Crystal Structure of a Bacterial Endospore Coat Component
2003; Elsevier BV; Volume: 278; Issue: 21 Linguagem: Inglês
10.1074/jbc.m301251200
ISSN1083-351X
AutoresFrancisco J. Enguita, Lı́gia O. Martins, Adriano O. Henriques, M.A. Carrondo,
Tópico(s)Biotin and Related Studies
ResumoEndospores produced by the Gram-positive soil bacterium Bacillus subtilis are shielded by a proteinaceous coat formed by over 30 structural components, which self-assemble into a lamellar inner coat and a thicker striated electrodense outer coat. The 65-kDa CotA protein is an abundant component of the outer coat layer. CotA is a highly thermostable laccase, assembly of which into the coat is required for spore resistance against hydrogen peroxide and UV light. Here, we report the structure of CotA at 1.7-Å resolution, as determined by x-ray crystallography. This is the first structure of an endospore coat component, and also the first structure of a bacterial laccase. The overall fold of CotA comprises three cupredoxin-like domains and includes one mononuclear and one trinuclear copper center. This arrangement is similar to that of other multicopper oxidases and most similar to that of the copper tolerance protein CueO of Escherichia coli. However, the three cupredoxin domains in CotA are further linked by external interdomain loops, which increase the packing level of the structure. We propose that these interdomain loops contribute to the remarkable thermostability of the enzyme, but our results suggest that additional factors are likely to play a role. Comparisons with the structure of other monomeric multicopper oxidases containing four copper atoms suggest that CotA may accept the largest substrates of any known laccase. Moreover, and unlike other laccases, CotA appears to have a flexible lidlike region close to the substrate-binding site that may mediate substrate accessibility. The implications of these findings for the properties of CotA, its assembly and the properties of the bacterial spore coat structure are discussed. Endospores produced by the Gram-positive soil bacterium Bacillus subtilis are shielded by a proteinaceous coat formed by over 30 structural components, which self-assemble into a lamellar inner coat and a thicker striated electrodense outer coat. The 65-kDa CotA protein is an abundant component of the outer coat layer. CotA is a highly thermostable laccase, assembly of which into the coat is required for spore resistance against hydrogen peroxide and UV light. Here, we report the structure of CotA at 1.7-Å resolution, as determined by x-ray crystallography. This is the first structure of an endospore coat component, and also the first structure of a bacterial laccase. The overall fold of CotA comprises three cupredoxin-like domains and includes one mononuclear and one trinuclear copper center. This arrangement is similar to that of other multicopper oxidases and most similar to that of the copper tolerance protein CueO of Escherichia coli. However, the three cupredoxin domains in CotA are further linked by external interdomain loops, which increase the packing level of the structure. We propose that these interdomain loops contribute to the remarkable thermostability of the enzyme, but our results suggest that additional factors are likely to play a role. Comparisons with the structure of other monomeric multicopper oxidases containing four copper atoms suggest that CotA may accept the largest substrates of any known laccase. Moreover, and unlike other laccases, CotA appears to have a flexible lidlike region close to the substrate-binding site that may mediate substrate accessibility. The implications of these findings for the properties of CotA, its assembly and the properties of the bacterial spore coat structure are discussed. Bacterial endospores are differentiated cell types that can withstand exposure to a wide range of physical agents including heat, desiccation, radiation, and UV light, and to chemicals such as hydrogen peroxide and lysozyme, at levels that would promptly destroy the corresponding vegetative cells (1Driks A. Microbiol. Mol. Biol. Rev. 1999; 63: 1-20Crossref PubMed Google Scholar, 2Henriques A.O. Moran C.P.J. Methods Companion Methods Enzymol. 