Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M512308200 on May 3, 2006

J. Biol. Chem., Vol. 281, Issue 27, 18626-18637, July 7, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/27/18626    most recent
M512308200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Van den Steen, P. E.
Right arrow Articles by Opdenakker, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Van den Steen, P. E.
Right arrow Articles by Opdenakker, G.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

The Hemopexin and O-Glycosylated Domains Tune Gelatinase B/MMP-9 Bioavailability via Inhibition and Binding to Cargo Receptors*

Philippe E. Van den Steen{ddagger}1, Ilse Van Aelst{ddagger}, Vibeke Hvidberg§, Helene Piccard{ddagger}, Pierre Fiten{ddagger}, Christian Jacobsen§, Soren K. Moestrup§, Simon Fry, Louise Royle, Mark R. Wormald, Russell Wallis||, Pauline M. Rudd, Raymond A. Dwek, and Ghislain Opdenakker{ddagger}2

From the {ddagger}Laboratory of Immunobiology, Rega Institute for Medical Research, University of Leuven, B-3000, Belgium, the §Department of Medical Biochemistry, University of Aarhus, Denmark, the Oxford Glycobiology Institute, University of Oxford, OX1 3QU, United Kingdom, and the ||MRC Immunochemistry Unit, Department of Biochemistry, University of Oxford, OX1 3QU, United Kingdom, and Department of Infection, Immunity and Inflammation, University of Leicester, LE1 9HN United Kingdom

Received for publication, November 16, 2005 , and in revised form, April 21, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Gelatinase B/matrix metalloproteinase-9 (MMP-9), a key regulator and effector of immunity, contains a C-terminal hemopexin domain preceded by a unique linker sequence of ~64 amino acid residues. This linker sequence is demonstrated to be an extensively O-glycosylated (OG) domain with a compact three-dimensional structure. The OG and hemopexin domains have no influence on the cleavage efficiency of MMP-9 substrates. In contrast, the hemopexin domain contains a binding site for the cargo receptor low density lipoprotein receptor-related protein-1 (LRP-1). Furthermore, megalin/LRP-2 is identified as a new functional receptor for the hemopexin domain of MMP-9, able to mediate the endocytosis and catabolism of the enzyme. The OG domain is required to correctly orient the hemopexin domain for inhibition by TIMP-1 and internalization by LRP-1 and megalin. Therefore, the OG and hemopexin domains down-regulate the bioavailability of active MMP-9 and the interactions with the cargo receptors are proposed to be the original function of hemopexin domains in MMPs.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Matrix metalloproteinases (MMPs)3 constitute a family of Zn2+-dependent proteases, which share a number of conserved protein domains (1). All MMPs consist of a propeptide and a Zn2+-containing active site, and most MMPs have a C-terminal hemopexin domain. Some MMPs contain additional domains, e.g. in MT-MMPs, the hemopexin domain is connected to a membrane anchor, and in MMP-2 (gelatinase A) and MMP-9 (gelatinase B), a fibronectin domain is inserted in the active site. In MMP-9, a unique linker sequence of unknown function connects the active site and the hemopexin domain, and is more than 50 amino acids long (in contrast to ~10 residues for the linker sequence in other MMPs). It has been considered to be an independent protein domain and has low homology with collagen type V (2), mainly because of its high proline content. Because we show in the present study that it contains O-linked glycans, it is henceforth named the O-glycosylated (OG) domain.

MMP-9 degrades extracellular matrix components and cleaves a variety of soluble proteins, including cytokines and chemokines. Overexpression of MMP-9 leads to inflammation, tissue destruction, and pathology (3, 4). Therefore, expression of MMP-9 is highly regulated at the levels of transcription, translation, and secretion. Once secreted, the activity is tightly kept in check by proenzyme latency, inhibition, and catabolism. The active site can be inhibited by interaction with the N-terminal domain of the tissue inhibitors of metalloproteinases (TIMPs) (5). In addition, the C-terminal part of TIMP-1 binds to the hemopexin domain of both the pro- and active MMP-9, leading to a high affinity interaction (6). MMP-9 may be internalized and subsequently degraded through the cargo receptor low density lipoprotein receptor related protein-1 (LRP-1 or CD91) (7).

Another regulatory system for MMP-9 activity is glycosylation (8). Structures of N- and O-linked sugars were determined for natural MMP-9 from human neutrophils (9, 10). Little information is available concerning the structures or functions of glycosylation of other MMPs, except that in MT1-MMP, the O-linked glycans on the linker peptide between the active site and the hemopexin domain are essential for the binding of TIMP-2 (11).

Besides the catalytic site, different domains of MMPs are important for the hydrolysis of particular substrates. Collagenolysis by interstitial collagenase is dependent on the hemopexin domain and linker peptide (12). Gelatinolysis by gelatinases is more than 100-fold enhanced by the gelatin-binding fibronectin domain (13). The hemopexin domain of MMP-2 binds monocyte chemoattractant protein-3 (MCP-3), leading to MCP-3 cleavage (14). These and other examples have led to the concept of "exosites," which bind the substrate and confer high efficiency on the subsequent cleavage (15). This concept is also interesting from a therapeutic point of view, because inhibition of the exosite interaction may be used to inhibit the cleavage of particular substrates while leaving other activities of the MMP unaffected.

Because no data are available on the structure and functions of the OG domain and limited data on the function of the hemopexin domain of MMP-9, we aimed at studying the structural and functional aspects of these two domains by the generation of recombinant MMP-9 variants, with deletions or mutations.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Expression and Purification of Recombinant MMP-9 in Sf9 Insect Cells
The cDNA of human MMP-9, a kind gift from Dr. G. I. Goldberg (2), was amplified by PCR with the following primers: ATATCTCGAGAGCCCCCAGACAGCGCCAGTCC ("hgelBfor", Xho-I site underlined) and AATTCCATGGCTAGTCCTCAGGGCACTGCAGG ("hgelBrev", Nco-I site underlined). After cleavage with Xho-I and Nco-I, the amplicon was inserted into the pMelBacB vector (Invitrogen) behind the polyhedrin promoter and the mellitin secretion signal, resulting in the vector pMelBacBMMP-9. The construct was checked by DNA sequence analysis. Sf9 cells were cotransfected with pMelBacBMMP-9 and the baculoviral Bac-N-BlueTM DNA (Invitrogen), according to the protocol of the manufacturer. This results in homologous recombination between the linearized baculoviral Bac-N-BlueTM DNA and pMelBacBMMP-9, generating viable recombinant viruses with the MMP-9 gene and a functional LacZ. The supernatant of the cotransfected cells was harvested after 72 h of incubation at 27 °C. This P1 transfection viral stock was subjected to a plaque assay in the presence of the chromogenic substrate Bluo-gal (Invitrogen). Blue recombinant virus clones were isolated and screened by PCR to detect the MMP-9 insert. Viral clones containing the insert of the correct length were expanded in Sf9 cells to form a P2 small scale high titer virus stock, which was further expanded to a large scale high titer P3 virus stock. Suspension cultures of Sf9 cells (2.106 cells/ml) were infected with 0.05 plaque-forming units (pfu)/cell, and the supernatant was harvested after 3 days of incubation at 27 °C in serum-free medium. MMP-9 was purified by gelatin-Sepharose chromatography as described (16).

Mutagenesis of Recombinant MMP-9
To generate deletion mutants of human MMP-9 lacking the hemopexin domain or the OG and hemopexin domains, a stop codon was inserted in the sequence of MMP-9 in pMelBacBMMP-9 behind the codon coding for Pro511 or for Pro447, respectively. This was performed by site-directed mutagenesis with the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). The following primers, which contain the mutations (underlined), were used for the mutagenesis: CTATGGTCCTCGCCCTTAACCTGAGCCACG and CGTGGCTCAGGTTAAGGGCGAGGACCATAG for MMP-9{Delta}OGHem, CTGTGCCTTTGAGTCCGTAGGACGATGCCTGC and GCAGGCATCGTCCTACGGACTCAAAGGCACAG for MMP-9{Delta}Hem. The same strategy was used to generate a vector coding for MMP-9MutEC, in which the catalytic Glu402 and Cys468 in the OG domain are mutated into Ala. The following primers, which contain the mutations (underlined), were used for this construct: GCTCCCCCGACGGTCGCCCCCACCGGACCC and GGGTCCGGTGGGGGCGACCGTCGGGGGAGC for mutation of Cys468 and CGTGGCGGCGCATGCGTTCGGCCACGCGCTGG and CCAGCGCGTGGCCGAACGCATGCGCCGCCACG for mutation of Glu402.

