Published online before print October 9, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Surgery and
Center for Inflammation and Regenerative Modeling, McGowan Institute for Regenerative Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
1 Correspondence: Department of Surgery, F-1200 PUH, University of Pittsburgh, 200 Lothrop Street, Pittsburgh, PA 15217, USA. E-mail: billiartr{at}upmc.edu
ABSTRACT
Endogenous damage-associated molecular pattern (DAMP) molecules are released from cells during traumatic injury, allowing them to interact with pattern recognition receptors such as the toll-like receptors (TLRs) on other cells and subsequently, to stimulate inflammatory signaling. TLR4, in particular, plays a key role in systemic and remote organ responses to hemorrhagic shock (HS) and peripheral tissue injury in the form of bilateral femur fracture. TLR4 chimeric mice were generated to investigate the cell lineage in which functional TLR4 is needed to initiate the injury response to trauma. Chimeric mice were generated by adoptive bone marrow (BM) transfer, whereby donor marrow was given to an irradiated host using reciprocal combinations of TLR4 wild-type (WT; C3H/HeOuJ) and TLR4 mutant (Mu; C3H/HeJ) mice. After a period of engraftment, chimeric mice were then subjected to HS or bilateral femur fracture. Control groups, including TLR4-WT mice receiving WT BM and TLR4-Mu mice receiving Mu BM, responded to injury in a similar pattern to unaltered HeOuJ and HeJ mice, and protection was afforded to those mice lacking functional TLR4. In contrast, TLR4-WT mice receiving Mu BM and TLR4-Mu mice receiving WT BM demonstrated intermediate inflammatory and cellular damage profiles. These data demonstrate that functional TLR4 is required in BM-derived cells and parenchymal cells for an optimal inflammatory response to trauma.
Key Words: hemorrhagic shock (HS) • femur fracture • chimerism • pattern recognition receptor (PRR) • damage-associated molecular pattern (DAMP)
INTRODUCTION
Morbidity and mortality following trauma result in part from the activation of inflammatory signaling pathways [1 2 3 4 5 6 ]. The most severely injured patients typically experience a combination of global hypoperfusion, resulting from hemorrhagic shock (HS) and tissue injury [7 8 9 ]. Studies in mice characterizing how these insults lead to activation of inflammatory signaling have implicated TLR4 as a key sensor of tissue hypoperfusion [10 , 11 ] and tissue injury [12 ]. It is important that evidence indicates that the release of endogenous molecules, referred to as damage-associated molecular pattern (DAMP) molecules, following tissue injury leads to TLR4 stimulation [13 ]. For example, neutralizing antibodies to the endogenous TLR4 ligand high-mobility group box 1 (HMGB1) inhibit early inflammatory signaling and mimic the TLR4-deficient state in hepatic ischemia/reperfusion (I/R) injury [14 ], HS [15 ], and bilateral femur fracture [16 ].
TLR4 is a member of a highly conserved family of pattern recognition receptors comprised of 10 known members in humans and 13 in mice [17 ]. The role of TLR family members in the recognition of microbial products is well established. TLR4 may be unique among TLR family members in its capacity to recognize a number of DAMP molecules and thus, play a role in sterile inflammatory processes such as trauma [18 ]. However, the cell type-specific expression of TLR4 is widespread and includes immune cells, such as macrophages and dendritic cells (DC), as well as parenchymal cells, such as hepatocytes and intestinal epithelial cells [19 20 21 22 23 24 25 ].
As HS and remote tissue injury activate inflammatory signaling in immune cells and parenchymal cells, we sought to determine the nature of TLR4 activation [26 ]. Specifically, our primary question was on what cell types is TLR4 required for the initiation of the inflammatory and hepatic damage responses after traumatic injury. We believe that answering this question is crucial to understanding the mechanism of the role of TLR4 in trauma. To address this question, chimeric mice were generated by adoptive bone marrow (BM) transfer using TLR4 wild-type (WT) mice (C3H/HeOuJ) and TLR4 mutant (Mu) mice (C3H/HeJ), which have a point mutation in the intracellular TLR4 signaling domain. Herein, we show that functional TLR4 on BM-derived cells and parenchymal cells is required for the early inflammatory response for HS and bilateral femur fracture.
