Wei Wang, Qiu-Yuan Gong, Lin Cai, Yao Jing, Dian-Xu Yang, Fang Yuan, Hao Chen, Heng-Li Tian

Abstract Sirtuin 2 (SIRT2) inhibition or Sirt2 knockout in animal models protects against the development of neurodegenerative diseases and cerebral ischemia. However, the role of SIRT2 in traumatic brain injury (TBI) remains unclear. In this study, we found that knockout of Sirt2 in a mouse model of TBI reduced brain edema, attenuated disruption of the blood-brain barrier, decreased expression of the nucleotide-binding oligomerization domain-like receptor protein 3 (NLRP3) inflammasome, reduced the activity of the effector caspase-1, reduced neuroinflammation and neuronal pyroptosis, and improved neurological function. Knockout of Sirt2 in a mechanical stretch injury cell model in vitro also decreased expression of the NLRP3 inflammasome and pyroptosis. Our findings suggest that knockout of Sirt2 is neuroprotective against TBI; therefore, Sirt2 could be a novel target for TBI treatment.

Key Words: blood-brain barrier; caspase-1; cerebral edema; neuroinflammation; neuroprotection; NLRP3; pyroptosis; Sirt2; tight junction protein; traumatic brain injury

Introduction 350 Methods 350 Results 352 Discussion 353

Graphical Abstract

Knockout of SIRT2 alleviates neurological dysfunction induced by TBI via NLPR3/caspase-1 pathway

Introduction

Traumatic brain injury (TBI) is a complex disease process that includes primary injury and secondary injury (Morganti-Kossmann et al., 2019). Primary injury is caused by an initial direct mechanical force; secondary injury is caused by a series of pathophysiological changes induced by the primary injury, including neuroinflammation, blood-brain barrier (BBB) disruption, oxidative stress, and excitotoxicity and etc. (Ma et al., 2017b; Sweeney et al., 2019; Jing et al., 2020). Neuroinflammation is a crucial biological process in central nervous system injury that leads to continuous neurological impairment that is characterized by the accumulation of immune cells and the abnormal production of proinflammatory cytokines (O’Brien et al., 2020; Henry and Loane, 2021; Shaheen et al., 2021; Li et al., 2022). Among the proinflammatory cytokines induced by TBI, interleukin (IL)-1β plays a crucial role in triggering the inflammatory cascade and is predominantly regulated by the nucleotidebinding oligomerization domain-like receptor protein 3 (NLRP3) inflammasome (Irrera et al., 2020).

The NLRP3 inflammasome is a multiprotein complex with three protein domains: NACHT, LRR, and PYD domains-containing protein 3 (NLRP3; sensor), apoptosis-associated speck-like protein containing a CARD (ASC; adapter), and caspase-1 (effector) (Sutterwala et al., 2006). When a cell is stimulated by damage-associated molecular pattern molecules, the NLRP3 protein oligomerizes and recruits ASC and procaspase-1, forming the NLRP3 inflammasome, which leads to the activation of procaspase-1 (Ma et al., 2017a; Kelley et al., 2019). Activated caspase-1 cleaves the precursors of IL-1β and IL-18 to form IL-1β and IL-18. IL-1β and IL-18 are the most potent cytokines that initiate inflammation; the increased inflammatory response leads to tissue damage, causing pyroptosis (Jha et al., 2010; Yang et al., 2019b; Chang et al., 2020).

Pyroptosis is an inflammatory form of programmed cell death mediated by caspase-1 activation and IL-1β and IL-18 release (Cookson and Brennan, 2001; Ding et al., 2016). A variety of cell types in multiple central nervous system diseases can die by pyroptosis, such as microglia, neurons, and astrocytes (de Rivero Vaccari et al., 2014; Gustin et al., 2015; Zhi et al., 2021). Gene knockout or inhibition of key molecules in the pyroptosis pathway, such as NLRP3, absent in melanoma 2, and caspase-1, can inhibit TBI damage and sequelae in animal models (Adamczak et al., 2014; Ge et al., 2018; Liu et al., 2018a).

Sirtuin 2 (SIRT2) is an NAD-dependent deacetylase involved in energy metabolism, oxidative stress, inflammation, cell apoptosis, and cell cycle regulation (Harting and Kn?ll, 2010). The role of SIRT2 in neurological diseases remains controversial. Some studies have shown that SIRT2 inhibition or

Sirt2

knockout has a protective effect on neurodegenerative diseases and cerebral ischemia (Xie et al., 2017; Fourcade et al., 2018; Wu et al., 2018), whereas others showed the opposite effect (Eshun-Wilson et al., 2019; Wang et al., 2019). However, few TBI studies have investigated the relationship between SIRT2 and NLRP3. Therefore, we aimed to investigate whether SIRT2 can affect the NLRP3 inflammasome, which affects neuroinflammation and pyroptosis of nerve cells in TBI, and to explore mechanisms of action.

According to 2017 statistics from the TBI Model Systems National Database (National Institute on Disability, Independent Living and Rehabilitation Research, U.S. Department of Health and Human Services), cases of TBI in men greatly outnumbered cases in women and accounted for more than 73% of all TBIs reported (Capizzi et al., 2020). Estrogen also has a neuroprotective effect on TBI (Brotfain et al., 2016). Therefore, female mice were excluded from the present study to reduce data variability.