The OG domain in MMP-9{Delta}OG was deleted using a splicing-by-overlap-extension method (17). The coding sequence for the part of MMP-9 preceding the OG domain was amplified by PCR using the following primers: ATATCTCGAGAGCCCCCAGACAGCGCCAGTCC ("hgelBfor") and ATCGTCCACAGGGCGAGGACCATAGAGGTGCCGG ("gbdelc5r", overlap with hemopexin domain is underlined). The hemopexin domain was amplified by PCR with the primers CCTCGCCCTGTGGACGATGCCTGCAACGTGAAC ("gbdel5-Fbis", overlap with OG domain is underlined) and AATTCCATGGCTAGTCCTCAGGGCACTGCAGG ("hgelBrev"). After purification, the two PCR products were mixed and again amplified using the hgelBfor and hgelBrev primers, resulting in a PCR product containing the sequence of MMP-9 without the OG domain. This PCR product was first subcloned in the pCR4-TOPO-vector (Invitrogen) in Escherichiacoli. Recombinant clones were screened and checked by DNA sequencing analysis. A clone without unwanted mutations was excised with Xho-I and Nco-I and ligated into pMelBacB, resulting in the vector pMelBacBMMP-9{Delta}OG. The insert was checked again by DNA sequencing. Each of these constructs was expressed and purified in the same way as for intact MMP-9 (see above).

Activation of Recombinant MMP-9 Variants
MMP-9 variants (10 µM) were activated with 0.1 µM recombinant catalytic domain of human stromelysin-1 (Calbiochem), without the presence of APMA, during 90 min at 37 °C in assay buffer (100 mM Tris/HCl, pH 7.4, 100 mM NaCl, 10 mM CaCl2). The activity at different time intervals was monitored by gelatin zymography and with a fluorogenic substrate conversion assay (18).

Binding of Recombinant MMP-9 to Lectins
Binding of MMP-9 variants to the lectins Helix pomatia agglutinin (HPA, specific for Ser- or Thr-linked GalNAc) and jacalin (specific for Gal-GalNAc) was tested using HPA- and jacalin-agarose (Sigma) chromatography. 100 µg of the MMP-9 variants in 50 mM Tris/HCl, pH 7.4, and 50 mM NaCl were loaded on the lectin-agarose column. Bound MMP-9 was eluted with a gradient of 0-400 mM GlcNAc for HPA-agarose and 0-100 mM lactose for jacalin-agarose. Binding and elution was monitored by UV absorption at 254 nm and by zymography analysis of the flow-through and eluent.

Detailed Glycosylation Analysis of Recombinant MMP-9
In-gel Enzymatic Release of N-Linked Glycans—10 µg of the MMP-9 variants were reduced and alkylated prior to SDS-PAGE. Gel bands were cut out and N-linked glycans extracted as described (19).

O-Glycan Release by Ammonia-based beta-Elimination—40 µg of the MMP-9 variants were applied to 0.45-µm HVHP protein-binding polyvinylidene difluoride Durapore membrane filters (Millipore, Bedford, MA). Filters were then washed with 2 x 2 ml of double distilled water and dried. O-Glycans were released based on the method of Huang et al. (20) by incubating filters in 1 ml of 29.2% aqueous ammonium hydroxide (Sigma) saturated with ammonium carbonate (Aldrich 99.999% pure) for 40 h at 60 °C. Ammonium hydroxide and ammonium carbonate were removed by repeated evaporation to dryness with double distilled water. 100 µl 0.5 M boric acid (BDH, Aristar) was added to sample tubes and incubated for 30 min at 37 °C. Boric acid was removed by repeated evaporation to dryness with methanol. Dried O-glycans were labeled.

2-Aminobenzamide Fluorescent Labeling of Released Glycans—Dried glycans were fluorescently labeled with 2-AB (21) using a Ludger Tag 2-AB labeling kit (Ludger Limited, Oxford, UK). Excess label was removed by ascending chromatography on Whatman 3MM chromatography paper in acetonitrile. Glycans were washed from the paper with water.

Exoglycosidase Array Digestions—2-AB-labeled glycans were digested in a volume of 10 µl for18 h at 37 °C in 50 mM sodium acetate buffer, pH 5.5, using the following exoglycosidases obtained from Prozyme (San Leandro, CA): bovine kidney {alpha}-fucosidase (EC 3.2.1.51 [EC] , BKF), 1 unit/ml; jackbean {alpha}-mannosidase (EC 3.2.1.24 [EC] , JBM), 50 units/ml. After digestions, samples were passed through a protein-binding filter (Micropure-EZ centrifugal filter devices, Millipore) to remove exoglycosidases. Glycans were eluted from filters with water.

Normal Phase HPLC—NP-HPLC separations were performed on a 2690 Alliance separation module (Waters, Milford, MA) equipped with a 474 fluorescent detector and fitted with a 4.6 x 250 mm TSKgel amide-80 column (Anachem, Luton, Beds, UK). Gradient conditions and detection were as described previously (22, 23). The system was calibrated against a standard dextran hydrolysate ladder so that the retention time for individual glycans could be converted to glucose units (GU) (22). Oligosaccharide structures were assigned by reference to the GU values of a data base of standard sugars (23).

Glycan Analysis by Liquid Chromatography-Electrospray Ionization-Mass Spectrometry (LC-ESI-MS)—2-AB-labeled glycans were analyzed using an LC packings Ultimate HPLC equipped with a Famos autosampler (Dionex Ltd, Leeds, UK) interfaced with a Q-Tof Ultima Global mass spectrometer (Waters-Micromass, Manchester, UK). Chromatographic separation was achieved using a 2 x 250 mm, microbore NP-HPLC TSK gel Amide-80 column (Hichrome), 50 mM ammonium formate, pH 4.4 (solvent A) and acetonitrile (solvent B), with a gradient of 20-70% solvent A over 100 min at a flow rate of 40 µl/min. The mass spectrometer was operated in positive ion mode with 3 kV capillary voltage; RF lens 60; source temp 100 °C; desolvation temperature 150 °C; cone gas flow 50 liters/h; and desolvation gas flow 450 liters/h (23).

Digestion of MMP-9 Substrates
To estimate the cleavage efficiency of various substrates by the different MMP-9 variants, denatured bovine collagen II (24), MMP-8-cleaved native collagen II (25), {alpha}1-antitrypsin inhibitor (R&D Systems, Abingdon, UK; 0.5 µM) (26), mouse eye extract containing betaB1-crystallin (27), recombinant intact mouse granulocyte chemotactic protein-2 (mGCP-2/LIX, Peprotech Inc, Rocky Hill, NJ; 2 µM) (28), recombinant interleukin-8 (IL-8, Peprotech, 2 µM) (29), galectin-3 (R&D Systems, 30 µg/ml) (30), soluble recombinant ICAM-1/Fc chimera (R&D Systems, 30 µg/ml) (31), soluble recombinant CD25/Fc chimera (R&D Systems, 30 µg/ml) (32), human plasminogen (R&D Systems, 30 µg/ml) (33), human platelet SPARC (Calbiochem, 1 µM) (34) or recombinant MCP-3 (Peprotech, 0.5 µM) was incubated in 100 mM Tris/HCl pH 7.4, 100 mM NaCl, 10 mM CaCl2 with different concentrations of activated MMP-9 variants at 37 °C. As negative control for the cleavage by MMP-9, the substrates were incubated under identical conditions with the corresponding concentration of recombinant catalytic domain of human MMP-3, used for the activation of MMP-9 (1/100 of the molar concentration of MMP-9). Digestion was stopped by the addition of reducing SDS-PAGE loading buffer, and the samples were analyzed by SDS-PAGE. To compare the cleavage efficiencies, it was estimated from the gels at which concentration of the MMP-9 variants the substrate was cleaved for 50%. For IL-8 and mGCP-2/LIX, the cleavage efficiency was determined by mass spectrometry analysis as described (28).