2000; 20: 95-110Crossref Scopus (177) Google Scholar). One of the parameters that decisively contributes to their remarkable resistance properties is the organization and composition of the protective layers that encase the mature spore (1Driks A. Microbiol. Mol. Biol. Rev. 1999; 63: 1-20Crossref PubMed Google Scholar, 2Henriques A.O. Moran C.P.J. Methods Companion Methods Enzymol. 2000; 20: 95-110Crossref Scopus (177) Google Scholar). The spore core that contains a copy of the genome is surrounded by a layer of modified peptidoglycan called the cortex, which is essential for heat resistance. The cortex is protected from the action of lytic enzymes by a proteinaceous coat, which also confers resistance to noxious chemicals and to UV light, and allows the prompt response of spores to germinants (1Driks A. Microbiol. Mol. Biol. Rev. 1999; 63: 1-20Crossref PubMed Google Scholar, 2Henriques A.O. Moran C.P.J. Methods Companion Methods Enzymol. 2000; 20: 95-110Crossref Scopus (177) Google Scholar, 3Hullo M.-F. Moszer I. Danchin A. Martin-Verstraete I. J. Bacteriol. 2001; 183: 5426-5430Crossref PubMed Scopus (323) Google Scholar). In the model organism Bacillus subtilis, the spore coat is composed of over 30 different protein components, which are arranged in a lamellar inner coat and a striated electrodense outer coat (1Driks A. Microbiol. Mol. Biol. Rev. 1999; 63: 1-20Crossref PubMed Google Scholar, 2Henriques A.O. Moran C.P.J. Methods Companion Methods Enzymol. 2000; 20: 95-110Crossref Scopus (177) Google Scholar). Synthesis of the coat polypeptides is temporally and spatially regulated by the successive appearance of four mother cell-specific transcriptional regulators, in the order σE, SpoIIID, σK, and GerE (1Driks A. Microbiol. Mol. Biol. Rev. 1999; 63: 1-20Crossref PubMed Google Scholar, 2Henriques A.O. Moran C.P.J. Methods Companion Methods Enzymol. 2000; 20: 95-110Crossref Scopus (177) Google Scholar, 4Ichikawa H. Kroos L. J. Biol. Chem. 2000; 275: 13849-13855Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar, 5Kroos L. Yu Y.T. Curr. Opin. Microbiol. 2000; 3: 553-560Crossref PubMed Scopus (51) Google Scholar). However, the ordered assembly of the coat components appears to rely mostly upon post-transcriptional and post-translational mechanisms such as alternative translation initiation, protein secretion, cross-linking, or proteolysis, which enforce the correct interactions among the various coat components (1Driks A. Microbiol. Mol. Biol. Rev. 1999; 63: 1-20Crossref PubMed Google Scholar, 2Henriques A.O. Moran C.P.J. Methods Companion Methods Enzymol. 2000; 20: 95-110Crossref Scopus (177) Google Scholar, 6Ozin A. Costa T.V. Henriques A.O. Moran C.P.J. J. Bacteriol. 2001; 183: 2032-2040Crossref PubMed Scopus (21) Google Scholar). Assembly of the spore coat also relies on the action of a class of unique morphogenetic proteins, which act by guiding the assembly of several coat components (1Driks A. Microbiol. Mol. Biol. Rev. 1999; 63: 1-20Crossref PubMed Google Scholar, 2Henriques A.O. Moran C.P.J. Methods Companion Methods Enzymol. 2000; 20: 95-110Crossref Scopus (177) Google Scholar, 7Ozin A.J. Henriques A.O. Hi H. Moran C.P.J. J. Bacteriol. 