{"http:\/\/dx.doi.org\/10.14288\/1.0400464":{"http:\/\/vivoweb.org\/ontology\/core#departmentOrSchool":[{"value":"Science, Faculty of","type":"literal","lang":"en"},{"value":"Chemistry, Department of","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/dataProvider":[{"value":"DSpace","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeCampus":[{"value":"UBCV","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/creator":[{"value":"Hedges, Jason","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/issued":[{"value":"2024-07-31T07:00:00Z","type":"literal","lang":"en"},{"value":"2021","type":"literal","lang":"en"}],"http:\/\/vivoweb.org\/ontology\/core#relatedDegree":[{"value":"Doctor of Philosophy - PhD","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeGrantor":[{"value":"University of British Columbia","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/description":[{"value":"Over time, evolution can result in enzymes developing novel functions. An emerging example of this phenomenon comes from studies of pyridoxal-5\u02b9-phosphate (PLP)-dependent enzymes, which catalyze a diverse set of chemical reactions on amino acid substrates. This thesis describes the discovery and characterization of some unusual PLP-dependent enzymes.\r\nSome PLP-dependent enzymes have been shown to catalyze challenging oxidations of an L-arginine substrate using O\u2082 as a co-substrate. In Chapter 2 I set out to describe new PLP-dependent arginine oxidases and study how they function. Using bioinformatics, I was able to hone in on one particular enzyme, named RohP. Biochemical characterization of RohP revealed that it is an arginine hydroxylase, which catalyzes the formation of (S)-4-hydroxy-2-ketoarginine. Furthermore, I was able to obtain several high-resolution X-ray crystal structures of RohP at different stages of its catalytic cycle. Together these results advance the understanding of how O\u2082- and PLP-dependent enzymes function.\r\nRohP was found in a conserved five gene biosynthetic gene cluster, with no known product. Therefore, in Chapter 3 I set out to determine what the product of this unusual biosynthetic gene cluster was. Using the studies of RohP as a starting point, additional in vitro biochemical investigations of four other enzymes encoded along with RohP in this biosynthetic gene cluster revealed that together they convert L-arginine to the antibiotic azomycin (2-nitroimidazole). As azomycin was first isolated over 50 years ago, the discoveries described in this chapter solve a longstanding biosynthetic mystery.\r\n\tInteresting PLP-dependent enzymes are found in many biosynthetic pathways. In Chapter 4 I report my characterization of BesB, an unrelated PLP-dependent enzyme which catalyzes the formation of a terminal alkyne bond. BesB has limited solubility in E. coli, which has hampered its study initially. Through use of a different heterologous expression system I was able to obtain soluble BesB. Through biochemical and X-ray crystallographic analysis, an active site phenylalanine substitution appears to be key to unlocking the novel reactivity of BesB. This study provides the first crystal structures of any alkyne-forming enzyme. Insights from the studies of RohP and BesB should prove useful in developing novel PLP-dependent biocatalysts.","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/aggregatedCHO":[{"value":"https:\/\/circle.library.ubc.ca\/rest\/handle\/2429\/79048?expand=metadata","type":"literal","lang":"en"}],"http:\/\/www.w3.org\/2009\/08\/skos-reference\/skos.html#note":[{"value":"BIOCHEMICAL AND STRUCTURAL STUDIES OF ENZYMES FROM THE AZOMYCIN AND BETA-ETHYNYLSERINE BIOSYNTHETIC PATHWAYS by  Jason Hedges  B.Sc. (Hons), The University of Western Ontario, 2013  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  July 2021  \u00a9 Jason Hedges, 2021  ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: BIOCHEMICAL AND STRUCTURAL STUDIES OF ENZYMES FROM THE AZOMYCIN AND BETA-ETHYNYLSERINE BIOSYNTHETIC PATHWAYS  submitted by Jason Hedges in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Chemistry  Examining Committee: Katherine Ryan, Associate Professor, Chemistry, UBC Supervisor  Martin Tanner, Professor, Chemistry, UBC Supervisory Committee Member  Harry Brumer, Professor, Chemistry and Michael Smith Laboratories, UBC Supervisory Committee Member David Perrin, Professor, Chemistry, UBC University Examiner Michael Fryzuk, Professor Emeritus, Chemistry, UBC University Examiner  Additional Supervisory Committee Members: Hongbin Li, Professor, Chemistry, UBC Supervisory Committee Member   iii  Abstract Over time, evolution can result in enzymes developing novel functions. An emerging example of this phenomenon comes from studies of pyridoxal-5\u02b9-phosphate (PLP)-dependent enzymes, which catalyze a diverse set of chemical reactions on amino acid substrates. This thesis describes the discovery and characterization of some unusual PLP-dependent enzymes. Some PLP-dependent enzymes have been shown to catalyze challenging oxidations of an L-arginine substrate using O2 as a co-substrate. In Chapter 2 I set out to describe new PLP-dependent arginine oxidases and study how they function. Using bioinformatics, I was able to hone in on one particular enzyme, named RohP. Biochemical characterization of RohP revealed that it is an arginine hydroxylase, which catalyzes the formation of (S)-4-hydroxy-2-ketoarginine. Furthermore, I was able to obtain several high-resolution X-ray crystal structures of RohP at different stages of its catalytic cycle. Together these results advance the understanding of how O2- and PLP-dependent enzymes function. RohP was found in a conserved five gene biosynthetic gene cluster, with no known product. Therefore, in Chapter 3 I set out to determine what the product of this unusual biosynthetic gene cluster was. Using the studies of RohP as a starting point, additional in vitro biochemical investigations of four other enzymes encoded along with RohP in this biosynthetic gene cluster revealed that together they convert L-arginine to the antibiotic azomycin (2-nitroimidazole). As azomycin was first isolated over 50 years ago, the discoveries described in this chapter solve a longstanding biosynthetic mystery.  Interesting PLP-dependent enzymes are found in many biosynthetic pathways. In Chapter 4 I report my characterization of BesB, an unrelated PLP-dependent enzyme which catalyzes the formation of a terminal alkyne bond. BesB has limited solubility in E. coli, which has hampered its study initially. Through use of a different heterologous expression system I was able to obtain soluble BesB. Through biochemical and X-ray crystallographic analysis, an active site phenylalanine substitution appears to be key to unlocking the novel reactivity of BesB. This study provides the first crystal structures of any alkyne-forming enzyme. Insights from the studies of RohP and BesB should prove useful in developing novel PLP-dependent biocatalysts.   iv  Lay Summary Natural products are compounds produced by living organisms and are the source of many of the medicines that are currently in use. Nature constructs these compounds using enzymes, protein catalysts that increase the rate of a chemical reaction. In this thesis I describe the discovery and characterization of several novel biosynthetic enzymes using a range of biochemical studies as well as X-ray crystallography. I discovered that one group of enzymes function together to produce the antibiotic azomycin. Azomycin was the lead molecule for the development of the nitroimidazoles, which are used today for the treatment of anaerobic bacterial infections. In a second study I characterize an enzyme which catalyzes an unprecedented type of oxidative reaction. Understanding this enzyme may lead to applications in chemical biology and biocatalysis.    v  Preface Chapter 1 and Chapter 5 are both original unpublished discussions written by the author (Jason Hedges), with editing and feedback provided by Katherine Ryan (supervisor).  A version of Chapter 2 has been published as: Hedges, J. B.; Kuatsjah, E.; Du, Y.-L.; Eltis, L. D.; Ryan, K. S. Snapshots of the Catalytic Cycle of an O2, Pyridoxal Phosphate-Dependent Hydroxylase. ACS Chem. Biol. 2018, 13 (4), 965\u2013974. The author (Jason Hedges), carried out purification of enzymes, enzyme activity assays, and crystallographic studies. Eugene Kuatsjah assisted with the collection of enzyme kinetic data, and oxygen consumption data via Oxygraph. The manuscript was written by Jason Hedges and Katherine Ryan (supervisor) with input and feedback from all authors.  A version of Chapter 3 has been published as: Hedges, J. B.; Ryan, K. S. In Vitro Reconstitution of the Biosynthetic Pathway to the Nitroimidazole Antibiotic Azomycin. Angew. Chem. Int. Ed. 2019, 58 (34), 11647\u201311651. The author (Jason Hedges), carried out all experimental work. Purified BphF and BphG were provided by Lindsay Eltis. The author (Jason Hedges) also wrote the manuscript with editing and feedback provided by Katherine Ryan (supervisor).  Chapter 4 contains original unpublished data. The author (Jason Hedges), carried out all experimental work under the supervision of Katherine Ryan (supervisor). vi  Table of Contents Abstract ......................................................................................................................................... iii Lay Summary ............................................................................................................................... iv Preface .............................................................................................................................................v Table of Contents ......................................................................................................................... vi List of Tables ............................................................................................................................... xii List of Figures ............................................................................................................................. xiii List of Equations ........................................................................................................................ xxi List of Abbreviations ................................................................................................................ xxii Acknowledgements .................................................................................................................. xxiv Chapter 1: Introduction ................................................................................................................1 1.1 General synopsis of introduction .................................................................................... 1 1.2 Nitro-containing natural products ................................................................................... 2 1.2.1 Nitroimidazoles ........................................................................................................... 3 1.2.1.1 Mechanism of action ........................................................................................... 5 1.2.1.2 Early studies on the biosynthesis of 2-nitroimidazole (azomycin) ..................... 6 1.2.2 Biosynthetic routes to nitro-containing natural products ............................................ 7 1.2.2.1 P450-catalyzed direct nitration: TxtE and RufO ................................................ 7 1.2.3 FAD-dependent oxidation ........................................................................................... 9 1.2.3.1 Nitrosuccinate biosynthesis: CreE, FzmM ......................................................... 9 1.2.3.2 Nitrosugar biosynthesis: Orf36, RubN8, KijD3 ............................................... 10 1.2.4 Non-heme iron dependent oxidases .......................................................................... 13 1.2.4.1 Iron and Rieske-dependent oxidase: PrnD ........................................................ 14 vii  1.2.4.2 Non-heme diiron oxidases: AurF ...................................................................... 15 1.2.4.3 Non-heme diiron oxidases: CmlI ...................................................................... 18 1.3 Alkyne-containing bioactive compounds ..................................................................... 21 1.3.1 Internal alkyne biosynthesis ...................................................................................... 23 1.3.1.1 Diiron-dependent acetylenases: Crep1 ............................................................. 23 1.3.1.2 Cytochrome P450-dependent 1,3-enyne formation: AtyI, BisI ........................ 26 1.3.2 Terminal alkyne biosynthesis ................................................................................... 27 1.3.2.1 ACP-tethered desaturation: JamB ..................................................................... 27 1.3.2.2 PLP-dependent desaturation: BesB................................................................... 29 1.4 PLP-dependent enzymes ............................................................................................... 31 1.4.1 Introduction ............................................................................................................... 31 1.4.2 Reactions catalyzed by PLP-dependent enzymes ..................................................... 32 1.4.3 Dunathan\u2019s stereoelectronic hypothesis .................................................................... 33 1.4.4 PLP-dependent enzyme classification ...................................................................... 34 1.4.5 PLP-dependent aminotransferases ............................................................................ 36 1.4.5.1 Structure of PLP-dependent aminotransferases ................................................ 36 1.4.5.2 PLP-dependent aminotransferase mechanism .................................................. 39 1.4.6 PLP-dependent \u03b3-synthases and \u03b3-lyases .................................................................. 41 1.4.6.1 Reaction mechanism for PF01053 .................................................................... 43 1.4.6.2 Mechanism of inhibition ................................................................................... 45 1.5 O2- and PLP-dependent enzymes .................................................................................. 49 1.5.1 O2- and PLP-dependent L-arginine oxidases ............................................................ 51 1.5.1.1 Indolmycin biosynthesis: Ind4 .......................................................................... 51 viii  1.5.1.2 Enduracididine biosynthesis: MppP ................................................................. 53 1.6 Thesis goals ................................................................................................................... 55 Chapter 2: Identification and characterization of the O2- and PLP-dependent arginine hydroxylase RohP ........................................................................................................................58 2.1 Introduction ................................................................................................................... 58 2.2 Materials and Methods .................................................................................................. 61 2.2.1 General methods ....................................................................................................... 61 2.2.2 Cloning and expression of RohP............................................................................... 61 2.2.3 Purification of RohP ................................................................................................. 61 2.2.4 In vitro biochemical assays and product analysis ..................................................... 62 2.2.5 NMR analysis of RohP reaction products ................................................................. 63 2.2.6 Steady-state kinetics for the RohP reaction .............................................................. 63 2.2.7 Stoichiometry of the RohP reaction .......................................................................... 64 2.2.8 Spectroscopic analysis of the RohP reaction ............................................................ 64 2.2.9 RohP crystallization .................................................................................................. 65 2.2.10 Data collection, structure determination and model refinement ........................... 65 2.3 Results ........................................................................................................................... 66 2.3.1 RohP is an L-arginine hydroxylase ........................................................................... 66 2.3.2 Kinetics and stoichiometry of the RohP-catalyzed reaction ..................................... 73 2.3.3 The hydroxyl group in the RohP product derives from water, not oxygen or H2O2 . 76 2.3.4 UV-Visible spectrum of RohP .................................................................................. 77 2.3.5 Crystal structure of holo-RohP ................................................................................. 78 2.3.6 Trapping RohP the first quinonoid ........................................................................... 80 ix  2.3.7 Trapping of a more conjugated quinonoid orders the N-terminus of RohP ............. 82 2.3.8 Structure of RohP-product complex ......................................................................... 85 2.3.9 Characterization of the RohP-His34Ala variant ....................................................... 88 2.4 Discussion ..................................................................................................................... 89 Chapter 3: In vitro biochemical characterization of the azomycin biosynthetic gene cluster........................................................................................................................................................93 3.1 Introduction ................................................................................................................... 93 3.2 Materials and Methods .................................................................................................. 94 3.2.1 General Methods ....................................................................................................... 94 3.2.2 Cloning and expression of recombinant proteins ...................................................... 94 3.2.3 Purification of recombinant proteins ........................................................................ 95 3.2.4 In vitro biochemical analysis of the RohP-RohR coupled reaction .......................... 96 3.2.5 In vitro biochemical analysis of the RohQ-catalyzed reaction ................................. 97 3.2.6 In vitro biochemical analysis of the KAzRohS-catalyzed reaction .......................... 97 3.2.7 KAzRohS kinetic methods ........................................................................................ 98 3.2.8 ICP-MS analysis of KAzRohS and RohT ................................................................. 99 3.2.9 UV Vis-spectroscopy of RohT.................................................................................. 99 3.2.10 In vitro biochemical analysis of RohT .................................................................. 99 3.2.11 Anaerobic reconstitution of RohT ...................................................................... 100 3.2.12 In vitro production of azomycin using four enzymes ......................................... 100 3.2.13 Cultivation and extraction of metabolites from Streptomyces cattleya .............. 101 3.2.14 Cultivation and extraction of metabolites from Pseudomonas ........................... 101 3.3 Results ......................................................................................................................... 102 x  3.3.1 Bioinformatic analysis of a cryptic gene cluster ..................................................... 102 3.3.2 In vitro analysis of the aldolase RohR .................................................................... 104 3.3.3 RohQ catalyzes a spontaneous cyclization reaction ............................................... 110 3.3.4 Analysis of nitroimidazole formation, catalyzed by RohS ..................................... 113 3.3.5 RohT may play a role in the RohS-catalyzed oxidation of 2-aminoimidazole ....... 117 3.3.6 Probing for azomycin production by Streptomyces cattleya .................................. 120 3.3.7 Probing for azomycin production by Pseudomonas ............................................... 121 3.3.8 Other proposed biosynthetic routes to azomycin .................................................... 123 3.4 Discussion ................................................................................................................... 124 Chapter 4: Characterization of the alkyne-forming PLP-desaturase BesB: Mechanism of pyridoxal phosphate-dependent alkyne formation .................................................................127 4.1 Introduction ................................................................................................................. 127 4.2 Materials and methods ................................................................................................ 132 4.2.1 General methods ..................................................................................................... 132 4.2.2 Cloning and protein expression .............................................................................. 132 4.2.3 Protein purification ................................................................................................. 134 4.2.4 Protein purification of halide free protein ............................................................... 134 4.2.5 Initial in vitro biochemical assays of BesB,............................................................ 135 4.2.6 NMR analysis of 4-bromoallylglycine and NaCl exchange product ...................... 136 4.2.7 Protein crystallization ............................................................................................. 137 4.2.8 Data collection, structural determination, and model refinement ........................... 137 4.3 Results ......................................................................................................................... 138 4.3.1 Expression and purification of BesB using Rhodococcus ...................................... 138 xi  4.3.2 In vitro biochemical activity assays ........................................................................ 142 4.3.3 Exploring additional in vitro BesB reactivity ......................................................... 150 4.3.4 Heavy atom derivatization and structural phasing of BesB .................................... 152 4.3.5 Overall tertiary structure of the BesB monomer ..................................................... 153 4.3.6 Refinement of unknown PLP-adduct ...................................................................... 157 4.3.7 Analyzing the active site of BesB ........................................................................... 159 4.3.8 Site-directed mutagenesis studies of BesB ............................................................. 163 4.3.9 Engineering BesB for improved expression and stability using PROSS ................ 166 4.4 Discussion ................................................................................................................... 168 Chapter 5: Conclusion ...............................................................................................................173 5.1 Chapter 2 Conclusions ................................................................................................ 173 5.2 Chapter 3 Conclusions ................................................................................................ 174 5.3 Chapter 4 Conclusions ................................................................................................ 176 5.4 Overall conclusions ..................................................................................................... 178 Bibliography ...............................................................................................................................179 Appendices ..................................................................................................................................196 Appendix A Supporting data for Chapter 2: Identification and Characterization of the O2- and PLP-dependent arginine hydroxylase RohP ........................................................................... 196 Appendix B Supporting data for Chapter 3: In vitro biochemical characterization of the azomycin biosynthetic gene cluster ........................................................................................ 210 Appendix C Supporting data for Chapter 4: Characterization and Crystallization of the unusual PLP-dependent desaturase BesB............................................................................................. 224 xii  List of Tables Table 1.1 Summary of PLP-dependent enzyme classification and activities. .............................. 34 Table 2.1 High-resolution ESI-MS analysis of RohP- L-arginine reaction products. ................... 73 Table 3.1 Bacterial strains with dual copies of the azomycin biosynthetic gene cluster. ........... 104 Table 4.1 Bacterial strains containing BesB homologs. ............................................................. 139  Table A.1 RohP X-ray data collection statistics. ........................................................................ 196 Table A.2 RohP X-ray data refinement statistics. ....................................................................... 197 Table A.3 Conserved residue numbering.................................................................................... 198 Table A.4 Comparison of the bond lengths generated from ARP\/wARP refinement to the bonds lengths of the Q2 ligand. ............................................................................................................. 199 Table B.1 Strains and vectors used in Chapter 3. ....................................................................... 210 Table B.2 PCR Primers used for studies in Chapter 3. ............................................................... 211 Table B.3 Protein purification buffer for azomycin biosynthetic enzymes. ............................... 211 Table B.4 KAzRohS ICP-MS results.......................................................................................... 212 Table B.5 RohT ICP-MS results. ................................................................................................ 212 Table B.6 Azomycin biosynthetic gene cluster in Streptomyces cattleya. ................................. 213 Table C.1 Strains and vectors used in Chapter 4. ....................................................................... 224 Table C.2 Primers used for PCR, cloning, and sequencing. ....................................................... 225 Table C.3 BesB X-ray data collection statistics.......................................................................... 227 Table C.4 BesB X-ray structure refinement statistics. ................................................................ 228 Table C.5 Major qualitative effects of mutations in Crep1 on chemoselectivity, stereoselectivity, and substrate specificity .............................................................................................................. 229 xiii  List of Figures Figure 1.1 Select examples of nitro-containing natural products. .................................................. 2 Figure 1.2 Structures of select 2- and 5-nitroimidazoles. ............................................................... 3 Figure 1.3 Reductive pathways of a) the 5-nitroimidazole metronidazole and b) 2-nitroimidazoles. ............................................................................................................................... 5 Figure 1.4 Eguchi\u2019s proposed biosynthetic route to azomycin. ...................................................... 6 Figure 1.5 Nitric oxide synthase and P450-coupled direct nitration of aromatic amino acids. ...... 7 Figure 1.6 The flavin-dependent oxidative pathway to nitrosuccinate. .......................................... 9 Figure 1.7 Nitrosugar biosynthesis by the FAD-dependent enzymes KijD3, RubN8, and Orf36.10 Figure 1.8 Superposition of the 3.15 \u00c5 structure of Orf36 and 2.05 \u00c5 structure of KijD3. ......... 11 Figure 1.9 2.10 \u00c5 structure of the FAD and TDP-sugar complex of KijD3................................. 13 Figure 1.10 Reaction catalyzed by PrnD. ..................................................................................... 14 Figure 1.11 Proposed mechanism for the oxidation of p-aminobenzoic acid by AurF. ............... 15 Figure 1.12 Di-iron active site of AurF complexed with its product 4-nitrobenzoic acid. ........... 16 Figure 1.13 Selected proposed diiron sites of arylamine oxygenases. ......................................... 17 Figure 1.14 Proposed mechanism for the oxidation of CAM by CmlI. ........................................ 18 Figure 1.15 Active sites of reduced and peroxo-bound CmlI. ...................................................... 19 Figure 1.16 Select examples of alkyne-containing bioactive molecules. ..................................... 21 Figure 1.17 General scheme for the CuI-catalyzed click reaction. ............................................... 22 Figure 1.18 Crep1-catalyzed oxidation to form internal alkynes. ................................................ 23 Figure 1.19 Various polyacetylenic compounds constructed using Crep1 homologs. ................. 25 Figure 1.20 Biosynthesis of 1,3-enyne containing compounds. ................................................... 26 Figure 1.21 Selected terminal alkyne-containing natural products. ............................................. 27 xiv  Figure 1.22 JamABC-catalyzed formation of the terminal alkyne moiety during the biosynthesis of jamaicamide B and related compounds. ................................................................................... 27 Figure 1.23 Biosynthesis of \u03b2-ethynylserine in Streptomyces cattleya. ....................................... 29 Figure 1.24 Schiff base formation in PLP-dependent enzymes. ................................................... 31 Figure 1.25 General reaction scheme for PLP-dependent enzymes. ............................................ 32 Figure 1.26 Depiction the empty \u03c0-orbitals of the PLP cofactor in Dunathan\u2019s stereoelectronic hypothesis. .................................................................................................................................... 34 Figure 1.27 Reaction catalyzed by aspartate aminotransferase (AAT). ....................................... 36 Figure 1.28 Structure of aspartate aminotransferase. .................................................................... 37 Figure 1.29 General mechanism of PLP-dependent aminotransferases. ...................................... 39 Figure 1.30 Reactions catalyzed by PF01053 synthases use serine and homoserine derivatives. 41 Figure 1.31 Reactions catalyzed by PF01053 lyases. ................................................................... 42 Figure 1.32 Crystal structure of MGL from Clostridium sporogenes. ......................................... 43 Figure 1.33 Mechanism for PLP-dependent \u03b3-elimination catalyzed by CGL. ............................ 44 Figure 1.34 Proposed mechanisms of inhibition for 4-chloroallylglycine and propargylglycine. 46 Figure 1.35 Structure of the propargylglycine-tyrosine adduct in human CGL. .......................... 48 Figure 1.36 Reactions catalyzed by O2- and PLP-dependent oxidative enzymes discussed in the text................................................................................................................................................. 50 Figure 1.37 Indolmycin biosynthesis in Streptomyces griseus. .................................................... 51 Figure 1.38 Summary of on-pathway Ind4- and Ind5-catalyzed reaction pathways. ................... 52 Figure 1.39 Structure of enduracidin (left) and biosynthetic pathway to L-enduracididine (right)........................................................................................................................................................ 53 Figure 1.40 Overall structure (top) and active site (bottom) of MppP. ........................................ 54 xv  Figure 1.41 Active site of the MppP (S)-4-hydroxy-2-ketoarginine complex. ............................. 55 Figure 2.1 Sequence alignment of RohP and homologous enzymes. ........................................... 67 Figure 2.2 Comparison of biosynthetic gene clusters containing O2- and PLP-dependent arginine oxidases. ........................................................................................................................................ 69 Figure 2.3 RohP-catalyzed consumption of L-arginine. ............................................................... 70 Figure 2.4 RohP-catalyzed oxidation of L-arginine. ..................................................................... 71 Figure 2.5 Summary of Ind4- and RohP-catalyzed reaction pathways. ....................................... 72 Figure 2.6 pH-dependence of the rate of the RohP-catalyzed reaction. ....................................... 73 Figure 2.7 Steady state consumption of O2 by RohP. ................................................................... 74 Figure 2.8 RohP steady-state kinetics with respect to L-arginine and oxygen. ............................ 76 Figure 2.9 UV-Vis absorption spectroscopy of the reaction between RohP and L-arginine. ....... 77 Figure 2.10 Comparison of the MppP and RohP internal aldimine structures. ............................ 79 Figure 2.11 Modeling of RohP-quinonoid 1 intermediates. ......................................................... 81 Figure 2.12 Distances between modelled intermediates and conserved residues Glu20 and His34 in Chain A of the RohP homodimer. ............................................................................................ 81 Figure 2.13 L-arginine soaking of RohP crystals. ......................................................................... 83 Figure 2.14 Quinonoid intermediates from a red, L-arginine soaked crystal. ............................... 83 Figure 2.15 Modeling of RohP-quinonoid 2 intermediates. ......................................................... 84 Figure 2.16 Active site of RohP-product complex. ...................................................................... 86 Figure 2.17 Possible stereoisomers of 4-hydroxy-2-ketoarginine. ............................................... 87 Figure 2.18 Refinement of alternative products. .......................................................................... 88 Figure 2.19 Reaction of the RohP-His34Ala variant with L-arginine. ......................................... 89 Figure 2.20 Proposed mechanism of the RohP-catalyzed reaction to 4-hydroxy-2-ketoarginine. 91 xvi  Figure 3.1 In vitro reactions linking L-arginine to azomycin. ...................................................... 94 Figure 3.2 Organization of the azomycin biosynthetic gene cluster in different soil-dwelling bacteria. ....................................................................................................................................... 103 Figure 3.3 Dual azomycin biosynthetic gene clusters in Streptomyces eurocidicus ATCC 27428...................................................................................................................................................... 103 Figure 3.4 Diverse reactions catalyzed by dihydrodipicolinate-like enzymes.192 ...................... 105 Figure 3.5 Sequence alignment of RohR compared to DHDPS and other characterized retro-aldolases. ..................................................................................................................................... 106 Figure 3.6 RohR- and RohQ-catalyzed production of 2-aminoimidazole and pyruvate. ........... 107 Figure 3.7 OPD analysis of the products of the RohP catalyzed oxidation of L-arginine. ......... 108 Figure 3.8 Possible mechanism for retro-aldol cleavage by RohR. ............................................ 109 Figure 3.9 RohQ catalyzed production of 2-aminoimidazole. .................................................... 110 Figure 3.10 LC-MS quantification of RohQ-catalyzed cyclodehydration of guanidinoacetaldehyde. ............................................................................................................... 111 Figure 3.11 Initial product distributions for the a) RohP + RohR coupled reactions and b) RohP + RohR + RohQ coupled reactions with L-arginine. ...................................................................... 112 Figure 3.12 Streptomyces griseocarnus strain 132, lacking a rohQ homolog. ........................... 113 Figure 3.13 N-oxygenases that have been experimentally characterized. .................................. 114 Figure 3.14 Multiple sequence alignment of RohS and other discussed N-oxygenases. ............ 115 Figure 3.15 In vitro production of azomycin. ............................................................................. 116 Figure 3.16 ESI-MS of KAzRohS in vitro reactivity in the presence of various transition metals...................................................................................................................................................... 117 xvii  Figure 3.17 KAzRohS steady state kinetic analysis with respect to its substrate 2-aminoimidazole. .......................................................................................................................... 117 Figure 3.18 Sequence alignment of Rieske proteins discussed in the text. ................................ 118 Figure 3.19 Purification and reduction of RohT. ........................................................................ 119 Figure 3.20 One pot production of azomycin. ............................................................................ 120 Figure 3.21 Ethyl acetate extractions of Streptomyces cattleya fermentations. ......................... 121 Figure 3.22 Azomycin gene cluster organization in Pseudomonas. ........................................... 122 Figure 3.23 LC-MS analysis of Pseudomonas brassicacearum DF41 ethyl acetate extracts. ... 123 Figure 3.24 Proposal for azomycin biosynthesis based on results from Graham and coworkers...................................................................................................................................................... 124 Figure 3.25 Diverse routes to arginine-derived heterocycles involving O2- and PLP-dependent arginine oxidases. ........................................................................................................................ 125 Figure 4.1 Characterized biosynthetic routes to acetylenic natural products. ............................ 128 Figure 4.2 Mechanistic proposal for the BesB-catalyzed reaction. ............................................ 130 Figure 4.3 IMAC purification of BesB. ...................................................................................... 140 Figure 4.4 Size-exclusion chromatography of BesB. ................................................................. 141 Figure 4.5 UV-Vis absorbance spectra of BesB. ........................................................................ 142 Figure 4.6 Reaction of 4-chloroallylglycine and BesB at different pHs..................................... 143 Figure 4.7 Reaction of 4-bromoallylglycine and BesB at different pHs. ................................... 144 Figure 4.8 Stability of 4-chloro- and 4-bromoallylglycine at pH 6. ........................................... 145 Figure 4.9 Isotopic distributions of 4-chloroallylglycine produced by BesB in D2O. ................ 147 Figure 4.10 Isotopic distribution of L-propargylglycine produced by BesB in D2O. ................. 148 xviii  Figure 4.11 Reaction of BesB with 4-bromoallylglycine in the absence of Cl\u2212 at different pHs...................................................................................................................................................... 149 Figure 4.12 Reactions of BesB with L-allylglycine at pH 6 and 9. ............................................ 150 Figure 4.13 Reactions of BesB with L-allylglycine at pH 6 and 9 in D2O. ................................ 151 Figure 4.14 BesB CGS activity. .................................................................................................. 152 Figure 4.15 BesB FPLC traces from expression in Rhodococcus with 1 mM selenomethionine...................................................................................................................................................... 153 Figure 4.16 Structure of BesB and alignments with two monomers of a homotetrameric cystathionine \u03b2-lyase. .................................................................................................................. 154 Figure 4.17 Structural alignments of BesB with related cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthases. .................................................................................................................................... 156 Figure 4.18 BesB crystals and corresponding UV-Vis absorbance spectra. ............................... 157 Figure 4.19 Fitting of unknown PLP-adduct into the density present in the active site of the wild-type BesB crystal. ....................................................................................................................... 159 Figure 4.20 Active site of wild-type BesB.................................................................................. 160 Figure 4.21 Comparison of the active site of BesB with the active sites of related enzymes. ... 161 Figure 4.22 Comparison of the structure of wild-type and F231Y BesB. .................................. 162 Figure 4.23 FPLC traces of purified BesB variants. ................................................................... 163 Figure 4.24 In vitro reactivity of F231Y BesB. .......................................................................... 164 Figure 4.25 F231Y BesB CGS activity....................................................................................... 164 Figure 4.26 In vitro activity of F62Y and F232Y BesB variants................................................ 165 Figure 4.27 In vitro reactivity of PROSS BesB variants. ........................................................... 167 xix  Figure 4.28 PROSS BesB variant-catalyzed 4-chloroallylglycine production from 4-bromoallylglycine. ...................................................................................................................... 168 Figure 4.29 Proposed catalytic mechanism for BesB. ................................................................ 171  Figure A.1 1H NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid. ............................................................................................................................................. 200 Figure A.2 13C NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid. ............................................................................................................................................. 201 Figure A.3 1H-1H COSY NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid.................................................................................................................. 202 Figure A.4 1H-13C HSQC NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid.................................................................................................................. 203 Figure A.5 1H-13C HMBC NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid.................................................................................................................. 204 Figure A.6 NMR assignment summary. ..................................................................................... 205 Figure A.7 Modeling of external aldimine (top) and quinonoid intermediates (bottom). .......... 206 Figure A.8 Modeling of possible oxidized Q2 intermediates. .................................................... 207 Figure A.9 ARP\/wARP refinement of oxidized Q2 intermediate .............................................. 208 Figure A.10 PLP-arginine adducts modelled into available density in the active site of the red crystal. Density maps and ligand fitting generated from a 2.0 \u00c5 data set. ................................. 209 Figure B.1 Phylogenetic analysis of azomycin biosynthetic gene clusters in soil-dwelling bacteria. ....................................................................................................................................... 214 Figure B.2 DNA sequence of codon optimized Kitasatospora azatica RohS homolog. ............ 216 xx  Figure B.3 UV-Vis absorbance spectra of RohT. ....................................................................... 217 Figure B.4 UV-Vis of anaerobically reduced RohT. .................................................................. 218 Figure B.5 Reaction of RohT with 2-aminoimidazole. ............................................................... 219 Figure B.6 Analysis of the KAzRohS + RohT coupled reaction. ............................................... 220 Figure B.7 KAzRohS reactions employing reduced RohT. ........................................................ 221 Figure B.8 The KAzRohS + RohT reaction with non-cognate ferredoxin reductases. .............. 222 Figure B.9 The KAzRohS + RohT reaction with a heterologous ferredoxin-ferredoxin reductase pair. ............................................................................................................................................. 223 Figure C.1 Tested synthetic substrate mimics of AtyI. ............................................................... 230 Figure C.2 Multiple sequence alignment of BesB with related PLP-dependent \u03b3-lyases. ......... 231 Figure C.3 1H NMR spectrum of exchanged 4-chloroallylglycine. ............................................ 232 Figure C.4 13C NMR spectrum of exchanged 4-chloroallylglycine. .......................................... 233 Figure C.5 1D NOESY spectrum of exchanged 4-chloroallylglycine. ....................................... 234 Figure C.6 Multiple sequence alignment of BesB homologs. .................................................... 235 Figure C.7 Complete tetrameric structure of a homotetrameric cystathionine \u03b3-lyase. .............. 236 Figure C.8 PROSS Construct amino acid and DNA sequences. ................................................ 241  xxi  List of Equations Equation 1 Michaelis \u2013 Menten equation. .................................................................................... 75 Equation 2 Hill Equation. ............................................................................................................. 75    xxii  List of Abbreviations ACN  acetonitrile ATP  adenosine triphosphate BLAST Basic Local Alignment Search Tool CBL  cystathionine-\u03b2-lyase CGL  cystathionine-\u03b3-lyase CGS  cystathionine-\u03b3-synthase CHES  N-cyclohexyl-2-aminoethanesulfonic acid DNS-Cl dansyl chloride E. coli  Escherichia coli EDTA  ethylenediaminetetraacetic acid EtOAc  ethyl acetate FA  formic acid FAD  flavin adenine dinucleotide FMN  flavin mononucleotide FPLC  fast protein liquid chromatography HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HEPPS 4-(2-hydroxyethyl)-piperazine-1-propanesulfonic acid HPLC  high-performance liquid chromatography ICP-MS inductively coupled plasma mass spectrometry IPTG  \u03b2-D-1-thiogalactopyranoside KB  King\u2019s B Medium KIE  kinetic isotope effect LB  Luria-Bertani LC-MS liquid chromatography\u2013mass spectrometry MeOH  methanol MES  2-(N-morpholino)-ethanesulfonic acid MGL  methionine-\u03b3-lyase MOPS  3-(N-morpholino)-propanesulfonic acid MPA  mercaptopropionic acid xxiii  NAD(P)H nicotinamide adenine dinucleotide (phosphate) NCBI  The National Center for Biotechnology Information NMR  nuclear magnetic resonance NRPS  non-ribosomal peptide synthetase OPA  o-phthaldialdehyde OPD  o-phenylenediamine PCR  polymerase chain reaction PDB  Protein Data Bank PIPES  1,4-piperazinediethanesulfonic acid PKS  polyketide synthase PLP  pyridoxal 5\u2019-phosphate PMS  phenazine methosulfate RMSD  root-mean-square deviation RT  room temperature SAM  S-adenosyl methionine SIM  selected ion monitoring TAPPS tris(hydroxymethyl)methylamino propanesulfonic acid TDP  thymidine diphosphate TFA  trifluoroacetic acid Tris  2-amino-2-(hydroxymethyl)propane-1,3-diol TLS  translation-liberation-screw Tsr   thiostrepton    xxiv  Acknowledgements I would like to begin by expressing my sincere gratitude to my supervisor, Dr. Katherine Ryan. It was a privilege to work with and learn from her, and her support and encouragement over the past few years was invaluable. I would also like to take the opportunity to thank Dr. Harry Brumer, Dr. Martin Tanner, and Dr. Hongbin Li for being part of my supervisory committee, as well as their helpful feedback over the course of my degree. I would also like to thank Dr. David Perrin, Dr. Michael Fryzuk, and Dr. Geoff Horsman for taking the time to be my PhD defense examiners. Additionally, I would like to thank Dr. Elena Polishchuk and Jessie Chen from Biological Services Laboratory, as well as Ben Herring from the Shared Instrument Facility for running great facilities in support of the graduate students in chemistry. I would also like to thank all the members of the Ryan lab (past and present) for making our lab a great place to work. I am also indebted to the now Dr. Linglan Fu for her support and making me eat my vegetables. I could not have done this without your help. Last, but not least I would also like to would also like to thank my family for their support and understanding, while I have spent the past few years so far away.1  Chapter 1: Introduction 1.1 General synopsis of introduction Natural products contain a vast array of molecular scaffolds which can be modified with numerous chemical functional groups. A recent study analyzed over 186,000 natural products found in the Dictionary of Natural Products database using an algorithm designed to identify and extract unique connections between heteroatoms revealed that these molecules contained 2,785 unique functional groups.1 This study also found that functional groups differ significantly in the frequency of their occurrence. For example, the most common functional group, aliphatic alcohols, are found in over 61% of natural products in the database. The prevalence of aliphatic alcohols is in stark contrast with many of the less common functional groups such as the nitro and alkyne, which are only present in 0.2% and 0.37% of the molecules, respectively.1 These two functional groups are more common in synthetic molecules and medical compounds,1,2 thus discovering novel biosynthetic enzymes can catalyze the formation of rare functional groups such as the nitro and alkyne is highly desirable. The work covered in this thesis will focus on characterizing steps in the biosynthesis of azomycin (2-nitroimidazole) (Chapters 2 and 3), and the terminal alkyne-containing amino acid \u03b2-ethynylserine (Chapters 4). This introductory chapter first introduces nitro-containing natural products, including azomycin, as well as other molecules with a nitro group. Second, known alkyne biosynthetic pathways, including \u03b2-ethynylserine, are introduced. As the biosynthesis of both azomycin and \u03b2-ethynylserine exhibit a shared dependence on highly unusual PLP-dependent enzymes, a discussion of PLP-dependent enzymes follows. This discussion will include relevant PLP-dependent enzyme mechanisms to set the stage for mechanistic discussions in Chapters 2 and 4 and will focus on the recent discoveries in O2- and PLP-dependent enzymology. The goals for each of these projects is then introduced.   