MATERIALS AND METHODS
Reagents
All reagents were from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise indicated.
Animals
Mice used in the experimental protocols were housed in accordance with University of Pittsburgh (Pittsburgh, PA, USA) and National Institutes of Health (NIH; Bethesda, MD, USA) animal care guidelines in specific pathogen-free conditions. The animals were maintained in the University of Pittsburgh Animal Research Center with a 12-h light-dark cycle and free access to standard laboratory feed and water. Male C3H/HeOuJ mice, C3H/HeJ mice, GFP-expressing mice, and C57Bl/6J mice (Jackson Laboratories, Bar Harbor, ME, USA), 8 weeks old and weighing 20–30 g, were used in chimera generation. All animals were fasted for 
12 h prior to experimental manipulation and were acclimatized for 7 days prior to being studied.
Generation of chimeric animals
Chimeric mice were generated by adoptive transfer of donor BM cells into irradiated recipient animals as described previously using combinations of TLR4 WT (C3H/HeOuJ) and TLR4 Mu (C3H/HeJ) mice [27
]. The following recipient/donor combinations were produced: WT/WT, WT/Mu, Mu/Mu, Mu/WT. GFP-chimeric animals were similarly generated by adoptive transfer of BM cells from GFP-expressing mice into C57Bl/6J WT mice. Chimeric animals were maintained under the same conditions as described above and underwent the experimental injury protocol or sham procedure 10–12 weeks after the adoptive transfer to ensure stable engraftment.
Murine fracture model
These animal research protocols complied with the regulations regarding the care and use of experimental animals published by NIH and were approved by the Institutional Animal Use and Care Committee of the University of Pittsburgh. As described previously [12
], animals were anesthetized with i.p. sodium pentobarbital (50 mg/kg) and inhaled isofluorane (Abbott Labs, Chicago, IL, USA). Using the sterile technique, a left groin exploration was performed, and the left femoral artery was cannulated with tapered polyethylene (PE)-10 tubing and connected to a blood pressure transducer (Micro-Med, Tustin, CA, USA) for continuous mean arterial pressure (MAP) monitoring for the duration of the experiment (6 h). A bilateral, closed mid-shaft femur fracture was then performed using two Hemostats applied to the hind-limb region. MAP was maintained above 60 mmHg throughout the experiment with the administration of Lactated Ringer’s solution (Baxter Corp., Deerfield, IL, USA) through the femoral cannula as needed in 0.1 mL boluses. This served to ensure that the animals were not in a state of circulatory shock. According to the manufacturer, the endotoxin content of the Lactated Ringer’s used was 0.008 EU/mL. Sham-operated mice underwent anesthesia and femoral cannulation only. All mice were re-anesthetized with i.p. sodium pentobarbital (20 mg/kg) as necessary throughout the experiment. Baseline MAP, total anesthetic dose, and volume of Lactated Ringer’s administered did not differ between species or experimental groups (sham vs. fracture). At the end of 6 h, mice were killed under inhalational anesthesia via cardiac puncture technique. Necropsy was performed to verify the presence of bilateral femur fractures and to ensure the absence of fracture site hematomas. Serum from postmortem blood samples was obtained for cytokine and blood chemistry analysis. Organs were snap-frozen in liquid nitrogen for molecular analysis.