2000; 182: 1828-1833Crossref PubMed Scopus (74) Google Scholar). CotE, for example, is a morphogenetic protein required for the assembly of the spore outer coat layer (8Zheng L. Donovan W.P. Fitz-James P.C. Losick R. Genes Dev. 1988; 2: 1047-1054Crossref PubMed Scopus (167) Google Scholar) and may act in part by directly interacting with and recruiting several of the outer coat proteins (9Little S. Driks A. Mol. Microbiol. 2001; 42: 1107-1120Crossref PubMed Scopus (69) Google Scholar).Despite its importance as a model system for studying the assembly of a multiprotein structure, as a platform for the display of heterologous enzymes or antigens, in pathogenesis and host immune response, and possibly in mediating the germination of spores in the gastrointestinal tract (2Henriques A.O. Moran C.P.J. Methods Companion Methods Enzymol. 2000; 20: 95-110Crossref Scopus (177) Google Scholar, 10Isticato R. Cangiano G. Tran H.T. Ciabattini A. Medaglini D. Oggioni M.R. Felice M.d. Pozzi G. Ricca E. J. Bacteriol. 2001; 183: 6294-6301Crossref PubMed Scopus (169) Google Scholar, 11Brossier F. Levy M. Mock M. Infect. Immun. 2002; 70: 661-664Crossref PubMed Google Scholar), our knowledge of the molecular mechanisms underlying the assembly of the bacterial spore coat is still scarce. The function of individual coat components is largely unknown, and in only a few cases have the interactions relevant for their assembly been unraveled. For example, a putative manganese catalase, CotJC, interacts with a smaller protein (CotJA) to form a complex that is targeted to the inner coat layers (12Henriques A.O. Beall B.W. Roland K. Moran C.P.J. J. Bacteriol. 1995; 177: 3394-3406Crossref PubMed Google Scholar, 13Seyler R. Henriques A.O. Ozin A. Moran C.P.J. Mol. Microbiol. 1997; 25: 955-966Crossref PubMed Scopus (47) Google Scholar). Another case involves the SafA and SpoVID morphogenetic proteins. SpoVID and SafA interact directly, and the targeting of SafA to the surface of the developing spore requires SpoVID (6Ozin A. Costa T.V. Henriques A.O. Moran C.P.J. J. Bacteriol. 2001; 183: 2032-2040Crossref PubMed Scopus (21) Google Scholar, 14Ozin A.J. Samford C.S. Henriques A.O. Moran C.P.J. J. Bacteriol. 2001; 183: 3041-3049Crossref PubMed Scopus (60) Google Scholar). However, the nature of these interactions and the structural basis for the assembly of the resulting complexes are unknown. Evidently, more detailed studies are needed to understand the mechanisms by which specific proteins or protein complexes are targeted to the nascent coat. In an attempt to begin addressing these questions, we have initiated the structural characterization of selected spore coat components. The CotA 1The abbreviations used are: CotA, B. subtilis CotA (PDB code 1GSK); CueO, E. coli CueO (PDB code 1KV7); CcLa, C. cinereus laccase (PDB code 1A65); TvLa, T. versicolor laccase (PDB code 1KYA); MaLa, M. albomyces laccase (PDB code 1GW0); Asox, zucchini ascorbate oxidase (PDB code 1AOZ); PDB, Protein Data Bank; MAD, multiwavelength anomalous dispersion; OSP, normalized occluded surface packing value.1The abbreviations used are: CotA, B. subtilis CotA (PDB code 1GSK); CueO, E. coli CueO (PDB code 1KV7); CcLa, C. cinereus laccase (PDB code 1A65); TvLa, T. versicolor laccase (PDB code 1KYA); MaLa, M. albomyces laccase (PDB code 1GW0); Asox, zucchini ascorbate oxidase (PDB code 1AOZ); PDB, Protein Data Bank; MAD, multiwavelength anomalous dispersion; OSP, normalized occluded surface packing value. protein is a 65-kDa abundant component of the outer coat layer (15Donovan W. Zheng L.B. Sandman K. Losick R. J. Mol. Biol. 1987; 196: 1-10Crossref PubMed Scopus (136) Google Scholar), recently shown to possess copper-dependent laccase activity and to be highly thermostable (3Hullo M.