2  1.2 Nitro-containing natural products  Figure 1.1 Select examples of nitro-containing natural products.  The nitro group is an important and versatile group in medicinal chemistry, with a long and continuing history of therapeutic use.3,4 The highly polar nature of the nitro group allows for strong interactions between nitro-containing small molecules and common biological components such as proteins or DNA.5,6 The nitro group serves as a prodrug, which is activated by enzymatic reduction to form various radicals, which are the biologically active species. However, reactions between drug radicals and proteins or DNA can also induce severe toxicity, and this toxicity is a common issue with nitro-containing drugs.3,7 Due their inherent toxicity in biological systems, nitro-containing compounds are not often associated with natural products, but in fact there are hundreds of isolated natural products that contain nitro groups, many of which have distinct natural biological activities.8,9 The structural diversity of these nitro-containing compounds is on display in Figure 1.1, which shows the structures of several relevant natural products that will be discussed 3  in more detail in the upcoming text. Investigating the biosynthetic origins of the various nitro groups found in these compounds has revealed several distinct biosynthetic routes the nitro functional group. These routes can involve direct nitration of the substrate, but most of the discovered biosynthetic pathways involve the direct oxidation of an amine to a nitro group. Three different classes of enzyme have been shown to catalyze oxidation of an amine to a nitro, and understanding the underlying mechanisms of nitro functional group biosynthesis is currently a robust area of research.10\u201312  1.2.1 Nitroimidazoles  Figure 1.2 Structures of select 2- and 5-nitroimidazoles. The 2- and 5-nitroimidazole moieties are highlighted in blue and red respectively.  One of the most notable naturally occurring nitro-containing compounds is azomycin (2-nitroimidazole) (6) (Figure 1.2). Azomycin was first isolated from Streptomyces eurocidicus in 1953,13 and its molecular structure was determined soon after in 1955.14 At the time of the discovery of 6, its simple 2-nitroimidazole structure represented a novel natural product scaffold, 4  and thus its biological properties were investigated. Azomycin (6) proved to be a highly efficacious treatment for combatting various anaerobic pathogens, most notably Trichomonas vaginalis, a pathogenic aerobic bacterium for which there were then no effective antibiotic treatments.15 These promising results inspired the chemical synthesis of many new nitroimidazoles along with the evaluation of the corresponding biological activities for these molecules. Thus, 6 is the founding member of the clinically important nitroimidazole class of drugs (Figure 1.2).  Despite the relatively simple structure of 6, at the time of its discovery it proved to be quite a challenge to synthesize and it was only in 1965 that a pair of groups reported the synthesis of 6.16,17 Both groups utilized NaNO2 and CuSO4\u03876H2O to catalyze the oxidation of 2-aminoimidazole to 6, albeit in relatively low yields.16,17 Despite the challenges of synthesizing 2-nitroimidazoles, some 2-nitroimidazoles were eventually developed into viable drugs. One such example is benznidazole (7), which was developed in 1972, and is still one of the few effective treatments for Trypanosoma cruzi, the cause of Chagas disease.3  In the intervening years, scientists also discovered that 5-nitroimidazoles (8) were much easier to synthesize, and more importantly, these compounds were found to be generally better tolerated by patients.15,18,19 Perhaps the most famous and widely used nitroimidazole drug is the 5-nitroimidazole metronidazole (9) (Figure 1.2). Metronidazole was first synthesized in 1959 and subsequently rapidly approved as a drug in 1960, and is still used to this day for the treatment of trichomoniasis and other anaerobic bacterial infections.19\u201321 Other important 5-nitroimidazoles include tinidazole (10), a broad spectrum antiparasitic compound developed in 1972.3,19 The synthesis and evaluation of novel 5-nitroimidazoles continues even today, with the 5-nitroimidazole scaffold being continually explored to produce novel nitroimidazole drugs with improved biological activities.22,23 One recent example is delamanid (11), which was first produced in 2006,24 and subsequently approved for the treatment of multidrug-resistant tuberculosis by the EU in 2014 (Figure 1.2).25  5  1.2.1.1 Mechanism of action  Figure 1.3 Reductive pathways of a) the 5-nitroimidazole metronidazole and b) 2-nitroimidazoles.  Due to the extensive use of nitroimidazoles over the past 60 years, the mode of action as well as their decomposition pathways under anaerobic conditions have been extensively studied (Figure 1.3). Although the specific mode of action depends on the nature of the substituents present on the nitroimidazole scaffold, all nitroimidazoles generally exert their antibiotic activity through similar reductive mechanisms. Once the nitroimidazole has diffused into the cell, the nitro-group is reduced to a reactive radical species by the anaerobic pyruvate\/ferredoxin reductase complex.15,18,19,21 The reduction potential of aerobic cells is too high to catalyze this reductive reaction, and any nitroimidazole that is reduced is re-oxidized by O2, in a process called futile cycling,19 revealing how nitroimidazoles specifically target anaerobic cells. The nitro radical anion is the primary damaging species, and can react with DNA, inducing the formation of DNA double-6  strand breaks triggering irreversible damage to the DNA.15,18\u201321 The nitro radical anion can also decompose to yield NO2- and an imidazole radical, that latter of which can also cause cellular damage. There is also evidence supporting the reduction of the nitro radical anion to corresponding nitroso, hydroxylamine, or even to the amine various cellular nitroreductases.20,21 However, reduction of the nitro radical anion is less prevalent for 5-nitroimidazoles such as tinidazole (10) and is more common with 2-nitroimidazoles such as 6 and benznidazole (7).19,26   1.2.1.2 Early studies on the biosynthesis of 2-nitroimidazole (azomycin)  Figure 1.4 Eguchi\u2019s proposed biosynthetic route to azomycin.  Scheme adopted from Nakane et al.27 Proposed compounds not detected during the experiments from Ref. 27 are highlighted in brackets and in grey.  While the 1960\u2019s and 70\u2019s saw intense efforts devoted to the synthesis and evaluation of novel nitroimidazoles, 6 was not fully relegated to the sideline, with several groups seeking to uncover the biosynthesis of this molecule. Early studies from Lancini and others in the 1960\u2019s were the first to investigate the biological origin of 6. In these studies they observed the accumulation of 2-aminoimidazole (12), and its oxidation to 6 by the azomycin producer Streptomyces sp. LE\/3342.28\u201330 The earlier stages of azomycin biosynthesis in another bacterial strain, Streptomyces eurocidicus SF506, were investigated several years later by Eguchi. His group demonstrated that L-arginine (13) was the initial precursor to 6, and the biosynthesis of 6 proceeded through an unusual hydroxylated arginine derivative 4-hydroxy-2-ketoarginine (14). The 7  production of the latter was contingent on the presence of both O2 and the enzymatic cofactor PLP.27 Furthermore, Eguchi also observed that 14 was converted into pyruvate (15) as well as the already established precursor 12 by S. eurocidicus. Combining the results of all the azomycin biosynthetic studies, Eguchi put forth a biosynthetic hypothesis to account for all of the biosynthetic intermediates detected their experiments (Figure 1.4).27 Eguchi proposed that azomycin biosynthesis begins with 13, which is oxidized in an O2- and PLP-dependent manner, to an enamine intermediate 16, which is in turn hydrolyzed to 14. Then 14 could be cleaved by an aldolase into 15 and 4-guanidinoacetaldehyde (17), the latter of which undergoes a spontaneous intramolecular cyclization to produce 12. Finally, 12 is oxidized directly to 6. Eguchi\u2019s thorough study answered several questions about the biosynthesis of 6, but also raised several new intriguing questions such as: Are there enzymes that require both O2 and PLP? And how does nature directly oxidize the amino group to a nitro group? The answers to these questions would not begin to come until several decades later.  1.2.2 Biosynthetic routes to nitro-containing natural products 1.2.2.1 P450-catalyzed direct nitration: TxtE and RufO  Figure 1.5 Nitric oxide synthase and P450-coupled direct nitration of aromatic amino acids. 8  While compounds such as chloramphenicol (5) and azomycin (6) have been known for several decades, it is only relatively recently that the biosynthesis of nitro-containing compounds has become well understood. The review of nitro group biosynthesis will begin by examining one of the less common biosynthetic routes, which involves the direct nitration of an aromatic substrate using NO. First discovered during the biosynthesis of the phytotoxin thaxtomin A (1)31\u201333 (Figure 1.1), the direct nitration route requires two enzymes: a nitric oxide synthase TxtD, and a unique type of cytochrome P450, TxtE. Bacterial nitric oxide synthases are slightly different than their mammalian counterparts, but nonetheless also catalyze the five electron oxidation of L-arginine (13) to L-citrulline (L-Cit) and NO.34 Accordingly, in vivo studies demonstrated that TxtD produces NO.32 Additionally, TxtE was able to catalyze the formation of 4-nitrotryptophan (18) from L-tryptophan (19) in vitro using the NO source NONOate (Figure 1.5).32 Crystal structures of TxtE indicates that it contains many structural features typically found in P450 enzymes, with its active site optimized for binding 19.35 However, TxtE also contains a proton transfer pathway which is different from other P450 enzymes, which may in part explain its unusual activity.36 A combination of stopped-flow spectroscopy and computational studies of TxtE indicate that the heme iron first reacts with O2 to form an iron(III)-superoxo species, which then reacts with NO, rapidly releasing NO2\u2022 that reacts with the indole ring of 19 to form 18.37  Furthermore, Zuo and Ding were able engineer a biocatalytic system for high level production of 18 using an engineered variant of TxtE.38 By coupling TxtE to the reductase domain of the P450 BM3, they were able to create a self-sufficient enzyme with improved catalytic properties.39 Engineered TxtE was heterologously expressed in E. coli along with an additional nitric oxidize synthase and glucose dehydrogenase for cofactor regeneration. Together, these three enzymes could produce up to 192 mg\/L of 18, an amount significantly higher than other engineered nitrotryptophan pathways.38  In addition to 18, the non-proteinogenic amino acid 3-nitrotyrosine (20), is also synthesized using a nitric oxide synthase-cytochrome P450 enzyme pair. Genes encoding putative bacterial nitric oxide synthase genes (ilaM, rufN) and putative cytochrome P450s (ilaN, rufO), were also identified in the biosynthetic gene clusters for the ilamycins40 and rufomycins,41 respectively. The activity of the cytochrome P450 RufO was verified in vitro and found to catalyze the formation of 20 using L-tyrosine (21) and the NO source NONOate in vitro (Figure 1.5).41 9  1.2.3 FAD-dependent oxidation The other characterized biosynthetic routes to nitro groups involve the oxidation of an amino group. Several different classes of enzyme are known to catalyze such as reaction, including FAD-dependent, Rieske-dependent, and diiron-dependent enzymes. Each group of enzymes will be reviewed individually, beginning with FAD-dependent oxidases. Nitrogen oxidizing FAD-dependent oxidases are well described in the literature, and can catalyze the oxidation of an amine to several different functional groups, including, but not limited to, hydroxylamines, oxazines, nitrones, and oximes.42 However, the following will focus on the two pathways where nitro group-containing compounds are produced.  1.2.3.1 Nitrosuccinate biosynthesis: CreE, FzmM  Figure 1.6 The flavin-dependent oxidative pathway to nitrosuccinate.  FAD-dependent oxidases such as CreE43 and FzmM44 are able to catalyze the oxidation of L-aspartic acid (22) to nitrosuccinate (23) using O2 and NADPH (Figure 1.6). CreE has be shown to consume approximately three equivalents of NADPH for every molecule of 23 produced, which indicates that this CreE-catalyzed oxidative pathway involves three individual iterative oxidations.43 The first oxidative intermediate, N-hydroxy aspartate (24), was detected during in vitro studies of FzmM, supporting the initial stages of the iterative oxidative pathway.44 Additionally, the off-pathway product 3-nitropropionic acid (25), was also detected in FzmM-containing in vitro assays, which suggests that 23 is unstable and must be quickly consumed by other enzymes before spontaneous degradation occurs.44 In these well-characterized biosynthetic 10  pathways, 23 serves as a source of nitrous acid, which is released from 23 by the actions of nitrosuccinate lyases such as CreD.43\u201345 The liberated nitrous acid is used to construct various diazo and N-N bond containing compounds. For example, nitrous acid liberated by CreD is used by the ATP-dependent enzyme CreM to form the diazo group of cremeomycin (26) (Figure 1.6).43,45,46  1.2.3.2 Nitrosugar biosynthesis: Orf36, RubN8, KijD3  Figure 1.7 Nitrosugar biosynthesis by the FAD-dependent enzymes KijD3, RubN8, and Orf36.  Nitrosugars like D-kijanose (27), D-rubranitrose (28), and L-evernitrose (29) (Figure 1.7), can all be found decorating many different natural product scaffolds. For example, the biologically active polyketide kijanimicin (2),47 (Figure 1.1) contains a single monomer of 27 appended to the polyketide core of the molecule. The biosynthesis of these sugars is quite complex, and up to ten enzymes can be required to synthesize the complete nitrosugar. This section will focus solely on 11  the steps required to synthesize the nitrosugar from the amino sugar precursor, and the extensive process of amino sugar precursor biosynthesis has been reviewed elsewhere.48,49 The enzymes responsible for catalyzing formation of the nitro group were first characterized by the Bachmann group.50 They observed homologous genes encoding putative FAD-dependent oxidases in biosynthetic gene clusters known to produce nitrosugar-containing compounds.50 Ultimately they cloned two of these genes, rubN8 and orf36 and produced the protein recombinantly using E. coli. Testing the in vitro activity of both RubN8 and Orf36 using the desmethyl analog of the natural TDP-substrate (30) demonstrated that both enzymes catalyze the formation of the TDP-hydroxylamino sugar (31) and the TDP-nitroso sugar (32) (Figure 1.7).50 Another set of experiments showed the incorporation of an oxygen atom from 18O2, indicative of oxygenase activity.51 However, in both of these studies, the final nitrosugar product (33) could not be detected under the in vitro reaction conditions tested. Using similar a similar TDP-aminosugar substrate (34) KijD3 was also found to catalyze the formation of a TDP-hydroxylamino sugar (35) and TDP-nitrososugar (36) intermediates, but again the final TDP-nitrosugar product (37) was not detected.52 The detection of various hydroxyamino and nitroso intermediates suggests that the nitrosugar synthases use the same three-step iterative oxidative pathway that is utilized by the nitrosuccinate synthases (Figure 1.6, Figure 1.7).   Figure 1.8 Superposition of the 3.15 \u00c5 structure of Orf36 and 2.05 \u00c5 structure of KijD3. The structure of Orf36 (PDB: 3MXL)51 is shown in cyan, while the structure of KijD3 (PDB: 3M9V)52 is shown in yellow. The dTDP-phenol in the KijD3 monomer is depicted as sticks. 12  The initial biochemical studies of nitrosugar synthases were followed by X-ray crystallographic studies, and in 2010 the crystal structures of both KijD352 and Orf3651 were solved (Figure 1.8). Both KijD3 and Orf36 adopt tetrameric assemblies and display the same fatty acyl-CoA dehydrogenase fold-type of fold. Orf36 was crystallized in the apo-form, lacking the FAD-cofactor,51 while KijD3 was co-crystallized with another substrate analog dTDP-phenol, which allowed for the active site to be identified.52 A third 2013 study reported a crystal structure of KijD3 bound with both FMN and its dTDP-linked sugar substrate (Figure 1.9).53 The amino-group of the substrate is located approximately 4.9 \u00c5 from C4a of the flavin, positioning it favorably for a reaction with a FMN-hydroperoxy intermediate. Subsequently, both the crystal structure of Orf36 and the crystal structure of the KijD3 dTP-phenol complex were used in molecular dynamics simulations to examine the catalytic mechanism utilized by these nitrososynthases.54,55 Modeling a FMN-hydroperoxy intermediate in the active site of the KijD3 complex revealed it can oxidize the amino group of the TDP-sugar substrate to form the experimentally detected TDP-hydroxylamine sugar intermediate.54 Furthermore, the second study demonstrated that the second oxidation catalyzed by Orf36 (hydroxylamine to nitroso) proceeds through three steps: hydroxylation, followed by hydrogen back transfer to the flavin, and then hydroxyl group elimination, with the last step being the rate limiting step in this series of reactions.55 Together, the empirical data from in vitro experiments with KijD3 and Orf36 coupled with the computational analysis provide compelling evidence to support the proposed nitrososugar biosynthetic pathway (Figure 1.7). As several nitrososugar-containing compounds have been isolated in high yields, it also is possible that the nitrososugar is in fact the true enzymatic product, which is then oxidized to the nitro product in light and air.56 The spontaneous oxidation of a hydroxylamine to nitro has also been observed in other biosynthetic pathways towards nitro-containing compounds.57  13   Figure 1.9 2.10 \u00c5 structure of the FAD and TDP-sugar complex of KijD3.  In the KijD3 structure (PDB: 4KCF)53 the dTDP-sugar is depicted as blue sticks, while the FMN is depicted as yellow sticks.  1.2.4 Non-heme iron dependent oxidases Some of the most powerful oxidative catalysts are iron-dependent enzymes, which are capable of catalyzing highly challenging oxidative reactions with high degrees of regio- and stereoselectivity.58 As shown previously, the oxidation of an amine to a nitro group can result in the formation of undesired reactive hydroxylamines and nitroso compounds. Studies of iron-dependent arylamine oxygenases have shown that these enzymes are well suited to the task and employ some uncommon oxidative chemistry to catalyze the six-electron oxidation of the arylamine group to an arylnitrogroup with a high degree of efficiency. Some of the best studied examples include the monoiron Rieske oxidase, PrnD, and the non-heme diiron enzymes AurF and CmlI.  14  1.2.4.1 Iron and Rieske-dependent oxidase: PrnD  Figure 1.10 Reaction catalyzed by PrnD.  Pyrrolnitrin (38), is a broad spectrum antifungal produced by various Pseudomonas strains.59,60 The amine to nitro oxidation of its precursor aminopyrrolnitrin (39) is catalyzed by the enzyme PrnD (Figure 1.10).61 Biochemical analysis of PrnD revealed that it contains both a mononuclear iron-binding site as well as a Rieske [2Fe-2S] cluster binding site, both of which are critical for catalysis.61 However, the catalytic mechanism utilized by PrnD still remains largely unknown. In the absence of PrnD crystal structures, molecular modeling coupled with site directed mutagenesis studies have been used probe the reactivity and substrate specificity of PrnD.62,63 These techniques demonstrated that replacement of the bulky active site residues Leu277 and Phe312 with smaller residues improved the catalytic turnover of these PrnD variants, and it was proposed that these changes to the active site could improve the orientation of the amine of 39 with respect to the iron center.62 Additionally, residues Asn180 and Asp183 may also be involved in electron transfer pathways between the Rieske cluster and iron-binding site, based on the close proximity of both residues to the iron center and Rieske cluster.63 It is proposed that in the likely dimeric PrnD complex, electrons from the Rieske cluster of one subunit are used to reduce the catalytic iron of the other subunit.63  15  1.2.4.2 Non-heme diiron oxidases: AurF  Figure 1.11 Proposed mechanism for the oxidation of p-aminobenzoic acid by AurF. Paths A and B were proposed based on initial experimental work from Hertweck66,68 and Zhao,69,70 respectively. Path C was proposed based on later experiments by Bollinger.71  The molecule p-nitrobenzoic acid (40) is a precursor required for the biosynthesis for several polyketides including obafluorin (3) and aureothin (4) (Figure 1.1).64,65 AurF was first linked to the production of 40 by He and Hertweck, who reported that deletion of aurF, a gene encoding a putative iron-dependent enzyme abolished production of auerothin (4) in the native aureothin producer S. thioluteus.64 This study also determined that production of aureothin (4) by S. thioluteus was contingent on the presence of 40 in the culture medium.64 Hertweck later reported that whole cell extracts of S. lividans heterologously expressing aurF could catalyze the oxidation of p-aminobenzoic acid (41) to 40 via a p-hydroxylaminobenzoic acid (42) intermediate (Figure 1.11).66 Furthermore, AurF was found to exhibit strict regio- and chemoselectivity for aromatic amines situated in para to a carboxylic acid.67 Hertweck also reported that AurF could bind both Fe and Mn, with an apparent preference for Mn over Fe, and the assembly of a complete bimanganese cluster was confirmed with ICP-MS.67,68 Both the dimanganese and diiron versions of AurF were fully catalytically competent. 16   Figure 1.12 Di-iron active site of AurF complexed with its product 4-nitrobenzoic acid.  In this AurF structure (PDB: 3CHT),69 the metal coordinating residues are depicted as green sticks, p-nitrobenzoic acid (40) as purple sticks, and the Fe-O-Fe complex is depicted as spheres.  In contrast to Hertweck, Zhao and coworkers demonstrated that Fe(NH4)2(SO4)2 could effectively reconstitute the AurF diiron center, and asserted that iron is preferentially bound by AurF.69 In another breakthrough, they were also able to fully reconstitute the activity of AurF in vitro through the use of PMS and ascorbate.69 Through use of this new in vitro reconstitution system, Zhao and coworkers detected 42 as did Hertweck, but also detected an additional compound, p-nitrosobenzoic acid (43) (Figure 1.11). Together, all the results supported a mechanism involving three iterative two-electron oxidations,69 a process analogous to the reactions catalyzed by FAD-dependent nitrosynthases (Chapter 1.2.3). The Zhao group was also able to obtain crystal structures of the oxidized form of AurF with both an atom of oxygen and 40 bound to the diiron center (Figure 1.12, Figure 1.13a).69 17   Figure 1.13 Selected proposed diiron sites of arylamine oxygenases.  a) Oxo-structure of diiron center in the crystal structure of AurF, and geometry of the proposed active peroxo intermediate with p-nitrobenzoic acid (40) bound. b) Peroxo-structure of diiron center in the crystal structure of AurF, and geometry of the proposed active peroxo intermediate.  Both groups offered competing hypotheses for the mechanism of nitro formation. Hertweck proposed that p-hydroxylaminobenzoic acid (42) is oxidized to p-dihydroxylaminobenzoic acid (44), which is followed by dehydration to yield p-nitrosobenzoic acid (43) (Path A, Figure 1.11).66,68 Based on the incorporation of only one atom of 18O from 18O2 into p-nitrobenzoic acid (40), Zhao and coworkers conversely proposed that 43 is formed directly via the dehydrogenation of the intermediate 42 (Path B, Figure 1.11).69,70 Both proposals have 43 being oxidized to the final product 40. To reconcile the discrepancies between the two competing mechanistic proposals, the Bollinger group first created the oxidized peroxo-Fe2 form of AurF, which later spectroscopic analysis assigned as a protonated \u03bc-1,2-peroxo-bridged Fe(III)2 intermediate (Figure 1.13a).71,72 The AurF-peroxo intermediate was able to fully oxidize 42 to 40, without requiring any additional reductant or O2. They also found that the proposed intermediates 43 and 44 do not substantially accumulate over the course of the AurF-catalyzed reaction. 18  Bollinger proposed that the AurF mechanism involves a two-electron oxidation followed by a four-electron oxidation which proceeds through the transient intermediate 44 (Path C, Figure 1.11). Furthermore they propose that 43 which was observed by Zhao and coworkers is instead the result of nonenzymatic oxidation or disproportionation reactions.71  1.2.4.3 Non-heme diiron oxidases: CmlI  Figure 1.14 Proposed mechanism for the oxidation of CAM by CmlI.  The chloramphenicol biosynthetic enzyme CmlI is another highly studied non-heme diiron oxidase. CmlI was also identified by Zhao, who noticed that it displayed iron-binding sites similar to those in found AurF.73 Similar to AurF, the activity of CmlI was reconstituted in vitro using PMS and NADH, where it oxidized NH2-Cam (45) to chloramphenicol (5) (Figure 1.14).73 Lipscomb and coworkers found that CmlI also binds two equivalents of iron, indicating that it requires a diiron cluster like AurF.74 They also observed that oxidation of diferrous CmlI by O2 resulted in the formation of an exceptionally long-lived peroxo intermediate with a half-life of over three hours.74 Comparing the crystal structures of both reduced CmlI and the CmlI peroxo intermediate allowed the peroxo adduct to be tentatively assigned as adopting a cis \u03bc-1,2-peroxo geometry (Figure 1.13b, Figure 1.15).75 Later spectroscopic studies of the CmlI peroxo intermediate revealed that it in fact adopts a novel (\u03bc-oxo)(\u03bc-1,1-peroxo)diferric core structure (Figure 1.13b).76 19   Figure 1.15 Active sites of reduced and peroxo-bound CmlI.  Reduced CmlI (PDB: 5HYH)75 is shown at the top in yellow. Peroxo-bound CmlI (PDB: 5HYG)75 is shown at the bottom in brown. Metal coordinating residues depicted as sticks, Fe-Fe and Fe-O-O-Fe centers depicted as spheres.    The role of the CmlI-peroxo intermediate was interrogated in vitro, where it can catalyze the full oxidation of NH2-Cam (45) to chloramphenicol (5) without any additional reductants.74 At a glance, the CmlI-peroxo intermediate should only be capable of a single two-electron oxidation, therefore the fact that it catalyzed the entire six electron oxidation raised questions about CmlI-20  catalyzed oxidative mechanism. The Lipscomb group devised a series of elegant isotopic labelling studies using 16O and 18O labelled O2 to probe the mechanism of CmlI.77 These studies revealed that both oxygen atoms in the nitro of 5 originate from O2 in the CmlI peroxo-intermediate, and not water. Synthetic hydroxylamine (46) and nitroso (47) intermediates were consumed by diferric and diferrous CmlI respectively, confirming that both were on pathway intermediates. Crucially, there is little dissociation of either of these intermediates during the course of the CmlI-catalyzed reaction, ensuring the reaction proceeds rapidly with high specificity, which protects other molecules in the cell from reactive intermediates.78  Ultimately, Lipscomb proposed a mechanism where first O2 reacts with CmlI to form a highly stable peroxo intermediate, which reacts with 45 to form 46. In a key step, 47 is formed from 46, by reduction of the now diferric CmlI back to the diferrous state. Formation of another CmlI peroxo intermediate can then oxidize 47 to 5 which incorporates the second atom of oxygen. The (\u03bc-oxo)(\u03bc-1,1-peroxo) diferric intermediate generated by CmlI proves crucial for this reactivity, as the geometry allows the iron center to act as an electrophilic oxidant during hydroxylation, as well as a nucleophilic oxidant during the oxidation of 47 to 5.76  To summarize, although AurF and CmlI are similar in structure and both catalyze arylamine oxidations with the same reaction outcome, they appear to do so using different mechanisms and different reductive intermediates. Experimental data for AurF supports a two + four-electron oxidative pathway (Figure 1.11), while the experimental data for CmlI supports a pathway involving three successive two-electron oxidations (Figure 1.14).   21  1.3 Alkyne-containing bioactive compounds  Figure 1.16 Select examples of alkyne-containing bioactive molecules.  Alkyne-containing or acetylenic compounds are molecules containing at least one carbon-carbon triple bond. Just as the nitro functional group, the alkyne is found throughout synthetic chemistry. However, as Minto and Blacklock note, acetylenic and polyacetylenic natural products are also produced by countless organisms from all branches of the tree of life.79 These acetylenic and polyacetylenic compounds tend to be unstable, and decompose through oxidative, photolytic, or pH-dependent degradative pathways, making it challenging to isolate and study these molecules.79 This reactivity is harnessed in some biosynthetic pathways to produce even more complex molecules. For example sulfur readily reacts with polyynes to generate thiophenes like xanthopappin C (48), which are insecticidal (Figure 1.16).80 Those acetylenic molecules that are stable, often display highly unusual structural motifs like Petroformyne-5 (49),81 which are highly 22  cytotoxic molecules often decorated with various functional substitutions including hydroxyl groups or halogens.79  One of the most notable classes of acetylenic molecules are the enediynes. All enediynes are defined by the striking presence of a nine- or ten-membered \u2018warhead\u2019 ring which contains the diagnostic triple bond-double bond-triple bond structure motif, exemplified by dynemycin A (50) (Figure 1.16).82,83 Enediynes are remarkably cytotoxic, to the point where entire self-sacrifice proteins are created by the producing organism to prevent it from killing itself.84,85 This potency has not gone unnoticed, and today several enediynes in the form of antibody-drug conjugates are used clinically as anticancer agents.83,86 The biosynthesis of these compounds has been extensively explored.87 Comparative bioinformatics of various enediyne producers has revealed a core set of five PKSs that are common to all enediyne biosynthetic gene clusters.83,86,88 Despite the interest in enediyne biosynthesis, much is still unknown about how the nine- or ten-membered ring is created. It does appear that PKS-bound polyketides are intermediates in enediyne biosynthesis,89 but it still not known how these intermediates are converted into the nine- or ten-membered enediyne core. The lack of knowledge about how enzymes catalyze the formation of alkynes is a reoccurring theme in the biosynthetic literature.  Figure 1.17 General scheme for the CuI-catalyzed click reaction.  An absence of well understood alkyne-forming enzymes is somewhat unfortunate, given the increasing value of alkynes in chemical biology. Much of this value is due to the continued advancements in biorthogonal click chemistry, and in particular the copper-catalyzed azide alkyne (CuAAC) reaction.90,91 The CuAAC reaction is a 1,3-dipolar cycloaddition between an azide and an alkyne, which produces solely the 1,4-isomer (Figure 1.17). Today the CuAAC reaction has countless applications in chemical biology and drug discovery.91,92 Therefore, routes that can generate alkynes via in vivo biocatalytic reactions could prove immensely valuable for chemical biologists. Additionally, several important synthetic drugs including antiretroviral drugs Islatravir (51) and Efavirenz (52) (Figure 1.16), which are used to treat HIV contain alkyne moieties.93,94 23  Therefore, better understanding the known routes of alkyne biosynthesis and discovering new biological routes of alkyne biosynthesis could enable chemists access to new alkynes, which could also facilitate the synthesis of other clinically important drugs.92  1.3.1 Internal alkyne biosynthesis 1.3.1.1 Diiron-dependent acetylenases: Crep1  Figure 1.18 Crep1-catalyzed oxidation to form internal alkynes.  While the enzymatic processes that assemble more complex molecular scaffolds like the enediynes are still not well understood, there are a few instances of acetylenases and bi-functional desaturases\/acetylenases whose activities have been verified experimentally both in vivo and in 24  vitro. The search for an acetylenase began in the seeds of the plant Crepis alpina, which are rich in the unsaturated fatty acid crepenynic acid (53) (Figure 1.18).95 To identify the enzyme which produces crepenynic acid a library of C. alpina cDNA was constructed. From this cDNA library Crep1, a gene encoding a putative non-heme diiron oxidase was identified as a prime candidate. Heterologous expression of crep1 in the host Arabidopsis thaliana led to a dramatic increase in the amount of acetylenic fatty acids in present in the seeds of A. thaliana. A detailed analysis of the fatty acid content of the seeds revealed that Crep1 catalyzes the oxidation of the C12-C13 double bond of linoleic acid (54) to produce crepenynic acid (53) (Figure 1.18).95 Interestingly, Crep1 was also found to have weak oleate \u039412 desaturase activity.95,96 To probe the Crep1-catalyzed reaction in more detail, Covello and coworkers synthesized several versions of linoleic acid (54) which were deuterated at the C12 and C13 positions to investigate the mechanism of alkyne formation using kinetic isotopic effects (KIE) (Figure 1.18).97 Incubating both deuterated linoleic acid substrates with Crep1, they observed a large KIE for the C12-deuterated substrate, and a negligible KIE for the C13-deuterated substrate. This result implies that it is a stepwise reaction with abstraction of the C12 proton being the first and kinetically important step, while the formation of the second degree of unsaturation much easier to catalyze. The Covello group continued their studies of Crep1 and a identified a number of residues involved in substrate recognition and catalysis through bioinformatic analysis of the sequences of Crep1 and other diiron desaturases.96 However, additional mechanistic studies of Crep1 and Crep1-like acetylenases remain hindered by the difficulties in working with membrane proteins, resulting in the absence of a Crep1 or Crep1-like crystal structure. The substrate scopes of Crep1 homologs from other higher plants have been investigated, and it appears that in some cases both the (E) and (Z) isomers at C12-C13 are accepted as substrates.98 The paucity of data for acetylenases is in contrast to the wealth of data available for other fatty acid desaturases,99,100 efforts which have culminated with a high impact report describing the X-ray crystal structure of the transmembrane mammalian stearoyl-CoA desaturase, a pivotal enzyme in fatty acid metabolism.101 25   Figure 1.19 Various polyacetylenic compounds constructed using Crep1 homologs.  Though much remains to be discovered about the mechanism of Crep1, its discovery and characterization has allowed for the presence of Crep1 homologs to be used as an identifier for biosynthetic gene clusters that produce acetylenic compounds. For example, the Crep1 homologs ACET1a and ACET1b are found in tomato plants and are involved in the biosynthesis of falcarindiol (55),102 whereas another Crep1 homolog, CL2, is involved in the biosynthesis of the insect pheromone dihydromatricaria acid (56).103 The acetylenase pair of CayB and CayC work together to create the conjugated polyacetylenic structure found in caryoynencin (57), which is produced by the phytopathogenic bacterium Burkholderia caryophylli (Figure 1.19).104 These examples also demonstrate empirically that acetylenic and polyacetylenic compounds can be produced by all types of organisms.   26  1.3.1.2 Cytochrome P450-dependent 1,3-enyne formation: AtyI, BisI  Figure 1.20 Biosynthesis of 1,3-enyne containing compounds. a) Discussed cyclohexanoid terpenoids containing a 1,3-enyne moiety. b) Proposed biosynthetic route to the crucial intermediate eutypinic acid (60).  A common structural motif amongst acetylenic natural products is the 1,3-enyne, which can be found in fungal-derived cyclohexanoid terpenes such as asperpentyn (58) and biscognienyne B (59) (Figure 1.20a).79,105,106 Based on the similar structures of both 58 and 59, it is not surprising that the biosynthesis of both molecules proceeds through the shared intermediate eutypinic acid (60), which is subjected to additional enzymatic modifications by downstream enzymes to generate the structural diversity seen in cyclohexanoid terpenes (Figure 1.20b). A pair of cytochrome P450s, AtyI and BisI, simultaneously reported in 2020 were shown to catalyze the oxidation of p-hydroxybenzoic acid (61) derived 4-hydroxy-3-prenyl-benzoic acid (62) substrate to eutypinic acid (60) in in vivo experiments using both native fungal producers.105,106 The discovery of these enzymes will accelerate the discovery of new enyne-containing terpenoids.  27  1.3.2 Terminal alkyne biosynthesis  Figure 1.21 Selected terminal alkyne-containing natural products.  There are also many examples of natural products that contain only terminal alkynes. Some examples of terminal alkyne-containing natural products include linear compounds such as the cyanobacterium-derived jamaicamide B (63),107 and carmabin A (64),108 as well as cyclic peptides like the depsipeptide dudawalamide A (65), from Moorea producens (Figure 1.21).109 These terminal alkynes can be produced using two different biosynthetic pathways, which are reviewed below. 1.3.2.1 ACP-tethered desaturation: JamB  Figure 1.22 JamABC-catalyzed formation of the terminal alkyne moiety during the biosynthesis of jamaicamide B and related compounds. 28  Both the jamaicamide B and carmabin A biosynthetic gene clusters were noted to contain a conserved three-gene operon in the otherwise unrelated biosynthetic gene clusters, leading to the proposal that the three encoded enzymes may be responsible for forming the terminal alkyne found in both molecules.107,108,110 The Zhang group investigated this hypothesis, cloning and heterologously expressing three genes from the jamaicamide biosynthetic gene cluster, jamA, jamB, and jamC in E. coli. Functionally, these genes were predicted to encode for a fatty acyl-CoA ligase, membrane-bound non-heme iron fatty acid desaturase, and an acyl carrier protein (ACP), respectively.110 JamA had been previously shown to activate several hexanoic acids including 5-hexenoic acid (66),107 and so JamA was used to create several different JamC-fatty acid tethered complexes which were tested as substrates for JamB. Zhang and coworkers observed that JamB efficiently catalyzed the desaturation of 5-hexenoyl-JamC (67) to 5-hexynoyl-JamC (68) (Figure 1.22). The JamB-catalyzed reaction was found only to occur with JamC-tethered substrate and not with free hexenoic acid (66). Unlike many Crep1-like acetylenases which display desaturase and acetylenase activity, JamB was unable to catalyze the four-electron oxidation of hexanoyl-JamC to 5-hexynoyl-JamC (68). In addition to also being found in the carmabin A biosynthetic gene cluster, bioinformatics revealed that the jamABC operon is widely distributed amongst other bacterial genomes, suggesting that many other bacterial strains may be capable of producing acetylenic natural products.111 As a proof of concept, the function of the ttuABC operon from Teredinibacter turnerae T7091 was interrogated in a follow-up study.111 As predicted, TtuA, TtuB, and TtuC were found to display similar activities to their JamA, JamB, and JamC counterparts. The most notable difference between the two sets of enzymes was that the membrane-bound non-heme iron fatty acid desaturase TtuB was found to have a strict specificity for C10 fatty acyl moieties, and also displayed bifunctional desaturase\/acetylenase activity, which JamB lacked.111 Therefore it is likely that different JamB homologs will have different substrate specificities and could be used to generate a wide range of desaturated short chain fatty acids. The JamABC system was also shown to be capable of interfacing with other multidomain enzyme complexes from different pathways. The feature allowed the JamABC system to be used in conjunction with type III polyketide synthases (PKSs),110\u2013112 type I PKSs,113 and even an NRPS,110 allowing for combinatorial biosynthesis of new terminal-alkyne containing molecules. 29   1.3.2.2 PLP-dependent desaturation: BesB  Figure 1.23 Biosynthesis of \u03b2-ethynylserine in Streptomyces cattleya.  Until very recently, all known biosynthetic routes to acetylenic and polyacetylenic compounds required membrane-bound non-heme iron-dependent acetylenases, discussed in the previous sections, which utilize fatty acid precursors or fatty acids as substrates.79 One of the rare exceptions to this phenomenon is the terminal alkyne-containing non-proteinogenic amino acid \u03b2-ethynylserine (69) produced by the soil bacterium Streptomyces cattleya.114 Michelle Chang\u2019s group observed that gene deletions for three non-essential fatty acid desaturase genes found in the genome of S. cattleya did not abolish production of 69.115 These results showed that the terminal alkyne of 69 was not generated using a Crep1-like acetylenase. Comparative bioinformatics coupled with further gene deletion studies eventually established that 69 is produced by a small six gene cluster, besA-F which lacked any acetylenase genes.115 Metabolomics and in vitro biochemical characterizations of the encoded enzymes were employed to deduce the biosynthetic route to 69, finding that 69 is synthesized from L-lysine (70) in a series of several remarkable transformations (Figure 1.23). First, 70 is converted to 4-chlorolysine (71) by BesD, a member of a group of iron, \u03b1-ketoglutarate dependent radical halogenases which catalyze mono- and di-chlorinations on amino acid substrates.116 The 71 30  produced by BesD is the substrate for another unusual non-heme iron-dependent enzyme, BesC, which catalyzes the oxidation of 71 to 4-chloroallylglycine (72), producing ammonia and formaldehyde as reaction byproducts. This reaction sets the stage for alkyne formation via elimination of a halide, as both 72 and the brominated vinylglycine 4-bromoallylglycine (73) were found to be substrates of the PLP-dependent lyase BesB, which produces L-propargylglycine (74) from both substrates. It was found that when this reaction was carried out in D2O, BesB incorporated two atoms of deuterium into 74, a result that was consistent with a typical cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like PLP-dependent reaction mechanism. Then 74 produced by BesB is converted into a glutamate-74 dipeptide (75) by BesA, and it is 75 which is hydroxylated by the Fe-, \u03b1-ketoglutarate-dependent enzyme BesE (76). The final 69 product is proposed to be the product of hydrolysis of the dipeptide 76. Unlike the previously reviewed alkynases which utilize metal cofactors to catalyze alkyne bond formation, in this pathway, a PLP-dependent enzyme is the critical alkynase. No other PLP-dependent enzyme had been previously linked to alkyne formation, and it remained unknown how the PLP-dependent enzyme BesB catalyzed this type of desaturation.  31  1.4 PLP-dependent enzymes 1.4.1 Introduction  Figure 1.24 Schiff base formation in PLP-dependent enzymes.  Enzymes are the tools by which nature constructs the wide array of complexity and structural diversity found in natural products. Many of these enzymes require additional cofactors to catalyze these reactions, and examples include FAD, heme, or iron (with specific examples reviewed in Chapters 1.2.3 and 1.2.4 respectively). One of the most common and versatile enzymatic cofactors is pyridoxal-5\u02b9-phosphate (PLP) (Figure 1.24).117 PLP-dependent enzymes are a vast protein family catalyzing a plethora of interesting and unique reactions, which in total comprise over 4% of all known enzymatic activites.118\u2013121 To utilize PLP, all PLP-dependent enzymes covalently bind to PLP using a conserved active site lysine residue (Figure 1.24). The terminal N\u03b5-amino group of the lysine sidechain reacts with the aldehyde of the PLP cofactor, forming a covalent PLP-enzyme Schiff base adduct. This is the holo-form of the PLP-dependent 32  enzyme, and in PLP-dependent enzymes is also commonly called the internal aldimine (IA). It is the IA imine that is most critical for reacting with enzyme\u2019s substrates and initiating catalysis.  1.4.2 Reactions catalyzed by PLP-dependent enzymes  Figure 1.25 General reaction scheme for PLP-dependent enzymes.  Many substrates of PLP-dependent enzymes are amine-containing molecules, and most often these molecules are amino acids. Importantly, all the different reactions catalyzed by PLP-dependent enzymes begin with the same transimination reaction. Here the neutral amine of the amino acid substrate reacts with the PLP-enzyme Schiff base, forming a geminal di-amine 33  intermediate. Transfer of a proton from the amino group of the substrate to the N\u03b5-amino group of the lysine sidechain catalyzes the release of the lysine sidechain from PLP. This process is coupled with the formation of a new Schiff base between the nitrogen of the amino acid substrate and PLP. As PLP is no longer covalently bound to the enzyme, this species is commonly called the external aldimine (EA) (Figure 1.25). The external aldimine (EA) is the common intermediate from which various PLP-dependent reaction pathways begin to diverge. Different PLP-dependent enzymes can catalyze the cleavage of all three bonds originating from the C\u03b1 carbon of the amino acid substrate (Figure 1.25). One reaction pathway involves the loss of CO2, which is catalyzed by PLP-dependent decarboxylases. A second series of reaction pathways can be initiated by the deprotonation of the external aldimine, and is the first step catalyzed by enzymes such as PLP-dependent transaminases and PLP-dependent racemases. Lastly, the side chain (R-group) can also be cleaved from the amino acid. Here the type of PLP-catalyzed reaction often depends on the chemical nature of the R-group that is lost and can include reactions such as \u03b2-eliminations, \u03b3-eliminations, and retro-aldol or retro-Claisen-like reactions. In all these reactions, the PLP serves as an \u2018electron sink\u2019, and the electron density from the broken bond can distributed into the conjugated \u03c0-system and positively charged pyridinium group of PLP, where it is stabilized by resonance. The resulting species is referred to as a quinonoid (Q) intermediate (Figure 1.25). All three of the aforementioned pathways produce different quinonoid intermediates based on the functional group that was previously lost.  1.4.3 Dunathan\u2019s stereoelectronic hypothesis  34  Figure 1.26 Depiction the empty \u03c0-orbitals of the PLP cofactor in Dunathan\u2019s stereoelectronic hypothesis. In this example the \u03c3-bond of the R1 group is aligned with the empty \u03c0-orbitals of the PLP cofactor and is the bond that will be broken.  As all PLP-dependent enzymes use similar mechanisms to initiate the reaction, the question of how PLP-dependent enzymes dictate reaction specificity arises. Dunathan was the first to suggest that the reaction specificity of PLP-dependent enzymes is dictated by the orientation of substrate \u03c3-bonds with respect to the \u03c0-orbital system of PLP, a concept which is referred to today as Dunathan\u2019s stereoelectronic hypothesis.122 In this hypothesis, PLP-dependent enzymes can favor certain reaction outcomes by placing the bond to be broken in the same plane as the adjacent \u03c0-orbital system, significantly weakening the desired \u03c3-bond (Figure 1.26). Dunathan proposed that PLP-dependent enzymes create this reaction specificity by careful placement of a positively charged amino acid sidechain (such as an arginine) positioned near the negatively charged carboxylic acid group, which would prevent free rotation of the substrate. Thus, precise organization of the ligand binding residues properly orient the substrate amino acid for a specific reaction and helps to explain why PLP-dependent enzymes are able to catalyze many different reactions from the same substrate using the same PLP cofactor.   1.4.4 PLP-dependent enzyme classification Table 1.1 Summary of PLP-dependent enzyme classification and activities. Fold Type Major activities I Aminotransferases, decarboxylases, lyases II Tryptophan synthases, dehydratases III Racemases, decarboxylases IV Aminotransferases V Glycogen phosphorylases VI Lysine 5,6-aminomutases VII Lysine 2,3-aminomutases  Today, extensive functional and structural studies of PLP-dependent enzymes support the ideas presented in Dunathan\u2019s stereoelectronic hypothesis and show that different substrates adopt 35  different orientations when binding with PLP.120 The countless studies of PLP-dependent enzymes have allowed enzymologists to create a detailed classification system for these enzymes. Despite the 1000\u2019s of different reactions catalyzed by PLP-dependent enzymes, these enzymes adopt a limited number of common overall structural folds, and are therefore classified based on what type of fold they adopt. Grishin was the first to systematically investigate the structural biology of the entire PLP-dependent enzyme family, classifying them into five main families.123 Percudani and Peracchi later proposed adding two new smaller families to give a total of seven PLP-dependent enzyme classes, which is the number most commonly referenced today.121 Furthermore, each individual fold type generally catalyzes a distinct set of reactions, and the major reactions catalyzed by each fold type are summarized in Table 1.1. The most common fold type of PLP-dependent enzymes is fold type I. This family includes various aminotransferases and lyases and are amongst the best characterized PLP-dependent enzymes. This class of PLP-dependent enzymes is reviewed in more extensive detail in Section 1.4.5. The second largest family of PLP-dependent enzymes is fold type II. This family most notably includes tryptophan synthase, which catalyzes formation of tryptophan via a \u03b2-elimination of H2O from serine, followed by replacement with an indole. Unlike in fold type I enzymes, the active sites of these enzymes are entirely composed of residues from one monomer.124 Fold type III PLP-dependent enzymes include racemases such as alanine racemase as well as a number of decarboxylases. They are defined by the presence of an \u03b1\/\u03b2 barrel fold and an additional pair of \u03b2-strands.123 Fold type IV includes several more aminotransferases like D-alanine aminotransferase and branched chain amino acid aminotransferases. These enzymes typically have a six-stranded antiparallel \u03b2-sheet and two \u03b1-helices.125 Fold type V includes glycogen phosphorylases, which are unique in that the PLP-phosphate group catalyzes proton transfer to the substrate as part of the enzymatic reaction. Finally, fold types VI and VII include lysine 5,6- and 2,3-aminomutases respectively, which also require an additional SAM cofactor and a [4Fe-4S] cluster for catalysis.121 These enzymes are structurally distinct from other PLP-dependent enzymes, and were therefore given separate enzyme families.121 The specific delineations of the PLP-dependent families have been reviewed in more extensive detail several times.118,121,124\u2013126  36  1.4.5 PLP-dependent aminotransferases  Figure 1.27 Reaction catalyzed by aspartate aminotransferase (AAT).  