cDNA cloning and expression of the gene encoding murine stromelysin-2 (MMP-10)

Gene 202 (1997) 75–81

cDNA cloning and expression of the gene encoding murine stromelysin-2 (MMP-10) Marianne Madlener, Sabine Werner * Max-Planck-Institut fu¨r Biochemie, Am Klopferspitz 18a, D-82152 Martinsried, Germany Received 19 May 1997; accepted 21 July 1997; Received by E. Boncinelli

Abstract Recently, we demonstrated a biphasic induction of the epithelial broad-spectrum matrix metalloproteinase (MMP) stromelysin-2 during cutaneous wound healing. Now we have generated a murine wound cDNA libary and have used it to isolate the putative cDNA of this murine matrix metalloproteinase. The predicted sequence of the protein shows 76 and 89% identity with its human and rat analogues, respectively. Stromelysin-2 and stromelysin-1 transcripts were both detected at very low levels in the lung and the heart of adult Balb/c mice, whereas stromelysin-2 mRNA expression alone was found at comparatively high levels in the small intestine, a tissue characterized by continuous epithelial renewal. Recombinant forms of murine stromelysin-1 and -2 produced in transfected COS cells were secreted and could be induced to undergo autocatalytic processing by addition of the organomercurial salt 4-aminophenylmercuric acetate (APMA). © 1997 Elsevier Science B.V. Keywords: Wound healing; Matrix metalloproteinase; cDNA library; APMA activation

1. Introduction The matrix metalloproteinases (MMPs) comprise a multi-gene family of proteolytic enzymes with the combined capacity of degrading virtually all components of the extracellular matrix and of basement membranes. They are generally considered to play key roles in extracellular remodeling and degradation processes under both physiological and pathological conditions. According to sequence homologies and related substrate specificities, the MMPs are generally divided into three major classes: the collagenases, gelatinases, and stromelysins (reviewed by Woessner, 1991; Matrisian, 1992; Birkedal-Hansen et al., 1993). Among these, the stromelysins are characterized by extremely broad substrate specificities. Stromelysin-1 (MMP-3) and stromelysin-2 * Corresponding author. Tel: 49 89 8578 2269/71; Fax: 49 89 8578 2814; e-mail: [email protected] Abbreviations: aa, amino acid; APMA, 4-aminophenylmercuric acetate; bp, base pair(s); cDNA, DNA complementary to mRNA; COS cells, SV40-transformed monkey cells; cpm, counts per minute; EDTA, ethylenediaminetetraacetic acid; IVH, influenza virus hemagglutinin; kDa, kilodalton; MMP, matrix metalloproteinase; nt, nucleotide(s); ORF, open reading frame; PCR, polymerase chain reaction; SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; UTR, untranslated region. 0378-1119/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S 03 7 8 -1 1 1 9 ( 9 7 ) 0 0 4 56 - 3

(MMP-10) have been shown to degrade fibronectin, laminin, elastin, proteoglycan core protein, gelatins, and collagen types III, IV, V, IX and X (Nagase et al., 1991). A putative matrix metalloproteinase originally identified in human invasive breast carcinomas was assigned to the stromelysins on behalf of sequence similarities (stromelysin-3, MMP-11; Basset et al., 1990). Its functional properties, however, differ so significantly from those of stromelysin-1 and -2 that an as yet unidentified physiological role other than that of a matrix-degrading enzyme seems likely (Noe¨l et al., 1995; Pei et al., 1995; Santavicca et al., 1996). Stromelysin-1 and -2, notwithstanding their overlapping substrate spectra, are anything else than abundant homologues of each other. Both in vitro and in vivo studies have demonstrated a cell-type specific expression and differential regulation of these proteolytic enzymes. Following stimulation of human keratinocytes and mucosal fibroblasts by phorbol esters and cytokines, stromelysin-1 transcripts were detected exclusively in the fibroblasts, whereas the expression of stromelysin-2 under identical conditions was restricted to the keratinocytes ( Windsor et al., 1993). Analysis of MMP expression in acute and chronic wounds has demonstrated a similar cell type specificity in vivo (Saarialho-Kere et al., 1994; Madlener et al., 1996, and unpublished observations).

