intracellular processing and activation of membrane type 1 matrix … · 2004. 11. 23. · *author...

IntroductionDegradation of the extracellular matrix (ECM) is a criticalprocess during cell invasion in both physiological andpathological processes such as morphogenesis, differentiation,cell migration, apoptosis, tumor invasion and neo-angiogenesis(reviewed in Basbaum and Werb, 1996). These events requirethe direct participation of released and exposed proteasesincluding matrix metalloproteases (MMPs); MMPs are afamily of related zinc-dependent proteases whose principalfunction appears to be the breakdown of extracellularmatrix (ECM) proteins during tissue remodeling, growth,development, repair and cell invasion, physiological andpathological alike.

A considerable body of data now indicates that theexpression and regulation of MMP activity is abnormal inmany common forms of human malignancy; in fact, severalMMPs are expressed in cancers at higher than normal levels(Edwards and Murphy, 1998; Egeblad and Werb, 2002) and,based on their characteristics, are believed to mediate theevents leading to tumor cell metastasis. More than 25members of this family have been identified that, as a group,degrade virtually all components of the ECM (Egeblad andWerb, 2002; Seiki, 2003). The MMP family can be broadlydivided into soluble or membrane types. The latter are theminority and are membrane-bound proteins featuring a single

type I transmembrane domain and a short cytoplasmicstretch (MMP14, MMP15, MMP16 and MMP24, or MT1-MMP, MT2-MMP, MT3-MMP and MT5-MMP), aglycosphingolipid anchor (MMP17 and MMP25, or MT4-MMP and MT6-MMP) or, as an exception, a type IItransmembrane domain (MMP23).

MT1-MMP plays a pivotal role in a number of physiologicaland pathological processes via mechanisms that go beyond thedegradation of ECM components (Seiki, 2003; Sounni et al.,2003). MT1-MMP displays proteolytic activity not onlyagainst a number of ECM components (d’Ortho et al., 1997)but also against other cell surface proteins (Belkin et al., 2001;Deryugina et al., 2002; Kajita et al., 2001) and has criticalfunctions in skeletal embryogenesis (Holmbeck et al., 1999;Zhou et al., 2000) and cell migration (Hotary et al., 2000;Kajita et al., 2001). In some cases MT1-MMP might functionirrespective of its proteolytic activity and might be involved insignal transduction events (Mu et al., 2002) triggered by cellinteraction or other factors such as ligand binding.

A recent exciting development has been the finding thatMT1-MMP confers tumor cells with an evident three-dimensional growth advantage in vitro and in vivo (Hotary etal., 2003). This requires pericellular proteolysis of the ECM.These observations emphasize the unique role of MT1-MMPin metastatis and reveal its function as a tumor-derived growth

6275

The integral membrane type 1 matrix metalloprotease(MT1-MMP) is a pivotal protease in a number ofphysiological and pathological processes and confers bothnon-tumorigenic and tumorigenic cell lines with a specificgrowth advantage in a three-dimensional matrix. Here weshow that, in a melanoma cell line, the majority (80%) ofMT1-MMP is sorted to detergent-resistant membranefractions; however, it is only the detergent-soluble fraction(20%) of MT1-MMP that undergoes intracellularprocessing to the mature form. Also, this processed MT1-MMP is the sole form responsible for ECM degradation invitro. Finally, furin-dependent processing of MT1-MMP isshown to occur intracellularly after exit from the Golgi

apparatus and prior to its arrival at the plasma membrane.It is thus proposed that the association of MT1-MMP withdifferent membrane subdomains might be crucial in thecontrol of its different activities: for instance in cellmigration and invasion and other less defined ones such asMT1-MMP-dependent signaling pathways.

Supplementary material available online athttp://jcs.biologists.org/cgi/content/full/117/26/6275/DC1

Key words: Matrix metalloproteases, Intracellular trafficking,Extracellular matrix, Cell invasion

Summary

Intracellular processing and activation of membranetype 1 matrix metalloprotease depends on itspartitioning into lipid domainsMarco Mazzone1, Massimiliano Baldassarre1, Galina Beznoussenko1, Giada Giacchetti1, Jian Cao2,Stanley Zucker2,3, Alberto Luini1 and Roberto Buccione1,*1Department of Cell Biology and Oncology, Consorzio Mario Negri Sud, S. Maria Imbaro, 66030, Chieti, Italy2Department of Medicine, State University of New York at Stony Brook, Stony Brook, NY 11794, USA3Department of Veterans Affairs Medical Center, Northport, NY 11768, USA*Author for correspondence (e-mail: [email protected])

Accepted 24 September 2004Journal of Cell Science 117, 6275-6287 Published by The Company of Biologists 2004doi:10.1242/jcs.01563

Research Article

6276

factor that regulates proliferation by controlling cell geometrywithin the confines of the three-dimensional ECM.

Similarly to most MMPs, MT1-MMP is synthesized as apro-enzyme. It is generally thought that the propeptide needsto be removed for the protein to function properly and that thisoccurs through cleavage at a short amino-acid stretch at the C-terminus of the propeptide by the action of a prohormoneconvertase [furin or similar (Sato et al., 1996; Yana and Weiss,2000)], although it has been proposed that MT1-MMP does notrequire the cleavage of its propeptide to activate MMP2 incertain cell types (Cao et al., 1996). In addition, activation ofMT1-MMP could also occur by plasmin at the cell surface(Okumura et al., 1997). The subcellular locations where MT1-MMP is activated by furin are still uncertain; however, it couldoccur intracellularly before exit from the Golgi apparatus in thetrans-Golgi network (TGN), similarly to MT3-MMP (Kang etal., 2002), or at the cell surface (Mayer et al., 2003). A furtherlayer of complexity derives from the fact that mechanisms asdiverse as autocatalytic processing, ectodomain shedding,homodimerization and internalization might all contribute tothe modulation of MT1-MMP activity on the cell surface(reviewed by Osenkowski et al., 2004). Most recently it wasshown that O-glycosylation might also contribute to theregulation of MT1-MMP substrate targeting (Wu et al., 2004).MT1-MMP has also been found to be partially located indetergent-resistant membrane fractions (DRM) (Annabi et al.,2001). Recently, new results have been published defining therole of DRM in MT1-MMP/CD44/caveolin interaction andhyaluronan cell surface binding (Annabi et al., 2004). Hence,MT1-MMP membrane partitioning could represent amechanism for the regulation of its activities. At variance withthese findings, it has been recently proposed that only adeletion mutant of MT1-MMP lacking the cytosolic C-terminaltail is partitioned into DRM (Rozanov et al., 2004).

Here, we investigate MT1-MMP intracellular processingand activation. We find that the association of MT1-MMPwith different membrane subdomains establishes divergentmetabolic fates and might be crucial in the control of itsdifferent activities. In particular, we show that, although themajority (80%) of MT1-MMP is sorted to detergent-resistantmembrane fractions, it is only the minor (20%) detergent-soluble fraction of MT1-MMP, that undergoes intracellularprocessing to the mature form. Also, this processed MT1-MMPis exclusively responsible for ECM degradation in vitro.Finally, we show that furin-dependent processing of MT1-MMP occurs intracellularly after exit from the Golgi apparatusprior to arrival at the plasma membrane and provide animmunoelectron microscopy analysis of the potential MT1-MMP processing compartment.

Materials and MethodsReagentsAll chemical reagents, unless otherwise stated, were purchased fromSigma Chemical Co. (St Louis, MO).

Cell culture and transfectionA375MM cells (hereafter referred to as A375 cells), an establishedhuman melanoma cell line (Kozlowski et al., 1984), were maintainedin DMEM/F12 (1:1) nutrient mixture (Life Technologies, Rockville,MD). Medium was supplemented with 10% fetal calf serum, 4 mM

L-glutamine, 100 U/ml penicillin and 100 µg/ml streptomycin. Fortransfection, cells were plated at 50% confluence, and the next daywere incubated for 1 hour at 37°C with 0.45 µg/cm2 DNA of interestand 2 µl/µg DNA TransFast (Promega, Madison, WI). Completemedium was then added. Experiments involving overexpression ofexogenous proteins were usually performed the day after transfection.

Constructs and antibodiesMT1-MMP cDNA encoding an open reading frame from amino acidresidues Met1-Val583 and GFP-tagged-MT1-MMP were cloned in apcDNA3.0 expression vector (Life Technologies) (Cao et al., 1996).FLAG-tagged α1-PDX cDNA (see Jean et al., 1998) was kindlyprovided by G. Thomas (Oregon Health Sciences University, Portland,OR) and subcloned in pcDNA3.0. GFP-tagged furin cDNA waskindly provided by J. Bonifacino (NIH, Bethesda, MD). Thedominant-negative dynamin 2 mutant Dyn2K44A and the ts045temperature-sensitive Vesicular Stomatitis Virus G glycoproteinmutant (VSV-G)-GFP chimera were generous gifts of Drs M.McNiven (Mayo Clinic, Rochester, MN) and J. Lippincott-Schwartz(NIH, Bethesda, MD), respectively.