-F. Moszer I. Danchin A. Martin-Verstraete I. J. Bacteriol. 2001; 183: 5426-5430Crossref PubMed Scopus (323) Google Scholar, 16Martins L.M. Soares C.M. Pereira M.M. Teixeira M. Jones G.H. Henriques A.O. J. Biol. Chem. 2002; 277: 18849-18859Abstract Full Text Full Text PDF PubMed Scopus (442) Google Scholar).Laccases are polyphenol oxidases, able to oxidize a wide range of substrates, including xenobiotic compounds such as methoxyphenols, anilines, and benzenethiols (17Xu F. Biochemistry. 1996; 35: 7608-7614Crossref PubMed Scopus (605) Google Scholar), and belong to the multicopper oxidase family, characterized by the presence of one mononuclear and one trinuclear copper site. The multicopper oxidase family also groups ascorbate oxidase (18Messerschmidt A. Ladenstein R. Huber R. Bolognesi M. Avigliano L. Petruzzelli R. Rossi A. Finazzi-Agro A. J. Mol. Biol. 1992; 224: 179-205Crossref PubMed Scopus (440) Google Scholar), ceruloplasmin (19Zaitsev I. Zaitsev V. Card G. Moshkov K. Bax B. Ralph A. Lindley P. J. Biol. Inorg. Chem. 1996; 1: 15-23Crossref Scopus (352) Google Scholar), various manganese oxidases (20Francis C.A. Tebo B.M. Appl. Environ. Microbiol. 2001; 67: 4272-4278Crossref PubMed Scopus (106) Google Scholar), and other enzymes involved in copper and iron metabolism, including CueO from Escherichia coli (21Roberts S.A. Weichsel A. Grass G. Thakali K. Hazzard J.T. Tollin G. Rensing C. Montfort W.R. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 2766-2771Crossref PubMed Scopus (279) Google Scholar) and Fet3p from Saccharomyces cerevisiae (22Blackburn N.J. Ralle M. Hassett R. Kosman D.J. Biochemistry. 2000; 56: 2316-2324Crossref Scopus (62) Google Scholar). In plants laccases are involved in cell wall formation, whereas in fungi they are involved in lignin degradation, detoxification and pathogenesis (23McGuirl M.A. Dooley D.M. Curr. Opin. Chem. Biol. 1999; 3: 138-144Crossref PubMed Scopus (109) Google Scholar). Besides B. subtilis, laccase activity was also found in two other bacterial species, the soil bacterium Azospirillum lipoferum (24Alexandre G. Bally R. FEMS Microbiol. Lett. 1999; 174: 371-378Crossref PubMed Google Scholar), and the marine bacteria Marinomonas mediterranea (25Sanchez-Amat A. Solano F. Biochem. Biophys. Res. Commun. 1997; 240: 787-792Crossref PubMed Scopus (94) Google Scholar). Moreover, putative laccase-like multicopper oxidases have been detected in the genomes of other bacterial species, suggesting that laccases are widespread in bacteria (26Alexandre G. Zhulin I.B. Trends Biotech. 2000; 18: 41-42Abstract Full Text Full Text PDF PubMed Scopus (249) Google Scholar). However, of the three laccases that have been structurally characterized, those from Coprinus cinereus (27Ducros V. Brzozowski A.M. Wilson K.S. Brown S.H. østergaard P. Schneider P. Yaver D.S. Pedersen A.H. Davies G.J. Nat. Struct. Biol. 1998; 5: 310-316Crossref PubMed Scopus (347) Google Scholar), from Trametes versicolor (28Bertrand T. Jolivalt C. Briozzo P. Caminade E. Joly N. Madzak C. Mougin C. Biochemistry. 2002; 41: 7325-7333Crossref PubMed Scopus (422) Google Scholar), and from Melanocarpus albomyces (29Hakulinen N. Kiiskinen L.-L. Kruus K. Saloheimo M. Paananen A. Koivula A. Rouvinen J. Nat. Struct. Biol. 2002; 9: 601-605PubMed Google Scholar), none is of bacterial origin. Laccases have been the subject of increasing attention because of their established or potential novel uses in biotechnology (see Ref. 28Bertrand T. Jolivalt C. Briozzo P. Caminade E. Joly N. Madzak C. Mougin C. Biochemistry. 2002; 41: 7325-7333Crossref PubMed Scopus (422) Google Scholar and references therein), but they are also an important system for studying the mechanism of oxidation reactions involving transfer of four single electrons from the substrate to the final acceptor (30Huang H.