76

M. Madlener, S. Werner / Gene 202 (1997) 75–81

So far, only the sequences of murine stromelysin-1 and -3 have been published (Hammani et al., 1992; Lefebvre et al., 1992). Therefore, regarding the unique properties of stromelysin-2 as an epithelial cell-specific broad-spectrum metalloproteinase and the importance of murine experimental models for further elucidation of MMP activity in normal and pathological states, we attempted to identify the missing murine stromelysin-2 cDNA. Recent data from our laboratory have demonstrated a strong biphasic induction of stromelysin-2 mRNA following skin injury in Balb/c mice (Madlener et al., 1996). We have, therefore, generated a cDNA library from murine full-thickness excisional skin wounds and screened it with a specific stromelysin-2 probe under high stringency conditions. In this study, we present the complete cDNA-sequence of murine stromelysin-2 and demonstrate that it codes for a secreted protein of approx. 52 kDa, which undergoes autoproteolytic cleavage after addition of aminophenylmercuric acetate (APMA).

2. Materials and methods 2.1. Preparation of wound tissue Female Balb/c mice (12 weeks of age) were anesthetized with a single intraperitoneal injection of Avertin. Six full-thickness wounds (6 mm in diameter, 3–4 mm apart) were made by excising the skin and panniculus carnosus. The wounds were allowed to dry in order to form a scab. The animals were sacrificed at 1 and 5 days after wounding, respectively, and the wound tissue was harvested by excising the complete wounded area, i.e. including 2 mm of the epithelial margins. Wound tissue was immediately frozen in liquid nitrogen and stored at −80°C.

fragment of murine stromelysin-2 cDNA generated by PCR as described previously (Madlener et al., 1996). Approximately 2×105 phages were plated on six 12×12 cm bacterial dishes, and phage plaques were transferred to Gene screen plus nylon membranes (NEN, Boston, MA) using standard protocols (Sambrook et al., 1989). The murine ST-2 cDNA fragment was radiolabeled with [a-32P]-dCTP to a specific activity of approximately 1×109 cpm/mg using the Rediprime DNA labeling system (Amersham Life Science). Filters were prehybridized in a buffer containing 50% formamide, 2×SSC, 4 mM EDTA, 1% Sarcosyl, 0.1% SDS, and 5×Denhardt’s solution for 15 min, then transferred to fresh prehybridization buffer containing 3×106 cpm/ml labeled murine ST-2 probe, and incubated at 37°C overnight. They were then washed twice in 2×SSC at room temperature for 5 min and once in prehybridization buffer lacking Denhardt’s solution for 30 min at 37°C and subsequently exposed to a BioMax MR-1 X-ray film ( Kodak) for 48 h. Forty positive plaques with varying signal intensities were identified and recovered from the bacterial dishes. The 12 plaques yielding the strongest signals were then replated and a further round of hybridization was performed as described above. A single positive phage plaque for each of these 12 clones was used for an in vivo excision reaction of the pBluescript SK(-) phagemid from the Uni-ZAP vector (according to Stratagene). The rescued phagemids were then used to transform the E. coli strain SOLR and preparations of the pBluescript II KS(-) plasmids were used for subsequent restriction and sequence analysis. 2.4. Sequence analysis Sequence analysis was performed using the T7 sequencing kit (Pharmacia) and by sequence comparison in the GenBank nucleic acid sequence data library.

