Unexpected Crucial Role of Residue 272 in Substrate Specificity of Fibroblast Collagenase
2002; Elsevier BV; Volume: 277; Issue: 30 Linguagem: Inglês
10.1074/jbc.m201367200
ISSN1083-351X
AutoresHiroki Tsukada, Tayebeh Pourmotabbed,
Tópico(s)Protease and Inhibitor Mechanisms
ResumoDegradation of type I collagen by collagenases is an important part of extracellular remodeling. To understand the role of the hinge region of fibroblast collagenase in its collagenolytic activity, we individually substituted the 10 conserved amino acid residues at positions 264, 266, 268, 296, 272, 277, 284, 289, 307, and 313 in this region of the enzyme by their corresponding residues in MMP-3, a noncollagenolytic matrix metalloproteinase. The general proteolytic and triple helicase activities of all of the enzymes were determined, and their abilities to bind to type I collagen were assessed. Among the mutants, only G272D mutant enzyme exhibited a significant change in type I collagenolysis. The alteration of the Gly272 to Asp reduced the collagenolytic activity of the enzyme to 13% without affecting its general proteolytic activity, substrate specificity, or the collagen binding ability. The catalytic efficiency of the G272D mutant for the triple helical peptide substrate [C6-(GP- Hyp)4GPL(Mca)GPQGLRGQL(DPN)GVR(GP-HYP)4-NH2]3and the peptide substrate Mca-PLGL(Dpa)AR-NH2 and its dissociation constant for the triple helical collagen were similar to that of the wild type enzyme, indicating that the presence of this residue in fibroblast collagenase is particularly important for the efficient cleavage of type I collagen. Gly272 is evidently responsible for the hinge-bending motion that is essential for allowing the COOH-terminal domain to present the collagen to the active site. Degradation of type I collagen by collagenases is an important part of extracellular remodeling. To understand the role of the hinge region of fibroblast collagenase in its collagenolytic activity, we individually substituted the 10 conserved amino acid residues at positions 264, 266, 268, 296, 272, 277, 284, 289, 307, and 313 in this region of the enzyme by their corresponding residues in MMP-3, a noncollagenolytic matrix metalloproteinase. The general proteolytic and triple helicase activities of all of the enzymes were determined, and their abilities to bind to type I collagen were assessed. Among the mutants, only G272D mutant enzyme exhibited a significant change in type I collagenolysis. The alteration of the Gly272 to Asp reduced the collagenolytic activity of the enzyme to 13% without affecting its general proteolytic activity, substrate specificity, or the collagen binding ability. The catalytic efficiency of the G272D mutant for the triple helical peptide substrate [C6-(GP- Hyp)4GPL(Mca)GPQGLRGQL(DPN)GVR(GP-HYP)4-NH2]3and the peptide substrate Mca-PLGL(Dpa)AR-NH2 and its dissociation constant for the triple helical collagen were similar to that of the wild type enzyme, indicating that the presence of this residue in fibroblast collagenase is particularly important for the efficient cleavage of type I collagen. Gly272 is evidently responsible for the hinge-bending motion that is essential for allowing the COOH-terminal domain to present the collagen to the active site. matrix metalloproteinase membrane-type fibroblast collagenase neutrophil collagenase (7-methoxycoumarin-4-yl) acetyl [3-(2′,4′-dinitrophenyl)-l-2,3-diaminopropionyl] 14 kDa-inhibitory domain of recombinant tissue inhibitor of metalloproteinase-2 Matrix metalloproteinases (MMP)1 are a family of enzymes capable of degrading a number of components of the extracellular matrix such as proteoglycans, type I collagen, fibronectin, and laminin. MMPs have been implicated to play significant roles in skeletal growth and remodeling (1Sellers A. Reynolds J.J. Meikle M.C. Biochem. J. 1978; 171: 493-496Crossref PubMed Scopus (111) Google Scholar, 2Vu T.H. Shipley J.M. Bergers G. Berger J.E. Helms J.A. Hanahan D. Shapiro S.D. Senior R.M. Werb Z. Cell. 1998; 93: 411-422Abstract Full Text Full Text PDF PubMed Scopus (1505) Google Scholar), wound healing (3Wolf C. Chernard M.P. Durand de Grossouvre P. Bellocq J.P. Chambon P. Basset P. J. Invest. Dermatol. 