Distinct roles for metalloproteinases during traumatic brain injury

Neurochemistry International xxx (2016) 1e10

Contents lists available at ScienceDirect

Neurochemistry International journal homepage: www.elsevier.com/locate/nci

Distinct roles for metalloproteinases during traumatic brain injury Si Zhang a, Luba Kojic a, Michelle Tsang a, Parampal Grewal a, Jie Liu b, Dhananjay Namjoshi c, Cheryl L. Wellington c, Wolfram Tetzlaff b, Max S. Cynader a, William Jia a, * a

Brain Research Center, University of British Columbia, Vancouver, BC, Canada International Collaboration on Repair Discoveries (ICORD), University of British Columbia, Vancouver, BC, Canada c Department of Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 August 2015 Received in revised form 16 February 2016 Accepted 25 February 2016 Available online xxx

Background: Significant protease activations have been reported after traumatic brain injury (TBI). These proteases are responsible for cleavage of transmembrane proteins in neurons, glial, and endothelial cells and this results in the release of their extracellular domains (ectodomains). Methods: Two TBI models were employed here, representing both closed head injury (CHI) and open head injury (OHI). In situ zymography, immunohistochemistry, bright field and confocal microscopy, quantification of immunopositive cells and statistical analysis were applied. Results: We found, using in situ zymography, that gelatinase activity of matrix metalloproteinases (MMP)-2 and MMP-9 was upregulated in cortex of both injury models. Using immunohistochemistry for several MPPs (Matrix metalloproteinases) and ADAMs (disintegrin and metalloproteinases), including MMP-2, -9, ADAM-10, -17, distinct patterns of induction were observed in the two TBI models. In closed head injury, an early increase in protein expression of MMP-2, -9 and ADAM-17 was found as early as 10 min post injury in cortex and peaked at 1 h for all 4 proteases examined. In contrast, after OHI the maximal expression was observed locally neighboring the impact site, at a later time-point, as long as 24 h after the injury for MMP-2 and MMP-9. Confocal microscopy revealed colocalization of the 4 proteases with the neuronal marker NeuN in CHI, but only MMP2 colocalized with NeuN in OHI. Conclusions: The findings may lead to a trauma-induced therapeutic strategy triggered soon after a primary insult to improve survival and to reduce brain damage following TBI. © 2016 Published by Elsevier Ltd.

Keywords: Traumatic brain injury Closed and open head injury Mice model Metalloproteinases Gelatinase activity Immunohistochemistry

1. Background Traumatic brain injury (TBI) is a leading cause of mortality and morbidity among young adults and children in the developed world. After traumatic brain injury the initial primary mechanical tissue damage quickly triggers a complex cascade of secondary injury, including vascular, metabolic, cellular, and molecular processes that exacerbate damage, limit recovery, and contribute to

* Corresponding author. Neurosurgery Division, Department of Surgery, Faculty of Medicine, University of British Columbia, 2211 Wesbrook Mall, Vancouver V6T 2B5, Canada. E-mail addresses: [email protected] (S. Zhang), [email protected] (L. Kojic), [email protected] (M. Tsang), [email protected] (P. Grewal), [email protected] (J. Liu), [email protected] (D. Namjoshi), [email protected] (C.L. Wellington), [email protected] (W. Tetzlaff), [email protected] (M.S. Cynader), [email protected] (W. Jia).

overall morbidity and mortality (Reilly, 2001). The secondary mechanisms of injury including oxidative stress, inflammatory response, and excitotoxicity lead to activation of several matrix metalloproteinases (Gold et al., 2009; Lo et al., 2002; Pineda et al., 2004; Siao and Tsirka, 2002; Vilalta et al., 2008b; Yong, 2005) that further exacerbate the injury resulting in opening of the bloodebrain barrier (BBB) (Cunningham et al., 2005), prevention of normal cell signaling, and eventually leading to cell death (Shlosberg et al., 2010). Current management, focused on preventing secondary brain injury, has demonstrated significant progress in recent decades leading to reduced mortality. Halting the evolution of the primary injury is a critical goal for management of TBI. Matrix metalloproteinases (MMPs) and the disintegrin and metalloproteinases (ADAMs) are families of zinc-binding proteolytic enzymes, and represent major regulatory systems in the brain. They normally remodel the extracellular matrix and play important roles in the development of the nervous system, regulating

http://dx.doi.org/10.1016/j.neuint.2016.02.013 0197-0186/© 2016 Published by Elsevier Ltd.

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013

2

S. Zhang et al. / Neurochemistry International xxx (2016) 1e10

proliferation, migration, differentiation and survival of various cells, as well as axonal growth and myelination (Yang et al., 2006).They also pathologically degrade substrates as part of the neuroinflammatory response during ischemia (Cunningham et al., 2005) and neurodegeneration (Rosenberg, 2009). By cleaving their substrates, metalloproteinases are responsible for shedding of transmembrane proteins in neurons, glial, and endothelial cells and the release of their extracellular domains (ectodomains) (Peschon et al., 1998). The injury-induced neuroinflammatory response increases the activity of metalloproteinases. Several key MMPs and ADAMs have been implicated in neuroinflammation after brain injury including MMP-2 and MMP-9 (gelatinases A and B), MMP-3 (stromelysin-1), membrane-type MMPs (MT1-MMP or MMP-14), ADAM-17 also known as tumor necrosis factor-a (TNFa) converting enzyme (TACE), and tissue inhibitor of metalloproteinases TIMP-3 (Candelario-Jalil et al., 2009; Grossetete and Rosenberg, 2008; Vilalta et al., 2008b; Walker and Rosenberg, 2009; Xue et al., 2009). MMP-2 and MMP-9 are secreted as latent enzymes and are involved in diverse homeostatic and pathological processes (Cunningham et al., 2005; Rosenberg, 2009; Yong, 2005). Both require activation by the proconvertase furin, itself activated by HIF1a, a hypoxia-induced transcription factor. MMP-2 is constitutively present in large quantities in normal astrocytes and CSF. Since it is tethered to the cell surface, MMP-2 demonstrates spacerestricted proteolytic activity, only in the vicinity of the membrane. On the other hand, MMP-9 is inducible and is released into the extracellular space without spaceeconstraints, to degrade multiple proteins in the extracellular matrix (ECM) surrounding neurons (Cunningham et al., 2005; Rosenberg, 2009; Yong, 2005). ADAMs are transmembrane proteins that mediate intercellular signaling by binding to integrins, but are also important for intracellular signaling and cell adhesion (Blobel, 2000). Although more than thirty mammalian ADAMs have been identified so far, only 17 are known to have some role in the brain (Rosenberg, 2009; Yang et al., 2006). ADAM-10 (Kuzbanian) and ADAM-17 (also known as TNF-a converting enzyme TACE) are membrane bound proteases expressed in cerebral and cerebellar cortex, hippocampus, and hypothalamus, in neurons, endothelial cells and astrocytes (Goddard et al., 2001; Skovronsky et al., 2001). In addition to their role in APP a-secretase activity (Asai et al., 2003; Buxbaum et al., 1998; Postina, 2008), both cleave many substrates including proTNFa, Notch, type-IV collagen, prion precursor, CD44, ephrins, cell adhesion molecule NCAM-L1, EGFR ligand, IL-1 receptor II, IL-6 receptor, L-selectin, type-XVII collagen, and chemokine fractalkine (Ludwig et al., 2005). The expression and activity of ADAM-17 increases under pathological conditions such as stroke and traumatic brain injury. It also promotes neural progenitor cell migration and contributes to stroke-induced neurogenesis (Katakowski et al., 2007). A growing body of evidence suggests that metalloproteinases are markedly upregulated in the brain in response to injury and are responsible for the propagation and regulation of neuroinflammatory processes that accompany many forms of CNS disease (Cunningham et al., 2005; Rosenberg, 2009; Yong et al., 2007b). However, information regarding their activation in the brain following traumatic brain injury is lacking. The first aim of this study was to examine the expression and activation pattern of various metalloproteinases after traumatic brain injury. Information on the activation of particular proteases shortly after a brain injury is required for the development of a new type of gene delivery vehicle that we are currently developing. This consists of a membrane-bound construct, embodying a specific cleavage site and a therapeutic amino acid sequence that can be released by specific proteases following neurotrauma. This type of construct

