| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Departments of Pharmacology and Pathology, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan;
Shigei Medical Institute, Okayama, Japan; and
Department of Anesthesiology and Resuscitology, Ehime University, Graduate School of Medicine, Ehime, Japan
3Correspondence: Department of Pharmacology, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama 700-8558, Japan. E-mail: mbori{at}md.okayama-u.ac.jp
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and iNOS, and suppressed the activity of MMP-9, whereas it had little effect on blood flow. Intracerebroventricular injection of HMGB1 increased the severity of infarction. Immunohistochemical study revealed that HMGB1 immunoreactivity in the cell nuclei decreased or disappeared in the affected areas, suggesting the release of HMGB1 into the extracellular space. These results indicate that HMGB1 plays a critical role in the development of brain infarction through the amplification of plural inflammatory responses in the ischemic region and could be an outstandingly suitable target for the treatment. Intravenous injection of neutralizing anti-HMGB1 mAb provides a novel therapeutic strategy for ischemic stroke.—Liu, K., Mori, S., Takahashi, H. K., Tomono, Y., Wake, H., Kanke, T., Sato, Y., Hiraga, N., Adachi, N., Yoshino, T., Nishibori, M. Anti-high mobility group box 1 monoclonal antibody ameliorates brain infarction induced by transient ischemia in rats.
Key Words: target therapy • inflammation • blood-brain barrier • matrix metalloproteinase • inducible nitric oxide synthase
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. Despite these extensive efforts, however, an effective therapy has not yet been successfully established. For fibrinolytic purpose, recombinant tissue plasminogen activator (rt-PA) has been used clinically; however, the narrow therapeutic window and adverse effects of rt-PA, such as brain hemorrhage and injury, limit its clinical usefulness (3)
.
High mobility group box-1 (HMGB1) is a highly conserved nonhistone nuclear protein that contributes to the architecture of chromatin DNA (4)
. A neurite promoting factor, amphoterin, purified from perinatal rat brain (5
, 6)
, was found to be identical to HMGB1 (6
, 7)
. Recently, HMGB1 was also recognized as a late mediator in septic shock (8
9
10)
as well as a proinflammatory factor (9
, 10)
. The cytokine profile of HMGB1 has shed new light on the role of nuclear proteins and promoted the studies on roles of this unique factor in different disease conditions that are accompanied by a variety of inflammatory responses (11
12
13
14
15)
. In the central nervous system, HMGB1 was shown to inhibit glutamate transport by glial glutamate-aspartate transporter 1 using the mouse glial membrane preparation (16)
, suggesting a contribution of HMGB1 to the elevation of excitotoxic glutamate in ischemic brain. Moreover, HMGB1 has affinity for t-PA and accelerates its proteolytic activity (7)
. Since therapeutically injected (17
18
19)
and endogenous t-PA (20)
have been implicated in the activation of matrix metalloproteinase-9 (MMP-9) during ischemic insult, HMGB1 may enhance the disruption of BBB structure through activation of MMP via t-PA.
In the present study, we demonstrated that anti-HMGB1 mAb dramatically ameliorated infarction after MCA occlusion in rats, and this effect was associated with a marked reduction in neurological deficits. The treatment resulted in the inhibition of increased BBB permeability, the activation of microglia, and the production of proinflammatory molecules but it did not improve blood flow. This treatment offers a novel therapeutic strategy for ischemic stroke.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The rats were anesthetized with 2% halothane in a mixture of 50% N2O and 50% O2 using a face mask. The root of the right MCA was occluded by insertion of a silicone-coated 4.0 nylon thread from the bifurcation of the internal and external carotid arteries. The tip of the thread was placed 18 mm distal from the bifurcation. The thermocouple needle probe was inserted into temporal muscle to maintain the temperature at 37.0 ± 0.1°C with a heating lamp during surgery. After the surgical incision was sutured, the rats were allowed to recover from anesthesia. All rats showed paralysis of the contralateral limbs after recovery from anesthesia. The rats were anesthetized again 5 min before reperfusion. After the skin was reopened, the cerebral blood flow was restored 2 h after MCA occlusion by pulling the thread by 5 mm. Anti-HMGB1 mAb (200 µg) or class-matched control mAb (IgG2a) against Keyhole Limpet hemocyanin was administered i.v. immediately and 6 h after reperfusion. For i.c.v. injection of HMGB1 (5 µg in 5 µl of 10 mM Tris-buffered saline, pH 7.4), the injection cannula was inserted into the right ventricle just after reperfusion under anesthesia. HMGB1 was immunoaffinity-purified from HMGB1/HMGB2 mixture (Wako, Osaka, Japan). Body temperature in control and anti-HMGB1-treated rats was within 37.2 ± 0.2°C throughout the experiments and no difference was found between 2 groups.
