Selective NLRP3 inflammasome inhibitor reduces neuroinflammation and improves long-term neurological outcomes in a murine model of traumatic brain injury

Xin Xua,b,1, Dongpei Yina,b,1, Honglei Renb,1, Weiwei Gaoa,b,c, Fei Lia,b,d, Dongdong Suna,b, Yingang Wua,b, Shuai Zhoua,b, Li Lyue, Mengchen Yanga,b, Jianhua Xionga,b, Lulu Hanb, Rongcai Jianga,b,⁎, Jianning Zhanga,b,⁎

Abstract

The nucleotide-binding oligomerization domain (NOD)-like receptor family pyrin domain containing 3 (NLRP3) inflammasome-mediated inflammatory response has emerged as a prominent contributor to the pathophysiological processes of traumatic brain injury (TBI). Recently, a potent, selective, small-molecule NLRP3 inflammasome inhibitor, MCC950, was described. Here, we investigated the effect of MCC950 on inflammatory brain injury and long-term neurological outcomes in a mouse model of TBI. Male C57/BL6 mice were subjected to TBI using the controlled cortical impact injury (CCI) system. Western blotting, flow cytometry, and immunofluorescence assays were utilized to analyze post-traumatic NLRP3 inflammasome expression and determine its cellular source. We found that NLRP3 inflammasome expression was significantly increased in the peri-contusional cortex and that microglia were the primary source of this expression. The effects of MCC950 on mice with TBI were then determined using post-assessments including analyses of neurological deficits, brain water content, traumatic lesion volume, neuroinflammation, blood-brain barrier (BBB) integrity, and cell death. MCC950 treatment resulted in a better neurological outcome after TBI by alleviating brain edema, reducing lesion volume, and improving long-term motor and cognitive functions. The therapeutic window for MCC950 against TBI was as long as 6 h. Furthermore, the neuroprotective effect of MCC950 was associated with reduced microglial activation, leukocyte recruitment, and pro-inflammatory cytokine production. In addition, MCC950 preserved BBB integrity, alleviated TBI-induced loss of tight junction proteins, and attenuated cell death. Notably, the efficacy of MCC950 was abolished in microglia-depleted mice. These results indicate that microgliaderived NLRP3 inflammasome may be primarily involved in the inflammatory response to TBI, and specific NLRP3 inflammasome inhibition using MCC950 may be a promising therapeutic approach for patients with TBI.

Keywords:
Traumatic brain injury
Neuroinflammation
NLRP3 inflammasome
Interleukin-1β
Microglia
MCC950

1. Introduction

Traumatic brain injury (TBI) is a worldwide health problem with high mortality and morbidity, and the effective clinical translation of pharmacotherapies for patients with TBI remains insufficient (Ge et al., 2014; Harrison et al., 2015). The pathophysiology of TBI involves a primary mechanical insult and multi-factorial secondary injury cascades (e.g., oxidative stress, apoptosis, and neuroinflammation) (Wang et al., 2016). Accumulating evidence suggests that innate immunity and neuroinflammation are involved in the pathogenesis of TBI (Simon et al., 2017). Upon brain injury, cellular damage results in the rapid release of damage-associated molecular patterns [DAMPs, e.g., ATP, DNA, reactive oxygen species (ROS)] (Corps et al., 2015). The recognition of DAMPs by pattern recognition receptors (PRRs) expressed on cells of the innate immune system then induces the local production of cytokines and chemokines that subsequently facilitate the activation, expansion, and recruitment of immune cells to the injury site (Corps et al., 2015; McKee and Lukens, 2016). Although moderate immune cell activation is essential for the removal of cellular debris and danger signals, excessive and uncontrolled inflammation can exacerbate neuronal damage and subsequent neurological impairment (Lee et al., 2014). In addition, the excessive inflammation can further compromise blood-brain barrier (BBB) integrity, facilitating the invasion of even more peripheral immune cells (da Fonseca et al., 2014; Gao et al., 2015). Accordingly, modulating post-traumatic neuroinflammation at an appropriate level may be of paramount importance for patients with TBI.
As subset of PRRs, the nucleotide-binding oligomerization domain (NOD)-like receptors (NLRs) have recently been demonstrated to be key mediators of innate immune responses via inflammasome signaling activation (Adamczak et al., 2012; de Rivero Vaccari et al., 2014). Among the inflammasome-forming NLRs, NLR pyrin domain containing 3 (NLRP3) is the most extensively studied and has been linked with acute central nervous system (CNS) injuries and several neurodegenerative diseases (de Rivero Vaccari et al., 2014; Zhou et al., 2016). The NLRP3 inflammasome is a cytosolic multi-protein complex that contains the sensor protein NLRP3, the adaptor protein apoptosis-associated speck-like protein containing a caspase recruitment domain (ASC), and the precursor enzyme pro-caspase-1 (Wang et al., 2014). Upon sensing danger stimuli, the NLRP3 inflammasome is assembled and activated to trigger caspase-1 activation, and subsequently the maturation and secretion of the pro-inflammatory cytokines interleukin (IL)-1β and IL-18 (Liu et al., 2013; McKee and Lukens, 2016). These cytokines are crucial in initiating or amplifying the innate immune response and neuroinflammation following TBI (Woodcock and Morganti-Kossmann, 2013). In addition, caspase-1 activation can initiate cell death directly via pyroptosis or indirectly via apoptosis (Sagulenko et al., 2013). Notably, blockade or inhibition of NLRP3 inflammasome activation could alleviate neuroinflammation and improve histopathological and functional outcomes in rodent models of TBI (Mortezaee et al., 2018). Although these studies indicated that the NLRP3 inflammasome might be a novel therapeutic target for patients with TBI, they do not reflect clinical application since they either used knockout models or used pharmacological therapy indirectly or nonspecifically prevented NLRP3 activation.
MCC950 is a recently developed, selective, small-molecule NLRP3 inflammasome inhibitor that has been found to attenuate the inflammatory response in vitro and in vivo (Coll et al., 2015; Mridha et al., 2017; van Hout et al., 2017). More recently, Ismael et al. reported that MCC950 could alleviate TBI and the consequent pro-inflammatory and in particular pro-apoptotic signals during the acute phase of TBI (Ismael et al., 2018). However, the effects of MCC950 treatment on microglia activation, leukocyte infiltration, BBB disruption, and the long-term neurological outcomes are unknown. In this study, we conducted a clinically relevant pharmacological analysis in an experimental TBI model using MCC950. We hypothesized that pharmacological inhibition of the NLRP3 inflammasome by administration of MCC950 would ameliorate inflammatory brain injury, thereby improving neurological outcomes after TBI. Furthermore, the expression profile and the cell type-specific expression of NLRP3 inflammasome in TBI were also confirmed.

