and DNA) re-lease (11). Consistently, antioxidant N-acetyl-L-cysteine protects against AP inpancreatic HMGB1 conditional knockoutmice (termed CH mice). Once released,extracellular histones (H3 and H4) can

recruit and activate immune cells such asmacrophages to secrete HMGB1 into thecirculation of CH mice. Moreover, anti-H3 antibody and anti-HMGB1 antibodyprotect against AP in CH mice (11).These findings suggest interplay be-tween intracellular and extracellularnDAMPs in the pathogenesis of AP (seeFigure 4).

ATP AND ACUTE PANCREATITISATP, the major energy currency mole-

cule of various cellular processes, is nor-mally present in the cell cytoplasm andis used as a substrate for protein kinaseactivation (151). Most ATP in mam-malian cells is generated within the mi-tochondria. The ability of autophagy todefend cells against various stressors isobtained partly through the generationof substrates to maintain ATP produc-tion. Intracellular ATP levels determinecell death fate (33). ATP depletion leadsto necrosis, whereas ATP production is

required for the induction of apoptosis(33). Further, ATP can be passively re-leased into the extracellular space dur-ing cell death, which mediates the im-mune response by P2X7 receptor (152).Several mechanisms have been reportedto regulate passive ATP release. For ex-ample, caspase and pannexin 1 regulateATP release in apoptosis (153). In addi-tion, active secretion of ATP has beenidentified in immune and cancer cells.The underlying mechanisms of ATP se-cretion include ion channel-mediatedconductive release as well as autophagy-mediated exocytosis of ATP-enrichedvesicles (154–156). Extracellular ATPregulates multiple cell processes, in-cluding cell migration, metabolism, in-flammation, immunity, differentiation,apoptosis and autophagy (151). How-ever, the signaling mechanism of theseATP-mediated activities remains to beelucidated.

In AP, extracellular ATP binds to its re-ceptor P2X7 and results in NLRP3 in-flammasome assembly, caspase-1 activa-tion and IL-1β secretion (10). Inhibitionof P2X7 through genetic deletion, ortreatment with P2X7 antagonists (for ex-ample, A-438079), limits pancreatic in-jury and the inflammatory response inan experimental animal AP model (10).In another study, mitochondria-mediatedintracellular ATP depletion was found tobe able to cause necrosis in AP (157). In-terestingly, the P2X7 receptor not onlymediates extracellular ATP activity, butalso regulates ATP release in several cells(158). Thus, it seems likely that the ATP-P2X7 signaling pathway is a novel regu-lating mechanism for AP.

CONCLUDING REMARKS AND FUTUREPERSPECTIVES

It has been widely accepted that in-flammation and cell death are key patho-logic responses of AP and determine dis-ease severity. Mitochondria-mediated celldeath is essential for the initiation of AP.As mentioned above, imbalanced produc-tion of mitochondrial calcium, ATP andROS are the most common causes of cel-lular injury and death. Various signaling

Figure 4. Role of intracellular and extracellular HMGB1 in acute pancreatitis. Under normalconditions, HMGB1 is located in the nucleus of pancreatic acinar cells to regulate nucle-osome stability. However, under conditions of increased pathologic insult, increased oxida-tive stress and calcium overload leads to HMGB1 release. Loss of intracellular HMGB1 inpancreatic acinar cells accelerates DNA damage, cell death and nDAMP (for example,histone) release. Subsequently, extracellular histone promotes immune cell (for example,macrophage and neutrophil) recruitment and activation, which leads to HMGB1 release.Consequently, circulating HMGB1 levels are increased and function as a mediator oflethal systemic inflammation in acute pancreatitis.

4 7 4 | K A N G E T A L . | M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4

D A M P s A N D A C U T E P A N C R E A T I T I S

molecules and proteins have related oroverlapping actions in the regulation ofcell death. This regulation not only modi-fies the magnitude of the death responsebut also results in switching between dif-ferent types of death. Such complexdeath events then progress to a systemicinflammatory response that is mediatedby the release of various intracellularcontent including DAMPs. DAMPs arenot only passively released during celldeath and tissue injury, but also activelysecreted by activated immune cells.DAMPs have multiple forms and theiractivities vary greatly depending on typeof death, posttranslational modification,and their receptors. In particular,HMGB1, one of the best-characterizedDAMPs, has a unique role in AP accord-ing to experimental and clinical studies.

