NOD1/RIP2 signalling enhances the microglia-driven inflammatory response and undergoes crosstalk with inflammatory cytokines to exacerbate brain damage following intracerebral haemorrhage in mice

Abstract

Background: Secondary brain damage caused by the innate immune response and subsequent proinflammatory factor production is a major factor contributing to the high mortality of intracerebral haemorrhage (ICH). Nucleotide-binding oligomerization domain 1 (NOD1)/receptor-interacting protein 2 (RIP2) signalling has been reported to participate in the innate immune response and inflammatory response. Therefore, we investigated the role of NOD1/RIP2 signalling in mice with collagenase-induced ICH and in cultured primary microglia challenged with hemin.

Methods: Adult male C57BL/6 mice were subjected to collagenase for induction of ICH model in vivo. Cultured primary microglia and BV2 microglial cells (microglial cell line) challenged with hemin aimed to simulate the ICH model in vitro. We first defined the expression of NOD1 and RIP2 in vivo and in vitro using an ICH model by western blotting. The effect of NOD1/RIP2 signalling on ICH-induced brain injury volume, neurological deficits, brain oedema, and microglial activation were assessed following intraventricular injection of either ML130 (a NOD1 inhibitor) or GSK583 (a RIP2 inhibitor). In addition, levels of JNK/P38 MAPK, IκBα, and inflammatory factors, including tumour necrosis factor-α (TNF-α), interleukin (IL)-1β, and inducible nitric oxide synthase (iNOS) expression, were analysed in ICH-challenged brain and hemin-exposed cultured primary microglia by western blotting. Finally, we investigated whether the inflammatory factors could undergo crosstalk with NOD1 and RIP2.

Results: The levels of NOD1 and its adaptor RIP2 were significantly elevated in the brains of mice in response to ICH and in cultured primary microglia, BV2 cells challenged with hemin. Administration of either a NOD1 or RIP2 inhibitor in mice with ICH prevented microglial activation and neuroinflammation, followed by alleviation of ICH-induced brain damage. Interestingly, the inflammatory factors interleukin (IL)-1β and tumour necrosis factor-α (TNF-α), which were enhanced by NOD1/RIP2 signalling, were found to contribute to the NOD1 and RIP2 upregulation in our study.

Conclusion: NOD1/RIP2 signalling played an important role in the regulation of the inflammatory response during ICH. In addition, a vicious feedback cycle was observed between NOD1/RIP2 and IL-1β/TNF-α, which could to some extent result in sustained brain damage during ICH. Hence, our study highlights NOD1/RIP2 signalling as a potential therapeutic target to protect the brain against secondary brain damage during ICH.

Keywords: Inflammatory response; Intracerebral haemorrhage; Microglial activation; NOD1; RIP2.

Fig. 1

Experimental design of the in vivo and vitro study. ICH, intracerebral hemorrhage; WB, western blot; IF, immunofluorescence; IHC, immunohistochemistry; ML, ML130; GSK, GSK583; iE, C12-iE-DAP; SP, SP600125; SB, SB202190; BAY, BAY11-7082; d, day; h, hour

Fig. 2

NOD1 and RIP2 expression were upregulated following ICH. Western blot of NOD1 and RIP2 a, b in the brains of mice on the indicated days following ICH (n = 3 mice for each group; *P < 0.01 vs. the sham group), c, d in primary microglia, and e, f in BV2 cells challenged with hemin at the indicated doses for 24 h (n = 3 experiments for each group; *P < 0.01 vs. the ctrl group). All data are the representative of three independent experiments

Fig. 3

NOD1/RIP2 inhibition alleviated ICH-induced brain damage and microglial activation. a, b Levels of injury volume, brain water content and neurological function in the indicated group (n = 6 mice/group for analysis of injury volume and brain water content; n = 10 mice/group for analysis of neurological function. #P < 0.01 vs. the DMSO-treated group). c, d CD68-, Iba-1-positive cells and the cell body size of microglia in the perihematomal tissue by immunofluorescence staining (n = 6 mice for each group, 3 images/mouse; *P < 0.01 vs. the sham group, #P < 0.01 vs. the DMSO-treated group). e The representative images of the four subtypes of microglia: type 1, the resting form of microglia; type 2, the initiated form of activated microglia; type 3, the activated form of microglia with non-phagocytic function; type 4, the overactivated form of microglia with phagocytic function. f, g Immunohistochemical analysis of microglial subtypes and quantitative data of activated microglia in the indicated groups (n = 6 mice for each group, 3 images/mouse; *P < 0.01 vs. the sham group, #P < 0.01 vs. the DMSO-treated group). All data are representative of at least six independent experiments