1992; 99: 870-872Abstract Full Text PDF PubMed Google Scholar, 4Wysocki A.B. Staiano-Coico L. Grinnell F. J. Invest. Dermatol. 1993; 101: 64-68Abstract Full Text PDF PubMed Google Scholar), periodontal disease (5Overall C.M. Sodek J. McCulloch A.G. Birek P. Infect. Immunol. 1991; 59: 4687-4692Crossref PubMed Google Scholar, 6Page R.C. J. Periodonal Res. 1991; 26: 230-242Crossref PubMed Scopus (683) Google Scholar), cancer (7Liotta L.A. Tryggvason K. Garbisa S. Hart I. Foltz C.M. Shafie S. Nature. 1980; 284: 67-68Crossref PubMed Scopus (1562) Google Scholar, 8Liotta L.A. Abe S. Robey P.G. Martin G. Proc. Natl. Acad. Sci. U. S. A. 1979; 76: 2268-2272Crossref PubMed Scopus (352) Google Scholar, 9Matrisian L.M. Bowden G.T. Krieg P. Furstenberger G. Briand J.-P. Leroy P. Breathnach R. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 9413-9417Crossref PubMed Scopus (205) Google Scholar, 10Bernhard E.J. Muschel R.J. Hughes E.N. Cancer Res. 1990; 50: 3872-3877PubMed Google Scholar, 11Bernhard E.J. Gruber S.B. Muschell R.J. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 4293-4297Crossref PubMed Scopus (391) Google Scholar, 12Bafetti L.M. Young T.N. Itoh Y. Stack M.S. J. Biol. Chem. 1998; 273: 143-149Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar), and arthritis (13Hayakawa T. Yamashita S. Kodama H. Iwata H. Iwata K. Biomed. Res. 1991; 12: 169-173Crossref Scopus (38) Google Scholar, 14Hirose T. Riefe R.A. Smith G.N., Jr. Stevens R.M. Mainardi C.L. Hasty K.A. J. Rheumatol. 1992; 19: 593-599PubMed Google Scholar, 15Yu L. Smith G.J., Jr. Brandt K. Capello W. Arthritis Rheum. 1990; 33: 1626-1633Crossref PubMed Scopus (18) Google Scholar, 16Harris E.D., Jr. N. Engl. J. Med. 1990; 322: 1277-1289Crossref PubMed Scopus (1565) Google Scholar). The MMP gene family is generally divided into four subgroups, collagenases (MMP-1, -8, and -13), gelatinases (MMP-2 and -9), stromelysins (MMP-3, -10, and -11), and membrane type (MT) MMPs (MMP-14–17). With a few exceptions, MMPs are composed of an amino-terminal prodomain, a catalytic domain containing the zinc binding active site, and a hemopexin-like carboxyl-terminal domain, which is separated from the catalytic domain by a proline-rich linker (hinge region). The length of the hinge region varies among the members of the MMP family (reviewed in Refs. 17Woessner J.F. FASEB J. 1991; 5: 2145-2154Crossref PubMed Scopus (3091) Google Scholar, 18Matrisian L.M. Bioessays. 1992; 14: 455-463Crossref PubMed Scopus (1332) Google Scholar, 19Birkedal-Hansen H. Moore W.G.I. Bodden M.K. Windsor L.J. Birkedal-Hansen B. DeCarlo A. Engler J.A. Crit. Rev. Oral Biol. Med. 1993; 4: 197-250Crossref PubMed Scopus (2647) Google Scholar, 20Massova I. Kotra L.P. Fridman R. Mobashery S. FASEB J. 1998; 12: 1075-1095Crossref PubMed Scopus (704) Google Scholar, 21Nagase H. Woessner J.F. J. Biol. Chem. 1999; 274: 21491-21494Abstract Full Text Full Text PDF PubMed Scopus (3903) Google Scholar). The turnover of type I collagen by mammalian collagenases MT1-MMP and gelatinase A is considered to be an important part of connective tissue remodeling. Collagenases cleave α1(I) and α2(I) chains of triple helical type I collagen at a unique bond, generating characteristic ¼- and ¾-length fragments (22Gross J. Hay E.D. Cell Biology Extracellular Matrix. Plenum Publishing Corp., New York1981: 217-258Google Scholar). Using collagen α1(I) mutants, Wuet al. (23Wu H. Byrne M.H. Stacy A. Goldring M.B. Birkhead J.R. Jaenisch R. Krane S.M. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5888-5892Crossref PubMed Scopus (94) Google Scholar) demonstrated that the conservation of the vicinal positions both upstream and downstream of the cleavage site is necessary for maintaining collagenase susceptibility. However, the means by which recognition of the triple helical collagen by the enzymes is coupled to cleavage remain speculative. It has been demonstrated that the binding of collagenases to type I collagen occur via the COOH-terminal domain, and this binding is required for the specific cleavage of type I collagen (24Murphy G. Allan J.A. Willenbrock F. Cockett M.I. O'Connell J.P. Docherty A.J.P. J. Biol. Chem. 1992; 267: 9612-9618Abstract Full Text PDF PubMed Google Scholar, 25Hirose T. Patterson C.E. Pourmotabbed T. Mainardi C.L. Hasty K.A. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 2569-2573Crossref PubMed Scopus (124) Google Scholar, 26Sanchez-Lopez R. Alexander C.M. Behrendsten O. Breathnach R. Werb Z. J. Biol. Chem. 1993; 268: 7238-7247Abstract Full Text PDF PubMed Google Scholar). Whether this domain is removed or replaced with that of the stromelysins, the resulting enzymes are not able to cleave type I collagen (24Murphy G. Allan J.A. Willenbrock F. Cockett M.I. O'Connell J.P. Docherty A.J.P. J. Biol. Chem. 1992; 267: 9612-9618Abstract Full Text PDF PubMed Google Scholar, 25Hirose T. Patterson C.E. Pourmotabbed T. Mainardi C.L. Hasty K.A. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 2569-2573Crossref PubMed Scopus (124) Google Scholar, 26Sanchez-Lopez R. Alexander C.M. Behrendsten O. Breathnach R. Werb Z. J. Biol. Chem. 1993; 268: 7238-7247Abstract Full Text PDF PubMed Google Scholar). However, it has been shown that the isolated COOH-terminal domain of fibroblast collagenase (FC) (MMP-1), despite its ability to bind to collagen, cannot cleave type I collagen (24Murphy G. Allan J.A. Willenbrock F. Cockett M.I. O'Connell J.P. Docherty A.J.P. J. Biol. Chem. 1992; 267: 9612-9618Abstract Full Text PDF PubMed Google Scholar, 27Knauper V. Osthus A. DeClerck Y.A. Langley K.E. Blaser J. Tschesche H. Biochem. J. 1993; 291: 847-854Crossref PubMed Scopus (99) Google Scholar, 28Bigg H.F. Clark I.M. Cawston T.E. Biochim. Biophys. Acta. 1994; 1208: 157-165Crossref PubMed Scopus (30) Google Scholar). In addition, chimeric enzymes containing the entire NH2-terminal domain of stromelysin-1 (MMP-3) (24Murphy G. Allan J.A. Willenbrock F. Cockett M.I. O'Connell J.P. Docherty A.J.P. J. Biol. Chem. 1992; 267: 9612-9618Abstract Full Text PDF PubMed Google Scholar) or stromelysin-2 (MMP-10) (26Sanchez-Lopez R. Alexander C.M. Behrendsten O. Breathnach R. Werb Z. J. Biol. Chem. 1993; 268: 7238-7247Abstract Full Text PDF PubMed Google Scholar) attached to the COOH-terminal domain of FC have been shown not to have type I collagenolytic activity. A further analysis of the COOH-terminal domain demonstrated that a 67 amino acid sequence (amino acid residues 259–326) consisting of the hinge region and a part of the first hemopexin-like repeat are essential for type I collagenolytic activity of neutrophil collagenase (NC) (MMP-8) (25Hirose T. Patterson C.E. Pourmotabbed T. Mainardi C.L. Hasty K.A. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 2569-2573Crossref PubMed Scopus (124) Google Scholar). The replacement of the hinge region of NC with that of MMP-3 substantially reduced the proteolytic activity of the enzyme toward type I collagen (25Hirose T. Patterson C.E. Pourmotabbed T. Mainardi C.L. Hasty K.A. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 2569-2573Crossref PubMed Scopus (124) Google Scholar). Moreover, the amino acid sequence RWTNNFREY in the catalytic domain of FC was recently identified as being involved in the collagenolytic activity of the enzyme (29Chung L. Shimokawa K. Dinakarpandian D. Grams F. Fields G.B. Nagase H. J. Biol. Chem. 2000; 275: 29610-29617Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar). However, the insertion of this sequence in MMP-3 did not render the enzyme type I collagenolytic activity (29Chung L. Shimokawa K. Dinakarpandian D. Grams F. Fields G.B. Nagase H. J. Biol. Chem. 2000; 275: 29610-29617Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar), suggesting that both the COOH- and NH2-terminal domains of collagenases contribute equally to their type I collagenolytic activity. Here, we have found that Gly272 in the hinge region of FC is one of the key elements required for the efficient utilization of type I collagen by the enzyme. Gly272 is located in the middle of the hinge region, suggesting that the structural flexibility of the hinge region led by Gly272 allows for the spatial cooperation between the NH2- and COOH-terminal domains necessary for catalysis. Substituting Gly272 limits the flexibility of the hinge region, leading to a lack of enzyme specificity. The construction of pET/FC containing the full-length cDNA for FC has been described previously (30O'Farrell T.J. Pourmotabbed T. J. Biol. Chem. 2000; 275: 27964-27972Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). A single-site mutation was introduced in and around the hinge region (amino acid residues 262–313) of FC by PCR using pET/FC as the template. The following 5′-mutagenic primers (substituted nucleotide is underlined) were used to substitute Ser264, Asn266, Val268, Gln269, Gly272, Asn307, and Val313 with Ala, Asp, Pro, Glu, Asp, Ser, and Pro, the corresponding amino acid residues in MMP-3, respectively (Fig. 1): S264A, 5′-ATATATGGACGTGCCCAAAATCCTGTCCAG-3′; N266D, 5′-TATATGGACGTTCCCAAGATCCTGTCCAG-3′; V268P, 5′-ATATGGACGTTCCCAAAATCCTCCTGTCGAGCCC-3′; Q269E, 5′-ATATGGACGTTCCCAAAATCCTGTCGAGCCC-3′; G272D, 5′-ATATGGACGTTCCCAAAATCCTGTCCAGCCCATCGACCCACAAACC-3′; T284S, 5′-AAAGCGTGTGACAGTAAGCTAAGCTTTGAT-3′; T289S, 5′-AAAGCGTGTGACAGTAAGCTAACCTTTGATGCTATAAGTACGATT-3′; N307S, 5′-ATTGTACATGCGCACAAGTCCCTTCTACC-3′; and V313P, 5′-CCCTTCTACCCGGAACCTGAGCTCAA- TTTCAC-3′. The 3′ primer, 5′-TTTGGACTCACACCATGTGTTTTCCATTCAAATTAG-3′, that was used to construct all of the above mutations was complementary to the 3′-untranslated region of the cDNA and contained a uniqueDraIII restriction site. The PCR products corresponding to S264A, N266D, V268P, Q269E, and G272D mutations were digested with MaeIII and DraIII restriction enzymes and ligated with SphI/DraIII and SphI/MaeIII fragments from pET/FC to produce the final expression vectors pET/FC/S264A, pET/FC/N266D, pET/FC/V268P, pET/FC/Q269E, and pET/FC/G272D, respectively. The pET/FC/T284S and pET/FC/T289S expression vectors were generated by digesting T284S and T289S PCR products with MaeIII and DraIII and ligating the resulting DNA fragments withBstXI/DraIII andBstXI/MaeIII DNA fragments of pET/FC. The N307S-mutagenic primer incorporated an FspI restriction site in the PCR product. Thus, pET/FC/N307S expression vector was generated by digesting the N307S PCR product with DraIII andFspI restriction enzymes and ligating the PCR fragment withSalI/DraIII and SalI/FspI DNA fragments of pET/FC. The V313P PCR product was digested with restriction enzymes DraIII and NciI (a unique restriction site at the 5′ end of the DNA fragment) and ligated withNciI/DraIII fragment from pET/FC to produce the expression vector