MitoQ Promotes M2-Like Macrophage/Microglia Polarization And Attenuates Apoptosis After Spinal Cord Injury Through Mitochondria-Dependent Pathway

Background: Mitochondrial dysfunction and macrophage/microglia mediating inammation are critical pathological process components of secondary injury in spinal cord injury (SCI), which lead to impaired regeneration and function recovery. Mitochondria is essential to macrophage/microglia metabolism and polarization. However, there are few research in SCI on mitochondrial dysfunction and macrophage/microglia polarization modulation by MitoQ, a mitochondrial-specic antioxidant. Therefore, we tried to investigate whether MitoQ could promote function recovery after SCI and its potential mechanism. Methods: A T10 spinal cord clip compression model was established in female C57BL/6 mice. MitoQ (1 mg/ml) was injected intraperitoneally on day 0, 1, 2 after SCI. Immunouorescence assay (iNOS/Arg-1) was performed on day 3 and 14 to evaluate macrophage/microglia polarization after SCI. Basso Mouse Scale (BMS) score and footprint analysis were used to evaluate function recovery. Haematoxylin-eosin (HE) and Luxol Fast Blue (LFB) staining were used to evaluate tissue preservation and myeline loss. Western-blotting (mfn-1/2/drp-1), ow cytometry, immunouorescence assay (iNOS/Arg-1) and uorescence assay (Mitosox Red/ DCFH-DA) were used to evaluate in-vitro mitochondrial function and macrophage/microglia polarization. Tunel staining, Hochest staining and Western-blotting (Bax/Bcl-2/Cleaved-caspase-3) were used to evaluate cell apoptosis in BV2 cells and established SCI animal models. Results: Compared with SCI group, MitoQ group showed reduced M1-like polarization with enhanced M2-like polarization on day 3 and day 14 after SCI. MitoQ group promoted function recovery and tissue preservation on day 28. MitoQ attenuated mitochondrial biodynamic imbalance and reduced mitochondrial-specic reactive oxygen species (ROS) production in BV2 cells. MitoQ inhibited cell apoptosis resulting from in-vitro LPS challenging and in-vivo SCI. Conclusion: MitoQ improves function recovery and tissue preservation after SCI through the promotion of M2-like polarization of macrophage/microglia, inhibition of cell apoptosis and following neuroinammation with the improvement of mitochondrial function and decreasing oxidative stress. Taken together, our results suggest that mitochondrial function modulation is a potential treatment option of post-SCI macrophage/microglia polarization dysfunction, following neuroinammation and function impairment.

mediated by macrophage/microglia, is one of the major causes of unsatis ed neuro-regeneration after SCI resulting from continuous pro-in ammatory factors releasing and microenvironment disruption.
Microglia, commonly considered as resident macrophage in central nervous system (CNS) 2 , is vital immune cell that mediates cell proliferation, scar formation and myeline remodeling following SCI 3 .
Microglia is highly exible as macrophage and it can be activated and shifted into different phenotypes, which can be generally divided into M1-like and M2-like subtypes in response to various stimuli in different pathological periods. M1-like microglia, commonly stimulated by TNF-α and IFN-γ, is de ned to be neurotoxic mainly through secretion of pro-in ammatory cytokines: TNF-α, IL-6 and IL-1β, disturbance of microenvironment and contribution to scar formation 4 . While M2-like microglia is considered as neuroprotector for its essential role in tissue repair and anti-in ammatory cytokines secretions 5 .
Therefore, macrophage/microglia polarization, the key contributor of post-SCI microenvironment formation, has been proved to be vital to SCI treatment 6,7 .Given that, research of macrophage/microglia phenotypes modulation have become a hot point recently [8][9][10] .
Macrophage/microglia metabolism including glucose, fatty acid and amino acid has been found to be essential to macrophage activation and polarization 11 . Given the importance of metabolism in macrophage regulation, mitochondria, the major energy factory of cells, shows enormous potential in macrophage regulation. SCI-leading mitochondrial dysfunction is a common pathological process in secondary injury which resulting in oxidative stress and cell apoptosis/death. Recently, mitochondrial function has been found to be essential to macrophage polarization 11 . However, there are few research have been done to investigate the relationship between mitochondrial dysfunction and macrophage/microglia polarization dysfunction in SCI. Mitoquinone (MitoQ) is a mitochondrial-targeted antioxidant protecting treated mitochondria from oxidative stress/reactive oxygen species'(ROS) damage to mitochondrial proteins and DNA 12 . MitoQ plays a neuroprotective role in neurologic disease 13 . It's bene cial to the recovery of intracerebral hemorrhage 14 and neurological pain 15 . The potential role of MitoQ in SCI neuroprotection and underlying mechanism requires further research to discover. Therefore, in this study we try to investigate MitoQ's role in modi cation of neuroin ammation and macrophage/microglia polarization phenotypes and whether it can be bene cial to post-SCI function recovery and tissue preservation.

