Receptor for advanced glycation end-products (RAGE) and TLR4 play an important role in the inflammatory response against High-mobility group box 1 protein (HMGB1), a late proinflammatory cytokine and a damage-associated molecular pattern. As cell surface receptors, both RAGE and TLR4 are constantly trafficking between the cytoplasm and plasma membrane. However, whether TLR4 is related to the intracellular transport of RAGE in HMGB1-induced inflammation remains unknown. In this study, we demonstrated that HMGB1 not only increased RAGE expression in both the cytoplasm and plasma membrane but also upregulated the expression of TLR4 in the plasma membrane. Knocking out of RAGE led to decreased MAPK activation, TLR4 cellular membrane expression, and corresponding inflammatory cytokine generation. Meanwhile, inhibiting MAPK activation also decreased TLR4 surface expression. These results indicated that HMGB1 may bind to cell surface RAGE receptors on the cell surface, leading to MAPK activation, thus promoting TLR4 translocation on the cell surface, but does not regulate its transcription and translation. In contrast, TLR4 can increase the transcription and translation of RAGE, which translocates to the cell surface and is able to bind to more HMGB1. The cell surface receptors TLR4 and RAGE bind to HMGB1, leading to the transcription and secretion of inflammatory cytokines. Finally, we also observed these results in the mice pseudofracture model, which is closely related to HMGB1-induced inflammatory response. All these results demonstrated that the interplay between RAGE and TLR4 are critical for HMGB1-induced inflammatory response.

HMGB1 is a widely expressed protein, which participates in the pathogenesis of infectious and noninfectious inflammatory diseases (1, 2). It is secreted by cells as a critical inflammatory mediator in the circumstance of sepsis, shock, autoimmune diseases, and chronic inflammation (35). Once released in the extracellular milieu, HMGB1 acts as a damage-associated molecular pattern (6), which binds to receptors such as RAGE and TLRs (7, 8), leading to massive generation of inflammatory cytokines, chemokines, and corresponding receptors (9). Extracellular HMGB1, especially as a preformed intracellular molecule, is transferred to the cell surface during cell activation (10). Researchers have reported extracellular HMGB1 to be involved in at least 14 different receptor systems. So far, only RAGE and TLR4 have been extensively investigated and have been identified as specific HMGB1 receptors in a large number of studies (11, 12).

As a member of the Ig gene superfamily, RAGE is an inflammatory type I transmembrane receptor and responsible for HMGB1 induced intracellular signaling of inflammation, chemotaxis, and NF-κB activation (13). TLR4 is a key receptor for innate immunity activation and cytokine release, which is indispensable for HMGB1-induced macrophages activation. The amount of cell surface TLR4 is determined by the transport of TLR4 from the Golgi to the cell membrane and the internalization of receptor on the cell surface to the endoplasmic chamber (14). Although the RAGE-mediated signaling pathway is well studied, there is insufficient investigation on the relationship between RAGE and TLR4 trafficking (7, 8, 15). The membrane trafficking of RAGE and TLR4 in HMGB1-induced inflammation and the relationship between RAGE and TLR4 still remains unclear.

In this study, we found that RAGE and TLR4 are both required for HMGB1-induced inflammation in bone marrow–derived macrophages (BMDMs) and in the pseudofracture (PF) mouse model (16). Knocking out of RAGE inhibited the activation of MAPKs. In addition, in BMDMs and in the PF model, RAGE and TLR4 cell surface expression were increased in HMGB1-induced inflammation, and decreased expression of RAGE and TLR4 attenuated HMGB1-induced inflammation. Furthermore, through RAGE or TLR4 knockout, we demonstrated that RAGE affected the cell surface expression of TLR4 by promoting its trafficking, rather than regulating its transcription and translation. TLR4 regulated RAGE cell surface expression by promoting its membrane trafficking, transcription, and translation. Likewise, in the mice PF model, we also observed that RAGE and TLR4 can interact with each other to regulate the surface expression of TLR4 and RAGE on macrophages. Overall, this article presents a novel interplay between the cell surface expression of RAGE and TLR4 in response to HMGB1-induced inflammation.

C57BL/6 male (wild-type; WT) mice (6–8 wk old) were purchased from Southern Medical University Animal Center (Guangzhou, China). TLR4, RAGE knockout male mice (6–8 wk old) were obtained from Animal Core Facility of Nanjing Medical University (Nanjing, China). All the animal experimental protocols were reviewed and conformed to the committees of Southern Medical University Animal Center for care and committees of Guangdong Medical University. Mice with a mean body weight of 25 g were used.

PF, a validated murine model of sterile musculoskeletal trauma, allows for evaluation of posttraumatic early and late immune response (1618). Mice (male, 8 wk old) were euthanized, and two femurs and two tibias were harvested and crushed using a sterile mortar in 2 ml sterile PBS under sterile conditions to create the “bone solution.” Mice were anesthetized with ketamine (50 mg/kg body weight) and xylazine (5 mg/kg body weight). Each posterior thigh musculature, midpoint along the femur, were clamped by an 18-cm hemostat for 30 s to induce a soft tissue injury, followed by injection of 0.15 ml bone solution into the injured posterior muscles of each thigh. Sham animal models had each posterior thigh musculature with 0.15 ml of sterile saline. After 12 h, peritoneal lavage fluid (PLF) was collected by 5 ml PBS after PF or Sham.

