| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Pathology, Yale University School of Medicine, New Haven, Connecticut, USA
1Correspondence: Department of Pathology, Yale University School of Medicine, 310 Cedar St., New Haven, CT 06520, USA. E-mail: joseph.Madri{at}yale.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: MMP-9 • MT1-MMP • EAE • T lymphocyte
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4)
. We previously defined important roles for 
4 integrin and MMP-2 in experimental autoimmune encephalomyelitis (EAE) (5
, 6)
. In both in vivo animal models and in vitro culture models of this disease, MMPs have been shown to participate in T lymphocyte transmigration and destruction of myelin sheaths in the CNS. Other pericellular protease systems, including uPA, have been determined to be involved in T lymphocyte transmigration during the development of EAE (3
, 7)
. To better elucidate the roles of MMP-2 in the development of EAE, we used MMP-2-deficient mice. These mice are viable but smaller than their wild-type and heterozygous littermates and exhibit reduced tumor-induced angiogenesis, glioma growth, and abnormal lung alveolarization (8)
. In this study we were surprised to observe that MMP-2-deficient mice exhibited an earlier onset and more severe clinical disease score than wild-type mice. Furthermore, MMP-2+/+ mice engrafted with MMP-2–/– bone marrow exhibited a similar earlier onset and more severe clinical disease score than MMP-2+/+ mice engrafted with MMP-2+/+ bone marrow, consistent with the concept that MMP profiles of the circulating immune cells are responsible in part for the differences in disease onset and severity. Using an in vitro transmigration assay, we observed that lymphocytes derived from MMP-2-deficient mice exhibited increased transmigration through endothelial cell monolayers and basal lamina components, which correlated with increased expression of MMP-9 and was abrogated by inhibition of MMP activity. Using lymphocytes derived from MT1-MMP+/– (MMP-14+/–) mice, we demonstrated a correlation between expression levels of MMP-14, MMP-9 expression, and transmigration rate, and suggest a signaling pathway involving tethering of MMP-2 to MT1-MMP as a modulator of MMP-9 expression. Last, we discuss other possible MMP-2-mediated mechanisms that may contribute to the observed phenotype.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(a kind gift from Dr. David Muir, University of South Florida, Gainesville, FL, USA); and MMP-14 heterozygous mice (MMP-14+/–) on a C57BL/6 background (10)
(a generous gift from Dr. Suneel Apte, Cleveland Clinic Research Foundation, Cleveland, OH, USA) were housed in the animal facilities of the Boyer Center for Molecular Medicine, Yale School of Medicine, according to Yale University and National Institutes of Health guidelines. All mice were used at 8–12 wk of age.
Bone marrow engraftment
A previously described method used for bone marrow engraftment in our laboratory was followed (11
, 12)
.
Induction of experimental autoimmune encephalomyelitis
EAE was induced using the myelin oligodendrocyte glycoprotein (MOG) immunization protocol described previously (12
, 13)
. The MOG peptide 35-55 (MEVGWYRSPFSRVVHLYRNGK) was synthesized by the W. M. Keck Biotechnology Resource Center at Yale University. Disease severity was monitored according to the following scale: 0, no disease; 1, flaccid tail; 2, hind limb weakness; 3, hind limb paralysis; 4, forelimb weakness; 5, moribund.
Histological examination was accomplished using standard hematoxylin and eosin staining of 5 µ-thick brain sections.
T lymphoblasts
Lymphocytes were obtained from spleens and cultured in RPMI 1640 (Invitrogen, San Diego, CA, USA) supplemented with 10% FCS (Invitrogen), 50 µg/mL gentamicin (Invitrogen), 10 ng/mL IL-2 (R&D Systems, Abingdon, Oxon, U), and 5 µg/mL concanavalin A (Calbiochem, San Diego, CA, USA) (14)
. Cells were used for experiments after 2 days in culture.
Quantitation of WT and MMP-2 KO CD3+,CD4+, and CD8+ lymphocyte numbers was performed by FACS analysis. Quantitation of proliferation of WT, MMP-2 KO, and MMP-14+/– lymphocytes was performed as described (15)
.
