- Email: [email protected]
High-mobility group box 1 contributes to mechanical allodynia and spinal astrocytic activation in a mouse model of type 2 diabetes
Brain Research Bulletin 88 (2012) 332–337
Contents lists available at SciVerse ScienceDirect
Brain Research Bulletin journal homepage: www.elsevier.com/locate/brainresbull
Research report
High-mobility group box 1 contributes to mechanical allodynia and spinal astrocytic activation in a mouse model of type 2 diabetes Peng-Cheng Ren a,1 , Yong Zhang a,1 , Xu-Dong Zhang a , Li-Jun An b , Hai-Gang Lv a , Jun He a , Chang-Jun Gao c,∗ , Xu-De Sun c,∗∗ a
Department of Orthopaedics, Tangdu Hospital, the Fourth Military Medical University, No. 1 Xinsi Road, Xi’an, Shaanxi Province 710038, China Department of Gerontology, Tangdu Hospital, the Fourth Military Medical University, No. 1 Xinsi Road, Xi’an, Shaanxi Province 710038, China c Department of Anesthesiology, Tangdu Hospital, the Fourth Military Medical University, No. 1 Xinsi Road, Xi’an, Shaanxi Province 710038, China b
a r t i c l e
i n f o
Article history: Received 21 November 2011 Received in revised form 13 February 2012 Accepted 11 March 2012 Available online 20 March 2012 Keywords: High-mobility group box 1 Diabetes Pain Spinal dorsal horn Astrocyte
a b s t r a c t Chronic pain is one of the most common complications of diabetes. However, current treatments for diabetic pain are usually unrealistic because the underlying mechanisms are far from being clear. Immerging studies have implicated immune factors as key players in the diabetic pain. High-mobility group box 1 (HMGB1) is an important mediator of inflammatory response, but its role in diabetic pain is unclear. In the present study, we observed that db/db mice (a model of type 2 diabetes) developed persistent mechanical allodynia from postnatal 2 months. Western blot showed that in postnatal 2–5 months, HMGB1 was significantly higher than that of the heterozygous littermates (db/+) mice. Intrathecal injection of a HMGB1 neutralizing antibody (anti-HMGB1) inhibited mechanical allodynia. Immunostaining data showed that compared with db/+ and C57 mice (postnatal 4 months), glial fibrillary acidic protein (GFAP) staining was significantly increased in the spinal cord of db/db mice. Anti-HMGB1 could effectively decrease GFAP expression. Real-time PCR showed that in postnatal 4 months, db/db mice induced significant increases of TNF-alpha, IL-1, IL-6 and monocyte chemoattractant protein-1 (MCP-1) in the spinal dorsal horn, while anti-HMGB1 (50 g) effectively inhibited the up-regulation of these inflammatory mediators. Our results indicate that HMGB1 is significantly up-regulated in the spinal cord of type 2 diabetes, and inhibiting HMGB1 may provide a novel treatment for diabetic pain. © 2012 Elsevier Inc. All rights reserved.
1. Introduction Chronic pain is one of the most common chronic complications of the type 2 diabetes [2,7]. Diabetic pain is a main course of decreased life quality in patients of diabetes. However, because of the poor understanding of the underlying mechanisms, current treatments for diabetic pain are usually unrealistic [2]. Like other types of chronic pain, diabetic pain has been described simply as a course of increased neuronal activities [3]. But immerging studies have implicated immune factors as key players in the induction and maintenance of chronic pain [9,27]. We previously observed that in a mouse model of type 2 diabetes (db/db mouse), spinal IL-1
Abbreviations: DRG, dorsal root ganglion; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GFAP, glial fibrillary acidic protein; HMGB1, high-mobility group box 1; MCP-1, monocyte chemoattractant protein-1; PWF, paw withdrawal frequency; RT-PCR, reverse transcription polymerase chain reaction; TLR, toll like receptor. ∗ Corresponding author. Tel.: +86 29 84777740; fax: +86 29 84777641. ∗∗ Corresponding author. Tel.: +86 29 84777539; fax: +86 29 84777641. E-mail addresses: [email protected] (C.-J. Gao), [email protected] (X.-D. Sun). 1 These authors contribute to the work equally. 0361-9230/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.brainresbull.2012.03.002
expression was significantly increased and astrocytes were obviously activated in db/db mice compared to normal mice. Intrathecal injection of an astrocytic toxin, cytokine inhibitor or interleukin-1 receptor antagonist could each significantly attenuate the allodynia [17]. These data indicate that neuroinflammation in the spinal dorsal horn may play an important role in diabetic pain. High-mobility group box 1 (HMGB1), a nonhistone DNA-binding molecule, was first discovered as a nuclear protein with rapid electrophoretic migration. However, HMGB1 is now implicated as an important mediator in inflammatory process [18]. In many disorders in the central nervous system, including brain ischemia, spinal cord injury and amyotrophic lateral sclerosis [1,15], HMGB1 has been found to play essential roles. In addition, it was reported that intrathecal administration of HMGB1 produced mechanical allodynia [5]. Very interestingly, recent studies suggest that HMGB1 plays an active role in bone cancer pain and peripheral nerve injuryinduced neuropathic pain [22,25]. However, it is still unknown whether HMGB1 is involved in diabetic pain. In the present study, based on the mouse model of type 2 diabetes (db/db mouse) we previously used [17], we firstly observed the time course of HMGB1 expression in the spinal dorsal horn. To further confirm the role of HMGB1 in diabetic pain, a
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
333
Table 1 Primer sequence for the rat genes characterized in this experiment. Genes
Primers
Sequences
Accession number
TNF-alpha
Forward primer Reverse primer
5 -TGATCGGTCCCAACAAGG A-3 5 -TGCTTG GTG GTTTGCTACGA-3
AY427675
IL-1
Forward primer Reverse primer
5 -TGCTGATGTACCAGTTGGGG-3 5 -CTCCATGAGCTTTGTACAAG-3
NM031512
IL-6
Forward primer Reverse primer
5 -GCCCTTCAGGAACAGCTATG-3 5 -CAGAATTGCCATTGCACAAC-3
NM012589
MCP-1
Forward primer Reverse primer
5 -CAGATCTCTCTTCCTCCACCACTAT-3’ 5 -CAGGCAGCAACTGTGAACAAC-3
M57441
GAPDH
Forward primer Reverse primer
5 -CCCCCAATGTATCCGTTGTG-3 5 -TAGCCCAGGATGCCCTTTAGT-3
NM01008
neutralizing antibody against HMGB1 was intrathecally injected into the diabetic mice and the behavioral consequences were observed. Furthermore, we detected the astrocytic activation and expression of inflammatory mediators in the spinal cord of diabetic mice after inhibiting HMGB1 expression. 2. Materials and methods 2.1. Animals All experimental procedures received prior approval from the Animal Use and Care Committee for Research and Education of the Fourth Military Medical University (Xi’an, PR China) and also the ethical guidelines to investigate experimental pain in conscious animals [30]. All efforts were made to minimize animals’ suffering and to reduce the number of animals used. Homozygous diabetic mice (db/db) (used as a model of type 2 diabetes) and heterozygous littermates (db/+) of C57BLKS strain (nondiabetic control) were purchased from Jackson Laboratory (stock number 000662; Bar Harbor, Maine, USA). The characterization (blood glucose and body weight) of the 2 diabetic mice was described in our previous study [17]. Age-matched C57BL6 mice were obtained from the Fourth Military Medical University (Xi’an, PR China). Mice were housed in plastic cages, three in each cage, and maintained on a 12:12 h light/dark cycle under conditions of 22–25 ◦ C ambient temperature with food and water available. 2.2. Intrathecal injection According to a previous report [13], we performed spinal cord puncture between L5 and L6 vertebrae with 10 l Gastight® syringe (Hamilton, USA) and a BD Precisionglide® 30G (1/2) in. needle (Becton Dickinson, USA) under sevofluorane anesthesia. Immediately after the needle entry into subarachnoid space, a brisk tail flick could be observed. Polyclonal neutralizing antibodies against HMGB1 B box were raised in rabbits and were affinity-purified using cyanogen bromide-activated Sepharose beads following standard procedures. Single injection of 3 l of antiHMGB1 (1, 10 and 50 g; n = 6 each) or nonimmuned rabbit IgG (Sigma, St. Louis, MO, USA; n = 6, vehicle control) was performed in the end of postnatal 4 months. 2.3. Behavioral testing Diabetic pain in mice was described as mechanical allodynia, which was evaluated by paw withdrawal frequencies (PWFs) to repeated mechanical stimuli. The procedures were described previously [29]. Mouse was placed in a Plexiglas chamber on an elevated mesh screen. Two calibrated von Frey monofilaments (0.24 and 4.33 mN; Stoelting Co., Wood Dale, IL, USA) were employed. Each filament was applied to the hind paw for approximately 1 s, and each trial was repeated 10 times to both hind paws. The data were expressed as a percentage of response frequency [(number of paw withdrawals/10 trials) × 100 = % response frequency], and this percentage was used as an indication of the amount of paw withdrawal. Behavioral testing was performed in a double-blind manner by those who were blind to the animals grouping. In the time-course study, PWF was performed at postnatal 1–5 months. To evaluate the effect of the HMGB1 antibody on pain behavior, PWF was performed at 1 h, 2 h, 4 h, 8 h and 1 d after intrathecal injection. 2.4. Western blot Mice were sacrificed and spinal cord was rapidly collected and put in liquid nitrogen. The L5 spinal cord segment was dissected on dry ice according to the termination of the L4 and L5 dorsal roots. The segment was split into the dorsal and ventral horns at the level of the central canal. The tissues were homogenized in chilled lysis buffer (50 mM Tris, 1 mM phenylmethylsulfonyl fluoride, 1 mM