Murine HS protocol
As described previously [11
], mice were anesthetized with i.p. sodium pentobarbital (50 mg/kg) and inhaled isofluorane (Abbott Laboratories). Unilateral groin dissections were performed, and femoral arteries were cannulated with tapered PE-10 tubing, flushed with heparin sulfate (Pharmacia, Uppsala, Sweden, and Upjohn, Kalamazoo, MI, USA), for a total of 2 U heparin per animal. The groin catheter was connected to a blood pressure transducer (Micro-Med) for continuous MAP readings. Mice were allowed to recover from the inhalational anesthesia for 10 min before initiation of hemorrhage. After baseline blood-pressure readings, repeated three times, mice were hemorrhaged to a MAP of 25 mmHg over 5 min. Total withdrawn blood was recorded every 10 min, and mice were maintained at a MAP of 25 mmHg for 150 min. The mice were then resuscitated over 10 min with their remaining shed blood plus two times the maximal shed blood amount in Lactated Ringer’s solution through the arterial catheter. After post-resuscitation blood pressure readings, catheters were removed, vessels were ligated, and groin incisions were closed with 4-0 nylon sutures. At 4 h, after the end of hemorrhage, the animals were killed under inhalational anesthesia. Serum from postmortem blood samples was obtained for cytokine and blood chemistry analysis. Organs were snap-frozen in liquid nitrogen for molecular analysis.
Hepatic nonparenchymal cell (NPC) isolation and cell culture
Hepatic NPC were isolated from mice by an in situ collagenase (type VI; Sigma-Aldrich) perfusion technique, modified as described previously [28
]. NPC were separated from hepatocytes by two cycles of differential centrifugation (50 g for 2 min) and purified further over a 30% Percoll gradient. The purity of these NPC cultures exceeded 98% by light microscopy, and viability was typically over 95% by trypan blue exclusion assay. NPC were plated onto 12-well culture plates and allowed to adhere overnight. Cells were then treated with LPS 100 ng/ml for 6 h, after which time, supernatants were harvested and frozen in –80°C. Culture medium was Williams medium E (Gibco-BRL, Grand Island, NY, USA), containing 10% calf serum, 15 mM HEPES, 10–6 M insulin, 2 mM L-glutamine, 100 U/ml penicillin, and 100 U/mL streptomycin. Before experiments, the cell culture media were changed to serum-free media.
Serum alanine aminotransferase (ALT) assay
To assess hepatocellular damage following bilateral femur fracture, serum ALT levels were measured using the Opera Clinical Chemistry System (Bayer Co., Tarrytown, NY, USA).
Serum IL-6, IL-10, and TNF ELISA
Serum IL-6 and IL-10 levels were used as a means of evaluating systemic inflammation and were quantified with ELISA kits (R&D Systems Inc., Minneapolis, MN, USA). IL-6 and TNF levels in supernatants were similarly analyzed.
EMSA
NF-
B DNA-binding activity was measured by EMSA using nuclear extracts prepared from liver tissue. Livers harvested at the conclusion of the experimental protocol were snap-frozen in liquid nitrogen and stored at –80°C. Preparation of nuclear extracts and performance of EMSA were done as described previously [12
].
Immunofluorescence microscopy
Harvested livers were cryoprotected in 2% paraformaldehyde for 4 h and then rehydrated in 30% sucrose overnight. Similarly, sections of ileum were harvested, flushed with ice-cold PBS (pH 7.4), and cryoprotected as above. Samples were then frozen in liquid nitrogen-cooled isopentane before transfer to –80°F for storage until ready to be cut. Samples were then embedded in OCT compound and cut into 6 µm sections using a cryostat machine and placed on slides, which were left to dry at room temperature overnight and then stored at –20°F. Immunofluorescence staining began by rehydrating slides with PBS. Nonspecific binding was blocked using 2% BSA for 45 min followed by rinses with 0.5% BSA. Samples were then treated with a CD68 antibody or a CD11c antibody diluted in 0.5% BSA for 60 min. After washes with 0.5% PBS, a secondary antibody in 0.5% BSA was applied for 60 min. A nuclear stain was then applied for 30 s. Slides were rinsed with PBS three times, and a cover slip was applied with gelvatol, a water-soluble mounting medium (21 g polyvinyl alcohol, 52 ml water, sodium azide, and 106 ml 0.2 M Tris buffer). Slides were visualized using an Olympus BX51 epifluorescence microscope and digitized with an Olympus color video camera. Sections were evaluated for immunofluorescence at 10–40x magnification. Efficiency of engraftment was calculated by counting donor (orange) and host (red) DC and macrophages in four high-power fields of liver and gut sections from each mouse. Results are expressed as a ratio of donor cells:total cells.
