Injurious non-microbial factors released from the stressed gut during shocked states contribute to the development of acute lung injury (ALI) and multiple organ dysfunction syndrome (MODS). Since Toll-like receptors (TLR) act as sensors of tissue injury as well as microbial invasion and TLR4 signaling occurs in both sepsis and noninfectious models of ischemia/reperfusion (I/R) injury, we hypothesized that factors in the intestinal mesenteric lymph after trauma hemorrhagic shock (T/HS) mediate gut-induced lung injury via TLR4 activation.
The concept that factors in T/HS lymph exiting the gut recreates ALI is evidenced by our findings that the infusion of porcine lymph, collected from animals subjected to global T/HS injury, into naïve wildtype (WT) mice induced lung injury. Using C3H/HeJ mice that harbor a TLR4 mutation, we found that TLR4 activation was necessary for the development of T/HS porcine lymph-induced lung injury as determined by Evan's blue dye (EBD) lung permeability and myeloperoxidase (MPO) levels as well as the induction of the injurious pulmonary iNOS response. TRIF and Myd88 deficiency fully and partially attenuated T/HS lymph-induced increases in lung permeability respectively. Additional studies in TLR2 deficient mice showed that TLR2 activation was not involved in the pathology of T/HS lymph-induced lung injury. Lastly, the lymph samples were devoid of bacteria, endotoxin and bacterial DNA and passage of lymph through an endotoxin removal column did not abrogate the ability of T/HS lymph to cause lung injury in naïve mice.
Citation: Reino DC, Pisarenko V, Palange D, Doucet D, Bonitz RP, Lu Q, et al. (2011) Trauma Hemorrhagic Shock-Induced Lung Injury Involves a Gut-Lymph-Induced TLR4 Pathway in Mice. PLoS ONE6(8): e14829. https://doi.org/10.1371/journal.pone.0014829
Editor: Rory Edward Morty, University of Giessen Lung Center, Germany
Received: November 23, 2009; Accepted: June 16, 2011; Published: August 4, 2011
Copyright: © 2011 Reino et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by NIH grants R01-GM059841 (EAD), 1P50GM069790 (to EAD and RF) and T32-GM069330 (to DCR, VP, DD and SUS). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Acute lung injury (ALI) and the development of the acute respiratory distress syndrome (ARDS) is one of the major causes of death in trauma patients and is associated with mortality rates as high as 40% in intensive care unit patients . The mortality rate of ALI and ARDS has only modestly improved over the last several decades, since the pathophysiology of this syndrome remains incompletely understood and thus therapy remains largely supportive. Consequently, extensive investigative efforts have been focused on clarifying the basic biology of trauma hemorrhagic shock (T/HS)-induced ALI and ARDS. This work has generated several working hypotheses, one of which is that acute lung injury after trauma-hemorrhage is secondary to factors released from the stressed gut . A key element in this gut hypothesis of ARDS is that a splanchnic ischemia-reperfusion (I/R) insult leads to gut inflammation and injury, which in turn results in a gut-induced systemic inflammatory state as well as lung injury. Although gut ischemia has been implicated in the development of ALI, the mechanisms by which a gut I/R event leads to lung injury remains to be fully clarified. Based on our studies showing that T/HS-induced lung injury can be prevented in both rodents and non-human primates by ligating the mesenteric lymph duct ,  and that the injection of T/HS mesenteric lymph, collected from animals subjected to T/HS, into naïve rodents recreates the lung injury observed after actual T/HS , we have proposed the hypothesis that gut-derived lymph is the vehicle by which gut injury leads to distant organ injury . The mechanisms by which T/HS lymph is sensed in the lung and the signaling pathways that are involved in T/HS lymph-induced inflammation and injury are not well defined. Since T/HS lymph is sterile and does not contain measurable levels of endotoxin , but does contain biologically active non-microbial protein and lipid species ,  it appears that T/HS mesenteric lymph-induced lung injury is being directly mediated by gut-derived endogenous ligands.
Based on studies indicating that the germline-encoded pattern-recognition Toll-like receptors (TLR) recognize molecular motifs shared by large groups of microorganisms ,  as well as by endogenous ligands released from stressed and/or injured tissues , , we hypothesized that T/HS lymph was exerting its lung injurious effects via a TLR-dependent pathway. In this study, we chose to focus on TLR4 as the putative signaling system by which T/HS lymph induces lung injury for the following reasons. First, TLR4 activation has been implicated in the pathogenesis of organ injury, increased inflammatory gene expression, innate immune system priming and microvascular dysfunction after hemorrhagic shock as well as mechanical and thermal trauma , , , , , , ,  . Secondly, TLRs are constitutively expressed on cells of the innate immune system, including macrophages and neutrophils . Lastly, TLR4 has recently been documented to recognize endogenous danger-type or ‘alarmin’ factors thereby implicating TLR4 as a sensor of tissue injury as well as microbial invasion . Therefore, the goal of this study was to determine whether non-microbial factors in the intestinal mesenteric lymph after T/HS mediate gut-induced lung injury via TLR4 activation.
Materials and Methods
Adult male non-castrated Yorkshire minipigs, weighing 25 to 35 kg were used in this study (Animal Biotech Industries). Male 10–12 week old outbred cesarean-derived (CD-1) mice, TLR4-mutated (C3H/HeJ) mice and their wild-type (WT) littermates (C3H/HeOuJ), TLR-2 deficient (TLR2−/−) mice and their WT (C57BL/6J) littermates, and C57BL/6J-Ticam1Lps2/J and their WT (C57BL/6J) littermates were obtained from The Jackson Laboratory. Mice on the C57BL/6J background with a specific deletion of the MyD88 gene (MyD88−/− mice) were kindly provided by Dr. Samuel J. Leibovich (UMDNJ-New Jersey Medical School, Newark, N.J). The genotyping of Myd88−/−, Myd88+/− and Myd88+/+ WT (C57BL/6J) was described previously . C3H/HeJ mice have a point mutation in the TLR4 gene  and C57BL/6J-Ticam1Lps2/J mice have a Lps mutation in the Trif/or Ticam-1 gene that inhibit TLR4- and TRIF-dependent signaling respectively. From herein C3H/HeJ mice will be referred to as TLR4mut and C57BL/6J-Ticam1Lps2/J mice will be referred to as TRIFmut mice. The animals were maintained in accordance with the guidelines of the National Institutes of Health Guide for Care and Use of Laboratory Animals and the experiments were approved by the Animal Care and Use Committee of the University of Medicine and Dentistry of New Jersey.