It is also worth mentioning that intra-cellular and extracellular DAMPs mayhave different mechanisms of action inthe pathogenesis of AP (see Figure 4). In-tracellular DAMPs such as HSP andHMGB1 inhibit cell death and the in-flammatory response during AP devel-opment. Extracellular DAMPs may exertsynergistic effects to accelerate the devel-opment of AP. The role of DAMP recep-tors such as TLR2 and TLR4 in AP ap-pears complex and even contradictory(159–161). The existence of multipleDAMP receptors likely provides distinctcontrol mechanisms for expanding or di-minishing inflammation at differentstages of AP. Metabolic changes such asaerobic glycolysis in cells participate inthe inflammatory response contributingto HMGB1 release in sepsis (162), al-though this change in AP remains un-known. Future studies are needed toconfirm this concept and define in detailDAMP release, biological activity, recep-tor signal transduction and cross-talk be-tween different DAMPs and their recep-tors in patients with AP. This area ofpathobiology offers great opportunitiesto define novel biomarkers for diseaseprogression as well as targeted therapiesto redirect the aggressive local and sys-temic inflammatory response that char-acterizes severe forms of AP.

ACKNOWLEDGMENTSWe apologize to the researchers who

were not referenced due to space limita-tions. We thank Christine Heiner (Depart-ment of Surgery, University of Pittsburgh,Pittsburgh, PA, USA) for her critical read-ing of the manuscript. This work wassupported by the National Institutes ofHealth (R01CA160417 to D Tang;R01CA181450 to HJ Zeh/MT Lotze) and a 2013 Pancreatic Cancer Action Network-AACR Career DevelopmentAward (grant number 13-20-25-TANG).The work supporting the findings re-viewed in this manuscript was aided bycore support from the University of Pittsburgh Cancer Institute (NIH grantP30CA047904).

DISCLOSURESThe authors declare that they have no

competing interests as defined by Molecu-lar Medicine, or other interests that mightbe perceived to influence the results anddiscussion reported in this paper.

REFERENCES1. Whitcomb DC. (2006) Clinical practice. Acute

pancreatitis. N. Engl. J. Med. 354:2142–50.2. van Geenen EJ, et al. (2010) Etiology and diagno-

sis of acute biliary pancreatitis. Nat. Rev. Gas-troenterol. Hepatol. 7:495–502.

3. Sand J, Lankisch PG, Nordback I. (2007) Alcoholconsumption in patients with acute or chronicpancreatitis. Pancreatology. 7:147–56.

4. Banks PA, et al. (2013) Classification of acute pan-creatitis—2012: revision of the Atlanta classifica-tion and definitions by international consensus.Gut. 62:102–11.

5. Sand J, Nordback I. (2009) Acute pancreatitis:risk of recurrence and late consequences of thedisease. Nat. Rev. Gastroenterol. Hepatol. 6:470–7.

6. Wu BU, Banks PA. (2013) Clinical managementof patients with acute pancreatitis. Gastroenterol-ogy. 144:1272–81.

7. Haydock MD, et al. (2013) Fluid therapy in acutepancreatitis: anybody’s guess. Ann. Surg. 257:182–8.

8. Whitcomb DC. (2013) Genetic risk factors for pan-creatic disorders. Gastroenterology. 144:1292–302.

9. Hoque R, Malik AF, Gorelick F, Mehal WZ.(2012) Sterile inflammatory response in acutepancreatitis. Pancreas. 41:353–7.

10. Hoque R, et al. (2011) TLR9 and the NLRP3 in-flammasome link acinar cell death with inflam-mation in acute pancreatitis. Gastroenterology.141:358–69.

11. Kang R, et al. (2014) Intracellular hmgb1 inhibitsinflammatory nucleosome release and limits

acute pancreatitis in mice. Gastroenterology.146:1097–107.

12. Zitvogel L, Kepp O, Kroemer G. (2010) Decodingcell death signals in inflammation and immunity.Cell. 140:798–804.

13. Bianchi ME. (2007) DAMPs, PAMPs and alarmins:all we need to know about danger. J. Leukoc. Biol.81:1–5.

14. Tang D, et al. (2012) PAMPs and DAMPs: signal0s that spur autophagy and immunity. Immunol.Rev. 249:158–75.

15. Lerch MM, Gorelick FS. (2013) Models of acute andchronic pancreatitis. Gastroenterology. 144:1180–93.

16. Saluja AK, Dudeja V. (2013) Relevance of animalmodels of pancreatic cancer and pancreatitis tohuman disease. Gastroenterology. 144:1194–8.

