Advertisement

The 1.92-Å Structure of Streptomyces coelicolor A3(2) CYP154C1

A NEW MONOOXYGENASE THAT FUNCTIONALIZES MACROLIDE RING SYSTEMS*
  • Larissa M. Podust
    Correspondence
    To whom correspondence should be addressed: Dept. of Biochemistry, Vanderbilt University, 23rd South at Pierce, Nashville, TN 37232-0146. Tel.: 615-343-4644; Fax: 615-322-4349
    Affiliations
    Department of Biochemistry, Vanderbilt University School of Medicine, Nashville, Tennessee 37232-0146
    Search for articles by this author
  • Youngchang Kim
    Affiliations
    Argonne National Laboratory, Structural Biology Center, Argonne, Illinois 60439
    Search for articles by this author
  • Miharu Arase
    Affiliations
    Department of Biochemistry, Vanderbilt University School of Medicine, Nashville, Tennessee 37232-0146
    Search for articles by this author
  • Benjamin A. Neely
    Affiliations
    Department of Biochemistry, Vanderbilt University School of Medicine, Nashville, Tennessee 37232-0146
    Search for articles by this author
  • Brian J. Beck
    Affiliations
    Department of Microbiology and BioTechnology Institute, University of Minnesota, Minneapolis, Minnesota 55455
    Search for articles by this author
  • Horacio Bach
    Affiliations
    Department of Microbiology and BioTechnology Institute, University of Minnesota, Minneapolis, Minnesota 55455
    Search for articles by this author
  • David H. Sherman
    Affiliations
    Department of Microbiology and BioTechnology Institute, University of Minnesota, Minneapolis, Minnesota 55455
    Search for articles by this author
  • David C. Lamb
    Affiliations
    Wolfson Laboratory of P-450 Biodiversity, Institute of Biological Sciences, University of Wales Aberystwyth, Aberystwyth, Wales SY23 3DA, United Kingdom
    Search for articles by this author
  • Steven L. Kelly
    Affiliations
    Wolfson Laboratory of P-450 Biodiversity, Institute of Biological Sciences, University of Wales Aberystwyth, Aberystwyth, Wales SY23 3DA, United Kingdom
    Search for articles by this author
  • Michael R. Waterman
    Affiliations
    Department of Biochemistry, Vanderbilt University School of Medicine, Nashville, Tennessee 37232-0146
    Search for articles by this author
  • Author Footnotes
    * This work was supported by National Institutes of Health Grants GM37942 and ES00267 (to M. R. W.), P30 ES00267 (to L. M. P.), GM48562 (to D. H. S.), by Biotechnology and Biological Sciences Research Council and a Welcome Trust Grant (to S. L. K. and D. C. L.), and by NCI Cancer Biology Training Grant CA09138 (to B. J. B.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked 舠advertisement舡 in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.The atomic coordinates and the structure factors (code ) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).
Open AccessPublished:January 07, 2003DOI:https://doi.org/10.1074/jbc.M212210200
      Evolutionary links between cytochrome P450 monooxygenases, a superfamily of extraordinarily divergent heme-thiolate proteins catalyzing a wide array of NADPH/NADH- and O2-dependent reactions, are becoming better understood because of availability of an increasing number of fully sequenced genomes. Among other reactions, P450s catalyze the site-specific oxidation of the precursors to macrolide antibiotics in the genus Streptomyces introducing regiochemical diversity into the macrolide ring system, thereby significantly increasing antibiotic activity. Developing effective uses forStreptomyces enzymes in biosynthetic processes and bioremediation requires identification and engineering of additional monooxygenases with activities toward a diverse array of small molecules. To elucidate the molecular basis for substrate specificity of oxidative enzymes toward macrolide antibiotics, the x-ray structure of CYP154C1 from Streptomyces coelicolor A3(2) was determined (Protein Data Bank code 1GWI). Relocation of certain common P450 secondary structure elements, along with a novel structural feature involving an additional ॆ-strand transforming the five-stranded ॆ-sheet into a six-stranded variant, creates an open cleft-shaped substrate-binding site between the two P450 domains. High sequence similarity to macrolide monooxygenases from other microbial species translates into catalytic activity of CYP154C1 toward both 12- and 14-membered ring macrolactonesin vitro.1GWI
      CYP
      cytochrome P450
      MES
      4-morpholineethanesulfonic acid
      The post-genomic era opens new opportunities for structural insight into the evolution of a single protein family within and between species. Cytochrome P450 (CYP)1 monooxygenases are a superfamily of heme-thiolate enzymes that are involved in a wide array of NADPH/NADH- and O2-dependent reactions (
      • Guengerich F.P.
      • MacDonald T.L.
      ). There are currently more than 2000 family members, including a large number of putative P450 open reading frames found in fully sequenced prokaryotic and eukaryotic genomes (drnelson.utmem.edu/CytochromeP450.html). Extensive studies have firmly established their role in the biosynthesis of sterols, fatty acids, and prostaglandins in animals, antibiotics, and other biologically active molecules in bacteria, fungi, and plants as well as in the metabolism of xenobiotic drugs and toxic chemicals (
      • Guengerich F.P.
      ). Accordingly, the extraordinary diversity in amino acid sequence enables wide variation in the substrates utilized and the patterns of oxidation catalyzed by these enzymes.
      A particularly rich source of CYPs is from Streptomycesspp., a group of developmentally complex, Gram-positive bacteria that are known for production of a broad array of biologically active secondary metabolites. Streptomyces coelicolor A3(2) has been investigated extensively as a model system for the study of morphological and physiological development of Streptomycesand for investigation of the genetic control of antibiotic production (
      • Hopwood D.A.
      ). Over the past decade, an increasing number ofStreptomyces spp. have been investigated because of their production of pharmaceutically important compounds including anti-cancer agents, immunosuppressants, and antibiotics. In addition,Streptomyces are being recognized as a source of versatile biocatalysts for the detoxification of hazardous chemicals (
      • Shelton D.R.
      • Khader S.
      • Karns J.S.
      • Pogell B.M.
      ,
      • Jendrossek D.
      • Tomasi G.
      • Kroppenstedt R.M.
      ,
      • Fadullon F.S.
      • Karns J.S.
      • Torrents A.
      ,
      • Obojska A.
      • Lejczak B.
      • Kubrak M.
      ,
      • Gallert C.
      ,
      • Amoroso M.J.
      • Castro G.R.
      • Duran A.
      • Peraud O.
      • Oliver G.
      • Hill R.T.
      ) in bioremediation processes.
      In an effort to analyze fully the complement of CYPs in the industrially important Streptomyces, we decided to focus on the 8.7 Mb S. coelicolor A3(2) genome whose sequence was recently completed by the Sanger Centre (
      • Bentley S.D.
      • Chater K.F.
      • Cerdeno-Tarraga A.M.
      • Challis G.L.
      • Thomson N.R.
      • James K.D.
      • Harris D.E.
      • Quail M.A.
      • Kieser H.
      • Harper D.
      • Bateman A.
      • Brown S.
      • Chandra G.
      • Chen C.W.
      • Collins M.
      • Cronin A.
      • Fraser A.
      • Goble A.
      • Hidalgo J.
      • Hornsby T.
      • Howarth S.
      • Huang C.H.
      • Kieser T.
      • Larke L.
      • Murphy L.
      • Oliver K.
      • O'Neil S.
      • Rabbinowitsch E.
      • Rajandream M.A.
      • Rutherford K.
      • Rutter S.
      • Seeger K.
      • Saunders D.
      • Sharp S.
      • Squares R.
      • Squares S.
      • Taylor K.
      • Warren T.
      • Wietzorrek A.
      • Woodward J.
      • Barrell B.G.
      • Parkhill J.
      • Hopwood D.A.
      ) (www.sanger.ac.uk/Projects/S_coelicolor/). Although 18 CYP open reading frames are found dispersed throughout the S. coelicolorA3(2) chromosome (
      • Bentley S.D.
      • Chater K.F.