Pig trauma-hemorrhagic shock and lymph collection model
Pigs underwent mesenteric lymph duct cannulation, followed by T/HS or trauma sham-shock (T/SS), as described previously . Briefly, male pigs were subjected to T/HS which involved a laparotomy with mesenteric lymph duct cannulation and withdrawal of blood in a staged fashion to a MAP of 40 mm Hg. The MAP was maintained until the base deficit reached -5 or the total shock period reached 3 hours. Animals were then resuscitated in a staged fashion to a MAP of 80–100 mm Hg. T/SS animals underwent laparotomy with lymph duct cannulation and without blood withdrawal. Mesenteric lymph was collected in sterile tubes for 30 minutes before shock, during shock and on an hourly basis during the postshock period. The collected lymph specimens were centrifuged at 500 g for 20 minutes at 4°C to remove all cellular components, tested for sterility on MacConkey and blood agar plates, aliquoted and stored at −80°C.
Male mice were anesthetized with pentobarbital (60–80 mg/kg IP) and under strict asepsis, a 2.5-cm midline laparotomy was performed. Isoflurane was given if needed to maintain the surgical level of anesthesia. Blood was withdrawn from the jugular vein until a mean arterial pressure (MAP) between 35–40 mmHg was obtained and maintained for 60 min. After 60 min, the mice were resuscitated with their shed blood. Animals subjected to T/SS underwent cannulation of the femoral artery and jugular vein followed by a laparotomy; however, no blood was withdrawn and the MAP was kept within normal limits. At 3 hr, after the end of shock or sham shock period, the Evans blue dye (EBD) lung permeability assay was performed.
Lymph infusion protocol
Mice underwent laparotomy as well as internal jugular vein cannulation. Laparotomy was closed after 15 minutes using two layers of 3–0 silk suture. WT, TLR4mut, Myd88−/−, TRIFmut and TLR2−/− mice were infused with T/HS and T/SS lymph from different pigs. The lymph used for each mouse was collected from an individual pig (pooled fractions collected during 1–3 hr post T/SS or T/HS). T/HS or T/SS pig lymph was infused via the jugular catheter at a rate of 10 µL/g body weight per hour for 3 hours. At the end of the 3 hour lymph infusion period, the mice were euthanized. The EBD lung permeability assay was performed and the lungs were harvested for myeloperoxidase (MPO) levels, and western blotting. The rationale for the lymph volume infused was based on the actual amount of lymph produced by the pigs (ml/kg/hr) subjected to T/HS or T/SS over the shock and resuscitation period which was approximately 28–30 mL/kg . Dose-response pilot studies of pig T/HS lymph documented that lung injury occurred with doses of pig lymph as low as 10 µL/g body weight (data not shown).
Evans blue dye (EBD) lung permeability assay
Mice were injected with EBD (30 mg/kg in 4% bovine serum albumin) via the internal jugular catheter 30 min prior to the end of the 3 hr reperfusion period following T/HS or T/SS or 30 min prior to the end of the 3 hr lymph infusion. After 10 minutes (allowing complete circulation of the dye), a blood sample (0.3 mL) was withdrawn from the femoral artery catheter, centrifuged and the resultant plasma was serially diluted to form a standard curve for EBD. After 20 min of EBD injection, a cardiopulmonectomy was performed after the mouse was fully exsanguinated. A bronchoalveolar lavage was then performed on the excised lungs. 1 ml of phosphate buffered saline (PBS) was instilled, rinsed in and out 3 times, repeated twice and the bronchoalveolar lavage fluid (BALF) was collected, centrifuged and the absorbance of EBD (620 nm) in the supernatant was measured spectrophotometrically. The percentage of EBD in the BALF was determined relative to the standard curve of EBD in the plasma.
Frozen lung tissue was homogenized and assayed for MPO activity as previously described . One unit of MPO activity represents the amount of enzyme that will reduce 1 mol/min of peroxide.
Western blot analysis
Whole cell extracts (WCE) prepared from lung tissue as previously described  were resolved by SDS-PAGE and detected by western blotting. The following antibodies were used: rabbit anti-iNOS (1∶3000; BD Biosciences) and rabbit anti-p42/p44 (1∶1000, Cell Signaling). As reported by others , total p42/p44 was used as our loading control since the expression of commonly used loading controls such as actin and tubulin have been shown to be altered in tissues of animals subjected to I/R injury . The western blots were developed with immoblin western (Millipore) and then analyzed by densitometry using an AlphaImager 3400 imaging system and AlphaEase FC software (Alpha Innotech).
Measurement of endotoxin in mesenteric lymph and endotoxin removal process
Lymph pooled from different pigs (n = 4–5) collected from the first 3 hr after T/HS were used in these assays. Endotoxin levels in the mesenteric lymph samples were measured using the limulus lysate assay pyrogent plus test kit according to the instructions provided by Lonza. Each sample was heat treated at 70°C for 5 minutes to inactivate potential inhibitory factors present in the mesenteric lymph and run in duplicate. The limit of detection was 0.06 endotoxin units (EU) /ml.
To further exclude the possibility that endotoxin is responsible for the lung injurious effects of the T/HS lymph samples, endotoxin removing columns containing immobilized polymyxin B that binds and removes endotoxin (Detoxi-Gel Endotoxin Removing Columns, Thermo Scientific Pierce) were used according to the manufacturer's instructions. To validate the efficacy of the columns, PBS and lymph specimens were spiked with 0.25 EU of endotoxin (E. coli 055:B5, Lonza) and endotoxin levels of spiked samples were measured using the limulus assay both prior to being placed on the column as well as after passage through the columns. Pre- and post-polymyxin-immobilized column eluted T/SS and T/HS lymph samples were infused in naïve C57BL6 mice for 3 hours as described above.