RAGE Ab (sc-365154) was purchased from Santa Cruz Biotechnology. Abs specific for TLR4 (14358), Phospho-p38 (9215), p-38 (8690), Phospho-Erk1/2 (4370), Erk1/2 (4695), Phospho-JNK (4668), JNK (9252), and GAPDH (5174) were obtained from Cell Signaling Technology. Recombinant mouse HMGB1 (764006) and neutralizing anti-HMGB1 (clone 3E8, 651402) were obtained from BioLegend. Mouse IgG Isotype Control (02-6502) was purchased from Thermo Fisher Scientific.

BMDM isolation was performed as previously described (19). In brief, bone marrow from male mice (6–8 wk old) were flushed out and cultured in BMDM culture medium (DMEM containing 10% FBS complemented with 50 μg/ml penicillin/streptomycin and 10 ηg/ml M-CSF).

For measuring cell surface expression of TLR4 and RAGE, BMDMs or PLF cells were stained with Ab to the following markers: 7-AAD (559925, 1:100; BD), F4/80-PE (563899, 1:200; BD), TLR4-BV650 (740615, 1:200; BD), RAGE-AF647 (FAB11795R, 1:200; R&D), TLR4-PE (12-9041-80, 1:200; eBioscience) or RAGE-AF488 (FAB11795G, 1:200; R&D) Ab for 30 min. Cells were washed three times with PBS, after suspension in PBS, and then cells were analyzed by flow cytometry (BD Biosciences). Data were analyzed using FlowJo (Tree Star).

BMDMs were lysed by RIPA buffer. The lysates were centrifuged and the supernatants were determined by BCA kit (Thermo Fisher Scientific). The proteins were separated by 10% SDS-PAGE and transferred to PVDF membranes (MilliporeSigma). After blocking with 5% skimmed milk for 1 h at room temperature, the membrane was incubated with primary Abs overnight at 4°C. After washing membranes three times with TBST, the membranes were incubated at room temperature for 1 h, followed by visualization using ECL reagent (MillporeSigma). The densitometric analysis of protein was performed by Image Pro Plus.

Total RNA was isolated from BMDMs using TRIzol RNA Isolation Reagents (Life Technologies, Pittsburgh, PA). RNA extraction and quantitative PCR was performed as previously described (19). The gene-specific primers are listed as below: RAGE forward primer: 5′-GGGACTGTGACCTTGACCTG-3′; reverse primer: 5′-ATAGGTGCCCTCATCCTCGT-3′. TLR4 forward primer: 5′-ATGGCATGGCTTACACCACC-3′; reverse primer: 5′-GAGGCCAATTTTGTCTCCACA-3′. TNF-α forward primer: 5′-CCCTCACACTCAGATCATCTTCT-3′; reverse primer: 5′-GCTACGACGTGGGCTACAG-3′. IL-1β forward primer: 5′-GCAACTGTTCCTGAACTCAACT-3′; reverse primer: 5′-ATCTTTTGGGGTCCGTCAACT-3′. IL-6 forward primer: 5′-TAGTCCTTCCTACCCCAATTTCC-3′; reverse primer: 5′-TTGGTCCTTAGCCACTCCTTC-3′.

The cell culture supernatants or PLF were harvested after indication treatment. ELISA kits for TNF-α, IL-1β (eBioscience), and IL-6 (R&D) were used to measure the cytokine levels in supernatants or PLF according to the manufacturer’s instructions.

Data analyses were done using SPSS19.0. Differences between two groups were compared using Student t test. One-way ANOVA was performed for multiple groups. Results are presented as mean ± SD. The p values of <0.05 was considered statistically significant.

To determine the role of HMGB1 in the inflammatory response, we tested the expression of inflammatory cytokines in BMDMs after incubating with 4 μg/ml HMGB1 for various periods of time. We found that production of TNF-α, IL-1β, and IL-6 in BMDMs were significantly increased in a time-dependent manner at the protein (Fig. 1A–C) and mRNA (Fig. 1D–F) levels after HMGB1 stimulation. Moreover, the protein phosphorylation levels of ERK1/2, P38, and JNK were also elevated in a time-dependent manner (Fig. 1G). These data suggest that HMGB1 can induce an inflammatory response in BMDMs.

FIGURE 1.

HMGB1 induces the inflammatory response and regulates RAGE and TLR4 expression in BMDMs. BMDMs were treated with 4 μg/ml HMGB1 for different time periods. (AC) The secretion of TNF-α, IL-1β, and IL-6 was determined by enzyme-linked immunoassay at the indicated times. (DF, I, and J) The mRNA levels of TNF-α, IL-1β, IL-6, TLR4, and RAGE were assayed by quantitative PCR. (G and H) The protein levels of MAPK signaling (total ERK, phospho-ERK, total P38, phospho-P38, total JNK, and phospho-JNK), TLR4, and RAGE were measured by Western blot. Total ERK, P38, JNK, and GAPDH were used as quantitative controls. (K and L) TLR4 or RAGE surface expression was analyzed by FACS analysis at indicated time. Data shown represent three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 1.

