help button home button
AJRCMB
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on March 27, 2003, doi:10.1165/rcmb.2003-0034OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2003-0034OCv1
29/3/283    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Owen, C. A.
Right arrow Articles by Shapiro, S. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Owen, C. A.
Right arrow Articles by Shapiro, S. D.
American Journal of Respiratory Cell and Molecular Biology. Vol. 29, pp. 283-294, 2003
© 2003 American Thoracic Society
DOI: 10.1165/rcmb.2003-0034OC

Inducible Expression of Tissue Inhibitor of Metalloproteinases–Resistant Matrix Metalloproteinase-9 on the Cell Surface of Neutrophils

Caroline A. Owen, Zhuma Hu, Brooke Barrick and Steven D. Shapiro

Division of Pulmonary and Critical Care Medicine, Brigham and Women's Hospital, Boston, Massachusetts; and Department of Internal Medicine, University of Utah Health Sciences Center, Salt Lake City, Utah

Address correspondence to: Caroline A. Owen, M.D., Ph.D., Division of Pulmonary and Critical Care Medicine, Brigham and Women's Hospital, 8th Floor Thorn Building, 75 Francis Street, Boston, MA 02115. E-mail: cowen{at}rics.bwh.harvard.edu


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Matrix metalloproteinase (MMP)-9 secreted by activated polymorphonuclear neutrophils (PMN) may play roles in mediating lung injury by degrading extracellular matrix proteins. However, the mechanisms by which MMP-9 retains activity in the presence of tissue inhibitors of metalloproteinases (TIMPs) are not known. We show that MMP-9 is also expressed on the cell surface of PMN, and proinflammatory mediators induce up to 10-fold increases in cell surface expression of MMP-9. Stimulated human PMN express active forms of cell surface MMP, which cleave the MMP substrate, McaPLGLDpaAR. Loss-of-function studies employing PMN from mice genetically deficient in MMP-9 (MMP-9-/-) demonstrate that membrane-bound MMP-9 contributes substantially to MMP-mediated surface-bound cleavage of McaPLGLDpaAR (~ 50%) and gelatin (~ 70%) by stimulated PMN. Like soluble MMP-9, membrane-bound MMP-9 cleaves McaPLGLDpaAR (Kcat/KM = 82,000 M-1s-1), gelatin, type IV collagen, elastin, and {alpha}1-proteinase inhibitor. However, in contrast to soluble MMP-9, membrane-bound MMP-9 is substantially resistant to inhibition by TIMPs. The IC50 for inhibition of membrane-bound MMP-9 by TIMP-1 and TIMP-2 are ~ 21-fold and ~ 68-fold higher, respectively, than those for inhibition of soluble MMP-9. The binding of MMP-9 to the plasma membrane of PMN enables it to evade inhibition by TIMPs, and thereby may alter the pericellular proteolytic balance in favor of extracellular matrix degradation. Membrane-bound MMP-9 on PMN may play pathogenetic roles in inflammatory lung diseases.

Abbreviations: 4-aminophenylmercuric acetate, APMA • N-formyl-leucyl-methionyl-phenylalanine, fMLP • interleukin, IL • lipopolysaccharide, LPS • 7-Methoxycoumarin-4-yl)-Acetyl-Pro-Leu-Gly-Leu-(3-[2,4-dinitrophenyl]-L2,3-diaminopropionyl)-Ala-Arg-NH2, McaPLGLDpaAR • matrix metalloproteinase(s), MMP • membrane-type MMP(s), MT-MMP • neutrophil gelatinase B-associated lipocalin, NGAL • platelet-activating factor (1-O-hexadecyl-2-acetyl-sn-glycero-3-phosphorylcholine), PAF • polymorphonuclear neutrophils, PMN • tissue inhibitor of metalloproteinases, TIMP • tumor necrosis factor-{alpha}, TNF-{alpha}


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Matrix metalloproteinase-9 (MMP-9, EC3.4.24.35) is a member of the MMP family of enzymes, which mediate connective tissue turnover during physiologic and pathologic processes (1). MMP-9 has a broad spectrum of catalytic activity against extracellular matrix components including denatured collagens (gelatins), type V and IX collagens, elastin, type IV collagen, and other basement membrane components (1). MMP-9 can also cleave non-matrix proteins, with important local consequences during the inflammatory response. For example, MMP-9 proteolytically processes interleukin (IL)-1ß precursor (2) and IL-8 (3), generating forms of the mediators with increased biologic activity. In contrast, MMP-9 cleaves and inactivates serpins (4), mature IL-1ß (5), connective tissue–activating peptide-III, platelet factor-4, and GRO-{alpha} (3). Thus, MMP-9 may play important physiologic roles in lung extracellular matrix remodeling and repair, and in regulating the lung inflammatory response to injury (6). However, MMP-9 has also been implicated in the pathogenesis of various lung diseases including chronic obstructive pulmonary diseases (7–9), asthma (10–12), immune complex–mediated acute lung injury (13), idiopathic pulmonary fibrosis, and alveolar bronchiolization (14, 15), as well as in diseases affecting other organ systems (1, 16, 17). MMP-9 also promotes tumor cell invasiveness and metastasis (1).

MMP-9 is expressed by many cell types, including polymorphonuclear neutrophils (PMN) (1), but PMN-derived MMP-9 differs from MMP-9 expressed by other cell types in two major ways. First, mature PMN do not synthesize MMP-9 de novo. Rather, MMP-9 is produced during the late stages of maturation of PMN precursors in the bone marrow. It is stored in its latent form (pro–MMP-9) within the gelatinase granules of PMN before its release into the extracellular space following PMN activation (18). Second, pro–MMP-9 is not released from activated PMN complexed to tissue inhibitor of metalloproteinases-1 (TIMP-1). Rather, MMP-9 is released from PMN in three different forms: 92-kD monomers, 200-kD homodimers, and 120-kD complexes of MMP-9 covalently bound to neutrophil gelatinase-B–associated lipocalin (NGAL), a 25-kD member of the lipocalin family of transport proteins (19). Following its release from PMN, pro–MMP-9 must be activated in the extracellular space to attain full catalytic activity. Activation of pro–MMP-9 is induced in vitro by mercurial compounds via conformational change and auto-cleavage. Although the physiologic activators are unclear, in vitro studies indicate that other proteinases and reactive oxygen species can activate pro–MMP-9 (1).

Uncertainties remain about the mechanism(s) permitting MMP-9 to retain its activity in the extracellular space following its release from PMN, because the extracellular space contains effective inhibitors of MMP-9, including TIMPs (1). PMN can degrade proteins in vitro, even when cells are bathed in media containing physiologic proteinase inhibitors (20). Inhibitors cannot eliminate the proteolytic activity associated with activated PMN; they only confine proteolytic activity to the immediate pericellular environment of PMN. These observations indicate that proteolytic events that occur at or near the cell surface are important in facilitating proteolysis in the presence of proteinase inhibitors.

Here, we show that MMP-9 is rapidly expressed on the cell surface of human and murine PMN in a highly inducible manner. Whereas cell surface–bound MMP-9 has catalytic activity similar to that of soluble MMP-9, membrane-bound MMP-9 differs from soluble MMP-9 in that it is substantially resistant to inhibition by TIMP-1 and TIMP-2. Our data provide novel insights into the mechanisms by which PMN-derived MMP-9 retains its activity in the extracellular space, and contributes to physiologic and pathologic processes of PMN in the lung.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
Goat anti-rabbit Fab2-Alexa Fluor 546 was obtained from Molecular Probes Inc., Eugene, OR. Purified human TIMP-1 and TIMP-2, polyclonal rabbit anti-human MMP-9 IgG (AB805), polyclonal rabbit anti-murine MMP-9 (AB19047) and gelatinase and type IV collagen assay kits were purchased from Chemicon International Inc., Temecula, CA. Biotinylated rat anti–Gr-1 antibody was obtained from PharMingen, San Diego, CA. Quenched FITC-conjugated gelatin was purchased from Molecular Probes. (7-Methoxycoumarin-4-yl)-Acetyl-Pro-Leu-Gly-Leu-(3-[2,4-dinitrophenyl]-L2,3-diaminopropionyl)-Ala-Arg-NH2 (McaPLGLDpaAR) was purchased from CalBiochem Novabiochem Corp., San Diego, CA. RS113456, RS104210–007, and purified human MMP-9 were generously provided by Roche Bioscience, Palo Alto, CA.

