Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Transmembrane-deletion Mutants of the Membrane-type Matrix Metalloproteinase-1 Process Progelatinase A and Express Intrinsic Matrix-degrading Activity

Transmembrane-deletion Mutants of the Membrane-type Matrix Metalloproteinase-1 Process... THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 271, No. 15, Issue of April 12, pp. 9135–9140, 1996 © 1996 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Transmembrane-deletion Mutants of the Membrane-type Matrix Metalloproteinase-1 Process Progelatinase A and Express Intrinsic Matrix-degrading Activity* (Received for publication, January 25, 1996) Duanqing Pei‡ and Stephen J. Weiss§ From the Division of Hematology/Oncology, Department of Internal Medicine, the University of Michigan Comprehensive Cancer Center, University of Michigan, Ann Arbor, Michigan 48109 Membrane-type matrix metalloproteinase-1 (MT- remodeling of the extracellular matrix (ECM) in events ranging MMP-1) has been proposed to play a critical role in from organogenesis to tumor metastasis (1–3). Very recently, regulating the expression of tissue-invasive phenotypes three new members of this family were discovered by screening in normal and neoplastic cells by directly or indirectly cDNA libraries for homologies to conserved regions of the known mediating the activation of progelatinase A. To begin MMP genes and named the membrane-type matrix metallopro- characterizing MT-MMP-1 structure-function relation- teinases-1, -2, and -3 (MT-MMP-1, -2, and -3; Refs. 4–6). Based ships, transmembrane-deletion mutants were con- on their predicted amino acid sequences, each of the MT-MMPs, structed, and the processing of the zymogens as well as like almost all previously characterized MMPs, contains (i) a the enzymic activity of the mature proteinases was an- candidate leader sequence, (ii) a propeptide region which in- alyzed. We now demonstrate that pro-MT-MMP-1 mu- cludes a highly conserved PRCGXPD sequence that helps stabi- tants are efficiently processed to active proteinases fol- lize the MMP zymogen in a catalytically inactive state, (iii) a lowing post-translational endoproteolysis immediately zinc-binding catalytic domain, and (iv) a hemopexin-like domain downstream of an Arg -Arg-Lys-Arg basic motif by a near their respective C termini (4–7). In addition, in a pattern proprotein convertase-dependent pathway. The se- similar to that described for stromelysin-3, each of the MT-MMPs creted form of active MT-MMP-1 not only displays an N contains a short amino acid insert sandwiched between their pro- terminus identical with that described for the processed 112 and catalytic domains that encodes a potential recognition motif wild-type enzyme at Tyr (Strongin, A. Y., Collier, I., for members of the proprotein convertase family (4–8). Despite Bannikov, G., Marmer, B. L., Grants, G. A., and Goldberg, their considerable similarity to other MMP family members, G. I. (1995) J. Biol. Chem. 270, 5331–5338), but also di- however, only the MT-MMPs contain ;75–100 amino acid exten- rectly mediated progelatinase A activation via a two- sions at their C termini, each of which includes a hydrophobic step proteolytic cascade indistinguishable from that ob- stretch consistent with the presence of a transmembrane (TM) served with intact cells. Furthermore, although the only function previously ascribed to MT-MMP-1 is its ability domain (4–6, 9). Thus, in contradistinction to all other MMPs, to act as a progelatinase A activator, purified transmem- the MT-MMPs are expressed as membrane-associated ectoen- brane deletion mutants also expressed proteolytic activ- zymes rather than soluble proteins. ities against a wide range of extracellular matrix mole- Although little is known with regard to the potential functions cules. Given recent reports that MT-MMP-1 ectodomains of the MT-MMPs, most attention has focused on the ability of may undergo intercellular transfer in vivo (Okada, A., MT-MMP-1 as well as MT-MMP-3 to induce the processing of the Bellocq, J.-P., Rouyer, N., Chenard, M.-P., Rio, M.-C., MMP zymogen, progelatinase A, to its activated form (i.e. Chambon, P., and Basset, P. (1995) Proc. Natl. Acad. Sci. 81 38 [Tyr ]gelatinase A) via a [Leu ]gelatinase A intermediate (4, 6, U. S. A. 92, 2730–2734), our data suggest that soluble 10). Given the ability of activated gelatinase A to cleave a wide forms of the proteinase confer recipient cells with the range of ECM substrates (including native types I, IV, V, VII, and ability to not only process progelatinase A, but also di- XI collagen, denatured collagens, elastin, proteoglycans, laminin, rectly degrade extracellular matrix components. and fibronectin) as well as the association of gelatinase A activa- tion with the expression of tissue-invasive phenotypes (1–3, 11– 15), MT-MMPs have been dubbed as possible “master switches” Members of the matrix metalloproteinase (MMP) gene family that control ECM remodeling (16). Nonetheless, the processes have been implicated in the physiologic as well as pathologic that regulate the activation of the MT-MMP zymogens them- selves to mature forms remain undefined as does the mechanism by which MT-MMPs mediate progelatinase A activation (4–6, 9, * This study was supported in part by Grant AI23876 from the 10). In large part, further progress in characterizing MT-MMP National Institutes of Health, the Susan G. Komen Breast Cancer activities has been hindered by the technical problems associated Foundation, and United States Army Medical Research Command Grant DAMD17-94-J-4322. The costs of publication of this article were with isolating and purifying membrane-associated molecules. defrayed in part by the payment of page charges. This article must Given that similar difficulties with other transmembrane en- therefore be hereby marked “advertisement” in accordance with 18 zymes have been negotiated by generating TM-deleted soluble U.S.C. Section 1734 solely to indicate this fact. mutants (17–19), we noted that, with the exception of the ex- ‡ Supported in part by an Oncology research training grant from NHLBI, National Institutes of Health. tended C-terminal domain, the modular organization of the MT- § To whom correspondence should be addressed: Division of Hema- MMPs is identical with that of the secreted MMPs (7). Hence, two tology/Oncology, 5220 MSRB III, 1150 W. Medical Center Dr., Ann Arbor, Michigan 48109-0640. Tel.: 313-764-0030; Fax: 313-764-0101. The abbreviations used are: MMP, matrix metalloproteinase; a PI, 1 , MT-MMP-1 ; TM, transmembrane; MDCK, Madin-Darby 1 DB 1–508 a -proteinase inhibitor; a PI , Pittsburgh mutant of a -proteinase canine kidney cells; PAGE, polyacrylamide gel electrophoresis; PCR, 1 1 Pitt 1 inhibitor; ECM, extracellular matrix; MT-MMP-1, -2, -3, membrane- polymerase chain reaction. type matrix metalloproteinase-1, -2, and -3; DMT-MMP-1, transmem- All MT-MMPs also contain a homologous 8-amino acid insert within brane-deleted MT-MMP-1; MT-MMP-1 , MT-MMP-1 ; MT-MMP- their catalytic domains whose function remains undefined (4–6). DA 1–538 This is an Open Access article under the CC BY license. 9136 Transmembrane-deletion Mutants of MT-MMP-1 TM-deletion mutants of MT-MMP-1 were constructed by either truncating the molecule (i) immediately upstream of the start site of the TM domain (i.e. MT-MMP-1 ; herein referred to as 1–538 MT-MMP-1 ) or (ii) at the conserved cysteinyl residue found at, DA or near, the terminus of all hemopexin domain-containing MMPs (i.e. MT-MMP-1 or MT-MMP-1 ). Utilizing these con- 1–508 DB structs, we now demonstrate that TM-deleted MT-MMP-1 (DMT- MMP-1) mutants undergo efficient post-translational endoprote- 111 112 olysis between Arg -Tyr by a proprotein convertase- dependent pathway to generate fully active proteinases. Furthermore, the purified DMT-MMP-1 mutants not only acti- vate recombinant progelatinase A directly via a two-step activa- tion cascade identical with that described for the membrane- associated enzyme, but they also express heretofore unsuspected ECM-degrading activities. EXPERIMENTAL PROCEDURES FIG.1. Expression and characterization of DMT-MMP-1. A, do- Cell Culture—COS-7 cells and MDCK cells (both obtained from main alignments of MT-MMP-1 and the DMT-MMP-1 mutants. Wild- ATCC) were maintained in Dulbecco’s modified Eagle’s medium (Life type MT-MMP-1 contains 582 amino acids arranged as a series of pre- Technologies, Inc.) containing 10% fetal bovine serum (HyClone) and 4 (shaded box), pro-, catalytic, hemopexin (bounded by a pair of highly mML-glutamine, 100 units/ml penicillin, and 100 mg/ml streptomycin as conserved cysteinyl residues indicated as C), TM (shaded in black and described (8, 20). Cells were transfected with purified plasmid DNA by marked TM), and cytosolic (residues 564 to 582) domains. MT-MMP-1 DA LipofectAMINE treatment (Life Technologies Inc.). MDCK clones is truncated at the edge of the TM domain while MT-MMP-1 ends at DB stably expressing MT-MMP-1 were established by transfection and DB the conserved cysteinyl residue that marks the extreme C terminus of subsequently selected with G418 (Life Technologies, Inc.) as described the hemopexin domain. B, Western blot analysis of DMT-MMP-1-trans- previously (8, 20). fected cells. Serum-free conditioned medium from COS-7 cells tran- Plasmids—Following screening of an HT-1080 cDNA library con- siently transfected with control (lane 1), MT-MMP-1 (lane 2), MT- DA structed in the pBK-CMV vector with a 0.45-kilobase canine MT- MMP-1 (lane 3), or wild-type MT-MMP-1 (lane 4) expression vectors DB MMP-1 cDNA isolated from an MDCK cDNA library, a human clone were analyzed by immunoblotting with MT-MMP-1-specific polyclonal antisera. The dark and clear arrowheads indicate the positions of the was obtained and sequenced. The predicted amino acid sequence of putative pro- and processed forms of DMT-MMP-1. In wild-type MT- full-length HT-1080 MT-MMP-1 was identical with that described by MMP-1-transfected cells, immunoblots of cell lysates identified the Okada et al. (13). To generate pro-MT-MMP-1 and pro-MT-MMP-1 , DA DB 63-kDa form of the enzyme as the major product (data not shown). C, a59 primer (ACCATGTCTCCCGCCCCAAGACCCTCCCGT) was paired gelatin zymography of MT-MMP-1-transfected cells. Serum-free condi- with the 39 primers, TTCAGCTCACCGCCCCGCCGCC or TTCAGAA- tioned medium from COS-7 cells transiently transfected with control GAAGAAGACTGCAAGGCC, respectively, in separate PCR reactions to (lane 1), MT-MMP-1 (lane 2), MT-MMP-1 (lane 3), or MT-MMP-1 DA DB generate cDNA fragments with the intended truncations at the C ter- (lane 4) expression vectors were depleted of endogenous progelatinase A minus (i.e. immediately prior to the TM domain as defined by Cao et al. by gelatin affinity chromatography and analyzed by zymography. In (9) at Ser or at the conserved Cys residue at position 508 (6, 7); see lane 5, the MT-MMP-1 zymogram was developed in the presence of DB Fig. 1A). The PCR fragments were then cloned into pCR3-Uni vector 1.0 mM BB-94. Identical results were obtained with b-casein or k-elastin (Invitrogen) and characterized by sequencing as described (8). Expres- zymograms (D. Pei and S. J. Weiss, unpublished observation). sion vectors for a -proteinase inhibitor (a PI), the Pittsburgh mutant of 1 1 a PI (a PI ), and a TM-deletion mutant of furin were provided by A. 1 1 Pitt SDS-PAGE, Western Blots, Zymography, and N-terminal Sequenc- Rehemtulla and R. Kaufman (University of Michigan). ing—Basic protocols for these techniques have been described (8, 20). Mutagenesis—Sequential PCR-based mutagenesis was performed as The MT-MMP antisera were raised in New Zealand rabbit using a described previously, and the mutagenized fragments were cloned into bacterially generated recombinant fusion protein between glutathione pCR3-Uni vector and characterized by sequencing (8). The mutagenic transferase and MT-MMP (20). 