HMGB1 induces the inflammatory response and regulates RAGE and TLR4 expression in BMDMs. BMDMs were treated with 4 μg/ml HMGB1 for different time periods. (AC) The secretion of TNF-α, IL-1β, and IL-6 was determined by enzyme-linked immunoassay at the indicated times. (DF, I, and J) The mRNA levels of TNF-α, IL-1β, IL-6, TLR4, and RAGE were assayed by quantitative PCR. (G and H) The protein levels of MAPK signaling (total ERK, phospho-ERK, total P38, phospho-P38, total JNK, and phospho-JNK), TLR4, and RAGE were measured by Western blot. Total ERK, P38, JNK, and GAPDH were used as quantitative controls. (K and L) TLR4 or RAGE surface expression was analyzed by FACS analysis at indicated time. Data shown represent three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

Extracellular HMGB1, as a damage-associated molecular pattern, can trigger signaling through several receptors, including TLR4 and RAGE. To determine the effect of RAGE and TLR4 on the HMGB1-induced inflammatory response, we tested the expression of RAGE and TLR4 at the protein, mRNA, and cell surface levels. We found that at both the protein and mRNA levels, RAGE expression was increased, beginning at 12 h, but was not significantly increased at 6 h (Fig. 1H, 1I). However, cell surface expression of RAGE was significantly increased in a time-dependent manner, beginning at 6 h (Fig. 1K). In addition, TLR4 expression showed no significant change at both the protein and mRNA levels (Fig. 1H, 1J), but the surface expression was significantly increased beginning at 6 h (Fig. 1L). These results suggest the potential involvement of RAGE and TLR4 in the regulation of HMGB1-induced inflammatory response.

To identify whether RAGE is involved in the regulation of HMGB1-induced inflammation, we compared the production of proinflammatory mediators in BMDMs from WT mice and RAGE knockout mice (RAGE−/−). We found that the production of proinflammatory mediators (TNF-α, IL-1β, and IL-6) in BMDMs upon HMGB1 stimulation were significantly inhibited at the protein (Fig. 2A–C) and mRNA (Fig. 2D–F) levels in RAGE knockout BMDMs, suggesting that RAGE is essential for HMGB1-induced inflammatory response.

FIGURE 2.

RAGE or TLR4 knockout reduces the production of proinflammatory mediators induced by HMGB1 in BMDMs. WT, RAGE knockout (RAGE−/−), or TLR4 knockout (TLR4−/−) BMDMs were treated with HMGB1 (4 μg/ml) for 4 h to measure the secretion and mRNA expression of TNF-α, IL-1β, and IL-6 by enzyme-linked immunoassay (AC and GI) and quantitative PCR (DF and JL). Data shown represent three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 2.

RAGE or TLR4 knockout reduces the production of proinflammatory mediators induced by HMGB1 in BMDMs. WT, RAGE knockout (RAGE−/−), or TLR4 knockout (TLR4−/−) BMDMs were treated with HMGB1 (4 μg/ml) for 4 h to measure the secretion and mRNA expression of TNF-α, IL-1β, and IL-6 by enzyme-linked immunoassay (AC and GI) and quantitative PCR (DF and JL). Data shown represent three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

Next, we investigated whether TLR4 is essential for HMGB1-induced inflammatory response. We compared the production of proinflammatory mediators in BMDMs from WT mice and TLR4 knockout mice (TLR4−/−). The proinflammatory mediators (TNF-α, IL-1β, and IL-6) expression levels at the protein (Fig. 2G–I) and mRNA (Fig. 2J–L) levels in TLR4 knockout BMDMs were significantly decreased compared with WT BMDMs, which suggests that TLR4 plays an important role in the production of proinflammatory mediators induced by HMGB1.

To confirm the effect of RAGE on HMGB1-induced MAPK activation, we tested the phosphorylation levels of ERK1/2, P38 and JNK. We found that the phosphorylation levels of ERK1/2, P38 and JNK were all significantly decreased by RAGE knockout, compared with WT BMDMs, after stimulation with HMGB1 (4 μg/ml) for 0.5 h (Fig. 3A). These data indicate that RAGE is essential for MAPK activation by HMGB1.

FIGURE 3.

RAGE regulates TLR4 cell surface expression. (A) WT or RAGE knockout (RAGE−/−) BMDMs were stimulated with HMGB1 (4 μg/ml) for 0.5 h. Phosphorylated and total ERK1/2, p-38, and JNK were detected by Western blot. WT or RAGE knockout (RAGE−/−) BMDMs were stimulated with HMGB1 (4 μg/ml) for indicated time periods. (BD) TLR4 cell surface expression was determined by FACS analysis after treatment with or without HMGB1 at the indicated time. (E and F) Expression of TLR4 was tested by Western blot (E) and quantitative PCR (F). (GJ) WT BMDMs were pretreat with 2 μg/ml BFA (G), 20 μM PD98059 (H), 10 μM SB203580 (I), or 10 μM SP600125 (J) for 1 h, followed by HMGB1 (4 μg/ml) for 24 h. TLR4 cell surface expression was tested by FACS analysis. Data shown represent three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3.

RAGE regulates TLR4 cell surface expression. (A) WT or RAGE knockout (RAGE−/−) BMDMs were stimulated with HMGB1 (4 μg/ml) for 0.5 h. Phosphorylated and total ERK1/2, p-38, and JNK were detected by Western blot. WT or RAGE knockout (RAGE−/−) BMDMs were stimulated with HMGB1 (4 μg/ml) for indicated time periods. (BD) TLR4 cell surface expression was determined by FACS analysis after treatment with or without HMGB1 at the indicated time. (E and F) Expression of TLR4 was tested by Western blot (E) and quantitative PCR (F). (GJ) WT BMDMs were pretreat with 2 μg/ml BFA (G), 20 μM PD98059 (H), 10 μM SB203580 (I), or 10 μM SP600125 (J) for 1 h, followed by HMGB1 (4 μg/ml) for 24 h. TLR4 cell surface expression was tested by FACS analysis. Data shown represent three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