Fig. 4

NOD1 inhibition significantly reduced the inflammatory response in response to ICH. Levels of iNOS, TNF-α, and IL-1β protein a, c in the brain in the indicated groups (n = 3 mice for each group; *P < 0.01 vs. the sham group, #P < 0.01 vs. the ICH/ICH + DMSO group) and b, d in the primary microglia from the indicated groups (n = 3 experiments for each group; *P < 0.01 vs. the ctrl group, #P < 0.01 vs. the Hemin group); RIP2, total and phosphorylated JNK/P38 MAPK, and IκBα protein levels e, g in the brain in the indicated groups (n = 3 mice for each group; *P < 0.01 vs. the sham group, #P < 0.01 vs. the ICH/ICH + DMSO) and f, h in the primary microglia from the indicated groups (n = 3 experiments for each group; *P < 0.01 vs. the ctrl group, #P < 0.01 vs. the hemin group). All data are representative of three independent experiments

Fig. 5

RIP2 inhibition alleviated the inflammatory response induced by NOD1 activation post ICH. Levels of iNOS, TNF-α, and IL-1β protein a, c in the brain in the indicated groups (n = 3 mice for each group; *P < 0.01 vs. the sham group, **P < 0.01 vs. the ICH + iE group, #P < 0.01 vs. the ICH + DMSO, ##P < 0.01 vs. the Hemin + iE group) and b, d in cultured primary microglia from the indicated groups (n = 3 experiments for each group; *P < 0.01 vs. the ctrl group, **P < 0.01 vs. the Hemin + iE group, #P < 0.01 vs. the Hemin + DMSO, ##P < 0.01 vs. the Hemin + iE group). Total and phosphorylated JNK/P38 MAPK, IκBα, and NOD1 protein levels e, g in the brain in the indicated groups (n = 3 mice for each group; *P < 0.01 vs. the sham group, **P < 0.01 vs. the ICH + iE group, #P < 0.01 vs. the ICH + DMSO, ##P < 0.01 vs. the Hemin + iE group) and f, h in cultured primary microglia from the indicated groups (n = 3 experiments for each group; *P < 0.01 vs. the ctrl group, **P < 0.01 vs. the Hemin + iE group, #P < 0.01 vs. the Hemin + DMSO, ##P < 0.01 vs. the Hemin + iE group). All data are representative of three independent experiments

Fig. 6

NOD1 and RIP2 expression can be upregulated by IL-1β and TNF-α. Total and phosphorylated JNK/P38 MAPK, IκBα, NOD1, and RIP2 protein levels a, c in the brain in ICH-induced mice treated with the indicated inhibitors (n = 3 mice for each group; *P < 0.01 vs. the sham group, #P < 0.01 vs. the ICH + DMSO group) and b, d in hemin-challenged BV2 cells pretreated with the indicated inhibitors (n = 3 experiments for each group; *P < 0.01 vs. the ctrl group, #P < 0.01 vs. the Hemin + DMSO group). NOD1 and RIP2 expression e, g in primary microglia challenged with either TNF-α or IL-1β alone or in combination for 24 h (n = 3 experiments for each group; *P < 0.01 vs. the ctrl group, #P < 0.01 vs. the IL-1β + TNF-α-challenged group) and f, h in mice challenged with either TNF-α or IL-1β alone or in combination for 24 h (n = 3 mice for each group; *P < 0.01 vs. the sham group, #P < 0.01 vs. the IL-1β + TNF-α-challenged group). All data are representative of three independent experiments

Fig. 7

Working model of the molecular mechanisms by which NOD1/RIP2 signalling mediates the ICH-induced inflammatory response and undergoes positive crosstalk with inflammatory factors during ICH in mice. More specifically, ICH-induced damage-associated molecular patterns (DAMPs) were able to initiate the activation of NOD1, which then resulted in the activation of its adaptor RIP2. Activated RIP2 in turn exerted its regulatory effect on the proinflammatory factors TNF-α and IL-1β via JNK/P38 MAPK- and NF-κB-dependent signalling, and the proinflammatory factors TNF-α and IL-1β that were induced by NOD1/RIP2 signalling also enhanced NOD1/RIP2 upregulation. Either the NOD1 inhibitor ML130 or RIP2 inhibitor GSK583 could suppress the inflammatory response induced by ICH. The P38 inhibitor SB202190, JNK inhibitor SP600125, and NF-κB inhibitor BAY11-7082 were used for intervention