Methods

Animals and experimental design

Adult male

Sirt2

knockout mice (

Sirt2

, RRID: IMSR_JAX:012772; 8–10 weeks, 20–25 g; Shanghai Model Organisms Center, Inc., Shanghai, China) and adult male wild type (WT) C57BL/6J mice (8–10 weeks, 20–25 g; Shanghai SLAC Laboratory Animal Corp., Shanghai, China) were used for our

in vivo

study. We also used 12 pregnant

Sirt2

mice and 12 pregnant WT C57BL/6J mice. The mice were housed under a standard 12-hour light/dark cycle at 23 ± 2°C with controlled humidity. All experimental procedures were approved by the Institutional Animal Care and Use Committee of Shanghai Sixth People’s Hospital affiliated to Shanghai Jiao Tong University, Shanghai, China (approved on March 2, 2016). All experiments were designed and reported according to the Animal Research: Reporting of

In Vivo

Experiments (ARRIVE) guidelines (Percie du Sert et al., 2020). Every necessary effort was taken to minimize the suffering and pain of the mice.The

Sirt2

and WT mice were randomly assigned to four groups: sham-WT, sham-

Sirt2

, TBI-WT, and TBI-

Sirt2

. A total of 244 mice were used in our study, and the experimental design is given in

Figure 1

. The mice were anesthetized intraperitoneally with xylazine (75 mg/kg; Sigma-Aldrich, St. Louis, MO, USA) and ketamine (10 mg/kg; Aladdin Biochemical Technology Co. Ltd., Shanghai, China) both for operations and for euthanasia in this study.

Figure 1｜Experimental design.ELSIA: Enzyme-linked immunosorbent assay; PCR: polymerase chain reaction; Sirt2: Sirtuin 2; TBI: traumatic brain injury; TUNEL: terminal deoxynucleotidyl transferase dUTP nick-end labeling; WT: wild type.

Establishment of the TBI model by controlled cortical impact

The TBI model was established by controlled cortical impact (CCI). Each anesthetized mouse was fixed in place with a stereotactic instrument (Stoelting Co., Wood Dale, IL, USA) and placed on a heat pad to maintain their body temperature at 37°C. A 1-cm incision was made in the middle of the scalp, and the incision was opened bilaterally with retractors. A bone window was made in the center of the right parietal bone using a trephine. Then, a precision cortical impactor (PCI3000, Hatteras Instruments, Inc., Cary, NC, USA) was used to establish TBI. The tip of the impactor was oriented perpendicular to the cortex. The velocity of impact was 1.5 m/s, the deformation depth was 1.5 mm, and the duration was 100 ms. The bone window was then fixed in place with bone wax, and the skin was sutured and disinfected with iodine. Each mouse recovered from anesthesia in a heated cage before being returned to its home cage. In the sham group, no CCI injury was performed after the bone window was opened.

Cell culture and mechanical stretch injury cell model

The pregnant

Sirt2

and WT C57BL/6J mice were killed by anesthesia to obtain primary cortical neurons from fetal mice (embryonic days 16–18). The skull, meninges, and superficial blood vessels were removed from each fetal mouse, and then the cortex was separated from both hemispheres and cut into pieces. The cortex was gently digested in 0.25% trypsin (Thermo Fisher Scientific, Waltham, MA, USA) at 37°C for 15 minutes, and digestion was terminated with 10% fetal bovine serum (Thermo Fisher Scientific). The cell suspension was mechanically dissociated using 70-μm nylon sieves (BD Biosciences, San Jose, CA, USA). After centrifugation, cells were resuspended in Dulbecco’s Modified Eagle Medium (Thermo Fisher Scientific) supplemented with 10% fetal bovine serum. The cells were seeded into poly-D-lysine–coated six-well BioFlex? culture plates (Flexcell International Corp., Burlington, NC, USA) at 1 × 10cells per well. Neurons were cultured in Neurobasal Medium (Thermo Fisher Scientific) with B-27 supplement (Thermo Fisher Scientific), GlutaMAX (Thermo Fisher Scientific), and penicillin and streptomycin (Thermo Fisher Scientific). Cells were cultured for 10–12 days in 5% COat 37°C under sterile conditions.

Mechanical stretch injury (SI) of primary cortical neurons was used to simulate rotational TBI and was generated using the Cell Injury Controller II system (Custom Design & Fabrication, Inc., Sandston, VA, USA). The control used was a burst of nitrogen of 50-ms duration, which caused 7.5-mm deformation of a silicone rubber membrane (Xu et al., 2017a).

Neurobehavioral tests

The modified neurological severity score was used to evaluate neurological deficits (Yang et al., 2019a) before TBI and at 1, 3, 7, and 14 days post-TBI. A score of 0 indicates normal behavior, and a score of 14 indicates maximal neurological deficit.

The rotarod test was used to evaluate the motor coordination of mice (Yang et al., 2019a). Briefly, each mouse was trained for 3 days with 3 trials per day at 5 minutes per trial. The speed of the rod was accelerated to 40 revolutions/minute. Mice that could not stay on the rod for 5 minutes were excluded from data analysis. The latency for mice that stayed on the rod was recorded before TBI and at 1, 3, 7, and 14 days post-TBI.

