Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The substrate specificity and conformational flexibility of ketosteroid hydroxylases Penfield, Jonathan 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2013_fall_penfield_jonathan.pdf [ 3.07MB ]
Metadata
JSON: 24-1.0103343.json
JSON-LD: 24-1.0103343-ld.json
RDF/XML (Pretty): 24-1.0103343-rdf.xml
RDF/JSON: 24-1.0103343-rdf.json
Turtle: 24-1.0103343-turtle.txt
N-Triples: 24-1.0103343-rdf-ntriples.txt
Original Record: 24-1.0103343-source.json
Full Text
24-1.0103343-fulltext.txt
Citation
24-1.0103343.ris

Full Text

  The substrate specificity and conformational flexibility of ketosteroid hydroxylases by Jonathan Penfield B.S., The University of Utah, 2010 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Biochemistry and Molecular Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   September 2013 ? Jonathan Penfield, 2013 ii  Abstract  3-Ketosteroid-9?-hydroxylase (KshAB) is a Rieske oxygenase involved in bacterial steroid degradation. Bacteria such as Rhodococcus rhodochrous DSM43269 harbor up to five KshA homologues (numbered 1 to 5) that appear to be involved in degrading different steroids. Previous work indicated that KshA5 (DSM43269) transforms an unusually broad range of 3-ketosteroids while KshA1 (DSM43269) and KshAMtb of Mycobacterium tuberculosis have strong preference for 3-ketosteroids with side chains at C-17. To better understand KshAs in general, KshA1 and KshA5 were purified anaerobically and characterized. Steady-state kinetic studies revealed that KshA1 has 10- to 100-fold higher apparent specificity constant for ketosteroids possessing long C17 side chain such as 3-oxo-23,24-bisnorchola-1,4-dien-22-oate (4-BNC), and is thus similar to KshAMtb. By contrast, KshA5 had highest specificity for substrates with C17-oxo (e.g., apparent kcat/Km > 105 s-1 M-1 for 4-estrendione and 5?-androstandione vs. 102 s-1 M-1 for 1,4-BNC-CoA). However, KshA5 displayed very strong substrate inhibition with 1,4-androstadiene-3,17-dione (ADD) and 4-BNC (KSS ~110 ?M) despite hydroxylation well coupled to O2 consumption and turnover occurring at reasonable rates (apparent kcat ~0.7 s-1). Crystallographic structures of four KshA:substrate complexes were determined: KshA1:ADD (2.4 ?), KshAMtb:ADD (2.3 ?), KshA5:ADD (1.8 ?) and KshA5:1,4-BNC-CoA (2.6 ?). In each, the substrate was bound in a similar orientation with the steroid?s C9 closest to the active site iron. In comparison to a structure of substrate-free KshA5 (2.6 ? resolution), the catalytic iron was displaced up to 3.1 ? in the complexes with ADD and 1,4-BNC-CoA. This was accompanied by similar magnitude shift in the helices harboring the iron-coordinating residues. The net effect was an unusually large distance between the iron and C9 of the substrate (6.3 ?). Additionally, the active site opening of KshA5 was occluded from bulk solvent by a loop comprising residues 217 to 233 in substrate-free and ADD-bound structures while the KshA5:1,4-BNC-CoA complex exhibited an open active site, as observed in KshA1 and KshAMtb structures, containing a similar disordered loop region. The loop conformation in these structures and the ability of KshA5 to turnover CoA thioesters demonstrate unexpected conformational flexibility in correlation with interesting kinetic behavior in a Rieske oxygenase.   iii  Preface   I was responsible for the work contained in this thesis including: constructing expression plasmids; producing and characterizing KshA1 and KshA5; screening, growing, and harvesting crystals; solving and refining crystallographic structures; and data analysis.  Crystal structures were determined in collaboration with Dr. Liam Worrall in the group of Prof. Natalie C. Strynadka (The University of British Columbia).     iv  Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iii Table of Contents ......................................................................................................................... iv List of Tables .............................................................................................................................. viii List of Figures ............................................................................................................................... ix List of Abbreviations ................................................................................................................... xi Acknowledgements ..................................................................................................................... xii Dedication ................................................................................................................................... xiii 1. Introduction ........................................................................................................................... 1 1.1 Reactivity of dioxygen in biology ............................................................................... 1 1.1.1 Oxygenases ................................................................................................ 1 1.1.2  Non-heme Fe activation of oxygen ............................................................ 1 1.2 Rieske oxygenases ....................................................................................................... 2 1.2.1 Phylogenetic diversity ............................................................................... 2 1.2.2 Reactions catalyzed ................................................................................... 3 1.2.3  Structure of Rieske oxygenases ................................................................. 4 1.2.4  Rieske oxygenase active site ..................................................................... 6 1.3. Catalysis by Rieske oxygenases...................................................................................... 7 1.3.1  Rieske oxygenase reaction cycle ............................................................... 8 1.3.2. Active O2 species ..................................................................................... 10 1.3.3.  Regulation of the catalytic cycle .............................................................. 10 1.4 3-Ketosteroid-9?-hydroxylase (KshAB) ................................................................... 11 1.4.1.  KshA structure ......................................................................................... 13 1.4.2  KshA1 and KshA5 ................................................................................... 14 v  1.5 Aim of this study ....................................................................................................... 17 2. Materials and Methods ....................................................................................................... 18 2.1 Chemicals and reagents ............................................................................................. 18 2.2   Bacterial strains and culture conditions ..................................................................... 18 2.3 DNA manipulation and plasmid construction ........................................................... 19 2.4 Protein production and purification ........................................................................... 20 2.4.1 KshA ........................................................................................................ 20 2.4.2  KshB production and purification ........................................................... 21 2.4.3  KstD production ....................................................................................... 22 2.5  Analytical methods .................................................................................................... 22 2.6  Steady-state kinetic analysis ...................................................................................... 23 2.7  HPLC analysis ........................................................................................................... 23 2.8  Coupling measurements ............................................................................................ 23 2.9  Protein crystallization ................................................................................................ 24 2.9.1 Crystal screening ..................................................................................... 24 2.9.2 KshA crystallization ................................................................................ 24 2.9.3  X-ray data collection and structure determination ................................... 25 3. Results ................................................................................................................................... 26 3.1  Purification of KshA ................................................................................................. 26 3.1.1 KshA1 preparation ................................................................................... 26 3.1.2 KshA5 preparation ................................................................................... 27 3.1.3 KshAMtb preparation ................................................................................ 28 3.2 Reaction products ...................................................................................................... 28 3.3 Coupling .................................................................................................................... 29 3.4 Steady-state kinetic analysis ...................................................................................... 29 vi  3.4.1 KshA1 kinetics ......................................................................................... 30 3.4.2  KshA5 kinetics ......................................................................................... 33 3.4.3 Oxygen dependency of KshA1 and KshA5 ............................................. 35 3.5 X-ray crystallography analyses ................................................................................. 36 3.5.1 Crystals .................................................................................................... 36 3.5.1.1  KshAMtb in complex with ADD .............................................................. 36 3.5.1.2  KshA1 and KshA5 hybrid crystal ........................................................... 37 3.5.1.3  KshA5 in complex with ADD ................................................................ 37 3.5.1.4  KshA5 in complex with 1,4-BNC-CoA .................................................. 37 3.5.2 Analysis of KshA structures .................................................................... 38 3.5.3 Structure of KshAMtb:ADD complex ....................................................... 39 3.5.4 Structure of KshA1 and KshA5 co-crystallization .................................. 41 3.5.5 Structure of KshA1:1,4-BNC-CoA complex ........................................... 42 3.5.6 Structure of KshA5 without substrate ...................................................... 43 3.5.7 Structure of KshA5:ADD complex ......................................................... 45 3.5.8 Structure of KshA5:1,4-BNC-CoA complex ........................................... 47 4. Discussion ............................................................................................................................. 49 4.1 Correlation of KshA specificities to previous investigations .................................... 49 4.1.1 KshA1 specificity and physiological implications .................................. 50 4.1.2 KshA5 specificity and inhibition ............................................................. 51 4.2 KshA structures ......................................................................................................... 53 4.2.1 KshA structural similarity ........................................................................ 53 4.2.2 KshAMtb in complex with ADD ............................................................... 54 4.2.3 KshA1 in complex with 1,4-BNC-CoA ................................................... 55 4.2.4 Similarity to other Rieske oxygenases ..................................................... 55 vii  4.3 KshA5 structural flexibility ....................................................................................... 56 4.3.1 Gating of the active site ........................................................................... 57 4.3.2 Structural correlation of broad specificity in KshA5 ............................... 58 4.3.3 Movement within the catalytic center ...................................................... 60 4.3.4 Movement of the bridging Asp ................................................................ 63 4.4 Concluding remarks .................................................................................................. 64 Bibliography ................................................................................................................................ 65 Appendix ...................................................................................................................................... 69 Appendix A  Data collection and refinement statistics ........................................................ 69    viii  List of Tables  Table 1.1   Example reactions catalyzed by Rieske oxygenases .................................................... 3 Table 1.2   Reactions catalyzed by NDO ........................................................................................ 4 Table 1.2   Activity of KshA1 and KshA5 of R.rhodochrous DSM 43269 (47) .......................... 15 Table 2.1   Oligonucleotides ......................................................................................................... 19 Table 2.2   Strains and plasmids used ........................................................................................... 20 Table 3.1   Coupling of NADH consumption to substrate turnover ............................................. 29 Table 3.2   Apparent steady-state kinetic parameters of KshA1 ................................................... 32 Table 3.3   Apparent steady-state kinetic parameters of KshA5 ................................................... 32 Table 3.4   Ligand-iron (Fe)-substrate  distances and structural alignment.................................. 38 Table 4.1   Inter-subunit distances of ?bridging? Asp-His of mononuclear and Rieske domains . 63    ix  List of Figures  Figure 1.1   Trimeric arrangement of Rieske oxygenases .............................................................. 5 Figure 1.2   Structure of OMO ........................................................................................................ 6 Figure 1.3   Active site channels observed in structurally characterized Rieske oxygenases ........ 7 Figure 1.4   Interaction of metallocenters in NDO ......................................................................... 8 Figure 1.5   Rieske oxygenase reaction cycle proposed for NDO (18) .......................................... 9 Figure 1.6   Possible routes for product formation (18) ................................................................ 10 Figure 1.7   Reaction catalyzed by KshAB ................................................................................... 12 Figure 1.8   Surface map of KshAMtb ............................................................................................ 13 Figure 1.9   Active site channel openings in Rieske oxygenases .................................................. 14 Figure 1.10   Structure of KshAMtb 2ZYL..................................................................................... 16 Figure 3.1   SDS-PAGE analysis of KshA preparations ............................................................... 26 Figure 3.2   KshA absorption spectra............................................................................................ 27 Figure 3.3   GC/MS fragmentation pattern ................................................................................... 28 Figure 3.4   Steroid substrates used for kinetic analysis. .............................................................. 30 Figure 3.5   Steady state kinetic curves of KshA1 ........................................................................ 31 Figure 3.6   Progress curves of KshA5 with initial concentrations of 8.3 ?M (A) and 50 ?M (B) ADD .............................................................................................................................................. 33 Figure 3.7   Steady state kinetic curves of KshA5 ........................................................................ 34 Figure 3.8   The steady-state utilization of O2 by KshA ............................................................... 35 Figure 3.9   Examples of KshA crystals ....................................................................................... 38 Figure 3.10   Electron density in the active site of KshAMtb:ADD ............................................... 39 Figure 3.11   Superposition of KshAMtb:ADD (cyan) and KshAMtb (gray) .................................. 40 Figure 3.12   KshA1:1,4-BNC-CoA (gray) and KshA5 (cyan) hybrid crystal packing arrangement................................................................................................................................... 41 Figure 3.13   Substrate binding in KshA1:1,4-BNC-CoA ............................................................ 43 Figure 3.14   Superposition of KshAMtb (gray) and KshA5 (cyan) ............................................... 44 Figure 3.15   Superposition of KshAMtb (gray) and KshA5 (cyan) ............................................... 45 Figure 3.16   The substrate-binding site of KshA5:ADD ............................................................. 46 x  Figure 3.17   Superposition of KshA5:ADD (cyan) and KshA5 (gray) ....................................... 47 Figure 3.18   Active site of KshA5:1,4-BNC-CoA ....................................................................... 48 Figure 4.1   Phylogenetic tree of KshAs (47). .............................................................................. 50 Figure 4.2   Superposition of KshAMtb and KshAMtb:ADD .......................................................... 54 Figure 4.3   Active site opening of KshA showing space available for secondary binding ......... 57 Figure 4.4   Mouth loop arrangements of KshA5 ......................................................................... 