17. Sah RP, Garg P, Saluja AK. (2012) Pathogenicmechanisms of acute pancreatitis. Curr. Opin.Gastroenterol. 28:507–15.

18. Pandol SJ, Saluja AK, Imrie CW, Banks PA. (2007)Acute pancreatitis: bench to the bedside. Gas-troenterology. 132:1127–51.

19. Dawra R, et al. (2011) Intra-acinar trypsinogen ac-tivation mediates early stages of pancreatic in-jury but not inflammation in mice with acutepancreatitis. Gastroenterology. 141:2210–7.

20. Tsai K, et al. (1998) Oxidative stress: an importantphenomenon with pathogenetic significance in theprogression of acute pancreatitis. Gut. 42:850–5.

21. Frick TW. (2012) The role of calcium in acutepancreatitis. Surgery. 152:S157–63.

22. Mulder DS, Brown RA, Thompson AG, Gurd FN.(1965) Acute hypotension in the pathogenesis ofpancreatitis. Surg. Forum. 16:380–2.

23. Bhoomagoud M, et al. (2009) Reducing extracel-lular pH sensitizes the acinar cell to secreta-gogue-induced pancreatitis responses in rats.Gastroenterology. 137:1083–92.

24. Gukovskaya AS, Gukovsky I. (2011) Which wayto die: the regulation of acinar cell death in pan-creatitis by mitochondria, calcium, and reactiveoxygen species. Gastroenterology. 140:1876–80.

25. Kubisch CH, et al. (2006) Early activation of en-doplasmic reticulum stress is associated witharginine-induced acute pancreatitis. Am. J. Phys-iol. Gastrointest. Liver Physiol. 291:G238–45.

26. Steer ML, Meldolesi J, Figarella C. Pancreatitis.(1984) The role of lysosomes. Dig. Dis. Sci. 29:934–8.

27. Chen X, et al. (2002) NF-kappaB activation inpancreas induces pancreatic and systemic inflam-matory response. Gastroenterology. 122:448–57.

28. Neuhofer P, et al. (2012) Deletion of IkappaBal-pha activates RelA to reduce acute pancreatitis inmice through up-regulation of Spi2A. Gastroen-terology. 144:192–201.

29. Huang H, et al. (2012) Activation of nuclear factor-kappaB in acinar cells increases the severity of pan-creatitis in mice. Gastroenterology. 144:202–10.

30. Rakonczay Z Jr, et al. (2008) The role of NF- kappaB activation in the pathogenesis of acutepancreatitis. Gut. 57:259–67.

31. Abdulla A, Awla D, Thorlacius H, Regner S.(2011) Role of neutrophils in the activation of

R E V I E W A R T I C L E

M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4 | K A N G E T A L . | 4 7 5

trypsinogen in severe acute pancreatitis. J. Leukoc.Biol. 90:975–82.

32. Zhang H, et al. (2013) IL-6 trans-signaling pro-motes pancreatitis-associated lung injury andlethality. J. Clin. Invest. 123:1019–31.

33. Galluzzi L, et al. (2012) Molecular definitions ofcell death subroutines: recommendations of theNomenclature Committee on Cell Death. CellDeath Differ. 19:107–20.

34. Bhatia M. (2004) Apoptosis versus necrosis inacute pancreatitis. Am. J. Physiol. Gastrointest.Liver Physiol. 286:G189–96.

35. Kaiser AM, et al. (1995) Relationship betweenseverity, necrosis, and apoptosis in five models ofexperimental acute pancreatitis. Am. J. Physiol.269:C1295–304.

36. Vernon PJ, Tang D. (2013) Eat-me: autophagy,phagocytosis, and reactive oxygen species signal-ing. Antioxid. Redox. Signal 18:677–91.

37. Shimada K, et al. (2012) Oxidized mitochondrialDNA activates the NLRP3 inflammasome duringapoptosis. Immunity. 36:401–14.

38. McGill MR, et al. (2012) The mechanism underly-ing acetaminophen-induced hepatotoxicity in hu-mans and mice involves mitochondrial damageand nuclear DNA fragmentation. J. Clin. Invest.122:1574–83.

39. Imaeda AB, et al. (2009) Acetaminophen-inducedhepatotoxicity in mice is dependent on Tlr9 and theNalp3 inflammasome. J. Clin. Invest. 119:305–14.