Antigen-specific T cell line
MOG-specific T cells were derived from immunized mice as described previously (12)
.
Endothelial cells
Lung-derived endothelial cells were obtained by positive selection using the Cellection® Biotin Binder kit (Dynal, Great Neck, NY, USA) according to the manufacturer’s instructions. The CD31 MEC 13.3-biotin antibody (BD Biosciences, Franklin Lakes, NJ, USA) was employed.
Endothelioma cell line
The culture of the endothelioma cell line bEnd.5 derived from primary brain microvascular endothelium was performed as described previously (16)
.
Cell adhesion assay
T cells were extensively washed and resuspended in RPMI 1640 containing 50 µg/mL gentamicin. Adhesion assays were performed in 96-well plates coated with collagen type IV (BD Biosciences) using 2 x 105 cells per well as described (17)
. Attached cells were stained using the CyQUANT® Cell Proliferation Assay kit (Molecular Probes, Eugene, OR, USA) according to the manufacturer’s instructions. Fluorescence was measured with the Victor2 multilabel counter (Wallac, Turku, Finland).
Chemotaxis assay
Freshly isolated spleen lymphocytes, in vitro activated lymphoblasts, and MOG-specific T cells were suspended in migration medium (RPMI 1640, 10 ng/mL IL-2, 50 µg/mL gentamicin) at 106 cells/mL. Laminin 1-, collagen type IV-, and collagen type I-coated transmigration inserts (diameter 6.5 mm; pore size 3 µm) were obtained from BD Biosciences. The cell line bEnd.5 was plated on collagen type IV inserts and, prior to the assay, was stimulated with 5 nM TNF- for 16 h. Chemokines CCL19 and CCL21 (R&D Systems) were diluted at 250 ng/mL in migration medium and added in triplicate to the lower chamber in a final volume of 500 µL. When transendothelial migration assays were performed, the migration medium in the lower chamber was RPMI 1640-10%FCS, containing no chemokines. T cells were added to the upper chamber in a final volume of 500 µL, and chemotaxis assays were conducted for 16 h in 5% CO2 at 37 C. In some experiments, T cells were preincubated for 20 min at 37 C with 1 mM of the broad spectrum MMP inhibitor 1, 10, o-phenanthroline, or 1 µM MMP-9 inhibitor I (Calbiochem) and the assay was carried out for 6 h. At the end of the assay, the inserts were discarded and the plates centrifuged at 200 g for 10 min. The media was aspirated and the cells fluorescently stained as described in the cell adhesion assay section. Counts of transmigrated cells are expressed as percentage of the input cells.
Zymography
Conditioned media from 107 resting T cells was concentrated 200-fold using Centricon Plus-20 devices (Millipore, Bedford, MA, USA) and subjected to gelatin zymography exactly as described before (18)
. In some experiments, lysates from 107 cells were obtained at 4°C using a 1% Triton-X, 50 mM Tris, 5 mM CaCl2, and 1 µM ZnCl2 buffer, mixed 1:1 with nonreducing Laemmli buffer, and analyzed by zymography.
Western blot
5 x 107 resting T cells were lysed in 1x Laemmli buffer and subjected to SDS-PAGE before transfer to PVDF membranes. After 1 h in blocking buffer, membranes were incubated overnight at 4°C with primary antibody in blocking buffer, washed with TBS/0.1% Tween20, and incubated 30 min with the appropriate HRP-conjugated secondary reagent at 1:10000 (Cell Signaling Technology, Beverly, MA, USA). Blots were washed as before and developed using the ECL system (Amersham, Arlington Heights, IL, USA) and exposed to X-ray film (Amersham). The following primary antibodies were used: biotinylated anti-proMMP-9 (R&D Systems) at 1:500; anti-MMP-14 hinge region (Triple Point Biologics Inc., Forest Grove, OR, USA) at 1:500; and anti-ERK-2 D-2 (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) at 1:5000. The blocking buffer was 5% fat-free milk in TBS/0.1% Tween20 except for the biotinylated antibody that was 3% BSA in TBS/0.1% Tween20.