      • Cerdeno-Tarraga A.M.
      • Challis G.L.
      • Thomson N.R.
      • James K.D.
      • Harris D.E.
      • Quail M.A.
      • Kieser H.
      • Harper D.
      • Bateman A.
      • Brown S.
      • Chandra G.
      • Chen C.W.
      • Collins M.
      • Cronin A.
      • Fraser A.
      • Goble A.
      • Hidalgo J.
      • Hornsby T.
      • Howarth S.
      • Huang C.H.
      • Kieser T.
      • Larke L.
      • Murphy L.
      • Oliver K.
      • O'Neil S.
      • Rabbinowitsch E.
      • Rajandream M.A.
      • Rutherford K.
      • Rutter S.
      • Seeger K.
      • Saunders D.
      • Sharp S.
      • Squares R.
      • Squares S.
      • Taylor K.
      • Warren T.
      • Wietzorrek A.
      • Woodward J.
      • Barrell B.G.
      • Parkhill J.
      • Hopwood D.A.
      ,
      • Lamb D.C.
      • Skaug T.
      • Song H.-L.
      • Jackson C.J.
      • Podust L.M.
      • Waterman M.R.
      • Kell D.B.
      • Kelly D.E.
      • Kelly S.L.
      ), the functions of the corresponding gene products and their role in various metabolic functions remain largely undefined.
      Most of the currently identified antibiotics are produced by complex biosynthetic systems comprised of clustered gene sets located contiguously on the Streptomyces chromosome (
      • Xue Y.
      • Zhao L.
      • Liu H.W.
      • Sherman D.H.
      ,
      • Bate N.
      • Butler A.R.
      • Gandecha A.R.
      • Cundliffe E.
      ,
      • Fouces R.
      • Mellado E.
      • Diez B.
      • Barredo J.L.
      ,
      • Ikeda H.
      • Nonomiya T.
      • Usami M.
      • Ohta T.
      • Omura S.
      ,
      • Pelzer S.
      • Sussmuth R.
      • Heckmann D.
      • Recktenwald J.
      • Huber P.
      • Jung G.
      • Wohlleben W.
      ,
      • Caffrey P.
      • Lynch S.
      • Flood E.
      • Finnan S.
      • Oliynyk M.
      ,
      • Huang J.
      • Lih C.J.
      • Pan K.H.
      • Cohen S.N.
      ). The clustering of secondary metabolite genes has been an aid in the isolation of P450 monooxygenases involved in antibiotic biosynthesis. Cytochrome P450 monooxygenases are particularly common in polyketide biosynthetic gene clusters, and they catalyze site-specific tailoring reactions leading to the macrolide antibiotics, including methymycin, neomethymycin, and pikromycin (
      • Xue Y.
      • Zhao L.
      • Liu H.W.
      • Sherman D.H.
      ,
      • Xue Y.
      • Wilson D.
      • Zhao L.
      • Liu H.
      • Sherman D.H.
      ,
      • Xue Y.
      • Sherman D.H.
      ,
      • Graziani E.I.
      • Cane D.E.
      • Betlach M.C.
      • Kealey J.T.
      • McDaniel R.
      ,
      • Betlach M.C.
      • Kealey J.T.
      • Ashley G.W.
      • McDaniel R.
      ), novamethymycin (
      • Zhang Q.
      • Sherman D.H.
      ), oleandomycin (
      • Rodriguez A.M.
      • Olano C.
      • Mendez C.
      • Hutchinson C.R.
      • Salas J.A.
      ), amphotericin (
      • Caffrey P.
      • Lynch S.
      • Flood E.
      • Finnan S.
      • Oliynyk M.
      ), and erythromycin (
      • Haydock S.F.
      • Dowson J.A.
      • Dhillon N.
      • Roberts G.A.
      • Cortes J.
      • Leadlay P.F.
      ,
      • Weber J.M.
      • Leung J.O.
      • Swanson S.J.
      • Idler K.B.
      • McAlpine J.B.
      ). Additionally, CYPs are involved in the formation of the anticancer agent epothilone (
      • Molnar I.
      • Schupp T.
      • Ono M.
      • Zirkle R.
      • Milnamow M.
      • Nowak-Thompson B.
      • Engel N.
      • Toupet C.
      • Stratmann A.
      • Cyr D.D.
      • Gorlach J.
      • Mayo J.M.
      • Hu A.
      • Goff S.
      • Schmid J.
      • Ligon J.M.
      ,
      • Tang L.
      • Shah S.
      • Chung L.
      • Carney J.
      • Katz L.
      • Khosla C.
      • Julien B.
      ), immunosuppressant rapamycin (
      • Molnar I.
      • Aparicio J.F.
      • Haydock S.F.
      • Khaw L.E.
      • Schwecke T.
      • Konig A.
      • Staunton J.
      • Leadlay P.F.
      ,
      • Chung L.
      • Liu L.
      • Patel S.
      • Carney J.R.
      • Reeves C.D.
      ), the growth promoter tylosin (
      • Fouces R.
      • Mellado E.
      • Diez B.
      • Barredo J.L.
      ), and the antiparasitic agent avermectin (
      • Ikeda H.
      • Nonomiya T.
      • Usami M.
      • Ohta T.
      • Omura S.
      ). These reactions typically occur during the late stages of biosynthesis after formation of the core ring system by a polyketide synthase. The hydroxyl or epoxide substituents provide an important layer of structural variability into the final natural product structures and often significantly influence biological activity (
      • Xue Y.
      • Zhao L.
      • Liu H.W.
      • Sherman D.H.
      ,
      • Betlach M.C.
      • Kealey J.T.
      • Ashley G.W.
      • McDaniel R.
      ). P450 monooxygenases are also involved in one of the initial steps in formation of the coumarin group of antibiotics (
      • Chen H.
      • Walsh C.T.
      ), and of the peptidyl nucleoside antibiotic nikkomycin (
      • Lauer B.
      • Russwurm R.
      • Schwarz W.
      • Kalmanczhelyi A.
      • Bruntner C.
      • Rosemeier A.
      • Bormann C.
      ), as well as in oxidative tailoring of the vancomycin-like glycopeptides balhimycin (
      • Pelzer S.
      • Sussmuth R.
      • Heckmann D.
      • Recktenwald J.
      • Huber P.
      • Jung G.
      • Wohlleben W.
      ) and complestatin (
      • Chiu H.T.
      • Hubbard B.K.
      • Shah A.N.
      • Eide J.
      • Fredenburg R.A.
      • Walsh C.T.
      • Khosla C.
      ). Ultimately, the power to manipulate macrolide metabolic systems using combinatorial biosynthetic technology (
      • Carreras C.W.
      • Santi D.V.
      ,
      • Hutchinson C.R.
      ,
      • Yoon Y.J.
      • Beck B.J.
      • Kim B.S.
      • Kang H.Y.
      • Reynolds K.A.
      • Sherman D.H.
      ) will be extended by identification and/or engineering of additional monooxygenases with versatile activities to provide novel biologically active natural products.
      Our current study was motivated by the intriguing amino acid sequence relationship between a number of S. coelicolor A3(2) CYPs that show significant similarity with P450 monooxygenases from other microorganisms involved in regio-specific oxidation of macrolide antibiotics. Among eight crystal structures reported for cytochrome P450 over the last 17 years, only one has been reported for a macrolide hydroxylase (P450 EryF from Saccharopolyspora erythraea) (
      • Cupp-Vickery J.R.
      • Poulos T.L.
      ). EryF is involved in hydroxylation of C-6 of the 14-membered ring macrolactone 6-deoxyerythronolide B in the erythromycin biosynthetic pathway. To elucidate structural, functional, and evolutionary aspects of monooxygenases that tailor macrolide and xenobiotic molecules, crystallographic analysis of the cytochrome P450 complement of S. coelicolor A3(2) (
      • Bentley S.D.
      • Chater K.F.
      • Cerdeno-Tarraga A.M.
      • Challis G.L.
      • Thomson N.R.
      • James K.D.
      • Harris D.E.
      • Quail M.A.
      • Kieser H.
      • Harper D.
      • Bateman A.
      • Brown S.
      • Chandra G.
      • Chen C.W.
      • Collins M.
      • Cronin A.
      • Fraser A.
      • Goble A.
      • Hidalgo J.
      • Hornsby T.
      • Howarth S.
      • Huang C.H.
      • Kieser T.
      • Larke L.
      • Murphy L.
      • Oliver K.
      • O'Neil S.
      • Rabbinowitsch E.
      • Rajandream M.A.
      • Rutherford K.
      • Rutter S.
      • Seeger K.
      • Saunders D.
      • Sharp S.
      • Squares R.
      • Squares S.
      • Taylor K.
      • Warren T.
      • Wietzorrek A.
      • Woodward J.
      • Barrell B.G.
      • Parkhill J.
      • Hopwood D.A.
      ,
      • Lamb D.C.
      • Skaug T.
      • Song H.-L.
      • Jackson C.J.
      • Podust L.M.
      • Waterman M.R.
      • Kell D.B.
      • Kelly D.E.
      • Kelly S.L.
      ) was initiated. We report here the first structural and functional analysis of the monooxygenase CYP154C1 from this organism determined to 1.92 Å resolution.