Isolation and Detection of Bacterial DNA in Mesenteric Porcine Lymph
DNA was isolated from 500 µL of porcine mesenteric lymph. The samples were centrifuged at 20,000xg for 10 min at 4°C and the pellet was lysed using MagNA Pure bacteria lysis buffer according to the manufacturer's instructions (Roche Diagnostics GmBH). DNA was isolated using the QIAamp DNA Mini Kit (Qiagen). DNA was also isolated from E. Coli as a positive control for bacterial DNA. PCR reactions for the amplication of the 16S ribosomal RNA wereperformed. The 16S rRNA primers were synthesized in the Molecular Resource Facility (New Jersey Medical School). The sequences of the 16S primer set included the forward primer (5′-TCCTACGGGAGGCAGCAGT-3′) and reverse primer (5′-GGACTACCAGGGTATCTAATCCTGTT-3′) as described in . The PCR reaction was performed with 5 ng of template and 50 pM of each primer using Platinum Taq polymerase (Invitrogen) in a thermal cycler 2720 (Applied Biosystems). DNA from E.coli served as a positive control and sterile water without template served as a negative control. Thermocycling conditions were: 95°C for 4 min, 25 cycles at 94°C for 30 s, 58°C for 30 s and 72°C for 30 s followed by a final extension at 72°C for 7 min. A 466 bp PCR 16S rRNA product was visualized on an 1.5% agarose gel stained with ethidium bromide.
Infusion of porcine mesenteric T/HS lymph induces lung injury in naïve CD1 mice
Our earlier studies demonstrated that the infusion of rat T/HS mesenteric lymph into naïve mice recreated the lung injury observed with actual T/HS . As reported in our in vitro studies using T/HS lymph from rats , pigs  and baboons  mesenteric lymph collected from male pigs subjected to T/HS contributed to priming of neutrophils and endothelial dysfunction. Thus, we first determined whether the infusion of sterile porcine mesenteric T/HS lymph into naïve CD1 mice caused similar lung injury as observed in CD1 mice subjected to T/HS. The rationale for using CD1 mice is that these mice have a robust immune system and do not have any of the potential confounding factors associated with inbreeding. As shown in Fig. 1, the infusion of pig T/HS lymph increased lung permeability to EBD by 2-fold as compared to CD-1 mice infused with T/SS pig lymph. Likewise, a 2.3-fold increase in percentage of EBD in the BALF was found in CD-1 mice subjected to actual T/HS as compared to their T/SS counterparts.
Lung permeability to Evans Blue dye (EBD) was measured in CD-1 mice subjected to actual T/HS or T/SS for 60 min and 3 hr reperfusion as well as CD1 mice infused with porcine T/SS or T/HS lymph for 3 hr. Data expressed as mean ± SE (n = 5–6 mice/ group).
TLR4 deficiency attenuates porcine T/HS lymph-induced lung injury
Having documented that T/HS lymph can recreate lung injury, we tested the hypothesis that non-microbial factors in the intestinal mesenteric lymph after T/HS mediate gut-induced lung injury via TLR4 activation. WT and TLR4mut mice were infused with porcine T/SS or T/HS lymph for 3 hours. Infusion of WT mice with T/HS lymph increased lung permeability by 2.5-fold as compared to mice infused with T/SS lymph (Fig. 2A). In contrast, there was no significant increase in lung permeability in TLR4 deficient mice infused with T/HS lymph relative to their T/SS lymph infused counterparts. Since the sequestration of polymorphonuclear neutrophils (PMNs) into the lung microvasculature has been characterized as a hallmark of ALI, MPO levels were measured in lung homogenates. As shown in Fig. 2B, a modest increase in MPO activity was evident in WT mice infused with T/HS lymph compared to WT mice infused with T/SS lymph. There was no difference between MPO levels in TLR4mut mice infused with T/HS or T/SS lymph. These findings suggest that factors in T/HS lymph mediate lung injury via TLR4 activation.
WT and TLR4mut mice were infused with porcine T/SS and T/HS lymph for 3 hr. A) Lung permeability to EBD was performed. Data expressed as mean ± SE (n = 3–12 mice/ group). B) MPO levels (U/g) were measured in lung homogenates. Data expressed as mean ± SE (n = 4–7 mice/group).
Since TLR4 mediates its downstream effects via two distinct signaling pathways, Myd88 and TRIF, we determined the effects of Myd88 and TRIF deficiency on T/HS lymph-induced lung injury. In both WT and Myd88−/− mice, infusion with T/HS lymph increased lung permeability by 3.3-fold (p<0.001) and 1.5-fold (p<0.01) respectively versus their T/SS lymph infused counterparts (Fig. 3A). The baseline level of lung permeability for the T/SS lymph infused Myd88−/− group was approximately 1.8-fold higher than the WT T/SS lymph infused group (p<0.05). In contrast, there was no significant increase in the percentage of EBD in the BALF of TRIFmut mice infused with T/HS lymph relative to their T/SS lymph infused group (Fig. 3B). Although both TRIF and Myd88 deficiency attenuated T/HS lymph-induced lung permeability (p<0.001 and p<0.05 respectively vs WT T/HS lymph infused mice), our results suggest that TRIF plays a more predominant role in T/HS lymph-induced lung permeability.
A) WT and Myd88−/− and B) WT and TRIFmut mice were infused with porcine T/SS and T/HS lymph for 3 hr and lung permeability to EBD was performed. Data expressed as mean ± SE (n = 5–8 mice/ group).
To further confirm the specificity of the TLR4 response in T/HS lymph-induced lung injury, we examined the effects of TLR2 deficiency on lung injury. In both the WT and TLR2 deficient mice, infusion with T/HS lymph increased lung permeability and pulmonary leukosequestration, as determined by EBD permeability (Fig. 4A) and MPO levels (Fig. 4B).
WT and TLR2−/− mice infused with porcine T/HS and T/SS lymph for 3 hr. A) Lung permeability to EBD was performed. Data expressed as mean ± SE (n = 4–6 mice/ group). B) MPO levels (U/g) were measured in lung homogenates. Data expressed as mean ± SE (n = 5–6 mice/group).
TLR4 deficiency blunts the pulmonary iNOS response associated with T/HS lymph-induced lung injury
Since increased iNOS activity has been implicated in T/HS-induced lung injury  and iNOS is a downstream target of TLR4 activation , we examined the effects of TLR4 deficiency on the pulmonary iNOS response after T/HS lymph infusion. Compared with WT mice infused with T/SS lymph, iNOS protein levels were 3-fold higher in WT mice infused with T/HS lymph (Fig. 5A and B). The induction of iNOS expression in lung homogenates from TLR4mut mice infused with T/HS lymph was markedly reduced as compared to their WT counterparts but remained elevated as compared to TLR4mut mice infused with T/SS lymph. Lungs from WT and TLR4mut mice infused with T/SS lymph expressed negligible levels of iNOS protein. These findings suggest an association between the T/HS lymph-induced pulmonary iNOS response and TLR4 activation.
