ST2825, a Small Molecule Inhibitor of MyD88, Suppresses NF-κB Activation and the ROS/NLRP3/Cleaved Caspase-1 Signaling Pathway to Attenuate Lipopolysaccharide-Stimulated Neuroinflammation

Neuroinflammation characterized by microglia activation is the mechanism of the occurrence and development of various central nervous system diseases. ST2825, as a peptide-mimetic MyD88 homodimerization inhibitor, has been identified as crucial molecule with an anti-inflammatory role in several immune cells, especially microglia. The purpose of the study was to investigate the anti-neuroinflammatory effects and the possible mechanism of ST2825. Methods: Lipopolysaccharide (LPS) was used to stimulate neuroinflammation in male BALB/c mice and BV2 microglia cells. The NO level was determined by Griess Reagents. The levels of pro-inflammatory cytokines and chemokines were determined by ELISA. The expressions of inflammatory proteins were determined by real-time PCR and Western blotting analysis. The level of ROS was detected by DCFH-DA staining. Results: In vivo, the improved levels of LPS-induced pro-inflammatory factors, including TNF-α, IL-6, IL-1β, MCP-1 and ICAM-1 in the cortex and hippocampus, were reduced after ST2825 treatment. In vitro, the levels of LPS-induced pro-inflammatory factors, including NO, TNF-α, IL-6, IL-1β, MCP-1, iNOS, COX2 and ROS, were remarkably decreased after ST2825 treatment. Further research found that the mechanism of its anti-neuroinflammatory effects appeared to be associated with inhibition of NF-κB activation and down-regulation of the NLRP3/cleaved caspase-1 signaling pathway. Conclusions: The current findings provide new insights into the activity and molecular mechanism of ST2825 for the treatment of neuroinflammation.


Introduction
Neuroinflammation is characterized by microglia activation, which is closely related to the pathogenesis of neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease, Multiple sclerosis, Huntington's disease, etc. Microglia are the resident immune cells of the central nervous system, which play an important role in maintaining tissue homeostasis under normal conditions. However, when there is a neuronal injury or other insult, microglia will be activated to secrete pro-inflammatory factors. Excessive microglial activation damages the surrounding healthy neural tissue [1][2][3]. In response to brain tissue damage, activated microglia release a high concentration of NO which, as a chemoattractant, recruits more microglia to migrate to the injury site, promotes its chronic activation and contributes to long-term neurodegeneration [3,4].
In general, microglia express numerous pattern recognition receptors such as the Toll-like receptor (TLR) 4, which can be activated by LPS, a vital molecular component of the cell outer membrane of gram-negative bacteria, followed by the activation of NF-kB and inflammatory cascades [5][6][7][8]. The Myeloid differentiation primary response protein 88 (MyD88) is an adaptor protein for the TLR4 and interleukin-1 receptor (IL-1R) families, leading to cytokine release, transformation of cell phenotype and migration of microglia and macrophages to injury sites [9]. Due to this fact, the identification of new molecules that regulate the TLR4/MyD88 signaling pathway has become a priority. ST2825 (Figure 1), a synthetic analogue of MyD88, is a MyD88-specific inhibitor, by virtue of interfering with MyD88 homodimerization. ST2825 has been applied in different models of human diseases including lymphoma, leukemia, human hepatocellular carcinoma and traumatic brain injury [6]. In particular, the anti-inflammatory effects of ST2825 have been demonstrated through the inhibition of various pro-inflammatory cytokines in PBMC [6] and RAW264.7 cells [10] stimulated by LPS. Given these findings, the purpose of this study was to investigate the effects of ST2825 against LPS-induced inflammation, and to analyze the possible molecular mechanism involved.

