The Pennsylvania State University the Graduate School the Huck

The Pennsylvania State University

The Graduate School

The Huck Institute of Life Sciences

RE-EVALUATING AND DELINEATING A PROTEIN FAMILY: BIOCHEMICAL AND BIOPHYSICAL INVESTIGATIONS OF WRBA, FOUNDING MEMBER OF A NAD(P)H:QUINONE OXIDOREDUCTASE FAMILY

A Dissertation in

Integrative Biosciences

Eric Vincent Patridge

Submitted in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

May 2009 The dissertation of Eric V. Patridge was reviewed and approved* by the following:

James G. Ferry Stanley Person Professor Dissertation Advisor Chair of Committee

J. Martin Bollinger, Jr. Associate Professor of Biochemistry and Molecular Biology Associate Professor of Chemistry

Squire Booker Associate Professor of Biochemistry and Molecular Biology Associate Professor of Chemistry

Ming Tien Professor of Biochemistry

Peter J. Hudson Willaman Professor of Biology Director of The Huck Institutes of the Life Sciences

* Signatures are on file in the Graduate School

ABSTRACT

WrbA (tryptophan W repressor binding protein) was discovered in Escherichia coli where it was proposed to play a role in regulation of the tryptophan operon; however, this has been put into question leaving the function unknown. Chapter 3 of this dissertation solidifies and defines the WrbA protein family across all domains of life. Presented with a phylogeny are the first biochemical investigations of WrbA proteins from E. coli and Archaeoglobus fulgidus, a thermophilic species from the Archaea domain. This research is the first to demonstrate that WrbA proteins have NAD(P)H:Quinone Oxidoreductase (NQO) activity. Physiologically relevant kinetic parameters presented here implicate WrbA in two-electron reduction of quinones, protecting against oxidative stress. Throughout the experiments, wild- type WrbA proteins demonstrate monomer-dimer-tetramer equilibria with one FMN cofactor per monomer in the holo-tetramer. Chapter 4 examines multimeric flavodoxin-like NAD(P)H:Quinone Oxidoreductases (NQOs) with many similarities to WrbA proteins, and a phylogeny suggests that WrbA is evolutionarily related to these NQOs. They appear to function in redox-linked processes and protect against environmental stressors, and previous research indicates that reduced dimeric NQO protein is required to protect protein targets from proteasomal degradation. Chapter 5 is the first comprehensive biophysical investigation of any flavodoxin-like NQO protein and it employs E. coli WrbA as the model protein. The study focuses on peptide interactions with the WrbA flavin cofactor, and it incorporates biochemical and biophysical characterizations of wild-type proteins and alanine variants of E. coli WrbA. Fluorescence quenching experiments demonstrate that flavin binding is cooperative and linked to multimerization in E. coli WrbA; monomer WrbA does not bind flavin. Redox potentials were determined with a gold-capillary spectroelectrochemical cell that I constructed and modified from previous investigations. (Construction of the electrochemical cell apparatus is presented in Appendix A.) The experiments show H133 directs electron accumulation to the flavin isoalloxazine ring in WrbA. Specific activities indicate that WrbA proteins retain activity with smaller electron acceptors and that the residues investigated by alanine-scanning are important for enzyme activity. Altogether, this dissertation redefines the WrbA protein family, and it has potential to significantly affect research with other flavodoxin-like NAD(P)H:Quinone Oxidoreductases.

iii

TABLE OF CONTENTS

List of Figures ...... vi List of Tables ...... ix

CHAPTER ONE: ELECTRON TRANSFER IN BIOLOGICAL SYSTEMS 1.1 Biological Roles of Electron Transfer ...... 1 1.2 Thermodynamics of Electron Transfer ...... 11 1.3 Energetic Cofactors ...... 16 1.3.1 Redox-Active Prosthetic Groups ...... 16 1.3.2 Cellular Energy Equivalents ...... 18 1.4 References ...... 24

CHAPTER TWO: FLAVINS AND FLAVOPROTEINS 2.1 Introduction to Flavin Cofactors ...... 37 2.2 Classical Flavodoxins ...... 41 2.3 Flavodoxin-like Proteins and other Flavoproteins ...... 45 2.3.1 Iron-Sulfur Flavoprotein ...... 45 2.3.2 NAD(P)H-dependent Reductases ...... 48 2.3.3 Cytochrome P450 Reductases ...... 50 2.3.4 Rubredoxin:Oxygen Oxidoreductase ...... 52 2.4 References ...... 54

CHAPTER THREE: WRBA FROM ESCHERICHIA COLI AND ARCHAEOGLOBUS FULGIDUS IS AN NAD(P)H:QUINONE OXIDOREDUCTASE 3.1 Abstract ...... 67 3.2 Introduction ...... 68 3.3 Materials and Methods ...... 70 3.4 Results ...... 76 3.5 Discussion ...... 96 3.6 Acknowledgements ...... 100 3.7 References ...... 101

CHAPTER FOUR: REVIEW: FLAVODOXIN-LIKE NAD(P)H:QUINONE REDUCTASES FUNCTIONING IN DETOXIFICATION AND REDOX-LINKED PROTEOLYSIS 4.1 Introduction ...... 108 4.1.1 NQO1 Proteins (Class I) ...... 110 4.1.2 FMN Reductase Proteins (Class II) ...... 113 4.1.3 Azoreductase Proteins (Class III) ...... 114 4.1.4 WrbA Proteins (Class IV) ...... 115 4.2 Bioinformatics Analysis of Type IV NQOs (Classes I-IV) ...... 116 4.3 Discussion ...... 121 4.3.1 Xenobiotic Detoxification ...... 122 4.3.2 Redox-Associated Proteasomal Degradation ...... 126 4.4 Future Directions ...... 129 4.5 References ...... 130

CHAPTER FIVE: CO-FACTOR BINDING AND REDOX-POTENTIAL MODULATION IN WRBA: A CLASS IV NAD(P)H REDUCTASE 5.1 Abstract ...... 143 5.2 Introduction ...... 144 5.3 Materials and Methods ...... 146 5.4 Results ...... 151 5.5 Equilibria Derivations ...... 172 5.6 Discussion ...... 180 5.7 Acknowledgements ...... 188 5.8 References ...... 189

APPENDIX A: UV-VISIBLE SPECTROELECTROCHEMISTRY OF REDOX-ACTIVE PROTEINS A.1 Introduction to Electrochemistry ...... 196 A.2 Classical Electrochemical Techniques in Research ...... 199 A.3 Design of a Fiber-Optic Gold-Capillary Electrochemical Cell ...... 204 A.4 References ...... 211

APPENDIX B: CRYSTAL STRUCTURE OF THE NADH:QUINONE OXIDOREDUCTASE WRBA FROM ESCHERICHIA COLI B.1 Personal Contribution to the Manuscript ...... 212 B.2 Article as Published ...... 213

APPENDIX C: TRYPTOPHAN REPRESSOR-BINDING PROTEINS FROM ESCHERICHIA COLI AND ARCHAEOGLOBUS FULGIDUS AS NEW CATALYSTS FOR 1,4-DIHYDRONICOTINAMIDE ADENINE DINUCLEOTIDE-DEPENDENT AMPEROMETRIC BIOSENSORS AND BIOFUEL CELLS C.1 Personal Contribution to the Manuscript ...... 220 C.2 Article as Published ...... 221

LIST OF FIGURES

Figure 1-1: One of many respiratory chains available to Escherichia coli ...... 3

Figure 1-2: Schematics of the redox-linked signaling mechanisms for I) ArcB (anoxic redox control) and for II) FNR (fumarate nitrate reductase) regulatory protein ...... 6

Figure 1-3: Mechanisms to address O2 or reactive oxygen species in anaerobes (I) and aerobes (II & III) ...... 8

Figure 1-4: Reductive half-reactions and reduction potentials of various chemicals and cofactors at pH 7.0 ...... 15

Figure 1-5: Example organic and organometallic, redox-active cofactors ...... 19

Figure 1-6: Reduced chemical structures of I) NADH and II) NADPH ...... 22

Figure 1-7: NAD+-dependent deacetylation activity of chromatin ...... 23

Figure 2-1: Flavin cofactors ...... 39

Figure 2-2: Various redox states of the typical isoalloxazine ring ...... 40

Figure 2-3: Crystal structure of flavodoxin 5NLL ...... 44

Figure 2-4: Crystal structure of Iron-sulfur flavoprotein (Isf) from Methanosarcina thermophila ...... 47

Figure 2-5: Crystal structure of WrbA from Escherichia coli ...... 49

Figure 2-6: Flavodoxin-domain incorporated into other proteins ...... 51

Figure 3-1: Alignment of WrbA, NAD(P)H:quinone oxidoreductase, and flavodoxin sequences ...... 77

Figure 3-2: Phylogenetic tree of selected WrbA, NAD(P)H:quinone oxidoreductase, and flavodoxin sequences ...... 79

Figure 3-3: Sequence alignment with WrbA sequences, Isf sequences, and misannotated WrbA sequences that have a compact cysteine motif ...... 83

Figure 3-4: Example of raw data for Dynamic Light Scattering (DLS) Analysis of WrbA Proteins ...... 85

Figure 3-5: UV-visible spectra of reconstituted, oxidized EcWrbA and AfWrbA ...... 88

Figure 3-6: DLS data from initial analyses of WrbA ...... 90

Figure 3-7: pH dependence of apparent initial velocities ...... 92

Figure 3-8: Arrhenius plot of apparent initial velocities ...... 93

Figure 3-9: Construct analysis of E. coli WT and E. coli wrbA- kan+ by internal and external primers ...... 95

Figure 4-1: Phylogram of Type IV NQOs, cured for representative sampling and for illustration ...... 117

Figure 4-2: Alignment of NQOs using sequences of the crystal structures in Figure 4-1 ..... 118

Figure 4-3: I) Redox potentials (pH 7.0) for each reactive oxygen species II) Scheme for quinone redox-cycling ...... 124

Figure 4-4: DNA and modifications caused by xenobiotics ...... 125

Figure 4-5: Equilibria that have been demonstrated for Type IV NQO proteins ...... 128

Figure 5-1: Stereo images illustrating the active site and locations of each variant residue .. 152

Figure 5-2: UV-visible absorbance spectra of reconstituted, oxidized wild type E. coli WrbA, F80A, and T116A ...... 154

Figure 5-3: UV-visible absorbance spectra of reconstituted, oxidized wild type E. coli WrbA, T116Y, and T116W ...... 155