A) and B) Western blot of iNOS in lung WCEs of WT and TLR4mut mice infused with porcine T/SS and T/HS lymph for 3 hr. B) Densitometry was performed to quantify iNOS and total p42/p44 MAPK expression. Data expressed as mean ± SE (n = 4–7 mice/group).
Endotoxin-and bacterial DNA- independent factors in T/HS lymph mediate lung injury
Although the lymph samples tested had no bacterial growth and were sterile (data not shown), we needed to rule out the possibility that the activation of TLR4 by T/HS lymph was due to gut-derived endotoxin contamination. Using a limulus lysate assay, neither T/SS nor T/HS lymph samples contained detectable levels of endotoxin by the limulus lysate assay (level of detection 0.06 EU/ml). To exclude the possibility that inhibitory factors in the lymph samples were interfering with the limulus assay, known quantities of endotoxin (0.06 EU, 0.125 EU or 0.25 EU) were added to T/SS and T/HS lymph samples and then assayed for endotoxin levels. The limulus assay tested positive for all the LPS concentrations tested, including 0.06 EU/ml of endotoxin, thereby validating that the limulus assay was able to detect endotoxin present in the lymph samples (data not shown). To further confirm that factors in T/HS lymph that mediate lung injury and activate TLR4 are devoid of endotoxin, pre- and post-polymyxin-immobilized column eluted T/SS and T/HS lymph samples were infused in naïve C57BL6 mice for 3 hours. As shown in Figure 6A, the magnitude of lung permeability was similar in mice infused with pre- and post-column T/HS lymph samples as compared to mice infused with T/SS lymph samples. Additionally, PCR analysis was performed to detect bacterial DNA in T/SS and T/HS lymph using a primer set which specifically detects 16S rDNA of the Domain Bacteria  and neither T/HS nor T/SS lymph tested positive for bacterial DNA contamination (Fig. 6B).
A) Pre- and post-polymyxin-immobilized column eluted T/SS and T/HS lymph samples were infused in naïve C57BL6 mice for 3 hours. Lung permeability to EBD. Data expressed as mean ± SE (n = 3–9 mice/ group). B) DNA isolated from lymph pooled from 3 pigs subjected to either T/HS or T/SS lymph was tested for bacterial DNA contaminants by PCR using 16S rDNA primers. DNA isolated from E.Coli served as our positive control was depicted as + and the negative control containing no DNA template was depicted as -. M denotes a 100 bp size marker.
Studies investigating the pathophysiology of trauma-induced SIRS, ARDS and multiple organ dysfunction syndrome (MODS) have been a major area of investigation. Work by us , , ,  and others , , , ,  have demonstrated that splanchnic ischemia leading to gut inflammation and loss of barrier function is the initial triggering event that turns the gut into the ‘motor of MODS’. Although, it was originally proposed that bacterial (endotoxin) translocation was the link between gut and distant organ injury, most recently, using rodent , porcine , and non-human primate trauma-hemorrhagic shock  as well as burn models , we have documented that acute ARDS and MODS observed after a major injury is due to immuno-inflammatory, tissue-injurious factors exiting the gut via the intestinal lymphatics. More support for the gut-lymph hypothesis of MODS stems from our recent study demonstrating that naïve rats and mice infused with mesenteric lymph from rats subjected to T/HS develop ALI . Our current study extends these results by showing that T/HS lymph-induced lung injury occurs via a TLR4-dependent pathway. The notion that TLR4, as well as other pattern recognition receptors, can respond to non-bacterial endogenous danger-type molecules as well as microbial products has expanded the role of pattern recognition receptors beyond that of sensors of microbial invasion . This concept also helps explain the many similarities between the host's septic response and the augmented inflammatory response to shock-trauma and other forms of major tissue injury.
Our results implicating TLR4, but not TLR2, as the proximal receptor for T/HS lymph-induced lung injury is novel in that we have implicated non-bacterial factors in mesenteric lymph that are able to directly activate TLR4. These findings are consistent with other studies showing that hemorrhagic shock-induced lung injury , and liver injury ,  is reduced in TLR4 deficient mice but not TLR2 deficient mice. Since other non-infectious models of injury, such as femur fracture ,  and thermal injury  also document that TLR4 deficient animals are protected, it appears that this immune-activating and tissue-injurious effect of TLR4 may be a common response to diverse shock and tissue-injury states. On the other hand, in certain conditions, such as hyperoxia-induced lung injury, TLR4 activation has been shown to be protective , . Thus, like many other systems, whether TLR4 activation is protective or deleterious may depend on the condition being studied as well as the timing and magnitude of its activation. Nonetheless, it is now clear that during tissue injury or proteolysis, extracellular matrix components undergo cleavage, exposing moieties that can act as ligands for TLRs and thereby initiate TLR-induced signal transduction. In this way, innate immune inflammatory responses may be activated without the presence of invading bacteria, but rather as a direct consequence of tissue injury .
Depending on the type, magnitude and duration of insult as well as the site of injury, the specificity of the TLR4 response is determined by the selective recruitment of its adapters, namely Myd88/TIRAP and TRIF/TRAM and their respective downstream signaling. Although there have been extensive studies delineating the Myd88-dependent and TRIF-dependent intracellular signaling pathways in the pulmonary innate immune response to microbial infections, our knowledge of their downstream signaling cascades in models of sterile/noninfectious ALI is very limited. For example, HMGB1/TLR4/Myd88 signaling activates NAD(P)H oxidase in neutrophils after hemorrhagic shock  whereas oxidized phospholipids mediate acid-induced ALI via TRIF signaling . Our study shows full and partial protection against T/HS lymph induced microvascular damage in TRIF and Myd88 deficient mice respectively. However, more work must be done to determine whether TRIF activation plays a more predominant role in the lung injurious effects of T/HS lymph than Myd88 activation.