In all experiments and unless indicated otherwise, MT1-MMPwas immunodetected or immunoprecipitated with a rabbitpolyclonal antiserum (MMR2) raised against a GST-fusedpolypeptide corresponding to amino acids 113-538 of MT1-MMPand produced with the help of G. Di Tullio (Consorzio Mario NegriSud, S. Maria Imbaro, Italy); where indicated, the commercial rabbitpolyclonal immunopurified antibody M3927 (Sigma) raised againsta synthetic peptide corresponding to the hinge domain of MT1-MMP was used. α1-PDX was immunodetected with a monoclonalanti-FLAG M2 antibody (Sigma). GFP-tagged proteins (inimmunoelectron microscopy experiments) were detected with arabbit polyclonal antibody to GFP (Abcam, Cambridge, UK).Caveolin-1 was detected with an anti-caveolin-1 antibody (SantaCruz Biotechnology, Santa Cruz, CA). In immunofluorescencecolocalizations studies anti-mannose 6-phosphate receptor (AffinityBioreagents, Golden, CO) and anti-transferrin receptor (Zymed, SanFrancisco, CA) antibodies were used. Polyclonal sheep anti-TGN46antibodies were a generous gift from S. Ponnambalan (University ofDundee, Dundee, UK). A rabbit polyclonal against the luminaldomain of VSV-G was kindly provided by M. A. De Matteis(Consorzio Mario Negri Sud, S. Maria Imbaro, Italy). In immuno-electron microscopy studies, furin was detected with a polyclonalantibody (Affinity Bioreagents).

Optiprep density gradientsOptiprep gradient analysis of Triton X-100-insoluble material wasperformed using previously published protocols with somemodifications (Arreaza et al., 1994). Briefly, cells were grown toconfluence in 100 mm dishes and labeled for 2 hours with 0.1 mCi/ml[35S]methionine (Amersham, Cologno Monzese, Italy). Cells werethen washed in PBS and lysed with 1% Triton X-100 in TNE buffer(50 mM Tris-HCl, 150 mM NaCl, 2 mM EDTA, pH 8.0) for 10minutes on ice. Lysates were scraped from dishes, adjusted withOptiprep to a final 35%, and then placed at the bottom of a centrifugetube. A discontinuous Optiprep gradient (5-30% in TNE) was layeredon the top of the lysates and the samples were ultracentrifuged at120,000 g overnight in a swinging-bucket rotor. Fractions of 1 mlwere harvested from the top of the gradient. MT1-MMPimmunoprecipitation was performed with MMR2 antiserum afteradding 1% Triton X-100 to the samples. Immunoprecipitates wererecovered with protein A-Sepharose (Pharmacia, Milan, Italy),washed twice in TNE, resuspended with sample buffer and separatedby SDS-PAGE; proteins were then electro-transferred to nitrocellulosefilters and detected by autoradiography with the Fujifilm bioimaginganalyzer system (BAS) 1800 II and relative software.

Journal of Cell Science 117 (26)

6277MT1-MMP transport and processing

Pulse-chase analysisMT1-MMP-transfected A375 melanoma cells were labeled fordifferent times, depending on the experiment, with 0.1 mCi/ml[35S]methionine (Amersham) and chased for different times. Cellswere cooled on ice and washed twice with ice-cold PBS (pH 8.0).After extraction with 1% Triton X-100 in TNE, soluble (S) andinsoluble (P) fractions were separated by ultracentrifugation at 4°C(120,000 g, 1 hour). The pellet, containing Triton X-100-insolublematerial, was further incubated with 0.2% SDS, 1% Triton X-100 inTNE at 37°C for 1 hour. MT1-MMP immunoprecipitation wasperformed with MMR2 antiserum. Immunoprecipitates wererecovered with protein A-Sepharose (Pharmacia), washed twice inTNE, resuspended with sample buffer and separated by SDS-PAGE;proteins were then transferred to nitrocellulose filters and detected byautoradiography. Quantitative analysis was performed with the FujiBAS software.

Biotinylation assayMT1-MMP-transfected A375 cells were cooled on ice, washed twicewith ice-cold PBS and incubated twice for 20 minutes each with 1mg/ml NHS-biotin (Pierce, Rockford, IL) in PBS at 4°C. Next, cellswere washed with PBS and excess NHS-biotin was quenched for 20minutes with 50 mM NH4Cl in PBS. After extraction with 1% TritonX-100 in TNE (pH 7.4) at 4°C, soluble and insoluble material wasseparated by ultracentrifugation at 4°C (120,000 g, 1 hour). The pellet,containing Triton X-100-insoluble material, was further incubatedwith 2% SDS, 1% Triton X-100 in TNE (pH 7.4), containing astandard antiprotease mix at 37°C for 1 hour. Biotinylated proteinswere captured by an overnight incubation in the cold room withstreptavidine-agarose beads (Sigma). Supernatants, containing non-biotinylated proteins, were recovered, precipitated in 10% trichloro-acetic acid (TCA) for 30 minutes on ice, washed twice with ice-coldacetone and resuspended in sample buffer. The pellets, whichcontained biotinylated proteins, were washed twice in TNE (pH 7.4),resuspended in sample buffer and subjected to SDS-PAGE. Proteinswere then electro-transferred to nitrocellulose filters and probed withthe immunopurified polyclonal anti-MT1-MMP MMR2 and anti-caveolin-1 polyclonal antibodies to reveal the MT1-MMP andcaveolin distribution profiles.

Gelatin degradation assayFluorescent gelatin coated coverslips were prepared and the assaycarried out as described (Baldassarre et al., 2003; Bowden et al.,2001). Cells were cultured on gelatin-coated coverslips for 16 hours.Most experiments were analyzed with a Zeiss LSM 510 laser scanningconfocal microscope (Esslingen, Germany). Immunofluorescenceimages were acquired at high confocality (pinhole=1 Airy unit) toachieve the thinnest possible optical slices at the substrate-cellinterface. To determine the number of degrading cells for eachexperiment we considered 100 random fields (containing at least 50transfected cells) at a 63� magnification as previously described(Baldassarre et al., 2003). In some experiments involving multi-colorlabeling, fluorescence was acquired by wide-field fluorescencemicroscopy in each channel along the z-axis. To remove the blurringcaused by the objective lens and reduce the background caused byfluorescence from out-of-focus regions, images were subjected tomathematical deconvolution with an acquired point spread function(Delta Vision, Issaquah, WA). Images were given a threshold toremove noise and enhance the contrast of structures of interest, andwere superimposed and saved.

ImmunofluorescenceAfter treatments, cells were fixed in 4% paraformaldehyde for 15minutes, permeabilized in PBS containing 0.02% saponin, 0.2% BSA

and 50 mM NH4Cl, incubated with the primary antibodies of interestfor 1 hour and then incubated with fluorophore-conjugated secondaryantibodies (Molecular Probes Europe BV, Leiden, The Netherlands)for 45 minutes. Finally, coverslips were mounted in the SlowFade(Molecular Probes) antifade reagent.

VSV-G transport assayA375 melanoma cells were plated on 12 mm glass coverslips,transfected with VSV-G-GFP and incubated overnight at 40°C toallow for expression while blocking exit from the endoplasmicreticulum. The following day, cells were washed three times withcomplete medium containing 20 mM Hepes and incubated at 20°Cfor 30 minutes in the same medium and chased at 32°C in the presenceor absence of 0.5% tannic acid. Cells were then fixed and processedfor immunofluorescence as described above, with the difference thatcells were not permeabilized prior to incubation with antibodies. Acombination of an antibody directed against the lumenal/extracellulardomain of VSV-G and a fluorophore-conjugated secondary antibodywas used to detect plasma membrane staining and compared with totalVSV-G measured as GFP fluorescence. For each treatment at least 20cells were quantified. The experiment was repeated three times.

Ultrathin cryosectioning and immunogold labelingMT1-MMP-transfected cells were fixed with 2% formaldehydeand 0.2% glutaraldehyde in PHEM buffer (60 mM PIPES, 25 mMHEPES, 2 mM MgCl2, 10 mM EGTA, pH 7.4) washed in the samebuffer and collected by centrifugation. Pellets were embedded in 10%gelatin, cooled in ice, and cut into 1 mm3 blocks in the cold room.The blocks were infused with 2.3 M sucrose at 4°C for at least 2 hours,and frozen in liquid nitrogen. 50-60-nm-thick sections were cut at–120°C using an Ultracut R/FCS (Leica, Milan, Italy) equipped withan antistatic device (Diatome, Fort Washington, PA) and a diamondknife. Ultrathin sections were picked up in a mix of 1.8%methylcellulose and 2.3 M sucrose (1:1) as described (Liou et al.,1996). Cryosections were collected on formvar/carbon-coated slotcopper grids and then incubated with different combinations of rabbitpolyclonal antibodies followed by protein-A-conjugated goldparticles of different sizes (Slot and Geuze, 1985). Doubleimmunolabeling was performed as described before (Slot and Geuze,1985; Webster, 1999) with optimal combinations of gold particlesizes. To reduce specificity problems with double labeling (when tworabbit secondary antibodies were used), following treatment with thefirst secondary antibody and protein A, sections were additionallyfixed with 1% glutaraldehyde for 5 minutes, washed with 0.12%glycine and incubated in blocking solution containing 0.01% BSA-c(Aurion, Wageningen, The Netherlands), as previously recommended(Webster, 1999). Additionally, all sequences of primary antibodyaddition and combinations of gold particle size were tested.After labeling, sections were treated with 1% glutaraldehyde,counterstained with uranyl acetate, and embedded in methyl celluloseuranyl acetate as described (Slot et al., 1991). Automated dataacquisition was carried out on a Tecnai 20 electron microscope at 200kV (FEI/Philips Electron Optics) equipped with a slow-scan CCDcamera.