58 Figure 4.5   Superposition of KshA5 and KshA5:ADD ............................................................... 61    xi  List of Abbreviations  ADD    1,4-androstadiene-3,17-dione BLAST  basic local alignment search tool 1,4-BNC   3-oxo-23,24-bisnorchol-1,4-dien-22-oic acid 4-BNC   3-oxo-23,24-bisnorchol-4-en-22-oic acid 1,4-BNC-CoA  3-oxo-23,24-bisnorchol-1,4-dien-22-oyl coenzyme A thioester CoA    coenzyme A ENDOR  electron nuclear double resonance FAD    flavin adenine dinucleotide FAS    ferrous ammonium sulfide GC-MS   gas chromatography-coupled mass spectrometry HPLC    high performance liquid chromatography IPTG    isopropyl "-D-1 thiogalactopyranoside KstD    3-ketosteroid-?1-dehydrogenase KshAB   3-ketosteroid 9?-hydroxylase LB    lysogeny broth MCD   magnetic circular dichorism Mtb    Mycobacterium tuberculosis NADH    nicotinamide adenine dinucleotide (reduced) NAD+    nicotinamide adenine dinucleotide (oxidized) NAPS    Nucleic Acid Protein Service Unit NMR    nuclear magnetic resonance 9-OHAD   9-hydroxy 4-androstene-3,17-dione P450    cytochrome P450 PCR    polymerase chain reaction PDB    Protein Data Bank RO    Rieske oxygenase RMSD    root mean square deviation SDS-PAGE   sodium dodecyl sulfate polyacrylamide gel electrophoresis TB   tuberculosis UV-vis   ultraviolet-visible xii  Acknowledgements   I offer my gratitude to my supervisor Dr. Lindsay Eltis for his unwavering support, encouragement, and guidance.   I thank my committee members Dr. Lawrence McIntosh and Dr. Natalie Strynadka for their valuable feedback and support.   I would like to thank all the members of the Eltis lab who have helped me grow and learn as a graduate student especially for their knowledge, support and positive attitudes. Many thanks to Jie Liu and Jenna Capyk for their technical assistance, especially in the beginning of my studies.   Thanks to Dr. Liam Worrall for his guidance and help in the technical aspects of the crystallographic work and for his insight into the structural data.     xiii  Dedication     To my family       1. Introduction 1.1 Reactivity of dioxygen in biology Dioxygen (O2) plays a pivotal role in many processes of life, serving as a terminal electron acceptor in aerobic respiration and to activate organic compounds in many anabolic and catabolic processes (1). The reaction between triplet O2 and singlet carbon is thermodynamically favorable but spin forbidden, making it kinetically slow (2). In general, this limits the reactivity of O2 with organic compounds. However, biological systems utilize enzymes to control the reactivity of this highly useful and abundant species. 1.1.1 Oxygenases Oxygenation is often the initial step in the bacterial catabolism of unactivated carbon substrates. Enzymatic catalysis by oxygenases is facilitated through the use of a transition metal, organic cofactor or both and accelerates the addition of oxygen by activating the substrate for easier attack by O2 or through an activated oxygen intermediate for addition to substrate (1). This approach enables oxygenases to utilize the high potential reactivity of O2 by reductively activating to the singlet or doublet species. This highly reactive intermediate must be well controlled and regulated by the enzyme to limit the production of reactive oxygen species (ROS), which cause oxidative damage to cellular components. Nevertheless, oxygenases have evolved to catalyze highly specific mono- and diooxygenation of a wide variety of substrates (3). Monooxygenases utilize one atom of O2 and one reducing equivalent for the oxygenation of a single substrate position, releasing H2O as a byproduct. Dioxygenases catalyze the addition of both atoms of O2 into the substrate, requiring two reducing equivalents. They typically affect a cis-dihydroxylation reaction. 1.1.2  Non-heme Fe activation of oxygen While numerous enzymes have evolved for specific oxygenase chemistry, many examples use iron as the redox metal (1). Those utilizing a mononuclear iron ion in the catalytic 12  center include extradiol or intradiol dioxygenases, 2-oxo acid-linked dioxygenases, and Rieske-type oxygenases. In these enzymes, the iron is coordinated on one face by two His and one Asp or Glu, leaving two or three coordination sites available for reaction with substrate or O2. This versatile ?2-His-1-carboxylate facial triad? is utilized by many enzymes for oxidation reactions, many of which share similar reaction mechanisms and intermediates. 1.2 Rieske oxygenases Rieske oxygenases (ROs) are multi-component enzymes that use NAD(P)H-derived electrons and O2 to effect a large variety of oxidative reactions utilizing the 2-His-1-carboxylate facial triad (4). While predominantly expressed in bacterial synthesis and for the degradation of organic compounds, they are also utilized in biosynthetic roles by animals and plants. ROs across different organisms exhibit a well conserved two-domain structure, although their sequences are highly divergent.  1.2.1 Phylogenetic diversity Previous research on ROs has mainly focused on aromatic dioxygenases (4-5). However, the latter represents a small fraction of RO functional and taxonomic diversity. High sequence diversity prevents identification of other ROs by techniques such as BLAST and has limited previous phylogenetic analyses to sequentially similar enzymes. Additionally, structural data for ROs is highly biased towards those expressed in easily cultureable bacteria. A phylogenetic analysis based on structurally conserved regions has revealed that most well characterized ROs cluster closely together, leaving much of the sequence landscape unexplored (6). Indeed, nearly all RO structures in the PDB come from only four orders of bacteria. Study of unique ROs will add to the understanding of this catalysis towards practical applications with numerous diverse and uncharacterized enzymes. Indeed, ROs characterized to date cover little of their potential substrate range and catalysis, with the majority of study focusing on the catabolism of aromatic hydrocarbons.  3        1.2.2 Reactions catalyzed The range of oxidations catalyzed by ROs (Table 1.1) in addition to the well characterized mono- and dioxygenation  include  N-demethylation (7), O-demethylation (8), desaturation (9), methyl group hydroxylation (10) and, more recently, alkane hydroxylation (11). ROs are able to catalyze a wide array of reactions from a conserved catalytic center to diverse substrates in a site- and stereo-specific manner. Indeed, ROs such as naphthalene dioxygenase (NDO) (12), toluene dioxygenase (TDO) (13) and biphenyl dioxygenase (BPDO) (14) have been shown to catalyze a variety of oxidations of a range of substrates (Table 1.2), demonstrating potential for stereo- and regio-specific reactions towards the generation of high-value chemicals or development of alternate synthesis routes (15-16) and have been shown to be conducive to enzyme engineering (17). Furthermore, the promiscuous substrate selectivity of ROs and ability to oxygenate recalcitrant aromatics makes them ideal candidates for the remediation of polluted environmental sites (16). Study of ROs outside the narrow range of well-characterized aromatic dioxygenases provides a high potential for novel catalysis and insight into oxidation mechanisms. Despite their considerable biocatalytic potential (5,14), critical aspects of RO function have yet to be elucidated. Table 1.1   Example reactions catalyzed by Rieske oxygenases Monohydroxylation  N-demethylation  O-demethylation  Desaturation  4    1.2.3  Structure of Rieske oxygenases ROs are composed of an oxygenase, a reductase and in some cases a ferredoxin. The latter transfer reducing equivalents from NAD(P)H to the oxygenase, which is usually trimeric (?3) (6). The oxygenase contains two domains: the N-terminal domain harbors a Rieske [2Fe-2S] cluster coordinated by two cysteines and two histidines; and the larger C-terminal domain harbors the mononuclear iron which is coordinated by the ?2-His-1-carboxylate facial triad? and orchestrates the oxidative chemistry as identified through spectroscopic studies demonstrating binding of substrate and activation of O2  (18). The latter catalytic domain minimally comprises a Table 1.2   Reactions catalyzed by NDO Reaction type Substrate Reaction Dioxygenation naphthalene  Monooxygenation indan  Benzylic Monooxygenation toluene  Desaturation indan  O-demethylation anisole  N-demethylation N-methylaniline  Sulfoxidation methyl phenyl Sulfide   5  central 7-stranded anti-parallel ?-sheet flanked by four ?-helices on the face distal to the Rieske domain (4,19). Each iron-coordinating residue is located on a helix with the two histidines on short helices and the carboxylate residue across the catalytic site located on the enzyme?s largest ?transverse? helix, spanning most of the catalytic domain. The Rieske center is responsible for accepting reducing equivalents and transferring them to the catalytic iron. The two centers are separated by 45 ? but the trimeric arrangement observed in ROs positions the monomers in a head-to-tail fashion (Figure 1.1). This provides a 13 ? separation between the metallocenters of adjacent subunits, a distance that is amenable to electron transfer.  Figure 1.1   Trimeric arrangement of Rieske oxygenases. 2-Oxoquinoline-8-monooxygenase from Pseudomonas putida 86 (OMO) is shown with two protomers colored grey and one blue and green representing the Rieske and catalytic domains, respectively. The irons are colored orange and distances between metallocenters are labeled. Figure generated using PyMol (20) and PDBID 1Z01 (21). 44 ? 13 ? 6    Figure 1.2   Structure of OMO. OMO is shown with the Rieske domain on the left and catalytic domain on the right. The ?transverse? helix is blue and the helices containing the His ligands to the orange mononuclear iron are colored green. The active site pocket opens to the right between the yellow ?-sheet and the green ?-helices as indicated by an arrow. Figure generated using PyMol (20) and PDBID 1Z01 (21). 1.2.4  Rieske oxygenase active site Protein crystallography demonstrates similar active site architecture for all the structurally characterized ROs. A mainly hydrophobic substrate pocket is composed primarily of residues from the catalytic subunit?s ?-sheet as well as residues from surrounding ?-helices (Figure 1.2). The above-mentioned coordination shell of the iron is completed by one or two additional solvent species. The active site channel is typically orthogonal to the transverse helix, opening to solvent from between the ?-sheet and nearby helices or adjacent loops (Figure 1.3). This channel has been observed through protein crystallography of ROs to exist in open and/or closed states with positioning of the backbone or individual residues blocking the opening (22-24). This is a common feature observed in oxygen-activating enzymes and may aid in occluding the active site from solvent and preventing the adventitious production of reactive oxygen species (ROS) which may be damaging to cellular components. Indeed, ?mouth? loop regions at 7  the entrance to the active site are typically disordered and may exhibit flexibility in accommodating substrates of different sizes and shapes (4). The two metallocenters are bridged by a conserved Asp (Figure 1.4) that has been observed in some crystallographic structures to H-bond between histidine ligands of both metal centers across the subunit interface (21), and has been demonstrated to be necessary for catalysis in NDO (25), phthalate dioxygenase (PDO) (26-27) and anthranilate 1,2-dioxygenase (AntDO) (28).  NDO   OMO      KshA Figure 1.3   Active site channels observed in structurally characterized Rieske oxygenases. Shown is the substrate pocket of the catalytic domain with the same color scheme as Figure 1.2. Substrate is shown in magenta. Viewing angle is orientated as looking down into the active site channel along the axis connecting the mononuclear and Rieske domains. Figure generated using PyMol (20) and PDBIDs 1NDO (left), 1Z01 (center) (21) and 2ZYL (right) (19) with an arrow indicating the active site channel. 1.3. Catalysis by Rieske oxygenases The coordination of multiple reaction components and creation and utilization of a highly reactive oxygen species necessitates a high degree of complexity in catalysis by ROs. Although the reactive intermediates of ROs are not known, structural and kinetic studies have developed a consensus (24) for the general regulation mainly from the study of NDO, benzene dioxygenase (BZDO), PDO, OMO, and carbazole dioxygenase (CARDO).  This has led to a hypothesized multi-step reaction cycle which is regulated by the presence of substrate and the oxidation state of the Rieske cluster. The reaction sequence leading to precise oxygenation was investigated by  8  single turnover studies of NDO demonstrating the ability of fully reduced oxygenase component alone to produce product at a 0.85:1 ratio to intact mononuclear iron leaving the enzyme in its fully oxidized state thus requiring two electrons for complete turnover (18). While the reductase component was not required, both substrate presence and a reduced state of the Rieske cluster were necessary for O2 reactivity. This indicated regulation of O2 reactivity as well as demonstrating the electron-supplying role of the reductase, providing initial clues as to the ability of Rieske oxygenases to control the activation and utilization of O2. 1.3.1  Rieske oxygenase reaction cycle At least three oxidation states between the two metallocenters are thought to exist in resting Rieske oxygenases (24): fully reduced, fully oxidized, and intermediate with an oxidized Rieske cluster and reduced mononuclear iron representing different points along the proposed reaction cycle (Figure 1.5). The mononuclear iron switches between ferrous and ferric oxidation states   Figure 1.4   Interaction of metallocenters in NDO.  The mononuclear iron on the left and the Rieske cluster on the right are in adjacent subunits. Asp205 ?bridges? the metallocenters by H-bonding to histidine ligands of each of them. Iron is orange, sulfur is yellow, carbon is gray, nitrogen is blue, and oxygen is red. This structure was determined in the absence of substrate, which would bind to the upper left of the mononuclear iron. Figure generated using PyMol (20) and PDBID 1NDO (29). 2.7 ? 2.8 ? Asp362 His208 Asp205 His83 His104 His213 9  while the cluster contains one ferric iron coordinated by cysteins and a second iron coordinated by histidines which switches between ferric and ferrous upon cluster reduction. The starting state of the reaction cycle contains a ferrous catalytic iron and oxidized Rieske cluster without substrate or O2 bound. Electrons are transferred to this un-reactive state from the reductase or ferredoxin, resulting in reduction of the Rieske cluster. Substrate is then bound to the active site and O2 coordinated to the mononuclear iron, which reduces and activates O2 to a reactive intermediate. This activated oxygen species then attacks substrate during which the Rieske center passes on an electron to the active site, producing the oxygenated product and leaving both iron centers in the oxidized state. Some evidence has been shown that the catalytic iron must be subsequently reduced in order to allow release of product and regeneration of the starting state (30). While the specifics of this reaction cycle likely differ between ROs, conservation of many important residues and the observation of similar enzymatic behavior indicate a common mechanism for the activation and utilization of oxygen.   Figure 1.5   Rieske oxygenase reaction cycle proposed for NDO (18). 10  1.3.2. Active O2 species The use of a ?2-His-1-carboxylate facial triad? by ROs suggests that an activated Fe-oxygen species is responsible for the reaction with substrate. Indeed, crystallographic structures of O2 bound to iron in the presence of substrate have been obtained for NDO (31) and CARDO (24) in a side-on fashion, suggesting a straightforward mechanism for the cis di-hydroxylation in dioxygenases. While the activated oxygen species may vary with different reactions catalyzed by ROs, it is generally thought to initially exist in the peroxo state [Fe(III)-(hydro)peroxo] before either reacting with substrate or cleaving O-O, resulting in a Fe(V)-oxo-hydroxo active species (Figure 1.6) (4). The activated species must be tightly controlled by the enzyme as uncoupling readily produces H2O2 or water in a one or two electron reduction of O2 resulting in futile consumption of NADH. Positioning the substrate adjacent to the iron and gating oxygen activation to substrate binding likely prevents the formation of uncoupled products. Binding and activation of O2 must be well controlled by ROs in order to remain active during turnover of substrate. For example, the generation of H2O2 during the uncoupled reaction of NDO with benzene has been observed to inhibit and inactivate the enzyme in its reduced form (32). 1.3.3.  Regulation of the catalytic cycle Kinetic and structural studies provide evidence that the reaction cycle is regulated by the presence of substrate and the oxidation state of the Rieske cluster. The use of NO as a O2 analogue has revealed that the small molecule does not bind iron until substrate is present and the Rieske cluster is reduced (33), suggesting that a mechanism for gating O2-binding and communication between subunits exists. Spectroscopic and crystallographic studies have  Figure 1.6   Possible routes for product formation (18). 11  demonstrated a reorganization of the catalytic iron center upon both substrate binding and reduction of the cluster in which the catalytic iron moves away from the substrate. Electron nuclear double resonance (ENDOR) of NDO established that reduction of the cluster results in a 0.5 ? increase in the distance between the mononuclear iron and substrate (33). Furthermore, an allosteric effect of the Rieske redox state on the positioning of the catalytic Fe has been observed in OMO (21) crystallographic structures in which the iron and a corresponding Histidine ligand move ~0.8 ? away from the substrate binding pocket upon cluster reduction. Additionally, crystallographic structures of BPDO  (14) and CARDO (24) demonstrate a shift in both the catalytic iron and the helices coordinating it away from substrate upon its binding. Regulation of the catalytic center?s electronic state has also been observed. Near-IR magnetic circular dichorism (MCD) and variable temperature, variable field (VTVH) MCD of NDO (34) and PDO (35-36) indicate a resting state 6-coordinate octahedral ferrous site with a weak axial ligand. Reduction of the cluster tightens the coordination of the sixth ligand whereas binding of naphthalene eliminates a water ligand from the coordination sphere, providing access for O2. Reduction of naphthalene-bound NDO causes a shift in 1/3 of the catalytic sites from square pyramidal to trigonal bipyramidal. Though this demonstrates regulation of the catalytic cycle through geometric and electronic reorganization of the mononuclear iron upon substrate binding, little is known of how the substrate causes this change and influences the reactivity of the catalytic iron with O2.  