Methods
Cell culture BV2 murine cells used in this experiment were purchased from Shanghai Cell Research Center (Shanghai, China). Cells were maintained in DMEM high glucose medium containing 10% fetal bovine serum (Gibico, USA) and 1% penicillin/streptomycin (Invitrogen, USA). Humidi ed incubator containing 5% CO2 at 37 ℃ was used to keep cells. After reaching about 80% in culturing plates, cells were considered to be suitable for following experiments.

Cell treatment
Prior to Lipopolysaccharide (LPS) challenging, cells in MitoQ group were pretreated with MitoQ (200 nM, MCE, USA) for 12 h. Then cells in LPS group were treated with 100 ng/ml LPS (Aladdin, China) for 24 and 48 h, and cells in MitoQ group were treated with the mixture of LPS (100 ng/ml) and MitoQ (200 nM).

Animals
Female C57BL/6 mice aged from 10 to 14 weeks old were purchased from Laboratory Animal Center of Shanghai Jiao Tong University. Animals were housed in facility which controlled its humidity at 55-65% and temperature at 24 ± 3 ℃ with 12hr period of light/dark cycle. Food and water were provided and animals had free access to the supply. All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Shanghai Jiao Tong University a liated Sixth People's Hospital, China. All study methods were in accordance with China's regulations on experimental animal usage, which were consistent with Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines.

SCI model establishment
The establishment of SCI clip-compression model was based on the research reported previously 16 . After animals were anesthetizing with intraperitoneally injection of ketamine (50 mg/kg) and xylazine (3 mg/kg), skin preparation was performed prior to surgery. A 1 cm incision was made after identifying thoracic spine at level 10, then surrounding skin and tissue were manually removed using ophthalmic scissors. After that, animals received a T10 laminectomy to expose the spinal cord. Animals in sham group only received the laminectomy. Next, a 30-g aneurysm clip was applied to the T10 level for 1 min to induce compression injury. Then the surgical site was closed by muscle, fascia and skin suturing. In MitoQ groups, MitoQ (1 mg/ml) was injected intraperitoneally immediately after SCI and on day 1, 2, respectively. After fully recovery on the heating pad (37 ℃), animals were housed and received manual bladder expression twice per day until a primary bladder re ex was re-established. Antibiotics and analgesic were used as routine on day 0, 1, 2, 3 to prevent infection and excessive pain.

Mitosox Red staining
For mitochondrial speci c superoxide radical generation measurement, MitoSOX™ Red Mitochondrial Superoxide Indicator (M36008, Invitrogen, USA) was used according to manufacturer's instruction. In brief, cells were plated and cultured in 24-well (for uorescence assay) or 6-well (for ow cytometry) plates. After receiving treatment as described previously, cells were washed for 3 times with Hank's solution (37 ℃) after removing culturing medium. Then cells were incubated with Mitosox Red (5 um) and Hoechst staining (5 ug/ml, Beyotime, China) for 30 min in light-proof humidi ed chamber at 37 °C. Then cells were ready for following experiments.

