USP8 protects against lipopolysaccharide-induced cognitive and motor deficits by modulating microglia phenotypes through TLR4/MyD88/NF-κB signaling pathway in mice

Ubiquitin-specific protease 8 (USP8) regulates inflammation in vitro; however, the mechanisms by which USP8 inhibits neuroinflammation and its pathophysiological functions are not completely understood. In this study, we aimed to determine whether USP8 exerts neuroprotective effects in a mouse model of lipopolysaccharide (LPS)-induced cognitive and motor impairment. We commenced intracerebroventricular USP8 administration 7 days prior to i.p. injection of LPS (750 μg/kg). All treatments and behavioral experiments were performed once per day for 7 consecutive days. Behavioral tests and pathological/biochemical assays were performed to evaluate LPS-induced hippocampal damage. USP8 attenuated LPS-induced cognitive and motor impairments in mice. Moreover, USP8 downregulated several pro-inflammatory cytokines [nitric oxide (NO), tumor necrosis factor α (TNF-α), prostaglandin E2 (PGE2), and interleukin-1β (IL-1β)] in the serum and brain, and the relevant protein factors [inducible nitric oxide synthase (iNOS) and cyclooxygenase 2 (COX-2)] in the brain. Furthermore, USP8 upregulated the anti-inflammatory mediators interleukin (IL)-4 and IL-10 in the serum and brain, and promoted a shift from pro-inflammatory to anti-inflammatory microglial phenotypes. The LPS-induced microglial pro-inflammatory phenotype was abolished by TLR4 inhibitor and in TLR4-/- mice; these effects were similar to those of USP8 treatment. Mechanistically, we found that USP8 increased the expression of neuregulin receptor degradation protein-1 (Nrdp1), potently downregulated the expression of TLR4 and myeloid differentiation primary response protein 88 (MyD88) protein, and inhibited the phosphorylation of IκB kinase (IKK) β and kappa B-alpha (IκBα), thereby reducing nuclear translocation of p65 by inhibiting the activation of the nuclear factor-kappaB (NF-κB) signaling pathway in LPS-induced mice. Our results demonstrated that USP8 exerts protective effects against LPS-induced cognitive and motor deficits in mice by modulating microglial phenotypes via TLR4/MyD88/NF-κB signaling.