The Morris water maze test was used to assess spatial memory (Xu et al., 2018). A pool was divided into four equal quadrants, and a 10-cm-diameter circular platform was placed in a quadrant 1 cm below the water. On days 15–19 post-TBI, each mouse was given 60 seconds to locate the hidden platform. If the mouse failed to find the platform within 60 seconds, it was guided onto the platform and kept on it for 15 seconds. On day 20, the platform was removed and each mouse was placed into the water in the quadrant diagonal to the platform’s prior location. The number of crossings at the platform location and the time spent in the target quadrant were recorded.

Brain water content measurement

The wet–dry weight method (Yang et al., 2019a) was used to determine brain edema at 3 days post-TBI. Briefly, the mouse brain was removed immediately after euthanasia without heart perfusion, and the ipsilateral injured cerebral tissue was separated and weighed using a precise analytical balance (Mettler Toledo, Columbus, OH, USA) to measure the wet weight. Then, the ipsilateral injured cerebral tissue was dried at 100°C for 72 hours before the dry weight was measured. Brain water content (%) was calculated as follows: (wet weight – dry weight)/wet weight × 100.

Blood-brain barrier integrity measureme

ntBBB integrity was determined by Evans blue (EB) dye extravasation. At 70 hours post-TBI, the mice were anesthetized, and 2% EB (Sigma-Aldrich) was injected intravenously and kept in circulation for 2 hours. The mice were euthanized and then transcardially perfused with normal saline to remove all dye. The brain was removed from each mouse and divided into the two hemispheres. The tissues were immediately weighed, homogenized in 50% trichloroacetic acid (Sigma-Aldrich) solution, and then centrifuged at 12,000 ×

g

for 20 minutes. The supernatant was transferred into a new tube for precipitation with 3 volumes of ethanol. EB extravasation was determined as the optical density at 610 nm using a microplate reader (Agilent Technologies, Santa Clara, CA, USA).

Nissl and terminal deoxynucleotidyl transferase dUTP nick-end labeling staining

Samples were collected at 3 days post-operation for terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) staining and at 14 days post-TBI or post-sham operation for Nissl staining.

Nissl staining was used to determine the brain lesion volume. After euthanasia and transcardial perfusion with normal saline, brain tissues were fixed with 4% paraformaldehyde and sectioned at 4 μm. Sections were stained with Nissl stain, and the lesion volume was calculated using ImageJ (National Institutes of Health, Bethesda, MD, USA) (Schneider et al., 2012).

The TUNEL assay was used to detect pyroptosis. TUNEL staining was performed using the

In Situ

Cell Death Detection Kit (Roche, Basel, Switzerland) in accordance with the manufacturer’s instructions. A fluorescence microscope (Leica, Wetzlar, Germany) was used to view and record images. For each mouse, three equally spaced coronal sections were selected from the injury site, and the number of pyroptotic cells was counted in each section in five random fields of view.

Immunofluorescence staining

Glial fibrillary acidic protein (GFAP), neuronal nuclei antigen (NeuN), and ionized calcium-binding adapter molecule 1 (Iba-1) were used to identify astrocytes, neurons, and microglia at 3 days post-TBI, respectively. The brain samples were quick-frozen in –80°C isopentane and cut into coronal sections of 30 μm using a freezing microtome (Leica). The cryosections and primary cortical neuron cells were fixed with 4% paraformaldehyde (Beyotime Biotechnology, Shanghai, China). Subsequently, the samples were permeabilized with Triton X-100 (Sigma-Aldrich), blocked with 10% bovine serum albumin (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) for 1.5 hours, and incubated with the primary antibodies at 4°C overnight as follows: goat anti-CD31 (1:200, R&D Systems, Minneapolis, MN, USA, Cat# AF3628, RRID: AB_2161028), rabbit anti-zonula occludens 1 (ZO-1; 1:200, Thermo Fisher Scientific, Cat# 61-7300, RRID: AB_2533938), mouse anti-NeuN (1:200, MilliporeSigma, Burlington, MA, USA, Cat# MAB377, RRID: AB_2298772), mouse anti-GFAP (1:200, MilliporeSigma, Cat# MAB3402, RRID: AB_94844), mouse anti-Iba-1 (1:200, Abcam, Cambridge, UK, Cat# ab283319), or rabbit anti-NLRP3 (1:200, Abcam, Cat# ab214185, RRID: AB_2819003). Then the samples were incubated with the corresponding secondary antibodies for 1 hour at 37°C as follows: donkey anti-goat IgG-Alexa Fluor 555 (1:500, Thermo Fisher Scientific, Cat# A-21432, RRID: AB_2535853), donkey anti-rabbit IgG-Alexa Fluor 594 (1:500, Thermo Fisher Scientific, Cat# A-21207, RRID: AB_141637), or donkey anti-mouse IgG-Alexa Fluor 488 (1:500, Thermo Fisher Scientific, Cat# A-21202, RRID: AB_141607). The samples were incubated with 4′,6-diamidino-2-phenylindole dihydrochloride (1:2000, Thermo Fisher Scientific) for 5 minutes at room temperature in the dark. Images were captured with a fluorescence microscope (Leica). Fluorescence intensity was quantified by LAS AF 2.8.0 software (Leica).