Evaluation of neurological deficits
Neurological deficits were evaluated by two methods. In the rota-rod test, rats were conditioned for 3 days before MCA occlusion on an accelerating rota-rod cylinder at 5 to 15 rpm. Rats that could stay on the rotating rod at 15 rpm for 180 s were subjected to MCA occlusion. After MCA occlusion, each rat was subjected to trials conducted at 3 different speeds (5, 10 and 15 rpm), and the mean duration of three trials at each speed on the rota-rod was recorded. The other test was a neurological scoring method, essentially as described by Bederson et al. (22)
. The signs of hemiparesis and postural reflex were analyzed as follows: 1) Forelimb flexion: rats were held by the tail on the flat surface. Paralysis of forelimbs was evaluated by the degree of flexion. 2) Hind-limb paralysis: both hind limbs were pulled caudally, and the difference in resistance was measured. 3) Torso twisting: rats were held by the tail on a flat surface, and the degree of body rotation was evaluated. 4) Lateral push: rats were pushed laterally, and resistance was evaluated. 5)Posture: the slope of the body axis was evaluated. The scores were categorized according to four grades (0, normal; 1, moderate; 2, considerable; 3, severe). The neurological evaluation was performed by an investigator blind to the treatment condition.
Triphenyltetrazolium chloride (TTC) staining
Coronal sections of the brain 2 mm thick were incubated with 2% TTC at 37°C for 30 min with gentle shaking and then were fixed with 10% formalin in PBS. The stained slices were photographed, and the size of the infarct was quantified using NIH image software.
Histological study
For the histological study, rats were anesthetized with an i.p. injection of sodium pentobarbital at a dose of 50 mg/kg and perfused through the left ventricle with 100 ml of saline, followed by 200 ml of 10% formalin in 0.01 M phosphate-buffered saline. The fixed brain was embedded in paraffin, and sections 6 µm thick were used for histological or immunohistochemical study. Anti-HMGB1 mAb was labeled with horseradish peroxidase and used for immunohistochemical staining. Griffonia Simplicifolia IB4 lectin-HRP conjugate was obtained from EY Laboratories (San Mateo, CA, USA) and anti-CD68 Ab was from R&D (San Diego, CA, USA). Immunohistochemical staining was performed using a Ventana system (Ventana Japan, Osaka, Japan). To determine the number of infiltrating neutrophils in coronal sections (1.7 mm and 0.7 mm rostral and 0.3 mm caudal to the bregma), cryostat sections were stained with antimyeloperoxidase antibody as described previously (21)
. The number of neutrophils was calculated as the number of myeloperoxidase-positive cells minus the number of CD68-positive cells.
Measurement of BBB leakage
To determine changes in vascular permeability, Evans blue dye (2% in saline, 2 ml/kg; Wako, Japan) was administered to rats via the tail vein immediately after the treatment with antibody. After 3 h, the rats were perfused via the left ventricle with 150 ml of saline, and the brains were removed, weighed, and dissected into hemispheres after removal of the brain stem and cerebellum. The hypothalamus and the remaining brain of 2 mm thick from rostral (no. 1) to caudal (no. 5) were dissected in each hemisphere, homogenized in 1 M KOH, and kept overnight. Brain homogenate was then mixed with 0.6 M H3PO4-acetone (5:13) at a 1:9 ratio. After centrifugation, the supernatant was discarded, and 50% trichloroacetic acid was added to the pellet to extract Evans blue. After centrifugation, the resultant supernatant was diluted with ethanol (1:3), and the dye concentration was determined spectrophotometrically at 620 nm. The tissue concentration of Evans blue was quantified from a standard curve that was derived from a known amount of the dye mixed with brain homogenate, and was normalized to tissue weight.