Flow cytometry
Flow cytometry was performed as described previously (18)
on a FACScanTM, and analyzed and rendered by means of the CellQuest® software (BD Biosciences). The following antibodies were used: anti-CD44-FITC IM7; anti-CD29-FITC Ha2/5; anti-CD18-FITC C71/16; anti-CD26-FITC H194-112; anti-Vß8.1/8.2 TCR-PE MR5-2; anti-CD8-FITC Ly-2; anti-CD4-TriColor (BD Biosciences); and anti-CCR7-PE (R&D Systems).
Quantitative PCR
Total RNA was extracted from 107 freshly isolated lymphocytes using the RNeasy Protect and the QIAshredder kits, then treated with the RNase-Free DNase Set to eliminate possible genomic DNA contamination, all according to the manufacturer’s instructions (Qiagen, Chatsworth, CA, USA). The total RNA obtained was reverse transcribed using the iScript cDNA Synthesis kit (Bio-Rad, Richmond, CA, USA) following the manufacturer’s procedures. Real-time PCR was performed on an iCyclerTM (Bio-Rad) using the iQ SYBR Green Supermix (Bio-Rad). The cycling conditions were 95°C 3 min, 45 cycles (95°C 10 s, 55°C 45 s). Under these conditions, no amplification was observed in the no-template and the no-reverse transcription controls. The specificity of the unique 100 bp amplification product was determined by melting curve analysis. ß-2 Microglobulin mRNA levels were used to normalize between samples. mRNA levels were quantitated by comparing experimental levels to standard curves generated using serial dilutions of a 1:1 mixture of WT and MMP-2 KO cDNAs. The primer sequences were as follows: ß-2mg 5'-tggtgcttgtctcactgacc-3' 5'-tatgttcggcttcccattct-3'; Mmp-9 5'-ctttgagtccggcagacaat-3' 5'-ttccagtaccaaccgtcctt-3'; Mmp-14 5'-tggcatccagcaactttatg-3' 5'-tttgggcttatctgggacag-3'; and Cd44 5'-atggctcatcatcttggcat-3' 5'-cagctttttcttctgcccac-3'.
Gelatin degradation assay
Enzymatic activity of the secreted MMP-9 present in conditioned media and the membrane-bound MMP-9 from PMA-activated (10 ng/mL for 6 h), 4% formaldehyde-fixed lymphocytes, were analyzed using a gelatin-FITC substrate-based assay, the EnzChek® Gelatinase/Collagenase Assay kit (Molecular Probes), according to the manufacturer’s directions. Experiments were carried out in triplicate using 106 fixed cells per well or 100 µL of conditioned media per well. Conditioned media was obtained as described in the zymography section.
Statistical analysis
Results were analyzed using Statview version 5 Software (SAS Institute, Inc., Cary, NC, USA), N-way ANOVA, and all pairwise multiple comparison procedures (Fisher’s PLSD, Bonferroni/Dunn, and Student-Newman-Keuls methods). For statistical analyses of the real-time PCR data, iCycler software (Bio-Rad) was used to determine the means and standard deviations of the triplicate samples. These data were then analyzed using the Statview software as described above.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
4ß1 (VLA-4) engagement (5
, 6)
. To further determine the role of MMP-2 in the etiology of multiple sclerosis, we induced EAE in MMP-2-deficient mice and studied the development of clinical symptoms. An unexpected earlier onset of disease and significantly higher clinical scores in the MMP-2–/– mice were consistently observed after immunization with MOG peptide (Fig. 1
A). The differences in the mean values among the groups (WT and KO, 0.385 vs. 0.846) were greater than would be expected by chance. An all pairwise multiple comparison procedure resulted in significant differences between the two groups from day 8 through 28 (for WT vs. KO, P=0.0003). Both groups of mice exhibited chronic disease, which is typical of MOG-induced EAE in the C57BL/6 model (6)
.
|
Similar results were obtained when EAE was induced in chimeric mice having MMP-2-deficient leukocytes in a WT background (Fig. 1D
). Here, too, the differences in the mean values among the groups (WT
WT and KOWT, 0.273 vs. 1.509) were greater than would be expected by chance. An all pairwise multiple comparison procedure resulted in significant differences between the two groups from days 12 through 32 (for WTWT vs. KOWT, P=0.0001). Again, both groups of mice exhibited chronic disease.