-W. Zoppellaro G. Sakurai T. J. Biol. Chem. 1999; 274: 32718-32724Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar).The exact function of CotA within the spore coat is still not fully understood, but the assembly of CotA is essential for the full complement of spore resistance properties. Expression of the cotA gene has been classically implicated in the biosynthesis of a brownish pigment that characterizes sporulating colonies of B. subtilis, and which has properties of a melanin and appears to confer protection against UV light (3Hullo M.-F. Moszer I. Danchin A. Martin-Verstraete I. J. Bacteriol. 2001; 183: 5426-5430Crossref PubMed Scopus (323) Google Scholar, 15Donovan W. Zheng L.B. Sandman K. Losick R. J. Mol. Biol. 1987; 196: 1-10Crossref PubMed Scopus (136) Google Scholar, 31Rogolsky M. J. Bacteriol. 1968; 95: 2426-2427Crossref PubMed Google Scholar). Expression of cotA is also required for the resistance of spores to hydrogen peroxide (32Riesenman P.J. Nicholson W.L. Appl. Environ. Microbiol. 2000; 66: 620-626Crossref PubMed Scopus (231) Google Scholar). Because of its importance for the resistance properties of the spore structure, because of the importance of laccases, and unique thermostability of CotA among this class of enzymes (16Martins L.M. Soares C.M. Pereira M.M. Teixeira M. Jones G.H. Henriques A.O. J. Biol. Chem. 2002; 277: 18849-18859Abstract Full Text Full Text PDF PubMed Scopus (442) Google Scholar), we began the structural characterization of relevant components of the spore coat, by focusing our attention on CotA. In this paper we describe the three-dimensional structure of the B. subtilis CotA laccase, as determined by x-ray crystallography. This is the first report on the structure of a bacterial endospore component, and also the first structure of a bacterial laccase.EXPERIMENTAL PROCEDURESProtein Purification and Crystallization—Purification of recombinant CotA was performed essentially as described previously (16Martins L.M. Soares C.M. Pereira M.M. Teixeira M. Jones G.H. Henriques A.O. J. Biol. Chem. 2002; 277: 18849-18859Abstract Full Text Full Text PDF PubMed Scopus (442) Google Scholar), using an E. coli overproducing host. Protein crystals were growth at 293 K, by the vapor diffusion method, using 10–15 mg/ml purified CotA protein, and a reservoir solution containing 100 mm sodium citrate buffer (pH = 5.5), 15% glycerol, 12–15% isopropanol, and 12–15% polyethylene glycol 4000 (33Enguita F.J. Matias P.M. Martins L.O. Plácido D. Henriques A.O. Carrondo M.A. Acta Crystallogr. Sect. D Biol. Crystallogr. 2002; 58: 1490-1493Crossref PubMed Scopus (19) Google Scholar).Data Collection and Structure Solution—CotA crystals reached maximum dimensions of 0.2 × 0.3 × 0.5 mm, showing a hexagonal prismatic shape, and the characteristic blue color caused by the presence of type I copper centers within the protein (33Enguita F.J. Matias P.M. Martins L.O. Plácido D. Henriques A.O. Carrondo M.A. Acta Crystallogr. Sect. D Biol. Crystallogr. 2002; 58: 1490-1493Crossref PubMed Scopus (19) Google Scholar). They belong to P3121 spacegroup with cell dimensions of a = b = 101.8 Å, c = 136.1 Å, and one protein molecule per asymmetric unit, corresponding to a solvent content of 55%.Diffraction data were collected from two crystals: one considered as a "native" for high resolution data collection, and the other one for structure solution by the multiwavelength anomalous dispersion method (MAD). Data collection for the "native" data set was performed at the ID-14-EH2 beamline (ESRF, Grenoble, France), and the MAD experiment was undertaken at the BW7A beamline (EMBL, Hamburg, Germany). The structure was solved by the MAD method at the copper K edge, using the anomalous signal of the copper atoms within the structure (33Enguita F.J. Matias P.M. Martins L.O. Plácido D. Henriques A.O. Carrondo M.A. Acta Crystallogr. Sect. D Biol. Crystallogr. 