Sepsis primer

Concise Definitive Review

Molecular biology of inflammation and sepsis: A primer*

Ismail Cinel, MD, PhD; Steven M. Opal, MD

I n sepsis, the expected and appropri-ate inflammatory response to an in-fectious process becomes amplifiedleading to organ dysfunction or risk

for secondary infection. A continuum existsfrom a low grade systemic response associ-ated with a self-limited infection to amarked systemic response with solitary ormultiorgan dysfunction, i.e., severe sepsis.As a clinical syndrome, sepsis occurs whenan infection is associated with the systemicinflammatory response.

The complex toll-like receptor signalingand associated downstream regulators ofimmune cell functions play a crucial role inthe innate system as a first line of defenseagainst pathogens (1). However, signalingis sometimes conflicting and a sustainedinflammatory response can result in tissue

damage. In addition, a reduction in theirantimicrobial capacity may set the stage foropportunistic and/or super infections (2).Severe sepsis may also be associated withan exaggerated procoagulant state. Thismay lead to ischemic cell injury, an effectthat further amplifies the damage causedby inappropriate inflammation. A micro-vasculature injured by inflammation andischemia, in turn, further deranges thehost response by altering leukocyte traffick-ing, generating apoptotic microparticles,and increasing cellular hypoxia (3). Mito-chondrial dysfunction, an acquired intrin-sic defect in cellular respiration termed “cy-topathic hypoxia,” also has an importantrole by decreasing cellular oxygen con-sumption in this chaotic process. In thisreview, we highlight the current under-standing of the basic molecular mecha-nisms that modulate these events so as toproduce sepsis.

Pattern Recognition Receptors,Pathogen-Associated MolecularPatterns (PAMPs) and Danger-Associated Molecular Patterns(DAMPs)

The initiation of the host responseduring sepsis or tissue injury involvesthree families of pattern recognition re-ceptors (PRRs): 1) toll-like receptors

(TLRs); 2) nucleotide-oligomerizationdomain leucine-rich repeat (NOD-LRR)proteins; and 3) cytoplasmic caspase ac-tivation and recruiting domain heli-cases such as retinoic-acid-induciblegene I (RIG-I)-like helicases (RLHs) (4,5). These receptors initiate the innateimmune response and regulate theadaptive immune response to infectionor tissue injury.

Gram-positive and Gram-negative bac-teria, viruses, parasites, and fungi all pos-sess a limited number of unique cellularconstituents not found in vertebrate ani-mals. These elements are now referred to asPAMPs, or more appropriately microbial-associated molecular patterns, as thesemolecules are also common in nonpatho-genic and commensal bacteria (6). PAMPsbind to PRRs, such as TLRs, expressed onthe surface of host cells. Cytoplasmic PRRsexist to detect invasive intracellular patho-gens (7). The NOD proteins recognize com-mon fragments of bacterial peptidoglycan.Diamino-pimelate from Gram-negativebacteria is the ligand for NOD1 and mu-ramyl dipeptide from peptidoglycan is theligand for NOD2 in the cytosol (see Figs. 1and 2). The PRRs also recognized damagesignals from the release of endogenous pep-tides and glycosaminoglycans from apopto-tic or necrotic host cells (8–10) Caspaseactivation and recruiting domain helicases

From the Division of Critical Care Medicine (IC), TheRobert Wood Johnson School of Medicine, The Universityof Medicine and Dentistry of New Jersey, Camden NJ;and The Infectious Disease Division (SMO), MemorialHospital of RI, The Warren Alpert School of Medicine ofBrown University, Providence, RI.

Dr. Opal has received grants from Wyeth and EisaiInc. Dr. Cinel has not disclosed any potential conflictsof interest.

For information regarding this article, E-mail:[email protected]

Copyright © 2008 by the Society of Critical CareMedicine and Lippincott Williams & Wilkins

DOI: 10.1097/CCM.0b013e31819267fb

Background: Remarkable progress has been made during thelast decade in defining the molecular mechanisms that underlieseptic shock. This rapidly expanding field is leading to newtherapeutic opportunities in the management of severe sepsis.

Aim: To provide the clinician with a timely summary of themolecular biology of sepsis and to better understand recentadvances in sepsis research.

Data Selection: Medline search of relevant publications inbasic mechanisms of sepsis/severe sepsis/septic shock, and se-lected literature review of other manuscripts about the signalo-some, inflammasome, apoptosis, or mechanisms of shock.

Data Synthesis and Findings: The identification of the toll-likereceptors and the associated concept of innate immunity based uponpathogen- or damage-associated molecular pattern molecules al-lowed significant advances in our understanding of the pathophys-iology of sepsis. The essential elements of the inflammasome andsignal transduction networks responsible for activation of the hostresponse have now been characterized. Apoptosis, mitochondrial

dysfunction, sepsis-related immunosuppression, late mediators ofsystemic inflammation, control mechanisms for coagulation, andreprogramming of immune response genes all have critical roles inthe development of sepsis.

Conclusions: Many of these basic discoveries have direct impli-cations for the clinical management of sepsis. The translation ofthese “bench-to-bedside” findings into new therapeutic strategies isalready underway. This brief review provides the clinician with aprimer into the basic mechanisms responsible for the molecularbiology of sepsis, severe sepsis, and septic shock.(Crit Care Med2009; 37:291–304)

KEY WORDS: sepsis; inflammation; toll like receptors; inflamma-some; signalosome; apoptosis; neutrophils; mitochondrial dys-function; reactive oxygen species; nitric oxide; peroxynitrite; ac-tivated protein C; a disintegrin-like and metalloproteinase withthrombospondin type-1 motifs-13; pathogen associated molecularpatterns; danger associated molecular patterns

291Crit Care Med 2009 Vol. 37, No. 1

primarily recognize viral nucleic acids andactivate antiviral measures including thetype I interferons.

TLR expression is significantly up-regulated in experimental models of sep-sis and in patients with sepsis (11–14).Trauma including thermal injury gener-

ates danger-associated molecular pat-terns (i.e., high mobility group box-1,heat shock proteins, S100 proteins, hya-luran, etc.) that augment TLR expressionlike PAMPs. It also primes the innate im-mune system for enhanced TLR reactiv-ity, resulting in excess lipopolysaccharide

(LPS)-induced mortality (15). Multiplepositive feedback loops between danger-associated molecular patterns andPAMPs, and their overlapping receptorstemporally and spatially drive these pro-cesses and may represent the molecularbasis for the observation that infections,as well as nonspecific stress factors, cantrigger flares in systemic inflammatoryresponse. The degree to which TLR reg-ulation mediates the ultimate outcome insepsis in individual patients remains elu-sive and is an active area of clinical in-vestigation (13, 16).

Inflammasome and SignalosomePathways

TLRs induce pro-interleukin(IL)-1betaproduction and prime NLR-containingmultiprotein complexes, termed “inflam-masomes,” to respond to bacterial prod-ucts and products of damaged cells (17,19). This results in caspase-1 activationand the subsequent processing of pro-IL-1� to its active extracellular formIL-1� (19). Caspases are a set of cysteineproteases that alter the enzymatic activityof target proteins at specific peptide se-quences adjacent to aspartate moieties.Caspases are important in process of ap-optosis, cellular regulation, and inflam-mation (Figs. 1 and 3). One of the manytargets of the caspase cascade is caspaseactivated DNase (CAD). CAD activationinduces DNA fragmentation characteris-tic of programmed cell death (apoptosis).

The posttranslational activation ofcaspase-1 is tightly regulated by inflam-masome, the known components ofwhich include caspase-1, ASC (apoptosis-associated speck-like protein containing acaspase activation and recruiting do-main), NALP1 (NACHT, leucine rich re-peat and pyrin domain containing 1), andcaspase-5 (20). Additionally, alternate in-flammasome constructions have beensuggested to contain pyrin, NALP3, andother members of the NOD-LRR family(21–23). ASC facilitates inflammasomeassembly thus triggering caspase-1 acti-vation and IL-1� processing. Interest-ingly, ASC can also regulate the nuclearfactor (NF)-�B pathway, thus linking theinflammasome to the signalosome(Fig. 1) (24).

The inflammasome pathways contrib-ute to the inflammatory response in sep-sis (25). Caspase-1 knock out mice areprotected from sepsis (26) while a natu-rally occurring polymorphism for humancaspase-12, a putative regulator of

Figure 1. Specific host immune response to each pathogen is mediated by various sets of pathogenassociated molecular patterns (PAMPs) and pattern recognition receptors (PRRs) as detailed in Figure1. PRRs are essential for initiating the host’s immune defenses against invading pathogens, yet theycan also contribute to persistent and deleterious systemic inflammation. PRRs also serve as receptorsfor endogenous danger signals, hemodynamic changes in sepsis (tissue hypoperfusion and ischemia/reperfusion phenomenon), thus same signaling systems that alerts the highly advantageous, hostdefense mechanisms, also contributes to the disadvantageous, pathologic events of systemic inflam-mation, coagulation, tissue damage in target organs in sepsis. Heat shock proteins, fibrinogen,fibronectin, hyaluran, biglycans and high mobility group box-1 (HMGB-1) have been defined as dangerassociated molecular patterns (DAMPs) which are likely relevant for sepsis. Toll-like receptors (TLRs),especially TLR4, are involved in the recognition of these endogenous or harmful self-antigens ligandswhich are released during noninfectious injury, such as trauma or ischemia/reperfusion suggestingtheir function may not be restricted to the recognition of extrinsic pathogens. NOD-LRR, nucleotide-oligomerization domain leucine-rich repeat; ASC, apoptosis-associated speck-like protein containingcaspase activation and recruiting domain; NF-��, nuclear factor ��.

Figure 2. Binding of toll-like receptors (TLRs) activates intracellular signal-transduction pathways thatlead to the activation of transcriptional activators such as interferon regulator factors, phosphoino-sitide 3-kinase (PI3K)/Akt, activator protein-1, and cytosolic nuclear factor-kappa � (NF-��). ActivatedNF-�� moves from the cytoplasm to the nucleus, binds to transcription sites and induces activationof an array of genes for acute phase proteins, inducible nitric oxide synthase, coagulation factors,proinflammatory cytokines, as well as enzymatic activation of cellular proteases. TLR9 DNA, TLR 3dsRNA, and TLR7/8 ss RNA are endosomal. TLR 10 ligand is not defined and TLR1 forms heterodimerswith TLR2. LPS, lipopolysaccharide; IRF, interferon regulatory factor; JNK; c Jun N-terminal kinase.

292 Crit Care Med 2009 Vol. 37, No. 1

caspase-1, has been linked to sepsis (27,28). Thus, caspase-1 activation appears tobe a prerequisite for a competent im-mune response (29). Recently, the in-flammasome components have beenshown to be significantly lower in septicshock patients during the early stages ofsystemic inflammatory response with el-evated plasma cytokines levels (30). This

monocyte deactivation process may bemaladaptive in the later phases of sepsisand predispose to secondary infection.

