UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Evolution of copper-containing nitrite reductase MacPherson, Iain 2007

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

Item Metadata

Download

Media
24-ubc_2008_spring_macpherson_iain.pdf [ 9.76MB ]
Metadata
JSON: 24-1.0066200.json
JSON-LD: 24-1.0066200-ld.json
RDF/XML (Pretty): 24-1.0066200-rdf.xml
RDF/JSON: 24-1.0066200-rdf.json
Turtle: 24-1.0066200-turtle.txt
N-Triples: 24-1.0066200-rdf-ntriples.txt
Original Record: 24-1.0066200-source.json
Full Text
24-1.0066200-fulltext.txt
Citation
24-1.0066200.ris

Full Text

EVOLUTION OF COPPER-CONTAINING NITRITE REDUCTASE  by  Iain Seido MacPherson B.Sc., Acadia University, 2001  A THESIS SUBMITTED TO THE FACULTY OF GRADUATE STUDIES IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in  THE FACULTY OF GRADUATE STUDIES (BIOCHEMISTRY AND MOLECULAR BIOLOGY)  THE UNIVERSITY OF BRITISH COLUMBIA  © Iain Seido MacPherson 2007  ii Abstract Copper-containing nitrite reductase (NiR) is a homotrimer of two cupredoxin domains and contains two spectroscopically distinct copper sites- a type-1 Cu that accepts electrons from a cupredoxin pseudoazurin and transfers them to a mononuclear type-2 Cu atom where NO2- is reduced to NO during denitrification. NiR has also been shown to reduce oxygen, for which the product is believed to be hydrogen peroxide (H2O2). Multicopper oxidases (MCOs) catalyze the reduction of oxygen to water and are homologous to NiR, containing multiple cupredoxin domains and a type-1 Cu site that transfers electrons to a trinuclear Cu active site, as opposed to a mononuclear active site of NiR. To investigate the evolution of NiR, methods of mutagenic library generation and high-throughput variant screening from E. coli colonies were developed. These methods allow for facile screening of 105 mutants for folding efficiency or substrate specificity. Initial proof of principle studies yielded several variants that oxidized the artificial substrate ο-dianisidine up to 8 times faster than wild type NiR, suggesting that this methodology has the potential to engineer NiR to acquire other reductase functions. A crystal structure was solved for a putative multicopper oxidase (MCO) and NiR homologue from Arthrobacter sp. (AMMCO) to 1.8 Å resolution. The overall folds of AMMCO and NiR are very similar (r.m.s.d. of 2.0 Å over 250 Cα atoms); Like NiR, AMMCO is a trimer with type-1 Cu sites in the N-terminal domain of each monomer; however, the active site of AMMCO contains trinuclear Cu site characteristic of MCOs instead of a the mononuclear type-2 Cu site found in NiR. Detailed structural analysis supports the theory that two-domain MCOs similar to AMMCO were intermediaries in  iii the evolution of NiR and the more common three-domain MCOs. The physiological function of AMMCO remains uncertain, but genomic, crystallographic and functional analysis suggests that the enzyme is involved in metal regulation. AMMCO homologues are co-localized within multiple genomes with a homologue of the ZIP family of zinc transporters. The x-ray crystal structure shows that a distinct loop of the enzyme (residues 184-192) bound calcium in distorted octahedral fashion. Calcium may not be the physiological metal bound at this site due to the high concentration (>200 mM) of Ca2+ required for crystallization. The AMMCO calcium binding loop was altered by sitedirected mutagenesis: Asp190Ala and Asp190Ala/Glu192Ala variants. These variants display altered electronic absorption spectra (shift in the 600 nm peak to 592 nm and increased absorbance at near-UV) and greater specific activity up to ~2X rates of wild type enzyme. The functional role of the loop is suggested to be in regulating oxidase activity. Considering the extensive similarity between AMMCO and NiR, particularly at the active site, engineering a trinuclear cluster into NiR appears feasible with a modest number of alterations to the polypeptide chain. With the aid of my newly developed highthroughput screening technique and site-directed mutagenesis, the mononuclear NiR active site was remodelled into a trinuclear Cu site similar to that of MCO. The substitutions Asp98His, Ile257His, Val304His, Asn305Thr and a 104-109 loop truncation were sufficient to allow for trinuclear cluster formation. A crystal structure of this variant was solved to 2.0 Å and the presence of three copper atoms at the engineered cluster was confirmed by Cu-edge anomalous diffraction data. Although the trinuclear copper cluster is present and catalyzes the reduction of oxygen, achieving rates of catalysis seen in  iv native MCOs has proven more difficult. With the framework provided, further engineering NiR into a robust MCO is likely to provide further insights into the structural basis of oxygen reduction by trinuclear copper sites.  v Table of Contents Abstract ............................................................................................................................... ii Table of Contents.................................................................................................................v List of Tables ................................................................................................................... viii List of Figures .................................................................................................................... ix List of Symbols, Abbreviations and Nomenclature........................................................... xi Acknowledgements........................................................................................................... xii Co-authorship Statement.................................................................................................. xiii CHAPTER ONE: INTRODUCTION..................................................................................1 1.1 Different copper types ...............................................................................................2 1.2 Type-1 Cu enzymes ...................................................................................................6 1.2.1 Nitrite reductase.................................................................................................6 1.2.2 Multicopper oxidase ........................................................................................10 1.3 Evolution of type-1 Cu proteins...............................................................................15 1.4 Directed evolution....................................................................................................20 1.4.1 Mutagenic library creation ..............................................................................21 1.4.1.1 Error-prone PCR ....................................................................................21 1.4.1.2 Site-specific randomization ...................................................................21 1.4.1.3 In vitro recombination ...........................................................................22 1.4.2 Screening or selection......................................................................................22 1.4.2.1 Mutagenic library size limitations .........................................................23 1.5 Thesis objectives......................................................................................................23 1.6 References................................................................................................................25 CHAPTER TWO: HIGH-THROUGHPUT SCREENING STUDIES OF NITRITE REDUCTASE ...........................................................................................................33 2.1 Introduction..............................................................................................................33 2.2 Materials and methods .............................................................................................35 2.2.1 Library construction ........................................................................................35 2.2.2 Screening method ............................................................................................37 2.2.3 Protein expression and purification .................................................................39 2.2.4 Activity assays.................................................................................................40 2.2.5 Electrochemistry..............................................................................................40 2.2.6 Electronic absorption spectroscopy.................................................................41 2.2.7 Crystal structures .............................................................................................41 2.3 Results......................................................................................................................43 2.3.1 Improved cloning efficiency............................................................................43 2.3.2 High-throughput screening..............................................................................45 2.3.3 ο-Dianisidine and pseudoazurin oxidation......................................................45 2.3.4 Type-1 Cu site reduction potentials and electronic spectra.............................47 2.3.5 Crystal structures .............................................................................................48 2.4 Discussion................................................................................................................52 2.4.1 Library generation and colony screen .............................................................52  vi 2.4.2 Oxygen reduction in wt NiR............................................................................53 2.4.3 Oxygen and nitrite reduction with DAB and o-dianisidine.............................55 2.5 References................................................................................................................59 CHAPTER THREE: CRYSTAL STRUCTURE AND FUNCTIONAL CHARACTERIZATION OF A TRIMERIC MULTICOPPER OXIDASE FROM ARTHROBACTER ........................................................................................65 3.1 Introduction..............................................................................................................65 3.2 Materials and methods .............................................................................................67 3.2.1 Protein expression ...........................................................................................67 3.2.2 Crystallography ...............................................................................................69 3.2.3 Cyclic voltammetry .........................................................................................69 3.2.4 Assays..............................................................................................................71 3.3 Results......................................................................................................................71 3.3.1 Crystal structure...............................................................................................71 3.3.2 Electronic absorption spectra ..........................................................................76 3.3.3 Electrochemical and activity assays ................................................................76 3.4 Discussion................................................................................................................79 3.4.1 AMMCO is a link between monomeric MCOs and NiR ................................79 3.4.2 The AMMCO type-1 Cu site...........................................................................86 3.4.3 AMMCO functional roles................................................................................88 3.5 References................................................................................................................93 CHAPTER FOUR: DIRECTED MOLECULAR EVOLUTION OF NITRITE REDUCTASE INTO A MULTICOPPER OXIDASE .............................................99 4.1 Introduction..............................................................................................................99 4.2 Materials and methods ...........................................................................................101 4.2.1 Mutagenesis, cloning and screening..............................................................101 4.2.2 Protein expression and purification ...............................................................107 4.2.3 Copper content...............................................................................................108 4.2.4 Electronic absorption spectroscopy...............................................................108 4.2.5 Activity assays...............................................................................................109 4.2.6 Crystallography .............................................................................................109 4.3 Results....................................................................................................................110 4.3.1 Trinuclear cluster design ...............................................................................110 4.3.2 Functional assessment of NiR variants..........................................................112 4.3.3 Crystal structure of ligand loop truncation (variant #2b) .............................125 4.3.4 Crystal structure of variant #8 .......................................................................126 4.4 Discussion..............................................................................................................129 4.4.1 Critical evaluation of the design and engineering .........................................131 4.4.1.1 Type-1 ligand loop length and trinuclear Cu binding..........................131 4.4.1.2 Residues 303 to 305.............................................................................132 4.4.1.3 Residue 104-109 loop truncation.........................................................133 4.4.1.4 Effectiveness of the high-throughput screen .......................................134 4.4.2 Comparison between the AMMCO and variant #8 structures ......................135  vii 4.4.3 Metal site engineering ...................................................................................137 4.4.4 Dioxygen binding by variant #2b ..................................................................137 4.4.5 Future directions for this engineering project ...............................................138 4.5 References..............................................................................................................140 CHAPTER FIVE: SUMMARY AND FUTURE DIRECTIONS....................................146 5.1 Summary................................................................................................................146 5.2 Future directions ....................................................................................................147 5.3 References..............................................................................................................151  viii List of Tables Table 2-1 Data collection and refinement statistics.......................................................... 42 Table 2-2 Reduction potentials and kcat values (relative to wt) of variant and wt NiR .... 46 Table 3-1 Data collection and refinement statistics for wt recombinant AMMCO......... 70 Table 3-2 Copper content and kcat’s of AMMCO and variants......................................... 80 Table 4-1 Grid showing the alterations present in variants #1 through #9..................... 103 Table 4-2 Primers used for site-directed mutagenesis, site-specific randomization, and random mutagenesis................................................................................................ 104 Table 4-3 Data collection and refinement statistics........................................................ 111 Table 4-4 kcat and copper content values for variants ..................................................... 116  ix List of Figures Figure 1-1 Type-1 Cu site of NiR (PDB entry 1SJM) and CuA site in nitrous oxide reductase (PDB entry 1FWX)..................................................................................... 3 Figure 1-2 Cu active sites in five enzyme families............................................................. 5 Figure 1-3 Type-1 and type-2 Cu sites in NiR.................................................................... 7 Figure 1-4 Proposed mechanism of Cu containing NiR ..................................................... 9 Figure 1-5 Proposed catalytic cycle of multicopper oxidases .......................................... 13 Figure 1-6 Stereo view of the superposition of the trinuclear cluster of the MCO laccase (PDB entry 1GYC) with the type-2 site of NiR (PDB entry 1SJM) ............ 16 Figure 1-7 Proposed scheme for the evolution of the type-1 copper proteins .................. 18 Figure 1-8 Cupredoxin domains in NiR............................................................................ 19 Figure 2-1 Typical megaprimer cloning reactions from various cycle numbers (12-24) . 36 Figure 2-2 Picture of a screened membrane. About 2000 clones .................................... 38 Figure 2-3 Schematic representation of the megaprimer-based cloning method ............. 44 Figure 2-4 Electronic spectra of wt and variant NiRs....................................................... 49 Figure 2-5 Residue positions of positively screened variants mapped to the wt NiR structure (PDB code 1SJM) ...................................................................................... 50 Figure 2-6 Crystal structures of variants........................................................................... 51 Figure 3-1 Overall structural comparison of AMMCO and AfNiR ................................. 73 Figure 3-2 Metal sites of AMMCO .................................................................................. 74 Figure 3-3 Electronic absorption spectra of wt recombinant AMMCO, E192A, and D190A/E192A .......................................................................................................... 77 Figure 3-4 Cyclic voltammogram (Current plotted against potential (vs. SCE)) of AMMCO compared to bare electrode....................................................................... 78 Figure 3-5 Sequence alignment of AMMCO, AniA, and AfNiR ..................................... 82 Figure 3-6 Stereo image of NiR and AMMCO active sites superposed........................... 83  x Figure 3-7 Surface architecture of AMMCO, NiR from Neisseria gonnorhea, and NiR from A. faecalis ......................................................................................................... 89 Figure 4-1 This flowchart represents the mutational lineage of this project. ................. 102 Figure 4-2 Superposition of NiR (PDB entry 1SJM) with modelled histidine substitutions and CotA (PDB entry 1GSK) ............................................................ 113 Figure 4-3 Superposition of the type-1 sites from wt NiR and the MCO CotA ............. 114 Figure 4-4 Electronic absorption spectra of the two characterized type-1 loop truncation variants, as isolated ................................................................................ 117 Figure 4-5 Electronic spectra of variants #4, #5, and #7 ................................................ 119 Figure 4-6 Electronic absorption spectra of variant #8 reduced with ascorbate............. 121 Figure 4-7 O2 consumption and H2O2 production by wt NiR and variant #8................. 122 Figure 4-8 Electronic absorption spectra of variant #9................................................... 123 Figure 4-9 Stereo views of the crystal structure of variant #2b. ..................................... 124 Figure 4-10 Cu-edge anomalous map (contoured to 3 σ) of variant #8 superposed against the native NiR structure showing significant density for four Cu atoms. .. 127 Figure 4-11 ∆104-109 component of variant #8 superposed against wt NiR................. 128 Figure 5-1 Schematic representation of an NiR biosensor ............................................. 149  xi List of Symbols, Abbreviations and Nomenclature Symbol A Å AfNiR AMMCO BCA DAB DFT DMSO dNTPs ε E1/2 E0 ’ EPR IPTG kcat LMCT MCO NADH NHE NiR O.D.600 OH· PCR PDB PHM S.C.E. SOD SSRL Taq TPQ U UV  Definition Hyperfine coupling constant Angstroms Nitrite reductase from Alcaligenes faecalis Arthrobacter multimeric multicopper oxidase bicinchoninic acid 3,3’diaminobenzidine Density functional theory Dimethylsulfoxide Deoxyribonucleotide triphosphates Extinction coefficient Midpoint reduction potential Standard reduction potential at pH 7 Electron paramagnetic resonance Isopropyl β-D-1-thiogalactopyranoside Catalytic rate constant Ligand-to-metal charge transfer Multicopper oxidase Protonated nicotinamide adenine dinucleotide Normal hydrogen electrode Copper-containing nitrite reductase Optical density at 600 nm Hydroxyl radical Polymerase chain reaction Protein databank Peptidylglycine α-hydroxylating monooxygenase Saturated calomel electrode Superoxide dismutase Stanford synchrotron radiation laboratory Thermostable polymerase from Thermophilus aquaticus Trihydroxyphenylalanine Units Ultraviolet  xii Acknowledgements I thank my supervisor Michael Murphy for his guidance throughout my graduate program. I am also grateful for my committee members Ross MacGillivray and Lindsay Eltis for their continuous support and help. I also thank Mike Page (MacGillivray lab) and Pascal Fortin (Eltis Lab) who taught me much of the knowledge and techniques that I used during my PhD. Thanks also to the rest of the Murphy lab for their help and support. Lastly, I thank my family and friends for their continuous support.  xiii Co-authorship Statement Scientific collaborations occurred throughout this thesis research. Below is a summary of the contributions of other scientists. Much of Chapter 1 (Introduction) was part of a review published in MacPherson, I. S. and M. E. P. Murphy. Type-2 Copper-containing enzymes. Cell. Mol. Life Sci. EPub ahead of print. 2007. Michael Murphy wrote the opening section and section 1-1 of the introduction, and prepared the panels for Figure 1-2. Elitza Tocheva produced Figure 1-4 (Mechanism of nitrite reductase). I produced all other text in the review, as well as figures not reprinted. Chapter 2 is a draft of a manuscript that will be submitted as: MacPherson, I. S., Rosell, F. I., Scofield, M., Mauk, A. G., and M. E. P. Murphy. High-throughput screening studies of nitrite reductase. Fred Rosell performed most of the cyclic voltammetry experiments to determine reduction potentials of the variants. I supervised a summer student, Melanie Scofield, who performed some protein preparations and helped with preliminary turnover assays. I performed all other experiments. Chapter 3 is a draft of a manuscript that will be submitted as: MacPherson, I. S., Lee, W. C., Liang, T. I., and M. E. P. Murphy. Crystal structure and functional characterization of a trimeric multicopper oxidase from Arthrobacter. Woo Cheol Lee supervised the x-ray data collection and solved the AMMCO crystal structure. I supervised a summer student, Teresa Liang, who produced native protein and crystals. Teresa Liang also performed preliminary activity assays of AMMCO. I performed all other experiments in the chapter.  xiv Chapter 4 contains experiments performed by myself exclusively, and is a draft that will be submitted as: MacPherson, I. S. and M. E. P. Murphy. Engineering a trinuclear Cu site into copper-containing nitrite reductase. Woo Cheol Lee provided helpful suggestions for refinement of the anomalous dataset.  1 Chapter One: Introduction Enzymes incorporate transition metal cofactors to perform essential metabolic reactions. Examples include proteins that comprise electron transport chains and catalyze difficult reactions such as nitrogen fixation and the reduction of ribonucleotides to deoxyribonucleotides. The most common metals employed are iron, zinc and copper; however, only iron and copper are suitable to perform redox reactions. The reduction potentials and coordination chemistries of both iron and copper are ideal for many biological processes and thus these metals are found broadly in nature. Moreover, these metals are able to bind and manipulate gaseous substrates such as oxygen and nitric oxide. For a particular system, the type of metal found in an enzyme is likely to be a consequence of its unique chemical properties and bioavailability in the course of evolution. The bioavailability of copper and iron has changed dramatically over the geological history of the earth (for a review see Williams and Frausto de Silva 1996) [1]. Before the advent of photosynthesis, an anaerobic earth favored reduced iron (Fe(II)) and sulfur (sulfides). Iron bioavailability was dominant since iron sulfides are much more soluble than copper sulfides. As the earth became oxygenated by early photosynthetic organisms, iron was oxidized to the less soluble Fe(III) form that precipitated as iron oxides. In contrast, as sulfides were oxidized to sulfates, copper was liberated since copper oxides are generally soluble, allowing organisms to incorporate this new metal into protein scaffolds to perform chemistry that previously was solely the domain of iron. The increased bioavailability of copper occurred at a time when organisms were adapting to a *A version of this chapter has been accepted for publication. MacPherson, I. S. and M. E. P. Murphy. Type2 Copper-containing enzymes. Cell. Mol. Life Sci. EPub ahead of print. 2007.  2 new oxic environment, which may explain why many copper proteins are found in systems that mediate oxygen chemistry. 1.1 Different copper types The Cu sites observed in proteins are classified into three types based on their structural and spectroscopic properties. Type-1 Cu proteins, traditionally known as blue Cu proteins, attracted attention as a consequence of their intense blue to green color [2]. Strong visible absorption (ε600 ≈ 5000 M-1cm-1) is characteristic of type-1 Cu sites when in the oxidized (Cu(II)) state. The reduction potentials of most type-1 Cu sites are sufficiently low that the oxidized state is favored in the presence of ambient oxygen. Structurally, the Cu atom of a typical type-1 site is coordinated by a Cys and two His residues in a trigonal planar arrangement (Figure 1-1). Often the thioether of a Met coordinates axially distorting the geometry towards tetrahedral. The color is due to charge transfer (LMCT) between the Cu and the S atom of the Cys ligand. Electron paramagnetic resonance (EPR) spectra of the oxidized site show unusually low coupling constants (A values) [3]. The function of these sites is exclusively single electron transfer reactions. Type-1 sites are found in small electron transfer proteins (cupredoxins) that ferry electrons between larger enzymes such as components of the denitrification pathway and photosynthesis. Also, type-1 sites are found in the larger enzymes nitrite reductase and multicopper oxidase and function in intramolecular electron transfer to copper active sites. Within some of the large enzymes, a CuA site functions as an electron entry point. This site is an expansion of the type-1 site by a second Cu to form a  3  Figure 1-1 Type-1 Cu site of NiR (left, PDB entry 1SJM) and CuA site in nitrous oxide reductase (right, PDB entry 1FWX) that function in electron transfer. In each panel, the copper atoms are large bronze spheres, amino acid ligands are drawn in ball and sticks with C atoms in green or purple and N and S atoms in blue and yellow, respectively.  