ADAMTS-9 expression is up-regulated following transient middle cerebral artery occlusion (tMCAo) in the rat

Neuroscience Letters 452 (2009) 252–257

Contents lists available at ScienceDirect

Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet

ADAMTS-9 expression is up-regulated following transient middle cerebral artery occlusion (tMCAo) in the rat Martin J. Reid a,∗ , Alison K. Cross a , Gail Haddock a , Stuart M. Allan c , Chris J. Stock c , M. Nicola Woodroofe a , David J. Buttle b , Rowena A.D. Bunning a a b c

Biomedical Research Centre, Sheffield Hallam University, Howard Street, Sheffield, S1 1WB, UK Academic Unit of Molecular Medicine/Rheumatology, E Floor, Medical School, University of Sheffield,Sheffield, S10 2RX, UK Faculty of Life Sciences, The University of Manchester, Michael Smith Building, Manchester, M13 9PT, UK

a r t i c l e

i n f o

Article history: Received 30 October 2008 Received in revised form 7 January 2009 Accepted 23 January 2009 Keywords: ADAMTS-9 Stroke Cerebral ischaemia tMCAo

a b s t r a c t The ADAMTS enzymes (a disintegrin and metalloproteinase with thrombospondin type 1-like motifs) have important roles in central nervous system (CNS) physiology and pathology. This current study aimed to analyse the expression of ADAMTS-9 following transient middle cerebral artery occlusion (tMCAo) in the rat, a model of focal cerebral ischaemia. Using real-time RT-PCR, ADAMTS-9 mRNA was demonstrated to be significantly up-regulated in tMCAo brain tissue compared to sham-operated at 24 h post-ischaemia. The mature form of the ADAMTS-9 protein was only detected by Western blotting in brains subjected to tMCAo at 24 h. In situ hybridisation demonstrated that ADAMTS-9 mRNA was expressed by neurones in tMCAo tissue. This study indicates that ADAMTS-9 expression is modulated in response to cerebral ischaemia in vivo and further research will resolve whether it plays a role in the subsequent degenerative or repair processes. © 2009 Elsevier Ireland Ltd. All rights reserved.

The ADAM (a disintegrin and metalloproteinase) and ADAMTS (a disintegrin and metalloproteinase with thrombospondin type 1like motifs) families of metalloproteinases are important candidates for contributing to proteolysis in central nervous system (CNS) disorders because of their substrate specificities and expression profiles [4,11,22]. The catalytic domains of ADAMs and ADAMTSs are structurally related to those of the matrix metalloproteinases (MMPs), which have been strongly implicated in post-cerebral ischaemic damage and/or repair, particularly MMP-2 and -9 [3,16,19]. ADAMTS-1, -4, -5, -9 and -15 are within the glutamyl endopeptidase subgroup of ADAMTSs [14,23]. The main substrates of these peptidases are the aggregating chondroitin sulphate proteoglycans (CSPGs), including aggrecan, versican and brevican, which are integral components of the CNS extracellular matrix (ECM) [8,30]. In normal physiological conditions, the ECM of the brain is a dynamic structure, which undergoes constant remodelling by processing of CSPGs [1,24]. Following neuronal injury, reactive gliosis occurs leading to the formation of glial scars around the lesion, comprised of reactive astrocytes

∗ Corresponding author at: University of Cambridge, Department of Medicine, Addenbrooke’s Hospital, Hills Road, Cambridge, CB2 2QQ. Tel.: +44 1223336742; fax: +44 1223336846. E-mail address: [email protected] (M.J. Reid). 0304-3940/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2009.01.058

and densely arranged CSPGs, which form a barrier to regeneration, prevent plasticity and seal off the damaged regions [8,32]. Animal studies conducted by our laboratory and others have shown that ADAMTS-1 and -4 are up-regulated in response to experimental stroke [4,28]. Recently, ADAMTS-9, which is abundantly expressed in the human adult and foetal brain [15,27], has emerged as a strong candidate for contributing to CNS pathology. Of all the ADAMTSs, ADAMTS-9 contains the highest number (14) of C-terminal TSP-1-like domains [27], suggestive of a protein with the ability to bind to heparan sulphate proteoglycans (HSPGs) of the blood-brain barrier (BBB) basement membrane and glial scars as well as to glycosphingolipid sulphatide, a lipid predominantly expressed in the CNS [2,33]. Levels of pro-inflammatory cytokines TNF and IL-1 are increased in the CNS in stroke [13] and ADAMTS-9 has been shown to be up-regulated in response to these in chondrocytes and chondrosarcoma cells [6,31], whereas anti-inflammatory mediators such as TGF␤1 down-regulate ADAMTS-9 expression in prostate cells [5]. In this present study we utilised the transient middle cerebral artery occlusion (tMCAo) rat model of focal cerebral ischaemia to demonstrate that ADAMTS-9 mRNA and protein were up-regulated within 24 h of experimental stroke. Furthermore our study shows that neurones express ADAMTS-9 in tMCAo tissue. Our data are consistent with ADAMTS-9 expression being modulated by cerebral