PLP-dependent aminotransferases are ubiquitous in biological pathways, and accordingly they have been extensively studied. In general, PLP-dependent aminotransferases catalyze the transfer of the amino group from one substrate, an amino acid, to a second substrate, an \u03b1-keto acid, overall forming a new \u03b1-keto acid and new amino acid. An example of a PLP-dependent aminotransferase reaction is shown in Figure 1.27. This reaction is performed by aspartate aminotransferase, which catalyzes the transfer of the amino group of L-aspartic acid (22) to \u03b1-ketoglutarate (77), producing both oxaloacetate (78) and L-glutamate (79) as products.127  1.4.5.1 Structure of PLP-dependent aminotransferases There is extensive structural data for the PLP-dependent aminotransferase family. A search of the current PDB database reveals that there are over 740 structural entries for transaminases (EC 2.6.1) as of March 1st, 2021. As briefly eluded to in Section 1.4.4 PLP-dependent aminotransferases are classified as fold type I PLP-dependent enzymes.118,121,123\u2013125 Fold type I PLP-dependent enzymes are usually oligomeric, and generally form either functional homodimers or homotetramers. As such, the active site of the enzyme is often located near the interface between two different subunits, and residues from both monomers are involved in PLP binding and catalysis (Figure 1.28).125 The individual monomers can be further divided into so-called small and large subunits. The large subunit is located in the N-terminal region of the protein, and is distinguished by a seven-stranded \u03b2-sheet, which is where the enzyme binds PLP.126 The small subunit is found in the C-terminal region of the protein, and folds into a small three or four-stranded \u03b2-sheet covered by helices on one side.125 Due to the location of the active site between both subunits, the assembly of the enzyme into homodimers or homotetramers is required for both enzyme stability as well as enzyme activity.128 37   Figure 1.28 Structure of aspartate aminotransferase.  Top) Structure of the aspartate aminotransferase fold type I homodimeric complex (PDB: 1X28),129 with small subunits and active site location indicated. Bottom) Structure of active site with PLP-aspartic acid (22) external aldimine and important conserved active site residues discussed in the text depicted as sticks.  Though there are minor differences between the active sites of individual aminotransferases, the active sites of various fold type I enzymes including aminotransferases have several conserved features which make them distinct from other fold types of PLP-dependent enzyme (Figure 1.28). First, there is an arginine residue (Arg386) which helps to coordinate the negatively charged carboxylate group of the amino acid substrate. On the same side of the active 38  site there is also an asparagine residue (Asn194) which coordinates with the hydroxyl group of PLP. At the base of the PLP is an aspartate residue (Asp222), and the negatively charged carboxylate group of aspartate coordinates the positively charged nitrogen of the PLP-pyridinium ring. There is often a bulky uncharged aromatic side chain located directly above the PLP (Trp140), where the two groups form stabilizing \u03c0-\u03c0 stacking interactions. Finally, residues from both side chains, often serine, tyrosine, or arginine are involved in PLP-phosphate binding (Arg266).130 In many aminotransferases including aspartate aminotransferase there is also a second positively charged residue (Arg292) which coordinates the other end of both amino acid and ketoacid substrates.129  39  1.4.5.2 PLP-dependent aminotransferase mechanism  Figure 1.29 General mechanism of PLP-dependent aminotransferases.  Due to the wealth of biochemical and structural data on PLP-dependent aminotransferases, the mechanism by which they catalyze reactions such as the one depicted in Figure 1.27 is well understood. PLP-dependent aminotransferases begin by forming an external aldimine between the substrate amino acid and PLP (Figure 1.29). With the amino acid destabilized by the PLP, the 40  amino group of the freed lysine residue is sufficiently basic to deprotonate the C\u03b1 proton of the external aldimine. This deprotonation results in the formation of a quinonoid intermediate. Now, the electron density in the PLP is transferred back up through PLP and deprotonates the now acidic lysine, and reprotonates the C4' position of PLP. The reprotonation at C4' breaks the conjugation between PLP and the amino acid substrate and forms a ketimine intermediate. The resulting imine is hydrolyzed by water, releasing the first product, an \u03b1-ketoacid. Importantly, this first half of the reaction also converts PLP into pyridoxamine phosphate (PMP). The free amino group of PMP is can now react with a second substrate, which in this reaction is a new \u03b1-ketoacid. The \u03b1-ketoacid binds, and the whole reaction essentially proceeds in reverse. The PMP and the \u03b1-ketoacid substrate react to form a second ketimine species. This ketimine is in turn deprotonated at the C4' position by the active site lysine, reforming a quinonoid intermediate. The lysine then reprotonates the second quinonoid on the same face from which it was deprotonated, which retains the stereochemistry that was present at \u03b1-carbon of the amino acid substrate in the product amino acid. Finally, the fully reformed external aldimine reacts once again with lysine, reforming the enzyme-PLP internal aldimine and releasing the second product, a new amino acid (Figure 1.29).   41  1.4.6 PLP-dependent \u03b3-synthases and \u03b3-lyases  Figure 1.30 Reactions catalyzed by PF01053 synthases use serine and homoserine derivatives.  In addition to aminotransferases, fold type I enzymes can also catalyze \u03b3-substitutions and \u03b3-eliminations of various amino acid substrates using mechanisms that are similar to those used by PLP-dependent aminotransferases. These types of \u03b3-substitutions are found in many primary metabolic pathways, such as the metabolism of L-cysteine, L-cystathionine, and L-methionine.131,132 Collectively the enzymes which are involved in this metabolic pathway are 42  designated as protein family 01053 (PF01053), as defined by the protein family online database.133 Members of PF01053 include the synthases O-acetylserine sulfhydrylase (OASS), O-acetylhomoserine sulfhydrylase (OAHS), O-succinylhomoserine sulfhydrylase (OSHS), and cystathionine \u03b3-synthase (CGS) (Figure 1.30). These synthases use activated derivatives of serine or homoserine in order to catalyze replacement reactions using H2S or cysteine in order to generate new C-S bonds. OASS catalyzes the formation of L-cysteine (80) from O-acetylserine (81), while OAHS and OSHS use H2S to catalyze the formation of L-homocysteine (82) from O-acetylhomoserine (83) and O-succinylhomoserine (84), respectively. Lastly, CGS catalyzes the formation of L-cystathionine (85) from 80 and 84.  Figure 1.31 Reactions catalyzed by PF01053 lyases.  Conversely, levels of sulfur-containing amino acids like L-cysteine (80), L-homocysteine (82), and L-cystathionine (85) are also maintained by the lyases cystathionine \u03b2-lyase (CBL), cystathionine \u03b3-lyase (CGL), which catalyze the cleavage of 85 to 82, pyruvate (15), and NH3 or 85 to 80, \u03b1-ketobutyrate (86), and NH3, respectively. Another common type of PF01053 lyase is 43  methionine \u03b3-lyase (MGL), which catalyzes the cleavage of L-methionine (87), releasing methanethiol (88), 86, and NH3 as byproducts (Figure 1.31).131,132 H2S is an important signaling molecule in cells and in order to maintain proper H2S levels the competing reactions catalyzed by the various synthases and lyases of PF01053 are tightly regulated.134  1.4.6.1 Reaction mechanism for PF01053  Figure 1.32 Crystal structure of MGL from Clostridium sporogenes.  The dimeric form of MGL is shown (PDB: 5DX5),135 demonstrating the interactions between the two monomers. The PLP-internal aldimine of each subunit is shown as sticks.  The molecular mechanisms used by PF01053 to catalyze the various reactions are also well understood. These enzymes generally exist as homotetrameric assemblies in solution, and structural studies have verified that they also adopt a fold type I structure (Figure 1.32).131,135 Despite catalyzing reactions with slightly different substrates, biochemical studies have shown that enzymes from PF01053 catalyze their respective reactions through essentially the same series of steps. To eliminate the redundancy of describing the mechanism used by each of the seven types of enzyme in this protein family, the mechanism for the reaction catalyzed by CGL will be reviewed, as it will be particularly relevant to subsequent discussions involving the reaction catalyzed by BesB. 44   Figure 1.33 Mechanism for PLP-dependent \u03b3-elimination catalyzed by CGL.  The first three steps in the CGL-catalyzed reaction are identical to those catalyzed by the fold type I PLP-dependent transaminases, whose mechanism was also covered in Section 1.4.5.2. Like all PLP-dependent enzymes, the CGL-catalyzed reaction begins with the formation of an external aldimine between PLP and the substrate (Figure 1.33). This step is followed by deprotonation of the C\u03b1 proton by lysine, generating quinonoid intermediate. This is in turn followed by reprotonation at C4' of the quinonoid to ketimine intermediate. It is from the ketimine 45  intermediate where the CGL catalyzed reaction diverges from the transaminase reaction pathway. In transaminases the ketimine intermediate is hydrolyzed by water, but in CGLs a basic residue instead deprotonates at C\u03b2 of the substrate, and the electron density is transferred to the positively charged imine nitrogen of PLP. The extra electron density on the nitrogen and resultant enamine are used to catalyze the release of the substituent located at the C\u03b3 position of the substrate, which in the case of CGL would be L-cysteine (80). As an aside, the intermediate resulting from elimination of the leaving group is highly electrophilic and is used by the sulfhydrylases in Figure 1.30 to catalyze the formation of new C-S bonds using H2S. However, CGL quickly converts the intermediate back into a quinonoid through removal of the C4' proton from PLP. This quinonoid with conjugation extending through the amino acid substrate is sufficiently nucleophilic to deprotonate another basic residue in the active site, which ultimately reforms the external aldimine, with the substrate now converted into an enamine. CGL then reforms the internal aldimine, releasing the product enamine. The released product can then tautomerize to the corresponding imine, which is subsequently hydrolyzed by water, producing ammonia and the corresponding \u03b1-keto acid, which for CGL would be \u03b1-ketobutyrate (86) (Figure 1.31, Figure 1.33).  1.4.6.2 Mechanism of inhibition PLP-dependent \u03b3-eliminations can also be found in secondary metabolic pathways. An example of such a reaction is the PLP-dependent \u03b3-elimination of Cl\u2212 catalyzed by BesB, which results in the formation of L-propargylglycine (74) (Figure 1.23). The incorporation of two atoms of deuterium into 74 was also observed when the reaction was carried out in D2O, which suggested that there were two enzyme-catalyzed reprotonations.115 Both results support that BesB is closely related to other members of PF01053. However, this classification is somewhat at odds with the fact that both the BesB substrate 4-chloroallylglycine (72) and BesB product 74 are well established irreversible inhibitors of several different members of the PF01053 enzyme family (Figure 1.34). 46   Figure 1.34 Proposed mechanisms of inhibition for 4-chloroallylglycine and propargylglycine.  This phenomenon of enzyme inhibition was first reported in the 1970\u2019s by Abeles, who discovered that CGL was irreversibly inhibited by 74.136,137 Soon after in 1982, Walsh and coworkers also reported that 74 irreversibly inhibited both CGS and MGL.138 In all three cases 47  reported here inhibition of the enzyme by 74 was observed to be stoichiometric, with each enzyme tetramer requiring four equivalents of 74 to become fully inactive.137,138 Systematic examination of inhibition kinetics, coupled radiochemical labeling data led both groups to propose similar mechanisms for CGL inhibition by 74, which are summarized on the right of Figure 1.34. Both groups proposed that deprotonation catalyzes an acetylenic to allenic rearrangement in the inhibitor molecule. The resulting allene is then attacked by a nucleophilic residue in the active site, triggering formation of a covalent bond between the PLP-74 adduct, irreversibly inhibiting the enzyme. Likewise, a 1989 report also describes the irreversible inhibition of MGL by 72.139 This study also observed that four equivalents of 72 were required to inactivate the one tetrameric MGL assembly. Inhibition was also accompanied by the stoichiometric release of Cl\u2212 from 72.139 The irreversible inhibition of MGL by 72 was also proposed to proceed through an allenic intermediate, and mechanistically looks very similar to the pathway used by 74 (Figure 1.34). In this route, the allenic intermediate is formed via elimination of Cl\u2212. An active site nucleophilic residue can then again react with the allenic intermediate to irreversibly the enzyme. 48   Figure 1.35 Structure of the propargylglycine-tyrosine adduct in human CGL. A) Active site of human CGL-propargylglycine (Pra) complex (PDB: 3COG).140 Selected active site residues shown as sticks. B) Refined 2Fo-Fc density of the Tyr114-Pra adduct contoured at 1.0 \u03c3. Figure is adapted from ref.140  These pioneering early studies paved the way for clinical use of L-propargylglycine (74), and today 74 finds use as a selective inhibitor of CGL.141 CGL, along with cystathionine-\u03b2-synthase (CBS) are responsible for producing most of the H2S in mammalian cells. Therefore, inhibition of CGL can be useful in alleviating disease states induced by improper levels of the signaling molecule H2S. As inhibition of CGL by 74 is a pharmacologically relevant interaction, Silvaraman et al. set out to determine the precise molecular mechanism of enzyme inhibition by 74 by obtaining a crystal structure of human CGL in complex with 74 (Figure 1.35).140 In the resultant crystal structure of the CGL-74 complex they observed additional electron density stemming from an active site tyrosine residue in located above the PLP-pyridinium ring. The available electron density perfectly fit the modelled tyrosine-propargylglycine adduct (Figure 49  1.35). This result validates the earlier mechanistic proposals put forth by several groups by unambiguously identifying the active site nucleophile as a tyrosine residue. Although the PLP-internal aldimine was regenerated, the covalent adduct formed by 74 (and 72 by extension) irreversibly inhibits CGL as well as other members of PF01053 Therefore, as BesB is not inhibited by either 72 or 74 it is a unique member of the protein family, and may have unique features which allow it to catalyze the formation of 74. Studies of BesB will be the focus of Chapter 4 in this thesis.  1.5 O2- and PLP-dependent enzymes Advances in genomic sequencing technologies have facilitated the growth of immense databases of sequenced DNA which can be mined with bioinformatics to look for novel natural products and corresponding enzymes. This field is continuing to expand the catalytic repertoire of many different types of enzymes. PLP-dependent enzymes are no exception, with examples such as BesB,115 as well as the discovery of several new PLP-dependent enzymes which were initially annotated as aminotransferases, but have been shown to adopt a type I fold and have been found to use O2 as a co-substrate in their reactions. It had been known for some time that the carbanionic character of quinonoid intermediates generated during the course of some PLP-catalyzed reactions are susceptible to off-pathway or so-called \u2018paracatalytic\u2019 reactions with O2.142 Paracatalytic reactions are those not considered to be the normal physiological reaction, and occur at slower rates, often with lower yields when compared to the normal physiological reaction. For example PLP-dependent decarboxylases such as DOPA decarboxylase can also catalyze oxidative deamination.143 Mutant versions of DOPA decarboxylase can stabilize the quinonoid intermediate, which can then undergo off-pathway reactions with O2. 50   Figure 1.36 Reactions catalyzed by O2- and PLP-dependent oxidative enzymes discussed in the text.  The first PLP-dependent enzyme that was found to use O2 as a genuine co-substrate was plant phenylacetaldehyde synthase (PAAS).144 PAAS displays significant sequence similarity to other PLP-dependent decarboxylases but uses O2 and PLP to catalyzes an oxidative deamination of L-phenylalanine (89) to phenylacetaldehyde (90), releasing CO2, H2O2, and NH3 as byproducts (Figure 1.36). Other more recently described PLP-dependent enzymes that catalyze similar oxidative deaminations include Cap15, which catalyzes the conversion of 5\u2032-C-glycyluridine (91) to uridine-5\u2032-carboxamide (92),145 and the lincosamide biosynthetic enzyme CcbF, which catalyzes 51  the oxidative deamination of the cysteine S-conjugate intermediate (93) to the corresponding aldehyde (94) (Figure 1.36).146Again, both Cap15 and CcbF-catalyzed reactions also release CO2, H2O2, and NH3 as byproducts. Most recently, the remarkable enzyme CuaB was shown to combine L-alanine (95) and the PCP-tethered fatty acid intermediate (96) via a Claisen condensation coupled with an O2-dependent oxidation to yield 97.147 As these examples illustrate, it appears that many PLP-dependent enzymes have evolved to use O2 as a substrate in order to catalyze challenging oxidative reactions. These examples, along with several more reactions catalyzed by other O2- and PLP-dependent enzymes have been recently reviewed.148,149  1.5.1 O2- and PLP-dependent L-arginine oxidases 1.5.1.1 Indolmycin biosynthesis: Ind4  Figure 1.37 Indolmycin biosynthesis in Streptomyces griseus. Biosynthetic scheme adapted from Du et al.150 The Ryan lab\u2019s interest in O2- and PLP-dependent enzymes stems from earlier work studying the biosynthesis of the antibiotic indolmycin (98).150 Previous group members characterized the enzymes responsible for the biosynthesis of 98 in the soil bacterium Streptomyces griseus.150 In this biosynthetic pathway 98 is produced using the amino acids L-arginine (13) and L-tryptophan (19) (Figure 1.37). In one branch of the pathway, 19 is converted into indolmycenic acid (99) by Ind1, Ind2, and an endogenous aminotransferase. While in the second branch of the pathway, 13 is converted into D-dehydroarginine (100) by Ind4 and Ind5. The two intermediates 99 and 100 are linked together by Ind3 using ATP, and the Ind5\/Ind6 complex directs the reaction 52  towards the formation of demethylindolmycin (101). In the final step, the SAM-dependent methyltransferase Ind7 methylates 101 to produce 98.150,151  Figure 1.38 Summary of on-pathway Ind4- and Ind5-catalyzed reaction pathways.   Several of the intriguing enzyme-catalyzed reactions found in indolmycin biosynthesis warranted further investigation. One question that remained unanswered was how the PLP-dependent enzyme Ind4 could catalyze an oxidation of unactivated C-H bonds. Further analysis of the Ind4-catalyzed reaction with 13 revealed that the answer to this question was O2. The initial products of the Ind4-catalyzed reaction are 4,5-didehydroarginine (102), the unstable arginine enamine 103, as well as H2O2 (Figure 1.38).152 Stoichiometric analysis revealed that these products are produced in ~1.7:1 ratio respectively, and consumption of 13 is coupled with the stoichiometric release of H2O2, indicating that no oxygen is incorporated into any reaction intermediates.152 Furthermore, less than one equivalent of 13 was consumed for every equivalent of O2 consumed, which suggested that the reaction proceeds through two oxidative pathways, with the more highly oxidized 102 requiring two equivalents of O2 to synthesize. The Ind4-catalyzed reaction products 102 and 103 will be spontaneously hydrolyzed under aqueous conditions to the unproductive products 4,5-dehydro-2-ketoarginine (104) and 2-ketoarginine (105) respectively. To prevent hydrolysis of the imine products from occurring, both 102 and 103 are intercepted by 53  the NAD(P)H-dependent imine reductase Ind5, and reduced to the corresponding D-enantiomers, D-dehydroarginine (100) and D-arginine (106) (Figure 1.38).150,153 The stereochemical inversion to 106 instead of 13 catalyzed by Ind5 prevents unproductive cycling of 13 by Ind4.   1.5.1.2 Enduracididine biosynthesis: MppP  Figure 1.39 Structure of enduracidin (left) and biosynthetic pathway to L-enduracididine (right).  Another non-proteinogenic amino acid L-enduracididine (107) is found in a few peptidic natural products, including the non-ribosomal peptide enduracidin (108) (Figure 1.39).154,155 Just like indolmycin (98), this unusual peptide is partially derived from L-arginine (13),156 but only recently have the enzymes responsible for synthesizing 107 been characterized. The first step in the biosynthesis of 107 is catalyzed by the PLP-dependent enzyme MppP, which catalyzes the 54  oxidation of 13 to 4-hydroxy-2-ketoarginine (14).157 Based on structural studies, it was proposed that MppR then catalyzes the dehydration and cyclization of 14 to form 2-ketoenduracididine (109).158 Lastly, it was proposed that an additional adjacent aminotransferase, MppQ, catalyzes the conversion of 109 to 107 (Figure 1.39). In some other biosynthetic pathways, 107 undergoes an additional \u03b2-hydroxylation, which is catalyzed by the Fe, \u03b1-ketoglutarate-dependent oxidase MppO.159  Figure 1.40 Overall structure (top) and active site (bottom) of MppP.  Top) Overall structure of MppP, displaying a typical type I dimeric fold (PDB: 5DJ3).157 Bottom) MppP active site, with conserved active site residues shown as sticks and the PLP-D-Arg (106) external aldimine shown as orange sticks. 55  Additionally, the MppP-catalyzed oxidation of 13 to 4-hydroxy-2-ketoarginine (14) was also observed to require O2.157 Furthermore, MppP also produces 105, and produces 14 and 105 at a ~1.7:1 ratio.157 This ratio happens to be same ratio as the products of the Ind4-catalyzed reaction with 13.152 Therefore it seems likely that both MppP and Ind4 utilize similar oxidative mechanisms to catalyze their respective reactions. The Silvaggi group was also able to obtain crystal structures of MppP complexed with 106 (Figure 1.40).157 However, 106 is not a substrate of MppP, and the first 23 residues of the N-terminus were absent in the crystal structures, leaving many questions about the structure of these kinds of enzymes. A later crystal structure of MppP complexed with 14 revealed that MppP catalyzes the formation of the (S)-enantiomer of 14 (Figure 1.39, Figure 1.41).160   Figure 1.41 Active site of the MppP (S)-4-hydroxy-2-ketoarginine complex. In this structure of MppP (PDB: 6C9B)160 conserved active site residues are shown as light blue or yellow sticks with (S)-4-hydroxy-2-ketoarginine (14) shown as orange sticks.  1.6 Thesis goals The central theme of this thesis is the discovery of characterization of new enzymes with rare or novel functions. Various bioinformatic programs can be used to identify genes of interest, and the encoded enzymes are produced through heterologous expression of these genes, and the 56  purified proteins can be used for in vitro biochemical studies and X-ray crystallographic analysis. As each chapter has distinct goals based on the enzyme(s) being characterized, the general goal(s) for each chapter will be discussed individually below. Ind4 and MppP are two recently described O2- and PLP-dependent enzymes which catalyze challenging oxidations of unactivated carbons on L-arginine substrates.152,157 These two enzymes are encoded by genes in two different biosynthetic gene clusters. Chapter 2 describes my search for homologous O2- and PLP-dependent enzymes encoded by other biosynthetic gene clusters. My aim was to discover and characterize novel arginine oxidases in order to gain additional insights into the mechanisms of O2 activation and catalysis used by O2- and PLP-dependent arginine oxidases. Furthermore, a search for these types of enzymes may lead to undescribed biosynthetic pathways. The goals in this chapter were to first characterize the activity of a novel arginine oxidase, RohP, and obtain complete X-ray crystal structures of RohP in complex with both its physiological substrates and enzymatic products in order to provide insight into the mechanisms of O2 activation and catalysis used by O2- and PLP-dependent arginine oxidases. Having characterized RohP as an O2- and PLP-dependent arginine hydroxylase in Chapter 2, Chapter 3 will focus on the characterization of the additional enzymes encoded by the cryptic five gene cluster containing rohP. Once again, I will employ a combination of bioinformatics, and in vitro biochemical studies to determine what the function of each enzyme is. By doing so I aimed to determine the final product of an unknown biosynthetic gene cluster and to continue to expand the biosynthetic pathways that utilize O2- and PLP-dependent arginine oxidases. Chapter 4 aims to characterize another unusual PLP-dependent enzyme, BesB. BesB catalyzes the \u03b3-elimination of Cl\u2212 from 4-chloroallylglycine (72) resulting in the formation of the alkyne-containing amino acid L-propargylglycine (74), which is a completely novel reaction for PLP-dependent enzymes.115 However, activity was only reported in vivo, due to the poor expression of BesB in E. coli. This was intriguing discovery on many levels, and I wanted to understand how BesB catalyzes this reaction as both 72 and 74 are known inhibitors of other PLP-dependent lyases. The lack of purified BesB has prohibited any more detailed biochemical or structural characterization of this enzyme. Through the use of novel protein constructs and different heterologous expression systems I aim to obtain soluble BesB for in vitro biochemical 57  studies, as well as obtain a crystal structure of BesB. These efforts provided the first mechanistic and structural characterization of an alkyne-forming enzyme in any biosynthetic pathway. This information could prove to be extremely useful for developing biocatalytic routes to alkynes. 58  Chapter 2: Identification and characterization of the O2- and PLP-dependent arginine hydroxylase RohP 2.1 Introduction Activation of carbon-hydrogen (C-H) bonds on sp3-hybridized carbons is an enduring challenge in chemical synthesis.161,162 Metalloenzymes, nature\u2019s solution to this challenge, have evolved the incredible capacity to catalyze the stereospecific functionalization of C-H bonds under mild conditions. Common cofactors for such enzyme-catalyzed reactions are heme and non-heme irons. For example, in the case of hydroxylation of L-arginine (13), the Fe(II)-, \u03b1-ketoglutarate-dependent enzymes VioC, YcfD, and OrfP catalyze hydroxylations of the sp3-hybridized carbons of the L-arginine side chain.163\u2013165 By contrast, organic cofactors such as PLP are not normally expected to catalyze installation of hydroxyl groups onto unactivated sp3-hybridized carbons.  As discussed in the introduction, PLP is one of the most widely used cofactors for enzymatic manipulation of amino acid substrates.166 In the course of catalysis, many PLP-dependent enzymes form a carbanionic intermediate, called a quinonoid, which can result from deprotonation or decarboxylation of an amino acid-PLP adduct (Figure 1.25).120 Normally, this quinonoid intermediate is sequestered in the active site of the enzyme and thus protected from reaction with O2. However, in several cases this quinonoid intermediate can be subject to off-pathway reactions with electrophiles, including O2.142 The O2-dependent oxidative deamination catalyzed by DOPA decarboxylase167 and the O2-dependent oxidative decarboxylation catalyzed by ornithine decarboxylase168 are two well-characterized examples of such paracatalytic enzymatic activities. More recently, the diversity of PLP-dependent enzymes has expanded with the discovery of enzymes that use O2 as a co-substrate to perform their primary enzymatic activities. For example, both petunia phenylacetaldehyde synthase and the celesticetin biosynthetic enzyme CcbF catalyze O2- and PLP-dependent decarboxylation-coupled oxidative deaminations,144,146 and Cap15 is an O2- and PLP-dependent monooxygenase-decarboxylase that generates uridine-5'-carboxamide on the pathway to capuramycin (Figure 1.36).145 The scope of such O2- and PLP-dependent reactions is not limited to oxidative decarboxylations but also includes functionalization of unactivated C-H bonds. In 1977, Eguchi and coworkers purified an active fraction from Streptomyces eurocidicus SF 506 that could convert 59  L-arginine (13) to 4-hydroxy-2-ketoarginine (14) and 2-ketoarginine (105) using PLP and O2 (Figure 1.4).27 This work suggested that a PLP-dependent enzyme could use O2 to catalyze a reaction that resulted in hydroxylation of an unactivated carbon center. However, the identity of the enzyme, the configuration of its product, and its likely mechanism involved remained unknown. More recently, Silvaggi and coworkers revealed that the fold type I PLP-dependent enzyme MppP is responsible for production of 105 and 14 (configuration unknown), from 13 on the pathway to the non-proteinogenic amino acid L-enduracididine (107), a precursor to compounds such as enduracidin (108) (Figure 1.39).157 They demonstrated that MppP consumes O2, suggesting that this enzyme has evolved the ability to activate O2 for a challenging hydroxylation reaction. They also solved two crystal structures of MppP. One of these structures shows PLP bound as an internal aldimine to Lys221, and the second structure has D-arginine (106), which is not a substrate, bound in an external aldimine with PLP (Figure 1.40). However, the N-terminus is disordered and the active site is exposed to solvent in all but one chain of the holo-enzyme, where it forms a small \u03b1-helix that points away from the active site and into the solvent. The lack of a closed active site and the lack of structures with native substrates bound in the active site raise questions about how catalysis might proceed past formation of the external aldimine. Furthermore, the specific role of oxygen in this reaction remains unknown. Previously the Ryan group had discovered a predicted fold type I PLP-dependent enzyme Ind4 from the indolmycin biosynthetic pathway.150,152 This enzyme, like MppP, catalyzes the two-electron oxidation of 13 to give 105. But, unlike MppP, Ind4 also catalyzes the four-electron oxidation of 13 to give 4,5-didehydroarginine (102) (Figure 1.38). In the imine form, 102 can be intercepted by the downstream enzyme Ind5 for a stereospecific NADH-dependent reduction to give D-dehydroarginine (100) (Figure 1.38). In the absence of Ind5, the imine tautomer is hydrolyzed to compound 104. The hydrolyzed compounds can react with H2O2 released from the reaction in a further degradative pathway.152,157  As this enzymology became apparent, the question of whether there are enzymes like Ind4 that are more widely distributed in natural product pathways arose, leading to the identification of a novel putative biosynthetic gene cluster encoding an Ind4 homolog. Here, I report the discovery and characterization of the new Ind4 homolog named RohP, and I demonstrate that it catalyzes an MppP-like reaction. To provide a mechanistic framework for understanding how RohP catalyzes 60  a hydroxylation \u2013 instead of an Ind4-like oxidation \u2013 I carried out extensive mass spectral, kinetic, stoichiometric, and X-ray crystallographic experiments. This work suggests that both the RohP hydroxylase and the Ind4 oxidase catalyze challenging reactions of 13 through similar oxidative pathways. The described crystallographic and stoichiometric studies support a mechanism where both the oxidase and hydroxylase carry out an O2-dependent four-electron oxidation on 13, giving a PLP-tethered didehydroarginine intermediate. Whereas Ind4 then releases 102 as a product, RohP additionally catalyzes a stereospecific alkene hydration on the PLP-tethered didehydroarginine, using water to give the hydroxylated product 14. 61  2.2 Materials and Methods 2.2.1 General methods Primers were purchased from Integrated DNA Technologies. DNA sequencing was carried out by NAPS Unit DNA Sequencing Facility (The University of British Columbia). Reagents were purchased from Anatrace, Bio Basic Inc., Gold Biotechnology, Hampton Research, New England Biolabs (NEB), Thermo Fisher Scientific Canada, and VWR International. 2.2.2 Cloning and expression of RohP The gene rohP, which encoded RohP as originally annotated, was amplified from genomic DNA of S. cattleya (NRRL 8057, DSM 46488) by the polymerase chain reaction using the primers RohP-F 5\uf0a2-AGCAGCCATATGAAGTACAACCTCGCCGACGCC-3\uf0a2 (NdeI site underlined) and RohP-R 5\uf0a2-AGTAGTCTCGAGTCAGCGGCCATGGCGGTC-3\uf0a2 (XhoI site underlined). The re-annotated rohP, which coded for the active form of RohP with an intact N-terminus, was similarly amplified using the primers RohP-F-2 5\uf0a2-AGCAGCCATATGCACCCGCAAG-CGACC-3\uf0a2 and RohP-R. The products were digested with NdeI and XhoI and ligated into similarly digested plasmid pET28a (EMD Millipore) to produce a N-terminal His6-tagged protein. Site-directed mutagenesis of RohP was performed using the Q5 Site-Directed Mutagenesis Kit (New England Biolabs), using the primers H34A-F 5\uf0a2-CGCCGACGCCGCCACCCACCAG-3\uf0a2 and H34A-R 5\uf0a2-GAGGTTGTACTTCATGGTCA-GCGCCTGGATCTCGTG-3\uf0a2. The nucleotide sequence of the cloned rohP was confirmed by sequencing, and the plasmid was transformed into E. coli BL21 (DE3) cells for protein production.  E. coli cell cultures were grown at 37 \u00b0C in LB medium containing 50 \u03bcg\/mL kanamycin to an OD600 of 0.8\u20131.0, and then cooled to 16 \u00b0C. Protein expression was induced with 0.1 mM IPTG, and the cultures were grown for an additional 16 h at 16 \u00b0C. Cells were harvested by centrifugation and frozen at \u201320 \u00b0C until protein purification. 2.2.3 Purification of RohP For purification, the cells were thawed and re-suspended in 20 mM HEPES, 50 mM NaCl (pH 7.5) buffer and sonicated to lyse the cells. The lysate was then centrifuged at 40,500 g for 45 min to remove insoluble material. The lysate was then applied to a column containing ~1 mL Chelating Sepharose\u2122 Fast Flow resin (GE Lifesciences) charged with NiSO4\u00b76H2O. The lysate 62  was gravity filtered through the resin, and the resin was washed with 20 mL of 20 mM HEPES, 50 mM NaCl (pH 7.5) buffer. The column was then washed with 5 mL portions of 20 mM HEPES, 50 mM NaCl (pH 7.5) containing 5, 10, 20, 50, 100, 200, 300, and 500 mM imidazole in a stepwise manner to elute the protein. The fractions containing RohP could be identified by the yellow color, and most of the protein eluted in the fractions that contained 200 and 300 mM imidazole. The RohP containing fractions were combined and concentrated to a volume of ~5 mL using an Amicon Ultra Centrifugal filter (10,000 molecular weight cut-off, EMD-Millipore). The concentrated fraction was loaded into a HiLoadTM Superdex 16\/600 Superdex column (GE Amersham Biosciences) pre-equilibrated with 20 mM HEPES, 50 mM NaCl (pH 7.5) buffer. The protein was eluted using a flow rate of 1 mL\/min. The fractions containing purified RohP were pooled to a final concentration of ~50 \u00b5M and dialyzed against HEPES buffer containing 250 \u00b5M PLP for 16 h. Excess PLP was removed by further dialysis with 20 mM HEPES, 50 mM NaCl (pH 7.5). Finally, RohP was concentrated to ~20 mg\/mL by centrifugation with an ultra-centrifugal filter (10,000 molecular weight cut-off, EMD-Millipore). At this concentration, the purified RohP remained stable at 4 \u00b0C and was stored in the dark to prevent PLP degradation. 2.2.4 In vitro biochemical assays and product analysis Initial in vitro reactions (100 \u00b5L) contained 30 \u00b5M RohP and 1 mM L-arginine (13), in 20 mM Tris, 50 mM NaCl (pH 7.5) buffer and proceeded for 16 h at room temperature. HPLC analysis of the reaction was carried out after pre-column derivatization with DNS-Cl. A reaction mixture of 50 \u00b5L was treated with 80 mM Li2CO3, 70 \u00b5L of ACN, and 30 \u00b5L of 5 mM DNS-Cl dissolved in ACN. The reaction was carried out at room temperature for 1 h and then 40 \u00b5L of 2% ethylamine was added to the mixture to react with excess DNS-Cl. The mixture was centrifuged, and 20 \u00b5L of the supernatant was subjected to HPLC analysis. HPLC analysis was carried out on a 1260 HPLC apparatus (Agilent), using a Luna C18(2), 5 \u00b5m, 4.6 mm ID \u00d7 250 mm column (Phenomenex). Elution was performed at 0.5 mL\/min using a mobile-phase consisting of a linear gradient of water and ACN ((v\/v): 95:5 to 50:50, 0 to 15 min; 0:100, 15 to 22 min), with both solvents containing 0.05% (v\/v) TFA. DNS-Arg was detected at a wavelength of 330 nm.  In vitro assays for ESI-MS analysis contained 10 \u00b5M RohP and 1 mM 13, in 20 mM HEPES, 50 mM NaCl (pH 7.5) buffer in a 100 \u00b5L reaction mixture. These reactions were carried out for 4 h at room temperature and then quenched with an equal volume of MeOH. Precipitated 63  protein was removed by centrifugation and 10 \u00b5L of the supernatant was subjected to ESI-MS analysis. MS analysis was performed with a 6120 Quadrupole LC\/MS (Agilent) system equipped with a Poroshell 120, EC-C18, 2.7 \u00b5m, 4.6 mm ID \u00d7 50 mm column (Agilent), and operated in positive ion mode. 2.2.5 NMR analysis of RohP reaction products The reaction mixture (5 mL) contained 20 \u00b5M RohP, and 10 mM 13 in 20 mM sodium phosphate buffer (pH 7.2). The mixture was incubated in an unsealed 50 mL vial at 25 \u00b0C and shaken at 120 rpm for 6 h. The solvent was then evaporated overnight using a SpeedVac plus vacuum concentrator. The dried solids were re-suspended in 600 \u00b5L D2O and centrifuged to remove any residual undissolved solids prior to NMR analysis. All spectra were acquired utilizing a Bruker Avance 600 MHz spectrophotometer. 2.2.6 Steady-state kinetics for the RohP reaction Kinetic assays were performed by monitoring the consumption of O2 using a Clark-type polarographic O2 electrode (Hansatech, Pentney, UK) similar to the method described previously.152 The electrode was calibrated daily using air-saturated water and sodium hydrosulfite according to the manufacturer\u2019s instructions. The standard assay was performed in 1 mL of air-saturated 40 mM MOPS (I = 0.1 M, pH 7.2) at 25 \u00b0C containing 500 \u03bcM 13. The reaction was initiated with the addition of RohP to 1 \u03bcM. The observed rates were corrected with background O2 consumption prior to reaction initiation. The effect of pH on the rate of the RohP-catalyzed reaction was evaluated using air-saturated 20 mM buffers (I = 0.1 M) of MES (pH 6.0), PIPES (pH 6.7), MOPS (pH 7.2), HEPES (pH 7.4), HEPPS (pH 8.0), and TAPS (pH 8.5). The steady-state kinetic parameters of RohP with respect to 13 were determined at ambient oxygen levels by varying the concentration of 13 (8 \u2013 500 \u00b5M) using 1 \u00b5M and 5 \u00b5M of RohP, respectively. The steady-state kinetic parameters of RohP with respect to oxygen were measured using 17 \u2013 685 \u00b5M O2 at an 13 concentration of 500 mM. O2 concentrations were established by bubbling mixtures of O2 and N2 into the reaction chamber prior to reaction initiation. The final oxygen concentrations were standardized to the level of air-saturated buffer before the gas bubbling. Kinetic parameters were determined either by least-squares fitting of the Michaelis-Menten equation to the data using LEONORA or by Hill equation using Origin 8.1 (OriginLab corp., Northampton, MA).  64  2.2.7 Stoichiometry of the RohP reaction The production of H2O2 was evaluated by comparing the reaction rates catalyzed by 2.5 \u00b5M RohP and 0.3 mM 13 in the presence or absence of ~4000 U of catalase. H2O2 production was also measured colorimetrically by supplementing the reaction with 0.1 mg\/mL horseradish peroxidase (Type-1) and 1 mM 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid. Upon the consumption of 100 \u00b5M oxygen, the reaction was quenched with 1 volume of 10% TCA. The experimental values were compared to a calibration curve of H2O2. For quantification of the 13:O2 stoichiometry, reaction mixtures (1 mL) containing 12 \u00b5M RohP and 250 \u00b5M 13 in 40 mM MOPS (I = 0.1 M, pH 7.2) buffer were incubated at 25 \u00b0C. The mixture was equilibrated to ambient O2 levels prior to the addition of RohP. Upon addition of RohP the O2 concentration was monitored by Oxygraph and the reaction was quenched with one volume of MeOH after 100 \u00b5M of O2 was consumed. The 13 remaining after the reaction with RohP by was quantified by HPLC after pre-column derivatization with OPA\/MPA. The quenched solution was incubated with 1 volume of 0.4 M borate buffer (pH 10.2) and 1 volume of OPA\/MPA solution (1 mg\/mL OPA, 0.1% (v\/v) MPA in 0.1 M borate buffer (pH 10.2)) for 3 min before HPLC analysis. HPLC analysis was carried out on a 1260 HPLC apparatus (Agilent), using a Poroshell 120, EC-C18, 2.7 \u00b5m, 4.6 mm ID \u00d7 50 mm column (Agilent). Elution was performed at 1.0 mL\/min using a mobile-phase consisting of a linear-gradient of water and ACN ((v\/v): 98:2, 0 to 1 min; 80:20, 1 to 5 min; 50:50, 5 to 6.5 min; 0:100, 6.5 to 8 min; 98:2, 8 to 10 min), with both solvents containing 0.05% (v\/v) TFA. Derivatized 13 was detected at a wavelength of 330 nm. The experimental values were compared to a calibration curve of 13 that was done using the same conditions. The data are reported as mean values \u00b1 s.d., with n = 5. 2.2.8 Spectroscopic analysis of the RohP reaction Aerobic reaction mixtures (1 mL) were prepared in a quartz cuvette with a path length of 1 cm and contained 30 \u00b5M of enzyme and 300 \u00b5M of substrate in 20 mM HEPES, 50 mM NaCl, pH 7.5. All solutions were room temperature and air saturated with oxygen. The reaction was initiated upon addition of substrate. Spectra were recorded using a Varian Cary 100 Bio UV-Vis spectrophotometer (Agilent). 65  2.2.9 RohP crystallization Initial crystallization conditions were identified by screening 4 mg\/mL RohP against the Index HT Screen (Hampton Research) and Top96 Crystal Screen (Anatrace). Optimization of the initial crystallization conditions was carried out at room temperature using hanging drop vapor diffusion. Diffraction-quality crystals were obtained by mixing 1.5 \u03bcL of 8 mg\/mL protein and an equal volume of a crystallization solution composed of 0.2 M sodium malonate (pH 7.0), and 16\u201320% (w\/v) PEG 3350, using hanging drop vapor diffusion over a 500 \u00b5L reservoir of the crystallization solution. Large, yellow crystals appeared after approximately 7 d, though some crystals took up to 4 weeks to form under these conditions.  The structures of RohP in complex with intermediates and product were obtained by adding solutions of PLP to the ~3 \u03bcL drops containing crystals of RohP, soaking overnight, and then adding 13 solutions for timed soaks. The structure of the holo-RohP resulted from adding 1 \u03bcL of 20 mM PLP to a drop containing crystals for an overnight soak, giving a final concentration of 5 mM PLP. The structure of the first RohP quinonoid intermediate (Q1) resulted from adding 0.5 \u03bcL 20 mM PLP for an overnight soak and then adding 1 \u03bcL 100 mM 13 for 90 s, giving final concentrations of 2.2 mM PLP and 22 mM 13. The structure of the second RohP quinonoid intermediate (Q2) resulted from adding 1.0 \u03bcL 20 mM PLP for an overnight soak and then adding 1 \u03bcL 50 mM 13 for 5 min, giving final concentrations of 4 mM PLP and 10 mM 13. The RohP-4-hydroxy-2-ketoarginine structure resulted from adding 1.0 \u03bcL 20 mM PLP and 1 \u03bcL 50 mM 13 for an overnight soak, giving final concentrations of 4 mM PLP and 10 mM 13. After soaking as described, the crystals were cryoprotected using solution composed of 0.2 M sodium malonate (pH 7.0), 15% (w\/v) PEG 3350, and 20% (v\/v) ethylene glycol and flash-frozen in liquid nitrogen prior to X-ray data collection. 2.2.10 Data collection, structure determination and model refinement X-ray diffraction data were collected at the Canadian Light Source (Saskatoon, Canada), beamline 08ID-1, at a wavelength of 0.97949 \u00c5 using MX300-HE and Pilatus 6M detectors. UV-Visible spectroscopic data of RohP crystals were collected using the microspectrophotometer at beamline BL9-2 at the Stanford Synchrotron Radiation Light Source (Menlo Park, United States). All data sets were integrated using iMOSFLM,169 and scaled using AIMLESS.170 RohP 66  crystallized with two units in the asymmetric unit, forming a homodimer in the space group C2. Complete data collection statistics are listed in Table A.1. The crystal structures of each RohP complex was phased by PHASER-MR in the Phenix software package,171 using the L-arginine, \u03b3-hydroxylase SwMppP as a model (PDB: 5DJ1 chain D, 32% identity across 393 amino acid residues). The initial results from molecular replacement were input into Phenix Autobuild172 for additional model building. The Autobuild output model was subjected to several rounds of manual inspection and building in COOT,173 and refinement in phenix.refine174 using TLS refinement. Alternate conformations of side chains and molecules from the crystallization solution were added to the model where appropriate. Solvent molecules were added automatically in phenix.refine and examined manually in COOT. Refinement statistics are listed in Table A.2. Non-standard ligand restraints were generated using Phenix eLBOW.175 The second quinonoid intermediate (Q2) was initially fit to the Fo-Fc omit density using ARP\/wARP version 7.6.176 The coordinates generated were used to produce a restraint file containing the ARP\/wARP optimized geometry, and the entire structure was subjected to additional refinement in phenix.refine. The holo-RohP structure has the N-terminus through residue 26 disordered in both chains. The first RohP quinonoid intermediate (Q1) structure has the N-terminus through residue 25 disordered in both chains. The second RohP quinonoid intermediate (Q2) structure has the N-terminus through residue 13 disordered in both chains. The structure of the RohP with 14 in the active site has the N-terminus through residue 13 disordered in Chain A and through residue 14 disordered in Chain B. In all structures, residues 391\u2013393 of the C-terminus are disordered in each monomer, except in Chain B of both quinonoid structures, where residues 392\u2013393 are disordered.  2.3 Results 2.3.1 RohP is an L-arginine hydroxylase Previous work from the Ryan lab described the O2- and PLP-dependent oxidase Ind4, which generates the conjugated 4,5-didehydroarginine (102) from L-arginine (13) (Figure 1.38).152 Using BLAST-P to search for enzymes similar to Ind4 in other bacteria led to the discovery of a new enzyme (accession number: AEW92768.1) from Streptomyces cattleya NRRL 8057 with 43% 67  sequence identity to Ind4,152 40% sequence identity to the Ind4-like enzyme Pel4,152 32% sequence identity to MppP,157 and 31% sequence identity to the MppP-like enzyme EndP (Figure 2.1).157   Figure 2.1 Sequence alignment of RohP and homologous enzymes. Clustal Omega (www.ebi.ac.uk\/Tools\/msa\/clustalo) was used to generate the alignment. MppP (accession number KDR62041), EndP (accession number ABD65947), Ind4 (accession number AJT38685), and Pel4 (accession number WP_010502631), are shown. AEW92768.1 corresponds 68  to the initial sequence of RohP as annotated, while RohP corresponds to the sequence of the enzyme used for this study. Conserved residues are highlighted in black, while similar residues are boxed in bold.  This newly identified enzyme is encoded in an unknown putative biosynthetic gene cluster that has little resemblance to either the indolmycin biosynthetic gene cluster from S. griseus, or the region surrounding mppP in the mannopeptimycin biosynthetic gene cluster from S. hygroscopicus (Figure 2.2). This close sequence identity of the encoded enzyme to both Ind4 and MppP suggested that it could have either Ind4-like or MppP-like activity, or perhaps catalyze a different reaction on 13. As a first step in elucidating the product of this gene cluster, the function of this enzyme was determined. Based on the deposited sequencing data, this enzyme is annotated with a ~20 amino acid truncation in the N-terminus when compared to the other, characterized enzymes (Figure 2.1). The purified recombinant enzyme from E. coli was tested whether it is also active on 13. However, the enzyme was inactive on 13 as purified or with exogenous PLP (Figure 2.3a). After determining that the enzyme was inactive with 13, the DNA from S. cattleya upstream from the start codon of the truncated enzyme was amplified and re-sequenced using Sanger sequencing. This sequencing revealed that this upstream region was highly GC rich, and that the deposited genomic sequencing data on NCBI had incorrectly contained a sequence of five consecutive cytosines which is actually a sequence of six consecutive cytosines. With the additional cytosine, an alternate annotation of the gene would have a different start site and encode an enzyme 25 amino acids longer that aligns along the full length of MppP, Ind4, and the other characterized enzymes (Figure 2.1). 69   Figure 2.2 Comparison of biosynthetic gene clusters containing O2- and PLP-dependent arginine oxidases. Annotated function listed below each gene. The arginine oxidases rohP from S. cattleya, ind4 from S. griseus, and mppP from S. hydroscopicus are shaded. 70  The gene including the new start site was recloned into pET28a, expressed and purified from E. coli, and the enzyme activity was assayed with 13. After 16 h of incubation, the reaction was derivatized with DNS-Cl and then analyzed by HPLC. This experiment revealed that 13 was consumed in the overnight assay (Figure 2.3b), suggesting that the enzyme \u2013 like Ind4, Pel4, and MppP \u2013 reacts directly with 13 under aerobic conditions, without the need for an exogenously provided keto-acid for regeneration of the PLP cofactor or any additional cofactors. Figure 2.3 RohP-catalyzed consumption of L-arginine. A) The RohP construct lacking 25 amino acid residues of the N-terminus does not consume L-arginine (13). B) The full-length RohP construct consumes L-arginine (13). Reaction mixtures were analyzed by HPLC at 330 nm after pre-column derivatization with dansyl-chloride (DNS-Cl).  To determine the product(s) of this enzyme, ESI-MS was employed to reveal that four new ions were generated: [M+H]+ 146, [M+H]+ 162, [M+H]+ 174, and [M+H]+ 190 (Figure 2.4). Previously, both the [M+H]+ 146 and the [M+H]+ 174 ions had been observed in the reaction of 13 with Ind4. The ion at [M+H]+ 174 was previously assigned as 105, while the ion at [M+H]+ 146 was previously assigned as 4-guanidinobutyric acid (110), arising from the non-enzymatic decarboxylation of 105 with H2O2 produced in the course of the reaction.152,177,178 The [M+H]+ ion of 190 observed in this reaction could correspond to the MppP product 14 (configuration 71  unknown),27,157 which could decarboxylate in the presence of H2O2 to give 3-hydroxy-4-guanidinobutyric acid (111, configuration unknown), having an [M+H]+ ion of [M+H]+ 162.  Figure 2.4 RohP-catalyzed oxidation of L-arginine. Liquid chromatography-mass spectrometry analysis of the products of the RohP reaction with L-arginine (13). The reaction conditions are indicated above each spectrum, and representative integrated total ion chromatograms for each reaction are shown.  