Zymography
Gelatin zymography was performed as follows. Briefly, the brain hemisphere was homogenized with 0.1 M Tris-HCl, pH 7.4, containing 5 mM CaCl2 and 0.05% Brij-35, and the homogenate was centrifuged at 10,000 g for 30 min. The resultant supernatant (40 µg protein) was mixed with 100 µl of gelatin-agarose for 1 h at 4°C with gentle shaking. The agarose beads were washed twice with 800 µl of 50 mM Tris-buffered saline, and then the bound proteins were eluted with 10% DMSO in PBS. The eluate was mixed with the sample buffer to give a final concentration of 2.5% glycerol, 5% SDS, and 125 mM Tris-HCl (pH 6.8) in the absence of reducing agent, and incubated at 25°C for 10 min. Protein samples were then electrophoresed on 10% SDS-polyacrylamide gel containing 1 mg/ml gelatin. After electrophoresis, the gels were incubated twice in 2.5% Triton X-100 at room temperature for 30 min to remove SDS from the gel. The gels were next incubated in 0.05 M Tris-buffered saline (pH 7.4) containing 5 mM CaCl2, 1 µM ZnCl2, and 0.035% Brij-35 for 10 min at room temperature and then sandwiched between 3 pieces of Whatman 3MM chromatography paper that had been saturated with buffer. The reaction was continued for 40 h at 37°C, and the gels were subsequently stained with Coomassie Brilliant Blue R-250 and appropriately destained.
Determination of cerebral blood flow
Cerebral blood flow (CBF) was measured on-line using a flexible-fiber probe linked to a Doppler blood flow device TBF-LC1 (Unique Medical Inc., Osaka, Japan). The scalp was incised at the midline and the skull was exposed. Probes were attached to the skull surface on both lateral sides 7 mm inferior from the bregma. CBF signals from both hemispheres were simultaneously monitored throughout the experiments while anesthesia was maintained with 2% halothane in a mixture of 50% N2O and 50% O2.
RT-PCR
The rats were treated with 50 mg/kg sodium pentobarbital (i.p.) 6 h after reperfusion and then perfused transcardially with 50 ml of ice-cold saline before decapitation. The striatum and cerebral cortex corresponding to the core infarct were dissected. Total RNA was isolated with BioRobot EZ1 (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions. Total RNA (1 µg) was reverse-transcribed with Avian Myeloblastosis virus reverse transcriptase XL and Oligo dT-primers (Takara RNA PCR Kit; Takara Biomedicals, Shiga, Japan). PCR was performed using Takara Ex taq HS DNA polymerase (Takara). The sense and antisense primers used for the analysis of expression of HIF-1, VEGF, iNOS, MMP-2, MMP-9 and β-actin were as follows. HIF-1: 5'-GCAGCACGATCTCGGCGAAGCAAA-3' and 5'-GCACCATAACAAAGCCATCCAGGG-3' (235 bp). VEGF: 5'-GCTCTCTTG GGTGCACTGGA-3' and 5'-CTTCTCTTGGGTGCACTGGA-3' (635 bp). iNOS: 5'-GCATCCCAAGTACGAGTGGT-3' and 5'-GAAGGCGTAGCTGAACAAGG-3' (700 bp). MMP-2: 5'-GATCTGCAAGCAAGACATTGTCTT-3' and 5'-GCCAAATAAACCGATCCTTGAA-3' (77 bp). MMP-9: 5'-GTAACCCTGGT CACCGGACTT-3' and 5'-ATACGTTCCCGGCTGATCAG-3' (80 bp). β-actin: 5'-CAGAGCAAGAGAGGCATCCT-3' and 5'-GGCAGCTCATAGCTCTTCTC-3' (535 bp). The annealing temperature was 55°C for all primer pairs.