1.4 3-Ketosteroid-9?-hydroxylase (KshAB) 3-Ketosteroid-9?-hydroxylase (KshAB) is a two-component RO involved in the bacterial catabolism of steroids (19,37-38). This catabolism and the enzyme have been studied for the biocatalytic production of high-value steroids (15,39) and for their contribution to the pathogenicity of Mtb (40). Steroid catabolism is an almost ubiquitous characteristic of mycolic acid-producing actinobacteria (41-42) with some strains containing paralogous pathways responsible for catabolizing different steroids. For example, Rhodococcus jostii RHA1 contains four clusters of steroid degradation genes, the first and third of which are responsible for the catabolism of cholesterol and cholate, respectively (41-42). Steroid catabolism appears to be organized according to three parts of the steroid molecule (43): the side chain at C17, rings A/B  12   and rings C/D, respectively. The side chain is degraded by a process similar to the ?-oxidation of fatty acids involving CoA thioester intermediates (44). KshAB is involved in degrading rings A/B, catalyzing the 9?-hydroxylation of 3-keto-4-ene and 3-keto-1,4-diene steroids (Figure 1.7) (37). The hydroxylation of the latter results in the non-enzymatic opening of ring B with the concomitant aromatization of ring A.  Based on the higher apparent specificity (kcat/Kmapp) of KshAMtb for substrates with a CoA thioesterified side chain, it has been proposed that side chain and rings A/B degradation occur concurrently at least to some extent in actinobacteria (38).  Nevertheless, the substrate specificities of the different steroid catabolic pathways, as well as those of key enzymes such as KshAB, remain largely unexplored.     Figure 1.7   Reaction catalyzed by KshAB. 9?-Hydroxylation of a 3-keto-4-ene steroid and 3-keto1,4-diene steroid resulting in concomitant aromatization of ring A and opening of ring B. A B C D 13  1.4.1.  KshA structure An X-ray crystallographic structure of KshAMtb (PDBID 2ZYL) (19), the oxygenase component from Mtb, solved to 2.3 ? revealed that KshA shares the functionally important elements of ROs while utilizing a smaller number of residues. A number of insertions seen in other ROs are absent in the catalytic domain. In addition, the substrate access channel is angled at ~90? with respect to those of ROs characterized to date and is significantly longer (~28 ?) as measured from the mononuclear iron to the surface of the protein (6). More specifically, the large channel entrance (Figure 1.8) of KshAMtb is located between the N-terminus of helix ?5 and the ?mouth loop?, which connects two ?-strands (?16 and ?17 in KshAMtb). By contrast, the channel entrance occurs between the second and third strands of the central ?-sheet in OMO (21) and CARDO (24) (corresponding to ?14 and ?15 in KshAMtb) and is similarly oriented in other ROs (Figure 1.9). The differently positioned substrate access channel of KshAMtb appears to be necessary to allow proper orientation of the large steroid substrate and C-17 tail within its binding pocket. Dicamba monooxygenase (DMO), which has a similar catalytic domain to KshA  Figure 1.8   Surface map of KshAMtb. The deep and spacious active site opening is in the center of the image with carbon, oxygen, nitrogen and sulfur colored gray red, blue, and yellow, respectively. The ?transverse? helix is directly left of the opening and ?mouth loop? to the right. Figure generated using PyMol (20) and PDBID 2ZYL (19). 14  exhibits a potential access channel opening in the same region (45), although it was only observed in the presence of a non-physiological metal and is therefore of unclear significance.    Figure 1.9   Active site channel openings in Rieske oxygenases. KshAMtb and CARDO surface representations showing catalytic iron in orange and active site opening in black. Carbon is colored gray, oxygen is red, nitrogen is blue, and sulfur is yellow. Both enzymes are in the same orientation with the Rieske domain to the left and catalytic domain to the right. The KshA opening points down and the CARDO opening points directly right. Figure generated using PyMol (20) and PDBIDs 2ZYL (19) and 1WW9 (46). 1.4.2  KshA1 and KshA5  Rhodococcus rhodochrous DSM43269 contains five KshA homologs: KshA1-A5 of 53-63% amino acid identity (Table 1.1) (47). While the genomic context of their genes is unknown, studies have provided some insight into their physiological roles (42). For example, KshA1 has been assigned to cholate catabolism based on its reciprocal best hit 72% amino acid sequence identity with the RHA1 homolog encoded by the cholate catabolic cluster (ro05811), the enzyme?s preference for a substrate with a side chain at C-17, the up-regulation of kshA1 during growth of DSM43269 on cholate, and the ability of kshA1 to restore growth of a kshA null mutant on cholate. Similar analyses revealed KshA5 to be remarkably versatile. This enzyme, with the RHA1 reciprocal best hit Ro02490 (72% amino acid sequence identity), transformed a remarkably broad range of ketosteroids with similar specific activity. Moreover, kshA5 expression was induced by each of four tested steroids: cholesterol, progesterone, 4-androstene-3,17-dione (AD), and cholate. Finally, a DSM43269 mutant containing only kshA5 (i.e.,  KshA CARDO 15  disrupted in each of the four other kshA genes) grew on each of these four steroids. Coexpression and copurification of KshA and KshB with E.coli was utilized to examine the substrate preference of KshA1 and KshA5 (Table 1.2). Remarkably similar activities were recorded for KshA5 for nearly all substrates tested whereas KshA1 showed a preference for steroids with a longer side chain. 9? Hydroxylation of steroid substrate was confirmed using 4-androstene-3,17-dione.   Table 1.3   Activity of KshA1 and KshA5 of R.rhodochrous DSM 43269 (47)  16  An alignment of KshA from DSM43269 shows highly similar sequence throughout the substrate-binding domain with the exception of a disordered loop observed in KshAMtb. Nearly all residues predicted to be involved in binding of substrate in KshAMtb (Val176, Gln204, Tyr232, Met238, Asn240, Phe301, and Trp308) were conserved in the Rrho homologues except for Val176 which is Ile in KshA1 and Asn240 which is Asp in KshA1 and KshA5. Substrate specificity in ROs has been observed to be primarily due to the fit and orientation of substrate within the active site, though some influence has been observed from residues at the opening of the active site (4). Although study of substrate binding and orientation has been mostly focused on regio-specificity of hydroxylation (48), some work has been done on substrate selectivity: Chimeric enzymes composed of sections exchanged between KshA1 and KshA5 have been produced and tested for substrate activities (17). Exchanging the beta sheets of the helix-grip fold resulted in activities resembling that of the donor enzyme whereas chimera of the variable loop   Figure 1.100   Structure of KshAMtb 2ZYL. Residues from sequence alignment with the ?-sheet residues exchanged between KshA1 and KshA5 from (17) highlighted in blue and iron in orange. Figure generated using PyMol (20) and PDBID 2ZYL (19). (Mouth loop) 17  region had a similar but smaller effect on substrate preference. Specifically, the exchange of residues 204-261 in KshA5 with 198-255 in KshA1 provided a KshA5 chimer (Figure 1.10) with greater activity for ADD, testosterone, progesterone, and 4-BNC while reducing or eliminating activity with other substrates. While KshA1 chimeras containing residues 204-261 of KshA5 exhibited lower activities for the latter substrates, an increase in activity for KshA5?s preferred substrates was not observed, suggesting that the ?-sheet and loop region are necessary but not sufficient for KshA5?s broad specificity. Although these altered activities are similar to the enzyme donating the ?-sheet, varying ratios of oxygenase to reductase between chimera complicate the interpretation. However, it is clear that the ?-sheet and mouth loop influence substrate selectivity. Similarly, a study of BPDO found that exchanging residues between BPDOLB400 ?type (broad range specificity) and BPDOKF707 ?type (narrow specificity) in four regions differing between the enzyme types yielded specificities resembling the donor enzyme for regions III and IV. These two regions are found at the opening of the active site to solvent (49). Further knowledge of the determinants of substrate preference will aid in the modification and design of Riekse oxygenases towards higher activities for specific substrates. 1.5 Aim of this study To better characterize KshA, we investigated the substrate specificity and structure of KshA1 and KshA5 from DSM43269 along with the structure of KshAMtb. KshA1 and KshA5 were purified anaerobically to maximize their activities. Their activities were reconstituted using KshB from M.tuberculosis.  Steady-state kinetic parameters for each of seven structurally diverse steroids, including a CoA thioester, were determined together with the coupling of NADH and O2 consumption with substrate hydroxylation. X-ray crystallographic structures were solved of KshA5 in a substrate-free state as well as in complex with each of ADD and 1,4-BNC-CoA. Similarly, the structures of KshA1 and KshAMtb were determined in complex with steroid substrate. The findings are discussed with respect to the substrate specificity of these unique monooxygenases and the mechanism of ROs.   18  2. Materials and Methods 2.1 Chemicals and reagents All reagents were of at least analytical grade unless otherwise noted. 4-Estren-3,17-dione, testosterone, ADD, 3-oxo-23,24-bisnorchol-4-en-22-oic acid (4-BNC), and 5?-androstan-3,17-dione were purchased from Steraloids, Inc. (Newport, RI). Steroids contained <5% impurities as judged by HPLC and GC-MS. Substrates 1,4-BNC and 1,4-BNC-CoA were enzymatically produced using CasI and KstD as described previously (38). Restriction enzymes and the Expand high fidelity PCR system were purchased from New England Biolabs (Ipswich, MA) and Roche Applied Science (Laval, Quebec, Canada), respectively. Water for buffers was purified using a Barnstead Nanopure DiamondTM system (Dubuque, Iowa) to a resistance of at least 18 megaohms.  2.2   Bacterial strains and culture conditions Escherichia coli BL21 (DE3) and GJ1158 were routinely used for protein production and E. coli DH5? was used for DNA production and propagation (Table 2.2). GJ1158 cells were grown on lysogeny broth (LB) without salt while all other strains were grown on standard LB at 37 ?C, 200 r.p.m. unless otherwise noted. Single colonies of bacteria were obtained by incubating cells on LB agar plates with appropriate antibiotics at 37 ?C and used for DNA propagation or enzyme production by incubating overnight in 50 mL LB followed by inoculation of 1 L of LB with cell suspension to an optical density at 600 nm (OD600) of 0.01. Electro-competent cells were produced by growing cultures to an OD600 of 0.5 and cooling cells on ice for 10 min. Cells were kept at 4 ?C or below, harvested by centrifugation for 10 min at 4700 x g. and washed three times with 20 mM potassium phosphate with 10% glycerol. Cells were stored as 30 ?L aliquots at -80 ?C until further use. Cells were transformed using a MicroPulser electroporation apparatus from Bio-Rad (Hercules, CA) with Bio-Rad 0.1 cm GenePulser Cuvettes. In a cold cuvette, 2 ?L of plasmid (10-50 ng ?L-1) was mixed with 20 ?L of freshly thawed competent cell suspension. Cells were 19  pulsed at 2.0 kV and transferred to 0.5 mL LB and incubated for 45 min at 37 ?C. Cells were transferred to LB agar plates containing appropriate antibiotic by spreading 10-50 ?L of cell suspension and were incubated overnight at 37 ?C to provide individual colonies. 2.3 DNA manipulation and plasmid construction DNA was propagated, digested, and ligated using standard protocols (50), and plasmids were purified as described previously (51). Oligonucleotides were purchased from Integrated DNA Technologies (San Diego, CA) through the Nucleac Acid Protein Service Unit (NAPS) at the University of British Columbia and are listed in Table 2.1.    Plasmids for protein expression were constructed by ligating PCR-amplified DNA fragments containing the gene of interest into the multiple cloning sites of restriction enzyme-digested vectors. The kshA1 and kshA5 and kshAMtb genes, previously isolated by Petrusma et al. (47) (kshA1 and kshA5) and Capyk et al. (19) (kshAMtb), were subcloned into the pT7HP20 expression vector (from Eltis et al. (52)) such that the protein is produced with an N-terminal polyhistidine tag that can be removed using Factor Xa. The nucleotide sequence of each gene in its resulting construct, pT7KA1, pT7KA5 and pT7KAMtb, was confirmed by NAPS. All constructs were free of mutations except for a single base pair substitution in pT7KAMtb which did not affect the translated protein sequence. Table 2.1   Oligonucleotides Gene Expression  plasmid Nucleotide sequence Restriction  site kshA1 pT7KA1      fwd 5?-TGAGCTAGCAGCCTCG CACTTCCGAACAATC-3? NheI  rev 5?- ACTAAGCTTCTAGCC CGCGGTGGTGGACTG-3? HindIII kshA5 pT7KA5      fwd 5?-CTTGCTAGCTCCATCG ACACCGCACGGTC-3? NheI  rev 5?-ACTAAGCTTCTAG GGGGTCGCGGTGGAGC-3? HindII Rv3526 pT7KAMtb fwd 5?-TAAGCTAGCAGTACC GACACGAGTGGGGTCG-3? NheI  rev 5?-ATCAAGCTTTCAGT GTTGCTCGGCGGGC-3? HindIII 20   2.4 Protein production and purification All KshA protein was heterologously produced using E. coli GJ1158 containing pT7KA1, pT7KA5, or pT7KAMtb. Cells were grown in low salt LB supplemented with 100 ?g/mL ampicillin and minerals (53). One liter of medium was inoculated with 3 mL of an overnight culture and was incubated at 25 ?C. When the culture reached an OD600 of 0.5, kshA expression was induced by adding 50 mL of 5 M NaCl. The culture was incubated a further 18 hours at 25 ?C and then harvested by centrifugation. Cell pellets were washed twice with 20 mM sodium phosphate, pH 8.0 containing 5% glycerol and frozen at -80 ?C until use. 2.4.1 KshA A cell pellet obtained from 3 L of culture was suspended to a final volume of 60 mL in binding buffer (20 mM sodium phosphate, pH 7.4, 300 mM NaCl, 20 mM imidazole). A small amount of DNaseI (~0.5 mg) was added before the cells were lysed using an Emulsiflex-05 homogenizer (Avestin, Ottawa, Canada) operated at 10,000 p.s.i. Cell debris were removed by ultracentrifugation at 10,000 x g for 45 min and the amber-colored supernatant was transferred to a Wheaton bottle and subsequent steps were performed anaerobically essentially as described for KshAMtb using a glove box maintained at <5 ppm O2 (19). The supernatant was filtered using a 30 ?m membrane. Table 2.2   Strains and plasmids used    Strain Relevant characteristic Plasmids Gene Characteristics E. coli DH5? DNA propagation pA1rho3 pA5rho2 KshA1 kshA5 Kanr Kanr E. coli BL21(DE3) IPTG induction, KstD & KshB expression pETKD1 pETKB3 KstD KshB Tetr, (His x 6) Kanr, N-(His x 6) Thrombin E. coli GJ1158 Salt induction, KshA expression pT7KA1 pT7KA5 pT7KAMtb KshA1 KshA5 KshAMtb Ampr, N-(His x 6) Factor Xa Ampr, N-(His x 6) Factor Xa Ampr, N-(His x 6) Factor Xa 21  The supernatant was loaded onto a Nickel-sepharose (Qiagen) column (1.8 x 4 cm) equilibrated with binding buffer. The column was washed with 4 CV?s (20 mL) of binding buffer, 2 CV?s of wash 2 (20 mM sodium phosphate, pH 6.4, 300 mM NaCl), 2 CV?s of wash 3 (20 mM sodium phosphate, pH 7.4, 300 mM NaCl, 200 mM imidazole) and eluted with elution buffer (20 mM sodium phosphate, pH 7.4, 300 mM NaCl, 500 mM imidazole). A brown fraction containing KshA was eluted and exchanged into 50 mM Tris, pH 7.5, 100 mM NaCl, 1 mM CaCl2 (plus 0.25 M urea for KshA5) and concentrated to ~2-5 mL using a stirred cell concentrator equipped with a YM30 membrane (Amicon, Oakville, Ontario).  His-tags were cleaved using Factor Xa (HTI, Essex Junction, VT) at a molar ratio of 1:1000 Factor Xa:KshA1, 1:500 Factor Xa:KshA5, and 1:500 Factor Xa:KshAMtb. KshA1 and KshA5 required overnight incubation at room temperature while KshAMtb required 4-6 hrs for complete cleavage.  The cleaved protein was loaded onto anion-exchange resin (1 x 10 cm; SourceTM15Q (GE Healthcare)) equilibrated with buffer A (25 mM HEPES, pH 7.5, 5% glycerol, 0.25 mM ferrous ammonium sulfate (FAS), and either 1 mM DTT or 1 mM TCEP). The protein was eluted with a 30 mL linear gradient from 150 to 300 mM NaCl using an AKTA Explorer (Amersham Biosciences). Brown-colored fractions eluting at ~200 mM NaCl (20-25 mS conductance) were combined, exchanged into buffer A, concentrated to 25 mg/mL, flash frozen as beads in liquid N2, and stored at -80 ?C. 2.4.2  KshB production and purification KshB was produced using E. coli BL21 (DE3) containing pETKB3 constructed by Capyk et al. and pPAISC-1. The latter carries the isc genes (54) of P. aeruginosa PA01 involved in FeS cluster assembly. Bacterial growth and protein purification were performed as described previously (19).   22  2.4.3  KstD production  KstD was produced using BL21 (DE3) cells containing the plasmid pETKD1. Bacteria were grown as previously described (38). Cell extract was prepared by subjecting cell suspension to five rounds on an MP Biomedicals FastPrep-24 bead beater (Solon, OH) at speed 5.0 for 20 s. Cell debris was pelleted by centrifugation at 16,100 x g for 10 min at 4 ?C. The resulting supernatant was used for the 3-ketosteroid-?1-dehydrogenation of 4-BNC in the production of 1,4-BNC and 1,4-BNC-CoA. 2.5  Analytical methods  SDS-PAGE was performed using a Bio-Rad MiniPROTEAN III apparatus and a 12% resolving gel. Gels were stained with Coomassie Blue according to standard protocols. Protein concentration was determined using the Micro BCATM protein assay kit (Pierce) using bovine serum albumin as a standard. Acid-labile sulfur content of samples was determined colorimetrically using the N,N-dimethyl-paraphenylene diamine assay and high-potential iron sulfur protein (HiPIP) from Chromatium vinosum as a standard using ?390 = 30,400 M-1cm-1 for the iron sulfur cluster (55). Iron content was determined using the Ferene S assay (56) adapted to a 96-well plate format as described previously (19). KshA5 UV-Visible spectra were recorded using a Cary 60 spectrophotometer. Sodium ferrocyanide and sodium dithionite were used to ensure oxidized and reduced states respectively.  