2. Materials and methods

2.1. Animals

Adult male C57BL/6 mice (8–10 weeks old, 20–25 g) were purchased from the Experimental Animal Laboratories of the Academy of Military Medical Sciences (Beijing, China). All procedures were conducted in strict accordance with the ARRIVE Guidelines (Animal research: reporting of in vivo experiments) and approved by the Ethics Committee of Tianjin Medical University (Tianjin, China). The mice were housed in animal facilities under a standardized 12 h light/dark cycle with controlled temperature and humidity and had free access to food and water. All efforts were made to minimize the number of mice used and their suffering. In all experiments, data were obtained by investigators blinded to the experimental design.

2.2. Experimental design and drug administration

In the present study, the following four separate experiments were conducted:
In Experiment 1, the post-TBI expression profiles of the NLRP3, ASC, and caspase-1 p20 subunit were determined by western blotting, and the specific cell-type distribution of NLRP3 inflammasome expression was determined by flow cytometric and immunofluorescence analyses. A total of 48 mice were randomly divided into 4 groups, including the sham controls and untreated TBI controls, for analyses at 3 time points (1, 3, and 7 d) after TBI (Fig. 1A).
In Experiment 2, we studied the effects of MCC950 on TBI-induced inflammatory brain damage. In total, 198 mice were randomly divided into 3 groups: the sham group, the TBI + Vehicle group, and the TBI + MCC950 group. MCC950 (MedChem Express, Shanghai, China) was dissolved in phosphate-buffered saline (PBS). After TBI induction, mice were randomly selected and treated with either MCC950 (10 mg/ kg body weight, intraperitoneal injection) or sterile PBS (vehicle) daily for the first 3 d and then every other day until the experiments ended (Coll et al., 2015; Ren et al., 2018). Post-treatment assessments at each time point are shown in Fig. 2A.
In Experiment 3, to evaluate the efficacy of MCC950 in microgliadepleted mice with TBI, we randomized 36 mice into the following 3 groups: the TBI + Vehicle group, the TBI + PLX3397 + Vehicle group, and the TBI + PLX3397 + MCC950 group. To deplete microglia, PLX3397 (Selleckchem, Houston, TX, USA; dissolved in PBS) was given by intragastric gavage at a dose of 40 mg/kg/day from 21 d before surgery to 3 d post-TBI (Li et al., 2017b). Post-assessments including modified neurological severity score (mNSS) and rota-rod tests, brain water content and flow cytometric analyses were performed at 3 d postTBI (Fig. 6A).
In Experiment 4, we explored the therapeutic window of MCC950 treatment for TBI. A total of 48 mice were randomly divided into 5 groups: the TBI + Vehicle group, the TBI + MCC950 (0 h) group, the TBI + MCC950 (3 h) group, the TBI + MCC950 (6 h) group, and the TBI + MCC950 (12 h) group. mNSS and rota-rod tests were performed, and the brain water content was measured (Fig. 7A).

2.3. Animal model of TBI

The procedure for inducing TBI in mice was described in our previous study (Xu et al., 2017). Briefly, general anesthesia was induced with 10% chloral hydrate (3 mL/kg, intraperitoneal injection). The mice were then placed in a stereotaxic apparatus (RWD Life Science, Shenzhen, China), and a 3.5-mm craniotomy was made over the right parietal cortex (2.0 mm posterior from the bregma and 2.0 mm lateral to the sagittal suture) with the dura intact. Mice in the TBI groups were impacted at 4.5 m/s velocity with a 200 ms dwell time and 2.0 mm depression using a 3-mm diameter impactor tip to produce a moderate TBI. Then, the scalp incision was closed with an interrupted 4–0 silk suture, and the mice were placed in heated cages to recover from anesthesia. In the sham group, the mice were anesthetized and only the right parietal craniotomy was performed. All operations were performed with strict aseptic technique.

2.4. Tissue preparation

Mice were sacrificed at each testing day, and the brains were quickly removed and used as follows: 1) For flow cytometric analysis, cerebral hemispheres were mechanically homogenized with 40-μm nylon cell strainers (BectonDickinson, Franklin Lakes, NJ, USA) in PBS. The cell suspensions were then centrifuged at 2000 rpm for 5 min, the resulting cell pellets were collected and resuspended in 5 mL of a 30% Percoll solution (Sigma-Aldrich, St. Louis, MO, USA), and the gradient was centrifuged at 700 ×g for 20 min at 18 °C. Single cells on the bottom of the tube were collected for antibody staining. 2) For immunofluorescence analysis, the brains were fixed in 4% paraformaldehyde overnight at 4 °C and immersed in 20% and then 30% phosphate-buffered sucrose solutions for dehydration. The brain samples were cut at a thickness of 6 μm using a microtome (Leica Microsystems, Wetzlar, Germany). For hematoxylin and eosin (H&E) staining, the fixed brains were dehydrated in graded alcohol and embedded in paraffin. The paraffin-embedded brains were also sectioned at a thickness of 6 μm. 3) The peri-contusional cortex of the brain samples from mice with TBI and the equivalent area form sham-injured mice were used for western blotting, quantitative real-time polymerase chain reaction (qRT-PCR), and enzyme-linked immunosorbent assay (ELISA) analyses. For western blotting analysis, total proteins were extracted using RIPA lysis buffer (Beyotime Biotech, Jiangsu, China) with the protease inhibitor phenylmethanesulfonyl fluoride (Beyotime Biotech). After incubation, the lysates were centrifuged at 12000 rpm for 20 min at 4 °C, and the supernatants were collected. For ELISA, the tissue was homogenized in the lysis buffer at a ratio of 1:10 (tissue/ buffer). The homogenates were shaken for 90 min on ice and then centrifuged at 1500 rpm for 15 min at 4 °C, and the supernatants were transferred to new tubes for further analysis.