Figure 5-4: Data handling for EcWT electrochemical titration ...... 158

Figure 5-5: Data handling for H133A electrochemical titration ...... 159

Figure 5-6: UV-visible absorbance spectra of several variant proteins, oxidized and reduced ...... 162

Figure 5-7: Fluorescence quenching behavior of E. coli WrbA (EcWrbA) and A. fulgidus WrbA (AfWrbA) ...... 164

Figure 5-8: Fluorescence quenching behavior of E. coli WrbA (EcWrbA), and alanine variants W98A, T78A, and F80A ...... 167

Figure 5-9: Fluorescence emission spectra of E. coli WrbA (EcWrbA) with limited flavin.. 168

Figure 5-10: Example of raw data collected for specific activities with each WrbA protein and each electron acceptor ...... 170

vii

Figure 5-11: Crystal structure of E. coli WrbA dimer ...... 183

Figure 5-12: Crystal structures of I) the E. coli WrbA active site and of II) the 5NLL active site ...... 184

Figure A-1: Schematic of the EG&G PAR 273 Potentiostat ...... 203

Figure A-2: Views of the electrochemical cell ...... 205

Figure A-3: Exploded top-view of the electrochemical cell and clamp ...... 206

Figure A-4: MicroCross ...... 207

Figure A-5: Gold Rod ...... 207

Figure A-6: Top, side, and front schematics for the clamp of the electrochemical cell ...... 208

Figure A-7: Side and top schematics for the flange bushing (made of carbonate) ...... 209

Figure B-1: Crystals of E. coli WrbA ...... 214

Figure B-2: Cartoon representation of WrbA from E. coli ...... 215

Figure B-3: Close-up of the active site of E. coli WrbA ...... 215

Figure B-4: Tetrameric arrangement of WrbA and binding of NADH ...... 216

Figure B-5: Binding of benzoquinone to the FMN site of WrbA ...... 217

Figure B-6: Superposition of the monomers of E. coli WrbA (red) and Methanosarcina thermophila ISF (black) ...... 217

Figure C-1: Response of AfWrbA/Os redox polymer modified electrodes blended with different fractions of SWCNTs after injection of 0.5 and 1 mM NADH...... 217

Figure C-2: Two representative cryoTEM images of SWCNTs of fraction 10 at different magnifications ...... 224

Figure C-3: CVs of an EcWrbA/Os redox polymer/SWCNTs fraction 10 modified spectrographic graphite electrode ...... 225

Figure C-4: Spectroelectochemical titration curves of EcWrbA ...... 225

Figure C-5: Polarization curve ...... 226

viii

LIST OF TABLES

Table 2-1: Variant investigations of classical flavodoxins focused in the active sites...... 43

Table 3-1: DLS data from initial analyses of WrbA ...... 85

Table 3-2: Specific activities for EcWrbA and AfWrbA conducted at 37°C and 65°C, respectively...... 89

Table 3-3: Apparent kinetic parameters for EcWrbA and AfWrbA conducted at 37°C and 65°C, respectively ...... 89

Table 4-1: Literature evidence for Type IV NQOs interacting with or conferring protection from xenobiotics ...... 111

Table 5-1: Redox potentials of wild type NQO proteins determined here and in previously published experiments ...... 160

Table 5-2: Redox potentials of the alanine variant NQO proteins, determined by potentiometric titrations as described in Materials and Methods ...... 161

Table 5-3: Binding parameters for alanine variants and wild type proteins A. fulgidus WrbA (AfWrbA) and E. coli WrbA (EcWrbA) ...... 165

Table 5-4: Specific activities for alanine variants and wild type proteins, A. fulgidus WrbA (AfWrbA) and E. coli WrbA (EcWrbA) ...... 171

Table B-1: Data collection statistics ...... 214

Table B-2: Refinement statistics ...... 214

DEDICATION

In matters of conscience, the law of majority has no place.

~ Mohandas Gandhi

It is my deepest hope that all who lack the opportunity or the permission to enhance the future might find the strength to ignore dissent when it is misplaced, to defy popular thought when necessary, to establish his/her/hir professional independence, and to build the confidence necessary to achieve constructive successes in all areas of life.

CHAPTER ONE: ELECTRON TRANSFER IN BIOLOGICAL SYSTEMS 1.1 Biological Roles of Electron Transfer 1.2 Thermodynamics of Electron Transfer 1.3 Energetic Cofactors 1.3.1 Redox-Active Prosthetic Groups 1.3.2 Cellular Energy Equivalents 1.4 References

Chapter 1

Electron Transfer in Biological Systems

There are numerous substrates available in the environment, including sunlight, and they provide various opportunities for capable organisms to derive bioenergy through substrate oxidation and light adsorption. However, organisms are only able to sustain themselves with these (sometimes minimally energetic) resources because they are notably efficient with energy transfer and they utilize the thermodynamic pressures coupled to various metabolic pathways.

Biological energy transfer mechanisms generally involve electrons; thus, investigation of electron transfer processes can provide significant insight into how organisms derive and utilize bioenergy [1-14].

1.1 Biological Roles of Electron Transfer

Electron transfer reactions function in a variety of biological roles. In the membrane they sequester bioenergy from the environment during respiration or photosynthesis, while electron transfer reactions in the cytosol derive bioenergy during fermentation reactions.

Peripheral to (and concomitant with) such processes are electron transfer reactions which function in cell signaling, detoxification, carbohydrate metabolism, and biosynthesis [1-3, 11,

14]. Processes involving chemical energy transfer and classical electron transfer are widely

1 exploited by organisms because they are so efficient; biological efficacy and efficiency provide significant advantages for an organism.

Respiration

By definition, respiratory chains function across membranes and serve to exploit the energetic redox gradients between environmental electron donors and electron acceptors.

Redox-active quinone compounds are integral to respiratory chains, since they translocate reducing equivalents across respiratory redox gradients, concomitantly translocating protons and generating a secondary energetic (proton) gradient across the plasma membrane that is later used for ATP synthesis (Figure 1-1) [1-3, 11, 14, 15]. In addition to moving electrons and protons across the membrane, quinones also serve as a collective junction along several electron transfer pathways. Capable organisms shuttle energy to the quinone pool from a variety of available energy sources such as NAD(P)H, FADH2, hydrogen (H2), glycerol-3- phosphate, metals, methane (CH4), amines (NH3), organic acids (acetate), and sulfur compounds. Electrons from the reduced quinone pool are eventually passed to various

- terminal electron acceptors such as oxygen (O2), fumarate, nitrate (NO3 ), carbon monoxide

(CO), carbon dioxide (CO2), metals, sulfur compounds, and other nitrogen compounds [1-3, 5-

7, 10, 11, 13, 14, 16-18]. Thus, via carefully managed respiratory pathways, and by taking advantage of spontaneous electron transfer to generate ion and charge gradients, organisms can derive and maximize chemical energy from oxidative environments.

[2x]

Figure 1-1. One of many respiratory chains available to Escherichia coli. Electron transfer is coupled from NADH to O2. Protons are concomitantly translocated during this electron transfer process, which creates a charge gradient useful for ATP synthesis via oxidative phosphorylation. Shown above (left to right): Type I NADH dehydrogenase, ubiquinone/ubiquinol, ubiquinol oxidase (PDB: 1FFT), and ATP synthase (PDBs: 1C17, 2L2P, 1E79, and 2A7U).

Photosynthesis

Photosystems are model protein complexes that efficiently utilize electromagnetic energy via electron excitation. In a thermodynamic exploit, electrons are spontaneously passed from water to chlorophyll in photosystem II, where light absorption augments the redox

gradient to nearly 1.8 volts as electrons are energized from the ground-state (Em, 7 = 0.78 V) to

the excited-state (Em, 7 = -1.07 V) [19, 20]. Analogous to respiratory chains, the excited electrons are then spontaneously passed to several quinone compounds, which exploit the redox gradient and translocate protons across the thylakoid membrane, forming a secondary energetic (proton) gradient useful for ATP synthesis. After passage down the redox gradient, the electrons are again excited and exploited – this time in photosystem I – and they proceed

+ down another redox gradient where they finally reduce NADP to NADPH [1, 14, 19-22].

Thus, by transforming electromagnetic energy into ion and charge gradients, photosynthesizing organisms are capable of deriving bioenergy from light.

Cell Signaling

In addition to manipulating and utilizing environmental energy, an organism must also endure the environment; to this end, cellular processes that involve electron transport and energy conversion are intimately coordinated with an organism‟s surroundings [11, 23-32].

Some well-known regulators controlling such processes function by coupling chemical oxidation/reduction and allosteric regulation. For example the FNR regulon, which regulates genes necessary for aerobic conditions, is controlled by the quaternary state of the O2- responsive FNR (fumarate nitrate reductase) regulatory protein (Figure 1-2). In the absence of

O2, insertion of [4Fe-4S] clusters induces dimerization of the FNR regulator, and in the presence of O2, the clusters collapse to [2Fe-2S] clusters and the FNR regulator returns to its

4 monomeric state. The quaternary state dictates activation and repression of genes in the FNR regulon; in this way, respiration with O2 as the terminal electron acceptor is intimately coordinated with the occurrence of O2 [31, 33-35]. In another example, transcription of genes in the ArcA-ArcB regulon is controlled by the quaternary state of the redox-responsive, membrane-bound ArcB (anoxic redox control B) protein (Figure 1-2). When the quinone pool is oxidized, intermolecular disulfide bonds induce dimer formation of the ArcB protein, which is spontaneously phosphorylated by ATP and subsequently activates the ArcA protein by phosphorylation [16, 26, 27, 29, 32]. Under reducing conditions, membrane quinols reduce the disulfide bonds and the ArcB protein returns to its monomeric state, which subsequently deactivates ArcA-P through phosphatase activity. Thus, electron transfer is coordinated with

ArcB via quinones that traverse the membrane. Functioning like physical switches, such mechanisms are effective ways to transduce signals for environmentally-dependent transitions

[33, 36].

II)

Binds DNA, Regulates Transcription

Figure 1-2. Schematics of the redox-linked signaling mechanisms for I) ArcB (anoxic redox control) and for II) FNR (fumarate nitrate reductase) regulatory protein.