Western blot assay

Proteins from brain tissue at 3 days post-TBI and primary cortical neurons at 24 hours post-SI were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes. The membranes were blocked with 5% nonfat milk for 1 hour and incubated in primary antibody overnight at 4°C and in the corresponding secondary antibody for 2 hours at room temperature. The antibodies were as follows: rabbit anti-ZO-1 (1:1000, Thermo Fisher Scientific, Cat# 61-7300, RRID: AB_2533938), rabbit anti-NLRP3 (1:500, Abcam, Cat# ab214185, RRID: AB_2819003), goat anti-ASC (1:500, Santa Cruz Biotechnology, Dallas, TX, USA, Cat# sc-22514-R, RRID: AB_2174874), rabbit anti-caspase-1 p45, p10 (1:1000, Cell Signaling Technology, Danvers, MA, USA, Cat# 24232, RRID: AB_2890194), rabbit-anti-caspase-1 p20 (1:1000, www.antibodiesonline.com, Cat# ABIN5675787), mouse anti-β-actin (1:1000, Cell Signaling Technology, Cat# 3700, RRID: AB_2242334), mouse anti-β-tubulin (1:1000, Cell Signaling Technology, Cat# 86298, RRID: AB_2715541), anti-mouse IgGhorseradish peroxidase (1:5000, Cell Signaling Technology, Cat# 7076, RRID: AB_330924), or anti-rabbit IgG-horseradish peroxidase (1:5000, Cell Signaling Technology, Cat# 7074, RRID: AB_2099233). Protein was visualized using an electrochemiluminescence reagent (Cat# WBKLS0100, MilliporeSigma) and a chemiluminescence instrument (Tanon, Shanghai, China). The relative protein levels were expressed by optical density ratio to β-actin.

Real-time reverse transcription-polymerase chain reaction

Total RNA was isolated from primary cultured neurons at 24 hours post-SI or brain tissue at 3 days post-TBI using TRIzol reagent (Thermo Fisher Scientific), and the RNA was reverse transcribed into a complementary DNA template using a PrimeScriptReverse Transcription Reagent Kit (Takara Bio, Kyoto, Japan). Real-time reverse transcription-polymerase chain reaction (PCR) was performed using a SYBR Premix Ex Taq Kit (Takara Bio) on an ABI 7900HT PCR instrument (Thermo Fisher Scientific). The following primers were used: NLRP-3 (forward: 5′-ATC AAC AGG CGA GAC CTC TG-3′ and reverse: 5′-GTC CTC CTG CAT ACC ATA GA-3′), ASC (forward: 5′-GAC AGT GCA ACT GGC GAG AAG-3′ and reverse: 5′-CGA CTC CAG ATA GTA GGC TGA C-3′), caspase-1 (forward: 5′-ACA AGG CAC GGA CCT ATG-3′ and reverse: 5′-TCC CAG TCA GTC CTG GAA ATG-3′), IL-1β (forward: 5′-GCA ACT GTT CCT GAA CTC AAC T-3′ and reverse: 5′-ATC TTT GGG GTC CGT CAA CT-3′), IL-18 (forward: 5′-GAC TCT TGC GTC AAC TTC AAG G-3′ and reverse: 5′-CAG GCT GTC TTT TGT CAA CGA-3′), and glyceraldehyde 3-phosphate dehydrogenase (forward: 5′-AGG TCG GTG TGA ACG GAT TTG-3′ and reverse: 5′-TGT AGA CCA TTA GTT GAG TCA-3′). Reverse transcription conditions were as follows: predenaturation at 95°C for 30 seconds, followed by denaturation at 95°C for 5 seconds and annealing and extension at 60°C for 30 seconds, for a total of 40 cycles. We used the comparative Ct (threshold cycle) method (Zhu et al., 2019) normalized to glyceraldehyde 3-phosphate dehydrogenase to analyze relative changes in gene expression.

Lactate dehydrogenase assay

The lactate dehydrogenase (LDH) detection kit (Roche) was used to evaluate cell membrane integrity. The primary neurons at 24 hours post-SI were cultured in six-well BioFlex? culture plates (Flexcell International Corp.). The amount of LDH released into the culture medium is represented as LDH activity. The optical density was read at 490 nm with a spectrophotometer (BioTek).

Caspase-1 activity measurement

A caspase-1 activity assay kit (Beyotime Biotechnology) was used to measure caspase-1 activity at 3 days post-TBI in accordance with the manufacturer’s instructions (Liu et al., 2018a). Briefly, brain tissues were lysed, and the supernatant was collected. A standard curve was prepared using a p-nitroaniline standard. Then, the optical density of the tissue lysates was read at 405 nm on a microplate reader (BioTek).

Enzyme-linked immunosorbent assay

An enzyme-linked immunosorbent assay (ELISA) kit (Beijing Solarbio Science & Technology Co., Ltd.) was used to measure the level of IL-1β in accordance with the manufacturer’s instructions. Mouse blood samples were collected using retro-orbital bleeding at 3 days post-TBI, centrifuged, and the supernatant was collected. Then the samples were added for 90 minutes to an ELISA plate coated with IL-1β antibody. After washing, the plate was sequentially treated with biotinylated IL-1β antibody, horseradish peroxidaselabeled streptavidin, and 3,3′,5,5′-tetramethylbenzidine. The optical density of the samples was read on a microplate reader (BioTek) at 450 nm.