ST2825 Reduced Pro-Inflammatory Factors Production in LPS-Stimulated Mice
In order to investigate whether ST2825 has an anti-inflammatory effect in vivo, we detected the levels of cytokines (such as TNF-α, IL-1β and IL-6), chemokine (such as MCP-1), and adhesion molecules (such as ICAM-1), in the cerebral cortex and hippocampus of BALB/c mice stimulated by LPS. Based on our previous experience in LPS-stimulated mice models [7], 5 mg/kg of ST2825 was selected for administration. For the administration group, mice were treated with ST2825 6 h before LPS stimulation, taking into consideration that ST2825 is a specific inhibitor of the MyD88 TIR domain. Our data revealed that, compared with the control group, these pro-inflammatory factors-especially IL-1β and MCP-1 (p < 0.01)-were significantly higher in the cortex and hippocampus of the LPS model group, and that ST2825 apparently reduced the levels of these inflammatory factors, except for ICAM-1 in the hippocampus (p < 0.01, Figure 2).

ST2825 Decreased Pro-Inflammatory Factors and Increased Anti-Inflammatory Factors in LPS-Stimulated BV2 Cells
The CCK-8 assay showed that ST2825 itself had no obvious cytotoxicity at 1, 3 and 10 µM in the BV2 microglial cells after 24 h treatment ( Figure 3A), therefore we chose concentrations at 10 µM or lower for all subsequent experiments. As an initial indicator of the anti-neuroinflammatory effects of ST2825 on LPS-stimulated BV2 cells, the concentration of NO was assessed. The results showed that ST2825 pretreatment markedly decreased the level of NO (p < 0.01, Figure 3B). In order to further explore the anti-neuroinflammatory effects of ST2825 on LPS-stimulated BV2 cells, we detected the levels of inflammatory cytokines, including TNF-α, IL-1β, IL-6 and chemokine MCP-1. As shown in Figure 3C-F, compared with the control group, the levels of these pro-inflammatory factors in the LPS model group increased significantly (p < 0.01), while ST2825 pretreatment observably decreased their levels (p < 0.01). The above results suggested that ST2825 inhibited the production of these pro-inflammatory factors. As shown in Figure 3G,H, compared with the control group, the levels of anti-inflammatory factors in the LPS model group decreased significantly (p < 0.01), while ST2825 pretreatment observably increased their levels (p < 0.01) in a dose-dependent manner, which suggested that ST2825 promotes the production of these anti-inflammatory factors.

ST2825 Decreased the Expression of iNOS and COX-2 in LPS-Stimulated BV2 Cells
As shown in Figure 4, the mRNA levels of iNOS and COX-2 in the LPS model group increased in varying degrees after LPS stimulation for 6 h (p < 0.01), and ST2825 pretreatment significantly decreased the transcriptional levels of iNOS (p < 0.01, Figure 4A) and COX-2 (p < 0.01, Figure 4B). This observation led us to further investigate the effect of ST2825 on the protein expression of iNOS and COX-2 in LPS-treated BV2 cells. The results showed that, compared with the control group, the protein expression levels of iNOS and COX-2 in the LPS model group increased to about eight times (p < 0.01), while ST2825 pretreatment observably decreased the protein expression of iNOS (p < 0.01, Figure 4C) and COX-2 (p < 0.01, Figure 4D).

ST2825 Inhibited ROS Accumulation in LPS-Stimulated BV2 Cells
The inhibitory effect of ST2825 on the production of ROS was also analyzed by flow cytometry. As shown in Figure 6, the level of ROS increased after LPS stimulation (p < 0.01), while in ST2825-pretreated BV2 cells, the production of LPS-stimulated ROS was almost completely blocked (p < 0.01). These findings indicated that ST2825 had a strong ROSscavenging effect.

Discussion
The central nervous system (CNS) is populated by neurons and glial cells, in which microglia account for 10-15% of the total number of adult brain glial cells [11]. Microglia play an important role in the regulation of innate immunity and neuroinflammation, and actively regulate the CNS development and maintain homeostasis [12]. However, under pathological conditions, neuronal cell death leads to microglia abnormal activation and the production of excessive neurotoxic factors, including NO, prostaglandin E2 (PGE 2 ), ROS and pro-inflammatory cytokines such as TNF-a, IL-1β and IL-6, which in turn leads to the gradual loss of neurons and the aggravation of inflammatory response [2,11]. Therefore, intervention in the activation process of microglia may be a promising method for the treatment of various CNS diseases.
Pathogens such as lipopolysaccharide (LPS) of Gram-negative bacteria, the peptidoglycan (PGN) of Gram-positive bacteria, β-amyloid, bacteria, and viruses can cause neuroinflammation, which is arguably the most destructive component of neurodegenerative processes initiated in part through the TLR4 signaling pathway activation [2,13,14]. LPS injection is a common method of performing an inflammatory response model, and is well-reproduced [7]. Upon stimulation with LPS, two classic signaling pathways are activated: MyD88-dependent and MyD88-independent pathways [15]. MyD88 is an adaptor protein common to all TLR signaling pathways, such that it forms a receptor complexexcept TLR3 [16,17]-while the Toll/IL-1R (TIR) domain-containing adaptor-inducing IFN-β (TRIF) is solely engaged by TLR3 and TLR4, which mediate the induction of type 1 interferon-inducible genes [18,19].
MyD88 consists of an N-terminal dead domain (DD) and a C-terminal TIR domain, separated by a short linker region [20]. It plays an important role in the pathogenesis of Alzheimer's disease, Parkinson's disease, amyotrophic lateral sclerosis, multiple system atrophy and related disorders [21]. MyD88 knockout mice lack the ability to respond to LPS [22]. The functional core of the MyD88 molecule is the ability of its TIR domain to heterodimerize with the TIR domain-containing adaptor protein (TIRAP) [23], and to homodimerize with another MyD88 molecule in the TIR domain [20,24]. By interacting with the IL-1R associated kinase (IRAK)-4 through its dead domain, MyD88 activates other members of the IRAK family, such as IRAK-4 and subsequently IRAK-1. The above process leads to the activation of the tumor necrosis factor (TNF), the receptor-associated factor 6 (TRAF6) and the E2 ubiquitin protein ligase, which activates the TAK1/TAK1 binding protein (TAB) complex to trigger MAPK and the NF-κB signaling pathway [25]. Mechanistically, ST2825 blocks MyD88-dependent signaling by binding to the BB ring and specifically interfering with MyD88 homodimerization of the TIR domains rather than homodimerization of the death domains [26,27], resulting in the reduction of Myddsome formation and pro-inflammatory cytokines production [20] (Figure 7).
The Cytokine and chemokine families mediate both immune cell recruitment and the complex intracellular signaling control mechanisms that characterize inflammation [28]. Cytokines including TNF-α, IL-1β and IL-6, participate in acute and chronic inflammation via signaling the type I cytokine receptor. It is generally believed that pro-inflammatory cytokines in the brain are synthesized in CNS after stimulation by peripheral LPS treatment, rather than cytokines released from the periphery being diffused through the blood-brain barrier [29]. Chemokines and their receptors are able to control the migration and residence of all immune cells [30]. As a typical chemokine, MCP-1 (CCL2) stimulates chemotaxis of monocytes, cytokine expression and ROS generation induced by respiratory burst [30]. ICAM-1, an adhesion molecule, regulates leukocyte rolling and adhesion, guides leukocyte crossing of the endothelial layer and promotes the regression of inflammatory response [31]. Our results showed that LPS induced the release of pro-inflammatory factors including TNF-α, IL-1β, IL-6, MCP-1 and ICAM-1 in the cortex and hippocampus of LPS-induced mice, while ST2825 pretreatment apparently reduced the levels of these inflammatory factors, except ICAM-1 in the hippocampus tissue. The reason may have been that the regulation of ICAM-1 in the hippocampus by ST2825 required a higher dose or longer duration of action. Similarly, the levels of pro-inflammatory factors TNF-α, IL-1β, IL-6 and MCP-1 were increased in the LPS-stimulated BV2 cells, while ST2825 pretreatment significantly decreased these pro-inflammatory factors in a dose-dependent manner. NO is produced when iNOS catalyzes the conversion of L-arginine to L-citrulline, and is another way in which neuroinflammation can directly influence neuronal apoptosis [32]. Excessive NO will lead to the production of pro-inflammatory cytokines, and then aggravate the inflammatory response and physiological dysfunction [33]. COX-2 has been identified as a key enzyme involved in inflammatory response due to its ability to catalyze the production of a variety of prostaglandins [3]. In this study, our data demonstrated that ST2825 pretreatment remarkably decreased LPS-induced NO release in a dose-dependent manner, as well as restraining the transcription and expression levels of iNOS and COX-2 protein induced by LPS. Based on the above results, we hold that ST2825 exhibited prominent anti-inflammatory effects on LPS-stimulated mice and BV2 cells. The expression levels of inflammatory cytokines, including TNF-α and IL-6, are regulated by various transcription factors, one of the most important of which is NF-κB [15]. NF-κB is a heterologous trimer consisting of p50, p65, and IκBα subunits, and is normally inactive in the cytoplasm. It has been reported that ST2825 prevents subarachnoid hemorrhage-induced activation of the NF-κB signaling pathway [25]. Nevertheless, when cells are stimulated, IκB is phosphorylated, ubiquitinated and degraded, releasing NF-κB into the nucleus and inducing the transcription of pro-inflammatory cytokines. In the present study, ST2825 prominently inhibited the activation of NF-κB and IκBα in LPSstimulated BV2 microglia.
Inflammasomes, cytosolic multiprotein complexes, are responsible for the cleavage of caspase-1-mediated pro-inflammatory cytokines, such as IL-1β and IL-18 [33]. NLRP3 inflammasome is composed of NLRP3, ASC and caspase-1, and is one of the most typical inflammasomes. It is well-established that the activation of the TLR4/MyD88/NF-κB signaling pathway makes a great contribution to the activation of NLRP3 inflammasome. In this study, compared with the control group, the expression of NLRP3 and cleaved caspases-1 proteins increased significantly after LPS stimulation, indicating that the NLRP3/cleaved caspase-1 signaling pathway was activated after LPS stimulation. However, ST2825 pretreatment observably inhibited the LPS-induced up-regulation of NLRP3, cleaved caspase-1, mature IL-18 and IL-1β proteins, which may in turn break the IL-1β/MyD88/NF-κB/IL-1β cycle.
Oxidative stress, caused by a decreased anti-oxidant system and increased production of ROS, plays a vital role in the development of various degenerative diseases [3]. As a secondary messenger, intracellular ROS promotes inflammation by upregulating kinase cascades, which has attracted extensive attention [3]. ROS regulates the activation of NLRP3 inflammasome, promotes the release of various pro-inflammatory cytokines and accelerates the process of inflammatory response by altering kinase cascades and activating transcription factors, including MAPK and NF-κB [34]. After LPS stimulation, ROS accumulation in BV2 cells was greatly increased, while pretreatment with ST2825 almost completely counteracted LPS-induced ROS production, which proved that ST2825 exhibited a strong antioxidant effect.

Animals and Treatment
All animal care and animal experiment procedures were approved by the ethics committees of the Institute of Materia Medica, and the Chinese Academy of Medical Sciences & Peking Union Medical College. A total of 18 male BALB/c mice (weighing 18-22 g, 6-8 weeks old) were included in the experiment, and they were obtained from the Laboratory Animal Centre of Beijing Hua-Fu-Kang Bioscience Co., Ltd., (Beijing, China; the animal certification number was SCXK (Jing) 2014-0004). All animals were kept under a 12-h light/dark cycle, with lights with controlled temperature at 23 ± 2 • C and controlled humidity at 55% ± 5%. Three days after acclimation, the mice were divided into three groups: the control group (i.g. administration of saline); the LPS group (i.g. administration of saline 6 h before LPS injection and i.p. LPS 5 mg/kg) and the LPS + ST2825 5 mg/kg group (i.g. administration of ST2825 5 mg/kg 6 h before LPS injection and i.p. LPS 5 mg/kg) according to random number table, with 6 mice in each group. After 6 h of LPS stimulation, the mice were euthanized under 5% isoflurane anesthesia and decapitated according to the order of LPS stimulation. The cerebral cortex and hippocampus of the mice were isolated on ice and immediately frozen at −80°C for further study [7,35,36].

BV2 Microglia Cells Culture and Treatment
BV2 microglia cells were maintained in DMED supplemented with 10% heat-inactivated FBS at 37°C in a humidified incubator under 5% CO 2 . The BV2 cells were seeded into 96-well or 6-well plates, and pre-treated with a series of indicated concentrations of ST2825 (1, 3 and 10 µM) for 12 h before being incubated with or without 200 ng/mL LPS for another 24 h or 6 h. The ST2825 was dissolved with DMSO and diluted to the correspond-ing concentration with fresh DMEM. The final DMSO concentration was <0.05% in all experiments [37].

Cell Viability Assay
To determine the cell viability, BV2 cells were seeded into 96-well plates at a density of 2.5 × 10 4 cells per well, and treated with different concentrations of ST2825 (1, 3, and 10 µM) for 24 h. Afterwards, the medium was removed and 10 µL CCK-8 (Dojindo Molecular Technologies, Inc., Kumamoto Techno Research Park, Kumamoto, Japan.) was added to each well. After incubation at 37 • C for 2 h, the optical density was measured at a wavelength of 450 nm using a microplate reader [37].

NO Assay
The BV2 cells were treated with different concentrations of ST2825 for 12 h, then stimulated with 200 ng/mL LPS for 24 h, and the culture supernatants were collected. The nitrite concentration in the medium, an indicator of NO production, was measured with the Griess reaction using NO analysis kit (Applygen Technologies Inc., Beijing, China), in accordance with the manufacturer's instructions [37].

ELISA Assay
The cortex and hippocampus tissues of mice were homogenized in cool saline and centrifugated at 4500 rpm for 10 min at 4 • C. The concentration of protein in supernatant was quantified by the BCA (Thermo, Waltham, MA, USA) method. The levels of inflammatory factors TNF-a, IL-1β, IL-6, MCP-1 and ICAM-1 into cell culture supernatant or tissue supernatant were detected by ELISA kits in accordance with the manufacturer's instructions [37].

Intracellular Reactive Oxygen Species (ROS) Assay
After 24 h of LPS stimulation, the BV2 cells in the 6-well plates were collected and stained with 5 µM DCFH-DA (Beyotime Institute of Biotechnology, Haimen, China) in the dark for 30 min at 37°C. After washing twice with PBS, 10,000 viable BV2 cells were immediately analyzed in a flow cytometer (BD Biosciences, San Jose, CA, USA) with an excitation wavelength of 488 nm and an emission wavelength of 525 nm [42].

Statistical Analysis
The data obtained from each mouse were included in the statistics. All data were analyzed with GraphPad Prism 7.0 software (GraphPad Software, Inc., La Jolla, CA, USA) using one-way ANOVA and Dunnett's multiple comparisons test, and p < 0.05 was taken to be significant. Only the experimenters performing the experimental intervention knew the allocations of the experimental group. The data analysts remained blinded to the experimental group allocations until all analyses were completed.

Conclusions
ST2825 suppressed the LPS-induced inflammation in BV2 microglia cells via decreasing the release of inflammatory factors, inactivating NF-κB and the ROS/NLRP3/cleaved caspase-1 signaling pathway. This study suggests MyD88 as a potential therapeutic target, and ST2825 as a potential therapeutic agent for neuroinflammation in CNS disease. However, further research is needed to determine which immune pathways or molecules involved in neuroinflammation should be targeted.