Quantitative immunoelectron microscopyTo determine the distribution of MT1-MMP throughout the secretorypathway in transfected A375 cells, quantification was performed onover 30 cross-sections of cells expressing average levels of MT1-MMP. In detail, labeling density of MT1-MMP protein in differentcompartments was calculated as previously described (Mayhew et al.,2003), micrographs were taken at a 46,000� magnification for eachmarker. Fields were taken randomly with the only criterion being awell-preserved morphology. Labeling densities in Table 1 were

6278

expressed as the number of gold particles per intersections ofthe morphometrical grid. Estimation of colocalization on EMcryosections was performed according to (Martinez-Menarguez et al.,2001).

ResultsMT1-MMP transport and membrane partitioningTo analyze MT1-MMP, we produced two new rabbitpolyclonal antisera (MMR1, MMR2) raised against the sameGST-fused polypeptide corresponding to amino acids 113-538of MT1-MMP. These were tested and compared with thecommercially available affinity-purified polyclonal antibodyM3927 (Sigma) by immunoprecipitation from untransfectedand MT1-MMP-transfected A375 melanoma cells (Fig. 1A).The three different antibodies displayed uneven efficiency inthe detection of a lower molecular weight (60 kDa) MT1-MMP

band, presumably the fully processed mature form (see below).This is of great practical importance because (1) it suggeststhat detection of mature MT1-MMP is generally critical, asexemplified by comparing the antisera generated in differentanimals but raised against the same polypeptide (MMR1 andMMR2); and (2) the interpretation of experimental datainvolving the maturation of MT1-MMP obtained with differentantibodies becomes a critical issue. In our experiments, sinceMMR2 appeared to be by far the most effective, it was used inall subsequent experiments

MT1-MMP-transfected A375 cells were pulse-labeled with[35S]methionine for 5 minutes and, after different chase times,processed for immunoprecipitation with the polyclonal anti-MT1-MMP antibody MMR2. Three molecular forms of MT1-MMP could be detected of 65, 63 and 60 kDa (Fig. 1B). Thehigher molecular mass 65 kDa form has been previouslyobserved (Maquoi et al., 1998; Yana and Weiss, 2000) and


Fig. 1. MT1-MMP processing and localizationin low-density Optiprep-gradient fractionsfrom A375 melanoma cells. (A) Untransfected(NT) and MT1-MMP-transfected (MT1-MMP) A375 cells were pulse-labeled with[35S]methionine for 2 hours and their clearedlysates were immunoprecipitated withdifferent antibodies, as indicated. (B) MT1-MMP-transfected A375 cells were pulse-labeled with [35S]methionine for 5 minutesand chased for the indicated times: theseMT1-MMP immunoprecipitates were obtainedwith polyclonal anti-MT1-MMP antiserumMMR2. Where indicated, transfected cellswere preincubated with 5 µM BB94 overnightand throughout the chase. MT1-MMPimmunoprecipitates in A and B weresubjected to SDS-PAGE, transferred tonitrocellulose membrane and visualized byautoradiography. The black arrowheads markthe positions of the MT1-MMP formsdetected: calculated molecular masses are,respectively, 65, 63 and 60 kDa. The whitearrowheads mark an unspecific band visible inall lanes that should not be confused with the43 kDa form shown to be a MT1-MMPdegradation product in some reports.(C) MT1-MMP-transfected A375 cells werelysed with 1% Triton X-100-containing TNEbuffer, as described in Materials and Methods.Insoluble (P) and soluble (S) fractions wereseparated by ultracentrifugation at 4°C(120,000 g, 1 hour), resuspended in samplebuffer and subjected to SDS-PAGE. Proteinswere then transferred to nitrocellulosemembrane and probed with polyclonal anti-MT1-MMP antiserum MMR2. MT1-MMPmolecular forms in each fraction werequantified with the public domain ImageJv.1.3 software and plotted on the reportedchart as a percentage of total MT1-MMP. (D) MT1-MMP-transfected A375 cells were labeled with [35S]methionine for 2 hours and lysed witha 1% Triton X-100-containing buffer, as described in Materials and Methods. Lysates were fractionated on a discontinuous Optiprep-densitygradient, as reported in the scheme on the bottom. MT1-MMP immunoprecipitates from each fraction and a sample of the initial preparation(Inp) were subjected to SDS-PAGE, transferred to nitrocellulose membrane and visualized by autoradiography. The amount of 63 kDa form ineach fraction was determined with Fuji BAS software and plotted on the reported chart. A small sample of each fraction was tested by westernblotting with a polyclonal anti-caveolin-1 antibody as a bona-fide marker for the DRM preparation.


proposed to be a post-translationally modified unprocessedform (Maquoi et al., 1998); we observed that this form isglycopeptidase F-resistant and is also present if cells are treatedwith inhibitors of N- and O-glycosylation (not shown). The 63kDa form (proMT1-MMP) had a slow turnover rate over a timeframe of 2 hours; only a minor fraction was converted to themature lower molecular weight 60 kDa form which insteadappeared to turn over much more rapidly (Fig. 1B). WhenBB94, a broad-range metalloprotease inhibitor (Davies et al.,1993), was included during preincubation and chase, MT1-MMP maturation occurred normally which, at least in ourconditions, contradicts previous results showing that MT1-MMP was functioning as a self-convertase (Rozanov andStrongin, 2003).

Both endogenous and overexpressed MT1-MMP have beenpartially located in detergent-resistant membranes (DRM)(Annabi et al., 2001). Insolubility in detergent is typical ofglycolipid-rich domains, also referred to as ‘rafts’ (Simons andEhehalt, 2002). It was thus possible that two populations ofMT1-MMP could coexist in transfected A375 cells and explainits peculiar processing pattern. We verified the Triton X-100solubility of MT1-MMP in total membrane preparations fromMT1-MMP-transfected A375 melanoma cells and found thatthe majority of MT1-MMP (at least 70-80%) was Triton X-100-insoluble and unprocessed (63 kDa pro-form; Fig. 1C). Toexclude the possibility that insolubility was due to associationwith cytoskeletal elements or to the formation of non-specificprotein aggregates, we further verified the distribution of MT1-MMP on an Optiprep-density gradient as described inMaterials and Methods. We found that the majority of the 63kDa proMT1-MMP floated towards the top of the gradient(Fig. 1D); the assay was validated by testing the distribution ofcaveolin-1, a bona-fide marker for raft-type membranedomains, and found that it co-fractionated with proMT1-MMP(Fig. 1D). Again, only the unprocessed 63 kDa proMT1-MMPwas detected in the DRM-enriched fraction (lane 5 in Fig. 1D).

Divergent processing of MT1-MMPWe thus addressed the possibility that differential MT1-MMPprocessing was occurring in association with its partitioning in

different membrane subdomains. To this end, MT1-MMP-transfected cells were pulse-labeled with [35S]methionine for 5minutes and, after different chase times, lysed with 1% TritonX-100. Soluble (S) and insoluble (P) fractions were separatedby ultracentrifugation and immunoprecipitated with polyclonalantibody MMR2 (Fig. 2). The detergent-soluble 63 kDaproMT1-MMP is rapidly processed to the mature form within20 minutes and, after 60 minutes, is severely reduced (Fig. 2).As a whole, the soluble MT1-MMP pool undergoes a rapidturnover (Fig. 2) and is degraded starting from 2 hours (Fig.1B); DRM-associated proMT1-MMP, instead, is stable andremains unprocessed even after a 2 hour-long chase (Fig. 2).This indicates that only the detergent-soluble pool of MT1-MMP undergoes maturation and rapid degradation; DRM-associated proMT1-MMP is, conversely, completely excludedfrom processing and is quite stable over time.