Real-time PCR
Real-time PCR was performed with the SYBR Premix EX Taq (Takara) in a Light Cycler instrument (Roche) according to the manufacturer’s instructions. The sense and antisense primers used for the analysis of the expression of iNOS, TNF-, and β-actin were the following. Inducible NOS: 5'-GCATCCCAAGTACGAGTGGT-3' and 5'-GAAGTCTCGGACTCCAATCTC-3' (154 bp). TNF-: 5'-GCCCAGACCCTCACACTC-3' and 5'-CCACTCCAGCTGCTCCTCT-3' (99 bp). β-Actin: 5'-CCCGCGAGTACAACCTTCT-3' and 5'-CGTCATCCATGGCGAACT-3' (72 bp). The expression of β-actin was used to normalize cDNA levels. The PCR products were analyzed by a melting curve to ascertain the specificity of amplification.
Production of HMGB1-specific monoclonal antibody
Rats were immunized with HMGB1/HMGB2 (Wako, Osaka, Japan) emulsified with Freund’s complete adjuvant and the booster injection with incomplete adjuvant was administered 3 wk later. After confirming elevation of anti-HMGB Ab levels, hybridomas were produced as described previously (23)
. The epitope recognized by each monoclonal antibody was determined by dot blotting using synthetic overlapping peptides derived from human HMGB1 sequence 15 amino acids in length. The clone (#10–22, subclass IgG2a) recognizing the C-terminal sequence of the HMGB1 molecule (DEDEEEE) and specific for HMGB1 but not for HMGB2 was used for the experiments. A monoclonal antibody (IgG2a subclass) against Keyhole Limpet hemocyanin was produced in the same way and was used as the control antibody.
Statistics
Statistical significance was evaluated by ANOVA followed by Dunnett’s test for multiple comparison, or by Student’s t test for comparisons between 2 groups. P values less than 0.05 were considered to be significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A and B show that mAb (#10–22, subclass IgG2a) recognized the C-terminal peptide sequence no. 41 (EDEEDEDEEEDDDE) of the HMGB1 molecule that was specific for HMGB1 but not for HMGB2. Since no. 40 peptide (DEEEEEDEEDEEE) was not recognized by mAb (#10–22) at all, the epitope was estimated to reside in the latter half of peptide no. 41. The other two mAbs (#4–1 and #11–19) recognized no. 30 and no. 31 peptides, indicating that the epitopes were overlapping sequence of both peptides (LKEKYEKDIA) that was also included in HMGB2. Therefore, we used mAb (#10–22) in the following experiments. The mAb detected a single band of 29 kDa on Western blotting using rat brain (Fig. 1C
). HMGB1 concentration-dependently (0.01–10 µg/ml) stimulated the expression of ICAM-1 on the surface of CD-14-positive monocytes in human peripheral mononuclear cell culture when the expression was determined 24 h after the start of stimulation (Fig. 1D
), as shown by FACS analysis. Addition of the mAb to culture medium concentration-dependently (0.1–100 µg/ml) inhibited the up-regulation of monocyte ICAM-1 expression induced by HMGB1 (10 µg/ml) (Fig. 1D
). Thus, anti-HMGB1 mAb (#10–22) was demonstrated to be a neutralizing activity for HMGB1.
|
Effects of Anti-HMGB1 mAb on brain infarction
Middle cerebral artery (MCA) occlusion for 2 h in rats produced massive infarction in the fronto-parieto-temporal regions of the cerebral cortex as well as the striatum in the ischemic hemisphere when the infarct areas were determined with tritetrazolium chloride (TTC) staining 24 h after reperfusion (Fig. 2
A). Treatment with i.v. injection of anti-HMGB1 mAb (200 µg) twice (immediately and 6 h after reperfusion) reduced the infarct volumes dramatically (Fig. 2A
). The infarction volumes were quantified in the cerebral cortex and the striatum, respectively, and the results are summarized in Fig. 2B
. Anti-HMGB1 mAb reduced the infarction volumes in the cerebral cortex and the striatum equally, leading to the 90% and 75% reduction at 24 and 48 h, respectively.