Absorption at 324 nm and sulfur content (corresponding to Rieske 2Fe-2S cluster) were used to determine an extinction coefficient of the oxidized enzyme. KshA1 concentrations were determined using ?324 = 19.6 mM-1, KshA5 concentrations were determined using ?324 = 15.9 mM-1, and KshAMtb concentrations were determined using ?324 = 24.8 mM-1. Prior to analysis, KshA was exchanged anaerobically into 0.1 M potassium phosphate, pH 7.0 by gel filtration chromatography to remove the FAS, DTT or TCEP, and glycerol from the sample. It was then stored in a sealed glass vial, kept on ice, and withdrawn using a gas-tight syringe for each assay. Gas chromatography-coupled mass spectrometry (GC-MS) was performed using an HP 6890 23  series GC system fitted with an HP-5MS 30 m x 250 ?m column (Hewlett-Packard, Palo Alto, CA) and an HP 5973 mass-selective detector. 2.6  Steady-state kinetic analysis Assays were performed in 1 mL of potassium phosphate (I = 0.1 M), pH 7.0 at 22 ?C, unless otherwise stated, using a Cary 60 spectrophotometer equipped with a thermostated cuvette holder (standard assay). Initial reaction velocities were measured by monitoring NADH oxidation at 339 nm (NADH ?339 = 6.22 mM-1 cm-1, NAD+ ?339 = 0 abs). Reaction mixtures containing 120-170 nM KshA1 or KshA5, 1.6 ?M KshB and 0.1 mM NADH were equilibrated for 3 min before the reaction was initiated by adding the steroid substrate. Solutions were prepared fresh daily. KshA was thawed and exchanged anaerobically as described above. The steady-state kinetic parameters for O2 were determined as described previously (19) using 4-BNC with KshA1 and 4-estren-3,17-dione with KshA5. Data were fitted using the program LEONORA to fit steady-state kinetic parameters using the Michaelis-Menten and substrate inhibition equations. 2.7  HPLC analysis  Substrates and products were analyzed using a Waters 2695 Separations HPLC module equipped with a Waters 2996 photodiode array detector and an 250 x 4.60-mm C18 Aqua 5? ODS-Prep column (Phenomenex, Torrance, CA). Reactions were quenched by adding acetic acid to a concentration of 5% and centrifuged to remove precipitated protein. Filtered reaction mixture (70 ?L) was injected onto the column equilibrated with 36% MeOH in water and 0.5% phosphoric acid and eluted with a flow rate of 0.9 mL/min using a multistep linear gradient of methanol from 36-90% over 14 min. 2.8  Coupling measurements Coupling of NADH consumption to substrate turnover was measured using ADD and 4-BNC for KshA1 and ADD and 4-estren-3,17-dione for KshA5. NADH was monitored 24  spectrophotometrically in air-saturated buffer under the conditions of the standard assay. Substrate turnover was assessed by integrating the peak area of the optical absorption of substrate (4-BNC at 243.2 nm or ADD and 4-estren-3,17-dione at 244.3 nm) remaining. Values were corrected using L-phenylalanine as an internal standard and a standard curve of substrate was used to determine the amount remaining in each assay. A coupling ratio was determined by dividing the amount of substrate turned over by the amount of NADH consumed. 2.9  Protein crystallization  Crystals of KshA were grown aerobically at room temperature (21?C) using the hanging drop method and a 24 well plate. Drops of 1 ?L contained a 1:1 ratio of 200-300 ?M KshA (in 25 mM HEPES, pH 7.5, substrate (ADD or 1,4-BNC-CoA), 0.25 mM FAS, 1 mM of reductant either DTT or TCEP) and crystallization well solution.  2.9.1 Crystal screening The Classics and Wizard sparse matrix crystallization screens (Quiagen Inc. Valencia, CA and Emerald BioSystems Inc. Bedford, MA respectively) were used to probe crystallization conditions which were further modified to obtain well diffracting crystals. Candidate well conditions were optimized using a grid of different precipitant or salt concentration and buffer pH. Sparse matrix and additive screens were used in addition to successful well conditions at 80:20, volume working condition to volume additive. For all substrates used in assays co-crystallization was attempted using sparse matrix screens and other successful well conditions.  2.9.2 KshA crystallization  KshA1 crystallized together with KshA5 in a drop containing a mixture of the two enzymes from 100-300 ?M each protomer. The well solution contained 0.5 M NaH2PO4, 0.125 M K2HPO4, 4% PEG-1000, 20 mM Tris, 100 mM phosphate-citrate, pH 4.2. 25  KshA5 without substrate was crystallized using 0.5 M NaH2PO4, 0.125 M K2HPO4, 4% PEG-1000, 20 mM tris, 100 mM phosphate-citrate, pH 4.2. KshA5 in complex with ADD was crystallized using 0.8 M NaH2PO4, 0.2 M K2HPO4, 2% PEG-3000, 20 mM CHES, 100 mM phosphate-citrate, pH 4.2. KshA5 in complex with 1,4-BNC-CoA was crystallized using 0.8 M NaH2PO4, 0.2 M K2HPO4, 0.25 M (NH4)2SO4, 20 mM Ches, 40 mM NaCl and 100 mM phosphate-citrate, pH 4.2.  KshAMtb was crystallized in complex with ADD using 0.5 M NaH2PO4, 0.125 M K2HPO4, 4% PEG-1000, 20 mM tris, 100 mM phosphate-citrate, pH 4.2. Candidate crystals were looped, washed in a cryo-solution containing 20-30% glycerol, and flash-frozen in liquid nitrogen before collection. 2.9.3  X-ray data collection and structure determination Crystals were screened for diffraction at an in-house rotating anode x-ray generator (CuKa radiation, ? = 1.542 ?). X-ray data were collected at the Canadian Light Source (Beamline CMCF1, ? = 1.000 ?). Data was processed using HKL2000 (57) and structures were solved by molecular replacement using KshAMtb together with the programs PHASER (58) and BUCCANEER (59) for model building as implemented in CCP4i. The model was then refined using REFMAC (60). Ligands were built using PRODRG (61) and modeled into obvious density after the arrangement of major structural features was completed. Electron density maps for figures were calculated using the FFT function of CCP4 (62).  Figures were generated using PyMOL (20).   26  3. Results 3.1  Purification of KshA Previous attempts to purify KshA1 and KshA5 revealed that they are O2-labile (63). Consequently, these had been purified in the presence of KshB to maintain their activity during aerobic purification. To obtain catalytically active preparations of KshA that were free of KshB, KshA was purified anaerobically. The yields of purified proteins were ~8 mg KshA1 per liter of culture, ~40 mg/L KshA5, and ~15 mg/L KshAMtb. Purified protein was analyzed by SDS-PAGE (Figure 3.1)    Figure 3.1   SDS-PAGE analysis of KshA preparations. Purified KshA1 (A) 1. Ion exchange column wash 2. Final, 2. Overloaded to view impurities, 4. Cell pellet. Purification of KshA5 (B) 1. Cell pellet pre-induction, 2. Overnight induction, 3. Ni-sepharose purification, 4. Molecular weight standard (kDa), 5. Final protein. Purification of KshAMtb (C) 1. Pre-induction 2. Final sample, 3. Overnight induction, 4. Molecular weight standard (kDa). 3.1.1 KshA1 preparation The KshA1 preparation contained 1.7 ? 0.3 mol of sulfur and 3.9 ? 0.2 mol of iron per KshA1 protomer. The spectrum of oxidized KshA1 had maxima at 455 and 280 nm with a A B C 1             2            3            4 1         2         3        4      5 1        2    3  4 200   116   97  66   45   31   200   116   97  66   45   31   27  shoulder at 324 nm (Figure 3.2A). The R-value (A280/A324) was 6.1 and the specific activity of the preparation was 0.36 ? 0.02 ?mol min-1mg-1 using 50 ?M ADD in 0.1 M potassium phosphate, pH 7.0 at 22 ?C. This activity compares with a value of 0.75 ?mol min-1 mg-1 reported for a mixed preparation of KshA1 and KshB using 200 ?M ADD in 50 mM Tris-HCl, pH 7.0 at  22 ?C (47). 3.1.2 KshA5 preparation The KshA5 preparation contained 1.4 ? 0.4 mol of sulfur and 3.8 ? 0.4 mol of iron per KshA5 protomer. The spectrum of oxidized KshA5 had maxima at 455 and 280 nm with a shoulder at 324 nm (Figure 3.2B). The R-value (A280/A324) was 7.4 and the specific activity of the preparation was 0.15 ? 0.02 ?mol min-1mg-1 using 50 ?M ADD in 0.1 M potassium phosphate, pH 7.0 at 22 ?C. This activity compares with a value of 0.073 ?mol min-1 mg-1 reported for a mixed preparation of KshA5 and KshB using 200 ?M ADD in 50 mM Tris-HCl, pH 7.0 at 22 ?C (47). 300 400 500 6000.00.10.20.30.4AbsorbanceWavelength (nm) reduced oxidized 3 0 4 0 5 0 6 00.00.10.20.3AbsorbanceWav length (nm) reduced oxidized Figure 3.2   KshA absorption spectra. Spectra of KshA1 (A) and KshA5 (B) in the reduced (thick line) and oxidized (thin line, higher shoulder) states. Anaerobic samples contained ~0.15 mg/ml KshA in 100 mM potassium phosphate, pH 7.0 at 22 ?C. A B Wavelength (nm) Wavelength (nm) Absorbance Absorbance 28  3.1.3 KshAMtb preparation The KshAMtb preparation contained 1.4 ? 0.2 mol of sulfur and 5.0 ? 0.3 mol of iron per KshAMtb protomer. The spectrum of oxidized KshAMtb had maxima at 455 and 280 nm with a shoulder at 324 nm. The R-value (A280/A324) was 5.0 and the specific activity of the preparation was 0.17 ? 0.08 ?mol min-1mg-1 using 50 ?M ADD in 0.1 M potassium phosphate, pH 7.0 at 22 ?C. This activity compares with a value of 0.34 ?mol min-1 mg-1 at 50 ?M under the same buffer, pH and temperature derived from Michaelis-Menten kinetic parameters and an R-value (A280/A324) of 5.8 previously reported for KshAMtb (19). 3.2 Reaction products KshA5 transformed each of ADD and 4-estren-3,17-dione to a single product as observed by GC-MS and ESI-MS/MS, respectively. The ADD transformation product had the same GC retention time and mass spectrum as 3-hydroxy-9,10-seconandrost-1,2,5(10)-triene-9,17-dione (3-HSA) (Figure 3.3) (19).   Figure 3.3   GC/MS fragmentation pattern. 9?-hydroxylation product of KshA1 and KshA5 with ADD. Derivatized with TCMS. 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 40005000001000000150000020 0000250000030000003500000400000045000005000000550000060000006500000700000075000008000000m/ z-->Abundanc eAv erage  o f 15 .274  to  15 .290  min .: 6501005.D \ da ta .ms206.2372 .373 .1163 .141 .2105 .1135 .1245 .1 339 .3301 .2273 .3 401 .129  3.3 Coupling Coupling of NADH consumption to product formation in KshA1 was assessed using 4-BNC at 20 and 50 ?M and ADD at 40 and 80 ?M. For KshA5, 4-estren-3,17-dione at 50 ?M and ADD at 20 and 40 ?M were used to determine coupling. In all cases, NADH consumption was well coupled to steroid hydroxylation (Table 3.1). Additionally, catalase was added to reaction mixtures while monitoring with a Clark-type oxygen electrode. In no case was a change in O2 concentration observed upon addition of catalase, signifying that H2O2 was not produced in significant amounts during steroid turnover. Finally, the reaction velocity was observed to reach zero when the amount of NADH consumed, as measured spectrophotometrically, corresponded to the amount of steroid added. These data demonstrate that the oxidation of NADH and consumption of O2 are well coupled to hydroxylation of the steroid substrate. This further demonstrates that the rates of NADH and O2 consumption are accurate measures of steroid hydroxylation.        3.4 Steady-state kinetic analysis KshA1 and KshA5 have previously been shown to have very different substrate preferences (47). To better understand their properties, the substrate specificities of these Table 3.1   Coupling of NADH consumption to substrate turnover Enzyme Substrate Conc. ( ?M) Rxn. time  (Substrate:NADH) KshA1 ADD 40 5 min 1.02 (0.07)  ADD 80 3 min 1.2 (0.2) KshA1 4-BNC 20 3 min 1.1 (0.1)  4-BNC 50 7 min 0.95 (0.05) KshA5 ADD 20 3 min 1.1 (0.1)  ADD 40 5 min 0.99 (0.01) KshA5 4-estren-3,17-dione 50 2 min 1.0 (0.1) 30  enzymes were investigated using seven steroids with C-17 side chains of different length: an alcohol or oxo (AD, ADD), isopropionyl (4-BNC and 1,4-BNC) or a CoA thioester Figure 3.4. In addition, the influence on specificity of ring A saturation and substitution were investigated.  Figure 3.4   Steroid substrates used for kinetic analysis. 3.4.1 KshA1 kinetics  KshA1 displayed Michaelis-Menten kinetics with all substrates except 5?-androstan-3,17-dione (Figure 3.5), which the enzyme did not detectably transform. Consistent with the previously reported substrate preference of KshA1, the enzyme had highest apparent specificity 31  (kcat/Km) for steroids possessing an isopropionyl sidechain at C-17 (Table 3.2). Indeed, the apparent kcat/Km for 4-BNC was three orders of magnitude higher than for 4-estren-3,17-dione and about twice that for 1,4-BNC-CoA.  This was also reflected by the apparent Km values which were 1-2 orders of magnitude higher for substrates with an isopropionyl at C-17 versus an alcohol or oxo. In contrast, the apparent kcat values for the different substrates varied by only five-fold. 0 200 400 600 800 1000 12000.000.020.040.06Rate (?M/sec)4-Estren-3,17-dione (?M)0 50 100. 00.050.10Rate (?M/sec)ADD (?M) 0 100 2000.000.050.10150.200.250.300.350.40Rate (?M/sec)4-BNC (?M)0 5 10 15 20 25 30 35 40 45. 0.020.040.060.080.10.120.140.16Rate (?M/sec)1,4-BNC (?M) Figure 3.5   Steady state kinetic curves of KshA1. Shown is the dependence of initial velocity of NADH consumption on 4-estren-3,17-dione (A), ADD (B), 4-BNC (C), and 1,4-BNC (D). Assays were performed in potassium phosphate (I = 0.1 M, pH 7.0) at 22 ?C and 100 ?M NADH. Curves represent the best fit of the Michaelis-Menten equation using the program LEONORA.  B A C D Rate (?M/s) Rate (?M/s) Rate (?M/s) Rate (?M/s) ADD (?M) 4-Estren-3,17-dione (?M) 4-BNC (?M) 1,4-BNC (?M) 32  Table 3.2   Apparent steady-state kinetic parameters of KshA1 Steroid Substrate kcat (s-1) Km (?M) kcat/Km (s-1mM-1) 5?-Androstan-3,17-dione No Activity - - 4-Estren-3,17-dione 0.5 (0.1) 400 (200) 1.2 (0.3) Testosterone 1.4 (0.3) 110 (20) 13 (2) ADD 1.3 (0.2) 50 (10) 27 (4) 4-BNC 2.6 (0.5) 2.2 (0.2) 1200 (20) 1,4-BNC 0.9 (0.2) 1.0 (0.2) 900 (200) 1,4-BNC-CoA 1.1 (0.3) 2 (1) 600 (400) O2 with 50 ?M 4-BNC 1.7 (0.3) 1.5 (0.2) 1100 (200) Assays were performed using potassium phosphate (I = 0.1 M, pH 7.0) at 22 ?C containing 100 ?M NADH and water at ~260 ?M. Errors in parentheses were calculated from the weighted mean of the error in the enzyme concentration and the standard errors in the parameters estimated using LEONORA.  Table 3.3   Apparent steady-state kinetic parameters of KshA5 Steroid Substrate kcat (s-1) Km (?M) kcat/Km (s-1mM-1) Kss(?M) 5?-Androstan-3,17-dione 0.8 (0.1) 6.1 (0.3) 140 (20)  4-Estren-3,17-dione 0.8 (0.1) 0.8 (0.1) 1000 (200)  Testosterone* 0.8 (0.1) 0.5 (0.1) 1500 (300) 130 (30) ADD1 0.7 (0.1) - -  4-BNC* 0.6 (0.1) 1.2 (0.5) 500 (200) 110 (20) 1,4-BNC* 0.5 (0.1) 0.5 (0.3) 1000 (500) 40 (20) 1,4-BNC-CoA 1.7 (0.6) 500 (200) 3.7 (0.8)  O2 with 5 ?M 4-estren-3,17-dione 0.8 (0.1) 49 (6) 17 (3)  Assays were performed using potassium phosphate (I = 0.1 M, pH 7.0) at 22 ?C containing 100 ?M NADH and water at ~260 ?M. Errors in parentheses were calculated from the weighted mean of the error in the enzyme concentration and the standard errors in the parameters estimated using LEONORA. 1kcat determined by averaging the maximal 3 rates from Figure 3.7C. * Parameters were calculated using substrate inhibition equation, where Km is replaced by Ks 33  3.4.2  KshA5 kinetics  KshA5 catalyzed the hydroxylation of all steroids tested. However, the observed steady-state kinetic behavior was steroid-dependent. Thus, 5?-androstan-3,17-dione, 4-estren-3,17-dione and 1,4-BNC-CoA showed Michaelis-Menten behavior (Fig. 3.7A,B,F). By contrast, the initial rates of ADD, testosterone, 4-BNC, and 1,4-BNC hydroxylation decreased significantly at substrate concentrations above 10 ?M (Fig. 3.7C-E). Inspection of the progress curves, examples of which are shown in Figure 3.6, revealed that the rate of the reaction increased as substrate was depleted. This indicates that the inhibition was not due to irreversible enzyme inactivation or product inhibition. Accordingly, the steady-state kinetic data were analyzed using the substrate inhibition equation. The combination of low apparent Ks and high substrate inhibition made the accurate determination of initial reaction rates at low substrate concentrations technically challenging. This was especially true for ADD, to the extent that it was not feasible to fit the equation to the data (Figure 3.7C).   3 4 50.000.020.040.060.08TIme (min)Rate (?M/s)0.0900.0950.100 NADH (mM) 4 6 8 100.0360.0380.0400.0420.044Time ( in)Rate (?M/s)0.0800.0850.0900.0950.100 NADH (mM) Figure 3.6   Progress curves of KshA5 with initial concentrations of 8.3 ?M (A) and 50 ?M (B) ADD. The rate of NADH consumption calculated over 10 s is displayed in black and the amount of NADH is shown in blue. Assays were performed using potassium phosphate (I = 0.1 M, pH 7.0) containing 100 ?M NADH at 22 ?C. A B Time (min) Time (min)      Rate (?M/s) Rate (?M/s) NADH (mM) NADH (mM)      Rate (?M/s) 34  0 50 1000.000.050.10Rate (?M/sec)5a-Androstan-3,17-dione (?M)0 2 4 6 8. 00.020.040.060.080.10Rate (?M/sec)4-Estren-3,17-dione (?M)0 50 1000.000.02.040.060.080.10Rate (?M/sec)ADD (?M)0 5 10 15 200.000.040.060.08Rate (?M/sec)1,4-BNC (?M)0 200 400 600. 00.050.10.15Rate (?M/sec)BNC-CoA (?M) Figure 3.7   Steady state kinetic curves of KshA5. Shown is the dependence of initial velocity of NADH consumption on 5?-androstan-3,17-dione (A), 4-estren-3,17-dione (B), ADD (C), 4-BNC (D), 1,4-BNC (E), and 1,4-BNC-CoA (F). Assays were performed using potassium phosphate (I = 0.1 M, pH 7.0) containing 100 ?M NADH at 22 ?C.  Curves represent the best fits of the Michaelis-Menten (A, B, F) or substrate inhibition (D, E) equations using LEONORA. 0 50 1 0. 0.02.04.06.080.10Rate (?M/sec)4-BNC (?M)A B C D E F Rate (?M/s) Rate (?M/s) Rate (?M/s) Rate (?M/s) Rate (?M/s) Rate (?M/s) ADD (?M) 4-Estren-3,17-dione (?M) 4-BNC (?M) 1,4-BNC (?M) 5?