2.2. Preparation of mouse tissues Testis, heart, kidney, lung, liver, and large and small intestine were dissected from Balb/c mice, frozen in liquid nitrogen, and stored at −80°C until further use. 2.3. Isolation and sequencing of a cDNA clone encoding murine stromelysin-2 A l ZAP II cDNA library was constructed (Stratagene) according to the instructions supplied by the manufacturer using 6 mg poly(A)+-RNA from murine full-thickness skin wounds, 1 and 5 days after injury. For this purpose, total cellular RNA was isolated from wound tissue as described (Chomczynski and Sacchi, 1987) and affinity-purified using oligo(dT )-cellulose columns (Pharmacia). This cDNA library was screened under high stringency conditions with a 212-bp

2.5. Generation of murine stromelysin-1 and stromelysin-2 expression constructs The ORFs of murine stromelysin-1 and -2 were ligated into the eukaryotic expression vector pCG, thus generating cDNAs encoding stromelysin-1 and -2 derivatives with a 17 aa COOH-terminal influenza hemagglutinin( IVH-) tag under the control of the CMV promoter. Truncated forms of the ORFs ( lacking the stop codons) flanked by XbaI and SmaI restriction sites were amplified by PCR using the pBluescript KS(−) plasmids containing the full-length murine stromelysin-1 and -2 cDNAs, respectively, as templates and with the following primers (the underlined sequences represent XbaI and SmaI sites, respectively; the start codon appears in bold letters): mST1-pCG-5∞: ACGTCTAGAATGAAAATGAAGGGTCTT [the last 18 nucleotides corresponding to

M. Madlener, S. Werner / Gene 202 (1997) 75–81

positions 19–36 of the full-length murine stromelysin-1 cDNA ( EMBL Sequence Data Library, accession numbers X66402; S46808)]; mST1-pCG-3∞:TGTCCCGGGACAATTAAACCAGCTATT (the last 18 nucleotides corresponding to positions 1438–1455 of the full-length murine stromelysin-1 cDNA); mST2-pCG-5∞: GACTCTAGAGTCTATATGGAGCCACTA (the last 18 nucleotides corresponding to positions 36–53 of the fulllength murine stromelysin-2 cDNA reported here); mST2-pCG-3∞: TATCCCGGGGCACAGCAGCCAGCTGTT (the last 18 nucleotides corresponding to positions 1452–1469 of the full-length murine stromelysin-2 cDNA). Amplification was performed in a 50 ml reaction volume by 25 cycles (1 min at 94°C, 2 min at 41°C, 2 min at 75°C ) preceded by one extended cycle (5 min at 94°C, 5 min at 49°C, 3 min at 75°C ) during the elongation phase, of which 2.5 units of Pfu polymerase (Stratagene) were added to each reaction tube. The PCR products were subsequently digested with XbaI/SmaI, purified by gel electrophoresis on a 2% (w/v) NuSieve Agarose ( FMC BioProducts) gel and ligated into the XbaI/SmaI-linearized pCG expression vector. 2.6. Expression of IVH-tagged murine stromelysin-1 and -2 derivatives in COS cells COS cells were used to transiently express the IVHtagged derivatives of murine stromelysin-1 and stromelysin-2. They were seeded in 6-cm dishes in Dulbecco’s modified Eagle’s medium (DMEM ) with 10% (v/v) fetal calf serum. Cells were transiently transfected at 50–70% confluency with 5 mg plasmid DNA using a DEAE-dextrane transfection method. After 36–48 h of incubation the serum-free conditioned media were collected and centrifuged at 1000×g for 5 min. The supernatants were frozen in liquid nitrogen immediately and stored at −80°C until further use. For preparation of cell lysates, each dish was incubated with 300 ml of lysate buffer (20 mM Tris/HCl, pH 8.0; 137 mM NaCl; 2 mM EDTA; 10% glycerin; 1% Triton X-100) for 5 min at 4°C. Concentrated Laemmli sample buffer (Laemmli, 1970) was added to the samples to a final concentration of 1×. Samples were stored at −20°C until analyzed by SDS–PAGE.