Cellular ROS assay
Generation of cellular ROS was assessed by DCFH-DA probe (Reactive Oxygen Species Assay Kit, Beyotime, China). All procedures were applied according to manufacturer's introduction. In brief, cells were plated and cultured in 24-well (for uorescence assay) or 6-well (for ow cytometry) plates. After receiving treatment as described previously, cells were washed 3 times with Hank's solution (37 ℃) after removing medium. Then DCFH-DA (10 uM) was added to each well following by 20-miniute incubation in light-proof humidi ed chamber at 37 °C. Then cells were ready for following experiments.

Flow cytometry
For Mitosox red and DCFH-DA assay, cells were plated and cultured in 6-well plates, then received treatment as described previously. For macrophage/microglia polarization assay, cells were plated and cultured in 6-well plates. After harvesting, cells were washed with PBS and stained with anti-iNOS/Arg-1 antibody (17-5920-80/12-3697-82, eBioscience, USA). Flow cytometry was performed using a BD FACS Aria (BD Bioscience, USA). And data was analyzed using FlowJo Software (Tree Star Inc, USA).

Western blotting
For the determination of protein expression, Western-blotting analysis was performed. BV2 cells cultured in six-well plates were treated as described previously. Firstly, cells were rinsed with pre-cold PBS for 3 times and then lysed with RIPA-Buffer (Beyotime, China) and protease inhibitor cocktail (Beyotime, China) on ice for 30 min. Then lysed extraction were centrifugated for 10 min, 12000g at 4 °C. BCA protein assay kit (Beyotime, China) was used to determine protein content in all samples. SDS-polyacrylamide gel electrophoresis was performed followed by transferring to polyvinylidene uoride (PVDF) membranes for 70 min at 250 mA. Then transferred PVDF membranes were blocked with 5% fat-free milk dissolved in TBST. Membranes were incubated with primary antibody overnight at 4 °C. On the next day, appropriate HRP-conjugated secondary antibody (1:5000) was added for one-hour incubation at room temperature. After several times of washing with TBST, bands were scanned and analyzed after adding enhanced chemiluminescence (ECL, Millipore, USA) through imaging system. Bands were normalized to β-actin.
And data were analyzed and presented through ImageJ and GraphPad Prism version 9 (GraphPad Software, USA).

Immuno uorescence staining
After receiving prior treatment, cells and tissue slides were applied with 30-miniute 4% paraformaldehyde xation at room temperature following by PBS washing for 3 times and 15-miniute 0.5% Triton X-100 permeabilization. Then cells and slides were incubated with 5% donkey serum for 1 h at room temperature. After incubation with primary antibody overnight at 4 °C, cells and slides were washed for 3 times by PBS and incubated with appropriate secondary antibodies for 1 h. Then incubated cells and slides were treated with DAPI staining medium (Beyotime, China) for 10 min at 37 ℃. Images were captured by uorescence microscope (DMI8, Leica, Germany) The primary antibodies used in this research included: iNOS (1:100, ab15323, Abcam, UK), Arg-1 (1:100, ab91279, Abcam, UK).

Histological staining
Hematoxylin and Eosin (HE) staining and Luxol-fast-blue (LFB) staining were performed to evaluate the tissue preservation and myeline loss on day 28 after SCI at the transection of lesion center. The protocol of HE and LFB staining was based on the well-established protocol reported previously 17,18 . Pictures were captured with DMI6 microscopy (Leica, Germany).

Function analysis
To evaluate function recovery of animals in SCI after MitoQ treatment, BMS test and footprint analysis were performed. BMS was used to evaluate the motor function recovery of the hind limbs. In brief, two independent researchers were trained and blind to the grouping of tested animals. BMS test were performed on the day before the surgery, day 0, 7, 14, 21 and 28.
Footprint analysis was performed through a CatWalk gait analysis system as reported previously 19 . Animals were pre-trained before the examinations. On day 28 after SCI, animals received the gait examinations at least three times per round. Pictures and videos were captured through camera of the system.