-androstan-3,17-dione (?M) BNC-CoA (?M) 35  Remarkably, despite the wide spectrum in steady-state kinetic behavior (Figure 3.7), nearly all the tested substrates were turned over with kcat values within the narrow range of 0.5 to 0.8 s-1 (Table 3.3) Furthermore, low apparent Km and Ks values and universally high specificities were observed for these substrates, as opposed to the wide variance seen in KshA1 activity. Additionally, KshA5 does not show a strong preference between short alcohol or oxo C-17 sidechains and longer isopropionyl sidechains. Finally, in contrast to KshA1, the addition of the CoA thioester sidechain provides a drastic difference in activity. Although 1,4-BNC-CoA provided a kcat double that of any other substrate, it exhibited the lowest specificity by a wide margin, providing an apparent Km value two orders of magnitude greater than the next highest observed.  3.4.3 Oxygen dependency of KshA1 and KshA5 The ability of the two rhodococcal KshAs to utilize O2 was investigated in the presence of saturating concentrations of their best steroid substrates: 50 ?M 4-BNC and 5 ?M 4-estren-3,17-dione in the cases of KshA1 and KshA5, respectively. As shown in Figure 3.8, KshA1 displayed a significantly better ability than KshA5 to utilize O2, with an apparent kcat/Km value almost two orders of magnitude higher.  This difference was also reflected in the respective apparent Km values.  Figure 3.8   The steady-state utilization of O2 by KshA. Kinetics were performed using 38 nM KshA1 and 50 ?M 4-BNC (A) or 122 nM KshA5 and 5 ?M 4-estren-3,17-dione (B). Assays were performed in potassium phosphate (I = 0.1 M, pH 7.0) at 22 ?C and 100 ?M NADH. Curves represent the best fit of the Michaelis-Menten equation using LEONORA. Oxygen (?M) Oxygen (?M) Rate (?M/s) Rate (?M/s) 36  3.5 X-ray crystallography analyses To understand the molecular basis of KshA function and substrate specificity, the enzymes? structures, alone and in the presence of selected steroid substrates, were studied using X-ray crystallography. These studies yielded a total of five structures: KshA5, KshA5:ADD, KshA5:1,4-BNC-CoA, KshAMtb:ADD, and KshA1:1,4-BNC-CoA.  3.5.1 Crystals Initial screening of all three KshA?s with and without substrate was performed using the Wizard sparse matrix screen. Small brown crystals of various shapes appeared over one to two weeks but diffracted poorly, even when the crystallization conditions were optimized. After 3 months, better diffracting crystals grew from a well solution of 1.6 M NaH2PO4, 1.2 M K2HPO4, 100 mM phosphate-citrate, pH 4.2. Although many other crystallization conditions were attempted, all of the best diffracting KshA crystals used for final structure determination were obtained through the use of additive screens to the above condition and optimization thereof. 3.5.1.1 KshAMtb in complex with ADD The first round of screening provided a roughly spherical brown crystal of KshAMtb which grew in the presence of ADD and diffracted to 3.5 ? at a home X-ray source (CuKa radiation ? = 1.542 ?). While optimization of these conditions didn?t provide diffraction better than 3 ?, additive screening combining NaH2PO4, K2HPO4, and phosphate-citrate with solutions of the Wizard screen at a ratio of 80:20 provided well diffracting crystals. A curved rectangular brown crystal of KshAMtb in complex with ADD grew over 2 weeks from a 1 ?L drop obtained by mixing 0.5 ?L 300 ?M KshAMtb in 25 mM HEPES, pH 7.5, 0.25 mM FAS, 1 mM DTT and 300 ?M ADD with 0.5 ?L well solution (0.5 M NaH2PO4, 0.125 M K2HPO4, 4% PEG-1000, 20 mM tris, 100 mM phosphate-citrate, pH 4.2). This crystal (Figure 3.9A) diffracted to 2.5 ? and belonged to space group P321. Molecular replacement using KshAMtb (PDBID 2ZYL) revealed one monomer in the asymmetric unit and trimeric arrangement of symmetry partners in the same fashion as other KshA crystals. 37  3.5.1.2 KshA1 and KshA5 hybrid crystal Attempts to crystallize each of KshA1 and KshA5 on their own and in the absence of substrate were unsuccessful. Unexpectedly, crystals were obtained from a mixture of KshA1 and KshA5. Pyramid-shaped light brown crystals grew over 2 weeks from a 1.5 ?L drop containing equal volumes of KshA1, KshA5 and well solutions. The final composition of the drop was 100 ?M KshA5, 100 ?M KshA1, 100 ?M 1,4-BNC-CoA in 17 mM HEPES, pH 7.5, 0.17 mM FAS, 0.67 mM TCEP, 0.17 M NaH2PO4, 0.01 M K2HPO4, 1.3% PEG-1000, 6.7 mM Tris, 33 mM phosphate-citrate, pH 4.2 (Figure 3.9B). The crystals diffracted to 2.5 ? and belonged to space group F23. Molecular replacement using KshAMtb (PDBID 2ZYL) revealed two monomers within the asymmetric unit: one of KshA5 without substrate and one of KshA1 in complex with 1,4-BNC-CoA. Crystal symmetry indicates that each monomer forms a homotrimer typical of ROs. 3.5.1.3 KshA5 in complex with ADD Further additive screening combining of the conditions in 3.5.1 with solutions of the Wizard screen at a ratio of 80:20 provided small columnar crystals. Optimization of this condition provided a rod-shaped dark brown crystal of KshA5 in complex with ADD, which grew over two weeks from a drop obtained by mixing 0.5 ?L 300 ?M KshA5 in 25 mM HEPES, pH 7.5, 0.25 mM FAS, 1 mM DTT and 300 ?M ADD with 0.5 ?L of well solution (0.8 M NaH2PO4, 0.2 M K2HPO4, 2% PEG-3000, 20 mM CHES, 100 mM phosphate-citrate, pH 4.2) (Figure 3.9C). The crystal diffracted to 2.1 ? and belonged to space group P63. Molecular replacement using KshAMtb (PDBID 2ZYL) revealed one monomer in the asymmetric unit and a trimeric arrangement of symmetry partners as expected in ROs. 3.5.1.4 KshA5 in complex with 1,4-BNC-CoA Further additive screening as described above provided a similar rod-shaped crystal of KshA5 in complex with 1,4-BNC-CoA from a drop obtained by mixing 0.5 ?L 300 ?M KshA5 in 25 mM HEPES, pH 7.5, 0.25 mM FAS, 1 mM TCEP and 300 ?M 1,4-BNC-CoA with 0.5 ?L 38  of well solution (0.5 M NaH2PO4, 0.125 M K2HPO4, 4% PEG-1000, 20 mM tris, 100 mM phosphate-citrate, pH 4.2). Crystals diffracted to 2.6 ? and belonged to space group P63. This crystal also exhibited a single copy of KshA and trimeric arrangement of symmetry partners.       Figure 3.9   Examples of KshA crystals. Crystals of KshAMtb:ADD ~0.4 mm long (A), KshA5 and KshA1 ~0.2 mm (B), KshA5:ADD ~0.5 mm long (C) obtained as described in the text. 3.5.2 Analysis of KshA structures  Phases were solved using molecular replacement with KshAMtb (PDBID: 2ZYL) for all structures except KshA5:1,4-BNC-CoA, which was solved using KshA5:ADD as a molecular replacement model. All structures were refined iteratively using Refmac5 by shaping the backbone, sidechains, obvious waters, and iron and sulfur atoms. Other solvent and substrate were placed into areas of clear density. Data collection and refinement statistics are listed in APPENDIX A. Table 3.4 lists relevant ligand- and substrate-mononuclear iron distances along with C? atom RMSD from PyMOL structural superpositioning. *Structure from (19). Superpositioning with 1KshAMtb and 2KshA5. Table 3.4   Ligand-iron (Fe)-substrate  distances and structural alignment Enzyme Substrate Fe-C9 (?) HisA (?) HisB (?) Asp distal (?) Asp prox. (?) H2O (?) RMSD (?) Aligned  C??s Mtb* -  2.2 2.1 2.5 2.1 2.1   Mtb ADD 3.9 2.4 2.6 2.5 2.4 2.3 0.209 3231 KshA5 -  1.9 2 2.4 2.2 1.6 0.452 2831 KshA5 ADD 6.4 2.2 2.4 2.8 2.3 2.4 0.863 2632 KshA5 1,4-BNC-CoA 6.1 2.3 2.2 2.5 2.7 1.9 0.799 2692 KshA1 1,4-BNC-CoA 4.3 2.2 2.2 2.6 2.2 1.8 0.490 2921 A. B. C. 39  3.5.3 Structure of KshAMtb:ADD complex  Similar to the previously solved structure of KshAMtb (19), residues 14-375, the Rieske cluster, mononuclear iron, and 132 waters were apparent from the observed electron density. All major structural features are retained in the substrate-bound structure. An obvious area of density in the active site resembling a steroid was modeled well with ADD at 100% occupancy. The substrate was positioned such that C-9 of the steroid was closest to the mononuclear iron at a distance of 3.8 ? and the C-17 carbonyl was orientated towards the solvent (Figure 3.10). This positioning is in agreement with a previous docking experiment (19), identifying primarily residues of the ?-sheet in contact with the substrate. Fully occupied iron provided a B-factor of 70 whereas surrounding residues and substrate were in the range of 30-45, thus mononuclear iron occupancy was refined at 50%, providing a B-factor of 38. A small density adjacent to the iron was modeled well as a fully occupied water 2.3 ? away and clearly that of a single atom.    Figure 3.10   Electron density in the active site of KshAMtb:ADD. Fo-Fc mesh is contoured at 2? for ADD, the mononuclear iron and its ligands. Iron and water are orange and red spheres, respectively, carbon is gray, nitrogen is blue, and oxygen is red. Figure generated using PyMol (20).  3.8 ? ADD His181 Asp304 His186 40  Residues 215-222 and 283-284 exhibited higher B-factors than observed in the substrate-free structure. ADD bound and free structures superimposed with PyMOL provide an RMSD of 0.209 ? over 323 C? atoms, with residues of the Rieske center and active site pocket positioned nearly identically in the two structures. Indeed, the structural superpositioning provided a maximum distance of 0.2 ? between C??s and 0.7 ? between sidechain atoms in the active site whereas most of the latter were within 0.5 ?.  The only significant difference in the backbone positioning of the ADD-bound and substrate-free enzymes was observed in regions that were disordered. Notably, the mouth loop residues between 216 and 221 shifted 2-3 ? away from the transverse helix in the substrate-bound structure (Figure 3.11) with respect to how they were modeled in the substrate-free structure. Nevertheless, both structures exhibited large B-factors in this loop, indicating that residue side chains cannot be positioned reliably in this region although placement of the backbone is possible. Outside of disordered regions, the only residues which altered position were at the subunit interface and near the entrance of the active site pocket. The ?bridging? Asp178 rotates and moves slightly towards each subunit?s iron-coordinating His. At the active site entrance some residues exhibit a slight shift in the substrate-bound form, but electron density is weak in this region for both structures indicating that the difference is likely insignificant.  Figure 3.11   Superposition of KshAMtb:ADD (cyan) and KshAMtb (gray) . ADD is shown in green, oxygen is red, nitrogen is blue, and iron is orange. Minimal changes are seen upon ADD binding in the KshA backbone. Figure generated using PyMol (20) and PDBID 2ZYL (19). 41  No significant changes were observed in the Rieske domain, which retained backbone and residue positioning in the ADD bound structure. The coordination of the mononuclear iron remains the same by the His181, His186 and bidentate Asp304 ligands.  However, upon binding ADD, the coordinated water shifts ~1.5 ? away from the substrate-binding pocket. Additionally, it moves farther from the iron, presumably to allow substrate positioning and open coordination space for O2.  3.5.4 Structure of KshA1 and KshA5 co-crystallization  Crystals grown in a mixture of KshA1, KshA5, and 1,4-BNC-CoA had two monomers in the asymmetric unit. The electron density observed clearly resembles that of the sequences of KshA1 for one monomer and KshA5 for the second monomer. Both were modeled using the structure of KshAMtb for molecular replacement. All amino acid side-chains were removed before further refinement and rebuilt into the density resembling the correct amino acids of each homologue. Inspection of the crystallographic symmetry revealed that each homologue is arranged in the ?3 structure typical of ROs to provide independent KshA1 and KshA5 trimers.   Figure 3.12   KshA1:1,4-BNC-CoA (gray) and KshA5 (cyan) hybrid crystal packing arrangement. Iron atoms are shown as orange spheres. Shown at center is the closest crystal contact between KshA1 and KshA5. Figure generated using PyMol (20).  42  In the crystalline lattice, the two homologues interact at the C-terminal helix with a closest contact of 3 ? between Gln373 of KshA5 and Thr360 of KshA1 (Figure 3.12). All other contacts are more than 3.5 ?. Superposition of KshAMtb with either protomer revealed similar positioning of the backbones and many residues at the interface between KshA1 and KshA5, indicating that there is little structural affect from crystal packing. The two structures are described independently in the following sections. 3.5.5 Structure of KshA1:1,4-BNC-CoA complex The structure of KshA1:1,4-BNC-CoA was refined to 2.5 ?. The refined model comprises residues 21-385, the Rieske cluster, mononuclear iron, and 149 waters. All major structural features of KshAMtb are conserved in KshA1. These include the structural fold, the metal coordination spheres, the 9 ?-helices and 19 ? strands. Active site electron density resembling a steroid nucleus with a C-17 extension of ~5 atoms out of the active site was modeled using 1,4-BNC-CoA with most of the CoA deleted, leaving a 6 atom side chain (Figure 3.13). Resulting B-factors were all in the range of 50-90. Lowering the substrate occupancy to 0.75 provided B-factors in the range of 30-50, similar to the nearby catalytic iron and its ligands. The lack of obvious electron density for the remaining portion of the CoA is likely due to its flexibility. However, it is also possible that the CoA thioester degraded during incubation of crystals. Superpositioning of KshA1:1,4-BNC-CoA to KshAMtb using PyMOL yielded an RMSD of 0.49 ? for 292 C? atoms and to KshAMtb:ADD provides an RMSD of 0.47 ? for 291 C? atoms showing that the overall folds of the proteins are very similar.   The substrate-bound active site more closely resembles that of KshAMtb than KshA5 with an iron-C9 distance of 4.3 ?. The first shell of residues surrounding the steroid nucleus are positioned as in the KshAMtb:ADD active site. Notably, these residues are conserved except for Asn240, which is Asp in KshA1 (3 ? from substrate C-8) and KshA5 and Val176, which is Ile in KshA1 (3.7 ? from substrate C-3 oxo). The main differences between the KshA1:1,4-BNC-CoA and KshAMtb:ADD structures occur at the active site opening where there are 4 non-conserved residues: Asn208, Leu226, Asn244 and Gly296 of KshAMtb are Gly209, Ser228, Tyr246 and 43  Phe298, respectively, in KshA1. In the latter, the active site channel is short and directed to the surface halfway along and perpendicular to the transverse helix. By contrast, the channel of KshAMtb runs the length of the transverse helix. Additional differences in amino acid positioning and identity are observed farther out the active site channel and within the mouth loop, which is directed more towards the active site in KshA1, although it is still fairly disordered. As would be expected with the larger substrate bound, the loop does not block the active site entrance of KshA1:1,4-BNC-CoA as in KshA5 or KshA5:ADD. Although the active site is more occluded from solvent in KshA1 than in KshAMtb.   Figure 3.13   Substrate binding in KshA1:1,4-BNC-CoA. The Fo-Fc mesh is contoured at 2? for the substrate, mononuclear iron, and its ligands. The ribbon representation of the KshA5 backbone and stick representation of the 1,4-BNC-CoA fragment are in gray with the mononuclear iron and water as orange and red spheres, respectively. Nitrogen, oxygen, and sulfur colored blue, yellow and red, respectively. Figure generated using PyMol (20). 3.5.6 Structure of KshA5 without substrate The structure of substrate-free KshA5 was refined to 2.5 ?. The refined model comprises residues 26-385, the Rieske cluster, mononuclear iron, and 265 waters. All major structural 4.3 ? 1,4-BNC-CoA His182 His187 Asp306 44  features of KshAMtb are conserved in KshA5. These include the structural fold, the metal coordination spheres, the 9 ?-helices and 19 ?-strands. A structural superpositioning in PyMOL between the substrate-free enzymes KshA5 and KshAMtb yielded an RMSD of 0.45 ? for 283 C? atoms showing that the structural folds of these two enzymes are very similar. Indeed, the substrate-binding residues within the active site are essentially identically positioned (Figure 3.14). While most residues thought to contact the substrate are conserved, those that are not retain a similar orientation with a net result of very little difference in substrate binding between the two enzymes. The active site iron is fully occupied as evidenced by similar B-factors in surrounding atoms (~20-40). Iron coordination is similar to that observed in KshAMtb with distorted tetragonal geometry by His188, His193, Asp312 (bidentate) and one water. Sparse electron density was observed in the active site pocket and attempts were made to model 1,4-BNC-CoA or other substrates into the active site, but the density was far too small for even partial occupancy by substrate or large solvent species. Thus, the electron density within the substrate pocket was modeled as 4 water molecules. While nearly all of the backbone is    Figure 3.14   Superposition of KshAMtb (gray) and KshA5 (cyan). Shown is the active site from inside the enzyme looking out the active site entrance. Iron is colored orange, oxygen is red, sulfur is yellow and nitrogen is blue. The substrate-binding site is at the center of the image. Figure generated using PyMol (20) and PDBID 2ZYL (19). Numbering is that of KshA5/(KshAMtb). Ala238 Phe189 Ser236 Gln211/204 Leu234 Leu250 Met246 Leu263 Asn265 Phe309 Trp316 His188 Asp312 His193 Met213/Leu206 Asp248/Asn240 Ser215/Asn208   Tyr227 45    Figure 3.15   Superposition of KshAMtb (gray) and KshA5 (cyan) . Shown are the mouth loops with that of KshA5 within the active site channel with Tyr227 labeled in the active site. View is facing that of Figure 3.14. Figure generated using PyMol (20) and PDBID 2ZYL (19). Numbering is that of KshA5. positioned similarly between the two enzymes throughout both the Rieske and catalytic domains, the mouth loop region represented by residues 217 to 233 in the KshA5 structure differs significantly. While the loop in KshAMtb (residues 210-224) is somewhat disordered and orientated away from the active site opening, the KshA5 loop (217-232) is well defined and plugs the substrate channel. Indeed, Tyr227 points into the substrate-binding pocket to completely occlude solvent access to the KshA5 active site (Figure 3.15). Additionally, crystallographic B-values representing the KshAMtb loop (C??s upwards of 100) are much higher than those in the corresponding KshA5 loop (C??s 50-70).  