Statistical analysis

No statistical methods were used to predetermine sample sizes; however, our sample sizes are similar to those reported in previous publications (Xu et al., 2017a, b). No animals or data points were excluded from the analysis. The evaluators were blinded to the assignment. All data were indicated as mean ± standard deviation. The comparisons between two groups were performed by Student’s

t

-test.

P

< 0.05 was considered statistically significant. SPSS (Version 21.0, IBM, Armonk, NY, USA) and GraphPad Prism 5.01 for Windows (GraphPad Software, San Diego, CA, USA, www.graphpad.com) were used for statistical analysis and visualization, respectively.

Results

The NLRP3 inflammasome is upregulated in the pericontusional area post-TBI

Western blot and PCR were performed to examine the NLRP3, ASC, and caspase-1 protein and mRNA levels, respectively, in pericontusional regions of WT mice at different time points post-TBI. The protein and mRNA expression levels of NLRP3, ASC, and caspase-1 all increased post-TBI; this increase peaked on day 3 and lasted at least 7 days (

Figure 2A–G

). Therefore, we selected WT mouse brain slices 3 days post-TBI to detect NLRP3 expression in neurons, astrocytes, and microglia using the immunofluorescence method. The results indicated that NLRP3 was expressed in neurons and microglia post-TBI (

Figure 2H

).

Knockout of

Sirt2

improves neurological function post-TBI in mice

There was no neurological deficit in WT or

Sirt2

mice after the sham operation. After TBI, the modified neurological severity scores increased significantly on day 1 in both groups of mice; the scores significantly reduced on days 3, 7, and 14 in

Sirt2

mice compared with those in WT mice (

P

< 0.05;

Figure 3A

). The

Sirt2

mice had longer latency than WT mice on the rotarod test on days 7 and 14 post-TBI (

P

< 0.05;

Figure 3B

). In the Morris water maze test, the

Sirt2

mice had a greater number of crossings at the platform location and stayed longer in the target quadrant on day 20 post-TBI compared with WT mice (

P

< 0.01;

Figure 3C–E

).

Knockout of

Sirt2

alleviates blood-brain barrier disruption and brain edema in mice after traumatic brain injury

TBI leads to BBB disruption. ZO-1 is a fundamental constituent of the BBB and plays a crucial role in maintaining BBB integrity (Cash and Theus, 2020). Immunostaining of brain tissues at 3 days post-TBI for ZO-1 and the vascular marker CD31 showed continuous costaining of ZO-1 and CD31 in the sham group, consistent with previous results (Xu et al., 2017b). ZO-1 gaps were present post-TBI, with fewer gaps in the TBI-

Sirt2

group than in the TBI-WT group (

Figure 4A

). Meanwhile, western blot analysis revealed a much higher expression level of ZO-1 in the TBI-

Sirt2

group than in the TBI-WT group at 3 days post-TBI (

P

< 0.05;

Figure 4B

and

C

).Nissl staining was used to calculate the volume of damaged brain tissue post-TBI in mice. The TBI-

Sirt2

group had significantly reduced volume of damaged tissue post-TBI compared with the TBI-WT group (

P

< 0.05;

Figure 4D

and

E

).EB extravasation and brain water content were used to estimate BBB integrity post-TBI.

Sirt2

mice had significantly decreased EB leakage in the ipsilateral cortex 3 days post-TBI compared with the WT mice post-TBI, indicating that BBB disruption induced by TBI was attenuated by knockout of

Sirt2

(

P

< 0.01;

Figure 4F

and

G

). The dry-wet weight results showed that the water content of the injured hemisphere increased post-TBI, and

Sirt2

mice had significantly reduced brain water content compared with the WT mice post-TBI (

P

< 0.01;

Figure 4H

).

Knockout of reduces the levels of NLRP3 inflammasome and nerve cell pyroptosis after traumatic brain injury

Western blot and PCR were used to detect the protein and mRNA expression levels, respectively, of each component in the NLRP3 inflammasome on day 3 post-TBI. The protein expression levels of NLRP3, ASC, and caspase-1 were increased in WT mice post-TBI compared with the sham-treated mice. The protein expression levels of NLRP3, ASC, and caspase-1 were significantly decreased in the

Sirt2

mice at 3 days post-TBI compared with the WT mice (all

P

< 0.05;

Figure 5A–F

). The protein expression levels were consistent with the mRNA levels (

Figure 5G–I

).Neuroinflammation is mediated by inflammatory cells and the cytokines secreted by the inflammatory cells (O’Brien et al., 2020). We used PCR to detect the mRNA expression levels of IL-1β and IL-18 in the brain tissue around the lesion 3 days post-TBI, and we used ELISA to detect the protein expression levels of IL-1β in blood 3 days post-TBI. The PCR results showed that

Sirt2

mice had significantly reduced IL-1β and IL-18 mRNA levels post-TBI compared with WT mice (

P

< 0.01;

Figure 5J

). ELISA results also showed that

Sirt2

mice had significantly reduced secretion of IL-1β in the blood compared with WT mice (

P

< 0.01;

