NMR Structure of the Netrin-like Domain (NTR) of Human Type I Procollagen C-Proteinase Enhancer Defines Structural Consensus of NTR Domains and Assesses Potential Proteinase Inhibitory Activity and Ligand Binding
2003; Elsevier BV; Volume: 278; Issue: 28 Linguagem: Inglês
10.1074/jbc.m302734200
ISSN1083-351X
AutoresEdvards Liepinsh, László Bányai, Guido Pintacuda, Mária Trexler, László Patthy, Gottfried Otting,
Tópico(s)Blood Coagulation and Thrombosis Mechanisms
ResumoProcollagen C-proteinase enhancer (PCOLCE) proteins are extracellular matrix proteins that enhance the activities of procollagen C-proteinases by binding to the C-propeptide of procollagen I. PCOLCE proteins are built of three structural modules, consisting of two CUB domains followed by a C-terminal netrin-like (NTR) domain. While the enhancement of proteinase activity can be ascribed solely to the CUB domains, sequence homology of the NTR domain with tissue inhibitors of metalloproteinases suggest proteinase inhibitory activity for the NTR domain. Here we present the three-dimensional structure of the NTR domain of human PCOLCE1 as the first example of a structural domain with the canonical features of an NTR module. The structure rules out a binding mode to metalloproteinases similar to that of tissue inhibitors of metalloproteinases but suggests possible inhibitory function toward specific serine proteinases. Sequence conservation between 13 PCOLCE proteins from different organisms suggests a conserved binding surface for other protein partners. Procollagen C-proteinase enhancer (PCOLCE) proteins are extracellular matrix proteins that enhance the activities of procollagen C-proteinases by binding to the C-propeptide of procollagen I. PCOLCE proteins are built of three structural modules, consisting of two CUB domains followed by a C-terminal netrin-like (NTR) domain. While the enhancement of proteinase activity can be ascribed solely to the CUB domains, sequence homology of the NTR domain with tissue inhibitors of metalloproteinases suggest proteinase inhibitory activity for the NTR domain. Here we present the three-dimensional structure of the NTR domain of human PCOLCE1 as the first example of a structural domain with the canonical features of an NTR module. The structure rules out a binding mode to metalloproteinases similar to that of tissue inhibitors of metalloproteinases but suggests possible inhibitory function toward specific serine proteinases. Sequence conservation between 13 PCOLCE proteins from different organisms suggests a conserved binding surface for other protein partners. Procollagen C-proteinase enhancer-1 (PCOLCE1) 1The abbreviations used are: PCOLCE, procollagen C-proteinase enhancer protein; APMA, p-aminophenylmercuric acetate; BMP-1, bone morphogenetic protein-1; BPTI, basic pancreatic trypsin inhibitor; CT-PCPE, C-terminal domain of procollagen C-terminal proteinase enhancer; CUB domain, complement-Uegf-BMP-1 domain; DABCYL, 4-{[(4-dimethylamino)phenyl]azo}benzoic acid; DNP, 2,4-dinitrophenyl; DTE, 1,4-dithioerythriol; EDANS, 5-[(2′-aminoethyl)-amino]naphthalenesulfonic acid; EST, expressed sequence tag; HSQC, heteronuclear single quantum coherence; MMP, matrix metalloproteinase; NOE, nuclear Overhauser effect; NOESY, NOE spectroscopy; NTR domain, netrin-like domain; PDB, protein data bank; PEST region, region rich in Pro, Glu, Ser, and Thr residues; PSTI, pancreatic secretory trypsin inhibitor; r.m.s.d., root mean square deviation; TACE/ADAM17, tumor necrosis factor-α converting enzyme/A disintegrin and metalloproteinase domain 17; TIMP, tissue inhibitor of metalloproteinase; TOCSY, total correlation spectroscopy; pNA, p-nitroanilide. is an extracellular matrix glycoprotein that yields up to 20-fold enhancement of the activities of procollagen C-proteinases such as bone morphogenetic protein-1 (BMP-1) (1Adar R. Kessler E. Goldberg B. Collagen Relat. Res. 1986; 6: 267-277Crossref PubMed Scopus (54) Google Scholar, 2Kessler E. Adar R. Eur. J. Biochem. 