91–243 primers used are as follows: ATCAAGGCCAATGTTGCAAGGAAGC- Enzymic Reactions—Enzyme assays of MT-MMP-1 and MT-MMP- DA GCTACGCC for R108A, GCCAATGTTCGAAGGGCGCGCTACGCCA- 1 were performed in buffer A supplemented with 150 mM NaCl at DB TCCAG for K110A, AATGTTCGAAGGAAGGCCTACGCCATCCAG- 37 °C unless noted otherwise. Matrix substrates (devoid of contaminat- GGT for R111A, and GTTCGAAGGAAGC GCTTTGCCATCCAGGG- ing progelatinase A activity as determined by gelatin zymography; data TCTC for Y112F (bold nucleotides indicate the altered codons). MT- not shown) were obtained from Collaborative Research (type I, IV, and MMP-1 expression constructs as well as those harboring the above DB V collagens, fibronectin, laminin, vitronectin, and dermatan sulfate mutations (1 mg each) were transiently transfected into COS-7 cells (2 proteoglycan). Processing of pro-MT-MMP-1 by purified soluble furin DB 3 10 /ml) by LipofectAMINE treatment, and, after a 24-h incubation, (specific activity 500 units/mg; Ref. 18) was performed as described (8). an aliquot of the cell-free supernatant (0.02 ml) was analyzed by West- Recombinant progelatinase A (purified as described in Ref. 22) was a ern blotting as described below. gift from R. Fridman (Wayne State University, Detroit, MI), recombi- Protein Purification—MDCK cells stably expressing MT-MMP-1 DB nant human TIMP-2 was supplied by Amgen, and soluble human furin were incubated in Opti-MEM (Life Technologies Inc.) supplemented was provided by R. Fuller, University of Michigan. with the synthetic MMP inhibitor, BB-94 (0.5 mM, British Biotechnol- ogy; Ref. 21), to trap the active form of the proteinase as a reversible RESULTS enzyme-inhibitor complex (20). After 48 h, ;2 liters of conditioned Expression of DMT-MMP-1 Mutants and Detection of Enzy- media were collected, dialyzed against buffer A (50 mM Tris, pH 7.5, 5 mic Activity—COS-7 cells transiently transfected with either mM CaCl ), and then loaded onto a Q-Sepharose column (1.5 3 10 cm). MT-MMP-1 or MT-MMP-1 cDNA each secreted a pair of Bound material was eluted with a NaCl gradient (0 to 1 M), and DA DB fractions containing MT-MMP-1 (identified by Western blotting) were major and minor products that were specifically recognized by DB combined and dialyzed against buffer A. A heparin-Sepharose column polyclonal antibodies to a truncated form of the bacterially (1 3 10 cm) was then loaded with the dialyzed materials and developed expressed protein (Fig. 1B). While the molecular mass of the with a NaCl gradient (0 to 1 M). Positive fractions were pooled and minor secreted proteins (;64 kDa for MT-MMP-1 and ;60 DA passed through a gelatin-Sepharose column (1 3 5 cm) followed by gel kDa for MT-MMP-1 ; lanes 2 and 3, respectively) were con- DB filtration chromatography (Ultrogel ACA44, 1 3 150 cm) in the absence sistent with those of the predicted proforms of the metallopro- of BB-94 to regenerate the active proteinase. In selected experiments, MT-MMP-1 was purified from a batch culture of transiently trans- teinases, the major soluble species detected with either TM- DA fected MDCK cells as described above. deletion mutant was a fragment ;10 kDa smaller in size. The generation of the major and minor forms was not specific to D. Pei and S. J. Weiss, unpublished observation. COS-7 cells since a similar profile was generated with trans- Transmembrane-deletion Mutants of MT-MMP-1 9137 FIG.2. Purification of MT-MMP-1 . A and B, fractionation of MT-MMP-1 on Q-Sepharose and heparin-Sepharose, respectively. Condi- DB DB tioned media from MDCK cells stably transfected with MT-MMP-1 were loaded onto a Q-Sepharose column and eluted with a NaCl gradient. DB The protein content of each fraction was monitored at A (dark squares) while MT-MMP-1 content in each fraction was monitored by 280 DB immunoblot analysis and reported as the percent recovered relative to the fraction containing the highest concentration of MT-MMP-1 (open DB squares). Fractions 9–16 were combined, dialyzed against buffer A, loaded onto a heparin-Sepharose column, and eluted with a NaCl gradient. C, characterization of the isolated MT-MMP-1 products. Conditioned media (lane 1), a pool of fractions 9–16 eluted from Q-Sepharose (lane 2), DB flow-through of fractions 9–16 that did not bind to heparin-Sepharose (lane 3), pool of fractions 10–15 eluted from heparin-Sepharose column (lane 4), and final purified form of MT-MMP-1 (4.5 pmol; lane 5) were separated by SDS-PAGE and visualized by Coomassie staining. In lanes 6 and DB 7, purified MT-MMP-1 (1.2 pmol) was analyzed by immunoblotting and gelatin zymography, respectively. D, the N terminus of MT-MMP-1 as DB DB determined after 10 cycles of sequencing (indicated by bold letters). The open box represents the MT-MMP-1 open reading frame with the amino DB 91 123 acid sequence of MT-MMP-1 from Pro to Glu . fected MDCK cells. As expected, when cells were transiently MMP-1 were subjected to a combination of gelatin-Sepharose DB transfected with wild-type MT-MMP-1, soluble forms of the affinity, Q-Sepharose affinity, and heparin-Sepharose affinity enzyme were not detected in the conditioned media (Fig. 1B). chromatography followed by gel filtration chromatography. Uti- In intact cell systems, MMPs can be recovered in conditioned lizing this protocol, a single immunoreactive ;50-kDa species medium as a mixture of zymogens, processed active enzymes, was isolated that co-migrated with the band of activity detected zymogen-inhibitor complexes, or enzyme-inhibitor complexes by gelatin-zymography (Fig. 2C). Following 10 cycles of N-termi- (1–3). In the case of almost all MMP family members, many of nal sequence analysis, the material was identified as MT-MMP-1 these forms can be detected following SDS-PAGE in substrate- with a single start site at Tyr (Fig. 2D). Interestingly, this N impregnated gels (1–3). Thus, serum-free conditioned media terminus not only aligns with that of the active forms of all other from control, MT-MMP-1 , MT-MMP-1 , or MT-MMP-1 MMPs (6), but it is also identical with that reported for the DA DB transfected cells were depleted of endogenous gelatinases by mature form of wild-type MT-MMP-1 recovered from the HT- gelatin affinity chromatography (DMT-MMP-1 does not bind to 1080 fibrosarcoma cell line (10). Thus, while almost all MMPs are gelatin; see below) and electrophoresed in gelatin- containing synthesized and secreted as inactive zymogens, MT-MMP-1 DB gels. Proteinases were then allowed to renature following the (as well as MT-MMP-1 ; see below) underwent further process- DA removal of SDS and then incubated overnight at 37 °C. As ing to its active form. shown in Fig. 1C, supernatants recovered from MT-MMP-1 - Like stromelysin-3, the only other membrane of the MMP DA or MT-MMP-1 -transfected cells each revealed the presence of family to be secreted as a fully processed active proteinase, DB a single band of gelatinolytic activity whose relative mobility MT-MMP-1 contains a motif of basic amino acids (i.e. RRKR) matched that of the major form of DMT-MMP-1 detected by immediately upstream of its catalytic domain (see Fig. 3A; Western blotting. Significantly, identical results were obtained Refs. 4 and 8). Recently, the RXKR array in stromelysin-3 (X 5 when zymograms were performed with either b-casein- or nonbasic amino acid) was shown to act as an endoproteolytic k-elastin-impregnated gels as well (data not shown). Regard- processing signal for an intracellular serine proteinase belong- less of substrate used, the band of proteolytic activity attrib- ing to the proprotein convertase family (8). Because specific uted to either DMT-MMP-1 mutant was completely inhibited proprotein convertases can display varying requirements for when zymograms were performed in the presence of the MMP- basic residues at positions 21, 22, and 24 relative to the 21 22 24 specific inhibitor, BB-94 (Fig. 1C). scissile bond (i.e. P ,P , and P , respectively; Refs. 23 and Purification of DMT-MMP-1 and Characterization of Zymogen 24), a potential role for this enzyme class in DMT-MMP-1 Processing—Because both MT-MMP-1 and MT-MMP-1 ap- processing was initially assessed by successively substituting DA DB peared to undergo a similar, if not identical, processing event to each basic residue with an Ala moiety in transient transfection generate active proteinases as assessed by zymography, one of assays. As shown in Fig. 3B, each of these substitutions almost the mutants (i.e. MT-MMP-1 ) was stably expressed in the completely blocked MT-MMP-1 processing (lanes 2–4). In DB DB MDCK cell line for further analyses. As shown in Fig. 2, serum- contrast, a Tyr 3 Ala substitution at the less critical P site free conditioned media from stable transfectants expressing MT- (23) did not affect MT-MMP-1 processing (Fig. 3, lane 5). DB 9138 Transmembrane-deletion Mutants of MT-MMP-1 FIG.4. DMT-MMP-1-dependent activation of recombinant progelatinase A. A, zymography and N-terminal sequence analysis of progelatinase A. Recombinant progelatinase A (140 nM) was incubated alone (lane 1) or with 14 nM,28nM,or56nM MT-MMP-1 (lanes 2–4, DB FIG.3. Proprotein convertase-dependent processing of MT- respectively) for2hat37 °Cin buffer A supplemented with 150 mM MMP-1 . A, mutational analysis of the putative proprotein convertase DB NaCl and 0.01% Brij 35 in a final volume of 0.02 ml. In lanes 5–7, recognition motif. The amino acids surrounding the cleavage site in progelatinase was incubated alone, with MT-MMP-1 or with MT- MT-MMP-1 are shown with the basic residue motif underlined. B, DB DB 108 MMP-1 and TIMP-2 (350 nM), respectively, for 16 h at 37 °C. Aliquots DB expression vectors for native MT-MMP-1 (lane 1), Arg 3 Ala DB 110 111 of each reaction mixture were analyzed by gelatin zymography or were (R108A; lane 2), Lys 3 Ala (K110A; lane 3), Arg 3 Ala (R111A; 112 electrophoresed and electroblotted for N-terminal sequence analysis. lane 4), and Tyr 3 Ala (Y112A; lane 5) were