Figure 5K

).Caspase-1 is an effector of the NLRP3 inflammasome, and caspase-1 activity determines the inflammatory response mediated by the NLRP3 inflammasome (Miao et al., 2011). We found that caspase-1 activity was significantly elevated in the brain tissues post-TBI, and its activity was increased less in

Sirt2

mice than in WT mice on day 3 post-TBI (

P

< 0.01;

Figure 5L

).We used TUNEL staining to detect DNA damage in nerve cells of brain tissue at 3 days post-TBI. The results showed very few TUNEL-positive cells in the sham group, but many TUNEL-positive cells were present 3 days post-TBI. Furthermore,

Sirt2

mice had significantly reduced numbers of TUNELpositive cells compared with WT mice post-TBI (

P

< 0.01;

Figure 5M

and

N

).

knockout reduces NLRP3 inflammasome expression and primary

neuron pyroptosis after mechanical stretch injury

We used primary cultured neurons to investigate the underlying mechanism of

Sirt2

knockout post-SI

in vitro

. Western blot analysis showed that the protein expression levels of NLRP3 (

P

< 0.05), ASC (

P

< 0.01), and caspase-1 (

P

< 0.01) were significantly decreased in the SI-

Sirt2

group compared with those in the SI-WT group (

Figure 6A–F

), which were similar to our

in vivo

results.LDH release tests are used to assess cell membrane integrity. Cytoplasmic LDH is released into the culture medium when the cell membrane integrity is compromised (Jing et al., 2020). Compared with the control group without intervention, LDH dramatically increased after SI, while knockout of

Sirt2

reduced the release of LDH from primary neurons induced by SI (

P

< 0.01;

Figure 6G

). The numbers of TUNEL-positive primary neuron cells were significantly reduced in the SI-

Sirt

2group compared with those in the SI-WT group (

P

< 0.01;

Figure 6H

and

I

).

Figure 2｜The NLRP3 inflammasome is upregulated in the pericontusional area of wild type mice after traumatic brain injury.(A) Western blot bands of the NLPR3 inflammasome in the pericontusional area. (B–D) Quantification of NLRP3, ASC, and caspase-1 protein expression, normalized to the sham group. (E–G) Quantification of NLRP3, ASC, and caspase-1 relative mRNA levels by PCR. Data are expressed as mean ± SD (n = 6 per group). **P < 0.01, vs. sham group (Student’s t-test). (H) Immunofluorescence staining of NLRP3 expression (red, stained with Alexa Fluor 594) in neurons (NeuN-positive cells, green, stained with Alexa Fluor 488), astrocytes (GFAP-positive cells, green, stained with Alexa Fluor 488), and microglia (Iba-1-positive cells, green, stained with Alexa Fluor 488) at 3 days post-TBI. Scale bar: 50 μm. ASC: Apoptosis-associated speck-like protein containing CARD; DAPI: 4′,6-diamidino-2′-phenylindole, dihydrochloride; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; GFAP: glial fibrillary acidic protein; Iba-1: ionized calcium binding adaptor molecule-1; NeuN: neuronal nuclei; NLRP3: nucleotide binding oligomerization domain-like receptor protein 3; PCR: polymerase chain reaction; TBI: traumatic brain injury.

Figure 3｜Knockout of Sirt2 improves neurological deficits after traumatic brain injury in mice.(A) mNSS. (B) Latency in the rotarod test. (C) Representative images of the exploration in the Morris water maze test. The small orange open circle represents the location of the platform. (D) Quantification of the number of crossings at the platform location in the Morris water maze test 20 days post-TBI. (E) Quantification of the time spent in the target quadrant in the Morris water maze test 20 days post-TBI. Data are expressed as mean ± SD (n = 10 per group). *P < 0.05, **P < 0.01, vs. TBI-Sirt2?/? group (Student’s t-test). mNSS: Modified neurological severity score; Sirt2: Sirtuin 2; TBI: traumatic brain injury; WT: wild type.

Figure 4｜Knockout of Sirt2 alleviates blood-brain barrier disruption and brain edema in mice after traumatic brain injury.(A) Representative costained immunofluorescence images of ZO-1 (green, stained with Alexa Fluor 488) and CD31 (red, stained with Alexa Fluor 555) at 3 days post-TBI. ZO-1 gaps are indicated by arrows. Scale bar: 10 μm. (B) Representative western blot bands of ZO-1 at 3 days post-TBI. (C) Quantification of ZO-1 protein expression normalized to β-tubulin. (D) Representative images of Nissl staining of brain tissue of mice 14 days post-TBI. Scale bar: 3 mm. (E) Quantification of brain lesion volume. (F) Representative images of the amount of Evans blue dye exudation at 3 days post-TBI. (G) Quantification of the amount of Evans blue dye exudation at 3 days post-TBI. (H) Quantification of brain water content. Data are expressed as mean ± SD (n = 6 per group). *P < 0.05, **P < 0.01 (Student’s t-test). Sirt2: Sirtuin 2; TBI: traumatic brain injury; WT: wild type; ZO-1: zonula occludens-1.

Figure 5｜Knockout of Sirt2 reduces the expression of the NLRP3 inflammasome and nerve cell pyroptosis after traumatic brain injury in vivo.(A) Representative western blot bands of the NLPR3 inflammasome components at 3 days post-TBI. (B–F) Quantification of protein expression of NLRP3, ASC, caspase-1 p45, caspase-1 p20, and caspase-1 p10 at 3 days post-TBI. The target protein expression was normalized to the sham-WT group. (G–J) Quantification of mRNA expression of NLRP3, ASC, caspase-1, IL-1β, and IL-18 at 3 days post-TBI by PCR. The target mRNA expression was normalized to the sham-WT group. (K) IL-1β secretion in blood by ELISA. (L) Caspase-1 activity at 3 days post-TBI. (M, N) Representative images and quantification of TUNEL-positive cells (red) at 3 days post-TBI. Double-positive TUNEL (red) and DAPI staining (blue) indicates the pyroptotic cells. Scale bar: 50 μm. Data are expressed as mean ± SD (n =6 per group). *P < 0.05, **P < 0.01 (Student’s t-test). ASC: Apoptosis-associated speck-like protein containing CARD; DAPI: 4′,6-diamidino-2-phenylindole; ELISA: enzyme-linked immunosorbent assay; IL: interleukin; NLRP3: nucleotide binding oligomerization domain-like receptor protein 3; PCR: polymerase chain reaction; Sirt2: Sirtuin 2; TBI: traumatic brain injury; TUNEL: terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling; WT: wild type.

Figure 6｜Knockout of Sirt2 reduces NLRP3 inflammasome expression and primary neuron pyroptosis after mechanical stretch injury.(A) Representative western blot bands of the NLPR3 inflammasome components at 24 hours post-SI. (B–F) Quantification of protein expression of NLRP3, ASC, caspase-1 p45, caspase-1 p20, and caspase-1 p10 at 24 hours post-SI. The target protein expression was normalized to the sham-WT group. (G) Release of LDH from primary neurons induced by SI at 24 hours. (H, I) Representative images and quantification of TUNEL-positive cells (red, arrow) at 24 hours post-SI. Scale bar: 30 μm. Data are expressed as mean ± SD. Each experiment was repeated three times. *P < 0.05, **P < 0.01 (Student’s t-test). ASC: apoptosis-associated speck-like protein containing CARD; LDH: lactate dehydrogenase; NLRP3: nucleotide binding oligomerization domain-like receptor protein 3; SI: stretch injury; Sirt2: Sirtuin 2; TBI: traumatic brain injury; TUNEL: terminal deoxynucleotidyl transferase dUTP nick-end labeling; WT: wild type.

Discussion

This study demonstrates that

Sirt2

knockout reduces the expression of the components of the NLRP3 inflammasome and the activity of caspase-1, which alleviates neuroinflammation and reduces pyroptosis, leading to reversal of neurological dysfunction and reductions in cerebral edema and BBB disruption.Secondary injury post-TBI is a gradual process that causes chemical and metabolic changes in nerve cells and leads to a series of pathological changes, such as apoptosis, neuroinflammation, and BBB disruption (Kulbe and Hall, 2017). The processes behind these pathological changes may be vital targets for TBI therapy (Jha et al., 2019). The BBB maintains brain homeostasis by transporting nutrients, removing toxic substances, and inhibiting the entry of peripheral immune cells. The BBB is composed of specialized endothelial cells closely joined by tight junction proteins. The loss of tight junction proteins, such as ZO-1 and occludin, allows typically restricted molecules, such as peripheral immune cells, to pass through the BBB, leading to vasogenic brain edema (Wang et al., 2018). We found that

Sirt2

knockout reduced EB extravasation and brain edema, indicating that

Sirt2

knockout partially reversed the BBB disruption induced by TBI. Furthermore, the results of ZO-1 immunofluorescence staining and western blot analysis showed that

Sirt2

knockout reduced the loss of the tight junction protein ZO-1. Therefore, we hypothesize that SIRT2 improves the integrity of the BBB by reducing the loss of tight junction proteins, thereby alleviating cerebral edema.

Widely used TBI animal models can be generated by fluid percussion injury, CCI injury, weight drop impact injury, blast injury, or penetrating ballisticlike injury (Xiong et al., 2013). We chose to use the CCI model, as it is easy to control the mechanical factors, such as time, speed, and impact depth (Saatman et al., 2006), and it has a wide range of simulation scenarios. We used the parameters of the moderate-TBI CCI model (Liu et al., 2018b).

When pyroptosis occurs, the cells swell, pores are formed in the cell membrane, and the integrity of the cell membrane is lost. The resultant release of IL-1β and IL-18 causes an inflammatory response (Cookson and Brennan, 2001; Ding et al., 2016). Pyroptosis is driven by canonical and noncanonical inflammasomes. In the canonical pathway, the caspase-1 precursor combines with NLRP3 protein and adapter protein ASC to form the NLRP3 inflammasome, which activates caspase-1 and cleaves and activates IL-1β and IL-18 precursors; IL-1β and IL-18 then cleave gasdermin D, which causes formation of holes in the cell membrane, causing cell pyroptosis (Kovacs and Miao, 2017; Sun et al., 2020).