4 metal-metal bond (Figure 1-1). Both type-1 and CuA sites are rigid and characterized by low reorganization energies to facilitate electron transfer [4,5]. Type-2 is used to designate Cu sites with a variety of amino acid ligands and geometries. Most type-2 sites are three to four coordinate and one or more of the Cu ligands are the imidazole side-chains of histidines (Figure 1-2). The coordination sphere may be completed by methionine, glutamate, glutamine or tyrosine. The absence of a thiol group results in weak visible absorption and thus no evident color. EPR spectra of type-2 sites are characterized by a weaker signal with larger A values and are clearly distinct from type-1 site spectra [3]. Coordination positions in type-2 Cu sites can either be vacant or occupied by exogenous ligands. Consequently, these sites can be catalytically active by interacting directly with enzyme substrates. When molecular oxygen is the substrate type-2 sites may function as: (1) oxidases, reducing oxygen to water or peroxide (Figure 1-2a&f); (2) monooxygenases, where one oxygen atom is inserted into the substrate and the other is reduced to water (Figure 1-2b) or (3) dioxygenases, where both oxygen atoms are incorporated into the substrate. In addition, type-2 sites are able to perform the dismutation of superoxide (SOD, Figure 1-2c&d) and reduce nitrite to nitric oxide (NiR, Figure 1-2e). More complex Cu sites such as the CuA site mentioned above are constructed from multiple metal centers. Type-3 sites consist of two antiferromagnetically coupled Cu atoms bridged by molecular oxygen or a hydroxyl. The type-3 pair plus a third Cu (type2) is part of the trinuclear cluster in multicopper oxidases (Figure 1-2f) [6]. A dinuclear type-3 Cu site is found in hemocyanins which function as oxygen carriers in invertebrates  5 c a  b  d  e  f  Figure 1-2 Cu active sites in five enzyme families. In each panel, amino acid ligands are drawn in ball and sticks with C, N, O, S atoms in orange, blue, red and yellow, respectively. The Cu atoms (brown) and solvent atoms (cyan) are depicted as spheres. Grey spheres in superoxide dismutase are zinc atoms. (a) Hydroperoxo-bound CuAO (PDB entry 1D6Z). Part of the TPQ cofactor is shown. (b) End-on superoxo-bound to PHM (PDB entry 1SDW). (c) Bridged SOD (PDB entry 2SOD). (d) Bridge-broken SOD (PDB entry 2JCW). (e) Resting state of the trinuclear active site of laccase (PDB entry 1GYC). (f) NO bound to NiR (PDB entry 1SNR).  6 [6]. A similar site is observed in tyrosinases; however these enzymes function as monooxygenases and activate oxygen for insertion in phenolic substrates [6]. 1.2 Type-1 Cu enzymes 1.2.1 Nitrite reductase Copper-containing nitrite reductase (NiR) catalyzes the one-electron reduction of nitrite (NO2-) to nitric oxide (NO) by the following reaction: NO2- + 2H+ + e-  NO + H2O  This process occurs during dissimilatory denitrification, in which nitrate and nitrite and their metabolites are used as electron acceptors when oxygen levels are low [7]. NiR is secreted in the periplasmic space between the inner and outer membrane of Gram negative bacteria. Two types of NiRs are known, one that uses copper and the other heme iron cofactors [7]. By sequence analysis, copper containing NiRs are found in unicellular organisms from all the main branches of life including some fungi. Most known examples are bacterial including all the lineages of proteobacteria, flavobacteria, eubacteria, as well as some archaea. In addition to fungi, the other eukaryotic examples are found in species of amoeba. The typical gene name for copper containing NiR is nirK. Due to the similarity between the multicopper oxidase family and NiR, many nirK sequences are misannotated in genome sequence projects; however, the sequences are easily distinguished by sequence motifs with the presence and absence of active site residues [8]. NiR is typically a 110 kDa trimer, containing a type-1 (blue) Cu site within each monomer and a type-2 Cu site located between monomers (Fig. 1-3). Each monomer is composed of two homologous domains with a Greek-key β-barrel (cupredoxin) fold [9].  7  Ile257 Cys136  Asp98  His255 His135 Type-2 Type-1  Figure 1-3 Type-1 and type-2 Cu sites in NiR. The type-1 site is embedded in the Nterminal domain of one monomer (green). The type-2 site is at the interface with a second monomer (orange). The Cu atoms are drawn as bronze spheres and are linked by the Cys136-His135 peptide bond. Nitrite (red and blue) is shown bound at the active site type-2 Cu which is coordinated by histidines from two adjacent monomers. Three catalytically important residues (Asp98, His255, and Ile257) are labelled. Carbon atoms are colored green or orange to match the monomer of origin; oxygen, nitrogen and sulphur atoms are red, blue and yellow, respectively.  8 The type-1 Cu site accepts electrons from a small protein electron donor. Electrons are subsequently transferred to the type-2 Cu site, where nitrite is reduced to nitric oxide gas [7]. The tetrahedral type-2 site is formed from three histidines, one of which is derived from an adjacent monomer (Figure 1-2e). In the oxidized resting state, the fourth position is occupied by a water molecule. The type-1 and type-2 Cu sites are linked by a Cys-His bridge such that a cysteine coordinates to the type-1 Cu and an adjacent histidine on the protein chain coordinates to the type-2 Cu [9]. This bridge is believed to facilitate rapid rates (>1000 s-1) of electron transfer between the type-1 and type-2 Cu sites [10]. Recently, a variant of the typical NiR from Hyphomicrobium denitrificans was shown to have an additional N-terminal cupredoxin domain [11]. The catalytic mechanism of NiR proceeds in a random sequential order [12] to give a Cu-nitrosyl intermediate [7,13] (Figure 1-4). Site-directed mutagenesis studies implicate Asp98 (numbering from Achromobacter cycloclastes NiR) and His255 directly in the catalytic mechanism [14,15]. Crystallographic structures reveal that the Asp98 forms an H-bond to both bound nitrite substrate and nitric oxide product [13,14]. His255 is proposed to participate in proton transfer to a catalytic intermediate either directly or via Asp98 (Figure 1-4). A large hydrophobic residue, usually an isoleucine (Ile257), partially occludes the type-2 active site limiting the binding of larger substrates (Figure 13). Mutation of this residue results in alternate non productive binding modes of nitrite to the Cu [16,17]. In the presence of chemical reductants, NiR can transform nitric oxide to nitrous oxide [7]. The concentration of nitric oxide reaches a steady state level of 80 nM suggesting that it is a potent inhibitor of catalysis [7]. In crystals of NiR exposed to nitric oxide, the diatomic molecule binds in a unique side-on fashion to the type-2 Cu [13].  9  a b  c  d  Figure 1-4 Proposed mechanism of Cu containing NiR. ET = electron transfer.  10 Side-on binding of NO is supported by DFT calculations on both model compounds [18] and the enzyme active site [19]. Nitric oxide is a structural analogue of O2 and has been shown to bind O2 reducing type-2 Cu enzymes such as amine oxidase [20] and laccase [21]. Thus, that NiR is able to reduce oxygen is not surprising [22], given the strong affinity of nitric oxide for the type-2 Cu site. The reduction product is believed to be hydrogen peroxide [22]. Continued reduction of hydrogen peroxide leads to the inactivation of NiR, suggesting that destructive hydroxyl radicals are being produced. This inactivation mechanism is supported by an abolition of enzyme inactivation when catalase is present in the reaction [22]. In vivo, the switch from anaerobic nitrate reduction to aerobic conditions in the absence of nitrite is shown to result in a rapid inactivation of NiR in A. faecalis, underlining the importance of oxygen reduction by NiR in nature [23]. The mechanism by which NiR reduces O2 remains poorly understood. 1.2.2 Multicopper oxidase Multicopper oxidases (MCOs) catalyze the oxidation of various small molecules and cations with the concomitant four-electron reduction of oxygen to water: O2 + 4H+ + 4e–  2H2O  Distribution of MCOs is widespread across both prokaryotes and eukaryotes. Two of the best studied prokaryotic MCOs are CueO from E. coli and CotA from Baccilus subtilus. In E. coli, CueO is proposed to function as a cuprous oxidase, oxidizing Cu(I) to Cu(II) and thus limiting Fenton chemistry that leads to the destructive oxygen radical species hydrogen peroxide (H2O2) and hydroxyl radical (OH·) [24]. CotA is involved in spore coat formation [25]. Ferroxidases catalyze the oxidation of Fe(II) to Fe(III) and have been  11 identified in mammals (humans) as ceruloplasmin and hephaestin or in yeast, as Fet3. Ferroxidase activity has been ascribed to iron transport pathways of both organisms [26]. Laccases constitute another kind of MCO and derive electrons from the oxidation of phenolic compounds [27]. These enzymes are common in wood rot fungi and have been shown to take part in the delignification process [27]. The word “laccase” has been freely extended to any MCO of unknown function capable of phenolic compound oxidation. However it should be noted that most MCOs, including ceruloplasmin and Fet3, are capable of phenolic compound oxidation. MCOs, like NiR, contain multiple Greek key β-barrel domains with multiple copper sites. Substrate oxidation occurs at or proximal to the type-1 Cu site. The reduction potential of MCOs range from 330-780 mV [28], significantly higher than that of NiR (240-260 mV [29,30]). The reduction potential of MCO type-1 Cu sites loosely correlates to the identity of the type-1 Cu axial ligand [28]. In the low potential MCOs, the axial ligand is often a weakly coordinating methionine ligand. In the mid-potential MCOs, the axial ligand is often a non-coordinating leucine, and in the high potential MCOs it is often a phenylalanine. Oxygen reduction to water in MCOs occurs at a trinuclear Cu cluster, coordinated by a total of 8 histidine ligands. The Cu atoms are located within 5 Å from each other. One of the Cu atoms, coordinated by two histidines and one solvent molecule, is EPRactive and has been given the designation of type-2. The other two atoms, connected to the type-1 Cu by separate Cys-His bridges, are EPR-silent, and this silence has been ascribed to strong antiferromagnetic coupling of the Cu sites by a bridging hydroxyl  12 (OH–) ligand [31]. The type-3 pair gives MCOs a distinctive electronic absorption shoulder at 330 nm (ε ≈ 4000 M-1cm-1) [32]. Catalysis by the trinuclear Cu active site of MCO has been studied extensively, particularly by the laboratory of Edward Solomon. A key characteristic of the trinuclear active site is, as termed by Solomon et al, “coordination unsaturation” [33]. Thus, the coordination spheres of the three Cu atoms may be completed by substrates, intermediates, and products during catalysis. The stabilization of this coordination unsaturation is attributed to the presence of negatively charged carboxylate side chains near the active site pocket, which stabilize the positively charged Cu cluster and are suggested to tune the trinuclear cluster for catalysis [33]. The four-electron reduction of oxygen to water corresponds to the four copper atoms (one type-1, two type-3, one type-2) in each MCO monomer (Fig. 1-5). The type-1 Cu was replaced with mercury, thus allowing a 3-electron reduction of oxygen [34]. This process is much slower (~107 times slower), allowing the observation of catalytic intermediates. In mercury-substituted MCO, Solomon et al observed a two-electron reduced intermediate, which they predicted to be a peroxide-level intermediate bridged between the reduced type-2 Cu and an oxidized type-3 Cu. Later, it was determined that the peroxide-level intermediate is bridged between all the copper atoms of the trinuclear cluster [35](Figure 1-5). Reduction by the third electron is proposed to yield hydroxyl radical (OH·) in the mercury-substituted enzyme, whereas the two electron reduction of the peroxide-level intermediate produces water. The differences in reaction rates by three and four electron reductions were attributed to the different reduction potentials of the  13  Figure 1-5 Proposed catalytic cycle of multicopper oxidases, reprinted from Acc Chem Res 40, Solomon, E.I., R. Sarangi, J.S. Woertink, A.J. Augustine, J. Yoon, and S. Ghosh, O2 and N2O activation by Bi-, Tri-, and tetranuclear Cu clusters in biology, p. 581-91, with permission. Copyright 2007 American Chemical Society.  14 three electron and four electron processes. The reduction of hydrogen peroxide to hydroxyl radical has an E0’ = 0.38 V, whereas the reduction of peroxide to water has an E0’ = 1.35 V [34]. Thus, the driving force for the four-electron process is greater and allows for greater reaction rates. The simultaneous reduction of the peroxide-level intermediate by two electrons results in the native intermediate, which is spectroscopically distinct from the restingstate enzyme [36]. In the presence of reducing equivalents, the native intermediate rapidly accepts electrons and protons to produce two molar equivalents of water [36]. It is proposed that an oxo species bridged between all three Cu atoms allows for rapid electron transfer through the trinuclear Cu cluster and reduction of the Cu sites is coupled to protonation [37]. In the absence of reducing equivalents, the native intermediate slowly returns to resting oxidized enzyme (0.34 sec-1) [36]. Much of the catalytic mechanism of MCOs is currently unknown. In particular, the source of protons for the overall reaction is unclear. In 2005, Solomon et al proposed that the proton required for O-O bond cleavage of the bridged peroxide intermediate may come from a type-2 Cu-water-water-Asp hydrogen bonding network [38]. The aspartate is residue 94 in Fet3, and is highly conserved among MCOs. This aspartate also forms a hydrogen bond with a type-3 Cu His ligand, which complicates mutational analysis for the essentiality of this residue. In a more recent review of MCO catalysis, Solomon et al. suggest a glutamate near the type-3 pair as a proton source [36]. Thus, literature to date is unclear and further experiments are likely taking place to elucidate the sources of protons in catalysis.  15 1.3 Evolution of type-1 Cu proteins Striking similarity exists between the Cu sites of NiR and those of the multicopper oxidases, both structurally and mechanistically. Furthermore, the fold and cupredoxin domain arrangement suggest that NiR and MCOs share a common ancestor [39,40]. As noted, both enzymes are capable of reducing oxygen. While NiR releases hydrogen peroxide, MCOs form a peroxide-level intermediate on the pathway towards the full oxygen reduction to water. Reminiscent of hydroxyl radical formation by NiR, both laccase and Fet3p are proposed to produce a hydroxyl radical when the type-1 Cu site is made nonfunctional resulting in a three electron reduction of oxygen [34,41]. A superposition of the amino acids that form the active Cu sites from Alcaligenes faecalis NiR (PDB entry 1SJM) and Trametes versicolor laccase (PDB entry 1GYC) is shown in Fig. 1-6. Firstly, in both NiRs and MCOs the blue type-1 Cu sites are connected to a second Cu site via a Cys-His bridge. Paradoxically, the type-1 sites are derived from different domains, the N-terminal domain for NiR and the C-terminal domain for laccase. The oxygen reducing site of MCO is a trinuclear copper cluster consisting of two antiferromagnetically coupled type-3 Cu atoms and one type-2 Cu. In the superposed structures, the type-2 Cu site of NiR overlaps with one of the type-3 Cu atoms of laccase, including the positions of the three histidine residues coordinating to the respective Cu atoms (Fig. 1-6). Of the remaining five His ligands that coordinate the type-2 and second type-3 Cu atoms, one aligns with His255, an essential residue in NiR catalysis. Furthermore, the other two identified catalytically important NiR residues, Asp98 and Ile257, both superpose with MCO His ligands. Lastly, if Ala137 and Val304 in NiR were changed to His residues, an approximate MCO trinuclear cluster would be complete. As  16  Ile257 His255  Asp98  Figure 1-6 Stereo view of the superposition of the trinuclear cluster of the MCO laccase (colored blue, PDB entry 1GYC) with the type-2 site of NiR (colored green, PDB entry 1SJM). Note the opposite orientation of the type-1 copper sites (far left and far right spheres). The NiR type-2 copper and laccase type-3 copper overlap. NiR residues Asp98, His255 and Ile257 are labeled.  17 mentioned, the type-2 Cu of NiR is located between monomers (His100 and 135 derived from one monomer, His306 from an adjacent monomer). Considering that laccase is monomeric, the active site structural similarity to NiR is remarkable. Multimeric small laccases are characterized functionally and spectroscopically [42,43] but await crystal structure determination. Nakamura and Go have recently proposed an evolutionary model for type-1 copper proteins, including the cupredoxins, nitrite reductase, and multimeric and monomeric multicopper oxidases. The simplest of the type-1 Cu proteins, the cupredoxins, are proposed to have spawned the more complex, multicopper enzymes as shown in Figure 1-7. A duplication of a single cupredoxin domain followed by multimerization and the addition of a mononuclear Cu site shared at the interface between monomers would yield a trimeric protein similar to NiR. Loss of type-1 Cu from the C-terminal domain would give NiR. By detailed structural comparison of the cupredoxin domains, Murphy et al confirm the hypothesis that all cupredoxin domains of cupredoxins, NiR, and MCO share a common ancestor [39]. Figure 1-8 shows that two molecules of the cupredoxin azurin can be superposed onto an NiR monomer. Three types of trimeric MCOs are proposed based on sequences found in sequence databanks [40,44]. The type [A] two-domain MCO contains type-1 Cu sites in both domains. Loss of the C-terminal type-1 Cu would yield the type [C] two domain enzyme. The type [C] MCO is similar to NiR in that it contains an N-terminal type-1 Cu. Thus, it is proposed that loss of two Cu atoms from the trinuclear site of type [C] also could have led to NiR enzymes. Loss of the N-terminal type-1 Cu from type [A] would  18  Figure 1-7 Proposed scheme for the evolution of the type-1 copper proteins, reprinted from Cell Mol Life Sci 62, Nakamura, K. and N. Go, Function and molecular evolution of multicopper blue proteins, p. 2050-66, Copyright 2005, with permission.  19  Figure 1-8 Cupredoxin domains in NiR. Two molecules of the cupredoxin azurin (PDB entry 4AZU) in orange and yellow ribbons are superposed onto the N (left) and C (right) termini of an NiR monomer (PDB entry 1SJM, gray). Spheres indicate the copper atoms.  20 yield the type [B] two domain enzyme. Type [B] two domain MCO likely underwent further domain duplication to yield the present day monomeric three domain MCOs that have been characterized extensively. A triplication of the two domain MCO is proposed to have led to the six domain ceruloplasmin and hephaestin. One ambiguity of the proposed model is the origin of NiR-type enzymes. There are two proposed routes, one proceeds directly after the first cupredoxin domain duplication. Loss of the C-terminal type-1 Cu (dashed line at top of Figure 1-7) would be required to resemble NiR. The other proposed route is from the type [C] two-domain MCO, in which two of the active site Cu atoms are lost to form a mononuclear type-2 active site. The most parsimonious evolutionary path is the latter because the type [A] two domain MCO could serve as the common ancestor to the type [B] and type [C] two domain MCO, as well as NiR and the monomeric MCOs. If the former origin of NiR is correct, type-1 Cu loss from the C-terminal domain would have to occur in two distinct events.  1.4 Directed evolution Directed evolution is a method whereby the mechanisms of natural selection are mimicked in the laboratory to produce variant biomolecules that are adapted to conditions defined by the experimenter. The methods of directed evolution include a wide variety of techniques and have been reviewed in great detail [45-55]. For the sake of simplicity, this section will focus on the basic steps of a typical protein directed evolution experiment that uses E. coli as an expression host.  21 There are two major components of a directed evolution experiment- variant library generation and screening/selection. In the case of protein directed evolution, there must be a physical linkage between the protein of interest and the gene sequence that codes for it. This requirement is fulfilled by expression of the protein in E. coli, in which a single cell (or colony resulting from plating of the bacterial cell) contains a mutated gene and produces the corresponding variant protein. By screening or selection, acquisition of the bacterial cell or colony producing the desired variant protein also results in the acquisition of the mutated gene, which can be amplified for recovery and sequenced for information on the mutant. 1.4.1 Mutagenic library creation 1.4.1.1 Error-prone PCR Mutagenic libraries have been generated by a number of methods. Most commonly used is error-prone PCR, which generates point mutations in the amplified sequence. Error-prone PCR mutagenesis can be accomplished by addition of manganese to a PCR mix using a polymerase that is deficient in 3’  5’ exonuclease activity for error  checking (such as Taq) [56]. There are also polymerases such as Mutazyme (Stratagene) that have been engineered to be error-prone. The resulting mutagenic PCR must be incorporated into an expression vector and transformed into E. coli for expression of the corresponding gene. 1.4.1.2 Site-specific randomization This technique requires the synthesis of oligonucleotides that contain randomized codons for specific residue positions in the protein. Any variation of site-directed mutagenesis techniques (e.g. Quickchange™ by Stratagene) can be used for the site-  22 specific randomization cloning reaction. Site-specific randomization is most useful when there is reasonable belief that specific residue positions have potential in adding desired function to the protein. Although similar variants may be obtained from random mutagenesis, site-specific randomization offers the advantage of smaller library size to achieve all the possible amino acid residues at the position of interest. Furthermore, the addition of deleterious variations elsewhere in the gene is avoided. 1.4.1.3 In vitro recombination Another method of mutagenic library creation is in vitro recombination, which mimics recombination events that are observed in nature, and allows the combination of advantageous mutations into one variant. In vitro recombination is effective with homologous genes with as little as 70% identity [57], or those genes obtained from prior directed evolution experiments employing random mutagenesis or site-specific randomization. The first method of in vitro recombination, called shuffling, was developed by Stemmer and involves the controlled fragmentation of variant genes by DNAse I followed by reassembly by fragment annealing and polymerase extension [58]. Since Stemmer’s invention, several other techniques have been developed, including StEP PCR and RACHITT (reviewed in [50]). Recombined product can be re-amplified and ligated into a vector for transformation into E. coli and subsequent screening/selection. 1.4.2 Screening or selection In any directed evolution experiment, experimental conditions must be controlled in a manner such that desired variants are identified from the mutagenic library. Two generalized methods are screening and selection. In a screen, the variant library is  23 propagated and the desired mutants are distinguished from the rest by an indicator. When enzymatic activity is the desired feature, catalytic turnover can be represented by color formation or luminescence of an E. coli colony or cell. In a selection, improved function of the biomolecule leads to the propagation of its gene. More specifically, in an E. coli-based selection, improved function of the variant protein confers growth of the E. coli (e.g. formation of colonies upon plating) under selective conditions. 1.4.2.1 Mutagenic library size limitations Screening and selection are characterized by the library size allowable for each strategy. In an E. coli-based screen, the size of the library that may be screened is limited by scalability of the screening resources (reagents, time, etc.). If improved protein function is tightly linked to growth of E. coli (i.e. strong selection), the efficiency of the cloning reaction rather than library size becomes limiting. Larger allowable libraries allow for a larger exploration of sequence space in the library (greater number of mutations per clone), which translates into greater chances of obtaining the desired function. For screening, an unsuitably high mutation rate (generally greater than two amino acid substitutions per variant in a random library) will incorporate too many deleterious mutations, resulting in a failed directed evolution experiment.  1.5 Thesis objectives Nitrite reductase has been studied for decades, with the first purification dating to the early 1980s [59]. The role of the Cu sites in catalysis in particular has been studied extensively. Less is known about NiR from an evolutionary standpoint. Further insight into the molecular evolution of the copper sites of NiR and related proteins will aid in our  24 understanding of evolutionary processes and structure-function relationships of type-1 Cu proteins. In particular, the evolutionary model of Nakamura and Go connecting NiR to three-domain MCOs is in need of experimental validation. Thus, the objectives of this thesis are to: a) formulate a high-throughput mutagenesis and screening method for nitrite reductase to enable directed evolution experiments; b) structurally and functionally characterize a two-domain MCO related to nitrite reductase to test the proposed model for evolution of type-1 Cu proteins; and c) evolve NiR into an MCO-like protein with a trinuclear active site in the laboratory, to determine structural requirements for trinuclear Cu binding to demonstrate an evolutionary path between MCO and NiR.  25 1.6 References  [1]  Williams, R.J.P. and Silva, J.J.R.F.d. (1996) The natural selection of the chemical elements : the environment and life's chemistry, Oxford University Press. Oxford.  [2]  Nersissian, A.M. and Shipp, E.L. (2002). Blue copper-binding domains. Adv Protein Chem 60, 271-340.  [3]  Lippard, S.J. and Berg, J.M. (1994) Principles of Bioinorganic Chemistry, University Science Books. Mill Valley.  [4]  Wijma, H.J., MacPherson, I., Farver, O., Tocheva, E.I., Pecht, I., Verbeet, M.P., Murphy, M.E. and Canters, G.W. (2007). Effect of the methionine ligand on the reorganization energy of the type-1 copper site of nitrite reductase. J Am Chem Soc 129, 519-25.  [5]  Farver, O., Hwang, H.J., Lu, Y. and Pecht, I. (2007). Reorganization Energy of the Cu(A) Center in Purple Azurin: Impact of the Mixed Valence-to-Trapped Valence State Transition. J Phys Chem B 111, 6690-6694.  [6]  Solomon, E.I., Chen, P., Metz, M., Lee, S.K. and Palmer, A.E. (2001). Oxygen Binding, Activation, and Reduction to Water by Copper Proteins. Angew Chem Int Ed Engl 40, 4570-4590.  [7]  Averill, B. (1996). Dissimilatory nitrite and nitric oxide reductases. Chem. Rev. 96, 2951-2964.  [8]  Boulanger, M.J. and Murphy, M.E. (2002). Crystal structure of the soluble domain of the major anaerobically induced outer membrane protein (AniA) from  26 pathogenic Neisseria: a new class of copper-containing nitrite reductases. J Mol Biol 315, 1111-27. [9]  Godden, J.W., Turley, S., Teller, D.C., Adman, E.T., Liu, M.Y., Payne, W.J. and LeGall, J. (1991). The 2.3 angstrom X-ray structure of nitrite reductase from Achromobacter cycloclastes. Science 253, 438-42.  [10]  Wherland, S., Farver, O. and Pecht, I. (2005). Intramolecular electron transfer in nitrite reductases. Chemphyschem 6, 805-12.  [11]  Yamaguchi, K., Kataoka, K., Kobayashi, M., Itoh, K., Fukui, A. and Suzuki, S. (2004). Characterization of two type 1 Cu sites of Hyphomicrobium denitrificans nitrite reductase: a new class of copper-containing nitrite reductases. Biochemistry 43, 14180-8.  [12]  Wijma, H.J., Jeuken, L.J., Verbeet, M.P., Armstrong, F.A. and Canters, G.W. (2006). A random-sequential mechanism for nitrite binding and active site reduction in copper-containing nitrite reductase. J Biol Chem 281, 16340-6.  [13]  Tocheva, E.I., Rosell, F.I., Mauk, A.G. and Murphy, M.E. (2004). Side-on copper-nitrosyl coordination by nitrite reductase. Science 304, 867-70.  [14]  Boulanger, M.J., Kukimoto, M., Nishiyama, M., Horinouchi, S. and Murphy, M.E. (2000). Catalytic roles for two water bridged residues (Asp-98 and His-255) in the active site of copper-containing nitrite reductase. J Biol Chem 275, 2395764.  [15]  Kataoka, K., Furusawa, H., Takagi, K., Yamaguchi, K. and Suzuki, S. (2000). Functional analysis of conserved aspartate and histidine residues located around  27 the type 2 copper site of copper-containing nitrite reductase. J Biochem (Tokyo) 127, 345-50. [16]  Boulanger, M.J. and Murphy, M.E. (2003). Directing the mode of nitrite binding to a copper-containing nitrite reductase from Alcaligenes faecalis S-6: characterization of an active site isoleucine. Protein Sci 12, 248-56.  [17]  Zhao, Y., Lukoyanov, D.A., Toropov, Y.V., Wu, K., Shapleigh, J.P. and Scholes, C.P. (2002). Catalytic function and local proton structure at the type 2 copper of nitrite reductase: the correlation of enzymatic pH dependence, conserved residues, and proton hyperfine structure. Biochemistry 41, 7464-74.  [18]  Wasbotten, I.H. and Ghosh, A. (2005). Modeling side-on NO coordination to type 2 copper in nitrite reductase: structures, energetics, and bonding. J Am Chem Soc 127, 15384-5.  [19]  Silaghi-Dumitrescu, R. (2006). Copper-containing nitrite reductase: a DFT study of nitrite and nitric oxide adducts. J Inorg Biochem 100, 396-402.  [20]  Wilmot, C.M., Hajdu, J., McPherson, M.J., Knowles, P.F. and Phillips, S.E. (1999). Visualization of dioxygen bound to copper during enzyme catalysis. Science 286, 1724-8.  [21]  Torres, J., Svistunenko, D., Karlsson, B., Cooper, C.E. and Wilson, M.T. (2002). Fast reduction of a copper center in laccase by nitric oxide and formation of a peroxide intermediate. J Am Chem Soc 124, 963-7.  [22]  Kakutani, T., Watanabe, H., Arima, K. and Beppu, T. (1981). A blue protein as an inactivating factor for nitrite reductase from Alcaligenes faecalis strain S-6. J Biochem (Tokyo) 89, 463-72.  28 [23]  Kakutani, T., Beppu, T. and Arima, K. (1981). Regulation of nitrite reductase in the denitrifying bacterium Alcaligenes faecalis S-6. Agric. Biol. Chem. 45, 2328.  [24]  Singh, S.K., Grass, G., Rensing, C. and Montfort, W.R. (2004). Cuprous oxidase activity of CueO from Escherichia coli. J Bacteriol 186, 7815-7.  [25]  Hullo, M.F., Moszer, I., Danchin, A. and Martin-Verstraete, I. (2001). CotA of Bacillus subtilis is a copper-dependent laccase. J Bacteriol 183, 5426-30.  [26]  Kosman, D.J. (2002). FET3P, ceruloplasmin, and the role of copper in iron metabolism. Adv Protein Chem 60, 221-69.  [27]  Riva, S. (2006). Laccases: blue enzymes for green chemistry. Trends Biotechnol 24, 219-26.  [28]  Shleev, S., Tkac, J., Christenson, A., Ruzgas, T., Yaropolov, A.I., Whittaker, J.W. and Gorton, L. (2005). Direct electron transfer between copper-containing proteins and electrodes. Biosens Bioelectron 20, 2517-54.  [29]  Kohzuma, T., Shidara, S. and Suzuki, S. (1994). Direct electrochemistry of nitrite reductase from Achromobacter cycloclastes IAM 1013. Bulletin of the Chemical Society of Japan 67, 138-43.  [30]  Wijma, H.J., Boulanger, M.J., Molon, A., Fittipaldi, M., Huber, M., Murphy, M.E.P., Verbeet, M.P. and Canters, G.W. (2003). Reconstitution of the type-1 active site of the H145G/A variants of nitrite reductase by ligand insertion. Biochemistry 42, 4075-4083.  [31]  Allendorf, M.D., Spira, D.J. and Solomon, E.I. (1985). Low-temperature magnetic circular dichroism studies of native laccase: spectroscopic evidence for exogenous  29 ligand bridging at a trinuclear copper active site. Proc Natl Acad Sci U S A 82, 3063-7. [32]  Reinhammar, B.R. and Vanngard, T.I. (1971). The electron-accepting sites in Rhus vernicifera laccase as studied by anaerobic oxidation-reduction titrations. Eur J Biochem 18, 463-8.  [33]  Quintanar, L., Yoon, J., Aznar, C.P., Palmer, A.E., Andersson, K.K., Britt, R.D. and Solomon, E.I. (2005). Spectroscopic and electronic structure studies of the trinuclear Cu cluster active site of the multicopper oxidase laccase: nature of its coordination unsaturation. J Am Chem Soc 127, 13832-45.  [34]  Shin, W., Sundaram, U.M., Cole, J.L., Zhang, H.H., Hedman, B., Hodgson, K.O. and Solomon, E.I. (1996). Chemical and spectroscopic definition of the peroxidelevel intermediate in the multicopper oxidases: relevance to the catalytic mechansim of dioxygen reduction to water. J. Am. Chem. Soc. 118, 3202-3215.  [35]  Sundaram, U.M., Zhang, H.H., Hedman, B., Hodgson, K.O. and Solomon, E.I. (1997). Spectroscopic investigation of peroxide binding to the trinuclear copper cluster site in laccase: Correlation with the peroxy-level intermediate and relvance to catalysis. J. Am. Chem. Soc. 119, 12525-12540.  [36]  Solomon, E.I., Sarangi, R., Woertink, J.S., Augustine, A.J., Yoon, J. and Ghosh, S. (2007). O2 and N2O activation by Bi-, Tri-, and tetranuclear Cu clusters in biology. Acc Chem Res 40, 581-91.  [37]  Yoon, J., Liboiron, B.D., Sarangi, R., Hodgson, K.O., Hedman, B. and Solomon, E.I. (2007). The two oxidized forms of the trinuclear Cu cluster in the multicopper  30 oxidases and mechanism for the decay of the native intermediate. Proc Natl Acad Sci U S A 104, 13609-14. [38]  Quintanar, L., Stoj, C., Wang, T.P., Kosman, D.J. and Solomon, E.I. (2005). Role of aspartate 94 in the decay of the peroxide intermediate in the multicopper oxidase Fet3p. Biochemistry 44, 6081-91.  [39]  Murphy, M.E., Lindley, P.F. and Adman, E.T. (1997). Structural comparison of cupredoxin domains: domain recycling to construct proteins with novel functions. Protein Sci 6, 761-70.  [40]  Nakamura, K. and Go, N. (2005). Function and molecular evolution of multicopper blue proteins. Cell Mol Life Sci 62, 2050-66.  [41]  Palmer, A.E., Quintanar, L., Severance, S., Wang, T.P., Kosman, D.J. and Solomon, E.I. (2002). Spectroscopic characterization and O2 reactivity of the trinuclear Cu cluster of mutants of the multicopper oxidase Fet3p. Biochemistry 41, 6438-48.  [42]  Endo, K., Hayashi, Y., Hibi, T., Hosono, K., Beppu, T. and Ueda, K. (2003). Enzymological characterization of EpoA, a laccase-like phenol oxidase produced by Streptomyces griseus. J Biochem (Tokyo) 133, 671-7.  [43]  Machczynski, M.C., Vijgenboom, E., Samyn, B. and Canters, G.W. (2004). Characterization of SLAC: a small laccase from Streptomyces coelicolor with unprecedented activity. Protein Sci 13, 2388-97.  [44]  Nakamura, K., Kawabata, T., Yura, K. and Go, N. (2003). Novel types of twodomain multi-copper oxidases: possible missing links in the evolution. FEBS Lett 553, 239-44.  31 [45]  Page, C.C., Moser, C.C., Chen, X. and Dutton, P.L. (1999). Natural engineering principles of electron tunnelling in biological oxidation-reduction. Nature 402, 47-52.  [46]  Zhao, H. (2007). Directed evolution of novel protein functions. Biotechnol Bioeng 98, 313-7.  [47]  Joyce, G.F. (2007). Forty years of in vitro evolution. Angew Chem Int Ed Engl 46, 6420-36.  [48]  Boersma, Y.L., Droge, M.J. and Quax, W.J. (2007). Selection strategies for improved biocatalysts. Febs J 274, 2181-95.  [49]  Woycechowsky, K.J., Vamvaca, K. and Hilvert, D. (2007). Novel enzymes through design and evolution. Adv Enzymol Relat Areas Mol Biol 75, 241-94, xiii.  [50]  Kaur, J. and Sharma, R. (2006). Directed evolution: an approach to engineer enzymes. Crit Rev Biotechnol 26, 165-99.  [51]  Griffiths, A.D. and Tawfik, D.S. (2006). Miniaturising the laboratory in emulsion droplets. Trends Biotechnol 24, 395-402.  [52]  Bloom, J.D., Meyer, M.M., Meinhold, P., Otey, C.R., MacMillan, D. and Arnold, F.H. (2005). Evolving strategies for enzyme engineering. Curr Opin Struct Biol 15, 447-52.  [53]  Otten, L.G. and Quax, W.J. (2005). Directed evolution: selecting today's biocatalysts. Biomol Eng 22, 1-9.  [54]  Jestin, J.L. and Kaminski, P.A. (2004). Directed enzyme evolution and selections for catalysis based on product formation. J Biotechnol 113, 85-103.  32 [55]  Lipovsek, D. and Pluckthun, A. (2004). In-vitro protein evolution by ribosome display and mRNA display. J Immunol Methods 290, 51-67.  [56]  Cadwell, R. and Joyce, G. (1992). Randomization of genes by PCR mutagenesis. PCR Methods Appl. 2, 28-33.  [57]  Joern, J.M., Meinhold, P. and Arnold, F.H. (2002). Analysis of shuffled gene libraries. J Mol Biol 316, 643-56.  [58]  Stemmer, W.P. (1994). Rapid evolution of a protein in vitro by DNA shuffling. Nature 370, 389-91.  [59]  Kakutani, T., Watanabe, H., Arima, K. and Beppu, T. (1981). Purification and properties of a copper-containing nitrite reductase from a denitrifying bacterium, Alcaligenes faecalis strain S-6. J. Biochem. (Tokyo) 89, 453-461.  33 Chapter Two: High-throughput screening studies of nitrite reductase 2.1 Introduction Under oxygen-limiting conditions, certain facultative anaerobic bacteria can respire using nitrate and nitrite as electron acceptors, producing the gases nitric oxide, nitrous oxide, and dinitrogen [1]. Copper-containing nitrite reductase (NiR) from Alcaligenes faecalis S-6 (AfNiR) catalyzes the single electron reduction of nitrite (NO2-) to nitric oxide (NO) during this process, called dissimilatory denitrification [2]. Energetically, respiratory denitrification is less efficient than oxygen respiration [3]. Furthermore, the switch from anaerobic to aerobic growth for A. faecalis leads to a rapid inactivation of nitrite reductase [4], a process that terminates the use of a less efficient terminal electron acceptor. NiR is a 110 kDa homotrimer, each monomer consisting of two Greek key βbarrels, with one type-1 and one type-2 copper site per monomer [5]. The type-1 copper is coordinated by His95, Cys136, His145, and Met150. Residues 136, 145, and 150 are part of a single “ligand loop” and His95 is located on an adjacent β-strand. The type-2 copper is located between monomers, and is coordinated by residues His100, His135, and His306, the last of which comes from an adjacent monomer. The type-1 and type-2 copper sites are connected via the peptide backbone between His135 and Cys136. In vivo, the small type-1 copper protein pseudoazurin is proposed to donate electrons to the type1 copper [6], followed by intramolecular electron transfer to the type-2 copper, the site of nitrite binding and reduction [5]. *A version of this chapter will be submitted for publication. MacPherson, I. S., Rosell, F. I., Scofield, M., Mauk, A. G., and M. E. P. Murphy. High-throughput screening studies of nitrite reductase.  34 NiR can catalyze other reactions. With the artificial reductants NADH and phenazine methosulfate, nitric oxide has been shown to be a substrate for NiR by the proposed overall reaction NO + NO2– + 4H+ + 3e-  N2O + 2H2O [7].  In addition, Kakutani et al reported that NiR is able to reduce oxygen to hydrogen peroxide and that prolonged production of H2O2 led to NiR inactivation [6]. Furthermore, the addition of catalase abolished this inactivation. Essential to the inactivation by H2O2 was the concurrent reduction of AfNir by pseudoazurin; therefore, Kakutani et al proposed that the inactivation of the enzyme is by hydroxyl radical, generated by further reduction of H2O2 at the type-2 copper [6]. Multicopper oxidases (MCOs) are homologous to NiR and contain type-1, type-2, and type-3 copper sites which are employed to reduce O2 to water. MCOs such as laccase and ascorbate oxidase have reduction potentials among the highest of known type-1 copper proteins, in the range of 330-780 mV [8]. These proteins catalyze the oxidation of various substrates at the Type-1 copper site, and the concomitant transfer of electrons to an active site trinuclear copper cluster [9]. Interestingly, studies of type-1 copper depleted laccase have shown the production of a bridged peroxide intermediate [10], paralleling what is seen with NiR. The nitrite reductase function of NiR has been the subject of numerous sitedirected mutagenesis studies targeting residues surrounding the two copper sites [5,1115]. These studies have been limited in scope due a lack of an effective functional screen. Here, a high throughput screen is developed that employs a chemical reductant of NiR to enable the study of oxygen reduction. This screening methodology coupled with a novel  35 process of mutational library generation is applicable to NiR engineering as a folding reporter or a tool to alter substrate specificity of the enzyme. Furthermore, these highthroughput tools could be used for other oxidoreductases such as MCOs.  2.2 Materials and methods 2.2.1 Library construction Variant libraries were generated in a pET28a construct (pAfNiR28a) containing the coding region for the soluble fragment of NiR [11]. Random PCR mutagenesis was performed by the method of Cadwell et al [16], using primers NIRFOR (5’GCAACTGCGGCAGAAATAGCA) and NIRREV (5’CGTGCCAGATGGTGCGA), which correspond to the first 21 and last 17 nucleotides of the coding sequence for soluble NiR. Reaction conditions were 10 mM Tris-Cl pH 8.3, 50 mM KCl, 7 mM MgCl2, 0.15 mM MnCl2, 200 µM dNTPs, 0.4 µM of each primer, 30 ng template, and 5 U Taq polymerase in 100 µl total volume. The PCR product was cloned into pET28a using a variation of the method described by Miyazaki et al [17]. Briefly, pAfNiR28a was used as template and the mutagenic PCR product used as a megaprimer for whole plasmid synthesis. The major variation was cutting of the template pAfNiR28a plasmid at a unique EcoR1 site within the NiR gene. Reactions contained 1X Platinum Pfx buffer (Invitrogen), 1 mM MgSO4, 300 µM dNTPs, 75 ng freshly cut template, 700 ng mutagenic PCR product, 4% DMSO, and 2.5 U Platinum Pfx polymerase (Invitrogen). Reactions were cycled between 95 °C (30 sec) and 68 °C (7 min) for 12-24 cycles (Figure 2-1). The reaction product was dialyzed, electroporated into E. coli  36  1 kb 12 ladder  15  18  21  24  nicked circular product linear product  megaprimer (~ 1 kb)  Figure 2-1 Typical megaprimer cloning reactions from various cycle numbers (12-24). Products were electrophoresed on 1% agarose gel. The band proposed to be nicked circular product corresponds to successfully cloned variant. Note that the megaprimer is consumed as the cycle number increases. The appearance of the proposed nicked circular product corresponds to colony forming units upon transformation.  37 HMS174(de3), and plated onto 2YT agar plates containing kanamycin (25 µg/ml) and IPTG (66 µM). Shuffling of variants was performed by the Stemmer method [18]. Briefly, the five fastest color-developing variants obtained from screening were subjected to controlled digestion by DNAaseI. Approximately 50 base pair fragments were reassembled by 30X thermal cycling of 94 °C (30 sec), 40 °C (30 sec), and 72 °C (4 min) extensions. A final PCR with NIRFOR and NIRREV amplified the shuffled library. Shuffled variants were ligated into pET28a using the above described modification of the Miyazaki method. 2.2.2 Screening method The colony lysis protocol was adapted from a method described by KadonoOkuda et al [19]. Plates were incubated for 15 hours at 33-35 °C. The resulting colonies were lifted onto Biodyne-A 0.45 µm nylon membranes (Pall) and placed, colony side up, onto Whatman filter paper saturated with a lysis solution (10 mM Tris pH 7, 2% SDS, 0.3% Tween-20, and 50 µM CuSO4) and incubated at 50 °C for 30 min. The membrane was then washed gently in 10 mM Tris pH 7.5, 100 mM NaCl for 5-10 minutes. The membranes were blotted dry and submerged in screening reagent (0.76 mM 3,3’diaminobenzidine tetrahydrochloride (Sigma) and 0.5 µM horseradish peroxidase (Sigma) in 100 mM sodium phosphate pH 7.4). Red spots, representing colonies expressing active NiR, were identified on the membrane (Figure 2-2). The fastest appearing spots were selected and mapped to the original plate. The original kanamycin/IPTG plates were incubated at 30 °C to allow colonies to re-grow and selected colonies were picked, grown overnight for plasmid isolation and sequencing for  38  Figure 2-2 Picture of a screened membrane. About 2000 clones are represented, half are not apparent due to loss of activity.  39 mutational analysis. 2.2.3 Protein expression and purification NiR variants were expressed in E. coli BL21(de3) from pET28a vectors as described previously [14]. One litre cultures were inoculated with 3 ml overnight culture and grown at 30 °C to an O.D.600 of 1.0, for a further 30 minutes at 25 °C, then induced with 0.5 mM IPTG and grown for 16-18 hours at 25 °C. The cells were pelleted and resuspended in nickel column binding buffer (20 mM sodium phosphate pH 7.8, 500 mM NaCl) supplemented with 1 mM CuSO4 and lysed with an Emulsiflex C-5 homogenizer (Avestin). The soluble fraction was applied to a nickel metal chelate sepharose column (GE Healthcare) for purification, resulting in greater than 95% purity. Typical yields are 250-350 mg NiR per litre culture. All variants expressed at a minimum of 200 mg/liter culture and were sufficiently stable to allow concentration to greater than 30 mg/ml with minimal precipitation (<1%). Pseudoazurin was expressed and isolated as described previously [20]. Briefly, E. coli BL21(de3) containing the expression construct ppAz24c [21] was grown at 30° to an O.D.600 of 2.0, followed by a further 30 minutes at 25 °C, then induced with 0.5 mM IPTG, followed by 12-14 hours further culturing. The cells were pelleted and resuspended in 20 mM sodium phosphate pH 6.3 supplemented with 10 mM CuSO4, lysed with an Emulsiflex C-5 homogenizer (Avestin). The soluble fraction was applied to CM-sepharose (GE Healthcare) for cation exchange chromatography. Typical yields were 100-150 mg of pseudoazurin per litre culture.  40 2.2.4 Activity assays Pseudoazurin-based assays for nitrite reduction were performed as described by Wijma et al. [20]. Pseudoazurin was reduced with excess ascorbate and then bufferexchanged against N2-saturated 100 mM MES-HEPES, pH 7.0. Reactions containing 315 µM reduced pseudoazurin and 2.5 mM NaNO2 in MES-HEPES pH 7.0 were started by the addition of NiR to the final concentration of 450 pM. The absorbance at 593 nm, corresponding to the amount of oxidized pseudoazurin (ε = 2900 M-1cm-1), was monitored by a Cary 50 Bio UV-Vis spectrophotometer in a cell maintained at 25 °C with circulating water. Pseudoazurin-based assays for oxygen reduction were performed under similar conditions as for nitrite reduction, but NO2– was omitted. NiR was added to a final concentration of 9-90 nM to initiate the assays. Catalase (from bovine liver) was added to a final concentration of 1 µM. o-Dianisidine (3,3′-dimethoxybenzidine dihydrochloride), a structural analogue of 3,3’-diaminobenzidine with a water-soluble oxidation product, also was used as an electron donor to NiR. Oxygen reduction assays were performed in 100 mM MESHEPES pH 7.0 containing 0.95 mM o-dianisidine and 1 µM catalase. Nitrite reduction assays contained 0.95 mM o-dianisidine and 2.5 mM sodium nitrite. Oxidation of odianisidine was monitored by the increase in absorbance at 460 nm (ε = 11300 M-1cm-1). 2.2.5 Electrochemistry Reduction potentials were measured for the variants by a method described by Kohzuma et al using apopseudoazurin as an electrode modifier [22]. Apopseudoazurin  41 was prepared by dialysis against 6 mM KCN overnight. This was accompanied by a loss of blue color from the sample, indicating a loss of type-1 Cu. The apopseudoazurin (30 µm) and variant NiR (10-100 µM) were added to a cell containing an edge-oriented pyrolytic graphite working electrode that had been polished with alumina (0.3 µM). The reference electrode was saturated calomel and the counter electrode was platinum wire. An Autolab potentiostat (Eco Chemie) was used to apply a potential of -300 mV to 300 mV to the graphite electrode. 2.2.6 Electronic absorption spectroscopy A Cary 50 spectrophotometer was used for measurements of protein sample in 20 mM MOPS pH 7.0. NiRs were centrifuged for 10 min to reduce light scattering. Scans were obtained for wavelength range 350-850 nm. 2.2.7 Crystal structures Crystals were grown in hanging drop format. Mother liquor consisted of 100 mM sodium acetate pH 4.5, 8-12% PEG 6000. Crystals were transferred to mother liquor supplemented with 30% glycerol as a cryoprotectant and immersed in liquid nitrogen. Home source x-rays were used for data collection of ISM42 (M62L). Sh1 and Sh10 datasets were collected at the Stanford Synchrotron Radiation Laboratory (SSRL). The crystal structures were solved for the double variants Sh1 (M94T/F312C), Sh10 (M150L/F312C), and ISM46 (M62L). For refinement, the starting point was the 1.4 Å resolution structure of nitrite-bound native NiR (PDB code 1SJM) after removal of nitrite and solvent molecules. Structures were refined with Refmac [23] from the CCP4 package [24]. Model building was performed with Xfit [24]. A summary of data collection and refinement statistics can be found on Table 2-1.  42  Table 2-1 Data collection and refinement statistics Crystal  M62L  M94T/F312C  M150L/F312C  P212121  P212121  R3  a = 61.13  a = 61.30  a = 126.72  b = 102.22  b = 102.08  b = 126.72  c = 146.15  c = 146.34  c = 65.06  84.5- 2.1  84.5 - 1.6  63.4 - 1.5  0.100 (0.356)a  0.068 (0.232)  0.084 (0.234)  9.8 (2.6)  21.0 (3.9)  21.3 (4.7)  Completeness (%)  97.3 (97.6)  89.4 (71.8)  99.6 (99.1)  Unique reflections  50095  103196  54872  Working R-factor  0.161  0.180  0.182  Free R-factor  0.217  0.227  0.209  Rmsd bond length (Å)  0.011  0.012  0.012  Overall B-factor (Å2)c  26.2  23.2  23.4  Water molecules  1122  1349  345  Space group  Cell dimensions (Å)  Resolution (Å) R-merge {I}/{σ(I)}b  a  Values in parenthesis are for the highest resolution shell. b {I}/{σ(I)} is the average  intensity divided by the average estimated error in intensity. c B-factors are an average from all three monomers where applicable.  43 2.3 Results 2.3.1 Improved cloning efficiency In this study, the library construction method typically yielded 1000-2000 colonies/µl reaction. Thus from a 50 µl reaction, up to 100,000 clones could be obtained. Since only ~20,000 variants were screened in this study, cloning efficiency was not a limiting factor. Based on sequencing of five randomly selected variants, a mutation rate of 3-4 substitutions per clone was obtained using a MnCl2 concentration of 0.15 mM in the mutagenic PCR reactions. The mutation rate defined by limited sequencing is consistent with the approximately 50% rate of inactivation observed in the screen. Cloning efficiency is often a limiting factor in high-throughput colony screening. Low efficiency limits the number of variants that can be screened in one experiment. This limitation is especially prevalent with notoriously difficult cloning vectors such as pET. For this reason, the megaprimer method by Miyazaki et al (2003) was used in this study with some modifications. Particularly, cleaving of the template plasmid DNA within the NiR gene significantly improved the polymerase-based cloning reaction (Figure 2-3). Rationale for this modification was the assumption that megaprimer annealing is rate limiting when the target template is coiled-circular DNA. Denaturation at 95 °C undoubtedly melts the strands of plasmid DNA, however it does not separate them due to their closed circular nature. Denaturation of singly cut template plasmid results in two separate ssDNA strands without restricted access to their termini (Figure 2-3). Thus, annealing of megaprimers is more efficient. As the megaprimers become consumed in the polymerization reaction, fewer will be available to bind both template and product, resulting in the circularization of the major product and formation of a viable plasmid.  44  EcoR1 cut site  5’  3’  5’  3’  Major product in brackets  NiR gene 5’  Strand denaturation, megaprimer annealing, polymerase extension  Cut with EcoR1 3’  5’ 3’  5’  +  3’  3’  5’  Major product annealing, excess megaprimer annealing (inhibits circularization)  5’  3’  3’  5’  Megaprimers consumed, promoting circularization  Mutagenic PCR product (megaprimers) 5’ 3’  3’ 5’  Figure 2-3 Schematic representation of the megaprimer-based cloning method. Green bands correspond to the NiR gene in the original pET28a plasmid. Orange bands correspond to the mutagenic PCR product (megaprimers). The plasmid is represented by black lines and arrows (not drawn to scale).  45 2.3.2 High-throughput screening NiR variants were obtained by high-throughput screening, based on their ability to catalyze the oxygen dependent oxidation of 3,3’-diaminobenzidine (DAB). A central component of the high-throughput screening was the colony lift and lysis method, which exposed the cytoplasmically expressed NiRs to Cu (free Cu ion is significantly limited in the cytoplasm [25]) and allowed for NiR interaction with DAB. DAB oxidation resulted in the deposition of red color at lysed colonies, indicating turnover by the variant NiRs. Plasmid DNA from the 12 initial variants selected from the screen was isolated and re-transformed, and a second round of colony screening yielded five single site variants. These variants were shuffled and screened yielding two double variants. A list of variants obtained in this study is found in Table 2-2. 