Detoxification

Organisms have also developed electron transfer processes that protect against and aid in recovery from various stressor compounds by directly addressing deleterious compounds

[24, 25, 37-46]. Oxygen is among the most abundant stressors, and while it is required for aerobic life, cells have learned to cope with and to remove O2 and reactive oxygen species

(ROS) from the cytosol. For example, once considered intolerant to aerobic environments, many classically „strict‟ anaerobes have evolved NAD(P)H-dependent systems that can protect against ROS [11, 47-67]. Across all domains of life, ROS accumulate in organisms under aerobic conditions via redox-cycling or electron siphoning within the membrane, and they then damage cells by disrupting cofactors and by peroxidating DNA, proteins, and lipids [11, 23,

25, 41, 42, 46, 68-81]. By coupling electron transfer from NAD(P)H, anaerobes can eliminate

ROS without regenerating O2, while aerobes regenerate O2 during catalase and dismutase activities (Figure 1-3) [52, 54, 57, 59-62, 66, 82-100]. Oxidative environments can be addressed in this way, but other environments may require unique solutions. For example, heavy metals can interrupt redox processes by replacing metal ligands, inhibiting membrane transport, producing ROS, and siphoning electrons away from electron transport systems [101-

107]. In contrast to the direct reduction of ROS, cells are able to protect against toxic metals by translocating them across the membrane or by reducing and sequestering them for storage in proteins such as siderophores and ferritins [73, 77, 103, 108-116]. Thus, measures which protect cells from xenobiotics include removing them, metabolizing them, or restricting their intracellular interactions.

Figure 1-3. Mechanisms to address O2 or reactive oxygen species in anaerobes (I) and aerobes (II & III).

flavoprotein rubredoxin rubredoxin:O2 oxidoreductase

NAD(P)H FADred Rbred FMNred Fe-Fered H2O

Fe(Cys)4

NAD(P)+ + H+ + FADox Rbox FMNox Fe-Feox O2 + 2H

Rbred peroxidasered H2O

Fe-Fe

+ Rbox peroxidaseox H2O2 + 2H

Rbred SORred H2O2

Fe(His)4Cys

.- + O2 + 2H Rbox SORox

I) NAD(P)H-dependent pathways that remove O2 and ROS in “strict” anaerobes. As illustrated, a flavoprotein couples electron transfer from NAD(P)H, and via rubredoxin,

electrons are passed to rubredoxin:O2 oxidoreductase, to a superoxide reductase (SOR), or

to a peroxidase. No oxygen is regenerated during the consumption of O2 or ROS.

.- + 2O2 + 2H H2O2 + O2 II) Superoxide dismutase activity. Superoxide is consumed during the production of

• hydrogen peroxide. One molecule of O2 is regenerated for every two molecules of O2 ˉ.

2H2O2 2H2O + O2 III) Catalase activity. Hydrogen peroxide is consumed during the production of water.

One molecule of O2 is regenerated for every two molecules of H2O2.

Carbohydrate Metabolism

Carbohydrates are peripherally involved in classic electron transfer and yet are pivotal for the energy derived during respiration and fermentation. They generally share the approximate formula CmH2nOn and are versatile organic compounds which store chemical energy through anabolic processes [11, 117-119]. In the catabolism of carbohydrates, transformable chemical energy is relocated between carbohydrates and energy equivalents such as ATP and NAD(P)H, but the amount of available energy varies between species since there is no pathway or set of metabolites common to all organisms [11, 15, 18, 119-122]. In the common glycolysis pathway, organisms use glucose to exploit various chemical rearrangements that serve to maximize the conversion of ADP to ATP and of NAD+ to NADH with the chemical energy stored in carbohydrates. In this way, carbohydrate catabolism is coupled to electron transfer via regeneration of ATP and NADH [15, 18, 117, 118, 120-124].

Further, the reduction of succinate to fumarate is an example that more directly links the citric acid cycle with electron transport through the membrane-bound succinate dehydrogenase

(Complex II), where electron transfer is coupled from succinate to quinones. As occurs with respiration, the quinones serve as a “redox junction” where electrons enter the electron transport chain, further contributing to ATP synthesis [11, 17, 119, 123].

Biosynthesis

Analogous to carbohydrate catabolism, the anabolism of macromolecular compounds is also peripherally tied to electron transfer via ATP and NAD(P)H. However, while the catabolism of carbohydrates generally liberates energy equivalents, the biosyntheses of macromolecules assimilates those energy equivalents that were liberated elsewhere. Several carbon fixation pathways including the Calvin Cycle, the Wood-Ljungdahl Pathway, and the

Reverse Krebs Cycle exemplify how energy is used to assimilate carbon dioxide for carbohydrate anabolism [11, 12, 18, 125, 126]. Nitrogen fixation, essential for amino acid biosynthesis, is another model pathway that uses energy equivalents to assimilate environmental compounds. After reduction of atmospheric N2 to NH3, organisms can use ammonium ions in biosynthesis. Once the basic macromolecular units are formed (such as nucleotides and amino acids), ATP is employed to polymerize the basic units into macromolecules, such as with DNA replication, RNA transcription, and peptide translation

[12, 127, 128]. Thus fixations and biosynthetic pathways represent major energetic liabilities, but they are necessary and contribute to the thermodynamic pressures that encourage efficient use of bioenergy.

1.2 Thermodynamics of Electron Transfer

The biological processes discussed in Section 1.1 broadly convey the importance and pervasiveness of biological electron transfer. Investigation of electron transfer mechanisms on a molecular level requires an introduction to the laws of thermodynamics. The first law explains that energy can be transformed but that it can be neither spontaneously created nor destroyed. This phrase may seem enigmatic when taken out of context, but it lays foundation for the important concepts of energy transformation, of energy transfer, and of total energy in a system [2, 129, 130]. By applying this law to the processes discussed in Section 1.1, we can further appreciate the extent to which life is entrenched in energy gradients, and how organisms are able to sequester energy and perform work with such efficiency. The second law of thermodynamics defines entropy in terms of concepts such as disorder, heat, noise, and probability, and it provides concepts which explain essential entropic energies that contribute to the efficacy and efficiency of electron transfer [1, 2, 129, 130]. Together, these laws offer context that is necessary in order to intimately understand biological electron transfer and bioenergy derivation.

Thermodynamic pressures are most easily depicted in the context of three types of idealized systems. Isolated systems are scenarios where material and energy are completely constrained within a defined space (i.e. an insulated vessel or the entire universe). Closed systems constrain material while energy is exchanged across a defined boundary (i.e. a bottle of water warming in the sun). Open systems constrain neither material nor energy, but rather exchange them across the boundaries of the working space (i.e. all organisms). When individual molecular interactions or groups of molecular interactions are considered as closed systems (i.e. scenarios where all material reactants and products are constrained within an idealized space), equilibrium thermodynamics become extremely useful [2, 130].

A reaction‟s spontaneity can be defined in terms of Gibbs energy (G). This variable is defined as the maximum amount of non-expansion work that can be extracted from a closed system. By making a few assumptions, we can quantify a reaction‟s spontaneity at equilibrium. No work occurs in a system at equilibrium, and the change in Gibbs energy is zero (G = 0). Many forces functioning in bioenergetics can be described and quantified in terms of Gibbs energy through several distinct equations including:

G = H – T · S [1.1] G = - R · T · ln Q [1.2] G = - n · F · E [1.3] G = m · F · p [1.4] G = - Z · F ·       G = N · h ·  = N · h · c  [1.6] Equations 1.1 – 1.6. Abbreviations: enthalpy (H), temperature (T), entropy (S), gas

constant (R), reaction quotient (Q), number of electrons transferred (ne), Faraday constant (F), cell potential (E), number of protons transferred (m), proton motive force (p), numerical charge of cation (Z), membrane potential ()* , Avogadro‟s number (N), Planck constant (h), and the frequency of (), speed of (c), and wavelength of () electromagnetic radiation. *Expressed with respect to the outside (outward = positive).

Again, it is generally not possible to calculate work energy from equations [1.1] to [1.6] as they are written, but manipulation of the collection can describe many aspects of bioenergetics and can ultimately be useful in calculating the amount of work accomplished between distinct states of a system [1, 2, 14, 130-133].

Equations [1.2] and [1.3] are particularly applicable to electron transfer reactions; they relate G to the reaction quotient (Q) and the cell potential (E). By manipulating equations [1.2] and [1.3], the actual redox potential (Eh) for the redox couple

n ̶ m + oxidized + e + H  reduced is given relative to a standard redox potential (퐸0 at pH = 0) by the relationship:

0 푅푇 [표푥푖푑푖푧푒푑 ] Eh = 퐸 + · 2.3 · log [1.7] 푛퐹 10 [푟푒푑푢푐푒푑 ] 12

Thus Q and E are inter-reliant. However, for pH ≠ 0 (퐸푚, 푝퐻= 푥 ) a standard redox potential decreases as a factor of the number of protons (m) and electrons (n) in the redox couple – by 60

푚 – = 1 0.5 mV or by 30 mV per pH unit when 푛 or , respectively, according to the equation:

푅푇 푚 퐸 = 퐸0 – 2.3 · · · 푥 [1.8] 푚, 푝퐻= 푥 퐹 푛

So for the widely referenced standard hydrogen redox couple:

̶ + 2e + 2H  H2

0 ̶ + 퐸 = 0 mV at pH = 0, but for 2e / 2H reaction(s), we should remember that 퐸푚, 푝퐻= 7 (퐸푚, 7) is ~420 mV more negative than 퐸0 [1, 2, 14, 130-132].

Referencing the hydrogen reduction potential (퐸푚, 7), we can predict the spontaneity and direction of electron transfer for the redox couples in a reaction quotient by employing equations [1.3] and [1.8] under „standard conditions.‟ Electron transfer simply proceeds spontaneously from redox components for which 퐸 is more negative (or G is more positive) to those for which 퐸 is more positive (or G is more negative) (Figure 1-4). However, most reactions occur under non-standard conditions, where the reaction quotient can significantly contribute to the spontaneity of a reaction. For example, at standard concentrations the hydrolysis of ATP yields ~30.5 kJ · mol-1,

ATP + H2O  ADP + Pi

5 but as the reaction proceeds in the forward direction – towards equilibrium (Keq = 2.3 × 10 M)

– the available chemical energy decreases towards 0 kJ · mol-1, according to equation [1.2].

Conversely, if the reverse direction is pursued – towards a reaction quotient of 10-5 M (i.e. in the cytoplasm) – the available chemical energy increases to ~57 kJ · mol-1. Thus, as equations

[1.3] and [1.7] indicate, the reaction quotient is essential in order to accurately describe Eh and

G for a closed system under non-standard conditions [1, 14, 130-132]. These examples demonstrate that cytosolic electron transfer reactions are affected by many environmental factors. It should be noted that, the ability to influence biochemical reactions extends beyond pH, chemical equilibria, and redox potentials. Energetic sources such as ion gradients and membrane potentials also impact electron transfer when reactions traverse boundaries such as the plasma membrane.