Statistical analysis
Results are expressed as the mean ± SEM. Group comparisons were assessed using the Student’s t-test or Mann-Whitney Rank Sum Test. The null hypothesis was rejected for P < 0.05 (
=0.05). Data were analyzed using SigmaStat® Version 3.1 (SPSS, Chicago, IL, USA). All experimental groups consisted of n = 8–12 mice per group.
RESULTS
GFP chimeric mice demonstrate engraftment
The TLR4 chimeric mice used in these studies are essentially "functional chimeras" in that the HeJ strain contains nonfunctional TLR4 protein containing a single base-pair mutation [29
]. As such, we are unable to evaluate the success of chimerism in these mice via protein-based assays or flow cytometry. To determine the efficiency of BM engraftment using our protocol, chimeric animals were generated by adoptive transfer of BM cells from GFP-expressing mice into C57Bl/6J WT hosts. Ten weeks after BM transfer, the animals were killed and tissues harvested. Immunohistochemistry for markers of DC (CD11c) and macrophages (CD68) was subsequently performed in the liver and ileum (Fig. 1
). The degree of colocalization of GFP-positive cells with either marker was quantified. This analysis demonstrated that 85.43 ± 8.62% of DC and 76.45 ± 7.11% of macrophages were of donor origin.
 View larger version (70K): [in a new window] |
Figure 1. GFP chimeric mice demonstrate engraftment. Chimeric mice were produced by adoptive BM transfer from GFP mice into irradiated C57Bl/6 hosts. To confirm chimeric status, liver and ileum sections were stained for CD11c (A and C) or CD68 (B and D). Images shown are representative sections from four GFP-chimeric mice.
|
 View larger version (12K): [in a new window] |
Figure 2. TLR4 chimera generation: functional assessment of chimerism. Chimeric mice were produced by adoptive BM transfer into irradiated hosts using the following combinations from TLR4-WT (C3H/HeOuJ) and TLR4-Mu (C3H/HeJ) mice: WT/WT, WT/Mu, Mu/Mu, and Mu/WT. NPC from representative mice of each chimeric group were harvested at 10 weeks and treated with LPS (100 ng/ml) for 6 h. IL-6 (A) and TNF (B) levels in supernatants were then analyzed by ELISA. Data represent means ± SEM. Data presented represent two experiments completed in triplicate.
|
 View larger version (27K): [in a new window] |
Figure 3. Maximal systemic inflammation after traumatic injury requires functional TLR4 on BM-derived cells and non-BM-derived cells. Chimeric mice underwent HS, bilateral femur fracture, or corresponding sham procedure. Serum IL-6 and IL-10 levels were analyzed for analysis of systemic inflammation. WT/WT mice subjected to fracture demonstrated significantly increased levels of IL-6 and IL-10 as compared with Mu/Mu or WT/Mu fracture mice (A and C, *, #, P<0.05). Levels were increased significantly in Mu/WT fracture mice as compared with Mu/Mu fracture mice (A and C, , P<0.05). IL-6 levels in Mu/WT fracture mice were decreased significantly compared with WT/WT fracture mice (A, ±, P<0.05). WT/WT mice subjected to HS demonstrated significantly increased IL-6 levels as compared with Mu/Mu or Mu/WT mice (B, *, #, P<0.05). Mu/WT mice demonstrated significantly increased IL-6 levels compared with Mu/Mu mice (B, , P<0.05). WT/WT mice subjected to HS demonstrated significantly increased IL-10 levels as compared with Mu/Mu, Mu/WT, or WT/Mu mice (D, *, #, , P<0.05) Data represent means ± SEM; n = 8–12 mice group. Results from unaltered HeJ and HeOuJ mice subjected to fracture were 211.3 ± 24.1 versus 1408 ± 397.5 for IL-6 and 35.7 ± 12.1 versus 143.6 ± 38.4 for IL-10, respectively. Results from unaltered HeJ and HeOuJ mice subjected to HS were 349.5 ± 48.1 versus 621.6 ± 153.2 for IL-6 and 40.6 ± 9.5 versus 75.7 ± 13.2 for IL-10, respectively. SHOCK, HS; Fx, bilateral femur fracture.