The detergent-soluble proMT1-MMP fraction alone isactivated and degrades ECM substrates whereas DRM-associated plasma membrane proMT1-MMP does notOverexpression of MT1-MMP in A375 cells led to, asexpected, a massive degradation of cross-linked fluorescentgelatin (Fig. 3A). When monitored in vivo, with a GFP-taggedMT1-MMP chimera (which produces the same phenotypeas wild-type MT1-MMP), this appeared to be an extremelyrapid process (confocal time-lapse series; see Movie 1 insupplementary material) (ecm.mov). The question thus arisesas to which molecular forms and populations of MT1-MMPare responsible for this biological activity. Clearly, in ourexperimental setting, proMT1-MMP is activated by furin. Infact, the serpin-like protein α1-PDX, a bioengineered α1-antitrypsin characterized as an extremely specific endocellularinhibitor of the pro-protein convertases furin and PC6 (Jean etal., 1998) completely abated MT1-MMP-induced (Fig. 3A)gelatin degradation by A375 cells.

A number of experiments were thus performed to understandbetter the processing of MT1-MMP and define the molecularform(s) responsible for the biological activity, a controversialissue in the literature. Since gelatin degradation relies on cellsurface presentation of enzymatically active MT1-MMP, we

verified the presence and membrane partitioningof MT1-MMP at the plasma membrane. To thisend, surface proteins were biotinylated on ice (toblock all membrane trafficking events) and then

Fig. 2. Time-course analysis of MT1-MMPprocessing in A375 melanoma cells and associationwith different membrane fractions. Untransfected(NT) and MT1-MMP-transfected A375 cells werepulse-labeled with [35S]methionine for 5 minutes and,after the indicated chase times, incubated with TNEcontaining 1% Triton X-100 on ice. Soluble (S) andinsoluble (P) fractions were separated byultracentrifugation, immunoprecipitated withpolyclonal anti-MT1-MMP antiserum MMR2,subjected to SDS-PAGE, transferred to nitrocellulosemembrane and visualized by autoradiography. Thechart shows the distribution in time of immature andmature MT1-MMP and association with differentmembrane fractions as determined with the Fuji BASsoftware.

6280

probed by western blotting analysis of the different membranefractions (detergent-resistant vs detergent-soluble) and cellularlocations (intracellular vs surface). We observed that, at steadystate, the vast majority of MT1-MMP was intracellular,immature and associated with DRM (Fig. 3B). Although totalplasma membrane MT1-MMP levels were relatively low, also

considering that surface biotinylation is not very efficient,DRM-associated proMT1-MMP remained the dominantfraction over non-DRM proMT1-MMP (Fig. 3B). Thisevidence also implies that mature 60 kDa MT1-MMP, shownhere to be associated with the detergent-soluble membranes, israpidly removed from the cell surface after presentation. This

is in good agreement with the rapid turnover ofMT1-MMP as verified in our pulse-chase analysis(Fig. 1A). We then proceeded to analyze the effectof furin inhibition on MT1-MMP partitioning atthe plasma membrane: although co-expressionwith α1-PDX totally inhibited the proteolyticprocessing of detergent-soluble MT1-MMP,proMT1-MMP was still present at the plasmamembrane (Fig. 3B). Interestingly, in theseconditions an increase of the 65 kDa unprocessedform of MT1-MMP was observed. As notedabove, this has been previously reported (Maquoiet al., 1998; Yana and Weiss, 2000) and alsooccurs when treatments affecting transport orprocessing are applied such as Brefeldin A (Fig.S2 in supplementary material).

In conclusion, these results clearly indicatethat: (1) only non-detergent-soluble proMT1-MMP is processed to the mature 60 kDa form;(2) this fully processed MT1-MMP alone isresponsible for gelatinolytic activity; (3) theprocessing enzyme is most likely furin or a furin-like convertase as previously suggested; and (4)removal of the pro-domain is not a prerequisitefor arrival of MT1-MMP at the plasmamembrane.

MT1-MMP processing takes place in anintracellular, post-Golgi compartmentThe MT1-MMP processing compartment stillneeds to be defined. To address this issue, we firstanalyzed the subcellular distribution of MT1-MMP at steady state in A375 melanoma cells.MT1-MMP presented a perinuclear patchydistribution with scattered cytoplasmic elementsand plasma membrane staining (Fig. 3A and Fig.4A,D,G,J). Next we compared the localization ofMT1-MMP to markers of the secretory pathway.Endoplasmic reticulum and Golgi apparatusmarkers showed no significant colocalizationwith MT1-MMP (not shown). When we tested ananti-TGN46 antibody in MT1-MMP-transfectedA375 cells, a patchy asymmetrical perinucleardistribution was observed, with an apparentlyhigh, but not complete, colocalization with MT1-MMP (Fig. 4A-C); TGN46 is a protein knownto cycle between the trans-Golgi network,the plasma membrane and the endosomalcompartment and, although generally consideredto be a trans-Golgi network marker, its actualdistribution depends on the cell type. By contrast,the distribution of transferrin receptor andmannose 6-phosphate receptor (bona fide earlyand late endosome markers, respectively) only


Fig. 3. α1-PDX blocks gelatin degradation and proMT1-MMP processing in MT1-MMP-transfected A375 melanoma cells. (A) A375 cells transfected with MT1-MMP and pcDNA 3.0 (MT1) or MT1-MMP and FLAG-tagged α1-PDX (MT1/α1-PDX) were cultured on surfaces coated with rhodamine-conjugated gelatin. After16 hours of incubation, cells were fixed and labeled with specific primaryantibodies: polyclonal immunopurified anti-MT1-MMP antibody MMR2 (MT1-MMP) and monoclonal anti-FLAG antibody M2 (α1-PDX). Degradation of theunderlying fluorescent gelatin by the MT1-MMP/pcDNA3.0 and MT1-MMP/α1-PDX cotransfectants is shown (Gelatin). Merged staining (MT1-MMP/Gelatin) forthe MT1-MMP/pcDNA3.0 cotransfectants is also shown. Transfected cells werevisualized by wide-field fluorescence microscopy and acquired images weredeconvoluted as described in Materials and Methods. Bars, 20 µm. (B) Steady-state distribution profile of MT1-MMP in A375 melanoma cells: A375 cellstransfected with MT1-MMP and pcDNA3.0 or MT1-MMP and α1-PDX werebiotinylated and incubated with TNE containing 1% Triton X-100 on ice. Soluble(S) and insoluble (P) fractions were separated by ultracentrifugation. Intracellularfractions (Int) were separated from plasma membrane fractions (PM) by capturingthe biotinylated proteins with streptavidin-conjugated agarose beads. Each fractionwas subjected to SDS-PAGE and transferred on nitrocellulose membrane. TheMT1-MMP distribution profile was analyzed with immunopurified polyclonal anti-MT1-MMP antibody MMR2. The white arrowheads mark the positions of the pro-and mature (63 and 60 kDa) forms in the MT1-MMP/pcDNA cotransfectants, theblack arrowheads indicate the 65 and 63 kDa immature forms in the MT1-MMP/α1-PDX cotransfectants. Staining for caveolin-1 in the same experiment isreported as a bona-fide marker for the DRM preparation.


partially overlapped with MT1-MMP (Fig. 4D-F and G-I,respectively). In which of these compartments, therefore, doesfurin-dependent activation actually occur? Our pulse-chaseanalysis showed that fully processed MT1-MMP appeared onlyafter at least 20 minutes, which suggests that it might occur ina late station along the biosynthetic-secretory pathway. Inaddition, a furin-dependent proteolytic activation is notexpected to occur before the trans-Golgi network (Molloy etal., 1999). Nevertheless, to formally rule out the possibility thatMT1-MMP activation occurs in an early station of thesecretory pathway, we took advantage of the fungal toxinBrefeldin A. This treatment leads to intermixing between theendoplasmic reticulum and the Golgi apparatus and blocks exitfrom the Golgi apparatus itself (Fujiwara et al., 1988;Lippincott-Schwartz et al., 1989). Proteins normally present inthe endoplasmic reticulum and the Golgi apparatus would stillbe modified by processing enzymes contained in thosecompartments. Despite this, there is no proteolytic processingof proMT1-MMP; moreover, its cell surface presentation isstrongly reduced as expected (see Fig. S2 in supplementarymaterial).

Next, we verified the extent of furin/MT1-MMPcolocalization. Furin is, depending on the cell type, typicallydistributed throughout the various proprotein processingcompartments including the trans-Golgi network, cell surfaceand endosomes (Mayer et al., 2004; Molloy et al., 1999). InA375 cells cotransfected with MT1-MMP and a furin-GFPchimera, there is an excellent correlation between the twoproteins (Fig. 4J-L). Considering this evidence, we furthersearched for the MT1-MMP activation compartment. The firstpossible station considered for MT1-MMP processing was thusthe trans-Golgi network, a complex structure located at thedistal-most region of the Golgi apparatus that plays a centralrole in the sorting, processing and targeting of proteins of thesecretory pathway en route to their final destination. WhenMT1-MMP was accumulated in the trans-Golgi networkthrough the application of a 20°C temperature block (Griffithset al., 1985), it remained unprocessed (Fig. 5A). Notably,previous reports have shown that, in these conditions, theenzyme activity of furin is not blocked (Song and Fricker,1995). Within 30 minutes after temperature shift to 37°C(release from the block), intracellular trafficking as well asMT1-MMP processing resumed (Fig. 5A). Hence, our resultsare consistent with a MT1-MMP processing event occurringafter exit from the trans-Golgi network. Considering all theabove and that proteins blocked in the trans-Golgi network at20°C are known to exit rapidly (Polishchuk et al., 2000), wesuggest that the processing of MT1-MMP takes place beyondthe trans-Golgi network during transport to the plasmamembrane.