Systems where many of these environmental factors converge for specific functions bring attention and clarity to the intricate bioenergetics that drive life in subtle and efficient manners. As already alluded to, the aerobic respiratory chain is one such system (Figure 1-1), and it facilitates production of ATP by exploiting several thermodynamic parameters including: the redox potentials of electron carriers that translocate protons across the membrane (equation [1.3]), the membrane potential which results from the translocation of protons (equation [1.5]), and the proton flux through ATP synthase (equation [1.4]). Finally

(and to reiterate), ATP synthase then makes itself useful in perpetuity by essentially buttressing the hydrolysis of ATP far from equilibrium while concomitantly elevating the energetic value of each ATP molecule as already described (equation [1.2]) [1, 14]. I would be remiss to refrain from mentioning that although Gibbs energy has classically been related to chemical equilibrium by equation [1.2], an article by Rosenberg and Klotz (1999) reminds us that Q is unpredictably temperature-dependent, and thus the available energy is difficult to predict via current methods without comprehensive understanding of the cellular conditions [134].

Em, pH = 7 (V)

+ - ½ O2 + 2H + 2e  H2O 0.816

- Chl a* + e  Chl a 0.780 - Fe3+ + e  Fe2+ 0.771

+ - O2 + 2H + 2e  H2O2 0.480 - + - - NO3 + 2H + 2e  NO2 + H2O 0.421

- dehydroascorbate + 2H+ + 2e  ascorbate 0.270

- ubiquinone + 2H+ + 2e  ubiquinol 0.113 - fumarate + 2H+ + 2e  succinate 0.031

- menaquinone + 2H+ + 2e  menaquinol ‒ 0.075

+ - FMN + 2H + 2e  FMNH2 ‒ 0.219 + - FAD + 2H + 2e  FADH2 ‒ 0.219 2- + - SO4 + 2H + 2e  SO2 + 2H2O ‒ 0.240

- NAD+ + 2H+ + 2e  NADH ‒ 0.320 - NADP+ + 2H+ + 2e  NADPH ‒ 0.320

+ - 2H + 2e  H2 ‒ 0.421

+ -* * Chl a + e  Chl a ‒ 1.07

Figure 1-4. Reductive half-reactions and reduction potentials of various chemicals and cofactors at pH 7.0. As values become more negative, more energy is needed for reduction.

1.3 Energetic Cofactors

The various roles for electron transfer in biology incorporate an array of distinct reaction mechanisms, including covalent catalysis, ionic interactions, dipole interactions, hydrogen bonds, charge transfer complexes, hydrophobic interactions, and van der Waals forces [39, 135]. In addition, cells have evolved the ability to use a variety of redox-active cofactors to couple electron transfer along spontaneous redox gradients towards their terminal acceptors, as in photosynthesis [1-3, 7, 11, 14, 21, 22, 39, 119, 123, 136, 137]. Moieties within active sites uniquely modulate activities and redox potentials – thus, understanding active-site cofactors and their environments is essential for a detailed investigation of electron transfer by redox-active cofactors.

1.3.1 Redox-Active Prosthetic Groups

Metal Cofactors

Transition metals function into one category of redox-active cofactors, and they play vital roles in electron transfer reactions. Their biological utility originates from their d-orbital electronic configurations, which enable them to participate in stable secondary valences.

Transition metals can be covalently coordinated by proteins through d-orbital electrons, and they are then able to use additional valences for enzyme catalysis [138-141]. The array of stable molecular geometries has enabled organisms to utilize highly-customized, redox-active, transition metal cofactors in a variety of distinct reaction mechanisms. Mononuclear iron centers, usually covalently coordinated by four cysteines [Fe(Cys)4], are among the simplest cofactors and are one-electron carriers with highly tunable redox potentials [11, 39, 142-147].

Metal cofactors generally have redox potentials between +400mV and -400mV. Other proteins can coordinate prosthetic groups such as iron-containing heme groups or complex,

16 multinucleate centers like the still enigmatic cofactor of nitrogenase, which contains seven iron atoms, one molybdenum atom, one bidentate homocitrate ligand, and a central atom that is most likely carbon, nitrogen, or oxygen [39, 148]. It is clear that metalloproteins provide organisms with numerous efficient redox reaction mechanisms under conditions that favor metal redox chemistry.

Flavin Cofactors

Flavins are another category of redox-active cofactors which function in a variety of electron transfer reactions and which sometimes replace metalloproteins when conditions are unfavorable for metal redox chemistry. As a result of substituted aromatic rings, flavin cofactors are able to participate in numerous reactions and can usually pass either one or two electrons at a time. Moreover, their redox potentials are highly tunable through covalent or electrostatic interactions at several positions around the aromatic isoalloxazine ring [39, 149-

153]. Flavin cofactors generally have redox potentials between +100mV and -500mV. In many cases, flavins are used in tandem with other prosthetic groups to perpetuate spontaneous electron transfer, such as in Complex I (NADH:Ubiquinone Oxidoreductase), where electron transfer is coupled from NADH to metal clusters through a non-covalent flavin cofactor functioning as a 1e-/2e- switch [1, 3, 11, 14, 17, 18, 37, 154]. As with metalloproteins, flavoproteins serve vital roles and provide organisms with numerous efficient reaction mechanisms, some of which will be explored with more detail in later chapters.

Other Cofactors

Although metal and flavin prosthetic groups are considered to be the major cofactors involved in electron transfer and catalysis, there are numerous examples of other redox-active

17 molecules. Such molecules include biopterin, various quinones, molybdopterin cofactors, folic acid, and a number of larger cofactors, like those containing porphyrins (Figure 1-5) [39, 124,

137, 142, 155, 156]. Collectively, the numerous common and specialized redox-active cofactors offer a significant toolset for organisms to utilize and manipulate for redox processes.

1.3.2 Cellular Energy Equivalents

Similar to, and occasionally intersecting with, redox-active cofactors are cellular energy equivalents. Generally considered as biological energetic capital, cellular energy equivalents encompass a range of molecules which traverse most metabolic processes, linking energetic processes (i.e. respirations, fermentations, photosynthesis, catabolism) to those such as anabolism, biosynthesis, catalysis, cell-signaling, and detoxification [1-3, 7, 10, 12, 14, 15, 18,

36, 39, 157, 158]. In several cases, redox-active proteins like ferredoxin and flavodoxin are considered to be energy equivalents since they provide electrons to various metabolic processes [3, 11, 159]. Some of the more widespread electron carriers (other than free flavins and quinones) include thiol compounds, phosphodiester compounds, and NAD(P)H [11, 25,

36, 39, 154, 158, 160-162].

O OH O I) III) N OH CH3O CH3 N CH3

H N N N CH3O ( CH C C CH ) 2 2 H 2 n O

NH2 N O II) O N IV) CH3 CH O N N 3 OR ( CH C C CH ) P 2 H 2 n O O O S O S Mo S O V) HO O O S O P O RO O N N NH O N O OH NH OH N N CH2 N H N VI) CH 2 H 3 C H2N N N H3C N N CH2 HC Fe CH N N H3C CH3 C H

HOOC COOH

Figure 1-5. Example organic and organometallic, redox-active cofactors. I) Tetrahydrobiopterin, II) Molybdopterin, III) Ubiquinone, IV) Menaquinone,

V) Folic acid, and VI) Heme B.

Phosphodiester Compounds

The most popular energy equivalent is ATP, and it is widely employed in a range of various molecular processes. The usefulness of phosphodiester compounds derives from the energy released in sequential hydrolyses of ATP, ADP, AMP, and poly- or pyro- phosphates

[163-165]. As already discussed, energy is stored in phosphodiester bonds by various energetic processes, and the energy they provide is augmented because cells buttress their hydrolysis far from equilibrium, favoring the reactants, according to equation [1.2]. Although phosphodiester compounds are not generally utilized in redox chemistry, several metabolic junctions exist between redox cofactors and ATP/ADP which translocate energy to classic redox cofactors such as NADH and Coenzyme A [1, 3, 11, 18, 25, 154]. Thus, phosphodiester compounds offer an opportune connection to the redox state of the cell, both because of their redox activity and because such compounds are extensively recognized and utilized in the cell.

Thiol Compounds

Thiol compounds also provide useful connections to the cellular redox state. Among the more common cellular thiols are free cysteine, glutathione, and protein thiols, and organisms typically employ them in biosynthesis, protein maintenance, protection from xenobiotic electrophiles, and cellular regulation [36, 39, 166-170]. Specifically, thiol compounds are vital for cofactor -construction and -regeneration in many metabolic processes; thus their synthesis and redox state are tightly regulated by molecules such as the NAD(P)H- dependent glutathione reductase [39, 169-171]. As with the redox-active prosthetic groups, there are also specialized thiols with unique roles in reductive catalyses, including trypanothione, mycothiol, and the popular coenzyme A [172]. Overall, thiols serve important roles in metabolism and their oxidation states are useful for indicating the cellular redox state

20 while they translocate energy equivalents from NADPH or ATP at various metabolic junctions

[158, 172, 173].

NAD(P)H

Arguably, NAD(P)H is the most important bioactive energy equivalent for cellular redox activity (Figure 1-6). As with both canonical phosphodiester and thiol compounds, the generation and utilization of reduced NAD(P)H provides a widely accessible linkage between metabolic processes and the redox state of the cell [124, 154, 160, 162, 174]. NADPH provides electrons for biosynthetic processes while NADH provides electrons for numerous metabolic processes, such as its role in driving ATP synthesis by translocating protons via Complex I and

Complex IV (Figure 1-1) [1, 3, 11, 12, 14, 18, 19, 21, 39, 154, 160, 162]. However, the role of

NADH doesn‟t stop at electron transfer – it functions in various processes. For example,

NADH molecules can provide reductive protection against xenobiotic compounds (Figure 1-3).

They can also directly or indirectly act to signal specific metabolic regulation such as by affecting the redox state of the quinone pool (Figure 1-2) [43, 158, 161]. Interestingly, sirtuin proteins directly link NAD+ metabolism with various cellular functions in a unique manner.

Sirtuins cleave NAD+ and concomitantly deacetylate proteins, but they are inhibited by either

NADH or nicotinamide. While cleaving NAD+, some sirtuins can also label proteins with

ADP-ribose. In moderation, these activities contribute to cellular processes such as gene silencing, regulation, and condensation of chromatin (Figure 1-7). When sirtuins are inhibited, longevity generally decreases and replicative senescence increases [158, 175-178]. Therefore, in conjunction with biosynthesis of NAD(P)+ from tryptophan or aspartate, the

NAD(P)+/NAD(P)H ratio offers an intricate relationship between cellular metabolism and the redox state of the cell.