3.5.7 Structure of KshA5:ADD complex The structure of KshA5:ADD was refined to 2.1 ?. The refined model comprises residues 16-381, 373 waters, PEG, ethanol, sulfate, and glycerol. Density observed in the active site adjacent to the mononuclear iron is clearly that of a steroid substrate, providing an excellent Asp312 His193 Tyr227 His188 46  fit for ADD with B-factors around 20 (Figure 3.16). While the majority of the KshA5:ADD structure is oriented identically to that of the substrate free enzyme, a large change in the helices harboring the iron-coordinating residues and the C-terminal helix is observed with ADD bound. Each helix shifts between 1 and 4 ? as measured from C? atoms within the stationary ?-sheet to C? of residues in the helices. In addition, the distance between the iron and C9, the hydroxylation site of the substrate, was 6.3 ? (Figure 3.17). PyMOL superpositioning between bound and unbound KshA5 (RMSD of 0.86 ? for 263 C? atoms) showed that substrate-binding residues of the ?-strands are nearly identically positioned as well as much of the backbone. The loop region in this structure is positioned slightly farther from the active site but still mostly occludes the channel, leaving a small opening of ~1 ? diameter. Similar to KshA5 without substrate, small B-values are observed in the range of 30-40 for C??s of this loop. The Rieske domain displays a small shift in its orientation, compensating for movement across the subunit interface and maintaining its interactions.    Figure 3.16   The substrate-binding site of KshA5:ADD. The Fo-Fc mesh is contoured at 2? for the mononuclear iron, its ligands and ADD. Carbons are gray, iron and water are orange and red spheres, respectively, oxygen is red and nitrogen is blue. Figure generated using PyMol (20). ADD Phe189 His188 Asp312 His193 6.4 ? 47     Figure 3.17   Superposition of KshA5:ADD (cyan) and KshA5 (gray). Shown are the iron and its ligands, the bridging Asp185, residues of the ?-sheet, and Tyr227 that blocks the active site channel in the substrate-free structure. ADD is cyan, oxygen is red, nitrogen is blue, and iron is orange. Arrows indicate movement upon binding ADD. Figure generated using PyMol (20). 3.5.8 Structure of KshA5:1,4-BNC-CoA complex The structure of KshA5:1,4-BNC-CoA was refined to 2.6 ?. The refined model comprises residues 15-385, 56 waters and one acetate molecule. All major structural features including ?-helices and ? strands are observed in the same orientation as KshA5:ADD. Electron density resembling a steroid was observed in the active site extending outwards. 1,4-BNC-CoA was fitted to the density by first placing the steroid nucleus in the same position as that of ADD in other structures and refining with different configurations of the C-17 CoA tail. The best structure observed exhibited a CoA side-chain curving out of the active site with a steroid nucleus fitting identically to that of ADD in the previous structure. Electron density is well defined for the steroid nucleus and the first few atoms of the C-17 sidechain before modeling with substrate. However, density is weaker for much of the CoA moiety as exemplified by high B-factors (50-90 mid-way and 90-100 at the end), indicative of flexibility or varying Asp185 Tyr227 His193 His188 Asp312 Gln211 Leu250 Ser236 ADD 48  orientations. A PyMOL structural superposition to KshA5 (RMSD 0.80 ? for 269 C? atoms) and KshA5:ADD (RMSD 0.53 ? for 318 C? atoms) shows a similar perturbation of the iron and helices containing its coordinating residues as observed with ADD bound. Indeed, the iron to C-9 distance of 5.9 ? is also high in this structure. Most notably is the positioning of the mouth loop (Figure 3.18): KshA5:1,4-BNC-CoA exhibits a loop oriented away from the active site channel and poorly defined between residues 223 and 228 (C? B-factors up to 100), providing a configuration similar to that observed in KshAMtb with a spacious substrate access channel. The Rieske domain shifts slightly as in KshA5:ADD, likely compensating for the shift in the catalytic center.    Figure 3.18   Active site of KshA5:1,4-BNC-CoA. Shown is Fo-Fc mesh contoured at 1? for 1,4-BNC-CoA, mononuclear iron, and its ligands. Carbons are colored gray with iron in orange, nitrogen in blue, oxygen in red, and sulfur in yellow. Figure generated using PyMol (20).  6.1 ? 1,4-BNC-CoA His188 Asp312 His193 49  4. Discussion  The steady-state kinetic studies of KshA1 and KshA5 corroborate and extend previous studies of the substrate preferences of these enzymes (47), the growth phenotypes of gene deletion strains, and phylogenetic analysis of KshAs (42). The structural characterization of KshA1 and KshA5 provides insight into the structural basis for their specificities and the conformational flexibility of ROs. The observed steroid specificity and utilization of O2 by KshA helps to identify roles these enzymes play in steroid catabolism, further explaining their apparent redundancy in rhodococci. The structural changes observed upon substrate binding, particularly in the catalytic iron site and the mouth loop?s closure, imply a previously unrecognized conformational flexibility in ROs, which have broader implications in the mechanism of substrate selection and catalysis. Overall, the insight gained in this study increases our understanding of ROs in terms of their catalytic abilities and structural diversity. 4.1 Correlation of KshA specificities to previous investigations   While this work details a substantial study into the structure of KshAs and their specificities for 3-ketosteroids, none of the data are truly representative of the physiological conditions under which the enzymes are utilized by bacteria. More particularly, the metabolic state of DSM43261 growing in its natural environment is difficult to duplicate in a laboratory setting. Thus mutational studies, while important for determining the use of particular genes, may have unforeseen effects such as the accumulation of toxic metabolites. Furthermore, the isolation and study of an individual enzyme, while indicative of its properties, does not reproduce the conditions under which it operates within the cell. Indeed, varying concentrations of reductase, ferredoxin, NADH, O2, or other molecules will necessarily have an effect on the enzyme?s catalysis (64). In the current studies, attempts were made to narrow the investigation of KshA down to substrate concentration as a single variable. Finally, while single products were observed from catalysis with KshA1 and KshA5 and are likely 9?-hydroxylation products based on GC/MS, only the product of the 9?-hydroxylation of ADD was confirmed by comparison to 50  an HSA (product) standard. Accordingly, it is not possible to definitively rule out that other substrates react differently. 4.1.1 KshA1 specificity and physiological implications KshA1?s higher apparent specificity (kcat/Km) for substrates with a side chain at C-17 is consistent with the enzyme?s previously determined preference for 4-BNC and 1,4-BNC as well as its predicted role in bile acid catabolism (47). Thus, transcriptional analysis of DSM43269 growing on different steroids revealed that KshA1 was only expressed during growth on cholic acid (47). Moreover, KshA1 restored growth of RG32, a kshA null mutant of DSM43269, on cholic acid. Finally, in a phylogenetic analysis, KshA1 clustered with the ortholog of RHA1, KshA3RHA1 (Figure 4.1), that is involved in cholate catabolism (42).    Figure 4.1   Phylogenetic tree of KshAs (47).  51  Interestingly, KshA1 does not show the same increased apparent specificity for a CoA thioester as KshAMtb despite the fact that cholesterol catabolism in Mtb (38) and RHA1 and cholate catabolism in RHA1 (65) all involve concurrent side chain and rings A/B degradation. The concurrent degradation in Mtb is supported by KshAMtb?s apparent specificity (kcat/Km) for CoA thioester substrates, which was 20?30 times that for the corresponding 17-keto compounds. Moreover, 13C-metabolite profiling in cholesterol-grown Mtb lacking the igr operon identified a two-ringed metabolite with a partially degraded side chain in which rings A/B had been degraded (66). Similarly, RHA1 mutants lacking genes responsible for ring A degradation accumulate cholesterol metabolites containing C-17 side chains of various lengths including isopropionyl and likely a 5 carbon side chain (A. Crowe, I. Casabon, L. Eltis, unpublished). During growth on cholate, RHA1 excretes two metabolites, THSBNC and HHIDP, which contain partially degraded C-17 side chains, before reassimilating and completely degrading them. Considering that concurrent side chain and rings A/B degradation occurs in these three instances, it is likely that it also occurs in cholate catabolism by DSM43269. KshA1?s specificities for 4-BNC and 1,4-BNC  are consistent with concurrent side chain and rings A/B degradation. They further suggest that not all intracellular side chain degradation catabolites are CoA thioesters and thus that cholate and cholesterol catabolism are organized differently, perhaps due to the cellular toxicity of cholate and/or its metabolites. The significance of KshA1?s low apparent KmO2 is not clear. Although this value is over an  order of magnitude lower than that of either KshA5 or KshAMtb (apparent KmO2 of 90 ?M in the presence of 1,4-BNC-CoA (38)), it is comparable to that of toluate dioxygenase measured in the presence of p-toluate (8.4 ?M (67)). The low apparent KmO2 would allow KshA1 to turnover substrate at significant rates when oxygen concentration is low, as might occur in the soil environments where rhodococci are found. Nevertheless, the different ability of the two paralogs to utilize O2 is striking and may be related to the different physiological roles of these enzymes.  4.1.2 KshA5 specificity and inhibition  KshA5 turned over steroids at remarkably similar rates, as implied in previous studies in which activities were measured at a single substrate concentration (47). However, the more in-depth analyses performed here revealed that the enzyme has a high specificity constant for some 52  steroids and is strongly inhibited by others. Nevertheless, the steady-state kinetic parameters only appear to rule out 1,4-BNC-CoA as a physiological substrate. Even ADD, for which a kcat/Km value could not be determined due to substrate inhibition, is turned over efficiently at very low concentrations. While concentrations of reaction components have a large effect on apparent specificities, the apparent kcat/Km of ROs for their physiological substrate ranges from 0.036 x 106 M-1s-1 (aminopyrrolnitrin oxygenase for aminopyrrolnitrin) to 2.4 x 106 M-1s-1 (BPDOLB400 for biphenyl) (38,54,64,67-70), spanning the range of the values observed for KshA5. Indeed, most of the latter values exceed that of KshAMtb for its best substrate, 1,4-BNC-CoA (0.16 x 106 M-1s-1 (38)). Overall, the relatively high apparent specificity of KshA5 for a range of substrates is consistent with its proposed role in degrading a range of steroids. Thus, kshA5 was expressed in DSM43269 during growth on each of AD, progesterone, cholic acid, and cholesterol. Furthermore, RG31, a mutant of DSM43269 containing only KshA5, grew on each of these four steroids (47). Finally, phylogenetic analyses revealed that KshA5 clusters with KshA2RHA1 and KshA4RHA1, neither of whose physiological roles has been determined. KshA5?s ability to support growth on cholesterol is somewhat unexpected given the enzyme?s low specificity for 1,4-BNC-CoA. It is possible that in the cholesterol catabolic pathway that concurrent ring and side chain degradation is more efficient but not obligate, and that KshA5 acts on a side chain-degraded intermediate. Nevertheless, the precise physiological role of KshA5 remains unclear. The significance of the substrate inhibition, which is extremely strong in the case of ADD and somewhat weaker in case of 4-BNC, 1,4-BNC and testosterone, is unclear. Substrate inhibition based on two-site binding fit all but ADD well, with data indicating that the Ks and Kss would be very low. All share a similar steroid core with methyl groups at C-10 and 13 along with ?4 or ?1,4 de-saturations, but have different C-17 side chains. Although 1,4-BNC-CoA has the same steroid core as ADD and 1,4-BNC, it shows no inhibitory effect at the concentrations measured. Thus, the addition of a bulky side chain at C-17 likely prevents binding at a secondary site. The inhibition data provide further evidence of remarkably similar low activities for all short side chained substrates. KshA5 was previously observed to show lower activity for ADD, 4-BNC, and 1,4-BNC (47), but substrate inhibition indicates that the amount of substrate used in the assay (200 ?M) did not reflect the maximum activity. Although inhibitory concentrations of substrate are unlikely to occur in a natural setting, it is possible that substrate inhibition plays a role in limiting the accumulation of toxic intermediates.  53  4.2 KshA structures  The structures solved with and without substrate provide examples of KshA in several different states. Previous studies have shown that high-powered X-ray sources reduce the iron-sulfur cluster of Rieske enzymes (65), making it likely that the structures presented here have reduced Rieske clusters. Additionally, the data could represent a mixture of states observed between KshA monomers within the trimeric unit, especially for those with lower occupancy atoms which may exist alternately or in only one or two protomers. The coordination of the mononuclear iron provides information as to the enzyme?s state as well. While some structures of ROs appear to have O2 bound to the iron either side-on or end-on, all the structures of KshA appear to have a single water molecule coordinating the iron. Furthermore, the presence of substrate affects the catalytic iron and indicates the enzyme is poised for binding and activation of O2. Finally, well conditions varied between each crystal, though all contained phosphate-citrate buffer, pH 4-5. Thus, the pH at which the KshA enzymes are crystallized may have an effect on the structures observed. For example, some residues may be protonated at lower pH, which may create or disrupt hydrogen bonds, potentially affecting key structural features. Indeed, the state of the Rieske cluster and bridging Asp is likely pH-dependent. While the redox state of the mononuclear iron is unknown, the presence of substrate and its influence on the structure is one key difference that these structures demonstrate. 4.2.1 KshA structural similarity The structures of KshA1, KshA5 and KshAMtb are very similar throughout the catalytically important areas responsible for binding substrate and coordinating the mononuclear iron and Rieske cluster. For the most part, the less conserved residues are located within disordered regions and/or less influential segments of the enzyme remote from metal centers or subunit interfaces. In key areas of the enzyme, most residues are conserved or are similar amino acids. The ?-sheet of the active site stands out most providing a highly similar substrate binding environment (Figure 3.12), while the Rieske and mononuclear iron coordination remains identical. A few areas with high structural divergence correspond to regions with low amino acid sequence similarity including the mouth loop (KshAMtb residues 210-224) and ?-turn at the 54  opening of the active site (KshAMtb residues 244-251). As the most substantial differences are located in these regions, they likely have significance in steroid recognition and transformation.  4.2.2 KshAMtb in complex with ADD  Comparison of the structure of KshAMtb:ADD with that of the previously solved substrate-free structure of KshAMtb (19) provides insight into how KshAs accommodate their substrate. The large active site pocket requires only a slight enlargement to bind ADD, providing a fit very similar to substrate docking experiments (19). Although little difference between structures is observed adjacent to the substrate, a 120? rotation of the ?bridging? Asp178 around its C?-C? bond provides a slightly closer contact to His89 coordinating the Rieske cluster (Figure 4.2). While in the substrate-free structure, distances are 2.8 ? from His181-N? to the backbone carbonyl of Asp178 and 2.9 ? from the Asp178-?O2 to the His89-N?, they change to 2.9 ? and 2.7 ? in the ADD-bound structure. Indeed, the reduction state of the Rieske cluster is thought to modulate the catalytic iron through the subunit interface, opening a coordination site for O2 to bind (21). However, it is not clear what causes this altered conformation of Asp178. While the binding of substrate may be responsible, the low pH of the ADD bound crystal may have an effect on the H-bonding at this location or nearby. Indeed, the substrate-free structure of KshAMtb was crystallized at neutral pH likely causing different H-bonding behavior.     Figure 4.2   Superposition of KshAMtb and KshAMtb:ADD. Shown are the mononuclear iron at left and its ligands, the bridging Asp178, and the Rieske center at right and its His ligands. The substrate-free structure is in gray and the ADD-bound structure is in cyan with iron colored orange and sulfur, yellow. Figure generated using PyMol (20) and PDBID 2ZYL (19) . Asp178 His181 His89 His69 Asp304 His186 2.9 ? 2.7 ? 55   Nevertheless, ADD is a poor substrate for KshAMtb (38). Thus, it is possible that the physiological substrate may cause different structural behavior upon binding. KshAMtb exhibits up to 30x greater specificity towards substrates with longer side chains such as 1,4-BNC-CoA (38). The large CoA tail would likely interact with residues at the entrance to the active site, conferring specificity and potentially modulating the enzyme?s activity. 