When the brains of the immunized mice were examined histologically at day 15 after immunization we observed a good correlation between the presence of perivascular and parenchymal mononuclear cell infiltrates in the MMP-2 KO mice and WT mice engrafted with MMP-2 KO bone marrow, but not in WT mice or WT mice engrafted with WT bone marrow (Fig. 1B, C, E, F
).
Thus, the absence of the MMP-2 enzymatic activity on bone marrow derived leukocytes seems to confer leukocytes with an enhanced ability for their successful entry to the CNS and to elicit a rapid and robust ascending paralysis which is the hallmark of this disease model.
MMP-2-deficient lymphocytes exhibit an increased transmigration through mouse brain endothelium in vitro
It is currently accepted that the migration of autoimmune T cells across the blood-brain barrier constitutes the key event in the initiation of EAE (3
, 7)
. Thus, based on the above in vivo result, we next asked whether MMP-2–/– lymphocytes display a higher transmigration rate across endothelial cells derived from brain microvasculature in a two-chamber system in vitro. Activated MOG-specific T cells from MMP-2-deficient mice displayed a significant increase in transmigration across TNF-stimulated bEnd.5 cell monolayers cultured on type IV collagen inserts (Fig. 2
A). Similar results were obtained with in vitro concanavalin A-activated lymphoblasts (our unpublished data). This increase in transmigration was not due to an increase in lymphocyte integrin receptor avidity, as reflected in cell adhesion assays to collagen type IV (Fig. 2B
), or an increase in ß1 and ß2 integrin surface expression, as assessed by flow cytometry analysis (Fig. 2C
). Other non-MMP, pericellular protease systems did not seem to be involved in this process. The level of the T cell activation marker dipeptidyl-peptidase IV (CD26) was analyzed by FACS and found to be the same in both types of lymphocytes (Fig. 2C
).
|
There is a compensatory exacerbated secretion of MMP-9 in MMP-2-deficient lymphocytes
In the CNS, MMPs have been implicated in the pericellular proteolysis of basal lamina components, contributing to the neuroinflammatory response (3
, 7)
. A notable basal up-regulation and activation of MMP-9 was observed in resting MMP-2–/– lymphocyte-derived conditioned media as assessed by gelatin zymography analysis (Fig. 3
A, media). An identical pattern of MMP-9 expression was observed when total cell lysates were used instead of cell supernatants by either zymography (Fig. 3A
, lysates) or Western blot (Fig. 3B
).
|
MMP-2 it is tethered to and activated at the cell membrane by a multimolecular complex composed of TIMP-2 and MMP-14 (4)
. In turn, MMP-14 can degrade several basal lamina constituents (4)
. Similarly, it is thought that MMP-9 activation takes place on the cell surface when it is bound to CD44, a transmembrane hyaluronan receptor (19)
. We studied the protein level of these key regulators of cell migration by Western and FACS analyses and found no significant changes in their expression (Fig. 3B, C
). The lack of appreciable MMP-2 activity in WT lymphocyte and endothelial cell preparations reflects the relative lower expression levels of MMP-2 compared with MMP-9 expression levels in these cells (18
, 20)
and the need to stop the zymographies at times when the differential MMP-9 activities could be discerned (Fig. 3A, D
). However, longer incubations of zymographies of WT lymphocyte and endothelial preparations did reveal MMP-2 expression in these samples (data not shown).
When dermal fibroblasts were isolated from MMP-2–/– mice, cultured and assessed for their MMP-9 expression by zymography, they did not exhibit any appreciable differences in expression compared with age- and sex-matched dermal fibroblasts similarly isolated from MMP-2+/+ mice (Fig. 3D
, left panel). In contrast, lung microvascular endothelial cells isolated from these mice, did exhibit an appreciable increase in MMP-9 expression (Fig. 3D
, right panel). These observations indicate that the molecular mechanisms involved in MMP-9 up-regulation in MMP-2 null mice are cell specific and may be restricted to cells of hematopoietic origin. Furthermore, in zymographies of plasma and sera harvested from MMP-2 KO mice, we did not note an appreciable increase in MMP-9 levels compared with WT mice (data not shown). This finding is consistent with our observations of a cell-specific (lymphocyte and endothelial cell) increased MMP-9 expression, which apparently is not robust enough to affect plasma/serum MMP-9 levels.