could have powerful protective potential, since it would spring into action rapidly after the traumatic insult even if the victim were alone or far from a medical facility. Our lab is in the process of developing such a trauma-inducible gene delivery system (Zhang et al., 2012, 2013) and understanding the pattern of protease activity following neurotrauma is an important step along this developmental path. 2. Material and methods 2.1. Traumatic brain injury All procedures were approved by the Animal Ethics Committee (AEC) of the University of British Columbia, BC, Canada. 2.2. Closed head injury A closed head injury (CHI) model of TBI (Flierl et al., 2009) was employed to assess the expression and activation of proteases induced immediately after traumatic brain injury. The CHI is a model of concussive and diffuse brain injury. This type of injury is difficult to reproduce with fluid percussion, controlled cortical impact, or focal brain contusion models that are more often associated with focal axonal damage. This model induces a CHI using a standardized weight-drop device inducing a focal blunt injury over an intact skull without pre-injury manipulations. Furthermore, the CHI weight drop TBI model triggers a profound neuroinflammatory response leading to neurological impairment and breakdown of the bloodebrain barrier. As a result, the CHI model can induce early brain edema followed by both apoptotic and necrotic neuronal cell death. Adult male C57Bl6 mice (Center for Disease Modeling, UBC) were housed under controlled environmental conditions with ambient temperature of 22
C, relative humidity of 65% and 12 h light/dark cycle, with free access to food and water. All surgical procedures were conducted using aseptic techniques and animals were kept warm using a heating pad. Adult (8e10 weeks old) C57Bl6 mice were anesthetized with isoflurane (induction: 3e4%, maintenance: 1.5e2%) in oxygen (0.9 L/min) delivered through a nose-cone. A surgical plane of anesthesia was maintained throughout the surgery, confirmed by testing for loss of papillary and corneal reflexes as well as loss of toe-pinch reflex. Lubricating eye ointment was applied to prevent corneal drying. Ketamine (15 mg/kg) and xylazine (1.75 mg/kg) mixture (both s.c.), meloxicam (1 mg/kg, s.c.), and bupivacaine (0.125 mg/kg, s.c., under the scalp) were administered for analgesia. Sterile, warm saline (1 ml/ 100 g body weight) was administered s.c. to prevent dehydration.The skin overlying the head of the animal was shaved and a midline longitudinal incision in the skin was performed. The skin was gently retracted to expose the calvarium. An area ~2 mm lateral to the sagittal suture and ~2 mm posterior to the coronal suture in the left parietal bone of the head was marked for injury. Anesthesia was momentarily discontinued, the animal was quickly moved under the weight-drop device, and a 95 g weight was dropped on the skull from a height of about 6 cm to induce a moderate trauma resulting in a focal injury to the left hemisphere.The tip of the Teflon-tipped cone of the injury device has a tip diameter of 2 mm. The precise drop height (cm) of the weight was adjusted to be onethird the weight (g) of the animal. Mean body weight and drop height did not differ significantly across the sham and injured groups. Anesthesia was reinstated following injury. The incision was closed and the animals were kept in a recovery chamber with a heating pad until fully recovered. They were then returned to their cages. Mice that sustained skull fractures were immediately euthanized. Sham controls received isoflurane anesthesia, pre-

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013

S. Zhang et al. / Neurochemistry International xxx (2016) 1e10

surgical analgesia, skin incision, and were placed under the injury device without the weight drop procedure. Mice of the injury group were randomly assigned to one of three assessment time points, i.e., 10 min, 1 h, and 24 h after the injury. 2.3. Control Cortical Impact A Control Cortical Impact (CCI) animal model of open head injury (OHI) was employed that allows for superior control over mechanical factors (Dixon et al., 1991; Lighthall, 1988). The model replicates a penetrating brain injury with skull deformation and cortical compression associated with direct cortical injury (Dixon et al., 1991; Smith et al., 1995). In general, the injury induced with CCI is usually restricted to the lower brainstem and produces a more focused injury compared with other models of TBI. CCI injury reproduces some changes reported in clinical head injuries such as brain edema that increases intracerebral pressure and reduces cortical perfusion and cerebral blood flow. OHI can cause significant neuroendocrine and metabolic changes in the brain that may lead to coma. The CCI was delivered to intact dura by an Infinite Horizon impactor causing deformation of the underlying cortex. Briefly, anesthetized adult C57Bl6 mice were placed in a stereotaxic frame fitted with a muzzle that delivered isoflurane throughout the surgery. The core body temperature was maintained at 37-38
degrees. A midline longitudinal incision in the skin was performed and the skin was gently retracted to expose the calvarium. An area ~2 mm lateral to the sagittal suture and ~2 mm posterior to the coronal suture in the left parietal bone was subjected to a 5 mm diameter craniotomy under surgical microscope to ensure the dura and pia mater kept intact. CCI was induced using an Ohio State University (OSU) impactor based on an electromagnetic device guiding a sensor regulated probewith a tip of 1.7 mm diameter. The electromagnetic device was mounted on a solid metal frame so that the position of the impactor could be adjusted. The tip was gently lowered down until it was 4 mm above the dura. The impact was then triggered to provide a force of 35 kdyn at 100 mm/s during the impact. In the injury groups, the force of the injury was 34.13 ± 2.29 kdyn and the depth of the cortical deformation was 1.779 ± 0.021 mm. Animals with dura damage were immediately euthanized. Sham animals participated in the entire surgical process except for the injury impact. The animal's body temperature was maintained at 37
C during the surgery. After the injury animals were put into a recovery chamber with a heating pad until they were fully awake and moving about. Then they were returned to their cages with free access to food and water. At 10 min, 1 h and 24 h after the injury, various groups of animals were euthanized and the brains were obtained. 2.4. Immunohistochemistry For immunohistochemical assays, frozen sections (30 mm) from mouse brain were prepared using a cryostat (Leica CM3050 S, Leica, Germany). The sections were incubated free floating in individual wells of a 24-well cell culture plate (Corning Inc., Corning, NY, USA) at 21
C on an orbital shaker. Sections were first washed 3 times for 10 min each in PBS to remove residual OCT. They were then incubated in a blocking solution of 4% bovine serum albumin (BSA) (SigmaeAldrich, St. Louis, MO, USA) in PBS-0.1% Triton X-100 (SigmaeAldrich, St. Louis, MO, USA) for one hour before overnight incubation with the primary antibody diluted in a solution of 1% BSA in PBS-0.1% Triton X-100: rabbit polyclonal anti-matrix metalloprotease-2 (MMP2) antibody (Abcam, Cambridge, MA, USA, dilution 1:100), rabbit polyclonal anti-matrix metalloprotease-9 (MMP9) antibody (Abcam, Cambridge, MA, USA, dilution 1:100),