40. Zong WX, Thompson CB. (2006) Necrotic deathas a cell fate. Genes. Dev. 20:1–15.

41. Gu H, et al. (2013) Necro-inflammatory responseof pancreatic acinar cells in the pathogenesis ofacute alcoholic pancreatitis. Cell Death Dis. 4:e816.

42. Ditsworth D, Zong WX, Thompson CB. (2007)Activation of poly(ADP)-ribose polymerase(PARP-1) induces release of the pro-inflammatorymediator HMGB1 from the nucleus. J. Biol. Chem.282:17845–54.

43. Mota RA, et al. (2005) Inhibition of poly(ADP- ribose) polymerase attenuates the severity ofacute pancreatitis and associated lung injury. Lab.Invest. 85:1250–62.

44. Criddle DN, et al. (2007) Calcium signalling andpancreatic cell death: apoptosis or necrosis? CellDeath Differ. 14:1285–94.

45. Shi C, Andersson R, Zhao X, Wang X. (2005) Po-tential role of reactive oxygen species in pancre-atitis-associated multiple organ dysfunction. Pan-creatology. 5:492–500.

46. Hengartner MO. (2000) The biochemistry of apo-ptosis. Nature. 407:770–6.

47. Riedl SJ, Shi Y. (2004) Molecular mechanisms ofcaspase regulation during apoptosis. Nat. Rev.Mol. Cell. Biol. 5:897–907.

48. Youle RJ, Strasser A. (2008) The BCL-2 proteinfamily: opposing activities that mediate celldeath. Nature Rev. Mol. Cell Biol. 9:47–59.

49. Jiang X, Wang X. (2004) Cytochrome C-mediatedapoptosis. Annu. Rev. Biochem. 73:87–106.

50. Du C, et al. (2000) Smac, a mitochondrial proteinthat promotes cytochrome c-dependent caspase

activation by eliminating IAP inhibition. Cell.102:33–42.

51. Gukovskaya AS, et al. (1996) Mechanisms of celldeath after pancreatic duct obstruction in theopossum and the rat. Gastroenterology. 110:875–84.

52. Cao Y, et al. (2007) Induction of apoptosis bycrambene protects mice against acute pancreatitisvia anti-inflammatory pathways. Am. J. Pathol.170:1521–34.

53. Booth DM, et al. (2011) Reactive oxygen speciesinduced by bile acid induce apoptosis and pro-tect against necrosis in pancreatic acinar cells.Gastroenterology. 140:2116–25.

54. Mareninova OA, et al. (2006) Cell death in pan-creatitis: caspases protect from necrotizing pan-creatitis. J. Biol. Chem. 281:3370–81.

55. Gukovskaya AS, et al. (2002) Cholecystokinin in-duces caspase activation and mitochondrial dys-function in pancreatic acinar cells. Roles in cellinjury processes of pancreatitis. J. Biol. Chem.277:22595–604.

56. Bhagat L, et al. (2002) Thermal stress-inducedHSP70 mediates protection against intrapancre-atic trypsinogen activation and acute pancreatitisin rats. Gastroenterology. 122:156–65.

57. Frossard JL, et al. (2002) Both thermal and non-thermal stress protect against caerulein inducedpancreatitis and prevent trypsinogen activationin the pancreas. Gut. 50:78–83.

58. Takeyama Y. (2005) Significance of apoptotic celldeath in systemic complications with severeacute pancreatitis. J. Gastroenterol. 40:1–10.

59. Klionsky DJ, Emr SD. (2000) Autophagy as a reg-ulated pathway of cellular degradation. Science.290:1717–21.

60. Kroemer G, Levine B. (2008) Autophagic celldeath: the story of a misnomer. Nat. Rev. Mol.Cell. Biol. 9:1004–10.

61. Thorburn J, et al. (2009) Autophagy regulates se-lective HMGB1 release in tumor cells that aredestined to die. Cell Death Differ. 16:175–83.

62. Levine B, Mizushima N, Virgin HW. (2011) Au-tophagy in immunity and inflammation. Nature.469:323–35.

63. Hou W, et al. (2013) Strange attractors: DAMPsand autophagy link tumor cell death and immu-nity. Cell Death Dis. 4:e966.

64. Zhang Q, et al. (2013) DAMPs and autophagy:cellular adaptation to injury and unscheduledcell death. Autophagy. 9:451–8.