|
 View larger version (19K): [in a new window] |
Figure 4. Maximal hepatic inflammation and hepatocellular damage after fracture require functional TLR4 on BM-derived cells and non-BM-derived cells. Chimeric mice underwent HS, bilateral femur fracture, or corresponding sham procedure. Serum ALT (A) levels were analyzed for analysis of hepatocellular injury. WT/WT mice subjected to fracture demonstrated significantly increased ALT levels as compared with Mu/Mu mice (A, *, P<0.05). WT/WT mice subjected to HS demonstrated significantly increased ALT levels as compared with Mu/Mu or Mu/WT mice (B, *, #, P<0.05). Mu/WT mice demonstrated significantly increased ALT levels compared with Mu/Mu mice (B, , P<0.05). Data represent means ± SEM; n = 8–12 mice group. Results from unaltered HeJ and HeOuJ mice subjected to fracture were 101.9 ± 18.9 and 486 ± 325.4, respectively. Results from unaltered HeJ and HeOuJ mice subjected to HS were 309.1 ± 76.8 versus 1056.4 ± 369.3, respectively.
|
B activation to assess the activation of inflammatory signaling pathways in the liver in the two injury models. EMSA. performed on nuclear extracts obtained at 6 h after fracture and 6.5 h after induction of HS, showed that femur fracture and HS led to NF-B activation, relative to sham-treated animals in the WT/WT group (Fig. 5
). Mu/Mu mice exhibited a significantly lower level of NF-B activation when subjected to femur fracture or HS, as compared with the WT/WT group. In the femur fracture group, the expression of functional TLR4 on parenchymal or BM-derived cells alone resulted in an intermediate level of NF-B activation, which did not differ significantly from either control group. A similar pattern was seen in HS, except that the expression of functional TLR4 on BM-derived cells alone was adequate to reconstitute the full NF-B response measured at this time-point. Thus, the activation of NF-B in the liver parallels the cytokine and injury pattern within our models, with the exception of the fact that TLR4 on BM-derived cells alone is adequate to activate NF-B in the liver in HS.
 View larger version (41K): [in a new window] |
Figure 5. Maximal liver inflammation after fracture requires functional TLR4 on BM-derived cells and non-BM-derived cells. Hepatic nuclear extracts show that WT/WT fracture mice demonstrate significantly increased hepatic NF-B activation as compared with Mu/Mu fracture mice (A, *, P<0.05). WT/Mu and Mu/WT fracture mice demonstrate intermediate levels of NF-B. In the setting of HS, WT/WT fracture mice demonstrate significantly increased hepatic NF-B activation as compared with Mu/Mu fracture mice (B, *, P<0.05). Mu/WT fracture mice demonstrate significantly increased NF-B activation as compared with Mu/Mu mice (B, #, P<0.05). Blots shown are representative images. Statistics are drawn from three to four animals per group, run from multiple blots.