Caspase-1 and its proinflammatory cy-tokine products are likely to contribute tothe pathogenesis of sepsis in overwhelm-ing inflammation. However, like manyother essential elements of innate immu-nity, caspase-1 also has a positive impact

on host defense against several infectionsup-regulating microbial killing mecha-nisms such as the production of reactiveoxygen and nitrogen species (ROS andRNS) (31, 32). Negative regulation ofTLRs and TLR-induced programmed celldeath has been defined (7). Apoptosis in-duction is used by Pseudomonas aerugi-nosa to inhibit the secretion of immuneand proinflammatory mediators by targetcells (33).

MyD88 protein and IL-1 receptor-associated kinase pathways activateNF-�B during the innate immune re-sponse (34). NF-�B is a transcription fac-tor that is constituted by homo- or het-erodimers of the Rel protein family with apivotal role in inflammation, cell sur-vival, and proliferation. In unstimulatedcells, NF-�B is maintained in a latentform in the cytoplasm by means of se-questration by inhibitory �B (I�B) pro-teins. NF-�B activating stimuli, such ascytokines, viruses, and lipopolysaccharide(LPS), induce the degradation of inhibi-tory �Bs by the proteasome, unmaskingthe nuclear localization signal of NF-�B,resulting in its nuclear translocation,binding to NF-�B motifs, and gene tran-scription. Dynamic redox control of NF-�B through glutaredoxin-regulated S-glutathionylation of inhibitory �Bs hasbeen demonstrated (35). NF-�B, subjectto regulation by redox changes, has beenshown to be involved in the transcrip-tional regulation of more than 150 geneswith a significant portion demonstratingproinflammatory properties (36). NF-�Bis readily activated upon intraperitoneallyLPS challenge within 4 hrs in lung, liver,and spleen (37). A variant IL-1 receptor-associated kinase �1 haplotype has re-cently been demonstrated to affect themagnitude of NF-�B activation and di-rectly correlates with an increased inci-dence of septic shock and significantlyreduced survival rates (38). The impact ofNF-�B signaling is tissue-specific as defi-cient NF-�B activation in intestinal epi-thelium is associated with increased in-flammation in vivo (39, 40). Both studiesdemonstrate that defects of NF-�B signal-ing cause immunosuppression whichtriggers and maintains inflammation.Thus, it can be suggested that inhibitionof massive NF-�B activation in vivo leadsto reduced inflammatory responses, atleast during certain phases and certaintissues (i.e., parenchymal tissues) in sep-sis. On the other hand, this activation isfollowed by negative NF-�B regulationfavoring apoptosis in immune cells which

Figure 3. Pathogenic mechanisms during sepsis or in response to tissue injury can lead to organdysfunction. PRRs, pattern recognition receptors; TLRs, toll-like receptors; NOD-LRR, nucleotide-oligomerization domain leucine-rich repeat protein receptors; RLHs; retinoic-acid-inducible gene I(RIG-I)-like helicases; TNF-�, tumor necrosis factor alpha; IL-1, interleukin 1; HMGB-1, high mobilitygroup box-1; LPS, lipopolysaccharide; LTA, lipoteichoic acid; PGN, peptidoglycan; VSMCs, vascularsmooth muscle cells; RBC, red blood cell; MPO, myeloperoxidase; XOR, xanthine oxidoreductase;iNOS, inducible nitric oxide synthase; COX-2, cyclooxygenase; RAGE, receptor for advanced glycationend products; ROCK; RhoA/Rho kinase; PARP-1, poly(ADP ribose) polymerase-1; PAR, proteaseactivated receptor; ROS/RNS, reactive oxygen and nitrogen species; NF, nuclear factor; NADPH,nicotimanide adenosine dinucleotide phosphate; PAMPs, pathogen-associated molecular patterns;DAMPs, danger-associated molecular patterns.


may lead to immunosuppression and fataloutcome in severe sepsis. Recently, usingspecific gene-targeted deletions, it hasbeen shown that deletion of MyD88caused a worsened survival in a model ofsevere peritonitis, despite the marked de-crease in sepsis-induced T and B lympho-cyte apoptosis suggesting MyD88, likeNF-�B, is also critical host survival insepsis (41).

Toll-Like Receptor Signaling

The Role of Phosphoinositide 3-Kinase.Phosphoinositide 3-kinase (PI3K), a sig-nal transduction enzyme, and the down-stream serine/threonine kinase Akt (alsoknown as protein kinase B) have beenreported in cellular activation, inflamma-tory responses, chemotaxis, and apoptosis(Figs. 2 and 3) (42). PI3K can functioneither as a positive or negative regulatorof TLR signaling. PI3Ks (three types) actat several steps downstream of TLRs, de-pending on the cell type and/or the en-gagement of a specific TLR regulatingdownstream signaling to NF-KB transac-tivation or to mitogen-activated phos-phokinase activation (43, 44). As a posi-tive mediator of TLR signaling, PI3Ktogether with p38 and extracellular reg-ulated kinase (ERK)1/2 mitogen-acti-vated phosphokinases, lead to productionof proinflammatory cytokines IL-1�,IL-6, and IL-8 upon microbial challenge(44, 45). In detail, PI3K activation ap-peared to have a significantly promotingfunction for these mediators in mono-cytes, whereas activation appeared tolimit the LPS response for generation ofthese cytokines in neutrophils (46). How-ever, PI3K inhibition resulted in im-paired oxidative burst and phagocytosisactivity in both neutrophils and mono-cytes. By limiting C5a-mediated effectson neutrophil cytokine generation, andpromoting oxidative burst and phagocy-tosis, PI3K activation seems to be a ther-apeutic approach for limiting inflamma-tion in sepsis.

On the other hand, PI3K/Akt signalingpathway acts as an endogenous negativefeedback mechanism that serves to limitproinflammatory and apoptotic events asseen in monocytes in response to endo-toxin (47). The suppressive role in in-flammation and coagulation was con-firmed in mouse model of endotoxemia(48). It can also promote the generationof anti-inflammatory cytokine IL-10 (49).PI3K/Akt has an effect in balancing Th1vs. Th2 responses (50). Overexpression of

Akt in lymphocytes decreases lymphocyteapoptosis, a Th1 cytokine propensity, andimprove outcome in cecal ligation andpuncture-induced sepsis (51). Peptide-mediated activation of Akt and extracel-lular regulated kinase signaling protectslymphocytes from numerous apoptoticstimuli both in vitro and in vivo (52).This suggests again the survival advan-tage of PI3K/Akt pathway activation forthe later stages of sepsis.

The Role of Rho GTPases. The Rhofamily of small GTPases is one of themaster regulators of cell motility, as theycontrol actin cytoskeleton remodeling.RhoA, Rac1, and CDc42 are the wellknown family members which act molec-ular switches regulating responses of in-nate immune cells related to pathogensensing, intracellular uptake, and de-struction (53). These GTPases play bothunique and overlapping roles in phago-cyte functions including migration, che-motaxis, and optimal bacterial killing(54). It has been shown that PAMPs mayutilize GTPases through TLRs (i.e.,TLR-2, -4, -3, and -9) in which Rac1 ac-tivation is required for PI3K activationupon TLR2 stimulation (55). RhoA regu-lates not only cytoskeletal events, whichmediate neutrophil migration, but alsocontributes to NF-��-dependent proin-flammatory gene transcription (Fig. 3)(56). It has been suggested that ROCKinhibition could attenuate cytoskeletalrearrangement of endothelial cells, lead-ing to decreased neutrophil emigrationinto the lung parenchyma in LPS-induced lung injury (57). An importantrole for rho kinase in leukocyte recruit-ment is also supported in an endotoxemicliver injury model (58). The recentlyemerged connections between TLR sig-naling and small Rho-GTPases seem toprovide new therapeutic avenue of re-search in sepsis.

Toll-Like Receptor Signalingand Regulatory T Cells (Tregs)

To balance self-tolerance and immu-nity against pathogens, the immune sys-tem depends on both up-regulatory anddown-regulatory mechanisms. Recentstudies have suggested that several lym-phocyte subpopulations (i.e., CD4�CD25�Foxp3� T regulatory-cell �Tregs�) mayhave the capacity to actively suppress anadaptive immune response and may po-tentially be involved in septic immunedysfunction. Naturally occurring Tregsexpress the transcription factor forkhead

box protein (FoxP3), which is induced bythe anti-inflammatory cytokine trans-forming growth factor-� (59, 60). TLRtriggering induces dendritic cell matura-tion also, which is essential for the induc-tion of adaptive immune responses (61).Studies have recently highlightened theimportance of TLRs on Tregs (62). In thisregard, the dominant role of TLR2 signal-ing on the Treg-mediated immune sup-pression has been demonstrated (63, 64).

Tregs control inflammatory reactionsto commensal bacteria and opportunistpathogens and play a major role in sup-pressing immune reactivity, rangingfrom autoimmunity to infectious disease(65) and to injury (66). Monneret et al(67) observed that sepsis increasesCD4�CD25� T cells in the peripheralblood of septic patients. This was subse-quently found to be a relative increase inTregs due to a decrease in the CD4�CD25–T effector cell populations (68). Further-more, Treg-mediated induction of “alter-natively activated” macrophages has beendemonstrated suggesting as one of thecauses of immune dysfunction in sys-temic inflammatory response syndromeand sepsis (69, 70). Targeting apoptosis ofTregs may be a new therapeutic approachin preventing the continuum of sepsis tosevere sepsis. However, it has been re-ported that depletion of CD25� cells be-fore inducing sepsis did not alter septicmortality pointing the need of more stud-ies to clarify the significance of this cellpopulation’s expansion in sepsis morbid-ity (71, 72).

Adoptive transfer of Tregs before orfollowing the initiation of polymicrobialsepsis improved survival by enhancingtumor necrosis factor (TNF)-� produc-tion and bacterial clearance (73). Tregsinhibit LPS-induced monocyte survivalthrough the Fas/FasL dependent pro-apoptotic mechanism, which might playa role in the resolution of exaggeratedinflammation (74). The positive and neg-ative effects of TLR on Tregs are curiousin sepsis, and the dynamics of TLR ex-pression on immune-suppressive Tregsupon inflammation or in relation to typeof pathogen are needed to illuminate.