1989; 186: 115-121Crossref PubMed Scopus (83) Google Scholar, 3Hulmes D.J.S. Mould A.P. Kessler E. Matrix Biol. 1997; 16: 41-45Crossref PubMed Scopus (49) Google Scholar), probably mediated by binding to the C-terminal propeptide of type I procollagen (4Takahara K. Kessler E. Biniaminaov L. Brusel M. Eddy R.L. Jani-Sait S. Shows T.B. Greenspan D.S. J. Biol. Chem. 1994; 269: 26280-26285Abstract Full Text PDF PubMed Google Scholar). Human PCOLCE1 is a 50-kDa protein (55 kDa when glycosylated) (2Kessler E. Adar R. Eur. J. Biochem. 1989; 186: 115-121Crossref PubMed Scopus (83) Google Scholar) of a rod-like shape of about 150 Å length (5Bernocco S. Steiglitz B.M. Svergun D.I. Petoukhov M.V. Ruggiero F. Ricard-Blum S. Ebel C. Geourjon C. Deleage G. Font B. Eichenberger D. Greenspan D.S. Hulmes D.J. J. Biol. Chem. 2003; 278: 7199-7205Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar). It is built of three structural modules, comprising two CUB domains followed by a C-terminal netrin-like (NTR) domain. The CUB2 and NTR domains are linked by a flexible PEST region of ∼40 residues. The enhancement in BMP-1 activity by PCOLCE1 has been assigned to the CUB domains, as the enhancer activity persists after removal of the C-terminal NTR domain by natural processing (2Kessler E. Adar R. Eur. J. Biochem. 1989; 186: 115-121Crossref PubMed Scopus (83) Google Scholar, 3Hulmes D.J.S. Mould A.P. Kessler E. Matrix Biol. 1997; 16: 41-45Crossref PubMed Scopus (49) Google Scholar). There is evidence that removal of the NTR domain can substantially increase the enhancer activity (2Kessler E. Adar R. Eur. J. Biochem. 1989; 186: 115-121Crossref PubMed Scopus (83) Google Scholar). Both PCOLCE1 and its structural and functional homologue PCOLCE2 are collagen-binding proteins, capable of binding at multiple sites on the triple helical portions of fibrillar collagens and also to its isolated C-propeptide trimer (6Steiglitz B.M. Keene D.R. Greenspan D.S. J. Biol. Chem. 2002; 277: 49820-49830Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 7Ricard-Blum S. Bernocco S. Font B. Moali C. Eichenberger D. Farjanel J. Burchart E.R. van der Rest M. Kessler E. Hulmes D.J.S. J. Biol. Chem. 2002; 277: 33864-33869Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar). This binding activity has been attributed to the CUB domains (5Bernocco S. Steiglitz B.M. Svergun D.I. Petoukhov M.V. Ruggiero F. Ricard-Blum S. Ebel C. Geourjon C. Deleage G. Font B. Eichenberger D. Greenspan D.S. Hulmes D.J. J. Biol. Chem. 2003; 278: 7199-7205Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar). Much less is known about the function of the NTR domain of PCOLCEs or any other protein. C-terminal fragments of PCOLCE1 comprising the NTR domain have been detected, however, in media conditioned by human brain tumor cells. The fragments, but not full-length PCOLCE1, were shown to be associated with inhibitory activity of matrix metalloproteinase-2 (MMP-2), suggesting that they present a new class of metalloproteinase inhibitors (8Mott J.D. Thomas C.L. Rosenbach M.T. Takahara K. Greenspan D.S. Banda M.J. J. Biol. Chem. 2000; 275: 1384-1390Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). NTR domains are characterized by a set of conserved disulfide bridges and amino acid sequence homologies. NTR domains have been identified in a wide range of proteins, including the netrin family of proteins, secreted frizzled-related proteins, complement proteins C3, C4, and C5, tissue inhibitors