SPR Analysis of the Binding of MMP-9 Variants to TIMP-1, LRP-1, and Megalin
The binding to TIMP-1, LRP-1, and megalin was studied by SPR analysis on a Biacore 3000 instrument (Biacore, Sweden) as follows. Biacore sensor chips type CM5 were activated with a 1:1 mixture of 0.2 M N-ethyl-N'-(3-dimethylaminopropyl)carbodiimide and 0.05 M N-hydroxysuccimide in water according the manufacturer. LRP was immobilized at 15 µg/ml in 10 mM sodium acetate, pH 3.0. Megalin was immobilized at 10 µg/ml in 10mM sodium acetate, pH 4.5. TIMP-1 was immobilized at 10 µg/ml in 10 mM sodium acetate, pH 6.0. Remaining binding sites were blocked with 1 M ethanolamine, pH 8.5. A control flow cell was made by performing the activation and blocking procedure only. The resulting receptor densities were 31 fmol LRP-1/mm2, 25 fmol megalin/mm2, and 70 fmol TIMP-1/mm2, respectively. Samples were dissolved and analyzed at various concentrations in 100 mM Tris/HCl, 100 mM NaCl, 10 mM CaCl2, and 0.005% Tween-20 pH 7.4. Sample and running buffer were identical. Regeneration of the sensor chip after each analysis cycle was performed with 1.6 M glycine-HCl buffer pH 3.0. The Biacore response is expressed in relative response units (RU) i.e. the difference in response between protein and control flow channel. Kinetic parameters were determined by BIAevaluation 4.1 software using a Langmuir 1:1 binding model and simultaneous fitting of all curves in a concentration range from 5-500 nM (global fitting).

Endocytosis Analysis
Cellular uptake and degradation of natural MMP-9 from human neutrophils and MMP-9fl was investigated in Brown Norway rat yolk sac sarcoma epithelial cells (BN16) and COS-1 cells as described previously (35, 36). BN16 and COS-1 cells endogenously express the megalin and LRP-1 receptor, respectively. Briefly, cells were grown to confluence in 24-well plates (Nunc, Denmark) in Dulbecco modified Eagle's medium (DMEM; Invitrogen) containing 10% fetal calf serum. After washing in phosphate-buffered saline, pH 7.4, cells were incubated with 125I-labeled MMP-9fl or 125I-labeled natural MMP-9 (3000 cpm/well ~ 0.5 ng) in serum-free DMEM containing 1% bovine serum albumin. In the inhibition studies, BN16 and COS-1 cells were incubated for 30 min with receptor-associated protein (RAP; 100 µg/ml), polyclonal rabbit anti-megalin (100 µg/ml) and anti-LRP-1 (300 µg/ml) antibody, respectively, or non-immune anti-myoglobin antibody (100 µg/ml or 300 µg/ml) prior to addition of radiolabeled protein. Degradation of 125I-labeled MMP-9fl or 125I-labeled natural MMP-9 in lysosomes was inhibited by addition of 300 µM chloroquine (Fluka) and leupeptin (Sigma). After incubation, the medium was harvested and trichloroacetic acid was added to it. In the meantime, the cells were washed and hereafter lysed to allow determination of cell-associated radioactivity. Radioactivity in the trichloroacetic acid-soluble fraction of the medium represented degraded protein.

Analytical Ultracentrifugation
Sedimentation velocity experiments were carried out using a Beckman Optima XL-A analytical ultracentrifuge equipped with absorbance optics with an An60Ti rotor. Samples were analyzed at concentrations between 0.075 and 0.3 mg/ml in 100 mM Tris/HCl, pH 7.4, 100 mM NaCl, and 10 mM CaCl2 at 50,000 rev./min. Scans were collected at 150-s intervals at 230 nm. Sedimentation coefficients were determined using the time derivative method (37) and were corrected for the effects of buffer composition. The partial specific volume of each protein was calculated from its amino acid and carbohydrate composition. The theoretical sedimentation coefficients of various model structures were calculated using the program HYDROPRO (38).

Molecular Modeling
Molecular modeling was performed on a Silicon Graphics Fuel workstation using InsightII and Discover software (Accelrys Inc., San Diego, CA). Figures were produced using the program Molscript (39). The molecular models of MMP-9 and the constructs were based on the crystal structure of the N-terminal domains of MMP-9 (40) and the hemopexin domain in the crystal structure of MMP-2 (41). Various models of the OG domain were generated to link the N- and C-terminal domains. The fully extended model of OG domain was built using InsightII and then restrained simulated annealing was used to generate structures of an average extension of 2.6 Å per residue (10). The extended model of OG domain allowing for the disulfide bond between the OG domain and the hemopexin domain was generated in a similar way, but in addition fixing the distance between Cys468 and the C-terminal residue of the OG domain. Compact models of the OG domain were generated by positioning the N-terminal and C-terminal domains as required to obtain a distance constraint between the first and last residue of the OG domain. In each case, the N-terminal and C-terminal domains were then added before a final round of energy minimization. N-Linked and O-linked glycan structures, chosen on the basis of the sequencing results, were generated using the data base of glycosidic linkage conformations (42) and in vacuo energy minimization to relieve unfavorable steric interactions. For the N-linked glycans, Asn-GlcNAc linkage conformations were based on the observed range of crystallographic values (43), the torsion angles around the Asn C{alpha}-Cbeta and Cbeta-C{gamma} bonds then being adjusted to eliminate unfavorable steric interactions between the glycans and the protein surface.


Figure 1
View larger version (42K):
[in this window]
[in a new window]
  		FIGURE 1.
Schematic domain structure and SDS-PAGE analysis of the recombinant MMP-9 variants. a, the first line shows full-length MMP-9 (MMP-9fl). N-Linked glycans are indicated with Y, and the conserved, unoccupied N-glycosylation sequon is indicated with a dotted Y. The probable attachment sites for O-linked sugars on the OG domain are indicated with vertical lines. The catalytic Glu402 (E) residue and the Cys468 (C) residue in the OG domain, which are mutated into Ala (A) in MMP-9MutEC, are also indicated. Pro, prodomain; Act, active site domain; FN, three repeat fibronectin domain; Zn2+, Zn2+-binding domain; OG, O-glycosylated domain; Hem, hemopexin domain. b, different recombinant MMP-9 variants, depicted in a, were produced in Sf9 insect cells with a baculoviral transfection system. After purification, the variants were analyzed by reducing and non-reducing SDS-PAGE to estimate the molecular masses and to visualize di- and multimerization, respectively. Nat MMP-9, natural MMP-9 isolated from human neutrophils; St, molecular mass reference standard.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Expression and Purification of Recombinant Domain Variants of MMP-9—Five recombinant protein variants of gelatinase B/MMP-9 were generated as described under "Experimental Procedures" (Fig. 1a). Each of the mutants retained its affinity for gelatin and was therefore purified to homogeneity in a similar way as the full-length enzyme, i.e. by substrate-affinity chromatography on gelatin-Sepharose.

Molecular Mass and Dimerization of the Recombinant Variants of MMP-9—Reducing and non-reducing SDS-PAGE tests were used to determine relative molecular weight and dimerization patterns of the different recombinant variants of MMP-9 (Fig. 1b). Full-length recombinant MMP-9 (MMP-9fl) migrated at a slightly lower position than natural MMP-9 from human neutrophils, probably because of differences in glycosylation (see below). Interestingly, deletion of the OG or the hemopexin domain resulted in a similar decrease in molecular mass, although the OG domain is only 64 amino acids long, whereas the hemopexin domain contains 196 amino acids. This finding might be explained by the O-glycosylation of the OG domain (see below).

An interesting feature of MMP-9 is the formation of dimers/multimers, which are SDS-resistant but can be dissociated by reduction. In contrast to the variants containing the OG domain, MMP-9{Delta}OG and MMP-9{Delta}OGHem showed only monomers on non-reducing SDS-PAGE and in analytical ultracentrifugation experiments (see below). This indicates that the OG domain is essential and sufficient for dimerization/multimerization. Though, the single cysteine residue in the OG domain, absent in MMP-9mutEC, is not essential for dimerization/multimerization. The exact mechanisms of dimerization remain therefore unknown.

