- Email: [email protected]
HMGB1 expression patterns during the progression of experimental autoimmune encephalomyelitis
Journal of Neuroimmunology 280 (2015) 29–35
Contents lists available at ScienceDirect
Journal of Neuroimmunology journal homepage: www.elsevier.com/locate/jneuroim
HMGB1 expression patterns during the progression of experimental autoimmune encephalomyelitis Yan Sun a,b, Huoying Chen a, Jiapei Dai b, Huijuan Zou a, Ming Gao a, Hao Wu a, Bingxia Ming a, Lin Lai a, Yifan Xiao a, Ping Xiong a, Yong Xu a, Feili Gong a, Fang Zheng a,⁎ a b
Department of Immunology, School of Basic Medicine, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China Wuhan Institute for Neuroscience and Neuroengineering, South-Central University for Nationalities, Wuhan 430074, China
a r t i c l e
i n f o
Article history: Received 8 January 2015 Received in revised form 9 February 2015 Accepted 23 February 2015 Keywords: High mobility group box 1 Experimental autoimmune encephalomyelitis Astrocyte Microglia Neuron
a b s t r a c t High mobility group box 1 (HMGB1), a nonhistone chromatin associated protein, plays different roles according to the expression pattern such as the amount, cell location and sub-cellular location. It has been recently demonstrated that the systemic HMGB1 is associated with autoimmune encephalomyelitis. However, the dynamic change of HMGB1 expression pattern in spinal cords that may be involved in the progression of disease is not fully understood. In this study, the amount, cell location and subcellular location of HMGB1 in adult mice spinal cords during various stages of experimental autoimmune encephalomyelitis (EAE) are investigated. HMGB1 is expressed in the nuclei of spinal cord resident cells such as some astrocytes, microglia and a few neurons in normal situation. During EAE progression, the total and extracellular HMGB1 in the spinal cord are increased, more HMGB1 positive astrocytes and microglia are observed, and the intra-neurons HMGB1 in the ventral horn and around the central canal localize majorly in the cytoplasm accompanied by the increasing extracellular HMGB1. Blockade of HMGB1 in central nervous system (CNS) locally attenuates the severity of EAE significantly. Our findings indicate that the HMGB1 expression pattern in the spinal cord is associated with the progression of EAE. HMGB1 may be a potential target for autoimmune encephalomyelitis (multiple sclerosis in human) therapy. © 2015 Elsevier B.V. All rights reserved.
1. Introduction High mobility group box 1 (HMGB1) is found initially as a nonhistone DNA-binding nuclear protein (Goodwin et al., 1973) in almost all eukaryotic cells (Matsuoka et al., 2010). It promotes chromatin function and gene regulation (Agresti et al., 2003). Now, a lot of studies have demonstrated that HMGB1 exerts different biological functions depending on its distribution and cellular localization. For example, upon proinflammatory mediators stimulation or cell damage, nuclear HMGB1 can be transported into the cytoplasm or released into extracellular space for the regulation of immunity and inflammation (Kim et al., 2009; Lotze and Tracey, 2005). Extracellular HMGB1 acting as a damage associated molecular pattern (DAMP) can induce the release of proinflammatory cytokines including IL-1β, TNF-α, IL-6 (Andersson et al., 2000) and chemokines (Rouhiainen et al., 2004) via binding with the receptor for advanced glycation end products (RAGE) or toll-like receptors (TLRs), triggering sterile inflammation in various diseases, such as arthritis, brain ischemia and traumatic brain injury (Laird et al., 2014; Pisetsky et al., 2008; Qiu et al., 2008). The administration of neutralizing anti-HMGB1 monoclonal and polyclonal antibodies (mAb and pAb), and other HMGB1 antagonists (e.g., glycyrrhizin) dampens inflammation ⁎ Corresponding author. E-mail address: [email protected] (F. Zheng).
http://dx.doi.org/10.1016/j.jneuroim.2015.02.005 0165-5728/© 2015 Elsevier B.V. All rights reserved.
(Andersson and Erlandsson-Harris, 2004; Kim et al., 2006; Yang et al., 2004), suggesting that HMGB1 could be considered as a therapeutic target. The central nervous system (CNS) has a highly specialized microenvironment. Because of the blood–brain barrier (BBB) that limits entry of immune cells and molecules, glial cells residing in the CNS are essential for the maintenance of organism homeostasis in normal situation (Neumann, 2001; Streit, 2002). Under pathological conditions, infiltrated immune cells and glial cells contribute to the disorder of CNS via inducing immune deregulation and neuroinflammation mediated by direct killing and releasing neurotoxic substances, free radicals and inflammatory mediators (Benn et al., 2001; Benveniste, 1997; Kroncke et al., 1998; Schwartz et al., 1998). Although HMGB1 protein is present in some subsets of CNS resident cells during development, with a very complex temporal, spatial and subcellular expression pattern (Guazzi et al., 2003), the detailed expression pattern of HMGB1 is not clear in the different types of neural cells including neurons, astrocytes, and microglia in the adult mouse CNS. The dynamic change of HMGB1 in inflammatory CNS is also needed to be addressed. Multiple sclerosis (MS) is an autoimmune-mediated inflammatory disease of CNS characterized by demyelination and axonal damage (Compston and Coles, 2008). Experimental autoimmune encephalomyelitis (EAE) is the most reliable experimental model of MS (Miller et al., 2010). Increased HMGB1 levels were found in the cerebrospinal fluid
30
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
(CSF) of MS patients (Andersson et al., 2008), and systemic treatment with neutralizing HMGB1 antibody can rescue mice from EAE (Robinson et al., 2013; Uzawa et al., 2013). These facts suggest that HMGB1 plays critical roles in MS/EAE. However, the proportion of CNS derived HMGB1 in the development of MS/EAE remains poorly understood. In the present study, we discover HMGB1 is nuclearly expressed in some astrocytes, microglia and a few of the neurons in adult normal mouse spinal cords. During EAE progression, the expression and release of HMGB1 in the spinal cord are increased, more astrocytes and microglia express HMGB1, and HMGB1 is detectable in the cytoplasm of neurons in the ventral horn and around the central canal. Blockade of HMGB1 locally in CNS with neutralizing anti-HMGB1 monoclonal antibody attenuates the progression of EAE. 2. Materials and methods 2.1. Animals 5–7 week female C57BL/6 mice were purchased from SLAC Laboratory Animal Co. Ltd. (Shanghai, China), and housed in a specific pathogen-free facility. All the animal experiments were performed in accordance with the guidelines and permission of Tongji Medical College Animal Care and Use Committee. 2.2. Antibodies Primary antibodies include anti-HMGB1 (rabbit, Abcam), antiCD11b (rat, AbD Serotec), anti-NeuN (mouse, Millipore), anti-glial fibrillary acidic protein (GFAP; mouse, Abcam), and anti-β-actin (mouse, ZSGB-Bio, Beijing, China). 2.3. Active induction of EAE Each mouse was subcutaneously (s.c.) immunized with a total of 200 μg of Myelin oligodendrocyte glycoprote protein (MOG35–55) peptide (CL Bio-Scientific Co. LTD, Xian, China) emulsified in complete Freund's adjuvant (Sigma, St. Louis, MO, USA), supplemented with a final concentration of Mycobacterium tuberculosis H37Ra (5 mg/ml; Difco Laboratories, Detroit, MI, USA). Each mouse was also injected intraperitoneally (i.p.) with 200 ng of pertussin toxin (PTX) (Sigma, St. Louis, MO, USA) i.p. on days 0 and 2 post-immunization. Clinical scores were recorded daily according to the standard EAE grading scale (Stromnes and Goverman, 2006): 0, no clinical sign; 0.5, weak tail; 1, complete loss of tail tone; 2, hind limb weakness; 2.5, partial hind limb paralysis; 3, both hind limb paralysis; 3.5, forelimb weakness and hind limb paralysis; 4, hind-limb and forelimb paralysis; 5, moribund or death state. 2.4. Administration of neutralizing anti-HMGB1 monoclonal antibody Anti-HMGB1 neutralizing monoclonal antibody (HMGB1 Ab; gift from Institute of Biophysics, Chinese Academy of Science, Beijing, China) with a concentration of 100 μg for i.p. injection or 10 μg for intracerebroventricular (i.c.v.) injection was administrated on every other day from days 12 to 22 post-EAE induction, respectively. For i.c.v. injection, HMGB1 Ab (10 μg in 10 μl sterile PBS) was injected into the left lateral ventricle (0.5 mm from bregma; lateral: 1.0 mm from bregma; depth: 2.5 mm from skull surface) at a rate of 2 μl/min via a lateral ventricle buried pipe. The same amount of mouse immunoglobulin (Ig) G (Sigma, St. Louis, MO, USA) was served as control treatment.