Neutrophils andMonocyte/Macrophages inInflammation

Myeloid cells including neutrophilsand elements of monocyte/macrophagelineage are heavily armed with largestores of proteolytic enzymes and with


the capacity to rapidly generate ROS andRNS to degrade internalized pathogens.The highly proapoptotic nature of neu-trophils is designed to maintain a balancebetween antimicrobial effectiveness andthe potential for neutrophil-associateddamage to the host in septic challenge orin other injurious processes, such astrauma or ischemia/reperfusion (75).Host tissue damage in severe sepsis mayarise via a variety of mechanisms includ-ing premature neutrophil activation dur-ing migration, extracellular release of cy-totoxic molecules and toxins duringmicrobial killing, removal of infected ordamaged host cells or debris during hosttissue remodeling, and failure to termi-nate acute inflammatory responses (76).Therefore, to maximize host defense ca-pabilities while minimizing damage tohost tissues, neutrophil microbial re-sponses are tightly regulated. On theother hand, quorum sensing (the abilityof bacteria to assess their population den-sity) has been found to have a crucial rolein regulating tissue invasion by bacterialpathogens, and inhibitors of quorumsensing system provide new avenues forintervention against invasive pathogens(1). Evidence now exists that quorumsensing system can even open up bidirec-tional lines of communication betweenbacteria and the human host.

Leukocyte-endothelial interactions,which may also contribute to inflamma-tion-mediated injury, involve two sets ofadhesion molecules, selectins, and inte-grins (77, 78). P-selectin is expressed onplatelets; E- and P-selectins are expressedby endothelial cells, whereas L-selectin isexpressed on leukocytes. Selectins mediateneutrophil rolling along activated endothe-lial surfaces as circulating neutrophils de-celerate to engage endothelial receptors.The beta-2 intergrins (CD11/CD18 com-plexes) mediate tight adhesion to endothe-lial membranes, allowing subsequentegress of neutrophils to extravascular sitesof inflammation (78). Activated neutrophilsstimulate transendothelial albumin trans-port through intracellular adhesion mole-cule-1 mediated, Scr-dependent caveolinphosphorylation. The role of caveolin ininflammatory response in sepsis has beenrecently defined (Fig. 3) (79–82).

Neutrophils contribute to blood coagu-lation in localized inflammation and ingeneralized sepsis (1, 3, 75). During sys-temic inflammation, homeostatic mecha-nisms are compromised in the microcircu-lation including endothelial hyperactivity,fibrin deposition, microvascular occlusion,

and cellular exudates that further impedeadequate tissue oxygenation. Neutrophilsparticipate in these rheologic changesthrough their augmented binding to bloodvessel walls and through the formation ofplatelet-leukocyte aggregates (83). Neutro-phil elastase, other proteases, glycases andinflammatory cytokines degrade endoge-nous anticoagulant activity, and impair fi-brinolysis on endothelial surfaces favoringa procoagulant state (1).

The Role of Apoptosis. Sepsis-inducedneutrophil-mediated tissue injury hasbeen demonstrated in a variety of organsincluding the lungs (84–86), diaphragm(87), kidneys (86), intestine (88, 89), and

liver (86). Apoptosis is a counter-regula-tor of the initial inflammatory responsein sepsis (90). Neutrophils are constitu-tively proapoptotic and apoptosis is fun-damental for the resolution of inflamma-tion and cell turnover. Neutrophils canundergo apoptosis via intrinsic and ex-trinsic pathways; the latter also requiresmitochondrial amplification (Fig. 4) (91).The role played by mitochondria in theregulation of neutrophil life span is morecrucial than in other cell types in thebody (92). As neutrophils kill pathogensusing ROS and RNS and a mixture of lyticenzymes, delayed clearance of neutro-phils in sepsis can potentially contribute

Figure 4. Summary of apoptotic signaling pathways as seen through activation of death receptor(extrinsic) or mitochondrial (intrinsic) pathway. Extrinsic signals bind to their receptors and triggerintracellular signaling, leading to caspase-8 activation. Activation of caspase-8 by extrinsic stimuli(such as tumor necrosis factor-� [TNF-�], Fas ligand) involves mitochondria-dependent signaling intype II cells. In type I cells, on the other hand, execution of apoptosis occurs without significantparticipation of mitochondria. MAPK, mitogen-activated phosphokinase; PI3K/Akt, phosphoinositide3-kinase/Akt; APAF-1, apoptosis protease activating factor 1; ER, endoplasmic reticulum; IL, interleukin.

Figure 5. The recognition of apoptotic cells by macrophages is largely dependent on the cell surfaceappearance of phosphatidylserine (PS). S-nitrosylation of critical cysteine residues inhibits aminophospho-lipid translocase (APLT), leading to PS externalization. It generates an “eat me” signal during apoptosis.iNOS, inducible nitric oxide synthase; .NO, nitric oxide; O2, superoxide; ONOO�, peroxynitrite.


to cell/organ injury. Importantly, thephagocytosis of bacteria and fungi accel-erates neutrophil apoptosis. Apoptoticcell clearance induces anti-inflammatoryeffects in tissues. It has been shown thatintratracheal administration of killed E.coli attenuated lung injury and improvedsurvival in an intestinal ischemia/reper-fusion injury associated with marked pul-monary neutrophil infiltration (93).

Cytokine-induced prolonged neutro-phil survival is accompanied by evidenceof increased neutrophil activation, in-cluding augmented respiratory burst ac-tivity (90). Neutrophils from patientswith sepsis manifest markedly prolongedsurvival in vitro in association with evi-dence of cellular activation (94). Phago-cytosis of apoptotic neutrophils bymacrophages inhibits the release ofproinflammatory cytokines and promotesthe secretion of anti-inflammatory cyto-kines (95). In contrast, inefficient apopto-tic cell clearance is proinflammatory andimmunogenic (96). The recognition ofapoptotic cells by macrophages is largelydependent on the cell surface appearanceof an anionic phospholipid, phosphatidyl-serine (PS), which is normally confinedto the inner leaflet of the plasma mem-brane (97). Asymmetric distribution ofPS across the plasma membrane ismainly because of the activity of a spe-cialized enzymatic mechanism, amin-ophospholipid translocase. S-nitrosyla-tion of critical cysteine residues inhibitsaminophospholipid translocase, leadingto PS externalization. PS expression dur-ing apoptosis generates an “eat-me” sig-nal (Fig. 5), which in turn triggers clear-ance of apoptotic cells and suppresses theinflammatory response (98). It has beendemonstrated that S-nitrosylation of crit-ical cysteine residues in aminophospho-lipid translocase using a cell-permeabletransnitrosylating agent, S-nitroso-acetyl-cysteine, resulted in egression of PS tothe outer surface of the plasma mem-brane, rendering these cells recognizableby macrophages (99). The therapeutic po-tential of regulating neutrophil life-spanin sepsis remains to be determined. Caremust be exercised in regulating of thispathway, as sepsis enhances the capacityof macrophages to clear expanded apo-ptotic populations, a mechanism contrib-uting to septic immune suppression(100).

The Role of ROS and RNS. ROS andRNS exert several beneficial physiologicfunctions, such as intracellular signalingfor several cytokines and growth factors,

second messengers for hormones and re-dox regulation. Despite their importanceas a defense mechanism against invadingpathogens, an overwhelming productionof ROS and RNS or a deficit in oxidantscavenger and antioxidant defenses resultin oxidative/nitrosative stress, a key ele-ment in the deleterious processes in sep-sis (Fig. 6) (101, 102).

Stimulated neutrophils produce ROSand RNS through the nicotinamide ade-nine dinucleotide phosphate oxidasecomplex, myeloperoxidase and xanthineoxidoreductase and represent a defensemechanism against invading microor-ganisms (103). Lipopolysaccharide andother proinflammatory mediators acti-vate nicotinamide adenine dinucleotidephosphate oxidase to produce superoxideradical (O2

�). In aqueous environments,superoxide radical is rapidly catalyzed bysuperoxide dismutase hydrogen peroxide(H2O2) and hydroxyl radicals. Myeloper-oxidase from neutrophil azurophilicgranules produces hypochlorous acidfrom hydrogen peroxide (H2O2) and chlo-ride anion (Cl�) during respiratory burst.These radicals are highly cytotoxic, andneutrophils used them to kill bacteriaand other pathogens. Expression of hu-man xanthine oxidoreductase is markedlyup-regulated by hypoxia, ischemia/reper-fusion, LPS, and TNF-�. Increased activ-ity of xanthine oxide, one the important

contributors of ROS production, has beenreported in adult and pediatric patientswith sepsis (104).

O2� in the presence of nitric oxide,

generates peroxynitrite (ONOO�), a keyplayer in the pathogenesis of sepsis-induced organ dysfunction. ONOO� cancause DNA strand breakage, which trig-gers the activation of DNA repair en-zymes such as poly (adenine dinucleotidephosphate-ribose) polymerase (Fig. 3).Poly (adenine dinucleotide phosphate-ribose) polymerase inhibitors protectagainst oxidative and nitrosative stress-induced organ dysfunction in endotox-emia (87–89). Recently, the potential roleof poly (adenine dinucleotide phosphate-ribose) polymerase activation has beenimplicated in the pathogenesis of myo-cardial contractile dysfunction associatedwith human septic shock (105).

Novel Cytokines in Inflammation

High Mobility Group Box-1 and Re-ceptor for Advanced Glycation End-Products. High mobility group box-1(HMGB-1) is a nonhistone, nuclear DNA-binding protein involved in nucleosomestabilization and gene transcription. How-ever, when HMGB-1 is released in largequantities into the extracellular environ-ment, it becomes a lethal mediator of sys-temic inflammation (Fig. 3) (106). Indeed,

Figure 6. Proinflammatory/anti-inflammatory cytokines and reactive oxygen and nitrogen species haveimportant effects within the microcirculatory unit: the arteriole, endothelial cell, capillary bed, and thevenule. The arteriole is where the characteristic intractable vasodilation of sepsis occurs. The capillary bedis where the effects of endothelial cell activation/dysfunction are most pronounced and microvascularthromboses are formed. The postcapillary venule is where leukocyte trafficking is most disordered. All ofthese causes impair flow through the microcirculation leading to microcirculatory dysfunction. TLR,toll-like receptor; NOD-LRR, nucleotide-oligomerization domain leucine-rich repeat; RLH, (RIG-I)-likehelicase; VSM, vascular smooth muscle; IL, interleukin; TGF, transforming growth factor; MIF, migrationinhibitory factor; TNF, tumor necrosis factor; HMGB, high mobility group box; PAF, protease activatingfactor; sTNFR, soluble tumor necrosis factor receptor.


it has been recently shown that it has aweak proinflammatory activity by itself andbinding to bacterial substances includinglipid molecules such as phophatidylserinestrengthens its effects (107). Interestingly,phosphatidylserine has been implicated inthe regulation of inflammation andHMGB-1 might thus regulate its anti-inflammatory activities (108).

HMGB-1 is released into the extracel-lular space through acetylation or phos-phorylation (109, 110). HMGB-1 is either“passively released” from necrotic cells,and a mechanism that represents a pro-cess adopted by the innate immune sys-tem to recognize damaged and necroticcells, or “actively secreted” by immunecells including macrophages and neutro-phils to trigger inflammation (106). Re-cently, HMGB-1 release from macro-phages has been shown during the courseof apoptosis as well as necrosis and de-fined as a downstream event of cell apo-ptosis during severe sepsis (111, 112).After treatment with LPS or various cy-tokines such as TNF-�, IL-1�, or IFN-�,HMGB-1 is released from activated mac-rophages within 4 hrs and reaches a pla-teau around 18–24 hrs. It binds to severaltransmembrane receptors such as recep-tor for advanced glycation end products(RAGE), TLR-2, and -4, activating NF-�Band extracellular regulated kinase 1/2(113, 114).

Although HMGB-1 was originally de-scribed as a late mediator of endotoxin-induced lethality (115), recent studies in-dicate a role for HMGB-1 in angiogenesis,tissue repair, and regeneration (116).HMGB-1 appears to be a novel myocardialdepressant factor upon release by resi-dent myocardial cells following tissue in-jury. HMGB-1 might decrease energy uti-lization in ischemic tissue, therebypreventing injured myocytes from wors-ening ATP depletion that eventuate innecrosis (117). On other hand, excessiverelease of HMGB-1 in sepsis might con-tribute to sustained inflammation and toprofound myocardial depression.

The cholinergic anti-inflammatorypathway is a neural mechanism that in-hibits the expression of HMGB-1 andother cytokines (118–121). Signals trans-mitted via the vagus nerve, the principalnerve of the parasympathetic nervoussystem, significantly attenuate the re-lease of HMGB-1 and other cytokines ininflammation in animal and human stud-ies (122).

The remarkable binding characteristicsof HMGB-1 suggest another important role

for this protein in the extracellular fluid.HMGB-1 might serve as a shuttle platformfor LPS and other microbial mediators fordocking to CD14 and recognition by theTLRs (108, 123) or through binding tohost-derived proinflammatory mediators,such as IL-1beta (124).

Elevated levels of HMGB-1 are measur-able in the majority of patients up to 1 wkafter the diagnosis of sepsis or septic shockand are correlated with the degree of organdysfunction (125, 126). However, serumHMGB-1 levels do not consistently identifynonsurvivors from survivors as a predictorof hospital mortality (127). As HMGB-1 islate inflammatory cytokine of sepsis, it pro-vides a wide therapeutic time window forclinical intervention and remains an attrac-tive target for sepsis treatment.

RAGE, a member of the immunoglobu-lin superfamily, is a pattern-recognition re-ceptor that binds diverse classes of endog-enous molecules including HMGB-1. It hasbeen defined as part of a newly appreciatedcomponent of the innate immune systemreferred to as the danger associated molec-ular pattern system (Fig. 3) (128). Mem-brane bound and soluble forms of RAGE(sRAGE) have been detected in plasma. Sol-uble RAGE, anti-RAGE antibody (Fab 2fragment) or data from RAGE�/� animalshave been shown to decrease inflammation,reduce neutrophil extravasation, and re-duce migration (129). Recently, the dem-onstration of survival benefit after delayedadministration of anti-RAGE antibody in amurine model of polymicrobial sepsis, evenwhen delayed up to 24 hrs, provides a ther-apeutic rationale for the use of anti-RAGEmAb as a salvage therapy for establishedsevere sepsis (130).

Macrophage Migration InhibitoryFactor. Migration inhibitory factor (MIF)acts as a stress response mediator andproinflammatory cytokine upon induc-tion by glucocorticoids (131). This pro-tein is readily measurable in patients withsepsis, and MIF probably contributes tothe pathogenesis of sepsis. Inhibition ofMIF or its targeted deletion attenuatesTNF-� and IL-1� expression and protectsmice from lethality is experimental sepsis(132, 133). Systemic challenge of animalswith MIF increases LPS-related lethality(133). MIF promotes the expression ofTLR4 on macrophages, and thereby sen-sitizes these immune effector cells to LPS(134).

The immunoregulatory effects of MIFmight be crucial to the control and resolu-tion of the inflammatory response as a con-sequence of its ability to regulate activa-

tion-induced apoptosis (135). High MIFlevels delay the removal of activated mono-cytes/macrophages by apoptosis. This pro-longs monocyte/macrophage survival, in-creases cytokine production, and sustainsan ongoing proinflammatory response.

Mitochondrial Dysfunction inInflammation

Although microvascular flow abnormal-ities occur, findings of decreased oxygenconsumption (136) and elevated tissue ox-ygen tension (137), yet minimal cell deathdespite functional and biochemical de-rangements (138), suggest that the prob-lem lies more in cellular oxygen utilizationrather than a problem with oxygen delivery(139). It is postulated that prolonged andsystemic inflammatory insult is accompa-nied by a basic tissue survival response me-diated by switching off its energy-consum-ing biophysiological processes. Recentevidence suggests that sepsis and septicshock severely impair the mitochondria(140, 141), and the severity and outcome oforgan dysfunction could be related to mi-tochondrial dysfunction (142). Depletedlevels of reduced glutathione, an importantintramitochondrial antioxidant, in combi-nation with excess generation of ROS andRNS severely inhibit oxidative phosphory-lation and ATP generation (142). This ac-quired intrinsic derangement in cellularenergy metabolism which has also beentermed “cytopathic hypoxia,” contributes toreduced activities of mitochondrial electrontransport chain enzyme complexes and im-paired ATP biosynthesis, potentially organdysfunction in sepsis (141, 143, 144).

Sepsis-related derangements in mito-chondrial function can activate the ubiq-uitin proteolytic pathway in skeletal mus-cle of septic patients (145). Mitochondrialpermeability transition pore seems to beinvolved in sepsis-induced mitochondrialdamage, since its inhibition significantlyimproved organ function and reducedmortality in rodents (146). Whether sep-sis-related reduction in energy supplycould result in a state of cellular shut-down analogous to myocardial “stun-ning” (or hibernation) following coronaryocclusion, allowing for eventual restora-tion of organ function and survival, hasnot yet been determined.

Impact of Inflammation onCoagulation

The clotting system is almost invari-ably activated by systemic microbial in-


vasion (147). Clotting is one most prom-inent features of sepsis. Coagulationcontributes significantly to the outcomein sepsis with concurrent down-regula-tion of anticoagulant systems and fibri-nolysis (Fig. 7). Inflammation-inducedcoagulation in turn contributes to fur-ther inflammation (148). Indeed, collab-oration between clotting and inflamma-tion accounts for the basic survivalstrategy of walling off the damaged andinfected tissues from the rest of the host(149). The key determinant of survival insepsis is to limit excess systemic inflam-matory and coagulopathic damage whileretaining the benefits of controlled anti-microbial clearance and localized clotformation (147, 150).

The inflammatory reaction to tissueinjury activates the clotting system, andcoagulation promotes inflammation (3,151, 152). The role of procoagulant apo-ptotic microparticles have also been dem-onstrated in sepsis (153–155). Linkage ofcoagulation enzymes with their serineprotease activity with protease activatedreceptors (PAR 1-4) on endothelial sur-faces increases P-selectin, cytokine produc-tion, and adhesion molecule expression,leading to microcirculatory dysfunction insevere sepsis (156, 157).

Activated Protein C. Activated proteinC (APC) is derived from its zymogen pro-tein C in contact with thrombin: thrombo-modulin complexes on endothelial sur-faces. Protein C was originally thought tobe synthesized exclusively by the liver(158). It has recently shown that it isstrongly expressed by the endothelium andkeratinocytes (159). The conversion to APCis augmented by endothelial PC receptor(EPCR) which is present on endothelialcells, neutrophils, monocytes, and keratin-ocytes (160, 161) whereas soluble EPCRinhibits APC anticoagulant activity (162).Soluble EPCR is released constitutively andlevels increase in patients with Gram-negative sepsis (163). Levels of APC, proteinC, and its cofactor protein S are depleted insepsis (164, 165). Furthermore, peripheralconversion of Protein C by the thrombin:thrombomodulin complex is impaired insepsis, further contributing to microvascu-lar thrombosis and vascular leakage (166).

APC is a central endogenous anticoag-ulant protein with antithrombotic, anti-inflammatory, antiapoptotic, and profi-brinolytic activities (Fig. 8) (167, 168).Although an anti-inflammatory role forAPC may be an indirect consequence ofits ability to reduce thrombin generation,APC also has direct anti-inflammatory

Figure 7. The extrinsic pathway (tissue factor pathway) is the primary mechanism by which thrombin isgenerated in sepsis. The intrinsic cascade (contact factor pathway) primarily serves an accessory role inamplifying the prothrombotic events that are initiated in sepsis. Thrombin, factor Xa and the TF-factor VIIacomplex interact with the PAR (protease activated receptors) system and directly activate endothelial cells,platelets and white blood cells, and induce a proinflammatory response. Platelet-derived microparticles(MPs) express functional adhesion receptors including P-selectin on their surface, attach to the site ofinjury on the vessel wall, and support the rolling of leukocytes in the presence of shear stress or severeinsult. When bearing appropriate counter-ligands, MPs can transfer their procoagulant potential to targetcells. Platelet-derived MPs can bind to soluble and immobilized fibrinogen, thus delivering procoagulantentities to the thrombus via the formation of aggregates. In vitro, interaction between endothelial MPs andmonocytes promotes TF mRNA expression and TF-dependent procoagulant activity. Activated platelets andplatelet-derived MPs thus amplify leukocyte-mediated tissue injury in thrombotic and inflammatorydisorders. EPCR, endothelial protein C receptor; PMN, polymorphonuclear leukocyte; APC, activatedprotein C; ADAMST13, a disintegrin and metalloproteinase with a thrombospondin type 1 motifs 13;ULVWF, unusually large von Willebrand factor; ICAM, intracellular adhesion molecule; VCAM, vascular celladhesion molecule-1; IL, interleukin; TNF, tumor necrosis factor; IFN, interferon.

Figure 8. The homeostatic balance between thrombin and activated protein C (APC) in coagulation andinflammation is given. APC is a central endogenous anticoagulant protein with antithrombotic, anti-inflammatory, anti-apoptotic and pro-fibrinolytic activities. The ligand occupancy of endothelial protein Creceptor (EPCR) switches the protease-activated receptor 1 (PAR-1)-dependent signaling specificity of thrombinfrom a permeability-enhancing to a barrier-protective response. Other pleiotropic effects are also explained. TAFI,thrombin activatable fibrinolysis inhibitor; NF-��, nuclear factor ��.


properties (169). APC can cleave and ac-tivate PAR1-dependent cellular pathways(170). APC competes for PAR-1 bindingwith thrombin, but in vitro studies sug-gest that APC is 103- to 104-fold less po-tent that thrombin in cleaving PAR-1.The critical receptors required for bothPC activation and APC cellular signaling(i.e., thrombomodulin, EPCR and PAR-1)are co-localized in lipid rafts on endothe-lial cells (171). EPCR is associated withcaveolin-1 on lipid rafts and EPCR bind-ing to the gamma-carboxyglutamic aciddomain of protein C/APC leads to its dis-sociation from caveolin-1 (Fig. 7). APCthen engages PAR-1 generating a protec-tive signaling pathway through couplingof PAR-1 to the pertussis toxin-sensitiveG(i)-protein. Thus, when EPCR is boundby protein C, the PAR-1-dependent pro-tective signaling responses in endothelialcells can be mediated by either thrombinor APC. These results explain how PAR-1and EPCR participate in protective sig-naling events in endothelial cells (172).

Therapeutic administration of recom-binant human APC (rhAPC) is currentlyin use as a treatment strategy for severesepsis patients with a high risk of death(173, 174). Genetically engineered vari-ants of APC have been designed withgreater antiapoptotic activity and reducedanticoagulant activity relative to wild-type APC to increase the risk/benefit ratio

of rhAPC regarding the bleeding compli-cation (175). A nonanticoagulant from ofAPC reduces mortality in experimentalmodels of endotoxemia and sepsis (176).

von Willebrand Factor, A Disintegrin-Likeand Metalloproteinase with ThrombospondinType-1 Motifs 13 and Ashwell Receptor.The discovery of a disintegrin-like and met-alloproteinase with thrombospondin type-1motifs 13 (ADAMTS-13) has provided newinsights in pathogenesis of thrombosis insepsis. ADAMTS-13, the principal physio-logic modulator of von Willebrand factor(VWF) is produced mainly stellate cells inthe liver (177). VWF is synthesized in vas-cular endothelial cells and released into theplasma as unusually large VWF multimerswhich are rapidly degraded into smallerVWF multimers by ADAMTS-13. Deficiencyof the ADAMTS-13, as observed in mostforms of thrombotic thrombocytopenicpurpura, increases the level of unusuallylarge VWF multimers in plasma and leadsto platelet aggregation and/or thrombusformation, especially in small arterioles, re-sulting in microvascular failure (178–180).Recently, inflammation associated-ADAMTS-13 deficiency has been de-scribed in patients with systemic inflam-matory response syndrome and severesepsis (Fig. 7) (181–185). Decreased lev-els of ADAMTS-13 have been reported inhealthy volunteers following endotoxininfusion (186). Furthermore, reduced

ADAMTS-13 levels are associated with dif-ferences in morbidity, mortality, andvariables of inflammation and endothelialdysregulation in severe sepsis patients(182, 187).

The Ashwell receptor, which is themajor lectin of hepatocytes, modulatesVWF homeostasis. It has been recentlydemonstrated that the marked thrombo-cytopenia associated with S. pneumoniaesepsis is the result of Ashwell receptor-dependent clearance of platelets (188).The ensuing reduction in platelet countsat the onset of sepsis protects the hostagainst the development of disseminatedintravascular coagulation. Homeostaticadaptation by this receptor moderates theonset and severity of disseminated intra-vascular coagulation during sepsis sug-gesting the improvement in host survivalprobability (189). At present, it remainsunclear whether blocking the Ashwell re-ceptor may have a beneficial effect insevere sepsis.

CONCLUSION

A unifying concept of innate immu-nity is based upon pathogen-, or damage-associated molecular pattern moleculesand downstream signaling pathways. Thishas facilitated significant advances in ourunderstanding of the pathophysiology ofsepsis and led to a multitude of clinical

Table 1. Ongoing and upcoming clinical studies with molecular targets

Target Sponsor or Institution Phase Comments

CytoFab (AZD9773) Fabs of IgG that bind to TNF-� AstraZeneca II RecruitingRecombinant human lactoferrin LPS-neutralization Agennix II RecruitingLipid emulsion (GR-270773) TLR4 Glaxo Smith Kline II SuspendedE5564 TLR4 Eiasi III OngoingTAK-242 TLR4 Takeda III RecruitingRecombinant human soluble thrombomodulin Coagulation Artisan Pharma II RecruitingRecombinant antithrombin Coagulation Leo Pharma II RecruitingRecombinant tissue factor pathway inhibitor Coagulation Novartis III CompletedRecombinant activated Protein C� Erythropoietin Coagulation and inflammation LHRI III RecruitingRecombinant activated Protein C Coagulation Lilly III RecruitingRecombinant plasma gelsolin Replacement of circulating

actin-binding proteinCBC II Recruiting

L-Citrulline supplementation NO pathway MU II Not yet open forparticipantrecruitment

Inhaled nitric oxide NO pathway NIMS III Not yet open forparticipantrecruitment

Simvastatin Pleitrophic effects MUV, Austria andChicago, USA

IV Recruiting

Rosuvastatin Pleitrophic effects UASLP, Mexico and BethIsrael

II Ongoing

Atorvastatin Pleitrophic effects HCPA, Brazil II Recruiting

TNF, tumor necrosis factor; LPS, lipopolysaccharide; NO, nitric oxide; TLR, toll-like receptor, LHRI, Lawson Human Resources Institute; CBC, CriticalBiologics Corporation; MU, Maastricht University; NIMS; National Institute of Medical Sciences; MUV, Medical University of Vienna; UASLP, UniversidadAutonoma de San Luis Potosi; HCPA, Hospital de Clinical de Porto Alegre.


trials (Table 1). However, further identi-fication of the critical elements in severesepsis that drives the transition from lo-calized inflammation to deleterious hostresponse remains incompletely illumi-nated. The limitations in our currentknowledge of the molecular mechanismsin sepsis have made the design of inter-vention trials in clinical sepsis challeng-ing (190). New findings as to biomarkersand measures to detect genetic signa-tures of sepsis are now making their wayinto clinical trial designs. It is anticipatedthat such innovations will improve theoutlook for successful development ofnew sepsis treatments tailored to individ-ual patient needs.

REFERENCES

1. van der Poll T, Opal SM: Host-pathogeninteractions in sepsis. Lancet Infect Dis2008; 8:32–43

2. Cinel I, Dellinger RP: Making a difference inoutcome in severe sepsis: The organism, thehost response and the treating clinician.Emerg Med Crit Care Rev 2007; 1:1–3

3. Cinel I, Dellinger RP: Advances in patho-genesis and management of sepsis. CurrOpin Infect Dis 2007; 20:345–352

4. Uematsu S, Akira S: Toll-like receptors andinnate immunity. J Mol Med 2007; 84:712–725

5. Creagh EM, O’Neill LA: TLRs, NLRs andRLRs: A trinity of pathogen sensors thatco-operate in innate immunity. TrendsImmunol 2006; 27:352–357

6. Granucci F, Foti M, Ricciardi-Castagnoli P:Dendritic cell biology. Adv Immunol 2005;88:193–233

7. Liew FY, Xu D, Brint EK, et al: Negativeregulation of toll-like receptor-mediatedimmune responses. Nat Rev Immunol 2005;5:446–458

8. Akira S, Uematsu S, Takeuchi O: Pathogenrecognition and innate immunity. Cell2006; 124:783–801

9. Bianchi ME: DAMPs PAM. Ps and alarmins:All we need to know about danger. J LeukocBiol 2007; 81:1–5

10. Mollen KP, Anand RJ, Tsung A, et al:Emerging paradigm: Toll-like receptor4-sentinel for the detection of tissue dam-age. Shock 2006; 26:430–437

11. Tsujimoto H, Ono S, Hiraki S, et al: He-moperfusion with polymyxin B-immobi-lized fibers reduced the number of CD16�CD14� monocytes in patients with septicshock. J Endotoxin Res 2004; 10:229–237

12. Armstrong L, Medford AR, Hunter KJ, et al:Differential expression of toll-like receptor(TLR)-2 and TLR-4 on monocytes in humansepsis. Clin Exp Immunol 2004; 136:312–319

13. Tsujimoto H, Ono S, Majima T, et al: Neu-trophil elastase, MIP-2, and TLR-4 expres-

sion during human and experimental sep-sis. Shock 2005; 23:39–44

14. Williams DL, Ha T, Li C, et al: Modulation oftissue toll-like receptor 2 and 4 during theearly phases of polymicrobial sepsis corre-lates with mortality. Crit Care Med 2003;31:1808–1818

15. Paterson HM, Murphy TJ, Purcell EJ, et al:Injury primes the innate immune systemfor enhanced toll-like receptor reactivity.J Immunol 2003; 171:1473–1483

16. Tsujimoto H, Ono S, Efron PA, et al: Role oftoll like receptors in the development ofsepsis. Shock 2008; 29:315–321

17. Ogura Y, Sutterwala FS, Flavell RA: Theinflammasome: First line of the immuneresponse to cell stress. Cell 2006; 126:659–662

18. Drenth JP, van der Meer JW: The inflamma-some—A linebacker of innate defense.N Engl J Med 2006; 355:730–732

19. Dinarello CA: Interleukin-1�, interleukin-18, and the interleukin-1� converting en-zyme. Ann NY Acad Sci 1998; 856:1–11

20. Martinon F, Burns K, Tschopp J: The in-flammasome. A molecular platform trigger-ing activation of inflammatory caspases andprocessing of proIL-1�. MolCell 2002; 10:417–426

21. Yu JW, Wu J, Zhang Z, et al: Cryopyrin andpyrin activate caspase-1, but not NF-kappaB, via ASC oligomerization. CellDeath Differ 2005; 13:236–249

22. Seshadri S, Duncan MD, Hart JM, et al:Pyrin levels in human monocytes andmonocyte-derived macrophages regulateIL-1� processing and release. J Immunol2007; 179:1274–1281

23. Tschopp J, Martinon F, Burns K: NALPs: Anovel protein family involved in inflamma-tion. NatRevMol Cell Biol 2003; 4:95–104

24. Sarkar A, Duncan M, Hart J, et al: ASCdirects NF-�B activation by regulating re-ceptor interacting protein-2 (RIP2)caspase-1 interactions. J Immunol 2006;176:4979 – 4986

25. Scott AM, Saleh M: The inflammatorycaspases: Guardians against infections andsepsis. Cell Death Differ 2007; 14:23–31

26. Sarkar A, Hall MW, Exline M, et al:Caspase-1 regulates E. coli sepsis andsplenic B cell apoptosis independently ofIL-1� and IL-18. Am J Respir Crit Care Med2006; 174:1003–1010

27. Saleh M, Vaillancourt JP, Graham RK, et al:Differential modulation of endotoxin re-sponsiveness by human caspase-12 poly-morphisms. Nature 2004; 429:75–79

28. Saleh M, Mathison JC, Wolinski MK, et al:Enhanced bacterial clearance and sepsis re-sistance in caspase-12-deficient mice. Na-ture 2006; 440:1064–1068

29. Joshi VD, Kalvakolanu DV, Hasday JD, et al:IL-18 levels and the outcome of innate im-mune response to lipopolysaccharide: Im-portance of a positive feedback loop withcaspase-1 in IL-18 expression. J Immunol2002; 169:2536–2544

30. Fahy RJ, Exline MC, Gavrilin MA, et al:Inflammasome mRNA expression in humanmonocytes during early septic shock. Am JResp Crit Care Med 2008; 177:983–988.

31. Lara-Tejero M, Sutterwala FS, Ogura Y, etal: Role of the caspase-1 inflammasome inSalmonella typhimurium pathogenesis.J Exp Med 2006; 203:1407–1412

32. Brightbill HD, Libraty DH, Krutzik SR, etal: Host defense mechanisms triggered bymicrobial lipoproteins through toll-like re-ceptors. Science 1999; 285:732–736

33. Bruno TF, Woods DE, Mody CH: Exoen-zyme S from Pseudomonas aeruginosa in-duces apoptosis in T lymphocytes. J LeukocBiol 2000; 67:808–816

34. Wesche H, Henzel WJ, Shillinglaw W, et al:MyD88: An adapter that recruits IRAK tothe IL-1 receptor complex. Immunity 1997;7:837–847

35. Reynaert NL, van der Vliet A, Guala AS, etal: Dynamic redox control of NF-kappaBthrough glutaredoxin-regulated S-gluta-thionylation of inhibitory kappaB kinasebeta. Proc Natl Acad Sci USA 2006; 103:13086–13091

36. Pahl HL: Activators and target genes of rel/NF-kappaB transcription factors. Oncogene1999; 18:6853–6866

37. Blackwell TS, Yull FE, Chen CL, et al: Mul-tiorgan nuclear factor kappa B activation ina transgenic mouse model of systemic in-flammation. Am J Respir Crit Care Med2000; 162:1095–1101

38. Arcaroli J, Silva E, Maloney JP, et al: VariantIRAK-1 haplotype is associated with in-creased nuclear factor-kappaB activationand worse outcomes in sepsis. Am J RespirCrit Care Med 2006; 173:1335–1341

39. Zaph C, Troy AE, Taylor BC, et al: Epithelialcell-intrinsic IKK-beta expression regulatesintestinal immune homeostasis. Nature2007; 446:552–556

40. Liu YJ, Soumelis V, Watanabe N, et al:TSLP: An epithelial cell cytokine that regu-lates T cell differentiation by conditioningdendritic cell maturation. Annu Rev Immu-nol 2007; 25:193–219

41. Peck-Palmer OM, Unsinger J, Chang KC, etal: Deletion of MyD88 markedly attenuatessepsis-induced T and B lymphocyte apopto-sis but worsens survival. J Leukoc Biol2008; 83:1009–1018

42. Cantley LC. The phosphoinositide 3-kinasepathway. Science 2002; 296:1655–1657

43. Okugawa S, Ota Y, Kitazawa T, et al: Januskinase2 is involved in lipopolysaccharide-induced activation of macrophages. Am JPhysiol Cell Physiol 2003; 285:C399–C408

44. Ojaniemi M, Glumoff V, Harju K, et al:Phosphatidylinositol kinase is involved inToll-like receptor 4-mediated cytokine ex-pression in mouse macrophages. Eur J Im-munol 2003; 335:97–605

45. Guillot L, Le GofficR, BlochS, et al: Involve-ment of toll-like receptor 3 in the immuneresponse of lung epithelial cells to double-


stranded RNA and influenza A virus. J BiolChem 2005; 280:5571–5580

46. Wrann CD, Tabriz NA, Barkhausen T, et al:The phosphatidylinositol 3-kinase signalingpathway exerts protective effects duringsepsis by controlling C5a-mediated activa-tion of innate immune functions. J Immu-nol 2007; 178:5940–5948

47. Guha M, Mackman N: The PI3K-Akt path-way limits LPS activation of signaling path-ways and expression of inflammatory medi-ators in human monocytic cells. J BiolChem 2002; 277:32124–32132

48. Schabbaue G, Tencati M, Pedersen B, et al:PI3K-Akt pathway suppresses coagulationand inflammation in endotoxemic mice. Ar-terioscler Thromb Vasc Biol 2004; 24:1963–1969

49. Pengal RA, Ganesan LP, Wei G, et al: Li-popolysaccharide-induced production of in-terleukin-10 is promoted by the serin thre-onine kinase Akt. Mol Immunol 2006; 43:1557–1156

50. Fukao T, Koyasu S: PI3K and negative reg-ulation of TLR signaling. Trends Immunol2003; 24:358–363

51. Bommhardt U, Chang KC, Swanson PE, etal: Akt decreases lymphocyte apoptosis andimproves survival in sepsis. J Immunol2004; 172:7583–7591

52. Mc Dunn JE, Muenzer JT, Rachdi L, et al:Peptide-mediated activation of Akt and ex-tracellular regulated kinase signaling pre-vents lymphocyte apoptosis. FASEB J 2008;22:561–568

53. Hall A: Rho GT. Pases and the actin cy-toskeleton. Science 1998; 279:509–514

54. Ruse M, Knaus UG: New players in TLR-mediated innate immunity. ImmunologicRes 2006; 34:33–48

55. Arbibe L, Mira JP, Teusch N, et al: Toll-likereceptor 2-mediated NF-kappa B activationrequires a Racl dependenpt pathway. NatImmunol 2000; 1:533–540

56. Honing H, van den Berg TK, van der PolSM, et al: Rho activation promotes transen-dothelial migration of monocytes viaROCK. J Leukoc Biol 2004; 75:523–528

57. Tasaka S, Koh H, Yamada W, et al: Attenu-ation of endotoxin-induced acute lung in-jury by the Rho-associated kinase inhibitor,Y-27632. Am J Respir Cell Mol Biol 2005;32:504–510

58. Thorlacius K, Slotta JE, Laschke MW, et al:Protective effect of fasudil, a Rho-kinaseinhibitor, on chemokine expression, leuko-cyte recruitment, and hepatocellular apo-ptosis in septic liver injury. J Leukoc Biol2006; 79:923–931

59. Hori S, Nomura T, Sakaguchi S: Control ofregulatory T cell development by the tran-scription factor FoxP3. Science 2003; 299:1057–1061

60. Zheng SG, Wang JH, Koss MN, et al: CD4�and CD8� regulatory T cells generated exvivo with IL-2 and TGF-beta suppress astimulatory Graft-versus- Host disease with

a Lupus-like syndrome. J Immunol 2004;172:1531–1539

61. Ausubel FM: Are innate immune signalingpathways in plants and animals conserved?Nat Immunol 2005; 6:973–979

62. Caramalho I, Lopes-Carvalho T, Ostler D, etal: Regulatory T cells selectively expresstoll-like receptors and are activated by lipo-polysaccharide. J Exp Med 2003; 197:403–411

63. Sutmuller RPM, den Brok MH, Kramer M,et al: Toll-like receptor 2 controls expansionand function of regulatory T cells. J ClinInvest 2006; 116:485–494

64. Liu H, Komai-Koma M, Xu D, et al: Toll-likereceptor 2 signaling modulates the func-tions of CD4�CD25� regulatory T cells.Proc Natl Acad Sci USA 2006; 103:7048–7053

65. McHugh RS, Shevach EM: The role of sup-pressor T cells in regulation of immuneresponses. J Allergy Clin Immunol 2002;110:693–702

66. Murphy TJ, Choileain NN, Zang Y, et al:CD4�CD25� regulatory T cells control in-nate immune reactivity after injury. J Im-munol 2005; 174:2957–2963

67. Monneret G, Debard AL, Venet F, et al:Marked elevation of human circulatingCD4�CD25� regulatory T cells in sepsis-induced immunoparalysis. Crit Care Med2003; 31:2068–2071

68. Venet F, Pachot A, Debard AL, et al: In-creased percentage of CD4�CD25� regula-tory T cells during septic shock is due to thedecrease of CD4�CD25-lymphocytes. CritCare Med 2004; 32:2329–2331

69. Takahashi H, Tsuda Y, Takeuchi D, et al:Influence of systemic inflammatory re-sponse syndrome on host resistance againstbacterial infections. Crit Care Med 2004;32:1879–1885

70. Tiemessen MM, Jagger AL, Evans HG, et al:CD4�CD25�Foxp3� regulatory T cells in-duce alternative activation of human mono-cytes/macrophages. Proc Natl Acad Sci USA2007; 104:19446–19451

71. Wisnoski N, Chung CS, Chen Y, et al: Thecontribution of CD4� CD25� T-regulato-ry-cells to immune suppression in sepsis.Shock 2007; 27:251–257

72. van Maren WW, Jacobs JF, de Vries IJ, et al:Toll-like receptor signalling on Tregs: Tosuppress or not to suppress? Immunology2008; 124:445–452.

73. Heuer JG, Zhang T, Zhao J, et al: Adoptivetransfer of in vitro-stimulated CD4�CD25�

regulatory T cells increases bacterial clear-ance and improves survival in polymicrobialsepsis. J Immunol 2005; 174:7141–7146

74. Venet F, Pachot A, Debard AL, et al: HumanCD4�CD25� regulatory T lymphocytes in-hibit lipopolysaccharide-induced monocytesurvival through a Fas/Fas ligand-depen-dent mechanism. J Immunol 2006; 177:6540–6547

75. Brown KA, Brain SD, Pearson JD, et al:Neutrophils in development of multiple or-

gan failure in sepsis. Lancet 2006; 368:157–169

76. Marshal JC: The pathogenesis and molecu-lar biology of sepsis. Crit Care Resusc 2006;8:227–229

77. Chandra A, Enkhbaatar P, Nakano Y, et al:Sepsis: Emerging role of nitric oxide andselectins. Clinics 2006; 61:71–76

78. Hu G, Predescu D, Vogel SM: ICAM-1 de-pendent neutrophil adhesion to endothelialcells increases caveolae-mediated pulmo-nary vascular albumin permeability. CritCare Med 2007; 34 (Suppl 12): A31

79. Garrean S, Gao X-P, Brovkovych V, et al:Caveolin-1 regulates NF-kB activation andlung inflammatory response to sepsis in-duced by lipopolysaccharide. J Immunol2006 177: 4853–4860

80. Wang XM, Kim HP, Song R, et al: Caveo-lin-1 confers antiinflammatory effects inmurine macrophages via the MKK3/p38MAPK pathway. Am J Respir Cell Mol Biol2006; 34:434–442

81. Medina FA, de Almeida CJ, Dew E, et al:Caveolin-1 deficient mice show defects ininnate immunity and inflammatory im-mune response during Salmonella entericaserovar typhimurium infection. Infect Im-mun 2006; 74:6665–6674

82. Medina FA, Cohen AW, de Almeida CJ, et al:Immune dysfunction in caveolin-1 nullmice following infection with Trypanosomacruzi (Tulahuen strain). Microbes Infect2007; 9:325–333

83. Astiz ME, DeGent GE, Lin RY, et al: Micro-vascular function and rheologic changes inhyperdynamic sepsis. Crit Care Med 1995;23:265–271

84. Ozdulger A, Cinel I, Koksel O, et al: Theprotective effect of N-acetylcysteine on ap-optotic lung injury in cecal ligation andpuncture-induced sepsis model. Shock2003; 19:366–372

85. Stehr SN, Knels L, Weissflog C, et al: Effectsof IgM enriched solution on polymorpho-nuclear neutrophil function, bacterialclearance, and lung histology in endotox-emia. Shock 2007; 29:167–172

86. Kabay B, Kocaefe C, Baykal A, et al: Inter-leukin-10 gene transfer: Prevention of mul-tiple organ injury in a murine cecal ligationand puncture model of sepsis. World J Surg2007; 31:105–115

87. Ozdulger A, Cinel I, Unlu A, et al: Poly(ADP-ribose) synthetase inhibition preventslipopolysaccharide-induced peroxynitritemediated damage in diaphragm. PharmacolRes 2002; 46:67–73

88. Taner S, Cinel I, Ozer L, et al: Poly (ADP-ribose) synthetase inhibition reduces bacte-rial translocation in rats after endotoxinchallenge. Shock 2001; 16:159–162

89. Cinel I, Buyukafsar K, Cinel L, et al: Therole of poly (ADP-ribose) synthetase inhibi-tion in preventing endotoxemia-induced in-testinal epithelial apoptosis. Pharmacol Res2002; 46:119–127


90. Savill J: Apoptosis in resolution of inflam-mation. J Leukoc Biol 1997; 61:375–380

91. Maianski NA, Geissler J, Srinivasula SM, etal: Functional characterization of mito-chondria in neutrophils: A role restricted toapoptosis. Cell Death Differ 2004; 11:143–153

92. van Raam BJ, Verhoeven AJ, Kuijpers TW:Mitochondria in neutrophil apoptosis. IntJ Hematol 2006; 84:199–204

93. Sookhai S, Wang JJ, McCourt M, et al: Anovel therapeutic strategy for attenuatingneutrophil-mediated lung injury in vivo.Ann Surg 2002; 235:285–291

94. Taneja R, Parodo J, Kapus A, et al: Delayedneutrophil apoptosis in sepsis is associatedwith maintenance of mitochondrial trans-membrane potential and reduced caspase-9activity. Crit Care Med 2004; 32:1460–1469

95. Fadok VA, Bratton DL, Konowal A, et al:Macrophages that have ingested apoptoticcells in vitro inhibit proinflammatory cyto-kine production through autocrine/para-crine mechanisms involving TGF, PGE2,and PAF. J Clin Invest 1998; 101:890–898

96. Sanford AN, Suriano AR, Herche D, et al:Abnormal apoptosis in chronic granuloma-tous disease and autoantibody productioncharacteristic of lupus. Rheumatology2006; 45:178–181

97. Daleke DL: Regulation of transbilayerplasma membrane phospholipid asymme-try. J Lipid Res 2003; 44:233–242

98. Bayir H, Kagan VE: Bench-to-bedside re-view: Mitochondrial injury, oxidative stressand apoptosis—there is nothing more prac-tical than a good theory. Crit Care 2008;12:206

99. Tyurina YY, Basova LV, Konduru NV, et al:Nitrosative stress inhibits the aminophos-pholipid translocase resulting in phosphati-dylserine externalization and macrophageengulfment: Implications for the resolutionof inflammation. J Biol Chem 2007; 282:8498–8509

100. Swan R, Chung CS, Albina J, et al: Polymi-crobial sepsis enhances clearance of apopto-tic immune cells by splenic macrophages.Surgery 2007; 142:253–261

101. Macdonald J, Galley HF, Webster NR: Oxi-dative stress and gene expression in sepsis.Br J Anaest 2003; 90:221–232

102. Matejovic M, Krouzecky A, Rokyta R Jr, etal: Effects of combining inducible nitric ox-ide synthase inhibitor and radical scavengerduring porcine bacteremia. Shock 2007; 27:61–68

103. Fialkow L, Wang Y, Downey GP: Reactiveoxygen and nitrogen species as signalingmolecules regulating neutrophil function.Free Radic Biol Med 2007; 42:153–164

104. Galley HF, Davies MJ, Webster NR: Xan-thine oxidase activity and free radical gen-eration in patients with sepsis syndrome.Crit Care Med 1996; 24:1649–1653

105. Soriano FG, Nogueira AC, Caldini EG, et al:Potential role of poly (adenosine 5-diphosphate-ribose) polymerase activation

in the pathogenesis of myocardial contrac-tile dysfunction associated with human sep-tic shock. Crit Care Med 2006; 34:1073–1079

106. Ulloa L, Tracey KJ: The “cytokine profile”: Acode for sepsis. Trends Mol Med 2005; 11:56–63

107. Zimmermann K, Volkel D, Pable S, et al:Native versus recombinant high-mobilitygroup B1 proteins: functional activity invitro. Inflammation 2004; 28:221–229

108. Rouhiainen A, Tumova S, Valmu L, et al:Pivotal advance: Analysis of proinflamma-tory activity of highly purified eukaryoticrecombinant HMGB1 (amphoterin). J Leu-kocyte Biol 2007; 81:49–58

109. Bonaldi T, Talamo F, Scaffidi P, et al: Mono-cytic cells hyperacetylate chromatin proteinHMGB1 to redirect it towards secretion.EMBO J 2003; 22:5551–5560

110. Youn JH, Shin JS: Nucleocytoplasmic shut-tling of HMGB1 is regulated by phosphory-lation that redirects it toward secretion.J Immunol 2006; 177:7889–7897

111. Bell CW, Jiang W, Reich CF III, et al: Theextracellular release of HMGB1 during ap-optotic cell death. Am J Physiol Cell Physiol2006; 291:C1318–C1325

112. Qin S, Wang H, Yuan R, et al: Role ofHMGB1 in apoptosis-mediated sepsis lethal-ity. J Exp Med 2006; 203:1637–1642

113. Hori O, Brett J, Slattery T, et al: The recep-tor for advanced glycation end products(RAGE) is a cellular binding site for ampho-terin: Mediation of neurite outgrowth andco-expression of rage and amphoterin in thedeveloping nervous system. J Biol Chem1995; 270:25752–25761

114. Park JS, Svetkauskaite D, He Q, et al: In-volvement of toll-like receptors 2 and 4 incellular activation by high mobility groupbox 1 protein. J Biol Chem 2004; 279:7370–7377

115. Wang H, Bloom O, Zhang M, et al: HMG-1as a late mediator of endotoxin lethality inmice. Science 1999; 285:248–251

116. Mitola S, Belleri M, Urbinati C, et al: Cut-ting edge: Extracellular high mobilitygroup box-1 protein is a proangiogenic cy-tokine. J Immunol 2006; 176:12–15

117. Tzeng HP, Fan J, Vallejo JG, et al: Negativeinotropic effects of high-mobility group box1 protein in isolated contracting cardiacmyocytes. Am J Physiol Heart Circ Physiol2008; 294:H1490–H1496

118. Wang H, Yu M, Ochani M, et al: Nicotinicacetylcholine receptor �7 receptor is an es-sential regulator of inflammation. Nature2003; 421:384–388

119. Pavlov VA, Wang H, Czura CJ, et al: Thecholinergic anti-inflammatory pathway: Amissing link in neuromodulation. Mol Med2003; 9:125–134

120. Wang H, Liao H, Ochani M, et al: Cholin-ergic agonists inhibit HMGB1 release andimprove survival in experimental sepsis.Nat Med 2004; 10:1216–1221

121. Huston JM, Gallowitsch-Puerta M, Ochani

M, et al: Transcutaneous vagus nerve stim-ulation reduces serum high mobility groupbox 1 levels and improves survival in mu-rine sepsis. Crit Care Med 2007; 35:2762–2768

122. Goldstein RS, Bruchfeld A, Yang L, et al:Cholinergic anti-inflammatory pathway ac-tivity and High Mobility Group Box-1(HMGB1) serum levels in patients withrheumatoid arthritis. Mol Med 2007;13(3–4):210–215

123. Youn JH, Oh YJ, Kim ES, et al: High Mo-bility Group Box 1 protein binding to lipo-polysaccharide facilitates transfer of lipo-polysaccharide to CD14 and enhanceslipopolysaccharide-mediated TNF-{alpha}production in human monocytes. J Immu-nol 2008; 180:5067–5074

124. Sha Y, Zmijewski J, Xu Z, et al: HMGB1develops enhanced proinflammatory activ-ity by binding to cytokines. J Immunol2008; 180:2531–2537

125. Sunden-Cullberg J, Norrby-Teglund A, Rou-hiainen A, et al: Persistent elevation of highmobility group box-1 protein (HMGB1) inpatients with severe sepsis and septic shock.Crit Care Med 2005; 33: 564–573

126. Gibot S, Massin F, Cravoisy A, et al: High-mobility group box 1 protein plasma con-centrations during septic shock. IntensiveCare Med 2007; 33:1347–1353

127. Karlsson S, Pettila V, Tenhunen J, et al:HMGB1 as a predictor of organ dysfunctionand outcome in patients with severe sepsis.Intensive Care Med 2008; 34:1046–1053.

128. Harris HE, Raucci A: Alarmin(g) news aboutdanger. EMBO Rep 2006; 7:774–778

129. Chavakis T, Bierhaus A, Al-Fakhri N, et al:The pattern recognition receptor (RAGE) isa counterreceptor for leukocyte integrins: Anovel pathway for inflammatory cell re-cruitment. J Exp Med 2003; 198:1507–1515

130. Lutterloh EC, Opal SM, Pittman DD, et al:Inhibition of the RAGE products increasessurvival in experimental models of severesepsis and systemic infection. Crit Care2007; 11:R122

131. Calandra T, Bernhagen J, Metz CN, et al:MIF as a glucocorticoid-induced modulatorof cytokine production. Nature 1995; 377:68–71

132. Calandra T, Echtenacher B, Roy DL, et al:Protection from septic shock by neutraliza-tion of macrophage migration inhibitoryfactor. Nat Med 2000; 6:164–170

133. Bozza M, Satoskar AR, Lin G, et al: Targeteddisruption of migration inhibitory factorgene reveals its critical role in sepsis. J ExpMed 1999; 189:341–346

134. Roger T, David J, Glauser MP, et al: MIFregulates innate immune responsesthrough modulation of Toll-like receptor4. Nature 2001; 414:920 –924

135. Mitchell RA, Liao H, Chesney J, et al: Mac-rophage migration inhibitory factor (MIF)sustains macrophage proinflammatoryfunction by inhibiting p53: Regulatory role


in the innate immune response. Proc NatlAcad Sci USA 2002; 99:345–350

136. Kreymann G, Grosser S, Buggisch P, et al:Oxygen consumption and resting metabolicrate in sepsis, sepsis syndrome, and septicshock. Crit Care Med 1993; 21:1012–1019

137. Rosser DM, Stidwill RP, Jacobson D, et al:Oxygen tension in the bladder epitheliumrises in both high and low cardiac outputendotoxemic sepsis. J Appl Physiol 1995;79:1878–1882

138. Hotchkiss RS, Swanson PE, Freeman BD, etal: Apoptotic cell death in patients with sep-sis, shock, and multiple organ dysfunction.Crit Care Med 1999; 27:1230–1251

139. Brealey D, et al: Mitochondrial dysfunctionin a long-term rodent model of sepsis andorgan failure. Am J Physiol Regul IntegrComp Physiol 2004; 286:R491–R497

140. Crouser ED: Mitochondrial dysfunction inseptic shock and multiple organ dysfunc-tion syndrome. Mitochondrion 2004;4:729–741

141. Fink MP: Cytopathic hypoxia. Mitochon-drial dysfunction as mechanism contribut-ing to organ dysfunction in sepsis. Crit CareClinics 2001; 17:219–237

142. Brealey D, Brand M, Hargreaves I, et al:Association between mitochondrial dys-function and severity and outcome of septicshock. Lancet 2002; 360:219–223

143. Crouser ED, Julian MW, Blaho DV, et al:Endotoxin-induced mitochondrial damagecorrelates with impaired respiratory activ-ity. Crit Care Med 2002; 30:276–284

144. Levy RJ, Vijayasarathy C, Raj NR, et al:Competitive and noncompetitive inhibitionof myocardial cytochrome C oxidase in sep-sis. Shock 2004; 21:110–114

145. Rabuel C, Renaud E, Brealey D, et al: Hu-man septic myopathy: Induction of cycloox-ygenase, heme oxygenase and activation ofthe ubiquitin proteolytic pathway. Anesthe-siology 2004; 101:583–590

146. Larche J, Lancel S, Hassoun SM, et al: In-hibition of mitochondrial permeabilitytransition prevents sepsis-induced myocar-dial dysfunction and mortality. J Am CollCardiol 2006; 48:377–385

147. Levi M, Ten Cate H: Disseminated intravas-cular coagulation. N Engl J Med 1999; 341:586–592

148. Schouten M, Wiersinga WJ, Levi M, et al:Inflammation, endothelium, and coagula-tion in sepsis. J Leukoc Biol 2008; 83:536–545

149. Opal SM: The nexus between systemic in-flammation and disordered coagulation insepsis. J Endotoxin Res 2004; 10:125–129

150. Esmon CT: Are natural anticoagulants can-didates for modulating the inflammatoryresponse to endotoxin? Blood 2000; 95:1113–1116

151. Johnson K, Choi Y, DeGroot E, et al: Po-tential mechanisms for a proinflamma-tory vascular cytokine response to coagu-lation activation. J Immunol 1998; 160:5130 –5135

152. Stouthard JM, Levi M, Hack CE, et al: In-terleukin-6 stimulates coagulation, not fi-brinolysis, in humans. Thromb Haemost1996; 76:738–742

153. Aupeix K, Hugel B, Martin T, et al: Thesignificance of shed membrane particlesduring programmed cell death in vitro, andin vivo, in HIV-1 infection. J Clin Invest1997; 99:1546–1554

154. Soriano AO, Jy W, Chirinos JA, et al: Levelsof endothelial and platelet microparticlesand their interactions with leukocytes neg-atively correlate with organ dysfunction andpredict mortality in severe sepsis. Crit CareMed 2005; 33:2540–2546

155. Fujimi S, Ogura H, Tanaka H, et al: Acti-vated polymorphonuclear leukocytes en-hance production of leukocyte micropar-ticles with increased adhesion molecules inpatients with sepsis. J Trauma 2002; 52:443–448

156. Amaral A, Opal SM, Vincent JL: Coagulationin sepsis. Intensive Care Med 2004; 30:1032–1040

157. Vervloet MG, Thijs LG, Hack CE: Derange-ments of coagulation and fibrinolysis incritically ill patients with sepsis and septicshock. Semin Thromb Hemost 1998; 24:33–44

158. Yan SB, Dhainut JF: Activated protein Cversus protein C in severe sepsis. Crit CareMed 2001; 29(Suppl):S69–S74

159. Xue M, Campbell D, Jackson CJ: Protein Cis an autocrine growth factor for humanskin keratinocytes, J Biol Chem 2007; 282:13610–13616

160. Xue M, March L, Sambrook PN, et al:Endothelial protein C receptor is overex-pressed in rheumatoid arthritic (RA) sy-novium and mediates the anti-inflamma-tory effects of activated protein C in RAmonocytes. Ann Rheum Dis 2007; 66:1574 –1580

161. Regan LM, Stearns-Kurosawa DJ, KurosawaS, et al: The endothelial cell protein C re-ceptor: Inhibition of activated protein C an-ticoagulant function without modulation ofreaction with proteinase inhibitors. J BiolChem 1996; 271:17499–17503

162. Kurosawa S, Stearns-Kurosawa DJ, CarsonCW, et al: Plasma levels of endothelial cellprotein C receptor are elevated in patientswith sepsis and systemic lupus erythemato-sus: Lack of correlation with thrombo-modulin suggests involvement of differentpathological processes. Blood 1998; 91:725–727

163. Yan SB, Helterbrand JD, Hartman DL, et al:Low levels of protein C are associated withpoor outcome in severe sepsis. Chest 2001;120:915–922

164. LaRosa SP, Opal SM, Utterback B, et al:Decreased protein C, protein S, and anti-thrombin levels are predictive of poor out-come in Gram-negative sepsis caused byBurkholderia pseudomallei. Int J Infect Dis2006; 10:25–31

165. Yan SB, Dhainaut JF: Activated protein C

versus protein C in severe sepsis. Crit CareMed 2001; 29(Suppl):S69–S74

166. Joyce DE, Gelbert L, Ciaccia A, et al: Geneexpression profile of antithrombotic proteinC defines new mechanisms modulating in-flammation and apoptosis. J Biol Chem2001; 14:11199–111203

167. Jackson CJ, Xue M: Activated protein C-Ananticoagulant that does more than stopclots. Int J Biochem Cell Biol 2008; 40:2692–2697

168. Dutt T, Toh CH: The yin-yang of thrombinand activated protein C. Br J Haematology2008; 140:505–515

169. Riewald M, Ruf W: Protease-activated recep-tor-1 signaling by activated protein C incytokine-perturbed endothelial cells is dis-tinct from thrombin signaling. J Biol Chem2005; 280:19808–19814

170. Bae JS, Yang L, Rezaie AR: Receptors of theprotein C activation and activated protein Csignaling pathways are colocalized in lipidrafts of endothelial cells. Proc Natl Acad SciUSA 2007; 104:2867–2872

171. Bae JS, Yang L, Manithody C, et al: Theligand occupancy of endothelial proteinC receptor switches the protease-activatedreceptor 1-dependent signaling specificityof thrombin from a permeability-enhanc-ing to a barrier-protective responsein endothelial cells. Blood 2007; 110:3909 –3916

172. Shorr AF, Bernard GR, Dhainaut JF, et al:Protein C concentrations in severe sepsis:An early directional change in plasma levelspredicts outcome. Crit Care 2006; 10:R92

173. Cinel I, Dellinger RP: Current treatment ofsevere sepsis. Curr Infect Dis Rep 2006;8:358–365

174. Dellinger RP, Levy MM, Carlet JM, et al:Surviving Sepsis Campaign: Internationalguidelines for management of severe sepsisand septic shock: 2008. Crit Care Med 2008;36:296–327

175. Bae JS, Yang L, Manithody C, et al: Engi-neering a disulfide bond to stabilize thecalcium binding loop of activated protein Celiminates its anticoagulant but not protec-tive signaling properties. J Biol Chem 2007;282:9251–9259

176. Kerschen EJ, Fernandez JA, Cooley BC, etal: Endotoxemia and sepsis mortality reduc-tion by non-anticoagulant activated proteinC. J Exp Med 2007; 204:2439–2448

177. Moake JL: Thrombotic microangiopathies.N Engl J Med 2002; 347:589–600

178. Ruggeri ZM: Von Willebrand factor, plate-lets and endothelial cell interactions.J Thromb Haemost 2003; 1:1335–1342

179. Furlan M, Robles R, Galbusera M, et al: vonWillebrand factor-cleaving protease inthrombotic thrombocytopenic purpura andthe hemolytic-uremic syndrome. New EnglJ Med 1998; 339:1578–1584

180. Arya M, Anvari B, Romo GM, et al: Ultra-large multimers of von Willebrand factorform spontaneous high-strength bonds


with the platelet glycoprotein Ib-IX com-plex: Studies using optical tweezers. Blood2002; 99:3971–3977

181. Bianchi V, Robles R, Alberio L, et al: VonWillebrand factor-cleaving protease (AD-AMTS13) in thrombocytopenic disorders: Aseverely deficient activity is specific forthrombotic thrombocytopenic purpura.Blood 2002; 100:710–713

182. Ono T, Mimuro J, Madoiwa S, et al: Severesecondary deficiency of von Willebrand fac-tor-cleaving protease (ADAMTS13) in pa-tients with sepsis-induced disseminated in-travascular coagulation: Its correlation withdevelopment of renal failure. Blood 2006;107:528–534

183. Nguyen TC, Liu A, Liu L, et al: AcquiredADAMTS-13 deficiency in pediatric pa-

tients with severe sepsis. Haematologica2007; 92:121–124

184. Kremer Hovinga JA, Zeerleder S, KesslerP, et al: ADAMTS-13, von Willebrand fac-tor and related parameters in severe sep-sis and septic shock. J Thromb Haemost2007; 5:2284 –2290

185. Bockmeyer CL, Claus RA, Budde U, et al:Inflammation-associated ADAMTS13 defi-ciency promotes formation of ultra-largevon Willebrand factor. Haematologica2008; 93:137–140

186. Reiter RA, Varadi K, Turecek PL, et al:Changes in ADAMTS13 (von-Willebrandfactor-cleaving protease) activity after in-duced release of von Willebrand factorduring acute systemic inflammation.Thromb Haemost 2005; 93:554 –558

187. Martin K, Borgel D, Lerolle N, et al: De-creased ADAMTS-13 (A disintegrin-like andmetalloprotease with thrombospondin type1 repeats) is associated with a poor progno-sis in sepsis-induced organ failure. CritCare Med 2007; 35:2375–2382

188. Grewal PK, Uchiyama S, Ditto D, et al: TheAshwell receptor mitigates the lethal coagulopa-thy of sepsis. Nat Med 2008; 14:648–655

189. van’t Veer C, van der Poll T. Keeping bloodclots at bay in sepsis. Nat Med 2008; 14:606–608

190. Treziack S, Cinel I, Dellinger RP, et al:Resuscitating the microcirculation in se-vere sepsis: The central role of nitric oxide,emerging concepts for novel therapies, andchallenges for clinical trials. Acad EmergMed 2008; 15:1–15


Sepsis primer

Health & Medicine

host response

systemic inammatory

innate immune response

severe sepsis

adaptive immune response

marked systemic response

sustained inammatory

molecular patterns pamps