EDTA, 1 M leupeptin). The crude homogenate was centrifuged at 4 ◦ C for 15 min at 1000 × g. The supernatant was collected and the protein concentration was measured. The electrophoresis samples were heated at 100 ◦ C for 5 min and loaded onto 10% SDS–polyacrylamide gels with standard Laemmli solutions (Bio-Rad Laboratories, CA, USA). The proteins were electroblotted onto a polyvinylidene difluoride membrane (PVDF, Millipore, Billerica, MA, USA). The membranes were placed in a blocking solution containing Tris-buffered saline with 0.02% Tween (TBS-T) and 5% non-fat dry milk, for 1 h, and incubated overnight under gentle agitation with primary antibodies: rabbit anti-HMGB1 (1:1000; BD Pharmingen, USA), mouse antiglial fibrillary acidic protein (GFAP; 1:2000; Millipore) or goat anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH, 1: 1000; Santa-Cruz Biotechnology, Santa Cruz, CA, USA). Bound primary antibodies were detected with the anti-rabbit, antimouse or anti-goat horseradish peroxidase (HRP)-conjugated secondary antibody (1:10,000; Amersham Pharmacia Biotech Inc., Piscataway, NJ, USA). All reactions were detected by the enhanced chemiluminescence (ECL) detection method (Amersham). The densities of protein blots were analyzed with Image G software. The densities of target proteins and GAPDH immunoreactive bands were quantified with background subtraction. The same size of square was drawn around each band to measure the density. Target protein levels were normalized against GAPDH levels and expressed as relative fold changes compared to the C57 control group. 2.5. Immunohistochemistry Mice were perfused with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.3) through the ascending aorta after deep anesthesia by pentobarbital (60 mg/kg, i.p.). The L5 spinal cords were collected, post-fixed in the same fixative solution for 2 h, cryoprotected for 24 h at 4 ◦ C in 0.1 M phosphate buffer containing 30% sucrose, and frozen-sectioned at 10 m. After being blocked in 0.01 M phosphate-buffered saline containing 0.3% Triton X-100 and 10% goat serum for 1 h at room temperature, the sections were incubated with primary mouse anti-GFAP (1:2000; Millipore) overnight at 4 ◦ C and then with FITC-conjugated horse anti-mouse IgG (1:200; Vector, Burlingame, CA, USA) for 1 h at room temperature. Images were obtained using a confocal laser microscope (FV1000; Olympus, Tokyo, Japan) and digital images were captured with Fluoview 1000 (Olympus). Five nonadjacent sections from the L5 segments were selected randomly. Images were evaluated by a computer-assisted image analysis program (MetaMorph 6.1). To measure the area of GFAP immunopositive somata, we used the Threshold Image function in Measure of MetaMorph 6.1 to set the low and high thresholds for the immunofluorescent intensity which was determined to be a signal. Our image data were collected using the same region and the same size of field within same lamina to avoid any variance and difference in staining between laminae. A standardized field area was sampled arbitrarily from regions within randomly selected dorsal horn sections. Then the immunoreactivities for GFAP within the superficial dorsal horn were averaged across the five spinal sections for each experimental group. 2.6. Real-time reverse transcription polymerase chain reaction (RT-PCR) After deeply anesthetized with sodium pentobarbital (60 mg/kg, i.p.), mice were sacrificed. L5 spinal dorsal horn was rapidly harvested and total RNA was extracted with Trizol (GIBCO/BRL Life Technologies Inc., Grand Island, NY, USA). Complementary DNA (cDNA) was synthesized with oligo (dT)12–18 using SuperscriptTM III Reverse Transcriptase for RT-PCR (Invitrogen, Carlsbad, CA, USA). The primers used were presented in Table 1. Equal amounts of RNA (1 g) were used to prepare cDNA using the SYBR® Premix Ex TaqTM (Takara, Tokyo, Japan) and analyzed by real-time PCR in a detection system (Applied Biosystems, Foster City, CA, USA). The amplification protocol was: 3 min at 95 ◦ C, followed by 42 cycles of 10 s at 95 ◦ C for denaturation and 45 s at 60 ◦ C for annealing and extension. Target cDNA quantities were estimated from the threshold amplification cycle number (Ct) using Sequence Detection System software (Applied Biosystems). Data were analyzed by those who
334
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
were blind with respect to group assignment. A Ct value was calculated for each sample by subtracting their Ct value from the Ct value for the corresponding GAPDH to normalize the differences in cDNA aliquots. Each cDNA quantity was then calculated with the following formula: 2Ct . GAPDH was served as an internal standard control for variations in RT-PCR efficiency. In each experiment, PCRs were done in triplicate and all experiments were repeated twice. 2.7. Statistical analysis All data were collected by experimenters blinded to the surgery and reagents treatments and statistical analyses were done by using SPSS® version 16.0 software (SPSS Inc., Chicago, IL, USA). Data were expressed as mean ± standard error mean (mean ± SEM). Two-way analysis of variance (ANOVA) (Group × time) was used for analysis of behavioral data as well as morphological analysis. Dunnett’s C post hoc test, assuming the variances were not equal, was employed whenever appropriate and significance was set at 0.05 for all statistical tests.
3. Results 3.1. Mechanical allodynia in the type 2 diabetic mice In our previous study, we reported that compared with db/+ and C57 mice, the body weight and blood glucose of db/db mice was significantly higher throughout the period tested, suggesting that db/db mice developed diabetic features [17]. Behavioral study showed that from postnatal 2 months, the PWF of db/db mice was significantly increased, compared with db/+ and C57 mice (P < 0.05; Fig. 1), suggesting that diabetic mice significantly developed mechanical allodynia. Then the mechanical allodynia continued thereafter throughout the period tested. Db/+ and C57 mice did not display mechanical pain hypersensitivity at any observed time points. In a recent study [21], locomotor function of the db/db mice was evaluated and these mice did not exhibit gross motor coordination defects, which confirmed that the performance in pain behavior tests was not caused by the locomotor functions. 3.2. HMGB1 is increased in the spinal dorsal horn of diabetic mice and contributes to diabetic pain To explore the potential roles of HMGB1 in diabetic pain, we examined HMGB1 expression in different postnatal time points with Western blot. HMGB1 expression in the C57 and db/+ mice was not significantly different at any observed postnatal time points (data not shown). Spinal expression of HMGB1 in diabetic mice was expressed as fold changes compared with that of C57 mice at postnatal 5 months. The protein level of HMGB1 was not increased until postnatal 2 months and gradually up-regulated till the end of the test. In postnatal 2–5 months, HMGB1 was 2.33 ± 0.42 folds, 2.41 ± 0.35 folds, 2.87 ± 0.42 folds, and 3.13 ± 0.64 folds of C57 mice, which were significantly higher than that of the C57 and db/+ mice (Fig. 1C, P < 0.05). In order to further confirm the involvement of HMGB1 in diabetic pain, we injected a neutralizing antibody of HMGB1 (anti-HMGB1) and observed the behavioral consequences. Treatment with anti-HMGB1 significantly reversed mechanical allodynia induced by diabetes. Intrathecally administering 1 g of antiHMGB1 could decrease PWF slightly, but there were no statistical differences compared to that of vehicle group at any time points (Fig. 2A, B, P > 0.05). However, 10 g of anti-HMGB1 could downregulate PWF significantly and this anti-allodynic effect lasted to 1 d after administration and 50 g of anti-HMGB1 produced a more significant effect (Fig. 2A, B, P < 0.05). To validate the efficacy of neutralizing antibody, at 4 h after antibody injection, rats were sacrificed and Western blot analysis was performed to evaluate the expression of HMGB1. Injection of 1 g of anti-HMGB1 could inhibit HMGB1 slightly, but not significantly compared to that of vehicle group. However, 10 g or 50 g of anti-HMGB1 significantly
Fig. 1. (A–B) Paw withdrawal frequencies in response to von Frey filaments with low (0.24 mN, A) and moderate (4.33 mN, B) intensities. C, Western blot analysis on the time course expression of HMGB1 in the spinal dorsal horn of C57 (postnatal 5 months), db/+ (postnatal 5 months) and db/db (postnatal 1–5 months) mice. The optic densities of HMGB1 are normalized to that of C57 mice at corresponding time points. All data are expressed as mean ± SEM. *P < 0.05 vs C57 mice. db/db: homozygous type 2 diabetic db/db mice; db/+: heterozygous nondiabetic littermates; C57: C57BL6 mice.
blocked the expression of HMGB1. These results indicated that antiHMGB1 treatment showed an effective and reliable anti-allodynic effect in a dose-dependent manner and this effect was through inhibiting HMGB1 expression.