Human PMN Isolation and Activation
Human PMN were isolated from peripheral blood of healthy donors (21), and then incubated for 30 min at 37°C in HBSS containing 10 mM HEPES (pH 7.4) with or without varying concentrations of 1-O-hexadecyl-2-acetyl-sn-glycero-3-phosphorylcholine (PAF), tumor necrosis factor-{alpha} (TNF-{alpha}), lipopolysaccharide (LPS) from E. coli 0111:B4, or N-formyl-leucyl-methionyl-phenylalanine (fMLP). We also primed PMN at 37°C for varying times with varying concentrations of LPS, PAF, or TNF-{alpha}, and then activated the PMN for 30 min with 10-8 M fMLP. To terminate the assays, PMN were fixed in phosphate-buffered saline containing 3% paraformaldehyde and 0.5% glutaraldehyde (pH 7.4 [21]).

Murine PMN Isolation and Activation
Murine PMN were isolated from the bone marrow of mice genetically deficient in MMP-9 (MMP-9-/-) or from wild-type mice (MMP-9+/+) in the same genetic background (129/SvEv [22]) by positive selection using biotinylated rat anti–Gr-1 antibody and MACs Microbeads conjugated to mouse anti-biotin (Miltenyi Biotec Inc., Auburn, CA [23]). PMN preparations were > 85% pure, as assessed by differential counting of Wright-stained cytocentrifuge preparations. Murine PMN were activated at 37°C for 15 min with LPS (100 ng/ml) or PAF (10-7 M) followed by fMLP (10-7 M for 30 min). PMN were fixed, washed, and immunostained for cell surface murine MMP-9 as outlined below.

Immunostaining of PMN, Image Analysis, and Confocal Microscopy
PMN were immunostained (21) for cell surface MMP-9 using rabbit anti-human MMP-9 IgG (AB805), rabbit anti-murine MMP-9 (AB19047) or rabbit IgG, as a control (1 µg/106 cells), followed by goat anti-rabbit Fab2 Alexa Fluor 546 (diluted 1:500). AB805 is a polyclonal IgG generated against a peptide region at the N-terminal region of the catalytic domain of MMP-9. Western blot analysis confirmed that AB805 recognizes both pro- and active MMP-9, and that it had no cross-reactivity with purified human MMP-1, -2, -3, -8, or -12. Cell surface immunofluorescence was quantified using incident light epifluorescence microscopy and image analysis (MetaMorph software; Universal Imaging Inc., West Chester, PA) as described previously (21). The data were corrected for nonspecific staining, as described previously (21).

Cell surface localization of MMP-9 was confirmed by staining unstimulated and TNF-{alpha}–primed and fMLP-activated PMN by the immunogold method (21) using AB805 or rabbit IgG followed by goat anti-rabbit IgG conjugated to 20 nm diameter colloidal gold particles. In addition, cells stained for MMP-9 using the immunofluorescence method above were examining with a Leica TCSNT confocal laser scanning microscope (Leica Inc., Exton, PA) fitted with air-cooled argon and krypton lasers. Fluorescent confocal micrographs were recorded under fluorescent imaging by exposing cells to 568 nm light attenuated by an acusto tunable optical filter using a long-pass 590 nm filter to detect the Alexa 546-labeled MMP-9.

Catalytic Activity and Susceptibility to Inhibition of MMP Activity Expressed on the Surface of Activated Human PMN
To assess whether activated human PMN express MMP in an active form, PMN were optimally activated (incubated for 15 min with 10-8 M PAF and then for 30 min with 10-8 M fMLP), washed, and then total cell-associated MMP activity was quantified using McaPLGLDpaAR, a general, sensitive, quenched fluorescent substrate for MMP (24). Viable, activated cells (106 cells in 100 µl of 0.05 M Tris buffer containing 0.15 M NaCl and 0.02M CaCl2; pH 7.4) were preincubated in triplicate for 10 min at 37°C with and without 10 µM RS113456 (a general, hydroxamate inhibitor of MMP), and then incubated for up to 60 min at 37°C with 1.8 µM McaPLGLDpaAR. Cleavage of the substrate was quantified in cell-free supernatant samples by fluorimetry ({lambda}ex 328 nm, {lambda}em 393 nm). MMP activity was quantified as the RS113456-inhibitable cleavage of McaPLGLDpaAR. To assess the proportion of the total cell-associated MMP activity due to free release of active MMP from cells, viable activated PMN were incubated under identical conditions in the absence of substrate, cell-free supernatant samples were harvested at intervals, 1.8 µM McaPLGLDpaAR was added, and RS113456-inhibitable hydrolysis of the substrate was quantified.

To assess the effectiveness of inhibitors against the McaPLGLDpaAR hydrolyzing activity associated with the cells, viable activated PMN (106 cells/assay) were also incubated in triplicate for 15 min at 37°C with or without: (i) 1 mM 1,10-phenanthroline (a general, synthetic inhibitor of MMP); (ii) 10 µM RS104210–007 (a general, hydroxamate inhibitor of MMP); (iii) 100 µM 4-(2-aminoethyl)-benzenesulfonyl fluoride (AEBSF, an inhibitor of serine proteinases); (iv) 1 µM TIMP-1; (v) 1 µM TIMP-2; or (vi) AEBSF (100 µM for 5 min) followed by TIMP-1 (1 µM) for 15 min. Residual MMP activity was then quantified using McaPLGLDpaAR, as described above.

Catalytic Activity of MMP-9 on the Cell Surface of Murine PMN
PMN isolated from MMP-9-/- and MMP-9+/+ mice were activated with PAF and fMLP, fixed, and then cells from both genotypes (106/assay) were pre-incubated at 37°C for 20 min in triplicate with and without 1,10-phenanthroline (1 mM, a synthetic inhibitor of MMP). Cells were then incubated for varying times with 1.8 µM McaPLGLDpaAR. Total cell surface-bound MMP activity was quantified as the 1,10-phenanthroline inhibitable, McaPLGLDpaAR-cleaving activity in cell-free supernatant samples using fluorimetry (Hitachi F2500 fluorescence spectrophotometer {lambda}ex 328 nm, {lambda}em 393 nm; Hitachi Ltd., Tokyo, Japan). To quantify MMP-mediated gelatinase activity bound to the cell surface of activated PMN, PAF- and fMLP-activated cells from MMP-9-/- and MMP-9+/+ mice (3 x 106) were preincubated in triplicate for 20 min at 37°C with 1 mM PMSF (to inactivate cell surface–bound serine proteinases that can cleave gelatin [11]) and with and without an MMP inhibitor (1 mM 1,10-phenanthroline). Cells were then incubated at 37°C with 50 µg/ml quenched FITC-conjugated gelatin, and MMP-mediated gelatinase activity was quantified as the 1,10-phenanthroline-inhibitable cleavage of gelatin-FITC in cell-free supernatant samples by fluorimetry ({lambda}ex 490 nm, {lambda}em 520 nm).

Human pro–MMP-9 Activation and Active Site Titration
Purified human pro–MMP-9 was incubated at 37°C for up to 4 h with 1 mM 4-aminophenylmercuric acetate (APMA). Complete activation of pro–MMP-9 was confirmed by SDS-PAGE. Active site titration of MMP-9 was performed using TIMP-2 (25).

Cleavage of McaPLGLDpaAR by Membrane-Bound MMP-9 on Human PMN
To study the catalytic activity and efficiency, and substrate specificity of cell surface–bound human MMP-9 (in isolation and in a quantitative manner), we bound exogenous MMP-9 to unstimulated PMN, as described previously for binding of proteinase-3 to PMN (26). Briefly, PMN (2 x 106 cells) were incubated with and without APMA-activated and active-site-titrated MMP-9 (9–300 nM), and then cells were washed and fixed. It was not possible to incubate PMN with concentrations of MMP-9 greater than 300 nM because they adversely affected PMN viability. PMN which bound MMP-9 (2 x 106 cells), 2 x 106 control PMN (which had no exogenous MMP-9 bound to their cell surface), or assay standards of soluble, active MMP-9 (3–400 ng) were incubated in Tris buffer at 37°C for 30 min with 1.8 µM McaPLGLDpaAR. Substrate cleavage was quantified in cell-free supernatant fluids by fluorimetry. The assay was calibrated using assay standards of either soluble, active-site-titrated MMP-9, or Mca-Pro-Leu-OH, and cell surface–bound MMP-9 activity on PMN was expressed as ng/106 PMN, or pmol substrate cleaved, respectively.

Effect of Fixatives on the Catalytic Activity of Membrane-Bound MMP-9
To assess whether the PMN fixation procedure activates pro–MMP-9, we incubated 200 ng of pro–MMP-9 for 3 min at 4°C in the presence and absence of fixatives at the same concentrations used for PMN fixation, then measured the activity of the samples (along with 200 ng APMA-activated MMP-9) against McaPLGLDpaAR, as outlined above. We also bound exogenous MMP-9 to PMN, then incubated aliquots of the cells with and without fixatives for 3 min at 4°C, washed the cells, then quantified cell surface–bound MMP-9 activity using McaPLGLDpaAR.