of metalloproteinases (TIMP), and a number of putative Caenorhabditis elegans proteins (9Bányai L. Patthy L. Protein Sci. 1999; 8: 1636-1642Crossref PubMed Scopus (149) Google Scholar). The putative C. elegans proteins K07C11.3 and K07C11.5 seem to consist only of single, isolated NTR domains, suggesting that NTR domains are stable structural entities. More frequently, however, NTR domains are found as modules in multidomain proteins, where they occur in association with epidermal growth factor-like repeats, CUB domains, fibulin-related domains, cysteine-rich domains related to the Wnt-binding domain of frizzled receptors, and domains homologous to the N-terminal domain VI of laminin B, whey acidic protein domains, follistatin domains, immunoglobulin domains, and Kunitz-type protease inhibitor domains (9Bányai L. Patthy L. Protein Sci. 1999; 8: 1636-1642Crossref PubMed Scopus (149) Google Scholar, 10Trexler M. Bányai L. Patthy L. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 3705-3709Crossref PubMed Scopus (48) Google Scholar, 11Trexler M. Bányai L. Patthy L. Biol. Chem. 2002; 383: 223-228Crossref PubMed Scopus (31) Google Scholar). Based on amino acid sequence homologies, NTR domains present the core structural element of tissue inhibitors of metalloproteinases (TIMP), although the sequence similarities are limited (9Bányai L. Patthy L. Protein Sci. 1999; 8: 1636-1642Crossref PubMed Scopus (149) Google Scholar). The only structural information about NTR domains to date comes from several TIMP structures and the laminin-binding domain of agrin. The laminin-binding agrin domain is structurally homologous to the NTR subdomain in TIMPs, but its sequence homology to TIMPs and other NTR domains is so low that it escaped identification as an NTR domain before structure determination (12Stetefeld J. Jenny M. Schulthess T. Landwehr R. Schumacher B. Frank S. Rüegg M.A. Engel J. Kammerer R.A. Nat. Struct. Biol. 2001; 8: 705-709Crossref PubMed Scopus (40) Google Scholar). Here we report the three-dimensional structure of the NTR domain of PCOLCE1, NTRPCOLCE1, in solution. The NTRPCOLCE1 domain is one of the smallest NTR domains known and, in contrast to TIMPs and the laminin-binding agrin domain, contains the canonical set of three disulfide bonds typical of classical NTR domains (9Bányai L. Patthy L. Protein Sci. 1999; 8: 1636-1642Crossref PubMed Scopus (149) Google Scholar). The structure establishes a set of characteristic residues that will assist the identification of similar folds from sequence alignments. Functional implications of the fold with regard to potential proteinase inhibitory activity are discussed. Materials—Restriction enzymes were purchased from Promega (Madison, WI) and New England Biolabs (Beverly, MA). DNA polymerase I Klenow fragment, thrombin, and M13 sequencing reagents were from Amersham Biosciences. PCR primers were obtained from Integrated DNA Technologies (Coralville, IA). Ammonium 15N chloride was from Sigma. Human matrix metalloproteinases (MMP-1 proenzyme, active MMP-2, MMP-3 catalytic chain, and active MMP-9) were purchased from Calbiochem; human TACE/ADAM17 was from R & D Systems (Minneapolis, MN). The TACE fluorogenic substrate (DABCYL-Leu-Ala-Gln-Ala-Val-Arg-Ser-Ser-Ser-Arg-EDANS) and the MMP fluorogenic substrate (DNP-Pro-Leu-Gly-Leu-Trp-Ala-d-Arg-NH2) were from Calbiochem; p-aminophenylmercuric acetate (APMA) was from Sigma. Human plasma kallikrein (Calbiochem) and bovine trypsin (Sigma) were commercial preparations. The synthetic substrates d-Pro-Phe-Arg-pNA and Bz-Phe-Val-Arg-pNA were from Serva (Heidelberg, Germany) and Bachem (Bubendorf, Switzerland), respectively. The pET15b expression vector was from Novagen Merck Kft (Budapest, Hungary). Escherichia