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
335
induction of a number of inflammatory mediators in astrocytes [20]. In the present study, we tried to demonstrate whether up-regulated HMGB1 contributed to astrocytic activation and mechanical allodynia. Immunohistochemistry indicated that compared with db/+ and C57 mice (postnatal 4 months), GFAP staining was significantly increased in the spinal cord of db/db mice at the same postnatal period. Intrathecally administering 1 g of antiHMGB1 could decrease GFAP expression slightly, although not significantly (Fig. 3A–G, P > 0.05). However, 10 g and 50 g of antiHMGB1 could inhibit GFAP immunodensity significantly (Fig. 3A–G, P < 0.05). We checked the expression of GFAP at different time following anti-HMGB1 (50 g) injection by Western blot analysis. The intensity of GFAP was slightly but not significantly decreased at 1 h but was remarkably down-regulated at 2–8 h. However, GFAP level returned to normal level at 1 d after anti-HMGB1 (50 g) injection (Fig. 3H, P < 0.05). These results showed that anti-HMGB1 treatment could inhibit astrocytic activation in the diabetic mice. Astrocytic activation in diabetic mice is likely to enhance gene transcription and consequently to modulate synthesis of inflammatory mediators such as IL-1, TNF-alpha, monocyte chemotactic protein (MCP-1), IL-6, which positively contribute to pain sensitization [8]. With real-time RT-PCR, we observed that in postnatal 4 months, db/db but not db/+ or C57 mice induced significant increases of TNF-alpha (2.16 ± 0.25 folds of C57, P < 0.05), IL-1 (3.28 ± 0.21 folds of C57, P < 0.05), IL-6 (2.23 ± 0.11 folds of C57, P < 0.05) and MCP-1 (1.83 ± 0.21 folds of C57, P < 0.05) in the spinal dorsal horn (Fig. 4A). Treatments with HMGB1 antibody (50 g) but not vehicle control, significant inhibited TNF-alpha (1.56 ± 0.14 folds of C57, P < 0.05), IL-1 (2.18 ± 0.15 folds of C57, P < 0.05), IL-6 (1.33 ± 0.18 folds of C57, P < 0.05) and MCP-1 (1.37 ± 0.22 folds of C57, P < 0.05). We also examined the mRNA levels of these inflammatory mediators at different time intervals following anti-HMGB1 (50 g) injection. TNF-alpha was significantly lower than the baseline value throughout the observing period (1 h to 1 d), while IL-1 was not obviously lower at 1 h. IL-6 and MCP-1 expressions were only remarkably decreased from 2 h to 8 h after anti-HMGB1 (50 g) injection (Fig. 4B). These results suggested that spinal HMGB1 up-regulation in diabetic mice contributed to synthesis of inflammatory mediators.
4. Discussion
Fig. 2. Behavioral consequences following intrathecal injection of the neutralizing antibody against HMGB1 (anti-HMGB1). Paw withdrawal frequencies in response to von Frey filaments with low (0.24 mN, A) and moderate (4.33 mN, B) intensities. Intrathecally injecting 10 and 50 g of anti-HMGB1 significantly reversed mechanical allodynia and this allodynia effect lasted to 1 d after injection. (C) Western blot analysis on the quality of neutralizing antibody on HMGB1 expression 4 h after injection. Ab, neutralizing antibody against HMGB1. *P < 0.05 vs vehicle control group.
3.3. HMGB1 contributes to spinal astrocytic activation and inflammatory response in diabetic mice We previously observed that astrocytic activation contributed to mechanical allodynia in the db/db mouse [17] and an in vitro study suggested that HMGB1 could induce transcriptional/translational
In the present study, we provide evidence of the novel contribution of spinal HMGB1 in diabetic pain. We observed that HMGB1 is significantly increased in the spinal dorsal horn of type 2 diabetic mice and HMGB1 expression correlates with mechanical allodynia. In addition, blocking HMGB1 could reverse diabetic pain with decreased astrocytic activation and down-regulated expression of inflammatory mediators. Cumulating data suggest that HMGB1 is not only a nuclear protein, but also an important mediator in inflammatory response. After stimulation, HMGB1 can be released into the extracellular space either actively or passively. Recent studies suggest that in conditions such as oxidative stress, HMGB1 could be released [24]. In diabetes, hyperglycemia was reported to not only directly induce oxidative stress [23], but also contribute to increased aldose reductase activity, nerve hypoxia/ischemia, activation of protein kinase C, and insulin-like growth factor deficiency [6]. These pathological pathways all in turn converge in producing dramatical oxidative stress [10]. Thus we hypothesize that hyperglycemia-induced oxidative stress may be a key mechanism for the initial increase of HMGB1. Released HMGB1 binds to its receptors, including the receptor for advanced glycation end products (RAGE), Toll-like receptor2 (TLR2), and TLR4, which are highly expressed on immune
336
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
Fig. 3. Diabetes-induced astrocytic activation in the spinal dorsal horn is revered by intrathecal anti-HMGB1 (Ab) treatment. (A–C) Compared to the C57 (postnatal 4 months) or db/+ (postnatal 4 months) mice, expression of glial fibrillary acidic protein (GFAP) in the spinal dorsal horn are significantly higher at the same postnatal period. Enhanced staining appeared in all areas of the spinal grey matter, with the most prominent increase in superficial dorsal horn laminae I–III. Diabetes-induced up-regulation of GFAP is significantly attenuated by the anti-HMGB1 (Ab) treatment (D–F), compared with the vehicle control. (G) Statistical results of the immunostaining data. *P < 0.05 vs vehicle control group; #P < 0.05 vs C57 group. (H) The expression of GFAP at different time points following anti-HMGB1 (10 g) injection indicated by Western blot analysis. *P < 0.05 vs control value (before injection).
cells [19]. Receptor activation leads to the consequent release of cytokines/chemokines, including TNF-alpha, IL-1, IL-6, and MCP1 [18,28]. In the present study, we also observed that inhibiting HMGB1 expression with the specific neutralizing antibody could significantly block the expression of these cytokines/chemokines. Previous studies have confirmed the positive contribution of these cytokines/chemokines to chronic pain. For example, it was previous reported that through binding with its receptor on the neurons, IL-1 increases calcium conductivity of neuronal NMDA receptor, possibly via intracellular signaling pathways leading to the phosphorylation of NMDA receptor NR1 subunit [12]. Activation of NMDA receptor in the spinal dorsal horn plays a pivotal role in central sensitization and thus chronic pain [16]. In addition to triggering release of proinflammatory cytokines in spinal dorsal horn, HMGB1 could also be released after cytokines stimulation [22]. So there is a positive feedback circuit between HMGB1 and proinflammatory cytokines. This positive feedback circuit enlarges the effect of HMGB1 on diabetic pain and makes it more difficult to be treated by clinical therapy. In the present study, we also observed a remarkable astrocytic activation in diabetic mice. Interestingly, inhibiting HMGB1 could significantly block astrocytic activation. As we have discussed above, HMGB1 could binds to TLR2 and TLR4, which are
also expressed on spinal astrocytes. We thus assume that HMGB1 may activate astrocytes through these two TLRs. Great deals of studies have been performed targeting TLRs as critical mediators of chronic pain [11]. It has been reported that after nerve injury, TLR2 is required for proinflammatory cytokines expression and pain hypersensitivity [14] and up-regulated TLR4 in the spinal cord are required for glial activation and pain response [4]. Another important receptor of HMGB1, RAGE, has not been reported to be expressed on spinal astrocytes. However, a recent study showed that HMGB1 is expressed on GFAP-positive satellite cells in dorsal root ganglia [15]. Previous studies have reported that RAGE is significantly increased in diabetes [26]. So RAGE may also mediate the pronociceptive effect of HMGB1 in diabetic pain. In summary, we observed that type 2 diabetes induced a significant increase of HMGB1, which could activate spinal astrocytes and expression of inflammatory cytokines and chemokines, which could modulate NMDAR-mediated synaptic transmission and pain response. In addition, cytokines and/or chemokines further increase release of HMGB1 in the spinal cord, and enhance nociceptive transmission in a manner of positive feedback. Our data thus indicate that HMGB1 may play an important role in diabetic pain and inhibiting HMGB1 expression may provide a novel treatment for diabetic pain.