transfected into COS-7 The bold sequence beginning with Leu represents the first 10 cycles of cells (2 3 10 /ml) by LipofectAMINE treatment, and the secreted prod- the N terminus of the ;64-kDa form of gelatinase A (indicated by ucts (0.02 ml of the cell-free supernatant) were analyzed by immuno- asterisk in lane 4) while the sequence beginning with Tyr represents blotting. C, inhibition of MT-MMP-1 processing by a PI . COS-7 DB 1 Pitt the first 5 cycles of the N terminus of the ;62-kDa form of gelatinase cells were transiently transfected as described above with MT-MMP- (indicated by dark arrow in lane 7). The faint bands of gelatinolytic 1 expression vector alone (lane 1) or co-transfected with MT-MMP- DB activity detected at ;50 kDa are due to MT-MMP-1 (lanes 2–4). In DB 1 and a PI (lane 2)or a PI expression vectors (lane 3), and the DB 1 1 Pitt lanes 8–10, progelatinase A (140 nM) was incubated alone (lane 8) with cell-free supernatants were analyzed by immunoblotting. MT-MMP-1 (14 nM; lane 9) or MT-MMP-1 (14 nM; lane 10)for3h DB DA in 20 mM Hepes/KOH (pH 7.5), 0.1 mM CaCl , and 0.02% Brij-35 as described (10). The asterisk, arrowhead, and circle indicate the posi- Given that COS are known to express only two members of the tions of the ;64-kDa, ;62-kDa, and ;42-kDa forms of gelatinase A, proprotein convertase family that recognize RXKR motifs (i.e. respectively. B, furin-mediated processing of pro-MT-MMP-1 . Pro- DB furin and PACE4; Ref. 24), cells were co-transfected with MT- MT-MMP-1 (100 nM) isolated from a PI co-transfected cells was DB 1 Pitt MMP-1 and the Pittsburgh mutant of a PI (a PI ), a reac- incubated alone (lane 1) or with soluble furin (10 nM; lane 2)for1hat DB 1 1 Pitt tive site variant that inhibits furin (but not PACE4) activity in 37 °C in a final volume of 0.02 ml and analyzed by Western blotting. For zymography (lanes 3–5), progelatinase A (140 nM) was incubated alone situ (25, 26). Under these conditions, a PI completely 1 Pitt (lane 3), with pro-MT-MMP-1 (28 nM; lane 4), or with furin-processed DB blocked MT-MMP-1 processing while wild-type a PI exerted DB 1 MT-MMP-1 (28 nM; lane 5) for 16 h at 37 °C. Furin alone did not DB no inhibitory effect (Fig. 3C). MT-MMP-1 processing was DB activate progelatinase A. similarly inhibited by a PI in MDCK cells (data not shown). 1 Pitt These results (together with the demonstration that soluble completely blocked by the addition of the MMP inhibitor, furin processes pro-DMT-MMP-1 to its active form under cell- TIMP-2 (Fig. 4A) or BB-94 (data not shown). Interestingly, free conditions; see below) indicate that pro-DMT-MMP-1 mat- while earlier studies have demonstrated that membrane-asso- uration is regulated by a furin-dependent pathway in an intact ciated forms of MT-MMP-1 can also process progelatinase A cell system. into a 42-kDa active species (10, 27, 28), this product was not Activation of Recombinant Progelatinase A by Purified DMT- detected under the standard conditions employed. However, MMP-1—Current evidence indicates that MT-MMP-1-express- when either MT-MMP-1 or MT-MMP-1 were incubated DA DB ing cells initiate progelatinase A activation via a two-step proc- with progelatinase A in a low ionic strength buffer identical ess that involves an initial cleavage of the Asn -Leu bond with that used previously (10, 27, 28), MT-MMP-1 -depend- DA/B followed by an autocatalytic conversion of the Leu interme- ent progelatinase A activation was significantly accelerated diate into a 62-kDa active enzyme with an N-terminal Tyr and the 42-kDa form of gelatinase generated (Fig. 4A, lanes residue (10, 27). Nonetheless, the ability of MT-MMP to di- 8–10). The ability of mature MT-MMP-1 to mediate progelati- rectly cleave progelatinase A is unclear, and it has been pos- DA/B tulated that additional intermediates may be involved (9, 10). nase A activation is consistent with a model wherein active Thus, purified active MT-MMP-1 was incubated with recom- MT-MMP-1 directly cleaves the gelatinase zymogen, but DB DA/B binant progelatinase A and processing monitored by gelatin the data do not rule out the possibility that the DMT-MMP-1 zymography and N-terminal sequence analysis. Following a zymogen only induces progelatinase A to undergo autocatalytic 2-h incubation at 37 °C, MT-MMP-1 initially cleaved proge- processing to its active form. Hence, purified pro-DMT-MMP-1 DB latinase A (which migrates as a ;68-kDa species) into a ;64- was isolated (i.e. from cells co-transfected with MT-MMP-1 DB kDa fragment (Fig. 4A). N-terminal sequence analysis of the and a PI ), and its ability to mediate progelatinase A activa- 1 Pitt ;64-kDa gelatinase A fragment yielded the Leu form of the tion was examined. As shown in Fig. 4B, pro-MT-MMP-1 was DB enzyme (Fig. 4A). Subsequently, the 64-kDa form of the enzyme unable to stimulate progelatinase A activation. However, when underwent further processing to a 62-kDa product whose N pro-MT-MMP-1 was processed to its active form ex situ with DB terminus confirmed the generation of [Tyr ]gelatinase A (Fig. a TM-deleted soluble form of furin (Fig. 4B, lane 2) and then 4A). As expected, the ability of MT-MMP-1 (as well as MT- incubated with progelatinase A, the gelatinase zymogen was DB MMP-1 ; data not shown) to activate progelatinase A was readily activated (lane 5). Thus, only the processed active form DA Transmembrane-deletion Mutants of MT-MMP-1 9139 generated by exchanging the TM and cytosolic domains of the metalloproteinase with those of the IL-2 receptor functioned normally in terms of its ability to mediate progelatinase A activation (9). Although this result is consistent with our con- clusion that the extracellular domain of MT-MMP-1 confers the proteinase with its distinct characteristics, these authors also reported that a TM-deletion mutant encoding residues 1–535 of MT-MMP-1 was unable to process progelatinase A (9). In com- paring our experimental approaches, it is important to note that the soluble MT-MMP-1 generated in their study was not isolated nor were its interactions with progelatinase A exam- FIG.5. Substrate specificity of MT-MMP-1 . Type I collagen (3 DB ined directly (9). Instead, Cao et al. (9) judged their TM-dele- mg; lane 1), type IV collagen (2 mg; lane 3), and type V collagen (2 mg; lane 5) were incubated alone or with 45 nM MT-MMP-1 (lanes 2, 4, tion mutant to be inactive on the basis of its inability to process DB and 6, respectively) at 25 °C for 16 h. Type I gelatin (3 mg; lane 7), endogenously derived progelatinase A secreted by COS-1 cells fibronectin (4 mg; lane 10), laminin (4 mg; lane 13), vitronectin (4 mg; in a transient transfection assay system (9). Under these con- lane 16), or dermatan sulfate proteoglycan (5 mg; lane 19) were incu- ditions, however, attempts to assess the activity of secreted bated alone, with MT-MMP-1 (45 nM; lanes 8, 11, 14, 17, and 20, DB MT-MMP-1 would be complicated by the presence of cell-de- respectively) or with MT-MMP-1 and 1 mM BB-94 (lanes 9, 12, 15, 18, DB and 21, respectively) at 37 °C in a final volume of 0.025 ml for 16 h. rived TIMPs which can interfere with progelatinase A process- Reaction mixtures were separated by SDS-PAGE and Coomassie- ing by either inhibiting DMT-MMP-1 activity directly, or, in the stained. The arrowhead by lane 15 indicates the position of the laminin case of TIMP-2, by binding to the C-terminal domain of the B chain, while the hatch marks at the margin of lanes 16 and 19 gelatinase zymogen (4, 27, 28). Indeed, when endogenous levels indicate the positions of the 97-, 69-, 45-, 32-, and 28-kDa molecular mass markers, respectively. Dermatan sulfate proteoglycan is recorded of TIMP are overwhelmed by co-transfecting COS cells with as DSPG. MT-MMP-1 and progelatinase A, gelatinase activation can DA/B be readily detected in the intact cell system as well as our of DMT-MMP-1 is able to mediate progelatinase A activation. purified system. Thus, while anchoring a proteinase to the cell ECM-degrading Activity of DMT-MMP-1—While previous membrane might be predicted to more effectively shield an attention has focused solely on the role of MT-MMP-1 in proge- active proteinase from soluble inhibitors (and to perhaps pro- latinase A activation (4, 9, 10, 13, 27), the ability of DMT- vide a surface more conducive for accelerating processing MMP-1 to degrade gelatin, b-casein, or k-elastin following zy- events) (e.g. Refs. 29–31), our data demonstrate that the TM- mography suggested that the proteinase might express activity deletion mutants retain the key functional properties of the against a wider range of targets. Thus, purified MT-MMP-1 wild-type enzyme. DA/B was incubated with either basement membrane- or intersti- As a consequence of our attempts to characterize the activity of DMT-MMP-1, we also discovered that the TM-deletion mu- tium-associated ECM molecules, and proteolysis was assessed in the absence or presence of BB-94. As shown in Fig. 5, while tants are capable of undergoing rapid processing to their ma- ture forms. This finding is noteworthy since, as a general rule, MT-MMP-1 did not degrade native type I, IV, or V collagens, DB the enzyme readily proteolyzed gelatin as well as fibronectin, MMPs are synthesized and secreted as inactive zymogens (1–3, 7). However, we recently reported that in a fashion similar to the B chain of laminin, vitronectin, and dermatan sulfate pro- teoglycan via a BB-94-sensitive process. Similar, if not identi- that observed for the DMT-MMP-1 mutants, prostromelysin-3 is secreted as an active enzyme following its intracellular proc- cal, results were obtained with purified MT-MMP-1 (data not DA shown). Given that none of these substrates were contaminated essing within the constituitive secretory pathway (8). In this with detectable quantities of progelatinase A (see “Experimen- case, activation was dependent upon a decapeptide insert that tal Procedures”), we conclude that DMT-MMP-1 mutants can is sandwiched between the pro- and catalytic domains of express intrinsic matrix-degrading activities. stromelysin-3 and encrypted with an extended furin recogni- tion motif (i.e. RXRXKR). At the time that these earlier studies DISCUSSION were completed, stromelysin-3 was the only member of the Sequence alignments of the 13 human MMPs that have been MMPs family known to contain this recognition sequence. characterized to date indicate that amino acids 1–508 of the However, with the recent cloning of MT-MMP-1, -2, and -3, it is 538-residue-long extracellular domain of MT-MMP-1 contain clear that all three of these enzymes contain homologous in- all of the major structural elements of the secreted members of serts which include an array of basic residues (i.e. RRKR) that this gene family (i.e. a propeptide and catalytic domain as well match the recognition motif of the proprotein convertases (i.e. as a hemopexin-like region that is bounded by a pair of highly RX(K/R)-R; Refs. 4–6). Consistent