pET/FC/V313P. To substitute Ala for Lys at position 277, the 3′-mutagenic primer (substituted nucleotide is underlined), 5′-TAGCTTACTGTCACACGCTGCTGGGGTTTG-3′, that was complementary to the nucleotides 821–850 of the FC cDNA and contained a unique MaeIII restriction site 5′ to the substituted nucleotide was used. The 5′-primer, 5′-GGGGATGCTCATTTTGATGAACATGAAAGGTGGACCAAC-3′, contained a uniqueBstXI restriction site and was complementary to the nucleotides 578–616 of the FC cDNA. The resulting 272-bp PCR product was used as a 3′ primer in the second round of PCR using pET/FC as template. The same 5′ primer used to construct all of the above mutations (264–313 mutations) was used in this reaction. The final PCR product was digested with BstXI and DraIII and finally ligated with DraIII/BstXI DNA fragment of pET/FC to produce pET/FC/K277A expression vector. All of the PCR-derived DNA for these constructs was sequenced to confirm that the desired mutations were the only mutations generated by the PCR. Plasmids encoding the mutant enzymes were introduced intoEscherichia coli BL21(DE3) cells and were maintained in 50 μg/ml carbenicillin. The cells bearing the plasmids were induced by 0.4–1.0 mmisopropyl β-d-thiogalactopyranoside. The recombinant enzymes were activated with 1.0 mm p-aminophenylmercuric acetate in the presence of 2.0 μm ZnCl2 at 37 °C for 4 h and purified on zinc-chelating columns as described previously (30O'Farrell T.J. Pourmotabbed T. J. Biol. Chem. 2000; 275: 27964-27972Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). Purified enzymes were dialyzed against 50 mm Tris-HCl, pH 7.5, 5 mm CaCl2, 150 mm NaCl, and 2 μm ZnCl2 and stored at −20 °C until use. Thekcat/Km values for the fluorogenic peptide substrate were obtained by assaying the enzymes at 23 °C using 4 μm substrate as described previously (30O'Farrell T.J. Pourmotabbed T. J. Biol. Chem. 2000; 275: 27964-27972Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). The assays were conducted under the condition whereKm ≫ [S] andkcat/Km values were calculated from the following equation:kcat/Km ≅vo/[E]T[S]. To demonstrate that 4 μm substrate concentration used in the above reaction mixture was adequate to fulfill the requirements for the kinetic analysis in which Km ≫ [S], all of the enzymes were initially assayed with various concentrations (2–36 μm) of Mca-PLGL(Dpa)AR-NH2 peptide. The initial rate of substrate hydrolysis was determined by measuring the increase in fluorescence (λex = 328 nm, λem= 393 nm) over the first 500 s of the reaction at 23 °C. The apparent Km and Vmax values were then calculated by Lineweaver-Burk plots using the Enzyme Kinetics program, version 1.0 (Trinity Software, Camton, NH). The concentrations of active enzymes, [E]T, were determined by active-site titration with the amino-terminal 14-kDa inhibitory domain of recombinant tissue inhibitor of metalloproteinase-2 (TIMP-2-ID) (a kind gift from Dr. Harold Tschesche, Lerstuhl für Biochemie, Universitat Bielefeld, Bielefeld, Germany) as described previously (31O'Farrell T.J. Pourmotabbed T. Arch. Biochem. Biophys. 1998; 354: 24-30Crossref PubMed Scopus (37) Google Scholar). The type I collagenolytic activity of the wild type and the mutant enzymes was initially determined by using soluble type I collagen as substrate. The active enzymes (0.6 μm) were incubated with 380 μg/ml triple helical native bovine type I collagen (Fluka Biochemika, Milwaukee, WI) in an assay buffer containing 50 mm Tris, pH 7.5, 5 mm CaCl2, 150 mm NaCl at 27 °C for 18 h. The reactions were then quenched by 10 mm EDTA, boiled for 5 min, and loaded onto a 6% SDS-PAGE under nonreducing conditions. Protein bands were visualized by staining the gel with Coomassie Blue. The