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
[5]
[6] [7]
[8] [9] [10] [11]
[12]
[13] [14]
[15]
[16] [17]
[18] Fig. 4. Effects of anti-HMGB1 (Ab) treatment on diabetes-induced cytokines/chemokine expression revealed by real-time RT-PCR. Data are normalized against GAPDH levels and expressed as relative fold changes compared to the C57 group. (A) Compare with C57 (postnatal 4 months) and db/+ (postnatal 4 months) mice, expression of TNF-alpha, IL-1, IL-6 and monocyte chemotactic protein (MCP-1) in db/db mice is significantly higher at the same postnatal period. *P < 0.05 vs C57 group; #P < 0.05 vs vehicle control group. (B) The expression of cytokines/chemokine at different time points following anti-HMGB1 (10 g) injection. *P < 0.05 vs control value (before injection).
[19]
[20]
[21]
[22]
Conflicts of interest The authors declare that there are no conflicts of interest. Acknowledgments
[23] [24] [25]
This work was supported by the National Natural Science Foundation of China (No. 81071527) and the Chinese Postdoctoral Science Foundation (No. 201003743). References [1] A. Andersson, R. Covacu, D. Sunnemark, A.I. Danilov, A. Dal Bianco, M. Khademi, E. Wallstrom, A. Lobell, L. Brundin, H. Lassmann, R.A. Harris, Pivotal advance: HMGB1 expression in active lesions of human and experimental multiple sclerosis, J. Leukoc. Biol. 84 (2008) 1248–1255. [2] A.M. Barrett, M.A. Lucero, T. Le, R.L. Robinson, R.H. Dworkin, A.S. Chappell, Epidemiology, public health burden, and treatment of diabetic peripheral neuropathic pain: a review, Pain Med. 8 (Suppl 2) (2007) S50–S62. [3] A.I. Basbaum, D.M. Bautista, G. Scherrer, D. Julius, Cellular and molecular mechanisms of pain, Cell 139 (2009) 267–284. [4] I. Bettoni, F. Comelli, C. Rossini, F. Granucci, G. Giagnoni, F. Peri, B. Costa, Glial TLR4 receptor as new target to treat neuropathic pain: efficacy of a new
[26]
[27]
[28]
[29]
[30]
337
receptor antagonist in a model of peripheral nerve injury in mice, Glia 56 (2008) 1312–1319. M. Chacur, E.D. Milligan, L.S. Gazda, C. Armstrong, H. Wang, K.J. Tracey, S.F. Maier, L.R. Watkins, A new model of sciatic inflammatory neuritis (SIN): induction of unilateral and bilateral mechanical allodynia following acute unilateral peri-sciatic immune activation in rats, Pain 94 (2001) 231–244. D. Choudhury, M. Tuncel, M. Levi, Diabetic nephropathy – a multifaceted target of new therapies, Discov. Med. 10 (2011) 406–415. C.J. Currie, C.D. Poole, A. Woehl, C.L. Morgan, S. Cawley, M.D. Rousculp, M.T. Covington, J.R. Peters, The financial costs of healthcare treatment for people with Type 1 or Type 2 diabetes in the UK with particular reference to differing severity of peripheral neuropathy, Diabet. Med. 24 (2007) 187–194. Y.J. Gao, R.R. Ji, Activation of JNK pathway in persistent pain, Neurosci. Lett. 437 (2008) 180–183. Y.J. Gao, R.R. Ji, Chemokines, neuronal–glial interactions, and central processing of neuropathic pain, Pharmacol. Ther. 126 (2010) 56–68. F. Giacco, M. Brownlee, Oxidative stress and diabetic complications, Circ. Res. 107 (2010) 1058–1070. L.H. Guo, H.J. Schluesener, The innate immunity of the central nervous system in chronic pain: the role of Toll-like receptors, Cell. Mol. Life Sci. 64 (2007) 1128–1136. W. Guo, H. Wang, M. Watanabe, K. Shimizu, S. Zou, S.C. LaGraize, F. Wei, R. Dubner, K. Ren, Glial–cytokine–neuronal interactions underlying the mechanisms of persistent pain, J. Neurosci. 27 (2007) 6006–6018. J.L. Hylden, G.L. Wilcox, Intrathecal morphine in mice: a new technique, Eur. J. Pharmacol. 67 (1980) 313–316. D. Kim, M.A. Kim, I.H. Cho, M.S. Kim, S. Lee, E.K. Jo, S.Y. Choi, K. Park, J.S. Kim, S. Akira, H.S. Na, S.B. Oh, S.J. Lee, A critical role of toll-like receptor 2 in nerve injury-induced spinal cord glial cell activation and pain hypersensitivity, J. Biol. Chem. 282 (2007) 14975–14983. J.B. Kim, C.M. Lim, Y.M. Yu, J.K. Lee, Induction and subcellular localization of high-mobility group box-1 (HMGB1) in the postischemic rat brain, J. Neurosci. Res. 86 (2008) 1125–1131. M. Larsson, J. Broman, Synaptic plasticity and pain: role of ionotropic glutamate receptors, Neuroscientist (2010). Y.H. Liao, G.H. Zhang, D. Jia, P. Wang, N.S. Qian, F. He, X.T. Zeng, Y. He, Y.L. Yang, D.Y. Cao, Y. Zhang, D.S. Wang, K.S. Tao, C.J. Gao, K.F. Dou, Spinal astrocytic activation contributes to mechanical allodynia in a mouse model of type 2 diabetes, Brain Res. 1368 (2010) 324–335. M.T. Lotze, K.J. Tracey, High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal, Nat. Rev. Immunol. 5 (2005) 331–342. J.A. Nogueira-Machado, C.M. Volpe, C.A. Veloso, M.M. Chaves, HMGB1, TLR and RAGE: a functional tripod that leads to diabetic inflammation, Expert Opin. Ther. Targets 15 (2011) 1023–1035. M. Pedrazzi, M. Patrone, M. Passalacqua, E. Ranzato, D. Colamassaro, B. Sparatore, S. Pontremoli, E. Melloni, Selective proinflammatory activation of astrocytes by high-mobility group box 1 protein signaling, J. Immunol. 179 (2007) 8525–8532. Y.S. Ren, N.S. Qian, Y. Tang, Y.H. Liao, Y.L. Yang, K.F. Dou, M. Toi, Sodium channel Nav1.6 is up-regulated in the dorsal root ganglia in a mouse model of type 2 diabetes, Brain Res. Bull. 87 (2012) 244–249. M. Shibasaki, M. Sasaki, M. Miura, K. Mizukoshi, H. Ueno, S. Hashimoto, Y. Tanaka, F. Amaya, Induction of high mobility group box-1 in dorsal root ganglion contributes to pain hypersensitivity after peripheral nerve injury, Pain 149 (2010) 514–521. D.K. Singh, P. Winocour, K. Farrington, Oxidative stress in early diabetic nephropathy: fueling the fire, Nat. Rev. Endocrinol. 7 (2011) 176–184. D. Tang, R. Kang, H.J. Zeh 3rd, M.T. Lotze, High-mobility group box 1, oxidative stress, and disease, Antioxid. Redox Signal. 14 (2011) 1315–1335. W. Tong, W. Wang, J. Huang, N. Ren, S.X. Wu, Y.Q. Li, Spinal high-mobility group box 1 contributes to mechanical allodynia in a rat model of bone cancer pain, Biochem. Biophys. Res. Commun. 395 (2010) 572–576. C. Toth, L.L. Rong, C. Yang, J. Martinez, F. Song, N. Ramji, V. Brussee, W. Liu, J. Durand, M.D. Nguyen, A.M. Schmidt, D.W. Zochodne, Receptor for advanced glycation end products (RAGEs) and experimental diabetic neuropathy, Diabetes 57 (2008) 1002–1017. W. Wang, X. Mei, J. Huang, Y. Wei, Y. Wang, S. Wu, Y. Li, Crosstalk between spinal astrocytes and neurons in nerve injury-induced neuropathic pain, PLoS ONE 4 (2009) e6973. W. Wang, X.P. Mei, Y.Y. Wei, M.M. Zhang, T. Zhang, L.X. Xu, S.X. Wu, Y.Q. Li, Neuronal NR2B-containing NMDA receptor mediates spinal astrocytic c-Jun N-terminal kinase activation in a rat model of neuropathic pain, Brain Behav. Immun. 25 (2011) 1355–1366. W. Wang, R.S. Petralia, K. Takamiya, J. Xia, Y.Q. Li, R.L. Huganir, Y.X. Tao, M. Yaster, Preserved acute pain and impaired neuropathic pain in mice lacking protein interacting with C Kinase 1, Mol. Pain 7 (2011) 11. M. Zimmermann, Ethical guidelines for investigations of experimental pain in conscious animals, Pain 16 (1983) 109–110.