|
TLR signaling is a key mechanism by which the innate immune system recognizes and responds to microbial products, initiating a potent inflammatory response [34 , 35 ]. More recently, endogenous DAMP molecules have been demonstrated to interact with members of this receptor family as a means of alerting the host to nominally sterile injury [13 ]. TLR4, in particular, has been shown to interact with and respond to a variety of endogenous molecules, including fibrinogen [36 ], heparan sulfate [37 ], heat shock proteins [38 ], and HMGB1 [39 , 40 ]. We and others have demonstrated profound protection from sterile injury among animals with absent or Mu TLR4 in the settings of HS [10 , 11 ], femur fracture [12 ], and hepatic I/R injury [14 ], as well as other forms of sterile injury [13 ]. Taken together, these studies point to an interaction between endogenous activators and TLR4 at the cell surface in target organs subsequent to injury.
Although the importance of TLR4 has been demonstrated in models of traumatic injury, the mechanism and location of TLR4 activation during injury have not. The purpose of this study was to investigate the cell lineages on which TLR4 signaling is required to initiate systemic inflammation and end organ damage after trauma. To do so, we created BM chimeric mice and subjected them to two clinically relevant models of trauma. In addition to demonstrating our capacity to generate chimeric animals with high efficiency, we showed that functional TLR4 is required on BM-derived cells and non-BM-derived cells for the initiation of the robust, systemic inflammatory and hepatic damage responses after bilateral femur fracture and HS. Other studies have demonstrated clear roles for TLR4 signaling on immune cells such as macrophages, DC, and neutrophils in the setting of sterile injury [27
, 41
, 42
]. More recently, the significance of TLR4 signaling on platelets has been demonstrated to play a role in the host response to septic insults [43
]. In addition, it has long been known that leukocyte activation and priming play an important role in the early response to trauma. Activated neutrophils, in particular, have been implicated in the post-trauma, inflammatory response within 24 h of injury [44
, 45
]. This process of priming is believed to contribute to the development of systemic inflammatory response syndrome (SIRS) and septic complications after injury [46
, 47
]. The high abundance of TLR4 on leukocytes makes it reasonable to suspect that early activation of leukocyte TLR4 may drive the early post-trauma inflammatory cascade [48
, 49
]. Further, we have demonstrated previously a central role for functional TLR4 on BM-derived cells of the liver in the initiation of hepatic inflammation and injury after I/R injury [27
]. However, it has also been demonstrated that TLR4 protein is present to a significant degree on parenchymal cells. For instance, we know that hepatocytes and intestinal endothelial cells express functional TLR4 [19
, 20
, 50
, 51
]. Hepatocyte TLR4 plays a role in LPS-mediated NF-B activation and LPS uptake (Scott, M., Billiar, T. R., unpublished results), and enterocyte TLR4 has been shown to be required for uptake of bacteria by intestinal epithelial cells [20
, 52
]. Further, significant crosstalk between TLR4-WT hepatocytes and nonparenchymal cells is required for an optimal response to LPS. These data point to the physiologic importance of TLR4 activation on BM-derived cells and parenchymal cells in the host response to LPS, as mediated through TLR4. Our findings in this study are significant in that they demonstrate for the first time a role for TLR4 signaling on nonhematopoietic cells in the initiation and propagation of the inflammatory and end organ damage responses to trauma. This surprising result has changed fundamentally the way in which we think about the host response to sterile injury. The relative importance of TLR4 on specific cell populations may vary depending on the specific nature of the experimental model and inflammatory endpoints. Nonetheless, our findings support the conclusion that TLR4 signaling on cell types other than circulating leukocytes, tissue macrophages, and DC is important to the systemic response following injury.