To analyze the relationship between RAGE and TLR4, we examined the expression of TLR4 at the protein, mRNA, and cell surface levels after RAGE knockout. Compared with WT BMDMs, RAGE knockout BMDMs decreased TLR4 cell surface expression at 6 and 24 h upon HMGB1 stimulation (Fig. 3B–D) but was NS at both the protein and mRNA levels at 6 and 24 h (Fig. 3E, 3F), which suggests that RAGE regulates TLR4 cell surface expression but does not regulate total protein and mRNA levels of TLR4 expression in HMGB1-induced inflammation. To confirm the effect of TLR4 trafficking to cell surface, we examined the TLR4 surface expression after brefeldin A (BFA) treatment, which inhibits the intracellular trafficking from endoplasmic reticulum to the Golgi. We found that BFA significantly decreased TLR4 cell surface expression (Fig. 3G). To further support the role of RAGE-regulated MAPK signaling in regulation of TLR4 cell surface expression, WT BMDMs were treated with PD98059, SB203580 or SP600125, which inhibit ERK1/2, P38, and JNK, respectively. We found that TLR4 cell surface expression was significantly inhibited by PD98059 (Fig. 3H), SB203580 (Fig. 3I) and SP600125 (Fig. 3J), indicating that MAPK signaling regulate TLR4 cell surface expression. Taken together, these results indicate that RAGE activates MAPK signaling, which affects TLR4 cell surface expression by promoting its trafficking rather than regulating its transcription and translation.

To further confirm the relationship between RAGE and TLR4, we tested the expression of RAGE at the protein, mRNA, and cell surface levels after TLR4 knockout. After treated with HMGB1 at different time points, the expression of RAGE at total protein and mRNA levels did not differ between WT and TLR4−/− group at 0 and 6 h, but the expression of RAGE in TLR4 knockout group was significantly decreased at 24 h (Fig. 4A, 4B) compared with the WT group, indicating that TLR4 does not affect total protein and mRNA levels of RAGE at 6 h but does affect the levels at 24 h, whereas TLR4 affects the RAGE cell surface expression at both 6 and 24 h (Fig. 4C–E). These results suggest that TLR4 influences RAGE cell surface expression by promoting its trafficking rather than regulating its transcription and translation at early phase; however, during the late phase, TLR4 could influence RAGE expression by promoting its trafficking, transcription, and translation.

FIGURE 4.

TLR4 regulates RAGE cell surface expression. BMDMs from WT or TLR4 knockout (TLR4−/−) were treated with or with HMGB1 (4 μg/ml) at different time points. Expression of TLR4 was tested by Western blot (A) and quantitative PCR (B). (CE) TLR4 cell surface expression was determined by FACS analysis. *p < 0.05, **p < 0.01.

FIGURE 4.

TLR4 regulates RAGE cell surface expression. BMDMs from WT or TLR4 knockout (TLR4−/−) were treated with or with HMGB1 (4 μg/ml) at different time points. Expression of TLR4 was tested by Western blot (A) and quantitative PCR (B). (CE) TLR4 cell surface expression was determined by FACS analysis. *p < 0.05, **p < 0.01.

Close modal

To determine the inflammatory response upon RAGE and TLR4 cell surface expression in macrophages within sterile infection, mice were subjected to PF, a validated model of long-bone fracture (1618). This model recapitulated both the sterile infection and proinflammatory response (16, 17). PLF and peritoneal macrophages were collected at 12 h after PF. As shown in Fig. 5A–E, PF significantly increased not only the concentration of proinflammatory mediators (TNF-α, IL-1β, and IL-6) in PLF but also the RAGE and TLR4 cell surface expression in peritoneal macrophages. To determine whether HMGB1 contributed to the effect of PF on proinflammatory mediators and the cell surface expression of RAGE and TLR4, HMGB1 neutralizing Ab (0.1 mg/kg) was directly injected into tail vein following PF. As shown in Fig. 5A–E, HMGB1-neutralizing Ab significantly decreased PF-induced production of proinflammatory mediators (TNF-α, IL-1β, and IL-6) and the cell surface expression of RAGE and TLR4. These data indicate that HMGB1 is required for the effect of PF on proinflammatory response and the peritoneal macrophages surface expression of RAGE and TLR4.

FIGURE 5.

RAGE and TLR4 regulates PF-induced inflammatory response. (AE) WT mice were injected with 0.1 mg/kg mouse IgG isotype control (Isotype control) or 0.1 mg/kg HMGB1 neutralizing Ab (anti-HMGB1) by tail vein, followed by Sham or PF for 12 h. Levels of TNF-α (A), IL-1β (B), and IL-6 (C) in PLF were measured by enzyme-linked immunoassay. RAGE (D) or TLR4 (E) cell surface expression in peritoneal macrophages (gated by 7AAD, F4/80+) were measured by FACS analysis. (FI) WT or RAGE knockout (RAGE−/−) mice were subjected to Sham or PF for 12 h. Levels of TNF-α (F), IL-1β (G), and IL-6 (H) in PLF were measured by enzyme-linked immunoassay. (I) TLR4 cell surface expression in peritoneal macrophages was measured by FACS analysis. (JM) WT or TLR4 knockout (TLR4−/−) mice were subjected to Sham or PF for 12 h. Levels of TNF-α (J), IL-1β (K), and IL-6 (L) in PLF were measured by enzyme-linked immunoassay. (M) RAGE cell surface expression in peritoneal macrophages was measured by FACS analysis. Data shown from groups of four individually analyzed mice and are representative of three independent experiments. *p < 0.05, **p < 0.01.

FIGURE 5.