Kcat/Km Determinations
MMP-9 was bound to PMN, and membrane-bound MMP-9 activity was quantified using McaPLGLDpaAR and assay standards of soluble, active-site-titrated MMP-9, as described above. Hydrolysis of McaPLGLDpaAR by 31–500 pM soluble MMP-9 or membrane-bound MMP-9 was measured in a total volume of 250 µl of Tris buffer as the increase in fluorescence ({lambda}ex 328 nm, {lambda}em 393 nm) over 6 min at 23°C, using 1.5 µM McaPLGLDpaAR, which fulfills the condition of [S] << KM required to permit the direct determination of kcat/KM (25).

Cleavage of Biologic Substrates by Membrane-Bound MMP-9
MMP-9 bound to the surface of PMN, control PMN (both at 6 x 106/assay), soluble MMP-9 (1 µg/assay) were incubated for up to 6 h at 37°C with FITC-conjugated particulate elastin (10 mg/ml, 200–400 mesh size) in Tris buffer, and solubilization of elastin was quantified in cell-free supernatant fluids by fluorimetry (27). MMP-9 bound to PMN, control PMN (both at 2 x 106 cells/assay), and soluble MMP-9 (4–500 ng) were incubated for 90 min at 37°C with biotinylated gelatin (denatured type I collagen) in Tris buffer. MMP-9 bound to PMN, control PMN (both at 5 x 106 cells/assay), and soluble MMP-9 standards (6.25–400 ng) were incubated for 4 h at 42°C with FITC-conjugated type IV collagen (83 µg/ml). Cell-free supernatant samples were then assayed for gelatinase or type IV collagenase activity, using commercially available kits. MMP-9 bound to PMN, control PMN (2 x 107/assay), or soluble MMP-9 (1 µg) were also incubated for 8 h at 37°C with 5 µg purified {alpha}1-proteinase inhibitor in Tris buffer. Cleavage of {alpha}1-proteinase inhibitor was assessed in cell-free supernatant samples subjected to 20% SDS-PAGE under reducing conditions.

Susceptibility of Membrane-Bound MMP-9 to Inhibition
MMP-9 was bound onto PMN, and its activity was quantified using McaPLGLDpaAR, and soluble, active-site-titrated MMP-9 standards. Soluble MMP-9 (10 nM) or MMP-9 bound to PMN (10 nM) were then incubated at 37°C for 15 min in Tris buffer with and without: (i) 1 mM 1,10-phenanthroline; (ii) 0.1 nM-1 µM RS113456; (iii) 15.6 nM-1 µM TIMP-1; (iv) 15.6 nM-1 µM TIMP-2; and (v) 1 mM PMSF. Residual MMP-9 activity was quantified in cell-free supernatant samples using 1.8 µM McaPLGLDpaAR. IC50 values were determined by nonlinear regression analysis using SigmaStat (SPSS Inc., Chicago, IL).

Statistics
The results for paired and unpaired data were compared using the Student's t test for parametric data and the Mann-Whitney rank sum test for nonparametric data; P values less than 0.05 were considered significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Inducible Expression of MMP-9 on the Cell Surface of Human and Murine PMN
Human PMN were incubated for 30 min with or without varying concentrations of fMLP, and cell surface–bound MMP-9 on PMN was quantified by immunostaining and image analysis. Unstimulated PMN expressed minimal amounts of MMP-9 on their cell surface (Figures 1 and 3A). Exposure of PMN to fMLP induced modest (up to 4-fold), concentration-dependent increases in cell surface MMP-9 expression (Figure 1A). LPS, PAF, and TNF-{alpha} also induced modest (2- to 4-fold) increases in cell surface expression of MMP-9 on PMN (Figures 1B–1D). However, priming of human PMN with these proinflammatory mediators, followed by activation with the optimal concentration of fMLP induced synergistic (up to 10-fold) increases in cell surface expression of MMP-9 on PMN (Figures 2A and 3B). These priming effects of proinflammatory mediators on cell surface expression of MMP-9 on human PMN in response to activation with fMLP, were concentration-dependent and time-dependent (Figures 2B and 2C).



View larger version (41K):
[in this window]
[in a new window]
  		Figure 1. Proinflammatory mediators upregulate cell surface expression of MMP-9 on human PMN. PMN were incubated for 30 min at 37°C with or without varying concentrations of fMLP (A), TNF-{alpha} (B), LPS (C), or PAF (D). Cells were immunostained for cell surface–bound MMP-9, cell surface fluorescence was quantified, and the data were corrected for nonspecific staining. The data are expressed as a percent of the mean integrated fluorescence value of unstimulated PMN (mean ± SEM; n = 100–200 cells). *P = 0.002; **P < 0.001 compared with unstimulated cells.

 


View larger version (34K):
[in this window]
[in a new window]
  		Figure 2. Cytokines prime human PMN for activation with fMLP for cell surface expression of MMP-9. In A–C, open bars are PMN, which were incubated for 30 min at 37°C with or without 100 ng/ml LPS, 10-8 M PAF, 100 U/ml TNF-{alpha}, or 10-8 M fMLP. In A–C, hatched bars are PMN which were variably primed with cytokines, then activated for 30 min with 10-8 M fMLP. In A, PMN were primed for 15 min with the above-listed concentrations of LPS, PAF, or TNF-{alpha}. In B, PMN were primed for 15 min with varying concentrations of TNF-{alpha}. In C, PMN were primed for varying times with 10-8 M PAF. Cell surface expression of MMP-9 was quantified, and the data were corrected for nonspecific staining. The data are expressed as a percent of the mean integrated fluorescence value of unstimulated PMN (mean ± SEM; n = 100–200 cells). *P < 0.001 compared with unstimulated PMN; **P < 0.001 compared with PMN incubated with agonists alone.

 



View larger version (293K):
[in this window]
[in a new window]
  		Figure 3. Immunostaining for cell surface MMP-9 on human PMN. In A and B, PMN were incubated for 30 min without agonists (A), or primed with PAF, then activated with fMLP (B). Cells were immunostained with rabbit anti-human MMP-9, followed by goat anti-rabbit IgG conjugated to Alexa-546. Cytocentrifuge preparations were examined by phase contrast (left panels) and incident-light fluorescence microscopy (right panels). Original magnification: x1,000. In C, PMN were primed with TNF-{alpha} and activated with fMLP (left) or incubated without agonists (right), then incubated with rabbit anti–MMP-9, followed by goat anti-rabbit conjugated to colloidal gold particles (20 nM diameter). Cytocentrifuge preparations were examined by reflection polarization microscopy to visualize the gold particles. Original magnification: x500. In D, PMN were activated and immunostained, as in B, then examined by confocal microscopy. Arrows indicate MMP-9 localized to the leading edge of polarized PMN. Original magnification: x1,000.

 
Immunostaining of unstimulated and activated murine PMN with an antibody to murine MMP-9 also demonstrated inducible expression of MMP-9 on their cell surface. Murine PMN activated with LPS and fMLP expressed ~ 4-fold more MMP-9 on their cell surface when compared with unstimulated murine PMN (386.2 ± SEM 36.7 versus 98.6 ± SEM 14.6 fluorescence units, respectively; P < 0.001).

To confirm that our immunostaining procedure detects cell surface–bound MMP-9 (rather than intracellular antigen), we also stained human PMN for MMP-9 by the immunogold method. This technique employs a second antibody conjugated to colloidal gold particles which are too large to penetrate cells, thereby localizing the antigen to the cell surface. Activated human PMN stained for MMP-9 were associated with numerous gold particles (Figure 3C, left panel). Unstimulated PMN (Figure 3C, right panel) and activated cells incubated with control antibody (not shown) were devoid of gold particles, confirming that MMP-9 is expressed on the cell surface of PMN activated with biologic mediators. Examination of chemoattractant-activated PMN using confocal microscopy showed that the distribution of MMP-9 on their cell surface is focal, and localized on the leading edge of many (but not all) polarized PMN (Figure 3D).