To determine whether these ions correspond to the previously identified MppP products, catalase was added to the reaction mixture to consume any H2O2 produced. With catalase present, only the ions at [M+H]+ 174 and [M+H]+ 190 were observed (Figure 2.4). This supports that 105 and 14 are the initial products of the enzymatic reaction that then decarboxylate in the presence of H2O2 to give 110 and 111, respectively (Figure 2.5). Furthermore, using high resolution ESI-MS 72  to analyze all four products, I was able to confirm their elemental compositions, which correspond to the formulas for the proposed products (Table 2.1). To further validate the structures of these molecules, the reaction of 13 with the S. cattleya enzyme was scaled up to provide sufficient material for NMR analysis. This reaction was carried out without catalase, and the 6 h incubation time allowed complete conversion of the initial products to the corresponding decarboxylated molecules. A combination of 1H-, 13C-, 1H-1H COSY, 1H-13C HSQC, and 1H-13C HMBC NMR were employed to characterize the products of this reaction, identifying them as 110 and 111 (Figure A.1-6). These results suggest that the enzymatic products of this reaction are 105 and 14, which non-enzymatically convert to the products observed by NMR, 110 and 111, respectively, in the presence of H2O2 (Figure 2.5). Despite the NMR and mass spectral analysis, the configuration of the hydroxyl in 111 (and by, extension, in 14) remained unknown. Because this new enzyme catalyzes hydroxylation on an arginine-derived molecule, it was named RohP to indicate its substrate (L-arginine = R), its activity (hydroxylation = OH), and its related activity to the \u201cP\u201d enzymes EndP and MppP.  Figure 2.5 Summary of Ind4- and RohP-catalyzed reaction pathways. [M + H]+ ions observed from both RohP- and Ind4-catalyzed reactions are shown in purple; those from only the RohP-catalyzed reactions are in red, and those from only the Ind4-catalyzed reaction are in blue. Molecules in gray arise from nonenzymatic reactions with H2O2.   73  Table 2.1 High-resolution ESI-MS analysis of RohP- L-arginine reaction products.  Reaction Conditions1 Experimental mass [M+H]+ Error (ppm) Chemical Formula Chemical Name Theoretical mass [M+H]+ RohP,  \u029f-Arg 146.0925 0.7 C5H12N3O2+ 4-guanidinobutyric acid 146.0924 162.0884 6.8 C5H12N3O3+ 3-hydroxy-4-guanidinobutyric acid 162.0873 174.0879 3.4 C6H12N3O3+ 2-ketoarginine 174.0873 190.0831 4.7 C6H12N3O4+ 4-hydroxy-2-ketoarginine 190.0822 RohP,  \u029f-Arg, catalase 174.0873 0 C6H12N3O3+ 2-ketoarginine 174.0873 190.0826 2.1 C6H12N3O4+ 4-hydroxy-2-ketoarginine 190.0822 1Reaction mixtures (250 \u00b5L) contained 10 \u00b5M RohP, 1 mM L-arginine (13) and 0.25 mg\/mL catalase (as indicated) in 20 mm HEPES, 50 mM NaCl (pH 7.5) and were allowed to react for 6 h at room temperature. Samples were analyzed using a Waters\/Micromass LCT TOF-MS spectrometer.   2.3.2 Kinetics and stoichiometry of the RohP-catalyzed reaction  Figure 2.6 pH-dependence of the rate of the RohP-catalyzed reaction. The following buffers were used MES (pH 6.0), PIPES (pH 6.7), MOPS (pH 7.2), HEPES (pH 7.4), HEPPS (pH 8.0), and TAPS (pH 8.5) and the rate of O2 consumption was monitored by Oxygraph. 0.10.20.30.40.50.60.70.80.91.04 6 8 10log(Vo) (\u03bcM\/min)pH74  It was hypothesized that RohP requires O2 for activity the same way Ind4 and MppP do. Accordingly, the RohP reaction with 13 was carried out in a series of buffers and the O2 consumption was monitored using an Oxygraph to develop a pH-dependent activity profile for RohP (Figure 2.6). No significant changes in the rate of oxygen consumption were detected at the tested pHs suggesting that the rate-limiting step in the reaction catalyzed RohP is not dependent on acid-base chemistry. Nonetheless it was found that RohP functioned best at a pH between 7 and 7.2. Therefore a 40 mM MOPS pH 7.2 buffer was used for subsequent kinetic studies.  Figure 2.7 Steady state consumption of O2 by RohP. Rate of oxygen consumption by 2.5 \u00b5M RohP with 300 \u00b5M L-arginine (13) in the absence and presence of 0.1 mg\/mL catalase in 40 mM MOPS (pH 7.2).  Similar experiments to monitor the effect of catalase on the RohP-catalyzed reaction were also carried out. In these experiments, O2 was consumed at a rate of 15 \u00b5M min-1 in 40 mM MOPS pH 7.2, a rate that decreased to ~8 \uf06dM min-1 in the presence of catalase (Figure 2.7). This approximate halving of the rate of O2 consumption in the presence of catalase \u2013 an enzyme that converts two molecules of H2O2 to one molecule of O2 \u2013 suggests that the RohP-catalyzed reaction consumes O2 with stoichiometric conversion of O2 to H2O2. To further confirm the stoichiometry of the reaction, 2.5 \uf06dM RohP was incubated with 300 \uf06dM 13 in the presence of 0.1 mg\/mL Horseradish Peroxidase (Type-1) and 1 mM 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) and the reaction was quenched when 100 \uf06dM of O2 was consumed. In this condition, the production of 96 \uf0b1 6 \uf06dM of H2O2 was detected, indicating stoichiometric conversion of O2 to H2O2. 75  It was also determined that 77 \uf0b1 4 \uf06dM of 13 was consumed when 100 \uf06dM of O2 was consumed. This result suggests there are two reaction pathways, as observed for Ind4. In one pathway, 13 undergoes a two-electron oxidation, with consumption of one equivalent of O2. In the second pathway, 13 undergoes a four-electron oxidation, with consumption of two equivalents of O2. The observed O2:13 stoichiometry can be rationalized if the first, less oxidizing pathway consumes twice as much 13 as the more oxidizing pathway. That is, for every 100 \u00b5M of O2 consumed in total, 50 \u00b5M is consumed by the first pathway to convert 50 \u00b5M of 13, and 50 \u00b5M is consumed by the more oxidizing pathway to convert 25 \u00b5M of 13. The first pathway leads 105, as observed for Ind4, whereas the second pathway should give a more oxidized product, such as 14.  \ud835\udc63 =\ud835\udc49\ud835\udc5a\ud835\udc4e\ud835\udc65[\ud835\udc46]\ud835\udc3e\ud835\udc5a + [\ud835\udc46] Equation 1 Michaelis \u2013 Menten equation.  \ud835\udc63 =\ud835\udc49\ud835\udc5a\ud835\udc4e\ud835\udc65[\ud835\udc46]\ud835\udc5b\ud835\udc3e\ud835\udc5a\ud835\udc5b + [\ud835\udc46]\ud835\udc5b Equation 2 Hill Equation.    Steady-state kinetic parameters for the RohP-catalyzed reaction were also determined for both 13 and O2 by fitting experimental data to the Michaelis \u2013 Menten equation (Equation 1) or the Hill Equation (Equation 2).179 In the presence of air-saturated buffer, RohP has a Km = 40 \u00b1 10 \u00b5M and a kcat = 7.8 \u00b1 0.6 min-1 for 13 (Figure 2.8a). These values are consistent with those previously reported for the related enzymes Ind4 (Km = 69 \u00b1 6 \u00b5M, kcat = 3.5 \u00b1 0.1 min-1) and MppP (Km = 50 \u00b1 8 \u00b5M and kcat = 13.2 \u00b1 0.6 min-1). Visual inspection of the non-linear fitting of the kinetic data for RohP with respect to 13 revealed that the best fit of the experimental data is achieved using a Hill coefficient of 0.7, indicative of slight negative cooperativity. Though uncommon in PLP-dependent enzymes, negative cooperativity has been previously reported for 4-aminobutyrate aminotransferase.180,181 The decreased affinity for 13 once bound may be a mechanism to decrease the amount of O2 consumed by the other RohP monomer, which would increase the likelihood of more highly oxidized products being produced. Additionally, in the presence of saturating amounts of 13, RohP has a Km = 78 \u00b1 6 \u00b5M and a kcat = 10.2 \u00b1 0.2 min-1 for 76  O2 (Figure 2.8b), which is comparable to the values previously obtained for Ind4 (Km = 90 \u00b1 10 \u00b5M, kcat = 4.6 \u00b1 0.1 min-1). A          B  Figure 2.8 RohP steady-state kinetics with respect to L-arginine and oxygen. a) Orange and blue lines indicate the data fit to the Michaelis \u2013 Menten equation with Hill coefficients of 1 and 0.7 \u00b1 0.1, respectively (Equation 2 Hill Equation.).179 (b) RohP steady-state kinetics for O2. Full details of the experiment are described in the Methods.  2.3.3 The hydroxyl group in the RohP product derives from water, not oxygen or H2O2 The observed stoichiometric conversion of O2 to H2O2 revealed that there are no remaining oxygen atoms that could be incorporated into the product from either O2 or H2O2. Thus, these stoichiometry results suggest that the hydroxyl oxygen atom in 14 derives from H2O. To further interrogate the source of this oxygen atom, the RohP reaction was carried out in 50% H218O in the absence of catalase (Figure 2.4), and the results were analyzed by LC-MS, revealing ten unique ions. Four of these ions, those at [M+H]+ 146, 162, 174, and 190, are also present in the reaction in unlabeled water, and correspond to the unlabeled compounds 110, 111, 105, and 14, respectively. Additionally, four ions, those at [M+H]+ 148, 164, 176, and 192, are two mass units higher, and should arise from hydrolysis of the likely imine products by H218O, giving the corresponding labeled carbonyl oxygens. Finally, the remaining two ions, [M+H]+ 166 and 194, are 4 Da heavier when compared to [M+H]+ 162 and [M+H]+ 190, meaning two labeled oxygens are present. One label should result from hydrolysis of the imine product by H218O, giving a labeled carbonyl, and the second label should result from incorporation of H218O into the hydroxyl group. 77  Altogether, the stoichiometric generation of H2O2 from 13 and H218O labeling studies support that the hydroxyl oxygen derives from water, not O2 or H2O2.  2.3.4 UV-Visible spectrum of RohP To further investigate the RohP-catalyzed reaction, the UV-Visible spectra of RohP was also analyzed. Initially, the spectrum of RohP (30 \u00b5M) contains a single peak with a \u03bbmax at 422 nm, typical of an enzyme-PLP internal aldimine. With the addition of excess 13 (300 \u00b5M) a decrease in the absorbance at 422 nm was observed (Figure 2.9). At the same time, a peak at 515 nm, which is diagnostic of a quinonoid, increased over the 10 min incubation. Finally, a peak at 563 nm increased initially and then decreased after 5 min over the course of the experiment. A similar peak at 567 nm in the Ind4 reaction was previously assigned as a more conjugated quinonoid intermediate,152 and Silvaggi and coworkers assigned a similar peak at 560 nm from the MppP reaction as a more conjugated quinonoid intermediate.157 This experiment suggests that RohP, like Ind4 and MppP, may transition through a more conjugated quinonoid intermediate with a \u03bbmax at 563 nm.  Figure 2.9 UV-Vis absorption spectroscopy of the reaction between RohP and L-arginine. The samples were prepared by mixing 30 \u00b5M of enzyme with 300 \u00b5M of substrate under aerobic conditions. The spectra were recorded at the timepoints as indicated.  78  2.3.5 Crystal structure of holo-RohP Stoichiometric, mass spectral, and spectroscopic results all suggest that RohP consumes two molecules of O2 to carry out a four-electron oxidation and incorporates water to yield 4-hydroxy-2-ketoarginine (14). However, several questions remained unanswered. First, what is the stereochemistry of the 14 product? Second, what intermediates arise in such a reaction scheme? To address these questions, I turned to an X-ray crystallographic investigation of RohP.  I obtained four X-ray crystallographic structures of RohP at resolutions from 1.50 to 1.55 \u00c5 (Table A.1). Each of these structures crystallized as homodimers in space group C2. First, I determined the initial structure of holo-RohP by soaking RohP crystals with 5 mM PLP for 4 h. Then the structure of MppP (PDB: 5DJ1) was used as a model for molecular replacement. Holo-RohP exhibits an overall structure that is typical of other fold type I PLP enzymes and is very similar to MppP (RMSD of 1.215 \u00c5 for C\u03b1 pairs, Figure 2.10a). The RohP monomer consists of a large domain (residues 25-265) and a small domain (residues 266-393) both of which exhibit an \u03b1-\u03b2-\u03b1 motif. Active sites are sandwiched between the large and small domains in individual monomers and exposed to solvent. Unlike other PLP enzymes of fold type I and similar to MppP, there are only minor contributions from one monomer to the active site of the other monomer. As was the case for MppP, the N-terminus of the RohP holoenzyme is incomplete and missing residues 1-25. 79   Figure 2.10 Comparison of the MppP and RohP internal aldimine structures. a) Overall structure of the MppP (PDB ID: 5DJ1) and RohP holoenzymes. b) The structure of MppP is depicted in blue, while RohP is depicted in green. Light and dark shades designate the two monomers composing each homodimer. The PLP of each RohP monomer is depicted as spheres to indicate the location of the active site. The structures were aligned with a RMSD of 1.215 \u00c5 for C\u03b1 pairs (Pymol). b) Overlay of the active site of the internal aldimine structures. Conserved residues are depicted in stick form with RohP numbering. The corresponding residue numbering for MppP can be found in Table A.3.  The RohP internal aldimine is formed between Lys235 and PLP, as was observed for Lys221 and PLP in MppP. Furthermore, all of the same residues that have stabilizing interactions with the internal aldimine in MppP are also observed in RohP (Figure 2.10b; Table A.3). The conserved residue Lys243 (Lys229 in MppP) provides positive charge to stabilize the PLP phosphate. The position of Ser95 (Ser91 in MppP) also causes the phosphate to rotate away from the plane of the pyridine ring, as observed in MppP. Additional interactions serve to stabilize the 80  pyridine ring of PLP. Asp198, conserved in fold type I PLP-enzymes is located near the pyridinyl nitrogen and stabilizes positive charges on the pyridinium cation through a hydrogen bond (Figure 2.10). Asn167 hydrogen bonds with the hydroxyl group of the PLP pyridine ring. As is the case with many PLP enzymes, an aromatic residue, Phe119, is located ~3.5 \u00c5 above the pyridine ring. Finally, the phosphate also forms a water-mediated hydrogen-bonding network with Asp241 (Asp227 in MppP) and Tyr92 (Tyr88 in MppP) of the other monomer of the homodimer. This structure of RohP provided a good initial model and was used to build structures produced from subsequent experiments.  2.3.6 Trapping RohP the first quinonoid The slow rate of RohP catalyzed oxidation of 13 afforded the opportunity to capture intermediates in the catalytic cycle by aerobic soaking of 13. A RohP quinonoid intermediate structure was produced by soaking a crystal with 22 mM 13 for 90 s. In the resulting structure, there is no evidence of a Schiff base linkage between Lys235 and PLP. Instead, positive Fo-Fc electron density in the shape of arginine was found to extend from the C4' aldehyde of PLP. Both the external aldimine (EA1) and quinonoid (Q1) intermediate were modelled into the electron density (Figure A.7). Although both EA1 and Q1 fit well to the density, Q1 was modelled into the final structure (Figure 2.11) because of its fit to the available density and the observation that a quinonoid intermediate with an absorbance at 515 nm accumulates in the reaction of RohP with 13 (Figure 2.9), suggesting that Q1 is more stable than EA1. Despite the formation of a quinonoid, the overall structure is very similar to that of holo-RohP, with the two structures superposing with an RMSD of 0.106 \u00c5 across C\u03b1 pairs. In the active site, only Asn121 exhibits a minor change in conformation, rotating towards the quinonoid (Figure 2.11). Furthermore, when the quinonoid is present, the guanidium of Arg367 forms a salt bridge with the carboxylic acid of the arginine substrate. Additionally, Leu266 of the second monomer, which is unchanged in position relative to holo-RohP, now forms part of the periphery of the active site and helps to orient the guanidium end of the quinonoid. Because the N-terminal amino acids are disordered in this structure, the sidechain of the 13 substrate is pointed into solvent. As such, C\u03b3 is 3.9 \u00c5 from His34 (Figure 2.12), which is a residue conserved among RohP, MppP, and Ind4 (Figure 2.1) but not observed in other fold type I aminotransferases. 81    Figure 2.11 Modeling of RohP-quinonoid 1 intermediates. Left: External aldimine species modelled shown in stick form. The fits to the available Fo-Fc density in the active site of RohP is shown below (positive density is contoured at 3.0 \u03c3). Right: Modelled quinonoid ligand in the active site of RohP. Residue side chain that interact with the ligand are shown as sticks.  Figure 2.12 Distances between modelled intermediates and conserved residues Glu20 and His34 in Chain A of the RohP homodimer. A) The RohP holoenzyme has only a water molecule (red sphere) positioned 2.4 \u00c5 from His34. B) The N-terminus and Glu20 are absent in the RohP-quinonoid I. His34 is positioned 3.9 \u00c5 from 82  the C\u03b3 atom of the bound substrate and 2.5 \u00c5 from a water molecule. C) In the RohP-quinonoid II, the N-terminus is ordered, including Glu20, which is now positioned 4.0 \u00c5 from the C\u03b2 of the bound substrate. Furthermore, the new positioning of the amino acid substrate places its C\u03b2 and C\u03b4 atoms 3.2 \u00c5 and 3.5 \u00c5, respectively, from His34. No ordered water was present near His34. D) The product 14 has its C\u03b2 3.3 \u00c5 from Glu20 and its C\u03b4 3.6 \u00c5 from His34. The 4' hydroxyl group is positioned 2.7 \u00c5 from His34.  2.3.7 Trapping of a more conjugated quinonoid orders the N-terminus of RohP When another RohP crystal was soaked with 4 mM PLP overnight and 10 mM 13 for 5 min, the crystal changed color from yellow to red (Figure 2.13). I cryoprotected and flash-froze this red crystal. Solving the structure of this red crystal again revealed unknown positive Fo-Fc electron density in the shape of arginine in the active site. Initially, several different intermediates were built using Phenix eLBOW and then modelled into the available Fo-Fc density (Figure A.8). However, none of these intermediates gave a satisfactory fit to the available density. Thus, I employed an alternate approach, utilizing ARP\/wARP,176 an online automated protein model building and refinement program to build a ligand structure based upon the unknown density. The initial ARP\/wARP output was adjusted in COOT, refined, and found to match available omit density well (Figure A.9). This modelled ligand has bond lengths that closely match what would be expected for the more conjugated quinonoid (Q2) intermediate, containing a double bond between C\u03b2 and C\u03b3 positions of the arginine (Table A.4). To interrogate whether modeling Q2 was reasonable for such a red crystal, I carried out spectroscopic analysis on a second set of crystals that had been soaked with 13 using the microspectrophotometer outfitted on beamline 9-2 at the Stanford Synchrotron Radiation Lightsource. This work revealed that such red crystals have an absorbance spectrum with a \u03bbmax of 515 and 563 nm (Figure 2.14), matching the peaks observed by UV-Visible spectroscopy in solution (Figure 2.9). 83   Figure 2.13 L-arginine soaking of RohP crystals. Representative results of soaking a single crystal (top) or several plate crystals (bottom) with L-arginine (13). Crystals were soaked for illustrative purposes and no additional data were obtained using the crystals depicted in these images.   Figure 2.14 Quinonoid intermediates from a red, L-arginine soaked crystal. Holo-enzyme (yellow) and a representative spectrum from crystals that had been soaked with \u029f-arginine (13) prior to being frozen (red). The baseline absorbance of the holo-enzyme was offset by -0.04 absorbance units to align the spectra for comparison.  The structure of one of these second set of red crystals, which diffracts to a lower, 2.0 \u00c5 resolution was solved, revealing that the electron density present in the active site of this second crystal is also supportive of an external aldimine adduct. However, the lower resolution (2.0 \u00c5) of 84  this second crystal did not allow for unambiguous assignment of the modelled quinonoid intermediates, as both Q1 and Q2 fit into the available density well (Figure A.10). As a more conjugated quinonoid intermediate was observed in the UV-visible spectra both in solution and in crystallo (Figure 2.9, Figure 2.14), it is likely that Q2 is a more stable intermediate than a more conjugated external aldimine (EA2). Therefore, Q2 was modeled into the final structure from the initial set of soaking experiments, based on the ARP\/wARP coordinates (Figure A.9, Figure 2.15). However, the ~1.5 \u00c5 resolution of this structure does not allow us to unambiguously assign the pattern of double bonds. Furthermore, a mixture of intermediates may contribute to the density that was observed.  Figure 2.15 Modeling of RohP-quinonoid 2 intermediates. Left: External aldimine species modelled shown in stick form. The fits to the available Fo-Fc density in the active site of RohP is shown below (positive Fo-Fc density is contoured at 3.0 \u03c3). Right: Modelled quinonoid ligand in the active site of RohP. Residue side chains that interact with the ligand are shown as sticks.  In this structure of the Q2 intermediate, electron density for residues 13-25 of the N-terminus is present, forming a small \u03b1-helix that closes off the active site, isolating the active site from the bulk solvent (Figure 2.15). With the ordering of the N-terminus, the amide nitrogen of Leu16 forms a hydrogen bond with Asp120. Additionally, the hydroxyl sidechain of Thr17 forms a hydrogen bond with the guanidium group of Q2 and pushes the guanidium deeper into the active site where it forms additional hydrogen bonds with Ser95 and the carbonyl oxygen of Val264. The conserved residue Glu20 becomes ordered near the carboxylic acid and C\u03b2 of Q2. Phe119 also 85  exhibits a dual conformation, with the additional conformer rotating from its previous position above the pyridine ring to a position ~3.5 \u00c5 above the carboxylic acid of Q2. Asn121 also moves ~180\u00b0 from its previous position to partially occupy the area that was vacated by the rotation of Phe119. Collectively, these changes push Q2 deeper into the active site, into a position where Glu20 is 4.0 \u00c5 away from C\u03b2 and His34 is 3.2 \u00c5 and 3.5 \u00c5 away from C\u03b2 and C\u03b4, respectively, of the 13 substrate (Figure 2.12).  2.3.8 Structure of RohP-product complex A final snapshot of the RohP catalytic cycle was obtained by soaking RohP crystals with 10 mM 13 overnight. During this experiment, the crystal changed from yellow to red and then back to yellow. The resulting yellow crystal was cryoprotected and flash-frozen. The structure I obtained has residues 14-25 of the N-terminus present, though the density of these residues in chain B is weaker. Therefore, these residues were modelled with an occupancy of ~0.7-0.8 in chain B, compared to full occupancy in chain A. The active site of the enzyme also remains largely unchanged compared to the Q2 structure, with Asn121 remaining flipped relative to its initial position, and Phe119 exhibiting a dual conformation away from the pyridine ring of PLP. However, in this structure, PLP has again formed an internal aldimine with Lys235 (Figure 2.16). 86   Figure 2.16 Active site of RohP-product complex. Residue side chains that interact with (S)-4-hydroxy-2-ketoarginine 14 (teal) are shown as sticks.  Strong Fo-Fc density was present above the internal aldimine, displaying a shape consistent with a product containing a 4-hydroxy group. As the stereochemistry of the 4-hydroxyl was still undetermined, both the R and S enantiomers of the molecule were modelled into the positive Fo-Fc density present in the active site (Figure 2.17). Refinement of the R-enantiomer produced strong negative Fo-Fc density around the hydroxyl group, while the same refinement of the S-enantiomer produced no negative Fo-Fc density around the hydroxyl group. I also modelled both enantiomers of the presumed enamine product into the available density to determine their fit as well (Figure 2.18). Once refined, I again observed a clear preference for the S-enantiomer based on the refined density. However, I also observed slightly more negative density over the enamine double bond, leading us to ultimately model (S)-14 into the final structure. It is likely that the dynamic nature of the N-terminus should allow for hydrolysis of the enamine product over the extended incubation period, and therefore it is more likely that the hydrolysis product is present in the active site. 87   Figure 2.17 Possible stereoisomers of 4-hydroxy-2-ketoarginine. A) The R-enantiomer modelled into Fo-Fc omit density present in the active site. B) The density present around the R-enantiomer after refinement using Refmac. C) The S-enantiomer modelled into Fo-Fc omit density present in the active site. D)The density present around the S-enantiomer after refinement using Refmac. The Fo-Fc maps are contoured at 3.0 \u03c3 with positive and negative density indicated as green and red, respectively. The 2Fo-Fc maps are contoured at 1.0 \u03c3 in gray. Ligands were built using eLBOW in the Phenix software suite.  88   Figure 2.18 Refinement of alternative products. A) The density surrounding the R-enantiomer after refinement. B) Density surrounding the S-enantiomer after refinement. Comparison of the refinement of C) (S)-14 and D) the modelled enamine. The Fo-Fc maps are contoured at 3.0 \u03c3 with positive and negative density indicated as green and red, respectively. The 2Fo-Fc maps are contoured at 1.0 \u03c3 in gray. Ligands were built using eLBOW in the Phenix software suite.  2.3.9 Characterization of the RohP-His34Ala variant Based upon the analysis of the four crystal structures of RohP, His34 appears likely to be involved in the installation of the 4-hydroxyl group, due to its position relative to the quinonoid intermediates and hydroxyl group of the final product (Figure 2.12). To probe the function of His34 I created a His34Ala variant of RohP with site-directed mutagenesis. ESI-MS analysis was again employed to determine the product(s) of the His34Ala variant. The production of 14 was abolished, however, the variant was still able to produce 105. The corresponding decarboxylation product 110 was also detected (Figure 2.19). That this variant was only able to perform one oxidation, and no other intermediates were detected, suggests that His34 may also be involved in catalyzing the second oxidation of the arginine substrate and possibly the final hydration reaction. 89   Figure 2.19 Reaction of the RohP-His34Ala variant with L-arginine. Liquid chromatography-mass spectrometry analysis of the products of (a) the reaction of L-arginine (13) and (b) the reaction with L-arginine (13) in the presence of catalase. Assays were carried out under the same reaction conditions as the wild type but using 2 mM L-arginine (13) and 20 \u00b5M of enzyme. Representative integrated total ion chromatograms are shown for each spectrum.  2.4 Discussion This work describes the investigation of RohP, an enzyme that uses pyridoxal phosphate to catalyze the transformation of L-arginine (13) and O2 to two products: 2-ketoarginine (105) and (S)-4-hydroxy-2-ketoarginine (14). While previous studies by Eguchi and Silvaggi highlighted this enzymatic reaction, the key question of how an enzyme uses PLP and O2 to hydroxylate an unactivated, sp3-hybridized carbon remained unaddressed. Here, I use detailed mass spectral, kinetic, stoichiometric, and X-ray crystallographic analysis to build a firm mechanistic framework for understanding this group of PLP-, O2-dependent hydroxylases. This work demonstrates that RohP and the oxidase Ind4 share many key features: both stoichiometrically convert O2 to H2O2, both generate quinonoid and conjugated quinonoid intermediates, and both produce the less oxidized product, 105, which results from hydrolysis of the corresponding imine. The enzymes differ only in whether they produce 4,5-didehydroarginine (102) or 14. The X-ray crystal structures of RohP containing trapped quinonoid and conjugated quinonoid intermediates, along with a structure with product bound, unveil a shared mechanism whereby both RohP and Ind4 can catalyze a four-electron oxidation of 13. However, only RohP can utilize water to carry out a stereospecific alkene hydration.  90  The crystallographic work described in this chapter shows that RohP forms an external aldimine at the \u03b1-amino group of arginine, exactly as would be expected for a PLP-dependent aminotransferase. The following mechanistic proposal (Figure 2.20), begins with the C\u03b1 proton of the external aldimine (EA1) being abstracted by Lys235, with the resulting anionic species stabilized by formation of a quinonoid intermediate (Q1). Q1 is a typical intermediate in PLP-dependent aminotransferases, and one that is also observed to accumulate by UV-Visible spectroscopy. It is here that Q1, like in the proposed Ind4 mechanism, reacts directly with O2, oxidizing the bound amino acid and releasing H2O2. The exact mechanism of how O2 interacts with Q1 is currently unknown, as is the mechanism for release of H2O2. Current speculative proposals for oxygen activation by arginine oxidases like MppP, Ind4, and RohP involve an electron transfer from Q1 (or Q2) to O2, which would generate superoxide,152,157,160 in a process similar to those used by both flavin-dependent enzymes182 and some cofactor-independent oxidases.183 The superoxide could then oxidize the quinonoid radical at particularly electron rich positions of the resultant quinonoid radical, such as C\u03b1 of the 13 substrate, or C4' of PLP.184 After oxidation, the resulting external aldimine intermediate (EA2) forms. This intermediate can undergo one of two fates: it can either be attacked by Lys235, which will reform the internal aldimine and release an enamine product, which tautomerizes to the imine and is then hydrolyzed to give 2-ketoarginine (105). Alternatively, the oxidized intermediate can remain in the active site. If it remains in the active site, the C\u03b3 proton can undergo rapid deprotonation by the adjacent His34, shuttling electron density into the cofactor to give the more conjugated quinonoid (Q2) intermediate. Then, a second molecule of O2 could react with Q2, again oxidize the substrate, and release a second molecule of H2O2. Now, the PLP-tethered didehydroarginine could undergo hydration. One possible scenario is that the double bond between C\u03b3-C\u03b4 is first protonated at the C\u03b4 position. The resulting carbocation at C\u03b3 could be stabilized through resonance with density from the PLP pyridine ring. Now His34 deprotonates an adjacent water to insert the hydroxyl group at C\u03b3 and give the resulting PLP-tethered final product. The stereospecificity of the hydration appears to be promoted by the positioning of His34, which is positioned to the si face of the alkene. At the same time, the phosphate of PLP occludes solvent access to the re face of the alkene. To complete the catalytic cycle, the 4-hydroxy product is then released when Lys235 91  attacks to release the enamine with the concomitant formation of the RohP-PLP internal aldimine. The imine tautomer is then hydrolyzed by water to produce the final product 14.   Figure 2.20 Proposed mechanism of the RohP-catalyzed reaction to 4-hydroxy-2-ketoarginine.  The work described in this chapter raises new questions about O2, PLP-dependent oxidases. First, what are the structural features that distinguish arginine oxidases like Ind4 from arginine hydroxylases like RohP and MppP? The data supports that the remarkable feat of RohP \u2013 installation of a hydroxyl group \u2013 could be the stereospecific hydration of a PLP-tethered didehydroarginine. If this possibility is true, what causes the hydroxylase RohP to catalyze the hydration, whereas the oxidase Ind4 releases didehydroarginine as a product? The two types of enzymes have all residues conserved in their active sites \u2013 including His34 \u2013 suggesting that other residues will be key to determining the product outcome (Table A.3). An Ind4 structure will be essential for pinpointing which residues are critical to determining product outcome. Second, in 92  both RohP-like and Ind4-like enzymes, 105 is produced. Is the production of 105 an unavoidable waste product for such 13, PLP-, O2-dependent enzymes, or is there a purpose for production of 105? Elucidation of the full biosynthetic pathway will begin to address this issue. Finally, an unresolved question is why a select group of PLP-dependent enzymes are able to use O2 to catalyze oxidation reactions and how O2 is activated during catalysis.142,144,146 Answers to these questions await further study.       93  Chapter 3: In vitro biochemical characterization of the azomycin biosynthetic gene cluster 3.1 Introduction Nitroimidazoles are an essential component of the modern antibiotic arsenal. For example, 5-nitroimidazole derivatives, exemplified by metronidazole (9) (Figure 1.2), are commonly used to treat both gram-positive and gram-negative anaerobic bacterial infections, including Clostridium difficile and Helicobacter pylori.18,19 The low redox potential of anaerobic bacterial cells allows the nitroimidazole drug to act as an electron sink for the bacterial pyruvate:ferredoxin oxidoreductase complex.18 The resultant radical species, whose specific identity varies depending on the particular nitroimidazole and target bacteria, cause DNA damage, eventually inducing bacterial cell death.18,185 Despite several decades of their use, incidents of nitroimidazole resistance remain relatively low, and nitroimidazoles are increasingly being utilized to treat multidrug-resistant bacteria.18 The development of the nitroimidazoles can be traced back to the discovery of azomycin (6) in 1953.13 The novel 2-nitroimidazole pharmacophore was found to be an effective treatment for Trichomonas vaginalis.15 Interest in this intriguing heterocycle led to several investigations into its biosynthetic origins.28,29 These early studies established that L-arginine (13) is converted to 6 via 2-aminoimidazole (12) in the producing strain Streptomyces eurocidicus. Building upon these studies, Eguchi and coworkers determined that in S. eurocidicus, O2 and PLP were required to convert 13 to the unusual oxidized intermediate 4-hydroxy-2-ketoarginine (14).27 They also observed that 14 was converted stoichiometrically to pyruvate (15) and 12. Based on these results, they proposed guanidinoacetaldehyde (17) as a key intermediate capable of spontaneously cyclizing to 12, which is in turn oxidized to produce 6 (Figure 1.4, Figure 3.1). However, no azomycin biosynthetic enzymes were ever identified. Perhaps the most intriguing aspect of Eguchi\u2019s biosynthetic proposal was an O2- and PLP-dependent enzymatic reaction. While PLP-dependent enzymes are extremely diverse, their reactions with oxygen are generally limited to so-called \u2018paracatalytic\u2019 reactions that produce unintended side products in low amounts.142,143 Recently, several PLP-dependent enzymes that use O2 as a co-substrate have been reported.148 These include decarboxylases such as plant phenylacetaldehyde synthase,144 as well as the L-arginine oxidases MppP from L-enduracididine 94  biosynthesis157,160 and Ind4 from indolmycin biosynthesis (Figure 1.38, Figure 1.39).150,152 Interest in these O2- and PLP-dependent enzymes led me to identify a third O2- and PLP-dependent arginine oxidase RohP, which was described in Chapter 2.186 RohP generates (S)-4-hydroxy-2-ketoarginine (14) and 2-ketoarginine (105) from 13 like MppP (Figure 2.5),157 but is encoded in a gene cluster unrelated to either the L-enduracididine or indolmycin gene clusters (Figure 2.2). Therefore my hypothesis is that this novel cryptic gene cluster containing RohP could be responsible for the biosynthesis of azomycin (6) (Figure 3.1).  Figure 3.1 In vitro reactions linking L-arginine to azomycin.  3.2 Materials and Methods 3.2.1 General Methods Primers were purchased from Integrated DNA Technologies. DNA sequencing was carried out by the NAPS Unit DNA Sequencing Facility and the DNA Sequencing Core Facility at the Center for Molecular Medicine and Therapeutics (The University of British Columbia). Gene synthesis and codon optimization was performed by Bio Basic Inc. (Markham, Canada). Reagents were purchased from Alfa Aesar, Bio Basic Inc., Enamine Ltd., Gold Biotechnology, Millipore Sigma, New England Biolabs (NEB), Thermo Fisher Scientific Canada, and VWR International. 3.2.2 Cloning and expression of recombinant proteins Genomic DNA of Streptomyces cattleya (DSM 46488) was isolated using phenol:chloroform extraction. The cell pellet of a 50 mL culture of S. cattleya was washed three times with 10 mL of 25 mM Tris, 25 mM EDTA, 0.3 M sucrose pH 8 buffer (TES). The cells were then lysed at 37 \u00b0C for 2 h with 5 mL of 3 mg\/mL lysozyme, and 100 \u03bcg\/mL RNase in TES buffer. Then 100 \u03bcL of 20 mg\/mL proteinase K and 1 mL of 10% sodium dodecyl sulfate were added, and this mixture was incubated at 55 \u00b0C for 2 h. After cooling on ice, 2.5 mL of 5 M sodium acetate was added, followed by 8 mL of a phenol:choroform mixture (1:1) and gently mixed. After 95  centrifugation of the mixture the DNA was isolated from the aqueous phase by precipitation with isopropanol and stored in 10 mM Tris, 1 mM EDTA, pH 8.0 buffer. The desired gene sequences were amplified from the purified genomic DNA by the polymerase chain reaction using primers found in Table B.2. The amplified DNA and pET28a were digested with NdeI and XhoI, and pET28a was also treated with Calf Intestinal Alkaline Phosphatase. These reactions were purified using the QIAquick PCR purification kit, and the digested products were ligated together using T4 ligase. The ligation mixture was used to transform chemically competent E. coli DH5\u03b1 cells. The nucleotide sequence of the insert was confirmed by sequencing. The plasmid containing the gene of interest was transformed into E. coli BL21 (DE3) cells for recombinant protein production. For recombinant production of RohP, RohQ, SCyRohS (RohS from S. cattleya, NCBI accession number: AEW92765.1), KAzRohS (RohS from Kitasatospora azatica, NCBI accession number: WP_063774763.1) and RohT, overnight cultures of E. coli BL21 (DE3) harboring the desired plasmid were used to inoculate 4 x 750 mL of LB medium containing 50 \u03bcg\/mL kanamycin. These cultures were grown at 37 \u00b0C at 200 rpm to an OD600 of between 0.7-1.0 and then cooled to 16 \u00b0C for 1 h. Protein expression was induced by adding 0.2 mM IPTG. For RohT, 1 mM (NH4)2Fe(SO4)2\u03876H2O and 1 mM L-cysteine were also added to the cultures at the time of induction with IPTG. All cultures were then grown for an additional 16 h at 16 \u00b0C. Cells were harvested by centrifugation at 5000 rpm and frozen at -20 \u00b0C until protein purification. Due to its low yields in LB media, RohR was instead produced using an autoinduction protocol capable of supporting high cell densities. The autoinduction media contained 20 g\/L tryptone, 10 g\/L yeast extract, 2.675 g\/L NH4Cl, 0.24 g\/L MgSO4, 8 g\/L glycerol, 2.31 g\/L KH2PO4, 12.54 g\/L K2HPO4, 0.05 g\/L glucose, 0.2 g\/L lactose, and 50 \u00b5g\/mL kanamycin. Overnight E. coli cell cultures in LB were prepared as described above and were used to inoculate 1 L of autoinduction media. These cultures were initially grown for 1 h at 37 \u00b0C at 200 rpm. After 1 h the growth conditions were changed to 16 \u00b0C and 130 rpm for ~72 hours. Cells were harvested by centrifugation at 5000 rpm and frozen at -20 \u00b0C until protein purification. 3.2.3 Purification of recombinant proteins All proteins were purified using the same general procedure. Briefly, the cells were thawed and resuspended in the appropriate binding buffer (see Table B.3). The cells were lysed with sonication, and then centrifuged at 12,000 rpm for 45 min. The clear lysate was gravity filtered 96  through ~1 mL of Chelating Sepharose\u2122 Fast Flow resin charged with NiCl2\u03876H2O. After the lysate had passed through the column, the resin was washed with 15 mL of binding buffer. The protein was then eluted using a stepwise gradient using binding buffer containing increasing concentrations of imidazole (5-500 mM) in 5 mL fractions. The fractions containing protein were identified using SDS-PAGE. These fractions were either combined and placed in dialysis overnight, or the protein was further purified using a HiLoadTM Superdex 16\/600 200 pg column equilibrated with the appropriate storage buffer (see Table B.3). After the proteins were exchanged into the appropriate storage buffer after dialysis or FPLC, they were concentrated by centrifugation with an ultra-centrifugal filter (10,000 molecular weight cut-off for RohP, RohQ, RohR, SCyRohS, and KAzRohS or 3000 molecular weight cut-off for RohT). The proteins were then stored at 4 \u00b0C or flash frozen and stored at -20 \u00b0C. 3.2.4 In vitro biochemical analysis of the RohP-RohR coupled reaction In vitro assays for the RohP-RohR coupled reaction were carried out in 100 \u00b5L reactions containing 4 mM 13, 10 \u00b5M RohP, 20 \u00b5M RohR and 25 \u00b5g\/mL catalase in 20 mM HEPES, 50 mM NaCl, pH 7.5 buffer. The reactions proceeded for 16 h at room temperature and were then quenched with 100 \u00b5L of MeOH. Precipitated protein was removed by centrifugation. All direct ESI-MS and LC-MS analysis of in vitro reactions was performed using a 1260 HPLC apparatus (Agilent) coupled to a 6120 Quadrupole LC\/MS system (Agilent) equipped with a Poroshell 120, EC-C18, 2.7 \u00b5m, 4.6 mm ID \u00d7 50 mm column (Agilent), operated in positive ion mode. For direct ESI-MS analysis of the reaction mixture, 10 \u00b5L of the supernatant was analyzed unless otherwise indicated. For DNS-Cl derivatization, 50 \u00b5L of the reaction mixture was reacted with 50 \u00b5L 80 mM Li2CO3, 70 \u00b5L ACN, and 30 \u00b5L of 5 mM DNS-Cl dissolved in ACN. The reaction was carried out at room temperature for 1 h and then 40 \u00b5L of 2 % ethylamine was added to the mixture to react with excess DNS-Cl. The mixture was centrifuged, and 10 \u00b5L of the supernatant was subjected to LC-MS, using a Luna C18, 5 \u00b5m, 4.6 mm ID \u00d7 250 mm column (Phenomenex). Elution was performed at 0.5 mL\/min using a mobile-phase consisting of a linear gradient of water and ACN ((v\/v): 95:5 to 50:50, 0 to 15 min; 0:100, 15 to 20 min, 95:5, 20 to 22 min; 95:5 22 to 26 min), with both solvents containing 0.1% (v\/v) FA. 97  For OPD derivatization, 100 \u00b5L of the reaction mixture was treated with 100 \u00b5L of H2O and 100 \u00b5L of 100 mM OPD dissolved in 2 M HCl for 20 minutes at 80 \u00b0C. The reaction mixture was then cooled on ice for 5 minutes, centrifuged, and 10 \u00b5L of supernatant was subjected to LC-MS analysis, using the same instrument and column as the DNS-Cl analysis. Elution was performed at 0.5 mL\/min using a mobile-phase consisting of a linear gradient of water and ACN ((v\/v): 95:5 to 50:50, 0 to 15 min; 0:100, 15 to 20 min, 95:5, 20 to 22 min; 95:5 22 to 25 min), with both solvents containing 0.1% (v\/v) FA. 3.2.5 In vitro biochemical analysis of the RohQ-catalyzed reaction In vitro assays for the RohP-RohR-RohQ coupled reaction were carried out in 100 \u00b5L reactions containing 4 mM 13, 10 \u00b5M RohP, 20 \u00b5M RohR, 8 \u00b5M of RohQ and 25 \u00b5g\/mL catalase in 20 mM MOPS, 50 mM NaCl, pH 7.0 buffer. For the initial assays the mixture reacted for 16 h at room temperature and was then quenched with an equal volume of MeOH. To monitor the initial product distributions in the presence and absence of RohQ, the reaction was scaled up to 600 \u00b5L. From this mixture 60 \u03bcL aliquots were taken at various time points and quenched immediately with an equal volume of MeOH. In both sets of experiments the precipitate was removed by centrifugation, and 1 \u00b5L of the resulting supernatant was then analyzed by ESI-MS. To directly test the activity of RohQ, its proposed substrate 17 was generated in situ using the RohP-RohR coupled reaction. In this case, six 200 \u00b5L reactions were set up in parallel, consisting of 1 mM 13, 10 \u00b5M RohP, 20 \u00b5M RohR, 25 \u00b5g\/mL catalase in 20 mM MOPS, 50 mM NaCl, pH 7.0 buffer. The reactions were incubated at room temperature for 2 h, combined, and then RohP and RohR were removed from the reaction mixture by centrifugation using an ultra centrifugal filter with a 10,000 Da molecular weight cut-off. Next, 10 \u00b5L of 80 \u00b5M RohQ (4 \u00b5M final concentration), and 20 \u00b5L of 20 mM MOPS, 50 mM NaCl, pH 7.0 were added to 170 \u00b5L of supernatant. A control reaction with an equivalent volume of boiled RohR was carried out in parallel. Aliquots of 20 \u00b5L were collected at the time points as indicated and quenched by the addition of 180 \u00b5L of ACN. Precipitation was removed by a brief centrifugation, and then 1 \u00b5L of the resulting supernatant was analyzed with ESI-MS. 3.2.6 In vitro biochemical analysis of the KAzRohS-catalyzed reaction In vitro assays for the KAzRohS-catalyzed reaction were carried out in 100 \u00b5L reactions containing 2 mM 12, 15 \u00b5M KAzRohS, 2 mM FeSO4\u00b77H2O, 50 \u00b5M PMS, and 5 mM NADH, in 98  20 mM HEPES (pH 7.5) buffer. Control reactions lacking each reaction component were carried out in parallel. To test the activity of KAzRohS in the presence of various metal ions the reaction was the same as described above except for the inclusion of 2 mM of the following metal hydrates: MnCl2\u00b74H2O, FeSO4\u00b77H2O, CuSO4\u00b75H2O, NiSO4\u00b76H2O, CoCl2\u00b72H2O, or ZnSO4\u00b75H2O. In all cases the reactions proceeded for 16 h at room temperature and were quenched by the addition of an equal volume of MeOH. The precipitate was removed by centrifugation, and 5 \u00b5L of the resulting supernatant was then analyzed by ESI-MS. The activity of KAzRohS was also tested in the presence of the characterized Rieske ferredoxin-ferredoxin reductase pair of BphF and BphG from the biphenyl degradation pathway.187,188 These assays were 100 \u00b5L in volume, and contained 500 12, 2 mM FeSO4\u00b77H2O, 3 mM NADH, 10 \u00b5M KAzRohS, 5 \u00b5M of BphF, and 5 \u00b5M of BphG in 20 mM HEPES (pH 7.5) buffer. The reactions proceeded for 16 h at room temperature and were then quenched with 100 \u00b5L of MeOH. Precipitated protein was removed by centrifugation and 10 \u00b5L of the resulting supernatant was analyzed by ESI-MS. 3.2.7 KAzRohS kinetic methods For determination of Michaelis-Menten kinetic parameters the reaction mixture contained 20 mM HEPES, pH 7.5, 5 mM NADH, 2 mM FeSO4\u03877H2O, 50 \u03bcM PMS in a total volume of 95 \u03bcL. The concentration of 12 was varied from 31 \u03bcM to 1 mM, and each chosen concentration was performed in triplicate. The reaction was initiated upon addition of 15 \u03bcM of KAzRohS (5 \u03bcL of 300 \u03bcM stock) to the reaction mixture. The reaction proceeded for 5 min at room temperate and was quenched with the addition of 100 \u03bcL of MeOH. The solution was briefly centrifuged to remove precipitated enzyme and the resulting supernatant was immediately analyzed by HPLC using an Agilent 1260 HPLC apparatus, equipped with a Luna C18, 5 \u00b5m, 4.6 mm ID \u00d7 250 mm column (Phenomenex). Elution was performed at 0.5 mL\/min using a mobile-phase consisting of a linear gradient of water and ACN ((v\/v): 90:10, 0 to 5 min, 90:10 to 0:100, 5 to 6 min; 0:100, 6 to 10 min, 90:10, 10 to 11 min; 90:10 11 to 15 min), with both solvents containing 0.05% (v\/v) TFA. The amount of 6 produced by the reaction was calculated by comparison to a standard curve of authentic 6 generated using the same elution gradient. The data points were plotted and subjected to a non-linear regression in GraphPad Prism v. 5.04 to obtain the final kinetic parameters. 99  3.2.8 ICP-MS analysis of KAzRohS and RohT KAzRohS and RohT were purified from E. coli BL21(DE3) as described in the protein purification section. For ICP-MS analysis, KAzRohS and RohT were transferred into 20 mM Tris, 50 mM NaCl, pH 8.0 buffer by dialysis overnight. The concentration of KAzRohS was adjusted to 105 \u03bcM using the ExPASy ProtParam calculated coefficient of 32,555 M-1\u0387cm-1 and measured by Nanodrop. The concentration of RohT was adjusted to 100 \u03bcM as calculated by the Bradford assay by comparison to a standard curve generated using bovine serum albumin. ICP-MS data were generated by ALS Limited in Burnaby, BC, Canada. Protein samples were treated with hydrochloric acid and nitric acid to release all metal ions prior to analysis. The metal:molar enzyme molar ratio was calculated using enzyme concentration. 3.2.9 UV Vis-spectroscopy of RohT A quartz cuvette with a path length of 1 cm and a Varian Cary 100 Bio UV-Vis spectrophotometer (Agilent) were used to record UV-Vis spectra of RohT. As purified RohT was used to obtain the oxidized spectrum, while RohT was incubated with 2 mM of reductant for 15 minutes prior to obtaining each reduced spectrum. The concentration of RohT in each spectrum was 110 \u00b5M as calculated using the Bradford assay by comparison to a standard curve generated using bovine serum albumin. 3.2.10 In vitro biochemical analysis of RohT In vitro assays for the reaction between 12 and RohT were carried out in 100 \u00b5L reactions containing 2 mM 12, 30 \u00b5M RohT, 2 mM FeSO4\u00b77H2O, 100 \u00b5M PMS, and 2 mM NADH in 20 mM HEPES, pH 7.5 buffer. A reaction containing 28 \u00b5M KAzRohS instead of RohT was used as a positive control. The reactions proceeded for 16 h at room temperature and were then quenched with 100 \u00b5L of MeOH. Precipitated protein was removed by centrifugation and 10 \u00b5L of the resulting supernatant was analyzed by ESI-MS. In vitro assays for the RohT and KAzRohS coupled reaction were carried out in 100 \u00b5L reactions containing 1 mM 12, 2 mM FeSO4\u00b77H2O, 100 \u00b5M PMS, 2 mM NADH, 15 \u00b5M KAzRohS, and 30 \u00b5M RohT in 20 mM HEPES, pH 7.5 buffer. Controls lacking RohT were carried out in parallel. The reactions proceeded for 16 h at room temperature and were then quenched with 100 \u00b5L of MeOH. Precipitated protein was removed by centrifugation and 10 \u00b5L of the resulting supernatant was analyzed by ESI-MS. 100  In vitro assays for the RohT and KAzRohS coupled reaction under reducing conditions were carried out in 100 \u00b5L reactions containing 2 mM 12, 2 mM FeSO4\u00b77H2O, 5 mM NADH, 4 mM DTT, 10 \u00b5M KAzRohS, and 10 \u00b5M RohT in 20 mM HEPES, pH 7.5 buffer. The reactions proceeded for 16 h at room temperature and were then quenched with 100 \u00b5L of MeOH. Precipitated protein was removed by centrifugation and 10 \u00b5L of the resulting supernatant was analyzed by ESI-MS. In vitro biochemical analysis of the RohT and KAzRohS coupled reaction including non-cognate ferredoxin reductases was carried out in 100 \u00b5L reactions containing 500 \u00b5M 12 FeSO4\u00b77H2O, 3 mM of NADH or NADPH, 5 \u00b5M KAzRohS, 10 \u00b5M RohT, and 5 \u00b5M of either spinach ferredoxin reductase, ferredoxin reductase from the cyanobacterium Synechococcus elongatus,189 seFDR or BphG188 in 20 mM HEPES, pH 7.5 buffer. The reactions proceeded for 16 h at room temperature and were then quenched with 100 \u00b5L of MeOH. Precipitated protein was removed by centrifugation and 10 \u00b5L of the resulting supernatant was analyzed by ESI-MS. 3.2.11 Anaerobic reconstitution of RohT Aerobically purified RohT (400 \u03bcL) was dialyzed into 200 mL of 20 mM Tris, 50 mM NaCl (pH 8) buffer, and the buffer was purged with N2 for 1 hour. After this time, sodium dithionite was dissolved in de-gassed buffer, and added to this solution to a final concentration of 4 mM and allowed equilibrate for 30 mins. Next, Na2S\u03879H2O was dissolved in de-gassed buffer, and then added to this solution to a final concentration of 1 mM. The solution was allowed to mix for an additional 30 mins. Then, (NH4)2Fe(SO4)2\u03876H2O was dissolved in de-gassed buffer, and then added to the anaerobic mixture to a final concentration of 1 mM. The protein was allowed to mix with this now black mixture for 3 hours. Then the protein was removed from this mixture and transferred to fresh, de-gassed 20 mM Tris, 50 mM NaCl (pH 8) buffer containing 1 mM sodium dithionite. Here it was allowed to equilibrate for an additional 2 h. At this point it was removed from the anaerobic conditions and then used for analysis. 