I) HO O OH O O P O P O HO N O OH O HO NH2 N N OH O N N NH 2

II)

HO O OH OH O O P O P O N HO P O O OH O HO O NH2 N N OH O N N

NH 2

Figure 1-6. Reduced chemical structures of I) NADH and II) NADPH. It is generally accepted that oxidation occurs through a hydride transfer from the 4-position of the substituted dihydropyridine (nicotinamide) ring.

NAD+

sirtuin activity

acetylation o-acetyl-ADP + nicotinamide

II)

Figure 1-7. NAD+-dependent deacetylation activity of chromatin. I) Active chromatin cycles between various states. Sirtuins affect chromatin structure and cellular processes by deacetylation or ADP-ribosylation of proteins. Sirtuin-mediated deacetylation affects regulation, increases longevity, and decreases senescence. However, when chromatin is severely hypoacetylated, sirtuins become detrimental for growth and can induce heterochromatin formation. II) Heterochromatin forms over multiple generations, perpetuates itself, and is generally hypoacetylated. This inactive chromatin decreases longevity and increases senescence.

1.4 References

1. Nicholls, D. and S. Ferguson, Bioenergetics 3. 2nd ed. (2002), San Diego: Elsevier Science, Ltd.

2. Morowitz, H.J., Foundations of Bioenergetics. (1978), New York: Academic Press. xix, 344 p.

3. Gottschalk, G., Bacterial metabolism. 2nd ed. Springer series in microbiology. (1986), New York: Springer-Verlag. xiii, 359 p.

4. Finneran, K.T., et al. (2002). Desulfitobacterium metallireducens sp. nov., an anaerobic bacterium that couples growth to the reduction of metals and humic acids as well as chlorinated compounds. Int J Syst Evol Microbiol. 52(Pt 6): p. 1929-35.

5. Cervantes, F.J., et al. (2002). Reduction of humic substances by halorespiring, sulphate- reducing and methanogenic microorganisms. Environ Microbiol. 4(1): p. 51-7.

6. Bond, D.R. and D.R. Lovley (2002). Reduction of Fe(III) oxide by methanogens in the presence and absence of extracellular quinones. Environ Microbiol. 4(2): p. 115-24.

7. Berry, S. (2002). The chemical basis of membrane bioenergetics. J Mol Evol. 54(5): p. 595-613.

8. Smith, A.J. and D.S. Hoare (1977). Specialist phototrophs, lithotrophs, and methylotrophs: a unity among a diversity of procaryotes? Bacteriol Rev. 41(2): p. 419- 48.

9. Taylor, R.R., K. Mohan, and R.J. Miles (1996). Diversity of energy-yielding substrates and metabolism in avian mycoplasmas. Vet Microbiol. 51(3-4): p. 291-304.

10. Thauer, R.K., K. Jungermann, and K. Decker (1977). Energy conservation in chemotrophic anaerobic bacteria. Bacteriol Rev. 41(1): p. 100-80.

11. Ljungdahl, L.G., Biochemistry and physiology of anaerobic bacteria. (2003), New York: Springer. xvii, 270 p.

12. Russell, J.B. and G.M. Cook (1995). Energetics of bacterial growth: balance of anabolic and catabolic reactions. Microbiol Rev. 59(1): p. 48-62.

13. Auernik, K.S. and R.M. Kelly (2008). Identification of components of electron transport chains in the extremely thermoacidophilic crenarchaeon Metallosphaera sedula through iron and sulfur compound oxidation transcriptomes. Appl Environ Microbiol. 74(24): p. 7723-32.

14. Lipmann, F.A. and Ciba Foundation., Energy transformation in biological systems. Ciba Foundation symposium, new ser., 31. (1975), Amsterdam ; New York: Elsevier. x, 416 p.

15. Koebmann, B.J., et al. (2002). The glycolytic flux in Escherichia coli is controlled by the demand for ATP. J Bacteriol. 184(14): p. 3909-16.

16. Malpica, R., et al. (2004). Identification of a quinone-sensitive redox switch in the ArcB sensor kinase. Proc Natl Acad Sci U S A. 101(36): p. 13318-23.

17. Iverson, T.M., et al. (1999). Structure of the Escherichia coli fumarate reductase respiratory complex. Science. 284(5422): p. 1961-6.

18. Anraku, Y. (1988). Bacterial electron transport chains. Annu Rev Biochem. 57: p. 101- 32.

19. Kê, B., Photosynthesis : photobiochemistry and photobiophysics. Advances in photosynthesis v. 10. (2001), Dordrecht ; Boston: Kluwer Academic Publishers. xxx, 763 p.

20. Metzner, H., Solar energy conversion by plant photosynthesis in Advances in Solid State Physics. (1978), Springer Berlin. p. 33-52.

21. Chitnis, P.R. (2001). Photosystem I: Function and Physiology. Annu Rev Plant Physiol Plant Mol Biol. 52: p. 593-626.

22. Fromme, P. and I. Grotjohann (2008). Structure of Photosystems I and II. Results Probl Cell Differ. 45: p. 33-72.

23. Vranova, E., D. Inze, and F. Van Breusegem (2002). Signal transduction during oxidative stress. J Exp Bot. 53(372): p. 1227-36.

24. Oh, J.I. and S. Kaplan (2000). Redox signaling: globalization of gene expression. EMBO J. 19(16): p. 4237-47.

25. Stubberfield, C.R. and G.M. Cohen (1989). Interconversion of NAD(H) to NADP(H). A cellular response to quinone-induced oxidative stress in isolated hepatocytes. Biochem Pharmacol. 38(16): p. 2631-7.

26. Bauer, C.E., S. Elsen, and T.H. Bird (1999). Mechanisms for redox control of gene expression. Annu Rev Microbiol. 53: p. 495-523.

27. Georgellis, D., O. Kwon, and E. Lin (2001). Quinones as the Redox Signal for the Arc Two-Component System of Bacteria. Science. 292: p. 2314-2316.

28. Withers, H., S. Swift, and P. Williams (2001). Quorum sensing as an integral component of gene regulatory networks in Gram-negative bacteria. Curr Opin Microbiol. 4(2): p. 186-93.

29. Liu, X. and P. De Wulf (2004). Probing the ArcA-P modulon of Escherichia coli by whole genome transcriptional analysis and sequence recognition profiling. J Biol Chem. 279(13): p. 12588-97.

30. Iuchi, S. and E.C. Lin (1988). arcA (dye), a global regulatory gene in Escherichia coli mediating repression of enzymes in aerobic pathways. Proc Natl Acad Sci U S A. 85(6): p. 1888-92.

31. Kang, Y., et al. (2005). Genome-wide expression analysis indicates that FNR of Escherichia coli K-12 regulates a large number of genes of unknown function. J Bacteriol. 187(3): p. 1135-60.

32. Mika, F. and R. Hengge (2005). A two-component phosphotransfer network involving ArcB, ArcA, and RssB coordinates synthesis and proteolysis of sigmaS (RpoS) in E. coli. Genes Dev. 19(22): p. 2770-81.

33. Demple, B. (1997). Study of redox-regulated transcription factors in prokaryotes. Methods. 11(3): p. 267-78.

34. Kiley, P.J. and H. Beinert (1998). Oxygen sensing by the global regulator, FNR: the role of the iron-sulfur cluster. FEMS Microbiol Rev. 22(5): p. 341-52.

35. Outten, F.W. (2007). Iron-sulfur clusters as oxygen-responsive molecular switches. Nat Chem Biol. 3(4): p. 206-7.

36. Lockwood, T.D. (2000). Redox control of protein degradation. Antioxid Redox Signal. 2(4): p. 851-78.

37. Cenas, N., et al. (1994). The electron transfer reactions of NADPH: cytochrome P450 reductase with nonphysiological oxidants. Arch Biochem Biophys. 315(2): p. 400-6.

38. Díaz, E., Microbial biodegradation : genomics and molecular biology. (2008), Norfolk, U.K.: Caister Academic Press. xiv, 402 p.

39. Banerjee, R., Redox biochemistry. (2008), Hoboken, N.J.: Wiley-Interscience. xx, 317 p.

40. Rotenberg Iu, S. (1982). Classification of xenobiotics according to their effect on mitochondrial enzymatic systems. Biull Eksp Biol Med. 94(9): p. 42-5.

41. Olczak, A.A., G. Wang, and R.J. Maier (2005). Up-expression of NapA and other oxidative stress proteins is a compensatory response to loss of major Helicobacter pylori stress resistance factors. Free Radic Res. 39(11): p. 1173-82.

42. Fang, Y.Z., S. Yang, and G. Wu (2002). Free radicals, antioxidants, and nutrition. Nutrition. 18(10): p. 872-9.

43. Boersma, M.G., et al. (1994). NADPH-cytochrome reductase catalysed redox cycling of 1,4-benzoquinone; hampered at physiological conditions, initiated at increased pH values. Biochem Pharmacol. 47(11): p. 1949-55.

44. Hukkanen, J., et al. (2002). Expression and regulation of xenobiotic-metabolizing cytochrome P450 (CYP) enzymes in human lung. Crit Rev Toxicol. 32(5): p. 391-411.

45. Rafii, F. and C.E. Cerniglia (1995). Reduction of azo dyes and nitroaromatic compounds by bacterial enzymes from the human intestinal tract. Environ Health Perspect. 103 Suppl 5: p. 17-9.

46. Rafii, F., J.D. Hall, and C.E. Cerniglia (1997). Mutagenicity of azo dyes used in foods, drugs and cosmetics before and after reduction by Clostridium species from the human intestinal tract. Food Chem Toxicol. 35(9): p. 897-901.

47. Gomes, C.M., et al. (2000). Spectroscopic studies and characterization of a novel electron-transfer chain from Escherichia coli involving a flavorubredoxin and its flavoprotein reductase partner. Biochemistry. 39(51): p. 16230-7.

48. Vicente, J.B., et al. (2002). Module fusion in an A-type flavoprotein from the cyanobacterium Synechocystis condenses a multiple-component pathway in a single polypeptide chain. Biochem Biophys Res Commun. 294(1): p. 82-7.

49. Coulter, E.D., N.V. Shenvi, and D.M. Kurtz, Jr. (1999). NADH peroxidase activity of rubrerythrin. Biochem Biophys Res Commun. 255(2): p. 317-23.

50. Frazao, C., et al. (2000). Structure of a dioxygen reduction enzyme from Desulfovibrio gigas. Nat Struct Biol. 7(11): p. 1041-5.