Activated Human PMN Express Catalytically Active Cell Surface–Bound MMP
Studies of the catalytic activity of human MMP-9 endogenously expressed on the cell surface of activated PMN are not feasible because: (i) activated human PMN express other MMP, including a membrane-type-MMP (MT6-MMP [28, 29]) and MMP-8 (C. A. Owen and S. D. Shapiro, unpublished observations), both of which overlap with MMP-9 in their substrate specificity; and (ii) substrates or inhibitors specific for MMP-9 are not available. Accordingly, we investigated whether activated human PMN express cell surface–bound MMP in catalytically active forms. We incubated viable activated human PMN with McaPLGLDpaAR (a general, sensitive, quenched fluorescent substrate for MMP) in the presence and absence of RS112456 (a hydroxamate inhibitor of MMP). The cells were associated with progressive MMP-mediated cleavage of McaPLGLDpaAR (Figure 4A). To assess the proportion of the total McaPLGLDpaAR-cleaving activity which was due to freely released forms of active MMP from stimulated PMN, viable activated PMN were also incubated at 37°C in buffer alone, and cell-free supernatant samples were harvested and assayed for MMP activity using McaPLGLDPaAR. Cell-free supernatant samples had less than 5% of the total PMN-mediated MMP activity at all time points tested (Figure 4A). Thus, active, extracellular forms of MMP are associated with stimulated PMN, and > 95% of this activity is associated in a stable manner with the PMN cell surface.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 4. Activated human PMN express MMP activity bound to their cell surface. In A, human PMN were optimally activated to induce cell surface expression of MMP-9 (10-8 M PAF for 15 min followed by 10-8 M fMLP for 30 min), then viable PMN were incubated (106/assay) in triplicate for up to 60 min at 37°C with 1.8 µM McaPLGLDpaAR in the presence and absence of 10 µM RS113456 (a hydroxamate inhibitor of MMP). Total MMP activity associated with the PMN (open circles) was quantified as the RS113456-inhibitable hydrolysis of McaPLGLDpaAR. To quantify the amount of MMP activity freely released by activated PMN, cells were incubated under identical conditions in the absence of substrate. Cell-free supernatant samples were harvested, and RS113456-inhibitable hydrolysis of McaPLGLDpa was quantified following the addition of the substrate (filled squares). Data are mean ± SD. Note that there is progressive MMP-mediated cleavage the substrate by the cells over time, but minimal free release of active MMP by PMN during the assay. In B, human PMM were activated then incubated in triplicate for 15 min at 37°C with and without: (i) 1,10-phenathroline (o-phen, 1 mM); (ii) RS104210–007 (RS104210, 10 µM); (iii) TIMP-1 (1 µM); (iv) TIMP-2 (1 µM); or (v) AEBSF (100 µM). Residual cell surface–associated MMP activity was quantified using McaPLGLDpaAR. Data are mean ± SD (n = 4). *P < 0.001 when compared with cells incubated with 1,10-phenanthroline or RS104210.

 
Activated Murine PMN Express Cell Surface–Bound MMP-9 in a Catalytically Active Form
To specifically assess whether MMP-9 endogenously expressed on the cell surface of activated PMN is catalytically active, we employed a loss-of-function strategy. We compared the total cell surface MMP activity of equal numbers of activated PMN from mice genetically deficient in MMP-9 (MMP-9-/-) and from wild-type mice (MMP-9+/+) against synthetic and biologic MMP substrates. When activated PMN from MMP-9+/+ mice were tested against the synthetic MMP substrate, McaPLGLDPaAR, they expressed substantial, time-dependent, cell surface–bound cleavage of this substrate (Figure 5A). This activity was mediated by MMP because a synthetic MMP inhibitor (1 mM 1,10-phenanthroline) produced 98 (± SD 3.3)% inhibition of this activity (n = 4); but PMSF, an inhibitor of serine proteinases, had no effect (0 ± SD 0% inhibition; n = 3). However, activated PMN from MMP-9-/- mice expressed substantially less cell surface MMP activity than cells from MMP-9+/+ mice at all time points tested (Figure 5A). On average, activated PMN from MMP-9-/- mice expressed ~ 50% of the total cell surface–bound MMP activity against McaPLGLDpaAR detected on the surface of activated PMN from MMP-9+/+ mice (Figure 5B). These data demonstrate that MMP-9 contributes substantially to PMN cell surface–mediated cleavage of this synthetic substrate. Membrane-bound MMP-9–mediated cleavage of biologic substrates was assessed by comparing 1,10-phenanthroline-inhibitable degradation of FITC-conjugate gelatin by activated PMN from MMP-9+/+ and MMP-9-/- mice. Activated PMN from MMP-9-/- mice expressed ~ 30% of the total MMP-mediated, cell surface–bound gelatinase activity of activated PMN from MMP-9+/+ mice (Figure 5B). Together, these data indicate that MMP-9 expressed on the cell surface of stimulated murine PMN is catalytically active against both synthetic and biologic substrates. In addition, MMP-9 accounts for most (~ 70%) of the MMP-mediated gelatinase activity associated with the cell surface of activated murine PMN.



View larger version (22K):
[in this window]
[in a new window]
  		Figure 5. Comparison of the cell surface McaPLGLDpaAR- and gelatin-cleaving activity of activated PMN from MMP-9+/+ versus MMP-9-/- mice. PMN from MMP-9+/+ (filled circles) and MMP-9-/- (filled squares) mice were activated at 37°C for 15 min with PAF (10-7 M) and then for 30 min with fMLP (10-7 M) for 30 min. In A, PMN (106 cells/assay) were incubated in triplicate with 1.8 µM McaPLGLDpaAR in the presence and absence of 1 mM 1,10-phenanthroline. Total cell surface MMP activity was quantified in cell-free supernatant samples as 1,10-phenanthroline-inhibitable cleavage of McaPLGLDpaAR by fluorimetry. Data are mean values ± SD. In B, equal numbers of activated PMN from pairs of MMP-9+/+ mice (hatched bars) and MMP-9-/- mice (open bars) were incubated at 37°C for 60 min with McaPLGLDpaAR (n = 5) or preincubated for 30 min with 1 mM PMSF (to inhibit cell surface–bound serine proteinases with gelatinase activity [11]), then incubated with gelatin-FITC (n = 6). Total cell surface MMP-mediated activity against the substrates was quantified in cell-free supernatant samples as 1,10-phenanthroline-inhibitable activity using fluorimetry. Data are mean ± SEM. *P = 0.002 and **P < 0.001 when compared with cells from MMP-9+/+ mice.

 
Comparison of the Catalytic Activity of Human Soluble MMP-9 versus Membrane-Bound MMP-9 on Human PMN
To study the catalytic activity and efficiency of human membrane-bound MMP-9 on PMN quantitatively and in isolation, we used a model cell system. We bound exogenous, active-site-titrated human MMP-9 onto the cell surface of unstimulated human PMN, and then we fixed the cells to prevent release of intracellular proteinases (including serine proteinases, which can contribute to cleavage of biologic substrates [10]). We first demonstrated that MMP-9 binds to the cell surface of PMN in a concentration-dependent manner (r = 0.99; Figure 6A), and that in the absence of binding of exogenous MMP-9, the cells have no detectable activity against McaPLGLDpaAR (control PMN, Figure 6B). We also assessed whether our fixation process affects the catalytic activity of cell surface–bound MMP-9. Exposure of soluble pro–MMP-9 to fixatives had minimal effect on its activity against McaPLGLDpaAR (2.4 ± SD 1.5% increase in activity; n = 5). To assess whether fixatives affect the catalytic activity of exogenous MMP-9 bound to PMN, we incubated aliquots of the cells in the presence and absence of fixatives, and then measured their activity against McaPLGLDpaAR. Cells exposed to fixatives expressed (91.1 ± SD 10.0%; n = 4) of the McaPLGLDpaAR-cleaving activity of cells not exposed to fixatives. These data confirm that our cell fixation process has minimal effect on the catalytic activity of cell surface–bound MMP-9. We then compared the activity of exogenous membrane-bound human MMP-9 to that of soluble MMP-9 and control PMN (PMN not exposed to MMP-9) against synthetic and biologic MMP substrates.



View larger version (35K):
[in this window]
[in a new window]
  		Figure 6. Human membrane-bound MMP-9 on PMN cleaves synthetic and biologic substrates. In A, unstimulated PMN were incubated with or without varying concentrations of exogenous, active MMP-9, and the amount of MMP-9 bound to the surface of PMN was quantified by immunostaining and image analysis. The results are expressed as a percent of the amount of MMP-9 detected on control PMN incubated without exogenous MMP-9. Data are mean ± SEM; n = 150–200 cells. *P < 0.001 compared with control PMN incubated without MMP-9. In B, MMP-9 was bound onto PMN, and membrane-bound MMP-9 activity was quantified, as described in MATERIALS AND METHODS. Soluble MMP-9 (0.5 nM, open circles), MMP-9 bound to the surface of PMN (0.5 nM, filled squares), or an equal number of control PMN incubated without MMP-9 (filled triangles) were incubated for varying times with 1.8 µM McaPLGLDpaAR, and cleavage of the substrate was quantified by fluorimetry using standards of purified cleavage product (Mca-pro-leu-OH). Data are mean values ± SD. C, Soluble MMP-9 (open circles), MMP-9 bound to the cell surface of PMN (filled squares), or an equal number of control PMN (filled triangles) were incubated at 37°C with FITC-conjugated particulate elastin. Solubilization of elastin was quantified by fluorimetry. Data are mean values ± SD. (D) MMP-9 bound to PMN (solid bars), or an equal number of control PMN (open bars) were incubated with FITC-conjugated type IV collagen (n = 10) or biotinylated gelatin (n = 6), and degradation of each substrate was quantified using assay standards of soluble, active MMP-9. Data are mean values ± SEM. *P < 0.002, **P < 0.001 when compared with control PMN. (E) Human {alpha}1-proteinase inhibitor was incubated at 37°C with soluble MMP-9 (lane 2), MMP-9 bound to the surface of PMN (lane 3), or the same number of control PMN (lane 4), or in the absence of soluble MMP-9 or cells (lane 1). Cell-free supernatant samples were subjected to 20% SDS-PAGE under reducing conditions. The arrowhead indicates intact {alpha}1-proteinase inhibitor, and the arrow indicates the 50-kD digestion product. The other cleavage product (Mr ~ 4 kD) is not visible on the gel.