65. Gukovsky I, et al. (2013) Inflammation, au-tophagy, and obesity: common features in thepathogenesis of pancreatitis and pancreatic can-cer. Gastroenterology. 144:1199–209.

66. Gukovsky I, et al. (2012) Impaired autophagy andorganellar dysfunction in pancreatitis. J. Gas-troenterol. Hepatol. 27 Suppl 2:27–32.

67. Mareninova OA, et al. (2009) Impaired au-tophagic flux mediates acinar cell vacuole forma-tion and trypsinogen activation in rodent modelsof acute pancreatitis. J. Clin. Invest. 119:3340–55.

68. Hashimoto D, et al. (2008) Involvement of au-tophagy in trypsinogen activation within the

pancreatic acinar cells. J. Cell Biol. 181:1065–72.69. Gukovskaya AS, Gukovsky I. (2012) Autophagy

and pancreatitis. Am. J. Physiol. Gastrointest. LiverPhysiol. 303:G993-G1003.

70. Fortunato F, et al. (2009) Impaired autolysosomeformation correlates with Lamp-2 depletion: roleof apoptosis, autophagy, and necrosis in pancre-atitis. Gastroenterology. 137:350–60, 360 e1–5.

71. Grasso D, et al. (2011) Zymophagy, a novel selec-tive autophagy pathway mediated by VMP1-USP9x-p62, prevents pancreatic cell death. J. Biol.Chem. 286:8308–24.

72. Ropolo A, et al. (2007) The pancreatitis-inducedvacuole membrane protein 1 triggers autophagyin mammalian cells. J. Biol. Chem. 282:37124–33.

73. Molejon MI, et al. (2013) The VMP1-Beclin 1 interac-tion regulates autophagy induction. Sci. Rep. 3:1055.

74. Kang R, Zeh HJ, Lotze MT, Tang D. (2011) TheBeclin 1 network regulates autophagy and apo-ptosis. Cell Death Differ. 18:571–80.

75. Ohmuraya M, et al. (2012) Role of intrapancreaticSPINK1/Spink3 expression in the developmentof pancreatitis. Front Physiol. 3:126.

76. Linkermann A, Green DR. (2014) Necroptosis.N. Engl. J. Med. 370:455–65.

77. Vandenabeele P, Galluzzi L, Vanden Berghe T,Kroemer G. (2010) Molecular mechanisms ofnecroptosis: an ordered cellular explosion. Nat.Rev. Mol. Cell. Biol. 11:700–14.

78. Sun L, et al. (2012) Mixed lineage kinase domain-like protein mediates necrosis signaling down-stream of RIP3 kinase. Cell. 148:213–27.

79. Degterev A, et al. (2005) Chemical inhibitor ofnonapoptotic cell death with therapeutic poten-tial for ischemic brain injury. Nat. Chem. Biol.1:112–9.

80. He S, et al. (2009) Receptor interacting protein ki-nase-3 determines cellular necrotic response toTNF-alpha. Cell. 137:1100–11.

81. Zhang DW, et al. (2009) RIP3, an energy metabo-lism regulator that switches TNF-induced celldeath from apoptosis to necrosis. Science.325:332–6.

82. Duprez L, et al. (2011) RIP kinase-dependentnecrosis drives lethal systemic inflammatory re-sponse syndrome. Immunity. 35:908–18.

83. Kaczmarek A, Vandenabeele P, Krysko DV. (2013)Necroptosis: the release of damage-associatedmolecular patterns and its physiological rele-vance. Immunity. 38:209–23.

84. Linkermann A, et al. (2012) Dichotomy betweenRIP1- and RIP3-mediated necroptosis in tumornecrosis factor-alpha-induced shock. Mol. Med.18:577–86.

85. Brennan MA, Cookson BT. (2000) Salmonella in-duces macrophage death by caspase-1-dependentnecrosis. Mol. Microbiol. 38:31–40.

86. Bergsbaken T, Fink SL, Cookson BT. (2009) Py-roptosis: host cell death and inflammation. Nat.Rev. Microbiol. 7:99–109.

87. Huang H, et al. (2013) Histones activate the NLRP3inflammasome in Kupffer cells during sterile in-flammatory liver injury. J. Immunol. 191:2665–79.

4 7 6 | K A N G E T A L . | M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4

D A M P s A N D A C U T E P A N C R E A T I T I S

88. Xu J, et al. (2014) Macrophage endocytosis ofhigh-mobility group box 1 triggers pyroptosis.Cell Death Differ. 21:1229-39.