4.2.3 KshA1 in complex with 1,4-BNC-CoA  Comparison of the KshA1:1,4-BNC-CoA and KshAMtb:ADD structures reveal a possible structural basis for the different substrate specificities of these enzymes. While the overall structures are quite similar including the backbones and orientation and the catalytic center, the structures differ most significantly in residue identity and loop positioning at the opening of the active site. Indeed, the long channel observed along the transverse helix in KshAMtb is constricted in KshA1 and directs the C-17 side chain immediately towards the enzyme surface (Figure 4.3A vs. B). Specifically, Asn244 and Gly296 in KshAMtb are Tyr246 and Phe298 in KshA1and block the channel in KshAMtb along the side of the transverse helix that likely accommodates the CoA side chain. Furthermore, the hydroxyl of Tyr246 is within 2.7 ? of the 1,4-BNC-CoA?s C-22 carbonyl, which is also present in 4-BNC and 1,4-BNC and likely serves a role in substrate selectivity. Indeed, KshA1?s apparent Kcat/Km are much greater than KshAMtb is for 4-BNC and 1,4-BNC while of a comparable magnitude for ADD.  4.2.4 Similarity to other Rieske oxygenases While KshA shares less than 15% amino acid sequence identity with other ROs (19), the architecture of the mononuclear iron, the subunit interface, and the Rieske domain are all very similarly arranged in these enzymes. Indeed, the distorted octahedral coordination geometry of the mononuclear iron observed in other ROs is conserved as well as ligation distances to the iron and Rieske cluster, with similar values in KshA as observed in other ROs. The ligand-iron distances in Table 3.4 are similar to those observed in other ROs. Distances between the mononuclear iron and both His-N? in most structurally characterized ROs are between 1.9 and 2.1 ?. Substrate-free KshAMtb and KshA5 and KshA1:1,4-BNC-CoA are within this range, 56  however the substrate-bound forms of KshAMtb and KshA5 exhibit longer distances. The significance of this increase is unknown. The distances from Asp-O?1 and O?2 to the mononuclear iron exhibit a wider range in structurally characterized ROs, between 1.9 and 3 ? with some structures providing a difference of 1 ? between distances of the Asp-O?1 and O?2 to the mononuclear iron. Although some correlation in the distances in the Rieske cluster have been cited to reflect the reduction state of the cluster, analysis of all structurally characterized ROs, including those with known oxidation states, does not show a strong correlation. Nevertheless, the distances and angles between iron, sulfur, and the coordinating residues observed in KshAs are consistent with those in the PDB. Finally, distances generally vary from 4 to 5 ? between the mononuclear iron and the site of catalysis of the bound substrate which is similar to that observed in the structures of KshA1 (4.2 ?) and KshAMtb (3.9 ?). However, the distances observed in KshA5:ADD (6.4 ?) and KshA5:1,4-BNC-CoA (6.1 ?) are significantly greater than any other structurally characterized RO.  4.3 KshA5 structural flexibility  The structural variation seen in the mouth loop and the catalytic center of KshA5 is greater than observed in any other RO and has implications into its substrate inhibition, substrate specificity, and catalytic mechanism. KshA5 has a wide substrate opening, gated by a flexible loop section and contains less bulky residues than either KshA1 or, to a lesser degree, KshAMtb, which exhibit no substrate inhibition even with a highly similar overall fold (Figure 4.3). While no other ROs have been observed to exhibit substrate inhibition, they all contain significantly smaller active site openings than KshA. Other ROs do contain a loop that appears to be flexible based on its elevated B-factors. However, this loop is situated between the helix containing the His ligands to the mononuclear iron and the ?-sheet whereas the loop of KshA originates from the subsequent ?-strand, positioning the active site channel adjacent to the transverse helix. The difference in the loop positions of the KshA5 structures solved here (Figure 4.4) is much greater than seen in other ROs. Indeed, the KshA5 active site opening can accommodate a second substrate, except for 1,4-BNC-CoA for which no inhibition was observed. Although a binding site remote from the active site could potentially inhibit catalysis, the effect of the C-17 side chain on inhibition suggests that the secondary binding is within the active site entrance and 57  facilitated by the flexible mouth loop, perhaps blocking the release of product or otherwise inhibiting turnover. This cleft is the most accommodating pocket observed on the enzyme surface and high inhibition seen with ADD indicates that it is due to interference with the active site. Indeed, binding near the active site would likely interfere with the flexibility of KshA5.       Figure 4.3   Active site opening of KshA showing space available for secondary binding. Shown is the surface representation of KshA with the disordered mouth loop represented by backbone and sticks in black. Enzyme surface is gray and substrate carbons is green. (A) KshA1:1,4-BNC-CoA , (B) KshAMtb:ADD, (C) KshA5:1,4-BNC-CoA. Figure generated using PyMol (20)  Although there is sufficient space for 4-BNC and other inhibitory substrates in the active site channel of KshA5, no electron density for a second substrate molecule was observed in X-ray crystallographic structures including KshA5:ADD. However, substrate was nearly equal to enzyme concentration in the crystallization drop, providing approximately one substrate molecule for each KshA monomer. Additionally, the catalytically relevant binding site would likely out-compete the secondary sites based on Ks and Kss. While two-site binding is a common mode of substrate inhibition, other more subtle and potentially mechanism-based substrate inhibition may be responsible. 4.3.1 Gating of the active site  The substrate-dependent conformations in KshA5?s mouth loop clearly influence substrate access. The KshA5 structures represent substantial changes in conformation, indicating that several mouth loop arrangements exist (Figure 4.4). There is a closed state without substrate, an intermediate state with ADD, and an open state with 1,4-BNC-CoA similar to that observed in KshA1 and KshAMtb, which may allow the entry and exit of substrate and product. With the A B C 58  exception of the cobalt-bound structure of DMO (45), all other RO?s exhibit only slight backbone or residue repositioning in the opening and closing of the active site to solvent. During catalysis, the enzyme must prevent the unproductive reaction of activated oxygen with water by occluding the solvent from the active site. The closed state of KshA5 in substrate-free and ADD-bound forms may serve to contain the reactive species, shielding the active site from solvent. While open and closed states are observed in other structurally characterized ROs, none are of the magnitude observed in KshA5, indicating that dynamics of the mouth loop may play a role greater than the omission of solvent from the active site.     Figure 4.4   Mouth loop arrangements of KshA5. Shown is a PyMOL structural alignment of the active site opening of KshA5 (gray), KshA5:ADD (cyan), and KshA5:1,4-BNC-CoA (green) with the mononuclear iron in orange. Figure generated using PyMol (20). 4.3.2 Structural correlation of broad specificity in KshA5 KshA5?s low apparent Km?s and highly similar kcat?s for all substrates except 1,4-BNC-CoA indicate that a similar binding and reaction mechanism that does not function with a CoA-thioester is responsible for the specificities observed. Although the steroids with alcohol, oxo, or isopropionyl C-17 side chains differ in shape and polarity, KshA5 has high kcat/Km values for each in contrast to KshA1 and KshAMtb, both of which exhibit high variance in all kinetic parameters between steroids with short C-17 side chains. However, the nearly conserved first shell of residues surrounding the steroid nucleus and similarity of steroid positioning in 59  KshA1:1,4-BNC-CoA and KshAMtb:ADD to KshA5:ADD and KshA5:1,4-BNC-CoA indicate that the fit of the substrate to the active site is not a determinant of broad specificity in KshA5. Additionally, the lack of broad specificity in the KshA1 chimer composed of the KshA5 ?-sheet (17) indicates that the residues binding the substrate are not responsible for KshA5?s unique kinetics. Indeed, the flexibility of the mouth loop in KshA5 may play a role in the selection and efficient turnover of substrate. Molecular dynamics studies of CYP3A4, a P450 enzyme responsible for the metabolism of more than 50% of clinically used drugs, indicate that the malleability of a loop obscuring the active site from bulk solvent is responsible for its broad specificity (71). High-amplitude motion of the loop was found to dampen in the presence of inhibitor and/or substrate in the active site. Similarly, the universally low Km and Ks values for all but 1,4-BNC-CoA in the KshA5 active site indicate that the flexibility of the mouth loop provides a highly adaptable binding site for numerous steroid nuclei containing short C-17 side chains. Indeed, a flexible mouth loop and catalytic iron capable of adapting to different substrates may result in broad substrate specificity. As a consequence of the movement regulating substrate access, turnover would potentially be slower than without such a gating mechanism. Indeed, kcat values observed with all substrates but 1,4-BNC-CoA are nearly the same intermediate rate as compared to other KshA or ROs, suggesting the existence of a rate-limiting step in the mechanism of binding, turnover or release. In contrast, the higher kcat and lower specificity observed for 1,4-BNC-CoA may be due to KshA5?s inability to enclose the large CoA tail, which may interfere with a mechanism providing high apparent affinity for shorter C-17 side-chained substrates while increasing kcat. Indeed, the steroid core of 1,4-BNC-CoA orients in a nearly identical manner to ADD indicating the CoA moiety is responsible for the difference in kinetic behavior.  While only the mouth loop open state has been observed in KshAMtb and KshA1, no structure has been solved of KshA1 capable of exhibiting a closed loop such as that in KshA5. However, the similar parameters for KshA1 with 1,4-BNC and 1,4-BNC-CoA indicate that both substrates bind and react in a similar manner and that the mouth loop does not have the same role in KshA1 or KshAMtb substrate selectivity as hypothesized here for KshA5. Indeed, different mechanism for excluding solvent may exist in KshA1 and KshAMtb but is likely dependent on the C-17 side chain obscuring the substrate access channel, not the mouth loop. While the channel is 60  most constricted in KshA1 with the 4-BNC isopropionyl large enough to isolate the active site from bulk solvent (Figure 4.3A) and position the carboxyl of the steroid?s C-17 side chain near the hydroxyl of Tyr246, it is less restricted in KshAMtb with the large CoA thioester necessary to obscure the active site (Figure 4.3B). Indeed, the high apparent specificities of KshA1 for isopropionyl at C-17 and KshAMtb for CoA thioesters indicate that the enzymes? maximum specificity is dependent on closure of the active site to solvent. Thus, the best substrates for each enzyme are those which provide a snug fit to the opening of the active site, whereas in KshA5 the mouth loop encloses the active site and is adaptable to different substrate shapes. Other more subtle mechanisms could be responsible for the kinetic behavior of KshA5. Indeed, the shift in iron positioning is unprecedented in KshAs and other ROs. Theoretically, the ability of the active site to adapt to the degree observed in KshA5 may allow the iron to position itself for catalysis with a wide range of substrate shapes. This may provide the enzyme with its broad steroid affinity; although it remains to be shown that this is a physiological behavior rather than an artifact of crystallographic observation. Regardless, it appears as though the structural flexibility of KshA5 is related to its intriguing catalytic behavior.  4.3.3 Movement within the catalytic center Substrate-induced shifts in the KshA5 mononuclear iron and its ligands are similar in nature but greater than previous observations in ROs, indicating that movement is relevant to the regulation of the catalytic cycle. The mononuclear iron was shifted away from the substrate binding pocket in response to Rieske cluster reduction in OMO (21) and following substrate binding in BPDO (14) and CARDO (24).  While these movements are in the same direction as those in KshA5 upon binding ADD and 1,4-BNC-CoA, the changes in KshA5 are more widespread. Indeed, the movement is not isolated to the iron and its ligands, but is evident in the helices containing the iron-coordinating residues including the transverse helix (Figure 4.5), which shifts longitudinally 4 to 4.6 ? toward its C-terminal end and slightly towards the Rieske domain, as determined in relation to the ?-sheet of the substrate-free structure, which shows very little movement. Movement of this type and magnitude has not been observed in any of the other 12 ROs structurally characterized to date. Furthermore, the lack of a shift in KshA1 and KshAMtb, which were prepared in the same manner, suggests that this flexibility is unique to  61    Figure 4.5   Superposition of KshA5 (gray) and KshA5:ADD (cyan). The mononuclear iron is represented as a sphere. Residues Glu310, Phe301 and Glu299 of the transverse helix and Tyr227 of the loop are displayed as sticks. Shifts of helical residue C?s upon binding substrate are shown with arrows indicating the distance between substrate-free and substrate-bound KshA5 in a PyMol structural alignment. Figure generated using PyMol (20).  KshA5 and may be related to its interesting kinetic behavior. While most residues of the helix remain in similar rotameric positions, Phe309 rotates to keep the ring in the same pocket on the edge of the substrate-binding site. KshA5:ADD and KshA5:1,4-BNC-CoA retain highly similar positioning of residues of the transverse helix with an RMSD of 0.4 ? over 35 atoms from a PyMOL superpositioning. Interestingly, the movement of the transverse helix exposes Thr261 of the ?-sheet in the KshA5:ADD structure, allowing its hydroxyl to form a 2.7 ? hydrogen-bond with the hydroxyl of Tyr227 (Figure 4.5). This demonstrates a potential communication link between the mouth loop and the presence of substrate, indicating that the substrate-induced shift in iron and ?-helix positioning exposes a H-bond partner on the ?-sheet for the loop to attach, promoting closure of the active site. Indeed this is the only H-bond to the loop throughout residues 221-228, which block the active site. The Thr261 residue is a serine in KshAMtb and KshA1 but is completely sequestered from solvent in KshA1:1,4-BNC-CoA. While it is exposed in KshAMtb and KshAMtb:ADD, it is surrounded by large residues, making it sterically Tyr227 Tyr227 Glu310 Phe301 Glu299 4.4 ? 4.4 ? ADD 62  inaccessible and indicating that such a H-bond and link between the loop and active site does not exist in KshA1 or KshAMtb. This single interaction point between the mouth loop and the active site demonstrate that the substrate?s presence may support the closing the active site during catalysis. The iron to C-9 distance of 6.4 ? in KshA5:ADD and 6.1 ? in KshA5:1,4-BNC-CoA is ~2 ? longer than observed in any other RO and may be of mechanistic relevance. While this distance would enable the binding and activation of O2 to the iron, it is likely too long for efficient hydroxylation at C-9. As it has been previously observed that the binding of O2 and analogue NO to the catalytic iron is gated by the presence of substrate (33), this long distance conformation may represent a state along the catalytic pathway open to the coordination of O2. Indeed, such a distant conformation is likely amenable to the binding of a wide range of steroid substrates, potentially helping to explain the broad specificity exhibited by KshA5. As described earlier, the Rieske cluster is likely in its reduced state, allowing O2 binding while substrate is in place. In theory, following the binding and activation of O2, the catalytic iron would move closer to substrate, bringing the activated oxygen species within attacking distance of C-9 while opening the active site. Thus, changes observed in the active site upon KshA5 binding substrate may be responsible for both wide substrate selectivity and regulating the O2 reactivity of the enzyme.  While KshA1 and KshAMtb were crystallized with 1,4-BNC-CoA and ADD, respectively in similar conditions to KshA5, neither shows a large substrate to iron distance. KshAMtb maintains a similar positioning of iron in both ADD-bound and substrate free structures. KshA1 has a slightly longer iron to C-9 distance than KshAMtb but no structure of KshA1 without substrate is available for comparison. While the large movement in the catalytic domain of KshA5 upon binding substrate likely does not occur in other KshA or ROs which have been structurally characterized, the crystallization of KshA5 may have captured an intermediate previously unobserved. The crystalline packing effects are likely not strong enough to distort the active site to such a degree, but might be sufficient to stabilize an intermediate state of KshA5?s catalytic pathway. Indeed, the shift is observed over three helices in addition to the catalytic iron and with two different substrates, both crystallized in the same spacegroup. Thus, it is likely that 63  the iron and helix shift observed is physiologically relevant, though further experimentation is necessary to demonstrate this concretely. 4.3.4 Movement of the bridging Asp  While the ?bridging? Asp178-O?2 of KshAMtb:ADD was positioned slightly closer to the His ligand of the Rieske cluster than the substrate-free form, similar distances were observed throughout the structures of KshA1 (Asp179) and KshA5 (Asp185) (Table 4.1). In all structures, the residue was in a similar conformation to that of KshAMtb:ADD (Figure 4.2).  The inter-subunit distances observed in KshA are similar to that observed in other ROs with most in the range of 2.6-2.8 ? from the His-N? of the mononuclear domain to the backbone carbonyl of the bridging Asp and the Asp-O?2 to His-N? of the Rieske cluster. Shifts observed in the catalytic center of KshA5 upon substrate binding were parallel to the subunit interface and thus did not affect H-bonding distances with the bridging Asp. Larger distances observed in some ROs were up to 3.7 ? for both of these spans, indicating that the interaction between the clusters is intact in each KshA crystal structure. However, it is difficult to draw any conclusions based on these observations as the oxidation state of the Rieske cluster is unknown and the pH of the crystallization condition and the crystalline lattice have an unknown effect on the H-bonding interactions between the catalytic sites.  Table 4.1   Inter-subunit distances of ?bridging? Asp-His of mononuclear and Rieske domains Structure His-N?-Asp backbone O (?) His-N?-Asp-O?1 (?) Asp-O?2-Rieske His-N? (?) KshAMtb 2.8 3.2 2.9 KshAMtb:ADD 2.9 3.2 2.7 KshA1:1,4-BNC-CoA 2.7 3.2 2.8 KshA5 2.8 3.6 2.7 KshA5:ADD 2.7 3.6 2.8 KshA5:1,4-BNC-CoA 2.7 3.2 2.6  64  4.4 Concluding remarks  This study has indicated that the interesting substrate specificity observed in KshA5 in comparison to KshA1 is likely due to the intriguing flexibility of KshA5, which has been revealed through structural analysis and suggests that further experimentation may confirm a correlation between the structural flexibility of KshA5 and its unusual kinetic activity. Solution studies utilizing NMR or fluorescence spectroscopy could be utilized to better characterize the movement of the mouth loop and the catalytic center. Confirmation of a connection between the large movement of the iron, ligands, and helices upon binding of substrate may help to describe the unique catalytic ability of KshA5. Furthermore, comparison to other KshA or ROs would confirm if the flexibility observed is unique to KshA5 and if these results imply previously unobserved dynamics in ROs. Additionally, mutational studies of key residues in KshA1 and KshA5 observed to provide steric hindrance or H-bonding partners could confirm the role that these residues play in substrate selectivity and loop conformations. Further work in mechanistic evaluation of KshA through peroxide shunt or single turnover studies would help in mechanism comparison of KshA to other ROs to further evaluate KshA?s unique behavior. Furthermore, insight into the ability of KshA5 to efficiently turnover a broad range of substrates may help engineer enzymes for broad-spectrum selectivity, having potential applications in the synthesis of pharmaceuticals and the degradation of environmental pollutants.   65  Bibliography  1. Kovaleva, E. G., and Lipscomb, J. D. (2008) Versatility of biological non-heme Fe(II) centers in oxygen activation reactions. Nat Chem Biol 4, 186-193 2. Sono, M., Roach, M. P., Coulter, E. D., and Dawson, J. H. (1996) Heme-containing oxygenases. Chem Rev 96, 2841-2887 3. Pazmino, D. E. T., Winkler, M., Glieder, A., and Fraaije, M. W. (2010) Monooxygenases as biocatalysts: Classification, mechanistic aspects and biotechnological applications. J Biotechnol 146, 9-24 4. Ferraro, D. J., Gakhar, L., and Ramaswamy, S. (2005) Rieske business: Structure-function of Rieske non-heme oxygenases. Biochem Bioph Res Co 338, 175-190 5. Wackett, L. P. (2002) Mechanism and applications of Rieske non-heme iron dioxygenases. Enzyme Microb Tech 31, 577-587 6. Capyk, J. K., and Eltis, L. D. (2012) Phylogenetic analysis reveals the surprising diversity of an oxygenase class. J Biol Inorg Chem 17, 425-436 7. Summers, R. M., Louie, T. M., Yu, C. L., and Subramanian, M. (2011) Characterization of a broad-specificity non-haem iron N-demethylase from Pseudomonas putida CBB5 capable of utilizing several purine alkaloids as sole carbon and nitrogen source. Microbiol-Sgm 157, 583-592 8. Priefert, H., Rabenhorst, J., and Steinbuchel, A. (1997) Molecular characterization of genes of Pseudomonas sp. strain HR199 involved in bioconversion of vanillin to protocatechuate. Journal of Bacteriology 179, 2595-2607 9. Yoshiyama-Yanagawa, T., Enya, S., Shimada-Niwa, Y., Yaguchi, S., Haramoto, Y., Matsuya, T., Shiomi, K., Sasakura, Y., Takahashi, S., Asashima, M., Kataoka, H., and Niwa, R. (2011) The Conserved Rieske Oxygenase DAF-36/Neverland Is a Novel Cholesterol-metabolizing Enzyme. Journal of Biological Chemistry 286, 25756-25762 10. Oster, U., Tanaka, R., Tanaka, A., and Rudiger, W. (2000) Cloning and functional expression of the gene encoding the key enzyme for chlorophyll b biosynthesis (CAO) from Arabidopsis thaliana. Plant J 21, 305-310 11. Li, P., Wang, L., and Feng, L. (2013) Characterization of a Novel Rieske-Type Alkane Monooxygenase System in Pusillimonas sp. Strain T7-7. J Bacteriol 195, 1892-1901 12. Resnick, S. M., Lee, K., and Gibson, D. T. (1996) Diverse reactions catalyzed by naphthalene dioxygenase from Pseudomonas sp strain NCIB 9816. J Ind Microbiol Biot 17, 438-457 13. Hudlicky, T., Gonzalez, D., and Gibson, D. T. (1999) Enzymatic dihydroxylation of aromatics in enantioselective synthesis: Expanding asymmetric methodology. Aldrichim Acta 32, 35-62 14. Mohammadi, M., Viger, J. F., Kumar, P., Barriault, D., Bolin, J. T., and Sylvestre, M. (2011) Retuning Rieske-type Oxygenases to Expand Substrate Range. Journal of Biological Chemistry 286, 27612-27621 15. Fernandes, P., Cruz, A., Angelova, B., Pinheiro, H. M., and Cabral, J. M. S. (2003) Microbial conversion of steroid compounds: recent developments. Enzyme Microb Tech 32, 688-705 16. Peng, R. H., Xiong, A. S., Xue, Y., Fu, X. Y., Gao, F., Zhao, W., Tian, Y. S., and Yao, Q. H. (2010) A Profile of Ring-hydroxylating Oxygenases that Degrade Aromatic Pollutants. Rev Environ Contam T 206, 65-94 17. Petrusma, M., Dijkhuizen, L., and van der Geize, R. (2012) Structural features in the KshA terminal oxygenase protein that determine substrate preference of 3-ketosteroid 9alpha-hydroxylase enzymes. J Bacteriol 194, 115-121 18. Wolfe, M. D., Parales, J. V., Gibson, D. T., and Lipscomb, J. D. (2001) Single turnover chemistry and regulation of O2 activation by the oxygenase component of naphthalene 1,2-dioxygenase. J Biol Chem 276, 1945-1953 66  19. Capyk, J. K., D'Angelo, I., Strynadka, N. C., and Eltis, L. D. (2009) Characterization of 3-ketosteroid 9{alpha}-hydroxylase, a Rieske oxygenase in the cholesterol degradation pathway of Mycobacterium tuberculosis. J Biol Chem 284, 9937-9946 20. The PyMOL Molecular Graphics System. Version 1.2.0 Ed., Schr?dinger, LLC. 21. Martins, B. M., Svetlitchnaia, T., and Dobbek, H. (2005) 2-Oxoquinoline 8-monooxygenase oxygenase component: active site modulation by Rieske-[2Fe-2S] center oxidation/reduction. Structure 13, 817-824 22. Furusawa, Y., Nagarajan, V., Tanokura, M., Masai, E., Fukuda, M., and Senda, T. (2004) Crystal structure of the terminal oxygenase component of biphenyl dioxygenase derived from Rhodococcus sp. strain RHA1. J Mol Biol 342, 1041-1052 23. Gakhar, L., Malik, Z. A., Allen, C. C. R., Lipscomb, D. A., Larkin, M. J., and Ramaswamy, S. (2005) Structure and increased thermostability of Rhodococcus sp naphthalene 1,2-dioxygenase. Journal of Bacteriology 187, 7222-7231 24. Ashikawa, Y., Fujimoto, Z., Usami, Y., Inoue, K., Noguchi, H., Yamane, H., and Nojiri, H. (2012) Structural insight into the substrate- and dioxygen-binding manner in the catalytic cycle of rieske nonheme iron oxygenase system, carbazole 1,9a-dioxygenase. BMC Struct Biol 12, 15 25. Parales, R. E., Parales, J. V., and Gibson, D. T. (1999) Aspartate 205 in the catalytic domain of naphthalene dioxygenase is essential for activity. Journal of Bacteriology 181, 1831-1837 26. Tarasev, M., Pinto, A., Kim, D., Elliott, S. J., and Ballou, D. P. (2006) The "bridging" aspartate 178 in phthalate dioxygenase facilitates interactions between the Rieske center and the iron(II)--mononuclear center. Biochemistry 45, 10208-10216 27. Pinto, A., Tarasev, M., and Ballou, D. P. (2006) Substitutions of the "bridging" aspartate 178 result in profound changes in the reactivity of the Rieske center of phthalate dioxygenase. Biochemistry 45, 9032-9041 28. Beharry, Z. M., Eby, D. M., Coulter, E. D., Viswanathan, R., Neidle, E. L., Phillips, R. S., and Kurtz, D. M. (2003) Histidine ligand protonation and redox potential in the Rieske dioxygenases: Role of a conserved aspartate in anthranilate 1,2-dioxygenase. Biochemistry 42, 13625-13636 29. Kauppi, B., Lee, K., Carredano, E., Parales, R. E., Gibson, D. T., Eklund, H., and Ramaswamy, S. (1998) Structure of an aromatic-ring-hydroxylating dioxygenase-naphthalene 1,2-dioxygenase. Structure 6, 571-586 30. Wolfe, M. D., Altier, D. J., Stubna, A., Popescu, C. V., Munck, E., and Lipscomb, J. D. (2002) Benzoate 1,2-dioxygenase from Pseudomonas putida: single turnover kinetics and regulation of a two-component Rieske dioxygenase. Biochemistry 41, 9611-9626 31. Karlsson, A., Parales, J. V., Parales, R. E., Gibson, D. T., Eklund, H., and Ramaswamy, S. (2003) Crystal structure of naphthalene dioxygenase: Side-on binding of dioxygen to iron. Science 299, 1039-1042 32. Lee, K. (1999) Benzene-induced uncoupling of naphthalene dioxygenase activity and enzyme inactivation by production of hydrogen peroxide. Journal of Bacteriology 181, 2719-2725 33. Yang, T. C., Wolfe, M. D., Neibergall, M. B., Mekmouche, Y., Lipscomb, J. D., and Hoffman, B. M. (2003) Substrate binding to NO-ferro-naphthalene 1,2-dioxygenase studied by high-resolution Q-band pulsed 2H-ENDOR spectroscopy. J Am Chem Soc 125, 7056-7066 34. Ohta, T., Chakrabarty, S., Lipscomb, J. D., and Solomon, E. I. (2008) Near-IR MCD of the nonheme ferrous active site in naphthalene 1,2-dioxygenase: correlation to crystallography and structural insight into the mechanism of Rieske dioxygenases. J Am Chem Soc 130, 1601-1610 35. Gassner, G. T., Ballou, D. P., Landrum, G. A., and Whittaker, J. W. (1993) Magnetic circular dichroism studies on the mononuclear ferrous active site of phthalate dioxygenase from Pseudomonas cepacia show a change of ligation state on substrate binding. Biochemistry 32, 4820-4825 36. Pavel, E. G., Martins, L. J., Ellis, W. R., Jr., and Solomon, E. I. (1994) Magnetic circular dichroism studies of exogenous ligand and substrate binding to the non-heme ferrous active site in phthalate dioxygenase. Chem Biol 1, 173-183 67  37. van der Geize, R., Hessels, G. I., van Gerwen, R., van der Meijden, R., and Dijkhuizen, L. (2002) Molecular and functional characterization of kshA and kshB, encoding two components of 3-ketosteroid 9 alpha-hydroxylase, a class IA monooxygenase, in Rhodococcus erythropolis strain SQ1. Molecular Microbiology 45, 1007-1018 38. Capyk, J. K., Casabon, I., Gruninger, R., Strynadka, N. C., and Eltis, L. D. (2011) Activity of 3-ketosteroid 9alpha-hydroxylase (KshAB) indicates cholesterol side chain and ring degradation occur simultaneously in Mycobacterium tuberculosis. J Biol Chem 286, 40717-40724 39. van der Geize, R., and Dijkhuizen, L. (2004) Harnessing the catabolic diversity of rhodococci for environmental and biotechnological applications. Curr Opin Microbiol 7, 255-261 40. Hu, Y., van der Geize, R., Besra, G. S., Gurcha, S. S., Liu, A., Rohde, M., Singh, M., and Coates, A. (2010) 3-Ketosteroid 9alpha-hydroxylase is an essential factor in the pathogenesis of Mycobacterium tuberculosis. Mol Microbiol 75, 107-121 41. Van der Geize, R., Yam, K., Heuser, T., Wilbrink, M. H., Hara, H., Anderton, M. C., Sim, E., Dijkhuizen, L., Davies, J. E., Mohn, W. W., and Eltis, L. D. (2007) A gene cluster encoding cholesterol catabolism in a soil actinomycete provides insight into Mycobacterium tuberculosis survival in macrophages. Proc Natl Acad Sci U S A 104, 1947-1952 42. Mohn, W. W., Wilbrink, M. H., Casabon, I., Stewart, G. R., Liu, J., van der Geize, R., and Eltis, L. D. (2012) A gene cluster encoding cholate catabolism in Rhodococcus spp. J Bacteriol  43. Casabon, I., Crowe, A. M., Liu, J., and Eltis, L. D. (2012) FadD3 is an acyl-CoA synthetase that initiates catabolism of cholesterol rings C and D in actinobacteria. Mol Microbiol  44. Nesbitt, N. M., Yang, X., Font?n, P., Kolesnikova, I., Smith, I., Sampson, N. S., and Dubnau, E. (2010) A thiolase of Mycobacterium tuberculosis is required for virulence and production of androstenedione and androstadienedione from cholesterol. Infect Immun 78, 275-282 45. D'Ordine, R. L., Rydel, T. J., Storek, M. J., Sturman, E. J., Moshiri, F., Bartlett, R. K., Brown, G. R., Eilers, R. J., Dart, C., Qi, Y., Flasinski, S., and Franklin, S. J. (2009) Dicamba monooxygenase: structural insights into a dynamic Rieske oxygenase that catalyzes an exocyclic monooxygenation. J Mol Biol 392, 481-497 46. Nojiri, H., Ashikawa, Y., Noguchi, H., Nam, J. W., Urata, M., Fujimoto, Z., Uchimura, H., Terada, T., Nakamura, S., Shimizu, K., Yoshida, T., Habe, H., and Omori, T. (2005) Structure of the terminal oxygenase component of angular dioxygenase, carbazole 1,9a-dioxygenase. Journal of Molecular Biology 351, 355-370 47. Petrusma, M., Hessels, G., Dijkhuizen, L., and van der Geize, R. (2011) Multiplicity of 3-Ketosteroid-9alpha-Hydroxylase enzymes in Rhodococcus rhodochrous DSM43269 for specific degradation of different classes of steroids. J Bacteriol 193, 3931-3940 48. Ferraro, D. J., Okerlund, A. L., Mowers, J. C., and Ramaswamy, S. (2006) Structural basis for regioselectivity and stereoselectivity of product formation by naphthalene 1,2-dioxygenase. Journal of Bacteriology 188, 6986-6994 49. Mondello, F. J., Turcich, M. P., Lobos, J. H., and Erickson, B. D. (1997) Identification and modification of biphenyl dioxygenase sequences that determine the specificity of polychlorinated biphenyl degradation. Appl Environ Microbiol 63, 3096-3103 50. Sambrook, J., Fritsch, E.F., Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 51. Pulleyblank, D., Michalak, M., Daisley, S. L., and Glick, R. (1983) A method for the purification of E. coli plasmid DNA by homogeneous lysis and polyethylene glycol precipitation. Mol Biol Rep 9, 191-195 52. Eltis, L. D., Iwagami, S. G., and Smith, M. (1994) Hyperexpression of a synthetic gene encoding a high potential iron sulfur protein. Protein Eng 7, 1145-1150 53. Vaillancourt, F. H., Han, S., Fortin, P. D., Bolin, J. T., and Eltis, L. D. (1998) Molecular basis for the stabilization and inhibition of 2, 3-dihydroxybiphenyl 1,2-dioxygenase by t-butanol. J Biol Chem 273, 34887-34895 68  54. Gomez-Gil, L., Kumar, P., Barriault, D., Bolin, J. T., Sylvestre, M., and Eltis, L. D. (2007) Characterization of biphenyl dioxygenase of Pandoraea pnomenusa B-356 as a potent polychlorinated biphenyl-degrading enzyme. Journal of Bacteriology 189, 5705-5715 55. Chen, J. S., and Mortenson, L. E. (1977) Inhibition of methylene blue formation during determination of the acid-labile sulfide of iron-sulfur protein samples containing dithionite. Anal Biochem 79, 157-165 56. Zabinski, R., Munck, E., Champion, P. M., and Wood, J. M. (1972) Kinetic and Mossbauer studies on the mechanism of protocatechuic acid 4,5-oxygenase. Biochemistry 11, 3212-3219 57. Otwinowski, Z., and Minor, W. (1997) Processing of X-ray diffraction data collected in oscillation mode. Method Enzymol 276, 307-326 58. McCoy, A. J. (2007) Solving structures of protein complexes by molecular replacement with Phaser. Acta Crystallogr D 63, 32-41 59. Cowtan, K. (2006) The Buccaneer software for automated model building. 1. Tracing protein chains. Acta Crystallogr D 62, 1002-1011 60. Murshudov, G. N., Vagin, A. A., and Dodson, E. J. (1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D 53, 240-255 61. Schuttelkopf, A. W., and van Aalten, D. M. F. (2004) PRODRG: a tool for high-throughput crystallography of protein-ligand complexes. Acta Crystallogr D 60, 1355-1363 62. Bailey, S. (1994) The Ccp4 Suite - Programs for Protein Crystallography. Acta Crystallogr D 50, 760-763 63. Petrusma, M., Dijkhuizen, L., and van der Geize, R. (2009) Rhodococcus rhodochrous DSM 43269 3-ketosteroid 9alpha-hydroxylase, a two-component iron-sulfur-containing monooxygenase with subtle steroid substrate specificity. Appl Environ Microbiol 75, 5300-5307 64. Parales, R. E., Huang, R., Yu, C. L., Parales, J. V., Lee, F. K., Lessner, D. J., Ivkovic-Jensen, M. M., Liu, W., Friemann, R., Ramaswamy, S., and Gibson, D. T. (2005) Purification, characterization, and crystallization of the components of the nitrobenzene and 2-nitrotoluene dioxygenase enzyme systems. Appl Environ Microbiol 71, 3806-3814 65. Swain, K., Casabon, I., Eltis, L. D., and Mohn, W. W. (2012) Two transporters essential for the reassimilation of novel cholate metabolites by Rhodococcus jostii RHA1. J Bacteriol  66. Thomas, S. T., VanderVen, B. C., Sherman, D. R., Russell, D. G., and Sampson, N. S. (2011) Pathway profiling in Mycobacterium tuberculosis: elucidation of cholesterol-derived catabolite and enzymes that catalyze its metabolism. J Biol Chem 286, 43668-43678 67. Ge, Y., and Eltis, L. D. (2003) Characterization of hybrid toluate and benzoate dioxygenases. J Bacteriol 185, 5333-5341 68. Lee, J., Simurdiak, M., and Zhao, H. (2005) Reconstitution and characterization of aminopyrrolnitrin oxygenase, a Rieske N-oxygenase that catalyzes unusual arylamine oxidation. J Biol Chem 280, 36719-36727 69. Imbeault, N. Y., Powlowski, J. B., Colbert, C. L., Bolin, J. T., and Eltis, L. D. (2000) Steady-state kinetic characterization and crystallization of a polychlorinated biphenyl-transforming dioxygenase. J Biol Chem 275, 12430-12437 70. Jouanneau, Y., Meyer, C., Jakoncic, J., Stojanoff, V., and Gaillard, J. (2006) Characterization of a naphthalene dioxygenase endowed with an exceptionally broad substrate specificity toward polycyclic aromatic hydrocarbons. Biochemistry 45, 12380-12391 71. Park, H., Lee, S., and Suh, J. (2005) Structural and dynamical basis of broad substrate specificity, catalytic mechanism, and inhibition of cytochrome P450 3A4. J Am Chem Soc 127, 13634-13642     69  Appendix Appendix A  Data collection and refinement statistics  A5:ADD A5:1,4-BNCCoA A5/A1:1,4-BNC-CoA KshAMtb:ADD Data collection     Wavelength (?) 0.92 0.98 0.98 0.92 Space group P63 P63 F23 P32 Cell dimensions         a, b, c (?) 162.2, 162.2, 46.97 163.18, 163.18,  47.09 273.6, 273.6, 273.6 115.8, 115.8, 80.81    ?, ?, ? (?)  90, 90, 120 90, 90, 120 90, 90, 90 90, 90, 120 Resolution (?) 46.82-1.92(1.97-1.92) 53.41-2.6(2.74-2.6) 68.41-2.45(2.52-2.45) 47.07-2.46(2.52-2.46) Rsym 0.124(1.318) 0.244(1.278) 0.176(1.524) 0.161(1.15) Rpim 0.043(0.494) 0.101(0.551) 0.067(0.586) 0.071(0.582) CC1/2 0.999(0.453) 0.983(0.197) 0.998(0.536) 0.997(0.412) I I 18(2.0) 7.1(1.5) 12.6(2.0) 13.77(1.61) Completeness (%) 88.1(88.1) 100(100) 100(100) 99.95(99.87) Redundancy 9.3(7.9) 6.7(6.2) 14.7(14.3) 9.7(9.4)      Refinement     Resolution (?) 1.92 2.6 2.45 2.46 No. reflections (Unique) 49275(1620) 22443(3231) 64010(4518) 23062(1502) Rwork / Rfree 0.183/0.216 0.199/0.256 0.169/0.213 0.182/0.253 No. atoms         Protein 2987 2961 5888 2947     Ligand/ion 50 81 112 28     Water 341 55 369 132 B-factors         Protein 28 44.16 47.29 46.39     Ligand/ion 35.3 73.18 65.39 34.01     Water 36.4 40.03 46.19 40.85 R.m.s. deviations         Bond lengths (?) 0.022 0.017 0.022 0.017     Bond angles (?) 2.103 2.151 2.318 1.975      Ramachandran Plot     Outliers 0.00% 5.15% 1.80% 2.49% Allowed 1.90% 16.80% 5.83% 4.43% Favored 98.10% 78.05% 92.37% 93.07% *Values in parentheses are for highest-resolution shell. aRsym = ?h?l |Ihl - ?Ih?|/?h?l ?Ih? bRpim = ?h/(nh-1)?l |Ihl - ?Ih?|/?h?l ?Ih? cRwork = ?||Fo| - |Fc||/?|Fo|, where Fo and Fc represent the observed and calculated structure factors, respectively. cRfree is the Rwork value for 5% of the reflections excluded from the refinement.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0103343/manifest

Comment

Related Items