 
Human Membrane-Bound Human MMP-9 Is Active against a Synthetic Substrate
When tested against McaPLGLDpaAR, both soluble and membrane-bound MMP-9 produced linear, time-dependent cleavage of McaPLGLDpaAR (r = 0.996 and 0.997, respectively; Figure 6B). This McaPLGLDpaAR-cleaving activity was not due to the release of proteinases from the cells, or detachment of proteinases from the cell surface, because cell-free supernatant fluids from cells incubated under identical conditions had no detectable activity when tested against McaPLGLDpaAR (n = 6). The amount of MMP-9 bound to PMN after cells were incubated with 300 nM MMP-9 was quantified using McaPLGLDpaAR as the substrate and standards of soluble, active-site-titrated MMP-9. PMN bound 57.9 (± SEM 8.3) ng MMP-9 activity per 106 cells (n = 6), and this activity was completely inhibited by 1 mM 1,10-phenanthroline (99.5 ± SD 1.2% inhibition; n = 5).

The catalytic efficiencies of soluble and membrane-bound human MMP-9 were calculated by measuring their Kcat/Km values for cleavage of McaPLGLDPaAR. Comparison of the initial reaction velocities for cleavage of McaPLGLDpaAR by soluble MMP-9 and membrane-bound MMP-9 (31–500 pM) over 6 min at 23°C under first order conditions yielded kcat/Km values of 192,000 ± SD 31,000 (n = 4) and 82,000 ± SD 14,000 M-1s-1 (n = 5) for soluble MMP-9 and membrane-bound MMP-9, respectively. Together, these data indicate that human membrane-bound MMP-9 cleaves a synthetic substrate with catalytic efficiency similar to that of the soluble form of MMP-9.

Human Membrane-Bound MMP-9 Cleaves Biologic Substrates
Like soluble MMP-9, human membrane-bound MMP-9 cleaves particulate elastin (Figure 6C), gelatin, and type IV collagen (Figure 6D). Soluble MMP-9 and membrane-bound MMP-9 also cleaved {alpha}1-proteinase inhibitor at a single locus, generating two cleavage products having Mr ~ 50 kD (Figure 6E), and ~ 4 kD (not shown), as reported previously for soluble MMP-9 (4). Control PMN had no activity against any of these biologic substrates (Figures 6B–6E). 1,10-phenanthroline (1 mM) completely inhibited (100 ± SD 0% inhibition; n = 6) the activity of membrane-bound MMP-9 against one of these substrates (gelatin). No substrate-degrading activity was detected in cell-free supernatants from cells, indicating that the observed activity was due only to cell surface–bound MMP-9. Together, these data indicate that human membrane-bound MMP-9 on PMN has similar substrate specificity and catalytic efficiency as the soluble form of MMP-9.

Susceptibility of Human Membrane-Bound MMP on PMN to Inhibition by Proteinase Inhibitors
To begin to assess the susceptibility of membrane-bound MMP-9 to inhibition, we first quantified the effectiveness of various proteinase inhibitors on the cell surface–bound MMP activity endogenously expressed on the cell surface of viable activated PMN (see Figure 4A). Low molecular mass, synthetic MMP inhibitors (1,10-phenanthroline, Mr 198 D; and RS104210–007, Mr 444 D) produced substantial inhibition of this endogenous cell–associated MMP activity (Figure 4B), but a control inhibitor (AEBSF, a general, synthetic inhibitor of serine proteinases) was ineffective. In marked contrast to the low molecular mass MMP inhibitors, the high molecular mass MMP inhibitors TIMP-1 (28 kD) and TIMP-2 (21 kD) produced minimal inhibition. Preincubation of the cells with AEBSF before addition of TIMP-1 did not increase the effectiveness of TIMP-1 against the MMP activity on the surface of viable activated PMN (not shown), indicating that the lack of effectiveness of TIMPs against this activity is not due to degradation of TIMPs by PMN-derived serine proteinases.

To study the susceptibility to inhibition of cell surface–bound human MMP-9 in isolation and in a quantitative manner, we bound active-site-titrated human MMP-9 to the surface of unstimulated human PMN, and then we quantified the effectiveness of various MMP inhibitors against equimolar amounts (10 nM) of soluble versus membrane-bound human MMP-9. As expected, all of the MMP inhibitors completely inhibited soluble MMP-9 (Figure 7A). The low molecular mass, synthetic MMP inhibitors (1,10-phenathroline and RS113456, Mr 426 D) also fully inhibited membrane-bound MMP-9. However, in marked contrast to soluble MMP-9, the high molecular mass TIMP-1 and TIMP-2 produced only partial inhibition of membrane-bound MMP-9, even when tested at a 100-fold molar excess over enzyme. The synthetic serine proteinase inhibitor, PMSF, did not inhibit either form of MMP-9, as expected.



View larger version (34K):
[in this window]
[in a new window]
  		Figure 7. Human membrane-bound MMP-9 is resistant to inhibition by TIMPs. In A, 10 nM soluble MMP-9 (solid bars) or 10 nM membrane-bound MMP-9 on PMN (open bars) were incubated with or without: (i) 1,10-phenanthroline (o-Phen,1 mM); (ii) RS113456 (1 µM); (iii) TIMP-1 (1 µM); (iv) TIMP-2 (1 µM); or (v) PMSF (1 mM). Residual MMP-9 activity was quantified in cell-free supernatant samples using McaPLGLDpaAR. Data are mean values ± SEM (n = 6). *P = 0.006, **P = 0.001 compared with soluble MMP-9 incubated with the same inhibitor. In B–D, 10 nM soluble MMP-9 (open circles) or 10 nM membrane-bound MMP-9 (filled squares) were incubated with or without varying concentrations of TIMP-1 (B), TIMP-2 (C), or RS113456 (D). Residual MMP-9 activity was quantified using McaPLGLDpaAR. IC50 values were calculated by nonlinear regression analysis. Data are mean values ± SD.

 
Inhibitor Kinetics
We investigated the ability of TIMPs to inhibit membrane-bound human MMP-9 as a function of time. Incubation of 10 nM membrane-bound MMP-9 on PMN with an equimolar amount of TIMP-1 or TIMP-2 at 37°C for 18 h resulted in only 10.4 ± SD 2.6% and 19.8 ± SD 6.2% inhibition of membrane-bound MMP-9 activity, respectively. The lack of complete inhibition of membrane-bound MMP-9 by TIMP-1 or TIMP-2 prevented determination of their second order association rate constant for membrane-bound MMP-9. However, we measured inhibition of membrane-bound MMP-9 and soluble MMP-9 as a function of TIMP-1 and TIMP-2 concentration. The results of representative experiments are shown in Figure 7B and 7C. The mean IC50 value for TIMP-1 when tested against 10 nM membrane-bound MMP-9 was ~ 21-fold higher than that for 10 nM soluble MMP-9 (428 ± SEM 122 nM [n = 3] versus 18 ± SEM 1 nM [n = 2], respectively). Similarly, the mean IC50 value for TIMP-2 tested against 10 nM membrane-bound MMP-9 was ~ 68-fold higher than that for TIMP-2 tested against 10 nM soluble MMP-9 (205 ± SEM 44 nM [n = 3] versus 3 ± SEM 1 nM [n = 3], respectively). Even when tested at 150-fold molar excess over enzyme, TIMP-1 and TIMP-2 did not completely inhibit membrane-bound MMP-9 (Figures 7B and 7C). In marked contrast, the low molecular mass MMP inhibitor, RS113456, effectively inhibited both soluble and membrane-bound MMP-9 as indicated by the similar, low IC50 values for the two forms of MMP-9 (Figure 7D). These data indicate that when bound to the cell surface of human PMN, MMP-9 is substantially resistant to inhibition by physiologic MMP inhibitors.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
We report for the first time that a member of the MMP family lacking a transmembrane spanning domain is rapidly expressed in an inducible manner on the cell surface of human and murine PMN activated with proinflammatory mediators. Following cellular activation, human PMN endogenously express substantial MMP activity on their cell surface, which is resistant to inhibition by TIMP-1 and TIMP-2. Studies of activated PMN from MMP-9-/- versus MMP-9+/+ mice revealed that MMP-9 is present on the PMN cell surface in a catalytically active form capable of cleaving synthetic and biologic substrates, accounting for ~ 70% of the total cell surface MMP-mediated gelatinase activity of PMN. Studies of human MMP-9 bound to the cell surface of human PMN demonstrated that membrane-bound MMP-9 has a similar spectrum of catalytic activity and similar catalytic efficiency when compared with the soluble form of MMP-9. However, in marked contrast to soluble MMP-9, membrane-bound MMP-9 on human PMN is substantially resistant to inhibition by TIMP-1 and TIMP-2. Cell surface–bound MMP-9 activity on activated PMN thus has the potential to contribute to the pericellular proteolytic activity of PMN, even in the presence of TIMPs.