89. Huang H, et al. (2011) Endogenous histonesfunction as alarmins in sterile inflammatoryliver injury through Toll-like receptor 9 in mice.Hepatology. 54:999–1008.

90. Schroder K, Tschopp J. (2010) The inflamma-somes. Cell. 140:821–32.

91. Ivanov S, et al. (2007) A novel role for HMGB1in TLR9-mediated inflammatory responses toCpG-DNA. Blood. 110:1970–81.

92. Yang H, et al. (2010) A critical cysteine is re-quired for HMGB1 binding to Toll-like receptor4 and activation of macrophage cytokine re-lease. Proc. Natl. Acad. Sci. U. S. A. 107:11942–7.

93. Liu L, et al. (2014) HMGB1-DNA complex-in-duced autophagy limits AIM2 inflammasomeactivation through RAGE. Biochem. Biophys. Res.Commun. 450:851–6.

94. Rathinam VA, Vanaja SK, Fitzgerald KA. (2012)Regulation of inflammasome signaling. Nat. Im-munol. 13:333–2.

95. Kang R, Tang D. (2012) PKR-dependent inflam-matory signals. Sci. Signal. 5:pe47.

96. Krysko DV, et al. (2011) Emerging role of damage-associated molecular patterns derivedfrom mitochondria in inflammation. TrendsImmunol. 32:157–64.

97. Kang R, et al. (2014) The HMGB1/RAGE in-flammatory pathway promotes pancreatictumor growth by regulating mitochondrialbioenergetics. Oncogene. 33:567–77.

98. Lange SS, Mitchell DL, Vasquez KM. (2008) Highmobility group protein B1 enhances DNA repairand chromatin modification after DNA damage.Proc. Natl. Acad. Sci. U. S. A. 105:10320–5.

99. Tang D, et al. (2010) Endogenous HMGB1 regu-lates autophagy. J. Cell Biol. 190:881–92.

100. Wang H, et al. (1999) HMG-1 as a late mediator ofendotoxin lethality in mice. Science. 285:248–51.

101. Scaffidi P, Misteli T, Bianchi ME. (2002) Releaseof chromatin protein HMGB1 by necrotic cellstriggers inflammation. Nature. 418:191–5.

102. Kazama H, et al. (2008) Induction of immuno-logical tolerance by apoptotic cells requires caspase-dependent oxidation of high-mobilitygroup box-1 protein. Immunity. 29:21–32.

103. Lu B, et al. (2012) Novel role of PKR in inflam-masome activation and HMGB1 release. Nature.488:670–4.

104. Tang D, et al. (2010) Endogenous HMGB1 regu-lates autophagy. J. Cell Biol. 190:881–892.

105. Kang R, et al. (2013) HMGB1 in cancer: good,bad, or both? Clin, Cancer Res. 19:4046–57.

106. Venereau E, et al. (2012) Mutually exclusiveredox forms of HMGB1 promote cell recruitmentor proinflammatory cytokine release. J. Exp. Med.209:1519–28.

107. Tang D, et al. (2010) HMGB1 release and redoxregulates autophagy and apoptosis in cancercells. Oncogene. 29:5299–310.

108. Tang D, Billiar TA, Lotze MT. (2012) A Janus

tale of two active HMGB1 redox states. Mol.Med. 18:1360–2.

109. Kang R, et al. (2014) HMGB1 in health and dis-ease. Mol. Aspects Med. 2014, July 8 [Epub aheadof print].

110. Yasuda T, et al. (2006) Significant increase ofserum high-mobility group box chromosomalprotein 1 levels in patients with severe acutepancreatitis. Pancreas. 33:359–63.

111. Yang ZY, et al. (2008) Delayed ethyl pyruvatetherapy attenuates experimental severe acutepancreatitis via reduced serum high mobilitygroup box 1 levels in rats. World J. Gastroenterol.14:4546–50.

112. Yuan H, et al. (2009) Protective effect of HMGB1a box on organ injury of acute pancreatitis inmice. Pancreas. 38:143–8.

113. Sawa H, et al. (2006) Blockade of high mobilitygroup box-1 protein attenuates experimental se-vere acute pancreatitis. World J. Gastroenterol.12:7666–70.

114. Zhang ZW, et al. (2010) Antioxidant inhibitsHMGB1 expression and reduces pancreas injuryin rats with severe acute pancreatitis. Dig. Dis.Sci. 55:2529–36.