3.2.12 In vitro production of azomycin using four enzymes  For the one-pot reaction to produce 6 the reaction mixture contained 4 mM 13, 50 \u00b5g\/mL catalase, 5 mM NADH, 50 \u00b5M PMS, 500 \u00b5M FeSO4, 20 \u00b5M RohP, 10 \u00b5M RohQ, 20 \u00b5M RohR, and 20 \u00b5M KAzRohS in 20 mM HEPES, pH 7.5 buffer in a final volume of 100 \u03bcL The reaction was initiated upon the addition of 13 and reacted for 16 h at room temperature. The reaction was 101  quenched with an equal volume of MeOH. The precipitate was removed by centrifugation, and 5 \u00b5L of the resulting supernatant was then analyzed by ESI-MS. 3.2.13 Cultivation and extraction of metabolites from Streptomyces cattleya  Frozen spore stocks of S. cattleya (in 10% glycerol) were used to inoculate 50 mL of seed culture medium containing 10 g\/L yeast extract, 10 g\/L glucose, 0.182 g KH2PO4, 0.190 Na2HPO4, 0.05 g\/L MgSO4\u03877H2O, adjusted to pH 6.5.190 The culture was grown for 3 d at 28 \u00b0C, 150 rpm. Next 20 \u00b5L of growing seed culture was used to inoculate 50 mL of growth medium. Three different growth mediums were used, glucose yeast maltose (GYM) medium containing 4 g\/L glucose, 4 g\/L yeast extract, 10 g\/L malt extract, adjusted to pH 7.2, peptone glycerol (PG) medium containing 10 g\/L peptone, 20 g\/L glycerol, 2.5 g\/L NaCl, 0.5 g\/L MgSO4\u03877H2O, 0.25 g\/L KH2PO4, 0.5 g\/L CaCl2\u03872H2O, adjusted to pH 7.0,30 and soybean glycerol (SG) medium containing 15 g\/L soybean meal, 20 g\/L glycerol, 2.5 g\/L NaCl, 0.5 g\/L MgSO4\u03877H2O, 2 g\/L NaNO3, adjusted to pH 7.0.27 The growth cultures were incubated at 28 \u00b0C, 150 rpm. After 5 d, 20 mg of 12 was added to each culture to directly feed into the azomycin biosynthetic pathway, grown for an additional 2 d at 28 \u00b0C, 150 rpm, and then analyzed for production of 6 as follows. To extract metabolites the culture was lowered to pH 2 using concentrated HCl and then centrifuged to remove the mycelium. The supernatant was extracted with 50 mL of ethyl acetate three times. The organic phases were collected and then the ethyl acetate was removed by rotary evaporation. The residual solid was resuspended in 1 mL DMSO and diluted ten-fold with MeOH. The resulting mixture was analyzed by LC-MS using a Luna C18, 5 \u00b5m, 4.6 mm ID \u00d7 250 mm column. Elution was performed at 0.5 mL\/min using a mobile-phase consisting of a linear gradient of water and ACN ((v\/v): 90:10 0 to 5min, 90:10 to 0:100 5 to 6 min, 0:100, 6 to 10 min, 0:100 to 90:10, 10 to 11 min, 90:10, 11 to 15 min), with both solvents containing 0.1% (v\/v) FA. Selected ion monitoring of m\/z 84 and m\/z 114 was used to specifically detect production of 12 and 6, respectively. 3.2.14 Cultivation and extraction of metabolites from Pseudomonas  The Pseudomonas strains P. syringae pv. tomato str. DC 3000 and P. brassicacearum DF41, both harbouring putative azomycin biosynthetic gene clusters were obtained from the lab of Cara Haney (Michael Smith Labs, UBC). To detect the production of 6 in these Pseudomonas strains, bacterial cells of both strains were used to inoculate 50 mL liquid cultures of both LB and 102  King\u2019s medium B (KB). The latter medium consisted of 20 g\/L peptone, 1.5 g\/L K2HPO4, 1% (v\/v) glycerol, and 5 mM MgSO4. The cell cultures were grown at 30 \u00b0C for either one, three, or five days. After the indicated time had elapsed the cultures were centrifuged to remove the cells, the supernatant was acidified to pH 2 with HCl, and then stored at -20 \u00b0C until extraction. For polar metabolite extraction, frozen supernatant samples were first thawed, then extracted with 50 mL of ethyl acetate twice. Both ethyl acetate fractions were combined, and then solvent was removed by rotary evaporation. The residual solids were resuspended in 500 \u03bcL DMSO. Prior to analysis samples were diluted two-fold with MeOH. For LC-MS analysis, 5 \u03bcL of the MeOH diluted sample was separated using a Luna C18, 5 \u00b5m, 4.6 mm ID \u00d7 250 mm column (Phenomenex), and was eluted at 0.5 mL\/min using a mobile phase consisting of a linear gradient of water and ACN ((v\/v): 90:10, 0 to 5 min, 90:10 to 0:100, 5 to 6 min; 0:100, 6 to 10 min, 90:10, 10 to 11 min; 90:10 11 to 15 min), with both solvents containing 0.1% (v\/v) FA.  3.3 Results 3.3.1 Bioinformatic analysis of a cryptic gene cluster To better understand the distribution and genomic context of RohP-like arginine hydroxylases I conducted a BLASTp analysis using RohP as a template. I screened out all the identified homologs from indolmycin or L-enduracididine gene clusters and used this curated sequence database to construct a phylogenetic tree (Figure B.1). This phylogenetic tree revealed that RohP homologs align into multiple subgroups, with each major branch of the tree linked to a distinct gene cluster (Figure 3.2). In all but one case, rohP was found in-frame with four other predicted genes: rohQ, a hypothetical protein; rohR, a dihydrodipicolinate synthase; rohS, an iron oxidase; and rohT, a Rieske ferredoxin (Figure 3.2, light blue, Table B.6). Two discrete copies of this cryptic five gene cluster were also found in the recently deposited genomic data for the azomycin (6) producer Streptomyces eurocidicus ATCC 27428 (Table 3.1, Figure 3.3), and copies of this gene cluster in genomes of several Pseudomonas and Burkholderia strains (Figure 3.2, Figure B.1). I hypothesized that this five-gene region could be the long-elusive azomycin biosynthetic gene cluster. 103   Figure 3.2 Organization of the azomycin biosynthetic gene cluster in different soil-dwelling bacteria. Putative azomycin biosynthetic genes are indicated in light blue with rohP-homologs indicated by diagonal striping.   Figure 3.3 Dual azomycin biosynthetic gene clusters in Streptomyces eurocidicus ATCC 27428. The NCBI accession numbers for the accessory genes shown and the genes comprising each cluster are indicated. 104  Table 3.1 Bacterial strains with dual copies of the azomycin biosynthetic gene cluster.  Bacterial strain NCBI accession number of rohP homolog 1 NCBI accession number of rohP homolog 2 Streptomyces eurocidicus ATCC 27428 WP_102919045.1 WP_102917515.1 (62) Streptomyces albireticuli MDJK11 WP_087925758.1 (97) WP_087924922.1 (80) Streptomyces albireticuli NRRL B-1670 WP_095581505.1 (94) WP_095584167.1 (78) Streptomyces griseocarneus strain 132 WP_121801107.1 (78) WP_121803594.1 (78) Streptomyces exfoliatus DSM 41693 WP_024758609.1 (77) WP_024754986.1 (76) Streptomyces olivioreticuli ATCC 31159 WP_116214779.1 (76) WP_116210246.1 (79) Values in parenthesis are the amino acid sequence percent identities to each corresponding S. eurocidicus seqeunce, WP_102919045.1 for homolog 1, and the percent identities to RohP from S. cattleya for homolog 2. Cells are shaded according to the type of cluster containing the rohP homolog, using the same color scheme as Figure 3.2 and Figure B.1, with blue being S. eurocidicus-type, green being S. cattleya-type, and red being a redox-type.  3.3.2 In vitro analysis of the aldolase RohR To determine the fate of the RohP product 14, I examined the in vitro activity of RohR. RohR is annotated as a dihydrodipicolinate synthase (DHDPS), which catalyzes the condensation of 15 and L-aspartate semialdehyde (112) to (2S,4S)-4-hydroxy-2,3,4,5-tetrahydrodipicolinate (HTPA) (113) in the biosynthesis of L-lysine (70).191 Several closely related enzymes have been shown instead to catalyze retro-aldol reactions, generating pyruvate (15) and the corresponding aldehyde from various \u03b1-ketoacid-containing substrates (Figure 3.4, Figure 3.5).192 105   Figure 3.4 Diverse reactions catalyzed by dihydrodipicolinate-like enzymes.192 106   Figure 3.5 Sequence alignment of RohR compared to DHDPS and other characterized retro-aldolases. The sequences of E. coli DHDPS (PDB ID: 1DHP), human HOGA (PDB ID: 3S5O), and Thermoproteus tenax KDPGA (PDB ID: 2R91) were used to construct the alignment. The conserved catalytic lysine residue is highlighted in red. Sequence alignment generated with Clustal Omega,193 and visualized with ESPript 3.0.194 107  To test whether RohR could catalyze a retro-aldol cleavage of 14 into pyruvate (15) and guanidinoacetaldehyde (17), purified RohR was added into the RohP + 13 assay with catalase. Relative to the control reaction between 13 and RohP (Figure 3.6a), I observed the disappearance 14 with the concurrent appearance of two new peaks with an m\/z 102 and m\/z 84, the expected [M+H]+ of 17 and 12 respectively (Figure 3.6b). To confirm that the peak at m\/z 84 was 12, the reaction mixture was derivatized with DNS-Cl and compared its retention time and mass to authentic dansylated 12 (Figure 3.6c). Additionally, OPD derivatization of the \u03b1-keto acids present in the RohP + RohR coupled reaction confirmed the presence of 15 (Figure 3.6d) and provided additional evidence for the consumption of 14 by RohR (Figure 3.7). These results demonstrate that RohR catalyzes the retro-aldol-like cleavage of 14. Like other enzymes in this family, RohR is likely to catalyze this reaction by formation of a Schiff base with 14, followed by deprotonation of C\u03b3-OH to initiate cleavage of the C\u03b2-C\u03b3 bond, releasing 17, which should spontaneously cyclize to 12 through dehydration (Figure 3.8)   Figure 3.6 RohR- and RohQ-catalyzed production of 2-aminoimidazole and pyruvate. a) ESI-MS analysis of the reaction between L-arginine (13) and RohP with catalase. b) ESI-MS analysis of the reaction between L-arginine (13), RohP, and RohR with catalase. c) LC-MS of the RohP + RohR reaction after derivatization with DNS-Cl. Selected ion monitoring (SIM) at m\/z 317. d) LC-MS of the RohP + RohR reaction after derivatization with OPD. SIM at m\/z 161. 108   Figure 3.7 OPD analysis of the products of the RohP catalyzed oxidation of L-arginine. i) LCMS analysis of the reaction containing RohP, ii) LCMS analysis of the reaction containing RohP + RohR. The product (S)-4-hydroxy-2-ketoarginine (14) is not detected when RohR is present in the reaction. A peak with an m\/z 244 that co-elutes with OPD-2-ketoarginine (105), which has been shown previously to be OPD-4,5-dehydro-2-ketoarginine (104) can also be detected here.152 Compound 14 and 105 are normally produced in a 1:2 ratio by RohP,186 and thus the lower than expected abundance of OPD-14 can be attributed to the dehydration of 14 to 104 during the derivatization process.  109   Figure 3.8 Possible mechanism for retro-aldol cleavage by RohR.  110  3.3.3 RohQ catalyzes a spontaneous cyclization reaction  Figure 3.9 RohQ catalyzed production of 2-aminoimidazole. a) LC-MS of the RohP + RohR reaction (solid line) and RohP + RohR + RohQ reaction (dashed line). Traces represent the SIM of m\/z 102. b) LC-MS of the RohP + RohR reaction (solid line) and RohP + RohR + RohQ reaction (dashed line). SIM of m\/z 84. c) Integrated peak area for m\/z 102 from the filtrate of the RohP + RohR catalyzed reaction with RohQ added. d) Integrated peak area for m\/z 84 from the filtrate of the RohP + RohR catalyzed reaction with RohQ added.  Despite lengthy overnight reaction times, complete conversion of 17 to 12 was not observed in the RohP + RohR reaction, as residual 17 remained present in the reaction mixtures (Figure 3.6b). As 17 was also not reported to accumulate in vivo,27 it seemed possible that there was a mechanism acting to prevent the build-up of 17. RohQ homologs are only encoded in putative azomycin gene clusters. This result and the absence of any functional annotations to guide 111  led to the hypothesis that the role of RohQ could be to convert 17 to 12. When RohQ, RohR, and RohP were all included with 13, a signal for 17 at m\/z 102 was no longer detected (Figure 3.9a). The absence of 17 coincided with a clear and reproducible increase in the signal at m\/z 84, which corresponds to 12 (Figure 3.9b, Figure 3.10). A   B  Figure 3.10 LC-MS quantification of RohQ-catalyzed cyclodehydration of guanidinoacetaldehyde. A) Selected-ion monitoring of the intermediate guanidinoacetaldehyde (17) in reactions lacking or containing RohQ (as indicated). No signal at m\/z 102 could be detected when RohQ was included in the assay, therefore the peak areas in these experiments were assigned an area of 0. B) Selected-ion monitoring of 2-aminoimidazole (12). Reactions were carried out in 20 mM MOPS, 50 mM 112  NaCl, pH 7.0 buffer, and 1 \u03bcL of the quenched sample was injected for MS analysis. Peak areas were extracted for each of the SIM signals. Average peak areas and standard deviations were calculated using the results from n = 5 samples.  A  B  Figure 3.11 Initial product distributions for the a) RohP + RohR coupled reactions and b) RohP + RohR + RohQ coupled reactions with L-arginine. Peak areas were calculated by integrating the signals from the extracted ion chromatograms for m\/z 84 and m\/z 102 in each experiment. 113  To better understand the cyclization of 17, the product distribution of RohR over time was also examined. When RohQ is absent, 17 is the most abundant product, while 12 slowly accumulates over time (Figure 3.11a). Conversely, when RohQ is present, the dominant product is always 12, with no significant build-up of 17 (Figure 3.11b). As the intrinsic reactivity of 17 prevents its isolation, 17 was generated in situ using RohP and RohR. RohP and RohR were then removed by filtration and RohQ was added to the filtrate. Under these conditions RohQ catalyzed a decrease in 17 (Figure 3.9c), which corresponded with a simultaneous increase in 12 (Figure 3.9d). Altogether, these results indicate that RohQ catalyzes the conversion of 17 to 12. The unusual role for RohQ helps to explain the absence of cofactor binding sites and the lack of RohQ homologs in other biosynthetic pathways. Among all rohP-containing gene clusters there was only a single occurrence of a putative azomycin gene cluster lacking a rohQ homolog (Figure 3.12), consistent with the importance of the role of RohQ in promoting a spontaneous reaction.   Figure 3.12 Streptomyces griseocarnus strain 132, lacking a rohQ homolog. NCBI accession numbers for the genes depicted are as follows: WP_121801246.1 (rhodanese), WP_121801106.1 (polyisoprenoid binding protein), WP_121801107.1 (rohP homolog), WP_121801108.1 (rohR homolog), WP_121801109.1 (rohT homolog), WP_121801247.1 (rohT homolog), WP_121801110.1, (methyltransferase), and WP_121801111.1 (copper oxidase).  3.3.4 Analysis of nitroimidazole formation, catalyzed by RohS Based on the results of Lancini and Eguchi, the next step in the biosynthesis of 6 is the oxidation of 12.27\u201329 The best candidate to catalyze such a transformation was the predicted non-heme diiron oxidase RohS, as other non-heme diiron enzymes AurF and CmlI have both been shown to catalyze similar aryl-amine to aryl-nitro oxidations.66,69,73 Interestingly, AurF and CmlI are proposed to use different oxidative pathways to catalyze their respective reactions (Sections 1.2.4.2, 1.2.4.3).10,195 Additionally, the Rieske oxygenase PrnD employs a Rieske [2Fe-2S] cluster and mononuclear iron center to catalyze an aryl-amine oxidation.61 However, functional annotations assign RohS to a different protein family than any of these N-oxygenases, placing 114  RohS in the same family as the diiron domain of StzF from the biosynthetic pathway to streptozocin (Figure 3.13). This domain hydroxylates the guanidium nitrogens of the arginine-derived compound N\u03c9-monomethyl- L-arginine (114).196,197 A sequence alignment between RohS, known aryl-amine oxygenases, and the diiron domain of StzF revealed that while there is limited overall homology between RohS and the other aryl-amine oxygenases, the putative diiron binding sites in StzF are closely related to those in RohS (Figure 3.14).  Figure 3.13 N-oxygenases that have been experimentally characterized. Diiron dependent aryl-amine oxygenases from PF11583 including AurF catalyze the direct oxidation of an aromatic amine to the corresponding aromatic nitro group.64\u201366,69 The mono-iron, Rieske oxygenase PrnD from PF00355 catalyzes a similar oxidation in the biosynthesis of pyrrolnitrin (38).61 RohS, also catalyzes direct oxidation of an amine to nitro group, but is a member another family, PF14518, the non-heme iron oxygenases. StzF which catalyzes the N-oxidation of an arginine derivative is also a member of PF14518.196,197  115   Figure 3.14 Multiple sequence alignment of RohS and other discussed N-oxygenases. This alignment includes the mono-iron binding domain of PrnD (residues 149-363), the di-iron domain of StzF (residues 176-354), and the full-length sequences of RohS and of the aryl-amine oxygenases CmlI, AurF, and ObiL. The predicted iron coordinating residues of StzF D214, E215, H225, E281, H311, H313, D315, and H318 are indicated by red arrows. Sequence alignment generated with Clustal Omega,193 and visualized with ESPript 3.0.194  116  Initial attempts to utilize RohS from S. cattleya (SCyRohS) in in vitro assays were unsuccessful as it readily precipitated upon inclusion in in vitro assays. Therefore, the gene encoding an RohS homolog from Kitasatospora azatica (KAzRohS, 71% identity to SCyRohS) was synthesized, and cloned into pET28a for recombinant expression in E. coli. KAzRohS was found to catalyze the oxidation of 12 to 6 when phenazine methosulfate (PMS), NADH, and FeSO4 were present in the reaction mixture (Figure 3.15). This reaction also exhibits a strict dependence on iron (Figure 3.15, Figure 3.16). ICP-MS analysis of KAzRohS revealed very low levels of bound iron (Table B.4), consistent with the requirement for the addition of exogenous iron to the reaction mixture and consistent with other studies on StzF.196\u2013198 Michaelis-Menten kinetic parameters (Equation 1) for the KAzRohS-catalyzed oxidation of 12 were also obtained, determining a KM of 214 \u03bcM, and a kcat of 0.314 min-1 for this reaction (Figure 3.17). Comparing KAzRohS to the aryl-amine oxygenases reveals that it catalyzes this oxidation ~20 fold slower than both AurF69 and PrnD.61 This lower turnover may be partially due to the low level of iron-binding by KAzRohS, even in the presence of excess iron.   Figure 3.15 In vitro production of azomycin. LC-MS analysis of the oxidation 2-aminoimidazole (12) to azomycin (6) by KAzRohS. The full reaction contains 2-aminoimidazole (12), FeSO4, PMS, and NADH.    117   Figure 3.16 ESI-MS of KAzRohS in vitro reactivity in the presence of various transition metals. Traces depict the extracted ion chromatograms for m\/z 114 for azomycin (6).  Figure 3.17 KAzRohS steady state kinetic analysis with respect to its substrate 2-aminoimidazole. Data points were fit using non-linear regression in GraphPad Prism v. 5.04 to obtain kinetic parameters.  3.3.5 RohT may play a role in the RohS-catalyzed oxidation of 2-aminoimidazole The last enzyme RohT was predicted to contain a Rieske [2Fe-2S] cluster (Figure 3.18). Multi-domain Rieske oxygenases utilize Rieske [2Fe-2S] clusters to transfer electrons to their iron-118  centers.199 Similarly, smaller independent Rieske ferredoxins are an integral component of many oxidative pathways, where they also transfer electrons directly to the iron center of another iron-dependent oxygenase.200,201 As such, I hypothesized that RohT and RohS may work together in the biosynthesis of 6.  Figure 3.18 Sequence alignment of Rieske proteins discussed in the text. Sequence alignment of the Rieske domain of PrnD (residues 26-148), the independent Rieske proteins StzG and RohT, and the characterized Rieske ferredoxin BphF.[2] The conserved residues of the Rieske motif CXHX15-17CXXH are highlighted in the red box. RohT displays an overall sequence similarity of 23% to the Rieske domain of PrnD, 35% to StzG, and 21% to BphF. Sequence alignment generated with Clustal Omega,193 and visualized with ESPript 3.0.194  I attempted to reconstitute the [2Fe-2S] cluster of RohT by co-expression in E. coli containing the plasmid pPAISC-1, which contains iron-sulfur cluster assembly genes.202 RohT purified using this expression system was brown (Figure 3.19a), however ICP-MS analysis revealed less than one equivalent of iron bound, indicating only partial reconstitution of the [2Fe-2S] cluster (Table B.5). Attempts to reconstitute the [2Fe-2S] cluster in vitro via dialysis under anaerobic conditions with Fe2+ and S2- were not successful (Figure B.3). Purified RohT was also reduced using several common reductants under aerobic conditions, but no changes in the UV-Vis absorbance spectrum of RohT were observed (Figure 3.19b). Nor were any changes observed in 119  the spectrum when the reductions took place in a glove box under anaerobic conditions (Figure B.4). All the obtained spectra look virtually identical, suggesting that chemical reductants are unable to alter the oxidation state of RohT when purified from aerobic cell cultures. Nonetheless, I prepared several assays to see if RohT changed the outcome of the oxidation of 12 catalyzed by RohS. Under these conditions it was found that RohT could not catalyze the oxidation of 12 to 6 on its own (Figure B.5). Furthermore, inclusion of RohT did not alter the amount of 6 produced by KAzRohS (Figure B.6). Production of 6 by KAzRohS in reaction mixtures with either RohT and reducing agents could also not be detected (Figure B.7). Additional non-cognate ferredoxin reductases were also tested to reduce RohT enzymatically. Again, the production of 6 was not observed under the tested conditions (Figure B.8; Figure B.9). A similar lack of in vitro activity was also reported for the related Rieske ferredoxin from the streptozotocin biosynthetic pathway.197  Figure 3.19 Purification and reduction of RohT. A) Purified RohT from co-expression with the iron-sulfur cluster assembly system. B) UV-Vis absorbance spectra of RohT (110 \u03bcM) as purified (oxidized), and after incubation with 2 mM of various reductants for 15 minutes under normal aerobic conditions.  Despite the inability to reconstitute the activity of RohT in vitro, the activities of the other four enzymes characterized should be sufficient for the production of 6 in a one-pot reaction. To test this all four enzymes were incubated together in a one-pot assay with 13, catalase, FeSO4, PMS, and NADH to reconstitute the full biosynthetic pathway from 13 to 6 in vitro (Figure 3.20). Detection of 6 in the reaction mixtures containing the four enzymes RohPQRS confirmed that these enzymes are sufficient to produce 6 in vitro. 120    Figure 3.20 One pot production of azomycin. LC-MS analysis of the one-pot reaction of L-arginine (13), FeSO4, PMS, NADH, catalase, RohP, RohQ, RohR, and KAzRohS. All traces are SIM m\/z 114 for azomycin (6).  3.3.6 Probing for azomycin production by Streptomyces cattleya To determine whether 6 is produced by S. cattleya during fermentation, S. cattleya was cultured in glucose yeast maltose (GYM), peptone glycerol (PG), or soy glycerol (SG) medium, which were also individually supplemented with 12. The bacterial cultures were extracted with ethyl acetate two days after supplementation with 12 (seven days in total), a timepoint previously used to isolate 6 from producing strains.28,29 However, 6 was not detected, even in cultures supplemented with 12 (Figure 3.21). The small peak in the PG + 12 was determined not to be 6, due to the shifted retention time and different absorbance spectrum. These results suggest that the gene cluster in S. cattleya is silent, or that another, yet-unknown molecule is produced instead, using the accessory enzymes encoded by genes adjacent to the five-gene cluster. It should also be noted that the 6-producing strain S. eurocidicus has two copies of the minimal five-gene cluster (Table 3.1, Figure 3.3), albeit each flanked by different accessory enzymes, suggesting that one of these two gene clusters is linked to 6 production in vivo. 121   Figure 3.21 Ethyl acetate extractions of Streptomyces cattleya fermentations. Signals represent SIM at m\/z 114 for azomycin (6).  3.3.7 Probing for azomycin production by Pseudomonas Pseudomonas are also well established as a rich source of secondary metabolites.203 Many of these compounds play critical roles in signalling and regulation of the complex processes of the soil rhizosphere.204 Coincidentally, one of these metabolites isolated from Pseudomonas is azomycin (6), having been detected in a Japanese isolate of P. fluorescens.205 This result is supported by my bioinformatic studies which revealed that putative azomycin biosynthetic clusters are distributed amongst multiple Pseudomonas strains (Figure 3.2; Figure B.1). Therefore, I was interested to see if 6 was more widely produced by other Pseudomonas strains.  In the absence of that particular Japanese P. fluorescens strain, two other Pseudomonas strains with putative copies of the azomycin biosynthetic gene cluster were obtained (Figure 3.22). In both P. syringae pv. tomato str. DC 3000 and P. brassicacearum DF41, the azomycin biosynthetic genes are flanked on either side by genes encoding enzymes that most likely occupy either regulatory or signaling roles and are unlikely to produce modified azomycin derivatives. 122     Figure 3.22 Azomycin gene cluster organization in Pseudomonas.   To see if either Pseudomonas strain produced 6, both P. syringae pv. tomato str. DC 3000 and P. brassicacearum DF41 were cultured in both LB and King\u2019s medium B (KB) for either 1, 3, or 5 days. The metabolites from these cultures were extracted with EtOAc, and the extracts analyzed using LC-MS. Compound 6 was not detected in any of the P. syringae pv. tomato str. DC 3000 isolates. However, isolates from P. brassicacearum DF41 proved more productive, as 6 was detected in several isolates from both KB and LB media (Figure 3.23). In both instances, the production of 6 appears to exhibit a moderate time dependence, with higher amounts detected upon longer 3- and 5-day growths.   These results indicate a link between the presence of the putative five gene azomycin biosynthetic gene cluster and production of 6 in vivo by Pseudomonas. Further gene deletion and genetic complementation studies will be required to confirm the link between the gene cluster and the production of 6. Nonetheless, these results are exciting as it demonstrates that 6 may be produced by additional strains of soil bacteria, not just a select few, and so 6 may be abundant in the rhizosphere. Additionally, if other strains of soil bacteria are producing 6, then it may have other role in addition to its antibiotic properties. 123   Figure 3.23 LC-MS analysis of Pseudomonas brassicacearum DF41 ethyl acetate extracts. a) EtOAc extracts of P. brassicacearum DF41 grown in KB medium. b) EtOAc extracts of P. brassicacearum DF41 grown in LB medium. Time of bacterial culture growth is indicated for each trace. Signal is the extracted ion chromatogram at m\/z 114 for detection of azomycin (6). 3.3.8 Other proposed biosynthetic routes to azomycin The five gene cluster that produces 6 is supported by additional data found in a report to the Department of Energy of the United States Government.206 In this report, David Graham and coworkers describe using combined genomic and proteomic approaches to identify one of these 124  gene clusters, designated aznABCDE, whose expression levels are correlated with the biosynthesis of 6 in S. eurocidicus. They show that heterologous expression of aznABC in E. coli gives 12, which is consistent with in vitro experiments using purified RohPQR reported in this thesis. Additionally, they report that cell-free lysates with AznD, AznE, and DTT convert 0.07 % of 12 to 6 in air. Overall, these data support that this azn cluster is likely linked to production of 6 in S. eurocidicus. However, their detection of ions in the E. coli cell lysates by mass spectrometry leads them to propose a different biosynthetic route to 6 (Figure 3.24). Most significant, they propose that AznB (RohQ) dehydrates 14 to give a substrate for AznC (RohR) and that 17 is an off-pathway intermediate. This pathway is inconsistent with the in vitro experiments reported in this thesis, which show that RohR (AznC) reacts with 14, through a retro-aldol reaction, giving 17, which is then an on-pathway substrate for RohQ (AznB) (Figure 3.1). They also do not delineate the roles of AznD and AznE in the oxidation of 12 to 6, while I establish that the iron-dependent oxygenase RohS can catalyze this reaction in vitro with PMS and NADH.  Figure 3.24 Proposal for azomycin biosynthesis based on results from Graham and coworkers.  Scheme adopted from the report Graham et al.206 Compounds identified in this study are indicated by numbers, consistent with the numbering scheme presented in Figure 3.1.  3.4 Discussion My in vitro characterization of a pathway to 6 was guided by the biosynthetic scheme proposed by Eguchi ~40 years ago. I was able to identify all the originally proposed intermediates and identify enzymes responsible for catalyzing each step in the biosynthesis of 6. The described 125  analysis identified RohP as an O2- and PLP-dependent arginine oxidase, RohR as a retro-aldolase, RohQ as a novel cyclodehydratase, and RohS as a nitro-forming enzyme. The inaccurate functional annotation of these enzymes partially explains why this gene cluster remained unidentified for so long and underscores the importance of characterizing cryptic enzymes to discover rare biological activities.  Figure 3.25 Diverse routes to arginine-derived heterocycles involving O2- and PLP-dependent arginine oxidases.  The discovery of the azomycin biosynthetic machinery opens the possibility of engineering bacteria to produce nitroaromatic compounds, as S. eurocidicus has been shown to be capable of converting alkylated 2-aminoimidazole derivatives to the corresponding nitro-derivatives in vivo.28,29 These reactions are catalyzed by RohS, making it an attractive target for biocatalytic applications towards the production of 6, related nitroimidazoles, and possibly new nitro-compounds. Interestingly, the indolmycin, L-enduracididine, and azomycin biosynthetic pathways all use O2- and PLP-dependent arginine oxidases as a first step to generate diverse, bioactive 126  heterocyclic scaffolds (Figure 3.25). Therefore, the discovery of new arginine oxidases may lead to the identification of more arginine-derived heterocycles. 127  Chapter 4: Characterization of the alkyne-forming PLP-desaturase BesB: Mechanism of pyridoxal phosphate-dependent alkyne formation 4.1 Introduction Acetylenic compounds are molecules containing at least one carbon-carbon triple bond, and are important components of modern synthetic and medicinal chemistry.207,208 Presently, acetylenic compounds are extensively used in Cu(I)-catalyzed azide-alkyne \u2018click\u2019 chemistry to label and investigate important biological targets.90,209 Acetylenic compounds are also widespread in nature, as well over 2000 acetylenic and polyacetylenic natural products have been isolated from all branches of the tree of life. Several examples of these compounds are shown in Figure 1.16, Figure 1.19, and Figure 1.21.79 These natural products exhibit an impressive array of biological activities, including roles as antimicrobials, antifungals, insecticides, and signaling molecules.79,208 There are now several enzymes which have been shown to catalyze the formation of an alkyne bond. The first characterized acetylenase was the membrane-bound fatty acid acetylenase Crep1 from the seeds of the plant Crepis alpine.96,97 Crep1 utilizes a diiron cluster to catalyze the oxidation of linoleic acid (54) to crepenynic acid (53), the latter of which contains an internal alkyne (Figure 1.18, Figure 4.1a).96,97 Crep1-like enzymes have since been found to be an integral component of many plant and insect biosynthetic pathways.102,103 In 2015 similar membrane-bound di-iron acetylenases were also discovered in bacteria.110,111 However, the bacterial enzyme JamB instead oxidizes the terminal alkene of PCP-tethered 5-hexenoic acid into the 5-hexynoyl derivative, as a part of jamaicamide biosynthesis (Figure 4.1b).110,111 PCP-tethered alkyne derivatives produced by JamB were also used to generate novel alkyne-tagged natural products in situ.110,113 In 2019 the enzyme families capable of forming acetylenic natural products was expanded to include PLP-dependent enzymes.115 BesB, a close homolog of the PLP-dependent cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase enzyme family, was found to convert 4-chloroallylglycine (72) to L-propargylglycine (74), a rare terminal alkyne-containing amino acid.115 Additional biosynthetic modifications then convert 74 to the final biosynthetic product \u03b2-ethynylserine (69) (Figure 1.23, Figure 4.1c). Finally, and most recently, a pair of 1,3-enyne-forming cytochrome P450s, AtyI and BisI were simultaneously reported in 2020.105,106 AtyI and BisI catalyze a highly unusual oxidative transformation, converting 4-hydroxy-3-prenylbenzoic 128  acid (62) to eutypinic acid (60), the latter of which is a key intermediate in the biosynthesis of a number of different cyclohexanoid terpenoids, including asperpentyn (58) and biscognienyne B (59) (Figure 4.1d).105,106 These advances in acetylenic enzymology are covered in more detail in a recently published review and in Chapter 1.210  Figure 4.1 Characterized biosynthetic routes to acetylenic natural products. 129  Despite the recent discoveries of several new enzymes which catalyze either internal or terminal alkyne formation, a detailed understanding of the how these enzymes catalyze their respective reactions remains limited. For example, a sequence alignment of 38 diiron desaturases with 12 putative Crep1-like acetylenases identified 11 amino acid positions which were specific to the Crep1 acetylenases.96 Each of these 11 residues were mutated individually, revealing that five of the residues are critical for determining the whether the outcome of the Crep1-catalyzed reaction is a desaturation or acetylation (Table C.5). However, in the absence of a crystal structure or additional mechanistic data, it is still not entirely clear how this \u2018improbable\u2019 reaction proceeds.96,97 The mechanism of the P450 AtyI-catalyzed reaction was probed using a number of synthetic substrates, designed to mimic possible intermediates in the AtyI-catalyzed reaction (Figure C.1). The synthesized molecules containing either a trans-alkene or allene were found to not be substrates of AtyI, while substrates containing a cis-alkene were.105 Based on these results the authors proposed that AtyI initially abstracts an allylic hydrogen from 4-hydroxy-3-prenyl benzoic acid, which triggers a bond rotation to form a cis-alkene containing intermediate.105 But once again, there is still much that is not known about how AtyI and BisI catalyze this reaction. Therefore, it is clear that there still remains much that can be learned about these acetylenic biosynthetic enzymes and discovering more about these enzymes will facilitate the development of these enzymes as novel biocatalysts. To further advance what is known about alkyne enzymology, I focused on the unusual PLP-dependent desaturase BesB, which catalyzes the conversion of 72 to 74 (Figure 4.1c).115 Based on the amino acid sequence of BesB, it was classified as a member of the PLP-dependent cystathionine-\u03b2-lyase (CBL)\/cystathionine-\u03b3-synthase (CBS) protein family,115 as it contains most of the conserved PLP-binding residues found in other members of the CGL\/CGS family (Figure C.2). It was observed that BesB incorporates two atoms of deuterium into 74 when the reaction was carried out in D2O, which is also consistent with the reaction proceeding through a typical PLP-dependent CBL\/CGS-like catalytic mechanism. In the proposed BesB mechanism, the conserved PLP-binding lysine catalyzes the initial steps of the reaction, and a second unknown base is proposed to deprotonate an allenic PLP-intermediate to form the alkyne bond (Figure 4.2).115  130   Figure 4.2 Mechanistic proposal for the BesB-catalyzed reaction. Proposed solvent deuterium incorporation pattern indicated with blue D\u2019s. Adopted from Marchand et al.115  131  However, the mechanistic proposal shown in (Figure 4.2) conflicts with data from several earlier studies. These studies from the 1970\u2019s and 80\u2019s established that PLP-dependent \u03b3-lyases (other members of this same protein family) are irreversibly inhibited by both 72 and 74.136,138,139,211 Based on the observed inhibition data, it was proposed that in the cases of both 72 and 74 the enzyme reacts with either molecule to generate a PLP-external aldimine intermediate containing an allene. The allene can then react with an active site nucleophile, forming a new covalent bond between the enzyme and the PLP-external aldimine, resulting in the enzyme becoming irreversibly inhibited (Figure 1.34). Additional support for this proposal can be found in a 2008 crystal structure of a cystathionine \u03b3-lyase, in which 74 is covalently bound to a tyrosine residue in the active site (Figure 1.35).140 Based on the available evidence, it seemed probable that BesB may have some unique structural features or utilize a modified catalytic mechanism that prevents it from becoming inhibited by any reactive intermediates which are generated over the course of the enzymatically catalyzed reaction. To obtain a better understanding of how BesB functions, I overcame the limited solubility of BesB in E. coli through use of a Rhodococcus heterologous expression system, which has been previously shown to be useful for other Streptomyces enzymes. Using Rhodococcus, the BesB homolog from S. achromogenes was expressed in the soluble form, enabling the characterization of this BesB homolog in vitro. I was able to determine that in vitro the BesB-catalyzed reaction is strongly affected by pH as well as the presence (or lack) of chloride ions in solution. Furthermore, BesB also retains some residual cystathionine-\u03b3-synthase activity. I also obtained high-resolution crystal structures of BesB and a catalytically deficient mutant, revealing that BesB adopts a typical cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like protein fold. BesB achieves this fold using a unique N-terminal sequence. An additional insight gleaned from the crystal structure of BesB is that the catalytic tyrosine residue conserved in other cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthases has been substituted with a phenylalanine residue, thereby preventing BesB from becoming inhibited by reaction intermediates. Taken together these changes make BesB a unique member of the CBL\/CGS protein family, perfectly suited to catalyze the formation of highly reactive unsaturated PLP intermediates without becoming inhibited. The improved understanding of this type of PLP-dependent chemistry could lead to the development of new PLP-dependent enzymes which can be utilized to generate unusual desaturated amino acids. 132  4.2 Materials and methods 4.2.1 General methods Primers were purchased from Integrated DNA Technologies. Gene constructs were synthesized by BioBasic Inc. DNA sequencing was performed by the DNA Sequencing Core Facility at the Center for Molecular Medicine and Therapeutics (The University of British Columbia). 4-bromoallylglycine (73) and L-propargylglycine (74) were acquired from AA blocks (San Diego, USA), and 4-chloroallylglycine (72) was generously provided by Michelle Chang. O-acetylhomoserine (83) was purchased from Carbosynth. O-succinylhomoserine (84), and L-cystathionine (85) were purchased from Millipore Sigma. Additional reagents and chemicals were purchased from Alfa Aesar, Anatrace, Bio Basic Inc., Chem Impex Ltd., Hampton crystallography, Millipore Sigma, and Thermo Fischer Scientific Canada.  4.2.2 Cloning and protein expression The DNA sequence of BesB was amplified by PCR directly from spores of S. achromogenes subsp. achromogenes B-2120 (DSMZ 40028) that had been suspended in DMSO. Primers for this PCR, and other DNA amplifications can be found in Table C.2. The amplified DNA sequence as well as vector were then separately digested with the appropriate restriction enzymes: NdeI and SacI for pET28a, or NdeI and HindIII for pET22b and pTIPQC1. Similarly, synthesized gene sequences were also digested with NdeI and HindIII. The vectors were also treated simultaneously with Calf Intestinal Alkaline Phosphatase. All digested DNA was purified by gel extraction prior to ligation. The digested products were ligated together overnight at 16 \u00b0C using T4 ligase. The ligation mixture was used to transform chemically competent E. coli DH5\u03b1 cells. The correct nucleotide sequence was verified by Sanger sequencing. Additionally, the site-directed mutagenesis primer pairs also listed in Table C.2 were used to create BesB variants using Quikchange PCR-based protocols,212 with pTIPQC1:BesB serving as the template for the PCR reaction. The sequences of each BesB variant were also verified using Sanger sequencing. The pET28a:BesB and pET22b:BesB constructs were then used to transform chemically E. coli BL21 (DE3) cells for recombinant protein production. All pTIPQC1:BesB constructs were transformed into Rhodococcus sp. RHA1 for recombinant protein production by means of electroporation using a 2 mM gapped cuvette and an electric pulse of 2.4-kV.213 The Rhodococcus 133  transformations were plated onto LB containing 25 \u03bcg\/mL chloramphenicol and grown for 3 days at 30 \u00b0C. E. coli cells containing pET28a:BesB and pET22b:BesB vectors were used to inoculate 50 mL of LB, and were grown overnight 37 \u00b0C, 200 rpm. Then 15 mL of the overnight culture was used to inoculate 750 mL of LB (3 L in total) containing the appropriate antibiotic (50 \u03bcg\/mL kanamycin or 100 \u03bcg\/mL ampicillin). These cultures were grown at 37 \u00b0C, 200 rpm to an OD600 of 0.8\u20131.0, and then cooled to 16 \u00b0C. Protein expression was induced with 0.1 mM IPTG, and the cultures were grown for an additional 16 h at 16 \u00b0C. Cells were harvested by centrifugation and frozen at \u201320 \u00b0C until protein purification.  Rhodococcus sp. RHA1 cells containing pTIPQC1:BesB vector were used to inoculate 100 mL of LB containing 25 \u03bcg\/mL chloramphenicol, and were grown at 30 \u00b0C, 200 rpm for 1 d. Then 15 mL of the starter culture was used to inoculate 750 mL of tryptic soy broth (TSB) (3 L in total), containing 17 g\/L tryptone, 3 g\/L soytone, 5 g\/L NaCl, 2.5 g\/L K2HPO4, and 2.5 g\/L glucose also containing 25 \u03bcg\/mL chloramphenicol. These cultures were grown for an additional day at 30 \u00b0C, 200 rpm. Then the cells were induced by adding 25 \u03bcg\/mL thiostrepton. The cultures were then further grown at 30 \u00b0C, 200 rpm for a third day. All variants were also expressed using this protocol as well. The cells were harvested by centrifugation at 6000 rpm, and stored at -20 \u00b0C until protein purification.  For recombinant expression of selenomethionine-derivatized BesB using Rhodococcus sp. RHA1, 15 mL overnight starter culture in LB containing 25 \u03bcg\/mL chloramphenicol was used to inoculate 750 mL of M9 minimal media (3 L in total), containing 6.8 g\/L Na2HPO4, 3 g\/L KH2PO4, 1 g\/L NH4Cl, 0.5 g\/L NaCl, 0.4 % (w\/v) glucose, 2 mM MgSO4, 1 mM CaCl2, and 25 \u03bcg\/mL chloramphenicol. These cultures were grown for two days at 30 \u00b0C, 200 rpm. Protein expression was then induced by the addition of 25 \u03bcg\/mL thiostrepton, and 0.11 g (1 mM) of selenomethionine, with an additional 10 mg of PLP also added to the cell cultures at the time of induction. The cultures were then further grown at 30 \u00b0C, 200 rpm for one day at which point the cells were harvested by centrifugation at 6000 rpm and stored at -20 \u00b0C until protein purification.  134  4.2.3 Protein purification For protein purification cells were thawed, and then resuspended in binding buffer consisting of 20 mM Tris, 50 mM NaCl, pH 8. For purifying protein from Rhodococcus sp. RHA1 the binding buffer also contained 1 mg\/mL lysozyme. Cells were then lysed by sonication or the combination of sonication and lysozyme, respectively. The lysate was centrifuged at 12000 rpm for 40 mins, inverted to mix the lysate, and centrifuged for an additional 20 mins at the same speed. The clarified lysate was passed through ~1 mL of Chelating Sepharose\u2122 Fast Flow resin charged with NiCl2\u03876H2O to isolate the tagged protein. The column was washed with a stepwise gradient of binding buffer containing 5-100 mM imidazole. The protein was eluted from the column using 5 mL of elution buffer containing 300 mM imidazole and was orange in color. The protein containing fraction was then loaded onto a HiLoadTM Superdex 16\/600 200 pg column equilibrated with 20 mM Tris, 50 mM NaCl, pH 8 buffer. The protein was eluted from the column using a flow rate of 1 mL\/min. Protein containing fractions were combined and dialyzed overnight against 20 mM Tris, 50 mM NaCl, pH 8 buffer containing excess PLP. The following day unbound PLP was removed by further dialysis against 20 mM Tris, 50 mM NaCl, pH 8 buffer. After size exclusion chromatography the enzyme was stable for several days when stored at 4 \u00b0C and could be frozen and thawed without much loss of soluble enzyme, which enabled in vitro activity assays on stored enzyme. The steps for purifying SeMet-containing BesB are the same as described above, only that all buffers used also contained 2 mM \u03b2-mercaptoethanol.   4.2.4 Protein purification of halide free protein For preparation of BesB in buffer solution containing no halides the BesB-containing Rhodococcus cells were lysed and purified as previously described in 20 mM Tris, 50 mM NaCl, pH 8 buffer. For the purification step involving FPLC the protein was loaded onto the same HiLoadTM Superdex 16\/600 200 pg column equilibrated instead in 20 mM phosphate, 50 mM NaCl pH 8 buffer. The protein containing fractions were combined and dialyzed extensively against 75 mM phosphate pH 8 buffer, containing 15 % (v\/v) glycerol in order to remove the Cl\u2212 from the purification buffers. The initial dialysis buffer also contained excess PLP. The pH of this phosphate buffer was adjusted using H3PO4 instead of HCl. After this the protein was concentrated by ultracentrifugal filters and used for in vitro assays. 135  4.2.5 Initial in vitro biochemical assays of BesB, The reaction mixtures (100 \u03bcL) contained 1 mM of 72 or 1 mM of 73, 5 \u03bcM of BesB purified normally in Tris buffer, in either 20 mM MES (pH 6), MOPS (pH 7), HEPES (pH 8), or CHES (pH 9) buffer with 50 mM NaCl. The reactions were allowed to proceed at overnight at room temperature, then quenched with 100 \u03bcL of ACN, and briefly centrifuged to remove precipitate. For LC-MS analysis of all samples was performed using a 1260 HPLC apparatus (Agilent) coupled to a 6120 Quadrupole LC\/MS system (Agilent), operated in positive ion mode. For these reactions, 5 \u03bcL of supernatant was injected into the system equipped with a Syncronis\u2122 HILIC column (5 \u00b5m, 4.6 mm ID \u00d7 250 mm), and eluted using an 20:80 (v\/v) mixture of 10 mM of NH4OAc and 0.15% (v\/v) FA in H2O: 100% ACN at a flow rate of 1 mL\/min. For monitoring the pH dependence of halide exchange, reaction mixtures (100 \u03bcL) contained 1 mM of either 72 and 10 mM NaBr or 73 and 10 mM NaCl, derived from stocks dissolved in pure H2O, 5 \u03bcM of BesB purified in halide free phosphate buffer, and either 20 mM BIS-Tris (pH 6), MOPS (pH 7), HEPES (pH 8), or CHES (pH 9) buffers, that were pH adjusted using H3PO4 or NaOH. Negative controls containing both substrate and salt mixtures in the absence of BesB, and boiled BesB were set up in parallel. All reactions proceeded overnight at room temperature, then quenched with 100 \u03bcL of ACN, and briefly centrifuged to remove precipitate. For halide exchange reactions at the optimal pH 6 reaction mixtures (100 \u03bcL) contained 1 mM of 72, 10 mM of NaBr, NaF, or NaI, 5 \u03bcM of BesB purified in halide free phosphate buffer, in 20 mM BIS-Tris, pH 6. The other set of reaction mixtures contained 1 mM of 73, 10 mM of NaCl, NaF, or NaI, 5 \u03bcM of BesB purified in halide free phosphate buffer, in 20 mM BIS-Tris, pH 6. The reactions were allowed to proceed at overnight at room temperature, then quenched with 100 \u03bcL of ACN, and briefly centrifuged to remove precipitate. For all the reactions described above, 5 \u03bcL of supernatant was injected into the system equipped with a HILIC column and eluted at a flow rate of 1 mL\/min as described previously. For monitoring the incorporation of deuterium reaction mixtures (100 \u03bcL) contained 1 mM of 72 or 73, 25 mM of MES (pH 6) or CHES (pH 9) buffer and 5 \u03bcM of BesB. The reaction was diluted to the final volume using D2O, resulting in reaction mixtures that were approximately 80% D2O. The reactions were allowed to proceed at overnight at room temperature, then quenched with 136  100 \u03bcL of ACN, and briefly centrifuged to remove precipitate. For all the reactions described above, 5 \u03bcL of supernatant was injected into the system equipped with a HILIC column and eluted at a flow rate of 1 mL\/min as described previously. Additional mutant enzymatic reaction mixtures (100 \u03bcL) contained 1 mM of 72, or 1 mM of 73, 5 \u03bcM of F62Y or F232Y BesB purified normally in Tris buffer, in CHES (pH 9). The reactions were allowed to proceed at overnight at room temperature, then quenched with 100 \u03bcL of ACN, and briefly centrifuged to remove precipitate. For all the reactions described above, 5 \u03bcL of supernatant was injected into the system equipped with a HILIC column and eluted at a flow rate of 1 mL\/min as described previously. Reaction mixtures (100 \u03bcL) to probe production of 85 contained 1 mM of 83 or 1 mM 84, 1 mM of 80, 15 \u03bcM of BesB or F231Y BesB in 25 mM MOPS (pH 7.0) buffer. The reactions proceeded overnight at room temperature and were then quenched with 100 \u03bcL of ACN, and briefly centrifuged to remove precipitate. LC-MS analysis was performed using a Luna C18, 5 \u00b5m, 4.6 mm ID \u00d7 250 mm column (Phenomenex). Elution was performed at 0.5 mL min-1 using a mobile-phase consisting of a linear gradient of water and acetonitrile ((v\/v): 90:10 to 50:50, 0 to 5 min; 0:100, 5 to 10 min ; 0:100, 10-12 min; 90:10, 12 to 13 min; 90:10, 13 to 15 min), with both solvents containing 0.05% (v\/v) formic acid. To analyze the PROSS BesB construct activity the reaction mixtures (100 \u03bcL) contained 1 mM of 72 or 1 mM of 73, 5 \u03bcM of each construct purified normally in Tris buffer, in either 20 mM MES (pH 6), MOPS (pH 7), HEPES (pH 8), or CHES (pH 9) buffer. The reactions were allowed to proceed at overnight at room temperature, then quenched with 100 \u03bcL of ACN, and briefly centrifuged to remove precipitate. For all the reactions described above, 5 \u03bcL of supernatant was injected into the system equipped with a HILIC column and eluted at a flow rate of 1 mL\/min as described previously.  4.2.6 NMR analysis of 4-bromoallylglycine and NaCl exchange product The reaction mixture (1 mL) contained 10 mM (~2 mg) of 73, 100 mM NaCl, and 15 \u03bcM BesB in 35 mM sodium phosphate, pH 6 buffer. The reaction was allowed proceed for 16 h at room temperature. After this time the precipitated protein was removed by centrifugation. The solvent was then removed by lyophilization for 72 hr. The residual powder was re-suspended in 700 \u03bcL D2O and then the sample was briefly centrifuged to remove any residual solids prior to 137  NMR analysis. All NMR spectra were acquired using a Bruker Avance 400 MHz NMR spectrometer at room temperature.  