We thus determined whether processing of MT1-MMP takesplace prior to or after delivery to the plasma membrane. Tannicacid has been used as an inactivator of fusion events at theplasma membrane (Newman et al., 1996; Polishchuk et al.,2004). We thus proceeded to use this experimental approach toprevent delivery of MT1-MMP to the cell surface after a 20°Ctemperature block. In the presence of tannic acid, proMT1-MMP processing still occurred, albeit not completely (Fig.5A), even though some toxic effects of the treatment can beinferred from the increased degradation of mature MT1-MMP.This is taken to indicate that MT1-MMP processing can occur

intracellularly in a post-trans-Golgi network compartment andprior to arrival at the plasma membrane. To confirm that tannicacid indeed functioned to block fusion at the plasma membranein our experimental system, we verified the delivery of a ts045temperature sensitive mutant of vesicular stomatitus virus Gprotein (VSV-G)-GFP chimera to the plasma membrane. Thisprotein is blocked in the endoplasmic reticulum while cells aremaintained at 40°C, is regularly transported through the Golgiapparatus up to the TGN (where it remains) at 20°C and isfinally transported to plasma membrane at the permissivetemperature of 32°C. As shown in Fig. 5B, control cells featurea clearly defined plasma membrane staining 60 minutes afterrelease from the 20°C block; by contrast, cells treated with0.5% tannic acid are unable to deliver VSV-G-GFP to thecell surface and present an accumulation of transportintermediates just below the plasma membrane. Thus, in A375cells treated with tannic acid, fusion of transport intermediateswith the plasma membrane is, in fact, completely impaired(Fig. 5B, graph). Complementary evidence was obtainedby coexpression of MT1-MMP with Dyn2K44A, a dominant

Fig. 4. Subcellular distribution of MT1-MMP in A375 melanomacells by immunofluorescence. A375 cells transfected with MT1-MMP and pcDNA3.0 (A-I) or MT1-MMP and furin-GFP (J-L) wereplated on coverlips for 16 hours, then fixed and analyzed at theconfocal microscope. Cells were double-labeled withimmunopurified polyclonal anti-MT1-MMP antibody MMR2(A,D,G,J) and with antibodies directed against TGN46 (B),transferrin receptor (E) and mannose 6-phosphate receptor (H).Merged staining is also shown (C,F,I,L). Bars, 10 µm.

6282

negative mutant of the GTPase dynamin 2 and a potentinhibitor of both caveolar and clathrin-dependent endocytosis(Cao et al., 2000; Pelkmans and Helenius, 2002); in suchconditions maturation of proMT1-MMP was not affected (notshown), which makes it unlikely that MT1-MMP is processedfollowing cell surface presentation, for example after anendocytic step.

The extracellular face of the plasma membrane is thought tobe the location where MT1-MMP processing by furin mightoccur (Mayer et al., 2003). Since our results are at variancewith this notion, we further tested this possibility. Furin is acalcium-dependent enzyme (Molloy et al., 1992) and as suchit would be completely inhibited in the absence of free (i.e.unchelated) Ca2+. Fig. 5C shows that MT1-MMP processingcan still occur when extracellular Ca2+ is chelated with asupramaximal concentration of EGTA (5 mM). Maturationdoes not appear to be complete (Fig. 5C, graph), however,possibly due to a plasma membrane component of processing(Mayer et al., 2003) or leakage of EGTA into the processingcompartment via fluid-phase endocytosis.

Ultrastructural features of a MT1-MMP/furin-positivepost-trans-Golgi network compartmentTo gain a better understanding of the processing compartment,MT1-MMP-transfected A375 cells were processed forcryoimmuno-electron microscopy and immunolabeled with animmunopurified anti-MT1-MMP antibody (MMR2). We thusdetermined the intracellular distribution of labeling densitiesfor MT1-MMP throughout the various compartments of thesecretory pathway identified on the basis of their morphology.

The distribution of MT1-MMP was not homogeneous andpresented a clear, high peak of 76.6% (Table 1) in tubulo-saccular structures located between the trans-Golgi networkand the plasma membrane (Fig. 6A-C). Notably, MT1-MMPexpression levels had no effect on the morphology of thesecretory structures (not shown). This compartment was thusdefined as a whole as a post-trans-Golgi network compartmenton the basis of our biochemical experiments and on themorphological observation that, at the ultrastructural level, itwas not specifically labeled by the TGN marker TGN46 atsteady state (see Fig. S1A,B in supplementary material). Thiscompartment might also contain endosomal elements assuggested by our immunofluorescence studies, consistent withrecent reports suggesting a recycling loop for MT1-MMP(Remacle et al., 2003; Wang et al., 2004).

When the labeling density of MT1-MMP in differentcompartments was compared with that of endogenous furin, itwas established that 64.6% of the MT1-MMP-positivestructures contained furin. Furthermore, 84.8% of the furin-positive structures contained MT1-MMP. This localizationcoincided with the post-trans-Golgi network compartment,where MT1-MMP was enriched, and which represented theforemost reservoir for endogenous furin in A375 cells (Fig.6A-C). Interestingly, the post-TGN compartment where furinand MT1-MMP colocalize was not significantly affected byour 20°C block protocol (see Fig. S1C in supplementarymaterial).

A previous report hypothesized that MT1-MMP is localizedto caveolar DRM (Annabi et al., 2001); it has also beensuggested that MT1-MMP is internalized via caveolae inaddition to the clathrin-mediated pathway (Remacle et al.,


Fig. 5. MT1-MMP processing takes place in anintracellular, post-Golgi compartment. (A) MT1-MMP-transfected A375 cells were pulse-labeled with[35S]methionine for 5 minutes. Labeled proteins wereaccumulated in the trans-Golgi network by incubatingthe cells at 20°C for 30 minutes, then released at 37°Cfor different lengths of time, as indicated. Triton X-100-soluble (S) and insoluble (P) fractions were separated byultracentrifugation, immunoprecipitated and analyzed bySDS-PAGE/autoradiography. Tannic acid 0.5% wasadded during the 20°C block, 10 minutes prior to releaseat 37°C for 60 minutes. (B) Tannic acid completelyblocks arrival of proteins to plasma membrane. A375cells were transfected with the ts045 temperaturesensitive mutant of VSV-G-GFP chimera, incubated at40°C overnight, blocked at 20°C for 30 minutes thenchased at 32°C for the indicated time in the absence orpresence of 0.5% tannic acid. Panels illustraterepresentative control (top) and treated (bottom) cells.VSV-G staining on PM detected from non permeabilizedcells with an antibody raised against its extracellulardomain is shown in red. Bar, 10 µm. The graph showsratio of VSV-G on plasma membrane to total VSV-G atindicated chase time. (C) EGTA treatment: afterincubating the cells at 20°C for 30 minutes, the MT1-MMP-transfected pulsed A375 cells were treated (lane3) or not (lane 2) with 5 mM EGTA for 30 minutes at37°C. Only the soluble fraction is shown. MT1-MMPmolecular forms in lanes 2 and 3 were quantified withthe public domain ImageJ v.1.3 software and plotted onthe reported chart as a ratio of MT1-MMP processing.


2003). We thus examined whether the protein caveolin-1, atypical component of caveolae, colocalized with MT1-MMP atthe plasma membrane. Fig. 6D clearly shows that MT1-MMPat the plasma membrane is not preferentially localized incaveolin-1-enriched domains. This is not in contrast with thebiochemical data in Fig. 1D; in fact, although both MT1-MMPand caveolin-1 partition into DRM, caveolae represent only asubset of the total DRM fraction (Schnitzer et al., 1995).

Furin is excluded from DRMOur data indicate that furin, the MT1-MMP-activating enzyme,is unable to process DRM-associated MT1-MMP which, as aresult, reaches the plasma membrane in its 63 kDa immature

form. A possible explanation for the inability of furin toactivate the large fraction of DRM-associated MT1-MMPcould be that they are physically segregated during transportand thus unable to interact. To test this, we examined themembrane partitioning of a furin-GFP chimera by transient co-transfection with MT1-MMP. As shown (Fig. 7A), despite theincreased cellular burden due to overexpression, furin-GFP, incontrast to MT1-MMP, was exclusively associated with thedetergent-soluble membrane fraction; notably, this partitioningwas not affected by incubation at 20°C (see Fig. S1D insupplementary material). GFP chimeras have been extensivelyused to study membrane partitioning of diverse proteins withno reported alterations in normal profiles (Rodgers, 2002).Furthermore, in our experimental system, the furin-GFP

Fig. 6. Characterization of an MT1-MMP- and furin-positive post-Golgi compartment in A375 melanoma cells by cryo-immuno-electronmicroscopy. MT1-MMP-transfected A375 cells were fixed and prepared for cryo-immunogold labeling as decribed in Materials and Methodsby using differently sized gold particles, as indicated. Subcellular colocalization of MT1-MMP with endogenous furin: (A-C): 10 nm, anti-furin; 15 nm, anti-MT1-MMP (immunopurified antibody MMR2). Arrows mark the positions of specific regions containing both MT1-MMPand furin. Subcellular and plasma membrane colocalization of MT1-MMP with endogenous caveolin (D): 10 nm, anti-caveolin-1; 15 nm, anti-MT1-MMP (immunopurified antibody MMR2). PM indicates plasma membrane rims. C indicates caveosome. Bars, 100 nm (A); 75 nm (B,C);and 150 nm (D).