2.3.3 ο-Dianisidine and pseudoazurin oxidation  ο-Dianisidine is structurally similar to DAB; however the oxidized form is also water soluble, enabling spectrophotometric determination of oxidation rates. οDianisidine can serve as the reductant for both oxygen and nitrite reduction by NiR (Table 2-2). With the exception of M94T, all variants obtained by screening were faster at oxidizing o-dianisidine than wt NiR (0.0042 s-1) when oxygen is the electron acceptor. The individual substitutions F312C NiR (2.7X wt) and M150L (3.4X wt) showed an additive effect in the shuffled variant F312C/M150L NiR (5.5X wt). The site-directed variant D98N NiR oxidized o-dianisidine at rates similar to wt NiR (0.9X wt). In the presence of 2.5 mM sodium nitrite, different rates were obtained for wt NiR and variants (Table 2-2). Firstly, wt NiR reduced nitrite at a significantly higher rate of 0.15 s-1 using o-dianisidine, roughly 35 times faster than with oxygen. All variants  46 Table 2-2 Reduction potentials and kcat values (relative to wt) of variant and wt NiR Variant  Oxygen  Nitrite  Oxygen  Nitrite  E1/2  reduction (o-  reduction (o-  reduction  reduction  (mV)  dianisidine)  dianisidine)  (pseudoazurin)  (pseudoazurin)  Wt  1a  1b  1c  1d  249.9  F312C  2.7 ± 0.04  0.54 ± 0.05  0.23 ± 0.02  0.22 ± 0.02  250.3  M94T  0.73 ± 0.03  0.92 ± 0.14  1.1 ± 0.07  0.59 ± 0.04  236.0  M62L  1.3 ± 0.02  3.1 ± 0.13  0.81 ± 0.04  2.1 ± 0.00  271.5  M150L  3.4 ± 0.69  8.5 ± 0.60  0.38 ± 0.03  0.30 ± 0.01  275.0  N96S  1.3 ± 0.11  5.9 ± 0.10  0.15 ± 0.01  0.41 ± 0.01  260.9  F312C/  2.7 ± 0.54  0.64 ± 0.06  0.20 ± 0.03  0.62 ± 0.07  215.1  5.5 ± 0.30  6.8 ± 0.33  0.12 ± 0.01  0.22 ± 0.01  265.3  0.9 ± 0.25  1.3 ± 0.13  0.05 ± 0.03  0.01 ± 0.00  e  M94T F312C/ M150L D98N  a. Wt rate of oxygen reduction using o-dianisidine as the electron donor: 0.0042 ± 0.0003 s-1 b. Wt rate of nitrite reduction using o-dianisidine as electron donor: 0.15 ± 0.01 s-1 c. Wt rate of oxygen reduction using pseudoazurin as the electron donor: 14.0 ± 0.3 s-1 d. Wt rate of nitrite reduction using pseudoazurin as the electron donor: 401 ± 23 s-1 e. Reduction potential not measured in this study but presumed to be similar to wt.  47 obtained by screening were faster at nitrite reduction than oxygen reduction using odianisidine, however F312C NiR showed lower relative activity (0.54X wt) for nitrite reduction, compared to 2.7X wt for oxygen reduction. The variant M150L NiR showed the largest increase in reducing nitrite with o-dianisidine (8.5X). However, when combined with F312C NiR, the double variant rate is 6.8X wt, consistent with the negative impact on nitrite reduction observed for F312C substitution. When oxygen reduction with pseudoazurin as a reductant was measured for wt NiR, the specific activity was 14 s-1, ~3000 times faster than with o-dianisidine as a reductant (Table 2-2). The specific activity for D98N NiR is 20X lower than for wt NiR. Interestingly, with the exception of M94T, all of the variants obtained from highthroughput screening displayed lower relative activity values at reducing oxygen with pseudoazurin. Most notably, M150L reduced oxygen with pseudoazurin at 38% of the wt rate. Nitrite reduction with reduced pseudoazurin showed wt NiR to be most proficient with the exception of M62L NiR. This variant showed activity levels 2.1 times that of wt NiR (Table 2-2). Again, F312C NiR showed diminished pseudoazurin/nitrite activity (22% of the wt activity). M150L NiR also showed decreased activity (30% of wt), and the double variant M150L/F312C NiR has 22% activity compared to wt NiR. 2.3.4 Type-1 Cu site reduction potentials and electronic spectra Reduction potentials of the type-1 copper sites of wt and variant NiRs were measured with cyclic voltammetry. The potential of wt NiR was 249.9 mV, in good agreement with previously published values of 240 mV [22] and 260 mV [14]. Reduction  48 potentials for the variants ranged from 215.1 mV for F312C/M94T to 275 mV for M150L. Spectra of the variants (as isolated) varied significantly compared to wt NiR. M150L NiR and the double variant, M150L/F312C are characterized by a large increase in the absorption maxima at around 600 nm, and an almost completely diminished absorption at 460 nm, when compared to wt NiR (Figure 2-4). M62L NiR also had a distinctly different absorption spectrum, with a maximum at 464 nm, compared to 458 nm for wt NiR. Also, the A464/A589 ratio for this variant was 1.7, compared to 1.3 for wt NiR. This change in absorbance gives M62L a grass green appearance, compared to the olive color of wt NiR. 2.3.5 Crystal structures The residue positions of all five single-variants are mapped to the wt NiR structure to show their relative positions (Figure 2-5). Crystal structures were solved for three variants- M62L, F312C/M94T and F312C/M150L to learn about the structural changes observed in the variants with improved o-dianisidine oxidation (Figure 2-6). In wt NiR, Met62 is located in close proximity to two type-1 copper site ligands, His145 and Met150. Two main chain changes occur in the M62L NiR structure. Specifically, the imidazole moiety of His145 is rotated 20° about χ2 such that the coordinating Nδ1 atom is displaced in the direction of Leu62 by ~0.2-0.3 Å (over the three monomers). Another change in the M62L NiR structure is at Trp144. The tryptophan side chain rotates about χ2 such that the indole moiety moves closer to the type-1 copper.  49  3.5  Extinction coefficient (mM -1cm -1)  3  2.5  2  1.5  1  0.5  0 350  450  550  650  750  850  w avelength (nm )  Figure 2-4 Electronic spectra of wt and variant NiRs. Wt; solid black. M62L; orange. F312C; pink. M94T; light blue. N96S; purple. M150L; red. M150L/F312C; green. M94T/F312C; dark blue.  50  Type-2 Cu site  N96S  M94T  M150L  M62L  F312C  Type-1 Cu site  Figure 2-5 Residue positions of positively screened variants mapped to the wt NiR structure (PDB code 1SJM). Labels at magenta-colored residues refer to the substitution obtained at that residue position. In addition to Met150, type-1 Cu site His and Cys ligands are shown as green sticks. Type-2 Cu His ligands are also shown as green sticks. Copper atoms are spheres.  51  A His145 Trp144 Met150  B Asn115  Resi 62  C  Glu113  His95  His145  Asn96  Resi 94  Cys136 Type-2  Type-1  Resi  Resi150  Figure 2-6 Crystal structures of variants. A) Stereo representation of the M62L NiR crystal structure at the site of the substitution. Wt NiR (PDB code 1SJM, green) and M62L (slate) are superposed. Cu atoms are spheres. B) Superposition of the wt NiR (green) with the double variant M94T/F312C (salmon) crystal structure. C) Superposition of the wt NiR (green) with double variant M150L/F312C (light blue) crystal structures. Oxygen atoms are colored red, and nitrogen atoms are colored dark blue. “Resi” refers to a residue position that is mutated.  52 The crystal structures of F312C/M94T (Figure 2-6) and F312C/M150L (Figure 2-6) NiR have the substitution F312C in common. Both variants show an oxidized cysteine (sulfinoalanine) at position 312. No other significant structural changes are attributed to this substitution. In wt NiR Met94 is located at the molecular surface close to the type-1 copper site. Replacement of this residue with threonine results in a new hydrogen bond (2.7 Å) formed between the threonine hydroxyl and Asn115 side chain carbonyl. To accomplish this, the Asn115 amide moiety rotates ~180 ° about χ2. Additionally, the conformation of Glu113 changes to occupy space taken by Met94 in wt NiR, forming a new hydrogen bond (2.9 Å) between the Glu113 carboxylate and the side chain amide nitrogen of Asn96. Met150 is the axial ligand to the type-1 copper site in wt NiR. Replacement with leucine results in a shift of the type-1 copper ~0.6 Å into the plane of the other three ligands, His95, Cys136, and His145 giving rise to a trigonal planar geometry (Figure 26).  2.4 Discussion 2.4.1 Library generation and colony screen Mutagenic library generation and high-throughput screening enabled isolation of variants with altered specificity for both the electron donor and acceptor. The use of DAB as the electron donor and as a screening agent fulfills the requirements for a colorimetric colony screen. In particular, DAB oxidation results in the deposition of red, waterinsoluble end product at lifted colonies with NiR-catalyzed oxygen reducing activity. The use of strong chemical reductants such as ascorbate and dithionite have a negative impact  53 on the oxidation of DAB as well as other chromogenic peroxidase substrates (data not shown); therefore, the screen is limited in terms of electron donors. However, reduced pseudoazurin has been shown to greatly increase the rate of red color formation in this colony lift screen, using DAB and horseradish peroxidase as H2O2 indicators (data not shown). Similar results are expected for protein electron donors such as reduced cytochrome c. Addition of horseradish peroxidase to the screening reagent ensured removal of H2O2, which inactivates NiR, and also enhanced the DAB oxidation leading to further deposition of red color in catalytically active colonies. The library generation and screening method can be used as a folding screen, provided that mutations do not influence the rate of DAB oxidation. Current research is aimed at randomizing specific regions of NiR and using the high-throughput method to screen for variants that allow proper folding and expression. The colony lift and lysis allows cytoplasmically expressed protein to bind copper. In addition to NiR, the screen can also be used for non-native multicopper oxidases expressed cytoplasmically in E. coli. Color formation by oxidation of DAB is expected to be a robust process for many MCOs due to their efficient oxidation of o-dianisidine [26-29]. 2.4.2 Oxygen reduction in wt NiR Facultative denitrifying bacteria transition between using oxygen and nitrate/nitrite as electron acceptors in oxidative phosphorylation [30]. The heme cd1 containing nitrite reductases are able to reduce oxygen to water and are oxygen insensitive [31]. In contrast, copper-containing NiRs are characterized as being oxygen sensitive under reducing conditions due to the production of hydrogen peroxide and possibly more reactive hydroxyl radicals [6]. Notably, catalase is protective of NiR  54 during turnover in the presence of oxygen. A significant oxygen reduction rate of 14 s-1 was determined for wt NiR using reduced pseudoazurin as the electron donor (Table 2-2). The nitrite reduction activity of wt NiR under the same conditions is 30X greater. If the coupling of electrons to H2O2 production is high, then the expected rate of H2O2 production is 7 s-1, as the reduction is a two-electron process. The presence of excess catalase results in the conversion H2O2 to water and oxygen and thus oxygen is partially replenished in the assay. Asp98 is an absolutely conserved residue that forms a hydrogen bond to nitrite and nitric oxide bound to the type-2 copper site [32]. According to the proposed mechanism of nitrite reduction [32], deprotonated Asp98 forms a hydrogen bond with HNO2, poising it for reduction and addition of a second proton to yield a copper nitrosyl intermediate and the release of a water molecule. The D98N substitution decreased the specific activity of NiR by 100X as compared to wt [11]. Additionally, crystal structures of D98N NiR show a loss of hydrogen bonding of this residue to nitrite or water (in the nitrite-free structure) [12], suggesting that this bonding is a key requirement for catalysis. To gain insight into the mechanism of oxygen reduction, D98N NiR was also assayed and found to have 20X less activity than wt NiR (Table 2-2). Strongly diminished oxygen reduction by D98N NiR suggests a role for the carboxylate group of Asp98 in this reaction as well. As suggested in the proposed nitrite reduction mechanism, Asp98 could serve to stabilize the singly protonated, two-electron reduced oxygen intermediate, Cu(II)-hydroperoxo. A second proton acquisition would result in hydrogen peroxide formation. Secondly, Asp98 could relay protons to reduced oxygen to form hydrogen peroxide.  55 2.4.3 Oxygen and nitrite reduction with DAB and o-dianisidine Wt NiR oxidizes o-dianisidine in the presence of oxygen at a rate of 0.0042 s-1, roughly 3 x 10-4X of the rate with reduced pseudoazurin. Variants obtained from high throughput screening improved the rate of o-dianisidine oxidation with oxygen as the electron acceptor up to 5.5 fold. The greatest increase in activity was for the double variant M150L/F312C. Met150 is the axial ligand for the type-1 copper site in NiR and replacement with leucine is expected to have a significant impact on the reduction potential of the type-1 copper site. Indeed, the reduction potential of M150L/F312C NiR was determined to be 265.3 mV, 15 mV greater than wt. A substitution of the axial methionine to leucine in the homologous NiR from Alcaligenes xylosoxidans was made by site-directed mutagenesis [33]. The reduction potential for this variant increased by 96 mV to 336 mV as determined by a titration method [33]. The discrepancy in reduction potentials between A. faecalis and A. xylosoxidans variant NiRs may be explained by the different methods used for redox potential determination. In A. xylosoxidans NiR the rate of nitrite reduction was found to be 1.7X that of the wt enzyme using azurin as the reductant. This increase is in sharp contrast to the ~3 fold loss of activity observed for A. faecalis NiR using pseudoazurin (Table 2-2). Note that absolute rates for A. xylosoxidans NiR were not provided and may be significantly lower than those measured for A. faecalis NiR. Of particular interest are the reduction potentials of the different NiR copper sites and substrates. By detailed electrochemical methods, the reduction potential of the type-1 copper of AfNiR has been determined to be 260 mV [14], a relatively low value compared to other type-1 copper proteins, but in good agreement with the standard  56 reduction potential of the half reaction NO2- + 2H+ + e-  NO + H2O (202 mV) and in  the same range as the cupredoxin pseudoazurin (270 mV) from which NiR receives electrons [20,34]. The reduction potential of the type-2 copper site has proven more difficult to measure. Studies on a homologous nitrite reductase from Rhodobacter sphaeroides have suggested that the reduction potential of the active site without nitrite bound is significantly lower than that of the type-1 site, by at least 40 mV [35]. It was suggested that the binding of nitrite to the type-2 copper could raise the reduction potential to favor electron transfer from the type-1 copper [35]. The standard potential for the reduction of oxygen to hydrogen peroxide is 280 mV [36], significantly higher than that for nitrite, but close to that measured for the NiR type-1 Cu. Leucine is preferentially found in place of the axial methionine in many laccases, which catalyze the oxidation of phenolic substrates as well as benzidine-based molecules such as DAB and o-dianisidine. These enzymes are homologous to NiR but instead of the mononuclear type-2 site possess a trinuclear copper site that reduces oxygen to water [9,37]. Several studies of type-1 Cu sites have shown that replacement of the axial methionine with leucine raises the reduction potential [33,38,39]. In NiR, the M150L substitution gives an improved catalytic rate with o-dianisidine but has detrimental effects with pseudoazurin as the electron donor. The Met150 residue is buried and the M150L substitution has minimal changes to the surface characteristics of the type-1 Cu site, suggesting that the substitution is not likely to affect the affinity for the chemical reductants. Notably, DAB and o-dianisidine have higher reduction potentials (480 mV and 340 mV, respectively) than the type-1 Cu of wt NiR [40]. The large decrease in overall rate observed with these reductants by wt NiR suggests that reducing the type-1  57 site is likely rate limiting. The increase in reduction potential as a result of M150L substitution likely increases the rate at which the type-1 site is oxidized by these reductants. Conversely, the same increase in reduction potential could be expected to slow electron transfer to the type-2 Cu, due to a decrease in positive driving energy between the Cu sites. This model explains the lower rates observed for oxygen and nitrite reduction with pseudoazurin. M150L NiR still catalyzes oxygen and nitrite reduction with pseudoazurin at high rates (approximately 4 sec-1 and 120 sec-1, respectively) compared to o-dianisine (0.014 sec-1 and 1.24 sec-1, respectively), which suggests that intramolecular electron transfer between type-1 and type-2 Cu sites is much less ratelimiting than DAB and o-dianisidine oxidation. Higher reduction potentials at the type-1 Cu site in multicopper oxidases are thermodyamically favorable because of the increased reduction potential of the trinuclear active site (390-780 mV) [41,42]. Phe312 is located at the surface of the deep cleft which houses the active site of NiR. This hydrophobic residue is also surrounded by several other hydrophobic residues, Val142, Val304, and Leu308. This hydrophobic patch has been proposed to be a route of nitric oxide egress during nitrite reduction in vivo [43]. Substitution with cysteine greatly increases the size of the pocket, as well as the hydrophilicity. The positive impact on odianisidine oxidation (2.7 times wt) with oxygen reduction and negative impact (0.54 X wt) with nitrite reduction suggests a specific role for the F312C substitution in oxygen reduction using o-dianisidine. This role could involve redox cycling of the cysteine, in which hydrogen peroxide reacts with the reduced form to generate the oxidized form, and then o-dianisidine re-reduces it. Another possible explanation is an opening of the  58 hydrophobic pocket for release of the more hydrophilic H2O2, although this does not explain why F312C and double variants are less effective than wt at oxygen reduction with reduced pseudoazurin. Opening of the hydrophobic pocket could aid in o-dianisidine interaction directly with an oxygen-bound intermediate at the active site. More specifically, faster sequential reduction of oxygen with o-dianisidine could be facilitated from two directions, one electron via the traditional type-1 site/type-2 site relay and the other via the F312C pocket.  59 2.5 References  [1]  Averill, B. (1996). Dissimilatory nitrite and nitric oxide reductases. Chem. Rev. 96, 2951-2964.  [2]  Kakutani, T., Watanabe, H., Arima, K. and Beppu, T. (1981). Purification and properties of a copper-containing nitrite reductase from a denitrifying bacterium, Alcaligenes faecalis strain S-6. J. Biochem. (Tokyo) 89, 453-461.  [3]  Tran, Q.H. and Unden, G. (1998). Changes in the proton potential and the cellular energetics of Escherichia coli during growth by aerobic and anaerobic respiration or by fermentation. Eur J Biochem 251, 538-43.  [4]  Kakutani, T., Beppu, T. and Arima, K. (1981). Regulation of nitrite reductase in the denitrifying bacterium Alcaligenes faecalis S-6. Agric. Biol. Chem. 45, 23-28.  [5]  Kukimoto, M., Nishiyama, M., Murphy, M.E., Turley, S., Adman, E., Horinouchi, S. and Beppu, T. (1994). X-ray structure and site-directed mutagenesis of a nitrite reductase from Alcaligenes faecalis S-6: roles of two copper atoms in nitrite reduction. Biochemistry 33, 5246-5252.  [6]  Kakutani, T., Watanabe, H., Arima, K. and Beppu, T. (1981). A blue protein as an inactivating factor for nitrite reductase from Alcaligenes faecalis strain S-6. J. Biochem. (Tokyo) 89, 463-472.  [7]  Jackson, M., Tiedje, J. and Averill, B. (1991). Evidence for a NO-rebound mechanism for production of N2O from nitrite by the copper-containing nitrite reductase from Achromobacter cyclosclastes. FEBS Letters 291, 41-4.  60 [8]  Shleev, S., Tkac, J., Christenson, A., Ruzgas, T., Yaropolov, A.I., Whittaker, J.W. and Gorton, L. (2005). Direct electron transfer between copper-containing proteins and electrodes. Biosens Bioelectron 20, 2517-54.  [9]  Solomon, E.I., Chen, P., Metz, M., Lee, S.K. and Palmer, A.E. (2001). Oxygen binding, activation, and reduction to water by copper proteins. Angew Chem Int Ed Engl 40, 4570-4590.  [10]  Shin, W., Sundaram, U.M., Cole, J.L., Zhang, H.H., Hedman, B., Hodgson, K.O. and Solomon, E.I. (1996). Chemical and Spectroscopic Definition of the Peroxide-Level Intermediate in the Multicopper Oxidases: Relevance to the Catalytic Mechanism of Dioxygen Reduction to Water. Journal of the American Chemical Society 118, 3202-3215.  [11]  Boulanger, M.J., Kukimoto, M., Nishiyama, M., Horinouchi, S. and Murphy, M.E. (2000). Catalytic roles for two water bridged residues (Asp-98 and His-255) in the active site of copper-containing nitrite reductase. J Biol Chem 275, 2395764.  [12]  Boulanger, M.J. and Murphy, M.E. (2001). Alternate substrate binding modes to two mutant (D98N and H255N) forms of nitrite reductase from Alcaligenes faecalis S-6: structural model of a transient catalytic intermediate. Biochemistry 40, 9132-41.  [13]  Boulanger, M.J. and Murphy, M.E. (2003). Directing the mode of nitrite binding to a copper-containing nitrite reductase from Alcaligenes faecalis S-6: characterization of an active site isoleucine. Protein Sci 12, 248-56.  61 [14]  Wijma, H.J., Boulanger, M.J., Molon, A., Fittipaldi, M., Huber, M., Murphy, M.E.P., Verbeet, M.P. and Canters, G.W. (2003). Reconstitution of the type-1 active site of the H145G/A variants of nitrite reductase by ligand insertion. Biochemistry 42, 4075-4083.  [15]  Wijma, H.J., Macpherson, I., Alexandre, M., Diederix, R.E., Canters, G.W., Murphy, M.E. and Verbeet, M.P. (2006). A rearranging ligand enables allosteric control of catalytic activity in copper-containing nitrite reductase. J Mol Biol 358, 1081-93.  [16]  Cadwell, R. and Joyce, G. (1992). Randomization of genes by PCR mutagenesis. PCR Methods Appl. 2, 28-33.  [17]  Miyazaki, K. (2003). Creating random mutagenesis libraries by megaprimer PCR of whole plasmid (MEGAWHOP). Methods Mol Biol 231, 23-8.  [18]  Stemmer, W.P. (1994). Rapid evolution of a protein in vitro by DNA shuffling. Nature 370, 389-91.  [19]  Kadono-Okuda, K. and Andres, D. (1997). An expression cloning method to identify monomeric GTP-binding proteins by GTP overlay. Anal. Biochem. 254, 187-191.  [20]  Wijma, H.J., Canters, G.W., de Vries, S. and Verbeet, M.P. (2004). Bidirectional catalysis by copper-containing nitrite reductase. Biochemistry 43, 10467-74.  [21]  Boulanger, M.J. (2001) The molecular mechanism of copper-containing nitrite reductase. University of British Columbia, Vancouver.  62 [22]  Kohzuma, T., Shidara, S. and Suzuki, S. (1994). Direct electrochemistry of nitrite reductase from Achromobacter cycloclastes IAM 1013. Bulletin of the Chemical Society of Japan 67, 138-43.  [23]  Murshudov, G.N., Vagin, A.A. and Dodson, E.J. (1997). Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr 53, 240-55.  [24]  Collaborative Computational Project, N. (1995). The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D 50, 760-763.  [25]  Rensing, C. and Grass, G. (2003). Escherichia coli mechanisms of copper homeostasis in a changing environment. FEMS Microbiol Rev 27, 197-213.  [26]  Quaratino, D., Federici, F., Petruccioli, M., Fenice, M. and D'Annibale, A. (2007). Production, purification and partial characterisation of a novel laccase from the white-rot fungus Panus tigrinus CBS 577.79. Antonie Van Leeuwenhoek 91, 5769.  [27]  Hassett, R.F., Yuan, D.S. and Kosman, D.J. (1998). Spectral and kinetic properties of the Fet3 protein from Saccharomyces cerevisiae, a multinuclear copper ferroxidase enzyme. J Biol Chem 273, 23274-82.  [28]  Schosinsky, K.H., Lehmann, H.P. and Beeler, M.F. (1974). Measurement of ceruloplasmin from its oxidase activity in serum by use of o-dianisidine dihydrochloride. Clin Chem 20, 1556-63.  [29]  Larrondo, L.F., Salas, L., Melo, F., Vicuna, R. and Cullen, D. (2003). A novel extracellular multicopper oxidase from Phanerochaete chrysosporium with ferroxidase activity. Appl Environ Microbiol 69, 6257-63.  63 [30]  Ingledew, W.J. and Poole, R.K. (1984). The respiratory chains of Escherichia coli. Microbiol Rev 48, 222-71.  [31]  Lam, Y. and Nicholas, D.J. (1969). A nitrite reductase with cytochrome oxidase activity from Micrococcus denitrificans. Biochim Biophys Acta 180, 459-72.  [32]  Tocheva, E.I., Rosell, F.I., Mauk, A.G. and Murphy, M.E. (2004). Side-on copper-nitrosyl coordination by nitrite reductase. Science 304, 867-70.  [33]  Hough, M.A., Ellis, M.J., Antonyuk, S., Strange, R.W., Sawers, G., Eady, R.R. and Samar Hasnain, S. (2005). High resolution structural studies of mutants provide insights into catalysis and electron transfer processes in copper nitrite reductase. J Mol Biol 350, 300-9.  [34]  Kukimoto, M., Nishiyama, M., Ohnuki, T., Turley, S., Adman, E.T., Horinouchi, S. and Beppu, T. (1995). Identification of interaction site of pseudoazurin with its redox partner, copper-containing nitrite reductase from Alcaligenes faecalis S-6. Protein Eng 8, 153-8.  [35]  Olesen, K., Veselov, A., Zhao, Y., Wang, Y., Danner, B., Scholes, C.P. and Shapleigh, J.P. (1998). Spectroscopic, kinetic, and electrochemical characterization of heterologously expressed wild-type and mutant forms of copper-containing nitrite reductase from Rhodobacter sphaeroides 2.4.3. Biochemistry 37, 6086-94.  [36]  Sawyer, D.T. (1991) Oxygen chemistry. New York.  [37]  Nersissian, A.M. and Shipp, E.L. (2002). Blue copper-binding domains. Adv Protein Chem 60, 271-340.  64 [38]  Durao, P., Bento, I., Fernandes, A.T., Melo, E.P., Lindley, P.F. and Martins, L.O. (2006). Perturbations of the T1 copper site in the CotA laccase from Bacillus subtilis: structural, biochemical, enzymatic and stability studies. J Biol Inorg Chem 11, 514-26.  [39]  Hall, J.F., Kanbi, L.D., Strange, R.W. and Hasnain, S.S. (1999). Role of the axial ligand in type 1 Cu centers studied by point mutations of met148 in rusticyanin. Biochemistry 38, 12675-80.  [40]  Chung, K.T., Chen, S.C., Wong, T.Y., Li, Y.S., Wei, C.I. and Chou, M.W. (2000). Mutagenicity studies of benzidine and its analogs: structure-activity relationships. Toxicol Sci 56, 351-6.  [41]  Reinhammar, B.R. (1972). Oxidation-reduction potentials of the electron acceptors in laccases and stellacyanin. Biochim Biophys Acta 275, 245-59.  [42]  Reinhammar, B.R. and Vanngard, T.I. (1971). The electron-accepting sites in Rhus vernicifera laccase as studied by anaerobic oxidation-reduction titrations. Eur J Biochem 18, 463-8.  [43]  Adman, E.T., Godden, J.W. and Turley, S. (1995). The structure of copper-nitrite reductase from Achromobacter cycloclastes at five pH values, with NO2- bound and with type II copper depleted. J Biol Chem 270, 27458-74.  65 Chapter Three: Crystal structure and functional characterization of a trimeric multicopper oxidase from Arthrobacter 3.1 Introduction Copper is an essential element for most organisms, acting as a cofactor of metalloproteins critical to survival. Copper is proposed to have emerged as a protein cofactor as the world’s atmosphere became oxygenated due to the advent of photosynthesis, increasing the solubility of Cu2+ and making it more readily available to organisms [1]. Thus, the role of copper metalloenzymes is often activation or utilization of molecular oxygen. An example are multicopper oxidases (MCOs) which catalyze the four-electron reduction of oxygen to water [2]. These enzymes contain three spectroscopically distinct copper sites – named type-1, type-2, and type-3. The type-1 Cu is responsible for oxidizing a variety of small molecules or metal ions and transferring the electrons to the type-2 and type-3 Cu-containing trinuclear cluster, the active site where oxygen is reduced to water [2]. The trinuclear Cu cluster is linked to the peptide chain by a total of eight histidine ligands. The type-2 Cu is coordinated by two histidines and each type-3 Cu is coordinated by three histidines. Nitrite reductase (NiR) is a related copper-containing enzyme that does not interact with O2 as its primary role, although it is capable of reducing oxygen [3]. This enzyme catalyzes the one-electron reduction of nitrite (NO2-) to nitric oxide (NO) gas during dissimilatory denitrification, a process that allows respiratory proton pumping utilizing nitrate and its reduction products as electron acceptors instead of oxygen [4]. *A version of this chapter will be submitted for publication. MacPherson, I. S., Lee, W. C., Liang, T. I., and M. E. P. Murphy. Crystal structure and functional characterization of a trimeric multicopper oxidase from Arthrobacter.  