Core-fucosylated Trimannosyl Chitobiose Core N-Linked Sugars Are Attached to the Propeptide and Active Site Domain— N-Glycans were analyzed by NP-HPLC, following their release from the intact protein by in-gel PNGase F digestion and fluorescent labeling. The elution times were compared with that of a dextran hydrolysate ladder and normalized as GU. By comparison of the obtained GU values with those in a data base (22), provisional structures could be assigned. Subsequently, these structures were confirmed by re-analysis of the glycans after specific exoglycosidase digests (Fig. 2). All the different variants expressed in Sf9 cells contained the N-linked glycan core structure (Man3GlcNAc2) with a core fucosylation. Three possible N-glycosylation sites are present in the sequence of MMP-9, one in the propeptide (Asn38-Leu-Thr) and two in the active domain (Asn120-Ile-Thr and Asn127-Tyr-Ser). To define the exact localization of the N-linked glycans, recombinant MMP-9fl was digested with trypsin and the resulting peptides were reduced, alkylated, separated by RP-HPLC, and analyzed by mass spectrometry and Edman degradation (data not shown). This revealed the presence of N-linked glycans on Asn38 in the propeptide and Asn120 in the catalytic domain, whereas Asn127, is unsubstituted. Fragmentation analysis (MS/MS and MS3) confirmed the structure and localization of the sugars of recombinant MMP-9. The localization of the N-linked sugars in recombinant MMP-9, produced in Sf9 cells, is in agreement with previous data with natural human MMP-9 (8) and recombinant MMP-9 expressed in HeLa-cells (44).


Figure 2
View larger version (24K):
[in this window]
[in a new window]
  		FIGURE 2.
N-Linked glycosylation analysis of the recombinant MMP-9 variants. N-Linked sugars of the recombinant MMP-9 variants were released from the protein, purified, and fluorescently labeled. After NP-HPLC analysis, a similar profile was obtained for the different variants. The N-glycan profile of MMP-9fl is shown on top. Confirmation of the structure was obtained by specific exoglycosidase digestion with bovine kidney fucosidase (BKF) and/or Jack bean mannosidase (JBM) and re-analysis with NP-HPLC (middle and lower panels). Symbol representation of glycans: N-acetylglucosamine, filled square; mannose, open circle; fucose, diamond with a dot inside; beta-linkage, solid line; {alpha}-linkage, dotted line. The orientation of lines showing the linkages between the monosaccharide residues correspond with the sugar linkage as follows: 1-4 linkage, -; 1-6 linkage, \; 1-3 linkage, /.

 
Presence of Abundant O-Linked Glycans Attached to the OG Domain—The OG domain of MMP-9 is unique, as it is absent in all other MMPs. Its sequence contains several Pro residues, explaining a low homology with collagen type V (2). However, this domain lacks the typical collagen motif with Gly repetition at every third residue. In contrast, it contains 12 repeats of the sequence Thr/Ser-Xxx-Xxx-Pro. Because O-glycosylation occurs often on Thr or Ser when a Pro residue is just in front of or three residues behind these Thr or Ser (45), such repeats are good candidates for O-glycosylation. Analysis of the MMP-9 sequence by the NetOGlyc server (46), and comparison with those of all other MMPs, also revealed that the OG domain, and only this domain in MMP-9, probably contains abundant attachment sites for O-linked oligosaccharides (data not shown).

To test this hypothesis, the binding of intact and deletion variants of recombinant MMP-9 to two lectins, specific for the most commonly found O-linked glycans on proteins expressed in Sf9 cells (H. pomatia agglutinin, HPA, specific for Ser- or Thr-linked GalNAc, and jacalin, specific for Gal-GalNAc), was analyzed. No binding to jacalin was observed, but all MMP-9fl bound to HPA-agarose. MMP-9{Delta}Hem bound similarly to HPA-agarose, but MMP-9{Delta}OG and MMP-9{Delta}OGHem did not bind to HPA (data not shown). This shows that the OG domain is O-glycosylated, whereas O-glycans are most likely absent on the other domains of MMP-9.


Figure 3
View larger version (29K):
[in this window]
[in a new window]
  		FIGURE 3.
NP-HPLC and LC-ESI-MS analysis of O-glycans from the recombinant MMP-9 variants. O-linked oligosaccharides were released from the proteins by ammonia-based beta-elimination. After purification and fluorescent labeling, the glycans were analyzed by: 1) NP-HPLC, shown in the lower trace in each panel, and 2) LC-ESI-MS, where the upper traces in each panel show the relative abundance of ions with m/z of 301, 342, and 504 corresponding to Hex-2AB, HexNAc-2AB, and HexNAc-Hex-2AB, respectively, against retention time.

 
These results were confirmed by a detailed glycosylation analysis of MMP-9fl and the different deletion mutants. The glycans were released from the MMP-9 variants by ammonia-based beta-elimination, fluorescently labeled with 2AB and analyzed by NP-HPLC and LC-MS analysis (Fig. 3). GU values of the eluted compounds a, b, and c were obtained by comparison of the elution times with those of a standard dextran hydrolysate ladder. Comparison of GU values to those in a data base (23), and the m/z data indicate that: Peak a is GalNAc-2AB with a GU value of 0.9 and a m/z of 342; Peak b is a hexose-2AB (most likely glucose-2AB) with a GU of 1.0 and a m/z of 301; Peak c is core1-2AB with a GU of 1.8 and a m/z of 504. Peak b is detected in all of the samples and is likely to have come from background contamination, since it was also detected in a blank run. Peaks a & c are detected only in the variants containing the OG domain (MMP-9fl and MMP-9{Delta}Hem) confirming the presence of O-linked GalNAc and small amounts of core-1 (Galbeta1-3GalNAc) on the OG domain, whereas the mutants without OG domain were not O-glycosylated.


Figure 4
View larger version (30K):
[in this window]
[in a new window]
  		FIGURE 4.
Analytical ultracentrifugation and in silico modeling of MMP-9 variants. Using the existing crystallographic data of the catalytic part and the hemopexin domain, a variety of models of the different variants of MMP-9 were constructed in silico, with different degrees of compactness of the OG domain. Theoretical sedimentation coefficients were calculated for each of these models and compared with experimentally obtained values (Table 1). Models which do not reproduce the experimental data are shown in gray. Yellow, N-linked glycans; orange, O-linked sugars; green, catalytic domains; blue, fibronectin repeat; light blue, OG domain; pink, hemopexin domain.

 
Sedimentation Analysis and Molecular Modeling of the MMP-9 Variants Show That the OG Domain Is Not Extended—Crystal structures of the N terminal part and the hemopexin domain of MMP-9 have been determined separately. However, nothing is known about the three-dimensional structure of the OG domain. Presumably, the multiple O-linked sugars strongly influence the three-dimensional structure of this domain. Furthermore, the presence of a cysteine (Cys468) in this domain may impose additional constraints to the three-dimensional structure. It seems reasonable to assume that, in monomers, Cys468 forms a disulfide bridge with Cys674 in the fourth blade of the hemopexin domain, because these two cysteines are the only ones in MMP-9 for which the coupling cysteine is not known. Different models were generated in silico with different degrees of extension/compactness of the OG domain (Fig. 4). To estimate which of the different models is most realistic, sedimentation coefficients of the different recombinant MMP-9 variants were determined by analytical ultracentrifugation and compared with the in silico calculated sedimentation coefficients of the different models (Table 1). The experimentally obtained sedimentation constants exclude the possibility of extended conformations and are consistent with the presence of a disulfide bridge between Cys468 and Cys674. This disulfide bridge, however, remains to be proven. In addition, mutation of Cys468 (in MMP-9MutEC) or deletion of the hemopexin domain (MMP-9{Delta}Hem) do not lead to a fully extended conformation, showing that the possible Cys468-Cys674 disulfide bridge is not essential for a (partially) folded conformation of the OG domain.