with this prediction, (i) the conserved cysteinyl residues; Refs. 4–7). Given that the C N terminus of DMT-MMP-1 was located at Tyr on the C- termini of virtually all secreted MMPs end at, or extend no terminal side of the Arg-Arg-Lys-Arg motif, (ii) DMT-MMP-1 more than 8 amino acids beyond, the final cysteinyl residue in processing could be inhibited by either inserting point muta- the hemopexin domain (7), we reasoned that TM-deletion mu- tions in the RRKR motif or by co-transfecting cells with the tants of MT-MMP-1 that retained this modular organization furin-specific inhibitor, a PI , and (iii) the DMT-MMP-1 zy- 1 Pitt would encode functional proteinases. Indeed, as demonstrated, mogen could be processed to its active form ex situ by soluble regardless of whether soluble mutants of MT-MMP-1 were furin. While we have not yet identified the intracellular/extra- truncated either at the edge of the TM domain or at the end of cellular compartments in which the DMT-MMP-1 zymogen un- the hemopexin domain, the expressed proteins displayed sim- dergoes processing in the intact cell, furin is a membrane- ilar, if not identical, activities as assessed by zymography, associated endoprotease that not only cycles between the trans- progelatinase A processing, or substrate specificity. By itself, our work does not rule out the possibility that the Although stromelysin-3 and MT-MMP-1 both contain proprotein TM or cytosolic domains of MT-MMP-1 convey additional struc- convertase recognition motifs, comparisons of their genomic organiza- tural information to the processed proteinase (i.e. beyond act- tion and chromosomal localization indicate that the two metalloprotein- ing as a membrane anchor). However, while our work was in ases are not closely related and belong to separate branches of the progress, Cao et al. (9) reported that an MT-MMP-1 chimera phylogenetic tree (D. Pei and S. J. Weiss, unpublished observation). 9140 Transmembrane-deletion Mutants of MT-MMP-1 Golgi network and the cell surface, but also undergoes been heightened by recent findings which suggest that soluble processing to a soluble form that accumulates extracellularly forms of MT-MMP-1 may be generated in vivo (13). Thus, while (32–34). Indeed, the possibility that DMT-MMP-1 may undergo the MT-MMP-1 antigen has been immunodetected on the sur- extracellular processing is further supported by our results face of cancer cells in vivo (4), RNA in situ hybridization studies with the TM-deleted form of soluble furin. Nonetheless, in spite have more recently demonstrated that MT-MMP-1 transcripts of the fact that furin is the most credible MT-MMP-1 activator are confined to the surrounding stromal cells (13). Should MT- identified to date, caution should be exercised in terms of MMP-1 undergo solubilization and intercellular transfer in situ extrapolating processing pathways that are operative for DMT- (4, 13), tumor cells could potentially use the stroma-derived MMP-1 to the wild-type enzyme. Indeed, in contrast to the enzyme to assemble a multicatalytic complex on their surface results obtained with DMT-MMP-1, we and others have found that would not only arm them with the ability to catalyze that COS cells transfected with wild-type MT-MMP-1 route progelatinase A activation, but also to express an additional most of the enzyme to the cell surface as the unprocessed repertoire of proteolytic activities. Additional studies will be 3,5 zymogen rather than the mature enzyme (4, 9, 31). Utilizing required to directly compare the soluble and membrane-an- chimeric constructs between stromelysin-3 and wild-type MT- chored forms of MT-MMP-1, but the established catalytic ac- MMP-1, it appears that while the furin recognition motif in tivity of the TM-deletion mutants should provide a useful tool either of the secreted metalloproteinases can be processed ef- for characterizing the enzymic properties of this new family of fectively, the TM domain of MT-MMP-1 appears to “shield” the membrane-anchored MMPs. recipient proteinase from undergoing rapid intracellular proc- Acknowledgments—We thank R. Fuller (University of Michigan) for essing. The mechanisms responsible for controlling the intra- assistance in purifying soluble furin, A. Galloway (British Biotechnol- cellular and extracellular processing of wild-type MT-MMP-1 ogy) for BB-94, and K. Langley (Amgen) for recombinant TIMP-2. require further analysis, but the fact that the active form of the full-length (10) and mutant enzyme display an identical N REFERENCES terminus directly downstream of the proprotein convertase- 1. Woessner, J. F., Jr. (1991) FASEB J. 5, 2145–21549 2. Matrisian, L. M. (1992) BioEssays 14, 455–463 recognition motif strongly suggests a role for furin or, perhaps, 3. Stetler-Stevenson, W. G., Aznavoorian, S., and Liotta, L. A. (1993) Annu. Rev. a related proprotein convertase (e.g. PC6; Refs. 24 and 35) in Cell Biol. 9, 541–573 zymogen activation. 4. Sato, H., Takino, T., Okada, Y., Cao, J., Shinagawa, A., Yamamoto, E., and Seiki, M. (1994) Nature 370, 61–65 In the presence of purified active DMT-MMP-1 (but not its 5. Will, H., and Hinzmann, B. (1995) Eur. J. Biochem. 231, 602–608 zymogen), progelatinase A was processed to its mature form 6. Takino, T., Sato, H., Shinagawa, A., and Seiki, M. (1995) J. Biol. Chem. 270, 81 38 23013–23020 (i.e. [Tyr ]gelatinase) via the formation of the Leu interme- 7. Birkedal-Hansen, H., Moore, W. G. I., Bodden, M. K., Windsor, L. J., Birkedal- diate. This two-step, TIMP-2-sensitive activation cascade is Hansen, B., DeCarlo, A., and Engler, J. A. (1993) Crit. Rev. Oral Biol. Med. identical with that previously established for crude prepara- 4, 197–250 8. Pei, D., and Weiss, S. J. (1995) Nature 375, 244–247 tions of plasma membrane-associated MT-MMP-1 (27) and al- 9. Cao, J., Sato, H., Takino, T., and Seiki, M. (1995) J. Biol. Chem. 270, 801–805 lows us to conclude that DMT-MMP-1 can initiate the process- 10. Strongin, A. Y., Collier, I., Bannikov, G., Marmer, B. L., Grants, G. A., and Goldberg, G. I. (1995) J. Biol. Chem. 270, 5331–5338 ing event independently of additional co-factors or substrates. 11. Brown, P. D., Bloxidge, R. E., Anderson, E., and Howell, A. (1993) Clin. Exp. Interestingly, the ability of DMT-MMP-1 to directly activate Metastasis 11, 183–189 progelatinase A under cell-free conditions contrasts with a 12. Brown, P. D., Bloxidge, R. E., Stuart, N. S. A., Gatter, K. C., and Carmichael, J. (1993) J. Natl. Cancer Inst. 85, 574–578 recent report by Strongin et al. (10) wherein an MT-MMP- 13. Okada, A., Bellocq, J.-P., Rouyer, N. Chenard, M.-P., Rio, M.-C., Chambon, P., 1zTIMP-2 complex (rather than MT-MMP-1 alone) was pro- and Basset, P. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 2730–2734 posed to function as the membrane-associated activator of the 14. Tournier, J. M., Polette, M., Hinnrasky, J., Beck, J., Werb, Z., and Basbaum, C. (1994) J. Biol. Chem. 269, 25454–25464 gelatinase zymogen. We were unable to reproduce this finding 15. Aimes, R. T., and Quigley, J. P. (1995) J. Biol. Chem. 270, 5872–5876 with MT-MMP-1 , but cannot rule out the possibility that 16. Vassalli, J.-D., and Pepper, M. S. (1994) Nature 370, 14–15 DA/B 17. Brenner, C., and Fuller, R. S. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 922–926 TIMP-2 plays a more complex role on the membrane surface. 18. Bravo, D. A., Gleason, J. B., Sanchez, R. I., Roth, R. A., and Fuller, R. S. (1994) However, an interpretation of the data presented by Strongin et J. Biol. Chem. 269, 25830–25837 19. de Vries, T., Srnka, C. A., Palcic, M. M., Swiedler, S. J., van den Eijnden, D. H., al. (10) is complicated by the fact that even in the apparent and Macher, B. A. (1995) J. Biol. Chem. 270, 8712–8722 absence of TIMP-2, MT-MMP-1 continued to process progela- 20. Pei, D., Majmuder, G., and Weiss, S. J. (1994) J. Biol. Chem. 269, tinase A to the [Leu ]gelatinase intermediate, but not the final 25849–25855 21. Davies, B., Brown, P. Q., East, N., Crimmin, M. J., and Balkwill, E. R. (1993) mature form. Thus, it remains possible that TIMP-2 exerts its Cancer Res. 53, 2087–2091 stimulatory effect by accelerating the inter- or intramolecular 22. Fridman, R., Fuerst, T. R., Bird, R. E., Hoyhtya, M., Oelkuct, M., Kraus, S., 38 81 autocatalytic conversion of [Leu ]gelatinase to [Tyr ]gela- Komarek, D., Liotta, L., Berman, M. L., and Stetler-Stevenson, W. G. (1992) J. Biol. Chem. 267, 15398–15405 tinase on the cell surface (31) rather than by stimulating MT- 23. Watanabe, T., Murakami, K., and Nakayama, K. (1993) FEBS Lett. 320, MMP-1 activity directly. 215–218 24. Seidah, N. G., Chretien, M., and Day, R. (1994) Biochimie (Paris) 76, 197–209 To date, the only function ascribed to MT-MMP-1 has been 25. Rehemtulla, A., and Kaufman, R. J. (1992) Blood 79, 2349–2355 its ability to activate progelatinase A (4, 9, 10, 13, 27, 28). 26. Wasley, C. L., Rehemtulla, A., Bristol, J. A., and Kaufman, R. J. (1993) J. Biol. However, we have demonstrated that purified MT-MMP-1 Chem. 268, 8458–8465 DA/B 27. Strongin, A. Y., Marmer, B. L., Grant, G. A., and Goldberg, G. I. (1993) J. Biol. can also degrade a number of extracellular matrix components. Chem. 268, 14033–14039 These data indicate that the ability of MT-MMP-1-transfected 28. Ward, R. V., Atkinson, S. J., Reynolds, J. J., and Murphy, G. (1994) Biochem. J. 304, 263–269 cells to express a heightened invasive potential may not nec- 29. Bangalore, N., and Travis, J. (1994) Biol. Chem. Hoppe-Seyler. 375, 659–666 essarily be linked to progelatinase A activation alone (4). Al- 30. Young, T. N., Pizzo, S. V., and Stack, M. S. (1995) J. Biol. Chem. 270, though our studies have employed a soluble form of MT- 999–1002 31. Atkinson, S. J., Crabbe, T., Cowell, S., Ward, R. V., Butler, M. J., Sato, H., MMP-1, we believe that the modular organization of MT- Seiki, M., Reynolds, J. J., and Murphy, G. (1995) J. Biol. Chem. 270, MMP-1 strengthens the likelihood that the membrane-tethered 30479–30485 32. Vidricaire, G., Denault, J.-B., and Leduc, R. (1993) Biochem. Biophys. Res. form displays a similar substrate specificity. Furthermore, the Commun. 195, 1011–1018 potential physiologic relevance of DMT-MMP-1 mutants have 33. Molloy, S. S., Thomas, L., VanSlyke, J. K., Stenberg, P. E., and Thomas, G. (1994) EMBO J. 13, 18–33 34. Vey, M., Schafer, W., Berghofer, B., Klenk, H.-D., and Garten, W. (1994) J. Cell However, a portion of the MT-MMP-1 zymogen does undergo proc- Biol. 127, 1829–1842 essing since the mature enzyme has been isolated from HT-1080 35. Horimoto, T., Nakayama, K., Smeekens, S. P., and Kawaoka, Y. (1994) J. Virol. plasma membranes (10) and is required for progelatinase activation. 68, 6074–6078 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Transmembrane-deletion Mutants of the Membrane-type Matrix Metalloproteinase-1 Process Progelatinase A and Express Intrinsic Matrix-degrading Activity

Pei, Duanqing; Weiss, Stephen J.
Journal of Biological Chemistry – Apr 1, 1996
Download PDF
6 pages

Loading next page...
 
/lp/unpaywall/transmembrane-deletion-mutants-of-the-membrane-type-matrix-03AnrovBR5

References (37)

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.271.15.9135
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 271, No. 15, Issue of April 12, pp. 9135–9140, 1996 © 1996 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Transmembrane-deletion Mutants of the Membrane-type Matrix Metalloproteinase-1 Process Progelatinase A and Express Intrinsic Matrix-degrading Activity* (Received for publication, January 25, 1996) Duanqing Pei‡ and Stephen J. Weiss§ From the Division of Hematology/Oncology, Department of Internal Medicine, the University of Michigan Comprehensive Cancer Center, University of Michigan, Ann Arbor, Michigan 48109 Membrane-type matrix metalloproteinase-1 (MT- remodeling of the extracellular matrix (ECM) in events ranging MMP-1) has been proposed to play a critical role in from organogenesis to tumor metastasis (1–3). Very recently, regulating the expression of tissue-invasive phenotypes three new members of this family were discovered by screening in normal and neoplastic cells by directly or indirectly cDNA libraries for homologies to conserved regions of the known mediating the activation of progelatinase A. To begin MMP genes and named the membrane-type matrix metallopro- characterizing MT-MMP-1 structure-function relation- teinases-1, -2, and -3 (MT-MMP-1, -2, and -3; Refs. 4–6). Based ships, transmembrane-deletion mutants were con- on their predicted amino acid sequences, each of the MT-MMPs, structed, and the processing of the zymogens as well as like almost all previously characterized MMPs, contains (i) a the enzymic activity of the mature proteinases was an- candidate leader sequence, (ii) a propeptide region which in- alyzed. We now demonstrate that pro-MT-MMP-1 mu- cludes a highly conserved PRCGXPD sequence that helps stabi- tants are efficiently processed to active proteinases fol- lize the MMP zymogen in a catalytically inactive state, (iii) a lowing post-translational endoproteolysis immediately zinc-binding catalytic domain, and (iv) a hemopexin-like domain downstream of an Arg -Arg-Lys-Arg basic motif by a near their respective C termini (4–7). In addition, in a pattern proprotein convertase-dependent pathway. The se- similar to that described for stromelysin-3, each of the MT-MMPs creted form of active MT-MMP-1 not only displays an N contains a short amino acid insert sandwiched between their pro- terminus identical with that described for the processed 112 and catalytic domains that encodes a potential recognition motif wild-type enzyme at Tyr (Strongin, A. Y., Collier, I., for members of the proprotein convertase family (4–8). Despite Bannikov, G., Marmer, B. L., Grants, G. A., and Goldberg, their considerable similarity to other MMP family members, G. I. (1995) J. Biol. Chem. 270, 5331–5338), but also di- however, only the MT-MMPs contain ;75–100 amino acid exten- rectly mediated progelatinase A activation via a two- sions at their C termini, each of which includes a hydrophobic step proteolytic cascade indistinguishable from that ob- stretch consistent with the presence of a transmembrane (TM) served with intact cells. Furthermore, although the only function previously ascribed to MT-MMP-1 is its ability domain (4–6, 9). Thus, in contradistinction to all other MMPs, to act as a progelatinase A activator, purified transmem- the MT-MMPs are expressed as membrane-associated ectoen- brane deletion mutants also expressed proteolytic activ- zymes rather than soluble proteins. ities against a wide range of extracellular matrix mole- Although little is known with regard to the potential functions cules. Given recent reports that MT-MMP-1 ectodomains of the MT-MMPs, most attention has focused on the ability of may undergo intercellular transfer in vivo (Okada, A., MT-MMP-1 as well as MT-MMP-3 to induce the processing of the Bellocq, J.-P., Rouyer, N., Chenard, M.-P., Rio, M.-C., MMP zymogen, progelatinase A, to its activated form (i.e. Chambon, P., and Basset, P. (1995) Proc. Natl. Acad. Sci. 81 38 [Tyr ]gelatinase A) via a [Leu ]gelatinase A intermediate (4, 6, U. S. A. 92, 2730–2734), our data suggest that soluble 10). Given the ability of activated gelatinase A to cleave a wide forms of the proteinase confer recipient cells with the range of ECM substrates (including native types I, IV, V, VII, and ability to not only process progelatinase A, but also di- XI collagen, denatured collagens, elastin, proteoglycans, laminin, rectly degrade extracellular matrix components. and fibronectin) as well as the association of gelatinase A activa- tion with the expression of tissue-invasive phenotypes (1–3, 11– 15), MT-MMPs have been dubbed as possible “master switches” Members of the matrix metalloproteinase (MMP) gene family that control ECM remodeling (16). Nonetheless, the processes have been implicated in the physiologic as well as pathologic that regulate the activation of the MT-MMP zymogens them- selves to mature forms remain undefined as does the mechanism by which MT-MMPs mediate progelatinase A activation (4–6, 9, * This study was supported in part by Grant AI23876 from the 10). In large part, further progress in characterizing MT-MMP National Institutes of Health, the Susan G. Komen Breast Cancer activities has been hindered by the technical problems associated Foundation, and United States Army Medical Research Command Grant DAMD17-94-J-4322. The costs of publication of this article were with isolating and purifying membrane-associated molecules. defrayed in part by the payment of page charges. This article must Given that similar difficulties with other transmembrane en- therefore be hereby marked “advertisement” in accordance with 18 zymes have been negotiated by generating TM-deleted soluble U.S.C. Section 1734 solely to indicate this fact. mutants (17–19), we noted that, with the exception of the ex- ‡ Supported in part by an Oncology research training grant from NHLBI, National Institutes of Health. tended C-terminal domain, the modular organization of the MT- § To whom correspondence should be addressed: Division of Hema- MMPs is identical with that of the secreted MMPs (7). Hence, two tology/Oncology, 5220 MSRB III, 1150 W. Medical Center Dr., Ann Arbor, Michigan 48109-0640. Tel.: 313-764-0030; Fax: 313-764-0101. The abbreviations used are: MMP, matrix metalloproteinase; a PI, 1 , MT-MMP-1 ; TM, transmembrane; MDCK, Madin-Darby 1 DB 1–508 a -proteinase inhibitor; a PI , Pittsburgh mutant of a -proteinase canine kidney cells; PAGE, polyacrylamide gel electrophoresis; PCR, 1 1 Pitt 1 inhibitor; ECM, extracellular matrix; MT-MMP-1, -2, -3, membrane- polymerase chain reaction. type matrix metalloproteinase-1, -2, and -3; DMT-MMP-1, transmem- All MT-MMPs also contain a homologous 8-amino acid insert within brane-deleted MT-MMP-1; MT-MMP-1 , MT-MMP-1 ; MT-MMP- their catalytic domains whose function remains undefined (4–6). DA 1–538 This is an Open Access article under the CC BY license. 9136 Transmembrane-deletion Mutants of MT-MMP-1 TM-deletion mutants of MT-MMP-1 were constructed by either truncating the molecule (i) immediately upstream of the start site of the TM domain (i.e. MT-MMP-1 ; herein referred to as 1–538 MT-MMP-1 ) or (ii) at the conserved cysteinyl residue found at, DA or near, the terminus of all hemopexin domain-containing MMPs (i.e. MT-MMP-1 or MT-MMP-1 ). Utilizing these con- 1–508 DB structs, we now demonstrate that TM-deleted MT-MMP-1 (DMT- MMP-1) mutants undergo efficient post-translational endoprote- 111 112 olysis between Arg -Tyr by a proprotein convertase- dependent pathway to generate fully active proteinases. Furthermore, the purified DMT-MMP-1 mutants not only acti- vate recombinant progelatinase A directly via a two-step activa- tion cascade identical with that described for the membrane- associated enzyme, but they also express heretofore unsuspected ECM-degrading activities. EXPERIMENTAL PROCEDURES FIG.1. Expression and characterization of DMT-MMP-1. A, do- Cell Culture—COS-7 cells and MDCK cells (both obtained from main alignments of MT-MMP-1 and the DMT-MMP-1 mutants. Wild- ATCC) were maintained in Dulbecco’s modified Eagle’s medium (Life type MT-MMP-1 contains 582 amino acids arranged as a series of pre- Technologies, Inc.) containing 10% fetal bovine serum (HyClone) and 4 (shaded box), pro-, catalytic, hemopexin (bounded by a pair of highly mML-glutamine, 100 units/ml penicillin, and 100 mg/ml streptomycin as conserved cysteinyl residues indicated as C), TM (shaded in black and described (8, 20). Cells were transfected with purified plasmid DNA by marked TM), and cytosolic (residues 564 to 582) domains. MT-MMP-1 DA LipofectAMINE treatment (Life Technologies Inc.). MDCK clones is truncated at the edge of the TM domain while MT-MMP-1 ends at DB stably expressing MT-MMP-1 were established by transfection and DB the conserved cysteinyl residue that marks the extreme C terminus of subsequently selected with G418 (Life Technologies, Inc.) as described the hemopexin domain. B, Western blot analysis of DMT-MMP-1-trans- previously (8, 20). fected cells. Serum-free conditioned medium from COS-7 cells tran- Plasmids—Following screening of an HT-1080 cDNA library con- siently transfected with control (lane 1), MT-MMP-1 (lane 2), MT- DA structed in the pBK-CMV vector with a 0.45-kilobase canine MT- MMP-1 (lane 3), or wild-type MT-MMP-1 (lane 4) expression vectors DB MMP-1 cDNA isolated from an MDCK cDNA library, a human clone were analyzed by immunoblotting with MT-MMP-1-specific polyclonal antisera. The dark and clear arrowheads indicate the positions of the was obtained and sequenced. The predicted amino acid sequence of putative pro- and processed forms of DMT-MMP-1. In wild-type MT- full-length HT-1080 MT-MMP-1 was identical with that described by MMP-1-transfected cells, immunoblots of cell lysates identified the Okada et al. (13). To generate pro-MT-MMP-1 and pro-MT-MMP-1 , DA DB 63-kDa form of the enzyme as the major product (data not shown). C, a59 primer (ACCATGTCTCCCGCCCCAAGACCCTCCCGT) was paired gelatin zymography of MT-MMP-1-transfected cells. Serum-free condi- with the 39 primers, TTCAGCTCACCGCCCCGCCGCC or TTCAGAA- tioned medium from COS-7 cells transiently transfected with control GAAGAAGACTGCAAGGCC, respectively, in separate PCR reactions to (lane 1), MT-MMP-1 (lane 2), MT-MMP-1 (lane 3), or MT-MMP-1 DA DB generate cDNA fragments with the intended truncations at the C ter- (lane 4) expression vectors were depleted of endogenous progelatinase A minus (i.e. immediately prior to the TM domain as defined by Cao et al. by gelatin affinity chromatography and analyzed by zymography. In (9) at Ser or at the conserved Cys residue at position 508 (6, 7); see lane 5, the MT-MMP-1 zymogram was developed in the presence of DB Fig. 1A). The PCR fragments were then cloned into pCR3-Uni vector 1.0 mM BB-94. Identical results were obtained with b-casein or k-elastin (Invitrogen) and characterized by sequencing as described (8). Expres- zymograms (D. Pei and S. J. Weiss, unpublished observation). sion vectors for a -proteinase inhibitor (a PI), the Pittsburgh mutant of 1 1 a PI (a PI ), and a TM-deletion mutant of furin were provided by A. 1 1 Pitt SDS-PAGE, Western Blots, Zymography, and N-terminal Sequenc- Rehemtulla and R. Kaufman (University of Michigan). ing—Basic protocols for these techniques have been described (8, 20). Mutagenesis—Sequential PCR-based mutagenesis was performed as The MT-MMP antisera were raised in New Zealand rabbit using a described previously, and the mutagenized fragments were cloned into bacterially generated recombinant fusion protein between glutathione pCR3-Uni vector and characterized by sequencing (8). The mutagenic transferase and MT-MMP (20). 