coli strains JM109 and BL21(DE3) were from New England Biolabs (Beverly, MA) and Novagen Merck Kft (Budapest, Hungary), respectively. The human liver cDNA library was purchased from Clontech Laboratories, Inc. (Palo Alto, CA). Cloning of NTRPCOLCE1—The DNA segment encoding the NTR domain of human PCOLCE1 was amplified by PCR from human liver cDNA, using the 5′-GCGAATTCATATGATATCTCCTGATGCACCCACCTG-3′ sense and 5′-GGCGAAGCTTGGATCCCTACACAGGTTGAGAGGGGCA-3′ antisense primers. The amplified DNA was digested with EcoRI and HindIII restriction endonucleases and ligated into M13mp18 Rf digested with the same enzymes. The sequence of the cloned DNA was determined by dideoxy sequencing on both strands. The expression vector pET15b was digested with XhoI, blunt-ended by filling in with DNA polymerase I Klenow fragment, and digested with BamHI restriction endonuclease. The DNA fragment encoding the NTRPCOLCE1 domain was excised from the sequencing vector with EcoRV and BamHI and ligated into the digested vector. E. coli BL21(DE3) cells were transformed with the ligation mixture and plated on 2TY medium containing 100 μg/ml ampicillin. The sequences of the NTRPCOLCE1 inserts of the plasmids were verified by dideoxy sequencing after PCR amplification. Expression of NTRPCOLCE1—Ampicillin-resistant E. coli colonies were screened for protein expression in small volume cultures. Recombinant clones of E. coli were grown in 2-ml aliquots of 2TY medium containing 100 μg/ml ampicillin at 37 °Ctoan A 600 of 0.4. Induction was started by addition of isopropyl-β-d-thiogalactopyranoside to a final concentration of 1 mm. The cultures were grown for3hat37 °C, and the expression of recombinant protein was monitored by analyzing the cell lysates with SDS-PAGE. The clone secreting the largest amount of recombinant protein was chosen for large scale protein expression. 3 liters of 2TY medium containing 100 μg/ml ampicillin were inoculated with one colony of the appropriate E. coli clone and grown at 37 °Ctoan A 600 = 0.6. Cells were induced by addition of isopropyl-β-d-thiogalactopyranoside to a final concentration of 1 mm and were grown for 3 h at 37 °C. Isotopic Labeling of NTRPCOLCE1—Cells from 100-ml overnight cultures of the appropriate E. coli clone were collected by centrifugation (15 min 5000 × g), and the pellet was suspended in 3000 ml of M9 minimal medium containing 10 mm lactose, 100 μg/ml ampicillin, and 1 g/liter 15NH4Cl and grown at 37 °C to an A 600 of about 1.2 to 1.6. Protein Purification—The cells were harvested by centrifugation with 5000 × g for 15 min, and the pellet was resuspended in 180 ml of 10 mm Tris-HCl, 5 mm MgCl2, pH 8.0. The cells were digested by addition of lysozyme to a final concentration of 0.1 mg/ml at 25 °C for 1 h and subsequently disrupted with an MSE ultrasonicator. Ribonuclease A was added to the suspension to a final concentration of 0.015 mg/ml, and the suspension was incubated for3hat37 °C. The insoluble material (containing the cloned protein in inclusion bodies) was collected by centrifugation (5,000 × g, 15 min) and washed three times with 150 ml of 10 mm Tris-HCl, 5 mm EDTA, 0.05% Triton X-100, pH 8.0 buffer. The pellet was dissolved in 25 ml of 100 mm Tris-HCl, 8 m urea, 5 mm EDTA, 100 mm DTE and was incubated overnight at 25 °C with constant stirring. Insoluble material was removed by centrifugation (5,000 × g, 15 min), and the solubilized proteins were chromatographed on a Sephacryl S300 column equilibrated with 100 mm Tris-HCl, 8 m urea, 10 mm EDTA, 0.1% 2-mercaptoethanol, pH 8.0. Fractions containing the pure protein were identified by SDS-PAGE, pooled, diluted 10-fold with 100 mm Tris-HCl, 8 m urea, 10 mm EDTA, 0.1% 2-mercaptoethanol, and dialyzed extensively