4.2.7 Protein crystallization Initial crystallization conditions were identified by screening 2.5 mg\/mL and 5 mg\/mL BesB against the MSCG-1 crystal screen (Anatrace). Screen crystallization conditions containing crystals were further optimized through hanging drop vapor diffusion at room temperature. The highest quality BesB crystals were produced by mixing 1.5 \u03bcL of BesB protein solution (4-10 mg\/mL), and an equal volume of well crystallization solution composed of 0.16 Mg(CH3COO)2, 0.08 M Tris, pH 8.5, 14\u201322% (w\/v) PEG 8000, and 15-20 % (w\/v) glycerol, and suspending the drops over a 500 \u03bcL reservoir of the crystallization solution. Under these conditions, small to medium yellow to yellow-orange crystals appeared after approximately 2 d. Selenomethionine-derivatized BesB was crystallized using hanging drop vapor diffusion at room temperature. These crystals were produced by mixing 1.5 \u03bcL of selenomethionine-BesB protein solution (2-8 mg\/mL), and an equal volume of well crystallization solution composed of 0.16 Mg(CH3COO)2, 0.08 M Tris, pH 8.5, 14-20% (w\/v) PEG 8000, and 20 % (w\/v) glycerol, and suspending the drops over a 500 \u03bcL reservoir of the crystallization solution. Under these conditions, light yellow crystals appeared after approximately 5 days.  4.2.8 Data collection, structural determination, and model refinement X-ray diffraction data were collected at the Canadian Light Source (Saskatoon, Canada), using beamline CMCF-BM (08B1) equipped with Pilatus 6M detectors, as well as at the Stanford Synchrotron Radiation Light Source (SSRL) (Menlo Park, United States) using beamline BL9-2 equipped with Pilatus 6M detectors. UV-Visible spectroscopic spectra of BesB and F231Y BesB crystals were collected using the microspectrophotometer at beamline BL9-2 of the SSRL. All crystallographic data sets were integrated using iMOSFLM,169 and scaled using AIMLESS.170 The anomalous signal for selenium at 0.97933 \u00c5 from a selenomethionine-derivatized F231Y crystal was used to phase the structure of BesB (Table C.3), and was phased using the Autosol wizard in Phenix.214 The initial model of BesB was further improved using Phenix Autobuild.172 At this point the initial BesB model was manually inspected, and subjected to several iterative rounds of manual model building in COOT,173 and subsequent structural 138  refinement using Phenix refine.174 This ultimately produced a low resolution model with approximately 70% of the residues in place. This initial model was used to solve a second data set at a higher resolution, enabling the majority of the missing parts of the structure to be built. The resulting structure was further manually inspected using COOT, and refined using Phenix refine.173,174 Non-standard ligand restraints were generated using Phenix eLBOW.175 Final refinement statistics for both crystal structures can be found in Table C.4. Overall, the majority of both the wild-type and F231Y crystal structures could be completed, with only the first 12 residues of the N-terminus and the final two residues of the C-terminus as regions that could not be modeled in both final structures. Additionally, a small, highly flexible region (residues 79-86) in both structures could not be fully modeled. Within this region only Glu83 and Arg84 could be modeled, as they are stabilized by close crystal contacts with an adjacent monomer in the crystal lattice, so the approximate location of this disordered loop can still be estimated.  4.3 Results 4.3.1 Expression and purification of BesB using Rhodococcus To search for new BesB homologs the amino acid sequence of S. cattleya BesB (NCBI accession number WP_014151495.1) was used as a template to conduct an updated BLASTp search. The BLASTp search identified 10 unique BesB homologs with e values above 5 x 10-149, located in putative L-propargylglycine or \u03b2-ethynylserine biosynthetic gene clusters, and all were found in various actinomycetes (Table 4.1). Furthermore, aligning the amino acid sequences of these 10 homologs revealed high overall sequence homology, and all are approximately 70% identical in amino acid sequence to the original BesB from S. cattleya (Figure C.6, Table 4.1). To decide which homolog(s) to investigate, the stability of each enzyme was calculated using Expasy ProtParam stability.215 This program uses the occurrence of certain dipeptides in a protein\u2019s amino acid sequence to predict the stability of the protein in a test tube (Table 4.1), with an instability index value of below 40 indicating that the protein is probably stable.215 Several sequences fell below this threshold. The BesB homolog from Streptomyces achromogenes (68% identical to S. cattleya BesB) had one of the lower instability indexes at 39.6, had not been extensively tested for solubility in other studies,115 and the bacterial strain was already present in the lab. Therefore, it was decided to first test the solubility of this homolog. Under the tested expression conditions, the 139  S. achromogenes BesB homolog was also found to be insoluble in E. coli (Figure 4.3a), an observation consistent with previous results.115 As BesB appeared to be recalcitrant to expression in E. coli, I opted to use a different host for heterologous expression, the gram-positive bacterium Rhodococcus jostii RHA-1. This approach has been successfully applied by previous members of the lab to produce difficult to express proteins.216 I cloned S. achromogenes BesB into the pTIPQC1 Rhodococcus expression vector,213 and transformed it into the host R. jostii RHA-1 for protein expression. This second heterologous expression system produced significant amounts of soluble BesB (Figure 4.3b), which remained stable after purification.  Table 4.1 Bacterial strains containing BesB homologs.  Bacterial Strain NCBI Accession number Length (aa) Instability Index* % identity to S. cattleya BesB Streptomyces cattleya WP_014151495.1 511 46.9 - Streptomyces catenulae WP_030285993.1 512 37.5 73 Streptomyces alboverticillatus WP_086569406.1 514 39.1 69 Streptomyces NRRL S-31 WP_030744376.1 516 40.0 69 Streptomyces NRRL S-1448 WP_030410683.1 512 39.8 73 Streptomyces lavenduligriseus WP_078637771.1 514 39.6 69 Actinomedura sp. KC06 WP_131963935.1 485 39.8 71 Streptomyces achromogenes WP_078844341.1 518 39.6 68 Streptomyces eurythermus WP_189751797.1 518 42.4 69 Streptomyces griseocarneus WP_190062244.1 516 38.6 72 *Instability index calculated using the Expasy webserver. 140   Figure 4.3 IMAC purification of BesB. a) Purification of BesB from E. coli BL21(DE3). b) Purification of BesB from R. jostii RHA-1. Concentration of imidazole in each fraction indicated above well. Sizes of relevant molecular-weight markers indicated to the right of their respective bands in the ladder well.  As a secondary purification step, S. achromogenes BesB (which will simply be referred to as BesB from now on, unless otherwise specified) was further purified using size exclusion chromatography. Comparing the retention time of BesB to that of the 66 kDa BSA molecular weight standard, suggested that BesB exists as a monomeric enzyme in solution (Figure 4.4). This oligomeric state is unusual, as most other cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like enzymes exist as pairs of catalytic dimers that form homotetramers in solution.217,218 One unique aspect of BesB is its extended N-terminus (Figure C.2), which may be contributing to the inability of BesB to adopt the typical cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like tetrameric quaternary protein assembly.  141   Figure 4.4 Size-exclusion chromatography of BesB. Retention time of wild-type BesB (~57 kDa), compared to that of a BSA standard (~66 kDa).  One other unusual observation was that BesB purified from R. jostii RHA-1 is orange, rather than the typical yellow color, which is normally associated with the internal aldimine of a holo-PLP-dependent enzyme (Figure 4.5a).219,220 The maximal absorbance of BesB purified from Rhodococcus is at ~470 nm, which suggests that in BesB PLP is bound to another metabolite in an external adduct and not to the enzyme as an internal aldimine.219,220 The ~470 nm wavelength suggests that this species may be a PLP-aminoacrylate-like intermediate,220,221 with the most likely scenario being that BesB is binding and partially reacting with an unknown metabolite from Rhodococcus. This adduct is remarkably stable, and remains present even after IMAC purification, size exclusion chromatography, and dialysis. Overnight dialysis of the protein with excess free PLP shifted the UV-Vis absorbance spectrum to ~420 nm, which is more typical for enzyme-bound PLP (Figure 4.5b). However, some of the orange version of the protein remained even after the additional dialysis steps, further demonstrating the stability of this unusual species. 142   Figure 4.5 UV-Vis absorbance spectra of BesB. a) Directly after purification from R. jostii RHA-1. b) After overnight dialysis with PLP, followed by additional dialysis in 20 mM Tris, 50 mM NaCl pH 8 buffer to remove excess PLP.  4.3.2 In vitro biochemical activity assays  Purified BesB from Rhodococcus was reacted with both its physiological substrate 72 as well as with a second verified halogenated substrate 73. Each substrate was tested at four different pHs in vitro. Under the in vitro conditions tested BesB was found to catalyze the conversion of 143  both 72 and 73 to 74 at all tested pHs (Figure 4.6, Figure 4.7a, b). The reaction was more efficient at higher pHs, with leftover substrate remaining at pH 6 and 7. Therefore, the orange PLP-adduct appeared to have no adverse effects on BesB activity. There was however, one unusual species that repeatedly appeared in assays with BesB and 73. Looking closely at the data obtained from the assays that used 73 as a substrate, I observed a reproducible signal with an m\/z of 150 and 152 in the assays that were run at pHs 6 and 7, but not at the higher pHs (Figure 4.7c). The observed masses of 150 and 152 are consistent with 72 and also had the same isotopic distribution that would be expected if a chlorine atom was present in the molecule. Furthermore, the peak also exhibited the same retention time as the 72 standard. After analysis of the data, the most probable scenario seemed to be that BesB was catalyzing a reaction between excess Cl\u2212 ions present in the reaction mixtures and 73.  Figure 4.6 Reaction of 4-chloroallylglycine and BesB at different pHs. a) Selected ion chromatograms for 4-chloroallylglycine (72) (m\/z 150 and 152). Reaction conditions are indicated beside each trace. b) Selected ion chromatograms for L-propargylglycine (74) (m\/z 114). Reaction conditions are indicated beside each trace. All reactions contained 1 mM 4-chloroallylglycine (72) and 5 \u03bcM BesB. 144   Figure 4.7 Reaction of 4-bromoallylglycine and BesB at different pHs. a) Selected ion chromatograms for 4-bromoallylglycine (73) (m\/z 194 and 196). Reaction conditions are indicated beside each trace. b) Selected ion chromatograms for L-propargylglycine (74) (m\/z 114). Reaction conditions are indicated beside each trace. c) Selected ion chromatograms at m\/z 150 and 152. Such a mass would be expected for a product like 4-chloroallylglycine (72). Reaction conditions are indicated beside each trace. All reactions contained 1 mM 4-bromoallylglycine (73) and 5 \u03bcM BesB. 145   To eliminate the possibility of non-enzymatic reactions or spontaneous breakdown of 72 and 73 at low pHs, negative controls using boiled enzyme were carried out at pH 6 (Figure 4.8). Additional controls also incubated both substrates in the presence of a 10-fold excess of the opposite sodium halide salt (Cl\u2212 or Br\u2212) at pH 6 (Figure 4.8). Both sets of control experiments demonstrated that both halogenated substrates were stable at low pH and in the presence of excess Cl\u2212 or Br\u2212. This result indicates that the unknown compound with m\/z 150 and 152 was formed through enzymatic reactions catalyzed by BesB.   Figure 4.8 Stability of 4-chloro- and 4-bromoallylglycine at pH 6. a) 4-bromoallylglycine (73) incubated overnight with 10 mM NaCl at pH 6. Reaction conditions indicated by each trace. b) 4-chloroallylglycine (72) incubated overnight with 10 mM NaBr at pH 6. Reaction conditions indicated by each trace. In both panels a) and b) the blue traces are the selected ion chromatograms at m\/z 150 and 152 for 4-chloroallylglycine (72), and the orange traces are the selected ion chromatograms at m\/z 194 and 196 for 4-bromoallylglycine (73). 146   Further evidence for the enzymatic origin of this additional compound came when the assays were repeated in ~80% D2O. Under these conditions two atoms of deuterium were incorporated into the m\/z 150 and 152 product, as peaks at m\/z 152 and 154 could now be detected (Figure 4.9b). Furthermore at pH 6, two atoms of deuterium were found in 72 that was not converted into 74, which indicated that 72 could still react with BesB under these conditions (Figure 4.9c). Again, both of these results are consistent with previous BesB data that showed BesB incorporates two atoms of deuterium into its products.115 This previous result was also verified by producing 74 from 73 at pH 9 (Figure 4.10). Under the reaction conditions tested, a peak at m\/z 116 corresponding to the incorporation of two atoms of deuterium was detected (Figure 4.10b). Unexpectedly, a small amount of the product was found to contain three atoms of deuterium (m\/z 117). However, this result is likely due to slow non-enzymatic exchange of the relatively more acidic acetylenic proton of 74, as this exchange was also seen in the control reaction at pH 9 with the synthetic standard of 74. Therefore, the exchange of multiple deuteriums and regeneration of starting 72 substrate indicated that though the initial steps of the BesB-catalyzed reaction can proceed at lower pH, the BesB-catalyzed desaturation does not occur at low pH. At this point, Cl\u2212 ions that are present in solution could react with the stalled intermediate to regenerate a chlorine-containing amino acid. Given the nature of this reaction, it was also possible that Cl\u2212 could react with the stalled intermediate at different locations. To determine the structure of this additional compound, I prepared a large-scale reaction to produce enough of the compound for NMR. A combination of 1H, 13C, and 1D-NOESY NMR revealed that the molecule being produced at low pH is indeed 72 and not an alternative constitutional isomer (Figure C.3-5).    147   Figure 4.9 Isotopic distributions of 4-chloroallylglycine produced by BesB in D2O. a) 4-chloroallylglycine (72) in D2O, b) 4-chloroallylglycine (72) produced from 4-bromoallylglycine (73) by BesB at pH 6, and c) 4-chloroallylglycine (72) incubated with BesB at pH 6.   148   Figure 4.10 Isotopic distribution of L-propargylglycine produced by BesB in D2O. a) L-propargylglycine (74) incubated in D2O at pH 9. b) Enzymatically produced L-propargylglycine (74) from the BesB reaction at pH 9. The peak at m\/z 118 is background signal present within the instrument.  The observation that the initial steps of the BesB-catalyzed reaction are reversible raised an additional scenario, whereby 73 may be first converted to 72 by BesB, which is then converted to 74 by a second BesB-catalyzed reaction. To exclude this possibility, BesB was purified in the absence of Cl\u2212 and all buffers were prepared without using HCl to repeat these in vitro assays in the absence of Cl\u2212. In the absence of Cl\u2212 BesB still catalyzes the conversion of 73 to 74, and exhibits the same pH dependence profile to that of the native reaction (Figure 4.11). Therefore, the results of these experiments indicate 73 is a genuine substrate of BesB. 149   Figure 4.11 Reaction of BesB with 4-bromoallylglycine in the absence of Cl\u2212 at different pHs. a) Selected ion chromatograms for 4-bromoallylglycine (73) (m\/z 194 and 196). Reactions contained 1 mM 4-bromoallylglycine (73) and 5 \u03bcM BesB. Reaction conditions are indicated beside each trace. b) Corresponding selected ion chromatograms for L-propargylglycine (74) (m\/z 114). Reaction conditions are indicated beside each trace.  The possibility of using the newly discovered BesB-catalyzed halide exchange to generate additional alternative 4-halogenated-allylglycine derivatives was also explored. The experiments were repeated once more at pH 6 for both 72 and 73 in the presence of 10 mM F- and I-. However, in both cases neither of the corresponding 4-halogenated allylglycine derivatives could be detected using LC-MS. Thus, it appears that this phenomenon is limited in scope, and while chlorination\/de-chlorination appears to be reversible, the reaction does not proceed with other common halides.   150  4.3.3 Exploring additional in vitro BesB reactivity The amino acid sequence of BesB closely resembles that of other PLP-dependent \u03b3-lyases (Figure C.2), and so the common PLP-dependent \u03b3-lyase substrates L-cysteine, L-methionine, and L-cystathionine (85) were each incubated with BesB in vitro, overnight at pH 8. However, BesB lacks \u03b3-lyase activity as none of these substrates were consumed, as determined by HPLC and LCMS analysis. The reactivity of BesB with a close substrate mimic L-allylglycine (115) was also tested (Figure 4.12). As 115 lacks the halide leaving group found in the substrates 72 and 73, it should be unable to fully react with BesB. To test this hypothesis 115 was incubated with BesB at several different pHs, but no significant changes in the amount of 115 present in the assays was observed (Figure 4.12). However, when the reactions were carried out in ~80% D2O, it was found that BesB can catalyze the incorporation of two atoms of deuterium into 115 (Figure 4.13).  Figure 4.12 Reactions of BesB with L-allylglycine at pH 6 and 9. Traces are the selected ion chromatograms at m\/z 116. All reactions contained 1 mM L-allylglycine (115), and reaction conditions are indicated beside each trace.  151   Figure 4.13 Reactions of BesB with L-allylglycine at pH 6 and 9 in D2O. a) boiled BesB, L-allylglycine (115), pH 6, b) boiled BesB, L-allylglycine (115), pH 9, c) BesB, L-allylglycine (115), pH 6, d) BesB, L-allylglycine (115), pH 9. Mass shifts due to the incorporation of deuterium are indicated in red.  The amino acid sequence of BesB also closely resembles that of other PLP-dependent cystathionine-\u03b3-synthases (Figure C.2). Therefore, the ability of BesB to synthesize 85 from either 83 and 80 or 84 and 80 was tested. Under the in vitro reaction conditions used BesB was able to catalyze the formation of small amounts of 85 using 83 and 80 as substrates, but not using 84 and 80 (Figure 4.14). The fact that BesB can catalyze such a reaction further illustrates its close relationship between enzymes of the cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase family. All in vitro data reported here for BesB is consistent with BesB sharing structural and mechanistic similarities with other members of the cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase family. In particular, the initial two deprotonations at the C\u03b1 and C\u03b2 positions of the substrate amino acid appear to be shared between BesB and the other related enzymes (Figure 1.33, Figure 4.2) 152   Figure 4.14 BesB CGS activity. a) Selected ion chromatograms for O-acetylhomoserine (83) (m\/z 162). Reaction conditions indicated. b) Corresponding selected ion chromatograms for L-cystathionine (85) (m\/z 223). Reaction conditions indicated. c) Selected ion chromatograms for O-succinylhomoserine (84) (m\/z 220). Reaction conditions indicated. d) Corresponding selected ion chromatograms for L-cystathionine (85) (m\/z 223). Reaction conditions indicated where L-cysteine (80) was added.  4.3.4 Heavy atom derivatization and structural phasing of BesB A second major focus of this project was to obtain a crystal structure of BesB, which would provide the molecular details to identify the distinguishing features of BesB. Despite the similarities between the C-terminal domain of BesB and other PLP-dependent \u03b3-lyases (Figure C.2), the crystal structure of BesB could not be solved using the structure of other \u03b3-lyases as a template for molecular replacement. Attempts to phase BesB using heavy metals failed, as crystals soaked with Hg-, Ho-, Pt-, or Sm-containing salts failed to produce sufficient anomalous signal to phase the structure. While Rhodococcus expression systems are commonly used to produce protein for crystallographic studies, there are only two reported instances of selenomethionine-derivatized proteins being produced using Rhodococcus expression systems.222,223 In both cases, the selenomethionine-derivatized protein was produced by supplementing the bacterial M9 minimal growth medium with 1 mM selenomethionine at the point of expression induction. Based on these results, I expressed wild-type BesB using this method, but found that the wild-type enzyme expressed very poorly (Figure 4.15). A BesB variant, F231Y, was observed to express at higher 153  levels than the wild-type enzyme. Therefore, the F231Y variant was also expressed in media supplemented with 1 mM selenomethionine. Under these conditions the F231Y variant was produced dramatically greater quantities than the wild type as estimated by FPLC (Figure 4.15). The F231Y variant also crystallized under similar conditions as the wild-type, and produced large, well diffracting selenomethionine-derivatized crystals. The anomalous signal from these crystals extended to ~ 3 \u00c5, a resolution which was sufficient to phase and solve the crystal structure of BesB (Table C.3, Table C.4).  Figure 4.15 BesB FPLC traces from expression in Rhodococcus with 1 mM selenomethionine.   4.3.5 Overall tertiary structure of the BesB monomer Solving the structure of BesB revealed that both wild-type BesB and the F231Y variant crystallize as monomers in the same P3221 spacegroup, with the same unit cell dimensions (Table C.3). The monomeric state is also consistent with FPLC data which supported that BesB exists as a monomer in solution (Figure 4.4). Furthermore, the monomeric state of BesB is unique, as normally cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like enzymes function as a pair of catalytic dimers that make up the overall homotetrameric assembly of the enzyme (Figure C.7).217,218 To determine what structural features of BesB trigger this shift in quaternary assembly, the structure of BesB was superposed with the structures of a cystathionine-\u03b2-lyase (PDB: 6CJA), a cystathionine-\u03b3-lyase (PDB: 3COG), a cystathionine-\u03b3-synthase (PDB: 3QHX), and a methionine-\u03b3-lyase (PDB: 5DX5) (Figure 4.16, Figure 4.17). The overall results are consistent 154  across each structural alignment, and for simplicity the following more detailed discussion will solely focus on the structural comparison between BesB and the cystathionine-\u03b2-lyase 6CJA.  Figure 4.16 Structure of BesB and alignments with two monomers of a homotetrameric cystathionine \u03b2-lyase. a) Overall structure of the BesB monomer with the N-terminal region colored orange, linker region colored blue, and C-terminal domain colored yellow. b) Alignment of BesB with a single monomer of the cystathionine \u03b2-lyase 6CJA (light blue). c) Alignment of BesB with a second monomer of 6CJA (green). d) Close up of the alignment of the linker region of BesB (blue), showing how it connects the N-terminus and C-terminus.  Individual monomers of each of these enzymes were superposed with BesB. This structural superposition revealed that the C-terminal region of BesB (assigned as residues 174-516) adopts the canonical cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like fold adopted by each of the other four enzymes (Figure 4.16a, b, Figure 4.17). The remaining residues of the BesB monomer (residues 13-173) did not align with any parts of the alignment monomer (Figure 4.16b). To determine what fold the unusual N-terminus of BesB adopts, the BesB monomer was also superposed with the second monomer from the catalytic dimer of each of the other four enzymes. 155  This second structural superposition revealed that the N-terminal domain of BesB (assigned as residues 13-159) is positioned in essentially the same location as the second monomer from the catalytic dimer (Figure 4.16c, Figure 4.17). The overall secondary structure of the aligned portions of the N-terminus of BesB closely matches the secondary structure present in the aligned sequences of the second 6CJA monomer (Figure 4.16c). Importantly, just as the second monomer of the 6CJA catalytic dimer interfaces with the first to form a complete active site,217,218 the N-terminus of BesB interfaces with the C-terminal domain in precisely the same orientation . This observation explains why truncations of the extended N-terminal sequence resulted in insoluble enzyme. The alignment of the N- and C-terminal domains with two separate monomers of the cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase catalytic dimer are consistent across multiple structural alignments with different enzymes (Figure 4.17). Finally, the multiple structural alignments of BesB revealed that the N- and C-terminus of BesB are linked together by a small sequence (residues 160-173), which lacks any defined secondary structure and does not align with any other structural features (Figure 4.16d). Thus, it appears that BesB has evolved to have its N-terminus adopt the same fold as a second monomer, retaining the overall fold and activity of other PLP-dependent \u03b3-lyases but doing so as a monomer. Protein oligomerization plays an important role in many cellular processes, and often increases the stability of a protein complex.224 As it has been demonstrated that BesB is unstable under many conditions, the N-terminus of BesB may not only be required for catalysis, but is also needed to keep the BesB monomer stable enough to function. A similar feature is also found in an unrelated thermophilic \u03b1\/\u03b2-hydrolase fold esterase, which contains a unique N-terminal extension.225 Disruptions to this region of the esterase greatly affected the stability and catalytic properties of the enzyme, leading the authors to propose that N-terminal extension was evolutionarily selected to preserve the stability of the enzyme.   156   Figure 4.17 Structural alignments of BesB with related cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthases. a) Structural alignment of BesB with a cystathionine-\u03b3-lyase (PDB: 3COG). b) Structural alignment of BesB with a cystathionine-\u03b3-synthase (PDB: 3QHX). c) Structural alignment of BesB with a methionine-\u03b3-lyase (PDB: 5DX5). In all figures BesB is colored orange and yellow, while the two monomers of the superposed enzyme are colored light blue and green. 157  4.3.6 Refinement of unknown PLP-adduct Having established that the monomeric assembly of BesB retains essentially all important structural features of the typical cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase tetramer, the next step was to compare the active site of BesB to the active sites of these other enzymes. Previously, I had observed that wild-type BesB purified from Rhodococcus contained an orange PLP-adduct (Figure 4.5a). This adduct proved to be remarkably stable, as it also remained present throughout crystallization and data collection, as UV-Vis scans of the orange wild-type crystals show a similar extended UV-Vis absorbance at ~470 nm to that of the protein in solution (Figure 4.18a). Accordingly, the electron density in the active site in the solved crystal structures of wild-type BesB showed no evidence for an internal aldimine with the enzyme. The available electron density in these structures did not connect PLP to the conserved active site lysine, and instead there was strong electron density present above the PLP aldehyde which indicated that a small molecule remained bound in a PLP-external aldimine. Electron density indicative of an external aldimine was present in the active site of all solved wild-type crystal structures, regardless of the crystallization conditions, the presence of other small molecules soaked into the crystals, or the resolution of the crystal.  Figure 4.18 BesB crystals and corresponding UV-Vis absorbance spectra. 158  A) Left panel: A representative image of typical BesB crystals. Right: A representative UV-Vis spectrum collected from a wild-type UV-Vis crystal. B) Left panel: A representative image of typical F231Y BesB crystals. Right: A representative UV-Vis spectrum collected from a F231Y BesB crystal. The insets in each right panel highlight the 300 \u2013 500 nm region of the spectrum, more clearly revealing the differing absorbance patterns found in each crystal.  Given that PLP-dependent enzymes predominantly catalyze reactions using amino acid substrates, I modeled a PLP-alanine external aldimine, as well as two different desaturated PLP-homoalanine adducts, fit them into the available density in the active site, and then refined the structures at a resolution of 1.29 \u00c5 (Figure 4.19). These results indicated that the one carbon side chain of the PLP-alanine external aldimine was too small to account for all of the available electron density (Figure 4.19a). Instead, the two carbon side chains of the other two adducts more closely fit the present electron density (Figure 4.19b, c). Ultimately, the \u03b1-\u03b2-unsaturated PLP-homoalanine species was chosen as the species to include in the wild-type structure (Figure 4.19c). This selection was based on its better fit to the available electron density and the fact that the UV-Vis absorbance spectrum of these crystals absorbs at ~470 nm (Figure 4.18a), a wavelength typically associated with an \u03b1-\u03b2-unsaturated PLP-aminoacrylate intermediate.220 However, despite the high 1.29 \u00c5 resolution of the crystal structure, this species could not be unambiguously assigned on the electron density alone. Therefore, other possibilities cannot be fully excluded, and the density may be a combination of multiple species or could possibly be a different molecule that was not modeled. In contrast to the crystals of the wild-type, the crystals of the F231Y variant were yellow and displayed the characteristic absorbance spectrum of a PLP-dependent enzyme (Figure 4.18b). Accordingly, the electron density in the active site clearly indicated the presence of an enzyme-PLP internal aldimine between lysine and PLP. 159   Figure 4.19 Fitting of unknown PLP-adduct into the density present in the active site of the wild-type BesB crystal. Top row: Chemical structure of each PLP-external aldimine adduct ligand. a) Alanine-PLP adduct, b) \u03b2-\u03b3 unsaturated homoalanine-PLP adduct in ketimine form, c) \u03b1-\u03b2 unsaturated homoalanine-PLP adduct. Middle row: Each ligand fit to unmodeled Fo-Fc electron density present in the active site of BesB. Fo-Fc electron density is shown in green and contoured at 3.0 \u03c3. Bottom row: Electron density after refinement with ligand in Phenix refine. 2Fo-Fc electron density is shown in blue, contoured at 1.0 \u03c3. Fo-Fc electron density is shown in green and red, and contoured at 3.0 \u03c3 and -3.0 \u03c3 respectively.  4.3.7 Analyzing the active site of BesB Modeling of the \u03b1-\u03b2-unsaturated PLP-homoalanine adduct completed the picture of the active site of the wild-type BesB structure (Figure 4.20). The active site of BesB is made up of residues from both the N-terminal and C-terminal domains, and both domains are involved in PLP binding. Residues Tyr59 and Arg61 from the N-terminus form hydrogen bonds with the phosphate of PLP, helping to stabilize PLP in the active site. PLP-binding residues from the C-terminal domain include Asp305, which forms a hydrogen bond with the pyridinium nitrogen of PLP and 160  Asn280, which forms a hydrogen bond with the hydroxyl group of PLP. In this structure the conserved lysine which would form the internal aldimine, Lys329, is instead hydrogen bonding with a nearby hydroxyl group from the side chain of Ser456. Though not interacting directly with PLP, nearby Arg456 is positioned to coordinate with the carboxylic acid group of the amino acid substrate and can be seen to do so in this structure. Importantly, all these PLP-binding residues are conserved amongst other enzymes which adopt a cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like fold (Figure C.2) and generally conserved in fold type I PLP-dependent enzymes. Therefore, when the active site of BesB is aligned with active sites of other related enzymes, both sets of active sites look remarkably similar (Figure 4.21). Furthermore, there was no additional electron density in the active site which would indicate the presence of additional co-factors, such as metals or other anions which could account for the unique reactivity of BesB.   Figure 4.20 Active site of wild-type BesB. Selected residues are labeled with their respective side chains are highlighted as sticks. The coloring is consistent with previous images, with the N-terminus in orange and the C-terminus in yellow. 161    Figure 4.21 Comparison of the active site of BesB with the active sites of related enzymes. a) Alignment of BesB and 6CJA. b) Alignment of BesB and 3COG. c) Alignment of BesB and 3HQX. d) Alignment of BesB and 5DX5. For simplicity, only the residues that differ between the two enzymes are labeled in red, with the residue and number of the non-BesB enzyme highlighted.  The most critical change that is observed in the active site of BesB is that Phe231 is positioned above the pyridine ring of PLP. Both sequence alignments (Figure C.2) and structural alignments (Figure 4.21) with other cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like enzymes reveal that this position is normally occupied by a conserved tyrosine residue. This tyrosine is critical for the proper positioning of substrates in the active site and is also involved certain stages of various \u03b3- and \u03b2-elimination reactions catalyzed by these other enzymes. Therefore, the Tyr to Phe substitution likely contributes significantly towards altering the activity of BesB. However, the Tyr to Phe substitution does not greatly affect the overall structure, as the structure of the F231Y variant is same as that of the wild type, with the two structures having an overall RMSD of 0.098 \u00c5 for C\u03b1 pairs. (Figure 4.22). The Tyr to Phe substitution also appears to have minimal effect on the overall orientation of the other conserved active site residues in BesB The only slight 162  differences within the active site between the two structures are the aforementioned PLP-internal aldimine of the F231Y structure, and the F231Y mutation itself, which results in the tyrosine sidechain shifting very slightly downwards towards the pyridine ring of PLP, relative to the position of Phe231 in the wild-type structure (Figure 4.22b). Additional residues which complete the active site of BesB are Phe62, and Phe232, which are positioned along with Phe231 to form a highly hydrophobic active site pocket. However, the phenylalanine side chains do not contain any exchangeable protons, and therefore are unable to participate as the catalytic base in the BesB-catalyzed reaction. Therefore, the purpose of the three phenylalanine residues may be to help stabilize any reactive species generated over the course of the reaction. The guanidinium of the side chain of Arg61 is also positioned near the phenylalanine pocket. However, as the pKa of the guanidium group is over 13,226 the guanidium group remains positively charged and therefore unable to serve as the catalytic base in the BesB-catalyzed reaction. This process of elimination leaves Asp233 as the potential catalytic base. While an Asp residue at this position is not uncommon amongst other cystathionine-\u03b2-lyases\/cystathionine-\u03b3-synthases, in BesB it is hydrogen bonding with His142, a feature which is unique to be unique to BesB. This hydrogen bonding pattern with His142 may allow Asp233 to act as the base which deprotonates the PLP-allene intermediate in the proposed reaction mechanism for BesB (Figure 4.2)  Figure 4.22 Comparison of the structure of wild-type and F231Y BesB. a) Overall alignment of both structures. b) Alignment of the active sites of both structures. In both images wild-type BesB is colored in the same color scheme as previous images, while the gray, 163  purple and white colored regions represent the N-terminal domain, linker, and C-terminal domain of F231Y, respectively.  4.3.8 Site-directed mutagenesis studies of BesB Based on the crystal structure of BesB, several variants were created using site-directed mutagenesis to interrogate the roles of several active site residues. Altogether seven variants were created, and all variants were expressed in Rhodococcus and purified using the same procedures as the wild-type enzyme (Figure 4.23).  Figure 4.23 FPLC traces of purified BesB variants.  Due to the crucial nature of the Tyr to Phe substitution, the reverse variant, F231Y, was created in order to assess what the effect of this change was on the reactivity of BesB. LCMS analysis of in vitro reaction mixtures revealed that the F231Y variant could still catalyze the formation of low levels of 74, and that the F231Y variant could also still catalyze the exchange of chloride at pH 6 (Figure 4.24). The F231Y variant could also catalyze the formation of 85 using 83 and 80 as substates, just as the wild-type (Figure 4.25). The other two active site phenylalanine residues were also mutated to tyrosine, creating the variants F62Y and F232Y. The F62Y and F232Y variants were both incubated with 72 and 73. It was found that F62Y and F232Y still efficiently converted both substrates to 74 at pH 9 (Figure 4.26). 164   Figure 4.24 In vitro reactivity of F231Y BesB. a) Selected ion chromatograms for 4-bromoallylglycine (73) (m\/z 194 and 196). Reaction conditions as indicated. b) Selected ion chromatograms for 4-chloroallylglycine (72) (m\/z 150 and 152). Reaction conditions as indicated. c) Selected ion chromatograms for L-propargylglycine (74) (m\/z 114). Reaction conditions as indicated.   Figure 4.25 F231Y BesB CGS activity. a) Selected ion chromatograms for O-acetylhomoserine (83) (m\/z 162). Reaction conditions indicated. b) Corresponding selected ion chromatograms for L-cystathionine (85) (m\/z 223). Reaction conditions indicated. c) Selected ion chromatograms for O-succinylhomoserine (84) (m\/z 220). Reaction conditions indicated. d) Corresponding selected ion chromatograms for L-cystathionine (85) (m\/z 223). Reaction conditions indicated where L-cysteine (80) was added. 165   Figure 4.26 In vitro activity of F62Y and F232Y BesB variants. a) Selected ion chromatograms 4-chloroallylglycine (72) (m\/z 150 and 152) consumption at pH 9. b) Selected ion chromatograms for 4-bromoallylglycine (73) (m\/z 194 and 196) consumption at pH 9. c) Corresponding selected ion chromatograms for L-propargylglycine (74) (m\/z 114) production at pH 9. Substrate of each reaction is indicated with brackets. All reactions contained 1 mM of substrate and 5 \u03bcM of each variant.  As Asp233 was the most likely candidate to act as a catalytic base in the reaction in the BesB-catalyzed reaction, both D233N and D233A variants were generated. Additionally, an H142A variant was also created, as His142 is hydrogen bonding to Asp233 in the structure of BesB. However, I was unable to purify any of these variants, as these single changes altered the expression levels of the enzyme, resulting in very low levels of protein being purified, as judged by FPLC chromatography (Figure 4.23). Therefore, the roles of either D233 or H142 in the BesB-catalyzed reaction cannot be interrogated directly, as changes to any of these residues result in an unstable protein. However, given the observed changes in protein stability and expression, it can be hypothesized that that both His142 and Asp233 play important roles for BesB.  166  4.3.9 Engineering BesB for improved expression and stability using PROSS One of the remaining challenges was to improve the overall protein yields of BesB, as wild-type BesB is produced at ~2 mg\/L of Rhodococcus cell culture. Increasing the yield to ~15 mg\/L of cell culture would allow us to examine BesB using stopped-flow spectroscopy and enable us to explore other biocatalytic reactions on a larger, more practical scale. There are many options available for protein engineering, and these methods are generally sequence based, structure based, or use random mutagenesis to improve the stability and catalytic properties of an enzyme.227,228 I elected to use the computational webserver Protein Repair One-Stop Shop (PROSS) to design BesB variants with improved stability.229,230 PROSS is a sequence and structure-based design method, which calculates and selects for stabilizing amino acid mutations within constraints generated from multiple sequence alignments. Of the twenty sequences generated by PROSS, five unique BesB sequences were chosen to be tested. These five sequences contain between 17 and 69 amino acid mutations, which corresponds to between ~3 % and ~13 % of all amino acid residues being modified (Figure C.8). The positions of these changes on the structure of BesB were also manually verified to ensure that they would not disrupt the active site. The corresponding genes were synthesized with codon usage patterns designed for high levels of expression in both E. coli and Rhodococcus (Figure C.8). The codon optimized genes were then cloned into the vector pET22b for heterologous expression in E. coli and into the vector pTIPQC1 for heterologous expression in Rhodococcus.  Unlike experiments with wild-type S. achromogenes BesB, all five of the PROSS BesB variants could be expressed and purified from E. coli. Despite the impressive improvements in solubility in E. coli, BesB was only produced at ~1-2 mg\/L of E. coli cell culture, slightly less than what is produced using Rhodococcus. Additionally, BesB purified from E. coli was yellow with a UV-Vis spectrum more typical of PLP-dependent enzymes, unlike when purified from Rhodococcus. This result suggests that presence of the stable orange aminoacrylate adduct bound to BesB is probably generated by a metabolite that is unique to Rhodococcus. The activity of four of the PROSS BesB variants were tested at pH 9 with both 72 and 73 (Figure 4.27). All four PROSS variants were able to convert both 4-halogenated substrates to 74, however the two more highly modified constructs containing 50 and 69 mutations appeared to have reduced levels of activity, as some of both substrates remained in the reaction mixture even after an overnight 167  incubation (Figure 4.27). Furthermore, 72 was detected assays with 73 at pH 6 for all four PROSS BesB variants tested, further underscoring the ease with which chloride ions can be reversibly exchanged in this reaction (Figure 4.28).  Figure 4.27 In vitro reactivity of PROSS BesB variants. a) Selected ion chromatograms for 4-chloroallylglycine (72) (m\/z 150 and 152). b) Corresponding selected ion chromatograms for L-propargylglycine (74) (m\/z 114). c) Selected ion chromatograms for 4-bromoallylglycine (73) (m\/z 194 and 196). d) Corresponding selected ion chromatograms for L-propargylglycine (74) (m\/z 114). All reactions contained 1 mM of substrate and 5 \u03bcM PROSS enzyme. The PROSS variant used is indicated beside each trace.   168    Figure 4.28 PROSS BesB variant-catalyzed 4-chloroallylglycine production from 4-bromoallylglycine. Selected ion chromatograms for 4-chloroallylglycine (72) (m\/z 150 and 152). All reactions contained 1 mM 4-bromoallylglycine (73) and 5 \u03bcM PROSS enzyme at pH 6. The PROSS variant used is indicated beside each trace.  4.4 Discussion The described studies of BesB were enabled by using Rhodococcus as the system for heterologous protein expression. Therefore, in addition to being a well-established versatile biocatalytic powerhouse,231 various strains of the actinobacteria Rhodococcus also represent an attractive alternative host system for the expression of difficult to isolate proteins from Streptomyces or other actinomycetes.216,232\u2013234 With BesB isolated from this system, I was able to more thoroughly investigate the unique PLP-dependent desaturation catalyzed by BesB. It was found that the desaturation reaction seems to be pH-dependent, as at lower pHs it does not occur. Based on these observations it seems likely that the residue(s) involved in deprotonation become protonated between pH 6 and 7 and therefore can no longer participate in the reaction. Based on what is known about BesB and related enzymes it seems most likely that this stalled intermediate is most likely an allene-PLP species. This hypothesis is based on the observation that the release of chloride appears to be relatively facile and reversible, as chloride present in solution can be readily re-incorporated into the reaction. This reaction is also the first reported instance of a PLP-dependent enzyme catalyzing a halogenation reaction. A related pair of amino acids one carbon 169  longer than 72 and 74, 2-amino-5-chloro-5-hexenoic acid and 2-amino-4,5-hexadienoic acid have been isolated from the fruiting bodies of Amanita miculifera, which support that the allene could be synthesized from a vinyl chloride.235 However, it is difficult to capture or verify this species experimentally, and there is no definitive evidence of this compound. Crystallographic studies have revealed that the structure of BesB is quite unique amongst enzymes that adopt a cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like fold. However, rather than adopting a new fold, BesB uses a modified version, forgoing the typical homotetrameric assembly (~1300 residues), and instead can fully function as a small independent monomer (~520 residues), with a unique N-terminus that serves the same role generally held by a second monomer. Such a change could have advantages, as it would be less resource intensive for a bacterium to synthesize the smaller monomer. The other most significant difference of BesB is that the active site contains a phenylalanine in place of the normal catalytic tyrosine. This change is crucial for preventing BesB from becoming irreversibly inhibited like other cystathionine-\u03b3-lyases.136,138,140,211 Interestingly, there is one reported instance of a cystathionine-\u03b3-synthase which contains a phenylalanine residue instead of the typical tyrosine.236 However, this enzyme was isolated from the thermoacidophilic archaeon Sulfolobus tokodaii, and accordingly shows maximal activity at 80 \u00b0C, a temperature at which perhaps the synthesis of 85 requires different energetic considerations. In most other cases where the catalytic tyrosine has been swapped to a phenylalanine, a dramatic reduction in catalytic proficiency towards the native reaction is observed in the phenylalanine variant.237\u2013239 This observation is consistent with the results from several experiments, as lyase activity 87, 80, 83, 84, or 85 could not be detected using LCMS. The F231Y mutant also demonstrated improved cystathionine-\u03b3-synthase activity with 83 and 80 when compared to the wild-type. Unfortunately, several other residues also thought to be important were unable to be directly interrogated experimentally, as the His142 and Asp233 variants were not expressed. However, based on the available data, including the chloride exchange and the fact that there are no other putative catalytic bases positioned nearby in the active site Asp233 may abstract a proton to produce the additional degree of unsaturation. Given that the reaction stalls at low pHs, the residue becomes protonated, thus preventing it from acting as a base. The other residues in the active site are either unlikely to be involved in such a reaction (phenylalanine and arginine), or not positioned close enough to catalyze such a transfer. 170   Given all the new data presented in this chapter, I am able to propose a slightly modified catalytic mechanism for BesB (Figure 4.29). It has been shown that BesB incorporates two atoms of deuterium can be incorporated into 74, 115 recycled 72 substrate (Figure 4.9), and allylglycine (Figure 4.13). Additionally, the crystal structure of BesB revealed that the active site of BesB is still highly similar to that of that of other cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase-like fold (Figure 4.16, Figure 4.17). Taken together, these observations suggest that the initial steps of the BesB-catalyzed desaturation of 72 are like also shared with other related enzymes,240 leading us to propose the following mechanism (Figure 4.29). As is the general case for PLP-dependent enzymes, the substrate amino acid 72 would react with the PLP-internal aldimine of BesB to form a new external aldimine. Then the now free PLP-binding lysine would first abstract the C\u03b1 proton from 72, producing a quinonoid intermediate. The quinonoid intermediate is then reprotonated at C-4', generating a ketimine intermediate. Now the lysine acts again and abstracts a proton from the C\u03b2 position of the sidechain of 72. How BesB catalyzes release of chloride and forms the terminal alkyne are less clear. The BesB-catalyzed chlorination reaction can be reversible, especially at low pH. This would suggest that this stage may be relatively easily to catalyze. Given that there do not appear to be any residues in the active site that would assist with catalyzing the release of chloride, the chloride leaving may be energetically favorable and proceed spontaneously, triggering the formation of an allene. While at this time there is no direct evidence supporting the generation of an allene, it seems possible based on the reversibility of the chloride, as well as the lack of reactive side chains in the active site. The three phenylalanines present in the active site may be positioned to stabilize this transition state, as well as shield this more reactive structure from undesirable off pathway reactions. The most likely candidate for the catalytic base that would deprotonate the allene to generate the alkyne is Asp233. An aspartate residue in this position is not an uncommon feature in related enzymes, and so this feature may have been retained, and its basicity enhanced by adding a nearby residue, His142, which seems to be unique for BesB. However, the nearby residues Tyr59, Arg61, or perhaps even Lys329 may also be involved in this step and should not be fully excluded at this stage. The roles of these residues are more difficult to interrogate, as they are all involved in binding PLP, any changes caused by disrupting these residues would most likely be due to reduced or altered PLP-binding. Finally, the latter two stages of the mechanism are most likely also shared with other related enzymes.240 Here 171  lysine catalyzes the reverse steps of the initial stages of the reaction, and in this process could incorporate two atoms of deuterium into 74 (Figure 4.29).  Figure 4.29 Proposed catalytic mechanism for BesB.   These studies of BesB have begun to provide significant insight into the PLP-dependent desaturation of 72. The crystal structure of the BesB monomer, the first crystal structure of an alkyne forming enzyme, shows that BesB adopts a typical cystathionine-\u03b2-lyase\/cystathionine-\u03b3-synthase fold. Its unique active site is modified to stabilize the reactive intermediates generated over the course of the reaction, while preventing undesirable inhibitory reactions from taking place. Further experimental, and computational studies will likely be required to work out the details of this reaction. For example, how favorable is the Cl\u2212 leaving? Does the presumed allene to alkyne isomerization take place on the enzyme or after product release? However, the unique reactions catalyzed by BesB show that is has potential to join the increasing number of PLP-dependent-\u03b3-lyases that display remarkable biocatalytic properties.241,242 The investigation of 172  BesB sets the stage for finding additional new monomeric cystathionine-\u03b3-synthase-like enzymes that catalyze unusual biosynthetic reactions.  