3

rabbit polyclonal anti-A disintegrin and metalloprotease-10 (ADAM10) antibody (Lifespan Biosciences, Seattle, WA, USA, dilution 1:100) and rabbit polyclonal anti-A disintegrin and metalloprotease 17 (ADAM-17) antibody (Lifespan Biosciences, Seattle, WA, USA, dilution 1:25), Hypoxyprobe™-1 mouse monoclonal antibody (MAb1) (HPI Inc., Burlington, MA, USA, dilution 1:50), mouse monoclonal anti-NeuN (Millipore, Temecula, CA, USA, dilution 1:500), rat monoclonal anti-CD11b (AbD Serotec, Raleigh, NC, USA, dilution 1:100). Sections were washed 3 times in PBS-0.1% Triton X-100 for 10 min each before incubation with the secondary antibody diluted in 1% BSA in PBS-0.1% Triton X-100: biotinylated goat anti-rabbit antibody (Millipore, Temecula, CA, USA, dilution 1:500), biotinylated goat anti-mouse antibody Ready-to-use (Abcam, Cambridge, MA, USA, no dilution), AlexaFluor-488 goat anti-rabbit antibody (Invitrogen, Carlsbad, CA, USA, dilution 1:200), AlexaFluor-568 goat anti-mouse antibody (Invitrogen, Carlsbad, CA, USA, dilution 1:250), AlexaFluor-555 goat anti-rat antibody (Invitrogen, Carlsbad, CA, USA, dilution 1:250). For staining with diaminobenzidine (DAB), all sections were incubated with the secondary antibody for 3 h, washed 3 times in PBS-0.1% Triton X100 for 10 min each, then incubated in a solution of preformed avidin and biotinylated horseradish peroxidase macromolecular complex (Vectastain ABC Kit, Vector Laboratories, Burlingame, CA, USA, dilution 1:100) for two hours. They were then washed two times in PBS-0.1% Triton X-100 for 10 min each, washed one time in 20 mM Tris-buffered saline (TBS) for 10 min and then incubated in a solution of diaminobenzidine (DAB) prepared from DAB Substrate Kit for Peroxidase (Vector Laboratories, Burlingame, CA, USA, dilution DAB 1:25, buffer 1:50, hydrogen peroxide 1:1000) for 2 min. Afterwards, sections were washed once for 10 min in distilled water followed by twice for 10 min in TBS. Finally, sections were then mounted onto glass slides, allowed to air dry overnight and covered with a mounting medium of 1:1 Permount (Fisher Scientific, Toronto, ON, CA): xylene and coverslips. For double-label fluorescent staining, sections were first incubated with AlexaFluor488 goat anti-rabbit antibody for two hours in the case of sections previously incubated with anti-MMP-2 and anti-MMP-9 and four hours in the case of sections previously incubated with anti-ADAM10 and anti-ADAM-17. Sections were then washed 3 times in PBS0.1% Triton X-100 for 10 min each before incubation for two hours with AlexaFluor-568 goat anti-mouse antibody in the case of sections previously incubated with anti-NeuN and AlexaFluor-555 goat anti-rat antibody in the case of sections previously incubated with anti-CD11b. After incubation, sections were washed three times in PBS-0.1% Triton X-100 for 10 min each, incubated for five minutes in a solution of 4’,-6-diamidino-2-phenylindole (DAPI) (SigmaeAldrich, St. Louis, MO, USA, dilution 1 mg/mL), washed three times in PBS-0.1% Triton X-100 for 10 min each again before mounting onto glass slides and air dried overnight. Sections were covered with Fluoromount-G (Southern Biotech, Birmingham, AB, USA) and coverslips. 2.5. In situ zymography Localization of gelatinase (MMP-2 and MMP-9) activity in brain sections was obtained using in situ zymography with the MMP fluorogenic substrate DQegelatineFITC (Molecular Probes, Eugene, OR, USA) as previously described (Amantea et al., 2008). OCTembedded fresh brain cryostat cut sections (20 mm thick) were obtained from mice 10 min, 1 h and 24 h after both closed and open head TBI using 5 animals for each experimental group and for control and sham-operated groups. Briefly, mouse brains were quickly dissected immediately embedded in OCT (Tissue-Tek, Sakura Finetek 4583, USA) and snap frozen in liquid nitrogen. 20 mm thick coronal sections were cut using a cryostat, air-dried for

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013

4

S. Zhang et al. / Neurochemistry International xxx (2016) 1e10

1 h at room temperature, then rehydrated in PBS and incubated overnight in PBS at 37
C with 40 mg/ml of quenched fluorogenic substrate DQegelatineFITC (Molecular Probes, Eugene, OR, USA) that remains quenched until it is cleaved by gelatinase activity. After washing out the excess of fluorogenic substrate the sections were fixed with 2% paraformaldehyde (PFA) in PBS for 5 min and the nuclei counterstained with DAPI for 15 min at room temperature. As a control, gelatinolytic activity was inhibited by EDTA (20 mM), a chelator for Ca2þ ions required for MMP activity. The green fluorescence induced by gelatinolytic activity was examined by confocal microscopy for each brain, and multiple regions across multiple coronal sections that covered the lesion. The patterns of gelatinase activity were quantified for each experimental group at each time point after TBI. To assess the cellular localization of gelatinolytic activity, in situ zymography was combined with immunohistochemistry for neuronal specific marker NeuN. After the DQegelatineFITC procedure brain slices were washed with PBS (3  5 min) and fixed with 2% PFA in PBS for 15 min. Sections were then incubated overnight at 4
C with the mouse anti-NeuN monoclonal antibody (1:100)and 0.3% Triton X-100. After 3  5 min rinses in PBS, sections were incubated for 2 h at RT with the secondary antibody Cy3-conjugated goat anti-mouse IgG (1:100) and counterstained DAPI, mounted on gelatin-coated slides, air-dried and coverslipped. The gelatinolytic activity was quantified with a laser confocal spectral microscope (Olympus FV-1000, USA) and images were acquired and processed using Olympus Fluoview software. Coronal brain sections were taken from each brain in the vicinity of the impact. Digitized images were acquired under identical microscope settings. The fluorescence intensity was analyzed off-line for dorsal and ventral cortex, hippocampus and thalamus. Nuclear (DAPI), cytosolic and ECM compartments were quantified for metalloproteinase activity.