The technique for chimera generation, which we used, has been validated by our group and others in the past [27 , 53 54 55 ]. These studies would suggest that a dose of 1000 cGy radiation is otherwise lethal and provides for rapid myloablation. Mice are able to survive only upon reconstitution of their BM with donor cells. The extent of engraftment in the blood over time has been evaluated extensively. Studies would suggest that the percent chimerism in the blood achieves a peak by 10–12 weeks at 80–90% chimerism and remains largely stable from then on [54 ]. Our TLR4 chimeric mice are essentially functional chimeras, in that we have exchanged cells containing a normal, functioning TLR4 molecule for cells, which express a nonfunctional TLR4 protein containing a single base-pair mutation and visa versa. Thus, we used a functional assay to evaluate the extent of chimerism in the liver parenchyma as a surrogate indicator for whole body chimerism. Previous studies have shown that Kupffer cells repopulate the liver within 14 days after their elimination by the administration of liposome-entrapped clodronate in mice [56 ]. In addition, Kupffer cells of donor origin repopulate the recipient livers within 14–21 days after mouse marrow transplantation [57 ]. Our in vitro assay experiments would suggest a successful conversion of hepatic NPC to the donor phenotype by 10 weeks. The GFP-chimeric mice offer us the opportunity to "see" the success of chimera generation, particularly at the tissue level. As we focus on the liver as an indicator of end-organ damage in our models, we evaluated the success of chimera formation in the livers of these mice. To be sure that the extent of engraftment in tissues is not a liver-specific response, we also evaluate the extent of chimerism in another organ known to express TLR4 to a significant degree—the ileum [19 ].
In analyzing the response to traumatic injury in chimeric mice, we focused primarily on the release of IL-6 and IL-10. Levels of these two cytokines are known to correlate with the severity of injury and the subsequent development of SIRS in human trauma patients [32 , 33 , 58 ]. Further, these cytokines have been found to be elevated significantly in HeOuJ mice subjected to HS or femur fracture as compared with HeJ mice [11 , 12 ] and correlate with hepatic damage in both models. Here, we demonstrate a role for TLR4 on donor and host cells in the initiation of the cytokine response to trauma. Systemic inflammation, as assessed by serum IL-6 levels after femur fracture, is minimal, if functional TLR4 is absent from either cell type. TLR4 signaling on BM-derived cells may contribute comparatively more to the IL-10 response to fracture, however. In contrast, the systemic IL-10 response to HS is minimal if functional TLR4 is absent from either cell type. It is interesting that the IL-6 response to HS was not statistically different between WT/WT and WT/Mu mice, indicating a strong role for TLR4 activation on non-BM-derived cells in the initiation of this response.
Although hepatocellular damage after trauma is most likely multifactorial, coming as a consequence of reduced tissue perfusion, the release of reactive oxygen species, as well as systemic cytokine release, it has also been found to be highly dependent on the presence of TLR4 [11
, 12
, 59
60
61
]. Using ALT levels as an index of cell damage and membrane breech, we find evidence of mild hepatocellular damage in the setting of fracture. This small but significant increase in ALT levels in the WT/WT group as compared with Mu/Mu mice is dependent on the presence of TLR4 on parenchymal and BM-derived cells. It is interesting that hepatic damage in the setting of HS is highly dependent on the presence of TLR4 on parenchymal cells. The observation that significant differences are seen between Mu/Mu animals and Mu/WT animals argues for some contribution from functional TLR4 on hematopoietic cells. We also assess the extent of NF-B DNA binding in the livers of chimeric mice. Studies have suggested that hepatic NF-B binding serves as an important hepatic inflammatory mediator in a number of injury models and that it correlates in general with hepatic damage [62
63
64
]. It is interesting that as with ALT, NF-B binding is dependent on the presence of TLR4 on parenchymal cells and BM-derived cells in the setting of fracture. However, the NF-B response in HS seems to be more dependent on the presence of TLR4 on hematopoeitic cells. Perhaps even more interesting is the fact that the NF-B response appears to be more robust after fracture. Thus, although the intra-injury group comparisons match those seen with ALT and systemic cytokine levels, there is discordance when comparing between models with a robust ALT response but a comparatively small NF-B response in hemorrhage. This may simply be a result of the different mechanisms of injury. It is unclear how this may occur, although at least one study suggests that the hepatic NF-B response is muted in the setting of an ischemic, perfused liver [64
]. Thus, the ischemic nature of injury may modulate the TLR4 response.