173  Chapter 5: Conclusion Given the broad nature of the work covered in the thesis, the conclusions of each chapter will be discussed individually in the following text.  5.1 Chapter 2 Conclusions In Chapter 2 I describe the discovery and characterization a new arginine hydroxylase RohP, from a cryptic biosynthetic gene cluster. I determined that RohP catalyzes the oxidation of L-arginine to (S)-4-hydroxy-2-ketoarginine, which happens to be the same reaction that is catalyzed by the previously reported enduracididine biosynthetic enzyme MppP.157,160 Using the studies of Ind4 and MppP as a guide, I determined the oxygen requirements, kinetics, and stoichiometry of the RohP-catalyzed reaction with L-arginine, data which was not available for arginine hydroxylases at the time of the study. It is now known that all three enzymes are highly similar, and differ only in the outcome of the reaction, (S)-4-hydroxy-2-ketoarginine for RohP and MppP and 4,5-didehydroarginine for Ind4. Based on these results, it is highly probable that all three enzymes use similar catalytic mechanisms. The second major goal of this chapter was to obtain crystal structures of RohP. At the start of this project, there were no crystal structures of Ind4, and all structures of MppP were missing critical parts of the N-terminus, making it is difficult to generate a mechanistic hypothesis for these enzymes. Not only was I able to obtain crystal structures RohP which included the N-terminus, but I was able to capture RohP bound to different intermediates of its catalytic cycle. The crystal structures revealed the importance of a dynamic N-terminal region, as well as an N-terminal histidine residue which is conserved in both Ind4 and RohP-like enzymes that coordinates a water and is optimally positioned to catalyze a hydration reaction. These studies highlight the ability of X-ray crystallography to elucidate enzyme mechanisms by capturing otherwise unstable intermediates. Despite the significant amount of new data from this study, there are some questions that remain unanswered. One of the defining characteristics of RohP is its ability to keep quinonoid intermediates stable for extended periods of time, but it is not clear what structural features accomplish this. Second, how does RohP activate O2, what reactive oxygen species is generated, and finally where does it react on the PLP-arginine quinonoid? Finally, it is still not clear what 174  features distinguish arginine hydroxylases from arginine desaturases. Based on the available data, it appears that the two types of enzyme share many features and would differ only it the latter stages of the reaction, which makes it difficult to experimentally probe these differences in the reaction might be attributed to subtle changes in the dynamics of both types of enzymes. The recent discoveries of other O2- and PLP-dependent enzymes such as Cap15145 and CcuB147 suggests that many different biosynthetic pathways use PLP-dependent enzymes utilize O2 to catalyze unusual oxidative transformations. The field of O2- and PLP-dependent enzymes is still expanding expanding.148,149 In the future it will be exciting to see what other unusual O2- and PLP-dependent enzymes are discovered, and what mechanisms they use to activate oxygen.  5.2 Chapter 3 Conclusions In Chapter 3 I describe the characterization of four novel biosynthetic enzymes, which together with RohP catalyze the conversion of L-arginine to azomycin. Though the discovery of azomycin was crucial for developing the nitroimidazole class of drugs, its biosynthetic gene cluster had not been identified and characterized until this work. The findings of this in vitro work also verifies the longstanding azomycin biosynthetic hypothesis brought forth by Eguchi over 50 years ago (Figure 1.4).27 Though azomycin has only been isolated from a select few strains of S. eurocidicus and P. fluorescens, the bioinformatic analysis carried out in this chapter revealed that the azomycin biosynthetic gene cluster is widely distributed amongst Streptomyces, Pseudomonas and other closely related families of soil-dwelling bacteria including Microbispora and Actinokinespora. Using this newfound knowledge, a limited analysis of extracts from two Pseudomonas strains nonetheless identified a novel azomycin producer, P. brassicacearum DF41. Together, these two observations suggest that azomycin may a relatively common metabolite and may be produced by additional strains of soil-dwelling bacteria. Widespread production of azomycin would be indicative that the potent biological activity of azomycin and\/or related nitroimidazoles also have an ecological role in the rhizosphere. A catabolic pathway for azomycin has also been discovered in Mycobacterium sp. JS330.243 This bacterial strain produces an enzyme, pJS901, which catalyzes the hydrolytic denitration of azomycin to produce imidazol-2-one and nitrite, and this strain is capable of growing solely on azomycin. The presence of an azomycin catabolic pathway suggests 175  that levels of azomycin in natural ecosystems may fluctuate in response to environmental conditions, or the bacterial strains which are present. Therefore, the discovery of the azomycin biosynthetic gene cluster provides a useful platform for studies to investigate the ecological effects of azomycin between competing soil bacteria, and its potential roles in signaling, or possible symbiotic interactions with plants.  Looking more closely at the azomycin biosynthetic machinery reveals that the four enzymes characterized in this chapter all have novel enzymatic activities. These enzymes, and their respective activites are: RohQ, a cyclodehydratase, RohR, an aldolase, RohS, a non-heme diiron oxidase, and RohT a Rieske ferredoxin-like protein. RohQ is small, cofactor-less enzyme, which currently is found exclusively in putative azomycin biosynthetic gene clusters. This unusual and highly specialized role for RohQ makes it difficult to envision potential applications for this enzyme. However, if RohQ was found in novel biosynthetic gene cluster along with RohP and RohR homologs, this could be used as an indicator for the production of 2-aminoimidazole by the biosynthetic gene cluster. On the other hand, the aldolase RohR may eventually find use in a biocatalytic role, due to the extensive use of aldolases as biocatalysts.93,244,245 Additionally, close homologs of RohP and RohR are found together the biosynthetic gene cluster for the nucleoside antibiotic mildiomycin.246 Though not experimentally verified, based on the amino acid sequence of the RohP homolog in this biosynthetic gene cluster it is highly likely that it also produces (S)-4-hydroxy-2-ketoarginine, which is then added to the nucleoside core of mildiomycin by the RohR homolog. Therefore, based on this knowledge, the co-localization of RohP and RohR together without RohQ may serve as a guide towards the discovery of other novel nucleoside antibiotics. Perhaps the most interesting enzyme of these four is RohS, which catalyzes the six-electron oxidation of 2-nitroimidazole to azomycin. AurF and CmlI catalyze similar amino to nitro oxidations using distinct mechanisms, however RohS is not closely related to either of them, and therefore may represent yet another distinct oxidative route to the nitro group. Bioinformatic analysis places RohS as a non-heme diiron oxidase, an emerging superfamily of iron-dependent enzymes which catalyze an impressive array of challenging oxidative transformations.11,12 Newer members of this protein family include BesC from the \u03b2-ethynylserine biosynthetic pathway,115,116 the streptozotocin biosynthetic enzyme StzF,196,197 and the oxidative decarboxylase UndA.247 RohS was observed to have a weak affinity for binding iron, and consequently low stability and 176  low turnover rates, which hindered the study of this enzymes. The weak affinity for iron displayed by RohS seems to be a trait that is shared amongst StzF, UndA and other members of this protein family, and therefore may be an intrinsic feature of these enzymes. As such, considerable efforts have been made to properly reconstitute the diiron centers in these other enzymes.198,248 Once properly reconstituted with iron, the reconstituted diiron centers of UndA and StzF have both been shown to capture O2 to form a peroxo-Fe2 (III\/III) intermediate.248,249 Therefore, determining how to properly reconstitute the diiron center of RohS will be critical understanding how it catalyzes the oxidation of 2-aminoimidazole to azomycin. Reconstituting the diiron cluster of RohS will also allow for its interactions with RohT to be studied, as I was not able to definitively assign a role for RohT in the azomycin biosynthetic pathway. These experiments would contribute to the understanding of the emerging non-heme diiron oxidase superfamily.  Additionally, once RohS has been properly reconstituted this would also open the avenue for generating novel 2-nitroimidazole derivatives, as azomycin producing strains have been shown be able to catalyze the oxidation of a range of substituted 2-aminoimidazoles in vivo. A fully catalytically competent version of RohS may enable biocatalytic production of novel 2-nitroimidazole derivatives, or possibly other nitroaromatic derivatives, all of which could be generated on a large scale or possibly in vivo.  5.3 Chapter 4 Conclusions There are a select few enzymes known to catalyze formation of an alkyne, and relatively little is known about how these enzymes catalyze the formation of this functional group. The experiments outlined in Chapter 4 describe my contributions towards unraveling the enzymatic mechanism utilized by the PLP-dependent desaturase BesB. BesB catalyzes the conversion of 4-chloroallylglycine to L-propargylglycine, a rare terminal alkyne-containing amino acid. One of the major challenges of working with BesB was its low solubility in E. coli. Therefore, my first goal was to find an alternative route to produce soluble BesB. I was able to circumvent the problem of low BesB solubility in E. coli by expressing a BesB homolog in a Rhodococcus heterologous expression system, which produced significant amounts of soluble enzyme. I utilized the purified enzyme to conduct an array of in vitro tests to build upon what was previously known about BesB. I revealed that the reaction of BesB is strongly dependent on pH, 177  with the desaturation reaction not occurring at low pHs. The Cl\u2212 elimination is also revealed to be facile, and in fact, the presence (or absence) of Cl\u2212 in solvent can greatly affect the outcome of the BesB-catalyzed reaction. Importantly, all results were consistent with previous results which demonstrated that the mechanism employed by BesB requires two deprotonations to occur.115 In the future, the ease at which Cl\u2212 can be displaced from 4-chloroallylglycine may allow for BesB to be used as a biocatalyst for the generation of unnatural substituted vinylogous amino acids.   The other major goal of this chapter was to obtain a crystal structure of BesB. Again, the Rhodococcus expression system was crucial to this goal. Using protein from this system, I was able to obtain high-resolution crystal structures of wild-type BesB as well as the F231Y variant. These crystal structures represent the first crystal structures of any alkyne-forming biosynthetic enzyme. The crystal structures revealed that BesB exists as a monomer and uses a unique extended N-terminal sequence to retain similar catalytic properties as other homotetrameric CGLs and CGSs. Though I was able to obtain multiple crystal structures of BesB, I was unable to obtain any crystal structures of BesB that had either substrate or product in the active site. Therefore, future efforts towards crystallizing BesB with these molecules may still prove valuable. In this absence of any small molecule bound structures, it is still difficult to identify the catalytic base which catalyzes the final desaturation step. Several active site residues can be ruled out based on the chemical nature of the side chains. Of the remaining residues in the active site, mutating them generally proved detrimental to the solubility of BesB. I was unable to obtain this protein and by extension I was unable to experimentally test the effects of mutations to several important active site residues. Given the difficulties in obtaining additional empirical data about the BesB-catalyzed reaction, it may be useful to probe alternative routes of generating data. Future studies focusing on the mechanism of BesB may require theoretical calculations, which could be used to support the limited experimental data. These computational studies could include computational ligand docking as well as calculating the energies of PLP-intermediates in the pathway to verify if the proposed reaction pathways are energetically feasible. Such studies would shed led on the expanded PLP-dependent catalytic repertoire, and may be useful for identifying other novel PLP-dependent desaturases which could form allene or alkyne-containing natural products.250 178  5.4 Overall conclusions The studies described in this thesis characterized key biosynthetic enzymes in the azomycin and \u03b2-ethynylserine biosynthetic pathways. These two compounds contain nitro and alkyne functional groups, respectively. Both functional groups are uncommon in natural products and biosynthetic routes to both of these functional groups are still and underexplored area of natural product chemistry. Interestingly, from what is now known about both of these biosynthetic pathways is that despite both biosynthetic pathways being remarkably different, they each utilize common types of enzymes in new and exciting ways. Both pathways require unusual diiron enzymes, RohS, and BesC, which catalyze uncommon oxidative transformations. Both pathways also use fold type I PLP-dependent enzymes in new and exciting ways. The biosynthesis of azomycin requires an unusual O2- and PLP-dependent hydroxylase, RohP, while the biosynthesis of \u03b2-ethynylserine requires a novel PLP-dependent desaturase, BesB. It is remarkable that nature can utilize roughly the same type of enzyme to catalyze entirely different types of reactions. Insights gained by studying the biosynthetic enzymes in this pathway will provide a starting point for discovering even more enzymatic diversity and allow this diversity to be harnessed to produce valuable small molecules biocatalytically.  179  Bibliography (1)  Ertl, P.; Schuhmann, T. A Systematic Cheminformatics Analysis of Functional Groups Occurring in Natural Products. J. Nat. Prod. 2019, 82 (5), 1258\u20131263. (2)  Ertl, P.; Altmann, E.; McKenna, J. M. The Most Common Functional Groups in Bioactive Molecules and How Their Popularity Has Evolved over Time. J. Med. Chem. 2020, 63 (15), 8408\u20138418. (3)  Nepali, K.; Lee, H.-Y.; Liou, J.-P. Nitro-Group-Containing Drugs. J. Med. Chem. 2019, 62 (6), 2851\u20132893. (4)  Olender, D.; \u017bwawiak, J.; Zaprutko, L. Multidirectional Efficacy of Biologically Active Nitro Compounds Included in Medicines. Pharmaceuticals 2018, 11 (2), 54. (5)  Bauz\u00e1, A.; Frontera, A.; Mooibroek, T. J. \u03a0\u2010Hole Interactions Involving Nitro Aromatic Ligands in Protein Structures. Chem. \u2013 Eur. J. 2019, 25 (58), 13436\u201313443. (6)  Col\u00f3n, I. G.; Gonz\u00e1lez, F. A.; Cordero, M.; Zayas, B.; Velez, C.; Cox, O.; Kumar, A.; Alegr\u00eda, A. E. Role of the Nitro Functionality in the DNA Binding of 3-Nitro-10-Methylbenzothiazolo[3,2-a]Quinolinium Chloride. Chem. Res. Toxicol. 2008, 21 (9), 1706\u20131715. (7)  Moreno, S. N. J.; Docampo, R. Mechanism of Toxicity of Nitro Compounds Used in the Chemotherapy of Trichomoniasis. Environ Health Perspect 1986, 64, 199\u2013208. (8)  Winkler, R.; Hertweck, C. Biosynthesis of Nitro Compounds. ChemBioChem 2007, 8 (9), 973\u2013977. (9)  Parry, R.; Nishino, S.; Spain, J. Naturally-Occurring Nitro Compounds. Nat Prod Rep 2011, 28 (1), 152\u2013167. (10)  Komor, A. J.; Jasniewski, A. J.; Que, L.; Lipscomb, J. D. Diiron Monooxygenases in Natural Product Biosynthesis. Nat. Prod. Rep. 2018, 35 (7), 646\u2013659. (11)  Rajakovich, L. J.; Zhang, B.; McBride, M. J.; Boal, A. K.; Krebs, C.; Martin Bollinger, J. Emerging Structural and Functional Diversity in Proteins With Dioxygen-Reactive Dinuclear Transition Metal Cofactors. In Comprehensive Natural Products III; Elsevier, 2020; pp 215\u2013250. (12)  N\u00f3bile, M. L.; Stricker, A. M.; Marchesano, L.; Iribarren, A. M.; Lewkowicz, E. S. N-Oxygenation of Amino Compounds: Early Stages in Its Application to the Biocatalyzed Preparation of Bioactive Compounds. Biotechnol. Adv. 2021, 107726. (13)  Maeda, K.; Osato, T.; Umezawa, H. A New Antibiotic, Azomycin. J. Antibiot. 1953, 6 (4), 182. (14)  Nakamura, S. Structure of Azomycin, a New Antibiotic. Pharm. Bull. 1955, 3 (5), 379\u2013383. (15)  Voogd, C. E. On the Mutagenicity of Nitroimidazoles. Mutat. Res. Genet. Toxicol. 1981, 86 (3), 243\u2013277. (16)  Beaman, A. G.; Tautz, W.; Gabriel, T.; Duschinsky, R. The Synthesis of Azomycin. J. Am. Chem. Soc. 1965, 87 (2), 389\u2013390. (17)  Lancini, G. C.; Lazzari, E. The Synthesis of Azomycin (2-Nitroimidazole). Experientia 1965, 21 (2), 83\u201383. (18)  Edwards, D. I. Nitroimidazole Drugs-Action and Resistance Mechanisms. I. Mechanisms of Action. J. Antimicrob. Chemother. 1993, 31 (1), 9\u201320. 180  (19)  Ang, C. W.; Jarrad, A. M.; Cooper, M. A.; Blaskovich, M. A. T. Nitroimidazoles: Molecular Fireworks That Combat a Broad Spectrum of Infectious Diseases. J. Med. Chem. 2017, 60 (18), 7636\u20137657. (20)  Leitsch, D. A Review on Metronidazole: An Old Warhorse in Antimicrobial Chemotherapy. Parasitology 2019, 146 (9), 1167\u20131178. (21)  Dingsdag, S. A.; Hunter, N. Metronidazole: An Update on Metabolism, Structure\u2013Cytotoxicity and Resistance Mechanisms. J. Antimicrob. Chemother. 2018, 73 (2), 265\u2013279. (22)  Miyamoto, Y.; Kalisiak, J.; Korthals, K.; Lauwaet, T.; Cheung, D. Y.; Lozano, R.; Cobo, E. R.; Upcroft, P.; Upcroft, J. A.; Berg, D. E.; Gillin, F. D.; Fokin, V. V.; Sharpless, K. B.; Eckmann, L. Expanded Therapeutic Potential in Activity Space of Next-Generation 5-Nitroimidazole Antimicrobials with Broad Structural Diversity. Proc. Natl. Acad. Sci. 2013, 110 (43), 17564\u201317569. (23)  Mandalapu, D.; Kushwaha, B.; Gupta, S.; Singh, N.; Shukla, M.; Kumar, J.; Tanpula, D. K.; Sankhwar, S. N.; Maikhuri, J. P.; Siddiqi, M. I.; Lal, J.; Gupta, G.; Sharma, V. L. 2-Methyl-4\/5-Nitroimidazole Derivatives Potentiated against Sexually Transmitted Trichomonas\u202f: Design, Synthesis, Biology and 3D-QSAR Study. Eur. J. Med. Chem. 2016, 124, 820\u2013839. (24)  Matsumoto, M.; Hashizume, H.; Tomishige, T.; Kawasaki, M.; Tsubouchi, H.; Sasaki, H.; Shimokawa, Y.; Komatsu, M. OPC-67683, a Nitro-Dihydro-Imidazooxazole Derivative with Promising Action against Tuberculosis In Vitro and In Mice. PLoS Med. 2006, 3 (11), e466. (25)  Sotgiu, G.; Pontali, E.; Centis, R.; D\u2019Ambrosio, L.; Migliori, G. B. Delamanid (OPC-67683) for Treatment of Multi-Drug-Resistant Tuberculosis. Expert Rev. Anti Infect. Ther. 2015, 13 (3), 305\u2013315. (26)  McClelland, R. A.; Panicucci, R.; Rauth, A. M. Products of Reductions of 2-Nitroimidazoles. J. Am. Chem. Soc. 1987, 109 (14), 4308\u20134314. (27)  Nakane, A.; Nakamura, T.; Eguchi, Y. A Novel Metabolic Fate of Arginine in Streptomyces Eurocidicus. J. Biol. Chem. 1977, 252 (15), 5267\u20135273. (28)  Lancini, G. C.; Kluepfel, D.; Lazzari, E.; Sartori, G. Origin of the Nitro Group of Azomycin. Biochim. Biophys. Acta 1966, 130 (1), 37\u201344. (29)  Lancini, C. C.; Lazzari, E.; Sartori, C. Microbial Oxidation of Aminoimidazoles to Nitroimidazoles. J. Antibiot. 1968, 21 (6), 387\u2013392. (30)  Seki, Y.; Nakamura, T.; Okami, Y. Accumulation of 2-Aminoimidazole by Streptomyces Eurocidicus. J. Biochem. 1970, 67 (3), 389\u2013396. (31)  Kers, J. A.; Wach, M. J.; Krasnoff, S. B.; Widom, J.; Cameron, K. D.; Bukhalid, R. A.; Gibson, D. M.; Crane, B. R.; Loria, R. Nitration of a Peptide Phytotoxin by Bacterial Nitric Oxide Synthase. Nature 2004, 429 (6987), 79\u201382. (32)  Barry, S. M.; Kers, J. A.; Johnson, E. G.; Song, L.; Aston, P. R.; Patel, B.; Krasnoff, S. B.; Crane, B. R.; Gibson, D. M.; Loria, R.; Challis, G. L. Cytochrome P450\u2013Catalyzed L-Tryptophan Nitration in Thaxtomin Phytotoxin Biosynthesis. Nat. Chem. Biol. 2012, 8 (10), 814\u2013816. (33)  Wang, L.; Wang, M.; Fu, Y.; Huang, P.; Kong, D.; Niu, G. Engineered Biosynthesis of Thaxtomin Phytotoxins. Crit. Rev. Biotechnol. 2020, 40 (8), 1163\u20131171. 181  (34)  Pant, K.; Bilwes, A. M.; Adak, S.; Stuehr, D. J.; Crane, B. R. Structure of a Nitric Oxide Synthase Heme Protein from Bacillus Subtilis. Biochemistry 2002, 41 (37), 11071\u201311079. (35)  Dodani, S. C.; Cahn, J. K. B.; Heinisch, T.; Brinkmann-Chen, S.; McIntosh, J. A.; Arnold, F. H. Structural, Functional, and Spectroscopic Characterization of the Substrate Scope of the Novel Nitrating Cytochrome P450 TxtE. ChemBioChem 2014, 15 (15), 2259\u20132267. (36)  Yu, F.; Li, M.; Xu, C.; Wang, Z.; Zhou, H.; Yang, M.; Chen, Y.; Tang, L.; He, J. Structural Insights into the Mechanism for Recognizing Substrate of the Cytochrome P450 Enzyme TxtE. PLoS ONE 2013, 8 (11), e81526. (37)  Louka, S.; Barry, S. M.; Heyes, D. J.; Mubarak, M. Q. E.; Ali, H. S.; Alkhalaf, L. M.; Munro, A. W.; Scrutton, N. S.; Challis, G. L.; de Visser, S. P. Catalytic Mechanism of Aromatic Nitration by Cytochrome P450 TxtE: Involvement of a Ferric-Peroxynitrite Intermediate. J. Am. Chem. Soc. 2020, 142 (37), 15764\u201315779. (38)  Zuo, R.; Ding, Y. Direct Aromatic Nitration System for Synthesis of Nitrotryptophans in Escherichia Coli. ACS Synth. Biol. 2019, 8 (4), 857\u2013865. (39)  Zuo, R.; Zhang, Y.; Jiang, C.; Hackett, J. C.; Loria, R.; Bruner, S. D.; Ding, Y. Engineered P450 Biocatalysts Show Improved Activity and Regio-Promiscuity in Aromatic Nitration. Sci. Rep. 2017, 7 (1), 842. (40)  Ma, J.; Huang, H.; Xie, Y.; Liu, Z.; Zhao, J.; Zhang, C.; Jia, Y.; Zhang, Y.; Zhang, H.; Zhang, T.; Ju, J. Biosynthesis of Ilamycins Featuring Unusual Building Blocks and Engineered Production of Enhanced Anti-Tuberculosis Agents. Nat. Commun. 2017, 8 (1), 391. (41)  Tomita, H.; Katsuyama, Y.; Minami, H.; Ohnishi, Y. Identification and Characterization of a Bacterial Cytochrome P450 Monooxygenase Catalyzing the 3-Nitration of Tyrosine in Rufomycin Biosynthesis. J. Biol. Chem. 2017, 292 (38), 15859\u201315869. (42)  M\u00fcgge, C.; Heine, T.; Baraibar, A. G.; van Berkel, W. J. H.; Paul, C. E.; Tischler, D. Flavin-Dependent N-Hydroxylating Enzymes: Distribution and Application. Appl. Microbiol. Biotechnol. 2020, 104 (15), 6481\u20136499. (43)  Sugai, Y.; Katsuyama, Y.; Ohnishi, Y. A Nitrous Acid Biosynthetic Pathway for Diazo Group Formation in Bacteria. Nat. Chem. Biol. 2016, 12 (2), 73\u201375. (44)  Huang, Z.; Wang, K.-K. A.; van der Donk, W. A. New Insights into the Biosynthesis of Fosfazinomycin. Chem. Sci. 2016, 7 (8), 5219\u20135223. (45)  Katsuyama, Y.; Sato, Y.; Sugai, Y.; Higashiyama, Y.; Senda, M.; Senda, T.; Ohnishi, Y. Crystal Structure of the Nitrosuccinate Lyase CreD in Complex with Fumarate Provides Insights into the Catalytic Mechanism for Nitrous Acid Elimination. FEBS J. 2018, 285 (8), 1540\u20131555. (46)  Waldman, A. J.; Balskus, E. P. Discovery of a Diazo-Forming Enzyme in Cremeomycin Biosynthesis. J. Org. Chem. 2018, 83 (14), 7539\u20137546. (47)  Zhang, H.; White-Phillip, J. A.; Melan\u00e7on, C. E.; Kwon, H.; Yu, W.; Liu, H. Elucidation of the Kijanimicin Gene Cluster: Insights into the Biosynthesis of Spirotetronate Antibiotics and Nitrosugars. J. Am. Chem. Soc. 2007, 129 (47), 14670\u201314683. (48)  Timmons, S. C.; Thorson, J. S. Increasing Carbohydrate Diversity via Amine Oxidation: Aminosugar, Hydroxyaminosugar, Nitrososugar, and Nitrosugar Biosynthesis in Bacteria. Curr. Opin. Chem. Biol. 2008, 12 (3), 297\u2013305. (49)  Lin, C.-I.; McCarty, R. M.; Liu, H. The Biosynthesis of Nitrogen-, Sulfur-, and High-Carbon Chain-Containing Sugars. Chem. Soc. Rev. 2013, 42 (10), 4377\u20134407. 182  (50)  Hu, Y.; Al-Mestarihi, A.; Grimes, C. L.; Kahne, D.; Bachmann, B. O. A Unifying Nitrososynthase Involved in Nitrosugar Biosynthesis. J. Am. Chem. Soc. 2008, 130 (47), 15756\u201315757. (51)  Vey, J. L.; Al-Mestarihi, A.; Hu, Y.; Funk, M. A.; Bachmann, B. O.; Iverson, T. M. Structure and Mechanism of ORF36, an Amino Sugar Oxidizing Enzyme in Everninomicin Biosynthesis,. Biochemistry 2010, 49 (43), 9306\u20139317. (52)  Bruender, N. A.; Thoden, J. B.; Holden, H. M. X-Ray Structure of KijD3, a Key Enzyme Involved in the Biosynthesis of D-Kijanose. Biochemistry 2010, 49 (17), 3517\u20133524. (53)  Thoden, J. B.; Branch, M. C.; Zimmer, A. L.; Bruender, N. A.; Holden, H. M. Active Site Architecture of a Sugar N-Oxygenase. Biochemistry 2013, 52 (19), 3191\u20133193. (54)  Pietra, F. Binding Pockets and Pathways for Dioxygen through the KijD3 N-Oxygenase in Complex with Flavin Mononucleotide Cofactor and a 3-Aminoglucose Substrate: Predictions from Molecular Dynamics Simulations. Chem. Biodivers. 2014, 11 (8), 1151\u20131162. (55)  Li, Y.; Ding, L.; Zhang, Q.; Wang, W. MD and QM\/MM Study on Catalytic Mechanism of a FAD-Dependent Enzyme ORF36: For Nitro Sugar Biosynthesis. J. Mol. Graph. Model. 2013, 44, 9\u201316. (56)  Bannister, B.; Zapotocky, B. A. Protorubradin, an Antibiotic Containing a C-Nitroso-Sugar Fragment Is the True Secondary Metabolite Produced by Streptomyces Achromogenes Var. Rubradiris. J. Antibiot. (Tokyo) 1992, 45 (8), 1313\u20131324. (57)  Franke, J.; Ishida, K.; Ishida-Ito, M.; Hertweck, C. Nitro versus Hydroxamate in Siderophores of Pathogenic Bacteria: Effect of Missing Hydroxylamine Protection in Malleobactin Biosynthesis. Angew. Chem. Int. Ed. 2013, 52 (32), 8271\u20138275. (58)  Nakamura, H.; Matsuda, Y.; Abe, I. Unique Chemistry of Non-Heme Iron Enzymes in Fungal Biosynthetic Pathways. Nat. Prod. Rep. 2018, 35 (7), 633\u2013645. (59)  Gordee, R. S.; Matthews, T. R. Systemic Antifungal Activity of Pyrrolnitrin. Appl Microbiol 1969, 17 (5), 690\u2013694. (60)  Pawar, S.; Chaudhari, A.; Prabha, R.; Shukla, R.; Singh, D. P. Microbial Pyrrolnitrin: Natural Metabolite with Immense Practical Utility. Biomolecules 2019, 9 (9), 443. (61)  Lee, J.; Simurdiak, M.; Zhao, H. Reconstitution and Characterization of Aminopyrrolnitrin Oxygenase, a Rieske N-Oxygenase That Catalyzes Unusual Arylamine Oxidation. J. Biol. Chem. 2005, 280 (44), 36719\u201336727. (62)  Lee, J.-K.; Ang, E.-L.; Zhao, H. Probing the Substrate Specificity of Aminopyrrolnitrin Oxygenase (PrnD) by Mutational Analysis. J. Bacteriol. 2006, 188 (17), 6179\u20136183. (63)  Tiwari, M. K.; Lee, J.-K.; Moon, H.-J.; Zhao, H. Further Biochemical Studies on Aminopyrrolnitrin Oxygenase (PrnD). Bioorg. Med. Chem. Lett. 2011, 21 (10), 2873\u20132876. (64)  He, J.; Hertweck, C. Biosynthetic Origin of the Rare Nitroaryl Moiety of the Polyketide Antibiotic Aureothin: Involvement of an Unprecedented N-Oxygenase. J. Am. Chem. Soc. 2004, 126 (12), 3694\u20133695. (65)  Schaffer, J. E.; Reck, M. R.; Prasad, N. K.; Wencewicz, T. A. \u03b2-Lactone Formation during Product Release from a Nonribosomal Peptide Synthetase. Nat. Chem. Biol. 2017, 13 (7), 737\u2013744. (66)  Winkler, R.; Hertweck, C. Sequential Enzymatic Oxidation of Aminoarenes to Nitroarenes via Hydroxylamines. Angew. Chem. Int. Ed. 2005, 44 (26), 4083\u20134087. 183  (67)  Winkler, R.; Richter, M. E. A.; Kn\u00fcpfer, U.; Merten, D.; Hertweck, C. Regio- and Chemoselective Enzymatic N-Oxygenation In Vivo, In Vitro, and in Flow. Angew. Chem. Int. Ed. 2006, 45 (47), 8016\u20138018. (68)  Winkler, R.; Zocher, G.; Richter, I.; Friedrich, T.; Schulz, G. E.; Hertweck, C. A Binuclear Manganese Cluster That Catalyzes Radical-Mediated N-Oxygenation. Angew. Chem. Int. Ed. 2007, 46 (45), 8605\u20138608. (69)  Choi, Y. S.; Zhang, H.; Brunzelle, J. S.; Nair, S. K.; Zhao, H. In Vitro Reconstitution and Crystal Structure of p-Aminobenzoate N-Oxygenase (AurF) Involved in Aureothin Biosynthesis. Proc. Natl. Acad. Sci. 2008, 105 (19), 6858\u20136863. (70)  Simurdiak, M.; Lee, J.; Zhao, H. A New Class of Arylamine Oxygenases: Evidence That p-Aminobenzoate N-Oxygenase (AurF) Is a Di-Iron Enzyme and Further Mechanistic Studies. ChemBioChem 2006, 7 (8), 1169\u20131172. (71)  Li, N.; Korboukh, V. K.; Krebs, C.; Bollinger, J. M. Four-Electron Oxidation of p-Hydroxylaminobenzoate to p-Nitrobenzoate by a Peroxodiferric Complex in AurF from Streptomyces Thioluteus. Proc. Natl. Acad. Sci. 2010, 107 (36), 15722\u201315727. (72)  Park, K.; Li, N.; Kwak, Y.; Srnec, M.; Bell, C. B.; Liu, L. V.; Wong, S. D.; Yoda, Y.; Kitao, S.; Seto, M.; Hu, M.; Zhao, J.; Krebs, C.; Bollinger, J. M.; Solomon, E. I. Peroxide Activation for Electrophilic Reactivity by the Binuclear Non-Heme Iron Enzyme AurF. J. Am. Chem. Soc. 2017, 139 (20), 7062\u20137070. (73)  Lu, H.; Chanco, E.; Zhao, H. CmlI Is an N-Oxygenase in the Biosynthesis of Chloramphenicol. Tetrahedron 2012, 68 (37), 7651\u20137654. (74)  Makris, T. M.; Vu, V. V.; Meier, K. K.; Komor, A. J.; Rivard, B. S.; M\u00fcnck, E.; Que, L.; Lipscomb, J. D. An Unusual Peroxo Intermediate of the Arylamine Oxygenase of the Chloramphenicol Biosynthetic Pathway. J. Am. Chem. Soc. 2015, 137 (4), 1608\u20131617. (75)  Knoot, C. J.; Kovaleva, E. G.; Lipscomb, J. D. Crystal Structure of CmlI, the Arylamine Oxygenase from the Chloramphenicol Biosynthetic Pathway. J. Biol. Inorg. Chem. 2016, 21 (5\u20136), 589\u2013603. (76)  Jasniewski, A. J.; Komor, A. J.; Lipscomb, J. D.; Que, L. Unprecedented (\u03bc-1,1-Peroxo)Diferric Structure for the Ambiphilic Orange Peroxo Intermediate of the Nonheme N-Oxygenase CmlI. J. Am. Chem. Soc. 2017, 139 (30), 10472\u201310485. (77)  Komor, A. J.; Rivard, B. S.; Fan, R.; Guo, Y.; Que, L.; Lipscomb, J. D. Mechanism for Six-Electron Aryl-N-Oxygenation by the Non-Heme Diiron Enzyme CmlI. J. Am. Chem. Soc. 2016, 138 (23), 7411\u20137421. (78)  Komor, A. J.; Rivard, B. S.; Fan, R.; Guo, Y.; Que, L.; Lipscomb, J. D. CmlI N-Oxygenase Catalyzes the Final Three Steps in Chloramphenicol Biosynthesis without Dissociation of Intermediates. Biochemistry 2017, 56 (37), 4940\u20134950. (79)  Minto, R. E.; Blacklock, B. J. Biosynthesis and Function of Polyacetylenes and Allied Natural Products. Prog. Lipid Res. 2008, 47 (4), 233\u2013306. (80)  Tian, Y.; Wei, X.; Xu, H. Photoactivated Insecticidal Thiophene Derivatives from Xanthopappus s Ubacaulis. J. Nat. Prod. 2006, 69 (8), 1241\u20131244. (81)  Guo, Y.; Gavagnin, M.; Trivellone, E.; Cimino, G. Further Structural Studies on the Petroformynes. J. Nat. Prod. 1995, 58 (5), 712\u2013722. (82)  Cohen, D. R.; Townsend, C. A. Characterization of an Anthracene Intermediate in Dynemicin Biosynthesis. Angew. Chem. Int. Ed. 2018, 57 (20), 5650\u20135654. 184  (83)  Van Lanen, S.; Shen, B. Biosynthesis of Enediyne Antitumor Antibiotics. Curr. Top. Med. Chem. 2008, 8 (6), 448\u2013459. (84)  Biggins, J. B. Resistance to Enediyne Antitumor Antibiotics by CalC Self-Sacrifice. Science 2003, 301 (5639), 1537\u20131541. (85)  Singh, S.; Hager, M. H.; Zhang, C.; Griffith, B. R.; Lee, M. S.; Hallenga, K.; Markley, J. L.; Thorson, J. S. Structural Insight into the Self-Sacrifice Mechanism of Enediyne Resistance. ACS Chem. Biol. 2006, 1 (7), 451\u2013460. (86)  Shen, B.; Hindra; Yan, X.; Huang, T.; Ge, H.; Yang, D.; Teng, Q.; Rudolf, J. D.; Lohman, J. R. Enediynes: Exploration of Microbial Genomics to Discover New Anticancer Drug Leads. Bioorg. Med. Chem. Lett. 2015, 25 (1), 9\u201315. (87)  Adhikari, A.; Teijaro, C. N.; Townsend, C. A.; Shen, B. Biosynthesis of Enediyne Natural Products. In Comprehensive Natural Products III; Elsevier, 2020; pp 365\u2013414. (88)  Yan, X.; Ge, H.; Huang, T.; Hindra; Yang, D.; Teng, Q.; Crnov\u010di\u0107, I.; Li, X.; Rudolf, J. D.; Lohman, J. R.; Gansemans, Y.; Zhu, X.; Huang, Y.; Zhao, L.-X.; Jiang, Y.; Van Nieuwerburgh, F.; Rader, C.; Duan, Y.; Shen, B. Strain Prioritization and Genome Mining for Enediyne Natural Products. mBio 2016, 7 (6), e02104-16. (89)  Belecki, K.; Townsend, C. A. Biochemical Determination of Enzyme-Bound Metabolites: Preferential Accumulation of a Programmed Octaketide on the Enediyne Polyketide Synthase CalE8. J. Am. Chem. Soc. 2013, 135 (38), 14339\u201314348. (90)  Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Click Chemistry: Diverse Chemical Function from a Few Good Reactions. Angew Chem Int Ed 2001, 40 (11), 2004\u20132021. (91)  Sokolova, N. V.; Nenajdenko, V. G. Recent Advances in the Cu(I)-Catalyzed Azide\u2013Alkyne Cycloaddition: Focus on Functionally Substituted Azides and Alkynes. RSC Adv. 2013, 3 (37), 16212. (92)  Jiang, X.; Hao, X.; Jing, L.; Wu, G.; Kang, D.; Liu, X.; Zhan, P. Recent Applications of Click Chemistry in Drug Discovery. Expert Opin. Drug Discov. 2019, 14 (8), 779\u2013789. (93)  Huffman, M. A.; Fryszkowska, A.; Alvizo, O.; Borra-Garske, M.; Campos, K. R.; Canada, K. A.; Devine, P. N.; Duan, D.; Forstater, J. H.; Grosser, S. T.; Halsey, H. M.; Hughes, G. J.; Jo, J.; Joyce, L. A.; Kolev, J. N.; Liang, J.; Maloney, K. M.; Mann, B. F.; Marshall, N. M.; McLaughlin, M.; Moore, J. C.; Murphy, G. S.; Nawrat, C. C.; Nazor, J.; Novick, S.; Patel, N. R.; Rodriguez-Granillo, A.; Robaire, S. A.; Sherer, E. C.; Truppo, M. D.; Whittaker, A. M.; Verma, D.; Xiao, L.; Xu, Y.; Yang, H. Design of an in Vitro Biocatalytic Cascade for the Manufacture of Islatravir. Science 2019, 366 (6470), 1255\u20131259. (94)  Bastos, M. M.; Costa, C. C. P.; Bezerra, T. C.; da Silva, F. de C.; Boechat, N. Efavirenz a Nonnucleoside Reverse Transcriptase Inhibitor of First-Generation: Approaches Based on Its Medicinal Chemistry. Eur. J. Med. Chem. 2016, 108, 455\u2013465. (95)  Lee, M. Identification of Non-Heme Diiron Proteins That Catalyze Triple Bond and Epoxy Group Formation. Science 1998, 280 (5365), 915\u2013918. (96)  Gagn\u00e9, S. J.; Reed, D. W.; Gray, G. R.; Covello, P. S. Structural Control of Chemoselectivity, Stereoselectivity, and Substrate Specificity in Membrane-Bound Fatty Acid Acetylenases and Desaturases. Biochemistry 2009, 48 (51), 12298\u201312304. (97)  Reed, D. W.; Polichuk, D. R.; Buist, P. H.; Ambrose, S. J.; Sasata, R. J.; Savile, C. K.; Ross, A. R. S.; Covello, P. S. Mechanistic Study of an Improbable Reaction: Alkene 185  Dehydrogenation by the \u039412 Acetylenase of Crepis Alpina. J. Am. Chem. Soc. 2003, 125 (35), 10635\u201310640. (98)  Carlsson, A. S.; Thomaeus, S.; Hamberg, M.; Stymne, S. Properties of Two Multifunctional Plant Fatty Acid Acetylenase\/Desaturase Enzymes: Stereochemistry of Acetylenase\/Desaturase Enzymes. Eur. J. Biochem. 2004, 271 (14), 2991\u20132997. (99)  Shanklin, J.; Guy, J. E.; Mishra, G.; Lindqvist, Y. Desaturases: Emerging Models for Understanding Functional Diversification of Diiron-Containing Enzymes. J. Biol. Chem. 2009, 284 (28), 18559\u201318563. (100)  Nagao, K.; Murakami, A.; Umeda, M. Structure and Function of \u03949-Fatty Acid Desaturase. Chem. Pharm. Bull. (Tokyo) 2019, 67 (4), 327\u2013332. (101)  Bai, Y.; McCoy, J. G.; Levin, E. J.; Sobrado, P.; Rajashankar, K. R.; Fox, B. G.; Zhou, M. X-Ray Structure of a Mammalian Stearoyl-CoA Desaturase. Nature 2015, 524 (7564), 252\u2013256. (102)  Jeon, J. E.; Kim, J.-G.; Fischer, C. R.; Mehta, N.; Dufour-Schroif, C.; Wemmer, K.; Mudgett, M. B.; Sattely, E. A Pathogen-Responsive Gene Cluster for Highly Modified Fatty Acids in Tomato. Cell 2020, 180 (1), 176-187.e19. (103)  Haritos, V. S.; Horne, I.; Damcevski, K.; Glover, K.; Gibb, N.; Okada, S.; Hamberg, M. The Convergent Evolution of Defensive Polyacetylenic Fatty Acid Biosynthesis Genes in Soldier Beetles. Nat. Commun. 2012, 3 (1), 1150. (104)  Ross, C.; Scherlach, K.; Kloss, F.; Hertweck, C. The Molecular Basis of Conjugated Polyyne Biosynthesis in Phytopathogenic Bacteria. Angew. Chem. Int. Ed. 2014, 53 (30), 7794\u20137798. (105)  Chen, Y.; Naresh, A.; Liang, S.; Lin, C.; Chein, R.; Lin, H. Discovery of a Dual Function Cytochrome P450 That Catalyzes Enyne Formation in Cyclohexanoid Terpenoid Biosynthesis. Angew. Chem. Int. Ed. 2020, 59 (32), 13537\u201313541. (106)  Lv, J.; Gao, Y.; Zhao, H.; Awakawa, T.; Liu, L.; Chen, G.; Yao, X.; Hu, D.; Abe, I.; Gao, H. Biosynthesis of Biscognienyne B Involving a Cytochrome P450\u2010Dependent Alkynylation. Angew. Chem. Int. Ed. 2020, 59 (32), 13531\u201313536. (107)  Edwards, D. J.; Marquez, B. L.; Nogle, L. M.; McPhail, K.; Goeger, D. E.; Roberts, M. A.; Gerwick, W. H. Structure and Biosynthesis of the Jamaicamides, New Mixed Polyketide-Peptide Neurotoxins from the Marine Cyanobacterium Lyngbya Majuscula. Chem. Biol. 2004, 11 (6), 817\u2013833. (108)  Jones, A. C.; Monroe, E. A.; Podell, S.; Hess, W. R.; Klages, S.; Esquenazi, E.; Niessen, S.; Hoover, H.; Rothmann, M.; Lasken, R. S.; Yates, J. R.; Reinhardt, R.; Kube, M.; Burkart, M. D.; Allen, E. E.; Dorrestein, P. C.; Gerwick, W. H.; Gerwick, L. Genomic Insights into the Physiology and Ecology of the Marine Filamentous Cyanobacterium Lyngbya Majuscula. Proc. Natl. Acad. Sci. 2011, 108 (21), 8815\u20138820. (109)  Almaliti, J.; Malloy, K. L.; Glukhov, E.; Spadafora, C.; Guti\u00e9rrez, M.; Gerwick, W. H. Dudawalamides A\u2013D, Antiparasitic Cyclic Depsipeptides from the Marine Cyanobacterium Moorea Producens. J. Nat. Prod. 2017, 80 (6), 1827\u20131836. (110)  Zhu, X.; Liu, J.; Zhang, W. De Novo Biosynthesis of Terminal Alkyne-Labeled Natural Products. Nat. Chem. Biol. 2015, 11 (2), 115\u2013120. (111)  Zhu, X.; Su, M.; Manickam, K.; Zhang, W. Bacterial Genome Mining of Enzymatic Tools for Alkyne Biosynthesis. ACS Chem. Biol. 2015, 10 (12), 2785\u20132793. 186  (112)  Zhu, X.; Zhang, W. Terminal Alkyne Biosynthesis in Marine Microbes. In Methods in Enzymology; Elsevier, 2018; Vol. 604, pp 89\u2013112. (113)  Porterfield, W. B.; Poenateetai, N.; Zhang, W. Engineered Biosynthesis of Alkyne-Tagged Polyketides by Type I PKSs. iScience 2020, 23 (3), 100938. (114)  Sanada, M.; Miyano, T.; Iwadare, S. \u03b2-Ethynylserine, an Antimetabolite of L-Threonine, from Streptomyces Cattleya. J. Antibiot. (Tokyo) 1986, 39 (2), 304\u2013305. (115)  Marchand, J. A.; Neugebauer, M. E.; Ing, M. C.; Lin, C.-I.; Pelton, J. G.; Chang, M. C. Y. Discovery of a Pathway for Terminal-Alkyne Amino Acid Biosynthesis. Nature 2019, 567 (7748), 420\u2013424. (116)  Neugebauer, M. E.; Sumida, K. H.; Pelton, J. G.; McMurry, J. L.; Marchand, J. A.; Chang, M. C. Y. A Family of Radical Halogenases for the Engineering of Amino-Acid-Based Products. Nat. Chem. Biol. 2019, 15 (10), 1009\u20131016. (117)  Du, Y.-L.; Ryan, K. S. Pyridoxal Phosphate-Dependent Reactions in the Biosynthesis of Natural Products. Nat. Prod. Rep. 2019, 36 (3), 430\u2013457. (118)  Percudani, R.; Peracchi, A. A Genomic Overview of Pyridoxal\u2010phosphate\u2010dependent Enzymes. EMBO Rep. 2003, 4 (9), 850\u2013854. (119)  Toney, M. D. Reaction Specificity in Pyridoxal Phosphate Enzymes. Arch. Biochem. Biophys. 2005, 433 (1), 279\u2013287. (120)  Toney, M. D. Controlling Reaction Specificity in Pyridoxal Phosphate Enzymes. Biochim. Biophys. Acta 2011, 1814 (11), 1407\u20131418. (121)  Percudani, R.; Peracchi, A. The B6 Database: A Tool for the Description and Classification of Vitamin B6-Dependent Enzymatic Activities and of the Corresponding Protein Families. BMC Bioinformatics 2009, 10 (1), 273. (122)  Dunathan, H. C. Conformation and Reaction Specificity in Pyridoxal Phosphate Enzymes. Proc. Natl. Acad. Sci. 1966, 55 (4), 712\u2013716. (123)  Grishin, N. V.; Phillips, M. A.; Goldsmith, E. A. Modeling of the Spatial Structure of Eukaryotic Ornithine Decarboxylases. Protein Sci. 1995, 4 (7), 1291\u20131304. (124)  Eliot, A. C.; Kirsch, J. F. Pyridoxal Phosphate Enzymes: Mechanistic, Structural, and Evolutionary Considerations. Annu. Rev. Biochem. 2004, 73 (1), 383\u2013415. (125)  Liang, J.; Han, Q.; Tan, Y.; Ding, H.; Li, J. Current Advances on Structure-Function Relationships of Pyridoxal 5\u2032-Phosphate-Dependent Enzymes. Front. Mol. Biosci. 2019, 6, 1\u201321. (126)  Schneider, G.; K\u00e4ck, H.; Lindqvist, Y. The Manifold of Vitamin B6 Dependent Enzymes. Structure 2000, 8 (1), R1\u2013R6. (127)  Kirsch, J. F.; Eichele, G.; Ford, G. C.; Vincent, M. G.; Jansonius, J. N.; Gehring, H.; Christen, P. Mechanism of Action of Aspartate Aminotransferase Proposed on the Basis of Its Spatial Structure. J. Mol. Biol. 1984, 174 (3), 497\u2013525. (128)  Ruggieri, F. Insight into the Dimer Dissociation Process of the Chromobacterium Violaceum (S)-Selective Amine Transaminase. Sci. Rep. 2019, 9, 16946. (129)  Islam, M. M.; Goto, M.; Miyahara, I.; Ikushiro, H.; Hirotsu, K.; Hayashi, H. Binding of C5-Dicarboxylic Substrate to Aspartate Aminotransferase: Implications for the Conformational Change at the Transaldimination Step. Biochemistry 2005, 44 (23), 8218\u20138229. 187  (130)  Denesyuk, A. I.; Denessiouk, K. A.; Korpela, T.; Johnson, M. S. Functional Attributes of the Phosphate Group Binding Cup of Pyridoxal Phosphate-Dependent Enzymes. J. Mol. Biol. 2002, 316 (1), 155\u2013172. (131)  Ferla, M. P.; Patrick, W. M. Bacterial Methionine Biosynthesis. Microbiology 2014, 160 (8), 1571\u20131584. (132)  Aitken, S. M.; Lodha, P. H.; Morneau, D. J. K. The Enzymes of the Transsulfuration Pathways: Active-Site Characterizations. Biochim. Biophys. Acta 2011, 1814 (11), 1511\u20131517. (133)  El-Gebali, S.; Mistry, J.; Bateman, A.; Eddy, S. R.; Luciani, A.; Potter, S. C.; Qureshi, M.; Richardson, L. J.; Salazar, G. A.; Smart, A.; Sonnhammer, E. L. L.; Hirsh, L.; Paladin, L.; Piovesan, D.; Tosatto, S. C. E.; Finn, R. D. The Pfam Protein Families Database in 2019. Nucleic Acids Res. 2019, 47 (D1), D427\u2013D432. (134)  Sbodio, J. I.; Snyder, S. H.; Paul, B. D. Regulators of the Transsulfuration Pathway: Modulating the Transsulfuration Pathway in Disease. Br. J. Pharmacol. 2019, 176 (4), 583\u2013593. (135)  Revtovich, S.; Anufrieva, N.; Morozova, E.; Kulikova, V.; Nikulin, A.; Demidkina, T. Structure of Methionine \u03b3-Lyase from Clostridium Sporogenes. Acta Crystallogr. Sect. F Struct. Biol. Commun. 2016, 72 (1), 65\u201371. (136)  Walsh, C. T.; Abeles, R. H. Acetylenic Enzyme Inactivators. Inactivation of \u03b3-Cystathionase, in Vitro and in Vivo, by Propargylglycine. J. Am. Chem. Soc. 1973, 95 (18), 6124\u20136125. (137)  Washtien, W.; Abeles, R. H. Mechanism of Inactivation of \u03b3-Cystathionase by the Acetylenic Substrate Analogue Propargylglycine. Biochemistry 1979, 16 (11), 2485\u20132491. (138)  Johnston, M.; Jankowski, D.; Marcotte, P.; Tanaka, H.; Esaki, N.; Soda, K.; Walsh, C. Suicide Inactivation of Bacterial Cystathionine \u03b3-Synthase and Methionine \u03b3-Lyase during Processing of L-Propargylglycine. Biochemistry 1979, 18 (21), 4690\u20134701. (139)  Esaki, N.; Takada, H.; Moriguchi, M.; Hatanaka, S.; Tanaka, H.; Soda, K. Mechanism-Based Inactivation of L-Methionine \u03b3-Lyase by L-2-Amino-4-Chloro-4-Pentenoate. Biochemistry 1989, 28 (5), 2111\u20132116. (140)  Sun, Q.; Collins, R.; Huang, S.; Holmberg-Schiavone, L.; Anand, G. S.; Tan, C.-H.; van-den-Berg, S.; Deng, L.-W.; Moore, P. K.; Karlberg, T.; Sivaraman, J. Structural Basis for the Inhibition Mechanism of Human Cystathionine \u03b3-Lyase, an Enzyme Responsible for the Production of H2S. J. Biol. Chem. 2009, 284 (5), 3076\u20133085. (141)  Asimakopoulou, A.; Panopoulos, P.; Chasapis, C. T.; Coletta, C.; Zhou, Z.; Cirino, G.; Giannis, A.; Szabo, C.; Spyroulias, G. A.; Papapetropoulos, A. Selectivity of Commonly Used Pharmacological Inhibitors for Cystathionine \u03b2 Synthase (CBS) and Cystathionine \u03b3 Lyase (CSE): Characterization of CSE and CBS Inhibitors. Br. J. Pharmacol. 2013, 169 (4), 922\u2013932. (142)  Bunik, V. I.; Schloss, J. V.; Pinto, J. T.; Dudareva, N.; Cooper, A. J. L. A Survey of Oxidative Paracatalytic Reactions Catalyzed by Enzymes That Generate Carbanionic Intermediates: Implications for ROS Production, Cancer Etiology, and Neurodegenerative Diseases. Adv. Enzymol. Relat. Areas Mol. Biol. 2011, 77, 307\u2013360. (143)  Bertoldi, M. Mammalian Dopa Decarboxylase: Structure, Catalytic Activity and Inhibition. Arch. Biochem. Biophys. 2014, 546, 1\u20137. 188  (144)  Kaminaga, Y.; Schnepp, J.; Peel, G.; Kish, C. M.; Ben-Nissan, G.; Weiss, D.; Orlova, I.; Lavie, O.; Rhodes, D.; Wood, K.; Porterfield, D. M.; Cooper, A. J. L.; Schloss, J. V.; Pichersky, E.; Vainstein, A.; Dudareva, N. Plant Phenylacetaldehyde Synthase Is a Bifunctional Homotetrameric Enzyme That Catalyzes Phenylalanine Decarboxylation and Oxidation. J. Biol. Chem. 2006, 281 (33), 23357\u201323366. (145)  Huang, Y.; Liu, X.; Cui, Z.; Wiegmann, D.; Niro, G.; Ducho, C.; Song, Y.; Yang, Z.; Van Lanen, S. G. Pyridoxal-5\u2032-Phosphate as an Oxygenase Cofactor: Discovery of a Carboxamide-Forming, \u03b1-Amino Acid Monooxygenase-Decarboxylase. Proc. Natl. Acad. Sci. 2018, 115 (5), 974\u2013979. (146)  Wang, M.; Zhao, Q.; Zhang, Q.; Liu, W. Differences in PLP-Dependent Cysteinyl Processing Lead to Diverse S-Functionalization of Lincosamide Antibiotics. J. Am. Chem. Soc. 2016, 138 (20), 6348\u20136351. (147)  Dai, G. Z.; Han, W. B.; Mei, Y. N.; Xu, K.; Jiao, R. H.; Ge, H. M.; Tan, R. X. Pyridoxal-5\u2032-Phosphate\u2013Dependent Bifunctional Enzyme Catalyzed Biosynthesis of Indolizidine Alkaloids in Fungi. Proc. Natl. Acad. Sci. 2020, 117 (2), 1174. (148)  Hoffarth, E. R.; Rothchild, K. W.; Ryan, K. S. Emergence of Oxygen\u2010 and Pyridoxal Phosphate\u2010dependent Reactions. FEBS J. 2020, 287 (7), 1403\u20131428. (149)  Bisello, G.; Longo, C.; Rossignoli, G.; Phillips, R. S.; Bertoldi, M. Oxygen Reactivity with Pyridoxal 5\u2032-Phosphate Enzymes: Biochemical Implications and Functional Relevance. Amino Acids 2020, 52 (8), 1089\u20131105. (150)  Du, Y.-L.; Alkhalaf, L. M.; Ryan, K. S. In Vitro Reconstitution of Indolmycin Biosynthesis Reveals the Molecular Basis of Oxazolinone Assembly. Proc. Natl. Acad. Sci. 2015, 112 (9), 2717\u20132722. (151)  Du, Y.-L.; Higgins, M. A.; Zhao, G.; Ryan, K. S. Convergent Biosynthetic Transformations to a Bacterial Specialized Metabolite. Nat. Chem. Biol. 2019, 15 (11), 1043\u20131048. (152)  Du, Y. L.; Singh, R.; Alkhalaf, L. M.; Kuatsjah, E.; He, H. Y.; Eltis, L. D.; Ryan, K. S. A Pyridoxal Phosphate-Dependent Enzyme That Oxidizes an Unactivated Carbon-Carbon Bond. Nat. Chem. Biol. 2016, 12 (3), 194\u2013199. (153)  Guo, J.; Higgins, M. A.; Daniel-Ivad, P.; Ryan, K. S. An Asymmetric Reductase That Intercepts Acyclic Imino Acids Produced in Situ by a Partner Oxidase. J. Am. Chem. Soc. 2019, 141 (31), 12258\u201312267. (154)  Atkinson, D. J.; Naysmith, B. J.; Furkert, D. P.; Brimble, M. A. Enduracididine, a Rare Amino Acid Component of Peptide Antibiotics: Natural Products and Synthesis. Beilstein J. Org. Chem. 2016, 12, 2325\u20132342. (155)  Yin, X.; Zabriskie, T. M. The Enduracidin Biosynthetic Gene Cluster from Streptomyces Fungicidicus. Microbiology 2006, 152 (10), 2969\u20132983. (156)  Hatano, K.; Nogami, I.; Higashide, E.; Kishi, T. Biosynthesis of Enduracidin: Origin of Enduracididine and Other Amino Acids. Agric. Biol. Chem. 1984, 48 (6), 1503\u20131508. (157)  Han, L.; Schwabacher, A. W.; Moran, G. R.; Silvaggi, N. R. Streptomyces Wadayamensis MppP Is a Pyridoxal 5\u2032-Phosphate-Dependent L-Arginine \u03b1-Deaminase, \u03b3-Hydroxylase in the Enduracididine Biosynthetic Pathway. Biochemistry 2015, 54 (47), 7029\u20137040. (158)  Burroughs, A. M.; Hoppe, R. W.; Goebel, N. C.; Sayyed, B. H.; Voegtline, T. J.; Schwabacher, A. W.; Zabriskie, T. M.; Silvaggi, N. R. Structural and Functional Characterization of MppR, an Enduracididine Biosynthetic Enzyme from Streptomyces 189  Hygroscopicus\u202f: Functional Diversity in the Acetoacetate Decarboxylase-like Superfamily. Biochemistry 2013, 52 (26), 4492\u20134506. (159)  Haltli, B.; Tan, Y.; Magarvey, N. A.; Wagenaar, M.; Yin, X.; Greenstein, M.; Hucul, J. A.; Zabriskie, T. M. Investigating \u03b2-Hydroxyenduracididine Formation in the Biosynthesis of the Mannopeptimycins. Chem. Biol. 2005, 12 (11), 1163\u20131168. (160)  Han, L.; Vuksanovic, N.; Oehm, S. A.; Fenske, T. G.; Schwabacher, A. W.; Silvaggi, N. R. Streptomyces Wadayamensis MppP Is a PLP-Dependent Oxidase, Not an Oxygenase. Biochemistry 2018, 57 (23), 3252\u20133264. (161)  Labinger, J. A.; Bercaw, J. E. Understanding and Exploiting C\u2013H Bond Activation. Nature 2002, 417 (6888), 507\u2013514. (162)  Roudesly, F.; Oble, J.; Poli, G. Metal-Catalyzed C\u2013H Activation\/Functionalization: The Fundamentals. J. Mol. Catal. Chem. 2017, 426, 275\u2013296. (163)  Helmetag, V.; Samel, S. A.; Thomas, M. G.; Marahiel, M. A.; Essen, L.-O. Structural Basis for the Erythro-Stereospecificity of the L-Arginine Oxygenase VioC in Viomycin Biosynthesis: High-Resolution Structures of VioC. FEBS J. 2009, 276 (13), 3669\u20133682. (164)  Ge, W.; Wolf, A.; Feng, T.; Ho, C.; Sekirnik, R.; Zayer, A.; Granatino, N.; Cockman, M. E.; Loenarz, C.; Loik, N. D.; Hardy, A. P.; Claridge, T. D. W.; Hamed, R. B.; Chowdhury, R.; Gong, L.; Robinson, C. V.; Trudgian, D. C.; Jiang, M.; Mackeen, M. M.; Mccullagh, J. S.; Gordiyenko, Y.; Thalhammer, A.; Yamamoto, A.; Yang, M.; Liu-Yi, P.; Zhang, Z.; Schmidt-Zachmann, M.; Kessler, B. M.; Ratcliffe, P. J.; Preston, G. M.; Coleman, M. L.; Schofield, C. J. Oxygenase-Catalyzed Ribosome Hydroxylation Occurs in Prokaryotes and Humans. Nat. Chem. Biol. 2012, 8 (12), 960\u2013962. (165)  Chang, C. Y.; Lyu, S. Y.; Liu, Y. C.; Hsu, N. S.; Wu, C. C.; Tang, C. F.; Lin, K. H.; Ho, J. Y.; Wu, C. J.; Tsai, M. D.; Li, T. L. Biosynthesis of Streptolidine Involved Two Unexpected Intermediates Produced by a Dihydroxylase and a Cyclase through Unusual Mechanisms. Angew. Chem. Int. Ed. 2014, 53 (7), 1943\u20131948. (166)  Phillips, R. S. Chemistry and Diversity of Pyridoxal-5\u2032-Phosphate Dependent Enzymes. Biochim. Biophys. Acta 2015, 1854 (9), 1167\u20131174. (167)  Bertoldi, M.; Cellini, B.; Montioli, R.; Borri Voltattorni, C. Insights into the Mechanism of Oxidative Deamination Catalyzed by DOPA Decarboxylase. Biochemistry 2008, 47 (27), 7187\u20137195. (168)  Sakai, K.; Yutaka, I.; Nagatomo, H.; Watanabe, H. L-Ornithine Decarboxylase from Hafnia Alvei Has a Novel L-Ornithine Oxidase Activity. J Biochem 1997, 122 (5), 961\u2013968. (169)  Battye, T. G. G.; Kontogiannis, L.; Johnson, O.; Powell, H. R.; Leslie, A. G. W. IMOSFLM\u202f: A New Graphical Interface for Diffraction-Image Processing with MOSFLM. Acta Crystallogr. D Biol. Crystallogr. 2011, 67 (4), 271\u2013281. (170)  Evans, P. R.; Murshudov, G. N. How Good Are My Data and What Is the Resolution? Acta Crystallogr. D Biol. Crystallogr. 2013, 69 (7), 1204\u20131214. (171)  McCoy, A. J.; Grosse-Kunstleve, R. W.; Adams, P. D.; Winn, M. D.; Storoni, L. C.; Read, R. J. Phaser Crystallographic Software. J. Appl. Crystallogr. 2007, 40 (4), 658\u2013674. (172)  Terwilliger, T. C.; Grosse-Kunstleve, R. W.; Afonine, P. V.; Moriarty, N. W.; Zwart, P. H.; Hung, L.-W.; Read, R. J.; Adams, P. D. Iterative Model Building, Structure Refinement and Density Modification with the PHENIX AutoBuild Wizard. Acta Crystallogr. D Biol. Crystallogr. 2008, 64 (1), 61\u201369. 190  (173)  Emsley, P.; Cowtan, K. Coot\u202f: Model-Building Tools for Molecular Graphics. Acta Crystallogr. D Biol. Crystallogr. 2004, 60 (12), 2126\u20132132. (174)  Afonine, P. V.; Mustyakimov, M.; Grosse-Kunstleve, R. W.; Moriarty, N. W.; Langan, P.; Adams, P. D. Joint X-Ray and Neutron Refinement with Phenix.Refine. Acta Crystallogr. D Biol. Crystallogr. 2010, 66 (11), 1153\u20131163. (175)  Moriarty, N. W.; Grosse-Kunstleve, R. W.; Adams, P. D. Electronic Ligand Builder and Optimization Workbench (ELBOW): A Tool for Ligand Coordinate and Restraint Generation. Acta Crystallogr. D Biol. Crystallogr. 2009, 65 (10), 1074\u20131080. (176)  Langer, G.; Cohen, S. X.; Lamzin, V. S.; Perrakis, A. Automated Macromolecular Model Building for X-Ray Crystallography Using ARP\/WARP Version 7. Nat. Protoc. 2008, 3 (7), 1171\u20131179. (177)  Li, C.; Lu, C.-D. Arginine Racemization by Coupled Catabolic and Anabolic Dehydrogenases. Proc. Natl. Acad. Sci. 2009, 106 (3), 906\u2013911. (178)  Vlessis, A. A.; Bartos, D.; Trunkey, D. Importance of Spontaneous Alpha-Ketoacid Decarboxylation in Experiments Involving Peroxide. Biochem. Biophys. Res. Commun. 1990, 170 (3), 1281\u20131287. (179)  Goutelle, S.; Maurin, M.; Rougier, F.; Barbaut, X.; Bourguignon, L.; Ducher, M.; Maire, P. The Hill Equation: A Review of Its Capabilities in Pharmacological Modelling. Fundam. Clin. Pharmacol. 