4% buffered paraformaldehyde (PFA). Spinal cords were removed and post-fixed in PFA at 4 °C for 24 h, equilibrated for 72 h with 30% sucrose in 0.05 M Tris-buffered saline (TBS, 0.05 M Tris, 0.9% NaCl, pH 7.6) at 4 °C. All spinal cords were cut to 20 μm sections for histological and immunofluorescent staining. 2.6. Immunohistochemistry staining The thoracic spinal cord sections were firstly incubated with the anti-HMGB1 antibody (1:800) in a mixture of TBS, 0.25% gelatin, and 0.5% Triton-X 100 overnight at 4 °C, then bonded with biotinylated anti-rabbit immunoglobulin (Ig)G antibody (1:400) for 2 h at room temperature. The color was visualized with the binding of the Avidin– Biotin Complex (1:800, ABC Elite Kit; Vector Laboratories, Burlingame, CA, USA) followed by 3, 3′-diaminobenzidine tetrahydrochloride (DAB) and H2O2 in TBS incubation. Observation and quantitative assessment of HMGB1 positive cells in total section, white matter or gray matter were performed under 40× magnification fields using a motorized microscope (ECLIPSE 90i, Nikon, Japan) with an Image-Pro Plus 6.0 software. 2.7. Immunofluorescence staining The thoracic spinal cord sections were incubated in 5% bovine serum albumin (BSA) solution for 2 h to block nonspecific binding, followed with primary antibodies mentioned above overnight at 4 °C. Then, the sections were incubated with fluorescent-labeled secondary antibodies at 37 °C for 1 h. Diamidino-2-phenylindole (DAPI; Invitrogen, Carlsbad, CA) was used to stain the cell nuclei. High resolution images were captured by a laser scanning confocal microscope (Olympus Inc., Japan) with FV10-ASW 2.1 image visualization system. Double immunofluorescent labeling cells were counted under 400 × magnification fields manually. 2–3 sections were used from each mouse, and five to six mice were included in each group. Total eleven fields of each spinal cord section were observed and counted. The data were presented as the number of positive cells per square millimeter (mm2). 2.8. Cerebrospinal fluids (CSF) and sera collection After anaesthetization, the skin of a mouse was incised, and the occipital bone was cleared of muscle to expose the atlanto-occipital membrane. The capillary tube was inserted into the cisterna magna through the dura mater, lateral to the arteria dorsalis spinalis and approximately 15 μl of CSF was withdrawn. Blood samples were collected by removing mouse eyeballs and the sera were separated. 2.9. Preparation of spinal cord homogenate The spinal cord was obtained freshly and homogenized with PBS on ice and then centrifuged with 12,000 rpm for 15 min at 4 °C. The supernatant was collected for enzyme-linked immunosorbent assay (ELISA). 2.10. Extraction of spinal cord protein The fresh spinal cords were homogenized on ice in RadioImmunoprecipitation Assay buffer (RIPA) (50 mM Tris–HCl, 150 mM NaCl, 0.02% sodium azide, 1% TritonX-100 and 1 mM PMSF, pH 7.4). After centrifugation, the supernatant was collected for Western Blotting. The protein concentration was measured using the bicinchoninic acid (BCA) protein assay. 2.11. Western Blot
2.5. Spinal cord collection Mice were general anesthetized with 2% sodium pentobarbital and rapidly perfused transcardially with 0.9% saline (37 °C), followed by
Polyacrylamide gel electrophoresis, electrophoretic transfer of protein to PVDF membrane, and immunoblotting were performed using routine techniques previously described (Zou et al., 2014). The primary
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
antibodies included rabbit anti-HMGB1 antibody (1:8000) and mouse anti-β-actin (1:500). Blots were visualized by an ECL system (Pierce Bio-Technology, Rockford, IL) after incubating with horseradish peroxidase (HRP) conjugated secondary antibody, and were quantified by densitometry using the Sysmex CHEMIX Image analysis software. 2.12. Enzyme-linked immunosorbent assay (ELISA) The cerebrospinal fluid, sera and spinal cord homogenate were measured for the level of HMGB1 with a commercial ELISA-assay (USCN Life Science Inc, Wuhan, China) according to the manufacturer's instructions. The minimum detectable dose of HMGB1 is typically less than 32.2 pg/ml. 2.13. Statistical analysis Experimental data are expressed as mean ± standard error of the mean (SEM), and comparisons between the values were performed using a two-tailed Student's t-test. The differences among groups were performed by one-way analysis of variance followed by Bonferroni correction. A P value b0.05 was considered to be statistically significant. 3. Results 3.1. HMGB1 increased during the progression of EAE The course of actively induced EAE in this study could be divided into four stages. The time points when the samples were collected in each stage were shown in Fig. 1A. Days 7–9 represent pre-onset, days 11– 13 for onset, days 16–21 for peak, and days 34–35 for remission stage. As shown in Fig. 1B, in sera, the concentrations of HMGB1 significantly increased during each stage of EAE as compared to that in normal situation (broken line), but no differences were found among stages (Fig. 1B left panel). In spinal cord homogenate, the level of HMGB1 during each stage of EAE markedly elevated and arrived at a peak value in the onset stage, and then declined (Fig. 1B middle panel). In CSF, HMGB1 gradually decreased from pre-onset to remission stage, but still remained higher than that in normal group (broken line) (Fig. 1B right panel). Histopathological examination of the spinal cord revealed inflammatory cell infiltration, especially in the onset and peak period of EAE (Fig. 1C). Immunohistochemical staining presented that HMGB1 was weakly expressed in normal spinal cord, but the expression was increased in the lateral column, dorsal column, ventral horn and dorsal horn of EAE mice in each stage, especially during the onset stage (Fig. 1D). Quantitative analysis of HMGB1+ cells in one whole spinal cord section, or in the white matter (WM) and gray matter (GM) separately showed that the number of HMGB1+ cells was the highest in the onset stage, but gradually decreased in the following stages (Fig. 1E). The amount of total HMGB1 protein in the spinal cord was significantly increased at each stage and reached the highest in the onset stage as compared to that in the normal group (Fig. 1F and G). These data suggest that the extracellular HMGB1 in CNS is highly relevant to the progression of EAE and could be derived from resident cells. 3.2. HMGB1 expression increased in astrocytes and microglia during the development of EAE To define the cell type of HMGB1 expression in adult spinal cords, the double-immunofluorescence staining was carried out by combining HMGB1 with cell specific markers, GFAP for astrocyte, CD11b for microglia and NeuN for neuron, respectively. As illustrated in Fig. 2A, the expression of HMGB1 (green) was detected in the nuclei of part of GFAP positive cells in the normal dorsal column of the spinal cord. During the progression of EAE, the subcellular location of HMGB1 in astrocytes still was the nuclei. The number of HMGB1 and GFAP double positive cells was gradually increased and reached a peak value in the
31
peak stage of EAE, as shown in Fig. 2A and B. Similarly, the expression of HMGB1 was also observed in some CD11b positive cells (microglia) located in the lateral column of the spinal cord. The number of HMGB1+CD11b+ cells was increased along with the development of EAE (Fig. 2C and D). The above data indicate that the nuclear expression pattern of HMGB1 in the astrocytes and microglia is not changed during EAE. Astrocytes and microglia are the potential resources of extracellular HMGB1. 3.3. HMGB1 subcellular location was changed in neurons during the development of EAE The positive staining for HMGB1 was also found in the nuclei of a few of NeuN positive cells (neurons) in the dorsal horn, ventral horn and around the central canal in the normal spinal cord (Fig. 3A, B and C). During EAE, in the dorsal horn, the HMGB1 was still majorly expressed in the nuclei of neurons. And in the pre-onset phase, more HMGB1 and NeuN double positive cells emerged, and then declined along with the damage of neurons presented as smeared NeuN staining in the onset and peak stages. In the remission phase, the count of HMGB1 and NeuN double positive cells was recovered to the normal level (Fig 3A and D, left panel). In the ventral horn, HMGB1 was observed majorly in the cytoplasm instead of in the nuclei of neurons during the onset phase, but in the remission phase, HMGB1 was found to be relocated in the nuclei of neurons. The numbers of HMGB1 positives neurons in each stage keep more than that in normal situation, although there was a dynamic change and the peak value in the onset phase (Fig 3B and D, middle panel). Around the central canal, the dynamic change of HMGB1 expression pattern in the neurons was similar with that in ventral horn, as shown in Fig 3C and D right panel. Especially in the onset stage, almost all of the neurons in this region expressed HMGB1 in the cytoplasm. In the stage of remission, the amount of HMGB1+ neurons was recovered accompanying with the restoration of neurons in all three regions. These data indicate that in the development of EAE, the neurons in distinguish regions present the different HMGB1 expression pattern and neurons may be another resource of extracellular HMGB1. 3.4. Blockade of extracellular HMGB1 in CNS attenuated the progression of EAE To determine the roles of local HMGB1 in the progression of EAE, one kind of HMGB1 blocker, anti-HMGB1 neutralizing monoclonal antibody (HMGB1 Ab) was used. Mice were treated with HMGB1 Ab in intraperitoneal (i.p.) injection (Fig. 4A up panel), or intracerebroventricular (i.c.v.) injection (Fig 4A down panel) from days 12 to 22 (onset and peak stages). It was found that HMGB1 Ab treatment via i.p. significantly ameliorated clinical disease progression (Fig. 4B and Table 1), and the treatment via i.c.v. almost totally inhibited the progression of EAE, decreased the disease incidence, and delayed EAE onset time (Fig. 4C and Table 1). These results suggest that local HMGB1 in CNS play critical roles in the progression of EAE. 4. Discussion Although HMGB1 is considered a very abundant, conserved and ubiquitous protein localized in the nucleus of most of cells (Muller et al., 2004), it is not a universally expressed, housekeeping protein in the central nervous system (Guazzi et al., 2003). HMGB1 is present in a subset of brain cells during development, but is undetectable in most cells in an adult mouse brain. In this study, we detected the nuclear expression of HMGB1 in some of astrocytes, microglia and a few neurons in adult mouse spinal cord, which supplies more information to characterize the expression of HMGB1 in CNS and helps in discovering the potential CNS specific functions of HMGB1.