66 NiR contains a type-1 Cu site that transfers electrons to the type-2 Cu, the site of nitrite binding and reduction to NO [4]. MCO and NiR are homologous proteins composed of cupredoxin domains [5]. The simplest cupredoxin domain proteins are the cupredoxins, small single domain proteins that function as electron carriers. When oxidized, type-1 Cu sites are characterized by EPR signals with low coupling constants and by strong electronic absorption in the 600 nm and sometimes 460 nm range that give the proteins an intense blue to green color [5]. NiRs are typically composed of homotrimers in which each monomer is constructed from two cupredoxin domains; whereas typical MCOs are generally composed of monomers of three cupredoxin domains and sometimes six cupredoxin domains such as ceruloplasmin. Duplication of a gene encoding a single cupredoxin domain has been suggested to bring about the two-domain NiR [6,7]. An additional duplication has been proposed to result in the three-domain MCO and a triplication of the two-domain protein has been proposed to give rise to the six-domain ceruloplasmin [6,8]. Whereas NiR has one type-1 Cu site in each of the N-terminal cupredoxin domains, all three-domain MCOs characterized have one type-1 Cu site in the C-terminal domain. Nakamura et al hypothesized and by sequence analysis discovered three types of two-domain MCOs that can explain the evolutionary relationships between NiR and three-domain MCOs [6]. These proteins contain four His-X-His motifs where in NiR these are His-X-X or X-X-His at homologous positions [6]. Modeling of these His-X-His motifs in NiR results in a metal binding site that has the potential to bind three coppers in an identical fashion to the multicopper oxidase trinuclear site, suggesting that these two-  67 domain proteins could function as multicopper oxidases. The type A two-domain MCOs contain type-1 Cu sites in both the C- and N-terminal domains. Loss of a type-1 Cu site in the N-terminal domain results in a type B two-domain MCO. Insertion of another cupredoxin domain between the domains of type B gives rise to the three-domain MCOs. Loss of the C-terminal type-1 Cu site in a type A MCO yields the type C variety, a protein more similar to NiR. Two type B two-domain MCO proteins have been characterized to date and shown to function as phenol oxidases [9,10]. We have expressed a recombinant form of a type C two-domain MCO from Athrobacter sp. (AMMCO). The x-ray structure of AMMCO is strikingly similar to nitrite reductase yet the presence of a trinuclear cluster and functional characterization show the enzyme to be a true MCO with a suggested role in metal homeostasis.  3.2 Materials and methods 3.2.1 Protein expression The locus Arth4419 from the genome sequence of Arthrobacter sp. FB24 (GenBank accession number NC_008537) was identified by BLAST searches of sequences similar to nitrite reductase yet possessing MCO motifs. The gene minus the first 75 amino acids was optimized for expression in E. coli, synthesized (Bio Basic Inc.) and cloned into pET28a (Novagen) allowing for an N-terminal His-tag with thrombin cleavage site (pET28a-AMMCO). This construct also contains an EcoR1 site immediately after the translation stop codon of the AMMCO gene but before the T7 terminator region. The E192A and D190A/E192A variants were produced using a  68 modified megaprimer method of Miyazaki et al. [11]. Briefly, mutagenic primers for E192A (5’ gacggtggtgacgacaatgcattttactctgttaacggc) or for D190A/E192A (5’ aacacggacggtggtgacgcaaatgcattttactctgttaacggc) were used along with a primer corresponding to the T7 terminator region of pET28a in a PCR, yielding the C-terminal region of the AMMCO gene with the desired substitution. pET28a-AMMCO was digested with EcoR1, allowing annealing of the mutagenic PCR megaprimer at both ends of the linearized plasmid. A 50 µl thermal cycling reaction consisting of the mutagenic PCR reaction (800 ng), linearized pET28a-AMMCO (50 ng), 1X reaction buffer, 1 mM MgSO4, 2.5 U Platinum Pfx (Invitrogen), 300 µM dNTPs, and 4% DMSO was cycled between 95° for 30 seconds and 68° for 7 minutes for 12-24 cycles. The reactions were electroporated into E. coli DH5α and plasmid was extracted from clones and sequenced. pET28a-AMMCO and mutants were transformed into E. coli BL21(DE3) and a single colony was used to inoculate a 5 ml overnight culture. One litre of 2YT media was inoculated with 4 ml overnight culture and grown at 30° to an O.D.600 of 1.0. The cultures were then grown at 25° for 30 minutes, followed by 0.5 mM IPTG induction and growth for 16 hours at 25°. The cells were harvested by centrifugation and resuspended in binding buffer (20 mM sodium phosphate pH 7.8, 500 mM NaCl) plus 1 mM CuSO4 and 50 mM imidazole pH 7. An Emulsiflex C5 homogenizer (Avestin) was used to lyse the cells, the debris was pelleted, and the extract applied to nickel-loaded Chelating Sepharose Fastflow (GE Healthcare). The column was washed and AMMCO eluted at 900 mM imidazole, pH 6. The N-terminal His-tag was then removed by bovine αthrombin cleavage and the protein applied to a Source 15Q column (GE Healthcare) and  69 eluted at 200 mM NaCl at pH 8. AMMCO was expressed with a yield of 40-60 mg of purified protein per liter culture. 3.2.2 Crystallography AMMCO crystals were grown with 35 mg/ml stock protein in 0.1 M imidazole pH 8, 0.2 M CaOAc, 0.1 M CaCl2, and 12% polyethyleneglycol 8000. Within 48 hours, large blue spherulites formed, the edges from which blue single crystals sprouted within 2 weeks. These crystals were used for hair microseeding, resulting in crystal formation within 72 hours. Crystals were looped and immersed in liquid nitrogen using the crystallization buffer supplemented with 25% (v/v) ethylene glycol as a cryoprotectant. A single wavelength anomalous dispersion dataset was collected at beamline 9-2 of the Stanford Synchrotron Radiation Laboratory (SSRL) at the Cu edge (~1.38 Å) as determined by x-ray fluorescence scan. The AMMCO crystals were identified to be in space group C2 (a = 133.95 Å, b = 50.71 Å, c = 134.67 Å, β = 107.9°). The programs Mosflm and Scala were used to integrate and scale the dataset [12]. In the phasing step, the program SOLVE located 11 out of 12 copper sites and the program RESOLVE was used for the density modification procedure [13]. The non-crystallographic 3-fold symmetry identified from the copper sites was used in averaging resulting in a map of exceptional quality that was readily traceable. The program ARP/WARP was used to build the initial model [14]. XtalView and Refmac5 were used to build and refine the final structure [12,15]. A list of data collection and refinement statistics can be found in Table 3-1. 3.2.3 Cyclic voltammetry Cyclic voltammograms were taken using an Autolab potentiostat (Eco Chemie).  70 Table 3-1 Data collection and refinement statistics for wt recombinant AMMCO Crystal Resolution (Å) Rmerge {I}/{σI} Completeness (%)  AMMCO 128.0 ~ 1.8 0.069 (0.222) 5.9 (2.7) 99.6 (100.0)  Redundancy  4.1 (4.0)  No. reflections  324746  Wilson B-factor (Å2) Rwork/Rfree  25.0 0.191/0.225  No. atoms / B-factors (Å2) Protein  6,816 / 25.0  Copper  12 / 27.0  Calcium  3 / 25.5  Water  428 / 31.8  r.m.s. deviations  a  Bond lengths (Å)  0.008  Bond angles (°)  1.182  Highest resolution shell is shown in parentheses  71 AMMCO (4.5 µM) was added to a cell containing a polished edge-oriented pyrolytic graphite working electrode. Buffer-saturated argon was bubbled through the cell. The reference electrode was a calomel electrode (Radiometer) and the counter electrode was a platinum wire. A voltage of -300 mV to 300 mV (vs. S.C.E.) was applied to the sample at a scan rate of 5 mV/sec. 3.2.4 Assays Copper contents of AMMCO and variants were determined using bicinchoninic acid with denatured samples reduced with ascorbate, which yields a purple complex measured at 562 nm[16]. AMMCO and variants were tested at optimum pH and 25° against a variety of phenolic substrates- 2,6-dimethoxyphenol (1 mM in 100 mM sodium phosphate pH 6.5, ε469 nm = 49.6 mM-1cm-1), ferrocyanide (10 mM in 100 mM sodium phosphate pH 7.0, ε405 nm = 900 mM-1cm-1), and ABTS (1 mM in 50 mM sodium acetate pH 4.5, ε405 nm = 36.6 mM-1cm-1). Ferroxidase activity was measured by the ferrozine method, which quantifies the remaining Fe2+ (ε570 nm = 27.9 mM-1cm-1) at different time points.[17]  3.3 Results 3.3.1 Crystal structure Recombinant wild-type AMMCO crystallized in space group C2 with a trimer in the asymmetric unit and the structure was solved to 1.8 Å resolution. The model is complete from residues Gly12 to Ala299 (chain B is from Ser14 to Ser298) and 90 % of the residues lie in the most favorable region of the Ramachandran plot [18]. The four residues in the disallowed regions are well defined in the electron density map. Asp41  72 (chain A) is at a β-turn between Pro40 and Gly42 and its side chain forms a salt bridge with the guanidinium group of Arg43. Tyr226 (chains A, B, C) is stacked between the two cupredoxin domains. The overall structure of the AMMCO trimer is a triangular prism (Figure 3-1) with ~70 Å edges for each side of the triangle and a height of ~45 Å. Each monomer forms two Greek key β-barrel domains. The N-terminal domain is composed of residues Gly12 to Glu164 followed by a linker region (residues 165-173) connected to the C-terminal domain (residues Asp174 to Ala299). The monomermonomer interface is extensive (~1200 Å2) and is mostly nonpolar in nature; 72% of the interfacial residues are nonpolar as defined using the Protein-protein Interaction Server [19]. A type-1 Cu site is found at one end of the N-terminal β-barrel. The Cu atom is coordinated by the side chains of His99, Cys144, and His152 in trigonal planar configuration and flanked axially by a non-coordinating Leu157 (Figure 3-2). The imidazole ring of His152 is surface-exposed. The molecular surface surrounding the type1 site is largely uncharged. A surface-exposed tryptophan residue (Trp68) is located within 3.5 Å from both the His99 and His152 ligands as well as a non-coordinating histidine (His97) that sits above the type-1 Cu, ~7 Å away. A trinuclear Cu cluster is coordinated by histidine residues 102, 104, 143, and 145 from the N-terminal domain of one monomer and histidines 233, 235, 280, and 282 from the C-terminal domain of an adjacent monomer (Figure 3-2). Each histidine coordinates a single Cu atom by the Nε2 atom of the side chain imidazole group. By homology with ascorbate oxidase (PDB entry 1AOZ), the two Cu atoms (5.6 Å apart) coordinated by histidines 104, 143 and 282 and histidines 145, 235 and 280, respectively, are type-3 Cu  73  Figure 3-1 Overall structural comparison of AMMCO (left) and AfNiR (right). Monomers are colored red, blue and green. The copper atoms are copper-colored spheres. Grey spheres are calcium atoms. The Cu active sites (trinuclear for AMMCO, mononuclear for NiR) are located at the interface of adjacent monomers.  74  A  B  C  Figure 3-2 Metal sites of AMMCO. Panel A: side-view of the trinuclear Cu site, type-1 Cu site, and calcium binding  loop. Cu atoms are copper spheres and calcium atoms are grey spheres. Two chains of the trimer are colored green and  blue. Panel B: Stereo view of the calcium binding loop. Asn184 ligates the calcium ion in foreground and Glu192  ligates the calcium ion in the background. Panel C: Stereo view of the trinuclear Cu site. The small red spheres are solvent molecules.  75 atoms. The type-2 Cu atom, positioned 4.4 and 4.0 Å from the Type-3 Cu atoms, is coordinated by His102 and His233. A solvent molecule weakly coordinates the type-2 Cu (3.1 Å) and another solvent molecule weakly coordinates all three Cu atoms with distances of 3.0, 3.0, and 2.7 Å from the type-2 and type-3 Cu atoms, respectively. Six of the eight trinuclear ligands (His102, His143, His145, His233, His280, and His282) form hydrogen bonds with backbone carbonyls of Asp112, Cys144, His145, Asp255, Ala281 and His282, respectively via their non-coordinating Nδ1 atoms. In addition, two aspartate residues, Asp112 and Asp255, interact with histidine Cu ligands. His102 and His235 form hydrogen bonds with the side chain carboxylate of Asp112 which also forms a hydrogen bond with a nearby non-coordinating His107. Asp255 forms a hydrogen bond with His104 and with a water brided interaction with the type-2 Cu atom. Adjacent to the type-1 Cu site is a narrow channel formed by residues Leu149, Met278, His280-Thr284, Ala287-Glu288, and Met292 which leads directly to the trinuclear active site and may serve as conduit for oxygen and water. Also, near Met278 and Met292 at the opening of the channel is one additional methionine (Met204). His143 and His145 of the trinuclear site are bridged to Cys144 of the type-1 Cu site; thus, there are short covalent linkages between the type-1 Cu and the trinuclear cluster to enable rapid electron transfer. Crystals were obtained in the presence of 0.3 M calcium ions. A loop in each monomer containing residues Asn184 to Glu192 coordinates a calcium ion, with a distorted octahedral geometry. These loops protrude from the molecular surface such that the calcium ion is ~18 Å from the type-1 Cu of the same monomer (Figure 3-2). For monomer A, the coordinating ligands (bond distances) are the side chains of Asn184 (2.4  76 Å), Asp196 (2.4 Å), Asp190 (2.3 Å), and Glu192 (2.5 Å, 2.5 Å). In addition, the backbone carbonyl atoms of Gly187 (2.4 Å) and Asp190 (2.4 Å) complete the coordination sphere. In monomer B, the main chain of the loop differs such that the backbone carbonyls of Gly188 (2.6 Å) and Asp190 (2.3 Å) coordinate the calcium. The B and C monomer calcium binding loops do not form crystal contacts but monomer A residues Asp186 and Asp190 form salt bridges with residues Arg82 and Lys166 from monomer A of an adjacent trimer in the crystal lattice. 3.3.2 Electronic absorption spectra Upon lysis of E. coli cells in the presence of 1 mM CuSO4, the lysate containing recombinant wild-type AMMCO remained colorless. However, for the E192A and D190A/E192A variants, the lysates began to darken within a few minutes. Purified recombinant wild-type and variant proteins were intensely blue in color. An electronic absorption spectrum of recombinant wild-type AMMCO is found in Figure 3-3. The intense absorbance at 600 nm (ε ~ 5000 M-1cm-1) is attributed to oxidized type-1 Cu sites. This peak is at 602 nm for the wild-type recombinant AMMCO, whereas the E192 and D190A/E192A variants have peaks shifted to 592 nm. In addition, the variant proteins both have significantly more absorbance in the near-UV region (325-425 nm). 3.3.3 Electrochemical and activity assays Direct electrochemistry of AMMCO was attempted using a polished edgeoriented pyrolytic graphite electrode. Cyclic voltammograms were obtained (Figure 3-4) from which an approximate midpoint potential of ~330 mV vs SHE could be estimated. Higher scan rates resulted in less well-defined voltammograms. At a lower applied  77  8000  Extinction coefficient (ε)  7000 6000 5000 4000 3000 2000 1000 0 300  400  500  600  700  800  wavelength (nm)  Figure 3-3 Electronic absorption spectra of wt recombinant AMMCO (thick black line), E192A (thick grey line), and D190A/E192A (thin black line). Samples were buffered in 20 mM MOPS pH 7.0.  78  2.00E-07 0.00E+00  i/A (A) Current  -2.00E-07 -4.00E-07 -6.00E-07 -8.00E-07 -1.00E-06 -1.20E-06 -0.4  -0.3  -0.2  -0.1  0  0.1  0.2  0.3  0.4  Potential E/V (V vs. SCE)  Figure 3-4 Cyclic voltammogram (Current plotted against potential (vs. SCE)) of AMMCO (blue plot) compared to bare electrode (red).  79 potential of -300 mV (vs. SCE), the current increased considerably suggesting that the presence of trace oxygen in the reaction cell resulting in catalysis by AMMCO. AMMCO and the E192A and D190A/E192A variants were assayed for specific activity using ABTS, 2,6-dimethoxyphenol and ferrocyanide as electron donors in the presence of ambient oxygen (Table 3-2). Oxidation of ABTS gave the highest specific activity for both wild-type and variant AMMCOs. Notably, the E192A and D190A/E192A variants were found to be more active than wild-type recombinant AMMCO. Copper occupancy of the protein samples was determined and shown not to be a significant factor in determining the specific activities (Table 3-2).  3.4 Discussion 3.4.1 AMMCO is a link between monomeric MCOs and NiR Mechanistically, NiR and MCO share many commonalities. First, electron routing through a type-1 Cu site is a feature shared by both types of enzyme. Second, both enzymes catalyze the reduction of substrate at a copper-containing active site. Although the primary role of NiR is the reduction of NO2- to NO, NiR has also been shown to reduce oxygen to hydrogen peroxide [3]. This parallels what is seen with MCO, which generates a peroxide-level intermediate from O2 before complete reduction to water [20]. The addition of two Cu atoms as compared to NiR to give a trinuclear cluster in MCO enables the four-electron reduction of oxygen to water [2]. The existence of key active site residues (Asp98, His255, and Ile257) in NiR at homologous positions to Cucoordinating histidine residues in MCO is an attractive evolutionary relationship [6,7,21,22].  80  Table 3-2 Copper content and kcat’s of AMMCO and variants  Copper content  ABTS  2,6-dimethoxyphenol  Ferrocyanide  (Cu/monomer)  (min-1)  (min-1)  (min-1)  WT  3.4  8.4 ± 0.3  1.8 ± 0.1  1.5 ± 0.0  E192A  3.8  14.7 ± 0.9  2.6 ± 0.0  2.3 ± 0.1  D190A/E192A  3.2  20.4 ± 0.8  3.1 ± 0.1  2.9 ± 0.3  81 The presence of a trinuclear Cu cluster in the crystal structure (Figure 3-2) and ABTS oxidation activity (Table 3-2) demonstrates that AMMCO is a true MCO. Not observed previously in structurally characterized MCOs, AMMCO forms a trimer with three trinuclear clusters located at subunit interfaces. This multimeric structure is a conserved feature of the homologous copper nitrite reductase family in which three mononuclear copper atoms are found at the subunit interfaces. Structural similarity of AMMCO to entries in the PDB was screened using the VAST server [23,24]. The most similar entries are copper nitrite reductases from Neisseria gonorrhea (1KBV, known as AniA), Hyphomicrobium denitrificans (2DV6) and Alcaligenes faecalis (1AS8). These alignments are over 250 residues with a r.m.s.d. of ~2.0 Å and sequence identities of 2430%. Next, two MCO crystal structures are identified, human ceruloplasmin (1KCW) and E. coli CueO (1N68), with alignments of ~230 residues at ~2.4 Å r.m.s.d. and sequence identities of 17 to 23%. Given that the overall fold, domain structure, and quaternary structure of AMMCO are most similar to those of the NiRs from N. gonorrhea (AniA) and A. faecalis AfNiR (Figure 3-1), these proteins will be a focus for further comparisons. A sequence alignment of these two NiRs (AfNiR and AniA) with AMMCO is presented in Figure 3-5. The amino acid residues in AMMCO that coordinate to the type1 Cu atom within the N-terminal domain and the residues that coordinate the Cu atoms of the trinuclear cluster are indicated in the alignment. The localization of the Type-1 Cu site in the N-terminal domain is a conserved feature of Cu NiRs [25], whereas in all previously characterized MCOs, the Type-1 Cu site is in the C-terminal most domain [5]. Furthermore, the sequence alignment shows that histidines 100, 135, and 306  82  AMMCO AniA AfNiR  10 20 30 40 50 60 ....|....|....|....|....|....|....|....|....|....|....|....| MTNHAGHAGFAGGSVLAERAGIDPTAILRDFDRGRTSTLPDGRTLREWDIVAVDKDFEIA 60 MAAQATAETPAGELPVIDAVTTHAPEVPPAIDRD-------YPAKVRVKMETVEKTMKMD 53 VRKATAAEIAALPRQKVELVDPPFVHAHSQVAEG-------GPKVVEFTMVIEEKKIVID 53  AMMCO AniA AfNiR  70 80 90 100 110 120 ....|....|....|....|....|....|....|....|....|....|....|....| PG-IIFKGWSYNGRIPGPTLWAREGDALRIHFTN--AGAHPHTIHFHGVHRATMDGTPGI 117 DG-VEYRYWTFDGDVPGRMIRVREGDTVEVEFSNNPSSTVPHNVDFH----AATGQGGGA 108 DAGTEVHAMAFNGTVPGPLMVVHQDDYLELTLINPETNTLMHNIDFH----AATGALGGG 109  AMMCO AniA AfNiR  130 140 150 160 170 180 ....|....|....|....|....|....|....|....|....|....|....|....| GAGSIAPGQSFTYEFDATPFGTHLYHCHQSPLA-PHIAKGLYGGFIVEPKEGRP------ 170 AATFTAPGRTSTFSFKALQPGLYIYHCAVAPVG-MHIANGMYGLILVEPKEGLP------ 161 GLTEINPGEKTILRFKATKPGVFVYHCAPPGMVPWHVVSGMNGAIMVLPREGLHDGKGKA 169  AMMCO AniA AfNiR  190 200 210 220 230 240 ....|....|....|....|....|....|....|....|....|....|....|....| -PADDEMVMVMNGYNTDGGDDN------------------------EFYSVNGLP-FHFM 204 -KVDKEFYIVQGDFYTKGKKGAQGLQPFDMDKAVAE--------QPEYVVFNGHVGALTG 212 LTYDKIYYVGEQDFYVPRDENGKYKKYEAPGDAYEDTVKVMRTLTPTHVVFNGAVGALTG 229  AMMCO AniA AfNiR  250 260 270 280 290 300 ....|....|....|....|....|....|....|....|....|....|....|....| DFPVKVKQHELVRIHLINVLEYDPINSFHIHGNFFHYYPTGT--MLTPSEYTDTISQVQG 262 DNALKAKAGETVRMYVGNGGP-NLVSSFHVIGEIFDKVYVEG--GKLINENVQSTIVPAG 269 DKAMTAAVGEKVLIVHSQANR-DTRP--HLIGGHGDYVWATGKFNTPPDVDQETWFIPGG 286  AMMCO AniA AfNiR  310 320 330 340 350 ....|....|....|....|....|....|....|....|....|....|....|... QRGILELRFPYPGKFMFHAH-KTEFAELGWMGFFEVSAS------------------- 300 GSAIVEFKVDIPGNYTLVDHSIFRAFNKGALGQLKVEGAENPEIMTQKLSDTAYAVPR 327 AAGAAFYTFQQPGIYAYVNHNLIEAFELGAAAHFKVTGEWNDDLMTSVLAPSGT---- 340  Figure 3-5 Sequence alignment of AMMCO, AniA, and AfNiR. Residues colored in red are the active site His-X-His (AMMCO) and His-X-X or X-X-His (NiR) motifs. Residues colored in blue contain the calcium binding loop (AMMCO) and homologous loops in AniA and AfNiR.  83  Figure 3-6 Stereo image of NiR (green) and AMMCO (blue) active sites superposed. The type-1 Cu-coordinating cysteine for both enzymes is located on the far right. NiR residue Asp98 is located in the back left, Ile257 is back middle, His255 is back right. Ala137 and Val304 are front left and front right, respectively.  84 coordinating the AfNiR type-2 Cu correspond to the AMMCO type-3 Cu coodinating histidines 104, 143 and 282 (Figures 3-5 & 3-6). In fact, the His residues of the AfNiR type-2 Cu site and those of one of AMMCO type-3 Cu sites superimpose well (r.m.s.d. of 0.74 Å over all 31 atoms). In addition, three residues (Asp98, His255, Ile257), which play critical roles in nitrite reduction [26,27], all align with trinuclear cluster ligands (His102, His233, His235) in AMMCO. Superposition of all four His-X-His motifs of AMMCO with the homologous positions in AfNiR (Asp98-X-His100, His135-X-Ala137, etc) results in an r.m.s.d. of 0.61 Å over the 12 Cα atoms (Figure 3-6). Thus, the active sites of AfNiR and AMMCO are strikingly similar despite the differences in Cu coordination and the specific reaction catalyzed. In addition to eight trinuclear histidine ligands, AMMCO contains second shell features at the active site that are similar to those found in three-domain MCOs. First, the hydrogen bonding potential of all His ligands at the non-coordinating nitrogen is fully satisfied. Backbone carbonyl hydrogen bonding is a characteristic shared by His102, His143, His145, His233, His280, and His282 and the homologous histidines in the threedomain MCOs such as CueO (PDB entry 1KV7). In addition, the hydrogen bonds between the carboxylates of Asp255 and Asp112 to the remaining His104 and His235, respectively, are also formed by homologous residues in three-domain MCOs. In addition to hydrogen bond formation with trinuclear Cu histidine ligands, these aspartate residues simultaneously form other hydrogen bonds. Asp112 is highly conserved in three domain MCOs and has been implicated in proton transfer to the peroxide-level intermediate during catalysis [28]. However, whereas the three-domain MCO aspartate forms a hydrogen bond with a water-water-type-2 Cu network, in  85 AMMCO Asp112 forms a hydrogen bond to a non-coordinating His107. Coincidentally, Asp255 of AMMCO forms the characteristic Asp-water-water-Type-2 Cu network implicated in proton donation [28]. Another marked difference between the trimeric proteins (NiR and AMMCO) and three-domain MCOs is centered on the Cu coordinating atom of a single histidine ligand at the active site. His100 in NiR and His104 in AMMCO coordinate a type-3 copper atom via the Nε2 atom, whereas all threedomain MCOs coordinate the same Cu via the Nδ1 atom. This difference can be rationalized by an insertion of an adjacent Gly residue C-terminal to the coordinating His in the three domain MCOs, shifting the His Cα and enabling coordination by the Nδ1 atom. Nakamura et al have proposed a model on the evolution of multimeric cupredoxin domain proteins including MCOs and Cu NiRs [22]. Of particular interest is the existence of common ancestors in this evolutionary scheme to explain the characterized and putative MCO sequences observed in the sequence databanks. From this analysis, the type-1 Cu site may exist in the N-terminal domain (type C), the C-terminal domain (type B), or in both domains (type A) of putative two-domain MCOs. A type A two-domain MCO has been proposed to be the common ancestor of the type B and type C MCOs. Also proposed is that a protein similar to AMMCO gave rise to nitrite reductase by loss of some type-2 and type-3 Cu ligands. Given the structural similarity between the NiR and AMMCO active sites, including key residues for catalysis, this evolutionary leap could have been surprisingly small. By a domain duplication, type B two-domain MCOs could give rise to the three domain MCO commonly found in prokaryotes and eukaryotes. This duplication represents a significant evolutionary leap that may have lead  86 to improved trinuclear cluster function or simpler expression since multimerization is not needed for these proteins. Two type B two-domain MCOs with the type-1 Cu ligands in the C-terminal domain have been characterized biochemically. SLAC, studied by Machczynski et al, and EpoA, studied by Endo et al, have been recombinantly expressed and their migration in SDS-PAGE gels suggested multimerization [9,10]. The amino acid sequences of SLAC (from Streptomyces ceolicolor) and EpoA (from Streptomyces griseus) are 69% identical [10]; however SLAC is suggested to be a dimer [10], while EpoA is proposed to be a trimer [9]. The function of EpoA is proposed to be cell morphogenesis in reaction to exogenous copper or related stimuli. SLAC has been proposed to participate in copper sequestration, based on the excess of histidine residues in the sequence (double the average histidine composition not counting the type-1 or trinuclear histidines). Considering the evolutionary distance between the type B and type C two-domain MCOs, their specific functions may be different. The lower activity of AMMCO compared to SLAC using the reductant DMP (greater than two orders of magnitude difference) confirms this difference. AMMCO, SLAC, and EpoA contain twin arginine signals in their native sequences, suggesting that the protein is completely folded before exportation [29]. In contrast, NiR is translocated via a leader sequence by a type 2 secretion system before extracellular folding and copper binding. 