2002; 58: 1490-1493Crossref PubMed Scopus (19) Google Scholar). Automated interpretation of anomalous Patterson maps by SOLVE (34Terwilliger T.C. Berendzen J. Acta Crystallogr. Sect. D Biol. Crystallogr. 1999; 55: 849-861Crossref PubMed Scopus (3219) Google Scholar) allowed the localization of two copper atoms:type I copper, and one of the copper atoms belonging to the binuclear type III center. The positions of the two remaining copper atoms were determined using the corresponding anomalous difference maps, calculated with the initial phases. Electron density maps were improved by density modification with RESOLVE (35Terwilliger T.C. Acta Crystallogr. Sect. D Biol. Crystallogr. 2001; 57: 1755-1762Crossref PubMed Scopus (166) Google Scholar).Model Building and Refinement—The original phases obtained with data to 2.65-Å resolution, derived from anomalous Patterson maps interpretation and density modification by SOLVE-RESOLVE (34Terwilliger T.C. Berendzen J. Acta Crystallogr. Sect. D Biol. Crystallogr. 1999; 55: 849-861Crossref PubMed Scopus (3219) Google Scholar, 35Terwilliger T.C. Acta Crystallogr. Sect. D Biol. Crystallogr. 2001; 57: 1755-1762Crossref PubMed Scopus (166) Google Scholar), were extended to 1.7-Å resolution using the maximum resolution data set available with DM (36Cowtan K. Joint CCP4 ESF-EACBM Newsl. Protein Crystallogr. 1994; 31: 34-38Google Scholar). Using this extended phase information, initial model building was performed automatically using ARP/WARP version 5.1 (37Perrakis A. Morris R. Lamzin V.S. Nat. Struct. Biol. 1999; 6: 458-463Crossref PubMed Scopus (2562) Google Scholar). After 50 cycles of refinement and 10 cycles of building, 93% (479 of 513 possible) of the protein residues were found, and placed in seven chains, with a global connectivity index of 0.97. At the end of the automatic building procedure, the refinement converged to an R factor of 20.2% and an Rfree factor of 24.5%. The amino acid side chains built with the side_dock script included in the ARP/WARP suite, had a confidence factor of 0.95. After this process, manual intervention was required to complete the model. Missing sections were built from 2Fo - Fc maps using Xtalview (38McRee D.E. J. Struct. Biol. 1999; 125: 156-165Crossref PubMed Scopus (2019) Google Scholar). The Xtalview suite was also employed to complete the solvent boundary of the protein model, including four glycerol molecules belonging to the crystallization buffer and located mainly at the surface of the protein.Protein chain and solvent molecules were input to REFMAC5 for refinement (39Murshudov G.N. Lebedev A. Vagin A.A. Wilson K.S. Dodson E.J. Acta Crystallogr. Sect. D Biol. Crystallogr. 1999; 55: 247-255Crossref PubMed Scopus (1006) Google Scholar). Positional and isotropic thermal parameters were refined individually for each atom to a resolution limit of 1.7 Å. During the initial refinement steps, isotropic thermal parameters of atoms belonging to the type II and III copper centers refined to values close to 80 Å2, probably because of the low copper content of the protein crystals. For this reason, occupancies for atoms belonging to the trinuclear T2/T3 copper center were maintained at 0.5. After several cycles of refinement and model checking the refinement converged to a R factor of 17.7% and a Rfree factor of 19.8%. No electron density was visible in the region comprising residues 90–96, which lie in an apparently disordered loop, the initial methionine, and also the C-terminal last three residues. The final model contained 502 of the 513 residues in the primary sequence of CotA. The stereochemistry of the final model was analyzed with PROCHECK (40Laskowsky R.A. MacArthur M.W. Moss D.S. Thornton J.M. J. Appl. Crystallogr. 