Among the cells that bear TLR4 are those of the innate immune system, including macrophages, neutrophils and dendritic cells . Since TLR4 activated inflammatory cells manifest increased nitric oxide production , , we asked whether TLR4 activation in T/HS lymph-induced lung injury was associated with the induction of the iNOS response. Our findings demonstrated that lung iNOS expression was markedly but not completely reduced in TLR4mut mice infused with T/HS lymph as compared to iNOS expression in WT mice. Although this observation suggests that the induction of the iNOS response by T/HS lymph is modulated by both TLR4-dependent and TLR4-independent pathways, the concept that up-regulated iNOS activity is involved in the pathogenesis of T/HS-induced lung injury is consistent with studies showing that excessive nitric oxide production contributes to trauma-shock-induced organ injury , , , . Thus, taken together, our previous work showing iNOS deficient mice are resistant to T/HS lymph-induced lung injury  and the current study documenting that TLR4 deficient mice are also resistant to T/HS lymph-induced lung injury provide important clues as to how T/HS is converted from a hemodynamic insult into an inflammatory and lung tissue injurious disease.
Having demonstrated that T/HS lymph is capable of recreating shock-induced lung injury through TLR4- and iNOS-dependent pathways, the question arises about the composition of T/HS lymph and what factors within it are triggering this response. Although T/HS lymph is collected from animals subjected to T/HS, in which gut barrier function is impaired, we do not believe that intestinal bacteria and/or endotoxin are responsible for the biologic activity of T/HS lymph. This conclusion is based on our current results showing that the pig lymph samples were devoid of bacteria, bacterial DNA and did not contain measurable levels of endotoxin. Furthermore, the biologic lung-injurious activity was not abrogated after passage through an endotoxin removal column. The notion that injury-induced TLR4 activation is not due to endotoxin is also supported by studies showing that TLR4 deficient but not CD14 deficient mice were resistant to liver injury after major trauma . Additionally, if gut-derived bacterial products in T/HS lymph are involved in lymph-induced lung injury, gram-positive (peptidoglycan) as well as gram-negative (endotoxin) bacterial products would be present in T/HS lymph. Since peptidoglycan activates TLR2 and TLR2−/− mice are not resistant to T/HS lymph-induced lung injury, it is not likely that gram-positive microbial factors were responsible for the lung-injurious effects of T/HS lymph. The fact that TLR4 has the capacity to recognize and respond to endogenous ligands is also consistent with recent observations showing that T/HS contains several incompletely identified and biologically-active non-cytokine protein and lipid factors , ,  including a modified albumin species . Thus, we believe that T/HS lymph contains modified proteins and lipids which could either act as endogenous TLR4 ligands or promote the release of danger signals from a stressed or activated cell population within the lung. Regardless of whether factors in the mesenteric T/HS lymph are acting directly or indirectly in the induction of the TLR4 response, our findings demonstrate that these factors play a role in the sterile inflammatory process that occurs after trauma-hemorrhage.
One advantage of the T/HS lymph infusion model is that it provides a unique opportunity to study the effects of shock-induced, gut-derived factors in vivo without the confounding systemic effects of trauma and hemorrhagic shock. Since T/HS lymph is produced in vivo and the amounts of lymph required to cause lung injury were equal to or less than what would be produced during the actual trauma-shock period, speaks to the physiologic relevance of the model. Furthermore, since both rat and pig T/HS lymph cause lung injury in mice, a cross-species event, it appears that the factor or factors within T/HS lymph may be conserved among different species and likely target the same pattern recognition receptors. This cross species observation is also important since pigs are physiologically closer to humans than rodents. Taken together, these results lend further support for the Danger model of immune activation, which states that our immune system is not only activated by foreign pathogens, but also by our own endogenous ligands, whose patterns are evolutionarily preserved .
In summary, our studies suggest that TLR4 activation by T/HS lymph is necessary for T/HS lymph-induced lung injury and an augmented pulmonary iNOS response. More importantly, our results suggest that non-bacterial factors in T/HS mesenteric lymph mediate cross talk between tissue injury in the gut and subsequent gut-induced lung injury. Future studies elucidating the contribution of TLR4 and its downstream signaling in both immune and parenchymal cell populations that mediate T/HS lymph-induced lung injury as well as identifying endogenous TLR4 ligands after T/HS will provide more insight in the pathogenesis of shock-induced ALI and identify potential therapeutic targets.
Conceived and designed the experiments: DCR VP EAD RF. Performed the experiments: DCR VP DP DD RPB QL IC SUS BC KBK MR DZX. Analyzed the data: DCR VP DD QL IC SUS KBK MR EAD RF. Wrote the paper: DCR VP EAD RF.
- 1. Deitch EA (1992) Multiple organ failure. Pathophysiology and potential future therapy. Ann Surg 216: 117–134.EA Deitch1992Multiple organ failure. Pathophysiology and potential future therapy.Ann Surg216117134