2008, 22 (6), 633\u2013648. (180)  Churchich, J. E.; Moses, U. 4-Aminobutyrate Aminotransferase. The Presence of Nonequivalent Binding Sites. J. Biol. Chem. 1981, 256 (3), 1101\u20131104. (181)  Burnett, G.; Yonaha, K.; Toyama, S.; Soda, K.; Walsh, C. Studies on the Kinetics and Stoichiometry of Inactivation of Pseudomonas \u03c9-Amino Acid:Pyruvate Transaminase by Gabaculine. J. Biol. Chem. 1980, 255 (2), 428\u2013432. (182)  Romero, E.; G\u00f3mez Castellanos, J. R.; Gadda, G.; Fraaije, M. W.; Mattevi, A. Same Substrate, Many Reactions: Oxygen Activation in Flavoenzymes. Chem. Rev. 2018, 118 (4), 1742\u20131769. (183)  Thierbach, S.; Bui, N.; Zapp, J.; Chhabra, S. R.; Kappl, R.; Fetzner, S. Substrate-Assisted O2 Activation in a Cofactor-Independent Dioxygenase. Chem. Biol. 2014, 21 (2), 217\u2013225. (184)  Caulkins, B. G.; Young, R. P.; Kudla, R. A.; Yang, C.; Bittbauer, T. J.; Bastin, B.; Hilario, E.; Fan, L.; Marsella, M. J.; Dunn, M. F.; Mueller, L. J. NMR Crystallography of a Carbanionic Intermediate in Tryptophan Synthase: Chemical Structure, Tautomerization, and Reaction Specificity. J. Am. Chem. Soc. 2016, 138 (46), 15214\u201315226. (185)  Edwards, D. I. Nitroimidazole Drugs-Action and Resistance Mechanisms II. Mechanisms of Resistance. J. Antimicrob. Chemother. 1993, 31 (2), 201\u2013210. (186)  Hedges, J. B.; Kuatsjah, E.; Du, Y.-L.; Eltis, L. D.; Ryan, K. S. Snapshots of the Catalytic Cycle of an O2, Pyridoxal Phosphate-Dependent Hydroxylase. ACS Chem. Biol. 2018, 13 (4), 965\u2013974. (187)  Couture, M. M.-J.; Colbert, C. L.; Babini, E.; Rosell, F. I.; Mauk, A. G.; Bolin, J. T.; Eltis, L. D. Characterization of BphF, a Rieske-Type Ferredoxin with a Low Reduction Potential. Biochemistry 2001, 40 (1), 84\u201392. (188)  Gomez-Gil, L.; Kumar, P.; Barriault, D.; Bolin, J. T.; Sylvestre, M.; Eltis, L. D. Characterization of Biphenyl Dioxygenase of Pandoraea Pnomenusa B-356 As a Potent Polychlorinated Biphenyl-Degrading Enzyme. J. Bacteriol. 2007, 189 (15), 5705\u20135715. 191  (189)  Ma, L.; Du, L.; Chen, H.; Sun, Y.; Huang, S.; Zheng, X.; Kim, E.-S.; Li, S. Reconstitution of the In Vitro Activity of the Cyclosporine-Specific P450 Hydroxylase from Sebekia Benihana and Development of a Heterologous Whole-Cell Biotransformation System. Appl. Environ. Microbiol. 2015, 81 (18), 6268\u20136275. (190)  Kahan, J. S.; Kahan, F. M.; Goegelman, R.; Currie, S. A.; Jackson, M.; Stapley, E. O.; Miller, T. W.; Miller, A. K.; Hendlin, D.; Mochales, S.; Hernandez, S.; Woodruff, H. B.; Birnbaum, J. Thienamycin a New \u03b2-Lactam Antibiotic 1. Discovery, Taxonomy, Isolation and Physical Properties. J. Antibiot. 1979, 32 (1), 1\u201312. (191)  Laber, B.; Gomis-R\u00fcth, F. X.; Rom\u00e3o, M. J.; Huber, R. Escherichia Coli Dihydrodipicolinate Synthase. Identification of the Active Site and Crystallization. Biochem. J. 1992, 288 (2), 691\u2013695. (192)  Riedel, T. J.; Johnson, L. C.; Knight, J.; Hantgan, R. R.; Holmes, R. P.; Lowther, W. T. Structural and Biochemical Studies of Human 4-Hydroxy-2-Oxoglutarate Aldolase: Implications for Hydroxyproline Metabolism in Primary Hyperoxaluria. PLoS ONE 2011, 6 (10), e26021. (193)  Sievers, F.; Wilm, A.; Dineen, D.; Gibson, T. J.; Karplus, K.; Li, W.; Lopez, R.; McWilliam, H.; Remmert, M.; Soding, J.; Thompson, J. D.; Higgins, D. G. Fast, Scalable Generation of High-Quality Protein Multiple Sequence Alignments Using Clustal Omega. Mol. Syst. Biol. 2014, 7 (1), 539\u2013539. (194)  Robert, X.; Gouet, P. Deciphering Key Features in Protein Structures with the New ENDscript Server. Nucleic Acids Res. 2014, 42 (W1), W320\u2013W324. (195)  Jasniewski, A. J.; Que, L. Dioxygen Activation by Nonheme Diiron Enzymes: Diverse Dioxygen Adducts, High-Valent Intermediates, and Related Model Complexes. Chem. Rev. 2018, 118 (5), 2554\u20132592. (196)  He, H.-Y.; Henderson, A. C.; Du, Y.-L.; Ryan, K. S. Two-Enzyme Pathway Links L-Arginine to Nitric Oxide in N-Nitroso Biosynthesis. J. Am. Chem. Soc. 2019, 141 (9), 4026\u20134033. (197)  Ng, T. L.; Rohac, R.; Mitchell, A. J.; Boal, A. K.; Balskus, E. P. An N-Nitrosating Metalloenzyme Constructs the Pharmacophore of Streptozotocin. Nature 2019, 566 (7742), 94\u201399. (198)  McBride, M. J.; Pope, S. R.; Hu, K.; Okafor, C. D.; Balskus, E. P.; Bollinger, J. M.; Boal, A. K. Structure and Assembly of the Diiron Cofactor in the Heme-Oxygenase\u2013like Domain of the N -Nitrosourea\u2013Producing Enzyme SznF. Proc. Natl. Acad. Sci. 2021, 118 (4), e2015931118. (199)  Perry, C.; de los Santos, E. L. C.; Alkhalaf, L. M.; Challis, G. L. Rieske Non-Heme Iron-Dependent Oxygenases Catalyse Diverse Reactions in Natural Product Biosynthesis. Nat. Prod. Rep. 2018, 35 (7), 622\u2013632. (200)  Ferraro, D. J.; Gakhar, L.; Ramaswamy, S. Rieske Business: Structure\u2013Function of Rieske Non-Heme Oxygenases. Biochem. Biophys. Res. Commun. 2005, 338 (1), 175\u2013190. (201)  Senda, M.; Kishigami, S.; Kimura, S.; Fukuda, M.; Ishida, T.; Senda, T. Molecular Mechanism of the Redox-Dependent Interaction between NADH-Dependent Ferredoxin Reductase and Rieske-Type [2Fe-2S] Ferredoxin. J. Mol. Biol. 2007, 373 (2), 382\u2013400. (202)  Capyk, J. K.; D\u2019Angelo, I.; Strynadka, N. C.; Eltis, L. D. Characterization of 3-Ketosteroid 9\u03b1-Hydroxylase, a Rieske Oxygenase in the Cholesterol Degradation Pathway of Mycobacterium Tuberculosis. J. Biol. Chem. 2009, 284 (15), 9937\u20139946. 192  (203)  Shahid, I.; Malik, K. A.; Mehnaz, S. A Decade of Understanding Secondary Metabolism in Pseudomonas Spp. for Sustainable Agriculture and Pharmaceutical Applications. Environ. Sustain. 2018, 1 (1), 3\u201317. (204)  Mhlongo, M. I.; Piater, L. A.; Madala, N. E.; Labuschagne, N.; Dubery, I. A. The Chemistry of Plant\u2013Microbe Interactions in the Rhizosphere and the Potential for Metabolomics to Reveal Signaling Related to Defense Priming and Induced Systemic Resistance. Front. Plant Sci. 2018, 9, 112. (205)  Shoji, J.; Hinoo, H.; Terui, Y.; Kikuchi, J.; Hattori, T.; Ishii, K.; Matsumoto, K.; Yoshida, T. Isolation of Azomycin from Pseudomonas Fluorescens. J. Antibiot. (Tokyo) 1989, 42 (10), 1513\u20131514. (206)  Graham, D. E.; Spain, J. C.; Parry, R. J.; Hettich, R. L.; Mahan, K. M.; Klingeman, D. M.; Giannone, R. J.; Gulvick, C. A.; Fida, T. T. Nitration Enzyme Toolkit for the Biosynthesis of Energetic Materials. 2018. (207)  Dorel, R.; Echavarren, A. M. Gold(I)-Catalyzed Activation of Alkynes for the Construction of Molecular Complexity. Chem. Rev. 2015, 115 (17), 9028\u20139072. (208)  Kuklev, D. V.; Domb, A. J.; Dembitsky, V. M. Bioactive Acetylenic Metabolites. Phytomedicine 2013, 20 (13), 1145\u20131159. (209)  Thirumurugan, P.; Matosiuk, D.; Jozwiak, K. Click Chemistry for Drug Development and Diverse Chemical\u2013Biology Applications. Chem. Rev. 2013, 113 (7), 4905\u20134979. (210)  Li, X.; Lv, J.-M.; Hu, D.; Abe, I. Biosynthesis of Alkyne-Containing Natural Products. RSC Chem. Biol. 2021, 10.1039.D0CB00190B. (211)  Washtien, W.; Abeles, R. H. Mechanism of Inactivation of \u03b3-Cystathionase by the Acetylenic Substrate Analogue Propargylglycine. Biochemistry 1977, 16 (11), 2485\u20132491. (212)  Liu, H.; Naismith, J. H. An Efficient One-Step Site-Directed Deletion, Insertion, Single and Multiple-Site Plasmid Mutagenesis Protocol. BMC Biotechnol. 2008, 8 (1), 91. (213)  Nakashima, N.; Tamura, T. Isolation and Characterization of a Rolling-Circle-Type Plasmid from Rhodococcus Erythropolis and Application of the Plasmid to Multiple-Recombinant-Protein Expression. Appl. Environ. Microbiol. 2004, 70 (9), 5557\u20135568. (214)  Terwilliger, T. C.; Adams, P. D.; Read, R. J.; McCoy, A. J.; Moriarty, N. W.; Grosse-Kunstleve, R. W.; Afonine, P. V.; Zwart, P. H.; Hung, L.-W. Decision-Making in Structure Solution Using Bayesian Estimates of Map Quality: The PHENIX AutoSol Wizard. Acta Crystallogr. D Biol. Crystallogr. 2009, 65 (6), 582\u2013601. (215)  Guruprasad, K.; Reddy, B. V. B.; Pandit, M. W. Correlation between Stability of a Protein and Its Dipeptide Composition: A Novel Approach for Predicting in Vivo Stability of a Protein from Its Primary Sequence. Protein Eng. Des. Sel. 1990, 4 (2), 155\u2013161. (216)  Du, Y.-L.; Dalisay, D. S.; Andersen, R. J.; Ryan, K. S. N-Carbamoylation of 2,4-Diaminobutyrate Reroutes the Outcome in Padanamide Biosynthesis. Chem. Biol. 2013, 20 (8), 1002\u20131011. (217)  Clausen, T.; Huber, R.; Laber, B.; Pohlenz, H.-D.; Messerschmidt, A. Crystal Structure of the Pyridoxal-5\u2019-Phosphate Dependent Cystathionine \u03b2-Lyase from Escherichia Coli at 1.83 \u00c5. J. Mol. Biol. 1996, 262 (2), 202\u2013224. (218)  Clausen, T.; Huber, R.; Prade, L.; Wahl, M. C.; Messerschmidt, A. Crystal Structure of Escherichia Coli Cystathionine \u03b3-Synthase at 1.5 \u00c5 Resolution. EMBO J. 1998, 17 (23), 6827\u20136838. 193  (219)  Metzler, C. M.; Metzler, D. E. Quantitative Description of Absorption Spectra of a Pyridoxal Phosphate-Dependent Enzyme Using Lognormal Distribution Curves. Anal. Biochem. 1987, 166 (2), 313\u2013327. (220)  Mozzarelli, A.; Bettati, S. Exploring the Pyridoxal 5\u2032-Phosphate-Dependent Enzymes. Chem. Rec. 2006, 6 (5), 275\u2013287. (221)  Griswold, W. R.; Toney, M. D. Role of the Pyridine Nitrogen in Pyridoxal 5\u2032-Phosphate Catalysis: Activity of Three Classes of PLP Enzymes Reconstituted with Deazapyridoxal 5\u2032-Phosphate. J. Am. Chem. Soc. 2011, 133 (37), 14823\u201314830. (222)  Hayashi, T.; Tanaka, Y.; Sakai, N.; Okada, U.; Yao, M.; Watanabe, N.; Tamura, T.; Tanaka, I. SCO4008, a Putative TetR Transcriptional Repressor from Streptomyces Coelicolor A3(2), Regulates Transcription of Sco4007 by Multidrug Recognition. J. Mol. Biol. 2013, 425 (18), 3289\u20133300. (223)  Hayashi, T.; Tanaka, Y.; Sakai, N.; Watanabe, N.; Tamura, T.; Tanaka, I.; Yao, M. Structural and Genomic DNA Analysis of a Putative Transcription Factor SCO5550 from Streptomyces Coelicolor A3(2): Regulating the Expression of Gene Sco5551 as a Transcriptional Activator with a Novel Dimer Shape. Biochem. Biophys. Res. Commun. 2013, 435 (1), 28\u201333. (224)  Hashimoto, K.; Panchenko, A. R. Mechanisms of Protein Oligomerization, the Critical Role of Insertions and Deletions in Maintaining Different Oligomeric States. Proc. Natl. Acad. Sci. 2010, 107 (47), 20352\u201320357. https:\/\/doi.org\/10.1073\/pnas.1012999107. (225)  Singh, M. K.; Shivakumaraswamy, S.; Gummadi, S. N.; Manoj, N. Role of an N-Terminal Extension in Stability and Catalytic Activity of a Hyperthermostable \u03b1\/\u03b2 Hydrolase Fold Esterase. Protein Eng. Des. Sel. 2017, 30 (8), 559\u2013570. (226)  Fitch, C. A.; Platzer, G.; Okon, M.; Garcia-Moreno E., B.; McIntosh, L. P. Arginine: Its pKa Value Revisited: PKa Value of Arginine. Protein Sci. 2015, 24 (5), 752\u2013761. (227)  Kazlauskas, R. J.; Bornscheuer, U. T. Finding Better Protein Engineering Strategies. Nat. Chem. Biol. 2009, 5 (8), 526\u2013529. (228)  Jones, B. J.; Lim, H. Y.; Huang, J.; Kazlauskas, R. J. Comparison of Five Protein Engineering Strategies for Stabilizing an \u03b1\/\u03b2-Hydrolase. Biochemistry 2017, 56 (50), 6521\u20136532. (229)  Goldenzweig, A.; Goldsmith, M.; Hill, S. E.; Gertman, O.; Laurino, P.; Ashani, Y.; Dym, O.; Unger, T.; Albeck, S.; Prilusky, J.; Lieberman, R. L.; Aharoni, A.; Silman, I.; Sussman, J. L.; Tawfik, D. S.; Fleishman, S. J. Automated Structure- and Sequence-Based Design of Proteins for High Bacterial Expression and Stability. Mol. Cell 2016, 63 (2), 337\u2013346. (230)  Weinstein, J. J.; Goldenzweig, A.; Hoch, S.; Fleishman, S. J. PROSS 2: A New Server for the Design of Stable and Highly Expressed Protein Variants. Bioinformatics 2021, 1\u20133. (231)  Busch; Hagedoorn; Hanefeld. Rhodococcus as A Versatile Biocatalyst in Organic Synthesis. Int. J. Mol. Sci. 2019, 20 (19), 4787. (232)  Nakashima, N.; Tamura, T. A Novel System for Expressing Recombinant Proteins over a Wide Temperature Range from 4 to 35\u00b0C: A Novel Recombinant Protein Expression System. Biotechnol. Bioeng. 2004, 86 (2), 136\u2013148. (233)  Saito, Y.; Kitagawa, W.; Kumagai, T.; Tajima, N.; Nishimiya, Y.; Tamano, K.; Yasutake, Y.; Tamura, T.; Kameda, T. Developing a Codon Optimization Method for Improved Expression of Recombinant Proteins in Actinobacteria. Sci. Rep. 2019, 9 (1), 8338. 194  (234)  DeLorenzo, D. M.; Rottinghaus, A. G.; Henson, W. R.; Moon, T. S. Molecular Toolkit for Gene Expression Control and Genome Modification in Rhodococcus Opacus PD630. ACS Synth. Biol. 2018, 7 (2), 727\u2013738. (235)  Hatanaka, S.-I.; Niimura, Y.; Takishima, K.; Sugiyama, J. (2R)-2-Amino-6-Hydroxy-4-Hexynoic Acid, and Related Amino Acids in the Fruiting Bodies of Amanita Miculifera. Phytochemistry 1998, 49 (2), 573\u2013578. (236)  Sato, D.; Shiba, T.; Mizuno, S.; Kawamura, A.; Hanada, S.; Yamada, T.; Shinozaki, M.; Yanagitani, M.; Tamura, T.; Inagaki, K.; Harada, S. The Hyperthermophilic Cystathionine \u03b3-Synthase from the Aerobic Crenarchaeon Sulfolobus Tokodaii\u202f: Expression, Purification, Crystallization and Structural Insights. Acta Crystallogr. Sect. F Struct. Biol. Commun. 2017, 73 (3), 152\u2013158. (237)  Inoue, H.; Inagaki, K.; Adachi, N.; Tamura, T.; Esaki, N.; Soda, K.; Tanaka, H. Role of Tyrosine 114 of L -Methionine \u03b3-Lyase from Pseudomonas Putida. Biosci. Biotechnol. Biochem. 2000, 64 (11), 2336\u20132343. (238)  Astegno, A.; Allegrini, A.; Piccoli, S.; Giorgetti, A.; Dominici, P. Role of Active-Site Residues Tyr55 and Tyr114 in Catalysis and Substrate Specificity of Corynebacterium Diphtheriae C-S Lyase: C-S Lyase: Role of Active-Site Residues. Proteins Struct. Funct. Bioinforma. 2015, 83 (1), 78\u201390. (239)  Anufrieva, N. V.; Faleev, N. G.; Morozova, E. A.; Bazhulina, N. P.; Revtovich, S. V.; Timofeev, V. P.; Tkachev, Y. V.; Nikulin, A. D.; Demidkina, T. V. The Role of Active Site Tyrosine 58 in Citrobacter Freundii Methionine \u03b3-Lyase. Biochim. Biophys. Acta BBA - Proteins Proteomics 2015, 1854 (9), 1220\u20131228. (240)  Brzovic, P.; Litzenberger Holbrook, E.; Greene, R. C.; Dunn, M. F. Reaction Mechanism of Escherichia Coli Cystathionine \u03b3-Synthase: Direct Evidence for a Pyridoxamine Derivative of Vinylglyoxylate as a Key Intermediate in Pyridoxal Phosphate Dependent \u03b3-Elimination and \u03b3-Replacement Reactions? Biochemistry 1990, 29 (2), 442\u2013451. (241)  Yee, D. A.; Kakule, T. B.; Cheng, W.; Chen, M.; Chong, C. T. Y.; Hai, Y.; Hang, L. F.; Hung, Y.-S.; Liu, N.; Ohashi, M.; Okorafor, I. C.; Song, Y.; Tang, M.; Zhang, Z.; Tang, Y. Genome Mining of Alkaloidal Terpenoids from a Hybrid Terpene and Nonribosomal Peptide Biosynthetic Pathway. J. Am. Chem. Soc. 2020, 142 (2), 710\u2013714. (242)  Hai, Y.; Chen, M.; Huang, A.; Tang, Y. Biosynthesis of Mycotoxin Fusaric Acid and Application of a PLP-Dependent Enzyme for Chemoenzymatic Synthesis of Substituted L -Pipecolic Acids. J. Am. Chem. Soc. 2020, 142 (46), 19668\u201319677. (243)  Qu, Y.; Spain, J. C. Catabolic Pathway for 2-Nitroimidazole Involves a Novel Nitrohydrolase That Also Confers Drug Resistance: Biodegradation of 2-Nitroimidazole. Environ. Microbiol. 2011, 13 (4), 1010\u20131017. (244)  Windle, C. L.; M\u00fcller, M.; Nelson, A.; Berry, A. Engineering Aldolases as Biocatalysts. Curr. Opin. Chem. Biol. 2014, 19, 25\u201333. (245)  Yang, J.; Zhang, T.; Tian, C.; Zhu, Y.; Zeng, Y.; Men, Y.; Chen, P.; Sun, Y.; Ma, Y. Multi-Enzyme Systems and Recombinant Cells for Synthesis of Valuable Saccharides: Advances and Perspectives. Biotechnol. Adv. 2019, 37 (7), 107406. (246)  Wu, J.; Li, L.; Deng, Z.; Zabriskie, T. M.; He, X. Analysis of the Mildiomycin Biosynthesis Gene Cluster in Streptoverticillum Remofaciens ZJU5119 and Characterization of MilC, a Hydroxymethyl Cytosyl-Glucuronic Acid Synthase. ChemBioChem 2012, 13 (11), 1613\u20131621. 195  (247)  Manley, O. M.; Fan, R.; Guo, Y.; Makris, T. M. Oxidative Decarboxylase UndA Utilizes a Dinuclear Iron Cofactor. J. Am. Chem. Soc. 2019, 141 (22), 8684\u20138688. (248)  Zhang, B.; Rajakovich, L. J.; Van Cura, D.; Blaesi, E. J.; Mitchell, A. J.; Tysoe, C. R.; Zhu, X.; Streit, B. R.; Rui, Z.; Zhang, W.; Boal, A. K.; Krebs, C.; Bollinger, J. M. Substrate-Triggered Formation of a Peroxo-Fe2 (III\/III) Intermediate during Fatty Acid Decarboxylation by UndA. J. Am. Chem. Soc. 2019, 141 (37), 14510\u201314514. (249)  McBride, M. J.; Sil, D.; Ng, T. L.; Crooke, A. M.; Kenney, G. E.; Tysoe, C. R.; Zhang, B.; Balskus, E. P.; Boal, A. K.; Krebs, C.; Bollinger, J. M. A Peroxodiiron(III\/III) Intermediate Mediating Both N -Hydroxylation Steps in Biosynthesis of the N -Nitrosourea Pharmacophore of Streptozotocin by the Multi-Domain Metalloenzyme SznF. J. Am. Chem. Soc. 2020, 142 (27), 11818\u201311828. (250)  Hoffmann-R\u00f6der, A.; Krause, N. Synthesis and Properties of Allenic Natural Products and Pharmaceuticals. Angew. Chem. Int. Ed. 2004, 43 (10), 1196\u20131216. (251)  Okonechnikov, K.; Golosova, O.; Fursov, M. Unipro UGENE: A Unified Bioinformatics Toolkit. Bioinformatics 2012, 28 (8), 1166\u20131167. (252)  Letunic, I.; Bork, P. Interactive Tree of Life (ITOL) v3: An Online Tool for the Display and Annotation of Phylogenetic and Other Trees. Nucleic Acids Res. 2016, 44 (W1), W242\u2013W245.    196  Appendices Appendix A  Supporting data for Chapter 2: Identification and Characterization of the O2- and PLP-dependent arginine hydroxylase RohP Table A.1 RohP X-ray data collection statistics.   Holo-RohP [6C3B] Quinonoid I (Q1) [6C3C] Quinonoid II (Q2) [6C3D] Int. Aldimine \u2013 Product [6C3A] Data Collection     Wavelength (\u00c5)  0.97949 0.97949 0.97949 0.97949 Space group C2 C2 C2 C2 Dimensions a = 79.38 b = 112.41 c = 102.16 \u03b2 = 95.77\u00b0 a = 79.15 b = 112.61 c = 101.61 \u03b2 = 95.98 a = 79.01 b = 112.56 c = 102.13 \u03b2 = 96.26 a = 78.80  b = 112.55 c = 101.97 \u03b2 = 96.39 Resolution (\u00c5)a 56.69-1.51  (1.54 -1.51) 56.62-1.50  (1.53-1.50) 64.41-1.55  (1.58-1.55) 52.17-1.53  (1.56-1.53) Rmerge a,b 0.069 (0.691) 0.086 (0.708) 0.067 (0.539) 0.067 (0.565) Rpim a,c 0.061 (0.591) 0.057 (0.497) 0.062 (0.473) 0.059 (0.479) CC1\/2 a,d 0.997 (0.660) 0.998 (0.637) 0.997 (0.702) 0.997 (0.700) Completeness (%)a 100.0 (99.9) 100.0 (100.0) 99.6 (98.7) 100.0 (100.0) Unique reflections a 139,369 (6,931) 141,151 (7,003) 127,676 (6,266) 132,732 (6,503) Multiplicity a 4.2 (4.0) 6.2 (5.7) 3.8 (3.6) 4.2 (4.1) I\/(\u03c3I) a 9.7 (1.9) 10.8 (2.2) 9.3 (2.1) 9.6 (2.1)  aData from the highest-resolution shell are indicated in parentheses. bRmerge=\u2211 \u2211 |\ud835\udc3c\ud835\udc56(\u210e\ud835\udc58\ud835\udc59)\u2212\ud835\udc3c(\u210e\ud835\udc58\ud835\udc59)|\ud835\udc5b?\u0307?=1\u210e\ud835\udc58\ud835\udc56\u2211 \u2211 \ud835\udc3c\ud835\udc56(\u210e\ud835\udc58\ud835\udc59)\ud835\udc5b?\u0307?=1\u210e\ud835\udc58\ud835\udc56 cRpim=\u2211 (1\ud835\udc5b\u22121)1\/2\u2211 |\ud835\udc3c\ud835\udc56(\u210e\ud835\udc58\ud835\udc59)\u2212\ud835\udc3c(\u210e\ud835\udc58\ud835\udc59)|\ud835\udc5b?\u0307?=1\u210e\ud835\udc58\ud835\udc56\u2211 \u2211 \ud835\udc3c\ud835\udc56(\u210e\ud835\udc58\ud835\udc59)\ud835\udc5b?\u0307?=1\u210e\ud835\udc58\ud835\udc56 dCC1\/2= \ud835\udc50\ud835\udc5c\ud835\udc63(\ud835\udc4b,\ud835\udc4c)\ud835\udf0e\ud835\udc65\ud835\udf0e\ud835\udc66= \ud835\udc38[\ud835\udc4b\u2212\ud835\udf07\ud835\udc65][\ud835\udc4c\u2212\ud835\udf07\ud835\udc66]\ud835\udf0e\ud835\udc65\ud835\udf0e\ud835\udc66    197  Table A.2 RohP X-ray data refinement statistics.   Holo-RohP [6C3B] Quinonoid I (Q1) [6C3C] Quinonoid II (Q2) [6C3D] Int. Aldimine \u2013 Product [6C3A] Refinement      Rworka 0.1522 (0.2332) 0.1624 (0.2409) 0.1599 (0.2326) 0.1528 (0.2251) Rfreea 0.1725 (0.2476) 0.1866 (0.2601) 0.1856 (0.2454) 0.1729 (0.2505) No. non-hydrogen atoms 6859 7079 7079 7121 Protein 5951 5942 6123 6140 Solvent 828 1025 847 904 Ligands 80 112 109 77 RMSD Bonds (\u00c5) 0.006 0.006 0.006 0.006 RMSD Angles (\u00b0) 0.87 0.83 0.82 0.92 Ramachandran favored (%) 98.75 98.62 98.27 98.66 Ramachandran allowed (%) 1.25 1.38 1.73 1.34 Ramachandran outliers (%) 0 0 0 0 Average B factor (\u00c52) 21.82 18.33 24.84 23.51    Protein 20.05 16.39 23.11 21.60    Solvent 34.22 29.33 36.41 35.48    Ligands 33.36 20.39 31.83 35.48 No. TLS groups 20 - 17 17      aData from the highest-resolution shell is indicated in parentheses.   198  Table A.3 Conserved residue numbering. Key residues discussed in the main text, and the corresponding residues in MppP and Ind4. Also refer to Figure 2.10 for additional information.  RohP MppP Ind4 Leu16 Leu11 Leu7 Thr17 Thr12 Thr8 Glu20 Glu15 Glu11 His34 His28 His24 Tyr92 Tyr88 Tyr83 Ser95 Ser91 Ser86 Phe119 Phe115 Phe110 Asn121 Asn117 Asn112 Asn167 Asn160 Asn156 Asp198 Asp188 Asp184 Lys235 Lys221 Lys217 Asp241 Asp227 Asp223 Lys243 Lys229 Asp225 Leu266 Leu252 Leu248 Arg367 Arg352 Arg350    199  Table A.4 Comparison of the bond lengths generated from ARP\/wARP refinement to the bonds lengths of the Q2 ligand. Arginine Bond Q2 eLBOW bond lengths (\u00c5)  ARP\/wARP initial bond lengths (\u00c5) ARP\/wARP refined bond lengths (\u00c5) NPLP-C\u03b1 1.25 1.24 1.32 C\u03b1-C\u03b2 1.53 1.54 1.55 C\u03b2-C\u03b3 1.32 1.34 1.33 C\u03b3-C\u03b4 1.53 1.54 1.51 C\u03b4-N\u03b5 1.45 1.43 1.43 The bond lengths highlighted in gray are the bond lengths present in the ligand in the final refined structure, which closely match the values expected for an oxidized Q2-like intermediate. Bond lengths are taken from intermediates in Figure A.9.   200   Figure A.1 1H NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid. 600 MHz, D2O, 298 K. \u03b4 4.08 (qd, J = 6.9, 3.3 Hz, 1H), 3.29 (dt, J = 14.5, 3.5 Hz, 1H), 3.17 \u2013 3.08 (m, 3H), 2.34 (dd, J = 6.9, 2.1 Hz, 2H), 2.18 (t, J = 7.3 Hz, 2H), 1.76 (pd, J = 7.2, 2.2 Hz, 2H). 201   Figure A.2 13C NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid. 150 MHz, D2O, 298 K. \u03b4 182.46, 179.41, 157.47, 156.87, 67.77, 46.63, 41.86, 40.80, 34.29, 24.96.  202   Figure A.3 1H-1H COSY NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid. 600 MHz, D2O, 298K. 203   Figure A.4 1H-13C HSQC NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid. 600 MHz, D2O, 298K.  204   Figure A.5 1H-13C HMBC NMR spectrum of 3-hydroxy-4-guanidinobutyric acid and 4-guanidinobutyric acid. 600 MHz, D2O, 298K.  205     Figure A.6 NMR assignment summary. Top) Assignment of 1H and 13C NMR data for the mixture of 4-guanidinobutyric acid (110) and 3-hydroxy-4-guanidinobutyric acid (111). Bottom) Assignment of 2D NMR data, with correlations shown as indicated.   206   Figure A.7 Modeling of external aldimine (top) and quinonoid intermediates (bottom). Left) Structure of intermediate modelled. Middle) Intermediate modelled into Fo-Fc omit density present in the active site. Model structure and restraints were produced using Phenix eLBOW. Right) Density after refinement with Phenix refine. The 2Fo-Fc maps are contoured at 1.0 \u03c3 as gray mesh, and the Fo-Fc maps are contoured at 3.0 \u03c3, with positive and negative density indicated as green and red mesh, respectively.  207   Figure A.8 Modeling of possible oxidized Q2 intermediates. Possible oxidized intermediates fit to electron density in active site of Chain B of the second RohP structure. Left) Structure of intermediate modelled. Middle) Intermediate modelled into Fo-Fc omit density present in the active site. Model structure and restraints were produced using Phenix eLBOW. Right) Density after refinement with Phenix refine. The 2Fo-Fc maps are contoured at 1.0 \u03c3 as gray mesh, and the Fo-Fc maps are contoured at 3.0 \u03c3, with positive and negative density indicated as green and red mesh, respectively.   208   Figure A.9 ARP\/wARP refinement of oxidized Q2 intermediate Q2 intermediate was fit to the available electron density using ARP\/wARP. Panel 1 displays the placement of the ligand resulting from the initial ARP\/wARP refinement. Panel 2 displays is the position of the ligand prior to refinement after manual adjustment in COOT. Panel 3 displays the position of the ligand after refinement with Phenix refine. The 2Fo-Fc maps are contoured at 1.0 \u03c3 as gray mesh, and the Fo-Fc maps are contoured at 3.0 \u03c3, with positive and negative density indicated as green and red mesh, respectively.   209   Figure A.10 PLP-arginine adducts modelled into available density in the active site of the red crystal. Density maps and ligand fitting generated from a 2.0 \u00c5 data set. Left) Structure of intermediate modelled. Middle) Intermediate modelled into Fo-Fc omit density present in the active site. Model structure and restraints were produced using Phenix eLBOW. Right) Density after refinement with Phenix refine. The 2Fo-Fc maps are contoured at 1.0 \u03c3 as gray mesh, and the Fo-Fc maps are contoured at 3.0 \u03c3, with positive and negative density indicated as green and red mesh, respectively.           210  Appendix B  Supporting data for Chapter 3: In vitro biochemical characterization of the azomycin biosynthetic gene cluster  Table B.1 Strains and vectors used in Chapter 3.  Strains Description Source E. coli DH5\u03b1 General cloning Laboratory stock E. coli BL21(DE3) Protein expression Laboratory stock S. cattleya DSM 46488 rohP-containing strain DSMZ P. syringae pv. tomato str. DC 3000 Contains a putative azomycin biosynthetic gene cluster Cara Haney (UBC) P. brassicacearum DF41 Contains a putative azomycin biosynthetic gene cluster Cara Haney (UBC)    Vectors Description Source pET28a Cloning and protein expression Laboratory stock pET28a:RohP rohP encoding region cloned in pET28a RohP study186 pET28a:RohQ rohQ encoding region cloned in pET28a This study pET28a:RohR rohR encoding region cloned in pET28a This study pET28a:SCyRohS SCyrohS encoding region cloned in pET28a This study pET28a:RohT rohT encoding region cloned in pET28a This study pUC57:KAzRohS Synthesized RohS homolog from Kitasatospora azatica  Bio Basic Inc. pET28a:KAzRohS K. azatica homolog cloned in pET28a This study pPAISC-1 Plasmid for Fe-S cluster assembly (TetR) Lindsay Eltis (UBC)202    211  Table B.2 PCR Primers used for studies in Chapter 3.  Primer Sequence (5' to 3') Description RohP NdeI-F AGCAGCCATATGCACCCGCAAGCGACC Cloning of pET28a:RohP RohP XhoI-R AGTAGTCTCGAGTCAGCGGCCATGGCGGTC RohQ NdeI-F AGCAGCCATATGGCCGCTGACCGCGAAG Cloning of pET28a:RohQ RohQ XhoI-R AGCATCCTCGAGTCACAGCTGGGCCGC RohR NdeI-F AGCAGCCATATGTTCCACGGAGTGACCGTCG Cloning of pET28a:RohR RohR XhoI-R AGCAGCCTCGAGTCATACGGGCGTTCCTTC SCyRohS NdeI-F AGCAGCCATATGACCACGGCGACCACCC Cloning of pET28a:SCyRohS SCyRohS XhoI-R AGCAGCCTCGAGTCACCGCGTCACCTC RohT NdeI-F AGCAGCCATATGAACGCCGAACTGCGGGAG Cloning of pET28a:RohT RohT XhoI-R AGCAGCCTCGAGTCAACGCGCTTCTCCTTC  Table B.3 Protein purification buffer for azomycin biosynthetic enzymes.  Protein Binding and storage buffer Stability (4 \u00b0C) RohP 20 mM HEPES, 50 mM NaCl, pH 7.5 Stable RohQ 20 mM Tris, 50 mM NaCl, pH 8.0 Stable (at low concentrations) RohR 20 mM CHES, 150 mM NaCl, pH 9.3 Moderately stable SCyRohS 20 mM phosphate, 100 mM NaCl, pH 8.0 Unstable KAzRohS 20 mM HEPES, 50 mM NaCl, pH 7.5 Stable RohT 20 mM Tris, 50 mM NaCl, pH 8.0 Stable Observed stabilities are based on the rate of precipitation during storage at 4 \u00b0C.   212  Table B.4 KAzRohS ICP-MS results.   Ni2+ column purified and FPLC purified protein Ni2+ column purified and dialyzed protein Metal Observed concentration (\u03bcM) Metal (mol): KAzRohS (mol) Observed concentration (\u03bcM) Metal (mol): KAzRohS (mol) Co <0.034 <3.2x10-4 <0.034 <3.2x10-4 Cu 13.7 0.13 0.393 3.7x10-3 Fe 6.4 0.061 17 0.16 Mg <4.1 <0.04 <4.1 <0.04 Mn 0.34 3.2x10-3 0.056 5.4x10-4 Mo <0.010 <9.5x10-5 <0.010 <9.5x10-5 Ni 78.5 0.74 88.3 0.84 Zn 24.0 0.23 7.22 0.069 Gray text denotes values below instrument detection limits. Table B.5 RohT ICP-MS results.   Ni2+ column purified protein, dialyzed in Tris buffer Metal Observed concentration (\u03bcM) Metal (mol): RohT (mol) Co <0.034 <3.2x10-4 Cu 3.49 0.035 Fe 73.2 0.73 Mg 4.1 0.041 Mn 0.082 8.2x10-3 Mo 0.018 1.9x10-4 Ni 47.4 0.47 Zn 36.2 0.36 Gray text denotes values below instrument detection limits.   213  Table B.6 Azomycin biosynthetic gene cluster in Streptomyces cattleya.  Gene Length (aa) Predicted function NCBI Accession number rohP 368 PLP-dependent aminotransferase (verified arginine hydroxylase) AEW92768.1 rohQ 170 Hypothetical protein AEW92767.1 rohR 268 Dihydrodipicolinate synthase AEW92766.1 rohS 290 Non-heme iron oxidase AEW92765.1 rohT 96 Rieske [2Fe-2S]-containing protein AEW92764.1   214   Figure B.1 Phylogenetic analysis of azomycin biosynthetic gene clusters in soil-dwelling bacteria. Branches containing conserved gene clusters are highlighted by color. Multiple occurrences of the gene cluster in the same strain are indicated by orange and black circles. The bacterial strains Paraburkholderia aspaltahi, Actinoalloteichus hymeniacidonis, and Amycolatopsis nigrescens all contain rohP homologs in putative azomycin gene clusters. In these three cases the accessory genes surrounding the cluster have been rearranged. As a result, they do not exhibit any conserved genomic orientation and have been left uncolored. Streptomyces ossamyceticus contains the 215  azomycin gene cluster but the accessory genes surrounding this gene cluster do not match any other identified gene cluster. To generate this figure RohP was used as the template for a BLASTp search of the NCBI non-redundant protein database. The genomic neighborhoods of the closest BLAST hits were manually inspected to identify those homologs contained in a putative azomycin biosynthetic gene cluster. These results were used to construct a sequence database of the azomycin-RohP homologs. These sequences were aligned using MUSCLE, and a phylogenetic tree was generated using the default parameters of the PhyML program as implemented in UGENE 1.30.0.251 The phylogenetic tree was visualized using the online service Interactive Tree of Life (iTOL).252 NCBI accession numbers of the sequences used in the phylogenetic analysis are indicated. Due to the high redundancy of genomic data for available Pseudomonas strains, a single representative sequence from the most commonly deposited strains was included. Accessory enzymes are colored in the same coloring scheme as in Figure 3.2.   216  WP_063774763.1 (KAzRohS) CATATGCGTCGTCTGTTCCTGCTGAACCGTGAACTGCCGTCTGTTGCTGGCCTGACCGCGATCCTGGAACTGGAACGTCAGTGGATTGTGCCGAAAATCTGCCAGCTGGCGCAGGCGATCGGTCCGGAAGTTACTAGCCGTGCTCAGTGGCTGGCTGCTCTGGACGGGCTGCTGGCAGCGGAACGTCCGGGTACCCCGGCTGGTGAGTACCTGGCTGAACAGGCGGACCGTGAACAGTTCGCTGCGGTTGTACGTGAATTCGCGCTGGACGGTCTGACCGAAGCGCAGAACTTCTTCCCGGCAGTTCCGCGTCTGCCGATCCGTGCCCAGATGGCGGTTATGCGTGTGCTGATCGATGAATTCGGCTGCGGTAACCTGCGTCAGACCCACTCTCAGCTGTACCTGGATCTGCTGGCTGAACTGGGTCTGCCGCAGGAACTGGAATCTTTCCTGGATACCACCTCTGAAGAAACCTTCGGTTTCCTGAACGTGTTCTACTGGCTGACTCAGCGTGCGCCGGCGGTTGAATACTTCCTGGGCGCGCTGGCTTACCTGGAAGCTTCTATCCCGGATGCGTTCACCGTTCAGGCGCGTGCGTGTGAACGTCTGGGTCTGGCGCACGGTCGTTACTACACCGAACACCTGCACATCGACACCTTCCACCGCGAAGAAATGCAGGTTGCTATCCGTGAACTGGAAGCGGCGCGTGGTCTGGACGCGGGTAAACTGTGGGTGGGTGCGCTGCTGCTGTCTGAACTGCTGGGTACCGCGTTCGAAGCTGCGGTTGAACGTGCTAGACGTGTTAAGGTTTAACTCGAG  (restriction enzyme sites underlined) Figure B.2 DNA sequence of codon optimized Kitasatospora azatica RohS homolog.  217   Figure B.3 UV-Vis absorbance spectra of RohT. RohT (200 \u03bcM) as purified \u2013 blue line. RohT (200 \u03bcM) after dialysis under anaerobic conditions with Fe2+ and S2- \u2013 grey line. There is a slight difference in absorbance between 350-400 nm between this spectrum and that of RohT as it is purified, but overall, the two spectra look highly similar.   218   Figure B.4 UV-Vis of anaerobically reduced RohT. Spectrum of RohT as purified (oxidized), and after incubation with 2 mM sodium dithionite and desalting under anaerobic conditions (reduced).   219   Figure B.5 Reaction of RohT with 2-aminoimidazole. RohT was incubated with 2-aminoimidazole (12) under the same conditions used in the KAzRohS in vitro studies. The reaction between 12 and KAzRohS was used as a positive control. No production of 6 was observed under the in vitro conditions tested. Traces depict the extracted ion chromatograms for m\/z 114 for azomycin (6).    220    Figure B.6 Analysis of the KAzRohS + RohT coupled reaction. a) The KAzRohS assay containing 2-aminoimidazole (12), FeSO4, PMS, NADH was repeated five times with 30 \u03bcM of RohT (red traces) and without RohT (blue) traces. The reaction mixture was analyzed by LC-MS and the results for SIM m\/z 114 for azomycin (6) for each sample are depicted. b) An overlay of the signal from one representative replicate, demonstrating that the same amount of 6 is produced in each assay, regardless of the presence of RohT. c) Average integrated peak area with error bars representing the standard deviation for each set of five samples presented in a).   221   Figure B.7 KAzRohS reactions employing reduced RohT. a) KAzRohS incubated with the indicated amount of 2-aminoimidazole (12) and 30 \u03bcM RohT that had been reduced under anaerobic conditions with 2 mM sodium dithionite. All reactions contained 2 mM FeSO4\u00b77H2O. No azomycin (6) could be detected with RohT as the sole reductant the reaction mixture. b) The reaction between 12, KAzRohS, and RohT under reducing conditions. Production of 6 could be observed in the presence of NADH and PMS, similar to KAzRohS on its own (Figure 3.15). However, once these components were removed, reduced RohT was insufficient to maintain the activity of KAzRohS. c) KAzRohS incubated with the indicated amount of 12 and 90 \u03bcM RohT that had been dialyzed with Fe2+ and S2- under anaerobic and reducing conditions. All reactions contained 2 mM FeSO4\u00b77H2O. All traces in each panel depict the extracted ion chromatograms at m\/z 114 for 6.  222   Figure B.8 The KAzRohS + RohT reaction with non-cognate ferredoxin reductases. The three ferredoxin reductases tested were spinach ferredoxin (spFDR), a ferredoxin-NADP+ reductase isolated from the cyanobacterium Synechococcus (seFDR),189 and the ferredoxin-NAD+ reductase BphG from Pandoraea pnomenusa B-356.188 The ferredoxin reductase and appropriate reducing agent utilized in each assay are indicated above each trace. In addition to the indicated components, all reactions contained 500 \u03bcM 2-aminoimidazole (12), 2 mM FeSO4\u00b77H2O, 3 mM NAD(P)H, 5 \u03bcM KAzRohS, and 10 \u03bcM RohT. All traces depict the extracted ion chromatograms for m\/z 114 for azomycin (6).   223   Figure B.9 The KAzRohS + RohT reaction with a heterologous ferredoxin-ferredoxin reductase pair. Biphenyl dioxygenase pair BphF187 and BphG188 were both supplied by the Eltis lab. All reactions contained 500 \u03bcM 2-aminoimidazole (12), 2 mM FeSO4\u00b77H2O, and 3 mM NADH as the reductant. The well-characterized BphF-BphG system was unable to reconstitute the activity of the di-iron oxygenase KAzRohS in vitro. All traces depict the extracted ion chromatograms at m\/z 114 for azomycin (6).   224  Appendix C  Supporting data for Chapter 4: Characterization and Crystallization of the unusual PLP-dependent desaturase BesB Table C.1 Strains and vectors used in Chapter 4.  Strains Description Source E. coli DH5\u03b1 General cloning Laboratory stock E. coli BL21(DE3) Protein expression Laboratory stock S. achromogenes subsp. achromogenes B-2120 (DSMZ 40028) sacBesB-containing strain DSMZ Rhodococcus sp. RHA-1 Protein expression Laboratory stock    Vectors Description Source pET28a Cloning and protein expression Laboratory stock pET22b Cloning and protein expression Laboratory stock pTIPQC1 Protein expression Need source    225  Table C.2 Primers used for PCR, cloning, and sequencing.   Primer Sequence Usage sacBesB NdeI-F AGTAGTCATATGAGCCAAGCGGTATCCGGC Cloning, pET28a, pET22b, and pTIPQC1 sacBesB SacI-R AGTAGTGAGCTCTCAGTCGGGGCGCAGGG Cloning, pET28a sacBesB \u2013 HindIII-R AGTAGTAAGCTTGTCGGGGCGCAGGG Cloning, pET22b, pTIPQC1 sacBesB- 61-518 AGTAGTCATATGCTGTCCGGGGTGCGGCTGCC Variant lacking first 60 N-terminal residues sacBesB- 121-518 AGTAGTCATATGCGCTTCCGTCCGCACCCCTACGTG Variant lacking first 120 N-terminal residues sacBesB-F62Y-F GCTACCCGCGCTACCGTCCGCACCC  F62Y mutant sacBesB-F62Y-R ACAGGTGCATGGTGTCGTAGAAGATCCAGCCGAGC F62Y mutant sacBesB-H142A-F CACGTCATGGCCACCGGCGGCCACCTGTC H142A mutant sacBesB-H142A-R CCGCCGGTGGCCATGACGTGGGCCC  H142A mutant sacBesB-F231Y-F CCAGCTCGGCTGGATCTACTTCGACACCATGCACC F231Y mutant sacBesB-F231Y-R GGTGCATGGTGTCGAAGTAGATCCAGCCGAGCTGG F231Y mutant sacBesB-F232Y-F TGGATCTTCTACGACACCATGCACCTGTTCG  F232Y mutant sacBesB-F232Y-R CATGGTGTCGTAGAAGATCCAGCCGAGCTG F232Y mutant 226  sacBesB-D233N-F GATCTTCTTCAACACCATGCACCTGTTCGAGA D233N mutant sacBesB-D233N-R GTGCATGGTGTTGAAGAAGATCCAGCCGAG  D233N mutant sacBesB-D233A-F GGATCTTCTTCGCCACCATGCACCTGTTCGAGAA D233A mutant sacBesB-D233A-R GGTGCATGGTGGCGAAGAAGATCCAGCCGAG D233A mutant sacBesB-K329A-F CTGCGAGAGCCTGACCGCGTACGCCACCGGATCGG K329A mutant sacBesB-K329A-R CCGATCCGGTGGCGTACGCGGTCAGGCTCTCGCAG K329A mutant pTIPQC1-F CGCTCATTCCAACCTCCGTGTGTTT Sequencing pTIPQC1-R TTGCACCTCACGTCACGTGAGGA Sequencing  Restriction enzyme sites indicated. Modified codons colored red.   227  Table C.3 BesB X-ray data collection statistics.   F231Y BesB SeMet collection holo-BesB (orange) holo-F231Y BesB (yellow) Data Collection    Wavelength (\u00c5)  0.97933 0.97946 0.97946 Space group P3221 P3221 P3221 Dimensions a, b = 59.07 \u00c5 c = 256.94 \u00c5 \u03b1, \u03b2 = 90\u00b0 \u03b3 = 120\u00b0 a, b = 58.64 \u00c5 c = 249.83 \u00c5 \u03b1, \u03b2 = 90\u00b0 \u03b3 = 120\u00b0 a, b = 58.52 \u00c5 c = 250.20 \u00c5 \u03b1, \u03b2 = 90\u00b0 \u03b3 = 120\u00b0 Resolution (\u00c5)a 51.39-3.00 (3.18-3.00) 50.78-1.29 (1.31-1.29) 46.97 \u2013 1.49 (1.52-1.49) Rmergea 0.087 (0.223) 0.107 (2.801) 0.089 (1.845) Rpima 0.030 (0.075) 0.037 (0.993) 0.046 (0.957) CC1\/2 a 0.991 (0.994) 0.999 (0.375) 0.999 (0.350) Completeness (%)a 100.0 (100.0) 99.9 (100.0) 99.8 (100.0) Anomalous completenessa 100.0 (100.0)   Unique reflectionsa 11199 (1758) 126887 (6174) 82643 (4034) Multiplicitya 16.8 (18.0) 17.0 (17.1) 8.7 (8.9) Anomalous multiplicitya 9.4 (9.7)   I\/(\u03c3I)a 26.9 (12.8) 12.1 (1.1) 11.8 (1.2) aData from the highest-resolution shell are indicated in parentheses.  228  Table C.4 BesB X-ray structure refinement statistics.   aData from the highest-resolution shell are indicated in parentheses.    Holo-BesB (orange) Holo-F231Y BesB (yellow) Refinement    Rworka 0.1735 (0.3218) 0.1716 (0.3094) Rfreea 0.1886 (0.3274) 0.1908 (0.3197) No. non-hydrogen atoms 4172 4092 Protein 3678 3672 Solvent 467 412 Ligands 27 8 RMSD Bonds (\u00c5) 0.005 0.005 RMSD Angles (\u00b0) 0.810 0.840 Ramachandran favored (%) 98.38 98.57 Ramachandran allowed (%) 1.62 1.43 Ramachandran outliers (%) 0.00 0.00 Clashscore 1.91 2.05 Average B factor (\u00c52) 22.85 26.46    Protein 21.23 24.97    Solvent 35,67 39.54    Ligands 21,29 37.17 No. TLS groups 7 7 229  Table C.5 Major qualitative effects of mutations in Crep1 on chemoselectivity, stereoselectivity, and substrate specificity             Table is adopted from Gagn\u00e9 et al.96  Mutant Desaturation: Acetylenation cis desaturation: trans desaturation total 16:2\/total 18:2 Y150F Increase Increase Decrease F183L Decrease Decrease Increase F183W Increase Increase Decrease F259L Increase Increase No change H266Q Increase Decrease No change V304I No change Increase No change 230   Figure C.1 Tested synthetic substrate mimics of AtyI. The trans-configured and allenic molecules are not substrates of AtyI (top), while the cis-configured allene and its methoxy derivative were both shown to be substrates of AtyI.     231   Figure C.2 Multiple sequence alignment of BesB with related PLP-dependent \u03b3-lyases.  232    Figure C.3 1H NMR spectrum of exchanged 4-chloroallylglycine. 400 MHz, D2O, 298K. \u03b4 5.39 \u2013 5.35 (m, 1H), 3.92 (dd, J = 9.0, 4.7 Hz, 1H), 2.92 (ddd, J = 15.1, 4.7, 1.1 Hz, 1H), 2.80 (ddd, J = 15.1, 8.9, 0.6 Hz, 1H).   233   Figure C.4 13C NMR spectrum of exchanged 4-chloroallylglycine. 101 MHz, D2O, 298 K. \u03b4 173.59, 135.97, 117.94, 52.62, 40.33.   234   Figure C.5 1D NOESY spectrum of exchanged 4-chloroallylglycine. 400 MHz, D2O, 298 K. Signals for C\u03b2-proton peaks at 2.92 and 2.80 ppm are saturated, demonstrating an NOE between 1H\u03b2 - 1H\u03b1 and 1H\u03b2 - 1H\u03b4. 235   Figure C.6 Multiple sequence alignment of BesB homologs. 236    Figure C.7 Complete tetrameric structure of a homotetrameric cystathionine \u03b3-lyase. Individual monomers in the structure (PDB: 6CJA) are distinctly colored. Active sites are composed of residues from two different monomers and are highlighted with black boxes.    237  3_17 amino acid sequence MSQAVSGTTGSADGLRHIAAGRPVPGSVHSVSVSIPDVASVIGYESNDAATLSRISWGYPRFRPHPYVVRVAELAAREDPAGEPGGALLLTRSARAARAAAAYAGLPPGAARDLTLGGHVLSGVRLPDRGPAAARARAFVQHTGGHLSSRQAEDVLWDAGLIDGRQVEETADDSPARAVRQALAGAYGVPGPRYVFLRNSGMNAVYAAIEAVTEIQRDRGRRHWLQLGWIFFDTMHLFEKKVVNVGHTTVPDPFDLAEVARVAAAHAGRLAGIITEIPSNPLMGVPDLPALREIADRAGCALVVDATIATPHNVDVVPYADVVCESLTKYATGSADVLAGAVVVNPGSPFAADLLTVLPRYGDEPYRRDTARVAARIRGYAERMRRVNANALALAECLRRHPDVVRDVSWALDTRSAANYRKVARDSGGPGGLLMVDLRVPLELVYDRLAVAKGPSFGAEFTMASPQVFVAHYDLLTTPRGRAELRARGLHRDMLRVSVGTEPPELIVETFERALRPD 3_17 optimized DNA sequence CATATGTCGCAGGCGGTCAGCGGTACCACCGGCTCCGCGGACGGTCTGCGCCACATCGCCGCGGGTCGCCCGGTGCCGGGTTCCGTGCACTCCGTGAGCGTCTCCATCCCGGACGTGGCGAGCGTGATCGGTTACGAATCCAACGACGCGGCCACCCTCTCCCGCATCAGCTGGGGCTACCCGCGCTTCCGCCCGCACCCGTACGTGGTGCGCGTGGCGGAGCTGGCGGCCCGCGAAGACCCGGCGGGCGAACCGGGGGGCGCGCTGCTGCTGACCCGCTCCGCGCGGGCCGCGCGCGCGGCGGCGGCGTACGCGGGTCTGCCGCCGGGCGCGGCGCGCGACCTGACGCTGGGGGGTCACGTGCTGTCGGGTGTGCGGCTGCCGGACCGCGGCCCGGCGGCCGCGCGCGCCCGGGCCTTCGTGCAGCACACCGGCGGTCACCTGTCGTCGCGCCAGGCGGAAGACGTGCTGTGGGACGCGGGTCTCATCGACGGCCGCCAGGTGGAGGAGACCGCCGACGACTCCCCGGCGCGCGCCGTCCGCCAGGCGCTGGCGGGCGCGTACGGGGTGCCGGGGCCGCGCTACGTCTTCCTGCGGAACAGCGGCATGAACGCCGTGTACGCGGCGATCGAAGCGGTGACCGAGATCCAGCGGGACCGGGGTCGGCGCCACTGGCTGCAGCTGGGCTGGATCTTCTTCGACACCATGCACCTGTTCGAGAAGAAGGTGGTGAACGTGGGTCACACCACCGTGCCGGACCCGTTCGACCTCGCGGAAGTGGCCCGCGTCGCGGCCGCCCACGCGGGCCGCCTGGCGGGTATCATCACCGAGATCCCGTCGAACCCGCTGATGGGGGTGCCGGACCTGCCGGCGCTGCGCGAAATCGCGGACCGCGCGGGGTGCGCGCTGGTGGTCGACGCGACGATCGCCACCCCGCACAACGTGGACGTGGTCCCGTACGCCGACGTGGTCTGCGAGTCCCTGACCAAGTACGCGACCGGCTCCGCGGACGTGCTGGCGGGCGCGGTGGTGGTGAACCCGGGGAGCCCGTTCGCGGCGGACCTCCTGACCGTGCTCCCGCGCTACGGCGACGAGCCGTACCGGCGGGACACCGCGCGGGTGGCGGCGCGCATCCGCGGCTACGCGGAGCGCATGCGGCGCGTGAACGCGAACGCCCTGGCCCTGGCCGAGTGCCTGCGGCGCCACCCGGACGTGGTGCGCGACGTCAGCTGGGCGCTGGACACCCGGTCCGCGGCGAACTACCGGAAGGTGGCGCGCGACTCGGGCGGTCCGGGTGGTCTGCTGATGGTGGACCTCCGCGTGCCGCTGGAGCTGGTGTACGACCGCCTGGCCGTGGCCAAGGGTCCGTCGTTCGGGGCGGAATTCACCATGGCGAGCCCGCAGGTGTTCGTGGCGCACTACGACCTGCTGACCACCCCGCGCGGGCGCGCGGAACTGCGCGCGCGGGGGCTCCACCGCGACATGCTGCGGGTGAGCGTGGGGACCGAACCGCCGGAGCTGATCGTGGAGACCTTCGAGCGGGCGCTGCGCCCGGACAAGCTT 5_30 amino acid sequence 238  MSQAVSGTTGSADGLRHIAAGRPVPGSVHSVSVSIPDVASVIGYESNDPATLSRISWGYPRFVPHPYVRRVAELAAREDPAGEPGGALLLTRSARAARAAAAYAGLPPGAARDLTLGGHVLSGVRLPDRGPAAARARAFVQHTGGHLSSRQAEDVLWDAGLIDGRQVEETADDSPARAVRQALARAYGVPGPRYVFLRNSGMNAVYAAIRAVTEIQRSRGRRHWLQLGWIFFDTMHLFEKKVVNVGHTTVPDPFDLAEVERVAAAHAGRLAGIITEIPSNPLMGVPDLPALREIADRAGCALVVDATIATPHNVDVVPYADVVCESLTKYATGSADVLAGAVVVNPGSPFAADLLTVLPRYGDEPYRRDLARVAARIRGYEERMKRVNANALALAECLRRHPDVVRDVHWALDTRSAANYRKVARPSGGPGGLLMVDLRVPLEKVYDRLAVAKGPSFGAEFTMASPQVFVAHYDLLTTPEGRAELRARGLHRDMLRVSVGTEPPELIVETFERALRPD 5_30 optimized DNA sequence CATATGTCGCAGGCCGTGTCGGGTACCACGGGGAGCGCGGACGGCCTCCGGCACATCGCGGCCGGGCGCCCGGTCCCGGGTTCGGTCCACTCGGTGTCGGTGTCGATCCCGGACGTGGCCAGCGTGATCGGCTACGAATCGAACGACCCGGCGACGCTGTCGCGGATCTCGTGGGGCTACCCGCGGTTCGTGCCGCACCCGTACGTGCGCCGGGTGGCGGAACTCGCCGCCCGCGAAGACCCGGCGGGTGAACCGGGTGGCGCCCTGCTGCTCACGCGCTCGGCGCGCGCCGCCCGGGCCGCGGCCGCCTACGCCGGTCTGCCGCCGGGCGCCGCGCGCGACCTGACGCTGGGTGGCCACGTGCTGTCCGGCGTCCGCCTGCCGGACCGGGGCCCGGCCGCGGCCCGCGCGCGGGCCTTCGTGCAGCACACGGGTGGTCACCTGTCGTCGCGCCAGGCCGAGGACGTGCTGTGGGACGCGGGCCTGATCGACGGGCGCCAGGTCGAAGAGACGGCGGACGACTCGCCGGCCCGGGCCGTGCGCCAGGCGCTGGCGCGCGCCTACGGTGTCCCGGGTCCGCGCTACGTGTTCCTCCGGAACTCGGGGATGAACGCGGTCTACGCCGCGATCCGCGCCGTGACGGAGATCCAGCGGTCCCGCGGTCGCCGCCACTGGCTGCAGCTCGGTTGGATCTTCTTCGACACCATGCACCTCTTCGAGAAGAAGGTGGTCAACGTGGGTCACACGACCGTGCCGGACCCGTTCGACCTCGCCGAGGTGGAGCGGGTGGCGGCCGCCCACGCGGGTCGCCTGGCCGGTATCATCACGGAAATCCCGTCCAACCCGCTCATGGGCGTGCCGGACCTCCCGGCGCTGCGCGAAATCGCGGACCGCGCGGGGTGCGCCCTCGTGGTCGACGCGACGATCGCCACGCCGCACAACGTCGACGTCGTGCCGTACGCGGACGTCGTGTGCGAATCGCTGACCAAGTACGCGACGGGTTCGGCCGACGTCCTCGCGGGCGCGGTGGTGGTCAACCCGGGCTCGCCGTTCGCGGCCGACCTCCTCACCGTCCTCCCGCGGTACGGTGACGAGCCGTACCGCCGGGACCTCGCGCGGGTCGCCGCGCGCATCCGCGGCTACGAAGAGCGCATGAAGCGCGTCAACGCCAACGCCCTGGCGCTGGCCGAATGCCTCCGCCGGCACCCGGACGTCGTGCGGGACGTGCACTGGGCGCTCGACACGCGCAGCGCGGCCAACTACCGCAAGGTGGCCCGGCCGTCGGGTGGGCCGGGGGGGCTGCTGATGGTCGACCTCCGCGTGCCGCTCGAAAAGGTCTACGACCGCCTGGCCGTGGCGAAGGGCCCGTCGTTCGGCGCCGAGTTCACGATGGCCTCGCCGCAGGTGTTCGTGGCGCACTACGACCTGCTGACGACCCCGGAGGGTCGGGCGGAACTGCGGGCCCGCGGTCTGCACCGGGACATGCTCCGCGTGTCGGTGGGGACCGAGCCGCCGGAACTCATCGTCGAAACCTTCGAGCGCGCCCTGCGCCCGGACAAGCTT   239  7_42 amino acid sequence MSQAVSGTTGSADGLRHIAAGRPVPGSVHSVSVSIPDVASVIGYESNDPATLSRISWGYPRFVPHPYVRRVAELAAREDPAGEPGGALLLTRSARAARAAAAYAGLPPSAARDLTLGGHVLSGVRLPDRGPAAARARAFVQHTGGHLSSRQAEDVLWDAGLIDGRQEEETEDDSPARAVRQALARAYGVPGPRYVFLRNSGMNAVYAAIRAVTEIQRSRGRRHWLQLGWIFFDTMHLFEKKVVDVGHTVVPDPFDLAEVERVAAAHAGRLAGIITEIPSNPLMGVPDLPALREIADRAGCALVVDATIATPHNVDVVPYADVVCESLTKYATGSGDVLAGAVVVNPGSPFAADLLKVLPRYGDPPYRRDAARVAARIRRYEERMRRVNANALALAECLRRHPDVVRDVHWALDTRSADNYRKVARPGGGPGGLLMVDLRVPLEKVYDRLAVAKGPSFGAEFTMASPQVFVAHYDLLTTPEGRAELRARGLHRDMLRVSVGTEPPEEIVETFERALRPD 7_42 optimized DNA sequence CATATGAGCCAGGCGGTGTCGGGGACCACCGGCTCGGCGGACGGGCTGCGGCACATCGCGGCGGGGCGCCCGGTGCCGGGCAGCGTGCACTCGGTGTCGGTGTCGATCCCGGACGTCGCGTCCGTCATCGGCTACGAGAGCAACGACCCGGCCACCCTGTCCCGCATCTCCTGGGGCTACCCGCGCTTCGTGCCGCACCCGTACGTGCGGCGGGTGGCCGAACTGGCGGCGCGCGAGGACCCGGCGGGGGAGCCGGGCGGGGCGCTGCTGCTGACCCGCAGCGCGCGGGCGGCGCGGGCGGCGGCCGCGTACGCGGGCCTGCCGCCGTCCGCCGCCCGCGACCTGACCCTCGGCGGGCACGTGCTGTCGGGCGTGCGGCTGCCGGACCGGGGCCCGGCGGCGGCCCGGGCGCGCGCGTTCGTCCAGCACACCGGGGGCCACCTGTCCTCCCGCCAGGCGGAAGACGTCCTGTGGGACGCGGGGCTCATCGACGGCCGGCAGGAGGAAGAAACGGAAGACGACAGCCCGGCGCGCGCGGTGCGCCAGGCGCTGGCGCGGGCGTACGGCGTGCCGGGCCCGCGGTACGTGTTCCTGCGCAACTCGGGCATGAACGCGGTGTACGCGGCCATCCGGGCGGTGACCGAGATCCAGCGCTCCCGGGGCCGGCGCCACTGGCTGCAGCTGGGCTGGATCTTCTTCGACACCATGCACCTGTTCGAGAAGAAGGTGGTCGACGTGGGCCACACCGTGGTCCCGGACCCGTTCGACCTGGCGGAGGTGGAGCGCGTGGCGGCCGCGCACGCCGGGCGCCTCGCGGGGATCATCACCGAAATCCCGTCGAACCCGCTGATGGGCGTGCCGGACCTGCCGGCCCTGCGCGAAATCGCGGACCGCGCCGGCTGCGCGCTGGTGGTGGACGCGACCATCGCGACGCCGCACAACGTCGACGTGGTCCCGTACGCGGACGTGGTGTGCGAGTCGCTGACCAAGTACGCGACGGGCAGCGGGGACGTGCTGGCGGGCGCGGTGGTCGTGAACCCGGGCTCCCCGTTCGCGGCGGACCTGCTGAAGGTGCTCCCGCGGTACGGCGACCCGCCGTACCGGCGCGACGCGGCGCGGGTGGCGGCGCGGATCCGCCGGTACGAGGAACGCATGCGCCGCGTCAACGCGAACGCGCTGGCCCTGGCGGAGTGCCTCCGGCGCCACCCGGACGTCGTGCGCGACGTGCACTGGGCGCTCGACACGCGCTCCGCGGACAACTACCGCAAGGTGGCGCGCCCGGGCGGGGGCCCGGGCGGGCTCCTGATGGTGGACCTGCGCGTGCCGCTGGAGAAGGTGTACGACCGCCTGGCGGTCGCGAAGGGCCCGTCGTTCGGCGCGGAGTTCACCATGGCCAGCCCGCAGGTCTTCGTGGCCCACTACGACCTGCTGACCACGCCGGAAGGCCGCGCGGAGCTGCGCGCGCGCGGCCTGCACCGCGACATGCTGCGGGTGTCCGTGGGCACCGAGCCGCCGGAAGAAATCGTGGAGACCTTCGAGCGCGCGCTGCGCCCGGACAAGCTT   240  8_50 amino acid sequence MSQAVSGTTGSADGLRHIAAGRPVPGSVHSVSVSIPDVASVIGYESNDPATLSRISWGYPRFVPHPYVRRVAELAAREDPAGGPGGTLLLTRSARAARAAAAYAGLPPSAARDLTLGGHVLSGVRLPDRGPAAERARAFVQHTGGHLSSRQAEDVLWDAGLIDGRQEEETEDDSPAEAVRQALARAYGVPGPRYVFLRNSGMNAVYAAIRAVTEIQRSRGRRHWLQLGWIFFDTMHLFEKKVVDVGHTIVPDPFDLAEVERVAAAHGGRLAGIITEIPSNPLMQVPDLPALREIADRAGCALVVDATIATPHNVDVVPYADVVCESLTKYATGSGDVLAGAVVVNPQSPFAADLLKVLPRYGDPPYRRDAARVAARIRRYEERMRRVNANALALAECLRRHPDVVRDVHWALDERSADNYRKVARPGGGPGGLLMVDLRVPLEKVYDRLAVAKGPSFGAEFTMASPQVFVAHYDLLTTPEGRAELRARGLHRDMLRVSVGTEPPEEIVETFERALRPD 8_50 optimized DNA sequence CATATGTCCCAGGCGGTGAGCGGGACCACCGGGTCCGCCGACGGCCTCCGCCACATCGCCGCCGGCCGCCCGGTGCCGGGCTCCGTCCACAGCGTCTCCGTCTCGATCCCGGACGTCGCGAGCGTCATCGGGTACGAAAGCAACGACCCGGCGACGCTCTCCCGCATCTCGTGGGGCTACCCGCGGTTCGTGCCGCACCCGTACGTGCGCCGCGTCGCGGAACTCGCGGCGCGCGAAGACCCGGCGGGCGGCCCGGGCGGCACGCTGCTGCTGACCCGCTCGGCGCGCGCGGCGCGCGCCGCGGCGGCGTACGCCGGCCTCCCGCCGAGCGCCGCCCGGGACCTGACGCTGGGGGGGCACGTCCTGTCCGGGGTCCGCCTCCCGGACCGCGGGCCGGCGGCGGAACGCGCCCGGGCGTTCGTCCAGCACACCGGCGGCCACCTGAGCAGCCGCCAGGCGGAAGACGTGCTCTGGGACGCGGGGCTCATCGACGGCCGCCAGGAGGAGGAGACCGAAGACGACTCGCCGGCGGAGGCGGTGCGCCAGGCGCTGGCCCGCGCGTACGGCGTCCCGGGCCCGCGCTACGTGTTCCTCCGCAACTCGGGCATGAACGCGGTCTACGCGGCGATCCGCGCGGTCACGGAGATCCAGCGGAGCCGCGGCCGCCGCCACTGGCTGCAGCTGGGGTGGATCTTCTTCGACACCATGCACCTGTTCGAAAAGAAGGTGGTGGACGTCGGCCACACGATCGTGCCGGACCCGTTCGACCTCGCGGAGGTCGAACGCGTGGCGGCCGCGCACGGCGGCCGCCTGGCGGGCATCATCACCGAAATCCCGAGCAACCCGCTGATGCAGGTGCCGGACCTCCCGGCGCTGCGCGAAATCGCGGACCGGGCGGGCTGCGCGCTGGTGGTGGACGCCACGATCGCGACCCCGCACAACGTCGACGTCGTGCCGTACGCCGACGTGGTCTGCGAGTCCCTGACGAAGTACGCGACGGGGTCCGGGGACGTGCTCGCGGGGGCGGTGGTCGTGAACCCGCAGAGCCCGTTCGCGGCGGACCTGCTGAAGGTCCTCCCGCGCTACGGCGACCCGCCGTACCGCCGCGACGCGGCGCGCGTGGCCGCGCGCATCCGCCGGTACGAAGAACGCATGCGGCGCGTCAACGCCAACGCGCTGGCGCTGGCGGAATGCCTGCGCCGCCACCCGGACGTGGTCCGCGACGTGCACTGGGCGCTGGACGAGCGCAGCGCGGACAACTACCGGAAGGTCGCCCGCCCGGGCGGCGGCCCGGGCGGCCTCCTCATGGTGGACCTGCGCGTCCCGCTGGAAAAGGTCTACGACCGGCTGGCGGTGGCCAAGGGCCCGTCCTTCGGGGCGGAATTCACCATGGCGAGCCCGCAGGTCTTCGTCGCGCACTACGACCTGCTGACCACGCCGGAGGGGCGGGCGGAGCTGCGGGCGCGCGGCCTCCACCGCGACATGCTGCGCGTGAGCGTCGGCACGGAGCCGCCGGAGGAAATCGTCGAAACCTTCGAGCGCGCGCTGCGGCCGGACAAGCTT   241  7_69 amino acid sequence MSQAVSGTTGSADGLRHIAAGRPVPGSVHSVSVSIPDVASVIGYEEKDPATLSRISWGYPRFVPHPYVVRVAELAAREDPAGEPGGALLLVRSARAARAAVAYAGLPPEAARDLTLGGHVLSGVRLPDRGPAAARARAFVQHTGGHLSSRQAEDVLWDAGLIDGRQEEETVDDSPAEIVRQVLARAYGVPGPDYVFLRNSGMNAVYAAFRAVTEIQRSRGRRHWLQLGWIFFDTIALFEKKIINVDHTIVPDPFDLAEIERIMAAHAGRLAGIITEIPSNPLMQVPDLPALRELCDRYGCVLVVDATIAGPVNVDVLPYADIVCESLTKYATGSADVLAGAVVVNPSSPWAADLLTVLPRYGDPPYPRDLARVAARIRRYEERMRRINANALALAECLRRHPDVVRDVFWALDERSAANYRKVARPSAGPGGLLMVDLRVPLEEVYDRLAVAKGPSFGAEFTMACPQVFVAHYDLLTTPEGRAELRARGLHRDLLRVSVGTEPPEEIVETFERALRPD 7_69 optimized DNA sequence CATATGTCGCAGGCCGTCTCGGGGACCACCGGCTCCGCCGACGGGCTGCGGCACATCGCGGCGGGTCGCCCGGTGCCGGGTAGCGTCCACAGCGTCTCGGTCTCGATCCCGGACGTCGCCTCGGTCATCGGTTACGAAGAAAAGGACCCGGCGACCCTCTCCCGCATCTCGTGGGGTTACCCGCGCTTCGTGCCGCACCCGTACGTCGTGCGCGTCGCCGAGCTGGCCGCCCGCGAGGACCCGGCGGGCGAGCCGGGCGGTGCCCTCCTGCTGGTGCGCTCCGCCCGCGCGGCCCGCGCCGCGGTCGCGTACGCCGGTCTGCCGCCGGAGGCCGCGCGCGACCTGACCCTCGGTGGTCACGTGCTGAGCGGCGTGCGCCTGCCGGACCGGGGTCCGGCGGCGGCGCGCGCGCGCGCCTTCGTGCAGCACACCGGGGGTCACCTGTCGTCCCGGCAGGCGGAGGACGTGCTGTGGGACGCCGGCCTGATCGACGGCCGGCAGGAAGAGGAGACCGTGGACGACTCCCCGGCGGAAATCGTCCGCCAGGTCCTGGCGCGGGCGTACGGTGTGCCGGGGCCGGACTACGTGTTCCTGCGGAACTCCGGTATGAACGCGGTCTACGCCGCGTTCCGGGCGGTGACCGAGATCCAGCGGTCGCGCGGCCGCCGGCACTGGCTGCAGCTGGGTTGGATCTTCTTCGACACCATCGCGCTGTTCGAGAAGAAGATCATCAACGTGGACCACACCATCGTGCCGGACCCGTTCGACCTCGCCGAAATCGAACGCATCATGGCCGCGCACGCGGGGCGCCTGGCCGGTATCATCACCGAGATCCCGTCGAACCCGCTGATGCAGGTCCCGGACCTCCCGGCGCTGCGCGAACTGTGCGACCGCTACGGCTGCGTGCTGGTGGTGGACGCCACCATCGCGGGGCCGGTCAACGTGGACGTGCTGCCGTACGCCGACATCGTGTGCGAATCGCTCACCAAGTACGCCACCGGTTCGGCCGACGTGCTGGCGGGTGCGGTGGTCGTCAACCCGTCGTCCCCGTGGGCGGCCGACCTCCTCACCGTGCTCCCGCGCTACGGTGACCCGCCGTACCCGCGCGACCTGGCGCGCGTGGCCGCCCGCATCCGCCGGTACGAAGAACGGATGCGCCGGATCAACGCCAACGCCCTGGCCCTCGCGGAATGCCTCCGCCGCCACCCGGACGTGGTGCGCGACGTCTTCTGGGCGCTGGACGAACGGTCGGCCGCGAACTACCGGAAGGTGGCGCGGCCGTCGGCGGGTCCGGGTGGTCTGCTGATGGTCGACCTGCGGGTGCCGCTGGAGGAAGTCTACGACCGCCTCGCGGTCGCCAAGGGTCCGTCCTTCGGCGCGGAGTTCACCATGGCGTGCCCGCAGGTGTTCGTCGCGCACTACGACCTCCTGACCACCCCGGAAGGCCGCGCGGAGCTCCGGGCGCGGGGCCTCCACCGGGACCTGCTGCGCGTGTCCGTCGGTACCGAGCCGCCGGAGGAGATCGTGGAGACGTTCGAACGCGCCCTGCGGCCGGACAAGCTT Figure C.8 PROSS Construct amino acid and DNA sequences. ","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/hasType":[{"value":"Thesis\/Dissertation","type":"literal","lang":"en"}],"http:\/\/vivoweb.org\/ontology\/core#dateIssued":[{"value":"2021-11","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/isShownAt":[{"value":"10.14288\/1.0400464","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/language":[{"value":"eng","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeDiscipline":[{"value":"Chemistry","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/provider":[{"value":"Vancouver : University of British Columbia Library","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/publisher":[{"value":"University of British Columbia","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/rights":[{"value":"Attribution-NonCommercial-NoDerivatives 4.0 International","type":"literal","lang":"*"}],"https:\/\/open.library.ubc.ca\/terms#rightsURI":[{"value":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/","type":"literal","lang":"*"}],"https:\/\/open.library.ubc.ca\/terms#scholarLevel":[{"value":"Graduate","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/contributor":[{"value":"Ryan, Katherine S.","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/title":[{"value":"Biochemical and structural studies of enzymes from the azomycin and beta-ethynylserine biosynthetic pathways","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/type":[{"value":"Text","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#identifierURI":[{"value":"http:\/\/hdl.handle.net\/2429\/79048","type":"literal","lang":"en"}]}}