2.5. Neurobehavioral training and evaluation

A panel of behavioral assays was used to assess neurological function by two investigators who were blinded to the drug administration. For data validation, each test was repeated twice with four different trials. To evaluate overall neurological deficits, the mNSS test was performed prior to and at 1, 3, 7, and 14 d after TBI or sham surgery. Briefly, the test included a composite of tasks to assess the motor, sensory, reflex, and balance abilities of mice post-injury. The mNSS test was graded on a scale of 0 (normal performance) to 18 (maximal deficit), and a higher score represented more severe neurological dysfunction. Mice with an abnormal score (score > 0) before surgery were excluded from the experiment.
The rota-rod test assessed fine motor coordination and balance by measuring the ability of the mice to remain on an accelerating rota-rod apparatus (RWD Life Science). On the day before surgery, the mice initially underwent an adaptation trial (rotational speed: 4 rpm/min) and then four additional test trials with an accelerating rotational speed (from 4 to 40 rpm within 5 min). The average time to fall off the rotating rod in the test trials was recorded to obtain stable pre-injury baseline values. On days 1, 3, 7, and 14 post-injury, each mouse was tested with the same speed four times per day with an inter-trial interval of 30 min between trials, and the average latency to falling was measured.
The Morris water maze (MWM) was used to evaluate spatial learning and memory on 14–18 d post-injury, followed by a probe trial 24 h after the last training trial as previously described (Vorhees and Williams, 2006). Briefly, the MWM apparatus was a stainless-steel circular tank (122 cm in diameter, 51 cm in height) filled with water (19–22 °C) that had been rendered opaque by white, non-toxic paint. The tank was subdivided into four equal-sized quadrants, and a round acrylic platform (8 cm in diameter) was submerged 1 cm below the waterline and was located at the center of a fixed quadrant. On each training day, the mice received four trials with an inter-trial interval of 10 min, and the start positions were semirandomly changed each day. During each trial, the mice were allowed 90 s to locate the submerged platform, and remained on it for 15 s. If a mouse failed to find the platform within the allotted time, it was picked up and placed on the platform for 15 s and received a latency score of 90 s for that trial. The entire MWM performance was monitored by a computerized tracking system (EthoVision 3.0, Noldus Information Technology, Wageningen, Netherlands). The time to platform (escape latency) and the swimming speed were recorded and then averaged over the four daily trials. On 19 d post-injury, the probe trial was conducted with the platform removed. The mice were released opposite from where the platform was placed and were allowed to swim for 30 s. The time spent in the goal platform quadrant was recorded using the video tracking system.

2.6. Western blotting

Western blotting was performed as previously described (Gao et al., 2015). The total protein concentrations were determined using a bicinchoninic acid (BCA) Protein Assay Kit (Thermo Fisher Scientific, Carlsbad, CA, USA). Equal amounts of protein (8 μg per lane) were loaded onto sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and electrophoretically transferred to 0.22-μm polyvinylidene fluoride (PVDF) membranes (pre-activated with methanol, Millipore, Temecula, CA, USA). The membranes were then blocked with 5% non-fat milk in Tris-buffered saline containing Tween-20 (TBST) for 2 h at room temperature (RT), followed by incubation overnight at 4 °C with the respective primary antibodies as follows: anti-NLRP3 antibody (1:1000, AdipoGen, San Diego, CA, USA), anti-ASC antibody (1:1000, Santa Cruz Biotechnology, CA, USA), anti-caspase-1 p20 antibody (1:1000, Millipore, Billerica, MA, USA), anti-IL-1β antibody (1:1000, Cell Signaling Technology, Danvers, MA, USA), anti-claudin-5 antibody (1:1000, Invitrogen, Carlsbad, CA, USA), anti-zonula occluden-1 (ZO-1) antibody (1:1000, Invitrogen), anti-cleaved caspase-3 antibody (1:1000, Cell Signaling Technology), and anti-β-actin antibody (1:1000, Cell Signaling Technology). After incubation with the species-appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies (1:5000, all from Cell Signaling Technology) for 2 h at RT, the blot bands were detected using a Bio-Rad Gel Doc Imager (Bio-Rad, Hercules, CA, USA) and quantified via densitometry with NIH ImageJ software (Version 1.46r, Wayne Rasband, USA). The expression level of each target protein was standardized to that of corresponding β-actin loading controls. All the experiments were repeated three times.

2.7. Flow cytometry analysis

To investigate the cell source of upregulated NLRP3 inflammasome after TBI, isolated cells were diluted to 1 × 106/mL, fixed and permeabilized with a commercial Intracellular Fixation & Permeabilization Buffer Set (eBioscience, San Diego, CA, USA), and then stained with fluorescently labeled antibodies as follows: anti-CD45-PE, anti-CD11bAPC, anti-GFAP-APC, anti-NeuN antibody (Abcam, Cambridge, MA, USA), anti-NLRP3 antibody (Abcam), and the appropriate isotype control. Alexa 488-conjugated donkey anti-goat IgG (for NLRP3) and Alexa 594-conjugated donkey anti-rabbit IgG (for NeuN) were the secondary antibody (Invitrogen), and fluorescence minus one (FMO) controls were stained respectively. To detect the cellular components in the injured brain, isolated cells were stained with fluorescently labeled antibodies as follows: anti-CD45-FITC, anti-Ly6G-PE, anti-CD3-PE-Cy7, anti-CD4-APC, anti-CD8-PE, anti-CD19-PE, anti-NK1.1-APC, and the appropriate isotype control. To analyze cell death post-injury, the isolated cells were stained for Annexin V (BD Biosciences, San Jose, CA, USA), as detailed in the manufacturer’s protocol. All antibodies were purchased from Biolegend (San Diego, CA, USA), unless otherwise indicated. The cells were sorted on an Accuri C6 instrument (BD Bioscience), and the data were analyzed by Flow Jo Software 7.6.1(Tree Star, US).

2.8. Immunofluorescence and image analysis

Brain cryosections were fixed with acetone at −20 °C for 20 min and incubated in 3% bovine serum albumin solution (BSA, SigmaAldrich) for 30 min at 37 °C to block nonspecific staining. The sections were then incubated with the following antibodies at 4 °C overnight: anti-Iba-1 (1:500, Wako, Osaka, Japan), anti-NLRP3 (1:100, AdipoGen), anti-claudin-5 (1:100, Invitrogen), and anti-ZO-1 (1:100, Invitrogen) antibodies. Thereafter, the cryosections were rinsed with PBS and incubated with corresponding secondary antibodies (Alexa Fluor 594: donkey anti-goat; Alexa Fluor 488: donkey anti-rabbit; Invitrogen) for 1 h at RT. Finally, nuclei were counterstained with 4′,6-diamidino-2phenyl-indole (DAPI; Abcam). Images of each section were captured using a fluorescence microscope (Olympus IX81, Tokyo, Japan), and the data were analyzed from 15 randomly selected microscopic fields (five fields per section × three sections per mouse) with NIH ImageJ software. The results were expressed as the proportion of the immunoreactive area. All analyses were performed in a blinded manner.