- 2. Magnotti L, Upperman J, Xu D, Lu Q, Deitch E (1998) Gut-derived mesenteric lymph but not portal blood increases endothelial permeability and potentiates lung injury following hemorrhagic shock. Ann Surg 228: 518–527.L. MagnottiJ. UppermanD. XuQ. LuE. Deitch1998Gut-derived mesenteric lymph but not portal blood increases endothelial permeability and potentiates lung injury following hemorrhagic shock.Ann Surg228518527
- 3. Deitch E, Forsythe R, Anjaria D, Livingston D, Lu Q, et al. (2004) The role of lymph factors in lung injury, bone marrow suppression, and endothelial cell dysfunction in a primate model of trauma-hemorrhagic shock. Shock 22: 221–228.E. DeitchR. ForsytheD. AnjariaD. LivingstonQ. Lu2004The role of lymph factors in lung injury, bone marrow suppression, and endothelial cell dysfunction in a primate model of trauma-hemorrhagic shock.Shock22221228
- 4. Senthil M, Watkins A, Barlos D, Xu DZ, Lu Q, et al. (2007) Intravenous injection of trauma-hemorrhagic shock mesenteric lymph causes lung injury that is dependent upon activation of the inducible nitric oxide synthase pathway. Ann Surg 246: 822–830.M. SenthilA. WatkinsD. BarlosDZ XuQ. Lu2007Intravenous injection of trauma-hemorrhagic shock mesenteric lymph causes lung injury that is dependent upon activation of the inducible nitric oxide synthase pathway.Ann Surg246822830
- 5. Deitch E, Xu D, Lu Q (2006) Gut lymph hypothesis of early shock and trauma-induced multiple organ dysfunction syndrome. J Org Dysf 2: 70–79.E. DeitchD. XuQ. Lu2006Gut lymph hypothesis of early shock and trauma-induced multiple organ dysfunction syndrome.J Org Dysf27079
- 6. Adams CJ, Xu D, Lu Q, Deitch E (2001) Factors larger than 100 kd in post-hemorrhagic shock mesenteric lymph are toxic for endothelial cells. Surgery 129: 351–363.CJ AdamsD. XuQ. LuE. Deitch2001Factors larger than 100 kd in post-hemorrhagic shock mesenteric lymph are toxic for endothelial cells.Surgery129351363
- 7. Kaiser V, Sifri Z, Dikdan G, Berezina T, Zaets S, et al. (2005) Trauma-hemorrhagic shock mesenteric lymph from rat contains a modified form of albumin that is implicated in endothelial cell toxicity. Shock 23: 417–425.V. KaiserZ. SifriG. DikdanT. BerezinaS. Zaets2005Trauma-hemorrhagic shock mesenteric lymph from rat contains a modified form of albumin that is implicated in endothelial cell toxicity.Shock23417425
- 8. Gonzalez R, Moore E, Biffl W, Ciesla D, Silliman C (2000) The lipid fraction of post-hemorrhagic shock mesenteric lymph (PHSML) inhibits neutrophil apoptosis and enhances cytotoxic potential. Shock 14: 404–408.R. GonzalezE. MooreW. BifflD. CieslaC. Silliman2000The lipid fraction of post-hemorrhagic shock mesenteric lymph (PHSML) inhibits neutrophil apoptosis and enhances cytotoxic potential.Shock14404408
- 9. Janeway CA, Medzhitov R (2002) Innate Immune Recognition. Annual Review of Immunology 20: 197–216.CA JanewayR. Medzhitov2002Innate Immune Recognition.Annual Review of Immunology20197216
- 10. Kawai T, Akira S (2007) TLR signaling. Seminars in Immunology 19: 24–32.T. KawaiS. Akira2007TLR signaling.Seminars in Immunology192432
- 11. Beutler B (2004) Inferences, questions and possibilities in Toll-like receptor signalling. 430: 257–263.B. Beutler2004Inferences, questions and possibilities in Toll-like receptor signalling.430257263
- 12. Seong S-Y, Matzinger P (2004) Hydrophobicity: an ancient damage-associated molecular pattern that initiates innate immune responses. 4: 469–478.S-Y SeongP. Matzinger2004Hydrophobicity: an ancient damage-associated molecular pattern that initiates innate immune responses.4469478
- 13. Prince JM, Levy RM, Yang R, Mollen KP, Fink MP, et al. (2006) Toll-like receptor-4 signaling mediates hepatic injury and systemic inflammation in hemorrhagic shock. J Am Coll Surg 202: 407–417.JM PrinceRM LevyR. YangKP MollenMP Fink2006Toll-like receptor-4 signaling mediates hepatic injury and systemic inflammation in hemorrhagic shock.J Am Coll Surg202407417
- 14. Paterson HM, Murphy TJ, Purcell EJ, Shelley O, Kriynovich SJ, et al. (2003) Injury primes the innate immune system for enhanced Toll-Like receptor reactivity. J Immunol 171: 1473–1483.HM PatersonTJ MurphyEJ PurcellO. ShelleySJ Kriynovich2003Injury primes the innate immune system for enhanced Toll-Like receptor reactivity.J Immunol17114731483
- 15. Meng X, Ao L, Song Y, Raeburn CD, Fullerton DA, et al. (2005) Signaling for myocardial depression in hemorrhagic shock: roles of Toll-like receptor 4 and p55 TNF-(α) receptor. Am J Physiol Regul Integr Comp Physiol 288: R600–606.X. MengL. AoY. SongCD RaeburnDA Fullerton2005Signaling for myocardial depression in hemorrhagic shock: roles of Toll-like receptor 4 and p55 TNF-(α) receptor.Am J Physiol Regul Integr Comp Physiol288R600606
- 16. Barsness KA, Arcaroli J, Harken AH, Abraham E, Banerjee A, et al. (2004) Hemorrhage-induced acute lung injury is TLR-4 dependent. Am J Physiol Regul Integr Comp Physiol 287: R592–599.KA BarsnessJ. ArcaroliAH HarkenE. AbrahamA. Banerjee2004Hemorrhage-induced acute lung injury is TLR-4 dependent.Am J Physiol Regul Integr Comp Physiol287R592599
- 17. Rajnik M, Salkowski C, Thomas K, Li Y, Rollwagen F, et al. (2002) Induction of early inflammatory gene expression in a murine model of nonresuscitated, fixed-volume hemorrhage. Shock 17: 322–328.M. RajnikC. SalkowskiK. ThomasY. LiF. Rollwagen2002Induction of early inflammatory gene expression in a murine model of nonresuscitated, fixed-volume hemorrhage.Shock17322328
- 18. Levy RM, Prince JM, Yang R, Mollen KP, Liao H, et al. (2006) Systemic inflammation and remote organ damage following bilateral femur fracture requires Toll-like receptor 4. Am J Physiol Regul Integr Comp Physiol 291: R970–976.RM LevyJM PrinceR. YangKP MollenH. Liao2006Systemic inflammation and remote organ damage following bilateral femur fracture requires Toll-like receptor 4.Am J Physiol Regul Integr Comp Physiol291R970976