51. Rodrigues, J.V., et al. (2005). Rubredoxin acts as an electron donor for neelaredoxin in Archaeoglobus fulgidus. Biochem Biophys Res Commun. 329(4): p. 1300-5.

52. Abreu, I.A., et al. (2001). The mechanism of superoxide scavenging by Archaeoglobus fulgidus neelaredoxin. J Biol Chem. 276(42): p. 38995-9001.

53. Rodrigues, J.V., et al. (2007). Superoxide reduction by Archaeoglobus fulgidus desulfoferrodoxin: comparison with neelaredoxin. J Biol Inorg Chem. 12(2): p. 248-56.

54. Abreu, I.A., et al. (2000). Oxygen detoxification in the strict anaerobic archaeon Archaeoglobus fulgidus: superoxide scavenging by neelaredoxin. Mol Microbiol. 38(2): p. 322-34.

55. Kawasaki, S., et al. (2009). O2 and reactive oxygen species detoxification complex, composed of O2-responsive NADH:rubredoxin oxidoreductase-flavoprotein A2- desulfoferrodoxin operon enzymes, rubperoxin, and rubredoxin, in Clostridium acetobutylicum. Appl Environ Microbiol. 75(4): p. 1021-9.

56. Wildschut, J.D., et al. (2006). Rubredoxin:oxygen oxidoreductase enhances survival of Desulfovibrio vulgaris hildenborough under microaerophilic conditions. J Bacteriol. 188(17): p. 6253-60.

57. Weinberg, M.V., et al. (2004). Rubrerythrin from the hyperthermophilic archaeon Pyrococcus furiosus is a rubredoxin-dependent, iron-containing peroxidase. J Bacteriol. 186(23): p. 7888-95.

58. Victor, B.L., et al. (2003). Docking and electron transfer studies between rubredoxin and rubredoxin:oxygen oxidoreductase. J Biol Inorg Chem. 8(4): p. 475-88.

59. Silaghi-Dumitrescu, R., et al. (2003). A flavodiiron protein and high molecular weight rubredoxin from Moorella thermoacetica with nitric oxide reductase activity. Biochemistry. 42(10): p. 2806-15.

60. Silva, G., et al. (2001). Analysis of the Desulfovibrio gigas transcriptional unit containing rubredoxin (rd) and rubredoxin-oxygen oxidoreductase (roo) genes and upstream ORFs. Biochem Biophys Res Commun. 280(2): p. 491-502.

61. Das, A., et al. (2001). Five-gene cluster in Clostridium thermoaceticum consisting of two divergent operons encoding rubredoxin oxidoreductase- rubredoxin and rubrerythrin-type A flavoprotein- high-molecular-weight rubredoxin. J Bacteriol. 183(5): p. 1560-7.

62. Chen, L., et al. (1993). Purification and characterization of an NADH-rubredoxin oxidoreductase involved in the utilization of oxygen by Desulfovibrio gigas. Eur J Biochem. 216(2): p. 443-8.

63. Moura, I., et al. (1990). Purification and characterization of desulfoferrodoxin. A novel protein from Desulfovibrio desulfuricans (ATCC 27774) and from Desulfovibrio vulgaris (strain Hildenborough) that contains a distorted rubredoxin center and a mononuclear ferrous center. J Biol Chem. 265(35): p. 21596-602.

64. May, S.W., B.J. Abbott, and A. Felix (1973). On the role of superoxide in reactions catalyzed by rubredoxin of Pseudomonas oleovorans. Biochem Biophys Res Commun. 54(4): p. 1540-5.

65. Vicente, J.B., et al. (2008). Kinetic characterization of the Escherichia coli nitric oxide reductase flavorubredoxin. Methods Enzymol. 437: p. 47-62.

66. Vicente, J.B. and M. Teixeira (2005). Redox and spectroscopic properties of the Escherichia coli nitric oxide-detoxifying system involving flavorubredoxin and its NADH-oxidizing redox partner. J Biol Chem. 280(41): p. 34599-608.

67. Mills, P.C., et al. (2005). Detoxification of nitric oxide by the flavorubredoxin of Salmonella enterica serovar Typhimurium. Biochem Soc Trans. 33(Pt 1): p. 198-9.

68. Buettner, G.R. (1993). The pecking order of free radicals and antioxidants: lipid peroxidation, alpha-tocopherol, and ascorbate. Arch Biochem Biophys. 300(2): p. 535- 43.

69. Massey, V. (1994). Activation of molecular oxygen by flavins and flavoproteins. J Biol Chem. 269(36): p. 22459-62.

70. Chignell, C.F. (1985). Structure-activity relationships in the free-radical metabolism of xenobiotics. Environ Health Perspect. 61: p. 133-7.

71. Park, S.-H., Y.-J. Ko, and C.-K. Kim (2001). Toxic Effects of Catechol and 4- Chlorobenzoate Stresses on Bacterial Cells. Journal of Microbiology. 39(3): p. 206- 212.

72. Tilley, G.J., et al. (2001). Influence of electrochemical properties in determining the sensitivity of [4Fe-4S] clusters in proteins to oxidative damage. Biochem J. 360(Pt 3): p. 717-26.

73. Ceci, P., et al. (2003). The Dps protein of Agrobacterium tumefaciens does not bind to DNA but protects it toward oxidative cleavage: x-ray crystal structure, iron binding, and hydroxyl-radical scavenging properties. J Biol Chem. 278(22): p. 20319-26.

74. Dukan, S. and T. Nystrom (1999). Oxidative stress defense and deterioration of growth- arrested Escherichia coli cells. J Biol Chem. 274(37): p. 26027-32.

75. Al-Maghrebi, M.A. and L.T. Benov (2001). Polyphosphate accumulation and oxidative DNA damage in superoxide dismutase-deficient Escherichia coli. Free Radic Biol Med. 31(11): p. 1352-9.

76. Sjölin, A. and D. Livingstone (1997). Redox cycling of aromatic hydrocarbon quinones catalysed by digestive gland microsomes of the common mussel (Mytilus edulis L.). Aquatic Toxicology. 38: p. 83-99.

77. Nair, S. and S.E. Finkel (2004). Dps protects cells against multiple stresses during stationary phase. J Bacteriol. 186(13): p. 4192-8.

.- 78. Fridovich, I. (1997). Superoxide anion radical (O2 ), superoxide dismutases, and related matters. J Biol Chem. 272(30): p. 18515-7.

79. Fridovich, I. (1998). Oxygen toxicity: a radical explanation. J Exp Biol. 201(Pt 8): p. 1203-9.

80. Jung, I.L. and I.G. Kim (2003). Transcription of ahpC, katG, and katE genes in Escherichia coli is regulated by polyamines: polyamine-deficient mutant sensitive to H2O2-induced oxidative damage. Biochem Biophys Res Commun. 301(4): p. 915-22.

81. Lim, P., et al. (2004). Peroxyl radical mediated oxidative DNA base damage: implications for lipid peroxidation induced mutagenesis. Biochemistry. 43(49): p. 15339-48.

82. Emerson, J.P., et al. (2003). Kinetics of the superoxide reductase catalytic cycle. J Biol Chem. 278(41): p. 39662-8.

83. Cruz, F. and J.G. Ferry (2006). Interaction of iron-sulfur flavoprotein with oxygen and hydrogen peroxide. Biochim Biophys Acta. 1760(6): p. 858-64.

84. Sieker, L.C., R.E. Stenkamp, and J. LeGall (1994). Rubredoxin in crystalline state. Methods Enzymol. 243: p. 203-16.

85. Guedon, E. and H. Petitdemange (2001). Identification of the gene encoding NADH- rubredoxin oxidoreductase in Clostridium acetobutylicum. Biochem Biophys Res Commun. 285(2): p. 496-502.

86. Nolling, J., et al. (1995). Characterization of a 45-kDa flavoprotein and evidence for a rubredoxin, two proteins that could participate in electron transport from H2 to CO2 in methanogenesis in Methanobacterium thermoautotrophicum. Eur J Biochem. 231(3): p. 628-38.

87. Gomes, C.M., et al. (1997). Studies on the redox centers of the terminal oxidase from Desulfovibrio gigas and evidence for its interaction with rubredoxin. J Biol Chem. 272(36): p. 22502-8.

88. Chen, C.J., et al. (2003). Preparation and X-ray crystallographic analysis of rubredoxin crystals from Desulfovibrio gigas to beyond ultra-high 0.68 A resolution. Biochem Biophys Res Commun. 308(4): p. 684-8.

89. Jin, S., et al. (2002). X-ray crystal structures of reduced rubrerythrin and its azide adduct: a structure-based mechanism for a non-heme diiron peroxidase. J Am Chem Soc. 124(33): p. 9845-55.

90. Rodrigues, R., et al. (2006). Desulfovibrio gigas flavodiiron protein affords protection against nitrosative stress in vivo. J Bacteriol. 188(8): p. 2745-51.

91. Lumppio, H.L., et al. (2001). Rubrerythrin and rubredoxin oxidoreductase in Desulfovibrio vulgaris: a novel oxidative stress protection system. J Bacteriol. 183(1): p. 101-8.

92. Sztukowska, M., et al. (2002). Role of rubrerythrin in the oxidative stress response of Porphyromonas gingivalis. Mol Microbiol. 44(2): p. 479-88.

93. LeGall, J., et al. (1988). Isolation and characterization of rubrerythrin, a non-heme iron protein from Desulfovibrio vulgaris that contains rubredoxin centers and a hemerythrin- like binuclear iron cluster. Biochemistry. 27(5): p. 1636-42.

94. Abreu, I.A., et al. (2002). Superoxide scavenging by neelaredoxin: dismutation and reduction activities in anaerobes. J Biol Inorg Chem. 7(6): p. 668-74.

95. Yeh, A.P., et al. (2000). Structures of the superoxide reductase from Pyrococcus furiosus in the oxidized and reduced states. Biochemistry. 39(10): p. 2499-508.

96. Clay, M.D., et al. (2003). Spectroscopic characterization of the [Fe(His)4(Cys)] site in 2Fe-superoxide reductase from Desulfovibrio vulgaris. J Biol Inorg Chem. 8(6): p. 671- 82.

97. Silva, G., et al. (1999). Desulfovibrio gigas neelaredoxin. A novel superoxide dismutase integrated in a putative oxygen sensory operon of an anaerobe. Eur J Biochem. 259(1-2): p. 235-43.

98. Jenney, F.E., Jr., et al. (1999). Anaerobic microbes: oxygen detoxification without superoxide dismutase. Science. 286(5438): p. 306-9.