View this table:
[in this window]
[in a new window]
  		TABLE 1
Comparison of the theoretical sedimentation coefficients of different models of the MMP-9 variants with the experimental sedimentation coefficients

 
Comparison of the Catalytic Activities of Domain Mutants of MMP-9 —Exosites are important for the efficient cleavage of a number of substrates by some MMPs (15). Therefore, it is possible that the hemopexin and OG domains of MMP-9 also contain exosites enhancing the cleavage of particular substrates. To test this hypothesis, the different variants were activated with the catalytic domain of MMP-3. The activation reaction progressed similarly for the different variants (data not shown). Subsequently, the cleavages of a number of known substrates by the different activated deletion variants were compared (Table 2). No significant differences were observed.


View this table:
[in this window]
[in a new window]
  		TABLE 2
Experimental substrate cleavage efficiencies by recombinant MMP-9 variants

 
Another theoretical possibility was that the OG domain might create a distance between the hemopexin domain and the catalytic site, thereby preventing the cleavage of e.g. MCP-3 bound to the hemopexin domain. Therefore, the cleavage of MCP-3 with the different variants was tested. No cleavage was observed, even with MMP-9{Delta}OG, although the latter mutant structurally resembles MMP-2, which efficiently cleaves MCP-3 (14).

Both the OG and Hemopexin Domains of MMP-9 Are Critical for Interaction with TIMP-1—The activity of MMP-9 is regulated in several ways, including inhibition by TIMPs. TIMP-1 efficiently inhibits MMP-9, because it does not only bind with its N terminus into the active site, but also with its C-terminal domain to the hemopexin domain of MMP-9 (6). The role of the OG domain in this context has not been studied yet. Therefore, the binding of the different recombinant proMMP-9 variants to TIMP-1 was compared by surface plasmon resonance (Fig. 5 and Table 3). As expected, proMMP-9{Delta}Hem and proMMP-9{Delta}OGHem showed no binding to TIMP-1. Unexpectedly, the binding affinity of proMMP-9{Delta}OG for TIMP-1 was ~4-fold lower. This result reinforces the idea that the OG domain of MMP-9 is important for optimal binding of TIMP-1, suggesting that in proMMP-9{Delta}OG, the hemopexin domain is too close to the N-terminal domains leading to steric hindering of TIMP-1 binding.


View this table:
[in this window]
[in a new window]
  		TABLE 3
Binding of MMP-9 variants to TIMP-1

 


Figure 5
View larger version (31K):
[in this window]
[in a new window]
  		FIGURE 5.
Binding of different MMP-9 variants to TIMP-1. The binding of proMMP-9 variants to TIMP-1 was analyzed by surface plasmon resonance to assess the role of the OG and hemopexin domains. The different variants are indicated in the graphs and included natural neutrophil MMP-9, which is partially complexed to NGAL, and the different recombinant variants. For each MMP-9 variant, a range of concentrations was analyzed (from top to bottom in each graph: 500, 200, 100, 50, 20, 10, and 5 nM). The dissociation constants (Kd) were calculated from the binding curves and are indicated in Table 3.

 


Figure 6
View larger version (23K):
[in this window]
[in a new window]
  		FIGURE 6.
Endocytosis of natural and recombinant MMP-9 through LRP-1. Both natural MMP-9 from human neutrophils and recombinant MMP-9fl were radioactively labeled and added to LRP-1 expressing COS-1 cells. A, time course of degraded (•) and cell-associated ({circ}) radioactivity in cells incubated with 125I-labeled recombinant MMP-9fl. B, same experiment as in A with the addition of 300 µM chloroquine and leupeptin. C, total uptake of 125I-labeled MMP-9 fl after 4 h of incubation in presence of RAP (100 µg/ml), rabbit anti-LRP IgG (300 µg/ml), or non-immune IgG (300 µg/ml). D-F, similar experiments as in A-C using natural 125I-labeled MMP as radioligand. All values are the measured radioactivity in percent of the added radioactivity and are the mean ± 1 S.D. of triplicate determinations. Total uptake in C and F is the sum of degraded and cell-associated radioactivity.

 
LRP-1 Binds MMP-9 through Its Hemopexin Domain—Another regulation mechanism for MMP-9 is its removal from extracellular fluids by binding to LRP-1, followed by internalization and degradation (7). LRP-1 expressing COS-1 cells were indeed found to bind, internalize, and degrade both natural and recombinant MMP-9. This could be inhibited by the receptor-associated protein (RAP) and by LRP-1-specific antibodies, confirming the involvement of LRP-1 (Fig. 6). To determine which domain(s) of MMP-9 are responsible for the binding to LRP-1, the different variants of MMP-9 were analyzed for binding to LRP-1 with the use of surface plasmon resonance (Fig. 7 and Table 4). MMP-9fl bound LRP-1 with high affinity. However, deletion of the hemopexin domain (in MMP-9{Delta}Hem and MMP-9{Delta}OGHem) resulted in the loss of binding. This indicates that the binding site for LRP-1 resides in the hemopexin domain of MMP-9.


View this table:
[in this window]
[in a new window]
  		TABLE 4
Binding of MMP-9 variants to LRP-1

 
The OG Domain Optimizes LRP-1 Binding—Similar to the binding of MMP-9 to TIMP-1, deletion of the OG domain in MMP-9{Delta}OG resulted in a significantly decreased affinity for LRP-1 (Table 4). This suggests that the OG domain is important for the correct orientation of the hemopexin domain. However, the binding sites for LPR-1 and TIMP-1 are unlikely to overlap, because proMMP-9 from THP-1 cells, which is complexed to TIMP-1 (47), binds with similar affinity to LRP-1 as uncomplexed MMP-9 (Fig. 7 and Table 4).


Figure 7
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 7.
Binding of MMP-9 variants to LRP-1. The binding of MMP-9 variants to LRP-1 was analyzed by surface plasmon resonance. The different variants are indicated in the graphs and included natural neutrophil MMP-9, which is partially complexed to NGAL, THP-1 MMP-9, which is complexed to TIMP-1, and the different recombinant variants. Different concentrations of each MMP-9 variant were used (from top to bottom in each graph: 500, 200, 100, 50, 20, 10, and 5 nM). The dissociation constants (Kd) were calculated from the binding curves and are indicated in Table 4.

 


Figure 8
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 8.
Identification of megalin as a new endocytosis receptor for MMP-9. Radioactively labeled natural neutrophil MMP-9 and recombinant MMP-9fl were added to cell cultures of an immortalized rat yolk sac cell line, BN16. A, time course of degraded (•) and cell-associated ({circ}) radioactivity in cells incubated with 125I-labeled recombinant MMP-9fl. B, same experiment as in A with the addition of 300 µM chloroquine and leupeptin. C, total uptake of 125I-labeled MMP-9fl after 4 h of incubation in presence of RAP (100 µg/ml), rabbit anti-megalin IgG (100 µg/ml) or non-immune IgG (100 µg/ml). D-F, similar experiments as in A-C using natural 125I-labeled MMP-9 as radioligand. All values are the measured radioactivity in percent of the added radioactivity and are the mean ± 1 S.D. of triplicate determinations. Total uptake in C and F is the sum of degraded and cell-associated radioactivity.

 
Identification of Megalin/LRP-2 as a New Functional Endocytotic Receptor for MMP-9—Megalin/LRP-2 is a receptor from the LDL receptor family, similar to LRP-1. Recently, it was shown to bind neutrophil gelatinase B-associated lipocalin (NGAL) (35). Because NGAL also occurs as a complex with MMP-9 in neutrophils, it was interesting to analyze whether megalin would also bind neutrophil MMP-9, and also recombinant human MMP-9, which does not contain NGAL. Therefore, BN16 rat yolk sac cells, which express megalin, were incubated with radioactively labeled MMP-9, and the internalization of the enzyme was analyzed. Significant internalization of both recombinant and natural MMP-9 was detected and was followed by degradation (Fig. 8). The internalization could be inhibited with RAP and with anti-megalin antibodies, indicating the essential role of megalin.