91–243 primers used are as follows: ATCAAGGCCAATGTTGCAAGGAAGC- Enzymic Reactions—Enzyme assays of MT-MMP-1 and MT-MMP- DA GCTACGCC for R108A, GCCAATGTTCGAAGGGCGCGCTACGCCA- 1 were performed in buffer A supplemented with 150 mM NaCl at DB TCCAG for K110A, AATGTTCGAAGGAAGGCCTACGCCATCCAG- 37 °C unless noted otherwise. Matrix substrates (devoid of contaminat- GGT for R111A, and GTTCGAAGGAAGC GCTTTGCCATCCAGGG- ing progelatinase A activity as determined by gelatin zymography; data TCTC for Y112F (bold nucleotides indicate the altered codons). MT- not shown) were obtained from Collaborative Research (type I, IV, and MMP-1 expression constructs as well as those harboring the above DB V collagens, fibronectin, laminin, vitronectin, and dermatan sulfate mutations (1 mg each) were transiently transfected into COS-7 cells (2 proteoglycan). Processing of pro-MT-MMP-1 by purified soluble furin DB 3 10 /ml) by LipofectAMINE treatment, and, after a 24-h incubation, (specific activity 500 units/mg; Ref. 18) was performed as described (8). an aliquot of the cell-free supernatant (0.02 ml) was analyzed by West- Recombinant progelatinase A (purified as described in Ref. 22) was a ern blotting as described below. gift from R. Fridman (Wayne State University, Detroit, MI), recombi- Protein Purification—MDCK cells stably expressing MT-MMP-1 DB nant human TIMP-2 was supplied by Amgen, and soluble human furin were incubated in Opti-MEM (Life Technologies Inc.) supplemented was provided by R. Fuller, University of Michigan. with the synthetic MMP inhibitor, BB-94 (0.5 mM, British Biotechnol- ogy; Ref. 21), to trap the active form of the proteinase as a reversible RESULTS enzyme-inhibitor complex (20). After 48 h, ;2 liters of conditioned Expression of DMT-MMP-1 Mutants and Detection of Enzy- media were collected, dialyzed against buffer A (50 mM Tris, pH 7.5, 5 mic Activity—COS-7 cells transiently transfected with either mM CaCl ), and then loaded onto a Q-Sepharose column (1.5 3 10 cm). MT-MMP-1 or MT-MMP-1 cDNA each secreted a pair of Bound material was eluted with a NaCl gradient (0 to 1 M), and DA DB fractions containing MT-MMP-1 (identified by Western blotting) were major and minor products that were specifically recognized by DB combined and dialyzed against buffer A. A heparin-Sepharose column polyclonal antibodies to a truncated form of the bacterially (1 3 10 cm) was then loaded with the dialyzed materials and developed expressed protein (Fig. 1B). While the molecular mass of the with a NaCl gradient (0 to 1 M). Positive fractions were pooled and minor secreted proteins (;64 kDa for MT-MMP-1 and ;60 DA passed through a gelatin-Sepharose column (1 3 5 cm) followed by gel kDa for MT-MMP-1 ; lanes 2 and 3, respectively) were con- DB filtration chromatography (Ultrogel ACA44, 1 3 150 cm) in the absence sistent with those of the predicted proforms of the metallopro- of BB-94 to regenerate the active proteinase. In selected experiments, MT-MMP-1 was purified from a batch culture of transiently trans- teinases, the major soluble species detected with either TM- DA fected MDCK cells as described above. deletion mutant was a fragment ;10 kDa smaller in size. The generation of the major and minor forms was not specific to D. Pei and S. J. Weiss, unpublished observation. COS-7 cells since a similar profile was generated with trans- Transmembrane-deletion Mutants of MT-MMP-1 9137 FIG.2. Purification of MT-MMP-1 . A and B, fractionation of MT-MMP-1 on Q-Sepharose and heparin-Sepharose, respectively. Condi- DB DB tioned media from MDCK cells stably transfected with MT-MMP-1 were loaded onto a Q-Sepharose column and eluted with a NaCl gradient. DB The protein content of each fraction was monitored at A (dark squares) while MT-MMP-1 content in each fraction was monitored by 280 DB immunoblot analysis and reported as the percent recovered relative to the fraction containing the highest concentration of MT-MMP-1 (open DB squares). Fractions 9–16 were combined, dialyzed against buffer A, loaded onto a heparin-Sepharose column, and eluted with a NaCl gradient. C, characterization of the isolated MT-MMP-1 products. Conditioned media (lane 1), a pool of fractions 9–16 eluted from Q-Sepharose (lane 2), DB flow-through of fractions 9–16 that did not bind to heparin-Sepharose (lane 3), pool of fractions 10–15 eluted from heparin-Sepharose column (lane 4), and final purified form of MT-MMP-1 (4.5 pmol; lane 5) were separated by SDS-PAGE and visualized by Coomassie staining. In lanes 6 and DB 7, purified MT-MMP-1 (1.2 pmol) was analyzed by immunoblotting and gelatin zymography, respectively. D, the N terminus of MT-MMP-1 as DB DB determined after 10 cycles of sequencing (indicated by bold letters). The open box represents the MT-MMP-1 open reading frame with the amino DB 91 123 acid sequence of MT-MMP-1 from Pro to Glu . fected MDCK cells. As expected, when cells were transiently MMP-1 were subjected to a combination of gelatin-Sepharose DB transfected with wild-type MT-MMP-1, soluble forms of the affinity, Q-Sepharose affinity, and heparin-Sepharose affinity enzyme were not detected in the conditioned media (Fig. 1B). chromatography followed by gel filtration chromatography. Uti- In intact cell systems, MMPs can be recovered in conditioned lizing this protocol, a single immunoreactive ;50-kDa species medium as a mixture of zymogens, processed active enzymes, was isolated that co-migrated with the band of activity detected zymogen-inhibitor complexes, or enzyme-inhibitor complexes by gelatin-zymography (Fig. 2C). Following 10 cycles of N-termi- (1–3). In the case of almost all MMP family members, many of nal sequence analysis, the material was identified as MT-MMP-1 these forms can be detected following SDS-PAGE in substrate- with a single start site at Tyr (Fig. 2D). Interestingly, this N impregnated gels (1–3). Thus, serum-free conditioned media terminus not only aligns with that of the active forms of all other from control, MT-MMP-1 , MT-MMP-1 , or MT-MMP-1 MMPs (6), but it is also identical with that reported for the DA DB transfected cells were depleted of endogenous gelatinases by mature form of wild-type MT-MMP-1 recovered from the HT- gelatin affinity chromatography (DMT-MMP-1 does not bind to 1080 fibrosarcoma cell line (10). Thus, while almost all MMPs are gelatin; see below) and electrophoresed in gelatin- containing synthesized and secreted as inactive zymogens, MT-MMP-1 DB gels. Proteinases were then allowed to renature following the (as well as MT-MMP-1 ; see below) underwent further process- DA removal of SDS and then incubated overnight at 37 °C. As ing to its active form. shown in Fig. 1C, supernatants recovered from MT-MMP-1 - Like stromelysin-3, the only other membrane of the MMP DA or MT-MMP-1 -transfected cells each revealed the presence of family to be secreted as a fully processed active proteinase, DB a single band of gelatinolytic activity whose relative mobility MT-MMP-1 contains a motif of basic amino acids (i.e. RRKR) matched that of the major form of DMT-MMP-1 detected by immediately upstream of its catalytic domain (see Fig. 3A; Western blotting. Significantly, identical results were obtained Refs. 4 and 8). Recently, the RXKR array in stromelysin-3 (X 5 when zymograms were performed with either b-casein- or nonbasic amino acid) was shown to act as an endoproteolytic k-elastin-impregnated gels as well (data not shown). Regard- processing signal for an intracellular serine proteinase belong- less of substrate used, the band of proteolytic activity attrib- ing to the proprotein convertase family (8). Because specific uted to either DMT-MMP-1 mutant was completely inhibited proprotein convertases can display varying requirements for when zymograms were performed in the presence of the MMP- basic residues at positions 21, 22, and 24 relative to the 21 22 24 specific inhibitor, BB-94 (Fig. 1C). scissile bond (i.e. P ,P , and P , respectively; Refs. 23 and Purification of DMT-MMP-1 and Characterization of Zymogen 24), a potential role for this enzyme class in DMT-MMP-1 Processing—Because both MT-MMP-1 and MT-MMP-1 ap- processing was initially assessed by successively substituting DA DB peared to undergo a similar, if not identical, processing event to each basic residue with an Ala moiety in transient transfection generate active proteinases as assessed by zymography, one of assays. As shown in Fig. 3B, each of these substitutions almost the mutants (i.e. MT-MMP-1 ) was stably expressed in the completely blocked MT-MMP-1 processing (lanes 2–4). In DB DB MDCK cell line for further analyses. As shown in Fig. 2, serum- contrast, a Tyr 3 Ala substitution at the less critical P site free conditioned media from stable transfectants expressing MT- (23) did not affect MT-MMP-1 processing (Fig. 3, lane 5). DB 9138 Transmembrane-deletion Mutants of MT-MMP-1 FIG.4. DMT-MMP-1-dependent activation of recombinant progelatinase A. A, zymography and N-terminal sequence analysis of progelatinase A. Recombinant progelatinase A (140 nM) was incubated alone (lane 1) or with 14 nM,28nM,or56nM MT-MMP-1 (lanes 2–4, DB FIG.3. Proprotein convertase-dependent processing of MT- respectively) for2hat37 °Cin buffer A supplemented with 150 mM MMP-1 . A, mutational analysis of the putative proprotein convertase DB NaCl and 0.01% Brij 35 in a final volume of 0.02 ml. In lanes 5–7, recognition motif. The amino acids surrounding the cleavage site in progelatinase was incubated alone, with MT-MMP-1 or with MT- MT-MMP-1 are shown with the basic residue motif underlined. B, DB DB 108 MMP-1 and TIMP-2 (350 nM), respectively, for 16 h at 37 °C. Aliquots DB expression vectors for native MT-MMP-1 (lane 1), Arg 3 Ala DB 110 111 of each reaction mixture were analyzed by gelatin zymography or were (R108A; lane 2), Lys 3 Ala (K110A; lane 3), Arg 3 Ala (R111A; 112 electrophoresed and electroblotted for N-terminal sequence analysis. lane 4), and