Matrix metalloproteinases (MMPs) play important roles in normal brain development and synaptic plasticity, although aberrant expression of MMPs leads to brain damage, including blood–brain barrier disruption, inflammation, demyelination, and neuronal cell death. In this article, we report that MMP-8 is upregulated in LPS-stimulated BV2 microglial cells and primary cultured microglia, and treatment of MMP-8 inhibitor (M8I) or MMP-8 short hairpin RNA suppresses proinflammatory molecules, particularly TNF-α secretion. Subsequent experiments showed that MMP-8 exhibits TNF-α–converting enzyme (TACE) activity by cleaving the prodomain of TNF-α (A74/Q75, A76/V77 residues) and, furthermore, that M8I inhibits TACE activity more efficiently than TAPI-0, a general TACE inhibitor. Biochemical analysis of the underlying anti-inflammatory mechanisms of M8I revealed that it inhibits MAPK phosphorylation, NF-κB/AP-1 activity, and reactive oxygen species production. Further support for the proinflammatory role of microglial MMP-8 was obtained from an in vivo animal model of neuroinflammatory disorder. MMP-8 is upregulated in septic conditions, particularly in microglia. Administration of M8I or MMP-8 short hairpin RNA significantly inhibits microglial activation and expression/secretion of TNF-α in brain tissue, serum, and cerebrospinal fluid of LPS-induced septic mice. These results demonstrate that MMP-8 critically mediates microglial activation by modulating TNF-α activity, which may explain neuroinflammation in septic mouse brain.

Matrix metalloproteinases (MMPs) are zinc-dependent endopeptidases that degrade extracellular matrix proteins. MMPs are involved in development, plasticity, and regeneration of the CNS (1, 2). However, in various neuropathological conditions, such as Parkinson’s disease, Alzheimer’s disease, stroke, and multiple sclerosis (MS), MMPs are aberrantly expressed and cause breakdown of the blood–brain barrier (BBB), infiltration of peripheral immune cells, demyelination, and neuronal cell death (36). In early stages of brain injury, MMPs are concurrently and massively upregulated, inducing neuronal cell death. However, at later stages, MMPs are involved in neuronal regeneration and recovery (2, 7). Thus, inhibition of MMPs in early stages of brain injury may be a therapeutic strategy for neuronal disorders (7, 8).

Although a number of studies report the role of MMPs in normal and diseased brains, their role in microglial activation remains obscure. Our group recently reported that MMP-3 and MMP-9 play critical roles as inflammatory mediators in LPS- or α-synuclein–stimulated microglia (9, 10). We demonstrated that MMPs secreted from activated microglia bind to protease activated receptor-1 (PAR-1) and activate microglia in an autocrine or paracrine manner. We also showed that MMP-8 is induced in activated microglia, and its inhibition suppresses various proinflammatory molecules, suggesting a role for MMP-8 in microglial activation.

MMP-8, known as collagenase-2 or neutrophil collagenase, plays an important role in various inflammatory disorders, such as asthma, rheumatoid arthritis, pulmonary diseases, diabetes, and cancer (11). Considering neurologic disorders, the role of MMP-8 has been studied in bacterial meningitis and experimental autoimmune encephalomyelitis (EAE), an animal model of human MS. In bacterial meningitis, MMP-8 degrades occludin, a component of the BBB, and induces neuroinflammation (12). In EAE, MMP-8 modulates Th1/Th2 polarization, causing inflammation and demyelination (13), and knockout of MMP-8 or administration of an MMP-8 inhibitor attenuates BBB breakdown and EAE progression. A recent study also demonstrated an increase in MMP-8 in cerebrospinal fluid (CSF) in rats with spinal cord injury, suggesting that MMP-8 is a biomarker for subacute spinal cord injury (14).

Although several recent studies reported the involvement of MMP-8 in neurologic disorders, it remains unclear whether microglial MMP-8 mediates neuroinflammation. In this article, we demonstrate a proinflammatory role for MMP-8 in in vitro and in vivo neuroinflammatory conditions. Moreover, we identified modulation of TNF-α activation in activated microglia as one of the major mechanisms of MMP-8. These findings point toward MMP-8 as a promising target in the treatment of neuroinflammatory disorders involving microglial activation.

Male ICR (7-wk-old) and C57BL/6 (10–11-wk-old) mice were purchased from Orient Bio (Seoul, Korea), a branch of Charles River Laboratories. All animal experiments were approved by the Institutional Animal Care and Use Committee at Lee Gil Ya Cancer and Diabetes Institute, Gachon University and at the School of Medicine, Ewha Womans University.

LPS (Escherichia coli serotype 055:B5) was obtained from Sigma-Aldrich (St. Louis, MO). MMP-8 inhibitor (M8I) was purchased from Calbiochem (La Jolla, CA). Recombinant TNF-α converting enzyme (TACE) protein was supplied by R&D Systems (Minneapolis, MN). TAPI-0 and recombinant MMP-8 protein were purchased from Enzo Life Sciences (Lausen, Switzerland), and all enzymes and chemicals for semiquantitative RT-PCR (semi-PCR) were purchased from Promega (Madison, WI). All reagents for cell culture were obtained from Invitrogen (La Jolla, CA). All other reagents were obtained from Sigma-Aldrich, unless otherwise stated.

The immortalized mouse BV2 (15) or rat HAPI microglial cell lines (16) were grown and maintained in DMEM with 10% heat-inactivated FBS, streptomycin (10 μg/ml), and penicillin (10 U/ml) at 37°C. Primary microglial cells were cultured from the cerebral cortex of 1–2-d-old Sprague-Dawley rat pups, as previously described (17). Briefly, cortices were triturated into single cells and cultured in MEM containing 10% FBS. After 10–14 d, microglial cells were separated by shaking the culture plate for 30 min at 200 rpm. Cells were then plated on different culture plates and allowed to settle for 1 h. Nonadhesive cells were removed by washing with the same medium before experiments.

Total RNA (1 μg) isolated from BV2 cells (4.5 × 105 cells on a 6-cm dish) or brain tissue of septic mice was reverse transcribed, and synthesized cDNA was used as a template for PCR. Semi-PCR was performed on a T100 Thermal cycler (Bio-Rad) with Go Taq polymerase (Promega). Quantitative real-time PCR was performed on an Mx3005p (Stratagene) with Brilliant III Ultra-fast SYBR Green mix (Agilent). The primer sets shown in Table I were used to detect specific PCR products, which were calculated as fold change relative to control after normalization to a reference gene (GAPDH and β-actin for semi-PCR and quantitative real-time PCR, respectively).

Table I.
Primer sets used for semi-PCR and quantitative real-time PCR
SpeciesGeneForward Primer (5′→3′)Reverse Primer (5′→3′)Size (bp)
Mouse TNF-α CCTATGTCTCAGCCTCTTCT CCTGGTATGAGATAGCAAAT 354 
 iNOS CAAGAGTTTGACCAGAGGACC TGGAACCACTCGTACTTGGGA 450 
 IL-6 CCACTTCACAAGTCGGAGGCTT CCAGCTTATCTGTTAGGAGA 395 
 IL-10 GCCAGTACAGCCGGGAAGACAATA GCCTTGTAGACACCTTGGTCTT 409 
 MMP-8a CCAAGGAGTGTCCAAGCCAT CCTGCAGGAAAACTGCATCG 180 
 GAPDH ATGTACGTAGCCATCCAGGC AGGAAGGAAGGCTGGAAGAG 420 
 β-actinb AGCCTTCCTTCTTGGGTATG CTTCTGCATCCTGTCAGCAA 158 
Rat MMP-8 TACAACCTGTTTCTCGTGGCTGC TCAACTGTTCTCAGCTGGGGATG 317 
 GAPDH GTGCTGAGTATGTCGTGGAGTCT ACAGTCTTCTGAGTGGCAGTGA 292 
SpeciesGeneForward Primer (5′→3′)Reverse Primer (5′→3′)Size (bp)
Mouse TNF-α CCTATGTCTCAGCCTCTTCT CCTGGTATGAGATAGCAAAT 354 
 iNOS CAAGAGTTTGACCAGAGGACC TGGAACCACTCGTACTTGGGA 450 
 IL-6 CCACTTCACAAGTCGGAGGCTT CCAGCTTATCTGTTAGGAGA 395 
 IL-10 GCCAGTACAGCCGGGAAGACAATA GCCTTGTAGACACCTTGGTCTT 409 
 MMP-8a CCAAGGAGTGTCCAAGCCAT CCTGCAGGAAAACTGCATCG 180 
 GAPDH ATGTACGTAGCCATCCAGGC AGGAAGGAAGGCTGGAAGAG 420 
 β-actinb AGCCTTCCTTCTTGGGTATG CTTCTGCATCCTGTCAGCAA 158 
Rat MMP-8 TACAACCTGTTTCTCGTGGCTGC TCAACTGTTCTCAGCTGGGGATG 317 
 GAPDH GTGCTGAGTATGTCGTGGAGTCT ACAGTCTTCTGAGTGGCAGTGA 292 
a

Primers used for semi-PCR and quantitative real-time PCR.

b

Primers used for quantitative real-time PCR.

Cells (1 × 105 cells/well in a 48-well plate) were pretreated with M8I for 1 h and further stimulated with LPS (100 ng/ml) for 24 h. Alternatively, BV2 cells were transduced with MMP-8 short hairpin RNA (shRNA) or control shRNA, followed by LPS stimulation. Concentrations of TNF-α, IL-6, and IL-10 in conditioned medium (CM) were measured by ELISA using mAbs and procedures recommended by the supplier (Pharmingen, San Diego, CA). Accumulated nitrite in CM and intracellular accumulation of reactive oxygen species (ROS) were measured using Griess reagent (Promega) and H2DCF-DA (Invitrogen), respectively, as previously described (18).

Mouse MMP-8–specific shRNA (MMP-8 Mission shRNA, 1 × 106 TU/ml, pLKO.1 vector) and nontarget shRNA control (Mission pLKO.1 puro nontarget shRNA control) lentiviral transduction particles were obtained from Sigma-Aldrich. The sequence for MMP-8 shRNA is 5′-CCGGGCCTTGATGTACCCAAACTATCTCGAGATAGTTTGGGTACATCAAGGCTTTTTG-3′. BV2 cells were transduced with MMP-8 shRNA lentiviral particles and incubated in serum- and antibiotic-free media. After 24 h, the media were changed to media containing serum and antibiotics, and incubation continued for an additional 6–8 h. Then, the cells were treated with LPS, and the effects of MMP-8 shRNA on NO and TNF-α production were determined. Knockdown efficiency of shRNAs was confirmed by Western blots.

Proteins isolated from total cell lysates, CM, and brain tissue were separated by SDS-PAGE, transferred to nitrocellulose membranes, and incubated with Abs against MMP-8 (1:1000; Epitomics), IL-1β (1:200; R&D Systems), TNF-α (1:200; Abcam), MAPKs (1:1000; Cell Signaling), or β-actin (1:1000; Santa Cruz). Membranes were incubated with HRP-conjugated secondary Abs (1:1000; New England Biolabs, Beverly, MA or Santa Cruz), and protein bands were visualized using an ECL detection kit (Pierce Biotechnology or Santa Cruz). For quantification, density of specific target bands was normalized by comparing blots for actin using the ImageJ 1.37v program (National Institutes of Health, Bethesda, MD).

Nuclear extracts from cells were prepared as previously described (19). dsDNA oligonucleotides containing the NF-κB or AP-1 consensus sequences (Promega) were end-labeled using T4 polynucleotide kinase (New England Biolabs) in the presence of γ-[32P]-ATP. Nuclear proteins (5 μg) were incubated with 32P-labeled NF-κB or AP-1 probe on ice for 30 min and resolved on a 5% acrylamide gel.

TACE activity was assayed using the SensoLyte 520 TACE activity assay kit (AnaSpec). Recombinant protein (recombinant human [rh]TACE or rhMMP-8; 250 ng) with M8I or TAPI-0 was incubated with TACE substrate. TACE activity was determined by continuous detection of peptide cleavage in wells for 30–60 min using a fluorescence plate reader. TACE activity was expressed as the change in fluorescence intensity at excitation of 490 nm/emission of 520 nm.

The two ligands of TAPI-0 and M8I were docked into the binding pocket of TACE using the AutoDock Vina program (20). The X-ray structure of TACE (PDB code 2FV5) was used to define the binding site for molecular docking studies. Ligand files were downloaded from the PubChem database (21) and prepared in the form of a three-dimensional structure using UCSF Chimera software (22), which immediately converts the two-dimensional chemical structure into a three-dimensional structure. Energy minimization was conducted for the protein and ligand with the Amber force field (23, 24). For more accurate docking analysis of protein–ligands, another docking program, UCSF DOCK 6.5 (2528), was used with a flexible algorithm. Before the refined docking experiment, complete energy minimization was performed for the ligands and TACE in Amber force field (23, 24). The APBS program (29) and Coulombic surface coloring method with UCSF Chimera were used to calculate the electrostatic surface potentials of the TACE model. Electrostatic potential maps were generated by numerically solving the Poisson–Boltzmann equation based on molecular mechanics embedded in APBS.

A liquid chromatography–mass spectrometry (LC-MS)–based pro–TNF-α cleavage assay was performed to identify interactions between pro–TNF-α and TACE or MMP-8 using residues 71–82 (Ac-S71PLAQAVRSSSR82-NH2) (30). For the reaction, 2 μM pro–TNF-α was digested by 0.5 nM TACE or 1 nM MMP-8. The effects of TAPI-0 or M8I on TNF-α cleavage were also determined by digesting pro–TNF-α for 1 h with TACE (0.5 nM) or MMP-8 (1 nM) in the absence or presence of TAPI-0 (5 nM) or M8I (80 nM).

Systemic inflammation was induced by LPS administration (5 mg/kg, i.p.) (17) into male C57BL/6 or ICR mice. M8I (1 or 5 mg/kg, i.p.), dissolved in vehicle solution (1% DMSO and normal saline), was administered daily for 4 d before LPS treatment. Samples were obtained 3 or 6 h after LPS treatment.

Mice were anesthetized and perfused with PBS (pH 7.4), followed by ice-cold 4% paraformaldehyde. Brains were removed, incubated in fixative and 30% sucrose solution, embedded in Tissue-Tek O.C.T. compound, and frozen on powdered dry ice. Immunohistochemistry was performed on cryostat brain sections (20 μm) to detect MMP-8, microglia, astrocytes, and TNF-α by labeling sections with rabbit anti-MMP8 (1:100), goat anti-Iba1 (1:500; Abcam or 1:500; Wako), mouse anti-GFAP (1:500; Sigma-Aldrich), and rabbit anti–TNF-α (1:100; Abcam) primary Abs, respectively, followed by labeling with secondary biotinylated Abs and avidin/biotin complex (Vector Labs). Signals were visualized with DAB staining. Alternatively, conjugated FITC (1:1000; Serotec, Raleigh, NC) was used as a secondary Ab. MMP-8 expression in microglia or reactive astrocytes was determined by double-immunolabeling of MMP-8 (brown) and Iba-1 or GFAP (blue-gray). In brief, tissue sections that were processed for MMP-8 detection (brown color) were rinsed, blocked with avidin/biotin followed by 5% normal goat serum in 0.5% Triton X-100, and incubated with goat anti-Iba1 or mouse anti-GFAP Ab. Sections were incubated with secondary biotinylated Ab, followed by avidin/biotin complex, and the signal (blue-gray) was visualized using the SG Substrate Kit (Vector Lab). For quantification, three images were taken at 20× magnification from each section, and the mean number of immunopositive cells in each group was calculated from the three images.

Lentiviral particles (3 μl/3 min, intracerebroventricular injection) for mouse MMP-8–specific shRNA (1 × 106 TU/ml; Sigma-Aldrich) or nontarget shRNA control (Sigma-Aldrich) were administered into the right lateral ventricle using stereotaxic coordinates from a mouse brain atlas, as previously described (31). Sepsis induction was performed 4 d postinfection.

M8I (5 mg/kg, i.p.) was administered to ICR mice, and blood samples (30 μl) were collected from the saphenous vein at different time points (2.5, 5, 10, 15, 30, 60, 90, or 120 min) after administration. Plasma obtained by centrifugation of blood samples (2000 × g, 10 min, 4°C) was analyzed for M8I content using liquid chromatography–tandem mass spectrometry (Agilent LC 1100 series).

The capillary depletion method (32), modified for mice (33), was performed to determine the amount of M8I in the brain. At different time points (30, 60, or 120 min) after administration of M8I (5 mg/kg, i.p.), mice were sacrificed after blood samples were collected from the right carotid artery, and brains were removed. Brain tissue was homogenized (10 strokes) in 0.8 ml HEPES buffer (10 mM HEPES, 141 mM NaCl, 4 mM KCl, 2.8 mM CaCl2, 1 mM MgSO4, 1 mM NaH2PO4, and 10 mM d-glucose [pH 7.4]). After adding 1.6 ml 26% dextran solution to the homogenates, another homogenization step followed (three strokes). All of the homogenization processes were performed on ice in <1 min to minimize the possible dissociation of M8I from the capillary upon generation of the homogenates, as previously reported (32). An aliquot of 100 μl homogenate (whole homogenate) was sampled, and another 2.2 ml homogenate was centrifuged (5400 × g, 15 min, 4°C) and divided into a brain parenchymal fraction (supernatant) and capillary fraction (pellet). M8I concentration (Cbr [ng/g] for brain or Cpl [ng/ml] for plasma) was determined using liquid chromatography–tandem mass spectrometry, and pharmacokinetic parameters were determined by fitting plasma concentration–time data to the first-order equation using WinNonlin 2.1 (Pharsight, Mountain View, CA). M8I concentration was converted to the percentage of injected dose (%ID/g for brain, %ID/ml for plasma) after organ distribution values were corrected for corresponding vascular volumes using the equation

%ID/g=[Vd(t)V0]Cpl(t),

where Vd(t) = (ng/g brain)/(ng/ml plasma) at time t, V0 = plasma volume of the brain (assumed to be 9.3 μl/g), and Cpl(t) = plasma concentration (%ID/ml) at time t.

Unless otherwise stated, all experiments were performed with triplicate samples and repeated at least three times. Data are presented as mean ± SEM, and statistical comparisons among groups were performed using one-way ANOVA, followed by the Newman–Keuls post hoc test or t test. Statistical significance was accepted for p values < 0.05.

To determine whether MMP-8 is induced in activated microglia, cells were exposed to LPS and processed for semi-PCR using primers indicated (Table I) and Western blot analyses. Upregulation of MMP-8 mRNA and protein was observed in activated BV2 microglia (Fig. 1A, 1B). In particular, the more apparent upregulation of MMP-8 in CM compared with lysates from LPS-exposed cells (Fig. 1B) indicates a facilitation of MMP-8 secretion from activated microglia. This upregulation and increased secretion of MMP-8 were recapitulated in LPS-stimulated primary microglia (Fig. 1C, 1D).

FIGURE 1.

MMP-8 is upregulated in LPS-stimulated BV2 cells and rat primary microglia. Cells were incubated with LPS (100 ng/ml) for the indicated amount of time. (A and B) BV2 cell data. MMP-8 mRNA (A) or protein (B) expression was detected by semi-PCR in BV2 cells or Western blot in CM and cell lysates. A representative gel (A, upper panel) or blot (B, upper panel) and quantification (A and B, lower panels). (C and D) Rat primary microglia data. (C) Semi-PCR. (D) Western blot.

FIGURE 1.

MMP-8 is upregulated in LPS-stimulated BV2 cells and rat primary microglia. Cells were incubated with LPS (100 ng/ml) for the indicated amount of time. (A and B) BV2 cell data. MMP-8 mRNA (A) or protein (B) expression was detected by semi-PCR in BV2 cells or Western blot in CM and cell lysates. A representative gel (A, upper panel) or blot (B, upper panel) and quantification (A and B, lower panels). (C and D) Rat primary microglia data. (C) Semi-PCR. (D) Western blot.

Close modal

To investigate the functional roles of MMP-8 in microglial activation, changes in neuroinflammatory responses were assessed in LPS-stimulated BV2 cells using M8I, a specific MMP-8 inhibitor (34). We confirmed that M8I efficiently inhibited LPS-induced MMP-8 activity but not MMP-3 or MMP-9 activities in BV2 cells (Supplemental Fig. 1). Pretreatment with M8I significantly inhibited the production of proinflammatory molecules TNF-α, IL-6, and NO and upregulated the anti-inflammatory cytokine IL-10 (Fig. 2A). Among these molecules, the inhibition of TNF-α production was most prominent (IC50, 1 μM; Fig. 2A). These results suggest that MMP-8 acts as an inflammatory mediator, with a particular effect on TNF-α production in activated microglia. In support of this possibility, MMP-8 knockdown by MMP-8 shRNA (Fig. 2B) suppressed TNF-α and NO production in LPS-stimulated BV2 cells (Fig. 2C). These specific roles for MMP-8 were confirmed by M8I treatment in LPS-stimulated rat primary microglia (Fig. 2D) and HAPI microglial cell lines (Fig. 2E).

FIGURE 2.

Pharmacological inhibition or knockdown of MMP-8 reduces neuroinflammatory responses in LPS-stimulated microglial cells. BV2 cells were incubated with LPS (100 ng/ml) for 24 h in the absence or presence of M8I. (A) Effects of M8I on levels of TNF-α, nitrite, IL-6, and IL-10 were determined in CM from LPS (24 h)–stimulated BV2 microglia. (B) BV2 microglia were transduced with MMP-8 shRNA, and knockdown efficiency was determined by Western blot. (C) Effect of MMP-8 shRNA on TNF-α and nitrite levels was determined in LPS (24 h)–stimulated BV2 microglia. n = 4–6/group. *p < 0.05 versus control shRNA-transduced cells. Effect of M8I on levels of TNF-α or nitrite was determined in CM from LPS-stimulated rat primary microglia (D) and HAPI cells, a rat microglial cell line (E). (F) Effects of M8I on mRNA levels of inflammatory molecules were determined by semi-PCR in LPS-stimulated BV2 microglia. Representative gel showing specific PCR products (left panel) and the fold change compared with control (-) after normalization to GAPDH (right panel). n = 4/group. *p < 0.05 versus LPS-treated cells.

FIGURE 2.

Pharmacological inhibition or knockdown of MMP-8 reduces neuroinflammatory responses in LPS-stimulated microglial cells. BV2 cells were incubated with LPS (100 ng/ml) for 24 h in the absence or presence of M8I. (A) Effects of M8I on levels of TNF-α, nitrite, IL-6, and IL-10 were determined in CM from LPS (24 h)–stimulated BV2 microglia. (B) BV2 microglia were transduced with MMP-8 shRNA, and knockdown efficiency was determined by Western blot. (C) Effect of MMP-8 shRNA on TNF-α and nitrite levels was determined in LPS (24 h)–stimulated BV2 microglia. n = 4–6/group. *p < 0.05 versus control shRNA-transduced cells. Effect of M8I on levels of TNF-α or nitrite was determined in CM from LPS-stimulated rat primary microglia (D) and HAPI cells, a rat microglial cell line (E). (F) Effects of M8I on mRNA levels of inflammatory molecules were determined by semi-PCR in LPS-stimulated BV2 microglia. Representative gel showing specific PCR products (left panel) and the fold change compared with control (-) after normalization to GAPDH (right panel). n = 4/group. *p < 0.05 versus LPS-treated cells.

Close modal

To determine whether M8I effects are linked to transcriptional regulation, we examined changes in mRNA levels of inducible NO synthase (iNOS) and cytokines. M8I suppressed expression of iNOS and IL-6 mRNA and increased expression of IL-10 mRNA, but it did not alter expression of TNF-α mRNA (Fig. 2F). These results indicate that M8I modulates iNOS, IL-6, and IL-10 expression at the transcriptional level, but it modulates TNF-α expression at the posttranscriptional level.

Because M8I most efficiently inhibited TNF-α production, we further investigated the underlying mechanisms of this effect. Western blot showed that M8I suppressed the secretion of the active form of TNF-α, but it only slightly increased or did not alter pro–TNF-α levels in BV2 cells (Fig. 3A, 3B) and primary microglia (Fig. 3C, 3D). TNF-α ELISA also showed that M8I blocked the release of TNF-α in a concentration-dependent manner (Fig. 3E). These results demonstrate that reduced TNF-α productivity by MMP-8 inhibition (Figs. 2A–E, 3A–E) is a consequence of reduced TNF-α secretion.

FIGURE 3.

MMP-8 modulates TNF-α secretion from microglia via its TACE activity. BV2 cells or rat primary microglia were incubated with LPS (100 ng/ml) for 6 h in the absence or presence of M8I. Effect of M8I on protein levels of TNF-α (pro form, pTNF-α; active form, aTNF-α) was determined by Western blot in cell lysates and CM. Representative blots showing pro forms (26 kDa) and active forms (17 kDa) of TNF-α (A and C) and fold change relative to control (-) after normalization to β-actin (B and D). Note that M8I inhibited the active form of TNF-α in supernatants. n = 3/group. (E) ELISA data for aTNF-α. *p < 0.05 versus LPS-treated cells. (F) TACE activity of rhTACE and rhMMP-8 and its inhibition by M8I or TAPI-0 were determined using a SensoLyte 520 TACE Assay Kit. *p < 0.05 versus TACE activity of rhTACE or rhMMP-8.

FIGURE 3.

MMP-8 modulates TNF-α secretion from microglia via its TACE activity. BV2 cells or rat primary microglia were incubated with LPS (100 ng/ml) for 6 h in the absence or presence of M8I. Effect of M8I on protein levels of TNF-α (pro form, pTNF-α; active form, aTNF-α) was determined by Western blot in cell lysates and CM. Representative blots showing pro forms (26 kDa) and active forms (17 kDa) of TNF-α (A and C) and fold change relative to control (-) after normalization to β-actin (B and D). Note that M8I inhibited the active form of TNF-α in supernatants. n = 3/group. (E) ELISA data for aTNF-α. *p < 0.05 versus LPS-treated cells. (F) TACE activity of rhTACE and rhMMP-8 and its inhibition by M8I or TAPI-0 were determined using a SensoLyte 520 TACE Assay Kit. *p < 0.05 versus TACE activity of rhTACE or rhMMP-8.

Close modal

TNF-α is produced as a proform (26 kDa) and secreted as an active form (17 kDa) after cleavage of its prodomain by TACE (3). To further dissect the mechanism of the effect of M8I on TNF-α secretion, we examined whether M8I inhibits TACE activity using recombinant proteins. M8I inhibited TACE activity of rhTACE more efficiently than TAPI-0, a general TACE inhibitor (Fig. 3F). Interestingly, we found that rhMMP-8 itself has TACE activity (Fig. 3F). Taken together, these results demonstrate that MMP-8 itself has TACE activity and that M8I is a strong inhibitor of TACE activity.

The efficient inhibition of TACE activity by M8I was supported by molecular docking analysis. M8I bound to the catalytic domain of TACE like TAPI-0 with similar docking scores (Fig. 4). M8I interacted with TACE deep inside positively charged binding pockets, leading to a complex with greater structural stability (Fig. 4A–D). Moreover, two hydrogen bonds were found between M8I and TACE (190His.TACE-Hε2⋅⋅⋅O2-M8I, 200His.TACE-Hε2⋅⋅⋅O2-M8I) (Fig. 4E), which presumably contribute to the stabilization of the complex as a result of the strong binding force.

FIGURE 4.

Molecular docking analysis indicates that M8I binds to the catalytic domain of TACE. (A and B) All ligands in the binding pocket of TACE are presented using a line-stick model, including TAPI-0 (yellow) and M8I (green). (C and D) Electrostatic surface potential of TACE is shown in blue (positively charged region) or red (negatively charged region) in ± 10 scaled energy kcal/(mole ⋅ e) units. (C) TAPI-0 ligand in binding pocket. (D) Hydrogen bonds of M8I are shown as green dotted lines. There are two hydrogen bonds between two histidine residues of TACE and oxygen of M8I. (E) Docking scores of TACE–TAPI-0 and TACE–M8I models. The scoring function of AutoDock Vina was represented as affinity units (kcal/mol). Grid score, Grid VDW (van der Waals energy), and Grid ES (electrostatic energy) are based on the nonbonded terms of the molecular mechanic force field of DOCK 6.5 (2528).

FIGURE 4.

Molecular docking analysis indicates that M8I binds to the catalytic domain of TACE. (A and B) All ligands in the binding pocket of TACE are presented using a line-stick model, including TAPI-0 (yellow) and M8I (green). (C and D) Electrostatic surface potential of TACE is shown in blue (positively charged region) or red (negatively charged region) in ± 10 scaled energy kcal/(mole ⋅ e) units. (C) TAPI-0 ligand in binding pocket. (D) Hydrogen bonds of M8I are shown as green dotted lines. There are two hydrogen bonds between two histidine residues of TACE and oxygen of M8I. (E) Docking scores of TACE–TAPI-0 and TACE–M8I models. The scoring function of AutoDock Vina was represented as affinity units (kcal/mol). Grid score, Grid VDW (van der Waals energy), and Grid ES (electrostatic energy) are based on the nonbonded terms of the molecular mechanic force field of DOCK 6.5 (2528).

Close modal

To address whether active MMP-8 directly cleaves the prodomain of TNF-α, we performed a TNF-α cleavage assay using LC-MS analysis. Both rhTACE and rhMMP-8 cleaved the A76/V77 residue, a conventional cleavage site in the N-terminal propeptide of TNF-α (Fig. 5). Interestingly, an additional cleavage site (A74/Q75) was identified with rhMMP-8 (Fig. 5B–D). To confirm that the cleavage reaction was specifically induced by MMP-8, M8I was added to the reaction mixture. As expected, no meaningful products were produced (Fig. 5D). These results indicate that MMP-8 has TACE activity by cleaving two specific sequences of pro–TNF-α.

FIGURE 5.

TNF-α cleavage assay using LC-MS analysis. (AD) The 12-residue peptide covering pro–TNF-α cleavage sites was incubated with active rhMMP-8 or rhTACE and analyzed by LC-MS. Cleavage profiles of pro–TNF-α by TACE (A) and MMP-8 (B). (C) Summary of pro–TNF-α sites cleaved by TACE and MMP-8. (D) Summary of enzyme activities of TACE and MMP-8 by comparing cleaved sequences of pro–TNF-α based on LC-MS data.

FIGURE 5.

TNF-α cleavage assay using LC-MS analysis. (AD) The 12-residue peptide covering pro–TNF-α cleavage sites was incubated with active rhMMP-8 or rhTACE and analyzed by LC-MS. Cleavage profiles of pro–TNF-α by TACE (A) and MMP-8 (B). (C) Summary of pro–TNF-α sites cleaved by TACE and MMP-8. (D) Summary of enzyme activities of TACE and MMP-8 by comparing cleaved sequences of pro–TNF-α based on LC-MS data.

Close modal

Several signaling molecules are linked to microglial activation, including MAPKs (JNK, ERK1/2, and p38), NF-κB, AP-1, and ROS (3538). To determine whether MMP-8 modulates the activation or production of these molecules, biochemical analyses were performed using M8I in activated microglia. M8I inhibited the phosphorylation of three types of MAPKs (Fig. 6A) and the DNA-binding activities of NF-κB and AP-1 (Fig. 6B). Furthermore, M8I significantly reduced intracellular levels of ROS, which are early signaling inducers of microglial activation (39) (Fig. 6C, 6D). The results suggest that MMP-8 mediates LPS-induced inflammatory reactions via modulating upstream-signaling molecules.

FIGURE 6.

MMP-8 modulates MAPK signaling, NF-κB/AP-1 DNA-binding activities, and ROS production. (A) Effect of M8I on the phosphorylation of three types of MAPKs was determined by Western blot 30 min after LPS treatment. Representative blot (left panel) and quantification (right panel). (B) Effect of M8I on NF-κB or AP-1 DNA-binding activity was determined by EMSA 3 h after LPS treatment. DNA–NF-κB and DNA–AP-1 protein complexes are identified. The autoradiography of the gel is representative of three independent experiments (upper panel), and the quantification data (lower panel) are mean ± SEM of three independent experiments. (C and D) BV2 microglia were treated with M8I for 1 h and then stimulated with LPS for 16 h. Effect of M8I on intracellular ROS levels was measured by the DCF-DA method. (C) Values for fluorescence intensity are expressed as fold increase relative to control. (D) Representative image of DCF-derived fluorescence (green). Original magnification ×200. n = 3–4/group. *p < 0.05 versus LPS-treated cells.

FIGURE 6.

MMP-8 modulates MAPK signaling, NF-κB/AP-1 DNA-binding activities, and ROS production. (A) Effect of M8I on the phosphorylation of three types of MAPKs was determined by Western blot 30 min after LPS treatment. Representative blot (left panel) and quantification (right panel). (B) Effect of M8I on NF-κB or AP-1 DNA-binding activity was determined by EMSA 3 h after LPS treatment. DNA–NF-κB and DNA–AP-1 protein complexes are identified. The autoradiography of the gel is representative of three independent experiments (upper panel), and the quantification data (lower panel) are mean ± SEM of three independent experiments. (C and D) BV2 microglia were treated with M8I for 1 h and then stimulated with LPS for 16 h. Effect of M8I on intracellular ROS levels was measured by the DCF-DA method. (C) Values for fluorescence intensity are expressed as fold increase relative to control. (D) Representative image of DCF-derived fluorescence (green). Original magnification ×200. n = 3–4/group. *p < 0.05 versus LPS-treated cells.

Close modal

To identify whether MMP-8 is upregulated in the mouse brain with LPS-induced sepsis, a well-known neuroinflammatory disorder, mRNA and protein levels of MMP-8 were determined 3 h after LPS challenge. LPS induced upregulation of MMP-8 mRNA and protein (Fig. 7A, 7B) in the brain, with MMP-8 induction particularly prominent in microglia (Fig. 7B) and reactive astrocytes (Supplemental Fig. 2).

FIGURE 7.

MMP-8 is critical for neuroinflammation induced by systemic LPS administration. (A and B) Changes in MMP-8 mRNA and protein expression were determined in LPS (3 h)–induced septic brains. (A) MMP-8 mRNA expression, as determined by quantitative real-time PCR. Representative gel showing quantitative real-time PCR products (upper panel). Vehicle (veh), n = 3; LPS, n = 5. (B) Representative images of MMP-8+ cells and double-positive cells (MMP-8, brown; Iba-1, blue-gray), as determined by immunohistochemistry (upper panels). Arrows indicate double-immunopositive cells. Scale bars, 50 μm. Quantification of data (lower panel). n = 4/group. (CE) Effects of M8I on microglial activation and TNF-α production were determined in LPS (3 h)—induced septic mice. (C) Representative images for immunofluorescence labeling for Iba1 (green) (upper panels) Scale bar, 20 μm. Quantification of data in the cortex (lower panel). n = 3 per group. (D) TNF-α levels in serum and CSF, as measured by ELISA (n = 3/group). (E) Changes in TNF-α and IL-1β protein levels in the mouse cortex. Representative Western blots (upper panel) and quantification of data (lower panels) (n = 3/group). (F) Effects of MMP-8 shRNA lentivirus on microglial activation and TNF-α expression were determined by immunohistochemistry in LPS (3 h)–induced septic mice. Representative images (left panels) and quantification of data (right panels). Brown DAB staining. Scale bars, 50 μm. Control (con), n = 4/group; LPS (veh), n = 3 for Iba-1 and n = 5 for TNF-α; LPS+control shRNA, n = 3/group; LPS+MMP-8 shRNA, n = 4/group. *p < 0.05 versus control mice (A and B), versus LPS-treated mice (C–E), or versus LPS+control shRNA–treated mice (F).

FIGURE 7.

MMP-8 is critical for neuroinflammation induced by systemic LPS administration. (A and B) Changes in MMP-8 mRNA and protein expression were determined in LPS (3 h)–induced septic brains. (A) MMP-8 mRNA expression, as determined by quantitative real-time PCR. Representative gel showing quantitative real-time PCR products (upper panel). Vehicle (veh), n = 3; LPS, n = 5. (B) Representative images of MMP-8+ cells and double-positive cells (MMP-8, brown; Iba-1, blue-gray), as determined by immunohistochemistry (upper panels). Arrows indicate double-immunopositive cells. Scale bars, 50 μm. Quantification of data (lower panel). n = 4/group. (CE) Effects of M8I on microglial activation and TNF-α production were determined in LPS (3 h)—induced septic mice. (C) Representative images for immunofluorescence labeling for Iba1 (green) (upper panels) Scale bar, 20 μm. Quantification of data in the cortex (lower panel). n = 3 per group. (D) TNF-α levels in serum and CSF, as measured by ELISA (n = 3/group). (E) Changes in TNF-α and IL-1β protein levels in the mouse cortex. Representative Western blots (upper panel) and quantification of data (lower panels) (n = 3/group). (F) Effects of MMP-8 shRNA lentivirus on microglial activation and TNF-α expression were determined by immunohistochemistry in LPS (3 h)–induced septic mice. Representative images (left panels) and quantification of data (right panels). Brown DAB staining. Scale bars, 50 μm. Control (con), n = 4/group; LPS (veh), n = 3 for Iba-1 and n = 5 for TNF-α; LPS+control shRNA, n = 3/group; LPS+MMP-8 shRNA, n = 4/group. *p < 0.05 versus control mice (A and B), versus LPS-treated mice (C–E), or versus LPS+control shRNA–treated mice (F).

Close modal

Next, we determined whether M8I distributes within the brain. Concentration–time profiles of M8I content in plasma following its injection were obtained using pharmacokinetic analysis (Supplemental Fig. 3A, 3C). Starting from 139 ± 90.1 ng/ml at 2.5 min, a Cmax of 430 ± 82.6 ng/ml was reached at a Tmax of 10 min, which corresponds to 0.235% of injected dose per ml (%ID/ml). After this peak, the concentration slowly decreased to 0.005 %ID/ml at 120 min. The amount of M8I in the brain, expressed as volume of distribution (Vd, μl/ml) in whole-brain homogenates, initially increased and then decreased (Supplemental Fig. 3B). Vd in the supernatant and pellets varied accordingly, but Vd in the supernatant was higher than that in pellets at all time points with a relatively constant ratio, indicating rapid equilibrium between the parenchyma and capillaries. The percentage of injected dose of M8I/g of brain tissue (%ID/g) reached 0.0143 at 60 min. These findings suggest that a large quantity of M8I crossed the BBB and entered the brain during the initial distribution phase and then was eliminated from the brain at a rate slower than that of plasma.

To determine whether MMP-8 acts as observed in cultured cells, mice were injected with M8I for 4 d before LPS treatment, followed by assessment of microglial activation and TNF-α production 3 h after LPS treatment. Microglial activation was first assessed by immunoreactivity of Iba1, a marker of activated microglia (40). The number of Iba1+ cells with a densely stained round shape, indicative of activated cells, was increased in the inflamed cortex of LPS-treated mice compared with control mice. However, M8I (5 mg/kg) significantly reduced the number of Iba1+ cells (Fig. 7C). Next, TNF-α productivity was assessed by ELISA in serum and CSF. LPS markedly increased TNF-α levels in both serum and CSF, and this effect was significantly inhibited by M8I (Fig. 7D). Also, increases in TNF-α and IL-1β protein levels in the LPS-treated brain were significantly reduced by M8I (5 mg/kg) (Fig. 7E). These results demonstrate a necessary role of MMP-8 in modulating microglial activation and TNF-α production in the septic mouse brain.

This identified role of MMP-8 was confirmed by genetic knockdown of MMP-8 in the brain using a lentiviral system. Because MMP-8 is induced by LPS stimulation, we checked knockdown efficiency in the LPS-injected mouse brain. Intracerebroventricular injection of shRNA lentivirus reduced MMP-8 expression by ∼60%, as assessed by quantitative real-time PCR (Supplemental Fig. 4). This knockdown of MMP-8 significantly reduced microglial activation and TNF-α expression in the septic mouse brain (Fig. 7F).

To our knowledge, this is the first study to demonstrate the role of MMP-8 in neuroinflammatory conditions in vitro and in vivo. MMP-8 was increased in activated microglia, and specific inhibition of MMP-8 reduced inflammatory responses in cultured microglia, as well as septic mouse brains. Mechanistic studies using in vitro systems showed that MMP-8 exhibited TACE activity by cleaving the prodomain of TNF-α. M8I inhibited this TACE activity and also suppressed proinflammatory signaling molecules, such as MAPKs, NF-κB/AP-1, and ROS, in LPS-stimulated microglia. Thus, our results provide evidence that MMP-8 is an important proinflammatory mediator in activated microglia and that its modulation has therapeutic benefits for neuroinflammatory disorders.

MMP-8 mediates inflammatory processes and is expressed in many cell types at sites of inflammation, such as neutrophils, macrophages, plasma cells, granulocytes, and epithelial choroid plexus cells (41, 42). In this study, we identified microglia as an additional locus of MMP-8 expression. Microglia are major immune cells in the brain that play important roles in neuroinflammation, such as phagocytosis and the production of various cytokines and chemokines (4345). Microglia activated by LPS exposure robustly expressed MMP-8, and neuroinflammatory responses were prevented by MMP-8 genetic knockdown or pharmacological inhibition, supporting the direct involvement of MMP-8 in inflammatory responses in activated microglia. The identified role of MMP-8 in cultured microglia was reaffirmed in an animal model of sepsis, which is characterized by neuroinflammation. In septic conditions, MMP-8 was upregulated in microglia at inflammatory brain sites, and its inhibition reduced neuroinflammation. These results point toward microglia as cellular targets in the treatment of CNS disorders, such as MS, spinal cord injury, and traumatic brain injury, all of which involve upregulation of MMP-8 as a pathological feature (13, 14, 46).

The in vivo effects of M8I and MMP-8 shRNA might not be confined to microglia. Rather, other cell types, such as infiltrated neutrophils and monocytes and reactive astrocytes, could also be involved. In particular, reactive astrocytes are observed in LPS-treated cortices and express MMP-8. However, we detected no clear differences in the presence of infiltrating neutrophils and monocytes in our in vivo model (data not shown). Although MMP-8–producing neutrophils and monocytes enter the CNS in response to neuroinflammatory stimuli, it seems that the infiltration of these cell types occurs rarely at early time points (i.e., 3 h) after LPS treatment. This notion is in line with several reports showing that microglial activation is a main event at early stages of LPS-induced neuroinflammation (4749). Further studies are necessary to identify the specific cell types that express MMP-8 and are important for neuroinflammation in the LPS-treated brain.

The site of action of M8I seems to be inside of the CNS. In this study, we demonstrated that M8I is distributed into the brain parenchyma, which indicates a significant transport of M8I across the BBB into the brain. It might be technically possible that M8I remains associated with the capillaries upon generation of the homogenates because mice were not perfused prior to homogenization. However, this may not affect the observed BBB permeability of M8I in this study. We did not perform the vascular washout to correct the contamination of the parenchymal fraction during brain homogenization. Instead, we did all homogenization processes on ice in < 1 min to minimize the dissociation of capillary-associated M8I as reported previously, where no measurable dissociation was detectable even without vascular washing out (32).

Our findings indicate that the neuroinflammatory actions of MMP-8 in microglia are due, in part, to its direct action on the proinflammatory cytokine TNF-α. Many cell types that contribute to inflammation, such as microglia, produce and secrete TNF-α, which, in turn, mediates tissue damage in various disorders, such as rheumatoid arthritis, ulcerative colitis, diabetes, atherosclerosis, Alzheimer’s disease, MS, and stroke (50, 51). An important regulator of TNF-α activity is TACE (3), which cleaves the 26-kDa membrane-bound protein (pro–TNF-α) to form a 17-kDa soluble cytokine. In this study, MMP-8 was revealed to have slightly different TACE activity: MMP-8 cleaves a prodomain of TNF-α at the A74/Q75 residue, in addition to the A76/V77 residue, a conventional cleavage site of TACE. Notably, MMP-8 inhibition primarily affected the secretion of TNF-α via inhibition of TACE activity, which may be an important consideration for therapeutic strategies. The development of agents that can control TNF-α and, hence, TACE (3) has been a major interest of pharmaceutical companies (52, 53). Therefore, the dual inhibition of MMP-8 and TACE activity by M8I may contribute to its strong anti-inflammatory and neuroprotective effects.

We found that M8I suppressed intracellular signaling molecules, such as MAPKs, NF-κB/AP-1, and ROS, which are largely involved in proinflammatory gene expression. Previous studies reported that soluble TNF-α binds to cell surface receptors on immune cells and evokes MAPK signaling and NF-κB/AP-1 activation, leading to cytokine production and inflammatory responses in an autocrine and/or paracrine fashion (3, 54, 55). Thus, the inhibition of soluble TNF-α by M8I may lead to the blockade of inflammatory signaling. Another possible mechanism of M8I is PAR-1 inhibition. Our previous study showed that MMP-8 activates PAR-1 on the cell surface of microglia and induces inflammatory reactions (9). Thus, the concurrent inhibition of TNF-α and PAR-1 by M8I may block proinflammatory signaling in activated microglia.

Several studies reported that MMP-8 is increased in the serum of sepsis patients and experimental animals and that inhibition of MMP-8 improves outcome, suggesting that MMP-8 is an important diagnostic marker for sepsis (34, 56). Furthermore, a recent report highlighted the functional importance of MMP-8 in neuroinflammation using genetic nulls for MMP-8 in a sepsis model, showing that MMP-8 in epithelial choroid plexus cells causes disruption of the blood–CSF barrier and results in brain inflammation (42). In this study, we identified additional meaningful events in sepsis, including a new locus for MMP-8 upregulation and mechanisms underlying the modulation of neuroinflammation. As observed in cultured microglia, MMP-8 was significantly upregulated in septic conditions, especially in microglia, and its pharmacological or genetic inhibition prevented microglial activation and downregulated TNF-α levels in serum, CSF, and brain tissue. Moreover, the specific inhibition of MMP-8 suppressed microglial activation and TNF-α expression in activated microglia. These results suggest that MMP-8 is a valuable diagnostic marker and therapeutic target for sepsis-related neuroinflammation. Furthermore, in light of the current findings, it is tempting to speculate that MMP-8 might be a therapeutic target for various neuroinflammatory disorders that are accompanied by microglial activation.

This work was supported by grants from the National Research Foundation funded by the Korean government (2012R1A2A2A01045821 and 2012R1A5A2A32671866 to H.-S.K. and 2012R1A2A2A02012332 to J.W.C.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

     
  • BBB

    blood–brain barrier

  •  
  • CM

    conditioned medium

  •  
  • CSF

    cerebrospinal fluid

  •  
  • EAE

    experimental autoimmune encephalomyelitis

  •  
  • iNOS

    inducible NO synthase

  •  
  • LC-MS

    liquid chromatography–mass spectrometry

  •  
  • M8I

    MMP-8 inhibitor

  •  
  • MMP

    matrix metalloproteinase

  •  
  • MS

    multiple sclerosis

  •  
  • PAR-1

    protease-activated receptor-1

  •  
  • rh

    recombinant human

  •  
  • ROS

    reactive oxygen species

  •  
  • semi-PCR

    semiquantitative RT-PCR

  •  
  • shRNA

    short hairpin RNA

  •  
  • TACE

    TNF-α converting enzyme.

1
Agrawal
S. M.
,
Lau
L.
,
Yong
V. W.
.
2008
.
MMPs in the central nervous system: where the good guys go bad.
Semin. Cell Dev. Biol.
19
:
42
51
.
2
Verslegers
M.
,
Lemmens
K.
,
Van Hove
I.
,
Moons
L.
.
2013
.
Matrix metalloproteinase-2 and -9 as promising benefactors in development, plasticity and repair of the nervous system.
Prog. Neurobiol.
105
:
60
78
.
3
Bahia
M. S.
,
Silakari
O.
.
2010
.
Tumor necrosis factor alpha converting enzyme: an encouraging target for various inflammatory disorders.
Chem. Biol. Drug Des.
75
:
415
443
.
4
Hansmann
F.
,
Herder
V.
,
Kalkuhl
A.
,
Haist
V.
,
Zhang
N.
,
Schaudien
D.
,
Deschl
U.
,
Baumgärtner
W.
,
Ulrich
R.
.
2012
.
Matrix metalloproteinase-12 deficiency ameliorates the clinical course and demyelination in Theiler’s murine encephalomyelitis.
Acta Neuropathol.
124
:
127
142
.
5
Morancho
A.
,
Rosell
A.
,
García-Bonilla
L.
,
Montaner
J.
.
2010
.
Metalloproteinase and stroke infarct size: role for anti-inflammatory treatment?
Ann. N. Y. Acad. Sci.
1207
:
123
133
.
6
Rosenberg
G. A.
2009
.
Matrix metalloproteinases and their multiple roles in neurodegenerative diseases.
Lancet Neurol.
8
:
205
216
.
7
Zhang
H.
,
Chang
M.
,
Hansen
C. N.
,
Basso
D. M.
,
Noble-Haeusslein
L. J.
.
2011
.
Role of matrix metalloproteinases and therapeutic benefits of their inhibition in spinal cord injury.
Neurotherapeutics
8
:
206
220
.
8
Yang
Y.
,
Thompson
J. F.
,
Taheri
S.
,
Salayandia
V. M.
,
McAvoy
T. A.
,
Hill
J. W.
,
Yang
Y.
,
Estrada
E. Y.
,
Rosenberg
G. A.
.
2013
.
Early inhibition of MMP activity in ischemic rat brain promotes expression of tight junction proteins and angiogenesis during recovery.
J. Cereb. Blood Flow Metab.
33
:
1104
1114
.
9
Lee
E. J.
,
Woo
M. S.
,
Moon
P. G.
,
Baek
M. C.
,
Choi
I. Y.
,
Kim
W. K.
,
Junn
E.
,
Kim
H. S.
.
2010
.
α-synuclein activates microglia by inducing the expressions of matrix metalloproteinases and the subsequent activation of protease-activated receptor-1.
J. Immunol.
185
:
615
623
.
10
Woo
M. S.
,
Park
J. S.
,
Choi
I. Y.
,
Kim
W. K.
,
Kim
H. S.
.
2008
.
Inhibition of MMP-3 or -9 suppresses lipopolysaccharide-induced expression of proinflammatory cytokines and iNOS in microglia.
J. Neurochem.
106
:
770
780
.
11
Dejonckheere
E.
,
Vandenbroucke
R. E.
,
Libert
C.
.
2011
.
Matrix metalloproteinase8 has a central role in inflammatory disorders and cancer progression.
Cytokine Growth Factor Rev.
22
:
73
81
.
12
Schubert-Unkmeir
A.
,
Konrad
C.
,
Slanina
H.
,
Czapek
F.
,
Hebling
S.
,
Frosch
M.
.
2010
.
Neisseria meningitidis induces brain microvascular endothelial cell detachment from the matrix and cleavage of occludin: a role for MMP-8.
PLoS Pathog.
6
:
e1000874
.
13
Folgueras
A. R.
,
Fueyo
A.
,
García-Suárez
O.
,
Cox
J.
,
Astudillo
A.
,
Tortorella
P.
,
Campestre
C.
,
Gutiérrez-Fernández
A.
,
Fanjul-Fernández
M.
,
Pennington
C. J.
, et al
.
2008
.
Collagenase-2 deficiency or inhibition impairs experimental autoimmune encephalomyelitis in mice.
J. Biol. Chem.
283
:
9465
9474
.
14
Light
M.
,
Minor
K. H.
,
DeWitt
P.
,
Jasper
K. H.
,
Davies
S. J.
.
2012
.
Multiplex array proteomics detects increased MMP-8 in CSF after spinal cord injury.
J. Neuroinflammation
9
:
122
.
15
Bocchini
V.
,
Mazzolla
R.
,
Barluzzi
R.
,
Blasi
E.
,
Sick
P.
,
Kettenmann
H.
.
1992
.
An immortalized cell line expresses properties of activated microglial cells.
J. Neurosci. Res.
31
:
616
621
.
16
Cheepsunthorn
P.
,
Radov
L.
,
Menzies
S.
,
Reid
J.
,
Connor
J. R.
.
2001
.
Characterization of a novel brain-derived microglial cell line isolated from neonatal rat brain.
Glia
35
:
53
62
.
17
Park
J. S.
,
Park
E. M.
,
Kim
D. H.
,
Jung
K.
,
Jung
J. S.
,
Lee
E. J.
,
Hyun
J. W.
,
Kang
J. L.
,
Kim
H. S.
.
2009
.
Anti-inflammatory mechanism of ginseng saponins in activated microglia.
J. Neuroimmunol.
209
:
40
49
.
18
Lee
K. M.
,
Kang
H. S.
,
Yun
C. H.
,
Kwak
H. S.
.
2012
.
Potential in vitro protective effect of quercetin, catechin, caffeic acid and phytic acid against ethanol-induced oxidative stress in SK-Hep-1 cells.
Biomol. Ther. (Seoul)
20
:
492
498
.
19
Woo
M. S.
,
Jang
P. G.
,
Park
J. S.
,
Kim
W. K.
,
Joh
T. H.
,
Kim
H. S.
.
2003
.
Selective modulation of lipopolysaccharide-stimulated cytokine expression and mitogen-activated protein kinase pathways by dibutyryl-cAMP in BV2 microglial cells.
Brain Res. Mol. Brain Res.
113
:
86
96
.
20
Trott
O.
,
Olson
A. J.
.
2010
.
AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading.
J. Comput. Chem.
31
:
455
461
.
21
Bolton, E. E., Y. Wang, P. A. Thiessen, and S. H. Bryant. 2008. Chapter 12 PubChem: Integrated Platform of Small Molecules and Biological Activities. Annu. Rep. Comput. Chem. 4: 217–241
.
22
Pettersen
E. F.
,
Goddard
T. D.
,
Huang
C. C.
,
Couch
G. S.
,
Greenblatt
D. M.
,
Meng
E. C.
,
Ferrin
T. E.
.
2004
.
UCSF Chimera—a visualization system for exploratory research and analysis.
J. Comput. Chem.
25
:
1605
1612
.
23
Case
D. A.
,
Cheatham
T. E.
 III,
,
Darden
T.
,
Gohlke
H.
,
Luo
R.
,
Merz
K. M.
 Jr.
,
Onufriev
A.
,
Simmerling
C.
,
Wang
B.
,
Woods
R. J.
.
2005
.
The Amber biomolecular simulation programs.
J. Comput. Chem.
26
:
1668
1688
.
24
Pearlman
D. A.
,
Case
D. A.
,
Caldwell
J. W.
,
Ross
W. S.
,
Cheatham
T. E.
,
DeBolt
S.
,
Ferguson
D.
,
Seibel
G.
,
Kollman
P.
.
1995
.
AMBER, a package of computer programs for applying molecular mechanics, normal mode analysis, molecular dynamics and free energy calculations to simulate the structural and energetic properties of molecules.
Comput. Phys. Commun.
91
:
1
41
.
25
Ewing
T. J.
,
Makino
S.
,
Skillman
A. G.
,
Kuntz
I. D.
.
2001
.
DOCK 4.0: search strategies for automated molecular docking of flexible molecule databases.
J. Comput. Aided Mol. Des.
15
:
411
428
.
26
Kuntz
I. D.
,
Blaney
J. M.
,
Oatley
S. J.
,
Langridge
R.
,
Ferrin
T. E.
.
1982
.
A geometric approach to macromolecule-ligand interactions.
J. Mol. Biol.
161
:
269
288
.
27
Lang
P. T.
,
Brozell
S. R.
,
Mukherjee
S.
,
Pettersen
E. F.
,
Meng
E. C.
,
Thomas
V.
,
Rizzo
R. C.
,
Case
D. A.
,
James
T. L.
,
Kuntz
I. D.
.
2009
.
DOCK 6: combining techniques to model RNA-small molecule complexes.
RNA
15
:
1219
1230
.
28
Moustakas
D. T.
,
Lang
P. T.
,
Pegg
S.
,
Pettersen
E.
,
Kuntz
I. D.
,
Brooijmans
N.
,
Rizzo
R. C.
.
2006
.
Development and validation of a modular, extensible docking program: DOCK 5.
J. Comput. Aided Mol. Des.
20
:
601
619
.
29
Baker
N. A.
,
Sept
D.
,
Joseph
S.
,
Holst
M. J.
,
McCammon
J. A.
.
2001
.
Electrostatics of nanosystems: application to microtubules and the ribosome.
Proc. Natl. Acad. Sci. USA
98
:
10037
10041
.
30
Minond
D.
,
Cudic
M.
,
Bionda
N.
,
Giulianotti
M.
,
Maida
L.
,
Houghten
R. A.
,
Fields
G. B.
.
2012
.
Discovery of novel inhibitors of a disintegrin and metalloprotease 17 (ADAM17) using glycosylated and non-glycosylated substrates.
J. Biol. Chem.
287
:
36473
36487
.
31
Lee
Y. W.
,
Kim
D. H.
,
Jeon
S. J.
,
Park
S. J.
,
Kim
J. M.
,
Jung
J. M.
,
Lee
H. E.
,
Bae
S. G.
,
Oh
H. K.
,
Son
K. H.
,
Ryu
J. H.
.
2013
.
Neuroprotective effects of salvianolic acid B on an Aβ25-35 peptide-induced mouse model of Alzheimer’s disease
[Published erratum appears in 2013 Eur. J. Pharmacol. 714: 345]
.
Eur. J. Pharmacol.
704
:
70
77
.
32
Triguero
D.
,
Buciak
J.
,
Pardridge
W. M.
.
1990
.
Capillary depletion method for quantification of blood-brain barrier transport of circulating peptides and plasma proteins.
J. Neurochem.
54
:
1882
1888
.
33
Gutierrez
E. G.
,
Banks
W. A.
,
Kastin
A. J.
.
1993
.
Murine tumor necrosis factor alpha is transported from blood to brain in the mouse.
J. Neuroimmunol.
47
:
169
176
.
34
Solan
P. D.
,
Dunsmore
K. E.
,
Denenberg
A. G.
,
Odoms
K.
,
Zingarelli
B.
,
Wong
H. R.
.
2012
.
A novel role for matrix metalloproteinase-8 in sepsis.
Crit. Care Med.
40
:
379
387
.
35
Gao
H. M.
,
Zhou
H.
,
Hong
J. S.
.
2012
.
NADPH oxidases: novel therapeutic targets for neurodegenerative diseases.
Trends Pharmacol. Sci.
33
:
295
303
.
36
Kaminska
B.
2005
.
MAPK signalling pathways as molecular targets for anti-inflammatory therapy—from molecular mechanisms to therapeutic benefits.
Biochim. Biophys. Acta
1754
:
253
262
.
37
Kaminska
B.
,
Gozdz
A.
,
Zawadzka
M.
,
Ellert-Miklaszewska
A.
,
Lipko
M.
.
2009
.
MAPK signal transduction underlying brain inflammation and gliosis as therapeutic target.
Anat. Rec. (Hoboken)
292
:
1902
1913
.
38
Smale
S. T.
2010
.
Selective transcription in response to an inflammatory stimulus.
Cell
140
:
833
844
.
39
Bedard
K.
,
Krause
K. H.
.
2007
.
The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology.
Physiol. Rev.
87
:
245
313
.
40
Ito
D.
,
Tanaka
K.
,
Suzuki
S.
,
Dembo
T.
,
Fukuuchi
Y.
.
2001
.
Enhanced expression of Iba1, ionized calcium-binding adapter molecule 1, after transient focal cerebral ischemia in rat brain.
Stroke
32
:
1208
1215
.
41
Van Lint
P.
,
Libert
C.
.
2006
.
Matrix metalloproteinase-8: cleavage can be decisive.
Cytokine Growth Factor Rev.
17
:
217
223
.
42
Vandenbroucke
R. E.
,
Dejonckheere
E.
,
Van Lint
P.
,
Demeestere
D.
,
Van Wonterghem
E.
,
Vanlaere
I.
,
Puimège
L.
,
Van Hauwermeiren
F.
,
De Rycke
R.
,
Mc Guire
C.
, et al
.
2012
.
Matrix metalloprotease 8-dependent extracellular matrix cleavage at the blood-CSF barrier contributes to lethality during systemic inflammatory diseases.
J. Neurosci.
32
:
9805
9816
.
43
Glass
C. K.
,
Saijo
K.
,
Winner
B.
,
Marchetto
M. C.
,
Gage
F. H.
.
2010
.
Mechanisms underlying inflammation in neurodegeneration.
Cell
140
:
918
934
.
44
Graeber
M. B.
,
Streit
W. J.
.
2010
.
Microglia: biology and pathology.
Acta Neuropathol.
119
:
89
105
.
45
McGeer
P. L.
,
McGeer
E. G.
.
2013
.
The amyloid cascade-inflammatory hypothesis of Alzheimer disease: implications for therapy.
Acta Neuropathol.
126
:
479
497
.
46
Roberts
D. J.
,
Jenne
C. N.
,
Léger
C.
,
Kramer
A. H.
,
Gallagher
C. N.
,
Todd
S.
,
Parney
I. F.
,
Doig
C. J.
,
Yong
V. W.
,
Kubes
P.
,
Zygun
D. A.
.
2013
.
A prospective evaluation of the temporal matrix metalloproteinase response after severe traumatic brain injury in humans.
J. Neurotrauma
30
:
1717
1726
.
47
Chen
Z.
,
Jalabi
W.
,
Shpargel
K. B.
,
Farabaugh
K. T.
,
Dutta
R.
,
Yin
X.
,
Kidd
G. J.
,
Bergmann
C. C.
,
Stohlman
S. A.
,
Trapp
B. D.
.
2012
.
Lipopolysaccharide-induced microglial activation and neuroprotection against experimental brain injury is independent of hematogenous TLR4.
J. Neurosci.
32
:
11706
11715
.
48
Jeong
H. K.
,
Jou
I.
,
Joe
E. H.
.
2010
.
Systemic LPS administration induces brain inflammation but not dopaminergic neuronal death in the substantia nigra.
Exp. Mol. Med.
42
:
823
832
.
49
Ji
K. A.
,
Yang
M. S.
,
Jeong
H. K.
,
Min
K. J.
,
Kang
S. H.
,
Jou
I.
,
Joe
E. H.
.
2007
.
Resident microglia die and infiltrated neutrophils and monocytes become major inflammatory cells in lipopolysaccharide-injected brain.
Glia
55
:
1577
1588
.
50
Aggarwal
B. B.
2003
.
Signalling pathways of the TNF superfamily: a double-edged sword.
Nat. Rev. Immunol.
3
:
745
756
.
51
McCoy
M. K.
,
Tansey
M. G.
.
2008
.
TNF signaling inhibition in the CNS: implications for normal brain function and neurodegenerative disease.
J. Neuroinflammation
5
:
45
.
52
DasGupta
S.
,
Murumkar
P. R.
,
Giridhar
R.
,
Yadav
M. R.
.
2009
.
Current perspective of TACE inhibitors: a review.
Bioorg. Med. Chem.
17
:
444
459
.
53
Moss
M. L.
,
Sklair-Tavron
L.
,
Nudelman
R.
.
2008
.
Drug insight: tumor necrosis factor-converting enzyme as a pharmaceutical target for rheumatoid arthritis.
Nat. Clin. Pract. Rheumatol.
4
:
300
309
.
54
Harms
A. S.
,
Lee
J. K.
,
Nguyen
T. A.
,
Chang
J.
,
Ruhn
K. M.
,
Treviño
I.
,
Tansey
M. G.
.
2012
.
Regulation of microglia effector functions by tumor necrosis factor signaling.
Glia
60
:
189
202
.
55
Sriram
K.
,
O’Callaghan
J. P.
.
2007
.
Divergent roles for tumor necrosis factor-α in the brain.
J. Neuroimmune Pharmacol.
2
:
140
153
.
56
Lauhio
A.
,
Hästbacka
J.
,
Pettilä
V.
,
Tervahartiala
T.
,
Karlsson
S.
,
Varpula
T.
,
Varpula
M.
,
Ruokonen
E.
,
Sorsa
T.
,
Kolho
E.
.
2011
.
Serum MMP-8, -9 and TIMP-1 in sepsis: high serum levels of MMP-8 and TIMP-1 are associated with fatal outcome in a multicentre, prospective cohort study. Hypothetical impact of tetracyclines.
Pharmacol. Res.
64
:
590
594
.

The authors have no financial conflicts of interest.

Supplementary data