Contents lists available at SciVerse ScienceDirect
Brain Research Bulletin journal homepage: www.elsevier.com/locate/brainresbull
Research report
High-mobility group box 1 contributes to mechanical allodynia and spinal astrocytic activation in a mouse model of type 2 diabetes Peng-Cheng Ren a,1 , Yong Zhang a,1 , Xu-Dong Zhang a , Li-Jun An b , Hai-Gang Lv a , Jun He a , Chang-Jun Gao c,∗ , Xu-De Sun c,∗∗ a
Department of Orthopaedics, Tangdu Hospital, the Fourth Military Medical University, No. 1 Xinsi Road, Xi’an, Shaanxi Province 710038, China Department of Gerontology, Tangdu Hospital, the Fourth Military Medical University, No. 1 Xinsi Road, Xi’an, Shaanxi Province 710038, China c Department of Anesthesiology, Tangdu Hospital, the Fourth Military Medical University, No. 1 Xinsi Road, Xi’an, Shaanxi Province 710038, China b
a r t i c l e
i n f o
Article history: Received 21 November 2011 Received in revised form 13 February 2012 Accepted 11 March 2012 Available online 20 March 2012 Keywords: High-mobility group box 1 Diabetes Pain Spinal dorsal horn Astrocyte
a b s t r a c t Chronic pain is one of the most common complications of diabetes. However, current treatments for diabetic pain are usually unrealistic because the underlying mechanisms are far from being clear. Immerging studies have implicated immune factors as key players in the diabetic pain. High-mobility group box 1 (HMGB1) is an important mediator of inflammatory response, but its role in diabetic pain is unclear. In the present study, we observed that db/db mice (a model of type 2 diabetes) developed persistent mechanical allodynia from postnatal 2 months. Western blot showed that in postnatal 2–5 months, HMGB1 was significantly higher than that of the heterozygous littermates (db/+) mice. Intrathecal injection of a HMGB1 neutralizing antibody (anti-HMGB1) inhibited mechanical allodynia. Immunostaining data showed that compared with db/+ and C57 mice (postnatal 4 months), glial fibrillary acidic protein (GFAP) staining was significantly increased in the spinal cord of db/db mice. Anti-HMGB1 could effectively decrease GFAP expression. Real-time PCR showed that in postnatal 4 months, db/db mice induced significant increases of TNF-alpha, IL-1, IL-6 and monocyte chemoattractant protein-1 (MCP-1) in the spinal dorsal horn, while anti-HMGB1 (50 g) effectively inhibited the up-regulation of these inflammatory mediators. Our results indicate that HMGB1 is significantly up-regulated in the spinal cord of type 2 diabetes, and inhibiting HMGB1 may provide a novel treatment for diabetic pain. © 2012 Elsevier Inc. All rights reserved.
1. Introduction Chronic pain is one of the most common chronic complications of the type 2 diabetes [2,7]. Diabetic pain is a main course of decreased life quality in patients of diabetes. However, because of the poor understanding of the underlying mechanisms, current treatments for diabetic pain are usually unrealistic [2]. Like other types of chronic pain, diabetic pain has been described simply as a course of increased neuronal activities [3]. But immerging studies have implicated immune factors as key players in the induction and maintenance of chronic pain [9,27]. We previously observed that in a mouse model of type 2 diabetes (db/db mouse), spinal IL-1
Abbreviations: DRG, dorsal root ganglion; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GFAP, glial fibrillary acidic protein; HMGB1, high-mobility group box 1; MCP-1, monocyte chemoattractant protein-1; PWF, paw withdrawal frequency; RT-PCR, reverse transcription polymerase chain reaction; TLR, toll like receptor. ∗ Corresponding author. Tel.: +86 29 84777740; fax: +86 29 84777641. ∗∗ Corresponding author. Tel.: +86 29 84777539; fax: +86 29 84777641. E-mail addresses: [email protected] (C.-J. Gao), [email protected] (X.-D. Sun). 1 These authors contribute to the work equally. 0361-9230/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.brainresbull.2012.03.002
expression was significantly increased and astrocytes were obviously activated in db/db mice compared to normal mice. Intrathecal injection of an astrocytic toxin, cytokine inhibitor or interleukin-1 receptor antagonist could each significantly attenuate the allodynia [17]. These data indicate that neuroinflammation in the spinal dorsal horn may play an important role in diabetic pain. High-mobility group box 1 (HMGB1), a nonhistone DNA-binding molecule, was first discovered as a nuclear protein with rapid electrophoretic migration. However, HMGB1 is now implicated as an important mediator in inflammatory process [18]. In many disorders in the central nervous system, including brain ischemia, spinal cord injury and amyotrophic lateral sclerosis [1,15], HMGB1 has been found to play essential roles. In addition, it was reported that intrathecal administration of HMGB1 produced mechanical allodynia [5]. Very interestingly, recent studies suggest that HMGB1 plays an active role in bone cancer pain and peripheral nerve injuryinduced neuropathic pain [22,25]. However, it is still unknown whether HMGB1 is involved in diabetic pain. In the present study, based on the mouse model of type 2 diabetes (db/db mouse) we previously used [17], we firstly observed the time course of HMGB1 expression in the spinal dorsal horn. To further confirm the role of HMGB1 in diabetic pain, a
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
333
Table 1 Primer sequence for the rat genes characterized in this experiment. Genes
Primers
Sequences
Accession number
TNF-alpha
Forward primer Reverse primer
5 -TGATCGGTCCCAACAAGG A-3 5 -TGCTTG GTG GTTTGCTACGA-3
AY427675
IL-1
Forward primer Reverse primer
5 -TGCTGATGTACCAGTTGGGG-3 5 -CTCCATGAGCTTTGTACAAG-3
NM031512
IL-6
Forward primer Reverse primer
5 -GCCCTTCAGGAACAGCTATG-3 5 -CAGAATTGCCATTGCACAAC-3
NM012589
MCP-1
Forward primer Reverse primer
5 -CAGATCTCTCTTCCTCCACCACTAT-3’ 5 -CAGGCAGCAACTGTGAACAAC-3
M57441
GAPDH
Forward primer Reverse primer
5 -CCCCCAATGTATCCGTTGTG-3 5 -TAGCCCAGGATGCCCTTTAGT-3
NM01008
neutralizing antibody against HMGB1 was intrathecally injected into the diabetic mice and the behavioral consequences were observed. Furthermore, we detected the astrocytic activation and expression of inflammatory mediators in the spinal cord of diabetic mice after inhibiting HMGB1 expression. 2. Materials and methods 2.1. Animals All experimental procedures received prior approval from the Animal Use and Care Committee for Research and Education of the Fourth Military Medical University (Xi’an, PR China) and also the ethical guidelines to investigate experimental pain in conscious animals [30]. All efforts were made to minimize animals’ suffering and to reduce the number of animals used. Homozygous diabetic mice (db/db) (used as a model of type 2 diabetes) and heterozygous littermates (db/+) of C57BLKS strain (nondiabetic control) were purchased from Jackson Laboratory (stock number 000662; Bar Harbor, Maine, USA). The characterization (blood glucose and body weight) of the 2 diabetic mice was described in our previous study [17]. Age-matched C57BL6 mice were obtained from the Fourth Military Medical University (Xi’an, PR China). Mice were housed in plastic cages, three in each cage, and maintained on a 12:12 h light/dark cycle under conditions of 22–25 ◦ C ambient temperature with food and water available. 2.2. Intrathecal injection According to a previous report [13], we performed spinal cord puncture between L5 and L6 vertebrae with 10 l Gastight® syringe (Hamilton, USA) and a BD Precisionglide® 30G (1/2) in. needle (Becton Dickinson, USA) under sevofluorane anesthesia. Immediately after the needle entry into subarachnoid space, a brisk tail flick could be observed. Polyclonal neutralizing antibodies against HMGB1 B box were raised in rabbits and were affinity-purified using cyanogen bromide-activated Sepharose beads following standard procedures. Single injection of 3 l of antiHMGB1 (1, 10 and 50 g; n = 6 each) or nonimmuned rabbit IgG (Sigma, St. Louis, MO, USA; n = 6, vehicle control) was performed in the end of postnatal 4 months. 2.3. Behavioral testing Diabetic pain in mice was described as mechanical allodynia, which was evaluated by paw withdrawal frequencies (PWFs) to repeated mechanical stimuli. The procedures were described previously [29]. Mouse was placed in a Plexiglas chamber on an elevated mesh screen. Two calibrated von Frey monofilaments (0.24 and 4.33 mN; Stoelting Co., Wood Dale, IL, USA) were employed. Each filament was applied to the hind paw for approximately 1 s, and each trial was repeated 10 times to both hind paws. The data were expressed as a percentage of response frequency [(number of paw withdrawals/10 trials) × 100 = % response frequency], and this percentage was used as an indication of the amount of paw withdrawal. Behavioral testing was performed in a double-blind manner by those who were blind to the animals grouping. In the time-course study, PWF was performed at postnatal 1–5 months. To evaluate the effect of the HMGB1 antibody on pain behavior, PWF was performed at 1 h, 2 h, 4 h, 8 h and 1 d after intrathecal injection. 2.4. Western blot Mice were sacrificed and spinal cord was rapidly collected and put in liquid nitrogen. The L5 spinal cord segment was dissected on dry ice according to the termination of the L4 and L5 dorsal roots. The segment was split into the dorsal and ventral horns at the level of the central canal. The tissues were homogenized in chilled lysis buffer (50 mM Tris, 1 mM phenylmethylsulfonyl fluoride, 1 mM
EDTA, 1 M leupeptin). The crude homogenate was centrifuged at 4 ◦ C for 15 min at 1000 × g. The supernatant was collected and the protein concentration was measured. The electrophoresis samples were heated at 100 ◦ C for 5 min and loaded onto 10% SDS–polyacrylamide gels with standard Laemmli solutions (Bio-Rad Laboratories, CA, USA). The proteins were electroblotted onto a polyvinylidene difluoride membrane (PVDF, Millipore, Billerica, MA, USA). The membranes were placed in a blocking solution containing Tris-buffered saline with 0.02% Tween (TBS-T) and 5% non-fat dry milk, for 1 h, and incubated overnight under gentle agitation with primary antibodies: rabbit anti-HMGB1 (1:1000; BD Pharmingen, USA), mouse antiglial fibrillary acidic protein (GFAP; 1:2000; Millipore) or goat anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH, 1: 1000; Santa-Cruz Biotechnology, Santa Cruz, CA, USA). Bound primary antibodies were detected with the anti-rabbit, antimouse or anti-goat horseradish peroxidase (HRP)-conjugated secondary antibody (1:10,000; Amersham Pharmacia Biotech Inc., Piscataway, NJ, USA). All reactions were detected by the enhanced chemiluminescence (ECL) detection method (Amersham). The densities of protein blots were analyzed with Image G software. The densities of target proteins and GAPDH immunoreactive bands were quantified with background subtraction. The same size of square was drawn around each band to measure the density. Target protein levels were normalized against GAPDH levels and expressed as relative fold changes compared to the C57 control group. 2.5. Immunohistochemistry Mice were perfused with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.3) through the ascending aorta after deep anesthesia by pentobarbital (60 mg/kg, i.p.). The L5 spinal cords were collected, post-fixed in the same fixative solution for 2 h, cryoprotected for 24 h at 4 ◦ C in 0.1 M phosphate buffer containing 30% sucrose, and frozen-sectioned at 10 m. After being blocked in 0.01 M phosphate-buffered saline containing 0.3% Triton X-100 and 10% goat serum for 1 h at room temperature, the sections were incubated with primary mouse anti-GFAP (1:2000; Millipore) overnight at 4 ◦ C and then with FITC-conjugated horse anti-mouse IgG (1:200; Vector, Burlingame, CA, USA) for 1 h at room temperature. Images were obtained using a confocal laser microscope (FV1000; Olympus, Tokyo, Japan) and digital images were captured with Fluoview 1000 (Olympus). Five nonadjacent sections from the L5 segments were selected randomly. Images were evaluated by a computer-assisted image analysis program (MetaMorph 6.1). To measure the area of GFAP immunopositive somata, we used the Threshold Image function in Measure of MetaMorph 6.1 to set the low and high thresholds for the immunofluorescent intensity which was determined to be a signal. Our image data were collected using the same region and the same size of field within same lamina to avoid any variance and difference in staining between laminae. A standardized field area was sampled arbitrarily from regions within randomly selected dorsal horn sections. Then the immunoreactivities for GFAP within the superficial dorsal horn were averaged across the five spinal sections for each experimental group. 2.6. Real-time reverse transcription polymerase chain reaction (RT-PCR) After deeply anesthetized with sodium pentobarbital (60 mg/kg, i.p.), mice were sacrificed. L5 spinal dorsal horn was rapidly harvested and total RNA was extracted with Trizol (GIBCO/BRL Life Technologies Inc., Grand Island, NY, USA). Complementary DNA (cDNA) was synthesized with oligo (dT)12–18 using SuperscriptTM III Reverse Transcriptase for RT-PCR (Invitrogen, Carlsbad, CA, USA). The primers used were presented in Table 1. Equal amounts of RNA (1 g) were used to prepare cDNA using the SYBR® Premix Ex TaqTM (Takara, Tokyo, Japan) and analyzed by real-time PCR in a detection system (Applied Biosystems, Foster City, CA, USA). The amplification protocol was: 3 min at 95 ◦ C, followed by 42 cycles of 10 s at 95 ◦ C for denaturation and 45 s at 60 ◦ C for annealing and extension. Target cDNA quantities were estimated from the threshold amplification cycle number (Ct) using Sequence Detection System software (Applied Biosystems). Data were analyzed by those who
334
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
were blind with respect to group assignment. A Ct value was calculated for each sample by subtracting their Ct value from the Ct value for the corresponding GAPDH to normalize the differences in cDNA aliquots. Each cDNA quantity was then calculated with the following formula: 2Ct . GAPDH was served as an internal standard control for variations in RT-PCR efficiency. In each experiment, PCRs were done in triplicate and all experiments were repeated twice. 2.7. Statistical analysis All data were collected by experimenters blinded to the surgery and reagents treatments and statistical analyses were done by using SPSS® version 16.0 software (SPSS Inc., Chicago, IL, USA). Data were expressed as mean ± standard error mean (mean ± SEM). Two-way analysis of variance (ANOVA) (Group × time) was used for analysis of behavioral data as well as morphological analysis. Dunnett’s C post hoc test, assuming the variances were not equal, was employed whenever appropriate and significance was set at 0.05 for all statistical tests.
3. Results 3.1. Mechanical allodynia in the type 2 diabetic mice In our previous study, we reported that compared with db/+ and C57 mice, the body weight and blood glucose of db/db mice was significantly higher throughout the period tested, suggesting that db/db mice developed diabetic features [17]. Behavioral study showed that from postnatal 2 months, the PWF of db/db mice was significantly increased, compared with db/+ and C57 mice (P < 0.05; Fig. 1), suggesting that diabetic mice significantly developed mechanical allodynia. Then the mechanical allodynia continued thereafter throughout the period tested. Db/+ and C57 mice did not display mechanical pain hypersensitivity at any observed time points. In a recent study [21], locomotor function of the db/db mice was evaluated and these mice did not exhibit gross motor coordination defects, which confirmed that the performance in pain behavior tests was not caused by the locomotor functions. 3.2. HMGB1 is increased in the spinal dorsal horn of diabetic mice and contributes to diabetic pain To explore the potential roles of HMGB1 in diabetic pain, we examined HMGB1 expression in different postnatal time points with Western blot. HMGB1 expression in the C57 and db/+ mice was not significantly different at any observed postnatal time points (data not shown). Spinal expression of HMGB1 in diabetic mice was expressed as fold changes compared with that of C57 mice at postnatal 5 months. The protein level of HMGB1 was not increased until postnatal 2 months and gradually up-regulated till the end of the test. In postnatal 2–5 months, HMGB1 was 2.33 ± 0.42 folds, 2.41 ± 0.35 folds, 2.87 ± 0.42 folds, and 3.13 ± 0.64 folds of C57 mice, which were significantly higher than that of the C57 and db/+ mice (Fig. 1C, P < 0.05). In order to further confirm the involvement of HMGB1 in diabetic pain, we injected a neutralizing antibody of HMGB1 (anti-HMGB1) and observed the behavioral consequences. Treatment with anti-HMGB1 significantly reversed mechanical allodynia induced by diabetes. Intrathecally administering 1 g of antiHMGB1 could decrease PWF slightly, but there were no statistical differences compared to that of vehicle group at any time points (Fig. 2A, B, P > 0.05). However, 10 g of anti-HMGB1 could downregulate PWF significantly and this anti-allodynic effect lasted to 1 d after administration and 50 g of anti-HMGB1 produced a more significant effect (Fig. 2A, B, P < 0.05). To validate the efficacy of neutralizing antibody, at 4 h after antibody injection, rats were sacrificed and Western blot analysis was performed to evaluate the expression of HMGB1. Injection of 1 g of anti-HMGB1 could inhibit HMGB1 slightly, but not significantly compared to that of vehicle group. However, 10 g or 50 g of anti-HMGB1 significantly
Fig. 1. (A–B) Paw withdrawal frequencies in response to von Frey filaments with low (0.24 mN, A) and moderate (4.33 mN, B) intensities. C, Western blot analysis on the time course expression of HMGB1 in the spinal dorsal horn of C57 (postnatal 5 months), db/+ (postnatal 5 months) and db/db (postnatal 1–5 months) mice. The optic densities of HMGB1 are normalized to that of C57 mice at corresponding time points. All data are expressed as mean ± SEM. *P < 0.05 vs C57 mice. db/db: homozygous type 2 diabetic db/db mice; db/+: heterozygous nondiabetic littermates; C57: C57BL6 mice.
blocked the expression of HMGB1. These results indicated that antiHMGB1 treatment showed an effective and reliable anti-allodynic effect in a dose-dependent manner and this effect was through inhibiting HMGB1 expression.
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
335
induction of a number of inflammatory mediators in astrocytes [20]. In the present study, we tried to demonstrate whether up-regulated HMGB1 contributed to astrocytic activation and mechanical allodynia. Immunohistochemistry indicated that compared with db/+ and C57 mice (postnatal 4 months), GFAP staining was significantly increased in the spinal cord of db/db mice at the same postnatal period. Intrathecally administering 1 g of antiHMGB1 could decrease GFAP expression slightly, although not significantly (Fig. 3A–G, P > 0.05). However, 10 g and 50 g of antiHMGB1 could inhibit GFAP immunodensity significantly (Fig. 3A–G, P < 0.05). We checked the expression of GFAP at different time following anti-HMGB1 (50 g) injection by Western blot analysis. The intensity of GFAP was slightly but not significantly decreased at 1 h but was remarkably down-regulated at 2–8 h. However, GFAP level returned to normal level at 1 d after anti-HMGB1 (50 g) injection (Fig. 3H, P < 0.05). These results showed that anti-HMGB1 treatment could inhibit astrocytic activation in the diabetic mice. Astrocytic activation in diabetic mice is likely to enhance gene transcription and consequently to modulate synthesis of inflammatory mediators such as IL-1, TNF-alpha, monocyte chemotactic protein (MCP-1), IL-6, which positively contribute to pain sensitization [8]. With real-time RT-PCR, we observed that in postnatal 4 months, db/db but not db/+ or C57 mice induced significant increases of TNF-alpha (2.16 ± 0.25 folds of C57, P < 0.05), IL-1 (3.28 ± 0.21 folds of C57, P < 0.05), IL-6 (2.23 ± 0.11 folds of C57, P < 0.05) and MCP-1 (1.83 ± 0.21 folds of C57, P < 0.05) in the spinal dorsal horn (Fig. 4A). Treatments with HMGB1 antibody (50 g) but not vehicle control, significant inhibited TNF-alpha (1.56 ± 0.14 folds of C57, P < 0.05), IL-1 (2.18 ± 0.15 folds of C57, P < 0.05), IL-6 (1.33 ± 0.18 folds of C57, P < 0.05) and MCP-1 (1.37 ± 0.22 folds of C57, P < 0.05). We also examined the mRNA levels of these inflammatory mediators at different time intervals following anti-HMGB1 (50 g) injection. TNF-alpha was significantly lower than the baseline value throughout the observing period (1 h to 1 d), while IL-1 was not obviously lower at 1 h. IL-6 and MCP-1 expressions were only remarkably decreased from 2 h to 8 h after anti-HMGB1 (50 g) injection (Fig. 4B). These results suggested that spinal HMGB1 up-regulation in diabetic mice contributed to synthesis of inflammatory mediators.
4. Discussion
Fig. 2. Behavioral consequences following intrathecal injection of the neutralizing antibody against HMGB1 (anti-HMGB1). Paw withdrawal frequencies in response to von Frey filaments with low (0.24 mN, A) and moderate (4.33 mN, B) intensities. Intrathecally injecting 10 and 50 g of anti-HMGB1 significantly reversed mechanical allodynia and this allodynia effect lasted to 1 d after injection. (C) Western blot analysis on the quality of neutralizing antibody on HMGB1 expression 4 h after injection. Ab, neutralizing antibody against HMGB1. *P < 0.05 vs vehicle control group.
3.3. HMGB1 contributes to spinal astrocytic activation and inflammatory response in diabetic mice We previously observed that astrocytic activation contributed to mechanical allodynia in the db/db mouse [17] and an in vitro study suggested that HMGB1 could induce transcriptional/translational
In the present study, we provide evidence of the novel contribution of spinal HMGB1 in diabetic pain. We observed that HMGB1 is significantly increased in the spinal dorsal horn of type 2 diabetic mice and HMGB1 expression correlates with mechanical allodynia. In addition, blocking HMGB1 could reverse diabetic pain with decreased astrocytic activation and down-regulated expression of inflammatory mediators. Cumulating data suggest that HMGB1 is not only a nuclear protein, but also an important mediator in inflammatory response. After stimulation, HMGB1 can be released into the extracellular space either actively or passively. Recent studies suggest that in conditions such as oxidative stress, HMGB1 could be released [24]. In diabetes, hyperglycemia was reported to not only directly induce oxidative stress [23], but also contribute to increased aldose reductase activity, nerve hypoxia/ischemia, activation of protein kinase C, and insulin-like growth factor deficiency [6]. These pathological pathways all in turn converge in producing dramatical oxidative stress [10]. Thus we hypothesize that hyperglycemia-induced oxidative stress may be a key mechanism for the initial increase of HMGB1. Released HMGB1 binds to its receptors, including the receptor for advanced glycation end products (RAGE), Toll-like receptor2 (TLR2), and TLR4, which are highly expressed on immune
336
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
Fig. 3. Diabetes-induced astrocytic activation in the spinal dorsal horn is revered by intrathecal anti-HMGB1 (Ab) treatment. (A–C) Compared to the C57 (postnatal 4 months) or db/+ (postnatal 4 months) mice, expression of glial fibrillary acidic protein (GFAP) in the spinal dorsal horn are significantly higher at the same postnatal period. Enhanced staining appeared in all areas of the spinal grey matter, with the most prominent increase in superficial dorsal horn laminae I–III. Diabetes-induced up-regulation of GFAP is significantly attenuated by the anti-HMGB1 (Ab) treatment (D–F), compared with the vehicle control. (G) Statistical results of the immunostaining data. *P < 0.05 vs vehicle control group; #P < 0.05 vs C57 group. (H) The expression of GFAP at different time points following anti-HMGB1 (10 g) injection indicated by Western blot analysis. *P < 0.05 vs control value (before injection).
cells [19]. Receptor activation leads to the consequent release of cytokines/chemokines, including TNF-alpha, IL-1, IL-6, and MCP1 [18,28]. In the present study, we also observed that inhibiting HMGB1 expression with the specific neutralizing antibody could significantly block the expression of these cytokines/chemokines. Previous studies have confirmed the positive contribution of these cytokines/chemokines to chronic pain. For example, it was previous reported that through binding with its receptor on the neurons, IL-1 increases calcium conductivity of neuronal NMDA receptor, possibly via intracellular signaling pathways leading to the phosphorylation of NMDA receptor NR1 subunit [12]. Activation of NMDA receptor in the spinal dorsal horn plays a pivotal role in central sensitization and thus chronic pain [16]. In addition to triggering release of proinflammatory cytokines in spinal dorsal horn, HMGB1 could also be released after cytokines stimulation [22]. So there is a positive feedback circuit between HMGB1 and proinflammatory cytokines. This positive feedback circuit enlarges the effect of HMGB1 on diabetic pain and makes it more difficult to be treated by clinical therapy. In the present study, we also observed a remarkable astrocytic activation in diabetic mice. Interestingly, inhibiting HMGB1 could significantly block astrocytic activation. As we have discussed above, HMGB1 could binds to TLR2 and TLR4, which are
also expressed on spinal astrocytes. We thus assume that HMGB1 may activate astrocytes through these two TLRs. Great deals of studies have been performed targeting TLRs as critical mediators of chronic pain [11]. It has been reported that after nerve injury, TLR2 is required for proinflammatory cytokines expression and pain hypersensitivity [14] and up-regulated TLR4 in the spinal cord are required for glial activation and pain response [4]. Another important receptor of HMGB1, RAGE, has not been reported to be expressed on spinal astrocytes. However, a recent study showed that HMGB1 is expressed on GFAP-positive satellite cells in dorsal root ganglia [15]. Previous studies have reported that RAGE is significantly increased in diabetes [26]. So RAGE may also mediate the pronociceptive effect of HMGB1 in diabetic pain. In summary, we observed that type 2 diabetes induced a significant increase of HMGB1, which could activate spinal astrocytes and expression of inflammatory cytokines and chemokines, which could modulate NMDAR-mediated synaptic transmission and pain response. In addition, cytokines and/or chemokines further increase release of HMGB1 in the spinal cord, and enhance nociceptive transmission in a manner of positive feedback. Our data thus indicate that HMGB1 may play an important role in diabetic pain and inhibiting HMGB1 expression may provide a novel treatment for diabetic pain.