32
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
Fig. 1. The expression and release of HMGB1 in various stages of EAE. (A) The development and stages of EAE. (B) The level of released HMGB1 in different body fluids. Broken line means the normal level. *P b 0.05, **P b 0.01, ***P b 0.001, vs Normal; +P b 0.05, ++P b 0.01, +++P b 0.001, vs Pre-onset stage; #P b 0.05, ##P b 0.01, ###P b 0.001, vs Onset stage; $$P b 0.01, vs Peak stage. (C) Thoracic spinal cord sections were stained by HE. Onset and peak period of EAE mice revealed intensive infiltration of mononuclear cells around the white matter of the spinal cord (arrowheads). Scale bars were 200 μm for low magnification and 20 μm for high magnification. (D) Expression of HMGB1 in thoracic spinal cords detected with immunohistochemistry. 1: lateral column; 2: dorsal column; 3: ventral horn; 4: dorsal horn. Scale bars were 100 μm for low magnification and 10 μm for high magnification. (E) The number of HGMB1 positive cells in one spinal cord section. At least six serial thoracic spinal cord sections were analysed from each mouse, and six mice were included in each group. Data were expressed as mean ± SEM. It is the quantification of panel D. **P b 0.01, ***P b 0.001, vs Normal; +++P b 0.001, vs Pre-onset stage; #P b 0.05, ###P b 0.001, vs Onset stage; $$$P b 0.001, vs Peak stage. (F) Expression of HMGB1 protein in the spinal cord detected by western blot. Two mice per group were shown. (G) Relative densitometry analysis of panel F. Nor: normal; Pr: pre-onset; O: onset; P: peak; R: remission. **P b 0.01, vs Normal. These data were representative of three independent experiments.
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
33
Fig. 2. The expression of HMGB1 in astrocytes and microglia cells in the thoracic spinal cord during EAE. HMGB1 was labeled with green fluorescence, GFAP and CD11b was labeled with red color. Nuclei were stained with DAPI in blue. (A) The expression of HMGB1 and GFAP in white matter. (B) The cell density of HMGB1 positive astrocytes in white matter. It is the quantification of panel A. (C) The expression of HMGB1 and CD11b in white matter. (D) The cell density of HMGB1 positive microglia in white matter. It is the quantification of panel C. The areas in white square in panel A and C were shown as high magnification. **P b 0.01, ***P b 0.001 vs Normal; +P b 0.05, ++P b 0.01, +++P b 0.001, vs Pre-onset stage; #P b 0.05, ###P b 0.001, vs Onset stage; $$P b 0.01, $$$P b 0.001, vs Peak stage. Scale bars were 50 μm. These data were representative of three independent experiments.
Extracellular HMGB1, which is released passively by death cells or actively by stimulated cells such as macrophages, has been reported to play a key role in autoimmune diseases (Magna and Pisetsky, 2014), allograft rejection (Xia et al., 2014), ischemic stroke (Hayakawa et al., 2012), traumatic brain injury (Laird et al., 2014), and neurodegenerative diseases (Fang et al., 2012) as an inflammatory mediator. MS/EAE is a CNS specific autoimmune disease, which happens in adult human being or is induced in adult animal. In the present study, extracellular HMGB1 was found to be increased in the sera, spinal cord homogenate and CSF from various stages of EAE, which is consistent with sera HMGB1 elevation observed in MS patients. The fluctuation line of HMGB1 in the sera of EAE mice is similar with that of EAE score, implicating a dynamic systemic inflammatory response including the immune response in secondary lymphoid organs and the inflammation in CNS. The peak of extracellular HMGB1 in the spinal cord homogenate emerges in the onset stage and is earlier than the peak of disease score, which may partly be due to the increase of total HMGB1 and the damage of endotheliocytes (ECs) and neurons that passively released HMGB1. In addition, some of the infiltrated immune cells, such as macrophages, also released HMGB1 actively. Previous studies have shown T cell infiltration and microglia activation in the CNS during the progression of EAE (Murphy et al., 2010). ELISA reveals increased HMGB1 in the CSF, implicating the intrathecal release of HMGB1. The levels of HMGB1 in
CSF that are gradually decreased from pre-onset to remission stage probably reflect the local release of HMGB1 decreased step by step, and tissue remodeling and healing progressively increased. These results suggest that the dynamic measurement of extracellular HMGB1 levels may be helpful in judging the progression and outcome of MS/ EAE. Presently, HMGB1 is regarded to play a deleterious role during the acute stages of EAE. However, a certain level of HMGB1, a little higher than normal value, may have beneficial effects on neuroregeneration. It has been reported that HMGB1 may also possess beneficial actions including wound healing, neovascularization, enhancement of neurite outgrowth and plastic forms of tissue remodeling (Biscetti et al., 2010; Horner and Gage, 2000; Huttunen et al., 2000). Therefore, extracellular HMGB1 may play biphasic roles in EAE progression. The distribution and localization of HMGB1 in spinal cord resident cells under normal condition are observed in the present study. During EAE, the number of HMGB1 positive astrocytes, microglia and neurons is increased. Most of the intracellular HMGB1 in both astrocytes and microglia is still in the nucleus. However, the intracellular HMGB1 in some neurons is in the cytoplasm, not in the nucleus. It is possible that the neurons can actively secrete HMGB1 to amplify CNS inflammatory response and exacerbate tissue damage. HMGB1 exerts its neuroinflammatory effects mainly through its receptors leading to the
34
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
Fig. 3. The expression of HMGB1 in neurons in the thoracic spinal cord during EAE. HMGB1 was labeled with green fluorescence and NeuN was labeled with red color. Nuclei were stained with DAPI in blue. (A) The expression of HMGB1 and NeuN in the dorsal horn. (B) The expression of HMGB1 and NeuN in the ventral horn. In (A) and (B), the areas in white boxes were highly magnified. Arrows pointed to HMGB1 and NeuN double positive cells. (C) The expression of HMGB1 and NeuN around the central canal. The cells directed by arrowheads were highly magnified. (D) The cell density of HMGB1 positive neurons. **P b 0.01, ***P b 0.001 vs Normal; ++P b 0.01, +++P b 0.001, vs Pre-onset stage; ##P b 0.01, ###P b 0.001, vs Onset stage; $$P b 0.01, $$$P b 0.001, vs Peak stage. Scale bars were 50 μm. These data were representative of three independent experiments.
Fig. 4. The effects of HMGB1 blocking on the development of EAE. (A) Schematic workflow of HMGB1 neutralizing antibody treatment. (B) The effects of HMGB1 neutralizing antibody i.p injection on disease score. 100 μg neutralizing anti-HMGB1 monoclonal antibody or control mouse IgG was injected each time. (C) The effects of HMGB1 neutralizing antibody i.c.v. injection on disease score. 10 μg/10 μl HMGB1 antibody or control IgG each time. **P b 0.01, EAE+HMGB1 Ab vs EAE group; ##P b 0.01, EAE+HMGB1 Ab vs the EAE+IgG group.
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35 Table 1 Clinical features of EAE mice with anti-HMGB1 neutralizing antibody treatment. Group
Incidence,%
Intraperitoneal (i.p.) injection EAE 100 EAE+IgG 100 EAE+HMGB1 Ab 75
Mean maximal score (mean ± SME)
Disease onset (mean ± SME)
2.79±0.20 2.88±0.16 1.38±0.31⁎⁎,##
14.17±0.67 15.92±0.74 17.33±1.99
Intracerebroventricular (i.c.v.) injection EAE pipes 100 2.25±0.31 EAE+IgG 100 2.40±0.19 EAE+HMGB1 Ab 63 0.63±0.25⁎⁎,##
13.38±0.89 11.18±0.37 14.80±0.37##
“EAE” means only EAE induction without treatment. “EAE+IgG” means EAE induction with control IgG treatment. “EAE+HMGB1 Ab” means EAE induction with anti-HMGB1 neutralizing antibody treatment. The unit of “Disease onset” is day. **P b 0.01 vs EAE group; ##P b 0.01 vs the EAE+IgG group.