Statistical analysis
All statistical analysis were performed through GraphPad Prism version 9. And all experiments were repeated at least three times. Data are expressed in the form of mean ± standard error of the mean from three independent examinations. Student's t test when performed to examine the statistical signi cance between two groups. One-way analysis of variance (ANOVA) with Bonferroni tests analysis was conducted to examine the differences among three groups. A p value lower than 0.05 was considered to be statistically signi cant.

Results
Macrophage/microglia polarization dysfunction after SCI Activation of macrophage/microglia is a vital pathological process in SCI, which is the major contributor of microenvironment disturbance after SCI. And in the acute period, M1-like macrophage/microglia are in the majority and remains predominant ( Figure. 1A, C, p 0.0001) and the expression of M2-like macrophage/microglia is transient and signi cantly lower than M1-like subtypes ( Figure. 1B, D, p 0.0001). And in subacute phase, the expression of M1-like macrophage/microglia is still signi cantly higher than that in Sham group ( Figure. 2A, C, p 0.001). However, the expression of M2-like macrophage/microglia decreased and showed no signi cant difference comparing to that in Sham group ( Figure. 2B, D). As previously reported, macrophage polarization in the late phase of organ injury repair is commonly transferred into an anti-in ammatory subtypes 20 (commonly considered as M2-like subtypes). The previous results showed that macrophage/microglia polarization dysfunction (prolonged M1 subtypes and transient M2 subtypes) leaded to a prolonged neuroin ammation and might be one of the major causes of impaired recovery of SCI.

MitoQ attenuates LPS-induced mitochondrial biodynamic imbalance in BV2 cells
To investigate MitoQ's potential role in macrophage/microglia function modulation, we rstly used LPS, a common M1-polarization-inducing factor, to induce M1-like polarization in BV2 cells. To identify whether mitochondrial biodynamic is disturbed in M1-like macrophage/microglia, the expression of mitochondrial biodynamic related protein (Mfn-1/2 and Drp-1) and their inner balance were measured through Western blotting assay (Figure. MitoQ reduces LPS-induced ROS production and following oxidative stress in BV2 cells Microenvironment oxidative stress is also an important secondary pathological process after SCI. Mitochondria, which is considered as the main source of cellular ROS in cells, is a potential target of ROS scavenger. To investigate MitoQ's effect on oxidative stress inhibition, the mitochondrial-speci c ROS production was assayed through Mitosox Red staining with uorescence assay and ow cytometry. In uorescence assay, the integrated density ratio to control group of Mitosox Red in MitoQ group signi cantly decreased comparing to LPS group ( Figure.  MitoQ attenuates LPS-induced and SCI-induced apoptosis through mitochondrial function improvement Cell death, apoptosis following primary mechanical SCI are important pathogenesis process contributing to impaired tissue preservation and function recovery of spinal cord. And mitochondria is the key to apoptotic process. In-vitro examinations including Western blotting and Hochest staining were implemented. As shown in Figure.  In-vivo examination was performed to identify MitoQ' anti-apoptosis effect in SCI animal models ( Figure. 6). To further con rm MitoQ's role in apoptosis inhibition, Tunel staining was performed on day 3 and day 14 after SCI. As performed in Figure. 5C-D, signi cant decrease in Tunel-positive area was found in MitoQ group comparing to SCI group (p 0.0001), which indicated that MitoQ could inhibit cell apoptosis in-vivo.
Taking these results together, we found that MitoQ could inhibit cell apoptosis through promotion of mitochondrial function in-vitro and in-vivo.

MitoQ promotes function recovery, tissue sparing and neural preservation after SCI
To assess the effect of MitoQ on function recovery after SCI, two behavioral tests were implemented. BMS test was applied weekly among Sham, SCI and MitoQ groups after SCI. On day 7 after SCI, no signi cant change was found between the BMS score of SCI and MitoQ groups. On day 14, 21, 28, MitoQ group showed signi cant increase in BMS score comparing to SCI group ( Figure. 7A, n=3, p 0.05). These data indicated that MitoQ might improve function recovery after the acute phase of SCI. And in footprint assay carried out through catwalk analyzer on day 28 after SCI, the footprint and hindlimb mean intensity were measured. (Figure. 7B). MitoQ groups showed clearer footprint of hindlimb comparing to SCI group ( Figure. 7B-a, b)  Arg-1) was performed. Similar result con rmed that MitoQ could inhibit the M1-polarization after 24-and 48-hour-LPS challenging and partially promote M2-like polarization (Figure. 11E). These data con rmed what we found in the SCI models that MitoQ could modulate macrophage/microglia polarization through a mitochondrial-dependent pathway, which indicated that mitochondrial dysfunction might be the core of the macrophage polarization dysfunction and cellular apoptosis ( Figure. 12).

Discussion
Increasing evidence have suggested the important role of mitochondria in the macrophage/microglia polarization modulation, but few related research have been performed in SCI. In the present study, we investigated the effect of MitoQ, a mitochondria-speci c antioxidant, on SCI and its underlying mechanism through in-vitro and in-vivo examinations. These results indicate that MitoQ could promote M2-like polarization and inhibit M1-like polarization which attenuate neuroin ammation after SCI and promote tissue preservation and neural repair through the mitochondrial-dependent pathway. MitoQ could inhibit cell apoptosis after SCI and LPS challenging in BV2 cells which resulted in the improved locomotor function recovery and tissue preservation. These data suggested that mitochondrial function could be considered as a potential therapeutic target and MitoQ's reliable role in SCI treatment.
In SCI, prolonged neuroin ammation, a major contributor to poor self-regeneration and ongoing damage of spinal cord 21 , has become an important treatment target. Recently, speci c treatments including nafamostat mesilate 22 , mulberrin 23 , matrix metalloproteinase-8 24 , zinc 25 have been used to reduce neuroin ammation in SCI and proved to be neuro-protective and contributive to locomotor recovery. Though prolonged and severe neuroin ammation has been considered to impair SCI self-regeneration and function recovery, mild neuroin ammation in SCI has been proved to be bene t to rehabilitative training result which indicates that controllable and mild in ammation can remove cell debris and dead cells to promotes recovery 26 . And microglia is considered as the crucial mediating cell in post-SCI neuroin ammation. The high exibility of macrophage/microglia has been paid much attention in central nervous system injury. Macrophage/microglia can be shifted into different subtypes including M1-like type and M2-like type in response to distinct stimuli. M1-like macrophage/microglia is considered to be neurotoxic while M2-like type is neuroprotective 27 . Thus, modulation of macrophage/microglia polarization is an important treatment target in SCI.
In our study, we found that LPS-challenging M1-like BV2 cells showed signi cantly impaired mitochondrial function and biodynamic imbalance with decreasing ratio of Mfn-1/2 to Drp-1 and increasing production of cellular and mitochondrial speci c ROS. Recent studies have revealed the important relationship between macrophage polarization and mitochondrial function. Yujia Yuan and colleagues found that mitochondrial ROS contributed to M1 macrophage polarization in diabetes 28 . Calcium uptake, an important physiological process of mitochondria, is essential to M2-like macrophage polarization 29 . Similar results were also found in skeletal muscle regeneration 30 . Given that, attenuation of mitochondrial damage/dysfunction has become a critical treatment target in macrophage phenotype modulation and following in ammation. IL-25, a member of IL-17 cytokine family, was found to regulate M2-like macrophage polarization and promote