2.9. qRT-PCR

Total RNA was extracted using TRIzol reagent (Invitrogen), and spectrophotometric analysis (OD260/280 > 1.8) was used to ensure the purity and quantity of RNA. Then, equal amounts of RNA (1.5 μg) were reverse-transcribed to cDNA with a SuperScript® III CellsDirect™ cDNA Synthesis Kit (Invitrogen). qRT-PCR analysis was conducted on an Opticon 2 Real-Time PCR Detection System (Bio-Rad) using SYBR® Green PCR Master Mix (Applied Biosystems, Waltham, MA, USA) and corresponding primers. Gene expression levels were normalized to the endogenous control GAPDH mRNA level and presented as the fold change relative to the sham group. All the experiments were repeated three times.
The specific primer sequences were as follows: NLRP3: forward 5′-GAAGAAGAGTGGATGGGTTTG-3′, reverse 5′-CTGCGTGTAGCGACTGTTGAG-3′. ASC: forward 5′-TGCTTAGAGACATGGGCTTAC-3′ reverse 5′-CTGTCCTTCAGTCAGCACACT-3′. Caspase-1: forward 5′-GACAAGGCACGGGACCTATGT-3′ reverse 5′-CAGTCAGTCCTGGAAATGTGC-3′. IL-1β: forward 5′-CTCGCAGCAGCACATCAACAA-3′ reverse 5′-AAGGTCCACGGGAAAGACACA-3′. GAPDH: forward 5′ CACTGAGCAAGAGAGGCCCTAT-3′ reverse 5′-GCAGCGAACTTTATTGATGGTATT-3′

2.10. Brain water content

Brain water content was measured using the dry–wet weight method (Gao et al., 2015). The mice were sacrificed under deep anesthesia at 72 h post-injury, and the brains were quickly removed and immediately separated into the left hemisphere (contralateral), right hemisphere (ipsilateral), and cerebellum. The tissue samples were then weighed to determine the wet weight (WW). The dry weight (DW) was obtained after the samples were dried in an oven at 100 °C for 24 h. The brain water content was calculated with the following formula: 100% × (WW − DW) / WW.

2.11. Measurement of lesion volume

To quantify brain lesion volume at 21 d post-injury, transverse sections at successive intervals of 120 μm were cut to cover the entire injured cortex as previously described (Yang et al., 2017). The sections were then stained with H&E (Zsgb-bio, Beijing, China) and imaged at 10 × magnification using a light microscope (Olympus). The ipsilateral and contralateral hemisphere volumes were calculated using NIH ImageJ software in a blinded fashion. Lesion volumes were computed by numeric integration of sequential regions, and the results were presented as the volume percentage of the lesion compared with the contralateral hemisphere.

2.12. ELISA

The concentrations of IL-1β, tumor necrosis factor (TNF)-α, IL-6, and IL-10 were measured with the specific ELISA kits (R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions.

2.13. Evans blue dye extravasation

Evans blue (EB) dye extravasation was used to evaluate BBB permeability as previously described (Gao et al., 2015). In brief, at 72 h after CCI, EB solution (4 mL/kg, 2% in PBS; Sigma-Aldrich) was injected over 2 min into the tail vein and allowed to circulate for 2 h. Thereafter, reanesthetized mice were perfused transcardially with icecold PBS to remove intravascular EB dye and were then decapitated immediately. The hemispheres were dissected, weighed, homogenized in N,N-dimethylformamide and incubated at 60 °C for 72 h. The homogenates were then centrifuged at 14000 rpm for 30 min, and absorbance of the supernatant was determined by a spectrophotometer at an excitation wavelength of 620 nm. The tissue EB concentration was quantified using a linear standard curve, and was expressed as micrograms per gram of brain tissue.

2.14. Statistical analysis

Data were presented as the mean ± standard error of the mean (SEM) and were analyzed using SPSS statistical software (version 22.0, IBM). Kruskal-Wallis H analysis followed by a Mann-Whitney U test were used to analyze data from mNSS tests (non-parametric data). Data from the rota-rod test and MWM test were analyzed using two-way ANOVA with repeated measures followed by a post-hoc Tukey multiple comparison test. Other data were analyzed using one-way ANOVA with Tukey post hoc comparisons or two-tailed Student’s t-test. A p value < 0.05 was considered significant.

3. Results

3.1. NLRP3 inflammasome expression was increased and was primarily observed in microglia after TBI

To determine whether NLRP3 inflammasome was involved in TBI, we first analyzed the protein expression profiles of NLRP3 inflammasome components (NLRP3, ASC, and caspase-1 p20) in the peri-contusional cortex at different time points. We found that NLRP3 (Fig. 1C), ASC (Fig. 1D), and caspase-1 p20 (Fig. 1E) were significantly increased at 1 d and peaked at 3 d post-TBI. Following this peak, the expression of these proteins decreased but remained higher than those in the sham group at 7 d post-TBI. Based on these results, we performed further studies focused on the time point of 3 d post-TBI to maximize NLRP3 inflammasome expression. In addition, we further found that NLRP3 was mainly expressed in CD11b + CD45int microglia, with low expression in CD11b + CD45hi infiltrated myeloid lineage cells, and no expression in GFAP+ astrocytes, NeuN+ neurons, and CD11b-CD45hi other leukocytes after TBI (Fig. 1F–G). Double immunofluorescence staining was then performed to confirm the cell source of post-TBI upregulated NLRP3 inflammasome. As indicated in Fig. 1H, TBI induced robust Iba-1-positive microglial activation, and microglial activation was accompanied by strong expression of NLRP3 in the pericontusional cortex. These results indicated that microglia-derived NLRP3 inflammasome may be primarily involved in the inflammatory response to TBI.

3.2. MCC950 administration inhibited NLRP3 inflammasome activation and IL-1β production after TBI

We further examined the role of MCC950 in NLRP3/ASC/caspase-1 activation and the subsequent secretion of mature IL-1β using western blotting and qRT-PCR analyses. As shown in Fig. 2B–C, compared with the vehicle-treated group, the protein level of NLRP3 was significantly decreased in the MCC950-treated group (p < 0.05). Similar trends were observed in the expression of ASC (p < 0.05), caspase-1 p20 (p < 0.05), and IL-1β (p < 0.05). Consistent with the western blotting results, qRT-PCR analysis revealed that the mRNA levels of NLRP3 (p < 0.01), ASC (p < 0.05), caspase-1 p20 (p < 0.05), and IL-1β (p < 0.05) were significantly downregulated in the MCC950-treated group compared with the vehicle-treated group (Fig. 2D). Furthermore, flow cytometric analysis revealed that MCC950-treated mice exhibited significantly lower counts of NLRP3-expressing microglia (CD11b + CD45intNLRP3+ cells) than the vehicle-treated mice (Fig. 2E, p < 0.001). Immunofluorescence analysis confirmed that MCC950 treatment significantly reduced the immunofluorescence staining of NLRP3 in microglia (Fig. 2F). These results indicated that MCC950 could also effectively inhibit the activation of microglia-derived NLRP3 inflammasome in the injured brain after TBI.

