High-Mobility Group Box 1 Protein Signaling in Painful Diabetic Neuropathy

Abstract

Diabetes is a global epidemic and more than 50% diabetic patients are also diagnosed with neuropathy, which greatly affects the quality of life of the patients. Available treatments are not always successful due to the limited efficacy and complications, such as addiction and dependency. Studies have implicated that high mobility group box1 (HMGB1) protein plays a crucial role in neuroinflammation and the development of neuropathic conditions. HMGB1 is a proinflammatory cytokine that can be released from necrotic cells in passive form or in response to inflammatory signals as an active form. HMGB1 is the ligand for the receptor for advanced glycation end products (RAGE), and toll-like receptors, (TLR)-2 and TLR4, which also indirectly activates C-X-C chemokine receptor type 4 (CXCR4). We investigated whether blocking of HMGB1 can reduce pain and inflammation in diabetic neuropathic animals to further understand the role of HMGB1 in diabetic neuropathy. Type 2 diabetic rats and mice were treated with natural inhibitor of HMGB1, glycyrrhizin (GLC) for five days/week for four weeks at a dose of 50 mg/kg per day by intraperitoneal injection. The animals were divided into three categories: naïve control, diabetic alone, diabetic with GLC treatment. All of the behavioral analyses were conducted before and after the treatment. The expression of inflammatory markers and changes in histone acetylation in the peripheral nervous system were measured by immunohistochemistry and Western blot analysis after the completion of the treatment. Our study revealed that TLR4, HMGB1, CXCR4, and Nod-like receptor protein 3 (NLRP3) levels were increased in the spinal and dorsal root ganglia (DRG) neurons of Type 2 diabetic mice and rats with painful neuropathy. GLC treatment inhibited the increases in TLR4, NLRP3, and CXCR4 expressions and improved the mechanical and thermal pain threshold in these animals. Immunohistochemical studies revealed that hyperglycemia mediated inflammation influenced HMGB1 acetylation and its release from the neurons. It also altered histone 3 acetylation in the microglial cells. The inhibition of HMGB1 by GLC prevented the release of HMGB1 as well as H3K9 acetylation. These findings indicate that the interruption of HMGB1 mediated inflammation could ameliorate diabetic neuropathy and might exhibit a unique target for the treatment.

Keywords: Diabetes; glycyrrhizin; high mobility group box1 (HMGB1); inflammation; neuropathy.

Figure 1

Alterations in mechanical and thermal pain behaviors in type 2 diabetic animals following treatment with Glycyrrhizin (GLC). (a) Thermal withdrawal latency (Hargreaves test) exhibited a reduction in latency in response to unpleasant thermal stimulus in diabetic animals when compared with control animals (p < 0.001). GLC treated animals showed significant amelioration in thermal hyperalgesia compared to diabetic only animals (p < 0.01). (b) Diabetic animals exhibited significant mechanical hyperalgesia (Randall-Selitto) compared to control animals (p < 0.0001). Animals treated with GLC showed significant alleviation of mechanical hyperalgesia compared to diabetic only animals (p < 0.01). Con: naïve control; Dia: diabetic only group; Dia+GLC: diabetic group treated with glycyrrhizin. The data presented in the graph indicates mean ± SEM, n = 6–8 per group. ** p < 0.01; *** p < 0.001; ## p < 0.0001.

Figure 2

Glycyrrhizin treatment ameliorated neuroinflammation in DRG of animals with PDN. (a–c) Type 2 diabetic animals with painful neuropathy demonstrated a significant increase in HMGB1, TLR4 and NLRP3 in DRG at eight weeks after hyperglycemia as compared to their control counterparts. Treatment with GLC confirmed alleviation of neuroinflammation with decreased expressions of NLRP3, TLR4, and HMGB1 compared to the diabetic animals with no treatment. Western blots are representation of one sample from each group. There are no significant differences between control vs diabetic+GLC treated groups. Con: naïve control; Dia: diabetic only group; Dia+GLC: diabetic group treated with glycyrrhizin. The data presented in the graph indicates mean ± SEM, n = 6–8 per group. * p < 0.05; ** p < 0.01.

Figure 3

GLC treatment inhibits hyperglycemia-induced HMGB1 cytoplasmic and extracellular release in spinal cord neurons of Type 2 diabetic mice. Immunohistochemical studies of HMGB1 expression in the spinal cord dorsal horn of ob/ob mice show cytoplasmic and extracellular HMGB1 release from neuronal cells and GLC treatment blocked the release. NeuN antibody, biomarker for neuron shows in green and DAPI (4’,6-diamidino-2-phenylindole), a blue-fluorescent dye stains DNA in the nucleus. Bar = 200 µM.

Figure 4

Hyperglycemia-induced cytoplasmic release of HMGB1 in DRG neurons of type 2 diabetic rats was hindered by GLC treatment. Immunohistochemical analysis indicated enhanced HMGB1 expression in sensory neurons and its translocation into the cytoplasm. GLC treatment abolished the increase in HMGB1 cytoplasmic release in DRG neurons of Zucker diabetic fatty (ZDF) rats. Bar = 200 µM.

Figure 5

Increased expression of TRPC6 and activation of c-Jun N-terminal kinase (JNK) in spinal cord neurons of type 2 diabetic animals was impeded by GLC treatment. (a,b) Type 2 ob/ob diabetic mice showed increased expression TRPC6 in the spinal cord neurons whereas activation of JNK was observed in the spinal cord of ZDF rats. GLC treatment alleviated the expression of TRPC6 and altered the JNK activation in both animal models. There are no significant differences between control vs diabetic+GLC treated groups. Con: naïve control; Dia: diabetic only group; Dia+GLC: diabetic group treated with glycyrrhizin. * p < 0.05.

Figure 6

Increased expression of CXCR4 in DRG neurons of ZDF diabetic animals was blocked by glycyrrhizin. Type 2 diabetic DRG neurons demonstrated increased expression of CXCR4 as displayed by immunohistochemical (a) and Western blot analysis (b). Treatment with GLC reduced the CXCR4 expression in DRG neurons. Bar = 200 µM. Con: naïve control; Dia: diabetic only group; Dia + GLC: diabetic group treated with glycyrrhizin. * p < 0.05; ** p < 0.01.

Figure 7

H3K9ac-dependent spinal HMGB1 expression in diabetic animals with painful neuropathy. (a) Diabetic ob/ob mice increased nuclear H3K9ac activity in non-neuronal glial cells of spinal cord. Treatment with GLC ameliorated the H3K9 acetylation in diabetic spinal cord neuronal and glial cells. (b) This increase was confirmed by western blot analysis of spinal cord of ZDF rats which also demonstrated an alleviation of H3K9ac expression in the GLC treated animals. (c) Further confirmation of results with ChIP analyses revealed that the hyperglycemia mediated increased acetylation of H3K9 is associated to HMGB1 promoter in the spinal dorsal horn of ZDF animals. Bar = 200 µM. Con: naïve control; Dia: diabetic only group; Dia+GLC: diabetic group treated with glycyrrhizin. * p < 0.05.