integrated intensities of the α1(I) and α2(I) collagen bands were determined by using an Alpha Imager 2000 documentation and analysis system (Alpha Innotech Corp., San Leandro, CA). The difference between the integrated intensities of the α1(I) collagen band in the presence and in the absence of the enzyme was used to calculate the activity of the enzyme as described previously (32Welgus H.G. Jeffrey J.J. Eisen A.Z. J. Biol. Chem. 1981; 259: 9511-9515Abstract Full Text PDF Google Scholar). The activity of the enzymes toward type I collagen was also measured using 14C-labeled fibril type I collagen at 35 °C for 1 h as described previously (33Cawston T.E. Mercer E. Tyler J.A. Biochim. Biophys. Acta. 1981; 657: 73-83Crossref PubMed Scopus (35) Google Scholar). Specific activity values for the enzymes were calculated by dividing the amount of14C-collagen degraded per hour by nanomoles of the full-length active enzyme used in the assay. The relationship between the concentration of the enzyme and the amount of collagen degraded per hour was linear. The concentrations of the full-length active enzymes were calculated by multiplying the percentages of their corresponding protein in the enzyme solution as determined by SDS-PAGE by the total protein in the solution. The total amount of the protein was calculated by the Bradford dye binding technique using bovine serum albumin as a standard. The activity of the wild type and FC/G272D enzymes was determined using the fluorogenic triple helical substrate [C6-(GP-Hyp)4GPL(Mca)GPQGLRGQL(DPN)GVR(GP-HYP)4-NH2]3(a kind gift from Dr. Gregg Fields, Department of Chemistry and Biochemistry, Florida Atlantic University) as described previously (34Lauer-Fields J.L. Broder T. Sritharan T. Chung L. Nagase H. Fields G.B. Biochemistry. 2001; 40: 5795-5803Crossref PubMed Scopus (82) Google Scholar). Type I collagen film was prepared by plating 50 μl of 100 μg/ml collagen in 10 mm Tris, pH 7.5, containing 200 mm NaCl onto the wells of XenobindTM covalent binding microwell plates (Xenopore, Saddle Brook, NJ). The films were allowed to dry at 37 °C. The wells were then washed with wash buffer, 25 mm Na2B4O7, pH 8.0, 5 mm CaCl2, 150 mm NaCl, 2 μm ZnCl2, 5% (v/v) ethylene glycol, 0.05% (v/v) Tween 20, and blocked with 300 μl of 10 mg/ml bovine serum albumin in the wash buffer for 1 h at 37 °C. The wells were washed for the second time with wash buffer and incubated with various concentrations (4–20 μg/ml) of either the wild type or the FC mutants in 50 μl of wash buffer containing 67 μg/ml bovine serum albumin for 1 h at 17 °C. The supernatants were then removed from the wells, and the films were washed five times with wash buffer. The bound enzymes were eluted by adding 50 μl of wash buffer containing 1 m NaCl, 0.05% (v/v) Brij-35, and 15% (v/v) Me2SO to the wells and by incubating the plates for 1 h at 17 °C. The eluted samples (bound fractions) were then assayed for proteolytic activity (initial rate of hydrolysis) using 10 μm Mca-PLGL(Dpa)AR-NH2 as described above. The assays at each applied enzyme concentration were performed in duplicate, and the initial rates were averaged. The amino acid substitutions in the crystal structure of porcine fibroblast collagenase (35Li J Brick P. O'Hare M.C. Skarzynski T. Lioyd L.F. Curry V.A. Clark I.M. Bigg H.F. Hazleman B.L. Cawstone T.E. Blow D.M. Structure. 1995; 3: 541-549Abstract Full Text Full Text PDF PubMed Scopus (239) Google Scholar) were made using Swiss Protein Data Bank viewer modeling system. The substitutions were made, and the energy of the resulting structure was minimized by the Biopolymer and Discover modules, respectively. Our previous experiments demonstrated that the substitution of hinge region and part of the first propeller of the COOH-terminal hemopexin-like domain of NC with that of MMP-3 substantially decreased type I collagenolytic activity of the enzyme (25Hirose T. Patterson C.E. Pourmotabbed T. Mainardi C.L. Hasty K.A. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 2569-2573Crossref PubMed Scopus (124) Google Scholar). To identify specific amino acid residues in this region of collagenases that might be important for enhancing their type I collagenolytic activity, a partial amino acid sequence alignment from this region of FC, NC, and MMP-3 was made (Fig. 1). Although MMP-3 binds type I collagen, it does not cleave native triple helical collagen. This alignment revealed that the conserved amino acid residues Ser264, Asn266, Val268, Gln296, Gly272, Lys277, Thr284, Thr289, Asn307, and Val313 in FC and NC have been substituted by Ala, Asp, Pro, Asp, Glu/Asp, Ala, Ser, Ser, Ser, and Pro in MMP-3, respectively. The significance of these conserved residues on the collagenolytic activity of the collagenases was studied by individually replacing these residues in FC with their corresponding residues in MMP-3. The latent wild type and mutant proteins were expressed in E. coli, activated with p-aminophenylmercuric acetate, purified, and recovered as 45-kDa species (Fig.2) as expected (36Windsor L.J. Bodden M.K. Birkedal-Hansen B. Engler J.A. Birkedal-Hansen H. J. Biol. Chem. 1994; 269: 26201-26207Abstract Full Text PDF PubMed Google Scholar). The catalytic efficiencies of recombinant FC and all of the mutant enzymes toward the peptide substrate Mca-PLGL(Dpa)AR-NH2 were then determined at 23 °C. In this experiment, the active site of each enzyme was titrated with TIMP-2-ID, and the same amount of active enzyme was assayed. The catalytic efficiency of all of the FC mutants toward the peptide substrate was found to be essentially equal to that of FC (Table I). These data suggest that all of the enzymes were properly folded and that the general proteolytic activity of FC was not compromised by the substitutions.Table ISubstrate specificity of FC and FC mutantsEnzymeMca-PLGL(Dpa)AR-NH21-aThe assay was carried out at 23 °C. The active enzyme concentrations were determined by TIMP-2-ID titration.Collagenase activity1-bThe collagenolytic activity of the enzymes was determined at 27 °C. The concentration of the full-length enzymes was determined by densitometric analysis of the 45-kDa protein bands as described under "Experimental Procedures." The activity of wild type FC was taken as 100%.kcat/Km mm−1s−1%FC1.90100S264A2.12105N266D1.99116V268P1.9396.8Q269E1.88127G272D1.9012.9K277A2.45102T284S2.11101T289S1.8498N307S2.14123V313P2.46106Experimental details are described under "Experimental Procedures."1-a The assay was carried out at 23 °C. The active enzyme concentrations were determined by TIMP-2-ID titration.1-b The collagenolytic activity of the enzymes was determined at 27 °C. The concentration of the full-length enzymes was determined by densitometric analysis of the 45-kDa protein bands as described under "Experimental Procedures." The activity of wild type FC was taken as 100%. Open table in a new tab Experimental details are described under "Experimental Procedures." The activity of the wild type and FC mutants toward collagen was initially assessed using soluble type I collagen as a substrate. Because upon activation some of the 45-kDa forms converted to lower molecular mass species (Fig. 2) and only the full-length FC expressed collagenolytic activity, the amount of each of the full-length enzyme was quantified by densitometric analysis after SDS-PAGE and staining of proteins with Coomassie Brilliant Blue R250. As shown in Fig.3 and Table I, all of the mutant enzymes with the