|
Consistent with the reduction in infarct volumes, treatment with i.v. injection of anti-HMGB1 mAb markedly improved the neurological deficits observed on the rota-rod test (Fig. 2C
). In particular, mAb treatment significantly ameliorated the deficits observed in control rats under the high-speed load (15 rpm). Moreover, high neurological deficit scores in the control rats persisted for up to 48 h; however, the scores in the anti-HMGB1 mAb treated-group time-dependently decreased and reached a considerably lower level at 48 h (Fig. 2D
). These findings as a whole indicated that anti-HMGB1 mAb dramatically ameliorated infarct formation and improved the accompanying neurological symptoms significantly. Furthermore, we found that intracerebroventricular (i.c.v.) injection of HMGB1 (5 µg) immediately after reperfusion aggravated infarction and exacerbated the neurological deficit induced by 1.5-h occlusion of the MCA (Fig. 3
). These findings were consistent with the effects of neutralizing anti-HMGB1 mAb.
|
Histological studies on the effects of anti-HMGB1 mAb
Hematoxylin-eosin staining of brain sections from the control MCA-occlusion rats showed that diffuse neuronal cell death occurred 12 h after reperfusion in the cerebral cortex and the striatum regions, and thereafter these regions were designated as the infarct areas (Fig. 4
). In contrast, these areas appeared to be intact in rats treated with anti-HMGB1 mAb (Fig. 4)
. Immunohistochemical studies revealed the presence of HMGB1 immunoreactivity in the nuclei of brain cells in the nonischemic side (Fig. 5
A). The immunoreactive nuclei were widespread, but not all (arrowheads in Fig. 5A
). The distribution of HMGB1 immunoreactivity in the intact rat brain was similar to that observed in the nonischemic side (data not shown).
|
|
The cytoplasm and neuronal processes were also positive in some neurons. In the ischemic side, strongly immunoreactive nuclei disappeared from the striatum and were less prominent in the cerebral cortex 12 h after reperfusion (Fig. 5A
). These changes in the distribution of HMGB1 immunoreactivity after ischemia strongly suggested that HMGB1 was translocated from nuclear stores to the extracellular space. Conversely, the immunoreactivity in the nerve fibers in the internal capsule appeared to be enhanced in the ischemic hemisphere as compared with the contralateral side (Fig. 5C
). The paraventricular nucleus and anterior hypothalamic area on the ischemic side were strongly immunoreactive with diffuse and moderate staining in the surrounding areas, whereas staining was weak on the nonischemic side (Fig. 5B
). There were also strongly immunoreactive neurons scattered in the reticular thalamic nucleus on the ischemic side (Fig. 5C
). The wall of the internal carotid artery was strongly immunoreactive on the ipsilateral hemisphere (Fig. 5B
), while no such immunoreactivity was observed on the contralateral side. The treatment with anti-HMGB1 mAb substantially diminished the changes observed in the ischemic hemisphere in control rats (Fig. 5A
–C). These results indicated that neutralization of extracellular HMGB1 by the administration of a specific Ab inhibited the disappearance of HMGB1 immunoreactivity from nuclei and the up-regulation of HMGB1 in specific neurons in the ischemic brain hemisphere.
Analysis on the action mechanism of anti-HMGB1 mAb
Leakage of Evans blue dye during 3 h after the reperfusion was remarkable in the ipsilateral hemisphere and hypothalamus of control rats, whereas Evans blue leakage in the contralateral hemisphere was much lower (Fig. 6
A). There was no detectable leakage of Evans blue in the hemispheres of sham-operated rats (Fig. 6A
). Treatment with anti-HMGB1 mAb significantly reduced the extravasation of Evans blue dye in the ischemic hemisphere including the hypothalamus (Fig. 6A, B
).
|
In coronal sections, high leakage of Evans blue dye was observed in the striatum and parietal and temporal cortex, which was consistent with the infarct areas determined 1 day after reperfusion (Figs. 2A
, 6A)
. Treatment with anti-HMGB1 mAb appeared to inhibit the vascular leakage evenly throughout all brain regions (Fig. 6A, B
).