against 100 mm ammonium bicarbonate buffer, pH 8.0, at 25 °C. Dialyzed protein was filtered through a membrane (pore size 0.2 μm) and chromatographed on a 10-ml nickel-chelate column. The column was washed first with 100 ml of binding buffer (20 mm Tris, 500 mm NaCl, 5 mm imidazole, pH 7.9), then with 100 ml of wash buffer (20 mm Tris, 500 mm NaCl, 60 mm imidazole, pH 7.9), and finally with 100 ml of elution buffer (20 mm Tris-HCl, 500 mm NaCl, 1 m imidazole, pH 7.9). Fractions of 1 ml were collected and analyzed by SDS-PAGE. Fractions containing pure recombinant NTRPCOLCE1 were pooled, dialyzed against 100 mm ammonium bicarbonate, pH 8.0 buffer, and lyophilized. The lyophilized sample was dissolved in 100 mm ammonium bicarbonate, pH 8.0, chromatographed on a Sephadex G-75 column, and desalted on a Sephadex G-25 column. For enzymatic activity tests, the N-terminal His tag was removed by thrombin digestion. Recombinant protein was dissolved in 5 ml of PBS buffer (140 mm NaCl, 2.7 mm KCl, 10 mm Na2HPO4, 1.8 mm KH2PO4, pH 7.3) at a final concentration of 1 mg/ml and incubated with 2 units of thrombin at 25 °C for 24 h. The reaction was arrested with 2 mm phenylmethylsulfonyl fluoride, and the digest was applied onto a nickel chelate column. Fractions containing the truncated protein were pooled, dialyzed against 100 mm ammonium bicarbonate buffer at pH 8.0, and lyophilized. The lyophilized sample was dissolved in 100 mm ammonium bicarbonate, pH 8.0, and desalted on a Sephadex G-25 column. The concentration of recombinant NTRPCOLCE1 was determined using a calculated extinction coefficient of 4200 m–1 cm–1 at 280 nm. N-terminal sequencing was performed on an Applied Biosystems 471A protein sequencer with an on-line ABI 120A phenylthiohydantoin analyzer. Typical yields before thrombin cleavage were 5 mg of purified NTRPCOLCE1 per liter of cell culture. Activation of MMP-1 Proenzyme—10 μl of 0.1 μg/μl MMP-1 proenzyme was diluted in 40 μl of 25 mm HEPES, 5 mm CaCl2, 20% glycerol, 0.005% Brij 35, pH 7.5 buffer, and 5 μl of 50 mm APMA dissolved in 0.1 m NaOH was added to the enzyme solution. The mixture was incubated at 37 °C for 3 h. Fluorescence Measurements—Fluorescence measurements were made with a Fluoromax-3 spectrofluorometer (Jobin Yvon, Edison, NJ) equipped with a thermoregulated sample chamber. The respective excitation (emission) wavelengths in the MMP assays were 280 (360) and 340 (490) nm in the TACE/ADAM17 measurements. 2-nm slits were used for both monochromators. Fluorescence assays were performed in 100 mm Tris-HCl, pH 8.0 (TACE/ADAM17), or 200 mm NaCl, 5 mm NaCl, 0.05% Brij 35, 50 mm Tris-HCl buffers (MMPs) at 25 °C. The concentrations of the enzymes were 2–3 nm, and the thrombin-digested recombinant protein was used at concentrations of 1–15 μm. The concentrations of the fluorogenic substrates of TACE/ADAM17 (DABCYL-Leu-Ala-Gln-Ala-Val-Arg-Ser-Ser-Ser-Arg-EDANS) and MMPs (DNP-Pro-Leu-Gly-Leu-Trp-Ala-d-Arg-NH2) were 5.3 and 5.6 μm, respectively. The enzyme reactions were followed for 10 min. Sequence Analyses—The sequences of human (Q15113; O14550), mouse (Q61398; O35113), and rat (O08628) PCOLCE1 sequences, human (Q9UKZ9) and mouse (Q8r4w6) PCOLCE2 sequences, and the sequence of the PCOLCE-related protein of the pufferfish Fugu rubripes (AF016494) were from the NCBI data bases. Iterative homology searches with PCOLCE1 and PCOLCE2 as query sequences identified several ESTs corresponding to PCOLCE1 of pig (e.g. BI345440); PCOLCE2 of cow (e.g. BE846217 and BE846216); PCOLCE1 of the frog Xenopus laevis (e.g. BC046734) and PCOLCE2 of the frog Xenopus tropicalis (e.g. AL628110T); and ESTs of PCOLCE-related proteins of zebrafish (e.g. BM316915, AW232258), salmon (e.g. CA055431, CA050779), and rainbow trout (e.g. BX082880 and BX081250). Multiple alignments of the amino acid sequences of NTR domains were constructed using ClustalW (13Thompson J.D. Higgins D.G. Gibson T. Nucleic Acids Res. 1994; 12: 4673-4680Crossref Scopus (56002) Google Scholar). Enzyme Assays—MMP-1 proenzyme was activated according to the protocol of the manufacturer as follows. 