M.J. Reid et al. / Neuroscience Letters 452 (2009) 252–257

ischaemic conditions in the rat and has potential implications in the cellular events (both degenerative and reparative) following stroke. All surgical procedures were carried out in accordance with the Guidance on the Operation of Animals (Scientific Procedures) Act (1986). Transient focal cerebral ischaemia was induced in twelve male Sprague-Dawley rats (290–320 g, Charles River, Margate, UK) by luminal thread occlusion of the middle cerebral artery as described previously [4]. The thread was withdrawn to allow reperfusion at 90 min post-occlusion. Nine sham-operated animals were exposed to the same surgical procedure, with the exception that the thread was inserted only briefly and not to its full extent. At 6, 24 and 120 h post-tMCAo or sham-operation, the animals were sacrificed by exposure to a rising concentration of CO2 and the brains were removed and frozen on dry ice. Brains were dissected on dry ice into ipsilateral hemispheres (IH, occluded in tMCAo) and contralateral hemispheres (CH, nonoccluded in tMCAo), which were subsequently cut in half coronally generating four quarters of each brain (CH/posterior quarter, CH/anterior quarter, IH/posterior quarter, IH/anterior quarter). Frozen sections of each quarter of each brain (10 ␮m thickness) were cut on a cryostat (Leica Microsystems, Wetzlar, Germany) and collected on polylysine slides (VWR International, Lutterworth, UK). Protein and RNA was extracted from 5 × 30 ␮m sections of each quarter of each brain with TRI Reagent (Sigma Aldrich, Poole, UK) in accordance with the manufacturer’s protocol. All tissue sections and protein/RNA samples were coded for blind experiments and stored at −80 ◦ C prior to analysis. Protein concentration was determined with the bicinchonic acid (BCA) protein assay (Sigma–Aldrich). Real-time RT-PCR was performed as described previously [4]. Briefly, RNA was reverse transcribed using Superscript II reverse transcriptase (RT) (Invitrogen Co., Paisley, UK). The resulting cDNA was used as a template for real-time PCR using SYBR green mastermix (Applied Biosystems, Warrington, UK) and primers that crossed intron–exon boundaries to preclude the amplification of genomic DNA (Table 1). Primers were designed using Primer Express software (Applied Biosystems) and synthesised by MWG Biotech (Eddersberg, Germany). All real-time PCR experiments were performed using the ABI Prism 7900HT sequence detection system (Applied Biosystems). Expression of ADAMTS-9 was normalised against expression of GAPDH, which was shown to give stable expression under the conditions used. Relative expression levels of ADAMTS-9 between samples were calculated using the following formula; 2−CT , where CT = CT (ADAMTS-9) − CT (GAPDH) [20]. The separation of tMCAo-extracted proteins and subsequent Western blotting for detection of ADAMTSs has been described previously [4]. In this study, each quarter of three brains from each time-point (tMCAo and sham) were blotted with a rabbit antiADAMTS-9 antibody (used at 2 ␮g/mL), which was a kind gift from Dr. Suneel Apte (Cleveland Clinic Foundation, USA). The antibody demonstrated a reactive band of 188 kDa (absent following incubation of the antibody with 30 ␮g ADAMTS-9 immunizing peptide), approximately the predicted molecular mass of the mature form of ADAMTS-9 [6,27]. Proteins of smaller sizes were also detected with the antibody, which represent non-specific cross-reactivity of the antibody [6]. Densitometric semi-quantitation of detected bands was performed as previously described [4]. Western blotting was