3.4.2 The AMMCO type-1 Cu site In general, type-1 Cu atoms are coordinated by a minimum of three side chain ligands (one cysteine, two histidines) [5]. At least two of these side chain ligands, a Cys and His, arise from a single loop, known as the “ligand loop”. In addition, the ligand loop  87 may contain a coordinating Met or Gln at the axial position. Otherwise, the axial position is replaced by non-coordinating Leu, Ile, or Phe residue. A second His ligand arising from an adjacent β-strand is always present in type-1 sites. The ligand loop varies in length among type-1 Cu proteins. AfNiR is known to have the longest ligand loop containing 15 residues from Cys136 to the axial ligand Met150 [5]. At the opposite extreme, amicyanin, a small 10 kDa cupredoxin, contains seven residues from Cys92 to Met98. An inspection of all structurally characterized MCOs with a C-terminal type-1 Cu reveals that they contain 11 residues from Cys to the residue at the axial position, whether it is a coordinating Met or non coordinating Leu or Phe. The amino acid sequences of three putative type A two-domain MCOs suggest an N-terminal ligand loop with 13 residues [6]. In contrast, AMMCO contains 14 residues in the ligand loop from Cys144 to the axial Leu157. Thus, the length of this functionally significant ligand loop is intermediary between those of the proposed ancestor type A MCOs and of NiR supporting the notion that AMMCO is an evolutionary link. Interestingly, the length of the ligand loop in AniA is 14 residues, shorter than typical NiRs and yet equal to that from AMMCO. The functionality of the ligand loop in tuning the reduction potential of the type-1 Cu site has been investigated recently in the cupredoxins azurin, amicyanin and plastocyanin [30]. By this analysis, the type-1 Cu site in AMMCO may have unique redox properties compared to other multicopper oxidases. The estimated reduction potential for AMMCO of 330 mV by cyclic voltammetry (Figure 3-4) is on the lower end of measured potentials for MCOs, which range from 340-780 mV [31]. The closest sequenced homologues for AMMCO, from Nocardia farcinica and Rhodococcus  88 erythropolis (both sharing ~60% amino acid sequence identity), contain an axial Met residue rather than Leu. Replacement of the axial Met with Leu in several type-1 sites has resulted in an increase of the reduction potential by 100 to 300 mV [32-34], suggesting that these AMMCO homologues could have reduction potentials as low as 250 mV, comparable to that measured for NiR [35]. Surface features surrounding the type-1 Cu site differ significantly between AfNiR and AniA [25]. In this region, the AMMCO architecture is more similar to AniA in that the surface is much flatter than that of AfNiR (Figure 3-7). This flatness is due largely to the difference in the length (33 amino acids) of a large loop in AfNiR (referred to the ‘tower loop’, residues 184-216). The corresponding loop in AMMCO is composed of only 9 residues (184-192). In AfNiR, the tower loop folds up and over the type-1 Cu site. As a result, Tyr203 protrudes over the type-1 site and is in close proximity to Trp144. This pair of aromatic residues has been implicated in facilitating rapid electron transfer to the NiR type-1 Cu from its cupredoxin redox partner [36,37]. In contrast, the much shorter homologous loop in AMMCO possesses all the residues that coordinate the calcium ion found in the crystal structure (Figure 3-2). Similar to AMMCO, AniA contains a homologous loop (residues 175-199) that is shorter (25 residues) than that for AfNiR and this loop does not reach over to the type-1 Cu site. 3.4.3 AMMCO functional roles Calcium was required for crystallization at a concentration of at least 200 mM. Indeed, the calcium binding loop makes several crystal contacts explaining the need for calcium in the crystallization buffer. The requirement of such high concentrations of calcium suggests that this protein-metal interaction is not likely to be physiologically  89  Calcium binding loop  Type-1 Cu  Tyr203 Trp144  Figure 3-7 Surface architecture of AMMCO (left), NiR from Neisseria gonnorhea (middle), and NiR from A. faecalis (right). The top left of the AMMCO surface shows the calcium binding loop. The homologous loop in NiR contains Tyr203, which contacts Trp144 of the type-1 loop.  90 relevant. However, the same site may function in the coordination of a physiologically relevant cation such as ferric iron. Thus, a plausible role for AMMCO is a ferroxidase, which oxidizes iron(II), analogous to the MCOs ceruloplasmin and Fet3 [38]. However, no measurable ferroxidase activity was detected by the wild-type recombinant or variant enzymes although activity using an iron complex, ferrocyanide, was observed (Table 32). In the Arthrobacter genome sequence, AMMCO is colocalized with a putative transmembrane protein, annotated as an iron/zinc transporter. The transporter annotation is based on homology to the ZIP family of proteins, originally found in plants (Arabidopsis) [39] but homologues have been characterized in mammals [40] as well as S. cerevisiae [41]. Homologues of both AMMCO and the putative iron/zinc transporter are found together in the genomes of several other bacteria including Nocardia farcinica (GenBank accession YP_121950 and YP_121951), Rhodococcus erythropolis (YP_345549 and YP_345550), Rubrobacter xylanophilus (YP_642876 and YP_642877), and Xanthobacter autotrophicus (ZP_01198820 and ZP_01198821 ) suggesting a role for AMMCO in metal transport. The E192A and E192A/D190A substitutions were made to explore the functional role of the calcium binding loop. The increase in activity for both variants compared to the wild-type recombinant protein (Table 3-2) suggests that the metal binding loop may play a regulatory role. A precedent for allosteric metal regulation of MCO activity has been demonstrated for CueO, involved in copper tolerance to E. coli [42]. When free copper invades the cell, it is spontaneously reduced due to intracellular reducing conditions, and may generate the reactive oxygen species hydrogen peroxide and hydroxyl radical. CueO lowers copper toxicity by oxidizing Cu(I) into Cu(II). In the  91 presence of excess Cu(II), the phenoloxidase activity (DMP, ABTS as substrates) of CueO increases 6-7 fold [42]. Also, the addition of excess Cu(II) induces ferroxidase activity in CueO [42,43], whereas there is little or no measurable ferroxidase activity in the absence of excess copper [42,43]. A labile regulatory copper binding site in CueO was observed in the structure of a CuCl2-soaked CueO crystal [44]. The binding site consists of two Met and two Asp residues, and a weakly bound water molecule [44]. Substitution of these residues renders the protein much less active for DMP oxidation and also renders the E. coli cell less Cu2+ tolerant [44]. Surprisingly, the calcium loop variants yield different electronic absorption spectra despite being located ~18 Å from the type-1 Cu centers. Most noticeably, the 602 nm peak in the recombinant wild-type spectrum is shifted to 592 nm for the variant proteins (Figure 3-3). A second feature is an increase in absorbance in the near-ultraviolet range, starting from approximately 450 nm and increasing as the wavelength approached 300 nm. The lack of a distinct shoulder near 330 nm in wild-type recombinant protein is puzzling, given that this spectral feature has been attributed to strong coupling between the type-3 Cu centers [45]. A bridging hydroxyl group is a central component to this coupling and in the AMMCO structure a bridging solvent molecule between the type-3 Cu centers (Figure 3-2) is present. However, the type-3 coppers are 5.6 Å apart (averaged over the three sites). In comparison with other MCOs, this distance ranges from 3.7 Å in ascorbate oxidase to 4.7 Å in CueO. Consequently, the Cu-bridging solvent distances in AMMCO are 2.6 Å and 3.1 Å for the His145-His235-His280 ligated Cu and His104His143-His292 ligated Cu, respectively, averaged over the three monomers. Cu-bridging hydroxyl distances for the other MCOs are shorter ranging from 2.0 to 2.4 Å and less  92 biased toward one Cu, consistent with stronger coupling of the type-3 Cu sites. The increased absorbance of E192A and E192A/D190A AMMCO variants at 330 nm suggests a model for regulating function. The calcium binding site may control structural features at the molecular surface near the type-1 Cu site and fine tune the trinuclear site by increasing coupling of the type-3 pair. This type of allostery is unprecedented; however, a substitution of a type-1 Cu site ligand in AfNiR has resulted in allosteric regulation [46].  93 3.5 References  [1]  Williams, R.J.P. and Silva, J.J.R.F.d. (1996) The natural selection of the chemical elements : the environment and life's chemistry, Oxford University Press. Oxford.  [2]  Solomon, E.I., Chen, P., Metz, M., Lee, S.K. and Palmer, A.E. (2001). Oxygen binding, activation, and reduction to water by copper proteins. Angew Chem Int Ed Engl 40, 4570-4590.  [3]  Kakutani, T., Watanabe, H., Arima, K. and Beppu, T. (1981). A blue protein as an inactivating factor for nitrite reductase from Alcaligenes faecalis strain S-6. J Biochem (Tokyo) 89, 463-72.  [4]  Averill, B. (1996). Dissimilatory nitrite and nitric oxide reductases. Chem. Rev. 96, 2951-2964.  [5]  Nersissian, A.M. and Shipp, E.L. (2002). Blue copper-binding domains. Adv Protein Chem 60, 271-340.  [6]  Nakamura, K. and Go, N. (2005). Function and molecular evolution of multicopper blue proteins. Cell Mol Life Sci 62, 2050-66.  [7]  Murphy, M.E., Lindley, P.F. and Adman, E.T. (1997). Structural comparison of cupredoxin domains: domain recycling to construct proteins with novel functions. Protein Sci 6, 761-70.  [8]  Ryden, L.G. and Hunt, L.T. (1993). Evolution of protein complexity: the blue copper-containing oxidases and related proteins. J Mol Evol 36, 41-66.  94 [9]  Endo, K., Hayashi, Y., Hibi, T., Hosono, K., Beppu, T. and Ueda, K. (2003). Enzymological characterization of EpoA, a laccase-like phenol oxidase produced by Streptomyces griseus. J Biochem (Tokyo) 133, 671-7.  [10]  Machczynski, M.C., Vijgenboom, E., Samyn, B. and Canters, G.W. (2004). Characterization of SLAC: a small laccase from Streptomyces coelicolor with unprecedented activity. Protein Sci 13, 2388-97.  [11]  Miyazaki, K. (2003). Creating random mutagenesis libraries by megaprimer PCR of whole plasmid (MEGAWHOP). Methods Mol Biol 231, 23-8.  [12]  Collaborative Computational Project, N. (1995). The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D 50, 760-763.  [13]  Terwilliger, T.C. and Berendzen, J. (1999). Automated MAD and MIR structure solution. Acta Crystallogr. D 55, 849-861.  [14]  Perrakis, A., Morris, R. and Lamzin, V.S. (1999). Automated protein model building combined with iterative structure refinement. Nat. Struct. Biol. 6, 458463.  [15]  McRee, D.E. (1999). XtalView/Xfit--A versatile program for manipulating atomic coordinates and electron density. J. Struct. Biol. 125, 156-165.  [16]  Brenner, A.J. and Harris, E.D. (1995). A quantitative test for copper using bicinchoninic acid. Anal Biochem 226, 80-4.  [17]  de Silva, D., Davis-Kaplan, S., Fergestad, J. and Kaplan, J. (1997). Purification and characterization of Fet3 protein, a yeast homologue of ceruloplasmin. J Biol Chem 272, 14208-13.  95 [18]  Laskowski, R.A., MacArthur, M.W., Moss, D.S. and Thornton, J.M. (1993). PROCHECK: A program to check the stereochemical quality of protein structures. J. Appl. Cryst. 26, 283-291.  [19]  Kukimoto, M., Nishiyama, M., Murphy, M.E., Turley, S., Adman, E., Horinouchi, S. and Beppu, T. (1994). X-ray structure and site-directed mutagenesis of a nitrite reductase from Alcaligenes faecalis S-6: roles of two copper atoms in nitrite reduction. Biochemistry 33, 5246-5252.  [20]  Shin, W., Sundaram, U.M., Cole, J.L., Zhang, H.H., Hedman, B., Hodgson, K.O. and Solomon, E.I. (1996). Chemical and spectroscopic definition of the peroxidelevel intermediate in the multicopper oxidases: relevance to the catalytic mechansim of dioxygen reduction to water. J. Am. Chem. Soc. 118, 3202-3215.  [21]  Godden, J.W., Turley, S., Teller, D.C., Adman, E.T., Liu, M.Y., Payne, W.J. and LeGall, J. (1991). The 2.3 angstrom X-ray structure of nitrite reductase from Achromobacter cycloclastes. Science 253, 438-42.  [22]  Nakamura, K., Kawabata, T., Yura, K. and Go, N. (2003). Novel types of twodomain multi-copper oxidases: possible missing links in the evolution. FEBS Lett 553, 239-44.  [23]  Gibrat, J.F., Madej, T. and Bryant, S.H. (1996). Surprising similarities in structure comparison. Curr Opin Struct Biol 6, 377-85.  [24]  Madej, T., Gibrat, J.F. and Bryant, S.H. (1995). Threading a database of protein cores. Proteins 23, 356-69.  [25]  Boulanger, M.J. and Murphy, M.E. (2002). Crystal structure of the soluble domain of the major anaerobically induced outer membrane protein (AniA) from  96 pathogenic Neisseria: a new class of copper-containing nitrite reductases. J Mol Biol 315, 1111-27. [26]  Boulanger, M.J., Kukimoto, M., Nishiyama, M., Horinouchi, S. and Murphy, M.E. (2000). Catalytic roles for two water bridged residues (Asp-98 and His-255) in the active site of copper-containing nitrite reductase. J Biol Chem 275, 2395764.  [27]  Boulanger, M.J. and Murphy, M.E. (2003). Directing the mode of nitrite binding to a copper-containing nitrite reductase from Alcaligenes faecalis S-6: characterization of an active site isoleucine. Protein Sci 12, 248-56.  [28]  Quintanar, L., Stoj, C., Wang, T.P., Kosman, D.J. and Solomon, E.I. (2005). Role of aspartate 94 in the decay of the peroxide intermediate in the multicopper oxidase Fet3p. Biochemistry 44, 6081-91.  [29]  Lee, P.A., Tullman-Ercek, D. and Georgiou, G. (2006). The bacterial twinarginine translocation pathway. Annu Rev Microbiol 60, 373-95.  [30]  Li, C., Banfield, M.J. and Dennison, C. (2007). Engineering copper sites in proteins: loops confer native structures and properties to chimeric cupredoxins. J Am Chem Soc 129, 709-18.  [31]  Shleev, S., Tkac, J., Christenson, A., Ruzgas, T., Yaropolov, A.I., Whittaker, J.W. and Gorton, L. (2005). Direct electron transfer between copper-containing proteins and electrodes. Biosens Bioelectron 20, 2517-54.  [32]  Hall, J.F., Kanbi, L.D., Strange, R.W. and Hasnain, S.S. (1999). Role of the axial ligand in type 1 Cu centers studied by point mutations of met148 in rusticyanin. Biochemistry 38, 12675-80.  97 [33]  Hough, M.A., Ellis, M.J., Antonyuk, S., Strange, R.W., Sawers, G., Eady, R.R. and Samar Hasnain, S. (2005). High resolution structural studies of mutants provide insights into catalysis and electron transfer processes in copper nitrite reductase. J Mol Biol 350, 300-9.  [34]  Durao, P., Bento, I., Fernandes, A.T., Melo, E.P., Lindley, P.F. and Martins, L.O. (2006). Perturbations of the T1 copper site in the CotA laccase from Bacillus subtilis: structural, biochemical, enzymatic and stability studies. J Biol Inorg Chem 11, 514-26.  [35]  Kohzuma, T., Shidara, S. and Suzuki, S. (1994). Direct electrochemistry of nitrite reductase from Achromobacter cycloclastes IAM 1013. Bulletin of the Chemical Society of Japan 67, 138-43.  [36]  Barrett, M.L., Harris, R.L., Antonyuk, S., Hough, M.A., Ellis, M.J., Sawers, G., Eady, R.R. and Hasnain, S.S. (2004). Insights into redox partner interactions and substrate binding in nitrite reductase from Alcaligenes xylosoxidans: crystal structures of the Trp138His and His313Gln mutants. Biochemistry 43, 16311-9.  [37]  Yamaguchi, K., Shuta, K. and Suzuki, S. (2005). Roles of Trp144 and Tyr203 in copper-containing nitrite reductase from Achromobacter cycloclastes IAM1013. Biochem Biophys Res Commun 336, 210-4.  [38]  Kosman, D.J. (2002). FET3P, ceruloplasmin, and the role of copper in iron metabolism. Adv Protein Chem 60, 221-69.  [39]  Grotz, N., Fox, T., Connolly, E., Park, W., Guerinot, M.L. and Eide, D. (1998). Identification of a family of zinc transporter genes from Arabidopsis that respond to zinc deficiency. Proc Natl Acad Sci U S A 95, 7220-4.  98 [40]  Kambe, T., Yamaguchi-Iwai, Y., Sasaki, R. and Nagao, M. (2004). Overview of mammalian zinc transporters. Cell Mol Life Sci 61, 49-68.  [41]  Kumanovics, A., Poruk, K.E., Osborn, K.A., Ward, D.M. and Kaplan, J. (2006). YKE4 (YIL023C) encodes a bidirectional zinc transporter in the endoplasmic reticulum of Saccharomyces cerevisiae. J Biol Chem 281, 22566-74.  [42]  Kim, C., Lorenz, W.W., Hoopes, J.T. and Dean, J.F. (2001). Oxidation of phenolate siderophores by the multicopper oxidase encoded by the Escherichia coli yacK gene. J Bacteriol 183, 4866-75.  [43]  Singh, S.K., Grass, G., Rensing, C. and Montfort, W.R. (2004). Cuprous oxidase activity of CueO from Escherichia coli. J Bacteriol 186, 7815-7.  [44]  Roberts, S.A., Wildner, G.F., Grass, G., Weichsel, A., Ambrus, A., Rensing, C. and Montfort, W.R. (2003). A labile regulatory copper ion lies near the T1 copper site in the multicopper oxidase CueO. J Biol Chem 278, 31958-63.  [45]  Reinhammar, B.R. and Vanngard, T.I. (1971). The electron-accepting sites in Rhus vernicifera laccase as studied by anaerobic oxidation-reduction titrations. Eur J Biochem 18, 463-8.  [46]  Wijma, H.J., Macpherson, I., Alexandre, M., Diederix, R.E., Canters, G.W., Murphy, M.E. and Verbeet, M.P. (2006). A rearranging ligand enables allosteric control of catalytic activity in copper-containing nitrite reductase. J Mol Biol 358, 1081-93.  99 Chapter Four: Directed molecular evolution of nitrite reductase into a multicopper oxidase 4.1 Introduction Nitrite reductase (NiR) catalyzes the one-electron reduction of nitrite to nitric oxide. In vivo, a type-1 copper site of NiR relays electrons from an external electron donor, pseudoazurin, to the type-2 copper site, where nitrite is reduced to nitric oxide. This reaction is mediated by three key residues at the active site. Asp98 forms a hydrogen bond with nitrite bound to the type-2 Cu and may serve to relay protons in the overall reaction [1-3]. His255 is part of a critical water hydrogen bonding network crucial to catalysis [1,2]. Ile257 has been shown to modulate the binding mode, and thus reactivity, of nitrite to the type-2 copper [4,5]. The ability of NiR to reduce oxygen has been documented, and the product is believed to be hydrogen peroxide (H2O2) [6]. A build-up of H2O2 in the presence of reductants inactivates NiR [6], which may be a mechanism of inactivation that occurs in vivo when switching between anaerobic and aerobic respiration [7]. Multicopper oxidases (MCOs) also catalyze the reduction of oxygen, by four electrons to produce water as an end-product [8,9]. A type-1 copper site oxidizes small molecules or metal cations and transfers electrons to a type-2/type-3 trinuclear Cu cluster, where the four-electron reduction takes place. The reaction is proposed to proceed by two consecutive two-electron reductions [8-10]. The first two-electron reduction produces a peroxide-level intermediate bound to all three Cu atoms [10,11]. The second two-electron reduction produces a native intermediate, an oxo species bridged between all three Cu *A version of this chapter will be submitted for publication. MacPherson, I. S. and M. E. P. Murphy. Engineering a trinuclear Cu site into copper-containing nitrite reductase.  100 atoms, that is rapidly converted to water in the presence of reducing equivalents [8,12]. NiR and MCO are both composed of cupredoxin domains and share a common ancestor [13-15]. A duplication event is proposed in which a single domain cupredoxin became a two-domain protein. Trimerization and the addition of copper binding sites at the interface between monomers gives rise to the NiR and MCO protein families [14,15]. NiR is a well-established example of a trimeric type-1 Cu protein, and two examples of characterized two domain MCOs are described in the literature [16,17] . The crystal structure of AMMCO (Chapter 3) provides direct evidence for the close evolutionary relationship between NiR and MCO. Three histidines in NiR coordinating the type-2 Cu are found at homologous positions in both two-domain and three-domain MCOs. NiR active site residues, Asp98, His255, and Ile257 as well as two other residues, Ala137 and Val304, occur at equivalent positions to the other five MCO trinuclear Cu- coordinating histidine residues. The striking similarity between the active sites of AMMCO and NiR suggests that a modest number of amino acid substitutions could effect the required structural changes to convert one enzyme into the other. To test this hypothesis, site-directed mutagenesis and targeted random mutagenesis were used to produce variants of NiR that might have MCO activity. A high-throughput screen was used to isolate variants with oxygen reducing activity. Crystallographic analysis shows that a trinuclear Cu cluster can be incorporated into NiR using this approach. Electronic absorption spectroscopy and activity assays suggest that some of these variants have oxygen reducing function similar to that of MCOs.  101 4.2 Materials and methods 4.2.1 Mutagenesis, cloning and screening Several mutagenesis experiments were carried out in this project. A flowchart (Figure 4-1) illustrates the steps taken towards converting NiR into an MCO. Table 4-1 describes the alterations present in each variant. Variant #1. The starting point was variant #1 (D98H, A137H, I257H, V304H) which was kindly provided by Dr. Makoto Nishiyama from the University of Tokyo. This NiR construct came in a pUC vector that yields much lower levels of expression than using pET, so the variant was transferred to pET via the megaprimer method [18]. Briefly, the primers NiRfor and NiRrev (Table 4-2), corresponding to the first 21 and last 17 bases of the gene for soluble NiR, were used to amplify variant #1. This PCR product was used as a megaprimer pair with EcoR1-digested pAfNiR28a [19]. Reactions contained 1X Platinum Pfx buffer (Invitrogen), 1 mM MgSO4, 300 µM dNTPs, 75 ng digested template, 700 ng mutagenic megaprimer, 4% DMSO, and 2.5 U Platinum Pfx polymerase (Invitrogen) in a total reaction volume of 50 µl. Reactions were cycled between 95° (30 sec) and 68° (7 min) for 12-26 cycles and electroporated into E. coli DH5α for plasmid recovery and sequencing. Variant #2, Type-1 ligand loop truncation was performed in wtNiR background. The random mutagenesis experiment replaced six residues (Pro138-Pro143) from the type-1 ligand loop (residues Cys136-Met150) with two random residues for a net loss of four residues. The mutagenic reverse primer used was loopremove2 (Table 4-2) which contains six randomized bases flanked by bases corresponding to residues F132-A137 and W144-G149. This primer was used in combination with NIRFOR in a PCR  102  NiR  Ligand loop truncation ∆P138A139  SDM  SSRS  #2a Ligand loop truncation ∆N138D139  #3 Comb.  SDM/SSRS  #1  4-His substitution (D98H/A137H/ I257H/Val304His)  #4  L106G/N305T  SSRS  #5 Loop truncation ∆104-109  #6 RMS SDM  #2  Ligand loop truncation  SDM  H137R  #7 H137A  #8  Wt ligand loop (H137A)  SDM  #9 A137H  Figure 4-1 This flowchart represents the mutational lineage of this project. Each colored box represents a specific alteration to NiR. Multiple boxes indicate combinations of variations. Text within the box indicates the most recently introduced variation. Text next to arrows represents the kind of experiment. SDM = site-directed mutagenesis. SSRS = sitespecific randomization/screening. RMS = random mutagenesis/screening. A that a crystal structure was obtained for this variant.  indicates  103  Table 4-1 Grid showing the alterations present in variants #1 through #9 Alterations to NiR 138-143 D98H, Variant  ligand A137H  I257H,  #  ∆104-109 L106G  H137R  N305T  loop  (G-P-G-D)  V304H truncation #1  X  X  #2  X  #3  X  X  X  #4  X  X  X  #5  X  X  X  X  X  #6  X  X  X  X  #7  X  X  X  X  #8  X  X  X  X  X  X  #9  X  X  X  X  104  Table 4-2 Primers used for site-directed mutagenesis, site-specific randomization, and random mutagenesis Primer name  Primer sequence  Nirfor  5’ gcaactgcggcagaaatagca  Nirrev  5’ cgtgccagatggtgcga  Loopremove2  5’gcccgatacgacatgccamnnmnntgcgcagtggtagacgaa  LRND  5’ cgatacgacatgccaatcattatggcagtggtagacgaa  L106Grev  5’ ggtcagcccgccgccgccccctgcaccggttgccgcatg  303&305rand  5’ ctcgatcagattgtgmnngtgmnnggcgtagatgccggg  Alpha1rev  5' cggattgatttcggtcagcccatcgccmnnmnnggttgccgcatggaa  HCAND  5’ tacgacatgccaatcatttgcgcagtggtagacgaagac  Nirtype1  5' gcccgatacgacatgccagggaaccattccgggaggtgcgcagtggtagacgaagac  Ala137his  5’ gggaaccattcccgggggatggcagtggtagacgaagac  Bold letters represent randomized bases: m = a or c. n = any base.  105 amplification to generate a 447 bp product to be used as a megaprimer. EcoR1digested pAfNiR28a was used as a template in the megaprimer whole plasmid synthesis reaction. Reactions contained 75 ng template and 1200 ng megaprimer in a 50 µl volume. All other conditions were identical to the reaction described above. Optimized reaction product was dialyzed, electroporated into E. coli HMS174(de3), plated onto kanamycin (25 µg/ml)-IPTG (66 µM)-2YT agar plates, and incubated for 15 hours at 33-35 °C. Colonies expressing active NiR were screened by the following procedure: a colony lift was performed using 0.45 µm BiodyneA nylon membrane (Pall). The original Kan-IPTG plates were then placed at 30° to allow the colonies to re-grow. The membrane was placed, colony side up, on Whatman paper saturated with lysis solution (10 mM Tris pH 7, 2% SDS, 0.3% Tween-20, and 50 µM CuSO4) and incubated at 50° C for 30 min followed by gentle washing in 10 mM Tris pH 7.5, 100 mM NaCl. The membranes were blotted dry and submerged in screening reagent (0.3 mg/ml 3,3’diaminobenzidine (DAB) tetrahydrochloride (Sigma) and 0.5 µM horseradish peroxidase (Sigma) in 100 mM sodium phosphate pH 7.0). Red spots on the membrane corresponding to colonies expressing NiR variants able to catalyze the oxidation of DAB and reduction of oxygen. These variants were mapped to the re-grown Kan-IPTG plates and the corresponding colonies were picked and grown overnight for plasmid extraction and sequencing. Variants #2a (∆N138D139) and #2b (∆P138A139) were selected for further testing and modification (Figure 4-1). Variant #3. The fastest appearing variant in the above screen, variant #2a, was combined with variant #1 (4-His substitutions) to generate a combination clone (variant #3 in Figure 4-1). This was performed using the primers LRND (Table 4-2) and NIRFOR  106 and the pAfNiR28a-variant #1 (pAfNiR28a-v1) as a PCR template to generate a megaprimer which was then used with EcoR1 digested pAfNiR-v1 in the whole plasmid synthesis reaction, as described above. Variant #4. The Leu106Gly substitution was made using the primer L106Grev (Table 4-2) in a megaprimer reaction as described above, using pAfNiR-v3 as template in both the megaprimer generation and whole plasmid synthesis steps. Also, residues 303 and 305 were randomized using the primer 303&305rand (Table 4-2) with the same method as described above. Colonies were screened as described for variant #2, but without horseradish peroxidase. The substitution Asn305Thr was selected (variant #4 in Figure 4-1). Variant #5. Starting with the variant #4 clone, a library was constructed in which the loop containing six residues (Gly104-Gly109) was replaced by the sequence (n-n-GD) resulting in an overall truncation by 2 residues. The mutagenic primer Alpha1rev (Table 4-2) was used with the primer NIRFOR to generate the megaprimer, which was then used with pAfNiR28a-v4 plasmid digested with EcoR1 in the whole plasmid synthesis reaction. Variants were screened and analyzed by the method described for variant #4. One variant with the sequence G-P-G-D was selected (variant #5) for further mutagenesis. Variant #6. Random mutagenesis and screening were performed on variant #5. Briefly, the NIRFOR and NIRREV primers were used in a random mutagenesis PCR as described by Cadwell and Joyce [20]. This PCR product was used as megaprimer for whole-plasmid synthesis with EcoR1-digested pAfNiR28a as template. The library was  107 screened as for variant #4. Sequencing of the fastest color-developing colony yielded the variant His137Arg (variant #6 in Figure 4-1). Variants #7 through #9. Site-directed mutagenesis was performed on variant #5 using the mutagenic primer HCAND (Table 4-2) encoding a His137Ala substitution in combination with the primer NIRFOR. The result is variant #7. The mutagenic primer NIRTYPE1 (Table 4-2) was also used with the primer NIRFOR to remove both the type1 loop truncation (introduced to produce variant #2) and to mutate His137 back to Ala. The result is variant #8 (pAfNiR28a-v8). The substitution Ala137His was reinstated in variant #8 with the primer A137H (Table 4-2) to generate variant #9 (pAfNiR28a-v9). 