exception of FC/G272D (Fig. 3, lane 7) were able to cleave type I collagen with similar specific activity to that of the wild type enzyme. FC/G272D mutant at the same concentration as FC (Fig.3, lane 7, and Tables I andII) was able to degrade only ∼13% of the soluble type I collagen. This finding was confirmed by assaying the enzyme using 14C-labeled type I collagen fibril as a substrate. FC/G272D also showed a low collagenolytic activity (∼10 μg of collagen degraded/h/pmol) toward the type I collagen fibril, whereas FC/V268P, FC/K277A, and FC/T289S had similar type I collagenolytic activity as that of FC (78–81 μg of collagen degraded/h/pmol). This finding indicated that the presence of Gly272 in the collagenase molecule is important for the type I collagenolytic activity of the enzyme. However, the Gly272 to Asp mutation did not affect the cleavage pattern of type I collagen (Fig. 3, lane 7), indicating that the low type I collagenolytic activity of the mutant was not because of a change in the specificity of cleavage sites within the collagen molecule. In addition, the digestion of gelatin by FC and FC/G272D and subsequent analysis of the products by SDS-PAGE indicated that these enzymes have similar substrate specificity. As shown in Fig.4, FC/G272D mutant (lane 3) digested gelatin in a pattern similar to that of FC (lane 2).Table IIKinetic parameters of FC and FC mutantsEnzymeMca-PLGL(Dpa)AR-NH22-aThe assay was carried out at 23 °C. The active enzyme concentrations were determined by TIMP-2-ID titration.Type I fibril collagen2-bThe collagenolytic activity of the enzymes was determined at 35 °C. The concentration of the full-length enzymes was determined by densitometric analysis of the 45-kDa protein bands as described under "Experimental Procedures." Specific activityKd for type I collagenkcat/Km mm−1s−1μg/h/pmolnmFC1.9079.091.37V268P1.9376.611.50G272D1.9010.191.35K277A2.4581.031.40T289S1.8477.531.39Experimental details are described under "Experimental Procedures."2-a The assay was carried out at 23 °C. The active enzyme concentrations were determined by TIMP-2-ID titration.2-b The collagenolytic activity of the enzymes was determined at 35 °C. The concentration of the full-length enzymes was determined by densitometric analysis of the 45-kDa protein bands as described under "Experimental Procedures." Open table in a new tab Figure 4Digestion of gelatin by FC and FC/G272D mutant. The purified enzymes (85 μm) were incubated with 2.0 μg/μl bovine type I gelatin in an assay buffer containing 50 mm Tris, pH 7.5, 5 mm CaCl2, and 2 μm ZnCl2 at 37 °C for 20 h. The reaction mixtures were then boiled and separated on a 10% SDS-PAGE. Protein bands were visualized by Coomassie Blue staining. Lane 1, type I gelatin alone; lane 2, wild type FC;lane 3, FC/G272D. Molecular mass markers are indicated on the right.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Experimental details are described under "Experimental Procedures." It has been suggested that collagenolysis requires the binding of the substrate followed by unwinding (helicase activity) and hydrolysis of the triple helical collagen molecule by FC. To investigate the role of Gly272 in triple helicase activity of the enzyme, the catalytic efficiencies of FC and FC/G272D mutant toward the fluorogenic triple helical peptide substrate [C6-(GP-Hyp)4GPL(Mca)GPQGLRGQL(DPN)GVR(GP-HYP)4-NH2]3were determined at 30 °C. The catalytic efficiency of FC/G272D toward the peptide substrate (kcat/Km = 9945m−1s−1) was found to be ∼1.5 times higher than that of FC (kcat/Km = 6430m−1s−1). This finding suggested that the triple helicase activity of the enzyme is not dependent on the presence of Gly at position 272. It has been proposed
Referência(s)