      EXPERIMENTAL PROCEDURES

      CYP154C1 Purification

      The DNA sequence corresponding to CYP154C1 with four histidine codons inserted at the 3′ end was generated by PCR and cloned into the Escherichia coliexpression vector pET17b as described elsewhere (
      • Lamb D.C.
      • Skaug T.
      • Song H.-L.
      • Jackson C.J.
      • Podust L.M.
      • Waterman M.R.
      • Kell D.B.
      • Kelly D.E.
      • Kelly S.L.
      ). The protein was expressed in HMS174(DE3) cells (Novagen) and purified to homogeneity by nickel-nitrilotriacetic acid (Qiagen) and Q-Sepharose (Amersham Biosciences) chromatography.

      CYP154C1 Crystallization and Data Collection

      Crystals grew in hanging drops from 0.4 mm CYP154C1 in 10 mmTris-HCl, pH 7.5, 450 mmNaCl, 0.5 mm EDTA mixed with an equal volume of 1.6 m MgSO4, 100 mm MES, pH 6.5, 10 mm 2-methylimidazole at 22 °C. Native diffraction data and multiple anomalous dispersion data at three wavelengths (TablesTable I, Table II, Table III) were collected at 100 K at the laboratory source on R-AXIS IV mounted on an RU-200 x-ray generator (Rigaku, Tokyo) and at the 19ID beamline of the Structural Biology Center (Advanced Photon Source, Argonne National Laboratory), respectively. Cryoprotectant contained 207 (v/v) glycerol plus mother liquor. The crystals belong in space groupP212121, with unit cell dimensions a = 62.728, b = 131.973,c = 134.886, α = ॆ = γ = 90°. There are two molecules/asymmetric unit with solvent content in the crystal 647.
      Table ICrystallographic data and statistics
      Dataλ1 (edge)λ2 (peak)λ3 (remote)Native
      Wavelength, Å1.741611.738411.610181.5418
      Resolution, Å2.87 (2.97−2.87)
      Data for high resolution bins are listed in parentheses.
      2.88 (2.98−2.88)2.66 (2.76−2.66)1.92 (1.99−1.92)
      Unique reflections27,08626,85533,99586,084
      Redundancy12.6 (12.1)12.6 (11.6)12.8 (12.3)6.3 (5.7)
      Coverage, 799.7 (99.1)100 (100)100 (100)99.9 (99.7)
      Rsym, 7
      Rsym = Σ ‖ Ii − 〈I〉 ‖ /ΣIi where Ii is the intensity of the ith observation, and 〈I〉 is the mean intensity of reflection.
      12.7 (42.7)13.5 (45.8)10.3 (39.1)6.2 (50.0)
      I/ς6.3 (5.1)5.9 (5.1)7.2 (5.9)15.8 (3.5)
      Data collection statistics are shown below.
      1-a Data for high resolution bins are listed in parentheses.
      1-b Rsym = Σ ‖ Ii − 〈I〉 ‖ /ΣIi where Ii is the intensity of the ith observation, and 〈I〉 is the mean intensity of reflection.
      Table IIMultiple anomalous dispersion data phasing statistics
      λ1λ2λ3Native
      Phasing power (anomalous)
      Phasing power = [Σ(FH)2]12 /[Σ(‖FPHo‖−‖FPHc‖)2], where FPHo and FPHc are the observed and calculated structure factors for the anomalous scatterer.
      1.511.521.56
      Phasing power (isomorphous)1.471.30reference
      FOM
      FOM is the mean figure of merit.
      after phasing
      0.60
      FOM
      FOM is the mean figure of merit.
      after solvent flattening and phase extension
      0.92
      2-a Phasing power = [Σ(FH)2]12 /[Σ(‖FPHo‖−‖FPHc‖)2], where FPHo and FPHc are the observed and calculated structure factors for the anomalous scatterer.
      2-b FOM is the mean figure of merit.
      Table IIIRefinement statistics
      Resolution, Å1.92
      Protein atoms6062
      Ligand atoms
      Ligand atoms include heme.
      86
      Water/ions atoms520/40
      R/Rfree, 7
      R = Σ‖ ‖Fo‖ − ‖Fc‖ ‖/Σ ‖Fo‖, whereFo and Fc are the observed and calculated structure factor amplitudes. Rfree is the same as R but calculated with using 107 of reflections omitted from the refinement.
      20.8/22.8
      rms deviation
      Bonds, Å0.005
      Bond angles, °1.2
      Dihedral angles, °20.8
      Improper angles, °0.81
      Ramachandran
      Program PROCHECK (41).
      A, 90.1;B, 90.7
      3-a Ligand atoms include heme.
      3-b R = Σ‖ ‖Fo‖ − ‖Fc‖ ‖/Σ ‖Fo‖, whereFo and Fc are the observed and calculated structure factor amplitudes. Rfree is the same as R but calculated with using 107 of reflections omitted from the refinement.
      3-c Program PROCHECK (
      • Laskowski R.A.
      • MacArthur M.W.
      • Moss D.S.
      • Thornton J.M.
      ).

      CYP154C1 Phasing and Refinement

      The images were integrated, and the intensities were merged using HKL2000 (
      • Otwinowski Z.
      • Minor W.
      ). The positions of two iron sites were determined with SnB (
      • Weeks C.M.
      • Miller R.
      ). The phases were calculated and improved by CNS (
      • Brunger A.T.
      • Adams P.D.
      • Clore G.M.
      • Delano W.L.
      • Gros P.
      • Grosse-Kunstleve R.W.
      • Jiang J.-S.
      • Kuszewski J.
      • Nilges M.
      • Pannu N.S.
      ) yielding an interpretable electron density map at 2.3 Å resolution. Crystallographic refinement was carried out with CNS (
      • Brunger A.T.
      • Adams P.D.
      • Clore G.M.
      • Delano W.L.
      • Gros P.
      • Grosse-Kunstleve R.W.
      • Jiang J.-S.
      • Kuszewski J.
      • Nilges M.
      • Pannu N.S.
      ) using native data to a resolution of 1.92 Å. The final atomic model (Tables Table I, Table II, Table III) with an R factor of 20.87 (22.87) was obtained after iterations of refinement (CNS (
      • Brunger A.T.
      • Adams P.D.
      • Clore G.M.
      • Delano W.L.
      • Gros P.
      • Grosse-Kunstleve R.W.
      • Jiang J.-S.
      • Kuszewski J.
      • Nilges M.
      • Pannu N.S.
      )), evaluation (PROCHECK (
      • Laskowski R.A.
      • MacArthur M.W.
      • Moss D.S.
      • Thornton J.M.
      )), and manual building (O (
      • Jones T.A.
      • Zou J.Y.
      • Cowan S.W.
      • Kjeldgaard M.
      )). Two protein molecules are present in the asymmetric unit. The quality of the final structure (Tables Table I, Table II, Table III) was assessed with the program PROCHECK (
      • Laskowski R.A.
      • MacArthur M.W.
      • Moss D.S.
      • Thornton J.M.
      ). For the first molecule, 90.17 of residues were found in the most-favored regions of the Ramachandran plot, and 9.97 were found in the allowed regions. For the second molecule, 91.77 of residues were found in the most-favored regions of the Ramachandran plot, and 9.37 were found in the allowed regions.