Tyr 3 Ala (Y112A; lane 5) were transfected into COS-7 The bold sequence beginning with Leu represents the first 10 cycles of cells (2 3 10 /ml) by LipofectAMINE treatment, and the secreted prod- the N terminus of the ;64-kDa form of gelatinase A (indicated by ucts (0.02 ml of the cell-free supernatant) were analyzed by immuno- asterisk in lane 4) while the sequence beginning with Tyr represents blotting. C, inhibition of MT-MMP-1 processing by a PI . COS-7 DB 1 Pitt the first 5 cycles of the N terminus of the ;62-kDa form of gelatinase cells were transiently transfected as described above with MT-MMP- (indicated by dark arrow in lane 7). The faint bands of gelatinolytic 1 expression vector alone (lane 1) or co-transfected with MT-MMP- DB activity detected at ;50 kDa are due to MT-MMP-1 (lanes 2–4). In DB 1 and a PI (lane 2)or a PI expression vectors (lane 3), and the DB 1 1 Pitt lanes 8–10, progelatinase A (140 nM) was incubated alone (lane 8) with cell-free supernatants were analyzed by immunoblotting. MT-MMP-1 (14 nM; lane 9) or MT-MMP-1 (14 nM; lane 10)for3h DB DA in 20 mM Hepes/KOH (pH 7.5), 0.1 mM CaCl , and 0.02% Brij-35 as described (10). The asterisk, arrowhead, and circle indicate the posi- Given that COS are known to express only two members of the tions of the ;64-kDa, ;62-kDa, and ;42-kDa forms of gelatinase A, proprotein convertase family that recognize RXKR motifs (i.e. respectively. B, furin-mediated processing of pro-MT-MMP-1 . Pro- DB furin and PACE4; Ref. 24), cells were co-transfected with MT- MT-MMP-1 (100 nM) isolated from a PI co-transfected cells was DB 1 Pitt MMP-1 and the Pittsburgh mutant of a PI (a PI ), a reac- incubated alone (lane 1) or with soluble furin (10 nM; lane 2)for1hat DB 1 1 Pitt tive site variant that inhibits furin (but not PACE4) activity in 37 °C in a final volume of 0.02 ml and analyzed by Western blotting. For zymography (lanes 3–5), progelatinase A (140 nM) was incubated alone situ (25, 26). Under these conditions, a PI completely 1 Pitt (lane 3), with pro-MT-MMP-1 (28 nM; lane 4), or with furin-processed DB blocked MT-MMP-1 processing while wild-type a PI exerted DB 1 MT-MMP-1 (28 nM; lane 5) for 16 h at 37 °C. Furin alone did not DB no inhibitory effect (Fig. 3C). MT-MMP-1 processing was DB activate progelatinase A. similarly inhibited by a PI in MDCK cells (data not shown). 1 Pitt These results (together with the demonstration that soluble completely blocked by the addition of the MMP inhibitor, furin processes pro-DMT-MMP-1 to its active form under cell- TIMP-2 (Fig. 4A) or BB-94 (data not shown). Interestingly, free conditions; see below) indicate that pro-DMT-MMP-1 mat- while earlier studies have demonstrated that membrane-asso- uration is regulated by a furin-dependent pathway in an intact ciated forms of MT-MMP-1 can also process progelatinase A cell system. into a 42-kDa active species (10, 27, 28), this product was not Activation of Recombinant Progelatinase A by Purified DMT- detected under the standard conditions employed. However, MMP-1—Current evidence indicates that MT-MMP-1-express- when either MT-MMP-1 or MT-MMP-1 were incubated DA DB ing cells initiate progelatinase A activation via a two-step proc- with progelatinase A in a low ionic strength buffer identical ess that involves an initial cleavage of the Asn -Leu bond with that used previously (10, 27, 28), MT-MMP-1 -depend- DA/B followed by an autocatalytic conversion of the Leu interme- ent progelatinase A activation was significantly accelerated diate into a 62-kDa active enzyme with an N-terminal Tyr and the 42-kDa form of gelatinase generated (Fig. 4A, lanes residue (10, 27). Nonetheless, the ability of MT-MMP to di- 8–10). The ability of mature MT-MMP-1 to mediate progelati- rectly cleave progelatinase A is unclear, and it has been pos- DA/B tulated that additional intermediates may be involved (9, 10). nase A activation is consistent with a model wherein active Thus, purified active MT-MMP-1 was incubated with recom- MT-MMP-1 directly cleaves the gelatinase zymogen, but DB DA/B binant progelatinase A and processing monitored by gelatin the data do not rule out the possibility that the DMT-MMP-1 zymography and N-terminal sequence analysis. Following a zymogen only induces progelatinase A to undergo autocatalytic 2-h incubation at 37 °C, MT-MMP-1 initially cleaved proge- processing to its active form. Hence, purified pro-DMT-MMP-1 DB latinase A (which migrates as a ;68-kDa species) into a ;64- was isolated (i.e. from cells co-transfected with MT-MMP-1 DB kDa fragment (Fig. 4A). N-terminal sequence analysis of the and a PI ), and its ability to mediate progelatinase A activa- 1 Pitt ;64-kDa gelatinase A fragment yielded the Leu form of the tion was examined. As shown in Fig. 4B, pro-MT-MMP-1 was DB enzyme (Fig. 4A). Subsequently, the 64-kDa form of the enzyme unable to stimulate progelatinase A activation. However, when underwent further processing to a 62-kDa product whose N pro-MT-MMP-1 was processed to its active form ex situ with DB terminus confirmed the generation of [Tyr ]gelatinase A (Fig. a TM-deleted soluble form of furin (Fig. 4B, lane 2) and then 4A). As expected, the ability of MT-MMP-1 (as well as MT- incubated with progelatinase A, the gelatinase zymogen was DB MMP-1 ; data not shown) to activate progelatinase A was readily activated (lane 5). Thus, only the processed active form DA Transmembrane-deletion Mutants of MT-MMP-1 9139 generated by exchanging the TM and cytosolic domains of the metalloproteinase with those of the IL-2 receptor functioned normally in terms of its ability to mediate progelatinase A activation (9). Although this result is consistent with our con- clusion that the extracellular domain of MT-MMP-1 confers the proteinase with its distinct characteristics, these authors also reported that a TM-deletion mutant encoding residues 1–535 of MT-MMP-1 was unable to process progelatinase A (9). In com- paring our experimental approaches, it is important to note that the soluble MT-MMP-1 generated in their study was not isolated nor were its interactions with progelatinase A exam- FIG.5. Substrate specificity of MT-MMP-1 . Type I collagen (3 DB ined directly (9). Instead, Cao et al. (9) judged their TM-dele- mg; lane 1), type IV collagen (2 mg; lane 3), and type V collagen (2 mg; lane 5) were incubated alone or with 45 nM MT-MMP-1 (lanes 2, 4, tion mutant to be inactive on the basis of its inability to process DB and 6, respectively) at 25 °C for 16 h. Type I gelatin (3 mg; lane 7), endogenously derived progelatinase A secreted by COS-1 cells fibronectin (4 mg; lane 10), laminin (4 mg; lane 13), vitronectin (4 mg; in a transient transfection assay system (9). Under these con- lane 16), or dermatan sulfate proteoglycan (5 mg; lane 19) were incu- ditions, however, attempts to assess the activity of secreted bated alone, with MT-MMP-1 (45 nM; lanes 8, 11, 14, 17, and 20, DB MT-MMP-1 would be complicated by the presence of cell-de- respectively) or with MT-MMP-1 and 1 mM BB-94 (lanes 9, 12, 15, 18, DB and 21, respectively) at 37 °C in a final volume of 0.025 ml for 16 h. rived TIMPs which can interfere with progelatinase A process- Reaction mixtures were separated by SDS-PAGE and Coomassie- ing by either inhibiting DMT-MMP-1 activity directly, or, in the stained. The arrowhead by lane 15 indicates the position of the laminin case of TIMP-2, by binding to the C-terminal domain of the B chain, while the hatch marks at the margin of lanes 16 and 19 gelatinase zymogen (4, 27, 28). Indeed, when endogenous levels indicate the positions of the 97-, 69-, 45-, 32-, and 28-kDa molecular mass markers, respectively. Dermatan sulfate proteoglycan is recorded of TIMP are overwhelmed by co-transfecting COS cells with as DSPG. MT-MMP-1 and progelatinase A, gelatinase activation can DA/B be readily detected in the intact cell system as well as our of DMT-MMP-1 is able to mediate progelatinase A activation. purified system. Thus, while anchoring a proteinase to the cell ECM-degrading Activity of DMT-MMP-1—While previous membrane might be predicted to more effectively shield an attention has focused solely on the role of MT-MMP-1 in proge- active proteinase from soluble inhibitors (and to perhaps pro- latinase A activation (4, 9, 10, 13, 27), the ability of DMT- vide a surface more conducive for accelerating processing MMP-1 to degrade gelatin, b-casein, or k-elastin following zy- events) (e.g. Refs. 29–31), our data demonstrate that the TM- mography suggested that the proteinase might express activity deletion mutants retain the key functional properties of the against a wider range of targets. Thus, purified MT-MMP-1 wild-type enzyme. DA/B was incubated with either basement membrane- or intersti- As a consequence of our attempts to characterize the activity of DMT-MMP-1, we also discovered that the TM-deletion mu- tium-associated ECM molecules, and proteolysis was assessed in the absence or presence of BB-94. As shown in Fig. 5, while tants are capable of undergoing rapid processing to their ma- ture forms. This finding is noteworthy since, as a general rule, MT-MMP-1 did not degrade native type I, IV, or V collagens, DB the enzyme readily proteolyzed gelatin as well as fibronectin, MMPs are synthesized and secreted as inactive zymogens (1–3, 7). However, we recently reported that in a fashion similar to the B chain of laminin, vitronectin, and dermatan sulfate pro- teoglycan via a BB-94-sensitive process. Similar, if not identi- that observed for the DMT-MMP-1 mutants, prostromelysin-3 is secreted as an active enzyme following its intracellular proc- cal, results were obtained with purified MT-MMP-1 (data not DA shown). Given that none of these substrates were contaminated essing within the constituitive secretory pathway (8). In this with detectable quantities of progelatinase A (see “Experimen- case, activation was dependent upon a decapeptide insert that tal Procedures”), we conclude that DMT-MMP-1 mutants can is sandwiched between the pro- and catalytic domains of express intrinsic matrix-degrading activities. stromelysin-3 and encrypted with an extended furin recogni- tion motif (i.e. RXRXKR). At the time that these earlier studies DISCUSSION were completed, stromelysin-3 was the only member of the Sequence alignments of the 13 human MMPs that have been MMPs family known to contain this recognition sequence. characterized to date indicate that amino acids 1–508 of the However, with the recent cloning of MT-MMP-1, -2, and -3, it is 538-residue-long extracellular domain of MT-MMP-1 contain clear that all three of these enzymes contain homologous in- all of the major structural elements of the secreted members of serts which include an array of basic residues (i.e. RRKR) that this gene family (i.e. a propeptide and catalytic domain as well match the recognition motif of the proprotein convertases (i.e. as a hemopexin-like