The MMP activities determined by gelatin zymography clearly showed that pro-MMP-9 band was increased significantly in the ischemic hemisphere of control rats, whereas it was increased only minimally in anti-HMGB1 mAb-treated rats (Fig. 6C
). MMP-2 activity was detected constitutively in both hemispheres and was not affected by ischemic insult (Fig. 6C
).
Doppler flowmetry measurements showed that blood flow on the ipsilateral side of the temporal cortex in MCA-occluded rats immediately after occlusion was decreased to 24% of the preoccluded level (Fig. 7
). By the end of the 2-h occlusion period, blood flow gradually recovered to 40% of the preoccluded level. Reperfusion slightly accelerated the recovery of blood flow. In the contralateral side, a change in blood flow was minimal. Treatment with anti-HMGB1 mAb (200 µg) did not affect the change in blood flow induced by MCA occlusion in either side of the cerebral cortex (Fig. 7)
.
|
Inhibition of microglia activation by anti-HMGB1 mAb
Activated microglia detected by Griffonia simplicifolia (IB4)-lectin were observed in the cerebral cortex, striatum, and hypothalamus in the ischemic side of the control rat brain (Fig. 8
A–D); however, there were few such microglia in the ischemic side of the brain in anti-HMGB1 mAb-treated rats (Fig. 8E
). However, no difference was seen between control and anti-HMGB1 mAb-treated rats with respect to the number of neutrophil infiltrates in the ischemic side of the striatum and remaining brain tissue 12 h after reperfusion (Fig. 8F
). There were few neutrophils in the contralateral side of the brain in the control and anti-HMGB1 mAb-treated groups.
|
Anti-HMGB1 mAb inhibits the expression of TNF- and iNOS
The expression of mRNA coding for inflammation-related molecules, including inflammatory cytokines, was determined by RT-PCR. MCA occlusion significantly up-regulated the expression of TNF- and iNOS in the cerebral cortex and striatum of the ischemic hemisphere as compared with the nonischemic side (Fig. 9
). Treatment with anti-HMGB1 mAb almost completely inhibited the expression of TNF- and iNOS in the cerebral cortex and striatum of the ischemic hemisphere (Fig. 9)
. In contrast, the ischemia-induced changes in the expression of HIF-1, IL-18, COX-2, MMP-2, and MMP-9 were marginal and anti-HMGB1 mAb had no apparent effect on the expression of these mRNAs. The quantification of iNOS and TNF- expression using real-time PCR confirmed marked increases in the expression of iNOS and TNF- in the ischemic cerebral cortex and striatum compared with the contralateral side in the control rats whereas the increases were not significant in the anti-HMGB1-treated rats (Fig. 10
). The treatment with anti-HMGB1 mAb significantly inhibited the TNF- expression in the ischemic cerebral cortex and striatum (Fig. 10)
.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 24)
contributes to disruption of the blood-brain barrier, leading to aggravation of brain infarction. Therefore, HMGB1 may play a key role in the development of brain injury under hypoxic and ischemic conditions by disrupting microvascular structure and neuron-matrix interactions. In addition, a recent study (16)
showed that HMGB1 inhibits glial glutamate transport by GLAST in mouse gliosomes and suggested that HMGB1 can increase extracellular glutamate levels in vivo. Thus, it is possible that anti-HMGB1 mAb also prevented the elevation of glutamate levels that leads to neuronal excitotoxicity.