2.6. Microscopy and imaging Sections stained with DAB were imaged with bright field microscopy using Northern Eclipse 8.0 (Empix Imaging Inc., Mississauga, ON, CA) and a CCD camera at 5 magnification for rat brain tissue and 10 magnification for mouse brain tissue. Sections stained with fluorescence were imaged with laser scanning confocal microscopy using the FV1000 (Olympus, Center Valley, PA, USA). After imaging with bright field microscopy, different regions of the brain (dorsal cortex, ventral cortex, hippocampus and thalamus for TBI and cortex, striatum, hippocampus and thalamus for MCAO) were rated on a 1e10 scale based on staining intensity. Immunopositive cells for MMP2, MMP-9, ADAM-10 and ADAM-17 were counted from 400  400 mm grid areas of confocal images. The total number of neurons and cells present in the same area were also counted based on NeuN and DAPI signals and used to calculate the proportion of protease immunopositive neurons and cells.

2.7. Statistical analysis Data were analyzed using ImageJ software (NIH). Statistical comparisons among groups were done using ANOVA with post hoc analysis for ANOVA or t-test (PRISM 4.0; GraphPad Software Incorporated, San Diego, CA, USA). All data are presented as mean ± SEM. Statistical significance was set at p < 0.05, <0.01, or <0.001, which are marked as *, **, or *** in the figures, respectively. Error bars in figures represent standard errors of the mean.

3. Results 3.1. Protease activity is upregulated in the brain after the traumatic injuries We first asked whether protease activity is changed in the cortex after either closed or open brain injuries. As quantitative methods to measure the specific activity of each of the metalloproteinases mentioned above are not currently available, we measured gelatinase activity instead as a general indication for changes of protease activity (Amantea et al., 2008). Fig. 1 illustrates gelatinase activity in the cortex of the Closed (Fig. 1A, B) and Open Head Injury (Fig. 1C, D) models at one hour post injury. Quantitative analysis was performed to compare the density of gelatinase-positive cells between sham and injured groups of CHI (Fig. 1E) and OHI (Fig. 1F) animals. Significantly upregulated gelatinase activity was found in the cortex ipsilateral to both the closed and open head injury. The gelatinase activities were markedly more diffuse in CHI than those seen in OHI. These results show that the gelatinase activities are significantly upregulated in injured cortex of both the closed head and the open head injury groups. 3.2. Protease expression increased underneath the impact areas after a closed head injury Since there are no specific markers for the activity of individual proteases, we looked at the expression changes of several of MPPs (Matrix metalloproteinases) and ADAMs (disintegrin and metalloproteinases), including MMP-2, MMP-9, ADAM-10, ADAM-17. We measured the time courses of changes of expression after the closed head injury for MMP-2, MMP-9, ADAM-10 and ADAM-17 in the cortices ipsilateral to the impact (Fig. 2). As shown in Fig. 2AeD and quantified in Fig. 2Q, elevated levels of MMP-2 could be observed as soon as 10 min after closed head injury, with a return to near baseline by 24 h following injury. A similar time course pattern was also seen for MMP-9 and ADAM-17 (Fig. 2EeH and MP, respectively; quantified in Fig. 2R and T, respectively), except that ADAM-17 did not return to the baseline by 24 h post trauma. Interestingly, ADAM-10 had a slightly delayed response but also returned to baseline at 24 h (Fig. 2IeL). Among the four proteases, ADAM-17 was the most responsive proteinase and it showed the most dramatic increase in the first 10 min. In all layers of the cortex, cells with increased density of MMP-2 staining were found (Fig. 2B) after CHI. The PAN-cortical increases of expression were also found for ADAM-17 (Fig. 2M vs N). Increased expression seemed more remarkable in layer V for MMP9. The higher expression level of MMP-9 in the deeper cortical layers was still seen at 24 h post trauma. Expression of ADAM-10 in sham treated brains was more concentrated in the superficial layers (Fig. 2I). After injury, this pattern remained at the early time point (Fig. 2J). However, a more dramatic increase of ADAM-10 expression in deeper layers was seen at 1 h post injury (Fig. 2K). Overall these results indicate that shortly after the closed head injury, the expression of MMP-2, MMP-9, ADAM-10, and ADAM-17 increased dramatically in the cortex near the impact site. 3.3. Protease expression increased topographically after an open head injury As a comparison, we examined protease expression in brains following induction of open head injury. Fig. 3 illustrates immunohistochemical staining for MMP-2, MMP-9, ADAM-10, and ADAM-17, on the sides of the brains in sham animals or in treated animals at 24 h post OHI. In all cases, we compared the immunohistochemical profiles at baseline (sham, with changes that

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013

S. Zhang et al. / Neurochemistry International xxx (2016) 1e10

5

Fig. 1. In situ Zymography in the cortex of the Closed and Open Head Injury mice. The photomicrographs reflecting the results of in situ zymography results for closed head injury (Sham and 1 h post injury, n ¼ 5 each) and open head injuries (Sham and 1 h post injury, n ¼ 5 each) are presented in Panels a and b, and Panels c and d, respectively. Quantitative results comparing the density of the positive cells in each model are shown under each panel. All data are presented as means ± SEM. Statistical significance was set at p < 0.05, <0.01, or <0.001, which are marked as *, **, or *** in the figures, respectively. Error bars in the figures represent standard errors of the mean.

Fig. 2. Immunohistochemistry of MMP2, MMP9, ADAM10, and ADAM17 following Closed Head Injury. Immunohistochemical staining in dorsal cortex of the Closed Head Injury hemisphere are presented in left panels of the figure. Results for MMP2 are shown in Panels a, b, c, d, for sham (n ¼ 5), 10 min (n ¼ 5), 1 h (n ¼ 5) and 24 h (n ¼ 5) groups, respectively. Similarly, the results for MMP9 are shown in Panels e, f, g and h. Those of ADAM10 and ADAM17 are shown in Panels i, j, k, l, and m, n, o, p, respectively. Quantitative analyses of the density of immunopositive cells are presented in diagrams, located to the right of the staining panels accordingly (q for MMP2, r for MMP9, s for ADAM10, and t for ADAM17). All data are presented as means ± SEM. Statistical significance was assessed as p < 0.05, <0.01, or <0.001, which are marked as *, **, or *** in the figures, respectively. Error bars in the figures represent standard errors of the mean.

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013

6

S. Zhang et al. / Neurochemistry International xxx (2016) 1e10

Fig. 3. Immunohistochemistry of MMP2, MMP9, ADAM10, and ADAM17 in the Open Head Injury model. The results of immunohistochemical staining in Sham animals (n ¼ 4) are presented in the left panels of the figure (a, c, e and g). Those for 24 h post injury ones (n ¼ 4) are shown in the right panels (b, d, f and h), in which the borders between zones lateral to and in the center of the injury are labeled by dotted lines. Results for MMP2 are shown in panels a and b, indicating staining in the ipsilateral cortex of Sham and the injury groups, respectively. Similarly, the results of MMP9 are shown in c and d. Those for ADAM10 and ADAM17 are shown in Panels e and f, g and h, respectively. Quantitative results on the density of the immunopositive cells are presented in the diagrams, lying to the right of representative photographic panels (i and j for MMP2, k and l for MMP9, m and n for ADAM10, o and p for ADAM17). All data are presented as means ± SEM. Statistical significance was set at p < 0.05, <0.01, or <0.001, indicated as *, **, or *** in the figures, respectively. Error bars in the figures represent standard errors of the mean.

occurred 10 min, 1 h, or 24 h after injury). In contrast to the widespread cortical increases observed in CHI, the change of the protease expression in OHI is more topographically restricted. As shown in Fig. 3B,D,F and H, the density and distribution of the immunopositive cells in the cortex underneath OHI were distinctly different from that seen in the cortex lateral to the injury site. We compared the time course and degree of alteration in the various proteases expression at the center of the injury with changes occurring 1mmlateral to the injury site. The most striking results were obtained for MMP-2 (Panels A, B, Graphed in I and J), with less marked changes observed for MMP-9 (Panel C, D). The increase in ADAM-10 only reached statistical significance at the 24 h time point near the center of the injury, (Panels E, F, M, N), while the changes for ADAM-17 were not statistically significant. Interestingly, statistically significant increases in expression levels of MMP-2 and MMP-9 could be observed as early as 10 min after injury in the injured hemisphere lateral to the injury but not at the center of the injury. The density of immunopositive cells for ADAM17 in the center of injury increased slightly at 10 min but was reduced within 24 h post injury (Fig. 3G and O).

3.4. Protease expression upregulated in cortical neurons post injury We further asked if the upregulation of proteases following the injury was associated with neuronal cells in the cortex. To answer this question, we double stained cells with antibodies for the neuronal cell marker NeuN and for specific proteases. Fig. 4 demonstrates colocalization of the proteases with NeuN in injured cortex in both Closed Head and Open Head models. As shown in Fig. 4, colocalization of the proteases with NeuN immunopositive cell can be found in both sham and 1 h post injury group of CHI (Fig. 4 A vs B, C vs D, E vs F, G vs H), and in both the center of, and lateral to the injury in OHI (Fig. 4 N vs O). Quantitative studies showed that the percentages of MMP-2, ADAM-10 and ADAM-17 immunopositive cells among NeuN immunopositive cells were significantly increased (Fig. 4I, K, L, respectively). It seems that many neuronal cells are MMP-9 positive in sham operated animals, but the level of expression was drastically increased 1 h after closed head injury although there was significant change in the number of MMP-9 positive neurons. A clear difference can be seen in comparisons of the center of the injury with the cortex lateral to the injury in the OHI groups. These results suggest that the brain injuries activate expression of MMP-2, ADAM-10 and ADAM-17 in a

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013

S. Zhang et al. / Neurochemistry International xxx (2016) 1e10

7

Fig. 4. Confocal microscopy of the proteases in the close head and open head injury groups. Microscopic images illustrating double-labeling of NeuN and the various proteases in dorsal cortex of the closed head injury group are presented in the left part of the figure. Results for MMP2 are shown in a and b, in the sham and 1 h post injury groups, respectively (n ¼ 5 for each group). Similarly, the results for MMP9 are shown in panels c and d. Those of ADAM10 and ADAM17 are shown in Panels e and f, and Panels g and h, respectively. Quantitative results on the density of immunopositive cells are presented in the diagrams, situated to the right of the picture panels accordingly (i for MMP2, j for MMP9, k for ADAM10, and l for ADAM17). For the open head injury, the microscopic images showing double-labeling of NeuN and MMP2 are presented on the right side of the figure. Results for MMP2 are shown in m, n and o, for the staining in sham, the center and peripheral areas of the injury, respectively (n ¼ 5 for each group). Quantitative results on the density of immunopositive cells are presented in the diagrams lying below the picture panel (p). All data are presented as means ± SEM. Statistical significance was set at p < 0.05, <0.01, or <0.001, which are marked as *, **, or *** in the figures, respectively. Error bars in the figures represent standard errors of the mean.

large neuronal population after the Closed Head Injuries, but that MMP2 is activated more specifically in the Open Head Injury group. 4. Discussion and conclusion Here we investigated the regional expression and activation of metalloproteinases, including MMP-2, MMP-9, ADAM-10, and ADAM-17 in two brain trauma models, involving open and closed head injury in adult mice at different time points post-injury. The distribution of each protease was demonstrated by immunohistochemistry and laser confocal microscopy. Upregulation of in situ gelatinase activity in the brain was confirmed with zymography. The findings are summarized in Table 1. Briefly stated, the closed

head injury model showed an earlier upregulation and earlier peak of expression than did the open head injury, although both models showed clear increases in both protease density and activity. In our CHI model, increased expression of MMP-2, MMP-9 and ADAM-17 was found in the dorsal cortex, as well as hippocampus and thalamus (data not shown) bilaterally as early as 10 min postinjury, with increased expression more pronounced on the side of brain ipsilateral to the injury. Our studies revealed that the intensity of immunopositivity for proteases after CHI was increased in NeuN positive cells (neurons) as well as the extracellular matrix (ECM) as compared to the sham controls. Abnormal expression of matrix metalloproteinases has been previously implicated in the initiation of the secondary injury mechanism, including breakdown

Table 1 Summary of protease changes after closed and open head injuries. Traumatic brain injuries Protease Activity (Zymography) Proteases Expression Topography Strong proteases Increased area (s) Peak time Proportion in Neuron with the protease expression

Close head injury Increase Increase Diffuse MMP2 MMP9 Diffused 10 min 1h Increase No

Open head injury

ADAM10

ADAM17

1h Increase

1h Increase

Center vs. Lateral MMP2 Center & Lateral 24 h Increase

MMP9 Lateral 24 h No

ADAM10 Center 24 h No

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013

8

S. Zhang et al. / Neurochemistry International xxx (2016) 1e10

of bloodebrain barrier (BBB), brain edema, and neuroinflammation that accompany most central nervous system diseases. Following brain trauma, a cascade of adverse events occurs, including excitotoxicity, oxidative stress, and inflammatory responses that lead to activation of several matrix metalloproteinases (Gold et al., 2009; Lo et al., 2002; Pineda et al., 2004; Siao and Tsirka, 2002) that further exacerbate the injury. Elevation of MMP-9 and -3 in CSF and blood have been reported in patients with severe traumatic brain injury (Grossetete et al., 2009; Kolar et al., 2008). In a group of patients with traumatic brain injury significantly higher levels of serum MMP-9 and MMP-2 were found during the first 3 days post injury that was correlated with ultrastructural changes in endothelial cells and perivascular hemorrhage (Vajtr et al., 2009). Gelatin and in situ zymography has demonstrated increased levels of MMP-2 and MMP-9 activity in brain contusions when compared to controls. MMP-9 activity was increased as early as 3 h after TBI (Wang et al., 2000), peaking at 24e96 h (Vilalta et al., 2008a), persisting for up to 1 week (Wang et al., 2000), and declining until day 14 when active MMP-9 was not detectable (Hadass et al., 2013). MMP-2 was only modestly increased between 24 and 96 h (Vilalta et al., 2008a). The high levels of gelatinase activity found in plasma and brain extracellular fluid in the early and later phases of TBI (Vilalta et al., 2008b), indicate that both local and systemic upregulation of gelatinases is likely associated with the development of delayed secondary brain injury that develops within hours to days or even weeks following the primary mechanical insult. This secondary injury after TBI results in a disruption of the bloodebrain barrier, inflammatory responses induced by activation of local microglia and bloodinfiltrated macrophages (Kumar and Loane, 2012), triggering brain edema and cell death (Loane et al., 2015). Our results show that this cascade of pathological events is initiated very early following the TBI. Locally, the events are presented distinctly with topographical and temporal specificity, according to the 2 kinds of physical impact forces and intactness of the skull. Multiple transcription factors regulate the expression and activation of metalloproteinases in normal and injured brain. In animal models of CHI, hypoxia-inducible factor 1 alpha (HIF1a) induction appears to be responsible for the increased expression of MMP-9 and BBB disruption (Higashida et al., 2011). Reactive oxygen species (ROS) including hydrogen peroxide and nitric oxide also regulate activation of MMPs to degrade occludin, a tight junction membrane protein (Alexander and Elrod, 2002), and indirectly activate matrix metalloproteinases (MMPs), that both contribute to open BBB (Abdul-Muneer et al., 2015). Activation of nuclear factor NF-kB, the proinflammatory transcription factor, induced transcription of MMP-9 and proinflammatory cytokines, such as TNF-a, IL-1b, and IL-6 in the brain after TBI (Chen et al., 2009; Li et al., 2009). Poly (ADP-ribose) polymerase-1 (PARP-1) activation can also contribute to tissue damage after brain trauma. PARP1 can stimulate NF-kB and/or AP-1 mediated release of matrix MMP-9 from activated microglia causing BBB damage (Haddad et al., 2008). Furthermore, the PARP-1 inhibitor PJ34 has been shown to attenuate the expression and activation of MMP-9 after cerebral ischemia (Haddad et al., 2008), and to decrease microglial activation, neuronal loss, and motor deficits after TBI (Stoica et al., 2014). The MMP-9 gene promoter contains a phorbol ester responsive region that binds AP-1 proteins, including dimers of early immediate response genes c-Fos and c-Jun (Rosenberg, 1995). Neutrophil infiltration is common to both brain and spinal cord injury, and these invading leukocytes represent an additional source of MMP-9 in the brain after brain injury (Yong et al., 2007a). We demonstrated the presence of gelatinase reactivity in the nuclei of some neurons with others showing enhanced cytoplasmic and membrane expression after TBI. The colocalization of

gelatinases with NeuN in apoptotic neurons may suggest a novel intranuclear gelatinase activity that may be linked to the pathophysiology of TBI. Similarly, it has been earlier suggested that intranuclear MMP-9 activity can facilitate the development of oxidative injury and neuronal apoptosis during early brain ischemia by cleaving PARP-1, and therefore, by interfering with DNA repair mechanisms (Yang et al., 2010). We observed a dramatic early upregulation of MMP-2, MMP-9, ADAM-10, and ADAM-17 near the impact site of the cortex that peaks within the first hour after closed head injury. In contrast, in open head injury the metalloproteinases expression peaked later, after 24 h. A hallmark of CHI is a massive diffuse axonal injury throughout the brain, with widespread and bilateral damage of the neurons, dendrites, and microvasculature (Foda and Marmarou, 1994; Marmarou et al., 1994). Diffusion-weighted magnetic resonance imaging showed a vasogenic edema rapidly developed within the first hour post injury, that was followed by a slower process of cellular swelling (Barzo et al., 1997). The OHI model is associated mostly with focal axonal pathology rather than the diffuse axonal injury present in CHI. We and others found that the histopathological changes after OHI were mostly restricted to the subcortical region, midbrain, pons, and medulla (Gennarelli, 1994). After TBI axonal injury, necrotic and apoptotic cell death contribute to the overall pathology of brain injury both in humans (Raghupathi, 2004) and animal models of open and closed head injury (Cernak, 2005). The cell death mechanisms after TBI are more likely a continuum between necrotic and apoptotic pathways where the apoptotic cell death is more delayed and prolonged (Raghupathi, 2004). Our study has demonstrated that CHI induced an early (1 h) upregulation peak of the neuronal MMP-2, ADAM-10, ADAM-17 and non-neuronal MMP-9 (Table 1), while the OHI induced a delayed (24 h) increase of neuronal MMP-2 and non-neuronal MMP-9. Results from our study are consistent with those of Lee et al. (2012) for aged and adult mouse brain where OHI increased MMP-9 expression and activity accompanied by disruption of BBB, more in aged than in adult brain, while decreasing the claudin-5 expression and blood brain barrier repair (Lee et al., 2012). In another model of CHI in infant rat brain, trauma significantly increased the protein levels for MMP-2 and MMP-9 in the ipsilateral thalamus at 6 h, with a peak at 24 h post trauma (Sifringer et al., 2007). In the group of OHI animals, in contrast to CHI animals, we found an increased expression of MMP-2 in NeuN positive cells in the dorsal cortex at the impact site and surrounding areas and therefore MMP-2 may be considered an appropriate therapeutic target in open head injury. MMP-9 also showed an increased expression after the OHI. We observed upregulation of gelatinase activity as early as 1 h post-OHI injury in the dorsal cortex, and also the CA3 area of hippocampus and the thalamus (Data not shown). This was confirmed with in situ zymography on the side of the brain ipsilateral to the injury. TBI has already become a leading cause of morbidity and mortality due the failure to translate experimental successes into clinical results and because of the difficulties of providing rapid treatment under difficult circumstances. After the immediate primary trauma the secondary injury evolves over minutes to days and include disturbed ionic homeostasis and energy metabolism, excitotoxicity, and initiation of brain and systemic inflammatory and immune responses resulting in delayed, widespread and progressive neurodegeneration. Despite the preclinical development of many promising therapeutics designed to reduce the secondary injury, over 30 late phase clinical trials have failed (Kabadi et al., 2015). The complexity of secondary injury has raised the need for urgent development of new treatment strategies that will target

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013

S. Zhang et al. / Neurochemistry International xxx (2016) 1e10

multiple secondary mechanisms (Loane et al., 2015). Our results suggest that both CHI and OHI induce upregulation of metalloproteinase expression and activity in the brain following distinct temporal and spatial patterns. The development of tissue damage that follows the injury implicates metalloproteinases as mediators of neuroinflammation (Candelario-Jalil et al., 2009; Rosenberg, 2009). This study further highlights the potential and importance of future prophylactic strategies that can exploit and manipulate the activation of metalloproteinase in the brain after injury and develop inducible- and multi-targeted gene therapies that will alleviate the damage and promote neural repair in the adult nervous system after injury. Competing interests disclosed The authors have read and understood the policy on declaration of interests and declare that we have no competing interests. References Abdul-Muneer, P.M., Pfister, B.J., Haorah, J., Chandra, N., 2015. Role of matrix metalloproteinases in the pathogenesis of traumatic brain injury. Mol. Neurobiol. Alexander, J.S., Elrod, J.W., 2002. Extracellular matrix, junctional integrity and matrix metalloproteinase interactions in endothelial permeability regulation. J. Anat. 200, 561e574. Amantea, D., Corasaniti, M.T., Mercuri, N.B., Bernardi, G., Bagetta, G., 2008. Brain regional and cellular localization of gelatinase activity in rat that have undergone transient middle cerebral artery occlusion. Neuroscience 152, 8e17. Asai, M., Hattori, C., Szabo, B., Sasagawa, N., Maruyama, K., Tanuma, S., Ishiura, S., 2003. Putative function of ADAM9, ADAM10, and ADAM17 as APP alpha-secretase. Biochem. Biophys. Res. Commun. 301, 231e235. Barzo, P., Marmarou, A., Fatouros, P., Hayasaki, K., Corwin, F., 1997. Biphasic pathophysiological response of vasogenic and cellular edema in traumatic brain swelling. Acta Neurochir. Suppl. 70, 119e122. Blobel, C.P., 2000. Remarkable roles of proteolysis on and beyond the cell surface. Curr. Opin. Cell Biol. 12, 606e612. Buxbaum, J.D., Liu, K.N., Luo, Y., Slack, J.L., Stocking, K.L., Peschon, J.J., Johnson, R.S., Castner, B.J., Cerretti, D.P., Black, R.A., 1998. Evidence that tumor necrosis factor alpha converting enzyme is involved in regulated alpha-secretase cleavage of the Alzheimer amyloid protein precursor. J. Biol. Chem. 273, 27765e27767. Candelario-Jalil, E., Yang, Y., Rosenberg, G.A., 2009. Diverse roles of matrix metalloproteinases and tissue inhibitors of metalloproteinases in neuroinflammation and cerebral ischemia. Neuroscience 158, 983e994. Cernak, I., 2005. Animal models of head trauma. NeuroRX 2, 410e422. Chen, G., Zhang, S., Shi, J., Ai, J., Qi, M., Hang, C., 2009. Simvastatin reduces secondary brain injury caused by cortical contusion in rats: possible involvement of TLR4/NF-kappaB pathway. Exp. Neurol. 216, 398e406. Cunningham, L.A., Wetzel, M., Rosenberg, G.A., 2005. Multiple roles for MMPs and TIMPs in cerebral ischemia. Glia 50, 329e339. Dixon, C.E., Clifton, G.L., Lighthall, J.W., Yaghmai, A.A., Hayes, R.L., 1991. A controlled cortical impact model of traumatic brain injury in the rat. J. Neurosci. Methods 39, 253e262. Flierl, M.A., Stahel, P.F., Beauchamp, K.M., Morgan, S.J., Smith, W.R., Shohami, E., 2009. Mouse closed head injury model induced by a weight-drop device. Nat. Protoc. 4, 1328e1337. Foda, M.A., Marmarou, A., 1994. A new model of diffuse brain injury in rats. Part II: morphological characterization. J. Neurosurg. 80, 301e313. Gennarelli, T.A., 1994. Animate models of human head injury. J. Neurotrauma 11, 357e368. Goddard, D.R., Bunning, R.A., Woodroofe, M.N., 2001. Astrocyte and endothelial cell expression of ADAM 17 (TACE) in adult human CNS. Glia 34, 267e271. Gold, M.S., Kobeissy, F.H., Wang, K.K., Merlo, L.J., Bruijnzeel, A.W., Krasnova, I.N., Cadet, J.L., 2009. Methamphetamine- and trauma-induced brain injuries: comparative cellular and molecular neurobiological substrates. Biol. Psychiatry 66, 118e127. Grossetete, M., Phelps, J., Arko, L., Yonas, H., Rosenberg, G.A., 2009. Elevation of matrix metalloproteinases 3 and 9 in cerebrospinal fluid and blood in patients with severe traumatic brain injury. Neurosurgery 65, 702e708. Grossetete, M., Rosenberg, G.A., 2008. Matrix metalloproteinase inhibition facilitates cell death in intracerebral hemorrhage in mouse. J. Cereb. Blood Flow. Metab. 28, 752e763. Hadass, O., Tomlinson, B.N., Gooyit, M., Chen, S., Purdy, J.J., Walker, J.M., Zhang, C., Giritharan, A.B., Purnell, W., Robinson 2nd, C.R., Shin, D., Schroeder, V.A., Suckow, M.A., Simonyi, A., Sun, G.Y., Mobashery, S., Cui, J., Chang, M., Gu, Z., 2013. Selective inhibition of matrix metalloproteinase-9 attenuates secondary damage resulting from severe traumatic brain injury. PLoS One 8, e76904. Haddad, M., Beray-Berthat, V., Coqueran, B., Palmier, B., Szabo, C., Plotkine, M., Margaill, I., 2008. Reduction of hemorrhagic transformation by PJ34, a poly(ADP-ribose)polymerase inhibitor, after permanent focal cerebral ischemia

9

in mice. Eur. J. Pharmacol. 588, 52e57. Higashida, T., Kreipke, C.W., Rafols, J.A., Peng, C., Schafer, S., Schafer, P., Ding, J.Y., Dornbos 3rd, D., Li, X., Guthikonda, M., Rossi, N.F., Ding, Y., 2011. The role of hypoxia-inducible factor-1alpha, aquaporin-4, and matrix metalloproteinase-9 in blood-brain barrier disruption and brain edema after traumatic brain injury. J. Neurosurg. 114, 92e101. Kabadi, S.V., Stoica, B.A., Zimmer, D.B., Afanador, L., Duffy, K.B., Loane, D.J., Faden, A.I., 2015. S100B inhibition reduces behavioral and pathologic changes in experimental traumatic brain injury. J. Cereb. Blood Flow. Metab. Katakowski, M., Chen, J., Zhang, Z.G., Santra, M., Wang, Y., Chopp, M., 2007. Strokeinduced subventricular zone proliferation is promoted by tumor necrosis factor-alpha-converting enzyme protease activity. J. Cereb. Blood Flow. Metab. 27, 669e678. Kolar, M., Pachl, J., Tomasova, H., Haninec, P., 2008. Dynamics of matrixmetalloproteinase 9 after brain traumaeresults of a pilot study. Acta Neurochir. Suppl. 102, 373e376. Kumar, A., Loane, D.J., 2012. Neuroinflammation after traumatic brain injury: opportunities for therapeutic intervention. Brain Behav. Immun. 26, 1191e1201. Lee, P., Kim, J., Williams, R., Sandhir, R., Gregory, E., Brooks, W.M., Berman, N.E., 2012. Effects of aging on blood brain barrier and matrix metalloproteases following controlled cortical impact in mice. Exp. Neurol. 234, 50e61. Li, B., Mahmood, A., Lu, D., Wu, H., Xiong, Y., Qu, C., Chopp, M., 2009. Simvastatin attenuates microglial cells and astrocyte activation and decreases interleukin1beta level after traumatic brain injury. Neurosurgery 65, 179e185 discussion 185e176. Lighthall, J.W., 1988. Controlled cortical impact: a new experimental brain injury model. J. Neurotrauma 5, 1e15. Lo, E.H., Wang, X., Cuzner, M.L., 2002. Extracellular proteolysis in brain injury and inflammation: role for plasminogen activators and matrix metalloproteinases. J. Neurosci. Res. 69, 1e9. Loane, D.J., Stoica, B.A., Faden, A.I., 2015. Neuroprotection for traumatic brain injury. Handb. Clin. Neurol. 127, 343e366. Ludwig, A., Hundhausen, C., Lambert, M.H., Broadway, N., Andrews, R.C., Bickett, D.M., Leesnitzer, M.A., Becherer, J.D., 2005. Metalloproteinase inhibitors for the disintegrin-like metalloproteinases ADAM10 and ADAM17 that differentially block constitutive and phorbol ester-inducible shedding of cell surface molecules. Comb. Chem. High. Throughput Screen 8, 161e171. Marmarou, A., Foda, M.A., van den Brink, W., Campbell, J., Kita, H., Demetriadou, K., 1994. A new model of diffuse brain injury in rats. Part I: pathophysiology and biomechanics. J. Neurosurg. 80, 291e300. Peschon, J.J., Slack, J.L., Reddy, P., Stocking, K.L., Sunnarborg, S.W., Lee, D.C., Russell, W.E., Castner, B.J., Johnson, R.S., Fitzner, J.N., Boyce, R.W., Nelson, N., Kozlosky, C.J., Wolfson, M.F., Rauch, C.T., Cerretti, D.P., Paxton, R.J., March, C.J., Black, R.A., 1998. An essential role for ectodomain shedding in mammalian development. Science 282, 1281e1284. Pineda, J.A., Wang, K.K., Hayes, R.L., 2004. Biomarkers of proteolytic damage following traumatic brain injury. Brain Pathol. 14, 202e209. Postina, R., 2008. A closer look at alpha-secretase. Curr. Alzheimer Res. 5, 179e186. Raghupathi, R., 2004. Cell death mechanisms following traumatic brain injury. Brain Pathol. 14, 215e222. Reilly, P.L., 2001. Brain injury: the pathophysiology of the first hours.'Talk and Die revisited'. J. Clin. Neurosci. 8, 398e403. Rosenberg, G.A., 1995. Matrix metalloproteinases in brain injury. J. Neurotrauma 12, 833e842. Rosenberg, G.A., 2009. Matrix metalloproteinases and their multiple roles in neurodegenerative diseases. Lancet Neurol. 8, 205e216. Shlosberg, D., Benifla, M., Kaufer, D., Friedman, A., 2010. Blood-brain barrier breakdown as a therapeutic target in traumatic brain injury. Nat. Rev. Neurol. 6, 393e403. Siao, C.J., Tsirka, S.E., 2002. Extracellular proteases and neuronal cell death. Cell Mol. Biol. Noisy-le-grand 48, 151e161. Sifringer, M., Stefovska, V., Zentner, I., Hansen, B., Stepulak, A., Knaute, C., Marzahn, J., Ikonomidou, C., 2007. The role of matrix metalloproteinases in infant traumatic brain injury. Neurobiol. Dis. 25, 526e535. Skovronsky, D.M., Fath, S., Lee, V.M., Milla, M.E., 2001. Neuronal localization of the TNFalpha converting enzyme (TACE) in brain tissue and its correlation to amyloid plaques. J. Neurobiol. 49, 40e46. Smith, D.H., Soares, H.D., Pierce, J.S., Perlman, K.G., Saatman, K.E., Meaney, D.F., Dixon, C.E., McIntosh, T.K., 1995. A model of parasagittal controlled cortical impact in the mouse: cognitive and histopathologic effects. J. Neurotrauma 12, 169e178. Stoica, B.A., Loane, D.J., Zhao, Z., Kabadi, S.V., Hanscom, M., Byrnes, K.R., Faden, A.I., 2014. PARP-1 inhibition attenuates neuronal loss, microglia activation and neurological deficits after traumatic brain injury. J. Neurotrauma 31, 758e772. Vajtr, D., Benada, O., Kukacka, J., Prusa, R., Houstava, L., Toupalik, P., Kizek, R., 2009. Correlation of ultrastructural changes of endothelial cells and astrocytes occurring during blood brain barrier damage after traumatic brain injury with biochemical markers of BBB leakage and inflammatory response. Physiol. Res. 58, 263e268. Vilalta, A., Sahuquillo, J., Poca, M.A., De Los Rios, J., Cuadrado, E., Ortega-Aznar, A., Riveiro, M., Montaner, J., 2008a. Brain contusions induce a strong local overexpression of MMP-9. Results of a pilot study. Acta Neurochir. Suppl. 102, 415e419. Vilalta, A., Sahuquillo, J., Rosell, A., Poca, M.A., Riveiro, M., Montaner, J., 2008b. Moderate and severe traumatic brain injury induce early overexpression of

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013

10

S. Zhang et al. / Neurochemistry International xxx (2016) 1e10

systemic and brain gelatinases. Intensive Care Med. 34, 1384e1392. Walker, E.J., Rosenberg, G.A., 2009. TIMP-3 and MMP-3 contribute to delayed inflammation and hippocampal neuronal death following global ischemia. Exp. Neurol. 216, 122e131. Wang, X., Jung, J., Asahi, M., Chwang, W., Russo, L., Moskowitz, M.A., Dixon, C.E., Fini, M.E., Lo, E.H., 2000. Effects of matrix metalloproteinase-9 gene knock-out on morphological and motor outcomes after traumatic brain injury. J. Neurosci. 20, 7037e7042. Xue, M., Fan, Y., Liu, S., Zygun, D.A., Demchuk, A., Yong, V.W., 2009. Contributions of multiple proteases to neurotoxicity in a mouse model of intracerebral haemorrhage. Brain 132, 26e36. Yang, P., Baker, K.A., Hagg, T., 2006. The ADAMs family: coordinators of nervous system development, plasticity and repair. Prog. Neurobiol. 79, 73e94. Yang, Y., Candelario-Jalil, E., Thompson, J.F., Cuadrado, E., Estrada, E.Y., Rosell, A., Montaner, J., Rosenberg, G.A., 2010. Increased intranuclear matrix

metalloproteinase activity in neurons interferes with oxidative DNA repair in focal cerebral ischemia. J. Neurochem. 112, 134e149. Yong, V.W., 2005. Metalloproteinases: mediators of pathology and regeneration in the CNS. Nat. Rev. Neurosci. 6, 931e944. Yong, V.W., Agrawal, S.M., Stirling, D.P., 2007a. Targeting MMPs in acute and chronic neurological conditions. Neurotherapeutics 4, 580e589. Yong, V.W., Zabad, R.K., Agrawal, S., Goncalves Dasilva, A., Metz, L.M., 2007b. Elevation of matrix metalloproteinases (MMPs) in multiple sclerosis and impact of immunomodulators. J. Neurol. Sci. 259, 79e84. Zhang, S., Kojic, L., Tsang, M., Cynader, M.S., Jia, W., 2012. Distinct Patterns of Protease Activation in the Acute Stage of Cerebral Ischemia, Society for Neuroscience, 13e17 October, New Orleans, USA. Zhang, S., Kojic, L., Wen, Y., Qiang, D., Wong, K.H., Cynader, M.S., Jia, W., 2013. Treatment of Ischemic Stroke Using a Protease-inducible Gene Delivery System in Rat MCAO Model, Society for Neuroscience, 8e13 November, San Diego, USA.

Please cite this article in press as: Zhang, S., et al., Distinct roles for metalloproteinases during traumatic brain injury, Neurochemistry International (2016), http://dx.doi.org/10.1016/j.neuint.2016.02.013