77

body and detected with 4-nitro blue tetrazolium chloride and 5-bromo-4-chloro-3-indolyl phosphate (Promega). 2.8. Activation of pro-stromelysins by the organomercurial salt 4-aminophenylmercuric acetate (APMA) Samples of conditioned media were concentrated as described, brought to 20–50 mM Tris, pH 7.5, 5 mM CaCl , and incubated with 2 mM 4-aminophenyl2 mercuric acetate (APMA) for 18 h at 37°C. To demonstrate inhibition of APMA-activation by EDTA, samples were pre-incubated in the presence of 5 mM of this chelating agent. 2.9. RNase protection analysis Isolation of total cellular RNA and RNase protection analysis were performed as previously described ( Werner et al., 1992). Briefly, DNA probes were cloned into the transcription vector pBluescript II KS (+) and linearized. An anti-sense transcript was synthesized in vitro by using T3 or T7 RNA polymerases and [32P]-rUTP (800 Ci/mmol ). RNA samples were hybridized at 42°C overnight with 100 000 cpm of the labeled anti-sense transcript. Hybrids were digested with RNases A and T1 for 1 h at 30°C. Under these conditions, every single mismatch is recognized by the RNases. Thus crossreaction of the individual probes with the other stromelysins can be excluded. Protected fragments were separated on 5% (w/v) acrylamide/8 M urea gels and analyzed by autoradiography. Fifty micrograms of tRNA were used as a negative control. The same RNA preparations were used for all protection assays. The DNA templates used for RNase protection analysis were as follows: a 182 bp fragment corresponding to nucleotides 1284–1455 bp of murine stromelysin-1 and the 212 bp fragment described previously (Madlener et al., 1996) that was also used for screening the cDNA library (corresponding to nucleotides 138–349 of the murine stromelysin-2 cDNA sequence reported here).

3. Results and discussion

2.7. Identification of IVH-tagged proteins by Western blotting

3.1. Isolation and sequence of a cDNA clone encoding murine stromelysin-2

Serum-free conditioned media were concentrated 40-fold by ultrafiltration in Centricon tubes. Samples were separated on 15% (w/v) SDS–PAGE gels according to the methods of Laemmli (1970). Proteins were transferred to nitrocellulose paper (Schleicher and Schu¨ll ) by electroblotting and probed with a monoclonal IVH antipeptide antibody. Alkaline phosphatase-conjugated goat anti-mouse IgG (Promega) was used as secondary anti-

In a previous study we provided evidence for a strong, biphasic induction of stromelysin-2 expression during the healing of full-thickness excisional wounds in Balb/c mice and localized stromelysin-2 mRNA specifically to keratinocytes of the migrating epithelium at the wound edge. For this purpose, we had cloned a 212 bp fragment of murine stromelysin-2 by PCR which was subsequently used as a template for the generation of anti-sense

78

M. Madlener, S. Werner / Gene 202 (1997) 75–81

transcripts for both RNase protection analysis and in situ hybridization studies (Madlener et al., 1996). In order to isolate the full-length murine stromelysin-2 cDNA, we constructed a l ZAP II cDNA library using RNA from full-thickness excisional wounds of Balb/c mice and screened it with a radiolabeled probe corresponding to the 212 bp stromelysin-2 PCR fragment. Out of 2×105 phage plaques, 40 positive clones were detected of which 12 were isolated by in vivo excision. In spite of the high stringency conditions under which this experiment was performed, most of these clones contained stromelysin-1 cDNAs and only one of them was shown to contain a 1744 bp cDNA with the putative ORF for murine stromelysin-2 ( Fig. 1). This abundance of stromelysin-1 clones is not altogether surprising. First of all, the template of the radioprobe used for screening was isolated by homology cloning and was, therefore, likely to contain highly conserved regions. And indeed, our sequence analyses have verified this hypothesis. Additionally, RNase protection analysis of acute murine skin wounds using stromelysin-1 and -2 riboprobes in parallel have revealed a significantly higher level of stromelysin-1 transcripts compared with stromelysin-2 mRNA ( Fig. 2), thus further increasing the probability of cross-reaction of the stromelysin-2 probe with stromelysin-1 cDNA. By sequence analysis of this isolated stromelysin-2 cDNA with known cDNA sequences of stromelysin-1 and -2 in various species we deduced that the ATG codon at position 41 codes for the amino-terminal methionine and that the ORF ends with a TGA stop codon at nt position 1470. Furthermore, one putative polyadenylation signal was detected at nt 1699–1705 in the 272 nt 3∞ UTR ( Fig. 1). 3.2. Comparison of the isolated murine stromelysin-2 cDNA with other stromelysin ORFs from different species The amino acid sequence deduced from the murine stromelysin-2 cDNA ORF exhibits 76% and 89% identity with human (Muller et al., 1988) and rat (Breathnach et al., 1987) stromelysin-2, respectively (data not shown). Comparison with murine stromelysin-1 (Hammani et al., 1992) revealed 73% identity and 84% similarity at the amino acid level. Sequence homologies imply that murine stromelysin-2 possesses a hydrophobic 17 aa signal peptide like the other stromelysins-1 and -2 identified so far (Sang and Douglas, 1996). The secreted murine stromelysin-2 proenzyme is deduced to consist of 459 residues, 99 of which should constitute the propeptide. The molecular weights for the secreted latent and the activated forms of murine stromelysin-2 were calculated to 52.2 and 42.8 kDa, respectively. The carboxyterminal half of the propeptide and the adjacent 72 amino-terminal residues of the catalytic domain exhibit 100% sequence homology

Fig. 1. Nucleotide and deduced amino acid sequence of murine stromelysin-2 (GenBank/EMBL Database accession number Y13185). The start codon at nt position 42 and the stop codon at position 1470 were determined by sequence comparison with known stromelysin-1 and -2 sequences from various species. The deduced amino-termini of the secreted proenzyme and the activated form are marked by arrows. The highly conserved carboxy-terminal region of the propeptide, the conserved positions of the zinc-binding motif, and the putative polyadenylation signal at nt 1699–1705 are underlined.

M. Madlener, S. Werner / Gene 202 (1997) 75–81

79

Fig. 2. Tissue distribution of murine stromelysin-1 and -2 transcripts in Balb/c mice. Twenty micrograms of total cellular RNA isolated from various tissues were used to assess stromelysin-1 and -2 mRNA expression by RNase protection analysis under conditions where cross-reactions could be excluded. 1000 cpm of the hybridization probes were loaded in the lanes labeled ‘probe’ and used as sized markers. The varying signal intensities of these probes reflect the different relative transcription levels.

with murine stromelysin-1. Furthermore, a stretch of about 20 residues including the three histidine residues which have been implicated to participate in the coordination of the active site Zn2+ (Dhanaraj et al., 1996) is also completely conserved. Beyond this point, murine stromelysin-1 and -2 exhibit an overall identity of only 67% (data not shown). 3.3. Tissue distribution of stromelysin-2 mRNA in comparison with stromelysin-1 To assess the expression of stromelysin-1 and -2 mRNAs in adult murine tissues we performed RNase protection assays, using 20 mg aliquots of total cellular RNA isolated from various tissues and organs. The experiments were performed under conditions where a single base mismatch could be detected, so that crossreaction of the individual probes with other stromelysins could be excluded. In most of the tissues analyzed there was no basal expression of either stromelysin-1 or -2 (Fig. 2). Weak signals indicating the presence of both stromelysin-1 and -2 transcripts were detected only in the heart and the lung and after long exposure times (as indicated by the intensities of the probes in the tissue assay compared with those in the wound assay in Fig. 2). Hereby, the expression levels of stromelysin-1 were higher than those of stromelysin-2. Stromelysin-2 transcripts alone were detected at a comparatively high level only in the small intestine, a tissue characterized by an ongoing process of epithelial self-renewal. Hereby, stem cells from the base of the crypts continuously differentiate to cells of epithelial lineages that then migrate upward to the villus tip, where finally exfoliation occurs (for review see Gordon and Hermiston, 1994). In this

context, the need for constitutive low-level expression of an epithelial protease seems plausible. 3.4. Eukaryotic expression and APMA activation of IVH-tagged murine stromelysin-1 and -2 derivatives The putative ORFs of murine stromelysin-1 and -2 lacking the stop codons were subcloned into the eukaryotic expression vector pCG, thereby linking their 3∞ ends to the coding sequence of a 17 aa influenza hemagglutinin- (IVH-) tag. The resulting constructs were used to transfect COS cells. Lysates and 40-fold concentrated conditioned media were analyzed by Western blotting using a monoclonal antibody directed against the IVH-tag. As observed before in analogous experiments (G. Hu¨bner, personal communication), IVH-tagged recombinant proteins subjected to SDS–PAGE generally appear to possess higher molecular weights than calculated. This is also true for the IVH-tagged stromelysin-1 and -2 derivatives (Fig. 3), which have calculated molecular weights of approx. 55 and 45 kDa, respectively, for the latent and activated forms. Both stromelysin-1 and -2 were readily detectable in the lysates. Whereas stromelysin-1 occurred intraIVH cellularly in one distinct form (Fig. 3A, lane 2), stromelysin-2 was represented by three bands IVH ( Fig. 3A, lane 3). The stromelysin-2 doublet IVH (>50 kDa) might be due to glycosylation, as observed for recombinant human stromelysin-1 ( Wilhelm et al., 1987; Nagase et al., 1990; Galazka et al., 1996), or to limited proteolysis. The lower stromelysin-2 band IVH (approx. 30 kDa), however, doubtlessly results from amino-terminal truncation of a full-length translation

80

M. Madlener, S. Werner / Gene 202 (1997) 75–81

indeed metalloproteinase mediated (Fig. 3B, lanes 3 and 6). These results demonstrate that both murine stromelysins generated here are functional matrix metalloproteinases. The molecular cloning of murine stromelysin-2 offers the possibility to generate transgenic animals, which will help to elucidate the role of this enzyme in development, repair, and disease. 3.5. Conclusions

Fig. 3. Expression, secretion, and activation of IVH-tagged recombinant forms of murine stromelysin-1 and -2. Stromelysin-1 and 2 fusion proteins with a 17 aa IVH epitope tag were transiently expressed in COS cells. Twenty-four hours after transfection, culture medium was changed to serum-free medium, and following a 2-day incubation, lysates (A) and conditioned media (B) were collected and subjected to Western blot analysis using a monoclonal IVH anti-peptide antibody. Conditioned media were concentrated 40-fold by ultrafiltration using Centricon tubes and additionally used to assess APMA activation and its inhibition by EDTA. (A) Lysates of non-transfected COS cells ( lane 1), stromelysin-1 expressing COS cells ( lane 2), and IVH stromelysin-2 expressing COS cells ( lane 3). (B) Concentrated conIVH ditioned media of stromelysin-1 expressing COS cells ( lanes 1–3) IVH and stromelysin-2 expressing COS cells ( lanes 4–6). The aliquots IVH represented by lanes 2 and 5 were incubated for 18 h at 37°C in the presence of 2 mM APMA, and those in lanes 3 and 6 in the presence of 2 and 5 mM EDTA.

product, since the antigenic tag is expressed at the carboxyterminus of the recombinant proteins. Again, a similar phenomenon has been reported for recombinant human stromelysin-1, although only following prolonged incubation with 1.5 mM APMA (Benbow et al., 1996). The secretion levels of the stromelysin-2 protein IVH (Fig. 3B, lanes 4–6) were reproducibly lower than those of stromelysin-1 (Fig. 3B, lanes 1–3). As the expresIVH sion plasmids for both stromelysins were constructed in an absolutely analogous manner, it seems likely that the different secretion efficiencies result from divergent post-transcriptional events, e.g. the amino-terminal truncation observed specifically for stromelysin-2 . IVH Nevertheless, activation by the organomercurial salt 4-aminophenylmercuric acetate (APMA) efficiently generated truncated forms of both stromelysin-1 and IVH -2 about 9–10 kDa shorter than the latent secreted IVH forms (Fig. 3B, lanes 2 and 5). This result is in agreement with the cDNA-predicted molecular weights of the prodomains. In accordance with findings by Nagase et al. (1990), APMA activation of the murine stromelysin-1 derivative does not seem to follow a IVH single-step mechanism, as indicated by the appearance of a double band following incubation with 2 mM APMA. Inhibition of the conversions by addition of the chelating agent EDTA to the reaction mixture during APMA treatment indicates that the cleavages were