RAGE and TLR4 regulates PF-induced inflammatory response. (AE) WT mice were injected with 0.1 mg/kg mouse IgG isotype control (Isotype control) or 0.1 mg/kg HMGB1 neutralizing Ab (anti-HMGB1) by tail vein, followed by Sham or PF for 12 h. Levels of TNF-α (A), IL-1β (B), and IL-6 (C) in PLF were measured by enzyme-linked immunoassay. RAGE (D) or TLR4 (E) cell surface expression in peritoneal macrophages (gated by 7AAD, F4/80+) were measured by FACS analysis. (FI) WT or RAGE knockout (RAGE−/−) mice were subjected to Sham or PF for 12 h. Levels of TNF-α (F), IL-1β (G), and IL-6 (H) in PLF were measured by enzyme-linked immunoassay. (I) TLR4 cell surface expression in peritoneal macrophages was measured by FACS analysis. (JM) WT or TLR4 knockout (TLR4−/−) mice were subjected to Sham or PF for 12 h. Levels of TNF-α (J), IL-1β (K), and IL-6 (L) in PLF were measured by enzyme-linked immunoassay. (M) RAGE cell surface expression in peritoneal macrophages was measured by FACS analysis. Data shown from groups of four individually analyzed mice and are representative of three independent experiments. *p < 0.05, **p < 0.01.

Close modal

To further confirm the effect and interrelationship of RAGE and TLR4 on PF, we compared WT, RAGE knockout, and TLR4 knockout mice after PF. We found that RAGE knockout mice prevented PF-induced production of proinflammatory mediators (TNF-α, IL-1β, and IL-6) in PFL (Fig. 5F–H). Moreover, TLR4 cell surface expression were significantly decreased in peritoneal macrophages of RAGE knockout mice after PF (Fig. 5I), indicating that RAGE is essential for PF-induced proinflammatory response and TLR4 cell surface expression in macrophages. Similarly, we found that TLR4 knockout mice decreased the production of proinflammatory mediators (TNF-α, IL-1β, and IL-6) in PFL (Fig. 5J–L) and RAGE cell surface expression in peritoneal macrophages (Fig. 5M) after PF, indicating that TLR4 is required for PF-induced proinflammatory response and RAGE cell surface expression in macrophages. Collectively, these results suggest that RAGE and TLR4 are essential for the mice PF model, which is closely related to HMGB1-induced inflammatory response, and that these receptors interact with each other to regulate the macrophages surface expression of TLR4 and RAGE.

FIGURE 6.

RAGE and TLR4 regulate HMGB1-induced inflammation with a complex network of feedback interactions. HMGB1 binds to RAGE receptors on the cell surface, leading to MAPK activation, which promotes TLR4 translocation on the cell surface but does not regulate its transcription and translocation. Here it can bind to HMGB1, leading to transcription and translation of RAGE, which translocates to cell surface to bind more HMGB1. TLR4 and RAGE bind to HMGB1 on the cell surface, which in turn leads to transcription and secretion of inflammatory cytokines, such as TNF-α, IL-1β, and IL-6.

FIGURE 6.

RAGE and TLR4 regulate HMGB1-induced inflammation with a complex network of feedback interactions. HMGB1 binds to RAGE receptors on the cell surface, leading to MAPK activation, which promotes TLR4 translocation on the cell surface but does not regulate its transcription and translocation. Here it can bind to HMGB1, leading to transcription and translation of RAGE, which translocates to cell surface to bind more HMGB1. TLR4 and RAGE bind to HMGB1 on the cell surface, which in turn leads to transcription and secretion of inflammatory cytokines, such as TNF-α, IL-1β, and IL-6.

Close modal

As one of the most important chromatin proteins, HMGB1 usually exists in the cell nucleus. In the case of cell stress and inflammation, it can be upregulated and mobilized into the cytoplasm and actively secreted outside the cell (5, 20). Several receptors have been verified in HMGB1 signaling, including RAGE, TLR2, and TLR4 (68, 21). Some studies suggested that RAGE and TLR4 functionally interact with each other to coordinate and regulate immune and inflammatory responses (6, 22). However, the influence and mechanism of RAGE and TLR4 interaction in HMGB1-induced inflammation still remained unclear. In this study, we explored the role of cell surface expression of RAGE and TLR4 in HMGB1-induced inflammation both in vitro and in vivo. Moreover, we expounded on the effect of functional interaction between RAGE and TLR4 on regulating the membrane trafficking of RAGE and TLR4 (Fig. 6).

As a polyligand protein, RAGE is found in mammals and belongs to the Ig superfamily and cell adhesion molecule family (23). RAGE is also a receptor for HMGB1 in immune cells, such as macrophages, dendritic cells, and neutrophils (6, 15). The effect of HMGB1 on dendritic cells relies on the interaction between HMGB1 and RAGE and its downstream activation of MAPKs and NF-κB (24). In this study, we found that RAGE regulated the activation of MAPKs at 30 min after HMGB1 treatment. The binding ability of HMGB1 depends on its redox state. The three redox states of HMGB1 vary with structure, half-life, and activity (25). The redox state determines the stability of HMGB1 and its binding ability with RAGE. Reduce state is the form that combines RAGE and activates the inflammation pathway (25). Our results showed that RAGE knockout decreased the production of inflammatory mediators and the activation of MAPKs at 30 min after HMGB1 treatment. PF is a validated murine model of sterile musculoskeletal trauma that allows for evaluation of posttraumatic immune responses (1618). Our in vivo studies showed that the inflammatory response of PF was induced by HMGB1 and increased cell surface expression of RAGE in macrophages. In addition, RAGE knockout mice attenuated the HMGB1-mediated inflammatory response in the pseudofrature. Therefore, the interaction between RAGE and HMGB1 activates MAPK signaling and, in turn, promotes the inflammatory response.