This study did not address the potential ligands involved in TLR4 activation. From the work of others [66
] and our own results (unpublished results), using germ-free animals, we know that the presence or absence of intestinal flora does not impact the level of systemic inflammation in either model. In contrast to microbial products, we have demonstrated a role for the DAMP and TLR4 ligand HMGB1 in the inflammatory response in the HS [15
] and bilateral femur fracture models [16
]. The source of HMGB1 is uncertain, and it is likely that other DAMP molecules contribute to TLR4 activation, especially in the HS model in which HMGB1 neutralization only partially inhibits inflammation and hepatocellular damage [15
]. Thus, we would hypothesize that HMGB1 is released from injured or stressed cells soon after trauma, allowing it to interact with the TLR4 signaling apparatus and initiate signaling terminating in NF-B activation and inflammatory cytokine release, although no direct interaction between TLR4 and HMGB1 has been demonstrated in vivo. The mechanism of the role of TLR4 in the setting of multiple models of sterile inflammation has been equally elusive [13
].
The TLR4 signaling cascade is surely only one of many factors contributing to the traumatic injury response, including a heightened adrenergic response, transient hypoxia, and the generation of free radicals, as well as the initiation of multiple proinflammatory signaling cascades leading to the release of inflammatory mediators [59 ]. However, our data would suggest that TLR4 is crucial to the initiation of this inflammatory response. It is interesting that a similar level of dependence on TLR4 is seen in other models of sterile inflammation, including I/R injury [14 , 67 , 68 ], ethanol-induced liver injury [69 ], and heparin sulfate-induced SIRS [70 ]. Data would also suggest that TLR4 activation after hemorrhage leads to up-regulation of other members of the TLR family including TLR2 and -9 [71 72 73 ]. Thus, TLR4 activation after traumatic injury appears to set in motion a number of different proinflammatory signaling cascades.
In summary, this study demonstrates that functional TLR4 on hematopoietic cells and non-BM-derived cells participates in the initiation of the systemic inflammatory response following peripheral tissue injury or HS. Although HS represents a complex form of injury consisting largely of global hypoperfusion, and bilateral femur fracture represents an isolated, nonischemic form of peripheral tissue injury, both forms of trauma exhibit similar requirements for functional TLR4 on BM and non-BM-derived cells. Therapeutic strategies targeting TLR4 signaling could be beneficial to limit the magnitude of the systemic inflammatory response in severely injured patients.
ACKNOWLEDGEMENTS
This work was supported by National Institutes of Health grant GM-53789-08 (T. R. B. and Y. V.). K. P. M., J. M. P., D. J. K., and R. V. are each recipients of American College of Surgeons Resident Research Scholarships. The authors acknowledge the technical assistance of Hong Liao, Lauryn Kohut, and Derek Barclay.
Received April 2, 2007; revised July 4, 2007; accepted September 1, 2007.
REFERENCES
B regulation and cytokine expression J. Clin. Invest. 99,1516-1524[Medline] B activation in mouse hepatocytes stimulated with lipopolysaccharide Infect. Immun. 70,3433-3442B through Toll-like receptor 4 (TLR-4) in cultured human dermal endothelial cells. Differential expression of TLR-4 and TLR-2 in endothelial cells J. Biol. Chem. 275,11058-11063B in cancer development and progression Nature 441,431-436[CrossRef][Medline] B and severe hepatic necrosis may mediate systemic inflammation in choline-deficient/ethionine-supplemented diet-induced pancreatitis Pancreas 33,260-267[CrossRef][Medline] B activity and 1-acid glycoprotein transcription do not predict outcome after cecal ligation and puncture in the rat Crit. Care Med. 29,589-596[CrossRef][Medline]B activation and TNF- bioactivity in perfused liver Am. J. Physiol. Regul. Integr. Comp. Physiol. 279,R99-R108
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