115. Hagiwara S, et al. (2009) Antithrombin III pre-vents cerulein-induced acute pancreatitis inrats. Pancreas. 38:746–51.

116. Hagiwara S, et al. (2009) Danaparoid sodiumprevents cerulein-induced acute pancreatitis inrats. Shock. 32:94–9.

117. Tang D, et al. (2014) High mobility group box 1(HMGB1) phenotypic role revealed with stress.Mol. Med. 2014;20:359–62.

118. Kaushik S, Cuervo AM. (2012) Chaperone- mediated autophagy: a unique way to enter thelysosome world. Trends Cell Biol. 22:407–17.

119. Tang D, et al. (2011) High-mobility group box 1is essential for mitochondrial quality control.Cell Metab. 13:701–11.

120. Li S, et al. (2011) Heat shock protein 72 enhancesautophagy as a protective mechanism inlipopolysaccharide-induced peritonitis in rats.Am. J. Pathol. 179:2822–34.

121. Ye Z, Gan YH. (2007) Flagellin contaminationof recombinant heat shock protein 70 is respon-sible for its activity on T cells. J. Biol. Chem.282:4479–84.

122. Gao B, Tsan MF. (2003) Endotoxin contamina-tion in recombinant human heat shock protein70 (Hsp70) preparation is responsible for the in-duction of tumor necrosis factor alpha releaseby murine macrophages. J. Biol. Chem. 278:174–9.

123. Tashiro M, et al. (2001) Arginine induced acutepancreatitis alters the actin cytoskeleton and in-creases heat shock protein expression in ratpancreatic acinar cells. Gut. 49:241–50.

124. Ethridge RT, et al. (2000) Acute pancreatitis re-sults in induction of heat shock proteins 70 and27 and heat shock factor-1. Pancreas. 21:248–56.

125. Feng JY, Li YY. (2010) Alteration and role ofheat shock proteins in acute pancreatitis. J. Dig.Dis. 11:277–83.

126. Lee HS, et al. (2000) Water immersion stress in-duces heat shock protein 60 expression and pro-tects against pancreatitis in rats. Gastroenterol-ogy. 119:220–9.

127. Kubisch C, et al. (2004) Overexpression of heatshock protein Hsp27 protects against cerulein-in-duced pancreatitis. Gastroenterology. 127:275–86.

128. Mahajan UM, et al. (2011) Alteration in inflam-matory/apoptotic pathway and histone modifi-cations by nordihydroguaiaretic acid preventsacute pancreatitis in Swiss albino mice. Apopto-sis. 16:1138–49.

129. Kim JN, et al. (2011) Heat shock proteins andautophagy in rats with cerulein-induced acutepancreatitis. Gut Liver. 5:513–20.

130. Song JM, et al. (2008) Extracellular heat-shockprotein 70 aggravates cerulein-induced pancre-atitis through toll-like receptor-4 in mice. Chin.Med. J. (Engl). 121:1420–5.

131. Liao WC, et al. (2009) Serum heat shock protein27 is increased in chronic pancreatitis and pan-creatic carcinoma. Pancreas. 38:422–6.

132. Pakendorf B, Stoneking M. (2005) Mitochondr-ial DNA and human evolution. Annu. Rev. Ge-nomics Hum. Genet. 6:165–83.

133. Sharma S, Fitzgerald KA. (2011) Innate immunesensing of DNA. PLoS. Pathog. 7:e1001310.

134. Gornik O, et al. (2011) Evaluation of cell-freeDNA in plasma and serum as early predictorsof severity in acute pancreatitis. Pancreas.40:787–8.

135. Kocsis AK, et al. (2009) Plasma concentrations ofhigh-mobility group box protein 1, soluble re-ceptor for advanced glycation end-productsand circulating DNA in patients with acutepancreatitis. Pancreatology. 9:383–91.

136. Gornik I, et al. (2009) Free serum DNA is anearly predictor of severity in acute pancreatitis.Clin. Biochem. 42:38–43.

137. Bagul A, et al. (2006) Quantitative analysis ofplasma DNA in severe acute pancreatitis. JOP.7:602–7.

138. Zeng YJ, et al. (2008) Toll-like receptor 9 is ex-pressed in rat pancreas and is involved incerulein-induced pancreatitis. Pancreas.36:212–4.

