Vascular MMP-9/TIMP-2 and Neuronal MMP-10 Up-Regulation in Human Brain after Stroke: A Combined Laser Microdissection and Protein Array Study Eloy Cuadrado,† Anna Rosell,† Anna Penalba,† Mark Slevin,‡ Jose´ Alvarez-Sabı´n,† Arantxa Ortega-Aznar,§ and Joan Montaner*,† Neurovascular Research Laboratory, Neurovascular Unit, Department of Neurology, Department of Internal Medicine, Universitat Auto`noma de Barcelona, Institut de Recerca, Hospital Vall d’Hebron, Barcelona, Spain, SBCHS, Manchester Metropolitan University, Manchester, United Kingdom, and Neuropathology Unit, Department of Pathology, Hospital Vall d’Hebron, Barcelona, Spain Received November 23, 2008
Abstract: Matrix Metalloproteinases (MMPs) play an important role in brain injury after ischemic stroke. In the present study, we aimed to assess the global expression of MMP-Family proteins in the human brain after stroke by using a combination of Searchlight Protein Array and Laser Microdissection to determine their cellular origin. This study demonstrated that MMP-1, MMP-2, MMP-3, MMP-8, MMP-9, MMP-10, MMP-13, and TIMP-1 were upregulated in the infarcted tissue compared to healthy control areas. Using laser microdissection we obtained specific neuronal and vascular populations from both infarcted and control areas. From these fractions, we showed that MMP-9 and TIMP-2 were highly produced in brain microvessels while MMP-10 was notably increased in neurons of the ischemic brain but not in healthy areas. These findings demonstrate a selective celldependent MMP secretion, opening the possibility of selectively targeting specific MMPs for neuroprotection or vasculoprotection following stroke. Keywords: matrix metalloproteinases • laser microdissection • protein array • neurons • vessels • brain • stroke
Introduction Ischemic stroke is a leading cause of death and disability worldwide that occurs after the occlusion of a cerebral artery, generally by a thrombus, and which can cause irreversible brain damage due to neuronal death, blood-brain barrier (BBB) breakdown, and cellular edema.1 Over the past few years, there has been increasing evidence that matrix metalloproteinases (MMPs) play a significant role in the pathobiological consequences that follow cerebral ischemia. MMPs are a family of zinc-binding proteolytic enzymes * To whom correspondence should be addressed. Dr Joan Montaner, Neurovascular Research Laboratory, Neurovascular Unit, Institut de Recerca, Hospital Vall d’Hebron, Pg Vall d’Hebron 119-129, 08035 Barcelona, Spain. Telephone number: +34934894073. Fax number: +34934894015. E-mail:
[email protected]. † Institut de Recerca, Hospital Vall d’Hebron. ‡ Manchester Metropolitan University. § Department of Pathology, Hospital Vall d’Hebron. 10.1021/pr801012x CCC: $40.75
2009 American Chemical Society
that normally remodel the extracellular matrix and pathologically attack substrates as part of the neuroinflammatory response.2 Animal models of cerebral ischemia have reported that both MMP-9 and MMP-2 are abnormally expressed after an ischemic insult.3,4 For example an increased in situ gelatinase/collagenase activity has been reported after cerebral ischemia in brain samples from mice,5 rats6-8 and humans9 related to MMP activity. Interestingly, mice lacking MMP-9 gene or treated with broad spectrum MMP inhibitors exhibit diminished cerebral damage after stroke.10 Recently, overexpression of several MMPs has been shown to be associated with BBB disruption and as a consequence, associated with brain hemorrhage after stroke. In this sense, MMP-9 activity has been related to a major basal lamina component destruction (collagen IV) in stroke patients.1 MMP-3, also appears to attack tight junctions11 and is up regulated in endothelial cells after thrombolytic treatment in mice.12 MMPs are either transmembrane or secreted proteins, however we have recently demonstrated the presence of MMP13 in the nuclei of brain cells after cerebral ischemia.13 Also, in previous studies from our group, we have demonstrated that blood levels of MMP-13 in stroke patients were related to infarct development and growth.14 However, data is lacking regarding other MMPs and their specific cellular production following brain ischemia. We hypothesized that Laser Microdissection (LMD) of the neurovascular unit followed by multiple protein-analysis techniques might help to identify their specific expression patterns after stroke. To our knowledge, there are no studies focusing on MMPs in cerebral ischemia using LMD at the protein level, indeed only a few investigations have used a combination of LMD and proteomic techniques in brain tissue. In two methodological studies, large brain areas (fixed or unfixed) were captured and proteins separated by two-dimensional electrophoresis (2-DE). Selected spots were analyzed with matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry (MALDITOF) showing that fixation had no effect on protein identification.15,16 Another group isolated much smaller amounts of brain microvessels and employed isotope-coded affinity tagging (ICAT), nano liquid chromatography and tandem mass specJournal of Proteome Research 2009, 8, 3191–3197 3191 Published on Web 03/24/2009
technical notes
Cuadrado et al. 17
trometry (nanoLC-MS/MS) for analysis. In this study the authors confirmed their results determining IL-1β levels in microdissected samples with a standard ELISA.17 Very recently, a proteomic study was done aiming to identify the protein profile of the mouse brain microvascular endothelium in situ. In this study, brain microvascular endothelial cells were captured and subsequent analysis was performed using linear ion trap coupled with Fourier transform-mass spectrometry (LTQ-FT MS); using this technology, 881 novel proteins were successfully identified.18 Pierce Searchlight protein array is a plate-based multiplex assay, which is in concept a standard ELISA, incorporating multiple capture/detection antibodies. In this study we aimed to examine the global MMPs expression in human brain samples after a fatal stroke. By using the combination of LMD and Searchlight Human MMP array we aimed to analyze the MMP expression profile of the neurovascular unit, that is, the neurons and the brain microvasculature, after stroke.