6284

chimera was enzymatically active, as its expression clearlyincreased the extent of MT1-MMP maturation (Fig. 7B).Taken together, these results further confirm thatpartitioning of MT1-MMP into different membrane sub-domains regulates its processing and biological activities.This also supports the notion that differential partitioninginto DRM, which is also seen in the Alzheimer precursorprotein (Ehehalt et al., 2003), reflects an intrinsic propertyof MT1-MMP.

DiscussionIn the present study we show that in a melanoma cell line,although the majority (80%) of MT1-MMP is sorted toDRM fractions, only the minor (20%) detergent-solublefraction undergoes intracellular processing to the matureform. Also, only this processed MT1-MMP is responsiblefor ECM degradation in vitro. Furthermore, differentexperimental approaches were employed to: (1) providedirect evidence that MT1-MMP processing occursintracellularly in a post-Golgi apparatus compartment, priorto arrival at the plasma membrane; and (2) define, at theultrastructural level, the subcellular compartment where themajority of MT1-MMP colocalizes with furin and whereprocessing presumably occurs.

MT1-MMP is clearly emerging as a multifunctional proteinplaying different physiological and pathological roles (for areview, see Seiki, 2003; Sounni et al., 2003). As a consequence,definition of MT1-MMP processing pathways, transport andthe subcellular site of activation is of central importance.

MT1-MMP partitioning and processing in A375melanoma cellsIn MT1-MMP-transfected A375 melanoma cells, the majorityof MT1-MMP is associated with DRM in accordance withprevious reports (Annabi et al., 2001; Annabi et al., 2004). Bycontrast, it has recently been reported that a MT1-MMPdeletion mutant lacking the cytosolic C-terminal tail, but notthe full-length protein, is stably partitioned into DRM(Rozanov et al., 2004). We extended these findings byanalyzing the relationship between proteolytic processing andmembrane subdomain association. We found that MT1-MMPundergoes proteolytic processing only if associated withdetergent-soluble membranes. The immature 63 kDa formremains stably associated with DRM-enriched membranes andpersists following the processing of the detergent-solubleproMT1-MMP and its degradation. In other words, themetabolic fate and thus the activity of MT1-MMP is directlydependent on its differential association with the twomembrane subdomains. A first consequence is that two cellularand possibly functionally distinct pools of MT1-MMP can thusbe defined: one that is normally and completely processed andthus available for ECM degradation and/or other proteolyticactivities, and a second, whose function remains to beunderstood, that is potentially available at the plasmamembrane for interactions with other molecules. A secondconsequence is that this phenomenon, if undetected (dependingon the experimental procedures used), could have generatedsome of the discrepancies in the interpretation of MT1-MMPprocessing.

Most recently MT1-MMP was found in DRM and thussuggested to be associated with caveolin-1 (Annabi et al.,2001); also, it was recently hypothesized that MT1-MMP isinternalized via caveolae as well as via the clathrin-mediatedpathway (Remacle et al., 2003). In our experiments, we did notfind morphological evidence of MT1-MMP association tocaveolae-like domains. This is not incongruous since caveolaerepresent only a subpopulation of the total DRM-fraction andcan actually be separately isolated (Schnitzer et al., 1995).

Intracellular activation of MT1-MMPPrevious findings (Yana and Weiss, 2000) suggested that MT1-MMP processing is dependent on a furin-like activity, althoughit has been proposed that MT1-MMP does not require thecleavage of its propeptide to activate MMP2 in certain celltypes (Cao et al., 1996). In addition, activation of MT1-MMPcould also occur by plasmin at the cell surface (Okumuraet al., 1997). To verify the activation mechanism in ourexperimental system, we co-transfected MT1-MMP with α1-PDX in A375 cells. In these conditions, MT1-MMP-dependentgelatin degradation was completely inhibited, indicating thatMT1-MMP needs to be processed to express direct or indirectgelatinolytic activity. In addition, since in such conditions themature 60 kDa form is undetectable, alternative pathways ofactivation, such as by extracellular plasmin (Okumura et al.,1997), can be excluded. Finally, BB94, a broad-rangemetalloprotease inhibitor (Davies et al., 1993), did not affectMT1-MMP processing which, at least in our conditions,contradicts previous results showing that MT1-MMP wasfunctioning as a self-convertase (Rozanov and Strongin, 2003).These data enhance the essential role of furin in the regulationof MT1-MMP processing and activity.

Despite the central role played by furin in manypathophysiological processes, the exact subcellular sites ofprocessing and activation of its substrates remain elusive. Thisis also the case for MT1-MMP. Hence, a major concern is to


Fig. 7. Furin is excluded from detergent-resistant membranes.(A) Untransfected (NT) and furin-GFP-transfected (FUR) A375 cellswere labeled with [35S]methionine for 2 hours and lysed with TNEbuffer containing 1% Triton X-100 on ice. Soluble (S) and insoluble(P) fractions were separated by ultracentrifugation, immunoprecipitatedwith a polyclonal anti-GFP antibody and analyzed by SDS-PAGE/autoradiography. (B) A375 cells transfected with MT1-MMP(MT1-MMP) or with MT1-MMP and furin-GFP (MT1+FUR) werelabeled with [35S]methionine for 2 hours and lysed with TNEcontaining 1% Triton X-100 on ice. Soluble fractions were collectedafter ultracentrifugation, immunoprecipitated with polyclonal anti-MT1-MMP antiserum MMR2 and analyzed by SDS-PAGE/autoradiography. The processing ratio of MT1-MMP (MT1-MMP/MT1-MMP+proMT1-MMP) was determined with the Fuji BASsoftware and plotted on the reported chart.


identify the cellular location(s) where endoproteolytic cleavageof MT1-MMP occurs. A number of studies have establishedthat furin is mainly localized in the trans-Golgi network butundergoes extensive cycling between the Golgi complex,endosome compartments and the plasma membrane (reviewedby Molloy et al., 1999). Notably, a high-resolutionimmunogold electron microscopy-based study has recentlyshown that, in addition to the Golgi apparatus, furin was clearlylocalized to endosomes and plasma membranes of a number ofcell types (Mayer et al., 2004). Our results that MT1-MMPprocessing takes place in an intracellular, post-Golgi apparatus,pre-plasma membrane compartment are consistent with thiswidespread distribution of furin.

The extracellular face of the plasma membrane was recentlyproposed to be the MT1-MMP activation compartment basedon experiments showing that, after chemical cross-linking,furin co-immunoprecipitates with proMT1-MMP at the plasmamembrane (Mayer et al., 2003). The same study also showeda colocalization of the two proteins on the plasma membrane.We show that, at least in A375 melanoma cells, furin-dependent processing of MT1-MMP can occur in a post-TGN,pre-plasma membrane compartment.

Exclusion of DRM-associated proMT1-MMP from furin-dependent processingAt steady state, we observed transfected proMT1-MMP to bemainly associated with DRM albeit with a significant fractionover the plasma membrane. This suggests that (1) intracellularremoval of the propeptide domain is not a prerequisite forMT1-MMP cell surface presentation; (2) DRM-associatedproMT1-MMP is not subject to intracellular furin-dependentactivation; and (3) proMT1-MMP is not functioning as agelatinase. A possible mechanicistical explanation is providedby our observation that furin is excluded from detergent-resistant membranes. In fact, physical interaction betweenproteins associated with different membrane subdomains isextremely unlikely, as recently suggested for other proteins(Ehehalt et al., 2003). In our context, colocalization of furinand MT1-MMP in common structures supports the idea thatdetergent-soluble and -resistant MT1-MMP pools mightcoexist but not necessarily interact. This suggests thatpartitioning of MT1-MMP into different lipid subdomains,which is also seen in the Alzheimer precursor protein (Ehehaltet al., 2003), could be a mechanism for the regulation of itsprocessing and activity. Interestingly, it was recently proposedthat the cholesterol content of tumor cells is critical for theregulation of MT1-MMP cell surface presentation and MMP2activation (Atkinson et al., 2004).