Recent studies in various models of systemic inflammation have indicated that HMGB1 has proinflammatory cytokine activity. HMGB1 has been identified as a late mediator of septic shock (8)
. Anti-HMGB1 mAb has been shown to decrease the lethality induced by LPS (8)
and cecal ligation puncture (25)
in mice, even though the mAb was administered after the induction of disease. There is mounting evidence that HMGB1 may be involved in other inflammatory diseases including acute lung injury (14)
, hepatic injury (15)
, and rheumatoid arthritis (12)
. In monocytes/macrophages, HMGB1 has also been reported to stimulate the production of IL-1/β, TNF-, IL-6, IL-8, and MIP-1/β (26)
; induce iNOS (27)
; and stimulate monocyte chemotaxis (28)
. In the present study, we demonstrated that ischemia-induced up-regulation of iNOS and TNF- was inhibited by the neutralization of HMGB1. The induction of iNOS (29)
and TNF- (30)
following an ischemic insult was reported to occur mainly in microglia. Thus, it is likely that HMGB1 activates microglia in the brain, leading to the up-regulation of iNOS and TNF- expression. In our study, the detection of activated microglia by IB4-lectin in the ischemic hemisphere supported this notion. Taken together, this evidence suggests that HMGB1 may play a key role in the development of brain injury under hypoxic and ischemic conditions by disrupting the structure of the microvasculature and by activating microglia to trigger the excess production of TNF- and NO-derived radicals, which can lead to further development of the inflammatory response and exacerbation of neuronal injury. In fact, the induction of iNOS (31)
, TNF- (32)
and MMP-9 (18
, 24)
has been reported to be involved in the inflammatory response and disruption of the blood-brain barrier, leading to aggravation of brain infarction. The regulation of any one of these factors has been postulated to reduce ischemic injury (24
, 31
, 32)
. Therefore, it is reasonable that anti-HMGB1 mAb, which has the ability to substantially reduce the expression of all three proinflammatory factors, exerted profound therapeutic effects on brain infarction.
Recently, Kim et al. (33)
reported that microinjection of short hairpin RNAi for HMGB1 into the striatum inhibited neuronal death in the transfected area after MCA occlusion in rats. They also showed that the RNAi inhibited the expression of TNF- and iNOS, in addition to IL-1β and COX-2, in the restricted area 24 h after MCA occlusion. Although their intervention modulated the expression of HMGB1, it was not possible to determine whether intracellular or extracellular HMGB1 was a more important mediator of brain injury. Moreover, the effect of hairpin RNAi was limited in the restricted micromilieu and, therefore, the evaluation of neurological deficits in the treated rats could not be performed (33)
. However, our findings on the expression of TNF- and iNOS were consistent with their results.
There are two main sources of extracellular HMGB1 (34
, 35)
. The first is released from the nuclei of necrotic cells and the second is secreted from cells of the monocyte/macrophage lineage in an activation-dependent manner. The lack of HMGB1-immunoreactivity in the nuclei of cells in necrotic areas supported that HMGB1 may also be released from necrotic cells in the brain. Also, based on the observation of the diffuse immunoreactive staining for HMGB1 in the paraventricular and anterior hypothalamic areas, it seemed that HMGB1 might be released from surviving neurons without inducing necrosis. Under such conditions, HMGB1 may bind to extracellular matrix components, i.e., neurocan and phosphacan (36)
, which have been reported to have high affinity for HMGB1. Thus, it is possible that the release of HMGB1 from different types of cells into the extracellular environment occurred during the ischemic insult in the ipsilateral hemisphere. Interestingly, treatment with anti-HMGB1 mAb also inhibited the ischemia-induced changes in the distribution pattern of HMGB1 immunoreactivity, suggesting that HMGB1 may be involved in the control of its own translocation and redistribution in the ischemic brain.
HMGB1 has plural receptors, RAGE (37)
and TLR-2/TLR-4 (38
, 39)
. It is not clear at this moment which receptor signaling system is most important for ischemic brain injury. The number of infiltrating neutrophils into the ischemic hemisphere 12 h after reperfusion did not differ between the control and anti-HMGB1-treated groups, suggesting that infiltrated neutrophil in the early phase of brain ischemia did not play a significant role in the development of the inflammatory response.