activation of MyD88 and MAPK cascades and mediated the NF-κB activation (Mc Guire et al., 2013). How HMGB1 interacts with CNS resident cell networks and pathways remains to be fully addressed. Other reports have shown the systemic beneficial effects using a neutralizing anti-HMGB1 monoclonal antibody for EAE mice previously (Robinson et al., 2013; Uzawa et al., 2013). In this study, HMGB1 Ab is injected intracerebroventricularly in EAE mice to confirm the roles of local HMGB1 in CNS. The results indicate that blockade of HMGB1 in CNS attenuates EAE more effectively than systemically blocking, which suggests that local HMGB1 plays critical roles in the progression of EAE via an autocrine or paracrine manner possibly. Taken together, in the present study we show that the expression and release of HMGB1 are significantly increased in various stages of EAE. HMGB1 expression pattern is dynamically changed during the development of EAE. Blocking of HMGB1 locally not only attenuated the disease severity and incidence but also delayed disease onset time. Our data suggest that HMBG1 in CNS acts as one of the key mediators in EAE neuropathology. It may trigger and amplify the local inflammatory cascade, resulting in axonal damage and demyelination, which is needed to be further addressed. Acknowledgment This work was supported by the National Natural Science Foundation of China (Grant No. 31470852) and the Major State Basic Research Development Program of China (973 Program) (Grant No. 2013CB530505). References Agresti, A., Lupo, R., Bianchi, M.E., Muller, S., 2003. HMGB1 interacts differentially with members of the Rel family of transcription factors. Biochem. Biophys. Res. Commun. 302, 421–426. Andersson, U., Erlandsson-Harris, H., 2004. HMGB1 is a potent trigger of arthritis. J. Intern. Med. 255, 344–350. Andersson, U., Wang, H., Palmblad, K., Aveberger, A.C., Bloom, O., Erlandsson-Harris, H., Janson, A., Kokkola, R., Zhang, M., Yang, H., Tracey, K.J., 2000. High mobility group 1 protein (HMG-1) stimulates proinflammatory cytokine synthesis in human monocytes. J. Exp. Med. 192, 565–570. Andersson, A., Covacu, R., Sunnemark, D., Danilov, A.I., Dal Bianco, A., Khademi, M., Wallstrom, E., Lobell, A., Brundin, L., Lassmann, H., Harris, R.A., 2008. Pivotal advance: HMGB1 expression in active lesions of human and experimental multiple sclerosis. J. Leukoc. Biol. 84, 1248–1255. Benn, T., Halfpenny, C., Scolding, N., 2001. Glial cells as targets for cytotoxic immune mediators. Glia 36, 200–211. Benveniste, E.N., 1997. Role of macrophages/microglia in multiple sclerosis and experimental allergic encephalomyelitis. J. Mol. Med. 75, 165–173. Biscetti, F., Straface, G., De Cristofaro, R., Lancellotti, S., Rizzo, P., Arena, V., Stigliano, E., Pecorini, G., Egashira, K., De Angelis, G., Ghirlanda, G., Flex, A., 2010. High-mobility group box-1 protein promotes angiogenesis after peripheral ischemia in diabetic mice through a VEGF-dependent mechanism. Diabetes 59, 1496–1505. Compston, A., Coles, A., 2008. Multiple sclerosis. Lancet 372, 1502–1517.
35
Fang, P., Schachner, M., Shen, Y.Q., 2012. HMGB1 in development and diseases of the central nervous system. Mol. Neurobiol. 45, 499–506. 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, 14–19. Guazzi, S., Strangio, A., Franzi, A.T., Bianchi, M.E., 2003. HMGB1, an architectural chromatin protein and extracellular signalling factor, has a spatially and temporally restricted expression pattern in mouse brain. Gene Expr. Patterns 3, 29–33. Hayakawa, K., Pham, L.D., Katusic, Z.S., Arai, K., Lo, E.H., 2012. Astrocytic high-mobility group box 1 promotes endothelial progenitor cell-mediated neurovascular remodeling during stroke recovery. Proc. Natl. Acad. Sci. U. S. A. 109, 7505–7510. Horner, P.J., Gage, F.H., 2000. Regenerating the damaged central nervous system. Nature 407, 963–970. Huttunen, H.J., Kuja-Panula, J., Sorci, G., Agneletti, A.L., Donato, R., Rauvala, H., 2000. Coregulation of neurite outgrowth and cell survival by amphoterin and S100 proteins through receptor for advanced glycation end products (RAGE) activation. J. Biol. Chem. 275, 40096–40105. Kim, J.B., Sig Choi, J., Yu, Y.M., Nam, K., Piao, C.S., Kim, S.W., Lee, M.H., Han, P.L., Park, J.S., Lee, J.K., 2006. HMGB1, a novel cytokine-like mediator linking acute neuronal death and delayed neuroinflammation in the postischemic brain. J. Neurosci. 26, 6413–6421. Kim, J.H., Kim, S.J., Lee, I.S., Lee, M.S., Uematsu, S., Akira, S., Oh, K.I., 2009. Bacterial endotoxin induces the release of high mobility group box 1 via the IFN-beta signaling pathway. J. Immunol. 182, 2458–2466. Kroncke, K.D., Fehsel, K., Kolb-Bachofen, V., 1998. Inducible nitric oxide synthase in human diseases. Clin. Exp. Immunol. 113, 147–156. Laird, M.D., Shields, J.S., Sukumari-Ramesh, S., Kimbler, D.E., Fessler, R.D., Shakir, B., Youssef, P., Yanasak, N., Vender, J.R., Dhandapani, K.M., 2014. High mobility group box protein-1 promotes cerebral edema after traumatic brain injury via activation of toll-like receptor 4. Glia 62, 26–38. Lotze, M.T., Tracey, K.J., 2005. High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal. Immunology 5, 331–342. Magna, M., Pisetsky, D.S., 2014. The role of HMGB1 in the pathogenesis of inflammatory and autoimmune diseases. Mol. Med. 20, 138–146. Matsuoka, N., Itoh, T., Watarai, H., Sekine-Kondo, E., Nagata, N., Okamoto, K., Mera, T., Yamamoto, H., Yamada, S., Maruyama, I., Taniguchi, M., Yasunami, Y., 2010. Highmobility group box 1 is involved in the initial events of early loss of transplanted islets in mice. J. Clin. Invest. 120, 735–743. Mc Guire, C., Prinz, M., Beyaert, R., van Loo, G., 2013. Nuclear factor kappa B (NF-kappaB) in multiple sclerosis pathology. Trends Mol. Med. 19, 604–613. Miller, S.D., Karpus, W.J., Davidson, T.S., 2010. Experimental Autoimmune Encephalomyelitis in the Mouse. Current protocols in immunology. In: Coligan, John E., et al. (Eds.), (Chapter 15, Unit 15 11). Muller, S., Ronfani, L., Bianchi, M.E., 2004. Regulated expression and subcellular localization of HMGB1, a chromatin protein with a cytokine function. J. Intern. Med. 255, 332–343. Murphy, A.C., Lalor, S.J., Lynch, M.A., Mills, K.H., 2010. Infiltration of Th1 and Th17 cells and activation of microglia in the CNS during the course of experimental autoimmune encephalomyelitis. Brain Behav. Immun. 24, 641–651. Neumann, H., 2001. Control of glial immune function by neurons. Glia 36, 191–199. Pisetsky, D.S., Erlandsson-Harris, H., Andersson, U., 2008. High-mobility group box protein 1 (HMGB1): an alarmin mediating the pathogenesis of rheumatic disease. Arthritis Res. Ther. 10, 209. Qiu, J., Nishimura, M., Wang, Y., Sims, J.R., Qiu, S., Savitz, S.I., Salomone, S., Moskowitz, M.A., 2008. Early release of HMGB-1 from neurons after the onset of brain ischemia. J. Cereb. Blood Flow Metab. 28, 927–938. Robinson, A.P., Caldis, M.W., Harp, C.T., Goings, G.E., Miller, S.D., 2013. High-mobility group box 1 protein (HMGB1) neutralization ameliorates experimental autoimmune encephalomyelitis. J. Autoimmun. 43, 32–43. Rouhiainen, A., Kuja-Panula, J., Wilkman, E., Pakkanen, J., Stenfors, J., Tuominen, R.K., Lepantalo, M., Carpen, O., Parkkinen, J., Rauvala, H., 2004. Regulation of monocyte migration by amphoterin (HMGB1). Blood 104, 1174–1182. Schwartz, D., Engelhard, D., Gallily, R., Matoth, I., Brenner, T., 1998. Glial cells production of inflammatory mediators induced by Streptococcus pneumoniae: inhibition by pentoxifylline, low-molecular-weight heparin and dexamethasone. J. Neurol. Sci. 155, 13–22. Streit, W.J., 2002. Microglia as neuroprotective, immunocompetent cells of the CNS. Glia 40, 133–139. Stromnes, I.M., Goverman, J.M., 2006. Active induction of experimental allergic encephalomyelitis. Nat. Protoc. 1, 1810–1819. Uzawa, A., Mori, M., Taniguchi, J., Masuda, S., Muto, M., Kuwabara, S., 2013. Anti-high mobility group box 1 monoclonal antibody ameliorates experimental autoimmune encephalomyelitis. Clin. Exp. Immunol. 172, 37–43. Xia, Q., Duan, L., Shi, L., Zheng, F., Gong, F., Fang, M., 2014. High-mobility group box 1 accelerates early acute allograft rejection via enhancing IL-17+ gammadelta T-cell response. Transpl. Int. 27, 399–407. Yang, H., Ochani, M., Li, J., Qiang, X., Tanovic, M., Harris, H.E., Susarla, S.M., Ulloa, L., Wang, H., DiRaimo, R., Czura, C.J., Wang, H., Roth, J., Warren, H.S., Fink, M.P., Fenton, M.J., Andersson, U., Tracey, K.J., 2004. Reversing established sepsis with antagonists of endogenous high-mobility group box 1. Proc. Natl. Acad. Sci. U. S. A. 101, 296–301. Zou, H., Yang, Y., Gao, M., Zhang, B., Ming, B., Sun, Y., Chen, H., Tang, X., Chen, Z., Xiong, P., Xu, Y., Fang, M., Tan, Z., Gong, F., Zheng, F., 2014. HMGB1 is involved in chronic rejection of cardiac allograft via promoting inflammatory-like mDCs. Am. J. Transplant. 14, 1765–1777.