10 μl of 0.1 μg/μl MMP-1 proenzyme was diluted in 40 μl of 25 mm HEPES, 5 mm CaCl2, 20% glycerol, 0.005% Brij 35 (Merck), pH 7.5 buffer; 5 μl of 50 mm APMA dissolved in 0.1 m NaOH was added to the enzyme solution, and the mixture was incubated at 37 °C for 3 h. The activities of MMP-1, MMP-2, MMP-3, and MMP-9 (2–3 nm) were determined using DNP-Pro-Leu-Gly-Leu-Trp-Ala-d-Arg-NH2 as substrate (5.6 μm) in 200 mm NaCl, 5 mm NaCl, 0.05% Brij 35, 50 mm Tris-HCl pH 8.0 buffer, at 25 °C. TACE/ADAM17 (2 nm) was assayed with DABCYL-Leu-Ala-Gln-Ala-Val-Arg-Ser-Ser-Ser-Arg-EDANS (5.3 μm) in 100 mm Tris-HCl, pH 8.0, at 25 °C. The reaction was followed for 10 min in the Fluoromax-3 spectrofluorometer. The activity of bovine trypsin and human plasma kallikrein on synthetic peptide-pNA substrates was monitored spectrophotometrically using a Cary 300 Scan spectrophotometer. Hydrolysis of peptide-pNA conjugates was monitored at 410 nm, and the initial rates of the reaction were determined. In the case of human plasma kallikrein the enzyme (3 nm) was preincubated for 30 min at 37 °Cin50 mm Tris, 100 mm NaCl, 2 mm CaCl2, 0.01% Triton X-100, pH 7.5 buffer, in the presence of increasing concentrations of thrombin-digested recombinant NTRPCOLCE1 (10–60 μm), and the reactions were initiated by adding the substrate d-Pro-Phe-Arg-pNA at 650 μm final concentration. Bovine trypsin (30 nm) was preincubated for 5 min at 37 °C in 25 mm Tris, 5 mm CaCl2, pH 7.5 buffer, with thrombin-digested recombinant NTRPCOLCE1 (1–10 μm), and the reaction was initiated by adding the substrate Bz-Phe-Val-Arg-pNA at 100 μm final concentration. NMR Measurements—NMR spectra were recorded at pH 5.5 and 7.5, 28 °C, using ∼ 1 mm solutions of the NTRPCOLCE1 construct of Fig. 1A including the His tag. Samples were prepared in 90% H2O/10% D2O or 100% D2O and measured at 1H NMR frequencies of 600 and 800 MHz on Bruker DMX 600 and Varian Unity INOVA 800 NMR spectrometers, respectively. The final structure determination was based on data recorded at pH 5.5. Sequence-specific resonance assignments (Fig. 2) were obtained from double-quantum filtered two-dimensional correlation spectroscopy (DQF-COSY), clean-TOCSY (70 ms mixing time), and NOESY (40 ms and 100 ms mixing time) spectra, recorded with unlabeled protein in H2O and D2O solution at 600 and 800 MHz. In addition, three-dimensional NOESY-15N-HSQC (60-ms mixing time) and TOCSY-15N-HSQC (80-ms mixing time) experiments were recorded in H2O solution on a 15N-labeled protein sample. Most of the NOE restraints were collected from a homonuclear NOESY spectrum recorded at 800 MHz with a sample of unlabeled PCOLCE1 in H2O at pH 5.5 (40-ms mixing time, t 1max = 100 ms and t 2max = 227.5 ms, 3:9:19 water suppression, 5 days total recording time). The NOE assignments were facilitated by comparison with data recorded at pH 7.5. Stereospecific assignments of CβH2-protons were obtained by a HNHB spectrum. 3 J(HN,Hα) couplings were measured by a CT-HMQC-HN experiment (14Ponstingl H. Otting G. J. Biomol. NMR. 1998; 12: 319-324Crossref PubMed Scopus (25) Google Scholar) as well as by using the program INFIT to fit the line shapes observed in a 15N-HSQC spectrum recorded with a purge-pulse (15Szyperski T. Güntert P. Otting G. Wüthrich K. J. Magn. Reson. 