Figure 9
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 9.
Binding of MMP-9 variants to megalin. The binding of MMP-9 to megalin/LRP-2 was analyzed by surface plasmon resonance. Different proMMP-9 variants, indicated in the graphs, were used, including the domain deletion mutants to assess the role of the OG and hemopexin domains. Furthermore, natural neutrophil MMP-9, which is partially complexed to NGAL, a known ligand of megalin, and THP-1 MMP-9, which is complexed to TIMP-1, was also included to evaluate the effects of the respective complex formations. Different concentrations of each MMP-9 variant were used (from top to bottom in each graph: 500, 200, 100, 50, 20, 10, and 5 nM). The dissociation constants (Kd) were calculated from the binding curves and are indicated in Table 5.

 
Megalin Binds MMP-9 Through Its Hemopexin Domain—To determine whether MMP-9 binds directly to megalin and to investigate which domain(s) of MMP-9 are responsible for this binding, the affinities of the different variants of MMP-9 for megalin were analyzed with the use of surface plasmon resonance (Fig. 9 and Table 5). Natural neutrophil MMP-9, partially complexed with NGAL, THP-1 MMP-9, and recombinant MMP-9fl bound to megalin with similar affinities. These data on the binding and internalization of MMP-9 define megalin as a new functional receptor for MMP-9. A similar pattern of hemopexin and OG domain-dependent interactions was observed as for the binding of TIMP-1 and LRP-1, indicating that the tertiary structure of the OG domain optimizes the functionality of MMP-9 binding to TIMP-1, LRP-1, and megalin.


View this table:
[in this window]
[in a new window]
  		TABLE 5
Binding of MMP-9 variants to Megalin

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Natural MMP-9 produced by neutrophils is a heavily glycosylated protein. Approximately 85% of the glycans on natural MMP-9 are O-linked, and the remaining 15% are N-linked sugars (9). In the supposition that the numbers of N- and O-linked glycans are comparable in natural and recombinant MMP-9, the combination of the relative ratio between N- and O-linked sugars (85:15) and the presence of 2 N-linked sugars would suggest that more than 10 O-linked sugars may be present. By using recombinant DNA technology and glycosylation analysis, a series of domain mutants were engineered and compared, indicating that the O-linked oligosaccharides are attached to the unique OG domain for which no function was known to date. This corresponds with the known primary sequence preferences of O-linked glycosylation, for which a Pro is preferred one position in front of or 3 positions behind Ser or Thr (45), as the OG domain is composed of 12 repeats of the sequence (Thr/Ser)-Xxx-Xxx-Pro. In natural MMP-9 isolated from human neutrophils, the O-linked sugars consist of the classical core 1 (Gal-GalNAc-O-Ser) or core 2 (Gal-(GlcNAc-)GalNAc-O-Ser) structures, which are further elongated to relatively large glycans, consisting of up to 10 monosaccharide residues (10). Such oligosaccharides may have various functions and may e.g. shield the protein backbone against proteolysis (45).

The O-linked glycans in the recombinant variants of MMP-9 contain, however, only basic structures with 1 or 2 monosaccharides (GalNAc-O-Ser and Gal-GalNAc-O-Ser). Because this could lead to functional differences, comparison with natural MMP-9 isolated from human neutrophils and, where applicable, with the MMP-9-TIMP-1 complex isolated from the monocytic leukemia cell line THP-1 was performed to detect possible functions of the full-length glycans. Comparison between the recombinant full-length variant and deletion mutants allowed for a further dissection of the role of the protein part of the OG domain with the basic monosaccharide structures.

Because little information was available on the overall three-dimensional structure of the OG domain, a combined strategy of in silico modeling and experimental analysis of sedimentation coefficients was used to gain information on the overall three-dimensional structure of the OG domain. This indicated that the structure of the OG domain is unique and not extended (such as in mucins) but rather more compact, and is consistent with the proposal that Cys468 in the OG domain may form a disulfide bridge with Cys674 in the hemopexin domain. This disulfide bridge, however, remains to be proven.

Deletion of the OG and/or hemopexin domains did not influence significantly the cleavage of a number of known substrates of MMP-9, excluding the possibility that these two domains contain exosites for these substrates. In contrast, both domains play a key role in the biological regulation of MMP-9. It has been known for some time that TIMP-1 binds to the C-terminal part of proMMP-9 (6, 48). The OG domain is shown here to influence the binding of TIMP-1 to the hemopexin domain, as deletion of the OG domain results in decreased affinity for TIMP-1. Down-regulation of MMP-9 also occurs by internalization and catabolism after binding to LRP-1 (7). By surface plasmon resonance analysis, the hemopexin domain is found to be essential for this binding. The OG domain is also important for this binding, since its deletion leads to a 10-fold decrease in affinity. The main differences between MMP-9{Delta}OG and the full-length forms in the binding to TIMP-1 and LRP-1 reside in the kon rate, which is concentration-dependent. This is not likely to be caused by misfolding of MMP-9{Delta}OG, because its catalytic activity is similar to that of MMP-9fl. It may rather indicate that the binding sites for TIMP-1 and LRP-1 in the hemopexin domain are less accessible, suggesting that the OG domain is required for the correct orientation of the hemopexin domain. However, TIMP-1 and LRP-1 probably do not compete for binding to MMP-9, since the proMMP-9-TIMP-1 complex binds to LRP-1 with the same affinity as intact proMMP-9.

In addition, internalization studies showed for the first time that megalin/LRP-2 is a functional endocytosis receptor for MMP-9. Binding, uptake, and chloroquine/leupeptin-sensitive degradation of MMP-9 by immortalized megalin-expressing BN16 rat yolk sac cells was inhibitable by RAP and anti-megalin antibodies. Whether endocytosis of MMP-9 is mediated in vivo by LRP-1 and/or by megalin, is likely to be dependent on the local expression levels of both receptors. It is therefore tempting to speculate that megalin will account for the endocytosis of MMP-9 in epithelial tissues such as the visceral yolk sac and renal proximal tubules, where megalin is predominantly expressed (49). LRP-1 is more predominant in the liver and on macrophages (50). By means of surface plasmon resonance, it was found that MMP-9 binds directly to megalin and that the hemopexin domain of MMP-9 is essential for this binding. Also the OG domain was important to confer high affinity to megalin/LRP-2, suggesting that the hemopexin domain orientation by the OG domain might be a common mechanism for efficient inhibition by TIMP-1 and receptor-mediated internalization of MMP-9 by LRP-1 and megalin. Both mechanisms may contribute as negative feedback in the regulation of the bioavailability of proteolytically active MMP-9.

Hemopexin domains in MMPs have a significant homology with hemopexin, which is abundantly present in plasma. The main function of hemopexin is to capture free heme that may be liberated in plasma upon red blood cell lysis, as free heme diplays strong inflammatory and toxic properties. Recently, the endocytotic receptor for the hemopexin-heme complex was identified as LRP-1 (36). The present findings show that the hemopexin domain of MMP-9 also binds LRP-1. Because this binding is the only known function shared by both hemopexin itself and the MMP hemopexin domains, it may be assumed that binding to LRP-1 was the primary function of the hemopexin domain in MMPs during evolution. In MMP-2 and MMP-13, which do not contain an OG domain, the binding of the hemopexin domain to LRP-1 involves indirect mechanisms and other molecules. The binding of gelatinase A/MMP-2 to LRP-1 is significantly enhanced by complex formation with either thrombospondin-1 or -2 or with TIMP-2 (51, 52). In the case of MMP-13, an unknown 170 kDa receptor first binds proMMP-13 to the cell surface and subsequently transfers the enzyme to LRP-1 for internalization and degradation (53). As a comparison, no such indirect or enhancing mechanisms have been described for MMP-9 to date, but the hemopexin domain function of MMP-9 may have been further refined by natural selection into a molecular domain with a defined orientation through the OG domain. Another function for the hemopexin domain was demonstrated recently and consists of the localization of gelatinase B/MMP-9 to the membrane by binding to a membrane-bound receptor, the DNA repair protein Ku (54).

As a conclusion, analysis of recombinant variants of MMP-9 shows that the hemopexin domain is essential for the interaction between MMP-9 and LRP-1 and megalin. In addition, it is documented for the first time that megalin is a new functional endocytosis receptor for MMP-9. Furthermore, a biological function of the unique OG domain of MMP-9 is found. This domain with abundant O-linked glycans and a single cysteine residue is important for an optimal binding of the hemopexin domain to the inhibitor TIMP-1 and the two internalization cargo receptors, LRP-1 and megalin, thus regulating the bioavailability of active MMP-9.