ConclusionsIt has clearly emerged that MT1-MMP (and perhaps othermembrane-type MMPs) is essential for the growth of humancancers and acts by disrupting the three-dimensional matrix thatwould otherwise impede cell proliferation (Hotary et al., 2003).Hence, full understanding of MT1-MMP physiology is crucialto develop strategies aimed at preventing escape of cells from aprimary tumor (Yamada, 2003). Here we show that only asubpopulation of MT1-MMP, associated with detergent solublemembranes, are activated and rapidly degraded, perhaps by

targeting to lysosomes (Takino et al., 2003). By contrast, DRM-associated MT1-MMP retains its propeptide and is stablyassociated with the plasma membrane. The functional role ofDRM-associated MT1-MMP is not yet known. A potentialindication stems from the findings that (1) hyaluronic acid cell-surface binding to its receptor CD44 is inhibited by theoverexpression of wild-type MT1-MMP; this inhibiton can beovercome by shifting part of the MT1-MMP pool out of DRMby cholesterol extraction (Annabi et al., 2004); and (2) wild-typeMT1-MMP overexpression triggers ERK activation (Annabi etal., 2004). A possibility that remains to be tested is whether theseevents are specifically triggered by the DRM-associatedproMT1-MMP pool; in fact, DRM domains represent a preferredplatform for a number of signaling receptors and cytoskeletoninteractors (Simons and Toomre, 2000).

In conclusion, our experiments clearly indicate that themetabolic fate of proMT1-MMP is dependent on itspartitioning into different lipid domains. In particular, furin-dependent activation, shown here to occur in a post-Golgi, pre-plasma membrane compartment, occurs exclusively at theexpense of the detergent-soluble fraction. These findingsestablish a novel mode of regulation of MT1-MMP activity andsuggest a potential molecular basis for its diverse functions.

We thank Arsenio Pompeo and Maria Inmaculada Ayala Grande forinsightful discussions. R.B. and A.L. were supported by grants fromthe Italian Association for Cancer Research (AIRC, Milano, Italy) andthe Italian Foundation for Cancer Research (FIRC, Milano, Italy).R.B. was also supported by the European Commission (contractLSHC-CT-2004-503049). J.C. was supported by a ScientistDevelopment Grant from the American Heart Association and a NewInvestigator grant from the US Army Medical Research and MaterielCommand. S.Z. was supported by a Merit Review grant and aResearch Enhancement Award Program grant from the Department ofVeterans Affairs. M.M. was supported by ‘Emilio Mussini’ (2003) andFIRC (2004) Fellowships.

ReferencesAnnabi, B., Lachambre, M., Bousquet-Gagnon, N., Page, M., Gingras, D.

and Beliveau, R. (2001). Localization of membrane-type 1 matrixmetalloproteinase in caveolae membrane domains. Biochem. J. 353, 547-553.

Annabi, B., Thibeault, S., Moumdjian, R. and Beliveau, R. (2004).Hyaluronan cell surface binding is induced by type I collagen and regulatedby caveolae in glioma cells. J. Biol. Chem. 279, 21888-21896.

Arreaza, G., Melkonian, K. A., LaFevre-Bernt, M. and Brown, D. A.(1994). Triton X-100-resistant membrane complexes from cultured kidneyepithelial cells contain the Src family protein tyrosine kinase p62yes. J. Biol.Chem. 269, 19123-19127.

Atkinson, S. J., English, J. L., Holway, N. and Murphy, G. (2004). Cellularcholesterol regulates MT1 MMP dependent activation of MMP 2 via MEK-1 in HT1080 fibrosarcoma cells. FEBS Lett. 566, 65-70.

Baldassarre, M., Pompeo, A., Beznoussenko, G., Castaldi, C., Cortellino,S., McNiven, M. A., Luini, A. and Buccione, R. (2003). Dynaminparticipates in focal extracellular matrix degradation by invasive cells. Mol.Biol. Cell 14, 1074-1084.

Basbaum, C. B. and Werb, Z. (1996). Focalized proteolysis: spatial andtemporal regulation of extracellular matrix degradation at the cell surface.Curr. Opin. Cell Biol. 8, 731-738.

Belkin, A. M., Akimov, S. S., Zaritskaya, L. S., Ratnikov, B. I., Deryugina,E. I. and Strongin, A. Y. (2001). Matrix-dependent proteolysis of surfacetransglutaminase by membrane-type metalloproteinase regulates cancer celladhesion and locomotion. J. Biol. Chem. 276, 18415-18422.

Bowden, E. T., Coopman, P. J. and Mueller, S. C. (2001). Invadopodia:unique methods for measurement of extracellular matrix degradation invitro. Methods Cell Biol. 63, 613-627.

6286

Cao, H., Thompson, H. M., Krueger, E. W. and McNiven, M. A. (2000).Disruption of Golgi structure and function in mammalian cells expressinga mutant dynamin. J. Cell Sci. 113, 1993-2002.

Cao, J., Rehemtulla, A., Bahou, W. and Zucker, S. (1996). Membrane typematrix metalloproteinase 1 activates pro-gelatinase A without furin cleavageof the N-terminal domain. J. Biol. Chem. 271, 30174-30180.

d’Ortho, M. P., Will, H., Atkinson, S., Butler, G., Messent, A., Gavrilovic,J., Smith, B., Timpl, R., Zardi, L. and Murphy, G. (1997). Membrane-type matrix metalloproteinases 1 and 2 exhibit broad-spectrum proteolyticcapacities comparable to many matrix metalloproteinases. Eur. J. Biochem.250, 751-757.

Davies, B., Brown, P. D., East, N., Crimmin, M. J. and Balkwill, F. R.(1993). A synthetic matrix metalloproteinase inhibitor decreases tumorburden and prolongs survival of mice bearing human ovarian carcinomaxenografts. Cancer Res. 53, 2087-2091.

Deryugina, E. I., Ratnikov, B. I., Postnova, T. I., Rozanov, D. V. andStrongin, A. Y. (2002). Processing of integrin alpha(v) subunit bymembrane type 1 matrix metalloproteinase stimulates migration of breastcarcinoma cells on vitronectin and enhances tyrosine phosphorylation offocal adhesion kinase. J. Biol. Chem. 277, 9749-9756.

Edwards, D. R. and Murphy, G. (1998). Cancer. Proteases – invasion andmore. Nature 394, 527-528.

Egeblad, M. and Werb, Z. (2002). New functions for the matrixmetalloproteinases in cancer progression. Nat. Rev. Cancer 2, 161-174.

Ehehalt, R., Keller, P., Haass, C., Thiele, C. and Simons, K. (2003).Amyloidogenic processing of the Alzheimer beta-amyloid precursor proteindepends on lipid rafts. J. Cell Biol. 160, 113-123.

Fujiwara, T., Oda, K., Yokota, S., Takatsuki, A. and Ikehara, Y. (1988).Brefeldin A causes disassembly of the Golgi complex and accumulation ofsecretory proteins in the endoplasmic reticulum. J. Biol. Chem. 263, 18545-18552.

Griffiths, G., Pfeiffer, S., Simons, K. and Matlin, K. (1985). Exit of newlysynthesized membrane proteins from the trans cisterna of the Golgi complexto the plasma membrane. J. Cell Biol. 101, 949-964.

Holmbeck, K., Bianco, P., Caterina, J., Yamada, S., Kromer, M.,Kuznetsov, S. A., Mankani, M., Robey, P. G., Poole, A. R., Pidoux, I. etal. (1999). MT1-MMP-deficient mice develop dwarfism, osteopenia,arthritis, and connective tissue disease due to inadequate collagen turnover.Cell 99, 81-92.

Hotary, K., Allen, E., Punturieri, A., Yana, I. and Weiss, S. J. (2000).Regulation of cell invasion and morphogenesis in a three-dimensional typeI collagen matrix by membrane-type matrix metalloproteinases 1, 2, and 3.J. Cell Biol. 149, 1309-1323.

Hotary, K. B., Allen, E. D., Brooks, P. C., Datta, N. S., Long, M. W. andWeiss, S. J. (2003). Membrane type I matrix metalloproteinase usurps tumorgrowth control imposed by the three-dimensional extracellular matrix. Cell114, 33-45.

Jean, F., Stella, K., Thomas, L., Liu, G., Xiang, Y., Reason, A. J. andThomas, G. (1998). alpha1-Antitrypsin Portland, a bioengineered serpinhighly selective for furin: application as an antipathogenic agent. Proc. Natl.Acad. Sci. USA 95, 7293-7298.

Kajita, M., Itoh, Y., Chiba, T., Mori, H., Okada, A., Kinoh, H. and Seiki,M. (2001). Membrane-type 1 matrix metalloproteinase cleaves CD44 andpromotes cell migration. J. Cell Biol. 153, 893-904.

Kang, T., Nagase, H. and Pei, D. (2002). Activation of membrane-type matrixmetalloproteinase 3 zymogen by the proprotein convertase furin in the trans-Golgi network. Cancer Res. 62, 675-681.

Kozlowski, J. M., Hart, I. R., Fidler, I. J. and Hanna, N. (1984). A humanmelanoma line heterogeneous with respect to metastatic capacity in athymicnude mice. J. Natl. Cancer Inst. 72, 913-917.