The present results strongly suggested that HMGB1 plays a crucial role in inducing massive infarction by activating multiple injurious cascades after an ischemic insult due to arterial thrombus or embolus. It should be pointed out that anti-HMGB1 mAb had little effect on blood flow to the ischemic regions. Therefore, blocking the deleterious effects of HMGB1 on ischemic stroke in humans might be effective for rescuing neurons and dramatically reducing the resultant infarct size without restoring blood flow to the ischemic regions. In other words, neurons may be much more resistant to hypoxic and ischemic insults if the activity or release of HMGB1 is minimized. Anti-HMGB1 mAb, with the ability to neutralize HMGB1, prevented the development of brain infarction induced by MCA occlusion by inhibiting very early inflammatory responses and possibly by inhibiting neurotoxicity due to elevated glutamate levels. It is noteworthy that for the treatment of acute brain infarction, anti-HMGB1 mAb had higher efficacy among several types of drugs, including FK506 (40)
, MMP inhibitors (41
, 42)
, radical scavengers (43)
, endothelin type A receptor antagonists (44)
, glutamate receptor antagonists (45)
, anti-4 integrin Abs (46)
, proteasome inhibitors (47)
, and PPAR-
inhibitors (48)
. The finding that i.c.v. injection of HMGB1 aggravated infarction was consistent with the effects of anti-HMGB1 mAb and implies the functional role of endogenous HMGB1. Concerning the accessibility of mAb to ischemic regions, it is likely that the mAb can be sufficiently delivered to the BBB-disrupted ischemic area spatially and temporally in the early phase of brain ischemia, resulting in the marked suppression of vascular permeability in the following phase. Therefore, anti-HMGB1 mAb appears to be a good therapeutic candidate for the treatment of ischemic stroke.
The clear demonstration of the effectiveness of anti-HMGB1 mAb against ischemic stroke beyond diverse types of candidates, strongly indicated that this treatment has opened a novel strategy for the treatment of ischemic stroke in humans. Monoclonal Abs constitute a new group of drugs, and these targeted therapies are expected to have profound potential for the treatment of cancer (49)
and inflammatory diseases (50)
. To promote the clinical application, the development of human mAbs against HMGB1 and the evaluation of their efficacy in animal models are important objectives for future research.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Current address: Shujitsu University, School of Pharmacy, Okayama, Japan.
Received for publication April 9, 2007. Accepted for publication June 14, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Kikuchi, K.-i. Kawahara, S. Tancharoen, F. Matsuda, Y. Morimoto, T. Ito, K. K. Biswas, K. Takenouchi, N. Miura, Y. Oyama, et al. The Free Radical Scavenger Edaravone Rescues Rats from Cerebral Infarction by Attenuating the Release of High-Mobility Group Box-1 in Neuronal Cells J. Pharmacol. Exp. Ther., June 1, 2009; 329(3): 865 - 874. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Watanabe, H. Keino, Y. Sato, A. Kudo, H. Kawakami, and A. A. Okada High Mobility Group Box Protein-1 in Experimental Autoimmune Uveoretinitis Invest. Ophthalmol. Vis. Sci., May 1, 2009; 50(5): 2283 - 2290. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-S. Wu, W.-M. Cheung, Y.-S. Tsai, Y.-T. Chen, W.-H. Fong, H.-D. Tsai, Y.-C. Chen, J.-Y. Liou, S.-K. Shyue, J.-J. Chen, et al. Ligand-Activated Peroxisome Proliferator-Activated Receptor-{gamma} Protects Against Ischemic Cerebral Infarction and Neuronal Apoptosis by 14-3-3{epsilon} Upregulation Circulation, March 3, 2009; 119(8): 1124 - 1134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Muhammad, W. Barakat, S. Stoyanov, S. Murikinati, H. Yang, K. J. Tracey, M. Bendszus, G. Rossetti, P. P. Nawroth, A. Bierhaus, et al. The HMGB1 Receptor RAGE Mediates Ischemic Brain Damage J. Neurosci., November 12, 2008; 28(46): 12023 - 12031. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