Regulation of Expression of MMP-9 on the Cell Surface of PMN
Proinflammatory mediators induce synergistic increases in the expression of MMP-9 on the cell surface of human and murine PMN. The association of MMP-9 with the plasma membrane is likely to be due to agonist-induced degranulation, followed by the binding of the released pro–MMP-9 to the external surface of the PMN plasma membrane. This concept is supported by the following observations: (i) agonists that upregulate PMN cell surface expression of MMP-9 also induce PMN degranulation; (ii) there is a direct relationship between the capacity of agonists to upregulate PMN cell surface expression of MMP-9 and their capacity to induce PMN degranulation (LPS and fMLP > fMLP alone > LPS alone [30]); (iii) immunogold studies showed that MMP-9 is bound to the external surface of activated PMN; and (iv) exogenous MMP-9 (and pro–MMP-9, not shown) binds to the external surface of PMN in a concentration-dependent manner.

Catalytic Activity of Membrane-Bound MMP-9 on PMN
Our studies of viable, activated human PMN tested against the general MMP substrate, McaPLGLDpaAR, established that MMP activity rapidly appears on the cell surface of PMN when they are activated with biologic mediators, and that this activity remains associated with the cell surface in a stable manner (Figure 4A). In contrast, minimal quantities of catalytically active MMP are freely released by activated PMN. These data indicate that MMP that bind to the cell surface of stimulated PMN are likely to be important bioactive forms of the proteinases in vivo.

The contribution of cell surface–bound MMP-9 to the total MMP activity on the surface of activated human PMN cannot be determined because activated human PMN also express other MMP on their cell surface that overlap with MMP-9 in their substrate specificity, including MT6-MMP (28, 29) and MMP-8 (C. A. Owen and S. D. Shapiro, unpublished observations). In addition, there are no MMP substrates or inhibitors available which are specific for MMP-9. To assess and quantify the catalytic activity of MMP-9 on the surface of activated PMN, we used two complementary strategies: (i) a loss-of-function strategy in which we compared the cell surface MMP activities endogenously expressed by activated PMN from MMP-9-/- versus MMP-9+/+ mice; and (ii) studies of the activity of exogenous, human MMP-9 bound to human PMN. Comparison of the endogenous cell surface MMP activity of activated PMN from MMP-9-/- versus MMP-9+/+ mice demonstrated that catalytically active MMP-9 is present on the surface of stimulated murine PMN, because PMN from MMP-9-/- mice expressed substantially less total MMP-mediated activity against McaPLGLDpaAR and gelatin when compared with PMN from MMP-9+/+ mice. Approximately 70% of the total cell surface MMP-mediated gelatinase activity on activated murine PMN is attributable to MMP-9, because activated PMN from MMP-9-/- mice expressed ~ 30% of the total MMP-mediated, cell surface gelatinase activity of activated cells from MMP-9+/+ mice.

To compare and quantify the catalytic activity and efficiency, and susceptibility to inhibition of soluble versus membrane-bound forms of human MMP-9, we studied the activity of exogenous active-site-titrated human MMP-9 bound to the surface of unstimulated human PMN (which expressed no detectable MMP activity in the absence of binding of exogenous MMP-9). This strategy enabled us to study the activity of human MMP-9 both in isolation and in a quantitative manner. We showed that when bound to the PMN cell surface, human MMP-9 has similar substrate specificity as soluble MMP-9, cleaving McaPLGLDPaAR (with high catalytic efficiency as indicated by its high kcat/Km), extracellular matrix proteins, and {alpha}1-proteinase inhibitor, an important inhibitor of serine proteinases.

Although membrane-bound MMP-9 and soluble MMP-9 exhibited similar catalytic activity when tested against extracellular matrix components, membrane-bound MMP-9 was less efficient than soluble MMP-9 in cleaving {alpha}1-proteinase inhibitor in that it consistently produced incomplete cleavage of this nonmatrix substrate (Figure 6E). The most likely explanation for this difference is that MMP-9 cleaves {alpha}1-proteinase inhibitor at only a single locus on the reactive site loop of this globular inhibitor, and MMP-9 sterically confined on the PMN cell surface likely has less favorable access to this peptide bond when compared with the soluble proteinase. Although cleavage of {alpha}1-proteinase inhibitor by membrane-bound MMP-9 was incomplete, it was always substantial, and cleavage of this substrate by either form of the proteinase is of particular interest, because studies of MMP-9-/- mice (31) have shown that this serpin is an important substrate for MMP-9 in vivo. The activities associated with the cells were not due to: (i) release of MMP activity from the cells, because cell-free supernatant samples harvested from the cells had no activity against any of the substrates tested; or (ii) cell surface–bound proteinases other than MMP-9, because control PMN (with no MMP-9 bound to their surface) had no significant activity against any of the substrates tested. These data indicate that when bound to the PMN cell surface, MMP-9 has the potential to contribute to pericellular proteolysis of PMN in vivo.

The mechanism(s) by which pro–MMP-9 is activated before or after its binding to the PMN cell surface are beyond the scope of the current study. It is noteworthy that the mechanisms by which pro–MMP-9 freely secreted by PMN and other cells is activated in vivo are also still unknown. Soluble pro–MMP-9 can be activated by oxidants and proteinases in vitro (1), and the possibility that membrane-bound MMP-9 on PMN is activated by similar mechanisms is the focus of ongoing studies in our laboratory.

Susceptibility of Membrane-Bound MMP-9 on Human PMN to Inhibition by TIMPs
MMP activity endogenously expressed on the cell surface of stimulated human PMN is effectively inhibited by low molecular weight, synthetic MMP inhibitors, but it is almost completely resistant to inhibition by TIMP-1 and TIMP-2 (Figure 4B). This lack of inhibition by TIMPs is not due to degradation of TIMPs by serine proteinases released by PMN or expressed on their cell surface (21), because preincubation of the cells with an inhibitor of serine proteinases (AEBSF) did not increase the capacity of TIMP-1 to inhibit cell surface–bound MMP activity on activated PMN. To quantify the susceptibility of membrane-bound MMP-9 to inhibition, we studied exogenous human MMP-9 bound to the surface of human PMN. Like MMP activity endogenously expressed on the cell surface of activated human PMN (Figure 4B), exogenous MMP-9 bound to the surface of PMN is also effectively inhibited by low molecular weight MMP inhibitors, but is resistant to inhibition by TIMPs (Figure 7). The effectiveness of 1 µM TIMPs against endogenous surface–bound MMP activity on stimulated PMN (< 5% inhibition) was substantially less than the effectiveness of the same concentration of TIMPs against exogenous MMP-9 bound to PMN (~ 60% inhibition). One explanation for this difference is the expression of other TIMP-resistant proteinases on the cell surface of stimulated PMN. This notion is supported by our ongoing studies showing that MMP-8 activity is expressed on the cell surface of stimulated PMN, which is also strikingly resistant to inhibition by TIMP-1 and TIMP-2 (C. A. Owen and S. D. Shapiro, unpublished observations). In addition, MMP-9 and MMP-8 may bind to cell surface molecules which are also expressed in an inducible manner on stimulated PMN. If this is the case, it is likely that PMN endogenously express greater amounts of MMPs than the maximal amount of exogenous MMP-9 that we can bind to unstimulated PMN, resulting in greater resistance to inhibition by any given concentration of TIMPs. Nevertheless, our data in both cell systems demonstrate that cell surface–bound MMPs and MMP-9 on PMN are strikingly less susceptible to inhibition by TIMPs than soluble MMP-9.

The indirect relationship between inhibitor size and its effectiveness against membrane-bound MMP-9 suggests that steric hindrance may be the mechanism underlying the resistance of membrane-bound MMP-9 to inhibition by TIMPs. An alternative explanation for the differential effectiveness of TIMP-1 against membrane-bound MMP-9 could be related to the mechanism of binding of the enzyme to the PMN cell surface. The synthetic inhibitors bind directly to the active site of MMP-9, whereas the COOH terminal domain of MMP-9 is important in its ability to form complexes with TIMP-1 (25). If MMP-9 binds to PMN plasma membranes via its COOH-terminal domain, this may prevent TIMP-1 from forming complexes with, and inhibiting membrane-bound MMP-9.

Expression of MMP on Leukocytes
There are only two previous reports of cell surface–bound MMP on inflammatory cells: MT6-MMP on the surface of PMN (32), and MT1-MMP, which is expressed at low levels on alveolar macrophages (33). In contrast, surface expression of catalytically active, serpin-resistant serine proteinases on inflammatory cells is well documented (21, 26, 34–36). Coordinate expression of MMP-9 and serine proteinases on the surface of activated PMN may serve to amplify degradation of susceptible proteins in the pericellular environment of PMN because (i) MMP-9 and serine proteinases degrade a similar spectrum of extracellular matrix proteins (20); and (ii) membrane-bound MMP-9 on PMN can cleave and inactivate {alpha}1-proteinase inhibitor, thereby facilitating serine proteinase activity in the pericellular environment of PMN.

Possible Functions of Membrane-Bound MMP-9 on PMN
MMP-9 is widely believed to be involved in PMN migration through tissue planes. However, there are conflicting reports on whether proteinases are directly involved in PMN migration (37–39). It has been suggested that limited and localized proteolysis may be sufficient to allow PMN to traverse tissue barriers (40). It is possible that membrane-bound MMP-9 permits this exquisitely localized degradation of matrix components in the immediate vicinity of an activated PMN (despite the presence of TIMPs in the extracellular space), thereby preserving tissue architecture as PMN migrate through tissues during the inflammatory response. It is of interest in this respect, that MMP-9 is focally distributed on the surface of activated PMN, and tends to be localized to the leading edge of chemoattractant-polarized PMN.

Coordinated expression of MMP-9 and serine proteinases on the cell surface of PMN may amplify pericellular proteolytic activities associated with activated PMN during the inflammatory response, at least in part, by cleavage and inactivation of serpins by membrane-bound MMP-9. It is noteworthy that catalytically active MMP-9 is expressed on the surface of tumor cells and keratinocytes, and that CD44 mediates the binding of MMP-9 to these cells (41, 42). Moreover, MMP-9 anchored to CD44 on these cells proteolytically cleaves and activates latent TGF-ß to generate fully active form of this multifunctional growth factor (41). This finding, together with the known spectrum of catalytic activity of soluble MMP-9 against cytokines and chemokines, raises the possibility that membrane-bound MMP-9 on PMN can also regulate inflammatory and fibrotic processes within the lung. This possibility is the focus of ongoing studies in our laboratory.

Conclusions
MMP-9 has been traditionally thought to function as a soluble proteinase followings its release from activated PMN. We now report that MMP-9 is also rapidly expressed on the cell surface of PMN following their activation with biologic mediators. The binding of MMP-9 to the PMN plasma membrane serves not only to focus and restrict its activity to the pericellular environment, but also enables it to retain its activity even in the presence of TIMPs, swinging the local proteolytic balance in favor of extracellular matrix degradation. Pericellular proteolysis mediated by membrane-bound MMP-9 may contribute to physiologic processes of PMN. However, if the activity of membrane-bound MMP-9 on PMN is excessive, inappropriate or prolonged, it is also well equipped to contribute to tissue injury.

Our data also have implications for current and future strategies to reduce lung injury in diseases in which PMN-derived MMP-9 may play pathogenetic roles, including chronic obstructive pulmonary disease, asthma, and idiopathic pulmonary fibrosis (6). Because our data indicate that membrane-bound MMP-9 on PMN is likely to be an important bioactive form of the proteinase in vivo, treatment strategies must be effective at inhibiting the activity of membrane-bound MMP-9 as well as soluble MMP-9. In this respect, our data indicate that low molecular weight proteinase inhibitors are likely to be an effective strategy for inhibiting PMN-derived MMP-9 in the lung.


   Acknowledgments
 
The authors thank Roche Bioscience for providing RS113456, RS104210-007, and MMP-9. They thank R. M. Senior, M.D., Washington University, St. Louis, MO; Clare Dollery, MRCP, Ph.D., University College of London, UK; and R. J. Riese, M.D., Ph.D., Brigham and Women's Hospital, Boston, MA, for critical reading of the manuscript. This work was supported by National Institutes of Health grant Ro1 HL63137 (C.A.O.). C.A.O. is a recipient of a Career Investigator Award from the Thoracic Society of Massachusetts.

Received in original form January 31, 2003

Received in final form March 20, 2003


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Vu, T. H., and Z. Werb. 1998. Gelatinase B: structure, regulation, and function. In Matrix Metalloproteinases. W. C. Parks and R. P. Mecham, editors. Academic Press, San Diego. 115–148.
  2. Schonbeck, U., F. Mach, and P. Libby. 1998. Generation of biologically active IL-1ß by matrix metalloproteinases: a novel caspase-1-independent pathway of IL-1ß processing. J. Immunol. 161:3340–3346.[Abstract/Free Full Text]
  3. Van Den Steen, P. E., P. Proost, A. Wuyts, J. Van Damme, and G. Opdenakker. 2000. Neutrophil gelatinase B potentiates interleukin-8 tenfold by aminoterminal processing, whereas it degrades CTAP-III, PF-4, and GRO-{alpha} and leaves RANTES and MCP-2 intact. Blood 96:2673–2681.[Abstract/Free Full Text]
  4. Sires, U. I., G. Murphy, H. G. Welgus, and R. M. Senior. 1994. Matrilysin is much more efficient than other metalloproteinases in the proteolytic inactivation of alpha 1-antitrypsin. Biochem. Biophys. Res. Commun. 204:613–620.[CrossRef][Medline]
  5. Ito, A., A. Mukaiyama, Y. Itoh, H. Nagase, I. B. Thogersen, J. J. Enghlid, Y. Sasaguri, and Y. Mori. 1996. Degradation of interleukin 1ß by matrix metalloproteinases. J. Biol. Chem. 271:14657–14660.[Abstract/Free Full Text]
  6. Atkinson, J. J., and R. M. Senior. 2003. Matrix metalloproteinase-9 in lung remodeling. Am. J. Respir. Cell Mol. Biol. 28:12–24.[Abstract/Free Full Text]
  7. Zheng, T., Z. Zhu, Z. Wang, R. J. Homer, B. Ma, R. J. Riese, Jr., H. A. Chapman, Jr., S. D. Shapiro, and J. A. Elias. 2000. Inducible targeting of IL-13 to the adult lung causes matrix metalloproteinase- and cathepsin-dependent emphysema. J. Clin. Invest. 106:1081–1093.[Medline]
  8. Betsuyaku, T., M. Nishimura, K. Takeyabu, M. Tanino, P. Venge, S. Xu, and Y. Kawakami. 1999. Neutrophil granule proteins in bronchoalveolar lavage fluid from subjects with subclinical emphysema. Am. J. Respir. Crit. Care Med. 159:1985–1991.[Abstract/Free Full Text]
  9. Lanone, S., T. Zheng, Z. Zhu, W. Liu, C. G. Lee, B. Ma, Q. Chen, R. J. Homer, J. Wang, L. A. Rabach, M. E. Rabach, J. M. Shipley, S. D. Shapiro, R. M. Senior, and J. A. Elias. 2002. Overlapping and enzyme-specific contributions of matrix metalloproteinases-9 and -12 in IL-13-induced inflammation and remodeling. J. Clin. Invest. 110:463–474.[CrossRef][Medline]
  10. Vignola, A. M., L. Riccobono, A. Mirabella, M. Profita, P. Chanez, V. Bellia, G. Mautino, P. D'accardi, J. Bousquet, and G. Bonsignore. 1998. Sputum metalloproteinase-9/tissue inhibitor of metalloproteinase-1 ratio correlates with airflow obstruction in asthma and chronic bronchitis. Am. J. Respir. Crit. Care Med. 158:1945–1950.[Abstract/Free Full Text]
  11. Hoshino, M., Y. Nakamura, J. Sim, J. Shimojo, and S. Isogai. 1998. Bronchial subepithelial fibrosis and expression of matrix metalloproteinase-9 in asthmatic airway inflammation. J. Allergy Clin. Immunol. 102:783–788.[CrossRef][Medline]
  12. Lee, Y. C., H. B. Lee, Y. K. Rhee, and C. H. Song. 2001. The involvement of matrix metalloproteinase-9 in airway inflammation of patients with acute asthma. Clin. Exp. Allergy 31:1623–1630.[CrossRef][Medline]
  13. Warner, R. L., L. Beltran, E. M. Younkin, C. S. Lewis, S. J. Weiss, J. Varani, and K. J. Johnson. 2001. Role of stromelysin 1 and gelatinase B in experimental acute lung injury. Am. J. Respir. Cell Mol. Biol. 24:537–544.[Abstract/Free Full Text]
  14. Suga, M., K. Iyonaga, T. Okamoto, Y. Gushima, H. Miyakawa, T. Akaike, and M. Ando. 2000. Characteristic elevation of matrix metalloproteinase activity in idiopathic interstitial pneumonias. Am. J. Respir. Crit. Care Med. 162:1949–1956.[Abstract/Free Full Text]
  15. Betsuyaku, T., Y. Fukuda, W. C. Parks, J. M. Shipley, and R. M. Senior. 2000. Gelatinase B is required for alveolar bronchiolization after intratracheal bleomycin. Am. J. Pathol. 157:525–535.[Abstract/Free Full Text]
  16. Liu, Z., J. M. Shipley, T. H. Vu, X. Zhou, L. A. Diaz, Z. Werb, and R. M. Senior. 1998. Gelatinase B–deficient mice are resistant to experimental bullous pemphigoid. J. Exp. Med. 188:475–482.[Abstract/Free Full Text]
  17. Dubois, B., S. Masure, U. Hurtenbach, L. Paemen, H. Heremans, J. van den Oord, R. Sciot, T. Meinhardt, G. Hammerling, G. Opdenakker, and B. Arnold. 1999. Resistance of young gelatinase B-deficient mice to experimental autoimmune encephalomyelitis and necrotizing tail lesions. J. Clin. Invest. 104:1507–1515.[Medline]
  18. Pugin, J., M.-C. Widmer, S. Kossodo, C.-M. Liang, H. L. Preas, II, and A. F. Suffredini. 1999. Human neutrophils secrete gelatinase B in vitro and in vivo in response to endotoxin and proinflammatory mediators. Am. J. Respir. Cell Mol. Biol. 20:458–464.[Abstract/Free Full Text]
  19. Opdenakker, G., P. E. Van den Steen, B. Dubois, I. Nelissen, E. Van Coillie, S. Masure, P. Proost, and J. Van Damme. 2001. Gelatinase B functions as regulator and effector in leukocyte biology. J. Leukoc. Biol. 69:851–859.[Abstract/Free Full Text]
  20. Owen, C. A., and E. J. Campbell. 1999. The cell biology of leukocyte-mediated proteolysis. J. Leukoc. Biol. 65:137–150.[Abstract]
  21. Owen, C. A., M. A. Campbell, P. L. Sannes, S. S. Boukedes, and E. J. Campbell. 1995. Cell-surface-bound elastase and cathepsin G on human neutrophils. A novel, non-oxidative mechanism by which neutrophils focus and preserve catalytic activity of serine proteinases. J. Cell Biol. 131:775–789.[Abstract/Free Full Text]
  22. Vu, T. H., J. M. Shipley, G. Bergers, J. E. Berger, J. A. Helms, D. Hanahan, S. D. Shapiro, R. M. Senior, and Z. Werb. 1998. MMP-9/gelatinase B is a key regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes. Cell 93:411–422.[CrossRef][Medline]
  23. Fleming, T. J., M. L. Fleming, and T. R. Malek. 1993. Selective expression of Ly-6G on myeloid lineage cells in mouse bone marrow. RB6–8C5 mAb to granulocyte-differentiation antigen (Gr-1) detects members of the Ly-6 family. J. Immmunol. 151:2399–2408.[Abstract]
  24. Murphy, G., and T. Crabbe. 1995. Gelatinases A and B. Methods Enzymol. 248:471–484.
  25. O'Connell, J. P., F. Willenbrock, A. J. P. Docherty, D. Eaton, and G. Murphy. 1994. Analysis of the role of the COOH-terminal domain the the activation, proteolytic activity, and tissue inhibitor of metalloproteinase interactions of gelatinase B. J. Biol. Chem. 269:14967–14973.[Abstract/Free Full Text]
  26. Campbell, E. J., M. A. Campbell, and C. A. Owen. 2000. Bioactive proteinase 3 on the cell surface of human neutrophils: quantification, catalytic activity, and susceptibility to inhibition. J. Immunol. 165:3366–3374.[Abstract/Free Full Text]
  27. Huebner, P. F. 1976. Determination of elastolytic activity with elastin-rhodamine. Anal. Biochem. 74:419–429.[CrossRef][Medline]
  28. Kang, T., J. Yi, A. Guo, X. Wang, C. Overall, W. Jiang, R. Elde, N. Borregaard, and D. Pei. 2001. Subcellular distribution and cytokine-/chemokine-regulated secretion of leukolysin/MT6-MMP/MMP-25 in neutrophils. J. Biol. Chem. 276:21960–21968.[Abstract/Free Full Text]
  29. English, W. R., G. Velasco, J. O. Stracke, V. Knauper, and G. Murphy. 2001. Catalytic activities of membrane-type 6 matrix metalloproteinase (MMP25). FEBS Lett. 491:137–142.[CrossRef][Medline]
  30. Fittschen, C., R. A. Sandhaus, G. S. Worthen, and P. M. Henson. 1988. Bacterial lipopolysaccharide enhances chemoattractant-induced elastase secretion by human neutrophils. J. Leukoc. Biol. 43:547–556.[Abstract]
  31. Liu, Z., X. Zhou, S. D. Shapiro, J. M. Shipley, S. S. Twining, L. A. Diaz, R. M. Senior, and Z. Werb. 2000. The serpin alpha1-proteinase inhibitor is a critical substrate for gelatinase B/MMP-9 in vivo. Cell 102:647–655.[CrossRef][Medline]
  32. Pei, D. 1999. Leukolysin/MMP25/MT6-MMP: a novel matrix metalloproteinase specifically expressed in the leukocyte lineage. Cell Res. 9:291–303.[CrossRef][Medline]
  33. Sato, T., M. del Carmen Ovejero, P. Hou, A.-M. Heegaard, M. Kumegawa, N. T. Foged, and J.-M. Delaisse. 1997. Identification of the membrane-type matrix metalloproteinase MT1-MMP in osteoclasts. J. Cell Sci. 110:589–596.[Abstract]
  34. Owen, C. A., M. A. Campbell, S. S. Boukedes, and E. J. Campbell. 1995. Inducible binding of cathepsin G to the cell surface of neutrophils: a mechanism for mediating extracellular proteolytic activity of cathepsin G. J. Immunol. 155:5803–5810.[Abstract]
  35. Owen, C. A., M. A. Campbell, S. S. Boukedes, and E. J. Campbell. 1997. Cytokines regulate membrane-bound leukocyte elastase on neutrophils: a novel mechanism for effector activity. Am. J. Physiol.(Lung Cell. Mol. Physiol.) 272:L385–L393.[Abstract/Free Full Text]
  36. Ellis, V., T.-C. Wun, N. Behrendt, E. Ronne, and K. Dano. 1990. Inhibition of receptor-bound urokinase by plasminogen-activator inhibitors. J. Biol. Chem. 265:9904–9908.[Abstract/Free Full Text]
  37. Delclaux, C., C. Delacourt, M.-P. D'Ortho, V. Boyer, C. Lafuma, and A. Harf. 1996. Role of gelatinase B and elastase in human polymorphonuclear neutrophil migration across basement membrane. Am. J. Respir. Cell Mol. Biol. 14:288–295.[Abstract]
  38. Betsuyaku, T., J. M. Shipley, Z. Liu, and R. M. Senior. 1999. Neutrophil emigration in the lungs, peritoneum, and skin does not require gelatinase B. Am. J. Respir. Cell Mol. Biol. 20:1303–1309.[Abstract/Free Full Text]
  39. Mackarel, A. J., D. C. Cottell, K. J. Russell, M. X. Fitzgerald, and C. M. O'Connor. 1999. Migration of neutrophils across human pulmonary endothelial cells is not blocked by matrix metalloproteinase or serine protease inhibitors. Am. J. Respir. Cell Mol. Biol. 20:1209–1219.[Abstract/Free Full Text]
  40. Price, B., C. Dennison, H. Tschesche, and E. Elliott. 2000. Neutrophil tissue inhibitor of matrix metalloproteinases-1 occurs in novel vesicles that do not fuse with the phagosome. J. Biol. Chem. 275:28308–28315.[Abstract/Free Full Text]
  41. Yu, Q., and I. Stamenkovic. 2000. Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-beta and promotes tumor invasion and angiogenesis. Genes Dev. 14:163–176.[Abstract/Free Full Text]
  42. Yu, Q., and I. Stamenkovic. 1999. Localization of matrix metalloproteinase 9 to the cell surface provides a mechanism for CD44-mediated tumor invasion. Genes Dev. 13:35–48.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2003-0034OCv1
29/3/283    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow