|Home | About | Journals | Submit | Contact Us | Français|
Traumatic brain injuries (TBI) are associated with complex inflammatory pathways that lead to the development of secondary injuries such as cerebral ischemia, elevated intracranial pressure, and cognitive deficits. The association between intracellular danger signaling involving nuclear chromatin-binding factor, high mobility group box-1 (HMGB1), and inflammatory pathways following TBI has not yet been fully understood.
To comprehensively review the available literature regarding the potential diagnostic, prognostic and therapeutic use of HMGB1 in TBI.
A systematic literature review of studies available in PubMed using human and animal subjects was performed. A total of eight studies were included in our results.
Comprehensive review of these reports demonstrated that following TBI, HMGB1 is released from damaged neurons and is elevated in patient’s serum and CSF. Furthermore, these studies showed the potential for HMGB1 to serve as a prognostic biomarker and therapeutic target in patients with TBI. Thus, HMGB1 is a prospective candidate for future studies as it shows promise in treating and/or predicting the sequelae of TBI.
Traumatic brain injury (TBI) contributes to approximately 50,000 deaths, 280,000 hospitalizations, and over 2 million emergency department visits, accounting for an estimated 76.5 billion dollars in health care costs in the US annually. TBI is a major public health issue as it contributes to a significant number of deaths and cases of permanent disability, particularly in age groups 0–4 years, 15–19 years, and 65 years and older . The pathophysiology of TBI remains elusive and greater research efforts must be made to investigate the progression of neurodegeneration and the ensuing inflammatory processes.
TBI consists of a primary and secondary injury. The primary injury is characterized as brain damage caused by a dramatic external force that results in structural damage and the disruption of the blood brain barrier (BBB). The initial damage leads to drastic change in the biochemical environment of the brain, leading to neuronal cell death, excitotoxicity, oxidative stress and the upregulation of various anti-inflammatory factors such as interleukin-4, -10 (IL-4, IL-10) and TGF-β and pro-inflammatory factors such as IL-1, IL-6 and tumor necrosis factor alpha (TNF-α) as well as various cytokines and chemokines . The inflammatory response contributes to cerebral edema, elevated intracranial pressure (ICP), ischemia, hematomas, infection, and seizures. These pathological processes comprise the secondary injury. The secondary injury can evolve over time and persist for months following the initial injury, but assuming appropriate treatment, may be reversed or improved. Interventions that inhibit the inflammatory responses may prove to be useful in the treatment of secondary damage and improve patient outcome.
Recent studies have directed attention toward high mobility group box-1 (HMGB1), a protein that propagates inflammation following TBI. Intracellular HMGB1 is a nuclear chromatin-binding factor that promotes protein assembly and binds DNA to facilitate gene transcription. Extracellular HMGB1 exhibits cytokine activity and acts as a prototypical danger associated molecular patterns (DAMP) molecule that stimulates the release of pro-inflammatory factors. Under inflammatory conditions, HMGB1 dissociates from the chromatin, translocates to the cytoplasm and moves into the extracellular space . Following TBI, HMGB1 is actively released by innate immune cells, such as macrophages, dendritic cells, and enterocytes, and is also passively released from injured and necrotic cells . The release of HMGB1 from necrotic cells serves as an endogenous “danger signal” that alerts the immune system to the presence of injured cells  and increases the secretion of neutrophils, monocytes, cytokines, and natural killer cells, and activates microglia, thereby propagating inflammation. Once released into the extracellular milieu, HMGB1 binds to the transmembrane toll-like receptors (TLR) 2 and 4, and the receptor for advanced glycation end products (RAGE) (Yang and Tracey, 2005). Several studies have shown that RAGE  and TLR [7,8] expressions become upregulated following TBI and are major mediators in the inflammatory response. The role of each receptor target in HMGB1 signaling may vary based on the type and severity of the inflammatory event, the surrounding microenvironment and cell type. Activation of RAGE, TLR2, and TLR4 by HMGB1 leads to the phosphorylation of several mitogen-activated protein kinases (MAPKs) that activate the downstream transcription factor nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) and generates an inflammatory cell response. This signaling cascade leads to further release of HMGB1 and ultimately the release of various pro-inflammatory cytokines such as TNF-α, IL-6, 10 and interferon gamma (INF-γ) . These cytokines, in turn, promote further secretion of HMGB1, thereby propagating inflammation . HMGB1-RAGE/TLR signaling plays a central role in inflammatory responses and may contribute to the major secondary complications of TBI that worsen patient outcome. Investigations into such signaling cascades may elucidate the mechanistic link between immune response and pathophysiological effects and introduce methods of therapeutic intervention. In the present study, we systematically review the available literature to assess state of research on HMGB1 in TBI. The current understanding of this molecule in TBI pathophysiology is critically reviewed together with the discussion of its potential for clinical utility.
The National Library of Medicine database was systematically searched using PubMed with the following terms: traumatic brain injury combined with high mobility group box-1 or HMGB1. Titles and abstracts of studies were screened for relevance. The full-text of potentially relevant articles was subsequently retrieved for review and inclusion in the study. Included articles met the following criteria: original research articles in the English language investigating the prognostic, diagnostic, and therapeutic roles of HMGB1 in patients TBI and/or animal models of TBI. Articles were excluded if not written in English, were conference abstracts, or did not use human or animal subjects or samples.
The initial PubMed search returned 85 articles, as shown in Figure 1. After screening of titles, 17 articles remained. Abstract review eliminated 2 studies for not meeting inclusion criteria. Full-text review eliminated 7 studies for not being applicable to the present review: used an inappropriate model (lung pathology following TBI, ischemic injury model, demyelination model) or presented unoriginal data (had overlapping data from included study or was a literature review). After full-text review, eight studies were ultimately included in this review. The final included articles were comprised of four human TBI studies [6,8,11,12] (summarized in Table 1) and six animal TBI studies [6–8,10,13,14] (summarized in Table 2). Two studies [6,8] included both human and animal model components. Results of the included studies are summarized in Table 3.
Four studies, summarized in Table 1, investigated the role of HMGB1 in TBI occurring in human subjects [6,8,11,12]. Study size ranged from 25 to 106 patients with TBI, for a total of 194 TBI subjects. Immunohistochemical examination of HMGB1 expression in brain tissue collected post-mortem from patients with TBI (n = 25) demonstrated immunoreactivity to be either lost or translocalized to the cytoplasm of intra-lesional cells up to 1 day following injury (i.e. patient died <1 day post-TBI). In cases resulting in death between 2–20 days post-TBI, HMGB1 was localized in the cytoplasm of infiltrating phagocytic microglia. HMGB1 in controls (n = 5; no brain disease and non-TBI cause of death) demonstrated a predominantly nuclear immunostaining pattern. Laird and colleagues demonstrated HMGB1 levels in CSF, as determined by Western blotting, to be elevated in adult Patients with TBI (n = 26; Glasgow coma scale (GCS) range: 3T-12T) and minimally expressed in normal pressure hydrocephalus (NPH) controls (argued to be ideal controls due to the significant edema exhibited without an inciting traumatic event). In Patients with TBI, levels of HMGB1 were highest over the first 72 hours and matched the time course of cerebrospinal fluid (CSF) levels of neuron-specific enolase, a cytoplasmic marker specific to neurons. These studies demonstrate HMGB1 to be released from injured neurons following TBI. Further investigations suggest this release of HMGB1 in injury may be exploited for use as a clinical biomarker. Levels of HMGB1 in either CSF or plasma could possibly be used to predict outcome.
Evaluation of ventricular CSF in a study of severe pediatric TBI cases  demonstrated HMGB1 levels to be associated with poor outcome in this cohort. Notably, the temporal profile of HMGB1 levels demonstrated no significant changes (examined at 24h, 48h, 72h, and >72h following injury). However, peak and mean levels were significantly higher than that of lumbar CSF from non-TBI pediatric patients used as controls. Prognostically, increased HMGB1 level was independently associated with Glasgow outcome scale (GOS) at six months, when controlled for initial GOS [odds ratio (OR), 0.49; 95% confidence interval (CI): 0.24–0.97; p = 0.041]. As a predictor of poor outcome (defined as GOS of 1 – 3, i.e. dead, vegetative state, or severe mental disability) at 6 months, HMGB1 levels analyzed using receiver operating characteristic (ROC) curves exhibited an area under the curve (AUC) of 0.70 (95% CI: 0.51–0.90). Accordingly, at a cutoff of 19.49 ng/ml, HMGB1 level was 30.77% sensitive and 100% specific for poor outcome at 6 months.
Wang and colleagues  demonstrated plasma HMGB1 levels to also be a potential prognostic biomarker in Patients with TBI. In the study population of 128 TBI cases, 48 (45.3%) suffered unfavorable outcome (defined as GOS of 1 – 3) while 21 patients (29.2%) died within 1 year. ROC curve analysis demonstrated plasma HMGB1 level >10.8 ng/ml to be predictive of unfavorable outcome of patients with 81.2% sensitivity and 84.5.0% specificity (AUC, 0.880; 95% confidence interval, 0.802–0.935). Plasma HMGB1 was also found to be an independent predictor of mortality in multivariate analysis (OR, 1.641; 95% CI: 1.128–2.726; p = 0.001). Predictive value of HMGB1 levels in 1-year mortality using a concentration cutoff of 11.7 ng/ml exhibited 90.3% sensitivity and 73.3% specificity (AUC, 0.883; 95% CI: 0.806–0.937).
Six studies examined the role of HMGB1 in TBI using animal models[6–8,10,13,14]. These studies utilized either mouse or rat subjects with TBI modeled using a modified Feeney’s weight drop, fluid percussion, or controlled cortical impact. HMGB1 levels were assessed using Western blotting, immunohistochemistry (IHC), enzyme-linked immunosorbent assays (ELISA), electrophoretic mobility shift assays (EMSA), or polymerase chain reaction (PCR). Behavioral tests, assessing cognitive changes post-TBI, included beam-walking balance tests and blinded rotarod tests.
Gao and colleagues  characterized expression of HMGB1 in rat brains. TBI was induced in 55 adult male rats (sham-injury controls, n = 5) using a modified method by Feeney et al.  in which a steel rod was dropped onto exposed parietal lobe producing a focal brain contusion. Western blotting demonstrated HMGB1 levels in contused areas to be significantly depressed from baseline at 6 hours (p < 0.01), followed by a gradual return to normal over the subsequent 2 days. Immunohistochemical analysis of HMGB1 similarly showed expression to have disappeared from the core of the lesion as early as thirty minutes after injury. Cells at the margin of the contused area exhibited translocation of HMGB1 from the nucleus to the cytoplasm. Expression of the receptor for advanced glycation endproducts (RAGE), a receptor for HMGB1, was found to increase from 6 hours post-injury until reaching peak levels at 1 day yet remaining elevated at upwards of 6 days post-injury. Immunoreactivity was primarily exhibited by microglial cells from 2 through 6 days after TBI.
The remaining studies investigated the efficacy of targeting HMGB1 in TBI treatment. One study  of rats with fluid percussion brain injuries administered neutralizing monoclonal antibodies (mAb) raised against HMGB1. Immunostaining and immunoblotting for HMGB1 protein revealed anti-HMGB1 mAb treatment to suppress the disappearance of HMGB1 from the site of injury after TBI. Additionally, mAb-treated animals exhibited a reduced loss of neurons and attenuated blood-brain barrier permeabilization after TBI. Brain edema after percussion injury, as evaluated by T2-weighted magnetic resonance imaging, was dramatic in the untreated TBI cohort. Rats treated with anti-HMGB1 mAb, on the other hand, exhibited minimal amounts of cerebral edema following injury. Lastly, impairment of motor function was assessed by rotarod test. Motor function in anti-HMGB1 mAb-treated animals did not significantly differ from sham operation controls and was significantly improved compared to injured controls.
Two studies[10,13] investigated the therapeutic utility of glycyrrhizin, a natural anti-inflammatory found in large quantities in licorice root that is able to bind HGMB1 and inhibit its cytokine-like signaling. Intravenous glycyrrhizin, administered in a rat TBI model, reduced brain edema and impairment of beam walking performance . Treatment additionally inhibited increased expression of the HMGB1 signaling axis, including receptors RAGE and TLR4, NF-κB DNA binding, and downstream inflammatory cytokines. Immunostaining also demonstrated a decrease in percent of cells positively expressing HMGB1, RAGE, and TLR4 with glycyrrhizin treatment. Separately, glycyrrhizin treatment in rats with fluid percussion-induced TBI  inhibited HMGB1 translocation within intralesional neurons and also maintained HMGB1 immunoreactivity in the nuclei.
Additionally, increase in serum HMGB1 levels was attenuated in treated TBI animals. Measurement of extravasation of serum albumin through the blood-brain barrier demonstrated a dose-dependent inhibition of increased blood-brain barrier permeability post-TBI. Motor impairment, assessed by the rotarod test, was significantly attenuated at 6 and 24 hours post-TBI in glycyrrhizin-treated animals, when compared to vehicle-treated controls. Glycyrrhizin treatment of RAGE −/− gene knockout mice with induced TBI demonstrated none of the protective effects of glycyrrhizin post-TBI.
Laird and colleagues  investigated inhibition of other receptors in the HMGB1 signaling axis in TBI. Using a TBI model in which mice were subjected to controlled cortical impact, antagonism of the NR2B subunit of the N-methyl-D-aspartate receptor (NMDA-R) significantly attenuated cerebral edema 24 hours post-TBI, when compared to sham operation controls. Additionally, this antagonism reversed NMDA-induced excitotoxicity, an essential inciting factor in secondary neurological injury, and reduced extracellular HMGB1 accumulation after NMDA-treatment.
Ethyl pyruvate has also been studied as a neuroprotective therapeutic agent in traumatic brain injury . In a rat model of TBI induced by a modification of the Feeney’s weight drop method, treatment with ethyl pyruvate improved motor function and ameliorated cerebral edema following TBI. Analysis of mRNA and protein levels demonstrated increased HMGB1 and TLR4 expression in the cortex surrounding the contusion of TBI rats, when compared to sham surgery controls. Treatment with ethyl pyruvate, on the other hand, suppressed this increased expression.
In summary, following TBI, HMGB1 is released from damaged and necrotic neurons. While intra-lesional HMGB1 decreases, levels dramatically rise in CSF and serum. This extracellular HMGB1 promotes inflammatory signaling and immune cell recruitment. Inhibition of HMGB1 in animal models attenuates adverse sequelae of TBI, such as cerebral edema, increased blood-brain barrier permeability, and motor function impairment. Targeting of HMGB1 and its signaling pathway has large therapeutic potential for clinical application and warrants further investigation.
The clinical studies cited in this paper demonstrate increased circulating levels of HMGB1 in plasma or CSF of Patients with TBI. Moreover, results suggest that HMGB1 contributes to the pathogenesis of TBI and is associated with poor outcome. HMGB1 shows potential as a biomarker in the treatment of TBI as an indicator of the degree of cell necrosis; a valuable property given that many of the symptoms experienced by Patients with TBI are largely due to neuronal cell death. TBI-induced necrosis is temporally related to immune response and edema formation ; edema and elevated ICP are major adverse events in Patients with TBI and are associated with poor outcome and increased risk of mortality. Increased levels of HMGB1 were found in patients with moderate/severe TBI presenting with elevated ICP . Currently, there is no method to predict elevations in ICP via neurological examination or computed tomography scans, nor are pharmacological interventions available due to lack of mechanistic understanding of the pathology of cerebral edema and ICP. Measurement of biomarkers such as HMGB1 may aid in the prognosis of ICP and would greatly improve patient management.
In addition to serving as a biomarker for TBI, research has looked into the potential of HMGB1 as a treatment target. Secondary TBI damage may be reversible and a substantial therapeutic window exists for intervention. HMGB1 is an attractive drug target because of its delayed kinetics of release and the larger therapeutic window of anti-HMGB1 therapies compared to those that target other inflammatory factors such as TNF or IL-1 β . Laird and group have shown that TLR activation by HMGB1 is related to edema formation and eventual raised ICP. Inhibition of HMGB1 translocation into extracellular space has been shown to attenuate cerebral edema and lower ICP. Animal studies have investigated the potential of ethyl pyruvate, glycyrrhizin, and anti-HMGB1 neutralizing antibody as therapeutics targeted toward HMGB1. Administration of glycyrrhizin, an anti-inflammatory that binds to HMGB1, reduced the upregulation of inflammatory cytokines, inhibited the over-expressions of HMGB1, TLR4, and RAGE, and reduced NF-κB DNA-binding activity . Ethyl pyruvate (EP) is an anti-inflammatory with neuroprotective effects that lessens edema, and improves motor outcome via inhibition of HMGB1/TLR4/NF-κB signaling . These results indicate that blocking the HMGB1 signaling axis leads to improved functional outcomes.
In the included studies, HMGB1 levels were measured via ELISA or Western blot of CSF or ELISA of plasma samples from Patients with TBI. It should be noted that detection of HMGB1 by Western blot or ELISA does not describe the functional activity of the protein. Moreover, the role of HMGB1 in inflammation may change over time; the first 48 hours approximately after trauma or infection, HMGB1 may act as a pro-inflammatory mediator, yet may become biologically inactive or anti-inflammatory in the later disease state . Furthermore, ELISA analysis of CSF does not differentiate between HMGB1 released by necrosis, or from macrophages and monocytes, or a combination of both . Further bench experiments are needed to determine the source of detected HMGB1 as this may heighten the accuracy of HMGB1 as an indicator of necrosis. Additional studies should also measure the ratio of nuclear and cytoplasmic HMGB1 levels to confirm the qualitatively observed translocation of HMGB1 as discussed in the available literature. TBI is a non-static injury whose pathology is driven by an ever-changing stream of biochemical cascades and interactions among various inflammatory mediators. The influx of inflammatory cells is time-dependent; neutrophils are generally the first cell types to recruit to lesion, followed by astrocytes and macrophages.This could be followed by a temporal pattern of, HMGB1 release by these inflammatory cells based on the extent of upregulation of cell type. A proposed mechanism for HMGB1 signaling in TBI pathology is shown in Figure 2. Furthermore, TBI is a heterogeneous injury and the patient outcome is affected by injury severity: mild, moderate or severe; injury type-focal versus diffuse, as well as location and prior history of TBI. Sex, age, race, ethnicity, concomitant illness/injury, resuscitation, hypotension, or body temperature can also potentially factor into pathological effects of TBI . It would be of interest to investigate how injury characteristics and patient factors correlate with the inflammatory processes and HMGB1 release.
Current biomarkers, such as glial fibrillary acidic protein (GFAP) and S100B, suffer from low specificity/sensitivity or low predictive value and in contrast to HMGB1, GFAP and S100B were not predictive of TBI patient outcome . Considering its profound role in inflammation and clear function in TBI pathology, HMGB1 can be used as a prognostic marker. Because it is released during cell necrosis and disperses throughout the CSF, levels in the CSF can be used as a marker. Detection of biomarkers such as HMGB1 in conjunction with the use of neuroimaging to assess injury and recovery  as well as GCS/GOS scores may prove effective in the diagnostic, prognostic and monitoring plan to improve TBI patient outcome.
The pathophysiology behind the complex inflammation cascades secondary to traumatic brain injuries (TBI) is not fully understood. However, resulting injuries and outcomes after TBI have been studied in the past and have suggested HMGB1 to be a major player in disease progression as well as a potential therapeutic target to reduce the injuries and improve outcome following TBI. Our literature review of both human and animal studies suggests an important association between TBI and increased levels of HMGB1 in serum and CSF. In turn, these studies support the potential for HMGB1 to be a target in therapeutic treatment of Patients with TBI. Future studies should be done to further assess the role of HMGB1 in the inflammatory pathways following TBI with the intention of utilizing it as a biomarker in future therapeutic treatments.
Declaration of Interests Statement
This work was supported by research grants R01 HL116042, R01 HL112597, and R01 HL120659 to DK Agrawal from the National Heart, Lung and Blood Institute, NIH USA. The content of this review is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
The authors have no potential conflicts of interest to declare. No writing assistance was used in the production of this work.