P.-C. Ren et al. / Brain Research Bulletin 88 (2012) 332–337
[5]
[6] [7]
[8] [9] [10] [11]
[12]
[13] [14]
[15]
[16] [17]
[18] Fig. 4. Effects of anti-HMGB1 (Ab) treatment on diabetes-induced cytokines/chemokine expression revealed by real-time RT-PCR. Data are normalized against GAPDH levels and expressed as relative fold changes compared to the C57 group. (A) Compare with C57 (postnatal 4 months) and db/+ (postnatal 4 months) mice, expression of TNF-alpha, IL-1, IL-6 and monocyte chemotactic protein (MCP-1) in db/db mice is significantly higher at the same postnatal period. *P < 0.05 vs C57 group; #P < 0.05 vs vehicle control group. (B) The expression of cytokines/chemokine at different time points following anti-HMGB1 (10 g) injection. *P < 0.05 vs control value (before injection).
[19]
[20]
[21]
[22]
Conflicts of interest The authors declare that there are no conflicts of interest. Acknowledgments
[23] [24] [25]
This work was supported by the National Natural Science Foundation of China (No. 81071527) and the Chinese Postdoctoral Science Foundation (No. 201003743). References [1] A. Andersson, R. Covacu, D. Sunnemark, A.I. Danilov, A. Dal Bianco, M. Khademi, E. Wallstrom, A. Lobell, L. Brundin, H. Lassmann, R.A. Harris, Pivotal advance: HMGB1 expression in active lesions of human and experimental multiple sclerosis, J. Leukoc. Biol. 84 (2008) 1248–1255. [2] A.M. Barrett, M.A. Lucero, T. Le, R.L. Robinson, R.H. Dworkin, A.S. Chappell, Epidemiology, public health burden, and treatment of diabetic peripheral neuropathic pain: a review, Pain Med. 8 (Suppl 2) (2007) S50–S62. [3] A.I. Basbaum, D.M. Bautista, G. Scherrer, D. Julius, Cellular and molecular mechanisms of pain, Cell 139 (2009) 267–284. [4] I. Bettoni, F. Comelli, C. Rossini, F. Granucci, G. Giagnoni, F. Peri, B. Costa, Glial TLR4 receptor as new target to treat neuropathic pain: efficacy of a new
[26]
[27]
[28]
[29]
[30]
337
receptor antagonist in a model of peripheral nerve injury in mice, Glia 56 (2008) 1312–1319. M. Chacur, E.D. Milligan, L.S. Gazda, C. Armstrong, H. Wang, K.J. Tracey, S.F. Maier, L.R. Watkins, A new model of sciatic inflammatory neuritis (SIN): induction of unilateral and bilateral mechanical allodynia following acute unilateral peri-sciatic immune activation in rats, Pain 94 (2001) 231–244. D. Choudhury, M. Tuncel, M. Levi, Diabetic nephropathy – a multifaceted target of new therapies, Discov. Med. 10 (2011) 406–415. C.J. Currie, C.D. Poole, A. Woehl, C.L. Morgan, S. Cawley, M.D. Rousculp, M.T. Covington, J.R. Peters, The financial costs of healthcare treatment for people with Type 1 or Type 2 diabetes in the UK with particular reference to differing severity of peripheral neuropathy, Diabet. Med. 24 (2007) 187–194. Y.J. Gao, R.R. Ji, Activation of JNK pathway in persistent pain, Neurosci. Lett. 437 (2008) 180–183. Y.J. Gao, R.R. Ji, Chemokines, neuronal–glial interactions, and central processing of neuropathic pain, Pharmacol. Ther. 126 (2010) 56–68. F. Giacco, M. Brownlee, Oxidative stress and diabetic complications, Circ. Res. 107 (2010) 1058–1070. L.H. Guo, H.J. Schluesener, The innate immunity of the central nervous system in chronic pain: the role of Toll-like receptors, Cell. Mol. Life Sci. 64 (2007) 1128–1136. W. Guo, H. Wang, M. Watanabe, K. Shimizu, S. Zou, S.C. LaGraize, F. Wei, R. Dubner, K. Ren, Glial–cytokine–neuronal interactions underlying the mechanisms of persistent pain, J. Neurosci. 27 (2007) 6006–6018. J.L. Hylden, G.L. Wilcox, Intrathecal morphine in mice: a new technique, Eur. J. Pharmacol. 67 (1980) 313–316. D. Kim, M.A. Kim, I.H. Cho, M.S. Kim, S. Lee, E.K. Jo, S.Y. Choi, K. Park, J.S. Kim, S. Akira, H.S. Na, S.B. Oh, S.J. Lee, A critical role of toll-like receptor 2 in nerve injury-induced spinal cord glial cell activation and pain hypersensitivity, J. Biol. Chem. 282 (2007) 14975–14983. J.B. Kim, C.M. Lim, Y.M. Yu, J.K. Lee, Induction and subcellular localization of high-mobility group box-1 (HMGB1) in the postischemic rat brain, J. Neurosci. Res. 86 (2008) 1125–1131. M. Larsson, J. Broman, Synaptic plasticity and pain: role of ionotropic glutamate receptors, Neuroscientist (2010). Y.H. Liao, G.H. Zhang, D. Jia, P. Wang, N.S. Qian, F. He, X.T. Zeng, Y. He, Y.L. Yang, D.Y. Cao, Y. Zhang, D.S. Wang, K.S. Tao, C.J. Gao, K.F. Dou, Spinal astrocytic activation contributes to mechanical allodynia in a mouse model of type 2 diabetes, Brain Res. 1368 (2010) 324–335. M.T. Lotze, K.J. Tracey, High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal, Nat. Rev. Immunol. 5 (2005) 331–342. J.A. Nogueira-Machado, C.M. Volpe, C.A. Veloso, M.M. Chaves, HMGB1, TLR and RAGE: a functional tripod that leads to diabetic inflammation, Expert Opin. Ther. Targets 15 (2011) 1023–1035. M. Pedrazzi, M. Patrone, M. Passalacqua, E. Ranzato, D. Colamassaro, B. Sparatore, S. Pontremoli, E. Melloni, Selective proinflammatory activation of astrocytes by high-mobility group box 1 protein signaling, J. Immunol. 179 (2007) 8525–8532. Y.S. Ren, N.S. Qian, Y. Tang, Y.H. Liao, Y.L. Yang, K.F. Dou, M. Toi, Sodium channel Nav1.6 is up-regulated in the dorsal root ganglia in a mouse model of type 2 diabetes, Brain Res. Bull. 87 (2012) 244–249. M. Shibasaki, M. Sasaki, M. Miura, K. Mizukoshi, H. Ueno, S. Hashimoto, Y. Tanaka, F. Amaya, Induction of high mobility group box-1 in dorsal root ganglion contributes to pain hypersensitivity after peripheral nerve injury, Pain 149 (2010) 514–521. D.K. Singh, P. Winocour, K. Farrington, Oxidative stress in early diabetic nephropathy: fueling the fire, Nat. Rev. Endocrinol. 7 (2011) 176–184. D. Tang, R. Kang, H.J. Zeh 3rd, M.T. Lotze, High-mobility group box 1, oxidative stress, and disease, Antioxid. Redox Signal. 14 (2011) 1315–1335. W. Tong, W. Wang, J. Huang, N. Ren, S.X. Wu, Y.Q. Li, Spinal high-mobility group box 1 contributes to mechanical allodynia in a rat model of bone cancer pain, Biochem. Biophys. Res. Commun. 395 (2010) 572–576. C. Toth, L.L. Rong, C. Yang, J. Martinez, F. Song, N. Ramji, V. Brussee, W. Liu, J. Durand, M.D. Nguyen, A.M. Schmidt, D.W. Zochodne, Receptor for advanced glycation end products (RAGEs) and experimental diabetic neuropathy, Diabetes 57 (2008) 1002–1017. W. Wang, X. Mei, J. Huang, Y. Wei, Y. Wang, S. Wu, Y. Li, Crosstalk between spinal astrocytes and neurons in nerve injury-induced neuropathic pain, PLoS ONE 4 (2009) e6973. W. Wang, X.P. Mei, Y.Y. Wei, M.M. Zhang, T. Zhang, L.X. Xu, S.X. Wu, Y.Q. Li, Neuronal NR2B-containing NMDA receptor mediates spinal astrocytic c-Jun N-terminal kinase activation in a rat model of neuropathic pain, Brain Behav. Immun. 25 (2011) 1355–1366. W. Wang, R.S. Petralia, K. Takamiya, J. Xia, Y.Q. Li, R.L. Huganir, Y.X. Tao, M. Yaster, Preserved acute pain and impaired neuropathic pain in mice lacking protein interacting with C Kinase 1, Mol. Pain 7 (2011) 11. M. Zimmermann, Ethical guidelines for investigations of experimental pain in conscious animals, Pain 16 (1983) 109–110.