Mdivi-1 | Dna Repair Signal

Mitochondrial division inhibitor 1 (Mdivi-1) offers neuroprotection through diminishing cell death and improving functional outcome in a mouse model of traumatic brain injury

Abstract

Mitochondria dysfunction, an enormous potential crisis, has attracted increasing atten- tion. Disturbed regulation of mitochondrial dynamics, the balance of mitochondrial fusion and ﬁssion, has been implicated in neurodegenerative diseases, such as Parkinson’s disease and cerebral ischemia/reperfusion. However the role of mitochondrial dynamics in traumatic brain injury (TBI) has not been illuminated. The aim of the present study was to investigate the role of Mdivi-1, a small molecule inhibitor of a key mitochondrial ﬁssion protein dynamin-related protein 1 (Drp1), in TBI-induced cell death and functional outcome deﬁcits. Protein expression of Drp1 was ﬁrst investigated. Outcome parameters consist of motor test, Morris water maze, brain edema and lesion volume. Cell death was detected by propidium iodide (PI) labeling, and mitochondrial morphology was assessed using transmission electron microscopy. In addition, the expression of apoptosis-related proteins cytochrome c (cyt-c) and caspase-3 was investigated. Our ﬁndings showed that up-regulation of Drp1 expression started at 1 h post-TBI and peaked at 24 h, but inhibition of Drp1 by Mdivi-1 signiﬁcantly alleviated TBI-induced behavioral deﬁcits and brain edema, reduced morphological change of mitochondria, and decreased TBI-induced cell death together with lesion volume. Moreover, treatment with Mdivi-1 remarkably inhibited TBI- induced the release of cyt-c from mitochondria to cytoplasm, and activation of caspase-3

1. Introduction

Traumatic brain injury (TBI), a major cause of morbidity and mortality (Jin et al., 2015), represents the quintessential neuropsychiatric paradigm with a combination of effects in cognition, personality, and the risk for psychiatric disorders (Santopietro et al., 2015). However, the hope for an effective treatment is derived from the fact that much of the post- traumatic damage to the injured brain is caused by a secondary injury cascade of pathochemical and pathophy- siological events that exacerbates the primary mechanical TBI (Mustafa et al., 2010).

Mitochondria, the primary energy-generating system in most eukaryotic cells, have been shown to be a crucial participant in TBI pathophysiology (Gajavelli et al., 2015). Furthermore, mitochondrial morphology is orchestrated by a well conserved cellular machinery comprised of dynamin- related GTPases, dynamin-related protein 1 (Drp1) for ﬁssion and mitofusions (Mfn1 and Mfn2) and optic atrophy-1 (OPA1) for fusion (Purnell and Ox, 2013; Wang et al., 2013a). Drp1, which is target to the outer mitochondrial membrane, is primarily found in the cytosol, and it localizes to discrete spots on mitochondrial surface to initiate ﬁssion by interaction with Fis1 (Chan, 2006; Qi et al., 2013; Sharp et al., 2015). Mdivi- 1, a mitochondrial division inhibitor, is a highly efﬁcacious small molecule acting as a selective inhibitor of Drp1 (Zhang et al., 2013b).

In the past few years, many researchers have demonstrated that TBI causes impaired mitochondrial function and impaired mitochondrial functional integrity (Watson et al., 2013), mito- chondrial ﬁssion and fragmentation play an active role in apoptotic cell death (Wang et al., 2012b; Yuan et al., 2007), and the mitochondrial dysfunction may trigger or exacerbate dama- ging secondary intracellular cascades after the primary injury (Watson et al., 2013). Emerging evidences have suggested blocking mitochondrial ﬁssion is protective in animal models of cerebral ischemia/reperfusion and various neurodegenerative disorders such as Parkinson’s disease (Knott and Bossy-Wetzel, 2008; Lackner and Nunnari, 2009; Ong et al., 2010). However the role of mitochondrial ﬁssion in TBI model has not been illuminated.

According to the above, we hypothesized that TBI results in mitochondrial ﬁssion which initiated by Drp1 and leads to cell death, but inhibition of Drp1 may prevent TBI-induced cell death, maintain normal mitochondrial morphology, and alleviate behavioral deﬁcits, brain edema and lesion volume. Therefore, by using Mdivi-1, a small molecule inhibitor of Drp1, we investigated that whether Mdivi-1 has neuroprotec- tive effect after TBI and the potential mechanism undergoing by which Drp1 regulates cell death. This was associated with reducing morphological changes of mitochondria, cyto- chrome c release and inhibiting activation of apoptosis.

2. Results

2.1. The time course of Drp1 expression in a mouse model of TBI

Drp1, the target protein of the outer mitochondrial mem- brane, was ﬁrst determined in the present study. Drp1 protein was analyzed by western blotting from 1 h to 7 d after TBI and the time course expression was shown in Fig. 1. Up-regulation of Drp1 started at 1 h post-TBI and peaked at 24 h, then slightly decreased afterward (Fig. 1A, Po0.05).

2.2. Mdivi-1 alleviated TBI-induced behavioral deﬁcits and brain water contents

Previous studies have indicated that TBI elicited a signiﬁcant decline in motor performance and cognitive function (Luo et al., 2011). To further determine whether the inhibition of Drp1 by Mdivi-1 was associated with improved neurologic outcome, we sought to perform behavior experiments. First, TBI elicited a signiﬁcant decline in motor performance on days 1–5 post-injury. In marked contrast, treatment with Mdivi-1 (Fig. 2A, Po0.05) accelerated the recovery of motor functional outcome on days 3–5 post-TBI.

Fig. 1 – Time dependent change in Drp1 protein induced by TBI. (A) TBI was induced for 1 h, 6 h, 12 h, 24 h, 48 h, 72 h, and 7 d. Extracts from injured cortex were separated on SDS-PAGEL and protein levels of Drp1 were detected with immunoblotting. (B) Optical densities of the protein bands were analyzed with ChemiScope analysis and normalized with loading control (β- actin). Bars represent mean7SD (n¼ 3). *Po0.05 vs. sham group.

Fig. 2 – Treatment with Mdivi-1 attenuated TBI-induced behavioral deﬁcits and brain edema. (A) Mice (n ¼ 8) were randomly assigned to sham, vehicle, and Mdivi-1 treated groups after TBI. Wire grip test was performed 1–7 d after injury. After TBI, motor impairment was observed as revealed by decreases in motor performance. #Po0.05 vs. sham group; *Po0.05 vs. vehicle group. (B) To ensure recovery from motor deﬁcits, MWM testing was performed on days 8–16 after TBI, and the animals in each group (n ¼ 8) were those who had suffered above wire grip test. #Po0.05 vs. sham group; *Po0.05 vs. vehicle group. (C) The water content of injured hemisphere and heterolateral hemisphere were measured from 1 to 7 d after TBI. #Po0.05 vs.
sham group (n¼ 5). (D) The effects of Mdivi-1 on brain edema were detected at 24 h after TBI. *Po0.05 vs. vehicle group (n ¼ 5).

To ensure recovery from motor deﬁcits, Morris-water maze testing was performed on days 8–16 after TBI. As shown in Fig. 2B, vehicle-treated animals displayed increased latencies in the ability to ﬁnd the hidden platform, versus sham group on days 15 and 16. After injury, animals subjected to Mdivi-1 treatment demonstrated a signiﬁcant decrease in the latencies, relative to vehicle mice on days 15 and 16 (Fig. 5B;Po0.05), thereby indicating Mdivi-1 treatment could result in cognitive functional recovery.

According to the time proﬁle of brain edema after TBI, the water content of the injured hemisphere increased 12 h, peaked 24 h to 48 h, and lasted to 72 h after TBI, compared with sham group (Fig. 2C). Intraperitoneal injection of Mdivi-1 ameliorated TBI-induced brain edema at 24 h post-TBI (Fig. 2D, Po0.05).

2.3. Mdivi-1 treatment decreased TBI-induced cell death and lesion volume

To elucidate the effect of Mdivi-1 treatment on TBI-induced cell death, PI-positive cells were counted as described previously (Luo et al., 2010). The number of PI-positive cells markedly up- regulated at 1 h post-TBI and peaked at 24 h, then declined but remained elevated on day 7 after TBI (Fig. 3A, Po0.05). Whereas, Mdivi-1 treatment markedly reduced the number of PI-positive cells, compared with the vehicle group at 24 h post-TBI (Fig. 3B). To visualize and analyze the overall density of total cells, sections were mounted with DAPI for nuclear staining. To indentify clearly the proportion of PI-positive cells, we combined PI labeling with DAPI nuclear staining (Fig. 3C).

To explore whether PI-positive cells can represent TBI- induced cell loss, the present study assessed the cumulative loss of brain tissue after TBI. The results showed that TBI caused profound tissue loss in the brain, but Mdivi-1 sig- niﬁcantly reduced the lesion volume compared with the vehicle-treated groups (Fig. 4).

2.4. Mdivi-1 reversed morphometric changes of mitochondria at 24 h after TBI

Morphometric changes were observed by TEM after TBI. As shown in Fig. 5, normal cells keep a balance between mitochondrial fusion and ﬁssion (Fig. 5A and D), and the abnormal mitochondria could be found in neurons at 24 h post-TBI obviously. The morphological abnormalities of mito- chondria were noted (Fig. 5B and E), including severe mitochondrial fragmentation, collapsed cristae, mitochon- drial swell and rupture of mitochondrial membrane, in addition to a decrease in mitochondrial density and increased heterogeneity in size and shape. The results suggest that mitochondria in neurons are deformed after TBI. Whereas, Mdivi-1 treatment dramatically attenuated mitochondrial fragmentation and showed lager diameter (Fig. 5C and F).

Fig. 3 – Mdivi-1 treatment decreased TBI-induced cell death detected by PI labeling. (A) The number of PI-positive cells markedly upregulated at 1 h post-TBI and reached its peaked at 24 h, then declined but remained elevated on day 7 after TBI. ##Po0.05 vs. sham group (n ¼ 5). (B) Mdivi-1 treatment markedly reduced the number of PI-positive cells, compared with the vehicle group at 24 h post-TBI. *Po0.05 vs. vehicle group (n¼ 5). (C) Double ﬂuorescent staining for PI, excluded from viable cells to indentify dead cells, and DAPI, binding selectively to DNA, in brain cortex. Representative images of PI (red) and DAPI (blue) from sham, injured animals receiving vehicle, and injured animals receiving Mdivi-1 at 24 h after TBI.

2.5. Mdivi-1 suppressed TBI-induced release of cytochrome c from mitochondria to cytosol and caspase-3 activation at 24 h after TBI

We next assessed blotting for cytochrome c in puriﬁed mitochondria. Western blot analysis (Fig. 6) showed that a signiﬁcant reduction in protein levels of cytochrome c in mitochondrial fraction and a concomitant increase in cyto- chrome c levels in cytosolic fraction were observed at 24 h after TBI. Whereas, treatment with Mdivi-1 (3 mg/kg) signiﬁ- cantly inhibited TBI-induced release of cytochrome c from mitochondria to cytosol.

The release of cytochrome c from mitochondria triggers activation of downstream caspases. Western blot analysis demonstrated that caspase-3 was activated at 24 h after TBI, but Mdivi-1 effectively inhibited caspase-3 activation (Fig. 7). Taken together, our data suggest that the inhibitory activation of caspase-3 may relate to the neuroprotective ability of Mdivi-1 against cell death in mouse TBI model.

Fig. 4 – Effects of Mdivi-1 on TBI-induced lesions. (A) All animals were sacriﬁced 17 days after TBI. 500-μm Brain sections from the entire brain stained with Cresyl Violet. (B) The areas of the lesion, both injured and noninjured hemisphere, were determined using an image analysis system. Lesion volume was quantitatively analyzed with Sigma Scan Pro5. *Po0.05 vs. vehicle group (n¼ 8).

Fig. 5 – Representative EM images of mitochondrial ultra-structures after TBI. (A) sham group; (B) vehicle group; (C) Mdivi-1 group; (D) Magniﬁcation of A showing normal mitochondrial morphology; (E) Magniﬁcation of B indicating mitochondrial fragmentation; (F) Magniﬁcation of C showing lager diameter mitochondria. The asterisks indicate mitochondrial fragment in (E). Arrows indicate enlarged mitochondria in (F). Scale bar¼ 2 μm.

Fig. 6 – TBI-induced the release of cyt-c from mitochondria to cytosol was inhibited by Mdivi-1. (A) Each sample at 24 h after TBI was homogenized and centrifuged for isolation of mitochondrial and cytosolic fractions. (B, C) Optical densities of the protein bands were analyzed with ChemiScope analysis and normalized with loading control (HSP-60, β-actin). Bars represent mean7SD (n¼ 3). #Po0.05 vs. sham group and *Po0.05 vs. vehicle group.

3. Discussion

Traumatic brain injury (TBI), a common cause of disability and death worldwide, causes cell death and behavioral defects (Al Nimer et al., 2015; Jin et al., 2015). There was limited advance in the therapeutic strategies to counter brain injury. Except for conservative management, neuroprotection and neurorecovery are still the main therapeutic strategies under development. Recent researches indicated that mito- chondrial ﬁssion, an important step in apoptosis, has been associated with cell death (Frank et al., 2001; Wang et al., 2013a). In addition, the GTPase dynamin-related protein 1 (Drp1) is an important protein that participates in mitochon- drial ﬁssion, and plays a role in mitochondrial fragmentation (Frank et al., 2001; Wang et al., 2013a). By blocking Drp1 activity with pharmacological approach, such as Mdivi-1 in this study, we attempted to investigate the role of Mdivi-1, a small molecule inhibitor of Drp1, in TBI-induced cell death and functional outcome deﬁcits. We examined the effect of Mdivi-1 on the overall excessive ﬁssion of mitochondria caused by TBI, and explored that Mdivi-1 signiﬁcantly pre- vented cell death and rebuilt the balance of mitochondrial ﬁssion and fusion in TBI model. Our results are consistent with the previous ﬁndings that blocking mitochondrial ﬁssion is protective in animal models of renal or myocardial ische- mia and various neurodegenerative disorders (Knott and Bossy-Wetzel, 2008; Lackner and Nunnari, 2009; Ong et al., 2010).

First of all, we investigated whether Mdivi-1 ameliorated behavioral deﬁcits. Previous studies demonstrate that typical TBI patient deﬁcits in attention, memory, and behavioral control (Santopietro et al., 2015). Moreover, Cerebral edema has been reported to be one of the major factors leading to the high mortality and long-term disability associated with patients with TBI (Cui and Zhu, 2015; Wang et al., 2015). We found that treatment with Mdivi-1 signiﬁcantly ameliorated traumatic-induced motor and resulted in a signiﬁcant ame- lioration in injured mice in hidden platform. In addition, the time proﬁle of brain edema after TBI was determined, and the effects of Mdivi-1 on brain edema were detected at 24 h after TBI in this study, at which time the brain water content reached the peak value in TBI model. Our data provided that treatment of Mdivi-1 signiﬁcantly improved the neurological outcome and mitigated brain edema after TBI.

Previous studies indicated that plasmalemma damage which portends fatal celluar injury occurs earlier than DNA damage, and loss of plasmalemma integrity, initiating dele- terious cascades, is a hallmark of cellular injury and death after TBI (Wang et al., 2013b; Whalen et al., 2008; Xiaoxia Zhu et al., 2012). Propidium iodide (PI)-labeling was used to identify injured cells (Whalen et al., 2008), and to follow the fate of cells injured by TBI (Luo et al., 2011). Therefore, the number of PI-positive cells was counted to assess the effect of Mdivi-1 on cell death after TBI. The result showed that the time course of the number of PI-positive cells was similar to that of Drp1 expression, indicating Drp1 may be involved in the pathophysiologic responses, such as celluar injury and death after TBI. Moreover, inhibition of Drp1 by Mdivi-1 resulted in a signiﬁcant decrease in the amount of PI- positive cells at 24 h post-TBI. Although PI positive cells predicted cell death (detected by PI labeling) and represented TBI-induced cumulative loss of cells (detected by lesion volume), what kind of pathway it underwent was not clear.

Fig. 7 – Inhibition of TBI-induced caspase-3 activation by Mdivi-1. (A) caspase-3 protein levels were detected with immunoblotting at 24 h post TBI. (B) Optical densities of the protein bands were analyzed with ChemiScope analysis and normalized with loading control (β-actin). Bars represent mean7SD (n ¼ 3). #Po0.05 vs. sham group and *Po0.05 vs.
vehicle group.

Evidences implied that apoptosis is an important type of cell death which may cause pathological changes and severe outcome post TBI (Wang et al., 2012a; Zhang et al., 2013a). In addition, mitochondria take part in intrinsic apoptotic path- way which involves a diverse of non-receptor-mediated stimuli producing intracellular signals (Elmore, 2007). It has also been reported that mitochondrial fragmentation resulted from an activation of physiological ﬁssion is an early phe- nomenon during apoptotic cell death (Li and Dewson, 2015; Frank et al., 2001). Therefore, we decided to determine mitochondrial morphological alterations and biochemical features changes at 24 h after TBI. A signiﬁcant improvement in the fraction of elongated mitochondria in Mdivi-1 treated group was observed compared to vehicle group with the large proportion of smaller mitochondria. Moreover, as is known to all, cytochrome c is a member of electron transport chain which participates in respiration. Besides, the release of cytochrome c from mitochondria is a vital signal in apoptosis initiation (Babbitt et al., 2015). It is also known that cyto- chrome c is normally sequestered within mitochondrial cristae folds and mitochondrial cristae rearrangement allow the release of cytochrome (Estaquier and Arnoult, 2007). This release in turn activates downstream caspase to initiate apoptosis. We chose caspase-3, regarded as executioner caspase, representative caspases family on account of that it has a predominant role in the nervous system (Boland et al., 2013; Wall and McCormick, 2014). Our data indicate that treatment with Mdivi-1 inhibited Drp1 causing mitochondrial ﬁssion, prevented cytochrome c release to cytoplasm and down-regulated the levels of active caspase-3 at 24 h post- TBI, sequentially, restrained apoptosis occurring at a rela- tively early stage after TBI.

Our data suggest that inhibition of mitochondrial ﬁssion (Drp1) might be a potential therapeutic target for TBI. In spite of this, there are still many deﬁciencies needed to explore in the future, such as the changes of mitochondrial function and the lasting effects of Mdivi-1 after TBI. Further studies examining the effects of Drp1 on more related signaling pathways are also needed to identify novel potential targets for treatment strategies in patients with traumatic brain injury.

In summary, our ﬁndings demonstrated that Drp1 inhibi- tion, using Mdivi-1, may represent a potential therapeutic agent for TBI treatment. Drp1 inhibition not only attenuated TBI-induced cell death, but also signiﬁcantly reduced lesion volume, behavioral outcome deﬁcits and brain water con- tents. This approach may be attributable to its inhibition of mitochondrial ﬁssion and apoptosis in TBI model. The under- lying mechanism may be associated with a reversal of the morphological change of mitochondria, inhibiting mitochon- drial ﬁssion, cytochrome c release and apoptosis activation induced by TBI.

4. Experimental procedure

4.1. TBI model and drug administration

All experiments were performed on male ICR mice, weighing 25–30 g upon arrival, and all procedures were in compliance with the NIH Guide for the care and Use of Laboratory Animals. Adult male mice were anesthetized with 4% chloral hydrate (0.4 mg/g) and mounted in a stereotaxic system (David Kopf Instruments, Tujunga, California). All animals were mounted in the stereotactic frame, a skin incision was performed to expose the skull, and a 3-mm craniotomy was made lateral to the sagittal suture and centered between bregma and lambda, the skull cap was carefully removed, without disruption of the underlying dura. Mice were sub- jected to TBI, which can causes serious cerebral cortex damage, in left part of the brain using a weight-drop model, a 40 g weight dropped from 20 cm onto a 2-mm-diameter footplate resting on the dura with a controlled depth of 1.0 mm (Bao et al., 2012; Luo et al., 2010, 2011). Otherwise, through the accurate location, hit pressure, depth, and hitting duration, the reproducibility and consistency of this TBI model were ensured and the craniotomy was closed imme- diately after TBI, which did not signiﬁcantly affect physiolo- gical parameters (arterial pressure, heart rate or body weight) (Luo et al., 2010). The experiments were divided into three groups: sham, Mdivi-1 and vehicle. Sham-operated rats only underwent the surgical procedure except for cortical impact. For the mice treated with Mdivi-1 (3 mg/kg), Drp1 inhibitor, was administered by intraperitoneal injection 10 min after TBI. Mdivi-1 was purchased from Tocris Bioscience Company and dissolved in 0.01 M PBS. Vehicle animals received an intraperitoneal injection of 0.01 M PBS.

4.2. Assessment of motor function

We use wire grip test to evaluate motor performance on days 1-7 after TBI as described previously (Luo et al., 2010; Mbye et al., 2012). Mice (n ¼ 8) were placed on a metal wire (35 cm long), suspended between two wooden bars 50 cm above a foam pad, and allowed to traverse the wire for 60 seconds.
The time mice remain on the wire and the manner mice held on the wire (holding, paws, tail, moving) were scored using a six-point scoring system. Each test was repeated three times for each mouse and the investigators were blinded to treat- ment conditions.

4.3. Assessment of Morris water maze

Morris water maze (MWM) was used to evaluate the spatial cognitive abilities of injured mice (Zohar et al., 2003). A white pool (90 cm in diameter and 50 cm deep) was ﬁlled with water, which was colored black, to a depth of 27 cm. A round, clear, plexiglass 10-cm-diameter platform was placed 1 cm
below the water surface and was positioned 10–15 cm from the southwest wall. The mice (n ¼ 8) were placed at a random starting position (N, S, E, W), facing the pool wall, and were given 90 s to ﬁnd to the platform and climb onto it. Experi- menter guided the mouse to the platform when it failed to reach there within 90 s. Mice were permitted to remain on the platform for 15 s, and then were placed in a warmed box. To ensure recovery from motor deﬁcits, testing was performed on days 8–16 after TBI and the animals in each group (n ¼ 8) were those who had suffered wire grip test. The performance was quantiﬁed by the latency to ﬁnd the hidden platform and the rate of learning over the trial period, which were shown to be more sensitive in detecting injury effects in mice. Animals were tracked by a computerized video system that measured path length.

4.4. Brain water content measurement

Brain edema was determined using the wet/dry method as previously described (Cui and Zhu, 2015). Anesthetized ani- mals (n ¼ 5) were sacriﬁced by decapitation, and the brain was removed and divided into ﬁve parts, injured cortex, injured medulla, contralateral cortex, contralateral medulla and cer- ebellum. Each part was weighed immediately after removal (wet weight) and weighed again after drying for 24 hours at 100 1C in an oven (dry weight). The following formula was used to calculate the brain water content: (wet weight— dry weight)/wet weight ~ 100%.

4.5. Propidium iodide labeling

Animals anesthetized using 4% chloral hydrate at a dose of 0.4 mg/kg, losing plasmalemma integrity, were evaluated by intraperitoneal injection of 0.4 mg/ml PI (Sigma, B1221) 1 h before sacriﬁcing. The brains were quickly removed and ten series of 12 μm sections (200 μm apart from each section)
from anterior to posterior hippocampus (bregma — 1.90 to — 3.00) were sliced using cryostat microtome (Thermo ﬁsher, Cryotome FSE, USA) and placed on poly-L-lysine-coated glass slides. All cortical brain regions were chosen from 200 ~ cortical ﬁelds from within contused cortex. Five to eight 200 ~ cortical ﬁelds were randomly chosen to count the numbers of PI positive cells with ﬂuorescence microscope (NIKON, ECLIPSEE Ti). To visualize and analyze the overall density of total cells, 4, 6-diamidino-2-phenyl-indole (DAPI) was used for nuclear staining. Five animals were used in each group, and the mean number of positive cells for a given time point was calculated by summing the cell count data from all of the brain sections (Wang et al., 2015).

4.6. Preparation of mitochondrial and cytosolic fractions

Isolation of mitochondria was performed as described pre- viously (Wang et al., 2011; Luo et al., 2013). Brieﬂy, each sample was homogenized in 0.3 mL icecold buffer A contain- ing 250 mM sucrose, 1 mM EDTA, 50 mM Tris–HCl, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl ﬂuoride, 1 mM benzamidine, 0.28 U/mL apotinin, 50 μg/mL leupeptin, and 7 μg/mL pepstatin A (pH adjusted to 7.4 with NaOH) and homogenized with a glass Pyrex microhomogenizer (30 strokes). The homogenate was centrifuged at 1000g at 4 1C for 10 min, and the resultant supernatant was transferred to a new tube to be centrifuged at 12,000g at 4 1C for 20 min to obtain the mitochondrial pellet and supernatant. The super- natant was centrifuged at 100,000g for 1 h at 4 1C to generate the cytosolic fraction. The mitochondrial pellet was washed three times in buffer B containing 250 mM sucrose, 1 mM EGTA, 10 mM Tris–HCl (pH 7.4), spun at 12,000g at 4 1C for 10 min, and then lysed in Western blot lysis buffer.

4.7. Western blot analysis

Injured cortical tissues were homogenized in Western blot analysis buffer containing 10 mM Tris–HCl (pH 7.4), 150 mM NaCl, 1% (v/v) triton X-100, 1% sodium deoxycholate, 0.1% SDS, 5 mM EDTA, 1 mM PMSF, 0.28 kU/I aprotinin, 50 mg/L leupeptin, 1 mM benzamidine, and 7 mg/L pepstain A (all chemicals were purchased from Sigma-Aldrich Inc., St. Louis, MO, USA). The homogenate was then centrifuged at 12,000 rpm for 10 min at 4 1C and the supernatant was retained and preserved at — 80 1C for later use. Proteins were extracted from brain tissue and protein concentrations were
determined using BCA kit (Pierce Chemical, Rockford, IL, USA). Each sample, twenty micrograms of protein, was loaded on a 10% and 15% sodium dodecyl sulfate- polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to polyvinylidene ﬂuoride (PVDF) membranes on a semidry electrotransferring unit (Bio-Rad, USA). The mem- branes were blocked with Tris-buffered saline containing 0.1% Tween-20 (TBST) containing 5% nonfat dry milk for 2 h, and then incubated with antibodies against Drp1(Milli- pore, ABT155, 1:1000), Caspase-3 (Bioworld, BS1511, 1:500), Cytochrome C (Abcam, ab110325, 1:500), COX IV (Abcam, ab16056, 1:1000) in TBST and 5% nonfat dry milk overnight at 1C. After the overnight incubation with the primary anti-
bodies, membranes were washed and incubated with horse-radish peroxidase-conjugated second antibody in TBST for 2 h. Immunoreactivity was detected using the ECL chemilu- minescence system (ChemiScope, Shanghai). The membranes were reprobed with β-actin (CMCTAG, AT0001,
1:2000) and Hsp60 (Abcam, ab46798, 1:1000) after striping. The signal intensity of primary antibody binding was quanti- tatively analyzed with ChemiScope analysis and was normalized to a loading control, β-actin or Hsp60.

4.8. Lesion volume measurements

Seventeen days after TBI, all animals (n ¼ 8) which had received MWM were anesthetized and sacriﬁced. The brains were removed, post-ﬁxed for 12 h, and then cryoprotected in 15% sucrose in PBS. The frozen brains were sectioned at 12 μm in the coronal plane, 500 μm apart from each section,and stained with Cresyl Violet. Lesion volume was expressed in cubic millimeters in the injured hemisphere and as percentage volume of the noninjured hemisphere. Lesion volume was quantitatively analyzed with Sigma Scan Pro 5 and average lesion volume in microliters for each group was calculated.

4.9. Electron microscopy and morphometry assay

The produce was divided into three groups: sham, vehicle and Mdivi-1, and 3 animals were used in each group. Brain tissues were ﬁxed with glutaraldehyde for transmission electron microscopy analysis. In brief, the brains, removed and cut into small blocks (1 mm3), were ﬁxed with 2.5% glutaraldehyde in 1 M phosphate buffer, pH 7.4. Before pro- cessed the blocks, they were washed by 0.1 M PBS for several times, and then removed and post-ﬁxed in 1% osmium tetraoxide for 1 h. Following dehydration with a graded series of acetone, the blocks were embedded in Epon 812 and cured overnight at 60 1C. Serial ultrathin sections were cut with the ultramicrotome and viewed through a transmission electron microscopy (HT-7700, HITACHI, Tokyo, Japan). The measure- ment of mitochondria was performed as described previously (Xu et al., 2008).

4.10. Statistic analysis

All data were given as mean values7standard deviation (SD). Statistic analysis was carried out by one-way ANOVA followed by Dunnett t-test or Student t-test (two means com- parison), using the related programs in SPSS 14.0.