Introduction
Neuroinflammation is a major cause of neurodegenerative diseases (Allison and Ditor, 2014), of which many remain incurable. Current clinical therapeutic strategies for progressive neurodegenerative disorders are designed to control symptoms rather than address the underlying cause of neurodegeneration (Herrero and Morelli, 2017). Un-production of neurotoxic mediators and inflammatory cytokines, which could aggravate the pathological symptoms of neurodegeneration (Nichols et al., 2019;Pimplikar, 2014). Mediators of neuroinflammation can directly affect cognition and memory, contributing to cognitive and motor impairment associated with several chronic neurological conditions (Jin et al., 2019). Strategies to prevent and/or ameliorate neuroinflammation are therefore essential to reduce brain damage and cognitive deficits (Verdile et al., 2015;Fan et al., 2014).
The neuroinflammatory response includes microglial activation resulting in a phagocytic phenotype, and subsequent release of inflammatory mediators (Cunningham, 2013). Microglial activation in the central nervous system is heterogeneous, and mainly results in the contrasting M1 and M2 phenotypes. M1 microglias are predominantly seen at injury sites at the end stage of disease, when the immune resolution and repair process of M2 microglia are dampened. (Tang and Le, 2016). The regulation of microglial polarization from the M1 to M2 phenotype could prove valuable in the development of therapeutic and preventive strategies against neurodegenerative diseases (Jin et al., 2019).
Microglial cells express TLR4 on their surface for actively monitoring the environment. TLR4 deficiency promotes a shift to the M2 microglial phenotype, which ameliorates neurological impairment (Yao et al., 2017). In response to lipopolysaccharide (LPS; a specific ligand for TLR4), microglia become hyper-activated, resulting in the production of cytotoxic factors such as NO, TNF-α, and PGE 2 (Hoogland et al., 2015). Although microglia are nervous system-specific immune cells that influence brain development, response to injury, and tissue repair, excessive cytokine production and microglial activation can cause systemic inflammatory responses including neuroinflammation, which could impair cognitive and motor function (Rice et al., 2017). Subsequent microglial activation promotes the release of more inflammatory factors, thus initiating a cycle that often leads to neuronal death (Frank-Cannon et al., 2009).
LPS primarily activates the MyD88-dependent and independent pathways, which involve recognition of the lipid A-region of LPS by TLR4, suggesting that MyD88 pathways play important roles in inflammation and immune response (Rahimifard et al., 2017). Microglia are the main immune cells of the central nervous system that express TLR4, while LPS is a widely expressed bacterial TLR4 ligand that activates the innate immune response to infections (Döring et al., 2017). TLR4 could modulate the NF-κB inflammatory cascade, leading to neurodegeneration (Lysakova-Devine et al., 2010). Inhibiting microglial activation and the resulting neuroinflammation could serve as an adjunctive treatment for the progressive neurodegenerative disorders. The ubiquitin proteasome system (UPS) is an efficient protein degradation pathway (Gallastegui and Groll, 2010), which regulates short half-life regulatory proteins in cells and structurally abnormal, misfolded, or damaged proteins. The UPS system includes ubiquitin and several other enzymes. USP8 is a member of DUBs, and regulates the stability of ubiquitin protein ligases (E3s) including Nrdp1, which is an E3 expressed primarily in the brain, heart, prostate, and skeletal muscle (Soares et al., 2004). Nrdp1, also known as FLRF or RBCC, is a newly defined ring finger (Printsev et al., 2014) that is involved in regulating cell growth, apoptosis, and oxidative stress in these tissues (Zhang et al., 2011). Nrdp1 inhibits the production of proinflammatory cytokines but increases interferon-β production in toll-like receptor-triggered macrophages by suppressing adaptor MyD88-dependent activation of NF-kB and activator protein-1 (AP-1) while promoting activation of the kinase TBK1 and transcription factor IRF3 ). Nrdp1 is a specific target for USP8 deubiquitinating enzyme, and USP8 could augment Nrdp1 activity by mediating its stabilization (Wu et al., 2004). We previously reported a significant reduction in the degradation of Nrdp1 in BV2-immortalized murine microglial cells after transfection with USP8. USP8 overexpression also reduced the production of LPS-induced proinflammatory mediators. USP8 could therefore be a novel candidate for the treatment of neuroinflammatory disorders (Zhu et al., 2015).
The mechanism by which USP8 inhibits inflammation in vivo remains unclear. We hypothesized that USP8 could increase the expression of Nrdp1 and inhibit neuroinflammation by modulating the TLR4mediated MyD88-dependent pathway. In this study, we used LPS (a specific ligand for TLR4) to induce neuroinflammation, VIPER (a specific TLR4 inhibitor), and TLR4 −/− mice to elucidate the mechanism of action and potential targets of USP8-mediated effects on cognitive and motor impairments. Subsequently, we studied the effects of USP8 on LPS-induced animal behaviors, microglial morphology, release of inflammatory factors, protein expression, and activation of inflammatory pathways.

Animals
C57BL/6J male mice (11 to 12 weeks old, Guangdong Medical Laboratory Animal Center) and transgenic TLR4 −/− male mice (Model Animal Research Center of Nanjing University) were maintained and handled in accordance with the guidelines of the Animal Ethics Committee of Jinan University (SYXK 2017-0174). All mice were housed in a room with automatically controlled temperature (21-25°C), relative humidity (45%-65%), and light-dark (12 h: 12 h) cycle.

Neurocognitive behavior tests
To determine the therapeutic effects of USP8 in LPS-induced mice, we used the Morris water maze (MWM) test (Foster, 2012) and stepthrough passive avoidance test (PAT) (Shan et al., 2009) to assess learning and memory abilities.
The MWM was performed by Cheng Du Technology & Market Co, LTD. A circular pool (height: 35 cm, diameter: 120 cm) was filled with water made opaque with whole milk kept at 21-25°C. An escape platform (height: 14 cm, diameter: 4-5 cm) was submerged 1-1.5 cm below the surface of the water at a specific position. The mice were trained with three trials per day for 7 days. After training, LPS or saline was administered 6 h prior to the beginning of the test every day, and the spatial probe test was conducted on the last day of testing. The tasks consisted of a place navigation test and spatial probe test.
The PAT, using a "step-through" apparatus (Cheng Du Technology & Market Co, LTD.) consisting of six reaction boxes, was used to test learning and memory ability. When mice entered the dark compartment, they received an electric shock (39 V, 3-s duration). Latency to enter the dark compartment was automatically recorded. Mice were placed in the illuminated compartment facing away from the dark compartment during the training trials for the first 3 days. Subsequently, LPS was injected 6 h before each daily test during the training phase (7 days).

