Glycyrrhizin Blocks the Detrimental Effects of HMGB1 on Cortical Neurogenesis After Traumatic Neuronal Injury

Abstract

Despite medical advances, neurological recovery after severe traumatic brain injury (TBI) remains poor. Elevated levels of high mobility group box protein-1 (HMGB1) are associated with poor outcomes; likely via interaction with receptors for advanced-glycation-end-products (RAGE). We examined the hypothesis that HMGB1 post-TBI is anti-neurogenic and whether this is pharmacologically reversible. Post-natal rat cortical mixed neuro-glial cell cultures were subjected to needle-scratch injury and examined for HMGB1-activation/neuroinflammation. HMGB1-related genes/networks were examined using genome-wide RNA-seq studies in cortical perilesional tissue samples from adult mice. Post-natal rat cortical neural stem/progenitor cell cultures were generated to quantify effects of injury-condition medium (ICM) on neurogenesis with/without RAGE antagonist glycyrrhizin. Needle-injury upregulated TNF-α/NOS-2 mRNA-expressions at 6 h, increased proportions of activated microglia, and caused neuronal loss at 24 h. Transcriptome analysis revealed activation of HMGB1 pathway genes/canonical pathways in vivo at 24 h. A 50% increase in HMGB1 protein expression, and nuclear-to-cytoplasmic translocation of HMGB1 in neurons and microglia at 24 h post-injury was demonstrated in vitro. ICM reduced total numbers/proportions of neuronal cells, but reversed by 0.5 μM glycyrrhizin. HMGB1 is activated following in vivo post mechanical injury, and glycyrrhizin alleviates detrimental effects of ICM on cortical neurogenesis. Our findings highlight glycyrrhizin as a potential therapeutic agent post-TBI.

Keywords: HMGB1; neurogenesis; neuroinflammation; traumatic brain injury.

Figure 1

Schematic diagram of the experimental design. Mixed glial and neural stem cell cultures were generated from the post-natal rat cortex (described in detail in the text). Gene expression, immunofluorescence, and western blot studies were performed on mixed glial cultures at multiple time points, post injury. Abbreviations: CCM—control condition media, Gly—glycyrrhizin, ICM—injury condition media.

Figure 2

Mechanical injury of cortical mixed glial cultures enhanced microglial activation and upregulated pro-inflammatory cytokine expression. (A) Representative immunofluorescence microscopy images of microglia in control and injury mixed glial cultures (IB4—green, DAPI—blue, HMGB1—red); (B) Injury significantly increased proportions of microglia with active/amoeboid (* p < 0.05) and reduced proportions of microglia with resting/ramified (* p < 0.05) morphologies, representative fluorescent microscopy images of different microglial morphologies (yellow arrows); (C,D) Significant upregulation of pro-inflammatory cytokine expression demonstrated by an acute increase in NOS-2 (**** p < 0.0001) and TNF-alpha mRNA levels (* p < 0.05); (E) Significant loss of neurons (Tuj1⁺ cells) (** p < 0.01) but no significant difference in total cell count (DAPI⁺ cells) at 24 h post-injury; (F) Representative immunofluorescence microscopy images of neurons in control and injury mixed glial cultures (Tuj1—green, DAPI—blue, HMGB1—red). Scale bars 100 µm, 20× magnification in microscopy images. The subfigures (B–D) show two-way ANOVA and the subfigure (E) shows the unpaired t-test used for statistical analysis with p < 0.05 considered significant. n: 3–4 animals, four wells per condition and three repeats per experiment. Abbreviations: HMGB1—high mobility group box protein-1

Figure 3

Schematic of the HMGB1 signaling pathway. The HMGB1 signaling pathway as derived from the IPA analysis results of RNA transcriptomics.

Figure 4

Upregulation of HMGB1 signaling pathways at 1 h and 24 h post needle injury. The heatmap with horizontal hierarchal clustering depicts differentially expressed genes (fragments per kilobase of exon per million—FPKM) associated with HMGB1 signaling across all time points (0 h: no injury, 1 h post-injury and 24 h post-injury). The red colour denotes upregulation of a molecule in HMGB1, signaling cascade, whilst the blue colour indicates downregulation of a molecule. Gene symbols with a single border represent single genes. Double borders represent a complex of genes or the possibility that alternative genes might act in the pathway. The node shapes denote enzymes, phosphatases, kinases, peptidases, G-protein coupled receptor, transmembrane receptor, cytokines, and other. The gradient colour degree means a slightly different expression tendency of that molecule, dark red > 2 fold change, light red 1–2 fold change, n: 4–5 per condition.

Figure 5

Changes in mitogen-activated protein kinase (MAPK) isoforms signaling at 1 h and 24 h post needle injury. The heatmap with horizontal hierarchal clustering depicts differentially expressed genes (FPKM) associated with HMGB1 signaling across all time points (0 h (no injury), 1 h post-injury and 24 h post-injury). The red colour denotes upregulation of a molecule involved in the HMGB1 signaling cascade, whilst the blue colour indicates downregulation.

Figure 6

HMGB1 cellular expression and release are increased after mechanical scratch injury of rat cortical mixed glial cell cultures. (A) HMGB1 mRNA expression was significantly elevated as early as 2 h post-injury when compared to the control (* p < 0.05); (B) HMGB1 protein levels significantly increased at 24 h post-injury relative to the control on the western blot (* p < 0.05); (C) Percentage HMGB1 cytoplasmic translocation was significantly increased in both neurons (Tuj1⁺ cells) and microglia (IB4⁺ cells) at 24 h post-injury (**** p < 0.0001), but not in astrocytes (GFAP⁺ cells). Subfigures (A,C) show two-way ANOVA and (B) one-way ANOVA used for statistical analysis with p < 0.05 considered significant. n: 3–4 animals, four wells per condition and 3 repeats per experiment.

Figure 7

Injury conditioned media (ICM) has a detrimental effect on neuronal progenitor cell survival and differentiation, via a RAGE-dependent mechanism. (A) Exposure to ICM does not result in a significant change in total cell count; (B) ICM exposure results in a significant decrease in both neuronal cell survival (Tuj1⁺ cells/mm2; * p < 0.05) and differentiation (%Tuj1/DAPI cells; * p < 0.05); (C) Representative immunofluorescence microscopy images of rat cortical neural stem cell/progenitor cell cultures demonstrate RAGE expression in Sox2⁺ cells (yellow arrows); (D,E) Addition of glycyrrhizin, a RAGE antagonist, rescued the anti-neurogenic effects seen with addition of ICM (** p < 0.01, **** p < 0.0001); (F) Representative immunofluorescence microscopy images of rat cortical neural stem cell/progenitor cell cultures exposed to CCM and ICM in the absence or presence of glycyrrhizin (Tuj1- green, DAPI- blue). Scale bars 100 µm, 20× magnification in microscopy images. Two-way ANOVA used for statistical analysis with p < 0.05 considered significant. n: 3–4 animals, four wells per condition and 2 repeats per experiment. Abbreviations: CCM- control condition media; ICM—injury condition media.