(1) The cDNA encoding murine stromelysin-2, a member of the matrix metalloproteinase (MMP) family, has been cloned. (2) The cDNA-predicted murine stromelysin-2 amino acid sequence is highly homologous to rat and human stromelysin-2 and displays significant similarity to murine stromelysin-1. (3) Stromelysin-2 transcripts were found in murine small intestine, suggesting a need for constitutive low level expression of an epithelial matrix-degrading enzyme in this tissue. (4) Transient expression of the murine stromelysins-1 and -2 in COS cells (as fusion proteins with a 17 aa IVH epitope tag at the carboxyterminus) yielded proteins that were secreted and could be induced to undergo proteolysis in the presence of the organomercurial salt APMA.

Acknowledgement We would like to thank Dr S. Frank for advice in generating the cDNA library, Drs C. Mauch and W.C. Parks for helpful discussions, and Dr P.H. Hofschneider for support. This work was supported by a grant from the Deutsche Forschungsgemeinschaft ( WE 1983/1-2) (to S.W.). S.W. was supported by a Herrmann-andLilly-Schilling award for medical research.

References Basset, P., Bellocq, J.P., Wolf, C., Stoll, I., Hutin, P., Limacher, J.M., Podjahcer, O.L., Chenard, M.P., Rio, M.C., Chambon, P., 1990. A novel metalloproteinase gene specifically expressed in stromal cells of breast carcinomas. Nature 348, 699–704. Benbow, U., Buttice`, G., Nagase, H., Kurkinen, M., 1996. Characterization of the 46-kDa intermediates of matrix metalloproteinase 3 (stromelysin 1) obtained by site-directed mutation of phenylalanine 83. J. Biol. Chem. 271, 10715–10722. Birkedal-Hansen, H., Moore, W.G.I., Bodden, M.K., Windsor, L.J., Birkedal-Hansen, B., DeCarlo, A., Engler, J.A., 1993. Matrix metalloproteinases: a review. Crit. Rev. Oral Biol. Med. 4, 197–250. Breathnach, R., Matrisian, L.M., Gesnel, M.C., Staub, A., Leroy, P., 1987. Sequences coding for part of oncogene-induced transin are highly conserved in a related rat gene. Nucl. Acids Res. 15, 1139–1151.