1993; 26: 283-291Crossref Google Scholar) and WHATIF (41Hooft R.W.W. Vriend G. Sander C. Abola E.E. Nature. 1996; 381: 272-276Crossref PubMed Scopus (1794) Google Scholar). The overall mean B factor of the structure after refinement was 25.82 Å2, and root mean square deviations from ideal values were 0.019 Å for bond lengths, and 1.837° for bond angles (see Table I for additional details of the refinement).Table IRefinement statistics obtained using REFMAC_5 (39Murshudov G.N. Lebedev A. Vagin A.A. Wilson K.S. Dodson E.J. Acta Crystallogr. Sect. D Biol. Crystallogr. 1999; 55: 247-255Crossref PubMed Scopus (1006) Google Scholar) for the final CotA modelSpacegroupP3121Cell dimensionsa = 102.051, b = 102.051, c = 136.393, α = 90.00, β = 90.00, γ = 120.00No. of reflexions86,412No. of protein atoms4044No. of solvent atoms480No. of heterogen atoms30Resolution range for refinement (Å)87.71-1.70 (1.78-1.70)Completeness for range99.8 (89.5)Rcryst17.76 (19.85)Rfree19.85 (22.10)Overall B value of the model (Å2)25.82Correlation coefficient Fo - Fc0.965Correlation coefficient Fo - Fc Free0.959Root mean square bond lengths deviation (Å)0.019Root mean square bond angles deviation (degrees)1.837 Open table in a new tab Other Methods—Surface calculations were performed by MSMS (42Sanner F. Olson A.J. Sehner J.-H. Biopolymers. 1996; 38: 305-320Crossref PubMed Google Scholar), and putative substrate binding pockets were determined using the CASTP server (43Liang J. Edelsbrunner H. Woodward C. Protein Sci. 1998; 7: 1884-1987Crossref PubMed Scopus (859) Google Scholar). All the graphical representations of CotA or other multicopper oxidases were made using PyMol (44DeLano W.L. PyMol. DeLano Scientific, San Carlos, CA2002Google Scholar). Domain analysis based on the quantification of local atomic interactions was performed by using DOMID server (45Lu, H. (1999) bioinfo1.mbfys.lu.se/Domid/domid.html.Google Scholar). Residue packing in CotA was estimated by calculating the normalized occluded surface packing value (OSP) using by the program OS (46Pattabiraman N. Ward K.B. Fleming P.J. J. Mol. Recog. 1995; 8: 334-344Crossref PubMed Scopus (108) Google Scholar). Fig. 2 was prepared by ESPript (47Gouet P. Courcelle E. Stuart D.I. Metoz F. Bioinformatics. 1999; 15: 305-308Crossref PubMed Scopus (2505) Google Scholar) using MULTALIN as alignment program (48Corpet F. Nucleic Acids Res. 1988; 16: 10881-10890Crossref PubMed Scopus (4265) Google Scholar).RESULTSOverall Structure of CotA—CotA is a monomeric protein (16Martins L.M. Soares C.M. Pereira M.M. Teixeira M. Jones G.H. Henriques A.O. J. Biol. Chem. 2002; 277: 18849-18859Abstract Full Text Full Text PDF PubMed Scopus (442) Google Scholar) and has the overall dimensions 70 × 50 × 20 Å (Fig. 1, panel A). The overall CotA fold comprises three cupredoxin-like domains, as shown in Figs. 1 (panel A) and 2. This fold was first observed in the small blue copper proteins plastocyanin and azurin (49Murphy M.E.P. Lindley P.F. Adman E.T. Protein Sci. 1997; 6: 761-770Crossref PubMed Scopus (113) Google Scholar), and subsequently detected in other more complex multicopper enzymes (21Roberts S.A. Weichsel A. Grass G. Thakali K. Hazzard J.T. Tollin G. Rensing C. Montfort W.R. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 2766-2771Crossref PubMed Scopus (279) Google Scholar, 27Ducros V. Brzozowski A.M. Wilson K.S. Brown S.H. østergaard P. Schneider P. Yaver D.S. Pedersen A.H. Davies G.J. Nat. Struct. Biol. 1998; 5: 310-316Crossref PubMed Scopus (347) Google Scholar). The cupredoxin fold is mainly formed by an eight-stranded Greek key β-barrel, comprising two β-sheets composed by four strands, arranged in a sandwich conformation (50Lindley P.F. Bertini I. Sigel A. Sigel H. Handbook of Metalloproteins. Marcel Dekker, Inc., New York2001: 763-811Google Scholar). The first (N-terminal, domain 1; represented in blue in Fig. 1A) cupredoxin-like domain of CotA (residues 2–176; Fig. 2) has a somewhat distorted conformation in comparison with the equivalent domain in other multicopper oxidases. It comprises eight strands organized in a β-barrel form, starting with a coiled section (residues 2–25; Fig. 2) that connects domains 1 and 2, and is stabilized by hydrogen bonds, contributing to the packing between these domains. This coiled section is absent in plant and fungal multicopper oxidases such the laccase from C. cinereus (CcLa; Fig. 1B) (27Ducros V. Brzozowski A.M. Wilson K.S. Brown S.H. østergaard P. Schneider P. Yaver D.S. Pedersen A.H. Davies G.J. Nat. Struct. Biol. 1998; 5: 310-316Crossref PubMed Scopus (347) Google Scholar) and ascorbate oxidase (Asox) (18Messerschmidt A. Ladenstein R. Huber R. Bolognesi M. Avigliano L. Petruzzelli R. Rossi A. Finazzi-Agro A. J. Mol. Biol. 1992; 224: 179-205Crossref PubMed Scopus (440) Google Scholar). However, a similar coiled section is present in the E. coli CueO protein (Figs. 1C and 2) (21Roberts S.A. Weichsel A. Grass G. Thakali K. Hazzard J.T. Tollin G. Rensing C. Montfort W.R. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 2766-2771Crossref PubMed Scopus (279) Google Scholar).Fig. 1Overall structure and putative substrate binding pockets of selected multicopper oxidases. Surface calculations were performed by MSMS (42Sanner F. Olson A.J. Sehner J.-H. Biopolymers. 1996; 38: 305-320Crossref PubMed Google Scholar) and putative substrate binding pockets were determined using the CASTP server (43Liang J. Edelsbrunner H. Woodward C. Protein Sci. 1998; 7: 1884-1987Crossref PubMed Scopus (859) Google Scholar) as described under "Experimental Procedures." All the molecular representations were generated using PyMol (44DeLano W.L. PyMol. DeLano Scientific, San Carlos, CA2002Google Scholar). Left column, a rainbow-colored (from N terminus in blue to C terminus in red) ribbon representation of CotA (A), CcLa (B), and CueO (C) is shown, including the localization of the copper atoms within the structure (plotted as orange balls). In panel A, the lidlike structure over the putative substrate-binding site in CotA is visible to the right. Right column, molecular surface representation of CotA (D), CcLa (E), and CueO (F) with the putative substrate binding pocket colored in green. The view in panels D and E represents a 45° clockwise rotation relative to the left column view, and was intended to facilitate the observation of the putative substrate binding site cavities.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The overall fold of the second cupredoxin-like domain of CotA (domain 2, represented in green in Fig. 1A) comprises a β-barrel composed by 12 strands (residues 183–340; Fig. 2), very similar to the fold of domain 2 in Asox (18Messerschmidt A. Ladenstein R. Huber R. Bolognesi M. Avigliano L. Petruzzelli R. Rossi A. Finazzi-Agro A. J. Mol. Biol. 1992; 224: 179-205Crossref PubMed Scopus (440) Google Scholar). Domain 2 of CotA acts as a bridge between domains 1 and 3 (Fig. 1A), but a short α-helical fragment, encompassing residues 177–182, makes the connection between domains 1 and 2, whereas a large loop segment including residues 341–368 links domains 2 and 3 (Figs. 1A and 2). In both the structures of CotA and CueO, this region represents an external connection between domai
Referência(s)