1992; 99: 552-560Google Scholar). Residual 1H-15N dipolar couplings were measured in a liquid crystal formed with 8% C12E6/n-hexanol (16Rückert M. Otting G. J. Am. Chem. Soc. 2000; 122: 7793-7797Crossref Scopus (541) Google Scholar), using a 15N-HSQC spectrum with α/β half-filter (17Andersson P. Weigelt J. Otting G. J. Biomol. NMR. 1998; 12: 435-441Crossref PubMed Scopus (106) Google Scholar).Fig. 215N-HSQC spectrum of the NTRPCOLCE1 module. The spectrum was recorded using a 1 mm sample of uniformly 15N-labeled NTRPCOLCE1 at pH 5.5 and 28 °C at a 1H NMR frequency of 600 MHz. Backbone resonances are assigned with the amino acid types and sequence numbers. Side chain resonances are labeled with lowercase letters.View Large Image Figure ViewerDownload Hi-res image Download (PPT) NMR Spectral Evaluation—The NMR data were processed with the program PROSA (18Güntert P. Dötsch V. Wider G. Wüthrich K. J. Biomol. NMR. 1992; 2: 619-629Crossref Scopus (280) Google Scholar). The cross-peaks in the NOESY spectra were assigned and integrated using the program XEASY (19Bartels C. Xia T. Güntert P. Billeter M. Wüthrich K. J. Biomol. NMR. 1995; 5: 1-10Crossref PubMed Scopus (1607) Google Scholar). 3 J(Hα,Hβ) couplings were estimated as 11.0 and 4.0 Hz (±3.0 Hz), respectively, when COSY, TOCSY, and NOESY cross-peaks indicated the presence of large and small couplings, respectively, together with staggered conformations around the Cα-Cβ bond. Structure Calculations and Evaluation—The NMR structure was calculated using the program DYANA (20Güntert P. Mumenthaler C. Wüthrich K. J. Mol. Biol. 1997; 273: 283-298Crossref PubMed Scopus (2558) Google Scholar) starting from 50 random conformers. As no long range NOE could be observed for the first 23 residues, only residues 24–154 were included in the structure calculations. The 20 conformers with the lowest residual restraint violations were energy-minimized in water using the program OPAL (21Luginbühl P. Güntert P. Billeter M. Wüthrich K. J. Biomol. NMR. 1996; 8: 136-146Crossref PubMed Google Scholar) with standard parameters. The program PALES was used for a best fit of experimental and residual 1H-15N dipolar couplings predicted from the structure (22Zweckstetter M. Bax A. J. Am. Chem. Soc. 2000; 122: 3791-3792Crossref Scopus (600) Google Scholar). The Ramachandran plot was analyzed using PROCHECK-NMR (23Laskowski M. Kato I. Ardelt W. Cook J. Denton A. Empie M.W. Kohr W.J. Park S.J. Parks K. Schatzley B.L. Schoenberger O.L. Tashiro M. Vichot G. Whatley H.E. Wieczorek A. Wieczorek M. Biochemistry. 1987; 26: 202-221Crossref PubMed Scopus (194) Google Scholar). Table I shows an overview of the restraints used and structural statistics. Secondary structure elements and root mean square deviation (r.m.s.d.) values were calculated using the program MOLMOL (24Koradi R. Billeter M. Wüthrich K. J. Mol. Graphics. 1996; 14: 51-55Crossref PubMed Scopus (6498) Google Scholar) which was also used to create Figs. 3 and 4. Side chain solvent accessibilities were measured with a spherical probe of 1.4 Å radius and calculated in percent of the accessibilities measured for a fully extended side chain of residue X in a helical Gly-X-Gly peptide (25Sevilla-Sierra P. Otting G. Wüthrich K. J. Mol. Biol. 1993; 235: 1003-1020Crossref Scopus (48) Google Scholar). The values obtained were averaged over the 20 NMR conformers.Table IStructural characteristics for the NMR conformers of the NTR PCOLCEI moduleNo. assigned NOE cross-peaks2260No. non-redundant NOE upper distance limits1486No. scalar coupling constantsa98 3 J(HN,Hα), 131 3 J(Hα,Hβ), 141 3 J(15N,Hβ).370No. dihedral-angle restraints341Intra-protein AMBER energy (kcal/mol)-4112 ± 72Maximum NOE-restraint violations (Å)0.10 ± 0.00Maximum dihedral-angle restraint violations (°)3.0 ± 0.6r.m.s.d. to the mean for N,Ca98 3 J(HN,Hα), 