3.3. MCC950 administration ameliorated neurological deficits, brain edema, and cortical lesion size after TBI

We performed mNSS, rota-rod, and MWM tests to determine whether MCC950 influenced the recovery of neurological functions in mice after TBI. In the mNSS test, as shown in Fig. 3A, mice in the TBI groups exhibited significantly higher scores than mice in the sham group at all tested time points. On the 3, 7, and 14 d post-TBI, the mNSS scores was significantly lower in the MCC950-treated group than in the corresponding vehicle-treated group (all p < 0.05). Similarly, at 3, 7 and 14 d post-TBI, significant improvements in the rota-rod test were detected in the MCC950-treated group compared with the vehicle-treated group (Fig. 3B, p < 0.05 or 0.01). Furthermore, the effects of MCC950 on spatial learning and memory were also explored. Escape latency, which represents the time to navigate and locate the hidden platform, was gradually decreased from the 14 to 18 d post-TBI, indicating that the development of spatial acquisition (Vorhees and Williams, 2006). Compared with the vehicle-treated group, escape latency was significantly reduced in the MCC950-treated group at 17 and 18 d after TBI (Fig. 3C, p < 0.05). No difference was detected in swim speed over the 6 consecutive testing days among the groups (data not shown), indicating that the changes in spatial acquisition were not due to TBIinduced motor impairments. The hidden platform was removed on the last testing day (19 d post-injury), and the probe trial was performed to evaluate the reference memory. The time spent in the goal quadrant was significantly increased in the MCC950-treated group compared with the vehicle-treated group (Fig. 3D, p < 0.05). These data indicated that MCC950 improved spatial learning and memory recovery after TBI. We then further determined the effects of MCC950 on brain edema by measuring brain water content. Treatment with MCC950 caused a significant reduction in the percent of water content within the ipsilateral hemisphere compared with the vehicle-treated group at 3 d post-TBI (Fig. 3E, p < 0.05). In addition, we also evaluated the effect of MCC950 on brain lesion volume at 21 d after TBI induction. As shown in Fig. 2f, sham-injured mice showed no gross lesions in the cerebral cortex. However, TBI resulted in a significant loss of brain tissue, which was significantly reduced with MCC950 administration (p < 0.05). Notably, this result was concomitant with the improvements in neurological outcome.

3.4. MCC950 administration reduced microglia activation and leukocyte infiltration, and altered inflammatory cytokine secretion after TBI

To determine the effect of MCC950 on neuroinflammation following TBI, we used flow cytometry to measure cellular components, including microglia and brain-infiltrating leukocytes, in the TBI-afflicted brains (Fig. 4). Flow cytometric analysis revealed that MCC950 treatment significantly reduced the counts of microglia (CD11b + CD45int; Fig. 4B, p < 0.05), macrophages (CD11b + CD45hiLy6G-; Fig. 4C, p < 0.01), neutrophils (CD11bCD45hi + Ly6G+; Fig. 4D, p < 0.05), CD4 + T cells (CD45hiCD3 + CD4+; Fig. 4E, p < 0.05), and CD8 + T cells (CD45hiCD3 + CD8+; Fig. 4F, p < 0.05), whereas no changes in the recruitment of B cells (CD45hiCD3-CD19+; Fig. 4G, p < 0.05) or NK cells (CD45hiCD3-NK1.1+; Fig. 4H, p < 0.05) were observed. ELISA analyses demonstrated that MCC950 treatment resulted in markedly reduced IL-6 (Fig. 4I, p < 0.05) secretion without affecting TNF-α (Fig. 4J, p < 0.05). In addition, we also found that MCC950 significantly upregulated the production of the anti-inflammatory cytokine IL-10 (Fig. 4K, p < 0.05).These results suggested that MCC950 could reduce neuroinflammation after TBI.

3.5. MCC950 administration attenuated BBB disruption after TBI

Post-traumatic inflammatory cascades have been shown to exacerbate TBI-induced BBB disruption (da Fonseca et al., 2014; Gao et al., 2015). We therefore explored whether NLRP3 inflammasome inhibition by MCC950 has a positive effect on BBB destruction. BBB permeability was determined by EB dye extravasation. We observed that the EB content in the ipsilateral hemisphere of the vehicle-treated TBI group was significantly increased compared with that in the sham group (Fig. 5A–B, p < 0.001), indicating severe breakdown of the BBB. However, this increase in EB content was significantly attenuated by MCC950 treatment at 3 d post-TBI (p < 0.01). Tight junction (TJ) proteins have been shown to have an important role in maintaining BBB integrity (Liu et al., 2017). To further investigate the impact of MCC950 on TJ proteins following TBI, we quantified the expression of claudin-5 and ZO-1 using immunofluorescence staining and western blotting at 3 d post-injury. The immunostaining results showed that expression of both proteins in the TBI groups was dramatically decreased compared with that in the sham group (Fig. 5C–D, p < 0.001). However, these apparent decreases in claudin-5 and ZO-1 expression in the lesioned boundary were rescued after MCC950 treatment (p < 0.01 or 0.001). Consistent with this finding, western blotting analysis confirmed that MCC950-treated mice exhibited lower reductions in the protein expression of these TJ proteins than the vehicle-treated mice (Fig. 5E–F, p < 0.05 or 0.001). Together, these results indicated that TBI-induced BBB disruption could be effectively rescued by MCC950 administration, possibly as a result of attenuated neuroinflammation.

3.6. MCC950 administration decreased cell apoptosis after TBI

The NLRP3 inflammasome is associated with the frequency of apoptosis (Cheng et al., 2017; Ismael et al., 2018; Sagulenko et al., 2013). We therefore investigated the effect of NLRP3 inhibition on cell death. Flow cytometry was used to measure cell apoptosis by determining the level of Annexin V-positive cells. Compared with vehicle treatment, MCC950 significantly reduced the percentage of Annexin Vexpressing cells in the brain at 3 d after TBI (Fig. 5G–H, p < 0.01). In addition, the protein level of cleaved (activated) caspase-3, a classic marker of apoptotic cells, was also detected using western blotting. The results confirmed that treatment with MCC950 significantly decreased the expression of cleaved caspase-3 compared with vehicle treatment at 3 d after TBI (Fig. 5I–J, p < 0.05). Taken together, these results suggested that MCC950 could attenuate caspase-3-mediated cell apoptosis, and this protective role may partially dependent on inflammatory inhibition.