- 19. Powers KA, Szaszi K, Khadaroo RG, Tawadros PS, Marshall JC, et al. (2006) Oxidative stress generated by hemorrhagic shock recruits Toll-like receptor 4 to the plasma membrane in macrophages. J Exp Med 203: 1951–1961.KA PowersK. SzasziRG KhadarooPS TawadrosJC Marshall2006Oxidative stress generated by hemorrhagic shock recruits Toll-like receptor 4 to the plasma membrane in macrophages.J Exp Med20319511961
- 20. Breslin J, Wu M, Guo M, Reynoso R, Yuan S (2008) Toll-like receptor 4 contributes to microvascular inflammation and barrier dysfunction in thermal injury. Shock 29: 197–216.J. BreslinM. WuM. GuoR. ReynosoS. Yuan2008Toll-like receptor 4 contributes to microvascular inflammation and barrier dysfunction in thermal injury.Shock29197216
- 21. Mollen KP AR, Tsung A, Prince JM, Levy RM, Billiar TR (2006) Emerging paradigm: toll-like receptor 4-sentinel for the detection of tissue damage. Shock 26: 430–437.AR Mollen KPA. TsungJM PrinceRM LevyTR Billiar2006Emerging paradigm: toll-like receptor 4-sentinel for the detection of tissue damage.Shock26430437
- 22. Adachi O, Kawai T, Takeda K, Matsumoto M, Tsutsui H, et al. (1998) Targeted Disruption of the MyD88 Gene Results in Loss of IL-1- and IL-18-Mediated Function. Immunity 9: 143–150.O. AdachiT. KawaiK. TakedaM. MatsumotoH. Tsutsui1998Targeted Disruption of the MyD88 Gene Results in Loss of IL-1- and IL-18-Mediated Function.Immunity9143150
- 23. Poltorak A, He X, Smirnova I, Liu M-Y, Huffel CV, et al. (1998) Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: Mutations in Tlr4 gene. Science 282: 2085–2088.A. PoltorakX. HeI. SmirnovaM-Y LiuCV Huffel1998Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: Mutations in Tlr4 gene.Science28220852088
- 24. Senthil M, Brown M, Xu DZ, Lu Q, Feketeova E, et al. (2006) Gut-lymph hypothesis of systemic inflammatory response syndrome/multiple-organ dysfunction syndrome: validating studies in a porcine model. J Trauma 60: 958–967.M. SenthilM. BrownDZ XuQ. LuE. Feketeova2006Gut-lymph hypothesis of systemic inflammatory response syndrome/multiple-organ dysfunction syndrome: validating studies in a porcine model.J Trauma60958967
- 25. Senthil M, Brown M, Xu D, Lu Q, Feketeova E, et al. (2006) Gut-lymph hypothesis of systemic inflammatory response syndrome/multiple-organ dysfunction syndrome: validating studies in a porcine model. J Trauma 60: 958–965.M. SenthilM. BrownD. XuQ. LuE. Feketeova2006Gut-lymph hypothesis of systemic inflammatory response syndrome/multiple-organ dysfunction syndrome: validating studies in a porcine model.J Trauma60958965
- 26. Koury J, Deitch EA, Homma H, Abungu B, Gangurde P, et al. (2004) Persistent HIF-1α activation in gut ischemia/reperfusion injury: potential role of bacteria and lipopolysaccharide. Shock 22: 270–277.J. KouryEA DeitchH. HommaB. AbunguP. Gangurde2004Persistent HIF-1α activation in gut ischemia/reperfusion injury: potential role of bacteria and lipopolysaccharide.Shock22270277
- 27. Kawagoe T, Sato S, Jung A, Yamamoto M, Matsui K, et al. (2007) Essential role of IRAK-4 protein and its kinase activity in Toll-like receptor-mediated immune responses but not in TCR signaling. J Exp Med 204: 1013–1024.T. KawagoeS. SatoA. JungM. YamamotoK. Matsui2007Essential role of IRAK-4 protein and its kinase activity in Toll-like receptor-mediated immune responses but not in TCR signaling.J Exp Med20410131024
- 28. Sutton TA, Wilkinson J, Mang HE, Knipe NL, Plotkin Z, et al. (2008) p53 regulates renal expression of HIF-1α and pVHL under physiological conditions and after ischemia-reperfusion injury. Am J Physiol Renal Physiol 295: F1666–1677.TA SuttonJ. WilkinsonHE MangNL KnipeZ. Plotkin2008p53 regulates renal expression of HIF-1α and pVHL under physiological conditions and after ischemia-reperfusion injury.Am J Physiol Renal Physiol295F16661677
- 29. Nadkarni MA, Martin FE, Jaques NA, Hunter N (2002) Determination of bacterial load by real-time PCR using a broad-range (universal) probe and primers set. Microbiology 148: 257–266.MA NadkarniFE MartinNA JaquesN. Hunter2002Determination of bacterial load by real-time PCR using a broad-range (universal) probe and primers set.Microbiology148257266
- 30. Deitch E (1990) Intestinal permeability is increased in burn patients shortly after injury. Surgery 107: 411–416.E. Deitch1990Intestinal permeability is increased in burn patients shortly after injury.Surgery107411416
- 31. Doig C, Sutherland L, Sandham J, Fick G, Verhoef M, et al. (1998) Increased intestinal permeability is associated with the development of multiple organ dysfunctino syndrome in the critically ill ICU patients. Am J Respir Crit Care Med 158: 444–451.C. DoigL. SutherlandJ. SandhamG. FickM. Verhoef1998Increased intestinal permeability is associated with the development of multiple organ dysfunctino syndrome in the critically ill ICU patients.Am J Respir Crit Care Med158444451
- 32. Ivatury R, Simon R, Islam S, Fueg A, Rohman M, et al. (1996) A prospective randomized study of end points of resuscitation after major trauma: global oxygen transport indices versus organ-specific gastric mucosal pH. J Am Coll Surg 183: 145–154.R. IvaturyR. SimonS. IslamA. FuegM. Rohman1996A prospective randomized study of end points of resuscitation after major trauma: global oxygen transport indices versus organ-specific gastric mucosal pH.J Am Coll Surg183145154
- 33. Border J, Hassett J, LaDuca J, Seibel R, Steinberg S, et al. (1987) The gut origin septic states in blunt multiple trauma (ISS = 40) in the ICU. Ann Surg 206: 427–448.J. BorderJ. HassettJ. LaDucaR. SeibelS. Steinberg1987The gut origin septic states in blunt multiple trauma (ISS = 40) in the ICU.Ann Surg206427448
- 34. Clark J, Coopersmith C (2007) Intestinal crosstalk: a new paradigm for understanding the gut as the “motor” of critical illness. Shock 28: 384–393.J. ClarkC. Coopersmith2007Intestinal crosstalk: a new paradigm for understanding the gut as the “motor” of critical illness.Shock28384393
- 35. Leaphart CL, Tepas JJ III (2007) The gut is a motor of organ system dysfunction. Surgery 141: 563–569.CL LeaphartJJ Tepas III2007The gut is a motor of organ system dysfunction.Surgery141563569
- 36. Sambol JT, Xu DZ, Adams CA, Magnotti LJ, Deitch EA (2000) Mesenteric lymph duct ligation provides long term protection against hemorrhagic shock-induced lung injury. Shock 14: 416–420.JT SambolDZ XuCA AdamsLJ MagnottiEA Deitch2000Mesenteric lymph duct ligation provides long term protection against hemorrhagic shock-induced lung injury.Shock14416420
- 37. Magnotti LJ, Xu D-Z, Lu Q, Deitch EA (1999) Gut-derived mesenteric lymph: A link between burn and lung injury. Arch Surg 134: 1333–1341.LJ MagnottiD-Z XuQ. LuEA Deitch1999Gut-derived mesenteric lymph: A link between burn and lung injury.Arch Surg13413331341
- 38. Kaczorowski DJ, Mollen KP, Edmonds R, Billiar TR (2008) Early events in the recognition of danger signals after tissue injury. J Leukoc Biol 83: 546–552.DJ KaczorowskiKP MollenR. EdmondsTR Billiar2008Early events in the recognition of danger signals after tissue injury.J Leukoc Biol83546552
- 39. Mollen KP, Levy RM, Prince JM, Hoffman RA, Scott MJ, et al. (2008) Systemic inflammation and end organ damage following trauma involves functional TLR4 signaling in both bone marrow-derived cells and parenchymal cells. J Leukoc Biol 83: 80–88.KP MollenRM LevyJM PrinceRA HoffmanMJ Scott2008Systemic inflammation and end organ damage following trauma involves functional TLR4 signaling in both bone marrow-derived cells and parenchymal cells.J Leukoc Biol838088
- 40. Zhang X, Shan P, Qureshi S, Homer R, Medzhitov R, et al. (2005) Cutting edge: TLR4 deficiency confers susceptibility to lethal oxidant lung injury. J Immunol 175: 4834–4838.X. ZhangP. ShanS. QureshiR. HomerR. Medzhitov2005Cutting edge: TLR4 deficiency confers susceptibility to lethal oxidant lung injury.J Immunol17548344838
- 41. Jiang D, Liang J, Fan J, Yu S, Chen S, et al. (2005) Regulation of lung injury and repair by Toll-like receptors and hyaluronan. Nat Med 11: 1173–1179.D. JiangJ. LiangJ. FanS. YuS. Chen2005Regulation of lung injury and repair by Toll-like receptors and hyaluronan.Nat Med1111731179
- 42. Fan J, Li Y, Levy RM, Fan JJ, Hackam DJ, et al. (2007) Hemorrhagic shock induces NAD(P)H oxidase activation in neutrophils: role of HMGB1-TLR4 signaling. J Immunol 178: 6573–6580.J. FanY. LiRM LevyJJ FanDJ Hackam2007Hemorrhagic shock induces NAD(P)H oxidase activation in neutrophils: role of HMGB1-TLR4 signaling.J Immunol17865736580
- 43. Imai Y, Kuba K, Neely GG, Yaghubian-Malhami R, Perkmann T, et al. (2008) Identification of oxidative stress and toll-like receptor 4 signaling as a key pathway of acute lung injury. Cell 133: 235–249.Y. ImaiK. KubaGG NeelyR. Yaghubian-MalhamiT. Perkmann2008Identification of oxidative stress and toll-like receptor 4 signaling as a key pathway of acute lung injury.Cell133235249
- 44. Eswarappa SM, Basu N, Joy O, Chakravortty D (2008) Folimycin (concanamycin A) inhibits LPS-induced nitric oxide production and reduces surface localization of TLR4 in murine macrophages. Innate Immunity 14: 13–24.SM EswarappaN. BasuO. JoyD. Chakravortty2008Folimycin (concanamycin A) inhibits LPS-induced nitric oxide production and reduces surface localization of TLR4 in murine macrophages.Innate Immunity141324
- 45. Sethi G, Sodhi A (2004) In vitro activation of murine peritoneal macrophages by ultraviolet B radiation: upregulation of CD18, production of NO, proinflammatory cytokines and a signal transduction pathway. Molecular Immunology 40: 1315–1323.G. SethiA. Sodhi2004In vitro activation of murine peritoneal macrophages by ultraviolet B radiation: upregulation of CD18, production of NO, proinflammatory cytokines and a signal transduction pathway.Molecular Immunology4013151323
- 46. Thiemermann C, Szabo C, Mitchell JA, Vane JR (1993) Vascular hyporeactivity to vasoconstrictor agents and hemodynamic decompensation in hemorrhagic shock is mediated by nitric oxide. Proc Natl Acad Sci USA 90: 267–271.C. ThiemermannC. SzaboJA MitchellJR Vane1993Vascular hyporeactivity to vasoconstrictor agents and hemodynamic decompensation in hemorrhagic shock is mediated by nitric oxide.Proc Natl Acad Sci USA90267271
- 47. Hierholzer C, Harbrecht B, Menezes JM, Kane J, MacMicking J, et al. (1998) Essential role of induced nitric oxide in the initiation of the inflammatory response after hemorrhagic shock. J Exp Med 187: 917–928.C. HierholzerB. HarbrechtJM MenezesJ. KaneJ. MacMicking1998Essential role of induced nitric oxide in the initiation of the inflammatory response after hemorrhagic shock.J Exp Med187917928
- 48. Hierholzer C, Kalff JC, Billiar TR, Bauer AJ, Tweardy DJ, et al. (2004) Induced nitric oxide promotes intestinal inflammation following hemorrhagic shock. Am J Physiol Gastrointest Liver Physiol 286: G225–233.C. HierholzerJC KalffTR BilliarAJ BauerDJ Tweardy2004Induced nitric oxide promotes intestinal inflammation following hemorrhagic shock.Am J Physiol Gastrointest Liver Physiol286G225233
- 49. Suzuki Y, Deitch EA, Mishima S, Lu Q, Xu D (2000) Inducible nitric oxide synthase gene knockout mice have increased resistance to gut injury and bacterial translocation after an intestinal ischemia-reperfusion injury. Crit Care Med 28: 3692–3696.Y. SuzukiEA DeitchS. MishimaQ. LuD. Xu2000Inducible nitric oxide synthase gene knockout mice have increased resistance to gut injury and bacterial translocation after an intestinal ischemia-reperfusion injury.Crit Care Med2836923696
- 50. Matzinger P (2002) The danger model: A renewed sense of self. Science 296: 301–305.P. Matzinger2002The danger model: A renewed sense of self.Science296301305