region that is bounded by a pair of highly RX(K/R)-R; Refs. 4–6). Consistent with this prediction, (i) the conserved cysteinyl residues; Refs. 4–7). Given that the C N terminus of DMT-MMP-1 was located at Tyr on the C- termini of virtually all secreted MMPs end at, or extend no terminal side of the Arg-Arg-Lys-Arg motif, (ii) DMT-MMP-1 more than 8 amino acids beyond, the final cysteinyl residue in processing could be inhibited by either inserting point muta- the hemopexin domain (7), we reasoned that TM-deletion mu- tions in the RRKR motif or by co-transfecting cells with the tants of MT-MMP-1 that retained this modular organization furin-specific inhibitor, a PI , and (iii) the DMT-MMP-1 zy- 1 Pitt would encode functional proteinases. Indeed, as demonstrated, mogen could be processed to its active form ex situ by soluble regardless of whether soluble mutants of MT-MMP-1 were furin. While we have not yet identified the intracellular/extra- truncated either at the edge of the TM domain or at the end of cellular compartments in which the DMT-MMP-1 zymogen un- the hemopexin domain, the expressed proteins displayed sim- dergoes processing in the intact cell, furin is a membrane- ilar, if not identical, activities as assessed by zymography, associated endoprotease that not only cycles between the trans- progelatinase A processing, or substrate specificity. By itself, our work does not rule out the possibility that the Although stromelysin-3 and MT-MMP-1 both contain proprotein TM or cytosolic domains of MT-MMP-1 convey additional struc- convertase recognition motifs, comparisons of their genomic organiza- tural information to the processed proteinase (i.e. beyond act- tion and chromosomal localization indicate that the two metalloprotein- ing as a membrane anchor). However, while our work was in ases are not closely related and belong to separate branches of the progress, Cao et al. (9) reported that an MT-MMP-1 chimera phylogenetic tree (D. Pei and S. J. Weiss, unpublished observation). 9140 Transmembrane-deletion Mutants of MT-MMP-1 Golgi network and the cell surface, but also undergoes been heightened by recent findings which suggest that soluble processing to a soluble form that accumulates extracellularly forms of MT-MMP-1 may be generated in vivo (13). Thus, while (32–34). Indeed, the possibility that DMT-MMP-1 may undergo the MT-MMP-1 antigen has been immunodetected on the sur- extracellular processing is further supported by our results face of cancer cells in vivo (4), RNA in situ hybridization studies with the TM-deleted form of soluble furin. Nonetheless, in spite have more recently demonstrated that MT-MMP-1 transcripts of the fact that furin is the most credible MT-MMP-1 activator are confined to the surrounding stromal cells (13). Should MT- identified to date, caution should be exercised in terms of MMP-1 undergo solubilization and intercellular transfer in situ extrapolating processing pathways that are operative for DMT- (4, 13), tumor cells could potentially use the stroma-derived MMP-1 to the wild-type enzyme. Indeed, in contrast to the enzyme to assemble a multicatalytic complex on their surface results obtained with DMT-MMP-1, we and others have found that would not only arm them with the ability to catalyze that COS cells transfected with wild-type MT-MMP-1 route progelatinase A activation, but also to express an additional most of the enzyme to the cell surface as the unprocessed repertoire of proteolytic activities. Additional studies will be 3,5 zymogen rather than the mature enzyme (4, 9, 31). Utilizing required to directly compare the soluble and membrane-an- chimeric constructs between stromelysin-3 and wild-type MT- chored forms of MT-MMP-1, but the established catalytic ac- MMP-1, it appears that while the furin recognition motif in tivity of the TM-deletion mutants should provide a useful tool either of the secreted metalloproteinases can be processed ef- for characterizing the enzymic properties of this new family of fectively, the TM domain of MT-MMP-1 appears to “shield” the membrane-anchored MMPs. recipient proteinase from undergoing rapid intracellular proc- Acknowledgments—We thank R. Fuller (University of Michigan) for essing. The mechanisms responsible for controlling the intra- assistance in purifying soluble furin, A. Galloway (British Biotechnol- cellular and extracellular processing of wild-type MT-MMP-1 ogy) for BB-94, and K. Langley (Amgen) for recombinant TIMP-2. require further analysis, but the fact that the active form of the full-length (10) and mutant enzyme display an identical N REFERENCES terminus directly downstream of the proprotein convertase- 1. Woessner, J. F., Jr. (1991) FASEB J. 5, 2145–21549 2. Matrisian, L. M. (1992) BioEssays 14, 455–463 recognition motif strongly suggests a role for furin or, perhaps, 3. Stetler-Stevenson, W. G., Aznavoorian, S., and Liotta, L. A. (1993) Annu. Rev. a related proprotein convertase (e.g. PC6; Refs. 24 and 35) in Cell Biol. 9, 541–573 zymogen activation. 4. Sato, H., Takino, T., Okada, Y., Cao, J., Shinagawa, A., Yamamoto, E., and Seiki, M. (1994) Nature 370, 61–65 In the presence of purified active DMT-MMP-1 (but not its 5. Will, H., and Hinzmann, B. (1995) Eur. J. Biochem. 231, 602–608 zymogen), progelatinase A was processed to its mature form 6. Takino, T., Sato, H., Shinagawa, A., and Seiki, M. (1995) J. Biol. Chem. 270, 81 38 23013–23020 (i.e. [Tyr ]gelatinase) via the formation of the Leu interme- 7. Birkedal-Hansen, H., Moore, W. G. I., Bodden, M. K., Windsor, L. J., Birkedal- diate. This two-step, TIMP-2-sensitive activation cascade is Hansen, B., DeCarlo, A., and Engler, J. A. (1993) Crit. Rev. Oral Biol. Med. identical with that previously established for crude prepara- 4, 197–250 8. Pei, D., and Weiss, S. J. (1995) Nature 375, 244–247 tions of plasma membrane-associated MT-MMP-1 (27) and al- 9. Cao, J., Sato, H., Takino, T., and Seiki, M. (1995) J. Biol. Chem. 270, 801–805 lows us to conclude that DMT-MMP-1 can initiate the process- 10. Strongin, A. Y., Collier, I., Bannikov, G., Marmer, B. L., Grants, G. A., and Goldberg, G. I. (1995) J. Biol. Chem. 270, 5331–5338 ing event independently of additional co-factors or substrates. 11. Brown, P. D., Bloxidge, R. E., Anderson, E., and Howell, A. (1993) Clin. Exp. Interestingly, the ability of DMT-MMP-1 to directly activate Metastasis 11, 183–189 progelatinase A under cell-free conditions contrasts with a 12. Brown, P. D., Bloxidge, R. E., Stuart, N. S. A., Gatter, K. C., and Carmichael, J. (1993) J. Natl. Cancer Inst. 85, 574–578 recent report by Strongin et al. (10) wherein an MT-MMP- 13. Okada, A., Bellocq, J.-P., Rouyer, N. Chenard, M.-P., Rio, M.-C., Chambon, P., 1zTIMP-2 complex (rather than MT-MMP-1 alone) was pro- and Basset, P. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 2730–2734 posed to function as the membrane-associated activator of the 14. Tournier, J. M., Polette, M., Hinnrasky, J., Beck, J., Werb, Z., and Basbaum, C. (1994) J. Biol. Chem. 269, 25454–25464 gelatinase zymogen. We were unable to reproduce this finding 15. Aimes, R. T., and Quigley, J. P. (1995) J. Biol. Chem. 270, 5872–5876 with MT-MMP-1 , but cannot rule out the possibility that 16. Vassalli, J.-D., and Pepper, M. S. (1994) Nature 370, 14–15 DA/B 17. Brenner, C., and Fuller, R. S. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 922–926 TIMP-2 plays a more complex role on the membrane surface. 18. Bravo, D. A., Gleason, J. B., Sanchez, R. I., Roth, R. A., and Fuller, R. S. (1994) However, an interpretation of the data presented by Strongin et J. Biol. Chem. 269, 25830–25837 19. de Vries, T., Srnka, C. A., Palcic, M. M., Swiedler, S. J., van den Eijnden, D. H., al. (10) is complicated by the fact that even in the apparent and Macher, B. A. (1995) J. Biol. Chem. 270, 8712–8722 absence of TIMP-2, MT-MMP-1 continued to process progela- 20. Pei, D., Majmuder, G., and Weiss, S. J. (1994) J. Biol. Chem. 269, tinase A to the [Leu ]gelatinase intermediate, but not the final 25849–25855 21. Davies, B., Brown, P. Q., East, N., Crimmin, M. J., and Balkwill, E. R. (1993) mature form. Thus, it remains possible that TIMP-2 exerts its Cancer Res. 53, 2087–2091 stimulatory effect by accelerating the inter- or intramolecular 22. Fridman, R., Fuerst, T. R., Bird, R. E., Hoyhtya, M., Oelkuct, M., Kraus, S., 38 81 autocatalytic conversion of [Leu ]gelatinase to [Tyr ]gela- Komarek, D., Liotta, L., Berman, M. L., and Stetler-Stevenson, W. G. (1992) J. Biol. Chem. 267, 15398–15405 tinase on the cell surface (31) rather than by stimulating MT- 23. Watanabe, T., Murakami, K., and Nakayama, K. (1993) FEBS Lett. 320, MMP-1 activity directly. 215–218 24. Seidah, N. G., Chretien, M., and Day, R. (1994) Biochimie (Paris) 76, 197–209 To date, the only function ascribed to MT-MMP-1 has been 25. Rehemtulla, A., and Kaufman, R. J. (1992) Blood 79, 2349–2355 its ability to activate progelatinase A (4, 9, 10, 13, 27, 28). 26. Wasley, C. L., Rehemtulla, A., Bristol, J. A., and Kaufman, R. J. (1993) J. Biol. However, we have demonstrated that purified MT-MMP-1 Chem. 268, 8458–8465 DA/B 27. Strongin, A. Y., Marmer, B. L., Grant, G. A., and Goldberg, G. I. (1993) J. Biol. can also degrade a number of extracellular matrix components. Chem. 268, 14033–14039 These data indicate that the ability of MT-MMP-1-transfected 28. Ward, R. V., Atkinson, S. J., Reynolds, J. J., and Murphy, G. (1994) Biochem. J. 304, 263–269 cells to express a heightened invasive potential may not nec- 29. Bangalore, N., and Travis, J. (1994) Biol. Chem. Hoppe-Seyler. 375, 659–666 essarily be linked to progelatinase A activation alone (4). Al- 30. Young, T. N., Pizzo, S. V., and Stack, M. S. (1995) J. Biol. Chem. 270, though our studies have employed a soluble form of MT- 999–1002 31. Atkinson, S. J., Crabbe, T., Cowell, S., Ward, R. V., Butler, M. J., Sato, H., MMP-1, we believe that the modular organization of MT- Seiki, M., Reynolds, J. J., and Murphy, G. (1995) J. Biol. Chem. 270, MMP-1 strengthens the likelihood that the membrane-tethered 30479–30485 32. Vidricaire, G., Denault, J.-B., and Leduc, R. (1993) Biochem. Biophys. Res. form displays a similar substrate specificity. Furthermore, the Commun. 195, 1011–1018 potential physiologic relevance of DMT-MMP-1 mutants have 33. Molloy, S. S., Thomas, L., VanSlyke, J. K., Stenberg, P. E., and Thomas, G. (1994) EMBO J. 13, 18–33 34. Vey, M., Schafer, W., Berghofer, B., Klenk, H.-D., and Garten, W. (1994) J. Cell However, a portion of the MT-MMP-1 zymogen does undergo proc- Biol. 127, 1829–1842 essing since the mature enzyme has been isolated from HT-1080 35. Horimoto, T., Nakayama, K., Smeekens, S. P., and Kawaoka, Y. (1994) J. Virol. plasma membranes (10) and is required for progelatinase activation. 68, 6074–6078

Journal

Journal of Biological Chemistry – Unpaywall

Published: Apr 1, 1996

There are no references for this article.