Motor coordination tests
Motor behavior was evaluated using the pole test and traction test (Zhu et al., 2018). For the pole test, we conducted 5 days of training on a pole (diameter: 1 cm, length: 60 cm). On the sixth day, LPS was injected 6 h before testing. The time taken for mice to climb down was recorded (finishing the upper and lower halves of the pole and total length of the pole). The following standards were used for scoring: crossing the three parts within 3 s was scored as 3, within 6 s was scored as 2, and > 6 s was scored as 1. Results were expressed as total scores.
For the traction test, mice were placed on a horizontal wire by their front paws and scored as follows: 3 when mice grasped the wire with both hind paws, 2 when mice grasped the wire with one hind paw, and 1 when mice could not grasp the wire with either hind paw, including falling down.

Nitrite and ELISA assay
NO production was determined by measuring nitrite levels in the serum. Nitrite is a stable oxidative product of NO and is assessed by the Griess reaction. For the ELISA assay, serum was collected after treatment. IL-1β and TNF-α were measured using an ELISA kit from eBioscience (Vienna, Austria), PGE 2 was measured using an ELISA kit from R&D Systems (Minneapolis, MN), and IL-4 and IL-10 were measured with an ELISA kit from USCN (Wuhan, China) according to manufacturers' instructions.

Statistical analysis
Data are presented as the mean ± SEM derived from three experiments. Comparisons between two groups were made using Student's t-test. Comparisons among multiple groups were made using one-way ANOVA followed by post hoc pairwise comparisons. The level of statistical significance was set at p < 0.05.
To determine whether USP8 is expressed in microglial cells in the brain, we detected the expression of USP8 and IBA-1 by immunofluorescence. USP8 was mainly co-localized with IBA1-positive cells (Fig. 1C), indicating that USP8 is expressed in microglia. Consistently, we observed that IBA1 expression was significantly reduced after LPS administration, and icv injection of USP8 restored microglial USP8 levels (Fig. 1B, C).

Inhibitory effects of USP8 on LPS-induced cognitive and motor dysfunction
To elucidate the effects of LPS and USP8 on hippocampus-dependent learning and memory, we conducted MWM and PAT. Control mice exhibited short escape latency to reach the platform. LPS-induced mice took a longer time to reach the platform location than did control mice (control, 9.31 ± 1.10 s; saline, 9.13 ± 1.05 s; LPS, 17.64 ± 0.92 s; P < 0.01; Fig. 2B), suggesting that LPS treatment caused memory deficits. Mice that received i.c.v injections of USP8 displayed significantly reduced escape latency after LPS treatment (11.50 ± 1.50 s; P < 0.01, Fig. 2B). In the spatial probe test, the mean incidences of crossing the removed platform and time in target section were increased in USP8-pretreated mice (P < 0.01, Fig. 2B).
In the PAT, time spent in the illuminated compartment indicates intact memory (Shan et al., 2009). USP8-pretreated mice exhibited longer latency (LPS, 14.76 ± 2.36 s; USP8 + LPS, 149.59 ± 21.77 s) and less errors (LPS, 5.62 ± 0.50 s; USP8 + LPS, 1.79 ± 0.21 s) compared to USP8-untreated mice (P < 0.01, Fig. 2C), suggesting that USP8 treatment alleviates LPS-induced memory deficits. USP8 protein level in mouse brain was significantly decreased at different time points in mice received i.p. injection of LPS. USP8 protein level was detected by western blot using anti-USP8 antibody. GAPDH was used as a loading control. Lower panel, quantification of USP8 protein levels, n = 6. (B) Intracerebroventricular injection of USP8 enhanced USP8 protein level in the brain. Lower panel, quantification of USP8 protein levels in mouse brains (C) Double immunofluorescence and confocal imaging confirmed the colocalization of USP8 (red) and IBA1 (green) in hippocampus. DAPI indicates nuclear staining in blue (Scale bar, 25 μm.). Data are presented as mean ± SEM. *P < 0.05, **P < 0.01 compared to control group and saline group; # P < 0.05, ## P < 0.01 compared to USP8 + LPS group and USP8 group, analyzed by one-way ANOVA. Error bars indicate SEM.
We used the pole and traction tests to determine whether USP8 could improve LPS-induced impairments in locomotion. As shown in Fig. 2D and E, test scores in USP8-pretreated mice were significantly higher than those in USP8-untreated mice (P < 0.01), indicating that USP8 treatment improves locomotive performance.

USP8 suppressed neuroinflammation in LPS-induced mice
Microglial activation and cytokines are essential for LPS-induced neuroinflammation. We therefore monitored the levels of select proinflammatory cytokines (TNF-α, IL-1β, PGE 2 , and NO) in serum and brain homogenates, and inflammation-related protein factors (iNOS and COX-2) in the brain. ELISA and Griess assay revealed that the levels (E) Traction test score, n = 15. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01 compared to control group and saline group; # P < 0.05, ## P < 0.01 compared to USP8 + LPS group and USP8 group, analyzed by one-way ANOVA. Error bars indicate SEM. Neurons and microglia were stained by MAP2 antibody (red) and IBA1 antibody (green) in the hippocampus after treatment. Images were acquired by double immunofluorescence with confocal microscopy. DAPI indicates nuclear staining (blue). n = 5 (Scale bar, 100 μm.). Data are presented as mean ± SEM. *P < 0.05, **P < 0.01 compared to control group and saline group; # P < 0.05, ## P < 0.01 compared to USP8 + LPS group and USP8 group, analyzed by one-way ANOVA. Error bars indicate SEM. of TNF-α, IL-1β, PGE 2 , and NO increased after LPS i.p. injection. USP8 pre-treatment significantly reduced their expression in serum and brain (P < 0.01, Fig. 3A). In addition, the levels of iNOS and COX-2 proteins in the brain of USP8-LPS-treated mice were significantly lower than those in LPS only-treated mice (P < 0.01, Fig. 3C). We subsequently used immunofluorescence to detect neuronal cells (MAP2-positive cells) and microglial cells (IBA1-positive cells) in the hippocampus. As expected, LPS induced prominent loss of cholinergic neurons (labeled by MAP2) and activation of microglia (labeled by IBA1; Fig. 3D). Treatment with USP8 significantly reduced neuronal loss and microglial activation (Fig. 3D).

USP8 increased anti-inflammatory factor levels in LPS-induced mice
We performed ELISA to detect the levels of anti-inflammatory markers (IL-4 and IL-10) in serum and brain homogenates to assess the anti-inflammatory effect of USP8 treatment. As shown in Fig. 3B, USP8 treatment significantly increased IL-4 and IL-10 levels in the serum and brain after LPS injection (P < 0.01, Fig. 3B). This result suggests that USP8 could facilitate the clearance of inflammation.

Inhibitory effects of USP8 on LPS-induced MyD88-dependent signaling pathway activation
We then studied the molecular mechanisms underlying the therapeutic effects of USP8 on neuroinflammation-associated cognitive impairment. LPS is a direct ligand of TLR4 (Laird et al., 2009). We confirmed that the expression of TLR4 was elevated in LPS-induced mice (0.68 ± 0.09) compared to control and saline groups, whereas USP8 pre-treatment significantly reduced TLR4 expression (control, 0 ± 0; saline, 0.02 ± 0.01; USP8 + LPS, 0.07 ± 0.06; P < 0.01). This result is consistent with those of other studies showing that increased inflammation in the brain is at least partly due to TLR4 activation in LPS-induced mice (Pardon, 2015). Importantly, USP8 treatment could inhibit TLR4 expression.
To determine whether USP8 could inhibit the TLR4 signaling pathway, nuclear extracts from mouse hippocampi were prepared and assayed for TLR4 signaling pathway activity. TLR4 interacts with the adapter protein MyD88 or/and TRIF to activate NF-κB, which regulates the expression of inflammatory mediators (Lin et al., 2012). In mice injected with LPS, we found no significant differences in TRIF protein expression compared with control, saline, and USP8-treated groups ( Fig. 4A and B); however, P65 protein was translocated to the nucleus, and IκBα was phosphorylated (control, 0 ± 0; saline, 0 ± 0; LPS, 0.87 ± 0.06; USP8 + LPS, 0 ± 0; P < 0.01). USP8 treatment inhibited LPS-induced translocation of P65 to the nucleus and IκBα phosphorylation (P < 0.01, Fig. 4A and B). Thus, USP8 suppresses the signaling activity of the MyD88-dependent pathway by inhibiting the expression of its signaling components.

USP8 increased Nrdp1 levels in LPS-induced mice
To provide insight into the potential molecular mechanism underlying the therapeutic effects of USP8, we examined its effect on Nrdp1 expression because it has been reported that USP8 augments Nrdp1 activity by mediating its stabilization (Wu et al., 2004), and that Nrdp1 is known to inhibit the production of proinflammatory cytokines in tolllike receptor-activated macrophages by suppressing its specific adaptor MyD88 . As shown in Fig. 4D, Nrdp1 expression in the brain of USP8-LPS-treated mice was significantly higher than that in LPS-treated mice (P < 0.01, Fig. 4D). This result suggests that USP8 increases the expression of Nrdp1 which is associated with the potent downregulation of TLR4 signaling activity in LPS-induced mice.
We used immunofluorescent staining of microglia and neurons to The inhibitory effect of USP8 on LPS-induced neuroinflammation could be mimicked by treatment of TLR4 inhibitor, VIPER. The expression of the indicated pro-inflammatory cytokines in the serum and brain were detected by ELISA and Griess assay after treatment, n = 5. (B) The expression of the indicated anti-inflammatory cytokines in the serum and brain were detected by ELISA after treatment, n = 5. (C) Percentage of neurons and microglia in the hippocampus were determined by confocal immunofluorescence microscopy by MAP2-labeling (red) and IBA1labeling (green), respectively. DAPI was used to stain nucleus (blue), n = 5 (Scale bar, 100 μm.). Data are presented as mean ± SEM. *P < 0.05, **P < 0.01 compared to control group and saline group; # P < 0.05, ## P < 0.01 compared to USP8 + LPS group, VIPER + LPS group, and USP8 + VIPER + LPS group, analyzed by one-way ANOVA. Error bars indicate SEM.
verify that VIPER inhibited microglial activation and neuronal damage. The number of IBA1-positive cells was lower and the number of MAP2positive cells higher in VIPER treated animals (Fig. 6C) than in animals not treated with VIPER. Furthermore, VIPER + LPS-treated animals exhibited significantly lower TLR4 expression in the brain than did the LPS group (Fig. 7A, B). In addition, VIPER abolished the effects of LPS on microglial M1 polarization under neuroinflammation (Fig. 7C).
These data indicated the involvement of the MyD88-dependent signaling pathway in activating inflammation, resulting in the activation of microglia and subsequent cognitive impairment. Notably, USP8 treatment inhibited TLR4-mediated neuroinflammation.

USP8 attenuated LPS-mediated cognitive deficits, motor impairment, and inflammation via TLR4
We determined whether TLR4 deficiency affects cognitive impairment after LPS-induced neuroinflammation by injecting TLR4 knockout mice (TLR4 −/− ) with LPS and evaluating memory and motor function. TLR4 −/− (10.87 ± 1.52 s) mice reached the platform faster than did WT mice (17.64 ± 0.92 s; P < 0.01, Fig. 8A), and spent more time in the platform quadrant and on the platform. In the PAT, TLR4 −/− mice spent less time in the dark compartment. The number of errors in the passive avoidance test was approximately reduced by half in TLR4 −/− mice (0.38 ± 0.18) compared to that in WT mice Fig. 7. VIPER inhibited MyD88-dependent signaling pathway activation and suppressed transition of microglial polarization from M2 to M1 phenotype under neuroinflammatory conditions. (A) Protein expression of the signaling components of MyD88-dependent pathway were analyzed using the indicated antibodies by western blot. (B) Quantification of immunoblots in A, n = 6. (C) Effect of VIPER on transition of microglial polarization from M1 to M2 phenotype was determined by confocal immunofluorescence microscopy. Microglia with M1 phenotype in the hippocampus was determined by the ratio of TNF-α (red) to IBA1 (purple) positive cells. M2 phenotype was determined by the ratio of YM-1 (green) to IBA1 (purple) positive cells. DAPI was used for nuclear staining, blue, n = 5 (Scale bar, 75 μm.). Lower panel, quantification of the percent M1 and M2 phenotypes after different treatments as described in C, n = 5 (Scale bar, 75 μm.). Data are presented as mean ± SEM. *P < 0.05, **P < 0.01 compared to control group and saline group; # P < 0.05, ## P < 0.01 compared to USP8 + LPS group, VIPER + LPS group, and USP8 + VIPER + LPS group, analyzed by one-way ANOVA. Error bars indicate SEM. Fig. 8. Knocking out TLR4 in mice protects against LPS-induced neuroinflammation and cognitive and motor impairments. (A) USP8 targeted TLR4 to ameliorate LPS-induced neuroinflammation. Morris water maze (MWM), passive avoidance test (PAT), pole test, and traction test were performed as described in the Methods section to test the memory ability and motor coordination in mice received the indicated treatments, n = 10. (B) The expression of pro-inflammatory cytokines, TNFα, IL-1β, and PGE 2 in mouse serum and brain were investigated using ELISA kits, n = 5. (C) The expression levels of the signaling components of TLR4-pathway in mouse hippocampus were investigated using western blot with the indicated antibodies, n = 6. Lower panel shows that quantification of the western blot result. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01 compared to control group; # P < 0.05, ## P < 0.01 compared to LPS (TLR4 −/− ) group, analyzed by oneway ANOVA. Error bars indicate SEM.
We then examined the effects of USP8 treatment on LPS-induced neuroinflammation and cognitive and motor impairment. USP8-treated mice exhibited recovery of cognitive and motor impairment similar to that in TLR4 −/− mice (Figs. 2, 5, 8A). To verify the relationship between TLR4 and USP8 in cognitive and motor impairment following LPS-induced neuroinflammation, we measured brain USP8 protein expression in TLR4 −/− mice. The expression of USP8 was reduced after LPS injection in WT mice; however, TLR4 knockout abolished LPS-induced USP8 repression (Fig. 8C). These results were similar to those observed (behavior, pro-inflammatory cytokines, and signaling pathway activity) in VIPER-treated mice (Figs. 5-7). Thus, USP8treated mice exhibited similar protective effect to that in TLR4 −/− mice and VIPER-treated mice.

Discussion
USP8 is a cysteine protease of the USP/UBP subfamily (Nijman et al., 2005), and is a growth-regulated enzyme essential for cell proliferation and survival. Conditional knock-out of murine USP8 promotes a dramatic loss in expression of receptor tyrosine kinases including EGFR, ErbB3, and c-Met (Niendorf et al., 2017); USP8 inactivation causes enhanced ubiquitination of ligand-activated EGFR (Alwan and Leeuwen, 2007). In conjunction with components of the ESCRT-0 complex, USP8 plays an integral role in the early endosomal sorting machinery that protects EGFR from lysosomal degradation (Berlin et al., 2010). In addition to regulating growth-related proteins, USP8 regulates the stability of its effector protein Nrdp1 under different stimuli (Wu et al., 2004). Moreover, Nrdp1 and USP8 may reciprocally regulate each other (Ceuninck et al., 2013). Nrdp1 regulates TLR signaling via inhibition of NF-κB and AP-1 and activation of TBK1 and IRF3, leading to attenuated production of pro-inflammatory cytokines ). In the present study, we showed that USP8 treatment ameliorates microglia-mediated cognitive and motor impairments following neuroinflammation. The detailed molecular mechanisms involved merit detailed study.
The hippocampus plays an important role in learning and memory consolidation (Bettio et al., 2017). LPS treatment caused learning and memory impairment as well as hippocampal microglial activation (labelled by IBA-1) and neuronal cell loss (labelled by MAP-2) in mice (Zhao et al., 2019). We observed that LPS-induced systemic inflammation caused cognitive impairment, which was ameliorated by USP8 pretreatment (Fig. 2B and C). Further, LPS-treated mice obtained lower scores on the pole and traction tests, and these effects were reduced by USP8 treatment (Fig. 2D and E). Thus, USP8 exerts neuroprotective effects that manifest as cognitive and motor improvements in mice. The neuroprotective mechanism of USP8 therefore merits further study.
Like peripheral macrophages, microglia includes heterogeneous populations of cells that display functional variability due to different polarization statuses (Kettenmann et al., 2011). "Classically activated" M1 phenotypes are characterized by the ability to release pro-inflammatory cytokines (Saijo and Glass, 2011), and "alternatively activated" M2 phenotypes are characterized by the ability to produce antiinflammatory and immunosuppressive factors (including Arg-1 and YM-1), and upregulate anti-inflammatory cytokines (Row et al., 2006).
The classical M1 state is neurotoxic and contributes to secondary neuronal damage and cell death, thereby leading to neurodegeneration (Crain et al., 2013). In this study, we showed that USP8 suppresses cognitive and motor impairments after neuroinflammation by significantly decreasing M1 phenotype-associated pro-inflammatory cytokines (TNF-α, IL-1β, PGE 2 , and NO), and increases anti-inflammatory cytokine (IL-4 and IL-10) expression ( Fig. 3A-C). Moreover, our immunofluorescence data suggested that USP8 may inhibit activated microglia and attenuate loss of neurons (Fig. 3D).
After LPS stimulation, TLR4 binds to MyD88 at the IL-1 receptor TIR cytoplasmic domain, leading to the recruitment of IL-1 receptor-associated kinase IRAK4 (Akira and Takeda, 2004). The phosphorylation of TAK-1 induces the activation of IKK, which in turn phosphorylates the IκB protein. This leads to their proteosome-mediated degradation, and the phosphorylated form of IκB dissociates from the NF-κB p50/p65 dimer, after which it enters the nucleus (Simon et al., 2015). NF-κB subsequently activates proinflammatory mediators, which activate iNOS and inducible COX-2 (Saha and Pahan, 2006). We found no significant differences in TRIF protein expression between the control, saline, and USP8-treated groups ( Fig. 4A and B). Nrdp1 is a specific target of USP8 deubiquitinating enzyme (Wu et al., 2004), and regulates TLR signaling, which leads to downregulation of pro-inflammatory cytokines ). Since we found that USP8 treatment increased the Nrdp1 levels in LPS-induced mice (Fig. 4D), we speculate that this may downregulate the expression of TLR4 and MyD88 protein, and subsequently inhibit the phosphorylation of IKKβ and IκBα, leading to reduction of nuclear translocation of p65 by inhibiting the activation of the NF-κB signaling pathway in LPS-induced mice. The detailed underlying mechanism will required further characterization and validation in future work.
Activation of M1 microglia results in increased iNOS expression. Ablation of iNOS in APP/PS1 mice can protect the mice from plaque formation and premature mortality (Tichauer and Von Bernhardi, 2012). Intracerebral injection of anti-inflammatory cytokines, such as IL-4 and IL-13, reduced Aβ plaque load in APP23 mice, which was accompanied by improved cognition and upregulation of Arg1 and YM1-positive M2 cells (Kawahara et al., 2012). Our results clearly showed that microglia in USP8-treated mice expressed the alternative phenotype. In USP8-treated mice, the number of microglia co-labeled with YM-1-positive cells increased. In contrast, in LPS-treated mice, there were almost no microglial cells co-labeled with YM-1-positive cells, and the number of microglial cells co-labeled with TNF-α-positive cells was higher (Fig. 4C). These results were consistent with our ELISA and Griess results showing pro-inflammatory (TNF-α, IL-1β, PGE 2 , and NO) and anti-inflammatory (IL-4 and IL-10) cytokine production.
We used VIPER and TLR4 −/− mice to compare the protective effects of USP8. The results from VIPER administration and the KO mouse model indicated that, similar to USP8 treatment, blocking TLR4 attenuated cognitive and motor impairments. Moreover, NF-κB activated by LPS triggered neuroinflammation in neuronal cells by activation of microglial cells via a series of inflammatory cytokines; however, this phenomenon was suppressed in VIPER-treated mice or TLR4 −/− mouse brains . These findings provide strong evidence that USP8treated mice exhibited similar protective effect to that in TLR4 −/− mice and VIPER-treated mice.
In summary, our in vivo study provided evidence that USP8 may act on the TLR4-mediated MyD88-dependent pathway to direct microglia into the M2 phenotype, ultimately reducing hippocampal inflammation and the consequent cognitive and motor impairments. USP8 could be a novel target to develop strategies to alleviate neuroinflammation-associated cognitive and motor impairments. Detailed studies on the signaling mechanisms underlying the interactions of USP8, Nrdp1, and TLR4 could lead to the identification of novel drug targets.