99. Rusnak, F., et al. (2002). Superoxide reductase activities of neelaredoxin and desulfoferrodoxin metalloproteins. Methods Enzymol. 349: p. 243-58.

100. Ascenso, C., et al. (2000). Desulfoferrodoxin: a modular protein. J Biol Inorg Chem. 5(6): p. 720-9.

101. Woodmansee, A.N. and J.A. Imlay (2002). Reduced flavins promote oxidative DNA damage in non-respiring Escherichia coli by delivering electrons to intracellular free iron. J Biol Chem. 277(37): p. 34055-66.

102. Netto, L.E. and E.R. Stadtman (1996). The iron-catalyzed oxidation of dithiothreitol is a biphasic process: hydrogen peroxide is involved in the initiation of a free radical chain of reactions. Arch Biochem Biophys. 333(1): p. 233-42.

103. Lambeth, D.O., et al. (1982). Implications for in vitro studies of the autoxidation of ferrous ion and the iron-catalyzed autoxidation of dithiothreitol. Biochim Biophys Acta. 719(3): p. 501-8.

104. Tripp, B.C., et al. (2004). A role for iron in an ancient carbonic anhydrase. J Biol Chem. 279(8): p. 6683-7.

105. Miller, R.T., et al. (1999). Zinc content of Escherichia coli-expressed constitutive isoforms of nitric-oxide synthase. Enzymatic activity and effect of pterin. J Biol Chem. 274(21): p. 14537-40.

106. Taylor, P.K., et al. (2001). Analysis of metal incorporation during overexpression of Clostridium pasteurianum rubredoxin by electrospray FTICR mass spectrometry. J Biol Inorg Chem. 6(2): p. 201-6.

107. Eidsness, M.K., et al. (1992). Expression of a synthetic gene coding for the amino acid sequence of Clostridium pasteurianum rubredoxin. Protein Eng. 5(4): p. 367-71.

108. Johnson, E., et al. (2005). Crystal structures of a tetrahedral open pore ferritin from the hyperthermophilic archaeon Archaeoglobus fulgidus. Structure. 13(4): p. 637-48.

109. Pulliainen, A.T., et al. (2005). Dps/Dpr ferritin-like protein: insights into the mechanism of iron incorporation and evidence for a central role in cellular iron homeostasis in Streptococcus suis. Mol Microbiol. 57(4): p. 1086-100.

110. Wiedenheft, B., et al. (2005). An archaeal antioxidant: characterization of a Dps-like protein from Sulfolobus solfataricus. Proc Natl Acad Sci U S A. 102(30): p. 10551-6.

111. Su, M., et al. (2005). The so-called Listeria innocua ferritin is a Dps protein. Iron incorporation, detoxification, and DNA protection properties. Biochemistry. 44(15): p. 5572-8.

112. Halsey, T.A., et al. (2004). The ferritin-like Dps protein is required for Salmonella enterica serovar Typhimurium oxidative stress resistance and virulence. Infect Immun. 72(2): p. 1155-8.

113. Ishikawa, T., et al. (2003). The iron-binding protein Dps confers hydrogen peroxide stress resistance to Campylobacter jejuni. J Bacteriol. 185(3): p. 1010-7.

114. Ilari, A., et al. (2002). Iron incorporation into Escherichia coli Dps gives rise to a ferritin-like microcrystalline core. J Biol Chem. 277(40): p. 37619-23.

115. Grant, R.A., et al. (1998). The crystal structure of Dps, a ferritin homolog that binds and protects DNA. Nat Struct Biol. 5(4): p. 294-303.

116. Martinez, A. and R. Kolter (1997). Protection of DNA during oxidative stress by the nonspecific DNA-binding protein Dps. J Bacteriol. 179(16): p. 5188-94.

117. Muller, M. (1988). Energy metabolism of protozoa without mitochondria. Annu Rev Microbiol. 42: p. 465-88.

118. Dills, S.S., et al. (1980). Carbohydrate transport in bacteria. Microbiol Rev. 44(3): p. 385-418.

119. Tielens, A.G. and J.J. Van Hellemond (1998). The electron transport chain in anaerobically functioning eukaryotes. Biochim Biophys Acta. 1365(1-2): p. 71-8.

120. Valle, F., et al. (1996). Basic and applied aspects of metabolic diversity: the phosphoenolpyruvate node. Journal of Industrial Microbiology. 17: p. 458-462.

121. Bailey-Serres, J. and L.A. Voesenek (2008). Flooding stress: acclimations and genetic diversity. Annu Rev Plant Biol. 59: p. 313-39.

122. Siebers, B. and P. Schonheit (2005). Unusual pathways and enzymes of central carbohydrate metabolism in Archaea. Curr Opin Microbiol. 8(6): p. 695-705.

123. Van Hellemond, J.J. and A.G. Tielens (1994). Expression and functional properties of fumarate reductase. Biochem J. 304 ( Pt 2): p. 321-31.

124. Duine, J.A. (2001). Cofactor diversity in biological oxidations: implications and applications. Chem Rec. 1(1): p. 74-83.

125. Detkova, E.N. and M.A. Pusheva (2006). Energy metabolism in halophilic and alkalophilic bacteria. Mikrobiologiia. 75(1): p. 5-17.

126. Furdui, C. and S.W. Ragsdale (2000). The role of pyruvate ferredoxin oxidoreductase in pyruvate synthesis during autotrophic growth by the Wood-Ljungdahl pathway. J Biol Chem. 275(37): p. 28494-9.

127. Carlier, M.F. (1990). Actin polymerization and ATP hydrolysis. Adv Biophys. 26: p. 51-73.

128. Stouthamer, A.H. (1973). A theoretical study on the amount of ATP required for synthesis of microbial cell material. Antonie Van Leeuwenhoek. 39(3): p. 545-65.

129. Clausius, R. (1851). On the Moving Force of Heat, and the Laws regarding the Nature of Heat itself which are deducible there-from. London, Edinburgh and Dublin Philosophical Magazine and Journal of Science. 2: p. 1-21.

130. Klotz, I.M., Introduction to biomolecular energetics : including ligand-receptor interactions. (1986), Orlando [Fla.]: Academic Press. xiv, 138 p.

131. Hamann, C.H., A. Hamnett, and W. Vielstich, Electrochemistry. (1998), Weinheim ; New York: Wiley-VCH. xvii, 423 p.

132. Bard, A.J. and L.R. Faulkner, Electrochemical methods : fundamentals and applications. 2nd ed. (2001), New York: Wiley. xxi, 833 p.

133. Savéant, J.M., Elements of molecular and biomolecular electrochemistry : an electrochemical approach to electron transfer chemistry. The George Fisher Baker non-resident lectureship in chemistry at Cornell University. (2006), Hoboken, N.J.: Wiley-Interscience. xviii, 485 p.

134. Rosenberg, R. and I. Klotz (1999). Spontaneity and the Equilibrium Constant: Advantages of the Planck Function. Journal of Chemical Education. 76(10): p. 1448- 1451.

135. Silverman, R.B., The organic chemistry of enzyme-catalyzed reactions. [Rev. ed. (2002), San Diego: Academic Press. xvii, 717 p.

136. Nelson, N. and C.F. Yocum (2006). Structure and function of photosystems I and II. Annu Rev Plant Biol. 57: p. 521-65.

137. Rothery, R.A., G.J. Workun, and J.H. Weiner (2008). The prokaryotic complex iron- sulfur molybdoenzyme family. Biochim Biophys Acta. 1778(9): p. 1897-929.

138. Hill, H.A.O., P.J. Sadler, and A.J. Thompson, Metal sites in proteins and models : redox centres. Structure and bonding,. (1998), Berlin ; New York: Springer. viii, 209 p.

139. Jean, Y., Molecular orbitals of transition metal complexes. (2005), Oxford ; New York: Oxford University Press. xii, 275 p.

140. Williams, R., Bioinorganic Chemistry: Trace Element Evolution from Anaerobes to Aerobes. Structure and bonding,. (1998), Berlin ; New York: Springer. viii, 207 p.

141. Winter, M.J., d-block chemistry. Oxford chemistry primers 27. (1994), Oxford ; New York: Oxford University Press. 90, [2] p.

142. Bott, A. (1999). Redox Properties of Electron Transfer Metalloproteins. Current Separations. 18(2): p. 47-54.

143. Sulpizi, M., et al. (2007). Calculation of redox properties: understanding short- and long-range effects in rubredoxin. J Phys Chem B. 111(15): p. 3969-76.

144. Sundararajan, M., I.H. Hillier, and N.A. Burton (2006). Structure and redox properties of the protein, rubredoxin, and its ligand and metal mutants studied by electronic structure calculation. J Phys Chem A. 110(2): p. 785-90.

145. Nanda, V., et al. (2005). De novo design of a redox-active minimal rubredoxin mimic. J Am Chem Soc. 127(16): p. 5804-5.

146. Eidsness, M.K., et al. (1999). Modulation of the redox potential of the [Fe(SCys)4] site in rubredoxin by the orientation of a peptide dipole. Biochemistry. 38(45): p. 14803-9.

147. Shenoy, V.S. and T. Ichiye (1993). Influence of protein flexibility on the redox potential of rubredoxin: energy minimization studies. Proteins. 17(2): p. 152-60.

148. Rubio, L.M. and P.W. Ludden (2008). Biosynthesis of the Iron-Molybdenum Cofactor of Nitrogenase. Annu Rev Microbiol.

149. Walsh, J.D. and A.-F. Miller (2003). Flavin reduction potential tuning by substitution and bending. Journal of Molecular Structure (Theochem). 623: p. 185-195.

150. Müller, F., Chemistry and biochemistry of flavoenzymes. (1991), Boca Raton: CRC Press. v. <1-3 >.

151. Mathews, F.S., L. Cunane, and R.C. Durley (2000). Flavin electron transfer proteins. Subcell Biochem. 35: p. 29-72.

152. Muller, F. (1983). The flavin redox-system and its biological function. Top Curr Chem. 108: p. 71-107.

153. Massey, V. (1995). Introduction: flavoprotein structure and mechanism. FASEB J. 9(7): p. 473-5.

154. Berg, J., J. Tymoczko, and L. Stryer, Biochemistry. 6th ed. (2006): W. H. Freeman.

155. Dunphy, P.J., P.G. Phillips, and A.F. Brodie (1971). Separation and identification of menaquinones from microorganisms. J Lipid Res. 12(4): p. 442-9.

156. Thony, B., G. Auerbach, and N. Blau (2000). Tetrahydrobiopterin biosynthesis, regeneration and functions. Biochem J. 347 Pt 1: p. 1-16.