MMP-9 mRNA level is elevated in MMP-2-deficient lymphocytes
The protein expression of MMP-9, MMP-14, and CD44 correlated with the levels of their respective mRNA as assessed by real-time quantitative PCR. In freshly isolated MMP-2–/– lymphocytes, a 2.76-fold increase (P<0.0001) was found in the relative quantity of the MMP-9 mRNA whereas the relative amounts of MMP-14 and CD44 mRNAs remained essentially invariant (P=NS) (Fig. 4
).
|
The enhanced in vitro transmigration through purified basal lamina constituents by MMP-2-deficient lymphocytes is MMP-9 dependent
To obtain direct proof of the functional involvement of MMP-9 in the increased transmigration exhibited by MMP-2–/– lymphocytes, activated T cells were allowed to transmigrate toward a chemotactic gradient of CCL19 (ELC) or CCL21 (SLC) through inserts coated with the purified MMP-9 substrates laminin 1 and collagen type IV (Fig. 5
A). MMP-2–/– lymphocytes consistently showed a 2- to 3-fold increase in transmigration in the presence of these chemokines, which have been implicated in the recruitment into the CNS of myelin-specific T cells during EAE (21)
. To assess the possibility that the increased transmigration exhibited by the MMP-2–/– lymphocytes was due to increased surface expression of the common CCL19 and CCL21 receptor CCR7, we performed FACS analysis which revealed no significant differences in CCR7 expression in WT and MMP-2–/– lymphocytes (Fig. 5B
). This increase in the transmigration rate was abrogated by specific pharmacological MMP inhibitors (Fig. 5C
).
|
To further evaluate the functional role for MMP-9 in MMP-2–/– lymphocyte extravasation, concentrated conditioned media and PMA-stimulated, fixed T cells were assessed for their capacity to degrade FITC-conjugated gelatin (Fig. 6
). In this experiment, the samples containing MMP-2-deficient, T cell-derived MMP-9 exhibited an 
3-fold increase in gelatinolytic activity (open bars) compared with MMP-2+/+ T cells. The membrane-associated MMP-9 in PMA-stimulated MMP-2-deficient T cells (filled bars) displayed a 2-fold increase in gelatin hydrolysis.
|
MMP-2+/+ and MMP-2–/– mice have the same CD3+ subpopulation composition
The adaptive immune response mounted against particular antigens depends on the interplay between effector and regulatory T cell populations. EAE is a CD4+ Th1-induced autoimmune condition in which CD8+ T cells may also play a regulatory role (22)
. To assess the possibility that WT and MMP-2 null mice have different CD4+ and CD8+ T cell numbers, we quantified the T cell population in our animal model by FACS analysis and found identical counts of CD3+/CD4+ (61.7% vs. 60%, respectively) and CD3+/CD8+ (18.1% vs. 15%, respectively) double positive cells in MMP-2+/+ and MMP-2–/– mice (FACS analyses not shown). WT, MMP-2 KO, and MMP-14 +/– lymphocytes were found to exhibit identical proliferation profiles upon concanavalin-A stimulation (data not shown).
MMP-2 interactions with MT1-MMP may modulate expression levels of MMP-9
Since MMP-2 is known to complex with MT1-MMP (MMP-14) and TIMP-2, forming a ternary complex, and MT1-MMP is a transmembrane MMP having a modest but functional cytoplasmic tail (23
24
25
26)
, we addressed the possibility that formation of the MT1-MMP/TIMP-2/MMP-2 complex, in addition to facilitating activation of pro-MMP-2, could be involved in regulating the expression of other membrane tethered MMPs, namely, MMP-9. To address this question we used MT1-MMP+/– mice. Surprisingly, when splenocyte lysates derived from MT1-MMP+/– mice were assessed for MMP activity by gelatin zymography, we observed that the MT1-MMP+/– splenocytes exhibited a significant increase (4.7-fold, ±0.6, n=6, P=4.3x10–6) in MMP-9 activity (Fig. 7
A, B). Transmigration assays revealed increased transmigration rates of MT1-MMP+/– (MMP-14+/–) lymphoblasts through both type IV (3.97-fold, ±0.08, n=4, P
0.001) (Fig. 7C
) and type I collagen (data not shown) -coated BD BIOCOAT inserts® compared with MMP-14+/+ lymphoblasts. This is consistent with a functional consequence of the decreased expression of MMP-14.