4.2.2 Protein expression and purification Wild type and variant proteins were expressed and purified from E. coli BL21(DE3) as described previously [21] with minor modifications. Briefly, 1 L cultures containing the pAfNiR28a-variant construct were shaken at 30 °C to an OD600 of 1, followed by 30 min at 25 °C and induction with 0.5 mM IPTG and a further 15 hr at 25 °C. The cells were harvested by centrifugation, resuspended in binding buffer (20 mM sodium phosphate pH 7.8, 500 mM NaCl) supplemented with 1 mM CuSO4 and 25 mM imidazole pH 7, and lysed with an Emulsiflex C5 homogenizer (Avestin). The protein was pumped over nickel-loaded chelating Sepharose fast-flow (GE Healthcare) and washed with increasing imidazole in binding buffer, pH 6. The protein eluted at 500 mM imidazole. Eluted protein was dialyzed against 10 mM MOPS pH 7, 50 µM CuSO4 followed by 10 mM MOPS pH 7 until the calculated CuSO4 concentration was < 5 µM. The 6-his tag was cleaved with either bovine factor Xa or thrombin (Hematologics) overnight at 4 °C at a ratio of 1:500 (protease:NiR (w/w)) and the protease was removed  108 by incubation with benzamidine-NTA beads (GE Healthcare). All spectroscopically characterized variants were expressed at high levels (at least 60 mg protein/liter culture) and sufficiently stable to be concentrated to levels of at least 30 mg/ml with minimal precipitation (<1%). Pseudoazurin was expressed [22] and purified [23] as described previously. Briefly, 1 L cultures were shaken at 30 °C to an OD600 of 2.0 after which the temperature was lowered to 25 °C for 30 minutes followed by induction with 0.5 mM IPTG and further shaking overnight. Cells were harvested by centrifugation, resuspended in 20 mM sodium phosphate pH 6.3 supplemented with 10 mM CuSO4, and lysed with an Emulsiflex C5 homogenizer (Avestin). The supernatant was applied to CM Sepharose fast-flow (GE Healthcare) and eluted with increasing concentrations of NaCl. 4.2.3 Copper content Copper content was determined by the bicinchoninic acid (BCA) method [24]. Briefly, 375 µl protein samples containing approximately 25 µM Cu were denatured with 250 µl 1.83 M trichloroacetic acid and centrifuged 5 min at 16000g to remove denatured protein. The top 500 µl of the supernatant was added to 100 µl of freshly prepared 2 mM ascorbate, after which 400 µl of HEPES-BCA (0.17 mM BCA, 0.9 M NaOH, 0.66 M HEPES free acid) was added. The absorbance at 562 nm corresponding to Cu(I)-BCA chelate was recorded and compared to a standard curve made with varying concentrations (5-25 µM) of copper sulfate standard (Sigma). 4.2.4 Electronic absorption spectroscopy Absorbance scans in the range of 300-850 nm were taken using a Cary 50 spectrophotometer and a 100 µl 1 cm path length quartz cuvette. Samples were buffered  109 in 20 mM MOPS pH 7. Reductant (1 mM sodium ascorbate or L-cysteine) was added to some samples. All samples were spun at 16000g for 10 minutes to reduce light scattering before readings were taken. 4.2.5 Activity assays Activity was measured using reduced pseudoazurin. Briefly, pseudoazurin was reduced with ascorbate and the ascorbate was removed by extensive ultrafiltration in nitrogen-saturated 100 mM MES, 100 mM HEPES pH 7. Reactions (220 µl) contained 315 µM reduced pseudoazurin. For oxygen reduction reactions, NiR was added to a final concentration of 9-90 nM. Nitrite reduction reactions included 2.5 mM sodium nitrite. The increase in absorbance at 593 nm (ε = 2900 M-1cm-1) corresponding to the oxidation of pseudoazurin was monitored. Hydrogen peroxide production and oxygen consumption were measured simultaneously with an Apollo 4000 free radical analyzer (World Precision Instruments). Pseudoazurin was used at a concentration of 100 µM in 1 ml of 100 mM MES, 100 mM HEPES pH 7.0. Reactions were started with the addition of NiR to a final concentration of 3 µM (wtNiR) or 25 µM (variant #8). 4.2.6 Crystallography Crystallization trials were attempted for all variants generated. Crystals were obtained for the ligand loop truncation variant #2b (∆P138A139). The crystal conditions were 100 mM sodium acetate pH 4.5, 2 mM CuSO4 and 12% PEG 4000. The protein solution was at a concentration of 30 mg/ml and the drop ratio was 2:1 (protein:mother liquor). Crucial to crystallization was hair seeding with wtNiR crystals. Crystals were  110 transferred to mother liquor plus 30% glycerol as a cryoprotectant and immersed in liquid nitrogen for data collection at the Stanford Synchrotron Radiation Laboratory (SSRL). Crystals were obtained for variant #8. The crystal conditions were 100 mM sodium acetate pH 4.5, 2 mM CuSO4, and 6% PEG 8000. The drop ratio used was 1:1 and seeding with wild type crystals was necessary for crystallization. Once obtained, crystals were transferred to soaking solution (100 mM MOPS pH 7, 2 mM CuSO4, and 32% PEG 8000) for 30 min and then to soaking solution supplemented with 10% glycerol as a cryoprotectant before immersion in liquid nitrogen. X-rays of wavelength 0.9877 Å and 1.38 Å (Cu-edge anomalous) were used to collect data to 2.0 Å resolution at the SSRL with separate crystals. Structures were refined with Refmac [25] from the CCP4 package [26] starting with the isomorphous wild type NiR structure (PDBID 1SJM) after removal of nitrite and mutated main chain and side chain residues. Anomalous difference maps were also generated with CCP4. Model building was performed with Xfit [26]. Data collection and refinement statistics are found in Table 4-3.  4.3 Results 4.3.1 Trinuclear cluster design The polypeptide chains of domains of NiR and MCOs can be superposed with RMSD < 1 Å over 65 Cα atoms [27,28]. Comparison of the NiR and MCO active sites reveals that the main chain of four NiR residues (Asp98, Ala137, Ile257, and Val304) overlaps with trinuclear Cu His ligands. The substitutions Asp98His, Ala137His,  111 Table 4-3 Data collection and refinement statistics Crystal  Truncated  Variant #8  ligand loop  Variant #8 anomalous  (Pro-Ala) Space group  P212121  P212121  P212121  Unit cell dimensions (Å)  a = 61.0  a = 61.3  a = 61.4  b = 102.1  b = 102.2  b = 102.2  c = 146.0  c = 146.5  c = 146.3  83.6 (2.0)  83.9 (2.0)  83.9 (1.9)  0.110 (0.442)  0.104 (0.318)  0.084 (0.289)  {I}/{σI}  17.2 (2.6)  52.54 (14.5)  58.5 (9.3)  Completeness (%)  98.8 (95.3)  99.9 (100)  99.1 (94.3)  3.8 (3.2)  7.1 (7.0)  6.6 (4.1)  58583  59385  68864  26.9  26.1  27.3  0.1654/0.2119  0.1577/0.2106  0.1613/0.2060  8619/26.89  8984/26.10  9093/27.27  Protein  7616/25.47  7700/23.84  7647/24.92  Copper  6/23.81  12/30.42  12/39.15  Water  992/37.78  1272/39.73  1327/42.0  Bond lengths (Å)  0.014  0.011  0.008  Bond angles (°)  1.45  1.30  1.15  Resolution (Å) Rmerge  Redundancy No. reflections Wilson B-factor (Å2) Rwork/Rfree No. atoms / B-factors (Å2)  r.m.s. deviations  a  Highest resolution shell is shown in parentheses  112 Ile257His, Val304His (4-His variant) were made in NiR (Dr. Makoto Nishiyama, University of Tokyo, personal communication); however, this variant did not exhibit oxidase activity and was unstable. A closer inspection of the superposition of the NiR and CotA active sites suggests that additional substitutions are required to construct a trinuclear Cu cluster in NiR (Figures 4-2 and 4-3). An overview of these additional substitutions to generate an intact trinuclear cluster is given in Figure 4-1 and Table 4-1. Three features the NiR main chain were identified as likely to prevent trinuclear cluster formation (Figures 4-2 and 4-3). First, the main chain conformation of Ala137 directs Cβ such that substitution to His does not enable Cu coordination by simple rotation of side chain torsional angles. Residue 137 is part of the ligand loop that in all structurally characterized three-domain MCOs is 11 residues long, shorter than the that of NiR (15 residues) (Figure 4-3). A truncation of the ligand loop in the 4-His variant is hypothesized to alter the main chain conformation of His137 to favour Cu coordination. Second, a trinuclear cluster is considerably larger than the mononuclear type-2 Cu site in NiR. To create space for a trinuclear cluster in the constrained active site of NiR, a loop comprising residues Gly104 - Gly109 was targeted for modification. Within this loop, Leu106 was targeted for site-directed mutagenesis since it is in direct contact with a His ligand modeled at position 257. Lastly, standard side chain rotamers for His304 were poorly positioned for coordination to the trinuclear Cu cluster (Figure 4-2). Therefore, residues 303 and 305 were targeted to allow for main chain conformational change. 4.3.2 Functional assessment of NiR variants The NiR ligand loop was truncated by a total of four residues. NiR has the longest documented ligand loop of 15 residues, largely due to an abundance of proline residues  113  104-109 loop Leu106  His257  His98  Ligand loop  His137  His304  Figure 4-2 Superposition of NiR (grey, PDB entry 1SJM) with modelled histidine substitutions and CotA (orange, PDB entry 1GSK). Spheres represent copper atoms. All four modelled histidine substitutions are labelled.  114  Trp144  Active site  Ala137  Cys136  ~9Å  Figure 4-3 Superposition of the type-1 sites from wt NiR (white) and the MCO cotA (orange). This superposition suggested that truncating the ligand loop of NiR could aid in trinuclear Cu incorporation. NiR residues 136, 137, and 144 are labelled.  115 (Pro138, Pro139, and Pro143) that enable the chain from residues 138 to 143 to form several tight turns while spanning a distance that could otherwise be covered by two residues (Figure 4-3). Thus, a library of ligand loop truncation variants was generated with two randomized codons to replace Pro138-Pro143 and screened. Ten ligand loop truncation variants were selected although these variants took considerably longer to develop color (1 hr), compared to wt NiR in the same screen (2 min). Indeed, several wt NiR clones, presumably originating from contaminating undigested template in the megaprimer whole-plasmid synthesis reaction, were recovered from the screen. Two variants, one with mutation ∆N138D139 (variant #2a) and the other with mutation ∆P138/A139 (variant #2b) were overexpressed and purified and their strong blue-green color (absorbance maxima near 600 nm and 460 nm) shows that they contained intact type-1 Cu sites (Figure 4-4). Whereas the ∆N138/D139 variant did not exhibit detectable activity with reduced pseudoazurin, the ∆P138/A139 variant contained significant nitrite reductase activity with pseudoazurin (170 s-1). Oxygen reducing activity was diminished to less than 1% of wt NiR (0.1 s-1) (Table 4-4). Both variants #2a and #2b had Cu content of 1.8 and 2.0 per monomer, respectively (Table 4-4). Variant #1 (4-His) was combined with variant #2b (loop truncation, ∆N138/D139) to make variant #3 (Figure 4-1, Table 4-1). To create space for the trinuclear cluster, the mutation L106G was made in variant #3. Then, residues 303 and 305 were randomized and the library was screened to improve the position of the side chain of His304. Two of seven clones selected from the screen and sequenced contained threonine at position 305 in addition to the L106G substitution (variant #4). Variant #4 has a copper content similar to that of the previous variants (1.6 Cu per monomer) and  116  Table 4-4 kcat and copper content values for variants  Variant  Substitutions  Wt  kcat (sec-1)a  Cu content (Cu/subunit)  14.0 ± 0.3  2.0  # 2a  ∆N138D139  <0.01  2.0  # 2b  ∆P138A139  0.10 ± 0.02  1.8  <0.01  1.8  <0.01  2.2-3.0  0.25 ± 0.03  3.0  <0.01  4.0  D98H, A137H, #4  I257H, V304H, ∆N138D139, L106G/N305T D98H, A137H,  #5  I257H, V304H, ∆N138D139, N305T, ∆104-109 D98H, I257H,  #8  V304H, N305T, ∆104-109 D98H, A137H,  #9  I257H, V304H, N305T, ∆104-109  a. Pseudoazurin oxidation in the presence of O2.  117  extinction coefficient (mM -1cm-1)  3  2.5  2  1.5  1  0.5  0 300  400  500  600  700  800  wavelength (nm) Figure 4-4 Electronic absorption spectra of the two characterized type-1 loop truncation variants, as isolated. The blue line represents the ∆138N139D variant (#2a) while the red line represents the ∆138P139A variant (#2b).  118 possesses an intact type-1 Cu site as judged by visible absorption spectroscopy (Figure 4-5). Reduction of variant #4 resulted in a uniform decrease in absorbance in the range of 300-800 nm (Figure 4-5). In the variant #4 background, a library was produced in which residues 104-109 (GAGGGG) were replaced with the sequence NNGD, shortening the loop by two residues. Two of the six variants recovered from screening contained the sequence GPGD (variant #5). This variant contained a significant increase in Cu content (3.0 Cu per monomer). Also, an increase in absorbance at 330 nm (Figure 4-5) is indicative of a trinuclear cluster formation [29]. Furthermore, reduction of variant #5 with ascorbate resulted in a shift of the shoulder peak to approximately 340 nm (ε ≈ 1800 M-1cm-1) that subsequently diminished over several minutes. A similar shoulder was observed for cysteine-reduced protein (Figure 4-5), suggesting that the feature was not reductantspecific. A shoulder at 340 nm suggested the formation of a bridged peroxide intermediate, as is observed for laccase [10]. At this stage, random mutagenesis and screening was employed in an attempt to improve oxidase function of variant #5, yielding the substitution H137R (variant #6). This variant was not characterized further; however, the substitution H137A was made in variant #5 by site-directed mutagenesis to determine the role, if any, of the His137 ligand in Cu binding. The resulting variant #7 exhibits a visible absorption spectrum similar to that of variant #5 (Figure 4-5) indicating that the type-1 Cu site is not greatly perturbed. Reduction with 1 mM ascorbate resulted in a shoulder at ~340 nm similar to that observed in variant #5 (Figure 4-5); however, reduction with 1 mM cysteine did not result in a shoulder at this wavelength (data not shown).  119  0.5  A  0.6 absorbance  absorbance  0.4  0.7  0.3 0.2 0.1  B  0.5 0.4 0.3 0.2 0.1  0 300  400  500  600  700  0.0 300  800  400  0.5  700  800  D  0.4 absorbance  absorban ce  0.5  C  0.3 0.2 0.1 0.0 300  600  w avelength (nm )  w avelength (nm )  0.4  500  0.3 0.2 0.1  400  500  600  w avelength (nm )  700  800  0.0 300  400  500  600  700  w avelength (nm )  Figure 4-5 Electronic spectra of variants #4, #5, and #7. A) Reduction of variant #4. The black line represents a scan of as-isolated variant #4 (5.0 mg/ml) before reduction with 1 mM ascorbate. The red line represents the scan taken 12 minutes after reduction. B) Reduction of as-isolated variant #5 with ascorbate. The resting protein (6.2 mg/ml) is shown in black. Protein reduced with 1 mM ascorbate for 2 minutes is shown in red. C) Variant #5 reduced with cysteine. Resting enzyme (black scan) is 6.2 mg/ml. Cysteine was added to a final concentration of 1 mM and the reading (red scan) was taken after 4 minutes. D) As-isolated variant #7 reduced with ascorbate. Resting enzyme (black scan) was reduced with 1 mM ascorbate and the reading was taken after 1 minute (red scan).  800  120 Because His at position 137 is not required for either the increased absorbance at near-UV or the ~340 nm shoulder in electronic absorption spectra in the resting and reduced variant #5 background, respectively, the truncated ligand loop was reverted to the wt NiR loop (variant #8). This variant includes an alanine at position 137. Reduction of variant #8 with ascorbate yields an absorbance spectrum with an expected increase at ~340 nm (Figure 4-6). The activity and copper content of variant #8 was dependent on copper concentration in the dialysis buffer. Dialysis of variant #8 (25 µM) against 25 µM CuSO4 resulted in 3.5 Cu per monomer and O2 reduction of 0.25 s-1 (Table 4-4) when added to reduced pseudoazurin. Dialysis against higher Cu2+ concentrations resulted in greater background oxidation of pseudoazurin but did not increase catalysis. O2 consumption and H2O2 production were simultaneously measured for wt NiR and variant #8. Whereas wt protein produced H2O2 with an apparent coupling of 0.54 ± 0.05 moles H2O2 per mole O2 consumed, variant #8 showed no H2O2 production despite an apparent reduction of O2 concentration by 20 µM (Figure 4-7a and 4-7b). Note that the amount of pseudoazurin present (100 µM) was only 4 times that of variant #8 (25 µM) but ~30 times that of wt NiR (3 µM). The A137H substitution was made in variant #8 to generate variant #9. Copper content of this protein was 4.0 Cu per monomer, the highest observed in this study. The most noticeable property of variant #9 was a shift in the A600/A460 ratio from less than to greater than one immediately after adding ascorbate (Figure 4-8). Following the first scan, uniform decreases in absorbance were observed (not shown). O2 reduction with reduced pseudoazurin was not detected (less than 0.01 s-1).  121  0.5 0.45 0.4  absorbance  0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 300  400  500  600  700  800  wavelength (nm)  Figure 4-6 Electronic absorption spectra of variant #8 reduced with ascorbate. Protein was 6.8 mg/ml. Readings were taken at 2 min intervals.  122  A  280  25  275  265  15  260  10  255 5  250 245  0 0  20  40  60  uM O2  100  120  Time (s)  B  [O2] (µM)  80  280  25  275  20  270  15  265 10  260 255  5  250  0 0  20  40  60  80  Relative uM [H2H O22O ] 2(µM)  [O2] (µM) uM O  2  270  H 2O2 Relative [HuM 2O2] (µM)  20  100  Time (s) Figure 4-7 O2 consumption and H2O2 production by wt NiR (panel A) and variant #8 (panel B). The O2 curve is represented by a red line and the H2O2 curve is represented by a blue line.  123  0.3  0.25  absorbance  0.2  0.15  0.1  0.05  0 300  400  500  600  700  800  w avelength (nm )  Figure 4-8 Electronic absorption spectra of variant #9. The black line is a scan of resting protein (3 mg/ml) and the red line is a scan taken two minutes after reduction with 1 mM ascorbate. Scans thereafter showed a uniform decrease in absorbance, compared to the red scan (not shown).  124 A Trp144  Ala137  B Ala137  Asp98  C  Asp98  Figure 4-9 Stereo views of the crystal structure of variant #2b. A) ∆P138A139 ligand loop truncation (cyan) superposed with the wt type-1 site (white). B) Asp98 conformation change in the truncation variant. C) Electron density map (contoured to 1 σ) of the variant #2b active site, with oxygen modeled at the type-2 Cu atom. Oxygen atoms are red, nitrogen atoms are dark blue, sulfur atoms are yellow.  125 4.3.3 Crystal structure of ligand loop truncation (variant #2b) The crystal structure of variant #2b, the ligand loop truncation with residues P138/A139, was solved to 2.0 Å resolution (Figure 4-9). The asymmetric unit consists of the variant #2 trimer, with each chain comprised of residues Ala4 to Gly339 (chain A) or Thr5 to Gly339 (chains B and C). The overall structure of the variant is similar to the wt NiR structure (PDB entry 1SJM, RMSD = 0.18 Å over 995 Cα atoms). The main differences are localized to the active site and the type-1 Cu site. Cys136 displays a displacement of the Sγ atom by 0.35 Å perpendicular to the thiolate-Cu bond (average over the three monomers in the asymmetric unit). There is also a consistent shift of the type-1 Cu atom by 0.2 Å towards the plane formed by the cysteine and histidine ligands. In addition, there is a consistent shift of the axial Met150 Sδ atom by 0.2 Å in the same direction such that the Sδ−Cu bond length is unchanged. As a result, the best description of the geometry of the type-1 site is changed from tetrahedral in wt NiR to more trigonal pyramidal in variant #2b. Several structural perturbations occur at the active site as a result of the ligand loop truncation in variant #2b. Elongated electron density is present adjacent to the type-2 copper sites (Figure 4-9c). After modeling this density by water, residual density for an additional atom remained in the monomer B and C sites. Dioxygen models well at these sites with a side-on coordination (Cu-O bond lengths of 2.1 and 2.4 Å) and O-O distance fixed at 1.1 Å. The average B-factors for the dioxygen molecules are 37 Å2 and 47 Å2 for the B and C chains, respectively. Concurrent with the presence of elongated density for dioxygen, is a ~120° rotation about χ1 in the side chain of Asp98 such that the carboxylate moiety is ~2.2 Å further away from the type-2 Cu. A second correlated  126 conformational change was observed for His255 in which χ1 rotates ~6° and the χ2 rotates ~6°, resulting in a displacement of the Nε2 atom of ~0.35 Å. 4.3.4 Crystal structure of variant #8 The overall crystal structure of variant #8 was similar to that of the wt NiR (RMSD of 0.37 Å over 998 Cα atoms). Most striking is the observation of two additional spherical densities in the active site of the variant #8 structure. The first electron density peak occurs between the residues His98 and His255 and the second is located between His257 and the His304. Each peak was modeled as a copper atom with 50% occupancy and the average B-factors of the His98-ligated and His257-ligated Cu atoms are 44 and 33 Å2, respectively. Anomalous difference maps from data collected at the Cu-edge confirm the presence of Cu at these sites (Figure 4-10). The type-2 copper site (Cu1) in wt NiR comprising His100, His135 and His306 is intact in the variant #8 structure. The coordination of the two additional Cu atoms is accomplished by His255 and three new His residues introduced into the active site (Figure 4-10). The conformation of His98 is similar to that of Asp98 in the wt NiR structure such that the Nε2 is 2.0 Å from the new Cu atom. The Νε2 atom of His255 is displaced by ~0.5 Å toward the Cu atom to give a 2.0 Å ligand bond. The large shift in His255 is accomplished by the main chain of residues 256-260 refolding and χ2 of His255 rotating by ~40°. Side chains for His257 and His304 were readily modeled into difference electron density maps, positioning the Nε2 atoms of His257 and His304 2.2 Å and 2.0 Å, respectively, from the second additional Cu atom. Each of the new His side chains in the variant #8 structure are well ordered (average B-factor of 38 Å2 or less).  127  D98H, I257H, V304H  104-109 loop truncation  N305T His257  His304  His98  Type-1 Cu  Figure 4-10 Cu-edge anomalous map (blue, contoured to 3 σ) of variant #8 (cyan) superposed against the native NiR structure (white) showing significant density for four Cu atoms. In the stick representation, oxygen atoms are colored red, nitrogen atoms are colored dark blue, and sulphur atoms are colored yellow.  128  Resi 105 Leu106 Resi 257  Resi 98  Figure 4-11 ∆104-109 component of variant #8 (cyan) superposed against wt NiR (white). The introduced Asp of ∆104-109 is modeled at hydrogen bonding distance from His98 and His257. The cα of the wt is closer than Van der Waals distance to His257 in the variant structure.  129 Strong electron density was observed for the main chain of the truncated 104109 loop in variant #8 and the refined structure for this loop has an average B-factor of 33 Å2. Comparison with the wt NiR structure shows that ∆104-109 resulted in refolding of the protein backbone to accommodate His98 and His257 as part of the trinuclear cluster (Figure 4-11). Electron density for the introduced Asp side chain (Asp109) was weaker and the B-factor for this side chain is elevated (~46 Å2). In the predominate conformation, the carboxylate of Asp109 forms a hydrogen bond with the Nδ1 of His255 and Nε2 of His257. The replacement of Asn305 by threonine in variant #8 results in an altered backbone conformation and a side chain that participated in a new hydrogen bond network. The protein backbone of His304 and Thr305 is altered considerably as compared to wt NiR (Figure 4-12). As a result, the carbonyl backbone atom of Thr305 forms a hydrogen bond (2.9 Å) with the Νδ1 atom of His304. The Cβ atoms of both His304 and Thr305 are shifted 0.5 Å and 0.9 Å, respectively. The side chain hydroxyl of Thr305 forms a hydrogen bond to a water molecule (2.8 Å) and the side chain Nη1 atom (3.2 Å) of Arg253 (Figure 4-12). The water molecule forms hydrogen bonds with residues Glu180, Arg253, and Glu310. The water occupies the same space and forms the same hydrogen bonds as Asn305 in the wt NiR structure. 4.4 Discussion The overall similarity of the active sites of NiR and MCOs has been well documented [15,27,30]. A detailed comparison of the active site of NiR with those of homologous MCOs revealed the challenges of building a functional trinuclear cluster  130  Glu180 Glu310 Resi 304 Resi 305  Arg253 His306  Figure 4-12 Hydrogen bonding of N305T in variant #8. Variant #8 carbon atoms are shown in cyan and wt NiR carbon atoms are shown in white. The small cyan sphere is a water molecule. In the stick diagram, oxygen atoms are red, nitrogen atoms are dark blue.  131 within NiR. In addition to endowing NiR with four additional histidine residues to coordinate two additional Cu atoms, key surrounding polypeptide chains needed to be targeted for mutagenesis. Indeed, through site-directed mutagenesis and carefully selected targets for random mutagenesis and screening, a trinuclear cluster could be engineered into NiR (Figure 4-10). The trinuclear site in NiR variant #8 exhibited oxygen reduction activity yet achieving high level function has proved to be more difficult. O2 consumption / H2O2 production assays for variant #8 in Figure 4-7, although promising, must be interpreted with caution. The amount of reduced pseudoazurin was roughly four times the amount of variant #8, allowing for minimal turnover of the enzyme. This said, no H2O2 production was detected and O2 was consumed, suggesting that variant #8 is capable of the reduction of O2 to water. To the best of my knowledge, the results shown in Figure 4-7 are the first direct measurement of H2O2 production from wt NiR. It should be noted that catalase was not present, and the enzyme likely inactivated due to further reduction of H2O2 [6]. Radicals formed from this reaction could react with O2, thus skewing the inferred stoichiometry of ~0.5 moles H2O2 per mole O2 consumed. 4.4.1 Critical evaluation of the design and engineering 4.4.1.1 Type-1 ligand loop length and trinuclear Cu binding Several features in native NiR were identified as possible hindrances for trinuclear Cu site formation. Firstly, the conformation of Ala137 in the native protein positioned this residue less than ideal for trinuclear site formation, if modeled as histidine (Figure 4-2). A superposition of the ligand loops of NiR and CotA (B. subtilus) (Figure 4-  132 3) shows that the distance between Cys136 and Trp144 Cα atoms in NiR (~9.5 Å) is similar to the distance between Cys492 and Glu496 Cα atoms from CotA (~9.2 Å), suggesting that truncation of four residues in the ligand loop of NiR could allow for ideal conformation of His137. Truncation of the ligand loop proved to be less important to trinuclear formation than anticipated. First, a complete trinuclear cluster is observed in variant #8, which contains neither the ligand loop truncation nor a His137 ligand. Second, the increase in copper content to 4.0 per monomer of variant #9 supports a role of His137 in trinuclear cluster coordination in the absence of ligand loop truncation. Likely, the main chain of His137 and of neighboring residues is sufficiently flexible to allow for Cu coordination. 4.4.1.2 Residues 303 to 305 A major change in the protein backbone conformation was hypothesized to be required to position His304 for Cu coordination. Therefore, the residues adjacent to this residue, Tyr303 and Asn305, were randomized and the library was screened. The resulting variant #4 (Tyr303-His304-Thr305) was chosen since two of the seven clones characterized contained this sequence. In native NiR, the Asn305 side chain forms hydrogen bonds with four surrounding residues: the carboxylate side chains of Glu180 and Glu310, the guanidinium of Arg253, and the backbone amide of Asn307 (Figure 412). Thus, Asn305 likely served to anchor the peptide backbone near His304 and the recovered Asn305Thr substitution may have allowed the main chain surrounding residue 304 to adopt a new conformation. This rationalization was confirmed by the crystal structure of variant #8, in which the main chain of His304 and Thr305 changes, positioning His304 for trinuclear Cu  133 coordination. An unexpected feature was movement of the Thr305 backbone carbonyl and hydrogen bond formation with His304 Nδ1. Furthermore, in variant #8, residue 305 has a similar interaction to anchor the polypeptide chain despite the N305T substitution. A hydrogen bond is formed between the Thr305 side chain and Arg253. As well, a water molecule is localized in the space formerly occupied by the Asn305 Nδ2 atom. This water forms hydrogen bonds with Glu180 and Glu310 and Thr305. 