139. Tian J, et al. (2007) Toll-like receptor 9-dependentactivation by DNA-containing immune com-plexes is mediated by HMGB1 and RAGE. Nat.Immunol. 8:487–96.

140. Julian MW, et al. (2012) Mitochondrial transcrip-tion factor A serves as a danger signal by aug-menting plasmacytoid dendritic cell responsesto DNA. J. Immunol. 189:433–43.

141. Park H, Boyington JC. (2010) The 1.5 A crystalstructure of human receptor for advanced gly-cation endproducts (rage) ectodomains revealsunique features determining ligand binding. J. Biol. Chem. 285:40762-70.

142. Urbonaviciute V, et al. (2008) Induction of in-flammatory and immune responses by HMGB1-nucleosome complexes: implications for thepathogenesis of SLE. J. Exp. Med. 205:3007–18.

R E V I E W A R T I C L E

M O L M E D 2 0 : 4 6 6 - 4 7 7 , 2 0 1 4 | K A N G E T A L . | 4 7 7

143. Campos EI, Reinberg D. (2009) Histones: anno-tating chromatin. Annu. Rev. Genet. 43:559–99.

144. Xu J, et al. (2009) Extracellular histones aremajor mediators of death in sepsis. Nat. Med.15:1318–21.

145. Xu J, et al. (2011) Extracellular histones are me-diators of death through TLR2 and TLR4 inmouse fatal liver injury. J. Immunol. 187:2626–31.

146. Wu D, et al. (2002) Apoptotic release of histonesfrom nucleosomes. J. Biol. Chem. 277:12001–8.

147. Brinkmann V, et al. (2004) Neutrophil extracellu-lar traps kill bacteria. Science. 303:1532–5.

148. Allam R, Darisipudi MN, Tschopp J, Anders HJ.(2013) Histones trigger sterile inflammation byactivating the NLRP3 inflammasome. Eur. J. Im-munol. 43:3336–42.

149. Allam R, Kumar SV, Darisipudi MN, AndersHJ. (2014) Extracellular histones in tissue injuryand inflammation. J. Mol. Med. 92:465–72.

150. Chen R, Kang R, Fan XG, Tang D. (2014) Re-lease and activity of histone in diseases. CellDeath Dis. 5:e1370.

151. Gordon JL. (1986) Extracellular ATP: effects,sources and fate. Biochem. J. 233:309–19.

152. Surprenant A, et al. (1996) The cytolytic P2Z re-ceptor for extracellular ATP identified as a P2Xreceptor (P2X7). Science. 272:735–8.

153. Chekeni FB, et al. (2010) Pannexin 1 channels me-diate ‘find-me’ signal release and membrane per-meability during apoptosis. Nature. 467:863–7.

154. Valera S, et al. (1994) A new class of ligand-gated ion channel defined by P2x receptor forextracellular ATP. Nature. 371:516–9.

155. Feranchak AP, et al. (2010) Initiation of purinergicsignaling by exocytosis of ATP-containing vesi-cles in liver epithelium. J. Biol. Chem. 285:8138–47.

156. Fader CM, Aguilera MO, Colombo MI. (2012)ATP is released from autophagic vesicles to theextracellular space in a VAMP7-dependentmanner. Autophagy. 8:1741-56.

157. Criddle DN, et al. (2006) Fatty acid ethyl esterscause pancreatic calcium toxicity via inositoltrisphosphate receptors and loss of ATP synthe-sis. Gastroenterology. 130:781–93.

158. Suadicani SO, Brosnan CF, Scemes E. (2006)P2X7 receptors mediate ATP release and ampli-fication of astrocytic intercellular Ca2+ signal-ing. J. Neurosci. 26:1378–85.

159. Awla D, Abdulla A, Regner S, Thorlacius H.(2011) TLR4 but not TLR2 regulates inflamma-tion and tissue damage in acute pancreatitis in-duced by retrograde infusion of taurocholate.Inflamm. Res. 60:1093–8.

160. Sharif R, et al. (2009) Impact of toll-like receptor4 on the severity of acute pancreatitis and pancreatitis-associated lung injury in mice. Gut.58:813–9.

161. Pastor CM, et al. (2004) Role of Toll-like receptor4 on pancreatic and pulmonary injury in a micemodel of acute pancreatitis associated with en-dotoxemia. Crit. Care Med. 32:1759–63.

162. Yang L, et al. (2014) PKM2 regulates the War-burg effect and promotes HMGB1 release insepsis. Nat. Commun. 2014;5:4436.