Molecular Mechanism Of HMGB1-induced Proinflammatory Cytokines Expression In Kupffer Cells From Severely Burned Rats

1. IntroductionDespite advances in burn prevention, treatment, and rehabilitation over the lastdecades, sepsis and subsequent multiple organ dysfunction syndrome (MODS)which were originated from systemic inflammatory response remain to be the mostfrequently reported causes of death in the severely burned patients. Being centralrole in metabolism and host defense mechanisms, the liver is thought to be a majororgan responsible for the initiation of multiple organ failure in patients with majorburns. Proinflammatory cytokines such as tumor necrosis factor (TNF)-α andinterleukin (IL)-1β have been demonstrated to be the two most important cytokinesin the early phase of burns and play an important role in producing hepatocelluardysfunction.Locating in the liver sinusoids, Kupffer cells (KCs) comprise the largestpopulation of tissue-fixed macrophages in the human organism. Studies havedocumented that Kupffer cell played a key role in producing the systemic changes inhost immune responses, namely through the up-regulation and release ofproinflammatory cytokines. Our previous study has demonstrated that Kupffer cellwas a significant source of TNF-α and IL-1β release during the early stage of severeburns, and thereby contributed to the liver injury following thermal injury.High-mobility group box1(HMGB1), a highly conserved non-histonechromosomal protein, was originally identified as a DNA-binding protein involvedin maintenance of nucleosome structure and regulation of gene transcription.Recently, HMGB1was found to act as a potent proinflammatory cytokine and a“late” mediator that participated in the development of systemic inflammatory response. Addition of purified recombinant HMGB1to human monocyte culturessignificantly stimulated the release of cytokines including TNF-α, IL-1α, IL-1β, IL-6,and IL-8. HMGB1can be either passively released from necrotic or damaged cells,or can be actively secreted by monocytes and macrophages under stressfulconditions. Recent data demonstrated that levels of HMGB1increased significantlyin plasma after extensive burn injury, which was associated with the development ofsepsis and fatal outcome of major burns. However, the role of HMGB1in the releaseof proinflammatory cytokines by KCs following thermal injury has not been fullyelucidated so far.Biological effects of extracellular HMGB1could be mediated by the activation ofsignaling pathways coupled to toll-like receptor (TLR)2, TLR4, TLR9, and thereceptor for advanced glycation end products (RAGE). RAGE has beendemonstrated to play only a minor role in macrophages activation by HMGB1,whereas signaling through TLRs, especially TLR2and TLR4, appears to be of muchgreater importance in the ability of HMGB1to generate inflammatory responses.TLR4-deficient mice were found to be less prone to liver injury following burntrauma and the expressions of TLR2and TLR4increased in rat macrophages afterthermal injury. Moreover, TLR2and TLR4could trigger intracellular signalingcascades in macrophages involving activation of p38mitogen-activated proteinkinase (MAPK), c-Jun NH(2)-terminal kinase (JNK), and nuclear factor-κB (NF-κB).Such signaling activation consequently leaded to the release of proinflammatorycytokines in monocytes including TNF-α and IL-1β. Augmented TLR2and TLR4reactivities in macrophages have been demonstrated to contribute to the developmentof heightened systemic inflammation after burn injury. However, there was littleinformation regarding the potential receptors and signaling mechanisms of HMGB1underlying immunological function of Kupffer cell after burn injury.2ObjectivesSince its crucial roles in pathophysiological process of inflammation, HMGB1might regulate the proinflammatory cytokines synthesis in KCs after burn injury. Therefore, There are three main purposes of this study:①to explore the effect offunction of HMGB1on the KCs after severe burn;②to explore the receptormechanism of after severe burns; whether there several receptors coexistence?Which play a predominant role in this process., RAGE receptor or TLR?;③toexplore receptor signal transduction mechanisms of KCs by stimulate with HMGB1after severe burns. Whether key molecules in “MAPK and NF-κB” signalingpathway involved in the activation of KCs regulation by HMGB1after burns?HMGB1could induce KCs to produce proinflammatory cytokines (TNF-α andIL-1β) through TLRs-dependent signaling after severe burn injury.3Materials and methodsPart ⅠThe role of HMGB1-induced proinflammatory cytokines expression in Kupffercells from severely burned ratsHealthy adult male Sprague-Dawley rats weighing200-250g were usedthroughout the study. All experimental manipulations were undertaken in accordancewith the National Institutes of Health’s Guide for the Care and Use of LaboratoryAnimals, with the approval of the animal experimental ethics committee of AnhuiMedical University, China. All animals were acclimatized to their environment for1week. Rats were fed a standard animal diet with food and tap water ad libitum for theentire study period.The animals were anesthetized with pentobarbital (30mg/kg) intraperitoneally, thedorsal and lateral surfaces were shaved, and rats were secured in a constructedtemplate device. The rats divided into2groups according to the random number table:A group(n=16) is burn trauma, B group(n=16) is sham burn. The surface area of theskin exposed through the template device was immersed in98°C water for12s on thedorsal surface. With the use of this technique, a full-thickness dermal burn comprising30%total body surface area (TBSA) were obtained. After immersion, the rats were immediately dried to avoid additional injury, and each animal was resuscitated with anintraperitoneal injection of lactated Ringer’s solution (30ml/kg). Sham burn rats weresubjected to an identical preparation except that they were immersed in roomtemperature water and no fluid resuscitation was administered. Twenty-four hoursafter burn injury (or sham burn), the rats were exsanguinated by cardiac puncture. Theviability of cells was greater than95%as determined by trypan blue exclusion. Cellswere cultured at a concentration of1×106cell/well in RPMI-1640mediumsupplemented with10%fetal calf serum,10mMN-(2-hydroxyethyl)piperazine-N’-(2-ethanesulfonic acid)(HEPES) and50μg/mlgentamycin at37°C in a humidified,5%CO2atmosphere.To examine the cell-specific effects of HMGB1on KCs secretion of TNF-α andIL-1β, KCs were harvested from rats24h after either burn trauma (n=16) or shamburn (n=16). Then, cells were stimulated with HMGB1(Sigma-Aldrich, St. Louis,MO, USA,0,50,100, or200ng/ml) for48h and supernatant was collected tomeasure KCs secretion of TNF-α and IL-1β. The TNF-α and IL-1β levels weremeasured in KCs supernatant using a “sandwich” enzyme-linked immunosorbentassay (ELISA) with TNF-α and IL-1β ELISA kits for rats (BioSource InternationalInc., Camarillo, CA, USA), according to the manufacturer’s instructions. All sampleswere run in duplicate and averaged.Part ⅡRole of receptor for advanced glycation end products in HMGB1-inducedproinflammatory cytokines expression in Kupffer cells from severely burnedratsHealthy adult male Sprague-Dawley rats weighing200-250g were usedthroughout the study. All experimental manipulations were undertaken in accordancewith the National Institutes of Health’s Guide for the Care and Use of LaboratoryAnimals, with the approval of the animal experimental ethics committee of AnhuiMedical University, China. All animals were acclimatized to their environment for1week. Rats were fed a standard animal diet with food and tap water ad libitum for theentire study period. The animals were anesthetized with pentobarbital (30mg/kg) intraperitoneally, thedorsal and lateral surfaces were shaved, and rats were secured in a constructedtemplate device. The surface area of the skin exposed through the template device wasimmersed in98°C water for12s on the dorsal surface. With the use of this technique,a full-thickness dermal burn comprising30%total body surface area (TBSA) wereobtained. After immersion, the rats were immediately dried to avoid additional injury,and each animal was resuscitated with an intraperitoneal injection of lactated Ringer’ssolution (30ml/kg). Sham burn rats were subjected to an identical preparation exceptthat they were immersed in room temperature water and no fluid resuscitation wasadministered. Twenty-four hours after burn injury, the rats were exsanguinated bycardiac puncture. The animals were anesthetized with pentobarbital (30mg/kg)intraperitoneally, the dorsal and lateral surfaces were shaved, and rats were secured ina constructed template device. The surface area of the skin exposed through thetemplate device was immersed in98°C water for12s on the dorsal surface. With theuse of this technique, a full-thickness dermal burn comprising30%total body surfacearea (TBSA) were obtained. After immersion, the rats were immediately dried toavoid additional injury, and each animal was resuscitated with an intraperitonealinjection of lactated Ringer’s solution (30ml/kg). Twenty-four hours after burn injury,the rats were exsanguinated by cardiac puncture. The viability of cells was greaterthan95%as determined by trypan blue exclusion. Cells were cultured at aconcentration of1×10~6cell/well in RPMI-1640medium supplemented with10%fetalcalf serum,10mM N-(2-hydroxyethyl)piperazine-N’-(2-ethanesulfonic acid)(HEPES)and50μg/ml gentamycin at37°C in a humidified,5%CO2atmosphere.KCs divided into4groups according to the random number table:1) negativecontrol (n=8);2) HMGB1only (n=8);3) HMGB1+anti-RAGE antibody (n=8); and4)HMGB1+rrRAGE/Fc(Chimeric)(n=8). The expression of RAGE were measuredby Western blot analysis. The levels of TNF-α and IL-1β proteins in culturesupernatant were measured by enzyme-linked immunosorbent assay (ELISA).Northern blot analysis was used to detect the expression of TNF-α and IL-1βmRNAs. Part ⅢThe role of toll-like receptor on high-mobiility group box-1(HMGB1)induction rat Kupffer cells (KCs) secretion of proinflammatory cytokinessuchfollowing burn trauma.Healthy adult male Sprague-Dawley rats weighing200-250g were usedthroughout the study. All experimental manipulations were undertaken in accordancewith the National Institutes of Health’s Guide for the Care and Use of LaboratoryAnimals, with the approval of the animal experimental ethics committee of AnhuiMedical University, China. All animals were acclimatized to their environment for1week. Rats were fed a standard animal diet with food and tap water ad libitum for theentire study period.The animals were anesthetized with pentobarbital (30mg/kg) intraperitoneally, thedorsal and lateral surfaces were shaved, and rats were secured in a constructedtemplate device. The surface area of the skin exposed through the template device wasimmersed in98°C water for12s on the dorsal surface. With the use of this technique,a full-thickness dermal burn comprising30%total body surface area (TBSA) wereobtained. After immersion, the rats were immediately dried to avoid additional injury,and each animal was resuscitated with an intraperitoneal injection of lactated Ringer’ssolution (30ml/kg). Twenty-four hours after burn injury, the rats were exsanguinatedby cardiac puncture. The viability of cells was greater than95%as determined bytrypan blue exclusion. Cells were cultured at a concentration of1×106cell/well inRPMI-1640medium supplemented with10%fetal calf serum,10mMN-(2-hydroxyethyl)piperazine-N’-(2-ethanesulfonic acid)(HEPES) and50μg/mlgentamycin at37°C in a humidified,5%CO2atmosphere.KCs were divided into4groups:1) negative control, KCs were left untreated andcultured for48h;2) HMGB1stimulation only, KCs were stimulated with100ng/mlHMGB1for48h;3) HMGB1+anti-TLR2antibody (InvivoGen, San Diego, CA,USA), KCs were pre-incubated (2h at37°C) with anti-TLR2monoclonal antibody(20μg/ml) and stimulated with100ng/ml HMGB1for48h; and4) HMGB1+anti-TLR4antibody (InvivoGen, San Diego, CA, USA), KCs were pre-incubated (2h at37°C) with anti-TLR4monoclonal antibody (20μg/ml) and stimulated with100ng/ml HMGB1for48h. After culture and stimulation, supernatant was removed foranalysis of TNF-α and IL-1β and cells were collected..The levels of Tumor necrosisfactor (TNF)-α and interleukin (IL)-1β in culture supernatant were measured byenzyme-linked immunosorbent assay. Northern blot analysis was performed to detectthe expressions of TNF-α and IL-1β mRNAs. The activities of p38MAPK and JNK(by Western blot analysis) as well as NF-κB (by EMSA) in KCs were also examined.Results:Part ⅠKCs from sham and burn rats were treated with various concentrations of HMGB1(50-200ng/ml) for48h. The TNF-α and IL-1β proteins in supernatant wereexamined by ELISA. As a result, KCs could produce constitutively few TNF-α andIL-1β under normal condition. HMGB1stimulated KCs to secrete TNF-α and IL-1βin a dose-dependent manner. Furthermore, HMGB1induced significantly greateramounts of TNF-α and IL-1β secretion by KCs from burn rats compared to thosefrom sham animals at a dose as low as50ng/ml. This discrepancy was even greaterwhen100ng/ml or higher concentrations of HMGB1was used.PartⅡThe expression of RAGE in rats KC were significantly increased after100ng/mlHMGB1stimulated24h (1.036±0.101vs0.191±0.024, t=-23.158, P=0.000). Thelevels of TNF-α and IL-1β in supernatant and expressions of TNF-α and IL-1βmRNAs in rats KC of HMGB1group were both significantly higher than negativecontrol. However, these were no significant difference among HMGB1group,HMGB1+anti-RAGE antibody group, and HMGB1+rrRAGE/Fc group (P>0.05).Part IIIPre-incubation of KCs with anti-TLR2or anti-TLR4antibody both significantlyattenuated HMGB1-induced TNF-α and IL-1β releases (p<0.01). Anti-TLR4antibody has more dramatic inhibitory effect than anti-TLR2antibody onHMGB1-induced TNF-α production (73%vs.53%, p<0.05) and the level of TNF-α in the supernatant of KCs of HMGB1+anti-TLR4group was similar to that found inunstimulated KCs from burn animals (p>0.05). Meanwhile, TLR2blockade causedmore dramatic inhibition of HMGB1-induced IL-1β production compared to thatcaused by TLR4blockade (80%vs.50%, p<0.05) and the level of IL-1β in thesupernatant of KCs of HMGB1+anti-TLR2group was similar to that found in thecells absent of HMGB1(p>0.05).In addition, Northern blot analysis of Kupffer cell RNA revealed that TNF-α andIL-1β mRNA levels were significantly augmented by48-h-stimulation with HMGB1in comparison with unstimulated controls as corrected by GAPDH transcripts(Figure3). Moreover, pre-incubation of anti-TLR2and anti-TLR4antibodies bothsignificantly inhibited HMGB1-induced expressions of TNF-α and IL-1β mRNAs inKCs of burn rats (p<0.01).phosphorylation of p38was significantly increased in HMGB1-stimulated KCscompared to control cells treated with medium only (Figure4). Anti-TLR2andanti-TLR4antibodies both significantly decreased phosphorylation of p38inHMGB1-stimulated cells, and reductions up to75%and71%respectively. Therewas no significant difference in total p38expression between groups. Furthermore,HMGB1also markedly upregulated JNK activity in KCs when compared tounstimulated KCs. Anti-TLR2and anti-TLR4antibodies performed the similarinhibitory effects on HMGB1-induced JNK activation as p38activation in KCs ofburn rats. Total JNK expression was not changed over time.Forty-eight hours after100ng/ml HMGB1stimulation, NF-κB was significantly activated in KCs incomparison with the unstimulated control (p<0.01). This up-regulation was greatlyattenuated in KCs pre-incubated with the anti-TLR2or anti-TLR4antibody (p<0.01).Conclusions;1: HMGB1in vitro upregulated expressions of TNF-α and IL-1β of KCs in adose-dependent manner, and HMGB1promoted KCs from burn rats to producesignificantly more TNF-α and IL-1β proteins than those from sham animals. 2: HMGB1can induced expression of proinflammatory cytokines TNF-α and IL-1βin rat KC after severe burn injury, but the RAGE receptor did not play apredominant role in this process.3: Anti-TLR2antibody has more dramatic inhibitory effect than anti-TLR4antibody on HMGB1-induced IL-1β production, suggesting HMGB1-induced IL-1βproduction mainly involving TLR2. These findings highlight the differential usagesof TLR2and TLR4in HMGB1signaling in cytokine release of KCs following burninjury. HMGB1induces proinflammatory cytokines production of KCs after severburn injury, and this process might be largely dependent on TLRs-dependentMAPKs/NF-κB signal pathway.