M. Madlener, S. Werner / Gene 202 (1997) 75–81 Chomczynski, P., Sacchi, N., 1987. Single step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162, 156–159. Dhanaraj, V., Ye, Q.-Z., Johnson, L.L., Hupe, D.J., Ortwine, D.F., Dunbar Jr., J.B., Rubin, J.R., Pavlovsky, A., Humble, C., Blundell, T.L., 1996. X-ray structure of a hydroxamate inhibitor complex of stromelysin catalytic domain and its comparison with members of the zinc metalloproteinase superfamily. Structure 4, 375–386. Galazka, G., Windsor, L.J., Birkedal-Hansen, H., Engler, J.A., 1996. APMA (4-aminophenylmercuric acetate) activation of stromelysin-1 involves protein interactions in addition to those with cystein-75 in the propeptide. Biochemistry 35, 11221–11227. Gordon, J.I., Hermiston, M.L., 1994. Differentiation and self-renewal in the mouse gastrointestinal epithelium. Curr. Opin. Cell Biol. 6, 795–803. Hammani, K., Henriet, P., Eeckhout, Y., 1992. Cloning and sequencing of a cDNA encoding mouse stromelysin 1. Gene 120, 321–322. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Lefebvre, O., Wolf, C., Limacher, J.M., Hutin, P., Wendling, C., LeMeur, M., Basset, P., Rio, M., 1992. The breast cancer-associated stromelysin-3 is expressed during mouse mammary gland apoptosis. J. Cell Biol. 119, 997–1002. Madlener, M., Mauch, C., Conca, W., Brauchle, M., Parks, W.C., Werner, S., 1996. Regulation of the expression of stromelysin-2 by growth factors in keratinocytes: implications for normal and impaired wound healing. Biochem. J. 320, 659–664. Matrisian, L.M., 1992. The matrix-degrading metalloproteinases. Bioessays 14, 455–463. Muller, D., Quantin, B., Gesnel, M.D., Millon-Collard, R., Abecassis, J., Breathnach, R., 1988. The collagenase gene family in humans consists of at least four members. Biochem. J. 253, 187–192. Nagase, H., Enghild, J.J., Suzuki, K., Salvesen, G., 1990. Stepwise activation mechanisms of the precursor of matrix metalloproteinase 3 (stromelysin) by proteinases and (4-aminophenyl )mercuric acetate. Biochemistry 29, 5783–5789. Nagase, H., Ogata, Y., Suzuki, K., Enghild, J.J., Salvesen, G., 1991.

81

Substrate specificities and activation mechanisms of matrix metalloproteinases. Biochem. Soc. Trans. 19, 715–718. Noe¨l, A., Santavicca, M., Stoll, I., L’Hoir, C., Staub, A., Murphy, G., Rio, M.C., Basset, P., 1995. Identification of structural determinants controlling human and mouse stromelysin-3 proteolytic activities. J. Biol. Chem. 270, 22866–22872. Pei, D., Majmudar, G., Weiss, S.J., 1995. Hydrolytic inactivation of a breast carcinoma cell-derived serpin by human stromelysin-3. J. Biol. Chem. 269, 25849–25855. Saarialho-Kere, U.K., Pentland, A.P., Birkedal-Hansen, H., Parks, W.C., Welgus, H.G., 1994. Distinct populations of basal keratinocytes express stromelysin-1 and stromelysin-2 in chronic wounds. J. Clin. Invest. 94, 79–88. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning. A Laboratory Manual. Cold Spring Harbour Laboratory Press, New York. Sang, Q.A., Douglas, D.A., 1996. Computational sequence analysis of matrix metalloproteinases. J. Protein Chem. 15, 137–160. Santavicca, M., Noe¨l, A., Angliker, H., Stoll, I., Segain, J.-P., Anglard, P., Chretien, M., Seidah, N., Basset, P., 1996. Characterization of structural determinants and molecular mechanisms involved in prostromelysin-3 activation by 4-aminophenylmercuric acetate and furin-type convertases. Biochem. J. 315, 953–958. Werner, S., Peters, K.G., Longaker, M.T., Fuller-Pace, F., Banda, M., Williams, L.T., 1992. Large induction of keratinocyte growth factor expression in the dermis during wound healing. Proc. Natl. Acad. Sci. USA 89, 6896–6900. Wilhelm, S.M., Collier, I.E., Kronberger, A., Eisen, A.Z., Marmer, B.L., Grant, G.A., Bauer, E.A., Goldber, G.I., 1987. Human skin fibroblast stromelysin: structure, glycosylation, substrate specificity, and differential expression in normal and tumorigenic cells. Proc. Natl. Acad. Sci. USA 84, 6725–6729. Windsor, L.J., Grenett, H., Birkedal-Hansen, B., Bodden, M.K., Engler, J.A., Birkedal-Hansen, H., 1993. Cell type-specific regulation of SL-1 and SL-2 genes. J. Biol. Chem. 268, 17341–17347. Woessner Jr., J.F., 1991. Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASEB J. 5, 2145–2154.