Published online before print November 21, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
1Correspondence: Department of Surgery, F1281, Presbyterian University Hospital, University of Pittsburgh Medical Center, 200 Lothrop St., Pittsburgh, PA 15213, USA. E-mail: billiartr{at}upmc.edu
ABSTRACT
The systemic inflammatory response observed in the setting of overwhelming infection bears striking similarities to that observed in the setting of severe traumatic injury from a clinical and physiologic standpoint. Recent observations have demonstrated that these disparate clinical entities share common mediators on a molecular level. TLRs, specifically TLR4, and the endogenous molecule high-mobility group box 1 are among the mediators that are known to play a role in inflammation in the setting of sepsis. Evidence is accumulating that demonstrates that these mediators also play a role in the host response to tissue injury. Here, we highlight findings from the 7th World Conference on Trauma, Shock, Inflammation and Sepsis in Munich, Germany, in the context of this growing body of literature.
Key Words: Toll-like receptors • HMGB1 • inflammation • trauma
TLRs AND IMMUNITY
The TLRs are a family of molecules that have recently been found to play a central role in the immune response to microbial pathogens [1 , 2 ]. Lamaitre et al. [3 ] first observed in 1996 that Toll, a protein involved in Drosophila development, also controls the antifungal response in adult flies. A family of homologous receptors—TLRs—was subsequently identified in higher organisms, including humans [4 ]. Up to 11 TLRs have been identified in humans and 13 in mice to date. These receptors contain an extracellular domain with leucine-rich repeats, a transmembrane domain, and a cytoplasmic Toll/IL-1R (TIR) signaling domain [5 6 7 ]. Although they share similar protein domains, the subcellular distribution of TLRs varies. Although TLR1, -2, -4, -5, and -6 are predominantly located at the cell surface, TLR3, -7, -8, and -9 are localized to the endosomal compartment [8 ].
TLRs are expressed on a number of different cell types that mediate the innate immune response, including monocytes, macrophages, and neutrophils [9 ]. Interestingly, they are also expressed on dendritic cells (DC) [10 ] and T lymphocytes [10 ], suggesting that in addition to their central role in innate immunity, they may serve as a bridge between the innate and adaptive immune systems. TLRs are also expressed on a variety of parenchymal and epithelial cell types, implicating broader and ligand recognition and cell signaling functions for the TLR family.
The TLRs serve as pattern recognition receptors (PRRs). These receptors recognize conserved molecules found in a broad array of microbes and alert the immune system to their presence. These conserved TLR ligands have been variously referred to as pathogen-associated molecular patterns or microbial-associated molecular patterns, as they are also found in commensal organisms as well as pathogens. A variety of macromolecules serves as microbial ligands of TLRs and includes lipids, carbohydrates, proteins, and nucleic acids [1 ].
TLR SIGNALING OVERVIEW
The precise molecular events that follow ligand binding by TLRs are not entirely understood and are currently under investigation. It appears that TLRs form heterodimers (TLR1 with TLR2 and TLR2 with TLR6, for example) or homodimerize (TLR4 and TLR9). TLR4 may be induced to form stable dimers and even tetramers after LPS binding [11
]. After ligand binding, these dimers likely undergo conformational changes [12
] that lead to association of the individual TIR domains. There are up to five different signaling adaptor molecules that can then potentially be recruited to the signaling complex. These adaptors, each of which also contain TIR domains, include MyD88, MyD88-adaptor-like (MAL; also referred to as TIR domain-containing adaptor protein), TIR domain-containing adaptor-inducing IFN-β (TRIF), TRIF-related adaptor molecule (TRAM), and sterile 
- and armadillo motif-containing protein (SARM) [13
].
MyD88 has been identified as a key adaptor for TLR signaling [14
]. It was subsequently determined that it serves as an adaptor for most TLRs, with the exception TLR3. MyD88 is recruited to the TIR domains of the TLRs after ligand binding. IL-1R-associated kinase 4 (IRAK4) is a crucial molecule that subsequently recruited in cooperation with MAL in the case of TLR2 and TLR4 [15
16
17
18
]. Activation of other intermediate signaling molecules, including IRAK1, TNFR-associated factor 6 (TRAF6), and TGF-β-activated kinase 1, leads to eventual downstream activation of NF-
B, as well as p38, JNK, and others [19
20
21
]. A number of regulatory molecules modulate this signaling pathway. For example, MyD88 signaling can be inhibited by a shorter form of the molecule, known as MyD88s, by preventing IRAK4 recruitment [22
, 23
]. Similarly, IRAK-M serves a negative regulatory role by preventing release of IRAK1 and IRAK4 [24
]. Recruitment of MAL is regulated by Bruton’s tyrosine kinase (BTK) and suppressor of cytokine signaling 1 [25
, 26
].
In addition to this MyD88-dependent pathway, TLR3 and in some settings, TLR4 can signal through another pathway that involves the adaptor TRIF [27
, 28
]. Ligand binding triggers the recruitment of TRIF. In the case of TLR4, TRAM is also recruited upon activation by protein kinase C-
(PKC) [29
]. Working through TRAF3 in concert with other molecules, TRIF stimulates IFN production through TANK-binding kinase 1 (TBK1) and IFN-regulatory factor 3 (IRF3) [30
, 31
]. It should be noted that there is considerable overlap between the two pathways. TLR signaling through TRIF can result in NF-B activation, and signaling through the MyD88 pathway can result in IRF activation. Stimulating apoptotic cell death through Fas-associated death domain (FADD) appears to be an aspect of TRIF-dependent signaling that is unique to this pathway, however [32
]. Regulation of the TRIF pathway occurs at multiple levels. The TLR adaptor SARM is one molecule that appears to exert negative regulatory control on the TRIF pathway [33
].
TLR4
TLR4 is among the most extensively studied members of the TLR family. TLR4 has been identified as the molecular sensor for bacterial LPS. Poltorak and colleagues [34 ] demonstrated that mutations in the tlr4 gene were responsible for defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice.
TLR4 signaling is complex and requires interactions with multiple extracellular molecules. LPS-binding protein binds to LPS in the serum and delivers LPS to CD14 [35 , 36 ], which exists in a soluble form in addition to a GPI-anchored, membrane-bound form [37 , 38 ]. At the cell surface, myeloid differentiation protein 2 (MD-2) binds to LPS and the ectodomain of TLR4 to initiate signaling. As described above, the intracellular signal can be mediated through a MyD88-dependent pathway involving MyD88 and MAL. TLR4 signaling is also mediated through a TRIF-dependent pathway involving the adaptors TRIF and TRAM. Downstream signaling results in potent activation of the immune system and included elaboration of proinflammatory cytokines such as TNF and up-regulation of costimulatory molecules. Endogenous controls may also exert influence over TLR4 signaling [39 ]. The PI-3K/Akt pathway may also serve as a regulator of TLR4 signaling [40 ].
RECOGNITION OF ENDOGENOUS MOLECULES BY TLR4
Although TLR4 was identified initially as the sensor of bacterial LPS, and its role in mediating inflammation in response to LPS is well established, TLR4 also appears to serve as a sensor for a number of endogenous molecules, which can be released from damaged or dying cells, secreted from stressed cells, or released from degradation of the extracellular matrix (ECM). Recent observations demonstrate that a number of molecules, including heparan sulfate, fibrinogen, hyaluronan, high-mobility group box 1 (HMGB1), and potentially others, are endogenous activators of TLR4 signaling [41 ]. The observation that endogenous molecules can trigger a TLR4-dependent response implies that in addition to its role as a sensor of bacterial invasion, TLR4 might be able to detect the presence of sterile tissue injury and initiate an inflammatory response. As might be predicted based on these observations, it has been demonstrated that TLR4 plays a central role in mediating the early inflammatory response in several models of sterile tissue injury, including warm and cold ischemia/reperfusion [42 43 44 ], hemorrhagic shock (HS), and traumatic injury, as described below.
TLR4 IN TRAUMA
TLR4 has been demonstrated to play a critical role in the early inflammatory response after traumatic injury. It has been demonstrated in models of HS and soft tissue injury, such as bilateral femur fracture.
TLR4 AND HS
A role for TLR4 signaling in the inflammatory response after injury was first suggested as early as 1993, when DeMaria and coinvestigators [45 ] examined the inflammatory response after HS in mice that were resistant to the effects of LPS. The mice (C3H/HeJ) used in the study are resistant to endotoxin as a result of a mutation in the gene encoding TLR4, but at the time, this was unknown. The investigators found that mice that were resistant to LPS also exhibited significantly lower levels of TNF after HS compared with wild-type counterparts. Furthermore, the endotoxin mutant mice also had improved survival when compared with wild-type mice. Interestingly, the authors found that pretreatment with an anti-TNF antibody abolished the increase in serum TNF levels after shock but did not improve survival. At the time, the results of these experiments were interpreted as evidence for translocation of bacteria or LPS across the gut epithelial barrier in the setting of HS. However, evidence for translocation of microbes or microbial products as a driving force behind the inflammatory response after HS is otherwise lacking.
More recently, our group has found that TLR4 mutant mice (C3H/HeJ) had lower levels of circulating IL-6 and IL-10 than wild-type, TLR4-competent mice (C3H/HeOuJ) after HS and resuscitation (HS/R) [46
, 47
]. A similar response was observed when TLR4 knockout mice were compared with their wild-type counterparts, indicating that TLR4 signaling is required for systemic cytokine release after HS/R. Furthermore, TLR4 mutant mice exhibited less hepatic NF-B activation and lower serum alanine transaminase (ALT) concentrations than wild-type mice, indicating less hepatocellular injury and diminished hepatic inflammation in the TLR4 mutant mice. In contrast, CD14 knockout mice and TLR2 knockout mice were not protected from hepatocellular injury [46
, 47
].
Evidence from other laboratories also indicates that TLR4 plays a role in the host response to HS. In a separate study, Barsness and colleagues [48 ] demonstrated that acute lung injury as a result of HS also involves TLR4 signaling. Fan and coinvestigators [49 ] demonstrated that shock prevents LPS-induced down-regulation of TLR4 gene expression in the lung. It was subsequently demonstrated by Powers et al. [50 ] that oxidative stress that occurs in the setting of HS primes resident tissue macrophages through recruitment of TLR4 to the cell surface.
TLR4 AND FEMUR FRACTURE
In addition to its role in mediating inflammation and end-organ injury in HS/R, we have observed that TLR4 plays a key role in systemic inflammation observed within the first 6 h following extensive soft tissue injury [51
, 52
]. TLR4 mutant mice (C3H/HeJ) demonstrate less systemic inflammation, as measured by serum IL-6 and IL-10 levels, compared with TLR4-competent wild-type mice (C3H/HeOuJ) in the setting of bilateral femur fracture. TLR4 signaling was also required for hepatic NF-B activation. Interestingly, CD14 knockout mice were not protected from systemic inflammation in this setting [51
, 52
]. These data strongly argue that TLR4 plays a role in the recognition of traumatic injury and in the initiation of a systemic inflammatory response to peripheral tissue damage.
Taken together, these observations implicate TLR4 as a central mediator in the early inflammatory response that occurs in the setting of traumatic injury. However, these studies did not provide insight into the cell lineages in which functional TLR4 is necessary to initiate the inflammatory response in the setting of acute trauma. To address this question, we created chimeric mice through adoptive bone marrow transfer. In these experiments, we generated wild-type mice with TLR4 mutant bone marrow and TLR4 mutant mice with wild-type bone marrow. The marrow was allowed to engraft, and then the mice were subjected to HS or femur fracture. We observed intermediate levels of inflammation in animals lacking functional TLR4 signaling in bone marrow-derived cells or parenchymal cells when compared with unaltered wild-type or mutant mice. The results of these experiments indicate that TLR4 signaling by bone marrow-derived cells and parenchymal cells is involved in the early inflammatory response after trauma [47 ].
ENDOGENOUS MEDIATORS OF INFLAMMATION IN TRAUMA
A number of ligands likely serve as endogenous mediators of inflammation in trauma. HMGB1 has long been known as a small DNA-binding protein. More recently, however, evidence has accumulated that suggests that HMGB1 has proinflammatory cytokine-like properties [53 ]. Although HMGB1 may be released as a late mediator in sepsis, it can also be released early after a traumatic insult.
We have observed elevated levels of HMGB1 in the serum of adult human trauma patients compared with healthy volunteers [54 ]. In a murine model of H/S, we found that treatment of mice with neutralizing antibodies against HMGB1 resulted in lower serum concentrations of IL-6 and IL-10, ameliorated gut hyperpermeabiliaty, and improved survival at 24 h [54 ].
We have also observed that TLR4 wild-type mice that undergo bilateral femur fracture after treatment with neutralizing antibodies to HMGB1 have lower serum IL-6 and IL-10 levels compared with animals that receive a control antibody. In addition, mice that are treated with neutralizing antibody against HMGB1 exhibit lower serum ALT levels, and decreased NF-B activation is observed in the liver and gut mucosa. Treatment with anti-HMGB1 neutralizing antibody offered no additional protection in TLR4 mutant mice, suggesting that these two molecules might act in the same pathway [55
].
These data strongly argue for a role for HMGB1 as a mediator of inflammation after traumatic injury. Although their role in the inflammatory response to traumatic injury is not yet clear, a number of other molecules have been implicated as endogenous mediators of inflammation. Evidence is accumulating that demonstrates that heparan sulfate, a glycosaminoglycan, is one such mediator [56 , 57 ]. Johnson and coinvestigators [58 ] found that DC matured in response to stimulation to heparan sulfate, as determined by costimulatory molecule expression as well as their ability to stimulate T lymphocytes [58 ]. Interestingly, this effect was absent in cells lacking functional TLR4. Furthermore, the same group found that injection of soluble heparan sulfate into mice resulted in a systemic inflammatory response that was dependent on the presence of functional TLR4 [59 ]. In similar experiments, injection of elastase also resulted in a systemic inflammatory response and resulted in mobilization of heparan sulfate from tissues [59 ]. Taken together, these observations support the hypothesis that enzymatic release of heparan sulfate from tissues can initiate an inflammatory response through TLR4.
It also appears that TLR4 stimulation by hyaluronan fragments may serve as an intrinsic mechanism for initiating and inflammatory response after tissue injury. Termeer et al. [60 ] observed that low molecular weight degradation products of hyaluronic acid, a glycosaminoglycan found in the ECM, also activate DC through TLR4. In another study, hyaluronan fragments were shown to result in up-regulation of a number of inflammatory mediators and release of IL-8 by endothelial cells [61 ]. Administration of hyaluronan fragments in vivo also resulted in increased circulating levels of the cytokines MIP-2 and keratinocyte-derived chemokine. Using TLR4-deficient mice and blocking antibodies, the authors demonstrated that these effects of hyaluronan administration were TLR4-dependent [61 ]. More recent data suggest that a complex consisting of TLR4, MD-2, and CD44 may be required for recognition of hyaluronan released after sterile injury [62 ]. The unique composition of this complex may help explain how disparate responses are elicited by LPS as compared with hyaluronan in cultured cells.
In addition to its well-documented role in coagulation, fibrinogen has been shown to serve as a potential mediator of inflammation through a mechanism involving TLR4. Smiley and coinvestigators [63 ] found that fibrinogen could stimulate monocytic cells or macrophage-like cell lines to secrete the chemokine MCP-1. This response occurred in the presence of polymyxin B and was inhibited by boiling the fibrinogen that was used to stimulate the cells, suggesting that the observation was not a result of LPS contamination. In addition, chemokine secretion after stimulation with fibrinogen was reduced in TLR4-deficient cells compared with wild-type cells, suggesting a role for functional TLR4 [63 ].
Heat shock proteins (Hsp) have attracted attention as potential endogenous ligands of TLR4 [64 65 66 ]. Ohashi and colleagues [67 ] found that Hsp60 stimulated macrophages to release TNF and synthesize NO in a TLR4-dependent manner. Others demonstrated that human Hsp60 stimulated cytokine release from monocytes and that this effect could be blocked by a neutralizing anti-TLR4 antibody [68 ]. However, it has also found that preincubation of low amounts of another HSP family member, Gp96, with LPS at concentrations unable to activate cells by themselves, resulted in a pronounced inflammatory response when two were administered to cells together [69 ]. These results suggest that HSPs may potentiate the effects of microbial TLR ligands and also highlight the necessity of rigorous controls during characterization of potential endogenous mediators of inflammation.
Recently, two proteins that are found in abundance in the cytoplasm of neutrophils and monocytes were also identified as endogenous activators of TLR4 [70 ]. These proteins, Mrp8 (S100A8) and Mrp14 (S100A9), are released upon phagocyte activation and appear to amplify the cellular response to endotoxin [71 ]. The authors found that mice lacking Mrp8-Mrp14 are protected from mortality as a result of LPS administration or sepsis. Further, Mrp8 appears to be the active component. Mrp8 appears to induce intracellular translocation of MyD88 to the TLR4 signaling complex though an interaction with TLR4-MD-2 [71 ].
Continued study of these and other potential endogenous molecules, such as uric acid [72 , 73 ], may yield further insight into the mechanisms of inflammation occurring in the setting of sterile injury. Furthermore, a better understanding of the molecular interactions required for the initiation of the inflammatory response by these molecules may allow for the development of therapeutics with the potential to alleviate the adverse consequences of injury-induced inflammation.
CELLULAR AND MOLECULAR MECHANISMS
Although it is becoming clear that TLR4 and HMGB1 are involved in the early inflammatory response that occurs in the setting of acute injury, the molecular interactions involved are less clear, and this represents an area of active investigation. HMGB1 release from cells may occur as a passive process as cells undergo necrotic death. However, active mobilization of HMGB1 out of the cell nucleus and into the extracellular environment also appears to occur in response to stressors [74 ]. This process appears to be regulated by modification of HMGB1 by acetylation and phosphorylation [75 ]. Recent data suggest that HMGB1, which is secreted into the extracellular milieu, can bind to the TLR4/MD-2 complex [76 , 77 ]. After binding to TLR4/MD-2, HMGB1 appears to activate TLR4 signaling through the MyD88-dependent pathway [78 ].
Data from our group presented at the 7th World Congress on Trauma, Shock, Inflammation and Sepsis (TSIS) yield some insight into the potential molecular events occurring at the cell surface during HMGB1 signaling though TLR4 [79 ]. Our evidence suggests that HMGB1 stimulates TLR4 and MD-2 clustering within lipid rafts and that optimal HMGB1-dependent activity through TLR4 requires MD-2 and CD14. Furthermore, preliminary data suggest that HMGB1 does not bind directly to TLR4 but signals though TLR4 after binding to CD14 and MD-2, as illustrated in Figure 1 [79 ].
 View larger version (33K): [in a new window] |
Figure 1. Recognition of endogenous molecules by TLR4. A number of molecules, including HMGB1, heparan sulfate, and hyaluronan, have recently been identified as endogenous mediators of inflammation. These molecules, referred to as damage-associated molecular pattern molecules (DAMPs), are secreted by cells under stress, released during cell death, or liberated upon degradation of the tissue matrix. The molecular mechanisms of TLR4 activation by DAMPs are poorly understood. However, in the TLR4 signaling complex, CD-14 and MD-2 may bind to endogenous molecules, triggering TLR4 clustering within lipid rafts. Adaptor molecules, such as MyD88 or TRIF, could then be recruited to the cytosolic TIR domains of TLR4, and signal transduction through corresponding downstream pathways could occur. IKK, IB kinase; MKK, MEK; RIP1, receptor-interacting protein 1.
|
In addition to highlighting multiple, novel findings that provide insight into the mechanisms that mediate the early inflammatory response after traumatic injury, several other potential areas of investigation that might contribute to our understanding of this response were presented at the TSIS in Munich. For example, in addition to TLRs, several other families of PRRs exist, such as nucleotide-binding oligomerization domain (NOD)-like receptors, and NACHT-, leucine-rich repeat-, and pyrin domain-containing proteins (NALPs) [80 , 81 ]. Their role in mediating inflammation after traumatic injury has not yet been elucidated. In addition to these pathways involving PRRs, other pathways may also be involved. One example is the IL-1 pathway [82 ]. The complement system and the coagulation system appear to play some role in the early response to injury, but further study is required [83 ]. Beyond the mechanisms that mediate the early inflammatory response to sterile injury, the consequences of this inflammatory response for intermediate and long-term immunity will also be of interest. Myeloid suppressor cells, through a TLR4-dependent mechanism, may exert influence on cellular immunity [84 ]. Prolonged immunosuppression may occur after trauma, and it appears that depressed T cell function may be responsible for this observation [85 ]. To better understand the immune response to injury, mathematical modeling of the inflammatory response may be useful [86 ]. Lastly, as the mechanisms behind the early events that mediate the inflammatory response after injury become elucidated, translational approaches directed at modulating these early events and the resultant immune response might be developed [87 88 89 ].
SUMMARY AND CONCLUSIONS
Recent findings from the literature and from the 7th World Conference on TSIS in Munich, Germany, demonstrate that the mediators that drive the systemic inflammatory response in the setting of sepsis or sterile tissue injury are strikingly similar. TLRs, specifically TLR4, and the endogenous DNA-binding protein HMGB1 are mediators that are central to these responses. Multiple other molecules, such as heparan sulfate, hyaluronan, fibrinogen, and others, may also serve as endogenous mediators of inflammation after tissue injury through stimulation of TLR4. Further understanding of these pathways on a molecular level may allow targeted, therapeutic intervention, which may ultimately reduce the morbidity associated with traumatic injury.
Received June 9, 2007; revised October 3, 2007; accepted October 22, 2007.
REFERENCES
B activation by Toll-like receptor (TLR)-2 and TLR4 J. Biol. Chem. 279,37227-37230 B J. Biol. Chem. 276,41661-41667B- and AP-1-dependent gene expression FEBS Lett. 548,103-107[CrossRef][Medline]} during Toll-like receptor 4 signaling Proc. Natl. Acad. Sci. USA 103,9196-9201 B and IFN-regulatory factor-3, in the Toll-like receptor signaling J. Immunol. 171,4304-4310This article has been cited by other articles:
 |
|
 |
|
  A. Piccini, S. Carta, S. Tassi, D. Lasiglie, G. Fossati, and A. Rubartelli ATP is released by monocytes stimulated with pathogen-sensing receptor ligands and induces IL-1{beta} and IL-18 secretion in an autocrine way PNAS, June 10, 2008; 105(23): 8067 - 8072. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