Liou, W., Geuze, H. J. and Slot, J. W. (1996). Improving structural integrityof cryosections for immunogold labeling. Histochem. Cell Biol. 106, 41-58.

Lippincott-Schwartz, J., Yuan, L. C., Bonifacino, J. S. and Klausner, R.D. (1989). Rapid redistribution of Golgi proteins into the ER in cells treatedwith brefeldin A: evidence for membrane cycling from Golgi to ER. Cell56, 801-813.

Maquoi, E., Noel, A., Frankenne, F., Angliker, H., Murphy, G. and Foidart,J. M. (1998). Inhibition of matrix metalloproteinase 2 maturation andHT1080 invasiveness by a synthetic furin inhibitor. FEBS Lett. 424, 262-266.

Martinez-Menarguez, J. A., Prekeris, R., Oorschot, V. M., Scheller, R.,Slot, J. W., Geuze, H. J. and Klumperman, J. (2001). Peri-Golgi vesiclescontain retrograde but not anterograde proteins consistent with the cisternalprogression model of intra-Golgi transport. J. Cell Biol. 155, 1213-1224.

Mayer, G., Boileau, G. and Bendayan, M. (2003). Furin interacts withproMT1-MMP and integrin alphaV at specialized domains of renal cellplasma membrane. J. Cell Sci. 116, 1763-1773.

Mayer, G., Boileau, G. and Bendayan, M. (2004). The proprotein convertasefurin colocalizes with caveolin-1 in the Golgi apparatus and endosomes ofhepatocytes. Cell Tissue Res. 316, 55-63.

Mayhew, T., Griffiths, G., Habermann, A., Lucocq, J., Emre, N. andWebster, P. (2003). A simpler way of comparing the labelling densities ofcellular compartments illustrated using data from VPARP and LAMP-1immunogold labelling experiments. Histochem. Cell Biol. 119, 333-341.

Molloy, S. S., Bresnahan, P. A., Leppla, S. H., Klimpel, K. R. and Thomas,G. (1992). Human furin is a calcium-dependent serine endoprotease thatrecognizes the sequence Arg-X-X-Arg and efficiently cleaves anthrax toxinprotective antigen. J. Biol. Chem. 267, 16396-16402.

Molloy, S. S., Anderson, E. D., Jean, F. and Thomas, G. (1999). Bi-cyclingthe furin pathway: from TGN localization to pathogen activation andembryogenesis. Trends Cell Biol. 9, 28-35.

Mu, D., Cambier, S., Fjellbirkeland, L., Baron, J. L., Munger, J. S.,Kawakatsu, H., Sheppard, D., Broaddus, V. C. and Nishimura, S. L.(2002). The integrin alpha(v)beta8 mediates epithelial homeostasis throughMT1-MMP-dependent activation of TGF-beta1. J. Cell Biol. 157, 493-507.

Newman, T. M., Tian, M. and Gomperts, B. D. (1996). Ultrastructuralcharacterization of tannic acid-arrested degranulation of permeabilizedguinea pig eosinophils stimulated with GTP-gamma-S. Eur. J. Cell Biol. 70,209-220.

Okumura, Y., Sato, H., Seiki, M. and Kido, H. (1997). Proteolytic activationof the precursor of membrane type 1 matrix metalloproteinase by humanplasmin. A possible cell surface activator. FEBS Lett. 402, 181-184.

Osenkowski, P., Toth, M. and Fridman, R. (2004). Processing, shedding, andendocytosis of membrane type 1-matrix metalloproteinase (MT1-MMP). J.Cell Physiol. 200, 2-10.

Pelkmans, L. and Helenius, A. (2002). Endocytosis via caveolae. Traffic 3,311-320.

Polishchuk, R. S., Polishchuk, E. V., Marra, P., Alberti, S., Buccione, R.,Luini, A. and Mironov, A. A. (2000). Correlative light-electron microscopyreveals the tubular-saccular ultrastructure of carriers operating betweenGolgi apparatus and plasma membrane. J. Cell Biol. 148, 45-58.

Polishchuk, R., di Pentima, A. and Lippincott-Schwartz, J. (2004).Delivery of raft-associated, GPI-anchored proteins to the apical surface ofpolarized MDCK cells by a transcytotic pathway. Nat. Cell Biol. 6, 297-307.

Remacle, A., Murphy, G. and Roghi, C. (2003). Membrane type I-matrixmetalloproteinase (MT1-MMP) is internalised by two different pathwaysand is recycled to the cell surface. J. Cell Sci. 116, 3905-3916.

Rodgers, W. (2002). Making membranes green: construction andcharacterization of GFP-fusion proteins targeted to discrete plasmamembrane domains. Biotechniques 32, 1044-1046.

Rozanov, D. V. and Strongin, A. Y. (2003). Membrane type-1 matrixmetalloproteinase functions as a proprotein self-convertase. Expression ofthe latent zymogen in Pichia pastoris, autolytic activation, and the peptidesequence of the cleavage forms. J. Biol. Chem. 278, 8257-8260.

Rozanov, D. V., Deryugina, E. I., Monosov, E. Z., Marchenko, N. D. andStrongin, A. Y. (2004). Aberrant, persistent inclusion into lipid rafts limitsthe tumorigenic function of membrane type-1 matrix metalloproteinase inmalignant cells. Exp. Cell Res. 293, 81-95.

Sato, H., Kinoshita, T., Takino, T., Nakayama, K. and Seiki, M. (1996).Activation of a recombinant membrane type 1-matrix metalloproteinase(MT1-MMP) by furin and its interaction with tissue inhibitor ofmetalloproteinases (TIMP)-2. FEBS Lett. 393, 101-104.

Schnitzer, J. E., McIntosh, D. P., Dvorak, A. M., Liu, J. and Oh, P. (1995).Separation of caveolae from associated microdomains of GPI-anchoredproteins. Science 269, 1435-1439.

Seiki, M. (2003). Membrane-type 1 matrix metalloproteinase: a key enzymefor tumor invasion. Cancer Lett. 194, 1-11.

Simons, K. and Ehehalt, R. (2002). Cholesterol, lipid rafts, and disease. J.Clin. Invest. 110, 597-603.

Simons, K. and Toomre, D. (2000). Lipid rafts and signal transduction. Nat.Rev. Mol. Cell Biol. 1, 31-39.

Slot, J. W. and Geuze, H. J. (1985). A new method of preparing gold probesfor multiple-labeling cytochemistry. Eur. J. Cell Biol. 38, 87-93.

Slot, J. W., Geuze, H. J., Gigengack, S., Lienhard, G. E. and James, D. E.(1991). Immuno-localization of the insulin regulatable glucose transporterin brown adipose tissue of the rat. J. Cell Biol. 113, 123-135.

Song, L. and Fricker, L. (1995). Processing of procarboxypeptidase E intocarboxypeptidase E occurs in secretory vesicles. J. Neurochem. 65, 444-453.



Sounni, N. E., Janssen, M., Foidart, J. M. and Noel, A. (2003). Membranetype-1 matrix metalloproteinase and TIMP-2 in tumor angiogenesis. MatrixBiol. 22, 55-61.

Takino, T., Miyamori, H., Kawaguchi, N., Uekita, T., Seiki, M. and Sato,H. (2003). Tetraspanin CD63 promotes targeting and lysosomal proteolysisof membrane-type 1 matrix metalloproteinase. Biochem. Biophys. Res.Commun. 304, 160-166.

Wang, X., Ma, D., Keski-Oja, J. and Pei, D. (2004). Co-recycling of MT1-MMP and MT3-MMP through the trans-Golgi network. Identification ofDKV582 as a recycling signal. J. Biol. Chem. 279, 9331-9336.

Webster, P. (1999). The production of cryosections through fixed andcryoprotected biological material and their use in immunocytochemistry. InMethods in Molecular Biology: Electron Microscopy Methods and

Protocols, Vol. 117 (ed. N. Hajibagheri), pp. 49-75. Totowa, NJ: HumanaPress.

Wu, Y. I., Munshi, H. G., Sen, R., Snipas, S. J., Salvesen, G. S., Fridman,R. and Stack, M. S. (2004). Glycosylation broadens the substrate profile ofmembrane type 1 matrix metalloproteinase. J. Biol. Chem. 279, 8278-8289.

Yamada, K. M. (2003). Cell biology: tumour jailbreak. Nature 424, 889-890.Yana, I. and Weiss, S. J. (2000). Regulation of membrane type-1 matrix

metalloproteinase activation by proprotein convertases. Mol. Biol. Cell 11,2387-2401.

Zhou, Z., Apte, S. S., Soininen, R., Cao, R., Baaklini, G. Y., Rauser, R. W.,Wang, J., Cao, Y. and Tryggvason, K. (2000). Impaired endochondralossification and angiogenesis in mice deficient in membrane-type matrixmetalloproteinase I. Proc. Natl. Acad. Sci. USA 97, 4052-4057.

intracellular processing and activation of membrane type 1 matrix … · 2004. 11. 23. · *author...

Documents