253

routinely performed with an anti-␤-actin antibody to assess protein loading levels (6 ␮g/well). Real-time RT-PCR data were analysed for statistical significance as follows using GraphPad Prism 3.0 software (GraphPad Software Inc, San Diego, CA, USA). For each brain (tMCAo and sham), the 2−CT values for the anterior and posterior regions were averaged (mean) for each hemisphere. The resulting 2−CT values for the IHs and CHs were averaged (mean) for each sham brain, whereas the tMCAo hemispheres were analysed separately. The means from the animals in each of the three groups (group 1: tMCAo IHs [6 h: n = 5, 24 h: n = 6, 120 h: n = 3], group 2: tMCAo CHs [6 h: n = 5, 24 h: n = 6, 120 h: n = 3] and group 3: sham IHs/CHs [6 h: n = 4, 24 h: n = 8, 120 h: n = 6]) were calculated for each time-point. To assess the variance of the data between the three groups at each time-point, a Kruskal–Wallis non-parametric test was performed. Data displaying variance were subjected to Dunn’s multiple comparison tests to determine which groups were significantly different from each other. Western blot densitometric data were analysed for statistical significance as follows using Microsoft Excel software (Microsoft Corp., Redmond, WA, USA). The total ADAMTS-9 protein detected in 24 h sham brains was compared with 24 h tMCAo brains. For each surgical procedure, the mean IODs were calculated from four quarters of three brains (n = 12). An unpaired Student’s t-test, assuming equal variance was conducted to compare the two data sets. A pBluescript KS + plasmid (Stratagene, Amsterdam, The Netherlands) containing a cDNA fragment encoding the ADAMTS-9 cysteine-rich module, spacer region and TSP-1s 2–4 was a kind gift from Dr. Suneel Apte and was used to generate an ADAMTS9 antisense riboprobe and a sense negative control probe, with digoxigenin (DIG) labelling as described and characterised previously [15]. The probe was used to detect ADAMTS-9 mRNA in tMCAo tissue sections by in situ hybridisation using the following protocol. Pre-hybridisation steps were performed with baked-glassware and buffers treated with 0.001% (v/v) diethyl pyrocarbonate (DEPC). Tissue sections were allowed to warm from −80 ◦ C to ambient temperature for 30 min prior to treatment with chloroform (Sigma–Aldrich) for 10 min. The tissue was fixed with 4% (w/v) paraformaldehyde (PFA) for 10 min prior to washing three times for 5 min each with phosphate buffered saline (PBS: 138 mM NaCl, 2.7 mM KCl, pH 7.4). Sections were treated with 100 ␮L proteinase K (2 ␮g/mL) (Novagen, Nottingham, UK) for 10 min. A further three washes in PBS for 5 min each were followed by a post-fixation step with 4% (w/v) PFA prior to rinsing the sections in dH2 O. The tissue was dehydrated with a series of 70%, 80%, 95% and 100% (v/v) molecular grade ethanol (Sigma–Aldrich) for 2 min each prior to air drying. Slides were incubated at 60 ◦ C in pre-hybridisation buffer (Biochain, Hayward, CA, USA) for 1 h in a humidified chamber. Probes were diluted (500–1500 ng/mL) in hybridisation buffer (Ambion, Austin, USA) as calculated by dot blot analysis. Probes were denatured at 70 ◦ C for 10 min prior to cooling on ice and incubation of 100 ␮L of probe/slide in a humidified chamber at 60 ◦ C for 18 h. The following washing steps were followed at 60 ◦ C unless otherwise stated: once in 15 mM sodium citrate, 150 mM NaCl, pH 7.0 (1× SSC) for 10 min, once in 1.5×SSC for 10 min, twice in 2× SSC at 37 ◦ C for 10 min each, once in 2× SSC 0.2 ␮g/mL RNAse A (Qiagen, Hilden, Germany) at 37 ◦ C for 30 min, once in 2× SSC at

Table 1 Real-time RT-PCR primer pairs. Gene

Accession number

Sequence

Location

Product size (bp)

ADAMTS-9

XM 232202

Forward: 5 GAAGTACATCACCGAGTTCTTAGACACT3 Reverse: 5 AGAAGGCCGGGCAGTTG3

1092 1193

101

GAPDH

M 17701

Forward: 5 TGATTCTACCCACGGCAAGT3 Reverse: 5 AGCATCACCCCATTTGATGT3

171 295

124

254

M.J. Reid et al. / Neuroscience Letters 452 (2009) 252–257

ambient temperature for 10 min, twice in 0.2× SSC for 30 min each. Slides were washed in 100 mM Tris–HCl, 150 mM NaCl (pH 7.5) (Buffer 1) for 1 min prior to incubation in 100 ␮L 10% (v/v) normal sheep serum (NSS) (Sigma–Aldrich) in PBS. Anti-DIG-alkaline phosphatase (AP)-conjugated antibody (Roche) (100 ␮L, 1:100 in 5% (v/v) NSS/PBS) was applied and incubated for 18 h at 4 ◦ C in a humidified chamber. Slides were washed in Buffer 1, 0.05% (v/v) Tween 20 for 10 min prior to a wash in 100 mM Tris–HCl, 100 mM NaCl (pH 9.5) for 10 min. Detection of hybridisation was performed by incubation with SIGMA FAST BCIP/NBT (Sigma–Aldrich) (∼2–3 h). The reaction was stopped by incubation in 10 mM Tris–HCl, 1 mM EDTA (pH 8.0). Hybridisation was routinely performed with a positive control poly(dT) probe (GeneDetect, Auckland, New Zealand). Sections were mounted with PBS/glycerol (1:1) and visualised under an Olympus BX60 microscope in bright-field mode. Images were captured and analysed with LabWorks Image Acquisition and Analysis Software (UVP). To confirm the identity of ADAMTS-9-positive cells, tMCAo sections were immunohistochemicaly stained for cell-specific markers: NeuN (neuronal nuclei), von Willebrand factor (vWF, endothelial cells) and glial fibrillary acidic protein (GFAP, astrocytes). Sections were allowed to warm from −80 ◦ C to ambient temperature for 30 min prior to fixation for 10 min with 4% (w/v) PFA. The tissue was washed twice for 5 min each in PBS and was blocked in 1% (w/v) bovine serum albumin (BSA)/PBS for 30 min prior to primary antibody (100 ␮L) incubations at 4 ◦ C. Primary and secondary antibodies were diluted in PBS/1% BSA. Primary antibodies: 1 ␮g/mL mouse anti-NeuN AlexaFluor 488-conjugated antibody (Chemicon Chandlers Ford, UK), rabbit anti-vWF antibody (1:200) (DakoCytomation, Glostrup, Denmark) and 5 ␮g/mL mouse anti-GFAP (Chemicon International, Temecula, USA). Unbound primary antibodies were removed by washing three times for 5 min each with PBS/0.05% Tween 20. The anti-vWF and anti-GFAP antibodies were detected by incubation with 100 ␮L goat anti-rabbit Alexa-Fluor 568-conjugated secondary antibody (1:500) (Molecular Probes, Eugene, USA) and rabbit anti-mouse IgG FITC-conjugated antibody (1:100) (DakoCytomation) respectively for 1.5 h. Unbound secondary antibodies were removed by the same method as primary antibodies. Sections were mounted in Vectashield Mounting media with DAPI (Vector Laboratories, Peterborough, UK) for nuclear staining prior to visualisation with an Olympus BX60 fluorescent microscope. Images were captured and analysed with LabWorks Image Acquisition and Analysis Software (UVP). Real-time RT-PCR showed that ADAMTS-9 mRNA expression was up-regulated rapidly in response to tMCAo (Fig. 1). At 6 h post-procedure, ∼3-fold and ∼9-fold higher levels (not statistically significant) of ADAMTS-9 mRNA were detected in IHs of tMCAo brains when compared to CHs of tMCAo brains and sham-operated tissue respectively. ADAMTS-9 mRNA expression levels remained high up to 24 h post-procedure, with ∼10-fold higher levels (statistically significant, p < 0.01) being detected in tMCAo IHs when compared to sham-operated tissue. Expression of ADAMTS-9 mRNA decreased in tMCAo brains between 24 and 120 h with only modestly increased levels being detected in tMCAo IHs when compared to tMCAo CHs (∼3-fold) and sham-tissue (∼4-fold) at 120 h. Western blotting analysis (Fig 2A) demonstrated that at 6 h post-sham and tMCAo, ADAMTS-9 protein levels were below the detectable limit. At 24 h post-procedure, the mature form of ADAMTS-9 protein (188 kDa) was detectable in tMCAo tissue but not in sham-operated brains. Densitometric analysis demonstrated that levels of the mature form of ADAMTS-9 were significantly higher (∼11-fold, p < 0.05) in total tMCAo brains when compared to total sham brains (Fig. 2B). Modestly higher levels (∼4-fold in posterior, ∼2-fold in anterior) of mature ADAMTS-9 protein were detected in CHs when compared to IHs in tMCAo tissue at 24 h, as calculated by densitometry (Fig 2B) (not statistically significant).

ADAMTS-9 protein was not detected at 120 h in either tMCAo or sham-operated brain tissue. In the cerebral cortex of tMCAo tissue, expression of ADAMTS-9 mRNA was detected in large cells of triangular/conical morphology, characteristic of pyramidal neurones as demonstrated by in situ hybridisation (Fig. 3A). ADAMTS-9 mRNA was also detected in the densely arranged granular neuronal cells of the dentate gyrus in the hippocampus of tMCAo brains (Fig. 3F). ADAMTS-9 in situ hybridisation signal was compared to serial immunohistochemically stained sections for cell-specific markers: NeuN, vWF and GFAP. The localisation of NeuN-positive large cells (Fig. 3C and H) was comparable to that of ADAMTS-9 mRNA-positive cells (Fig 3B and G) in tMCAo tissue, suggesting a neuronal source of ADAMTS-9. Throughout brains from tMCAo animals, cells smaller than neurones were also ADAMTS-9 mRNA-positive in regions with high levels of GFAP expression (Fig. 3D and I), suggesting expression of ADAMTS-9 by astrocytes in vivo. By comparison of the localisation of vWF staining (Fig. 3E and J) and ADAMTS-9 in situ hybridisation signal (Fig 3B and G), endothelial cells did not express ADAMTS-9 mRNA in tMCAo tissue. This study provides strong evidence that ADAMTS-9 expression is modulated by pathological conditions that occur in response to cerebral ischaemia. These ADAMTS-9 data are consistent with ADAMTS-1 and -4 expression, which were both up-regulated posttMCAo and in contrast to ADAMTS-5 and -8, levels of which were not significantly raised in response to experimental stroke [4,28]. ADAMTS-9 has the potential to be involved in the cellular events that follow a stroke because the mature form of the protein was present in tMCAo brains at 24 h, which is likely to be a direct result of the rapid up-regulation of ADAMTS-9 mRNA expression, detected at 6 and 24 h post-tMCAo. The activity of ADAMTS-9 at 24 h in tMCAo tissue may not have been neutralised by TIMP-3, a known inhibitor of ADAMTSs [12,26], levels of which have been shown by our laboratory to remain relatively constant between tMCAo and sham brains at all time-points [4]. A potential contributing factor to the presence of the mature form in tMCAo brains is that the proprotein convertase furin, which processes ADAMTS-9 by removal of the prodomain [17,18,27], has been shown to be up-regulated at the mRNA level following transient global cerebral ischaemia in the rat [29]. The detection of slightly higher levels of the mature form of ADAMTS-9 protein in CHs when compared to IHs of tMCAo brains (statistically non-significant) was surprising given the mRNA data. The reason for this discrepancy is not known but may relate to the protein having a shorter half-life in the IHs because of its breakdown by up-regulated proteinases e.g. MMPs [9,25]. Alternatively, mRNA levels may have been significantly raised in tMCAo CHs at timepoints not included in this study between 6 and 24 h. It is feasible given the mRNA data that the zymogen of ADAMTS-9 was present in higher levels than the mature form in tMCAo IHs when compared to tMCAo CHs and sham brains. This is perhaps significant because it has been demonstrated that the zymogen of ADAMTS-9 is more effective than the mature form at cleaving versican [18]. However, the accurate resolution of the ADAMTS-9 zymogen by SDS-PAGE was problematic because of its size (∼216 kDa) and extensive glycosylation [27], as documented previously [17]. The relatively high S.E.Ms. calculated for the tMCAo brains when compared to shamoperated (both real-time RT-PCR and western blotting) was likely a result of variability in cerebral ischaemic injury severity between tMCAo rats at each time-point [7]. ADAMTS-9 expression has been shown to be modulated by inflammatory mediators, many of which have been implicated in stroke pathogenesis. The time-course of expression of such mediators, post-ischaemia, is comparable to that of ADAMTS-9 detected in this study. TNF, which up-regulates ADAMTS-9 mRNA expression in

M.J. Reid et al. / Neuroscience Letters 452 (2009) 252–257

255

Fig. 1. Real-time RT-PCR analysis of relative ADAMTS-9 mRNA expression levels in rat brain tissue following tMCAo and sham-operation at 6, 24 and 120 h post-procedure. (A) Melt-curve analysis of ADAMTS-9 and GAPDH primers, showing no primer-dimer or amplification of non-specific products. (B) Sigmoidal amplification plots demonstrating the stable expression of GAPDH (CT ≈ 15) and the differentially regulated expression of ADAMTS-9 (CT ≈ 21–25) in rat brains subjected to either sham or tMCAo surgery. A and B depicts data from five brains (four quarters of each) at different time-points/treatments and is representative of experiments throughout the study. (C) Quantification of data from all brains expressed as mean 2-CT ± S.E.M. ** denotes a statistically significant (p < 0.01) difference (Kruskal–Wallis test followed by Dunn’s test, as described in ‘Statistics’). Rn, reporter signal normalised to fluorescent signal of passive reference dye (ROX) (Rn) – baseline fluorescence; C, contralateral; I, ipsilateral.

chondrocytes and chondrosarcoma cells [6], generally increases at ∼2 h post-ischaemia prior to peaking at 12 h and remaining high for 24 h before returning to normal at 120 h [21]. IL-1␤, which also up-regulates ADAMTS-9 mRNA expression in chondrocytes and chondrosarcoma cells [6,31], generally starts to increase at ∼4 h, peaking at 24 h before returning to normal at 120 h [21]. Furthermore, our laboratory has previously shown that IL-1␤ and TNF mRNA were increased in tMCAo IHs when compared to tMCAo CHs and sham brains at 24 h [4]. However, the impact of IL-1␤ may be counteracted to some extent by IL-1 receptor antagonist (IL1ra), a competitive inhibitor of IL-1 [10], levels of which were also higher in tMCAo brains [4]. The finding that tMCAo CH as well as IH ADAMTS-9 protein expression was increased compared to shamoperated controls may be due to a global response of the brain tissue to increased pro-inflammatory cytokine expression levels in tMCAo animals. It appears that increased ADAMTS-9 expression in response to the surgical procedure may not be restricted entirely to the site of injury.

Perineuronal net and glial scar CSPGs aggrecan and versican inhibit neuronal outgrowth post-stoke and are substrates of ADAMTS-9 in vivo [27,32]. There is increasing evidence to suggest that ADAMTS-9 has novel mechanisms of processing and secretion, which result in the enzyme having maximal activity as a proteinase on the surface of cells as opposed to extracellularly [17,18]. Therefore, our data demonstrating that neurones expressed ADAMTS-9 in tMCAo rat brains is potentially important in that ADAMTS-9 could be involved in processing the extracellular matrix (including CSPGs of perineural nets and glial scars) for axonal regeneration. Similarly to ADAMTS-9, MMP-9 was previously shown to be expressed by pyramidal neurones following a rodent model of cerebral ischaemia [19]. This study has provided evidence that ADAMTS-9 is a potentially important proteinase with respect to cerebral ischaemia. Further elucidation of the complexities of the ADAMTS proteinase system in the CNS could lead to the development of novel therapeutics.

256

M.J. Reid et al. / Neuroscience Letters 452 (2009) 252–257

Fig. 2. Western blot analysis of ADAMTS-9 protein expression levels in rat brain tissue following tMCAo and sham-operation. (A) Representative blots demonstrating the detection of the mature form of ADAMTS-9 protein in 24 h tMCAo brains only. The corresponding ␤-actin blot for the 24 h tMCAo protein is also presented. (B) Semi-quantitative densitometric analysis of the 188 kDa band in brains at 24 h post-operation (sham and tMCAo) expressed as mean integrated optical density (IOD) ± S.E.M. Data represents three independent experiments from three different brains at each time-point. * denotes a statistically significant (p < 0.05) difference in mature ADAMTS-9 protein expression between total 24 h sham and total 24 h tMCAo brains (unpaired Student’s t-test, as described in ‘Statistics’). CA, contralateral anterior; CP, contralateral posterior; IA, ipsilateral anterior; IP, ipsilateral posterior.

Fig. 3. Analysis of the cellular origin of ADAMTS-9 mRNA expression in tMCAo rat brain tissue. ADAMTS-9 mRNA-positive cells were detected by in situ hybridisation in tMCAo tissue at all time-points with an ADAMTS-9 antisense riboprobe as demonstrated by the representative series of images (A, B, F and G). High-power (×400) microscopy confirmed neurones were ADAMTS-9 mRNA-positive (A and F), specifically pyramidal neurones (arrows) and granular neurones (arrowheads). Low-power (×100) microscopy demonstrated that ADAMTS-9 mRNA-positive cells (B and G) had comparable localisation patterns to cells stained positive for NeuN protein (neurones) (C and H) but not vWF (endothelial cells of blood vessels [BV]) (E and J). ADAMTS-9 mRNA-positive cells of non-neuronal morphology (A and F) were detected. Such cells were potentially astrocytes because of their smaller size and comparable localisation to GFAP protein-positive cells (D and I). Representative images capture the cerebral cortex (CC)/leptomeninges (LM)/thalamus (T) region (A–E) and the hippocampus (F–J) of 24 h tMCAo tissue (CH/posterior quarter). In situ hybridisation performed with the ADAMTS-9 sense-strand probe showed only background staining (A and B insets). CA3, Cornu Ammonis field 3, DG, dentate gyrus.

Acknowledgement Martin Reid was the recipient of a Biomedical Research Centre PhD studentship. References [1] C. Brakebusch, C.I. Seidenbecher, F. Asztely, U. Rauch, H. Matthies, H. Meyer, M. Krug, T.M. Böckers, X. Zhou, M.R. Kreutz, D. Montag, E.D. Gundelfinger, R. Fässler, Brevican-deficient mice display impaired hippocampal CA1 long-term potentiation but show no obvious deficits in learning and memory, Mol. Cell. Biol. 22 (2002) 7417–7427. [2] H. Chen, M.E. Hernden, J. Lawler, The cell biology of thrombospondin-1, Matrix Biol. (2000) 597–614.

[3] J.-C. Copin, M.-C. Goodyear, J.M. Gidday, A.R. Shah, E. Gascon, A. Dayer, D.M. Morel, Y. Gasche, Role of matrix metalloproteinases in apoptosis after transient focal cerebral ischemia in rats and mice, Eur. J. Neurosci. 22 (2005) 1597–1608. [4] A.K. Cross, G. Haddock, C.J. Stock, S. Allan, J. Surr, R.A.D. Bunning, D.J. Buttle, M.N. Woodroofe, ADAMTS-1 and -4 are up-regulated following transient middle cerebral artery occlusion in the rat and their expression is modulated by TNF in cultured astrocytes, Brain Res. 1088 (2006) 19–30. [5] N.A. Cross, S. Chandrasekharan, N. Jokonya, A. Fowles, F.C. Hamdy, D.J. Buttle, C.L. Eaton, The expression and regulation of ADAMTS-1, -4, -5, -9, and -15 and TIMP-3 by TGF␤1 in prostate cells: relevance to the accumulation of versican, Prostate 63 (2005) 269–275. [6] K. Demircan, S. Hirohata, K. Nishida, O.F. Hatipoglu, T. Oohashi, T. Yonezawa, S.S. Apte, Y. Ninomiya, ADAMTS-9 is synergistically induced by interleukin-1␤ and tumour necrosis factor alpha in OUMS-27 chondrosarcoma cells and in human chondrocytes, Arthritis Rheum. 52 (2005) 1451–1460.

M.J. Reid et al. / Neuroscience Letters 452 (2009) 252–257 [7] U. Dirnagl, Bench to bedside: the quest for quality in experimental stroke research, J. Cereb. Blood Flow Metab. 26 (2006) 1465–1478. [8] C.M. Galtrey, J.W. Fawcett, The role of chondroitin sulfate proteoglycans in regeneration and plasticity in the central nervous system, Brain Res. Rev. 54 (2007) 1–18. [9] G. Gao, A. Plaas, V.P. Thompson, S. Jin, F. Zuo, J.D. Sandy, ADAMTS4 (aggrecanase-1) activation on the cell surface involves C-terminal cleavage by glycosylphosphatidyl inositol-anchored membrane type 4-matrix metalloproteinase and binding of the activated proteinase to chondroitin sulfate and heparan sulfate on syndecan-1, J. Biol. Chem. 279 (2004) 10042–10051. [10] R.M. Gibson, N.J. Rothwell, R.A. Le Feuvre, CNS injury: the role of the cytokine IL-1, Vet. J. 168 (2004) 230–237. [11] G. Haddock, A.K. Cross, J. Plumb, J. Surr, D.J. Buttle, R.A.D. Bunning, M.N. Woodroofe, Expression of ADAMTS-1, -4, -5 and TIMP-3 in normal and multiple sclerosis CNS white matter, Mult. Scler. 12 (2006) 386–396. [12] G. Hashimoto, T. Aoki, H. Nakamura, K. Tanzawa, Y. Okada, Inhibition of ADAMTS4 (aggrecanase-1) by the tissue inhibitors of metalloproteinases (TIMP1, 2, 3 and 4), FEBS Lett. 494 (2001) 192–195. [13] J. Huang, U.M. Upadhyay, R.J. Tamargo, Inflammation in stroke and cerebral ischemia, Surg. Neurol. 66 (2006) 232–245. [14] G.C. Jones, G.P. Riley, ADAMTS proteinases: a multi-domain, multi-functional family with roles in extracellular matrix turnover and arthritis, Arthritis Res. Ther. 7 (2005) 160–169. [15] K.A. Jungers, C. Le Goff, R.P.T. Somerville, S.S. Apte, Adamts 9 is widely expressed during mouse embryo development, Gene Expr. Patterns 5 (2005) 609–617. [16] M.A. Kelly, A. Shuaib, K.G. Todd, Matrix metalloproteinase activation and bloodbrain barrier breakdown following thrombolysis, Exp. Neurol. 200 (2006) 38–49. [17] B.-H. Koo, J.-M. Longré, R.P.T. Somerville, J.P. Alexander, R. Leduc, S.S. Apte, Cell-surface processing of pro-ADAMTS9 by furin, J. Biol. Chem. 281 (2006) 12485–12494. [18] B.-H. Koo, J.-M. Longpré, R.P.T. Somerville, J.P. Alexander, R. Leduc, S.S. Apte, Regulation of ADAMTS9 secretion and enzymatic activity by its propeptide, J. Biol. Chem. 282 (2007) 16146–16154. [19] S.-R. Lee, H.-Y. Kim, J. Rogowska, B.-Q. Zhao, P. Bhide, J.M. Parent, E.H. Lo, Involvement of matrix metalloproteinase in neuroblast cell migration from the subventricular zone after stroke, J. Neurosci. 26 (2006) 3491–3495. [20] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using realtime quantitative PCR and the 2−CT method, Methods 25 (2001) 402–408.

257

[21] M.N. Perera, H.K. Ma, S. Arakawa, D.W. Howells, R. Markus, C.C. Rowe, G.A. Donnan, Inflammation following stroke, J. Clin. Neurosci. 13 (2006) 1–8. [22] J. Plumb, A.K. Cross, J. Surr, G. Haddock, T. Smith, R.A.D. Bunning, M.N. Woodroofe, ADAM-17 TIMP3 protein and mRNA expression in spinal cord white matter of rats with acute experimental autoimmune encephalomyelitis, J. Neuroimmunol. 164 (2005) 1–9. [23] S. Porter, I.M. Clark, L. Kevorkian, D.R. Edwards, The ADAMTS metalloproteinases, Biochem. J. 386 (2005) 15–27. [24] U. Rauch, X.-H. Zhou, G. Roos, Extracellular matrix alterations in brains lacking four of its components, Biochem. Biophys. Res. Commun. 328 (2005) 608–617. [25] J.C. Rodríguez-Manzaneque, A.B. Milchanowski, E.K. Dufour, R. Leduc, M.L. Iruela-Arispe, Characterization of METH-1/ADAMTS-1 processing reveals two distinct active forms, J. Biol. Chem. 275 (2000) 33471–33479. [26] J.C. Rodríguez-Manzaneque, J. Westling, S.N.-M. Thai, A. Luque, V. Knauper, G. Murphy, J.D. Sandy, M.L. Iruela-Arispe, ADAMTS1 cleaves aggrecan at multiple sites and is diferentially inhibited by metalloproteinase inhibitors, Biochem. Biophys. Res. Commun. 293 (2002) 501–508. [27] R.P.T. Somerville, J.-M. Longpre, K.A. Jungers, J.M. Engle, M. Ross, S. Evanko, T.N. Wight, R. Leduc, S.S. Apte, Characterization of ADAMTS-9 and ADAMTS-20 as a distinct ADAMTS subfamily related to Caenorhabitis elegans GON-1, J. Biol. Chem. 278 (2003) 9503–9513. [28] Y.F. Tian, P.B. Zhang, X.L. Xiao, J.S. Zhang, T.T. Zhao, Q.Y. Kang, X.L. Chen, F. Qiu, Y. Liu, The quantification of ADAMTS expression in an animal model of cerebral ischemia using real-time PCR, Acta. Anaesth. Scand. 51 (2007) 158–164. [29] N. Yakota, M. Uchijima, S. Nishizawa, H. Namba, Y. Koide, Identification of differentially expressed genes in rat hippocampus after transient global cerebral ischemia using Subtractive cDNA cloning based on polymerase chain reaction, Stroke 32 (2001) 168–174. [30] Y. Yamaguchi, Lecticans: organizers of the brain extracellular matrix, Cell. Mol. Life Sci. 57 (2000) 276–289. [31] K.O. Yaykasli, T. Oohashi, S. Hirohata, O.F. Hatipoglu, K. Inagawa, K. Demircan, Y. Ninomiya, ADAMTS9 activation by interleukin 1b via NFATc1 in OUMS-27 chondrosarcoma cells and in human chondrocytes, Mol. Cell. Biochem, in press. [32] G. Yiu, Z. He, Glial inhibition of CNS axon regeneration, Nat. Rev. Neurosci. 7 (2006) 617–626. [33] W. Zeng, C. Corcoran, L.A. Collins-Racie, E.R. LaVallie, E.A. Morris, C.R. Flannery, Glycosaminoglycan-binding properties and aggrecanase activities of truncated ADAMTSs: comparative analyses with ADAMTS-5, -9, -16 and -18, Biochim. Biophys. Acta 1760 (2006) 517–524.