Material and Methods Brain Tissue Samples. Ten deceased patients (6 women and 4 men) who had an ischemic stroke within the previous 4 days (range 10-87 h) were included in the study. On autopsy and during macroscopic examination, the infarcted area was identified by an experienced neuropathologist as previously described by our group.9 Neuroradiological images were used to guide brain tissue sampling from the infarct core (n ) 10) and the contralateral hemisphere (n ) 10). All samples were obtained within the first six hours after death and snap frozen in liquid nitrogen or fixed with 4% paraformaldehyde for immunohistochemistry and stored at -80 °C until use. This study was approved by the Ethics Committee of the Hospital Universitari Vall d’Hebron [PR(HG)85/04] and informed consent was acquired from relatives prior to the autopsy following Spanish governmental guidelines. Frozen brain samples (0.2 g) were homogenized and mixed with 0.7 mL of cold lysis buffer (50 mM Tris-HCl pH 7.6, 150 Mm NaCl, 5 mM CaCl2, 0.05% Brij-35, 0.02% NaN3 and 1% Triton X-100) containing protease inhibitors (1 mM PMSF and 7 µg/mL aprotinin) and centrifuged at 12 000g for 10 min. Total protein content was determined by the standard bicinchoninic acid assay (BCA, Pierce, Rockford, IL, U.S.A.). Preparation of Sections for LMD. Frozen brain samples were embedded in Tissue-Tek O.C.T (Sakura Finetek Europe, The Netherlands) at -20 °C, and 10 µm-thick sections were cut using a cryostat (Leica CM3050 S; Leica Microsystems, Germany). Sections were mounted on 2 µm PEN-Membrane slides (MicroDissect GmbH, Germany) and stored at -80 °C. Neurons were stained using a mouse anti-NeuN antibody (1:50; Chemicon, U.S.A.). Briefly, sections were fixed in ice-cold acetone for 30 s and then washed in tween phosphate buffer saline (TPBS) for 1 min. Preparations were air-dried for 5 min at 37 °C. The sections were then blocked with 10% BSA (w/v)/ 0.3% Triton X-100 (v/v) in PBS (blocking buffer) for 3 min at room temperature, and then incubated for 10 min at 37 °C with the primary antibody diluted in blocking buffer. After washing in TPBS (1 min), sections were incubated with a goat AlexaFluor 488 antimouse IgG (Invitrogen, U.S.A.) diluted 1:50 in blocking buffer for 10 min at 37 °C. Finally sections were washed in TPBS for 1 min and air-dried for 10 min at 37 °C prior to microdissection. 3192
Journal of Proteome Research • Vol. 8, No. 6, 2009
Brain microvessels were stained using Ulex europeaus Agglutinin I (UEA I) lectin (1:20; Sigma-Aldrich, U.S.A.) as previously described.19 Laser Microdissection. LMD was performed on a Leica LMD6000 microscope (Leica Microsystems, Germany). Approximate areas of 2 000 000 µm2 of each cell type (4000-5500 cells) were dissected from 10 µm cryostat cut sections containing either infarct or contralateral brain tissue from 5 stroke deceased patients (n ) 10 per cell type). Cells were dissected into dry 0.2 mL tube caps coated with silicon (MicroDissect GmbH, Germany) at a power of 90-92 and a speed of 6 using a 20× objective. Cells were recovered in 130 µL of cold lysis buffer (described above), vortexed for 5 min and centrifuged at 12 000g for 10 min. Total protein content was determined using the micro bicinchoninic acid assay (microBCA, Pierce, U.S.A.). Protein yield was on average, 0.115 µg/uL. Multiplexed Searchlight MMP Protein Array. A SearchLight Human MMP Array 1 (Pierce, USA) was used to measure MMP expression; this assay consists of a multiplexed sandwich enzyme-linked immunosorbent assay, and was used for the simultaneous quantitative measurement of the following 9 proteins: gelatinases (MMP-2 and MMP-9), collagenases (MMP1, MMP-8, and MMP-13), stromelysines (MMP-3 and MMP10), and endogenous inhibitors (TIMP-1 and TIMP-2). Each sample was assayed twice and the mean value of both measurements was used. The mean intra assay coefficients of variation were