3.7. The neuroprotective effect of MCC950 against TBI required the presence of microglia

Since microglia were the predominant cell type expressing NLRP3 following TBI, we sought to explore whether MCC950 induced neuroprotection by targeting microglia. The survival of microglia in the adult brain is fully dependent upon colony-stimulating factor 1 receptor (CSF1R) signaling (Elmore et al., 2014). Therefore, we used a selective CSF1R inhibitor, PLX3397, to eliminate microglia before TBI induction. The efficacy of microglial depletion by PLX3397 was determined using flow cytometric analysis (Fig. 6B). We found that PLX3397 treatment resulted in the elimination of ~90% of microglia (CD11b + CD45int cells) and NLRP3-expressing microglia (CD11b + CD45intNLRP3+ cells) before or after TBI (Fig. 6C–D). MCC950 treatment did not alter the counts of CD11b + CD45int, CD11b + CD45int NLRP3+, CD11b + CD45hi, and CD11b-CD45hi cells in PLX3397-treated mice with TBI (Fig. 6C–E, all p > 0.05). More importantly, the neuroprotective effect of MCC950 and its ability to attenuate brain edema were abolished in mice with TBI receiving PLX3397 (Fig.6F–H, all p > 0.05). These results demonstrated that microglia contributed to the neuroprotective effect of MCC950 after TBI. Of interest, compared with TBI + vehicle group, we also found that microglial elimination by PLX3397 treatment dramatically attenuated neurological deficits and brain edema at 3 d post-TBI (Fig. 6F–H, all p < 0.05).

3.8. The neuroprotective effect of MCC950 was limited to within 6 h after TBI

To investigate whether the delayed post-TBI treatment with MCC950 could still confer neuroprotection, MCC950 was given to mice 3 h, 6 h, and 12 h after TBI. As shown in Fig. 7B–D, we found that delayed administration of MCC950 at 3 h and 6 h post-TBI significantly attenuated neurological deficits and brain edema at 3 d post-TBI compared with the vehicle-treated group (all p < 0.05), which was similar to that observed when the drug was administered immediately after TBI. However, treatment initiated 12 h post-TBI did not result in any significant improvement (all p > 0.05). These results suggested that the therapeutic window for MCC950 in protecting against TBI was as long as 6 h.

4. Discussion

The present study provided the in vivo evidence that the selective small-molecule NLRP3 inflammasome inhibitor MCC950 significantly ameliorated TBI-induced brain inflammation and long-term neurological deficits. The major findings were that (1) NLRP3 inflammasome expression was significantly increased in the peri-contusional cortex in mice post-TBI, and microglia were the primary source of NLRP3 inflammasome expression; (2) administration of MCC950 attenuated microglia-derived NLRP3 inflammasome activation and subsequent secretion of mature IL-1β; (3) MCC950 treatment reduced brain edema, tissue loss, microglia activation, leukocyte infiltration, BBB disruption, and cell death, and improved long-term neurological function following TBI; (4) the neuroprotective effect of MCC950 against TBI was abolished when microglia were depleted; and (5) the therapeutic window for MCC950 in protecting against TBI was as long as 6 h.
Recently, the NLRP3 inflammasome, one of the key components of the innate immune system, has been implicated in the sterile inflammatory response via processing caspase-1 and IL-1β in the setting of TBI (Liu et al., 2013; Mortezaee et al., 2018). It has been reported that potassium efflux, mitochondrial ROS generation, and cathepsin release after lysosomal destabilization are the three major signals that trigger the NLRP3 inflammasome activation (Gao et al., 2017). Once activated, the NLRP3 inflammasome can form a molecular platform for caspase-1 activation, which leads to the maturation and release of IL-1β and IL-18 and induces pyroptosis, eventually amplifying the inflammatory responses (Liu et al., 2013; Zhou et al., 2016). The protein and mRNA levels of NLRP3 inflammasome components and subsequent IL-1β and IL-18 have been shown to increase within hours in patients with TBI and in murine TBI models (Irrera et al., 2017; Lin et al., 2017; Liu et al., 2013; Ma et al., 2017). In addition, an increased level of NLRP3 was found in the cerebrospinal fluid (CSF) of children with severe TBI and was associated with poor prognosis (Wallisch et al., 2017). Co-immunoprecipitation studies showed the NLRP3 interacts with ASC following TBI, suggesting formation of the inflammasome complex (Liu et al., 2013; Ma et al., 2017). In the present study, our results confirmed that NLRP3 inflammasome components (NLRP3, ASC, and caspase-1 p20) were significantly increased from 1 d to 7 d and peaked at 3 d postTBI. Our results are different from those of a study conducted by Liu et al. in a rat TBI model, which indicated that NLRP3 inflammasome protein levels were enhanced in a time-dependent manner until 7 d post-TBI (Liu et al., 2013). The different TBI models, the magnitude of brain injury, and the different animals used could be possible explanations for this discrepancy. In addition, the specific cell-type distribution of NLRP3 inflammasome following TBI remains controversial (Mortezaee et al., 2018). Liu et al. (2013) reported that the NLRP3 inflammasome was detected in neurons, microglia, and astrocytes, while data from other studies highlighted that NLRP3 was mainly expressed in microglia, and not in astrocytes or neurons (Gustin et al., 2015; Qian et al., 2017). Here, using flow cytometry and double immunofluorescence staining, we confirmed that microglia were the primary source of the NLRP3 inflammasome expression. As caspase-1 activation in inflammasomes has been confirmed as a major mechanism responsible for IL-1β production (McKee and Lukens, 2016; Netea et al., 2015), our results may support the notion that microglia are the main cell type in the brain responsible for IL-1β and IL-18 secretion (Gustin et al., 2015; Pan et al., 2014).
MCC950 is a novel, potent, selective, small-molecule NLRP3 inflammasome inhibitor that was reported to inhibit NLRP3 inflammasome formation, reduce IL-1β and IL-18 production, and attenuate pyroptosis both in vitro and in vivo (Coll et al., 2015; Ismael et al., 2018; Mridha et al., 2017; Ren et al., 2018; van Hout et al., 2017). Ismael et al. indicated that MCC950 was also effective in the setting of TBI (Ismael et al., 2018). In line with these data, we confirmed that intraperitoneal administration of MCC950 dramatically inhibited the activation of NLRP3 inflammasome and the subsequent IL-1β production in experimental TBI, without affecting TNF-α levels. In addition, we found that MCC950 treatment exerted long-lasting neuroprotective effects in mice with TBI, as demonstrated by an improvement in longterm neurobehavioral functions, attenuation of brain edema, and a reduction in lesion volume. Notably, we showed that delayed MCC950 treatment (up to 6 h post-TBI) could still confer neuroprotection, suggesting that this compound has a wide therapeutic time window, which is useful in clinical practice.
More importantly, we found that the protective effects of MCC950 were abolished in microglia-depleted mice. In view of this result and the finding that microglia were the primary source of NLRP3 inflammasome after TBI, we postulated that the benefit of MCC950 involved its action on microglia to dampen brain damage post-TBI. In other words, microglia may be the key cellular mediators of NLRP3 inflammasome-mediated tissue damage. Moreover, the finding of low NLRP3 inflammasome expression in CD11b + CD45hi infiltrated myeloid lineage cells indicated that MCC950 may also exert an effect on these blood-borne leukocytes to reduce inflammatory brain injury. Nevertheless, the exact protective mechanisms of MCC950 in TBI require further study. Additionally, the impact of microglia on brain injury is highly controversial and disease-dependent (Szalay et al., 2016). Using the selective, brain-penetrant microglia inhibitor PLX3397, it has been reported that microglia exert protective roles in the setting of ischemic stroke and Parkinson’s disease (Jin et al., 2017; Szalay et al., 2016; Yang et al., 2018), while exerting detrimental roles in hemorrhagic stroke (Li et al., 2017a). In the present study, we found that microglial elimination by PLX3397 treatment dramatically reduced TBI-induced neurodeficits and brain edema, indicating a potential detrimental role of microglia during the acute phase of TBI. However, additional research is required to clarify the exact role of microglia in TBI and the underlying mechanisms.
Trauma to the brain results in cellular injury and BBB disruption, CNS-resident microglia and peripheral immune cells are immediately activated and infiltrate into the site of injury, contributing to the inflammatory brain damage via the generation and release of inflammatory mediators, such as IL-1β and IL-18 (Corps et al., 2015; McKee and Lukens, 2016). The involvement of IL-1β in post-traumatic neuroinflammation has long been speculated, and it has been shown to be elevated within hours in the brain parenchyma and CSF in humans and rodent models (Woodcock and Morganti-Kossmann, 2013). During neuroinflammation, IL-1β has been reported to have profound effects on increasing other pro-inflammatory cytokines such as IL-6, activating and recruiting microglial and leukocytes, disrupting the BBB by destroying the endothelial TJ proteins, and inducing apoptosis (McKee and Lukens, 2016; Woodcock and Morganti-Kossmann, 2013). In the present study, accompanied by the selective NLRP3 inflammasome and IL-1β inhibition, MCC950 administration significantly attenuated microglia activation, leukocyte (macrophages, neutrophils, and T lymphocytes) infiltration, BBB disruption, and cell death.
In pre-clinical studies, numerous pharmacologic agents targeting the NLRP3 signaling pathways have been explored to ameliorate brain injury after TBI. To date, omega-3 fatty acids (Lin et al., 2017), apocynin (Ma et al., 2017), telmisartan (Wei et al., 2016), propofol (Ma et al., 2016), and resveratrol (Zou et al., 2018) have been used to suppress NLRP3 activation after TBI. However, these pharmacological therapies either indirectly or non-specifically prevented NLRP3 activation, which may result in a higher incidence of side effects when translated to clinical practice (Ren et al., 2018). In addition, neutralizing or antagonizing the activity of IL-1β has been proven to exert neuroprotective effects following TBI (Kumar and Loane, 2012). Despite their notable efficacy, anti-IL-1 biologicals have no effect on caspase-1-mediated pyroptosis and may increase the risk of immunosuppression (Ren et al., 2018; Song et al., 2017). MCC950 specifically inhibits activation of the NLRP3 inflammasome but not the major antimicrobial NLRC4 or NLRP1 inflammasome (Coll et al., 2015), indicating that MCC950 treatment will not lead to complete blockade of IL-1β and that the antimicrobial responses will remain intact, thus avoiding increased susceptibility to infection. Moreover, our data showed that treatment with MCC950 after TBI was well-tolerated, as there was no body weight loss or other discernible adverse behavioral effects (data not shown). Importantly, MCC950 has been reported to have better CNS penetration (Chen et al., 2017). These findings suggested that MCC950 may be a promising drug candidate for future clinical trials. However, this study had some limitations. We selected a dose that was protective in the setting of EAE (Coll et al., 2015) and ICH (Ren et al., 2018) without performing a dose–response study. To improve the clinical relevance of MCC950, further studies are needed to determine the optimal dose, the frequency of administration, and the therapeutic window of MCC950 treatment for TBI and the associated underlying mechanisms of action.

5. Conclusions

The present study demonstrated that microglia-derived NLRP3 inflammasome may be primarily involved in the post-traumatic inflammatory response. Blocking the NLRP3 inflammasome pathway with MCC950 attenuated inflammatory brain injury and behavior deficits in a mouse TBI model. These results suggested that MCC950 might serve as a promising therapeutic candidate for intervention in patients with TBI.

References

Adamczak, S., et al., 2012. Inflammasome proteins in cerebrospinal fluid of brain-injured patients as biomarkers of functional outcome: clinical article. J. Neurosurg. 117, 1119–1125.
Chen, W., et al., 2017. Specific inhibition of NLRP3 in chikungunya disease reveals a role for inflammasomes in alphavirus-induced inflammation. Nat Microbiol. 2, 1435–1445.
Cheng, Y., et al., 2017. Cordycepin confers neuroprotection in mice models of intracerebral hemorrhage via suppressing NLRP3 inflammasome activation. Metab.Brain Dis. 32, 1133–1145.
Coll, R.C., et al., 2015. A small-molecule inhibitor of the NLRP3 inflammasome for the treatment of inflammatory diseases. Nat. Med. 21, 248–255.
Corps, K.N., et al., 2015. Inflammation and neuroprotection in traumatic brain injury.JAMA Neurol. 72, 355–362.
da Fonseca, A.C., et al., 2014. The impact of microglial activation on blood-brain barrier in brain diseases. Front. Cell. Neurosci. 8, 362. de Rivero Vaccari, J.P., et al., 2014. Activation and regulation of cellular inflammasomes: gaps in our knowledge for central nervous system injury. J. Cereb. Blood Flow Metab. 34, 369–375.
Elmore, M.R., et al., 2014. Colony-stimulating factor 1 receptor signaling is necessary for microglia viability, unmasking a microglia progenitor cell in the adult brain. Neuron 82, 380–397.
Gao, W., et al., 2015. VEGI attenuates the inflammatory injury and disruption of bloodbrain barrier partly by suppressing the TLR4/NF-kappaB signaling pathway in experimental traumatic brain injury. Brain Res. 1622, 230–239.
Gao, L., et al., 2017. NLRP3 inflammasome: a promising target in ischemic stroke.Inflamm. Res. 66, 17–24.
Ge, X.T., et al., 2014. miR-21 improves the neurological outcome after traumatic brain injury in rats. Sci. Rep. 4, 6718.
Gustin, A., et al., 2015. NLRP3 inflammasome is expressed and functional in mouse brain microglia but not in astrocytes. PLoS One 10, e0130624.
Harrison, J.L., et al., 2015. Resolvins AT-D1 and E1 differentially impact functional outcome, post-traumatic sleep, and microglial activation following diffuse brain injury in the mouse. Brain Behav. Immun. 47, 131–140.
Irrera, N., et al., 2017. Lack of the Nlrp3 inflammasome improves mice recovery following traumatic brain injury. Front. Pharmacol. 8, 459.
Ismael, S., et al., 2018. MCC950, the selective inhibitor of nucleotide oligomerization domain-like receptor protein-3 inflammasome protects mice against traumatic brain injury. J Neurotrauma. 35, 1294–1303.
Jin, W.N., et al., 2017. Depletion of microglia exacerbates postischemic inflammation and brain injury. J. Cereb. Blood Flow Metab. 37, 2224–2236.
Kumar, A., Loane, D.J., 2012. Neuroinflammation after traumatic brain injury: opportunities for therapeutic intervention. Brain Behav. Immun. 26, 1191–1201.
Lee, J., et al., 2014. Altering leukocyte recruitment following traumatic brain injury with ghrelin therapy. J. Trauma Acute Care Surg. 77, 709–715.
Li, M., et al., 2017a. Colony stimulating factor 1 receptor inhibition eliminates microglia and attenuates brain injury after intracerebral hemorrhage. J. Cereb. Blood Flow Metab. 37, 2383–2395.
Li, M., et al., 2017b. A TSPO ligand attenuates brain injury after intracerebral hemorrhage. FASEB J. 31, 3278–3287.
Lin, C., et al., 2017. Omega-3 fatty acids regulate NLRP3 inflammasome activation and prevent behavior deficits after traumatic brain injury. Exp. Neurol. 290, 115–122.
Liu, H.D., et al., 2013. Expression of the NLRP3 inflammasome in cerebral cortex after traumatic brain injury in a rat model. Neurochem. Res. 38, 2072–2083.
Liu, Y.L., et al., 2017. Sesamin alleviates blood-brain barrier disruption in mice with experimental traumatic brain injury. Acta Pharmacol. Sin. 38, 1445–1455.
Ma, J., et al., 2016. Propofol inhibits NLRP3 inflammasome and attenuates blast-induced traumatic brain injury in rats. Inflammation 39, 2094–2103.
Ma, M.W., et al., 2017. NADPH oxidase 2 regulates NLRP3 inflammasome activation in the brain after traumatic brain injury. Oxidative Med. Cell. Longev. 2017, 6057609.
McKee, C.A., Lukens, J.R., 2016. Emerging roles for the immune system in traumatic brain injury. Front. Immunol. 7, 556.
Mortezaee, K., et al., 2018. Inflammasome: its role in traumatic brain and spinal cord injury. J. Cell. Physiol. 233, 5160–5169.
Mridha, A.R., et al., 2017. NLRP3 inflammasome blockade reduces liver inflammation and fibrosis in experimental NASH in mice. J. Hepatol. 66, 1037–1046.
Netea, M.G., et al., 2015. Inflammasome-independent regulation of IL-1-family cytokines.Annu. Rev. Immunol. 33, 49–77.
Pan, Y., et al., 2014. Microglial NLRP3 inflammasome activation mediates IL-1beta-related inflammation in prefrontal cortex of depressive rats. Brain Behav. Immun. 41, 90–100.
Qian, H., et al., 2017. Hyperbaric oxygen alleviates the activation of NLRP3inflammasomes in traumatic brain injury. Mol. Med. Rep. 16, 3922–3928.
Ren, H., et al., 2018. Selective NLRP3 (pyrin domain-containing protein 3) inflammasome inhibitor reduces brain injury after intracerebral hemorrhage. Stroke 49, 184–192.
Sagulenko, V., et al., 2013. AIM2 and NLRP3 inflammasomes activate both apoptotic and pyroptotic death pathways via ASC. Cell Death Differ. 20, 1149–1160.
Simon, D.W., et al., 2017. The far-reaching scope of neuroinflammation after traumatic brain injury. Nat. Rev. Neurol. 13, 572.
Song, L., et al., 2017. NLRP3 inflammasome in neurological diseases, from functions to therapies. Front. Cell. Neurosci. 11, 63.
Szalay, G., et al., 2016. Microglia protect against brain injury and their selective elimination dysregulates neuronal network activity after stroke. Nat. Commun. 7, 11499.
van Hout, G.P., et al., 2017. The selective NLRP3-inflammasome inhibitor MCC950 reduces infarct size and preserves cardiac function in a pig model of myocardial infarction. Eur. Heart J. 38, 828–836.
Vorhees, C.V., Williams, M.T., 2006. Morris water maze: procedures for assessing spatial and related forms of learning and memory. Nat. Protoc. 1, 848–858.
Wallisch, J.S., et al., 2017. Cerebrospinal fluid NLRP3 is increased after severe traumatic brain injury in infants and children. Neurocrit. Care. 27, 44–50.
Wang, X., et al., 2014. RNA viruses promote activation of the NLRP3 inflammasome through a RIP1-RIP3-DRP1 signaling pathway. Nat. Immunol. 15, 1126–1133.
Wang, Z.G., et al., 2016. bFGF protects against blood-brain barrier damage through junction protein regulation via PI3K-Akt-Rac1 pathway following traumatic brain injury. Mol. Neurobiol. 53, 7298–7311.
Wei, X., et al., 2016. Telmisartan reduced cerebral edema by inhibiting NLRP3 inflammasome in mice with cold brain injury. J. Huazhong Univ. Sci. Technolog. Med. Sci. 36, 576–583.
Woodcock, T., Morganti-Kossmann, M.C., 2013. The role of markers of inflammation in traumatic brain injury. Front. Neurol. 4, 18.
Xu, X., et al., 2017. Anti-inflammatory and immunomodulatory mechanisms of atorvastatin in a murine model of traumatic brain injury. J. Neuroinflammation 14, 167.
Yang, L., et al., 2017. DRalpha1-MOG-35-55 treatment reduces lesion volumes and improves neurological deficits after traumatic brain injury. Metab. Brain Dis. 32, 1395–1402.
Yang, X., et al., 2018. Depletion of microglia augments the dopaminergic neurotoxicity of MPTP. FASEB J. http://dx.doi.org/10.1096/fj.201700833RR.
Zhou, K., et al., 2016. Recent advances of the NLRP3 inflammasome in central nervous system disorders. J Immunol Res 2016, 9238290.
Zou, P., et al., 2018. Resveratrol pretreatment attenuates traumatic brain injury in rats by suppressing NLRP3 inflammasome activation via SIRT1. Mol. Med. Rep. 17, 3212–3217.