HMGB1 can also bind to TLR4, which belongs to the transmembrane pattern recognition receptor family, and plays a key role in the regulation of immune and inflammatory response (26). Recently, HMGB1 was found to bind to MD2 in the TLR4R complex, indicating the possible role of TLR4 in HMGB1-induced inflammation (27). Both our in vitro and in vivo results also revealed that TLR4 is essential for HMGB1-induced inflammation. HMGB1 can bind to both TLR4 and RAGE, and both RAGE and TLR4 are essential for HMGB1-induced inflammation. However, the mechanism of interaction between RAGE and TLR4 was unclear. Hence, we investigated the interrelationship between RAGE and TLR4 in HMGB1-induced inflammation.

Cell surface expression of receptors is determined by the balance between receptor trafficking from the Golgi apparatus to the cell membrane and internalization of the cell surface receptor into endosomal compartments. Chaperones and transport proteins regulate TLR4 trafficking and activation (14). The amount of receptor present on the cell surface regulates different responses, including inflammation. In LPS-induced endotoxemia, cell surface expression of the TLR4R is increased in BMDMs (28). Interestingly, we found that the cell surface expression of TLR4 was also increased in HMGB1-induced inflammation in BMDMs (in vitro) or in peritoneal macrophages (in vivo). TLR4 cell surface expression can be inhibited through interrupting HMGB1/RAGE-induced MAPK activation. We also found that RAGE knockout led to decreased membrane TLR4 expression but did not affect the total protein and mRNA levels. These results suggested that RAGE regulated HMGB1-induced inflammation by promoting the trafficking of TLR4 to the cell surface rather than affecting TLR4 transcription or translation. TLR4 is synthesized in the intracellular matrix and folded in the endoplasmic reticulum (29). Previous studies have reported that the TLR4 membrane expression is regulated by protein associated with TLR 4 (PRAT4A) (29, 30). A recent study showed that TMED7 is necessary for the transporting of TLR4 from the Golgi to the cell surface (31, 32). Wang et al. (28) found that small GTPase rab10 colocalizes with TLR4 in the Golgi and enhances the signal activity of TLR4 by increasing the transportation rate of TLR4 from Golgi to the cell surface. In our study, we found that RAGE and its downstream molecular MAPKs are also required for promoting trafficking of TLR4 to the cell surface from the Golgi.

Although we found that RAGE regulates TLR4 cell surface expression in response to HMGB1 treatment, the effect of TLR4 on the surface expression of RAGE remains unclear. RAGE is expressed in many cell types, especially as an intracellular molecule transported to the cell surface during cell activation (10). In human microvascular endothelial cells (31) and umbilical vein endothelial cells (32), HMGB1 can increase the expression of RAGE on the cell surface. Similarly, in our results, we found that RAGE surface expression was increased in BMDMs after HMGB1 stimulation. SRC-mediated CAV-1 phosphorylation was reported to be necessary for RAGE translocation to the plasma membrane, and inhibition of SRC activation led to RAGE targeting to lysosomes (33). Fiuza et al. (33) found that HMGB1 promotes the phosphorylation of ERK1/2 and two stress-activated MAPK pathways (JNK and P38); the release of inflammatory factors (TNF-α), chemokines (IL-8 and MCP-1), and adhesion molecules (ICAM-1 and VCAM-1); and increased RAGE membrane surface expression in endothelial cells. Interestingly, in HMGB1-stimulated BMDMs, we also showed that HMGB1–RAGE could activate MAPK signaling, which is necessary for RAGE surface expression. Furthermore, we found that TLR4 knockout led to a decrease in the membrane expression, total protein, and mRNA levels of RAGE, indicating that TLR4 may influence RAGE cell surface expression by promoting its trafficking in the early phase and influence RAGE expression by promoting its trafficking, transcription, and translation in the late phase. TLR4 activation increased RAGE expression both at total protein and mRNA levels at the late phase, creating a positive feedback loop for RAGE trafficking to the cell surface after ligand binding. However, it is unclear what mediates TLR4 in the regulation of transcription and translocation of RAGE. The exact mechanism of the interaction between RAGE and TLR4 needs to be further explored.

In conclusion, our in vitro and in vivo studies demonstrated that the expression of RAGE and TLR4 on cell surface are critical important in HMGB1-induced inflammation. More importantly, our study provides a novel discovery in HMGB1-induced inflammation, in that HMGB1 may bind to cell surface RAGE receptor, leading to MAPK activation, thus promoting TLR4 translocation to the cell surface but not regulating its transcription and translation. Furthermore, surface TLR4 can also bind to HMGB1, leading to transcription and translation of RAGE, thus promoting RAGE translocates to the cell surface to interact with additional HMGB1. The interplay of RAGE and TLR4, and subsequent binding with HMGB1, leads to the transcription and secretion of inflammatory cytokines. These findings provide further insight into the collaborative role of RAGE and TLR4 in HMGB1-induced inflammation.

We thank Xuegang Sun, Zaisheng Qin, Yuanliang Liu, and Zhiyun Zeng (Southern Medical University) for their technical expertise.

This work was supported by National Natural Science Foundation of China 81671957 and 81873951 (to J.T.), key projects of Guangdong Natural Science Foundation 2018B030311038 (to J.T.), Science and Technology Planning Project of Guangdong Province 2016A020215212 (to J.T.), Guangdong Medical Research Fund A2017483 (to Y.N.), and Guangzhou Science and Technology Project 201707010244 (to Y.N.).

Abbreviations used in this article:

BFA

brefeldin A

BMDM

bone marrow–derived macrophage

PF

pseudofracture

PLF

peritoneal lavage fluid

WT

wild-type.

1
Venereau
,
E.
,
F.
De Leo
,
R.
Mezzapelle
,
G.
Careccia
,
G.
Musco
,
M. E.
Bianchi
.
2016
.
HMGB1 as biomarker and drug target.
Pharmacol. Res.
111
:
534
544
.
2
Deng
,
M.
,
Y.
Tang
,
W.
Li
,
X.
Wang
,
R.
Zhang
,
X.
Zhang
,
X.
Zhao
,
J.
Liu
,
C.
Tang
,
Z.
Liu
, et al
.
2018
.
The endotoxin delivery protein HMGB1 mediates caspase-11-dependent lethality in sepsis.
Immunity
49
:
740
753.e7
.
3
Lei
,
C.
,
S.
Lin
,
C.
Zhang
,
W.
Tao
,
W.
Dong
,
Z.
Hao
,
M.
Liu
,
B.
Wu
.
2013
.
Effects of high-mobility group box1 on cerebral angiogenesis and neurogenesis after intracerebral hemorrhage.
Neuroscience
229
:
12
19
.
4
Yang
,
H.
,
D. J.
Antoine
,
U.
Andersson
,
K. J.
Tracey
.
2013
.
The many faces of HMGB1: molecular structure-functional activity in inflammation, apoptosis, and chemotaxis.
J. Leukoc. Biol.
93
:
865
873
.
5
Tang
,
D.
,
R.
Kang
,
H. J.
Zeh
III
,
M. T.
Lotze
.
2011
.
High-mobility group box 1, oxidative stress, and disease.
Antioxid. Redox Signal.
14
:
1315
1335
.
6
Andersson
,
U.
,
K. J.
Tracey
.
2011
.
HMGB1 is a therapeutic target for sterile inflammation and infection.
Annu. Rev. Immunol.
29
:
139
162
.
7
Tafani
,
M.
,
L.
Schito
,
L.
Pellegrini
,
L.
Villanova
,
G.
Marfe
,
T.
Anwar
,
R.
Rosa
,
M.
Indelicato
,
M.
Fini
,
B.
Pucci
,
M. A.
Russo
.
2011
.
Hypoxia-increased RAGE and P2X7R expression regulates tumor cell invasion through phosphorylation of Erk1/2 and Akt and nuclear translocation of NF-{kappa}B.
Carcinogenesis
32
:
1167
1175
.
8
Cohen
,
M. J.
,
M.
Carles
,
K.
Brohi
,
C. S.
Calfee
,
P.
Rahn
,
M. S.
Call
,
B. B.
Chesebro
,
M. A.
West
,
J. F.
Pittet
.
2010
.
Early release of soluble receptor for advanced glycation endproducts after severe trauma in humans.
J. Trauma
68
:
1273
1278
.
9
Lotze
,
M. T.
,
K. J.
Tracey
.
2005
.
High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal.
Nat. Rev. Immunol.
5
:
331
342
.
10
Ramasamy
,
R.
,
S. F.
Yan
,
A. M.
Schmidt
.
2011
.
Receptor for AGE (RAGE): signaling mechanisms in the pathogenesis of diabetes and its complications.
Ann. N. Y. Acad. Sci.
1243
:
88
102
.
11
Andersson
,
U.
,
H.
Yang
,
H.
Harris
.
2018
.
High-mobility group box 1 protein (HMGB1) operates as an alarmin outside as well as inside cells.
Semin. Immunol.
38
:
40
48
.
12
He
,
Q.
,
H.
You
,
X. M.
Li
,
T. H.
Liu
,
P.
Wang
,
B. E.
Wang
.
2012
.
HMGB1 promotes the synthesis of pro-IL-1β and pro-IL-18 by activation of p38 MAPK and NF-κB through receptors for advanced glycation end-products in macrophages.
Asian Pac. J. Cancer Prev.
13
:
1365
1370
.
13
Sternberg
,
D. I.
,
R.
Gowda
,
D.
Mehra
,
W.
Qu
,
A.
Weinberg
,
W.
Twaddell
,
J.
Sarkar
,
A.
Wallace
,
B.
Hudson
,
F.
D’Ovidio
, et al
.
2008
.
Blockade of receptor for advanced glycation end product attenuates pulmonary reperfusion injury in mice.
J. Thorac. Cardiovasc. Surg.
136
:
1576
1585
.
14
Saitoh
,
S.
2009
.
Chaperones and transport proteins regulate TLR4 trafficking and activation.
Immunobiology
214
:
594
600
.
15
Sims
,
G. P.
,
D. C.
Rowe
,
S. T.
Rietdijk
,
R.
Herbst
,
A. J.
Coyle
.
2010
.
HMGB1 and RAGE in inflammation and cancer.
Annu. Rev. Immunol.
28
:
367
388
.
16
Darwiche
,
S. S.
,
P.
Kobbe
,
R.
Pfeifer
,
L.
Kohut
,
H. C.
Pape
,
T.
Billiar
.
2011
.
Pseudofracture: an acute peripheral tissue trauma model.
J. Vis. Exp.
2074
.
17
Kobbe
,
P.
,
D. J.
Kaczorowski
,
Y.
Vodovotz
,
C. H.
Tzioupis
,
K. P.
Mollen
,
T. R.
Billiar
,
H. C.
Pape
.
2008
.
Local exposure of bone components to injured soft tissue induces toll-like receptor 4-dependent systemic inflammation with acute lung injury.
Shock
30
:
686
691
.
18
Menzel
,
C. L.
,
R.
Pfeifer
,
S. S.
Darwiche
,
P.
Kobbe
,
R.
Gill
,
R. A.
Shapiro
,
P.
Loughran
,
Y.
Vodovotz
,
M. J.
Scott
,
M. S.
Zenati
, et al
.
2011
.