4.4.1.3 Residue 104-109 loop truncation Truncation of the 104-109 loop was performed to create more space for the idealized position of His257. This variation proved to be critical to trinuclear cluster formation. Early indication of the importance of this variation came from electronic absorption spectra of the resting and reduced variants #5 and #7, which contained increased absorbance at near-UV in the resting state and formed a distinct shoulder at 340 nm upon the addition of reductant. These observations are reminiscent of the peroxidelevel intermediate observed by Shin et al after reduction of type-1 mercury-substituted laccase [10]. In contrast, variant #4 contained neither increased absorbance at near UV nor a shoulder at 340 nm upon reduction. Increased copper content of variant #5 also supported that copper cluster formation was taking place. The crystal structure of variant #8 confirmed the role of 104-109 truncation in trinuclear cluster formation. Specifically, overlap of His257 with the space formerly occupied by the Leu106 main chain suggests that the native 104-109 loop prevented Cu coordination by His257. The loop truncation was designed with Asp positioned for hydrogen bonding with His98 and His257. Although the poor electron density for the carboxylate of Asp109 implies a weak interaction with these His ligands, the negatively  134 charged aspartate may also serve to stabilize the overall positive charge of the trinuclear cluster. A role of trinuclear cluster charge stabilization by surrounding acidic residues is supported by DFT calculations [31]. 4.4.1.4 Effectiveness of the high-throughput screen The high-throughput functional screen, developed in Chapter 2 was instrumental in obtaining functional variants from randomized libraries. Despite the low level of activity in variants #2, #4, #5, and #6 when assayed using pseudoazurin as the reductant (Table 4-4), the screen still served to recover variants that folded and were capable of some function. Deposition of red, insoluble oxidized DAB at active colonies is a central component of the screen and enabled the accumulation of signal for periods of up to five hours. Alternative soluble substrates such as o-dianisidine or ABTS would likely fail to detect any activity in a colony screen, due to diffusion of the oxidized product away from the site of activity. A potential rate limiting step in the development of color in the high-throughput screen is DAB oxidation. The native type-1 site likely oxidizes DAB at a maximum rate near 0.15 sec-1, based on the rates determined for o-dianisidine with nitrite as the terminal electron acceptor (Chapter Two). Therefore, directed evolution techniques may only be able to achieve this maximal rate with a native type-1 Cu site. The M150L substitution could provide a boost in DAB oxidation by as much as 10-fold, thus putting rate-limiting pressure on catalysis and not substrate oxidation. However, DAB oxidation is not a current issue as color development of the variants during screening fails to rival that of native NiR.  135 4.4.2  Comparison between the AMMCO and variant #8 structures Crystal structures of AMMCO and variant #8 allow us to compare their trinuclear  Cu sites. The most obvious difference is the absence of one trinuclear ligand in variant #8 (recall that this variant possesses a reversion of the A137H substitution). Thus, without this third type-3 Cu ligand, the resulting coordination of the Cu to His257 and His304 is approximately linear. The lack of a third ligand is expected to lower overall stability of the trinuclear site, which may explain the need for the addition of excess Cu2+ to achieve full activity for variant #8. Another striking difference between the variant #8 and AMMCO structures is the coordination of His257 in NiR and the homologous residue, His235, in AMMCO. Whereas AMMCO His235 coordinated the Cu atom with the Nε2 atom, variant #8 His257 coordinates the Cu via its Nδ1 atom. This difference is reminiscent of the threedomain MCOs which coordinate a type-3 Cu atom with the Nδ1 atom of a histidine residue (noted in Chapter 3). In fact, if the type-1 sites of variant #8 and the three-domain MCOs were aligned, variant #8 His257 would superpose with the MCO Nδ1coordinating histidine. Negatively charged residues around the trinuclear site of MCOs have been proposed to aid in trinuclear Cu cluster stability and function by charge stabilization of the highly positively charged Cu cluster and proton relay to oxygen intermediates of catalysis, respectively [8,31,32]. Variant #8 has an introduced Asp residue in the 104-109 loop. A Glu residue in both the wt and variant #8 enzymes (Glu279) forms a hydrogen bond with a trinuclear cluster ligand, His100. In addition, variant #8 contains 4 acidic residues (Glu113, Glu180, Glu310, Glu313) located 11-15 Å from the nearest trinuclear  136 cluster Cu atom. In comparison, AMMCO contains four nearby acidic sidechains, two that form hydrogen bonds with trinuclear cluster ligand His sidechains (Asp112 and Asp255) and two within 10 Å of a trinuclear Cu atom (Glu225 and Glu288). The abundance of negatively charged residues near the engineered cluster in variant #8 could serve to stabilize the positively charged trinuclear cluster; however, the distance, solvent exposure and orientation are likely to be important factors and would require in depth characterization to determine their role, if any, in cluster stabilization. Also, the lack of an Asp-water-water-type-2 Cu network observed in both AMMCO (noted in Chapter 3) and MCOs [32] could explain the poor catalysis by variant #8. Engineered residues of variant #8 and the homologous residues of AMMCO show some similarities. First, His304 in variant #8 forms a hydrogen bond with the backbone carbonyl of Thr305. In AMMCO as well as all structurally characterized MCOs, an analogous hydrogen bond is formed between the atoms of homologous residues. Though supportive of the N305T substitution, a reversion back to an Asn as position 305 and crystallographic or copper content analysis would better indicate the effect of this substitution for trinuclear cluster formation. Another similarity between the structures are the orientations of His98 and His255 (variant #8) and His102 and His233 (AMMCO), which are the type-2 Cu ligands by classical definition. The imidazole groups of variant #8 His98 and AMMCO His102 have a similar conformation (position, torsional angles). Also, in variant #8, the imidazole ring of His257 stacks on that of the His98, analogous to the interaction between His233 and His102 in AMMCO. These commonalities suggest potential structural feature crucial for trinuclear cluster coordination.  137 4.4.3 Metal site engineering There are many examples of engineering metal sites (heme, non-heme iron, and copper) onto existing protein scaffolds (reviewed in [33]). In fact, type-1 Cu proteins have been either used as scaffolds or engineered from other scaffolds. A histidine ligand to the zinc site of CuZn superoxide dismutase was replaced with cysteine, resulting in electronic absorption and EPR spectra, as well as redox characteristics similar to that of a type-1 Cu site [34]. Type-1 ligand loops, in particular, have been the targets for mutagenesis experiments to yield novel Cu proteins. The type-1 ligand loops of both azurin and amicyanin have been replaced with an expanded loop from the CuA site of cytochrome C oxidase to yield dinuclear purple CuA sites [35,36]. To the best of my knowledge, variant #8 is the first example of a trinuclear Cu cluster engineered into any scaffold. Unlike the engineered CuA sites which involve a surface loop, the active site of NiR is nearly buried at the bottom of a narrow channel. Therefore, even conservative substitutions at the active site could have detrimental effects on protein folding and the extra charge resulting from the binding or additional Cu atoms is likely to be destabilizing. Thus, successful engineering of the trinuclear cluster may have been dependent on the high-throughput screen, which enabled the selection of stably folding variants from large libraries. 4.4.4 Dioxygen binding by variant #2b The most surprising feature of the variant #2b structure was not the ∆ P138A139 truncation but the diatomic molecule binding to the active site Cu of the enzyme. Dioxygen was modelled as the diatomic molecule at full occupancy for several reasons. First, the enzyme is known to reduce oxygen to hydrogen peroxide [6]. The substitution  138 D98N results in significantly lowered oxygen reducing ability of the enzyme (Chapter 2), indicating that oxygen reduction is likely taking place at the type-2 Cu site of NiR. By analogy with the D98N substitution, the substantially lowered oxygen reducing activity of variant #2b (<1% of wt) may be explained by the swinging out of Asp98 into a nonproductive conformation upon oxygen binding (Figure 4-8). Superposition of the new Asp98 conformation onto wt NiR reveals that the Asp98 side chain carboxylate has more freedom of motion due to the altered position of the Cβ of Ala137 (Figure 4-8b). Oxygen has been observed bound to type-2 Cu sites in crystal structures of peptidylglycine α-hydroxylating monooxygenase (PHM) [37] and amine oxidase [38]. In PHM, this species is proposed to be superoxo, in accordance with the proposed reaction mechanism in which a Cu-superoxo performs the hydrogen abstraction [39]. Unprecedented in PHM structure is the end-on nature of the oxygen adduct. In amine oxidase, the oxygen species is proposed to be Cu-hydroperoxo and is believed to be an intermediate in the reduction of oxygen to hydrogen peroxide catalyzed by the enzyme. The oxygen species of variant #2b is likely to be molecular oxygen, since no reducing equivalents were present in the crystallization buffer. Further investigation is necessary to determine the identity of the Cu adduct and the mechanism of oxygen reduction at the type-2 Cu site of NiR. 4.4.5 Future directions for this engineering project This project has opened up various avenues for further investigation. An immediate question is whether the rates of catalysis can be increased significantly (100 fold) to rival those of the traditional MCOs. In theory, the Cys136-His137-Cu bridge would likely be necessary for rapid electron transfer to the trinuclear cluster. The lack of  139 this bridge in variant #8 could partially explain the slow rates of catalysis in this variant. Therefore, a completely randomized approach may improve upon variant #9. An iteration of the mutagenesis processes used to obtain variants #2 through 5 could also improve on variant #9. Engineering of acidic residues either by random mutagenesis or site-specific randomization could improve the MCO acitivity of the engineered trinuclear cluster.  140 4.5 References  [1]  Boulanger, M.J., Kukimoto, M., Nishiyama, M., Horinouchi, S. and Murphy, M.E. (2000). Catalytic roles for two water bridged residues (Asp-98 and His-255) in the active site of copper-containing nitrite reductase. J Biol Chem 275, 2395764.  [2]  Boulanger, M.J. and Murphy, M.E. (2001). Alternate substrate binding modes to two mutant (D98N and H255N) forms of nitrite reductase from Alcaligenes faecalis S-6: structural model of a transient catalytic intermediate. Biochemistry 40, 9132-41.  [3]  Kataoka, K., Furusawa, H., Takagi, K., Yamaguchi, K. and Suzuki, S. (2000). Functional analysis of conserved aspartate and histidine residues located around the type 2 copper site of copper-containing nitrite reductase. J Biochem (Tokyo) 127, 345-50.  [4]  Boulanger, M.J. and Murphy, M.E. (2003). Directing the mode of nitrite binding to a copper-containing nitrite reductase from Alcaligenes faecalis S-6: characterization of an active site isoleucine. Protein Sci 12, 248-56.  [5]  Zhao, Y., Lukoyanov, D.A., Toropov, Y.V., Wu, K., Shapleigh, J.P. and Scholes, C.P. (2002). Catalytic function and local proton structure at the type 2 copper of nitrite reductase: the correlation of enzymatic pH dependence, conserved residues, and proton hyperfine structure. Biochemistry 41, 7464-74.  141 [6]  Kakutani, T., Watanabe, H., Arima, K. and Beppu, T. (1981). A blue protein as an inactivating factor for nitrite reductase from Alcaligenes faecalis strain S-6. J Biochem (Tokyo) 89, 463-72.  [7]  Kakutani, T., Beppu, T. and Arima, K. (1981). Regulation of nitrite reductase in the denitrifying bacterium Alcaligenes faecalis S-6. Agric. Biol. Chem. 45, 23-28.  [8]  Solomon, E.I., Sarangi, R., Woertink, J.S., Augustine, A.J., Yoon, J. and Ghosh, S. (2007). O2 and N2O activation by Bi-, Tri-, and tetranuclear Cu clusters in biology. Acc Chem Res 40, 581-91.  [9]  Solomon, E.I., Chen, P., Metz, M., Lee, S.K. and Palmer, A.E. (2001). Oxygen Binding, Activation, and Reduction to Water by Copper Proteins. Angew Chem Int Ed Engl 40, 4570-4590.  [10]  Shin, W., Sundaram, U.M., Cole, J.L., Zhang, H.H., Hedman, B., Hodgson, K.O. and Solomon, E.I. (1996). Chemical and spectroscopic definition of the peroxidelevel intermediate in the multicopper oxidases: relevance to the catalytic mechansim of dioxygen reduction to water. J. Am. Chem. Soc. 118, 3202-3215.  [11]  Sundaram, U.M., Zhang, H.H., Hedman, B., Hodgson, K.O. and Solomon, E.I. (1997). Spectroscopic investigation of peroxide binding to the trinuclear copper cluster site in laccase: Correlation with the peroxy-level intermediate and relvance to catalysis. J. Am. Chem. Soc. 119, 12525-12540.  [12]  Yoon, J., Liboiron, B.D., Sarangi, R., Hodgson, K.O., Hedman, B. and Solomon, E.I. (2007). The two oxidized forms of the trinuclear Cu cluster in the multicopper oxidases and mechanism for the decay of the native intermediate. Proc Natl Acad Sci U S A 104, 13609-14.  142 [13]  Nersissian, A.M. and Shipp, E.L. (2002). Blue copper-binding domains. Adv Protein Chem 60, 271-340.  [14]  Nakamura, K. and Go, N. (2005). Function and molecular evolution of multicopper blue proteins. Cell Mol Life Sci 62, 2050-66.  [15]  Nakamura, K., Kawabata, T., Yura, K. and Go, N. (2003). Novel types of twodomain multi-copper oxidases: possible missing links in the evolution. FEBS Lett 553, 239-44.  [16]  Endo, K., Hayashi, Y., Hibi, T., Hosono, K., Beppu, T. and Ueda, K. (2003). Enzymological characterization of EpoA, a laccase-like phenol oxidase produced by Streptomyces griseus. J Biochem (Tokyo) 133, 671-7.  [17]  Machczynski, M.C., Vijgenboom, E., Samyn, B. and Canters, G.W. (2004). Characterization of SLAC: a small laccase from Streptomyces coelicolor with unprecedented activity. Protein Sci 13, 2388-97.  [18]  Miyazaki, K. (2003). Creating random mutagenesis libraries by megaprimer PCR of whole plasmid (MEGAWHOP). Methods Mol Biol 231, 23-8.  [19]  Boulanger, M.J. (2001) The molecular mechanism of copper-containing nitrite reductase. University of British Columbia, Vancouver.  [20]  Cadwell, R. and Joyce, G. (1992). Randomization of genes by PCR mutagenesis. PCR Methods Appl. 2, 28-33.  [21]  Wijma, H.J., Macpherson, I., Alexandre, M., Diederix, R.E., Canters, G.W., Murphy, M.E. and Verbeet, M.P. (2006). A rearranging ligand enables allosteric control of catalytic activity in copper-containing nitrite reductase. J Mol Biol 358, 1081-93.  143 [22]  Wijma, H.J., Canters, G.W., de Vries, S. and Verbeet, M.P. (2004). Bidirectional catalysis by copper-containing nitrite reductase. Biochemistry 43, 10467-74.  [23]  Kukimoto, M., Nishiyama, M., Ohnuki, T., Turley, S., Adman, E.T., Horinouchi, S. and Beppu, T. (1995). Identification of interaction site of pseudoazurin with its redox partner, copper-containing nitrite reductase from Alcaligenes faecalis S-6. Protein Eng 8, 153-8.  [24]  Brenner, A.J. and Harris, E.D. (1995). A quantitative test for copper using bicinchoninic acid. Anal Biochem 226, 80-4.  [25]  Murshudov, G.N., Vagin, A.A. and Dodson, E.J. (1997). Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr 53, 240-55.  [26]  Collaborative Computational Project, N. (1995). The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D 50, 760-763.  [27]  Murphy, M.E., Lindley, P.F. and Adman, E.T. (1997). Structural comparison of cupredoxin domains: domain recycling to construct proteins with novel functions. Protein Sci 6, 761-70.  [28]  Macpherson, I.S. and Murphy, M.E. (2007). Type-2 copper-containing enzymes. Cell Mol Life Sci  [29]  Reinhammar, B.R. and Vanngard, T.I. (1971). The electron-accepting sites in Rhus vernicifera laccase as studied by anaerobic oxidation-reduction titrations. Eur J Biochem 18, 463-8.  144 [30]  Godden, J.W., Turley, S., Teller, D.C., Adman, E.T., Liu, M.Y., Payne, W.J. and LeGall, J. (1991). The 2.3 angstrom X-ray structure of nitrite reductase from Achromobacter cycloclastes. Science 253, 438-42.  [31]  Quintanar, L., Yoon, J., Aznar, C.P., Palmer, A.E., Andersson, K.K., Britt, R.D. and Solomon, E.I. (2005). Spectroscopic and electronic structure studies of the trinuclear Cu cluster active site of the multicopper oxidase laccase: nature of its coordination unsaturation. J Am Chem Soc 127, 13832-45.  [32]  Quintanar, L., Stoj, C., Wang, T.P., Kosman, D.J. and Solomon, E.I. (2005). Role of aspartate 94 in the decay of the peroxide intermediate in the multicopper oxidase Fet3p. Biochemistry 44, 6081-91.  [33]  Lu, Y., Berry, S.M. and Pfister, T.D. (2001). Engineering novel metalloproteins: design of metal-binding sites into native protein scaffolds. Chem Rev 101, 304780.  [34]  Lu, Y., Gralla, E.B., R., J.A. and Valentine, J.S. (1992). Redesign of a type 2 into a type 1 copper protein: construction and characterization of yeast copper-zinc superoxide dismutase mutants. J Am Chem Soc 114, 3560-3562.  [35]  Hay, M., Richards, J.H. and Lu, Y. (1996). Construction and characterization of an azurin analog for the purple copper site in cytochrome c oxidase. Proc Natl Acad Sci U S A 93, 461-4.  [36]  Dennison, C., Vijgenboom, E., de Vries, S., van der Oost, J. and Canters, G.W. (1995). Introduction of a CuA site into the blue copper protein amicyanin from Thiobacillus versutus. FEBS Lett 365, 92-4.  145 [37]  Prigge, S.T., Eipper, B.A., Mains, R.E. and Amzel, L.M. (2004). Dioxygen binds end-on to mononuclear copper in a precatalytic enzyme complex. Science 304, 864-7.  [38]  Wilmot, C.M., Hajdu, J., McPherson, M.J., Knowles, P.F. and Phillips, S.E. (1999). Visualization of dioxygen bound to copper during enzyme catalysis. Science 286, 1724-8.  [39]  Klinman, J.P. (2005). The copper-enzyme family of dopamine betamonooxygenase and peptidylglycine alpha-hydroxylating monooxygenase: resolving the chemical pathway for substrate hydroxylation. J Biol Chem 281, 3013-3016.  146 Chapter Five: Summary and future directions 5.1 Summary This thesis shows that site-directed and directed evolution can provide insights into the evolutionary pathways of NiR. Chapter two introduces a robust library generation and versatile screening method that can be used not only for NiR but also may be extended to multicopper oxidases. These methods were used to obtain NiR variants that were more proficient at utilizing the artificial reductant, o-dianisidine. The type-1 site axial methionine replacement with leucine was isolated from screens of random libraries and shown to provide the greatest increase in o-dianisidine oxidation (Table 2-2). Interestingly, a leucine at the axial position is found in many multicopper oxidases in their native sequences. Drawbacks to the screening method are exemplified by the slow rates of catalysis observed for o-dianisidine (and presumably DAB). The rate limiting step in the screen is substrate oxidation and this was considered in future work using the screen. Chapter two also included two key findings about oxygen reduction by NiR. First, the rate of oxygen reduction as measured by pseudoazurin reoxidation by wt NiR is 14 s-1 at pH 7 (Table 2-2). Second, Asp98 was shown to be important in oxygen reduction in addition to nitrite reduction. The most likely role of this residue is proton donation to a reduced oxygen species, paralleling its proposed role for nitrite catalysis in which protons are donated from the side chain carboxylate group [1-3]. Chapter three structurally characterizes AMMCO, a multidomain MCO with protein fold similar to NiR. By structural comparison and the assumption of the most parsimonious evolutionary path, NiR is proposed to be a result of trinuclear cluster loss from the active site of a type [C] two-domain MCO, such as AMMCO. AniA is more  147 similar to AMMCO than AfNiR, indicating that AfNiR likely arose from an AniAlike ancestor. Based on the structural and functional data, AMMCO is suggested to be part of a metal homeostasis pathway. The calcium-binding loop present in the structure is proposed to regulate AMMCO activity. Cooperation of structure-based design and directed evolution has been recognized as the most powerful method of protein engineering largely because library size is a commonly encountered limitation in projects exclusively using directed evolution [4]. Chapter Four demonstrates that a combined directed evolution and structure-based strategy could successfully engineer an active trinuclear copper site into nitrite reductase framework. Besides the structure-based histidine substitution of residues homologous to trinuclear cluster ligands, two other variations, ∆104-109 loop truncation and the N305T substitution were identified as important contributors to the assembly of the novel trinuclear cluster. This engineering experiment demonstrates the interchangeability of NiR and MCO active sites and supports the hypothesis that they are closely related. While the function of the engineered site is far from optimized, this research should inspire further investigation into the structural determinants of a functional trinuclear cluster.  5.2 Future directions Hydrogen peroxide production by NiR has been studied in this thesis, but the mechanism of oxygen reduction is still largely unknown. Furthermore, the mechanism of inactivation is also a remaining unanswered question. Investigation into these questions  148 may shed light on the greater subjects of oxygen activation at copper sites, as well as oxidative damage by free radical species. The biological role of AMMCO and related trimeric MCOs is still unclear. Investigation into the physiology of an Arthrobacter sp. FB24 AMMCO deletion mutant could shed light on the role of this enzyme in vivo. In addition, expression reporter assays under different growth stimuli could indicate AMMCO function. Crystal structures of the calcium binding loop mutants could also shed light onto the potential regulatory mechanism of this interesting feature of AMMCO. The ground-up construction of a highly functional trinuclear active site in NiR can aid in elucidatating the catalytic mechanisms of trinuclear site function in MCOs. In particular, the source of protons in the reduction of oxygen to water is still an open question.  NiR-based biosensor Enzyme-based biosensors and biofuel cells utilize the ability of redox enzymes to catalyze electron transfer between electrodes and small substrate molecules [5]. A schematic of an amperometric nitrite biosensor is shown in Figure 5-1. In such a biosensor, the concentration of substrate present is directly related to the rate of catalysis, and therefore the electric current afforded at the electrode by the enzyme, which can be measured and interpolated to determine the substrate levels of a test solution. Nitrite reductase has the ability to catalyze the reduction of nitrite to nitric oxide with electrodefed electrons or the oxidation of nitric oxide to nitrite by catalyzing the reverse reaction [6]. The most promising use of a NiR-based biosensor is the continuous, real-time measurement of nitrite or nitric oxide levels in humans, and the devices could also rival  149  H+ NO2NO2A  Electrode  H+ NO2H+  NO2-  potentiostat  +  current  2H+  e-  e-  NO  H+ NO2-  + H2O  H+ - H+ NO2 NO2-  NiR layer Figure 5-1 Schematic representation of an NiR biosensor  e-  150 those currently used on the bench top. Several studies have attempted to use copper nitrite reductase in an amperometric biosensor [7-10], with some success; however, with nitrite detection limits orders of magnitude higher than non-enzymatic methods. Effective coupling of enzyme to the electrode without loss of function is required for the optimal sensitivity of an enzyme-based biosensor. This can be accomplished by either direct contact between the enzyme and electrode (direct electron transfer), or via a small molecule mediator to carry electrons between the enzyme and electrode [5]. Direct electron transfer is the preferred method because otherwise redox mediators and enzyme can be easily lost from the system. Directed evolution has great potential for the improvement of enzyme-electrode contacting. This idea has been postulated in a review of protein engineering for bioelectrocatalysis [11], but to date, has never been used for the acquisition of variants proficient in direct electron transfer with electrodes. Nitrite reductase represents a promising enzyme for such high-throughput studies, because the enzyme is capable of producing hydrogen peroxide, which is easy to indicate visually with peroxidase and a variety of colorimetric substrates. Chapter two illustrates the sensitivity of the NiR screen, even with a poor electron donor such as DAB. Turnover levels of less than 0.004 s-1 have been effectively detected in a single E. coli colony. Given that the optimal rate of H2O2 production by NiR is 14 sec-1 (~ 3000 times faster), the detection of H2O2 efficiently produced by a NiR-electrode couple should be possible. While peroxidaseDAB will not be a suitable screening agent, as DAB serves as a substrate for NiR, peroxidase with 4-chloronaphthol or luminol could represent a method for detection of peroxide production.  151 5.3 References  [1]  Boulanger, M.J., Kukimoto, M., Nishiyama, M., Horinouchi, S. and Murphy, M.E. (2000). Catalytic roles for two water bridged residues (Asp-98 and His-255) in the active site of copper-containing nitrite reductase. J Biol Chem 275, 2395764.  [2]  Boulanger, M.J. and Murphy, M.E. (2001). Alternate substrate binding modes to two mutant (D98N and H255N) forms of nitrite reductase from Alcaligenes faecalis S-6: structural model of a transient catalytic intermediate. Biochemistry 40, 9132-41.  [3]  Kataoka, K., Furusawa, H., Takagi, K., Yamaguchi, K. and Suzuki, S. (2000). Functional analysis of conserved aspartate and histidine residues located around the type 2 copper site of copper-containing nitrite reductase. J Biochem (Tokyo) 127, 345-50.  [4]  Zhao, H. (2007). Directed evolution of novel protein functions. Biotechnol Bioeng 98, 313-7.  [5]  Habermuller, K., Mosbach, M. and Schuhmann, W. (2000). Electron-transfer mechanisms in amperometric biosensors. Fresenius J Anal Chem 366, 560-8.  [6]  Wijma, H.J., Canters, G.W., de Vries, S. and Verbeet, M.P. (2004). Bidirectional catalysis by copper-containing nitrite reductase. Biochemistry 43, 10467-74.  [7]  Astier, Y., Canters, G.W., Davis, J.J., Hill, H.A., Verbeet, M.P. and Wijma, H.J. (2005). Sensing nitrite through a pseudoazurin-nitrite reductase electron transfer relay. Chemphyschem 6, 1114-20.  152 [8]  Sasaki, S., Karube, I., Hirota, N., Arikawa, Y., Nishiyama, M., Kukimoto, M., Horinouchi, S. and Beppu, T. (1998). Application of nitrite reductase from Alcaligenes faecalis S-6 for nitrite measurement. Biosens Bioelectron 13, 1-5.  [9]  Wu, Q., Storrier, G.D., Pariente, F., Wang, Y., Shapleigh, J.P. and Abruna, H.D. (1997). A nitrite biosensor based on a maltose binding protein nitrite reductase fusion immobilized on an electropolymerized film of a pyrrole-derived bipyridinium. Anal Chem 69, 4856-63.  [10]  Zhang, J., Welinder, A.C., Hansen, A.G., Christensen, H.E.M. and Ulstrup, J. (2003). Catalytic monolayer voltammetry and in situ scanning tunneling microscopy of copper nitrite reductase on cysteamine-modified Au(111) electrodes. J. Phys. Chem. B 107, 12480-12484.  [11]  Wong, T.S. and Schwaneberg, U. (2003). Protein engineering in bioelectrocatalysis. Curr Opin Biotechnol 14, 590-6.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0066200/manifest

Comment

Related Items