131 3 J(Hα,Hβ), 141 3 J(15N,Hβ). and C′ (Å)bFor residues 30-149.0.52 ± 0.09r.m.s.d. to the mean for all heavy atoms (Å)bFor residues 30-149.0.86 ± 0.09Ramachandran plot appearancecFrom PROCHECK-NMR (23).Most favored regions (%)85.4Additionally allowed regions (%)13.6Generously allowed regions (%)1.0Disallowed regions (%)0.0a 98 3 J(HN,Hα), 131 3 J(Hα,Hβ), 141 3 J(15N,Hβ).b For residues 30-149.c From PROCHECK-NMR (23Laskowski M. Kato I. Ardelt W. Cook J. Denton A. Empie M.W. Kohr W.J. Park S.J. Parks K. Schatzley B.L. Schoenberger O.L. Tashiro M. Vichot G. Whatley H.E. Wieczorek A. Wieczorek M. Biochemistry. 1987; 26: 202-221Crossref PubMed Scopus (194) Google Scholar). Open table in a new tab Fig. 4Comparison of the NTRPCOLCE1 domain with BPTI and PSTI. A, stereo view of backbone traces of the NTRPCOLCE1 domain (red) superimposed onto the trypsin inhibitors BPTI (cyan; PDB code 1BTH) (37van de Locht A. Bode W. Huber R. Le Bonniec B.F. Stone S.R. Esmon C.T. Stubbs M.T. EMBO J. 1997; 16: 2977-2984Crossref PubMed Scopus (89) Google Scholar) and PSTI (magenta; PDB code 1TGS) (38Bolognesi M. Gatti G. Menagatti E. Guarneri M. Marquart M. Papamokos E. Huber R. J. Mol. Biol. 1982; 162: 839-868Crossref PubMed Scopus (179) Google Scholar). The trypsinogen molecule present in the 1TGS coordinate set is shown truncated (blue). The superposition of BPTI and PSTI was achieved by superposition of the proteinases in the BPTI·thrombin-E192Q and PSTI·trypsinogen complexes, respectively. The NTRPCOLCE1 domain was superimposed for best fit of the backbone surrounding Lys-32. The arrow identifies the P1 site in BPTI and PSTI. In addition, two selected residues in the NTRPCOLCE1 domain are labeled. B, stereo view of a superposition of the peptide segment 30–36 of the NTRPCOLCE1 domain (using the conformer closest to the mean structure in this segment; backbone in red), the proteinase-binding peptide segment 13–19 of BPTI (backbone in cyan), and the proteinase-binding peptide segment 16–22 of PSTI (backbone in magenta). The Cα atoms of the residues in the P1–P3 and P1′–P3′ sites of the inhibitors are identified. Spheres mark the N- and C-atoms of the N- and C-terminal ends, respectively, of the polypeptide segments. The following colors were used for the side chains: blue, Arg, Lys; yellow, Ala, Cys, Ile, Pro; gray, Asn, Gln, Thr, Tyr. C, sequence alignment of the inhibitor segments shown in B. Boxes identify residues with closely superimposable backbones.View Large Image Figure ViewerDownload Hi-res image Download (PPT) NMR Spectra—The line widths observed in the NMR spectra of the NTRPCOLCE1 domain were characteristic of a monomeric protein (Fig. 2). Particularly narrow resonances were observed for the N-terminal 24 and the C-terminal 4 residues, indicating increased mobility for these polypeptide segments on a nanosecond time scale. Pro-25 and Pro-150 thus mark the boundaries of the structured domain. The amide proton resonances of the N-terminal 11 residues and of His-20 could not be observed, presumably because of exchange broadening. No NOEs could be observed for the side chain NH2 resonances of Gln-33, Gln-100, and Gln-152, prohibiting their assignment. The protein contains 17 proline residues. Because of spectral overlap, only incomplete resonance assignments were obtained for the proline residues 25, 86, 103, 125, and 126. Pro-102 is the only residue for which a cis-peptide bond was identified. Besides Pro-126, this is the only proline residue which is totally conserved in all known PCOLCE proteins (Fig. 1B). Structural Statistics—The number of restraints and the final r.m.s.d. values are characteristic of a well determined structure. No residue was found in the forbidden region of the R
Referência(s)