   FOOTNOTES
 
* This study was supported by the Geconcerteerde OnderzoeksActies 2002-2006 and 2006-2010, the National Fund for Scientific Research (F.W.O.-Vlaanderen), the Centre of Excellence EF/05/015, the Charcot Foundation, and the Federation against Cancer (Belgium). This work was also supported by funding from the Oxford Glycobiology Institute. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 A postdoctoral fellow of the F.W.O.-Vlaanderen. Back

2 To whom correspondence should be addressed: Laboratory of Immunobiology, Rega Institute for Medical Research, University of Leuven, Minderbroedersstraat 10, 3000 Leuven, Belgium. Tel.: 32-16-337341; Fax: 32-16-337340; E-mail: ghislain.opdenakker{at}rega.kuleuven.be.

3 The abbreviations used are: MMPs, matrix metalloproteinases; 2-AB, 2-aminobenzamide; fl, full-length; GU, glucose units; HPA, helix pomatia agglutinin; IL, interleukin; LC-ESI-MS, liquid chromatography-electrospray ionization-mass spectrometry; LRP, low density lipoprotein receptor-related protein; MCP-3, monocyte chemotactic protein-3; mGCP-2/LIX, mouse granulocyte chemotactic protein-2; MT-, membrane type-; NGAL, neutrophil gelatinase B-associated lipocalin; OG, O-glycosylated; RU, response units; SPR, surface plasmon resonance; TIMP, tissue inhibitor of MMPs. Back


   ACKNOWLEDGMENTS
 
We thank S. Husson for generating one of the domain deletion mutants.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Nagase, H., and Woessner, J. F. (1999) J. Biol. Chem. 274, 21491-21494[Free Full Text]
  2. Wilhelm, S. M., Collier, I. E., Marmer, B. L., Eisen, A. Z., Grant, G. A., and Goldberg, G. I. (1989) J. Biol. Chem. 264, 17213-17221[Abstract/Free Full Text]
  3. Opdenakker, G., Van den Steen, P. E., and Van Damme, J. (2001) Trends Immunol. 22, 571-579[CrossRef][Medline] [Order article via Infotrieve]
  4. Parks, W. C., Wilson, C. L., and Lopez-Boado, Y. S. (2004) Nat. Rev. Immunol. 4, 617-629[CrossRef][Medline] [Order article via Infotrieve]
  5. Maskos, K., and Bode, W. (2003) Mol. Biotechnol. 25, 241-266[CrossRef][Medline] [Order article via Infotrieve]
  6. O'Connell, J. P., Willenbrock, F., Docherty, A. J., Eaton, D., and Murphy, G. (1994) J. Biol. Chem. 269, 14967-14973[Abstract/Free Full Text]
  7. Hahn-Dantona, E., Ruiz, J. F., Bornstein, P., and Strickland, D. K. (2001) J. Biol. Chem. 276, 15498-15503[Abstract/Free Full Text]
  8. Van den Steen, P. E., Opdenakker, G., Wormald, M. R., Dwek, R. A., and Rudd, P. M. (2001) Biochim. Biophys. Acta 1528, 61-73[Medline] [Order article via Infotrieve]
  9. Rudd, P. M., Mattu, T. S., Masure, S., Bratt, T., Van den Steen, P. E., Wormald, M. R., Küster, B., Harvey, D. J., Borregaard, N., Van Damme, J., Dwek, R. A., and Opdenakker, G. (1999) Biochemistry 38, 13937-13950[CrossRef][Medline] [Order article via Infotrieve]
  10. Mattu, T. S., Royle, L., Langridge, J., Wormald, M. R., Van den Steen, P. E., Van Damme, J., Opdenakker, G., Harvey, D. J., Dwek, R. A., and Rudd, P. M. (2000) Biochemistry 39, 15695-15704[CrossRef][Medline] [Order article via Infotrieve]
  11. Wu, Y. I., Munshi, H. G., Sen, R., Snipas, S. J., Salvesen, G. S., Fridman, R., and Stack, M. S. (2004) J. Biol. Chem. 279, 8278-8289[Abstract/Free Full Text]
  12. Lauer-Fields, J. L., Juska, D., and Fields, G. B. (2002) Biopolymers 66, 19-32[CrossRef][Medline] [Order article via Infotrieve]
  13. Allan, J. A., Docherty, A. J., Barker, P. J., Huskisson, N. S., Reynolds, J. J., and Murphy, G. (1995) Biochem. J. 309, 299-306[Medline] [Order article via Infotrieve]
  14. McQuibban, G. A., Gong, J. H., Tam, E. M., McCulloch, C. A., Clark-Lewis, I., and Overall, C. M. (2000) Science 289, 1202-1206[Abstract/Free Full Text]
  15. Overall, C. M. (2002) Mol. Biotechnol. 22, 51-86[CrossRef][Medline] [Order article via Infotrieve]
  16. Masure, S., Proost, P., Van Damme, J., and Opdenakker, G. (1991) Eur. J. Biochem. 198, 391-398[Medline] [Order article via Infotrieve]
  17. Horton, R. M., Hunt, H. D., Ho, S. N., Pullen, J. K., and Pease, L. R. (1989) Gene (Amst.) 77, 61-68[CrossRef][Medline] [Order article via Infotrieve]
  18. Knight, C. G., Willenbrock, F., and Murphy, G. (1992) FEBS Lett. 296, 263-266[CrossRef][Medline] [Order article via Infotrieve]
  19. Küster, B., Wheeler, S. F., Hunter, A. P., Dwek, R. A., and Harvey, D. J. (1997) Anal. Biochem. 250, 82-101[CrossRef][Medline] [Order article via Infotrieve]
  20. Huang, Y., Konse, T., Mechref, Y., and Novotny, M. V. (2002) Rapid Commun. Mass Spectrom. 16, 1199-1204[CrossRef][Medline] [Order article via Infotrieve]
  21. Bigge, J. C., Patel, T. P., Bruce, J. A., Goulding, P. N., Charles, S. M., and Parekh, R. B. (1995) Anal. Biochem. 230, 229-238[CrossRef][Medline] [Order article via Infotrieve]
  22. Guile, G. R., Rudd, P. M., Wing, D. R., Prime, S. B., and Dwek, R. A. (1996) Anal. Biochem. 240, 210-226[CrossRef][Medline] [Order article via Infotrieve]
  23. Royle, L., Mattu, T. S., Hart, E., Langridge, J. I., Merry, A. H., Murphy, N., Harvey, D. J., Dwek, R. A., and Rudd, P. M. (2002) Anal. Biochem. 304, 70-90[CrossRef][Medline] [Order article via Infotrieve]
  24. Van den Steen, P. E., Proost, P., Grillet, B., Brand, D. D., Kang, A. H., Van Damme, J., and Opdenakker, G. (2002) FASEB J. 16, 379-389[Abstract/Free Full Text]
  25. Van den Steen, P. E., Proost, P., Brand, D. D., Kang, A. H., Van Damme, J., and Opdenakker, G. (2004) Biochemistry 43, 10809-10816[CrossRef][Medline] [Order article via Infotrieve]
  26. Desrochers, P. E., Mookhtiar, K., Van Wart, H. E., Hasty, K. A., and Weiss, S. J. (1992) J. Biol. Chem. 267, 5005-5012[Abstract/Free Full Text]
  27. Descamps, F. J., Martens, E., Proost, P., Starckx, S., Van den Steen, P. E., Van Damme, J., and Opdenakker, G. (2005) FASEB J. 19, 29-35[Abstract/Free Full Text]
  28. Van den Steen, P. E., Wuyts, A., Husson, S. J., Proost, P., Van Damme, J., and Opdenakker, G. (2003) Eur. J. Biochem. 270, 3739-3749[Medline] [Order article via Infotrieve]
  29. Van den Steen, P. E., Proost, P., Wuyts, A., Van Damme, J., and Opdenakker, G. (2000) Blood 96, 2673-2681[Abstract/Free Full Text]
  30. Ochieng, J., Fridman, R., Nangia-Makker, P., Kleiner, D. E., Liotta, L. A., Stetler-Stevenson, W. G., and Raz, A. (1994) Biochemistry 33, 14109-14114[CrossRef][Medline] [Order article via Infotrieve]
  31. Fiore, E., Fusco, C., Romero, P., and Stamenkovic, I. (2002) Oncogene 21, 5213-5223[CrossRef][Medline] [Order article via Infotrieve]
  32. Sheu, B. C., Hsu, S. M., Ho, H. N., Lien, H. C., Huang, S. C., and Lin, R. H. (2001) Cancer Res. 61, 237-242[Abstract/Free Full Text]
  33. Patterson, B. C., and Sang, Q. A. (1997) J. Biol. Chem. 272, 28823-28825[Abstract/Free Full Text]
  34. Sasaki, T., Gohring, W., Mann, K., Maurer, P., Hohenester, E., Knaüper, V., Murphy, G., and Timpl, R. (1997) J. Biol. Chem. 272, 9237-9243[Abstract/Free Full Text]
  35. Hvidberg, V., Jacobsen, C., Strong, R. K., Cowland, J. B., Moestrup, S. K., and Borregaard, N. (2005) FEBS Lett. 579, 773-777[CrossRef][Medline] [Order article via Infotrieve]
  36. Hvidberg, V., Maniecki, M. B., Jacobsen, C., Hojrup, P., Moller, H. J., and Moestrup, S. K. (2005) Blood 106, 2572-2579[Abstract/Free Full Text]
  37. Stafford, W. F., III (1992) Anal. Biochem. 203, 295-301[CrossRef][Medline] [Order article via Infotrieve]
  38. Garcia De La Torre, J., Huertas, M. L., and Carrasco, B. (2000) Biophys. J. 78, 719-730[Medline] [Order article via Infotrieve]
  39. Kraulis, P. J. (1991) J. Appl. Crystallogr. 24, 946-950[CrossRef]
  40. Elkins, P. A., Ho, Y. S., Smith, W. W., Janson, C. A., D'Alessio, K. J., McQueney, M. S., Cummings, M. D., and Romanic, A. M. (2002) Acta Crystallogr. D. Biol. Crystallogr. 58, 1182-1192[CrossRef][Medline] [Order article via Infotrieve]
  41. Morgunova, E., Tuuttila, A., Bergmann, U., Isupov, M., Lindqvist, Y., Schneider, G., and Tryggvason, K. (1999) Science 284, 1667-1670[Abstract/Free Full Text]
  42. Wormald, M. R., Petrescu, A. J., Pao, Y. L., Glithero, A., Elliott, T., and Dwek, R. A. (2002) Chem. Rev. 102, 371-386[CrossRef][Medline] [Order article via Infotrieve]
  43. Petrescu, A. J., Milac, A. L., Petrescu, S. M., Dwek, R. A., and Wormald, M. R. (2004) Glycobiology 14, 103-114[Abstract/Free Full Text]
  44. Kotra, L. P., Zhang, L., Fridman, R., Orlando, R., and Mobashery, S. (2002) Bioorg. Chem. 30, 356-370[CrossRef][Medline] [Order article via Infotrieve]
  45. Van den Steen, P., Rudd, P. M., Dwek, R. A., and Opdenakker, G. (1998) Crit. Rev. Biochem. Mol. Biol. 33, 151-208[Medline] [Order article via Infotrieve]
  46. Julenius, K., Molgaard, A., Gupta, R., and Brunak, S. (2005) Glycobiology 15, 153-164[Abstract/Free Full Text]
  47. Van Ranst, M., Norga, K., Masure, S., Proost, P., Vandekerckhove, F., Auwerx, J., Van Damme, J., and Opdenakker, G. (1991) Cytokine. 3, 231-239[CrossRef][Medline] [Order article via Infotrieve]
  48. Murphy, G., Houbrechts, A., Cockett, M. I., Williamson, R. A., O'Shea, M., and Docherty, A. J. (1991) Biochemistry 30, 8097-8102[CrossRef][Medline] [Order article via Infotrieve]
  49. Moestrup, S. K., and Verroust, P. J. (2001) Annu. Rev. Nutr. 21, 407-428[CrossRef][Medline] [Order article via Infotrieve]
  50. Moestrup, S. K., Gliemann, J., and Pallesen, G. (1992) Cell Tissue Res. 269, 375-382[CrossRef][Medline] [Order article via Infotrieve]
  51. Yang, Z., Strickland, D. K., and Bornstein, P. (2001) J. Biol. Chem. 276, 8403-8408[Abstract/Free Full Text]
  52. Emonard, H., Bellon, G., Troeberg, L., Berton, A., Robinet, A., Henriet, P., Marbaix, E., Kirkegaard, K., Patthy, L., Eeckhout, Y., Nagase, H., Hornebeck, W., and Courtoy, P. J. (2004) J. Biol. Chem. 279, 54944-54951[Abstract/Free Full Text]
  53. Barmina, O. Y., Walling, H. W., Fiacco, G. J., Freije, J. M., Lopez-Otin, C., Jeffrey, J. J., and Partridge, N. C. (1999) J. Biol. Chem. 274, 30087-30093[Abstract/Free Full Text]
  54. Monferran, S., Paupert, J., Dauvillier, S., Salles, B., and Muller, C. (2004) EMBO J. 23, 3758-3768[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
I. Bertini, M. Fragai, C. Luchinat, M. Melikian, E. Mylonas, N. Sarti, and D. I. Svergun
Interdomain Flexibility in Full-length Matrix Metalloproteinase-1 (MMP-1)
J. Biol. Chem., May 8, 2009; 284(19): 12821 - 12828.
[Abstract] [Full Text] [PDF]