      Spectral Substrate and Inhibitor Binding Assays

      YC-17-induced (
      • Djerassi C.
      • Zderic J.A.
      ) and narbomycin-induced (
      • Djerassi C.
      • Halpern O.
      ) spectral shifts were monitored at 23.5 °C by using a Shimadzu UV-2401 spectrophotometer. The sample contained 1 ml of 5 ॖmCYP154C1 in 10 mm Tris-HCl, pH 7.5, 107 glycerol. YC-17 and narbomycin were dissolved in ethanol at a stock concentration of 25 mm. Spectral binding assay was carried out by difference spectra with the addition of substrate dissolved in ethanol to the sample cuvette and ethanol to the reference cuvette by 2-ॖl aliquots. The assays were performed by using eight YC-17 or narbomycin concentrations ranging from 50 to 390 ॖm. The data were linearized in the form of the S0A versus S0 plot, whereS0 is a total concentration of substrate in the reaction mixture. The difference in absorbance between 384 nm (peak) and 418 nm (trough) for each spectrum was taken as the ΔAof the reaction. Kd was estimated from the intercept of the linear plot on the S0 axis.

      Catalytic Activity

      The CYP154C1 and PikC conversion of YC17 or narbomycin was compared using a modification of the assay developed for PikC (
      • Xue Y.
      • Wilson D.
      • Zhao L.
      • Liu H.
      • Sherman D.H.
      ). The assay mixture contained 1 ॖm enzyme, 3.5 mm spinach ferredoxin, 0.1 unit of spinach ferredoxin-NADP+ reductase, 1 mm NADPH, and ∼500 ॖm YC17 or narbomycin in a total volume of 1 ml of conversion buffer (100 mm NaH2PO4, pH 7.3, 1 mm EDTA, 0.2 mm dithiothreitol, and 107 glycerol). The reaction was carried out at 37 °C for 20 min and terminated by extraction with ethyl acetate (3 × 1 ml). The extracts were combined, dried, and resuspended in 100 ॖl of ethyl acetate. The extracts were compared with purified compounds using silica TLC developed with chloroform:methanol:NH4OH (9:1:0.1) and stained with vanillin.

      RESULTS AND DISCUSSION

      Sequence Homology with Other Monooxygenases

      CYP154C1(Q9L142; accession numbers are according to SWISS-PROT/TrEMBL Protein Knowledgebase, tw.expasy.org/sprot/) is one of the 18 cytochromes P450 revealed in theS. coelicolor A3(2) genome (
      • Bentley S.D.
      • Chater K.F.
      • Cerdeno-Tarraga A.M.
      • Challis G.L.
      • Thomson N.R.
      • James K.D.
      • Harris D.E.
      • Quail M.A.
      • Kieser H.
      • Harper D.
      • Bateman A.
      • Brown S.
      • Chandra G.
      • Chen C.W.
      • Collins M.
      • Cronin A.
      • Fraser A.
      • Goble A.
      • Hidalgo J.
      • Hornsby T.
      • Howarth S.
      • Huang C.H.
      • Kieser T.
      • Larke L.
      • Murphy L.
      • Oliver K.
      • O'Neil S.
      • Rabbinowitsch E.
      • Rajandream M.A.
      • Rutherford K.
      • Rutter S.
      • Seeger K.
      • Saunders D.
      • Sharp S.
      • Squares R.
      • Squares S.
      • Taylor K.
      • Warren T.
      • Wietzorrek A.
      • Woodward J.
      • Barrell B.G.
      • Parkhill J.
      • Hopwood D.A.
      ), and one of the two assigned to the CYP154 family (
      • Lamb D.C.
      • Skaug T.
      • Song H.-L.
      • Jackson C.J.
      • Podust L.M.
      • Waterman M.R.
      • Kell D.B.
      • Kelly D.E.
      • Kelly S.L.
      ). The closest homolog of CYP154C1 identified in the data base using program BLAST (
      • Altschul S.F.
      • Madden T.L.
      • Schaffer A.A.
      • Zhang J.
      • Zhang Z.
      • Miller W.
      • Lipman D.J.
      ) is a cytochrome P450 CYP154B1 from Streptomyces fradiae with 447 sequence identity (Q9XCC6) localized within the tylosin biosynthetic gene cluster (
      • Bate N.
      • Butler A.R.
      • Gandecha A.R.
      • Cundliffe E.
      ). However, because macrolactone ring hydroxylations at C-20 and C-23 that occur during tylosin biosynthesis are catalyzed by the products of different genes, one of which encodes another cytochrome P450, CYP105L1 (Q9ZHQ1) (
      • Fouces R.
      • Mellado E.
      • Diez B.
      • Barredo J.L.
      ), the role of the CYP154C1 homolog in S. fradiae remains unknown. CYPs with lower identity include the second member of CYP154 family in S. coelicolor A3(2), CYP154A1 (Q9KZR7) (with 427 identity), as well as CYP107B1 (B42606) and EryF, or CYP107A1 (Q00441), from S. erythraea (both with 377 identity and 54 and 517 homology, respectively). There are also a number of P450s from different Streptomyces species having about 367 identity, including CYP107C1 from S. thermotolerans (Q60005), CYP107L1 (PikC) from Streptomyces venezuelae (O87605), PTED from Streptomyces avermitilis (Q93H80), and CYP107D1 (OLEP) from Streptomyces antibioticus (Q59819). This last group of related P450s is known to be involved in modification of macrolide antibiotics such as carbomycin (
      • Arisawa A.
      • Tsunekawa H.
      • Okamura K.
      • Okamoto R.
      ), methymycin, neomethymycin, pikromycin (
      • Xue Y.
      • Wilson D.
      • Zhao L.
      • Liu H.
      • Sherman D.H.
      ,
      • Graziani E.I.
      • Cane D.E.
      • Betlach M.C.
      • Kealey J.T.
      • McDaniel R.
      ,
      • Betlach M.C.
      • Kealey J.T.
      • Ashley G.W.
      • McDaniel R.
      ), novamethymycin (
      • Zhang Q.
      • Sherman D.H.
      ), avermectin (
      • Ikeda H.
      • Nonomiya T.
      • Usami M.
      • Ohta T.
      • Omura S.
      ), and oleandomycin (
      • Rodriguez A.M.
      • Olano C.
      • Mendez C.
      • Hutchinson C.R.
      • Salas J.A.
      ).
      Two S. erythraea enzymes, CYP107B1 and EryF (467 identical to each other) show the same level of identity to CYP154C1 fromS. coelicolor A3(2) (Fig. 1). Having 467 identity, the enzymes have different enzymatic activity. Specifically, EryF catalyzes C-6 hydroxylation of the 6-deoxyerythronolide B (Scheme FS1) in the biosynthesis of erythromycin (
      • Andersen J.F.
      • Tatsuta K.
      • Gunji H.
      • Ishiyama T.
      • Hutchinson C.R.
      ). CYP107B1 shows no detectable activity toward 6-deoxyerythronolide B (
      • Andersen J.F.
      • Hutchinson C.R.
      ), and its function remains unknown. Interestingly, CYP107B1 has 517 identity and 657 homology with the PikC from S. venezuelae (Fig. 1A). Previous studies have revealed that PikC has remarkable substrate flexibility. It is capable of accepting 12- and 14-membered ring macrolides as substrates and catalyzes conversion of the 12-membered ring macrolide intermediate YC-17 to methymycin, neomethymycin (
      • Xue Y.
      • Zhao L.
      • Liu H.W.
      • Sherman D.H.
      ,
      • Xue Y.
      • Wilson D.
      • Zhao L.
      • Liu H.
      • Sherman D.H.
      ,
      • Graziani E.I.
      • Cane D.E.
      • Betlach M.C.
      • Kealey J.T.
      • McDaniel R.
      ), and novamethymycin (
      • Zhang Q.
      • Sherman D.H.
      ). PikC also converts the 14-membered ring macrolide narbomycin to pikromycin (
      • Xue Y.
      • Zhao L.
      • Liu H.W.
      • Sherman D.H.
      ,
      • Xue Y.
      • Wilson D.
      • Zhao L.
      • Liu H.
      • Sherman D.H.
      ,
      • Graziani E.I.
      • Cane D.E.
      • Betlach M.C.
      • Kealey J.T.
      • McDaniel R.
      ,
      • Betlach M.C.
      • Kealey J.T.
      • Ashley G.W.
      • McDaniel R.
      ) (Scheme FS1).
      Figure thumbnail gr1
      Figure 1Sequence alignment between CYP154C1 fromS. coelicolor A3(2) and homologous P450 monooxygenases. A, sequence alignment between CYP154C1 from S. coelicolor A3(2), EryF (CYP107A1) from S. erythraea, and PikC from S. venezuelae. B, fragment of sequence alignment between CYP154C1 from S. coelicolor A3(2) and CYP105 proteins from S. griseusaccession numbers for CYP105A1, B1, C1, and D1 in the SWISS-PROT data base are P18326, P18327, P23296, and P26911, respectively. Theblack shading shows regions of conservation, whereasgray shading denotes similarity of amino acid residues in a given position. Residue numbers for each protein correspond to their sequences deposited in the data base. Secondary structure elements are assigned based on crystal structures for CYP154C1 determined here and for EryF (Protein Data Bank code 1OXA). The α- and 310-helices are marked by open bars, ॆ-strands are marked by arrows, and T indicates turns in the protein chain. The BC loop region is enclosed in a box. The star indicates the conserved Cys.
      Figure thumbnail gr7
      Figure FS1Reactions catalyzed by cytochromes P450 EryF and PikC.

      The Biological Role of CYP154C1

      The endogenous function and biological role of CYP154C1, as with all other S. coelicolorA3(2) CYPs, remain unclear. CYP154C1 lies on the chromosome adjacent to CYP157A1 in an operon that does not contain polyketide synthases or nonribosomal peptide synthetase gene sequences. Adjacent sequences include a sensor kinase, two open reading frames of unknown function, and an ATP-binding protein. The expression of this operon and the function of CYP154C1 may be related to secondary metabolism through a regulatory cascade induced by an environmental stress or to modification of xenobiotics.

      Functional Analysis of CYP154C1: Conversion of YC-17 and Narbomycin in Vitro

      Sequence similarity of CYP154C1 with monooxygenases involved in tailoring of macrolide antibiotics (Fig. 1A) led us to investigate its catalytic activity toward these important compounds. Thus, the protein was tested against 12- and 14-membered ring macrolide intermediates YC-17 and narbomycin, which are converted to methymycin, neomethymycin, novamethymycin, and pikromycin by PikC inS. venezuelae (
      • Xue Y.
      • Zhao L.
      • Liu H.W.
      • Sherman D.H.
      ,
      • Xue Y.
      • Wilson D.
      • Zhao L.
      • Liu H.
      • Sherman D.H.
      ,
      • Xue Y.
      • Sherman D.H.
      ,
      • Graziani E.I.
      • Cane D.E.
      • Betlach M.C.
      • Kealey J.T.
      • McDaniel R.
      ,
      • Betlach M.C.
      • Kealey J.T.
      • Ashley G.W.
      • McDaniel R.
      ,
      • Zhang Q.
      • Sherman D.H.
      ) (Scheme FS1). Interestingly, CYP154C1 showed catalytic activity toward both substrates. The reaction with the 14-membered ring macrolide narbomycin was notably faster and resulted in a more complete conversion to pikromycin compared with conversion of the 12-membered ring macrolide, YC-17. However, transformation of YC-17 was accomplished with notable position selectivity at the C12 position, leading predominantly to neomethymycin. This substrate selectivity is in marked contrast to PikC, which converts YC-17 with equal efficiency to methymycin and neomethymycin (
      • Xue Y.
      • Wilson D.
      • Zhao L.
      • Liu H.
      • Sherman D.H.
      ). CYP154C1 binds YC-17 and narbomycin with similar affinity (Kd = 405 and 403 ॖm, respectively), producing type I binding spectra resulting from expulsion of a water molecule from the iron coordination sphere followed by transition of the heme iron to a pentacoordinate high spin state (Fig. 2, A andC). The similarities in binding characteristics of CYP154C1, while at the same time showing significant substrate specificity toward the two macrolactone ring systems, represent a significant opportunity to explore structure-function relationships within and between other members of the CYP enzyme class.
      Figure thumbnail gr2
      Figure 2Substrate binding and catalytic activity of CYP154C1. A, type I binding spectra resulting from CYP154C1 titration with increasing concentrations of YC-17 or narbomycin ranging from 50 to 390 ॖm. B, linearization of the titration data in the form ofS0A versus S0 plot. C, thin layer chromatography analysis of products of the catalytic conversion of YC-17 or narbomycin by PikC and CYP154C1. MM is for methymycin, NMMis for neomethymycin, NBM is for narbomycin, andPKM is for pikromycin.

      X-ray Structure Determination of CYP154C1

      The growing number of cytochrome P450 structures in the Protein Data Bank (eight are currently available) provides a pool of search models to aid in solving new P450 structures by the molecular replacement technique. However, amino acid sequence diversity of enzymes within the superfamily has limited use of the molecular replacement approach for cytochrome P450 structure determination (
      • Yano J.K.
      • Koo L.S.
      • Schuller D.J.
      • Li H.
      • Ortiz de Montellano P.R.
      • Poulos T.L.
      ). Fortunately, the intrinsic heme-iron atom of cytochromes P450 can be used as an anomalous scatterer to facilitate protein structure determination using the multiple anomalous dispersion technique.
      CYP154C1 has 377 sequence identity and 517 sequence similarity to the macrolide monooxygenase EryF, the highest homology to date between P450s with reported x-ray structure. However, this similarity is insufficient to successfully use EryF atomic coordinates to find a molecular replacement solution for CYP154C1, even when poorly conserved regions are omitted from the search model. Thus, CYP154C1 is the first cytochrome P450 whose structure has been determined solely by the multiple anomalous dispersion technique using the endogenous heme-iron as an anomalous scatterer. The final model (Tables Table I, Table II, Table III) consists of two molecules in an asymmetric unit each bearing a heme group and residues 8–407, plus three residues from the C-terminal His tag for the first molecule, along with 520 water molecules and eight sulfate ions. Nine residues in the first molecule and three residues in the second molecule were truncated to Ala because of insufficient electron density.

      Overall CYP154C1 Crystal Structure

      Although P450 monooxygenases have been studied extensively by x-ray crystallography, new superfamily members continue to reveal structural features as yet unpredictable from primary sequence. Despite relatively high homology to EryF, the three-dimensional appearance of the substrate-free CYP154C1 is quite distinct from substrate/inhibitor bound EryF. CYP154C1 appears to be separated along the interface between the ॆ-sheet and α-helical domains from the distal surface all the way down to the heme (Fig. 3A; see also Fig. 6). The two domains remain separated by a slit about 10 Å wide without additional stabilizing forces. Despite the distinct packing environment in the crystal, two molecules in an asymmetric unit show minor deviations (root mean square deviation of 0.3 Å as calculated using an algorithm implemented in SWISS PDB VIEWER (
      • Guex N.
      • Peitsch M.C.
      )) in the major part of the structure, whereas more significant deviations up to 1.8 Å are observed in the FG region (blue in Fig.3A). Repositioning of the FG loop results in slightly more open access to the active site in the one molecule compared with the second.
      Figure thumbnail gr3
      Figure 3Superimposition of structures. A, superimposition between two CYP154C1 molecules in an asymmetric unit. B, superimposition between CYP154C1 and EryF. Superimposition was performed by using SWISS PDB VIEWER (
      • Guex N.
      • Peitsch M.C.
      ) according to an algorithm implemented in the program. The diagrams here and in Figs. and were prepared using SETOR (
      • Evans S.V.
      ). CYP154C1 is shown in gray with the FG region highlighted inblue, the BC loop and the B′ helix in violet, and the ॆ-sheet 1 in magenta. EryF (Protein Data Bank code1OXA) is in green with the BC loop highlighted inorange. The EryF substrate 6-deoxyerythronolide B is denoted by gray balls. Heme is in red.
      Figure thumbnail gr6
      Figure 6Surface electrostatic potential of CYP154C1 and EryF. The deepest shades of red and bluecorrespond to potentials of −22.6 and 27.2 kcal, respectively, whereas neutral points are white. Heme partially seen through the cleft is green. Cyan three-dimensional vectors show the direction and relative size of molecular dipoles for CYP154C1 and EryF.
      The open conformation in CYP154C1 is achieved by relocation of certain secondary structure elements constituting the common P450 fold, as well as being due to a novel ॆ-strand unique to CYP154C1. As seen from of the alignment (Fig. 1A) and from stereo views of the structure (Fig. 3), ॆ-sheet 1 in CYP154C1 consists of six strands instead of the five observed in EryF and other P450s. The new strand (indicated for CYP154C1 as ॆ1–1 (Fig. 1A)) precedes the invariant helix A. At residues Pro14-Phe15 the polypeptide chain makes a 180° turn so that the A helix runs in a direction anti-parallel to the ॆ1-1 strand. The result of this new arrangement is absence of the A′ 310 helix present in EryF or a loop preceding the A helix in some other P450s that normally contribute to interactions between the α-helical and ॆ-sheet domains. In the α-helical domain, significant dislocation of the F, G (blue in Fig. 3A), and B′ (violet) helices relative to their positions in substrate-bound EryF (greenin Fig. 3B) occurs. The F and G helices are translocated away from the distal and toward the proximal surface by about one α-helical turn. Together with the shorter length of the G helix in CYP154C1 and the FG loop being open compared with EryF, a significant separation between the two protein domains is achieved. Additionally, the B′ helix in CYP154C1 is repositioned relative to its EryF counterpart so that it becomes almost parallel to the G helix (Fig.3B). These two helices complete the formation of an opening, which we believe is a substrate-binding site in CYP154C1.
      An important issue regarding the open conformation of CYP154C1 is whether its presence is due to the empty active site pocket. If so, CYP154C1 might oscillate between open/closed conformations triggered by substrate binding/product release. Conformational changes leading to closure of the active site entrance caused by substrate binding have been demonstrated previously for P450BM3 (
      • Li H.
      • Poulos T.L.
      ,
      • Li H.
      • Poulos T.L.
      ). S. coelicolor A3(2) CYP154C1 is the second P450 (after mycobacterial CYP51 (
      • Podust L.M.
      • Poulos T.L.
      • Waterman M.R.
      )) whose active site is accessible from the surface in substrate-free form. Obtaining structures for P450s in substrate-free and substrate-bound forms are expected to address the complex conformational changes that occur when substrate binds, and following hydroxylation the product is released from the enzyme. These dynamic affects could account for the open structure of substrate-free CYP154C1 and the closed structure of substrate-bound EryF.

      Analysis of the Putative CYP154C1 Substrate-binding Site

      The putative CYP154C1 substrate-binding site in the absence of substrate is shaped as a cleft open from the distal surface, with both walls built by hydrophobic residues (Fig. 4). The back side of the cleft is lined by polar residues and has negative potential because of Asp398, the only charged residue residing in the substrate-binding site. The left cleft wall is predominantly built by the residues from the BC loop, whereas the right wall is comprised of residues from the FG region and the I helix (Fig. 4).
      Figure thumbnail gr4
      Figure 4Putative substrate-binding site of CYP154C1. For a better view of the cleft entrance, the CYP154C1 molecule in Fig. was rotated toward the viewer approximately along the horizontal axis in the plane of the drawing. Residues building the left wall of the binding site cleft are from the BC loop (violet), and residues building the right wall are from the FG region (blue) and the I helix (green). Residues lining the back side of the cleft (gray) are from the last turn within ॆ-sheet 3 and from the junction between the K helix and the strand ॆ1–5.
      Sequence alignment of CYP154C1 with other highly related monooxygenases shows variation in the BC loop region (enclosed in box in Fig. 1A). Significantly, this region is two residues shorter in CYP154C1 than in EryF and is substantially shorter (11 residues) in PikC that has broad substrate specificity having ability to hydroxylate both 12- and 14-membered ring macrolide antibiotics. It is worth noting that sequence alignment of CYP154C1 with CYP105C1 fromStreptomyces griseus also shows significant shortening of the BC loop (20 residues shorter in CYP105C1 and five residues shorter in three other family members) (Fig. 1B). It seems likely that the shortened BC loop would allow more space above the heme plane and confer more freedom to accommodate substrates of variable size. This is consistent with the biological function of monooxygenases with broad substrate specificity, including PikC and xenobiotic functionalizing enzymes, which may be inherently more flexible.
      A water molecule bound to the iron in the sixth coordination site (W1 in Fig. 5) was modeled into electron density in the active site of each molecule in an asymmetric unit. Although iron-bound water has a relatively high temperature factor, 40.1 Å2 with the occupancy of 1.0 in both molecules, omitting this water from the model results in appearance of extra electron density in a FoFc map (dark blue in Fig. 5). The iron-oxygen distances in CYP154C1 are refined to 2.42 and 2.54 Å for the two molecules, which is considerably longer than the 2.28 Å observed in substrate-free P450cam (
      • Poulos T.L.
      • Finzel B.C.
      • Howard A.J.
      ). High B-factors for water at the sixth ligand position are likely a result of partial occupancy of this site. The relatively long iron-oxygen distance also indicates that this bond is weakened in CYP154C1 compared with P450cam.
      Figure thumbnail gr5
      Figure 5Oxygen scission site of CYP154C1. The fragment of the I helix shown represents the partially conserved I helix motif positioned relative to the heme plane. Atoms are colored according to elements: oxygen in red, nitrogen inblue, sulfur in yellow, and carbon ingray. Red was also used for the heme iron. Two water molecules in the active site are represented by the red spheres: W1 is for water in the sixth heme ligand position, and W2 is for water in the water-binding cleft of the I helix. Fragments of the 2FoFc electron density composite omit map are contoured at 1.0ς (cyan) and 2.5ς (red). Fragment of the FoFc electron density map generated with W1 omitted is contoured at 1.5 ς (dark blue).

      The I Helix Motif in CYP154C1

      Water in the sixth coordination site (W1) is hydrogen-bonded to the carbonyl oxygen of Ala242 (with the distances 2.65 and 2.78 Å in both molecules), which in turn is hydrogen bonded to the side chain of Thr246 occupying the n + 4 position along the same face of the I helix (Fig. 5). A similar hydrogen bond is formed between Gly243 and Thr247. A motif occurring in the CYP154C1 I helix as Ala242-Gly243-His244-Glu245-Thr246-Thr247is partially conserved among cytochromes P450, although the growing number of P450 enzymes in the data base show variations of amino acids in every single position. For S. coelicolor A3(2) CYPs, this motif is summarized as Ala(Gly)242-Gly(Ala)243-Xaa244-Glu(Asp,Gln,Ala,Val)245-Thr(Ala,Pro,Val)246-Thr(Met,Leu,Ile)247(residue numbers are according to CYP154C1). To a large degree the motif conservation seems to be determined by the position of the I helix relative to the heme plane. The I helix is positioned so that residues 242 and 246 face the heme plane to the exclusion of bulky side chains because of space constraints (Fig. 5). Because of the small size of these side chains, the I helix approaches near the heme plane and enables a carbonyl oxygen of residue 242 (usually Ala, sometimes Gly) to hydrogen bond a water molecule bound to the heme iron. This interaction violates the hydrogen bonding pattern and the geometry of the I helix. To compensate for missing helical hydrogen bonds, the hydroxyl of Thr246 (or Ser in many P450s) hydrogen bonds with the carbonyl oxygen of Ala242, whereas the hydroxyl of Thr247 hydrogen bonds to the carbonyl oxygen of Gly243. This unconventional α-helix hydrogen bonding pattern is observed in the I helix of most structurally defined CYPs (
      • Podust L.M.
      • Poulos T.L.
      • Waterman M.R.
      ).
      Gly in position 243 is the most highly conserved within the I helix motif, and we believe is related to the molecular dynamics of P450 molecules. Gly confers flexibility to the I helix, which might allow the I helix water binding cleft (Fig. 5) to adjust its size expelling or accepting a water molecule (
      • Cupp-Vickery J.R.
      • Garcia C.
      • Hofacre A.
      • McGee-Estrada K.
      ) and also allow the I helix N terminus to perform swinging motions in response to substrate or inhibitor binding in the active site (
      • Li H.
      • Poulos T.L.
      ,
      • Podust L.M.
      • Poulos T.L.
      • Waterman M.R.
      ,
      • Podust L.M.
      • Stojan J.
      • Poulos T.L.
      • Waterman M.R.
      ). Conservation of residue functionalities in positions 245, partially 246, and perhaps 247 is likely due to mechanistic reasons for electron and/or proton transfer and for oxygen activation (
      • Vaz A.D.N.
      • Pernecky S.J.
      • Raner G.M.
      • Coon M.J.
      ,
      • Vidakovic M.
      • Sligar S.G.
      • Li H.
      • Poulos T.L.
      ,
      • Obayashi E.
      • Shimizu H.
      • Park S.Y.
      • Shoun H.
      • Shiro Y.
      ). Apparently, structural reasons dominate conservation of this motif because positions 242, 243, and 246 are never occupied by residues with bulky side chains, whereas functionalities of residues in 245, 246, and 247 alternate. Whether these variations correlate with reaction mechanism is unknown because most P450s have not been studied in sufficient detail to address this question.
      The CYP154C1 structure illustrates that functional diversity within the cytochrome P450 family arises from the structural diversity of a few externally positioned elements: the F and G helices with the FG loop in between and the highly variable B′ helix, whereas the oxygen scission site remains largely intact. Relocation of the structural elements allows an almost endless number of configurations for P450 substrate-binding sites, particularly in the absence of substrate, and might be a serious obstacle in attempts to predict new cytochrome P450 structures based on homology modeling approaches.

      Surface Electrostatic Potential of CYP154C1

      Electrostatic potential distribution shows the presence of a cationic patch on the proximal surface of CYP154C1 (Fig. 6), the region that has been implicated in redox partner interactions in P450s (
      • Graham S.E.
      • Peterson J.A.
      ). A similar cationic patch is present on the surface of EryF, although it has a larger dipole moment (694 versus 305 Debye in CYP154C1) pointed in a different direction (Fig. 6). The dipole vector of CYP154C1 points directly in the middle of the patch on a proximal surface almost perpendicular to the heme plane, whereas the dipole vector of EryF is strongly inclined toward the C and C′ helices. A key landmark of the proximal surfaces of both CYP154C1 and EryF is a protruding Arg344 (Arg339 in EryF) coming from the region known as the meander. Together with other positively charged residues that form similar patterns on the proximal surfaces of both CYP154C1 and EryF (Fig. 6), the prominent Arg might play a role in anchoring the electron donor on the cationic patch.

      Biological Implications

      This effort represents the first structural analysis of S. coelicolor CYP enzymes and is expected to provide information on hydroxylation patterns of endogenous as well as exogenous substrates. Cytochrome P450 monooxygenases fromStreptomyces catalyze site-specific oxidations of the precursors to many macrolide antibiotics and degrade a wide range of xenobiotic compounds. Use of Streptomyces in the production of natural products of pharmaceutical importance as well as in bioremediation technologies provides an important incentive to identify native enzymes and engineer novel monooxygenases with activities toward alternative substrates. Together with the opportunity to manipulate entire biosynthetic gene clusters, it allows design of biosynthetic pathways giving rise to novel polyketides and other natural product structures. CYP154C1 is the first three-dimensional structure for P450 monooxygenase with activity toward polyketides of diverse structures and will contribute to understanding the molecular basis for the specificity of oxidative tailoring in macrolide antibiotic biosynthesis, as well as in development of approaches that lead to fully elaborated and biologically active macrolide structures.

      Acknowledgments

      We thank Jarrod A. Smith and the Vanderbilt University Center for Structural Biology computing facilities for expert technical assistance, Dr. Stanley N. Cohen and Dr. Jianqiang Huang (Stanford University) for providing microarray information on expression of individual CYP genes, and Dr. Thomas Poulos (University of California, Irvine) for interest in the work, helpful discussions, and critical reading of the manuscript.

      References

        • Guengerich F.P.
        • MacDonald T.L.
        FASEB J. 1990; 4: 2453-2459
        • Guengerich F.P.
        Nat. Rev. Drug. Discov. 2002; 1: 359-366
        • Hopwood D.A.
        Microbiology. 1999; 145: 2183-2202
        • Shelton D.R.
        • Khader S.
        • Karns J.S.
        • Pogell B.M.
        Biodegradation. 1996; 7: 129-136
        • Jendrossek D.
        • Tomasi G.
        • Kroppenstedt R.M.
        FEMS Microbiol. Lett. 1997; 150: 179-188
        • Fadullon F.S.
        • Karns J.S.
        • Torrents A.
        J. Environ. Sci. Health. B. 1998; 33: 37-49
        • Obojska A.
        • Lejczak B.
        • Kubrak M.
        Appl. Microbiol. Biotechnol. 1999; 51: 872-876
        • Gallert C.
        Syst. Appl. Microbiol. 2000; 23: 433-441
        • Amoroso M.J.
        • Castro G.R.
        • Duran A.
        • Peraud O.
        • Oliver G.
        • Hill R.T.
        J. Ind. Microbiol. Biotechnol. 2001; 26: 210-215
        • Bentley S.D.
        • Chater K.F.
        • Cerdeno-Tarraga A.M.
        • Challis G.L.
        • Thomson N.R.
        • James K.D.
        • Harris D.E.
        • Quail M.A.
        • Kieser H.
        • Harper D.
        • Bateman A.
        • Brown S.
        • Chandra G.
        • Chen C.W.
        • Collins M.
        • Cronin A.
        • Fraser A.
        • Goble A.
        • Hidalgo J.
        • Hornsby T.
        • Howarth S.
        • Huang C.H.
        • Kieser T.
        • Larke L.
        • Murphy L.
        • Oliver K.
        • O'Neil S.
        • Rabbinowitsch E.
        • Rajandream M.A.
        • Rutherford K.
        • Rutter S.
        • Seeger K.
        • Saunders D.
        • Sharp S.
        • Squares R.
        • Squares S.
        • Taylor K.
        • Warren T.
        • Wietzorrek A.
        • Woodward J.
        • Barrell B.G.
        • Parkhill J.
        • Hopwood D.A.
        Nature. 2002; 417: 141-147
        • Lamb D.C.
        • Skaug T.
        • Song H.-L.
        • Jackson C.J.
        • Podust L.M.
        • Waterman M.R.
        • Kell D.B.
        • Kelly D.E.
        • Kelly S.L.
        J. Biol. Chem. 2002; 277: 24000-24005
        • Xue Y.
        • Zhao L.
        • Liu H.W.
        • Sherman D.H.
        Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 12111-12116
        • Bate N.
        • Butler A.R.
        • Gandecha A.R.
        • Cundliffe E.
        Chem. Biol. 1999; 6: 617-624
        • Fouces R.
        • Mellado E.
        • Diez B.
        • Barredo J.L.
        Microbiology. 1999; 145: 855-868
        • Ikeda H.
        • Nonomiya T.
        • Usami M.
        • Ohta T.
        • Omura S.
        Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 9509-9514
        • Pelzer S.
        • Sussmuth R.
        • Heckmann D.
        • Recktenwald J.
        • Huber P.
        • Jung G.
        • Wohlleben W.
        Antimicrob. Agents Chemother. 1999; 43: 1565-1573
        • Caffrey P.
        • Lynch S.
        • Flood E.
        • Finnan S.
        • Oliynyk M.
        Chem. Biol. 2001; 8: 713-723
        • Huang J.
        • Lih C.J.
        • Pan K.H.
        • Cohen S.N.
        Genes Dev. 2001; 15: 3183-3192
        • Xue Y.
        • Wilson D.
        • Zhao L.
        • Liu H.
        • Sherman D.H.
        Chem. Biol. 1998; 5: 661-667
        • Xue Y.
        • Sherman D.H.
        Metab. Eng. 2001; 3: 15-26
        • Graziani E.I.
        • Cane D.E.
        • Betlach M.C.
        • Kealey J.T.
        • McDaniel R.
        Bioorg. Med. Chem. Lett. 1998; 8: 3117-3120
        • Betlach M.C.
        • Kealey J.T.
        • Ashley G.W.
        • McDaniel R.
        Biochemistry. 1998; 37: 14937-14942
        • Zhang Q.
        • Sherman D.H.
        J. Nat. Prod. 2001; 64: 1447-1450
        • Rodriguez A.M.
        • Olano C.
        • Mendez C.
        • Hutchinson C.R.
        • Salas J.A.
        FEMS Microbiol. Lett. 1995; 127: 117-120
        • Haydock S.F.
        • Dowson J.A.
        • Dhillon N.
        • Roberts G.A.
        • Cortes J.
        • Leadlay P.F.
        Mol. Gen. Genet. 1991; 230: 120-128
        • Weber J.M.
        • Leung J.O.
        • Swanson S.J.
        • Idler K.B.
        • McAlpine J.B.
        Science. 1991; 252: 114-117
        • Molnar I.
        • Schupp T.
        • Ono M.
        • Zirkle R.
        • Milnamow M.
        • Nowak-Thompson B.
        • Engel N.
        • Toupet C.
        • Stratmann A.
        • Cyr D.D.
        • Gorlach J.
        • Mayo J.M.
        • Hu A.
        • Goff S.
        • Schmid J.
        • Ligon J.M.
        Chem. Biol. 2000; 7: 97-109
        • Tang L.
        • Shah S.
        • Chung L.
        • Carney J.
        • Katz L.
        • Khosla C.
        • Julien B.
        Science. 2000; 287: 640-642
        • Molnar I.
        • Aparicio J.F.
        • Haydock S.F.
        • Khaw L.E.
        • Schwecke T.
        • Konig A.
        • Staunton J.
        • Leadlay P.F.
        Gene (Amst.). 1996; 169: 1-7
        • Chung L.
        • Liu L.
        • Patel S.
        • Carney J.R.
        • Reeves C.D.
        J. Antibiot. 2001; 54: 250-256
        • Chen H.
        • Walsh C.T.
        Chem. Biol. 2001; 8: 301-312
        • Lauer B.
        • Russwurm R.
        • Schwarz W.
        • Kalmanczhelyi A.
        • Bruntner C.
        • Rosemeier A.
        • Bormann C.
        Mol. Gen. Genet. 2001; 264: 662-673
        • Chiu H.T.
        • Hubbard B.K.
        • Shah A.N.
        • Eide J.
        • Fredenburg R.A.
        • Walsh C.T.
        • Khosla C.
        Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 8548-8553
        • Carreras C.W.
        • Santi D.V.
        Curr. Opin. Biotechnol. 1998; 9: 403-411
        • Hutchinson C.R.
        Curr. Opin. Microbiol. 1998; 1: 319-329
        • Yoon Y.J.
        • Beck B.J.
        • Kim B.S.
        • Kang H.Y.
        • Reynolds K.A.
        • Sherman D.H.
        Chem. Biol. 2002; 9: 203-214
        • Cupp-Vickery J.R.
        • Poulos T.L.
        Nat. Struct. Biol. 1995; 2: 144-153
        • Otwinowski Z.
        • Minor W.
        Methods Enzymol. 1997; 276: 307-326
        • Weeks C.M.
        • Miller R.
        Appl. Crystallogr. 1999; 32: 120-124
        • Brunger A.T.
        • Adams P.D.
        • Clore G.M.
        • Delano W.L.
        • Gros P.
        • Grosse-Kunstleve R.W.
        • Jiang J.-S.
        • Kuszewski J.
        • Nilges M.
        • Pannu N.S.
        Acta Crystallogr. 1998; D54: 905-921
        • Laskowski R.A.
        • MacArthur M.W.
        • Moss D.S.
        • Thornton J.M.
        J. Appl. Crystallogr. 1993; 26: 283-291
        • Jones T.A.
        • Zou J.Y.
        • Cowan S.W.
        • Kjeldgaard M.
        Acta Crysallogr. 1991; A47: 110-119
        • Djerassi C.
        • Zderic J.A.
        J. Am. Chem. Soc. 1956; 78: 6390-6395
        • Djerassi C.
        • Halpern O.
        Tetrahedron. 1958; 3: 255-268
        • Altschul S.F.
        • Madden T.L.
        • Schaffer A.A.
        • Zhang J.
        • Zhang Z.
        • Miller W.
        • Lipman D.J.
        Nucleic Acids Res. 1997; 25: 3389-3402
        • Arisawa A.
        • Tsunekawa H.
        • Okamura K.
        • Okamoto R.
        Biosci. Biotechnol. Biochem. 1995; 59: 582-588
        • Andersen J.F.
        • Tatsuta K.
        • Gunji H.
        • Ishiyama T.
        • Hutchinson C.R.
        Biochemistry. 1993; 32: 1905-1913
        • Andersen J.F.
        • Hutchinson C.R.
        J. Bacteriol. 1992; 174: 725-735
        • Yano J.K.
        • Koo L.S.
        • Schuller D.J.
        • Li H.
        • Ortiz de Montellano P.R.
        • Poulos T.L.
        J. Biol. Chem. 2000; 275: 31086-31092
        • Guex N.
        • Peitsch M.C.
        Electrophoresis. 1997; 18: 2714-2723
        • Li H.
        • Poulos T.L.
        Biochimie (Paris). 1996; 78: 695-699
        • Li H.
        • Poulos T.L.
        Nat. Struct. Biol. 1997; 4: 140-146
        • Podust L.M.
        • Poulos T.L.
        • Waterman M.R.
        Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 3068-3073
        • Poulos T.L.
        • Finzel B.C.
        • Howard A.J.
        Biochemistry. 1986; 25: 5314-5322
        • Cupp-Vickery J.R.
        • Garcia C.
        • Hofacre A.
        • McGee-Estrada K.
        J. Mol. Biol. 2001; 311: 101-110
        • Podust L.M.
        • Stojan J.
        • Poulos T.L.
        • Waterman M.R.
        J. Inorg. Biochem. 2001; 87: 227-235
        • Vaz A.D.N.
        • Pernecky S.J.
        • Raner G.M.
        • Coon M.J.
        Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 4644-4648
        • Vidakovic M.
        • Sligar S.G.
        • Li H.
        • Poulos T.L.
        Biochemistry. 1998; 37: 9211-9219
        • Obayashi E.
        • Shimizu H.
        • Park S.Y.
        • Shoun H.
        • Shiro Y.
        J. Inorg. Biochem. 2000; 82: 103-111
        • Graham S.E.
        • Peterson J.A.
        Arch. Biochem. Biophys. 1999; 369: 24-29
        • Evans S.V.
        J. Mol. Graphics. 1993; 11: 134-138