Contents lists available at ScienceDirect
Journal of Neuroimmunology journal homepage: www.elsevier.com/locate/jneuroim
HMGB1 expression patterns during the progression of experimental autoimmune encephalomyelitis Yan Sun a,b, Huoying Chen a, Jiapei Dai b, Huijuan Zou a, Ming Gao a, Hao Wu a, Bingxia Ming a, Lin Lai a, Yifan Xiao a, Ping Xiong a, Yong Xu a, Feili Gong a, Fang Zheng a,⁎ a b
Department of Immunology, School of Basic Medicine, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China Wuhan Institute for Neuroscience and Neuroengineering, South-Central University for Nationalities, Wuhan 430074, China
a r t i c l e
i n f o
Article history: Received 8 January 2015 Received in revised form 9 February 2015 Accepted 23 February 2015 Keywords: High mobility group box 1 Experimental autoimmune encephalomyelitis Astrocyte Microglia Neuron
a b s t r a c t High mobility group box 1 (HMGB1), a nonhistone chromatin associated protein, plays different roles according to the expression pattern such as the amount, cell location and sub-cellular location. It has been recently demonstrated that the systemic HMGB1 is associated with autoimmune encephalomyelitis. However, the dynamic change of HMGB1 expression pattern in spinal cords that may be involved in the progression of disease is not fully understood. In this study, the amount, cell location and subcellular location of HMGB1 in adult mice spinal cords during various stages of experimental autoimmune encephalomyelitis (EAE) are investigated. HMGB1 is expressed in the nuclei of spinal cord resident cells such as some astrocytes, microglia and a few neurons in normal situation. During EAE progression, the total and extracellular HMGB1 in the spinal cord are increased, more HMGB1 positive astrocytes and microglia are observed, and the intra-neurons HMGB1 in the ventral horn and around the central canal localize majorly in the cytoplasm accompanied by the increasing extracellular HMGB1. Blockade of HMGB1 in central nervous system (CNS) locally attenuates the severity of EAE significantly. Our findings indicate that the HMGB1 expression pattern in the spinal cord is associated with the progression of EAE. HMGB1 may be a potential target for autoimmune encephalomyelitis (multiple sclerosis in human) therapy. © 2015 Elsevier B.V. All rights reserved.
1. Introduction High mobility group box 1 (HMGB1) is found initially as a nonhistone DNA-binding nuclear protein (Goodwin et al., 1973) in almost all eukaryotic cells (Matsuoka et al., 2010). It promotes chromatin function and gene regulation (Agresti et al., 2003). Now, a lot of studies have demonstrated that HMGB1 exerts different biological functions depending on its distribution and cellular localization. For example, upon proinflammatory mediators stimulation or cell damage, nuclear HMGB1 can be transported into the cytoplasm or released into extracellular space for the regulation of immunity and inflammation (Kim et al., 2009; Lotze and Tracey, 2005). Extracellular HMGB1 acting as a damage associated molecular pattern (DAMP) can induce the release of proinflammatory cytokines including IL-1β, TNF-α, IL-6 (Andersson et al., 2000) and chemokines (Rouhiainen et al., 2004) via binding with the receptor for advanced glycation end products (RAGE) or toll-like receptors (TLRs), triggering sterile inflammation in various diseases, such as arthritis, brain ischemia and traumatic brain injury (Laird et al., 2014; Pisetsky et al., 2008; Qiu et al., 2008). The administration of neutralizing anti-HMGB1 monoclonal and polyclonal antibodies (mAb and pAb), and other HMGB1 antagonists (e.g., glycyrrhizin) dampens inflammation ⁎ Corresponding author. E-mail address: [email protected] (F. Zheng).
http://dx.doi.org/10.1016/j.jneuroim.2015.02.005 0165-5728/© 2015 Elsevier B.V. All rights reserved.
(Andersson and Erlandsson-Harris, 2004; Kim et al., 2006; Yang et al., 2004), suggesting that HMGB1 could be considered as a therapeutic target. The central nervous system (CNS) has a highly specialized microenvironment. Because of the blood–brain barrier (BBB) that limits entry of immune cells and molecules, glial cells residing in the CNS are essential for the maintenance of organism homeostasis in normal situation (Neumann, 2001; Streit, 2002). Under pathological conditions, infiltrated immune cells and glial cells contribute to the disorder of CNS via inducing immune deregulation and neuroinflammation mediated by direct killing and releasing neurotoxic substances, free radicals and inflammatory mediators (Benn et al., 2001; Benveniste, 1997; Kroncke et al., 1998; Schwartz et al., 1998). Although HMGB1 protein is present in some subsets of CNS resident cells during development, with a very complex temporal, spatial and subcellular expression pattern (Guazzi et al., 2003), the detailed expression pattern of HMGB1 is not clear in the different types of neural cells including neurons, astrocytes, and microglia in the adult mouse CNS. The dynamic change of HMGB1 in inflammatory CNS is also needed to be addressed. Multiple sclerosis (MS) is an autoimmune-mediated inflammatory disease of CNS characterized by demyelination and axonal damage (Compston and Coles, 2008). Experimental autoimmune encephalomyelitis (EAE) is the most reliable experimental model of MS (Miller et al., 2010). Increased HMGB1 levels were found in the cerebrospinal fluid
30
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
(CSF) of MS patients (Andersson et al., 2008), and systemic treatment with neutralizing HMGB1 antibody can rescue mice from EAE (Robinson et al., 2013; Uzawa et al., 2013). These facts suggest that HMGB1 plays critical roles in MS/EAE. However, the proportion of CNS derived HMGB1 in the development of MS/EAE remains poorly understood. In the present study, we discover HMGB1 is nuclearly expressed in some astrocytes, microglia and a few of the neurons in adult normal mouse spinal cords. During EAE progression, the expression and release of HMGB1 in the spinal cord are increased, more astrocytes and microglia express HMGB1, and HMGB1 is detectable in the cytoplasm of neurons in the ventral horn and around the central canal. Blockade of HMGB1 locally in CNS with neutralizing anti-HMGB1 monoclonal antibody attenuates the progression of EAE. 2. Materials and methods 2.1. Animals 5–7 week female C57BL/6 mice were purchased from SLAC Laboratory Animal Co. Ltd. (Shanghai, China), and housed in a specific pathogen-free facility. All the animal experiments were performed in accordance with the guidelines and permission of Tongji Medical College Animal Care and Use Committee. 2.2. Antibodies Primary antibodies include anti-HMGB1 (rabbit, Abcam), antiCD11b (rat, AbD Serotec), anti-NeuN (mouse, Millipore), anti-glial fibrillary acidic protein (GFAP; mouse, Abcam), and anti-β-actin (mouse, ZSGB-Bio, Beijing, China). 2.3. Active induction of EAE Each mouse was subcutaneously (s.c.) immunized with a total of 200 μg of Myelin oligodendrocyte glycoprote protein (MOG35–55) peptide (CL Bio-Scientific Co. LTD, Xian, China) emulsified in complete Freund's adjuvant (Sigma, St. Louis, MO, USA), supplemented with a final concentration of Mycobacterium tuberculosis H37Ra (5 mg/ml; Difco Laboratories, Detroit, MI, USA). Each mouse was also injected intraperitoneally (i.p.) with 200 ng of pertussin toxin (PTX) (Sigma, St. Louis, MO, USA) i.p. on days 0 and 2 post-immunization. Clinical scores were recorded daily according to the standard EAE grading scale (Stromnes and Goverman, 2006): 0, no clinical sign; 0.5, weak tail; 1, complete loss of tail tone; 2, hind limb weakness; 2.5, partial hind limb paralysis; 3, both hind limb paralysis; 3.5, forelimb weakness and hind limb paralysis; 4, hind-limb and forelimb paralysis; 5, moribund or death state. 2.4. Administration of neutralizing anti-HMGB1 monoclonal antibody Anti-HMGB1 neutralizing monoclonal antibody (HMGB1 Ab; gift from Institute of Biophysics, Chinese Academy of Science, Beijing, China) with a concentration of 100 μg for i.p. injection or 10 μg for intracerebroventricular (i.c.v.) injection was administrated on every other day from days 12 to 22 post-EAE induction, respectively. For i.c.v. injection, HMGB1 Ab (10 μg in 10 μl sterile PBS) was injected into the left lateral ventricle (0.5 mm from bregma; lateral: 1.0 mm from bregma; depth: 2.5 mm from skull surface) at a rate of 2 μl/min via a lateral ventricle buried pipe. The same amount of mouse immunoglobulin (Ig) G (Sigma, St. Louis, MO, USA) was served as control treatment.
4% buffered paraformaldehyde (PFA). Spinal cords were removed and post-fixed in PFA at 4 °C for 24 h, equilibrated for 72 h with 30% sucrose in 0.05 M Tris-buffered saline (TBS, 0.05 M Tris, 0.9% NaCl, pH 7.6) at 4 °C. All spinal cords were cut to 20 μm sections for histological and immunofluorescent staining. 2.6. Immunohistochemistry staining The thoracic spinal cord sections were firstly incubated with the anti-HMGB1 antibody (1:800) in a mixture of TBS, 0.25% gelatin, and 0.5% Triton-X 100 overnight at 4 °C, then bonded with biotinylated anti-rabbit immunoglobulin (Ig)G antibody (1:400) for 2 h at room temperature. The color was visualized with the binding of the Avidin– Biotin Complex (1:800, ABC Elite Kit; Vector Laboratories, Burlingame, CA, USA) followed by 3, 3′-diaminobenzidine tetrahydrochloride (DAB) and H2O2 in TBS incubation. Observation and quantitative assessment of HMGB1 positive cells in total section, white matter or gray matter were performed under 40× magnification fields using a motorized microscope (ECLIPSE 90i, Nikon, Japan) with an Image-Pro Plus 6.0 software. 2.7. Immunofluorescence staining The thoracic spinal cord sections were incubated in 5% bovine serum albumin (BSA) solution for 2 h to block nonspecific binding, followed with primary antibodies mentioned above overnight at 4 °C. Then, the sections were incubated with fluorescent-labeled secondary antibodies at 37 °C for 1 h. Diamidino-2-phenylindole (DAPI; Invitrogen, Carlsbad, CA) was used to stain the cell nuclei. High resolution images were captured by a laser scanning confocal microscope (Olympus Inc., Japan) with FV10-ASW 2.1 image visualization system. Double immunofluorescent labeling cells were counted under 400 × magnification fields manually. 2–3 sections were used from each mouse, and five to six mice were included in each group. Total eleven fields of each spinal cord section were observed and counted. The data were presented as the number of positive cells per square millimeter (mm2). 2.8. Cerebrospinal fluids (CSF) and sera collection After anaesthetization, the skin of a mouse was incised, and the occipital bone was cleared of muscle to expose the atlanto-occipital membrane. The capillary tube was inserted into the cisterna magna through the dura mater, lateral to the arteria dorsalis spinalis and approximately 15 μl of CSF was withdrawn. Blood samples were collected by removing mouse eyeballs and the sera were separated. 2.9. Preparation of spinal cord homogenate The spinal cord was obtained freshly and homogenized with PBS on ice and then centrifuged with 12,000 rpm for 15 min at 4 °C. The supernatant was collected for enzyme-linked immunosorbent assay (ELISA). 2.10. Extraction of spinal cord protein The fresh spinal cords were homogenized on ice in RadioImmunoprecipitation Assay buffer (RIPA) (50 mM Tris–HCl, 150 mM NaCl, 0.02% sodium azide, 1% TritonX-100 and 1 mM PMSF, pH 7.4). After centrifugation, the supernatant was collected for Western Blotting. The protein concentration was measured using the bicinchoninic acid (BCA) protein assay. 2.11. Western Blot
2.5. Spinal cord collection Mice were general anesthetized with 2% sodium pentobarbital and rapidly perfused transcardially with 0.9% saline (37 °C), followed by
Polyacrylamide gel electrophoresis, electrophoretic transfer of protein to PVDF membrane, and immunoblotting were performed using routine techniques previously described (Zou et al., 2014). The primary
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
antibodies included rabbit anti-HMGB1 antibody (1:8000) and mouse anti-β-actin (1:500). Blots were visualized by an ECL system (Pierce Bio-Technology, Rockford, IL) after incubating with horseradish peroxidase (HRP) conjugated secondary antibody, and were quantified by densitometry using the Sysmex CHEMIX Image analysis software. 2.12. Enzyme-linked immunosorbent assay (ELISA) The cerebrospinal fluid, sera and spinal cord homogenate were measured for the level of HMGB1 with a commercial ELISA-assay (USCN Life Science Inc, Wuhan, China) according to the manufacturer's instructions. The minimum detectable dose of HMGB1 is typically less than 32.2 pg/ml. 2.13. Statistical analysis Experimental data are expressed as mean ± standard error of the mean (SEM), and comparisons between the values were performed using a two-tailed Student's t-test. The differences among groups were performed by one-way analysis of variance followed by Bonferroni correction. A P value b0.05 was considered to be statistically significant. 3. Results 3.1. HMGB1 increased during the progression of EAE The course of actively induced EAE in this study could be divided into four stages. The time points when the samples were collected in each stage were shown in Fig. 1A. Days 7–9 represent pre-onset, days 11– 13 for onset, days 16–21 for peak, and days 34–35 for remission stage. As shown in Fig. 1B, in sera, the concentrations of HMGB1 significantly increased during each stage of EAE as compared to that in normal situation (broken line), but no differences were found among stages (Fig. 1B left panel). In spinal cord homogenate, the level of HMGB1 during each stage of EAE markedly elevated and arrived at a peak value in the onset stage, and then declined (Fig. 1B middle panel). In CSF, HMGB1 gradually decreased from pre-onset to remission stage, but still remained higher than that in normal group (broken line) (Fig. 1B right panel). Histopathological examination of the spinal cord revealed inflammatory cell infiltration, especially in the onset and peak period of EAE (Fig. 1C). Immunohistochemical staining presented that HMGB1 was weakly expressed in normal spinal cord, but the expression was increased in the lateral column, dorsal column, ventral horn and dorsal horn of EAE mice in each stage, especially during the onset stage (Fig. 1D). Quantitative analysis of HMGB1+ cells in one whole spinal cord section, or in the white matter (WM) and gray matter (GM) separately showed that the number of HMGB1+ cells was the highest in the onset stage, but gradually decreased in the following stages (Fig. 1E). The amount of total HMGB1 protein in the spinal cord was significantly increased at each stage and reached the highest in the onset stage as compared to that in the normal group (Fig. 1F and G). These data suggest that the extracellular HMGB1 in CNS is highly relevant to the progression of EAE and could be derived from resident cells. 3.2. HMGB1 expression increased in astrocytes and microglia during the development of EAE To define the cell type of HMGB1 expression in adult spinal cords, the double-immunofluorescence staining was carried out by combining HMGB1 with cell specific markers, GFAP for astrocyte, CD11b for microglia and NeuN for neuron, respectively. As illustrated in Fig. 2A, the expression of HMGB1 (green) was detected in the nuclei of part of GFAP positive cells in the normal dorsal column of the spinal cord. During the progression of EAE, the subcellular location of HMGB1 in astrocytes still was the nuclei. The number of HMGB1 and GFAP double positive cells was gradually increased and reached a peak value in the
31
peak stage of EAE, as shown in Fig. 2A and B. Similarly, the expression of HMGB1 was also observed in some CD11b positive cells (microglia) located in the lateral column of the spinal cord. The number of HMGB1+CD11b+ cells was increased along with the development of EAE (Fig. 2C and D). The above data indicate that the nuclear expression pattern of HMGB1 in the astrocytes and microglia is not changed during EAE. Astrocytes and microglia are the potential resources of extracellular HMGB1. 3.3. HMGB1 subcellular location was changed in neurons during the development of EAE The positive staining for HMGB1 was also found in the nuclei of a few of NeuN positive cells (neurons) in the dorsal horn, ventral horn and around the central canal in the normal spinal cord (Fig. 3A, B and C). During EAE, in the dorsal horn, the HMGB1 was still majorly expressed in the nuclei of neurons. And in the pre-onset phase, more HMGB1 and NeuN double positive cells emerged, and then declined along with the damage of neurons presented as smeared NeuN staining in the onset and peak stages. In the remission phase, the count of HMGB1 and NeuN double positive cells was recovered to the normal level (Fig 3A and D, left panel). In the ventral horn, HMGB1 was observed majorly in the cytoplasm instead of in the nuclei of neurons during the onset phase, but in the remission phase, HMGB1 was found to be relocated in the nuclei of neurons. The numbers of HMGB1 positives neurons in each stage keep more than that in normal situation, although there was a dynamic change and the peak value in the onset phase (Fig 3B and D, middle panel). Around the central canal, the dynamic change of HMGB1 expression pattern in the neurons was similar with that in ventral horn, as shown in Fig 3C and D right panel. Especially in the onset stage, almost all of the neurons in this region expressed HMGB1 in the cytoplasm. In the stage of remission, the amount of HMGB1+ neurons was recovered accompanying with the restoration of neurons in all three regions. These data indicate that in the development of EAE, the neurons in distinguish regions present the different HMGB1 expression pattern and neurons may be another resource of extracellular HMGB1. 3.4. Blockade of extracellular HMGB1 in CNS attenuated the progression of EAE To determine the roles of local HMGB1 in the progression of EAE, one kind of HMGB1 blocker, anti-HMGB1 neutralizing monoclonal antibody (HMGB1 Ab) was used. Mice were treated with HMGB1 Ab in intraperitoneal (i.p.) injection (Fig. 4A up panel), or intracerebroventricular (i.c.v.) injection (Fig 4A down panel) from days 12 to 22 (onset and peak stages). It was found that HMGB1 Ab treatment via i.p. significantly ameliorated clinical disease progression (Fig. 4B and Table 1), and the treatment via i.c.v. almost totally inhibited the progression of EAE, decreased the disease incidence, and delayed EAE onset time (Fig. 4C and Table 1). These results suggest that local HMGB1 in CNS play critical roles in the progression of EAE. 4. Discussion Although HMGB1 is considered a very abundant, conserved and ubiquitous protein localized in the nucleus of most of cells (Muller et al., 2004), it is not a universally expressed, housekeeping protein in the central nervous system (Guazzi et al., 2003). HMGB1 is present in a subset of brain cells during development, but is undetectable in most cells in an adult mouse brain. In this study, we detected the nuclear expression of HMGB1 in some of astrocytes, microglia and a few neurons in adult mouse spinal cord, which supplies more information to characterize the expression of HMGB1 in CNS and helps in discovering the potential CNS specific functions of HMGB1.
32
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
Fig. 1. The expression and release of HMGB1 in various stages of EAE. (A) The development and stages of EAE. (B) The level of released HMGB1 in different body fluids. Broken line means the normal level. *P b 0.05, **P b 0.01, ***P b 0.001, vs Normal; +P b 0.05, ++P b 0.01, +++P b 0.001, vs Pre-onset stage; #P b 0.05, ##P b 0.01, ###P b 0.001, vs Onset stage; $$P b 0.01, vs Peak stage. (C) Thoracic spinal cord sections were stained by HE. Onset and peak period of EAE mice revealed intensive infiltration of mononuclear cells around the white matter of the spinal cord (arrowheads). Scale bars were 200 μm for low magnification and 20 μm for high magnification. (D) Expression of HMGB1 in thoracic spinal cords detected with immunohistochemistry. 1: lateral column; 2: dorsal column; 3: ventral horn; 4: dorsal horn. Scale bars were 100 μm for low magnification and 10 μm for high magnification. (E) The number of HGMB1 positive cells in one spinal cord section. At least six serial thoracic spinal cord sections were analysed from each mouse, and six mice were included in each group. Data were expressed as mean ± SEM. It is the quantification of panel D. **P b 0.01, ***P b 0.001, vs Normal; +++P b 0.001, vs Pre-onset stage; #P b 0.05, ###P b 0.001, vs Onset stage; $$$P b 0.001, vs Peak stage. (F) Expression of HMGB1 protein in the spinal cord detected by western blot. Two mice per group were shown. (G) Relative densitometry analysis of panel F. Nor: normal; Pr: pre-onset; O: onset; P: peak; R: remission. **P b 0.01, vs Normal. These data were representative of three independent experiments.
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
33
Fig. 2. The expression of HMGB1 in astrocytes and microglia cells in the thoracic spinal cord during EAE. HMGB1 was labeled with green fluorescence, GFAP and CD11b was labeled with red color. Nuclei were stained with DAPI in blue. (A) The expression of HMGB1 and GFAP in white matter. (B) The cell density of HMGB1 positive astrocytes in white matter. It is the quantification of panel A. (C) The expression of HMGB1 and CD11b in white matter. (D) The cell density of HMGB1 positive microglia in white matter. It is the quantification of panel C. The areas in white square in panel A and C were shown as high magnification. **P b 0.01, ***P b 0.001 vs Normal; +P b 0.05, ++P b 0.01, +++P b 0.001, vs Pre-onset stage; #P b 0.05, ###P b 0.001, vs Onset stage; $$P b 0.01, $$$P b 0.001, vs Peak stage. Scale bars were 50 μm. These data were representative of three independent experiments.
Extracellular HMGB1, which is released passively by death cells or actively by stimulated cells such as macrophages, has been reported to play a key role in autoimmune diseases (Magna and Pisetsky, 2014), allograft rejection (Xia et al., 2014), ischemic stroke (Hayakawa et al., 2012), traumatic brain injury (Laird et al., 2014), and neurodegenerative diseases (Fang et al., 2012) as an inflammatory mediator. MS/EAE is a CNS specific autoimmune disease, which happens in adult human being or is induced in adult animal. In the present study, extracellular HMGB1 was found to be increased in the sera, spinal cord homogenate and CSF from various stages of EAE, which is consistent with sera HMGB1 elevation observed in MS patients. The fluctuation line of HMGB1 in the sera of EAE mice is similar with that of EAE score, implicating a dynamic systemic inflammatory response including the immune response in secondary lymphoid organs and the inflammation in CNS. The peak of extracellular HMGB1 in the spinal cord homogenate emerges in the onset stage and is earlier than the peak of disease score, which may partly be due to the increase of total HMGB1 and the damage of endotheliocytes (ECs) and neurons that passively released HMGB1. In addition, some of the infiltrated immune cells, such as macrophages, also released HMGB1 actively. Previous studies have shown T cell infiltration and microglia activation in the CNS during the progression of EAE (Murphy et al., 2010). ELISA reveals increased HMGB1 in the CSF, implicating the intrathecal release of HMGB1. The levels of HMGB1 in
CSF that are gradually decreased from pre-onset to remission stage probably reflect the local release of HMGB1 decreased step by step, and tissue remodeling and healing progressively increased. These results suggest that the dynamic measurement of extracellular HMGB1 levels may be helpful in judging the progression and outcome of MS/ EAE. Presently, HMGB1 is regarded to play a deleterious role during the acute stages of EAE. However, a certain level of HMGB1, a little higher than normal value, may have beneficial effects on neuroregeneration. It has been reported that HMGB1 may also possess beneficial actions including wound healing, neovascularization, enhancement of neurite outgrowth and plastic forms of tissue remodeling (Biscetti et al., 2010; Horner and Gage, 2000; Huttunen et al., 2000). Therefore, extracellular HMGB1 may play biphasic roles in EAE progression. The distribution and localization of HMGB1 in spinal cord resident cells under normal condition are observed in the present study. During EAE, the number of HMGB1 positive astrocytes, microglia and neurons is increased. Most of the intracellular HMGB1 in both astrocytes and microglia is still in the nucleus. However, the intracellular HMGB1 in some neurons is in the cytoplasm, not in the nucleus. It is possible that the neurons can actively secrete HMGB1 to amplify CNS inflammatory response and exacerbate tissue damage. HMGB1 exerts its neuroinflammatory effects mainly through its receptors leading to the
34
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35
Fig. 3. The expression of HMGB1 in neurons in the thoracic spinal cord during EAE. HMGB1 was labeled with green fluorescence and NeuN was labeled with red color. Nuclei were stained with DAPI in blue. (A) The expression of HMGB1 and NeuN in the dorsal horn. (B) The expression of HMGB1 and NeuN in the ventral horn. In (A) and (B), the areas in white boxes were highly magnified. Arrows pointed to HMGB1 and NeuN double positive cells. (C) The expression of HMGB1 and NeuN around the central canal. The cells directed by arrowheads were highly magnified. (D) The cell density of HMGB1 positive neurons. **P b 0.01, ***P b 0.001 vs Normal; ++P b 0.01, +++P b 0.001, vs Pre-onset stage; ##P b 0.01, ###P b 0.001, vs Onset stage; $$P b 0.01, $$$P b 0.001, vs Peak stage. Scale bars were 50 μm. These data were representative of three independent experiments.
Fig. 4. The effects of HMGB1 blocking on the development of EAE. (A) Schematic workflow of HMGB1 neutralizing antibody treatment. (B) The effects of HMGB1 neutralizing antibody i.p injection on disease score. 100 μg neutralizing anti-HMGB1 monoclonal antibody or control mouse IgG was injected each time. (C) The effects of HMGB1 neutralizing antibody i.c.v. injection on disease score. 10 μg/10 μl HMGB1 antibody or control IgG each time. **P b 0.01, EAE+HMGB1 Ab vs EAE group; ##P b 0.01, EAE+HMGB1 Ab vs the EAE+IgG group.
Y. Sun et al. / Journal of Neuroimmunology 280 (2015) 29–35 Table 1 Clinical features of EAE mice with anti-HMGB1 neutralizing antibody treatment. Group
Incidence,%
Intraperitoneal (i.p.) injection EAE 100 EAE+IgG 100 EAE+HMGB1 Ab 75
Mean maximal score (mean ± SME)
Disease onset (mean ± SME)
2.79±0.20 2.88±0.16 1.38±0.31⁎⁎,##
14.17±0.67 15.92±0.74 17.33±1.99
Intracerebroventricular (i.c.v.) injection EAE pipes 100 2.25±0.31 EAE+IgG 100 2.40±0.19 EAE+HMGB1 Ab 63 0.63±0.25⁎⁎,##
13.38±0.89 11.18±0.37 14.80±0.37##
“EAE” means only EAE induction without treatment. “EAE+IgG” means EAE induction with control IgG treatment. “EAE+HMGB1 Ab” means EAE induction with anti-HMGB1 neutralizing antibody treatment. The unit of “Disease onset” is day. **P b 0.01 vs EAE group; ##P b 0.01 vs the EAE+IgG group.
activation of MyD88 and MAPK cascades and mediated the NF-κB activation (Mc Guire et al., 2013). How HMGB1 interacts with CNS resident cell networks and pathways remains to be fully addressed. Other reports have shown the systemic beneficial effects using a neutralizing anti-HMGB1 monoclonal antibody for EAE mice previously (Robinson et al., 2013; Uzawa et al., 2013). In this study, HMGB1 Ab is injected intracerebroventricularly in EAE mice to confirm the roles of local HMGB1 in CNS. The results indicate that blockade of HMGB1 in CNS attenuates EAE more effectively than systemically blocking, which suggests that local HMGB1 plays critical roles in the progression of EAE via an autocrine or paracrine manner possibly. Taken together, in the present study we show that the expression and release of HMGB1 are significantly increased in various stages of EAE. HMGB1 expression pattern is dynamically changed during the development of EAE. Blocking of HMGB1 locally not only attenuated the disease severity and incidence but also delayed disease onset time. Our data suggest that HMBG1 in CNS acts as one of the key mediators in EAE neuropathology. It may trigger and amplify the local inflammatory cascade, resulting in axonal damage and demyelination, which is needed to be further addressed. Acknowledgment This work was supported by the National Natural Science Foundation of China (Grant No. 31470852) and the Major State Basic Research Development Program of China (973 Program) (Grant No. 2013CB530505). References Agresti, A., Lupo, R., Bianchi, M.E., Muller, S., 2003. HMGB1 interacts differentially with members of the Rel family of transcription factors. Biochem. Biophys. Res. Commun. 302, 421–426. Andersson, U., Erlandsson-Harris, H., 2004. HMGB1 is a potent trigger of arthritis. J. Intern. Med. 255, 344–350. Andersson, U., Wang, H., Palmblad, K., Aveberger, A.C., Bloom, O., Erlandsson-Harris, H., Janson, A., Kokkola, R., Zhang, M., Yang, H., Tracey, K.J., 2000. High mobility group 1 protein (HMG-1) stimulates proinflammatory cytokine synthesis in human monocytes. J. Exp. Med. 192, 565–570. Andersson, A., Covacu, R., Sunnemark, D., Danilov, A.I., Dal Bianco, A., Khademi, M., Wallstrom, E., Lobell, A., Brundin, L., Lassmann, H., Harris, R.A., 2008. Pivotal advance: HMGB1 expression in active lesions of human and experimental multiple sclerosis. J. Leukoc. Biol. 84, 1248–1255. Benn, T., Halfpenny, C., Scolding, N., 2001. Glial cells as targets for cytotoxic immune mediators. Glia 36, 200–211. Benveniste, E.N., 1997. Role of macrophages/microglia in multiple sclerosis and experimental allergic encephalomyelitis. J. Mol. Med. 75, 165–173. Biscetti, F., Straface, G., De Cristofaro, R., Lancellotti, S., Rizzo, P., Arena, V., Stigliano, E., Pecorini, G., Egashira, K., De Angelis, G., Ghirlanda, G., Flex, A., 2010. High-mobility group box-1 protein promotes angiogenesis after peripheral ischemia in diabetic mice through a VEGF-dependent mechanism. Diabetes 59, 1496–1505. Compston, A., Coles, A., 2008. Multiple sclerosis. Lancet 372, 1502–1517.
35
Fang, P., Schachner, M., Shen, Y.Q., 2012. HMGB1 in development and diseases of the central nervous system. Mol. Neurobiol. 45, 499–506. 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, 14–19. Guazzi, S., Strangio, A., Franzi, A.T., Bianchi, M.E., 2003. HMGB1, an architectural chromatin protein and extracellular signalling factor, has a spatially and temporally restricted expression pattern in mouse brain. Gene Expr. Patterns 3, 29–33. Hayakawa, K., Pham, L.D., Katusic, Z.S., Arai, K., Lo, E.H., 2012. Astrocytic high-mobility group box 1 promotes endothelial progenitor cell-mediated neurovascular remodeling during stroke recovery. Proc. Natl. Acad. Sci. U. S. A. 109, 7505–7510. Horner, P.J., Gage, F.H., 2000. Regenerating the damaged central nervous system. Nature 407, 963–970. Huttunen, H.J., Kuja-Panula, J., Sorci, G., Agneletti, A.L., Donato, R., Rauvala, H., 2000. Coregulation of neurite outgrowth and cell survival by amphoterin and S100 proteins through receptor for advanced glycation end products (RAGE) activation. J. Biol. Chem. 275, 40096–40105. Kim, J.B., Sig Choi, J., Yu, Y.M., Nam, K., Piao, C.S., Kim, S.W., Lee, M.H., Han, P.L., Park, J.S., Lee, J.K., 2006. HMGB1, a novel cytokine-like mediator linking acute neuronal death and delayed neuroinflammation in the postischemic brain. J. Neurosci. 26, 6413–6421. Kim, J.H., Kim, S.J., Lee, I.S., Lee, M.S., Uematsu, S., Akira, S., Oh, K.I., 2009. Bacterial endotoxin induces the release of high mobility group box 1 via the IFN-beta signaling pathway. J. Immunol. 182, 2458–2466. Kroncke, K.D., Fehsel, K., Kolb-Bachofen, V., 1998. Inducible nitric oxide synthase in human diseases. Clin. Exp. Immunol. 113, 147–156. Laird, M.D., Shields, J.S., Sukumari-Ramesh, S., Kimbler, D.E., Fessler, R.D., Shakir, B., Youssef, P., Yanasak, N., Vender, J.R., Dhandapani, K.M., 2014. High mobility group box protein-1 promotes cerebral edema after traumatic brain injury via activation of toll-like receptor 4. Glia 62, 26–38. Lotze, M.T., Tracey, K.J., 2005. High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal. Immunology 5, 331–342. Magna, M., Pisetsky, D.S., 2014. The role of HMGB1 in the pathogenesis of inflammatory and autoimmune diseases. Mol. Med. 20, 138–146. Matsuoka, N., Itoh, T., Watarai, H., Sekine-Kondo, E., Nagata, N., Okamoto, K., Mera, T., Yamamoto, H., Yamada, S., Maruyama, I., Taniguchi, M., Yasunami, Y., 2010. Highmobility group box 1 is involved in the initial events of early loss of transplanted islets in mice. J. Clin. Invest. 120, 735–743. Mc Guire, C., Prinz, M., Beyaert, R., van Loo, G., 2013. Nuclear factor kappa B (NF-kappaB) in multiple sclerosis pathology. Trends Mol. Med. 19, 604–613. Miller, S.D., Karpus, W.J., Davidson, T.S., 2010. Experimental Autoimmune Encephalomyelitis in the Mouse. Current protocols in immunology. In: Coligan, John E., et al. (Eds.), (Chapter 15, Unit 15 11). Muller, S., Ronfani, L., Bianchi, M.E., 2004. Regulated expression and subcellular localization of HMGB1, a chromatin protein with a cytokine function. J. Intern. Med. 255, 332–343. Murphy, A.C., Lalor, S.J., Lynch, M.A., Mills, K.H., 2010. Infiltration of Th1 and Th17 cells and activation of microglia in the CNS during the course of experimental autoimmune encephalomyelitis. Brain Behav. Immun. 24, 641–651. Neumann, H., 2001. Control of glial immune function by neurons. Glia 36, 191–199. Pisetsky, D.S., Erlandsson-Harris, H., Andersson, U., 2008. High-mobility group box protein 1 (HMGB1): an alarmin mediating the pathogenesis of rheumatic disease. Arthritis Res. Ther. 10, 209. Qiu, J., Nishimura, M., Wang, Y., Sims, J.R., Qiu, S., Savitz, S.I., Salomone, S., Moskowitz, M.A., 2008. Early release of HMGB-1 from neurons after the onset of brain ischemia. J. Cereb. Blood Flow Metab. 28, 927–938. Robinson, A.P., Caldis, M.W., Harp, C.T., Goings, G.E., Miller, S.D., 2013. High-mobility group box 1 protein (HMGB1) neutralization ameliorates experimental autoimmune encephalomyelitis. J. Autoimmun. 43, 32–43. Rouhiainen, A., Kuja-Panula, J., Wilkman, E., Pakkanen, J., Stenfors, J., Tuominen, R.K., Lepantalo, M., Carpen, O., Parkkinen, J., Rauvala, H., 2004. Regulation of monocyte migration by amphoterin (HMGB1). Blood 104, 1174–1182. Schwartz, D., Engelhard, D., Gallily, R., Matoth, I., Brenner, T., 1998. Glial cells production of inflammatory mediators induced by Streptococcus pneumoniae: inhibition by pentoxifylline, low-molecular-weight heparin and dexamethasone. J. Neurol. Sci. 155, 13–22. Streit, W.J., 2002. Microglia as neuroprotective, immunocompetent cells of the CNS. Glia 40, 133–139. Stromnes, I.M., Goverman, J.M., 2006. Active induction of experimental allergic encephalomyelitis. Nat. Protoc. 1, 1810–1819. Uzawa, A., Mori, M., Taniguchi, J., Masuda, S., Muto, M., Kuwabara, S., 2013. Anti-high mobility group box 1 monoclonal antibody ameliorates experimental autoimmune encephalomyelitis. Clin. Exp. Immunol. 172, 37–43. Xia, Q., Duan, L., Shi, L., Zheng, F., Gong, F., Fang, M., 2014. High-mobility group box 1 accelerates early acute allograft rejection via enhancing IL-17+ gammadelta T-cell response. Transpl. Int. 27, 399–407. Yang, H., Ochani, M., Li, J., Qiang, X., Tanovic, M., Harris, H.E., Susarla, S.M., Ulloa, L., Wang, H., DiRaimo, R., Czura, C.J., Wang, H., Roth, J., Warren, H.S., Fink, M.P., Fenton, M.J., Andersson, U., Tracey, K.J., 2004. Reversing established sepsis with antagonists of endogenous high-mobility group box 1. Proc. Natl. Acad. Sci. U. S. A. 101, 296–301. Zou, H., Yang, Y., Gao, M., Zhang, B., Ming, B., Sun, Y., Chen, H., Tang, X., Chen, Z., Xiong, P., Xu, Y., Fang, M., Tan, Z., Gong, F., Zheng, F., 2014. HMGB1 is involved in chronic rejection of cardiac allograft via promoting inflammatory-like mDCs. Am. J. Transplant. 14, 1765–1777.