Home page
 Sci SignalHome page
M. D. Bass, M. R. Morgan, and M. J. Humphries
Syndecans Shed Their Reputation as Inert Molecules
Sci. Signal., March 31, 2009; 2(64): pe18 - pe18.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
H. Zhao, A. Sohail, Q. Sun, Q. Shi, S. Kim, S. Mobashery, and R. Fridman
Identification and Role of the Homodimerization Interface of the Glycosylphosphatidylinositol-anchored Membrane Type 6 Matrix Metalloproteinase (MMP25)
J. Biol. Chem., December 12, 2008; 283(50): 35023 - 35032.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
E. Mantuano, G. Inoue, X. Li, K. Takahashi, A. Gaultier, S. L. Gonias, and W. M. Campana
The Hemopexin Domain of Matrix Metalloproteinase-9 Activates Cell Signaling and Promotes Migration of Schwann Cells by Binding to Low-Density Lipoprotein Receptor-Related Protein
J. Neurosci., November 5, 2008; 28(45): 11571 - 11582.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
J. Redondo-Munoz, E. Ugarte-Berzal, J. A. Garcia-Marco, M. H. del Cerro, P. E. Van den Steen, G. Opdenakker, M. J. Terol, and A. Garcia-Pardo
{alpha}4{beta}1 integrin and 190-kDa CD44v constitute a cell surface docking complex for gelatinase B/MMP-9 in chronic leukemic but not in normal B cells
Blood, July 1, 2008; 112(1): 169 - 178.
[Abstract] [Full Text] [PDF]

Home page
 J. Leukoc. Biol.Home page
H. Piccard, P. E. Van den Steen, and G. Opdenakker
Hemopexin domains as multifunctional liganding modules in matrix metalloproteinases and other proteins
J. Leukoc. Biol., April 1, 2007; 81(4): 870 - 892.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
D. Minond, J. L. Lauer-Fields, M. Cudic, C. M. Overall, D. Pei, K. Brew, R. Visse, H. Nagase, and G. B. Fields
The Roles of Substrate Thermal Stability and P2 and P1' Subsite Identity on Matrix Metalloproteinase Triple-helical Peptidase Activity and Collagen Specificity
J. Biol. Chem., December 15, 2006; 281(50): 38302 - 38313.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/27/18626    most recent
M512308200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Van den Steen, P. E.
Right arrow Articles by Opdenakker, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Van den Steen, P. E.
Right arrow Articles by Opdenakker, G.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2006 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement