(Journal of Leukocyte Biology. 2002;72:1084-1091.)
© 2002
by Society for Leukocyte Biology
HMGB1 as a DNA-binding cytokine
Ulf Andersson*,
,
Helena Erlandsson-Harris*,
Huan Yang
and
Kevin J. Tracey
* Department of Medicine, Rheumatology Research Unit, Karolinska Hospital, Stockholm, Sweden;
Department of Woman and Child Health, Karolinska Institutet, Astrid Lindgren Children’s Hospital, Stockholm, Sweden; and
Laboratory of Biomedical Science, North Shore–Long Island Jewish Research Institute, Manhasset, New York
Correspondence: Professor Ulf Andersson, Department of Rheumatology, Astrid Lindgren Children’s Hospital, Q1:02,171 76, Stockholm, Sweden. E-mail: ulf{at}mbox313.swipnet.se
 |
ABSTRACT
|
|---|
HMGB1 (high mobility group box chromosomal protein 1), historically known as an abundant, nonhistone architectural chromosomal protein, is extremely conserved across species. As a nuclear protein, HMGB1 stabilizes nucleosomes and allows bending of DNA that facilitates gene transcription. Unexpectedly, recent studies identified extracellular HMGB1 as a potent macrophage-activating factor, signaling via the receptor for advanced glycation end-products to induce inflammatory responses. It is released as a late mediator during inflammation and participates in the pathogenesis of systemic inflammation after the early mediator response has resolved. HMGB1 occupies a critical role as a proinflammatory mediator passively released by necrotic but not apoptotic cells. Necrotic Hmgb1-/- cells mediate minimal inflammatory responses. Stimulated macrophages actively secrete HMGB1 to promote inflammation and in turn, stimulate production of multiple, proinflammatory cytokines. HMGB1 mediates endotoxin lethality, acute lung injury, arthritis induction, activation of macrophages, smooth muscle cell chemotaxis, and epithelial cell barrier dysfunction. HMGB1 is structurally composed of three different domains: two homologous DNA-binding sequences entitled box A and box B and a highly, negatively charged C terminus. The B box domain contains the proinflammatory cytokine functionality of the molecule, whereas the A box region has an antagonistic, anti-inflammatory effect with therapeutic potential. Administration of highly purified, recombinant A box protein or neutralizing antibodies against HMGB1 rescued mice from lethal sepsis, even when initial treatment was delayed for 24 h after the onset of infection, establishing a clinically relevant therapeutic window that is significantly wider than for other known cytokines.
Key Words: inflammation • nuclear cytokine • necrosis • RAGE ligand • sepsis • arthritis
|
INTRODUCTION
|
|---|
HMGB1 (high mobility group box chromosomal protein 1) was isolated 30 years ago from calf thymus as an abundant, chromosomal protein with important structural functions in chromatin organization [1
]. The name derives from its characteristic electrophoretic mobility in polyacrylamide gels. The HMGB family includes the three nuclear proteins HMGB1 (previously named amphoterin or HMG1), HMGB2 (previously HMG2), and HMGB3 (previously HMG4 or HMG2b; reviewed in refs. [2
3
4
]). The structure of these proteins is highly conserved and is made up by three distinct domains [5
6
7
]. The two DNA-binding elements, termed A and B box, are each made up of 80 amino acid residues arranged in three 
-helices. These two domains are strongly, positively charged, and the third C-terminal domain is extremely negatively charged, as it contains 30 repetitive aspartic and glutamic acid residues. HMGB proteins are thus bipolarly charged, almost like detergents; this feature led to the designation of HMGB1 as "amphoterin," but amphoterin is the product of the HMGB1 gene [8
]. The amino acid sequence within HMGB family members exhibits 85% similarity, but the proteins have a distinctly different tissue-expression pattern. HMGB1 is ubiquitously present in all vertebrate nuclei, but the expression of HMGB2 and HMGB3 is more restricted. HMGB2 is widely present during embryonic development, and expression is limited to only testis and lymphoid tissue in the adult mouse [9
]. HMGB3 appearance has only been demonstrated during embryogenesis [10
]. Quite surprisingly, extracellular HMGB1 was recently identified as a cytokine, a new observation that provides the background for this review [11
12
13
]. It is presently unknown whether HMGB2 and HMGB3 are also capable of mediating extracellular cytokine bioactivity.
HMGB1 is a 215 amino acid protein with a uniquely conserved sequence among species. Mouse and rat HMGB1 are identical; they differ from human HMGB1 by only two substitutions in the continuous C-terminal stretch of glutamate and aspartate residues [14
15
16
17
]. HMGB1 binds double-stranded DNA without sequence specificity and interacts with other DNA binding proteins, which facilitate chromatin bending. This architectural function facilitates the binding of several transcription factors including some steroid hormone receptors and recombination activating gene recombinase (reviewed in ref. [3
]). The phenotype of Hmgb1-deficient mice underscores the important role of HMGB1 as a regulator of transcription, as these animals die within a few hours from birth, possibly as a result of a defect in activation of glucocorticoid receptor-responsive genes [18
]. Importantly, it has proven feasible to propagate cell lines from Hmgb1-deficient mice, although the gene abolition is not compatible with the post-delivery life of the animal.
|
DISCOVERY OF HMGB1 AS A CYTOKINE
|
|---|
The unexpected identification of HMGB1 as a cytokine originates from studies of endotoxemia and sepsis [11
]. Numerous attempts have been made during the last decade to treat sepsis with antibodies or antagonists against several well-known, proinflammatory cytokines including tumor necrosis factor (TNF). Although these approaches have led to the development of successful therapeutics for the treatment of rheumatoid arthritis and Crohn’s disease, they have not proven to be effective in treating heterogenous patients with sepsis. One major difficulty in targeting TNF in the treatment of sepsis is that most of this and other early proinflammatory cytokines are released acutely during the septic response, usually within minutes to hours after exposure to the invasive stimuli. This kinetic pattern of cytokine release does not allow time long enough for successful clinical intervention, although experimental therapy studies have indicated strong protection using a very early approach. The fact that death in sepsis frequently occurs at time points long after serum TNF and interleukin (IL)-1 levels have returned to basal levels prompted one of us (K. J. T.) to systematically search for additional explanations. Could sepsis lethality be caused by a previously unrecognized factor released by activated macrophages in a delayed temporal response as compared with TNF and IL-1? If so, could that molecule be successfully, therapeutically targeted in a clinically relevant setting?
This search culminated in the unexpected identification of HMGB1 as a late mediator of lipopolysaccharide (LPS) lethality [11
]. The strategy that enabled these findings was based on in vitro stimulation of murine macrophages with LPS, TNF, or IL-1 and subsequent analysis of the conditioned culture medium by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Released proteins that appeared in the supernatants later than 6–8 h from initiation of cultures were identified by molecular weight and amino acid sequencing. These methods verified the unexpected demonstration of extracellular HMGB1 in the culture supernatants at time points later than 8 h post-LPS induction, peaking at 18 h. Furthermore, serum concentrations of HMGB1 increased significantly 8–32 h after in vivo administration of LPS or TNF in mice. Systemic administration of purified recombinant HMGB1 (rHMGB1) protein produced in Escherichia coli was lethal to LPS-sensitive C3H/HeN mice as well as LPS-resistant C3H/HeJ mice, indicating that HMGB1 may mediate lethal toxicity in the absence of LPS signal transduction. Injection of toxic doses of HMGB1 in mice leads to the development of fever, weight loss, piloerection, shivering, and microthrombi formation in liver and lungs. Importantly, passive administration of anti-HMGB1 antibodies protected against LPS lethality in mice even when the therapy was delayed until after the early proinflammatory cytokine response [11
]. Increased serum concentrations of HMGB1 have been observed in patients with sepsis, and higher levels were observed in patients that succumbed to sepsis. HMGB1 serum levels were also observed to be increased in a patient with hemorrhagic shock, suggesting that HMGB1 might play a role in inadequate tissue perfusion [19
]. Together, these data indicate that HMGB1 is a late mediator of endotoxin lethality that may be a therapeutic target in patients.
Apart from being a secreted product by activated macrophages and monocytes, HMGB1 release has also been demonstrated in cultured pituicytes exposed to LPS, IL-1, or TNF in a time- and dose-dependent manner [12
]. HMGB1 secretion from pituicytes occurred as early as 3 h and reached peak levels 8–10 h after TNF exposure. IFN-
alone did not stimulate release of HMGB1, but synergistically enhanced TNF-mediated HMGB1 released by three- to fivefold from pituicytes and monocytes. The regulation of HMGB1 release and the relevant control mechanisms are enigmatic and under investigation. HMGB1 lacks a classic signal sequence, although there is homology to a nuclear localization sequence that may participate in the nuclear functions of the protein. It is also unclear whether HMGB1 secreted from activated macrophages resides in the cell membrane before release.
It was later demonstrated that HMGB1 is not only a product released from activated macrophages/monocytes but also by itself acts as a strong mediator of macrophage activation. HMGB1, like other cytokine mediators of endotoxemia such as TNF and IL-1, activates a cascade of proinflammatory cytokine release from human monocytes and macrophages [11
, 13
, 20
, 21
]. Addition of HMGB1 to monocyte cultures activated the formation of TNF, IL-1, IL-1ß, IL-1ra, IL-6, IL-8, macrophage-inflammatory protein-1 (MIP-1), and MIP-1ß but not IL-10 or IL-12 [13
]. HMGB1 induced de novo cytokine synthesis, as cytokine mRNA levels were increased after HMGB1 stimulation. Notably, TNF synthesis after HMGB1 stimulation was significantly delayed as compared with LPS-stimulated TNF production (Fig. 1
). HMGB1-induced TNF release was biphasic with the first peak at 3 h, followed by a second peak 8–10 h after HMGB1 exposure. HMGB1-induced TNF mRNA levels in monocytes were still elevated 8–10 h after HMGB1 stimulation. Intraperitoneal administration of rHMGB1 to mice significantly increased serum TNF levels in vivo. Altogether, the data indicate that as a cytokine, HMGB1 acts as a potent stimulus of proinflammatory cytokine release, which occurs in a delayed manner compared with LPS. This cytokine-stimulating effect of HMGB1 is independent of LPS signal transduction, as HMGB1 stimulates cytokine release in LPS-resistant C3H/HeJ mice. The HMGB1-induced cytokine production in cultured human mononuclear blood cells is cell type-specific, as cytokine synthesis was not increased in lymphocytes stimulated with HMGB1 [13
].
View larger version (121K):
[in this window]
[in a new window]
|
Figure 1. TNF expression in HMGB1- or LPS-stimulated, cultured human blood mononuclear cells. Cells were cultured alone with rHMGB1 (1 µg/ml) or LPS (100 ng/ml) for 1 and 8 h, respectively, and were then stained for intracellular TNF synthesis. TNF-producing monocytes can be detected as cells containing an intracellular, rounded brown dot, representing the accumulation of TNF in the Golgi organelle during synthesis. The number of TNF-producing monocytes in the LPS-stimulated cultures was significantly increased in cultures harvested within 1 h compared with HMGB1-stimulated or unstimulated cultures. In contrast, HMGB1 stimulation significantly increased TNF expression at 8 h compared with LPS activation. No TNF synthesis occurred in unstimulated cultures.
|
|
HMGB1 activation of macrophages occurs with a significantly delayed kinetic as compared with LPS-activated macrophages. TNF production in primary monocyte cultures stimulated by LPS has well stopped after a culture period of 24 h and will not reappear by addition of LPS and fresh medium to the cells. In contrast, after HMGB1 exposure, there is a strong induction of TNF synthesis that occurs after 2 h and persists for as long as 8 h, even when "LPS-exhausted" cells are exposed and cocultured with HMGB1 (U. A. and H. E-H., manuscript in preparation). Few, if any, factors have to our knowledge previously been described that can overcome endotoxin resistance. In summary, HMGB1 is not only secreted by activated macrophages/monocytes, but it also provokes a delayed response in the same cells and thereby prolongs the inflammatory response.
|
THE CYTOKINE-INDUCING ACTIVITY RESIDES IN THE B BOX DOMAIN
|
|---|
The structural basis of HMGB1 cytokine activity has recently been elucidated by expression of full-length and truncated forms of HMGB1. The purified proteins have been screened for stimulating inflammatory responses in macrophage cultures and by using in vivo experiments (K. J. T. and H. Y., manuscript in preparation). The truncated protein containing the B box motif (B box protein) was demonstrated to be a potent inducer of TNF production in cultured macrophages. These observations were supported by the fact that the chemically synthesized B box protein could also stimulate TNF release under similar conditions. Affinity-purified anti-B box antibodies significantly inhibited TNF release induced by full-length HMGB1 or B box protein, giving evidence that the macrophage-stimulating effects of HMGB1 were B box-specific.
In vivo effects of injection of HMGB1 B box protein were studied in Balb/c mice sensitized with D-galactosamine (D-gal), a model widely used to study cytokine toxicity [22
, 23
]. Administration of B box protein was lethal, and dose-dependent death occurred within 7–8 h. Injection of B box protein to LPS-resistant C3H/HeJ mice was also lethal, indicating that HMGB1 B box protein is toxic in the absence of LPS signal transduction. Administration of HMGB1 B box in vivo significantly stimulated serum levels of TNF, IL-1ß, and IL-6. Histological examination of sections obtained from mice exposed to lethal doses of B box protein showed liver injury with polymorphonuclear leukocyte accumulation and apoptotic hepatocytes. Cardiac tissue showed signs of ischemia and loss of cross striation in myocardial fibers, and no abnormality was seen in kidneys and lungs. These signs of specific organ injury are similar to the tissue injury mediated by LPS in the D-gal model [24
, 25
] but differ significantly from the organ damage caused by TNF treatment [26
]. This fact combined with the absence of elevated TNF levels at 7–8 h after HMGB1 challenge indicate that HMGB1 kills through mechanisms distinct from TNF. Mice passively immunized with anti-HMGB1 B box antibodies were significantly protected against lethal endotoxemia, indicating that selective inhibition of HMGB1 B box attenuates the toxicity of endogenous HMGB1. Together, these data indicate that the inflammatory domain of HMGB1 maps to the B box and that B box alone is sufficient to recapitulate the cytokine-stimulating effects of full-length HMGB1.
|
THE A BOX ANTAGONIZES HMGB1-INDUCED INFLAMMATION
|
|---|
As described previously, truncation of HMGB1 into individual structural domains revealed that the B box induced strong proinflammatory cytokine production in cultured macrophages. The A box protein did not mediate such effects, which was somewhat surprising, as there is a 30–40% shared homology between the two domains (K. J. T. and H. Y., manuscript in preparation). This unexpected observation suggested a testable hypothesis: that the A box protein might function as a competitive antagonist of HMGB1. To determine whether A box protein may act as an antagonist of HMGB1, TNF and IL-1ß release were measured in macrophage cultures exposed to A box protein and stimulated with HMGB1. It was demonstrated that the A box protein dose-dependently inhibited HMGB1-induced TNF and IL-1ß release in the cultures. A box protein significantly displaced saturable 125I-labeled HMGB1 binding to macrophage cultures at 4°C, indicating that A box protein is a competitive antagonist of HMGB1 binding to macrophages.
To examine the role of administering A box protein to animals with sepsis, it was first necessary to determine whether HMGB1 is produced during sepsis, as has been described for endotoxemia, and whether it exhibits a "late" kinetic profile. HMGB1 levels were measured in a standardized model of sepsis, in which mice are subjected to cecal ligation and puncture. A perforation is created in a surgically induced cecal diverticulum, which leads to polymicrobial peritonitis and sepsis [27
]. Serum HMGB1 levels were undetectable for the first 8 h after cecal perforation but increased significantly after 18 h. Raised serum HMGB1 levels remained at elevated plateau levels for at least 72 h, a time course that is quite similar to the previously described, delayed HMGB1 kinetics in endotoxemia [11
]. Significant increases in HMGB1 levels were also observed in the peritoneal exudate fluids 48 h after cecal perforation. The kinetic profile is atypical from other proinflammatory mediators that have been implicated in causing lethality. The timing of increased serum HMGB1 levels corresponded closely to the onset of clinical signs of sepsis. During the first 8 h after cecal perforation, all animals appeared mildly ill, with some diminished activity and loss of exploratory behavior. Over the ensuing 18 h, all animals became gravely ill, huddled together in groups with piloerection, did not seek water or food, and became minimally responsive to external stimuli. Administration of highly purified, recombinant A box protein to mice with established clinical signs significantly rescued the animals from the lethal effects of sepsis, even when treatment was initiated as late as 24 h after cecal perforation. These results establish a clinically relevant, therapeutic window that is significantly wider than for other known cytokines.
|
HMGB1 RECEPTOR AND INTRACELLULAR SIGNALING
|
|---|
The signaling mechanisms by which HMGB1 interacts with target cells are still incompletely understood. However, work by Rauvala and colleagues [28
] in identifying HMGB1 expression in the developing brain has provided important clues. They demonstrated that cell membrane-associated HMGB1 signals neurite outgrowth by interaction with the multiligand receptor RAGE (the receptor for advanced glycation end products) [29
]. A similar, extracellular proteolytic activity induced by HMGB1, expressed on the leading edge of motile cells, has recently been confirmed in an experimental tumor system [30
]. Tumor growth and metastasis formation were demonstrated to be suppressed when HMGB1 was prevented from interacting with RAGE using RAGE-blocking antibodies or neutralizing anti-HMGB1 antibodies. RAGE is a transmembrane protein that is a member of the immunoglobulin (Ig) superfamily and is homologous to a neural cell-adhesion molecule [31
]. RAGE is expressed in the central nervous system on endothelial cells, smooth muscle cells, and mononuclear phagocytes. This receptor has been implicated in the pathogenesis of multiple diseases such as diabetes, atherosclerosis, and Alzheimer’s disease (reviewed in ref. [32
]). HMGB1 is a specific and saturable ligand for RAGE binding with a higher affinity than the receptor’s other known ligands, advanced glycation end products (AGEs) [29
]. HMGB1-induced intracellular signaling through RAGE can activate two different cascades, one involving the small GTPases Rac and Cdc42, leading to cytoskeletal reorganization, and a second that involves the Ras-mitogen-activated protein (MAP) kinase pathway and subsequent nuclear factor (NF)-
B nuclear translocation-mediating inflammation [33
]. It is interesting that RAGE is expressed on mononuclear phagocytes, where its interaction with AGEs enhances cellular oxidant stress [34
], including generation of thiobarbituric acid reactive substances and activation of NF-B [33
]. RAGE signaling stimulates an inflammatory response, as AGE-modified ß2 microglobulin binds RAGE in mononuclear phagocytes to mediate monocyte chemotaxis and induce TNF release [34
]. Several groups have shown that HMGB1-RAGE interaction will also lead to phosphorylation of MAP-kinases p38, p42/p44, and c-jun NH2-terminal kinase, resulting in NF-B activation [30
, 35
]. Furthermore, HMGB1-mediated migration of smooth muscle cells involves activation of the MAP-kinase pathway and a nuclear translocation of phosphorylated extracellular regulated kinase-1 and -2. This process also involves signaling via a yet-unidentified Gi/o protein [35
]. The recent finding that HMGB1-induced differentiation of erythroleukaemia cells is independent of RAGE signaling indicates that there are additional signaling HMGB1 receptors to be identified [36
]. Apart from HMGB1-receptor interactions, it has been demonstrated that HMGB1, being a "sticky" molecule, binds to many membrane molecules such as heparin, proteoglycans including syndecan-1, sulfoglycolipids, and phospholipids [37
, 38
]. A local binding of HMGB1 will restrict the diffusion of extracellular HMGB1 and inhibit systemic release, which might be potentially dangerous.
|
NECROSIS-INDUCED INFLAMMATION IS CAUSED BY HMGB1
|
|---|
Necrosis is the mode of cell death characterized by subsequent induction of an inflammatory response; necrotic cells, in contrast to apoptotic cells, lose their cell membrane integrity and release intracellular contents that cause inflammation. Until now, the specific molecular background for necrosis-induced inflammation has been obscure. Recent work by Bianchi and his group [35
, 39
, 40
] demonstrated that HMGB1 is a critical factor connecting necrotic cell death to inflammation. They postulated this hypothesis based on the novel information that extracellular HMGB1 is a potent proinflammatory molecule [11
] combined with the fact that HMGB1 is only loosely associated with the chromatin of interphase or mitotic cells [41
, 42
]. HMGB1 binds tightly to nucleosomes such as histones but much weaker to the cellular DNA. This suggested HMGB1 to be a candidate for providing a diffusible nuclear signal from a disintegrated cell that might mediate inflammation. Coculture of damaged or necrotic nucleated cells with macrophages induced nuclear NF-B translocation in the macrophages leading to a necrosis-driven inflammatory response. Further studies with necrotic or damaged Hmgb1-/- cells in cocultures with macrophages revealed a greatly reduced ability to promote necrosis-induced inflammation, proving the importance of the release of HMGB1 as the proinflammatory signal from the necrotic cell [40
]. Apoptotic cells failed to release HMGB1 even after undergoing secondary necrosis and did not mediate any inflammatory response. It was demonstrated that HMGB1 was strongly bound to chromatin in apoptotic cells because of generalized underacetylation of the histones. If chromatin deacetylation was actively prevented during the apoptosis process, HMGB1 was readily released extracellularly and mediated inflammation. The importance of HMGB1 mediating necrosis-induced inflammation was also confirmed in vivo using animal experiments. Taken together, these results very convincingly demonstrate that unprogrammed cell death leads to inflammation and tissue repair via release of nuclear HMGB1.
Recent studies by Li and co-workers [43
] showed that necrotic but not apoptotic cells liberate a nuclear factor that induced expression of genes involved in inflammation in macrophages by interacting with the Toll-like receptor 2 (TLR2). These receptors have until now been considered to be meant for interaction with exogenous microbial agents exclusively (reviewed in ref. [44
]). The released nuclear factor could not be identified in the report from Li and his co-workers, but it is now plausible that this putative factor is indeed identical to HMGB1. The bioactivity was lost during purification "using conventional separation technology." We have observed that the cytokine-inducing activity by HMGB1 is very sensitive to acidic pH. Natural or rHMGB1 exposed to pH 5.5 have in our hands lost the cytokine-inducing capacity, and the bactericidal effect was retained [45
]. Conventional high-pressure liquid chromatography (HPLC) fractionation of rHMGB1 led in our experiments to a complete loss of the stimulatory activity of cytokine synthesis. The purification process applied to recover HMGB1 is thus critical in protecting the inflammation-inducing potential of HMGB1. We suspect that this fact may explain why the proinflammatory role of HMGB1 has been undiscovered until recently. A putative role of TLR2 as an alternative signaling receptor for HMGB1 needs to be further evaluated.
|
CELLULAR SOURCES OF HMGB1
|
|---|
HMGB1 is expressed in the nucleus of all vertebrate cells. Activated mononuclear phagocytes and pituicytes have developed a capacity to translocate their nuclear HMGB1 to the cytoplasm. Whether there are additional cell lineages that have developed this capacity is presently unknown. Resting human platelets express cytoplasmic HMGB1, which is exported to the cell surface during platelet activation [46
]. The functional role of this is not understood today, but it is well established that HMGB1 associates with plasminogen and tissue plasminogen activator on cell surfaces and enhances plasminogen generation and proteolysis [47
, 48
]. This interaction is of particular interest in light of the recent and intense focus on the design of sepsis therapeutics that interferes with activation of blood-clotting systems. It will be important to delineate the connection between neutralization of HMGB1 and coagulation mechanisms, as these two systems occupy a critical, final common pathway to tissue injury and death from sepsis.
|
EXTRACELLULAR HMGB1 RELEASE
|
|---|
There are two distinctly separate ways for HMGB1 to be secreted from a cell (Fig. 2
). HMGB1 can be passively released from the nuclei of necrotic or disintegrating, damaged cells or actively secreted from activated macrophages/monocytes or pituicytes, which does not involve cell death. However, it was not known until recently how HMGB1 is actively secreted from these cells [49
]. HMGB1 does not have a leader sequence and is thus not processed via the endoplasmatic reticulum/Golgi pathway. This is a characteristic shared with a few other secreted proteins such as fibroblast growth factor and IL-1ß. Pulse-chase labeling with 35S-methionine revealed that most of the HMGB1 secreted during the first 12 h after TNF stimulation of macrophages was from a preformed pool; thereafter, the extracellular HMGB1 was newly synthesized [11
]. Recent data demonstrate that cultured, activated macrophages will translocate their nuclear HMGB1 to the cytoplasm before extracellular release via lysosomal exocytosis [49
]. We have analogous observations from in vivo studies of joint tissue in experimental arthritis in mice and rats. The cellular expression of HMGB1 in immunohistochemically stained sections from normal joint tissue showed a strictly nuclear pattern (Fig. 3
). This staining appearance changed dramatically in inflamed synovial tissue during the course of collagen-induced or adjuvant arthritis. Additional cytoplasmic HMGB1 expression could be demonstrated in many macrophage-like cells, and the contributing nuclear HMGB1 staining in a subset of these cells was clearly reduced or absent [50
].
View larger version (28K):
[in this window]
[in a new window]
|
Figure 2. Schematic illustration of potential pathways for HMGB1 release leading to inflammatory responses. HMGB1 can be extracellularly released by passive secretion from any necrotic cell or by active secretion from activated macrophages/monocytes.
|
|
View larger version (74K):
[in this window]
[in a new window]
|
Figure 3. Immunohistochemical staining for HMGB1 (brown) in synovial tissue from a healthy Dark Agouti rat (left picture) compared with inflamed synovial tissue from collagen-induced arthritis in a DA rat (right picture). The blue color reflects a nuclear counterstaining with hematoxylin. HMGB1 staining in the normal tissue was mainly restricted to cell nuclei. Additional cytoplasmic HMGB1 staining was evident in numerous cells with macrophage appearance in the inflamed synovial tissue (right picture).
|
|
|
BACTERICIDAL ACTIVITY OF HMGB1
|
|---|
Recent studies of antibacterial factors purified by tissue extraction from human adenoid glands without inflammatory signs revealed that HMGB1 may mediate strong, direct bactericidal effects [45
]. Reversed-phase HPLC purification of adenoid-derived HMGB1 or rHMGB1 demonstrated that HMGB1 functionally belongs to the growing family of antibiotic peptides. The mechanism for this activity is presently unknown but must be different from the cytokine-inducing activity. HPLC fractionation of HMGB1 is excellent to recover the bactericidal action but kills the cytokine-promoting effects. HMGB1 could not be detected as an extracellular factor in these studies of noninfected material.
|
HMGB1 AND DISEASE
|
|---|
A detailed investigation of the role of HMGB1 in human diseases is just beginning. Several important observations from animal studies have been made during the last 2–3 years.
Sepsis
The novel demonstration of HMGB1 as an important, delayed mediator in the pathogenesis of experimental sepsis and endotoxemia has already been discussed in this review. It is of great interest that there is a unique opportunity to successfully treat experimental sepsis by a "late" administration of neutralizing HMGB1 antibodies or the recombinant antagonistic A box protein. The relevance for treatment of human sepsis needs to be addressed in controlled, clinical trials.
Lung inflammation
Intratracheally administered HMGB1 is a mediator of acute inflammatory lung injury as manifested by neutrophil accumulation, edema, and increased pulmonary cytokine production [20
]. HMGB1 stimulated synthesis of TNF, IL-1ß, and MIP-2 in the lung tissue. Administration of anti-HMGB1 antibodies to animals exposed to LPS intratracheally significantly decreased the migration of neutrophils to the lungs and attenuated the edema formation. Anti-HMGB1 antibodies failed to significantly suppress the LPS-inducible up-regulation of pulmonary TNF, IL-1ß, and MIP-2, indicating that the protective effects of HMGB1 antibodies against endotoxin-induced pulmonary injury were specific. Taken together, these observations indicate that HMGB1 acted as a distal mediator of acute pulmonary injury
Arthritis
Macrophages play a pivotal role in the pathogenesis of chronic arthritis. We are currently studying whether HMGB1 release from activated macrophages has a pathogenetic role in the development of experimental arthritis in mice and rats [50
]. Normal synovial tissue displayed a strictly nuclear HMGB1 localization as analyzed by immunohistochemistry, and tissue sections from synovitis in expressed disease showed strong, extracellular HMGB1 depositions and additional cytoplasmic staining in macrophages and vascular endothelial cells. Therapeutic intervention with systemic A box protein injections in established collagen-induced arthritis in mice has been tried by us with very promising results (manuscript in preparation).
Biopsy samples from patients with rheumatoid arthritis have also demonstrated cytoplasmic HMGB1 staining in macrophage-like cells, and 12 out of 14 studied synovial fluid samples from rheumatoid arthritis patients contained detectable HMGB1 levels of 1–10 µg/ml. No correlation between recorded TNF and HMGB1 levels could be demonstrated in the synovial fluids [50
].
|
FUTURE PERSPECTIVES
|
|---|
To elucidate the importance of extracellularly released HMGB1 in the pathogenesis of human diseases and to plan for future therapeutic intervention, there are a number of basic questions to resolve. One issue will be to provide simple and quantitative assays for measuring HMGB1 levels in clinical and research samples. The fact that HMGB1 is a highly conserved molecule between species has retarded the establishment of IgG monoclonal antibodies against HMGB1. Such antibodies will be mandatory for analytic and therapeutic aims. Quantitative assessments of HMGB1 in human disease are presently based on Western blot assays or immunohistochemical methods, which do not allow processing of a great enough number of samples and only provide semiquantitative analysis results. Enzyme-linked immunosorbent assay methods will help to dissect the role of HMGB1 in human diseases but need to be complemented by bioassays to evaluate the bioactivity of the detected extracellular HMGB1. It is clear that any "neutralizing antibodies" should be screened for inhibition of HMGB1-activated TNF release in macrophages. These basic tools will enable assessment of fundamental questions that still need to be addressed; e.g., What is the quantitative relationship between the extracellular HMGB1 pool from necrotic cells versus active secretion from macrophages, platelets, vascular endothelial cells, and other putative HMGB1 sources? Are there other HMGB1 receptors than RAGE? To what extent are TNF and other proinflammatory molecules needed for the proinflammatory role of HMGB1? Will HMGB1 be validated as a clinical target, like TNF or IL-1, to modulate acute or chronic inflammation, or will it be too dangerous to interfere with a molecule that is so central for the interplay between necrotic cell death with subsequent inflammation and repair responses? These are indeed very exciting questions that need to be addressed and elucidated in the near future.
Received May 27, 2002;
revised July 19, 2002;
accepted July 25, 2002.
|
REFERENCES
|
|---|
- Goodwin, G. H., Sanders, C., Johns, E. W. (1973) A new group of chromatin-associated proteins with a high content of acidic and basic amino acids Eur. J. Biochem. 38,14-17[Medline]
- Bustin, M. (1999) Regulation of DNA-dependent activities by the functional motifs of the high mobility group chromosomal proteins Mol. Cell. Biol. 19,5237-5246[Free Full Text]
- Bianchi, M. E., Beltrame, M. (2000) Upwardly mobile proteins. The role of HMG proteins in chromatine structure, gene expression and neoplasia EMBO Rep. 1,109-119[Medline]
- Bustin, M. (2001) Revised nomenclature for high mobility group (HMG) chromosomal proteins Trends Biochem. Sci. 26,152-153[Medline]
- Read, C. M., Cary, P. D., Crane-Robinson, C., Driscoll, P. C., Norman, D. G. (1993) Solution structure of a DNA-binding domain from HMG1 Nucleic Acids Res. 21,3427-3436[Abstract/Free Full Text]
- Weir, H. M., Kraulis, P. J., Hill, C. S., Raine, A. R. C., Laue, E. D., Thomas, J. O. (1993) Structure of the HMG box motif in the B-domain of HMG1 EMBO J. 12,1311-1319[Medline]
- Hardman, C. H., Broadhurst, R. W., Raine, A. R. C., Grasser, K. D., Thomas, J. O., Laue, E. D. (1995) Structure of the A-domain of HMG1 and its interaction with DNA as studied by heteronuclear three- and four-dimensional NMR spectroscopy Biochemistry 34,16596-16607[Medline]
- Merenmies, J., Pihlaskari, R., Laitinen, J., Wartiovaara, J., Rauvala, H. (1991) 30-kDa Heparin-binding protein of brain (amphoterin) involved in neurite outgrowth. Amino acid sequence and localization in the filopodia of the advancing plasma membrane J. Biol. Chem. 266,16722-16729[Abstract/Free Full Text]
- Ronfani, R., Ferraguti, M., Croci, L., Ovitt, C. E., Schöler, H. R., Consalez, G. G., Bianchi, M. E. (2001) Reduced fertility and spermatogenesis defects in mice lacking chromosomal protein Hmgb2 Development 128,1265-1273[Abstract]
- Vaccari, T., Beltrame, M., Ferrari, S., Bianchi, M. E. (1998) Hmg4a new member of the Hmg 1/2 gene family Genomics 49,247-252[Medline]
- Wang, H., Bloom, O., Zhang, M., Vishnubhakat, J. M., Ombrellino, M., Che, J., Frazier, A., Yang, H., Ivanova, S., Borovikova, L., Manogue, K. R., Faist, E., Abraham, E., Andersson, J., Andersson, U., Molina, P. E., Abumrad, N. N., Sama, A., Tracey, K. J. (1999) HMG-1 as a late mediator of endotoxin lethality in mice Science 285,248-251[Abstract/Free Full Text]
- Wang, H., Vishnubhakat, J. M., Bloom, O., Zhang, M., Ombrellino, M., Sama, A., Tracey, K. J. (1999) Proinflammatory cytokines (tumor necrosis factor and interleukin 1) stimulate release of high mobility group protein-1 by pituicytes Surgery 126,389-392[Medline]
- Andersson, U., Wang, H., Palmblad, K., Aveberger, A-C., Bloom, O., Erlandsson-Harris, H., Janson, A., Kokkola, R., Yang, H., Tracey, K. J. (2000) HMG-1 stimulates proinflammatory cytokine synthesis in human monocytes J. Exp. Med. 192,565-570[Abstract/Free Full Text]
- Paonessa, G., Frank, R., Cortese, R. (1987) Nucleotide sequence of rat liver HMG1 cDNA Nucleic Acids Res. 15,9077-9081[Free Full Text]
- Wen, L., Huang, J. K., Johnson, B. H., Reeck, G. R. (1989) A human placental cDNA clone that encodes nonhistone chromosomal protein HMG-1 Nucleic Acids Res. 17,1197-1214[Abstract/Free Full Text]
- Rauvala, H., Merenmies, J., Pihlaskari, R., Korkolainen, M., Huhtala, M-L., Panula, P. (1988) The adhesive and neurite-promoting molecule p30: analysis of the amino-terminal sequence and production of antipeptide antibodies that detect p30 at the surface of neuroblastoma cells and of brain neurons J. Cell Biol. 107,2293-2305[Abstract/Free Full Text]
- Ferrari, S., Finelli, P., Rocchi, M., Bianchi, M. E. (1996) The active gene that encodes human high mobility group 1 protein (HMG1) contains introns and maps to chromosome 13 Genomics 35,367-371[Medline]
- Calogero, S., Grassi, F., Aguzzi, A., Voigtlander, T., Ferrier, P., Ferrari, S., Bianchi, M. E. (1999) The lack of chromosomal protein Hmg1 does not disrupt cell growth but causes lethal hypoglycemia in newborn mice Nat. Genet. 22,276-280[Medline]
- Ombrellino, M., Wang, H., Ajemian, M. S., Talhouk, A., Scher, L. A., Friedman, S. G., Tracey, K. J. (1999) Increased serum concentrations of high-mobility-group protein 1 in hemorrhagic shock Lancet 354,1446-1447[Medline]
- Abraham, E., Arcaroli, J., Carmody, A., Wang, H., Tracey, K. J. (2000) HMG-1 as a mediator of acute lung injury J. Immunol. 165,2950-2954[Abstract/Free Full Text]
- Yang, H., Wang, H., Tracey, K. J. (2001) HMG-1 rediscovered as a cytokine Shock 15,247-253[Medline]
- Galanos, C., Freudenberg, M. A., Reutter, W. (1979) Galactosamine-induced sensitization of the lethal effects of endotoxin Proc. Natl. Acad. Sci. USA 76,5939-5943[Abstract/Free Full Text]
- Lehmann, V., Freudenberg, M. A., Galanos, C. (1987) Lethal toxicity of lipopolysaccharide and tumor necrosis factor in normal and D-galactosamin-treated mice J. Exp. Med. 165,657-663[Abstract/Free Full Text]
- Morikawa, A., Sugijama, T., Kato, Y., Koide, N., Jiang, G. Z., Takahashi, K., Tamada, Y., Yokochi, T. (1996) Apoptotic cell death in the response of D-galactosamin-sensitized mice to lipopolysaccharide as an experimental endotoxic shock model Infect. Immun. 64,734-738[Abstract]
- Hishinuma, I., Nagakawa, J., Hirota, K., Miyamoto, K., Tsukidate, K., Yamanaka, T., Katayama, K., Yamatsu, I. (1990) Involvement of tumor necrosis factor-alpha in development of hepatic injury in galactosamine-sensitized mice Hepatology 12,1187-1191[Medline]
- Tracey, K. J., Beutler, B., Lowry, S. F., Merryweather, J., Wolpe, S., Milsark, I. W., Hariri, R. J., Fahey, T. J., Zentella, A., Albert, J. D. (1986) Shock and tissue injury induced by recombinant human cachectin (1986) Science 234,470-474[Abstract/Free Full Text]
- Wichmann, M. W., Haisken, J. M., Ayala, A., Chaudry, I. H. (1996) Melatonin administration following hemorrhagic shock decreases mortality from subsequent septic challenge J. Surg. Res. 65,109-114[Medline]
- Rauvala, H., Huttunen, H. J., Fages, C., Kaksonen, M., Kinnunen, T., Imai, S., Raulo, E., Kilpelainen, I. (2000) Heparin-binding proteins HB-GAM (pleiotrophin) and amphoterin in the regulation of cell motility Matrix Biol. 19,377-387[Medline]
- Hori, O., Brett, J., Slattery, T., Cao, R., Zhang, J., Chen, J. X., Nagashima, M., Lundh, E. R., Vijay, S., Nitecki, D., et al (1995) The receptor for advanced glycation end products (RAGE) is a cellular binding site for amphoterin. Mediation of neurite outgrowth and co-expression of rage and amphoterin in the developing nervous system J. Biol. Chem. 270,25752-25761[Abstract/Free Full Text]
- Taguchi, A., Blood, D. C., del Toro, G., Canet, A., Lee, D. C., Qu, W., Tanji, N., Lu, Y., Lalla, E., Fu, C., Hofmann, M. A., Kislinger, T., Ingram, M., Lu, A., Tanaka, H., Hori, O., Ogawa, S., Stern, D. M., Schmidt, A. M. (2000) Blockade of RAGE-amphoterin signalling suppresses tumour growth and metastases Nature 405,354-360[Medline]
- Neeper, M., Schmidt, A. M., Brett, J., Yan, S. D., Wang, F., Pan, Y. C., Elliston, K., Stern, D., Shaw, A. (1992) Cloning and expression of a cell surface receptor for advanced glycosylation end products of proteins J. Biol. Chem. 267,14998-15004[Abstract/Free Full Text]
- Schmidt, A. M., Yan, S. D., Yan, S. F., Stern, D. M. (2000) The biology of the receptor for advanced glycation end products and its ligands Biochem. Biophys. Acta 1498,99-111[Medline]
- Huttunen, H. J., Fages, C., Rauvala, H. (1999) Receptor for advanced glycation end products (RAGE)-mediated neurite outgrowth and activation of NF-kappaB require the cytoplasmic domain of the receptor but different downstream signaling pathways J. Biol. Chem. 274,19919-19924[Abstract/Free Full Text]
- Miyata, T., Hori, O., Zhang, J., Yan, S. D., Ferran, L., Iida, Y., Schmidt, A. M. (1996) The receptor for advanced glycation end products (RAGE) is a central mediator of the interaction of AGE-beta2microglobulin with human mononuclear phagocytes via an oxidant-sensitive pathway. Implications for the pathogenesis of dialysis-related amyloidosis J. Clin. Invest. 98,1088-1094[Medline]
- Degryse, B., Bonaldi, T., Scaffidi, P., Muller, S., Resnati, M., Sanvito, F., Arrigoni, G., Bianchi, M. E. (2001) The high mobility group (HMG) boxes of the nuclear protein HMG1 induce chemotaxis and cytoskeleton reorganization in rat smooth muscle cells J. Cell Biol. 152,1197-1206[Abstract/Free Full Text]
- Sparatore, B., Pedrazzi, M., Passalacqua, M., Gaggero, D., Patrone, M., Pentremoli, S., Melloni, E. (2002) Stimulation of erythroleukaemia cell differentiation by extracellyular high-mobility group-box protein 1 is independent of the receptor for advanced glycation end-products Biochem. J. 363,529-535[Medline]
- Bianchi, M. E. (1988) Interaction of a protein from rat liver nuclei with cruciform DNA EMBO J. 7,843-849[Medline]
- Salmivirta, M., Rauvala, H., Elenius, K., Jalkanen, M. (1992) Neurite growth-promoting protein (amphoterin, p30) binds syndecan Exp. Cell Res. 200,444-451[Medline]
- Muller, S., Scaffidi, P., Degryse, B., Bonaldi, T., Ronfani, L., Agresti, A., Beltrame, M., Bianchi, M. E. (2001) New EMBO members’ review: the double life of HMGB1 chromatin protein: architectural factor and extracellular signal EMBO J. 16,4337-4340
- Scaffidi, P., Misteli, T., Bianchi, M. E. (2002) The release of chromatin protein HMGB1 by necrotic cells triggers inflammation Nature 418,191-195[Medline]
- Falciola, L., Spada, F., Calogero, S., Längst, G., Voit, R., Grummt, I., Bianchi, M. E. (1997) High mobility group 1 (HMG1) protein is not stably associated with the chromosomes of somatic cells J. Cell Biol. 137,19-26[Abstract/Free Full Text]
- Spada, F., Brunet, A., Mercier, Y., Renard, J. P., Bianchi, M. E., Thompson, E. M. (1998) High mobility group 1 (HMG1) protein in mouse preimplantation embryos Mech. Dev. 76,57-66[Medline]
- Li, M., Carpio, D. F., Zheng, Y., Bruzzo, P., Singh, V., Ouaaz, F., Medzhitov, M. R., Beg, A. A. (2001) An essential role of the NF-B/Toll-like receptor pathway in induction of inflammatory and tissue-repair gene expression by necrotic cells J. Immunol. 166,7128-7135[Abstract/Free Full Text]
- Medzhitov, R. (2001) Toll-like receptors and innate immunity Nat. Rev. Immunol. 1,135-145[Medline]
- Zetterström, C. K., Bergman, T., Rynnel-Dagöö, B., Erlandsson-Harris, H., Söder, O., Andersson, U., Boman, H. G. (2002) High mobility group box chromosomal protein 1 (HMGB1) is an antibacterial factor produced by the human adenoid Pediatr. Res. 52,148-154[Medline]
- Rouhiainen, A., Imai, S., Rauvala, H., Parkkinen, J. (2000) Occurrence of amphoterin (HMG1) as an endogenous protein of human platelets that is exported to the cell surface upon platelet activation Thromb. Haemost. 84,1087-1094[Medline]
- Parkkinen, J., Rauvala, H. (1991) Interactions of plasminogen and tissue plasminogen activator (t-PA) with amphoterin. Enhancement of t-PA-catalyzed plasminogen activation by amphoterin J. Biol. Chem. 266,16730-16735[Abstract/Free Full Text]
- Parkkinen, J., Raulo, E., Merenmies, J., Nolo, R., Kajander, E. O., Baumann, M., Rauvala, H. (1993) Amphoterin, the 30-kDa protein in a family of HMG1-type polypeptides. Enhanced expression in transformed cells, leading edge localization, and interactions with plasminogen activation J. Biol. Chem. 268,19726-19738[Abstract/Free Full Text]
- Gardella S., Andrei, C., Ferrera D., Lotti, L. V., Torrisi, M. R., Bianchi, M. E., Rubartelli, A. (2002) The nuclear protein HMGB1 is secreted by monocytes via a non-classical, vesicle-mediated secretory pathway EMBO Rep. 3,995-1001[Medline]
- Kokkola, R., Sundberg, E., Ulfgren, A.-K., Palmblad, K., Li, J., Wang, H., Ulloa, L., Yang, H., Yan, X.-J., Furie, R., Chiorazzi, N., Tracey, K. J., Andersson, U., Erlandsson-Harris, H. () High mobility group box chromosomal protein 1: A novel proinflammatory mediator in synovitis Arthritis Rheum. 46,2598-2603
This article has been cited by other articles:
|
|
|
|
 
N. Hamada, T. Maeyama, T. Kawaguchi, M. Yoshimi, J. Fukumoto, M. Yamada, S. Yamada, K. Kuwano, and Y. Nakanishi
The Role of High Mobility Group Box1 in Pulmonary Fibrosis
Am. J. Respir. Cell Mol. Biol.,
October 1, 2008;
39(4):
440 - 447.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. Kitahara, Y. Takeishi, M. Harada, T. Niizeki, S. Suzuki, T. Sasaki, M. Ishino, O. Bilim, O. Nakajima, and I. Kubota
High-mobility group box 1 restores cardiac function after myocardial infarction in transgenic mice
Cardiovasc Res,
October 1, 2008;
80(1):
40 - 46.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
L. M. Alleva, A. C. Budd, and I. A. Clark
Systemic Release of High Mobility Group Box 1 Protein during Severe Murine Influenza
J. Immunol.,
July 15, 2008;
181(2):
1454 - 1459.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Y. Sha, J. Zmijewski, Z. Xu, and E. Abraham
HMGB1 Develops Enhanced Proinflammatory Activity by Binding to Cytokines
J. Immunol.,
February 15, 2008;
180(4):
2531 - 2537.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. Ponsuksili, E. Murani, C. Walz, M. Schwerin, and K. Wimmers
Pre- and postnatal hepatic gene expression profiles of two pig breeds differing in body composition: insight into pathways of metabolic regulation
Physiol Genomics,
May 11, 2007;
29(3):
267 - 279.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Yang, Q. Chen, H. Yang, K. J. Tracey, M. Bustin, and J. J. Oppenheim
High mobility group box-1 protein induces the migration and activation of human dendritic cells and acts as an alarmin
J. Leukoc. Biol.,
January 1, 2007;
81(1):
59 - 66.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. W. Bell, W. Jiang, C. F. Reich III, and D. S. Pisetsky
The extracellular release of HMGB1 during apoptotic cell death
Am J Physiol Cell Physiol,
December 1, 2006;
291(6):
C1318 - C1325.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