mitochondrial function 31 . Inhibition of AMP Kinase could enhance M1-like macrophage polarization and reduce mitochondrial biogenesis in chronic kidney disease 32 . Pyruvate dehydrogenase kinase 1 was found to take part in macrophage polarization through the regulation of glucose metabolism (partial function of mitochondria) 33 . And MitoQ is considered as a mitochondrial-targeting antioxidant, which can quickly protect mitochondria from oxidative stress. In invitro examinations, we found that MitoQ could rebalance the mitochondrial biodynamic imbalance, inhibit mitochondrial-speci c and cellular ROS production which indicated MitoQ's role in mitochondrial function modulation. These results are consistent with former research. However, there are few research on the relationship between mitochondrial function modulation and macrophage/microglia polarization in SCI.
In this study, it's the rst time to evaluate the association between mitochondrial dysfunction rescue with MitoQ and its modulation of macrophage/microglia polarization in SCI. As mentioned above, mitochondria function is vital to macrophage/microglia polarization modulation. Classi cation of microglia is similar to that in macrophage. Microglia can be classi ed into M1-like type and M2-like type. The polarization statement of microglia is vital to pathogenesis process of multiple diseases. And in neurological disorders like Parkinson's disease and Alzheimer's Disease, modulation of microglia phenotypes is still the hot zone. Further cellular and molecular mechanisms were investigated in microglia modulation. Triggering receptor expressed on myeloid cells-2 (TREM2) was found to be essential to microglia polarization in Parkinson's disease that it could altering M1 microglia to M2 subtypes 34 . Ion channel including potassium channel/ Kir6.1-containing ATP-sensitive potassium is the modulation of microglia phenotypes 35 . As uncontrollable neuroin ammation impairs self-regeneration ability of spinal cord following SCI. Recently, macrophage metabolism, which contributing to immunometabolism, has become increasingly important and highlighted. Innate immunity can be characterized by metabolic reprogramming change. A recent review concluded 6 major metabolic pathways in macrophage subsets including amino acid, glycolysis, oxidative phosphorylation (OXPHOS), pentose phosphate pathway (PPP), fatty acid synthesis (FAS) and fatty acid oxidation(FAO) 11 . As mitochondria is a vital organelle participating in energy production, it takes part in multiple pathways among those six pathways. Inducible NO synthase (iNOS) and arginase-1(Arg-1) expression is typical marker of M1 and M2 macrophage. Arginine is coverted into nitric oxide(NO) through iNOS in M1 macrophage and it's metabolized by Arg-1 in M2 macrophage 36 .While in energy production, M1-like macrophage is characterized by enhanced PPP and disrupted TCA cycle and M2-like macrophage is mainly dependent on glycolysis 37 . In accordance with the result of in-vitro study, MitoQ could inhibit M1like polarization and promote M2-like polarization on day 3 and day 14 in established SCI animal model. Signi cant function recovery and tissue preservation were observed in MitoQ treatment group. These results indicate mitochondria is an important treatment target for prolonged neuroin ammation resulting from macrophage/microglia polarization dysfunction in SCI.
Cell apoptosis can be considered as the ultimate destiny of neuron and surrounding cells at severely injured site of SCI which leads to neural de ciency and tissue loss. Mitochondria is considered as the multifaceted regulators of cell death 38 . Thus, modulation of mitochondrial function can be bene cial to cell apoptosis after SCI. And our results showed that rescue of mitochondrial function through MitoQ treatment could signi cantly inhibit the apoptosis after SCI and LPS challenging.
Limitation of this research must be declared that the underlying molecular mechanism of how mitochondria modulate macrophage/microglia polarization hasn't been mentioned. As mitochondria is the vital regulator of cell metabolism and destiny, there are various pathway related to mitochondria modulation including mechanistic/mammalian target of rapamycin (mTOR) pathway 39,40 and NF-κB pathway 41,42 . Molecular mechanism of this process requires further research to shed light on the inner relationship.