The role of the NLRP3 inflammasome post-TBI has been confirmed in experimental TBI models and patients with moderate or severe TBI. Liu et al. (2013) reported that NLRP3, ASC, and caspase-1 mRNA increased 6 hours after fluid percussion injury in rats, and the protein levels of NLRP3 and caspase-1 remained increased 24 hours after injury. Chen et al. (2019) showed that NLRP3 mRNA began to increase within 6 hours after TBI and exhibited peaks at 24 and 72 hours post-TBI in mice; on day 7, the expression of NLRP3 decreased to a level that remained higher than the control group. TBI can activate the NLRP3 inflammasome, so we hypothesized that targeting this pathway may effectively alleviate neuroinflammation and promote the recovery from TBI. In animal models of other neuroinflammatory diseases, such as Alzheimer’s disease, NLRP3 knockout has been shown to reduce neuroinflammation and improve functional prognosis (Feng et al., 2020). MCC950 is a highly selective and potent NLRP3 inhibitor. A study found that intraperitoneal injection of MCC950 in mice reduced the expression of NLRP3, caspase-1, ASC, and IL-1β 24 hours after CCI and decreased IL-1β levels 72 hours post-TBI (Ismael et al., 2018). In our study, expression of the components of the NLRP3 inflammasome increased post-TBI, peaked on day 3, and lasted at least 7 days; these components were expressed in neurons and microglia 3 days post-TBI.

Caspase-1 activity determines the inflammatory response mediated by the NLRP3 inflammasome. Caspase-1 knockout mice are partially resistant to stroke, and intracerebroventricular administration of caspase-1 inhibitors provides protective effects in experimental stroke models (Ross et al., 2007). A study showed that caspase-1 and ASC levels were elevated in the serum of patients post-TBI and were related to poor prognosis post-TBI (Kerr et al., 2018). Sun et al. (2020) demonstrated that VX765, a caspase-1 inhibitor, inhibits IL-1β and IL-18 production, which decreases pyroptosis in the TBI acute phase, decreases LDH release, reduces neuroinflammation, and inhibits microglial death. Thus, the NLRP3 inflammasome plays a key role in regulating neuroinflammation and pyroptosis. Activation of the NLRP3 inflammasome is a two-step process of priming and activation (Swanson et al., 2019). Priming upregulates the expression of the NLRP3 inflammasome and is induced by various signals, such as damage-associated molecular patterns, Toll-like receptors, and tumor necrosis factor, that result in nuclear factor-κB activation (Bauernfeind et al., 2009; Xing et al., 2017). In our study,

Sirt2

mice post-TBI had decreased expression of the NLRP3 inflammasome components and decreased caspase-1 activity. We suggest that knockout of

Sirt2

may alleviate neuroinflammation and reduce pyroptosis via the NLPR3/caspase-1 pathway.The role of SIRT2 in neurological diseases is still controversial. The Parkinson’s disease (PD) model generated by long-term administration of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine can replicate most of the clinical features of PD and produce reliable and reproducible PD. The SIRT2 inhibitor AK7 can prevent dopamine depletion and dopaminergic neuron loss caused by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine

in vivo

(Chopra et al., 2012).

Sirt2

mice had reduced neurodegeneration caused by long-term administration of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, which was caused by a reduction in apoptosis by an increase in FOXO3a acetylation and a reduction in Bcl-2-like protein 11 levels (Liu et al., 2014). In contrast, in the PD model induced by 6-hydroxydopamine treatment, the decrease in SIRT2 activity lead to an increase in acetylated α-tubulin, and the function of tubulin deacetylase was improved (Patel and Chu, 2014). AK7 treatment can improve motor function, prolong survival, reduce brain atrophy, and significantly reduce the accumulation of mutant huntingtin protein and improve Huntington’s disease symptoms (Chopra et al., 2012); however, another study reached the opposite conclusion (Capizzi et al., 2020).

Sirt2

knockout or use of the SIRT2 inhibitor AGK2 showed neuroprotection against cerebral ischemia (Wang et al., 2016). Another study showed that SIRT2 determines the level of α-tubulin acetylation, which regulates microtubule assembly. Acetylated α-tubulins are necessary for the activation of the NLRP3 inflammasome (Misawa et al., 2013). Thus, the relationship between SIRT2 and NLRP3 in TBI needs to be further investigated.There are several limitations to this study. First, the role of SIRT2 in microglial cells was not investigated. Second, we did not use overexpression technology or rescue experiments to validate our hypothesis. Thus, further research needs to be performed. In future studies, we plan to investigate gender differences in the response to

Sirt2

knockout post-TBI.In conclusion,

Sirt2

knockout can improve the neurological dysfunction of mice post-TBI, improve memory, reduce the loss of tight junction proteins, reduce the disruption of the BBB, and reduce cerebral edema.

Sirt2

knockout may reduce neuroinflammation and pyroptosis by reducing the expression of the NLRP3 inflammasome and the activity of caspase-1.

Author contributions:

Study design: WW, HLT; manuscript writing: WW. All authors participated in experiment impementation, analyzed data, and approved the final version of manuscript for publication.

Conflicts of interest:

No competing financial interests exist.

Availability of data and materials:

All data generated or analyzed during this study are included in this published article and its supplementary information files.

Open access statement:

This is an open access journal, and articles are distributed under the terms of the Creative Commons AttributionNonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.

Open peer reviewers:

Chao Weng, Renmin Hospital of Wuhan University, China; Fei Ding, Affiliated Hospital of Nantong University, China; Rosario Donato, Interuniversity Institute of Myology, Italy.

Additional file:

Open peer review reports 1–3.