|
Taken together, these findings indicate that, in MMP-2 and MT1-MMP-deficient mice, a compensatory MMP-9 up-regulation occurs that may account for the increased transmigration of lymphocytes, and thus the exacerbated inflammatory reaction observed in models of autoimmune conditions such as EAE (as evidenced in this report) and antibody-induced arthritis as previously reported (27)
.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
, 7)
. Lymphoblasts require high-affinity VLA-4 to be tethered on VCAM-1 expressing CNS microvessels under flow conditions in vivo (5
, 6
, 28)
. The crucial role of MMP-2 in the successful recruitment of autoimmune lymphocytes into the brain tissue is suggested by the fact that the sole expression of MMP-2 in T cells devoid of VLA-4 is responsible for inducing a rapid, although transient, clinical course in adoptively transferred EAE (6)
. In vitro these cells display increased transmigration rates and in vivo enter but do not establish residency in the CNS. Pharmacological inhibition of MMPs results in decreased T cell infiltration into brain parenchyma. To further elucidate the function of MMP-2 in the pathogenesis of autoimmune conditions, we induced EAE in MMP-2 null mice and noted earlier onset and higher clinical disease scores, a counterintuitive result considering the function of MMP-2 in degrading ECM substrates.
MMP-2–/– mice have smaller stature, blunted tumor-related angiogenesis, and abnormal lung alveolarization but are not been known to exhibit skeletal or joint abnormalities (8)
. To better characterize the leukocytic effects on the onset and severity of active EAE, we used chimeric mice in which bone marrow from MMP-2-deficient mice was engrafted onto WT mice, and WT bone marrow was engrafted onto WT mice. As observed in the MMP-2–/– mice, a significant early, robust, persistent EAE condition was noted in animals engrafted with MMP-2-deficient bone marrow in a WT background. Analogous results were obtained in vitro using a modified Boyden chamber transmigration assay using brain-derived endothelioma cells and resting T cells or either concanavalin A-activated or myelin-specific lymphoblasts. Thus, MMP-2–/– lymphocytes exhibit a higher rate of transendothelial migration that may account for the increased neuroinflammation observed in vivo in this study.
The leukocyte-endothelial cell interaction that takes place during tissue-specific homing occurs through sequential steps regulated by cell adhesion molecules, chemokine receptors and cell surface proteases (1)
. Therefore, the potential participation of key molecules belonging to these superfamilies was studied. The increased transmigration rate exhibited by MMP-2-deficient lymphocytes across endothelium and ECM constituents was not due to changes in the avidity of ß1 or ß 2 integrins. The membrane expression of these integrins was assessed by flow cytometry, whereas the affinity state of ß1 integrins was assessed by cell adhesion assays on collagen type IV. In both cases, no significant differences were found.
The profile of MMPs secreted in the media by T cells was also analyzed by gelatin zymography to better assess functionality. Herein, we report that the lack of MMP-2 in null mice is compensated by a significant MMP-9 up-regulation in lymphocytes. The MMP-9 overexpression in C57BL/6 mice was found to be sex and age independent in mice 8 to 12 wk old (our unpublished data; ref 29
). No other gelatinases were detected with this technique, neither in cytokine-, chemokine-, concanavalin A- or fibronectin-stimulated T cells (our unpublished data). When MMP-2-deficient dermal fibroblasts and lung-derived microvascular endothelial cells were isolated, cultured and assayed for MMP-9 expression, the fibroblasts exhibited no appreciable changes in MMP-9 expression compared with MMP-2+/+ fibroblasts, but the MMP-2–/– endothelial cells exhibited a marked increase in MMP-9 expression. These findings suggest that the dysregulation of MMP-9 expression in the MMP-2 null mouse is cell-specific and not a pan-cellular phenomenon. This is consistent with the concept that the complex phenotype observed in the MMP-2 null mouse is the result of the differential responses of many distinct cell types, with the significant early, robust persistent EAE being mediated by the dysregulation of MMP-9 expression in bone marrow derived leukocytes.
MMP-9 cleaves myelin basic protein in vitro (4)
and high levels of this enzyme correlate spatially and temporally with the EAE symptomatology in susceptible mouse models (30)
. While adult MMP-9 null mice exhibit similar EAE susceptibility and severity after induction of the disease, younger MMP-9–/– mice exhibit a reduced susceptibility to EAE induction, suggesting a causal relationship between the expression of MMP-9 and the susceptibility and severity of EAE. These observations indicate that although leukocyte MMP-9 expression is a critical regulatory factor in the outcome of chronic inflammatory conditions, other molecules are likely involved in these processes. In vitro, we found that the MMP-9 overexpression was solely responsible for the exacerbated transmigration rates observed in MMP-2–/– lymphocytes, and the blockade of MMP-9 activity restored the baseline transmigration level exhibited by MMP-2+/+ T cells.
However, our results do not rule out that other phenomena may act in conjunction with the MMP-9 compensation, aiding or enhancing the EAE increased susceptibility observed in the MMP-2 null mice. Thus, while the up-regulation of MMP-9 may indeed be responsible for partially mediating the observed phenotype it is possible that the loss of MMP-2 expression may play a significant role in the early onset and severity of the induced EAE in the MMP-2-deficient mice due to pleiotropy. Indeed, MMP-2 has been recently reported to process and inactivate several chemokines giving rise to inflammation antagonists (31
32
33
34
35)
. This negative feedback of the inflammatory reaction may be absent or diminished in our animal model and may be of crucial importance in the phenotype described herein, as well as in the antibody-induced arthritis model recently reported (27)
.
MMP-2–/– lymphocytes could be more prone to transmigration by presenting on their surface higher expression levels of the chemokine receptors involved in increasing T cell adhesion to the CNS vasculature. So far, the only chemokines known to be expressed by the brain microvasculature in vivo are CCL19 and CCL21 (21)
. Thus, their common receptor, CCR7, and the alternative CCL21 receptor CXCR3, could exhibit an altered expression in MMP-2-deficient leukocytes, leading to increased transmigration and earlier onset and more severe disease. However, this is not the case for CCR7 as we showed in our study.
Other possibilities including proteolytic compensation by ADAM family members and other metalloproteinases, other classes of proteases, as well as dysregulation of amoeboid movement (36
37
38)
are worthy of consideration and future investigation.
Recently, a hereditary deficiency in MMP-2 has been described in middle eastern families who develop a vanishing bone disorder (39
, 40)
. In this disorder a missense mutation in the MMP-2 gene affects the MMP-2 prodomain while a nonsense (stop) mutation affects the catalytic, the fibronectin type II-like and the hemopexin/TIMP-2 binding domains. Because only small numbers of unaffected and affected individuals of the three families were studied, it is unclear whether the absence of MMP-2 has an effect on MMP-9 expression in humans. If enough individuals could be tested, perhaps one would see a correlation of the loss of MMP-2 expression observed with nonsense mutation with increased MMP-9 activity (which should affect MMP-2 binding to MMP-14, thus affecting MMP-9 expression), but not with the missense mutation (which should not affect MMP-2 binding to MMP-14).
Recent studies elucidating complex, dynamic interactions between MT1-MMP and CD44 (41)
are consistent with the possibility that such interactions could be modulated by the absence or presence of MMP-2, which could ultimately lead to modulation of MMP-9 expression. The levels of CD44 and MT1-MMP were not found to be significantly changed in MMP-2–/–-deficient lymphocytes, neither at protein nor at mRNA levels. MMP-9 was also found to be dramatically up-regulated in MT1-MMP+/– mice by zymography analysis. This MMP compensation phenotype (Fig. 8
) has not been previously reported in these null animals to our knowledge and may be restricted to MMPs that either are tethered to the cell surface by transmembrane proteins (MMP-2/MT1-MMP and MMP-9/CD44), or are themselves membrane-spanning proteins (MT1-MMP). These possibilities may be of importance in the development of lymphocyte infiltrates during EAE and deserve further investigation, as does the potential role of MMP-2 as a signaling ligand modulating the expression of MMP-9 via the MT1-MMP/TIMP-2 complex.
|
|
ACKNOWLEDGMENTS |
|---|
Received for publication June 7, 2004. Accepted for publication August 2, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. R. Folgueras, A. Fueyo, O. Garcia-Suarez, J. Cox, A. Astudillo, P. Tortorella, C. Campestre, A. Gutierrez-Fernandez, M. Fanjul-Fernandez, C. J. Pennington, et al. Collagenase-2 Deficiency or Inhibition Impairs Experimental Autoimmune Encephalomyelitis in Mice J. Biol. Chem., April 4, 2008; 283(14): 9465 - 9474. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Tian, J. Lappalainen, M. Autero, S. Hanninen, H. Rauvala, and C. G. Gahmberg Shedded neuronal ICAM-5 suppresses T-cell activation Blood, April 1, 2008; 111(7): 3615 - 3625. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Segarra, A. Garcia-Martinez, M. Sanchez, J. Hernandez-Rodriguez, E. Lozano, J. M Grau, and M. C Cid Gelatinase expression and proteolytic activity in giant-cell arteritis Ann Rheum Dis, November 1, 2007; 66(11): 1429 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Tian, M. Stefanidakis, L. Ning, P. Van Lint, H. Nyman-Huttunen, C. Libert, S. Itohara, M. Mishina, H. Rauvala, and C. G. Gahmberg Activation of NMDA receptors promotes dendritic spine development through MMP-mediated ICAM-5 cleavage J. Cell Biol., August 9, 2007; 178(4): 687 - 700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Dell'Aica, R. Niero, F. Piazza, A. Cabrelle, L. Sartor, C. Colalto, E. Brunetta, G. Lorusso, R. Benelli, A. Albini, et al. Hyperforin Blocks Neutrophil Activation of Matrix Metalloproteinase-9, Motility and Recruitment, and Restrains Inflammation-Triggered Angiogenesis and Lung Fibrosis J. Pharmacol. Exp. Ther., May 1, 2007; 321(2): 492 - 500. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-A. Renault and D. W. Losordo The Matrix Revolutions: Matrix Metalloproteinase, Vasculogenesis, and Ischemic Tissue Repair Circ. Res., March 30, 2007; 100(6): 749 - 750. [Full Text] [PDF]
|
|
|
|
 |
|
 K. Nakaji, M. Ihara, C. Takahashi, S. Itohara, M. Noda, R. Takahashi, and H. Tomimoto Matrix Metalloproteinase-2 Plays a Critical Role in the Pathogenesis of White Matter Lesions After Chronic Cerebral Hypoperfusion in Rodents Stroke, November 1, 2006; 37(11): 2816 - 2823. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-Y. C. Hsu, R. McKeon, S. Goussev, Z. Werb, J.-U. Lee, A. Trivedi, and L. J. Noble-Haeusslein Matrix Metalloproteinase-2 Facilitates Wound Healing Events That Promote Functional Recovery after Spinal Cord Injury J. Neurosci., September 27, 2006; 26(39): 9841 - 9850. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Garg, M. Rojas, A. Ravi, K. Bockbrader, S. Epstein, M. Vijay-Kumar, A. T. Gewirtz, D. Merlin, and S. V. Sitaraman Selective Ablation of Matrix Metalloproteinase-2 Exacerbates Experimental Colitis: Contrasting Role of Gelatinases in the Pathogenesis of Colitis J. Immunol., September 15, 2006; 177(6): 4103 - 4112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Agrawal, P. Anderson, M. Durbeej, N. van Rooijen, F. Ivars, G. Opdenakker, and L. M. Sorokin Dystroglycan is selectively cleaved at the parenchymal basement membrane at sites of leukocyte extravasation in experimental autoimmune encephalomyelitis J. Exp. Med., April 17, 2006; 203(4): 1007 - 1019. [Abstract] [Full Text] [PDF]
|
|
|
|