Models of lower extremity damage in mice: time course of organ damage and immune response.
J. Surg. Res.
166
:
e149
e156
.
19
Chen
,
W.
,
H.
Zhong
,
X.
Wang
,
Q.
Pang
,
J.
Zhuang
,
J.
Hu
,
Y.
Chen
,
J.
Hu
,
J.
Liu
,
J.
Tang
.
2018
.
Mig6 reduces inflammatory mediators production by regulating the activation of EGFR in LPS-induced endotoxemia
.
J. Cell. Physiol
.
233
:
6975
6983
.
20
Tang
,
D.
,
T. R.
Billiar
,
M. T.
Lotze
.
2012
.
A Janus tale of two active high mobility group box 1 (HMGB1) redox states.
Mol. Med.
18
:
1360
1362
.
21
Park
,
J. S.
,
D.
Svetkauskaite
,
Q.
He
,
J. Y.
Kim
,
D.
Strassheim
,
A.
Ishizaka
,
E.
Abraham
.
2004
.
Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 protein.
J. Biol. Chem.
279
:
7370
7377
.
22
Ibrahim
,
Z. A.
,
C. L.
Armour
,
S.
Phipps
,
M. B.
Sukkar
.
2013
.
RAGE and TLRs: relatives, friends or neighbours?
Mol. Immunol.
56
:
739
744
.
23
Sessa
,
L.
,
E.
Gatti
,
F.
Zeni
,
A.
Antonelli
,
A.
Catucci
,
M.
Koch
,
G.
Pompilio
,
G.
Fritz
,
A.
Raucci
,
M. E.
Bianchi
.
2014
.
The receptor for advanced glycation end-products (RAGE) is only present in mammals, and belongs to a family of cell adhesion molecules (CAMs).
PLoS One
9
: e86903.
24
Dumitriu
,
I. E.
,
P.
Baruah
,
B.
Valentinis
,
R. E.
Voll
,
M.
Herrmann
,
P. P.
Nawroth
,
B.
Arnold
,
M. E.
Bianchi
,
A. A.
Manfredi
,
P.
Rovere-Querini
.
2005
.
Release of high mobility group box 1 by dendritic cells controls T cell activation via the receptor for advanced glycation end products.
J. Immunol.
174
:
7506
7515
.
25
Weber
,
D. J.
,
Y. M.
Allette
,
D. S.
Wilkes
,
F. A.
White
.
2015
.
The HMGB1-RAGE inflammatory pathway: implications for brain injury-induced pulmonary dysfunction.
Antioxid. Redox Signal.
23
:
1316
1328
.
26
Janssens
,
S.
,
R.
Beyaert
.
2003
.
Role of toll-like receptors in pathogen recognition.
Clin. Microbiol. Rev.
16
:
637
646
.
27
Trøseid
,
M.
,
A.
Lind
,
P.
Nowak
,
B.
Barqasho
,
B.
Heger
,
I.
Lygren
,
K. K.
Pedersen
,
T.
Kanda
,
H.
Funaoka
,
J. K.
Damås
,
D.
Kvale
.
2013
.
Circulating levels of HMGB1 are correlated strongly with MD2 in HIV-infection: possible implication for TLR4-signalling and chronic immune activation.
Innate Immun.
19
:
290
297
.
28
Wang
,
D.
,
J.
Lou
,
C.
Ouyang
,
W.
Chen
,
Y.
Liu
,
X.
Liu
,
X.
Cao
,
J.
Wang
,
L.
Lu
.
2010
.
Ras-related protein Rab10 facilitates TLR4 signaling by promoting replenishment of TLR4 onto the plasma membrane.
Proc. Natl. Acad. Sci. USA
107
:
13806
13811
.
29
Takahashi
,
K.
,
T.
Shibata
,
S.
Akashi-Takamura
,
T.
Kiyokawa
,
Y.
Wakabayashi
,
N.
Tanimura
,
T.
Kobayashi
,
F.
Matsumoto
,
R.
Fukui
,
T.
Kouro
, et al
.
2007
.
A protein associated with toll-like receptor (TLR) 4 (PRAT4A) is required for TLR-dependent immune responses.
J. Exp. Med.
204
:
2963
2976
.
30
Wakabayashi
,
Y.
,
M.
Kobayashi
,
S.
Akashi-Takamura
,
N.
Tanimura
,
K.
Konno
,
K.
Takahashi
,
T.
Ishii
,
T.
Mizutani
,
H.
Iba
,
T.
Kouro
, et al
.
2006
.
A protein associated with toll-like receptor 4 (PRAT4A) regulates cell surface expression of TLR4.
J. Immunol.
177
:
1772
1779
.
31
Liaunardy-Jopeace
,
A.
,
C. E.
Bryant
,
N. J.
Gay
.
2014
.
The COP II adaptor protein TMED7 is required to initiate and mediate the delivery of TLR4 to the plasma membrane.
Sci. Signal.
7
:
ra70
.
32
Doyle
,
S. L.
,
H.
Husebye
,
D. J.
Connolly
,
T.
Espevik
,
L. A. J.
O’Neill
,
A. F.
McGettrick
.
2012
.
The GOLD domain-containing protein TMED7 inhibits TLR4 signalling from the endosome upon LPS stimulation.
Nat. Commun.
3
:
707
.
33
Fiuza
,
C.
,
M.
Bustin
,
S.
Talwar
,
M.
Tropea
,
E.
Gerstenberger
,
J. H.
Shelhamer
,
A. F.
Suffredini
.
2003
.
Inflammation-promoting activity of HMGB1 on human microvascular endothelial cells.
Blood
101
:
2652
2660
.

The authors have no financial conflicts of interest.