157. Neijssel, O.M. and M.J. Teixeira de Mattos (1994). The energetics of bacterial growth: a reassessment. Mol Microbiol. 13(2): p. 172-82.

158. Ying, W. (2008). NAD+/NADH and NADP+/NADPH in cellular functions and cell death: regulation and biological consequences. Antioxid Redox Signal. 10(2): p. 179- 206.

159. Ferry, J.G., Methanogenesis : ecology, physiology, biochemistry & genetics. Chapman & Hall microbiology series. (1993), New York: Chapman & Hall. x, 536 p.

160. Kim, S., et al. (2009). Comparison of the effects of NADH- and NADPH-perturbation stresses on the growth of Escherichia coli. Curr Microbiol. 58(2): p. 159-63.

161. Berger, F., M.H. Ramirez-Hernandez, and M. Ziegler (2004). The new life of a centenarian: signalling functions of NAD(P). Trends Biochem Sci. 29(3): p. 111-8.

162. Rizzi, M. and H. Schindelin (2002). Structural biology of enzymes involved in NAD and molybdenum cofactor biosynthesis. Curr Opin Struct Biol. 12(6): p. 709-20.

163. Kornberg, A. (1995). Inorganic polyphosphate: toward making a forgotten polymer unforgettable. J Bacteriol. 177(3): p. 491-6.

164. Kulaev, I.S. (1975). Biochemistry of inorganic polyphosphates. Rev Physiol Biochem Pharmacol. 73: p. 131-58.

165. Westheimer, F.H. (1987). Why nature chose phosphates. Science. 235(4793): p. 1173- 8.

166. Satoh, T. and S.A. Lipton (2007). Redox regulation of neuronal survival mediated by electrophilic compounds. Trends Neurosci. 30(1): p. 37-45.

167. Netto, L.E., et al. (2007). Reactive cysteine in proteins: protein folding, antioxidant defense, redox signaling and more. Comp Biochem Physiol C Toxicol Pharmacol. 146(1-2): p. 180-93.

168. Meyer, A.J. and R. Hell (2005). Glutathione homeostasis and redox-regulation by sulfhydryl groups. Photosynth Res. 86(3): p. 435-57.

169. Rouhier, N., S.D. Lemaire, and J.P. Jacquot (2008). The role of glutathione in photosynthetic organisms: emerging functions for glutaredoxins and glutathionylation. Annu Rev Plant Biol. 59: p. 143-66.

170. Xu, X.M. and S.G. Moller (2008). Iron-sulfur cluster biogenesis systems and their crosstalk. Chembiochem. 9(15): p. 2355-62.

171. Fontecave, M. and S. Ollagnier-de-Choudens (2008). Iron-sulfur cluster biosynthesis in bacteria: Mechanisms of cluster assembly and transfer. Arch Biochem Biophys. 474(2): p. 226-37.

172. Fahey, R.C. (2001). Novel thiols of prokaryotes. Annu Rev Microbiol. 55: p. 333-56.

173. Stirnimann, C.U., et al. (2006). nDsbD: a redox interaction hub in the Escherichia coli periplasm. Cell Mol Life Sci. 63(14): p. 1642-8.

174. Yang, M.S., et al. (2008). Increase in intracellular free/bound NAD(P)H as a cause of Cd-induced oxidative stress in the HepG(2) cells. Toxicology. 247(1): p. 6-10.

175. Frye, R.A. (2000). Phylogenetic classification of prokaryotic and eukaryotic Sir2-like proteins. Biochem Biophys Res Commun. 273(2): p. 793-8.

176. Smith, B.C. and J.M. Denu (2007). Sir2 deacetylases exhibit nucleophilic participation of acetyl-lysine in NAD+ cleavage. J Am Chem Soc. 129(18): p. 5802-3.

177. Zhao, K., et al. (2004). Structural basis for nicotinamide cleavage and ADP-ribose transfer by NAD+-dependent Sir2 histone/protein deacetylases. Proc Natl Acad Sci U S A. 101(23): p. 8563-8.

178. Blander, G. and L. Guarente (2004). The Sir2 family of protein deacetylases. Annu Rev Biochem. 73: p. 417-35.

CHAPTER TWO: FLAVINS AND FLAVOPROTEINS 2.1 Introduction to Flavin Cofactors 2.2 Classical Flavodoxins 2.3 Flavodoxin-like Proteins 2.3.1 Iron-Sulfur Flavoprotein 2.3.2 NAD(P)H-dependent Reductases 2.3.3 Cytochrome P450 Reductases 2.3.4 Rubredoxin:Oxygen Oxidoreductase 2.4 References

Chapter 2

Flavins and Flavoproteins

Flavoproteins are ubiquitous in nature and serve a variety of roles that generally involve electron transfer. Most flavoproteins catalyze two separate half-reactions, utilizing their redox-active flavin cofactor to oxidize one substrate and subsequently reduce a second substrate. While peptide interactions modulate the physicochemical properties of substrates and cofactors, the activity of a flavoprotein is largely dictated through modulation of the flavin redox potentials via the covalent and electrostatic interactions inherent to flavin coordination in the active site [1-8].

2.1 Introduction to Flavin Cofactors

Flavin compounds were first isolated from cow milk in 1879 when Alexander Wynter

Blyth purified and concentrated a compound he named lactochrome, which he described as having “a simple spectrum, allowing most of the red and yellow rays to pass through" [9].

Following his discovery, several compounds with properties analogous to lactochrome were identified (i.e. lycochrome, hepatoflavin, verdoflavin), and all were eventually classified as

‘riboflavin,’ a term designated by The Council on Pharmacy and Chemistry of the American

Medical Association in 1937 [2, 5]. Riboflavin (Vitamin B2) and two of its chemical

derivatives, flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD), are the most common flavin cofactors, though specialized flavins such as lumichrome and coenzyme

F420 also have important physiological roles (Figure 2-1) [1-5].

The primary usefulness of flavins is a result of their ability to participate in redox reactions with one- or two-electron acceptors. The aromatic isoalloxazine ring, central to the functionality of flavins, enables significant modulation of the flavin redox potential. This occurs via electrostatic or covalent interactions with its three fused aromatic rings, which incorporate two carbonyl groups and two methyl groups protruding from the aromatic rings.

Flavins are frequently described as amphoteric molecules with extensive capacity to participate in multiple charge states, resonance forms, and tautomer structures for each redox state [1-5, 7, 8, 10-19].

Flavin compounds have three canonical redox states: the fully oxidized flavoquinone, the one-electron reduced flavosemiquinone, and the two-electron reduced flavohydroquinone

(Figure 2-2). For each of these states, there are three protonated forms involving the nitrogen atoms of the isoalloxazine ring. However, both carbonyl groups can also be protonated, further contributing to possible resonance and tautomer structures. Thus, each position on the ring affords opportunity for unique modulation of the cofactor’s electrostatic characteristics, and each position can be exploited for fine-tuning of the redox-potential or for covalent catalysis with the flavin. For example, free flavins cycle between their quinone and hydroquinone states (Em,7 = -195 to -220 mV), while enzyme-bound flavins generally have a redox potential range of +100 to -500mV (pH 7.0) [1-5, 7, 8, 17]. As a result, flavins are able to participate in an impressive variety of reactions – more than any other coenzyme known in nature [2].

I) O R1 = OH N O NH O P OH R2 = N N O OH OH O O OH R3 = O P O P O N HO O NH OH OH N 2 R N N HO OH

O III) II) O N NH NH

N N O HO N N O

HO OH O O COOH O P H O N OH N COOH H O COOH

Figure 2-1. Flavin cofactors: I) R = R1, Riboflavin; R=R2, FMN; R = R3, FAD.

II) Coenzyme F420. III) Lumichrome.

R 1 9 10 8 2 H3C N N O

NH 7 3 H3C 6 5 N 4 O

A. Flavoquinone R R R H H C N N O pK = 0 H C N N O H3C N N O 3 a 3 pKa = 10 + NH NH N H C N - H3C N H3C N 3 O O O

B. Flavosemiquinone R R R H H C N N O pK = 2.3 H C N N O H C N N O 3 a 3 pKa = 8.3 3 . NH . NH . NH H C N H C N H3C N 3 +H 3 H O O O -

C. Flavohydroquinone R R R H H - H C N N O pK < 0 H C N N O H C N N O 3 a 3 pKa = 6.7 3 + NH NH NH H C N H C N H3C N 3 3 H H H H O O O

Figure 2-2. Various redox states of the typical isoalloxazine ring.

TOP: Illustration of the numerical positions on the isoalloxazine ring, shown with the re-side up.

A. Free flavoquinone is stable in solution and is neutral between pH 1 and 9, with deprotonation of N(3) near pH 10. Protonation of N(1) occurs at pH values below pH 0 and a dicationic form follows with protonation of N(5), possible only in concentrated acidic environments.

B. The one-electron reduced, neutral semiquinone is protonated at N(5) and appears blue or green in solution, while deprotonation (pKa = 8.3) yields a red semiquinone and protonation (pKa = 2.3) yields the cationic semiquinone.

C. In the two-electron reduced 1,5-flavohydroquinone N(5) is protonated with a pKa of <0, and while N(1) of the oxidized flavoquinone is an acidic position, N(1) of the reduced flavohydroquinone is a basic position (pKa = 6.7).

2.2 Classical Flavodoxins

Flavodoxins are small electron-transferring flavoproteins involved in a variety of

constitutive and specific cellular reactions, mediating various electron transfer processes.

For example, in an environment where iron is sparse, flavodoxins are known to substitute for

ferredoxin, an iron-containing protein functioning in pathways that lead to the reduction of

+ NADP and N2. In addition to this activity, flavodoxins are known to function in methionine

synthesis, biotin synthesis, pyruvate oxidation, and activation of the ribonucleoside reductase

[2, 20-37]. Further, the flavodoxin fold has been incorporated into many other proteins as

functional domains and as flavodoxin-like proteins with unique modifications [25, 38-52].

The biological importance of this enzyme and the unique modulation of its flavin cofactor

have brought much attention to flavodoxins. In addition, they were the first flavoproteins for

which crystal structures became available, providing an early opportunity to investigate

flavin binding (Figure 2-3) [2, 24].

The most prominent characteristic of flavodoxin is the ability of its FMN cofactor to

stabilize the one-electron reduced flavosemiquinone state over the oxidized flavoquinone or

two-electron reduced flavohydroquinone – this is distinct from free FMN, which does not

stabilize the flavosemiquinone [2, 23, 24, 51, 53-66]. Upon binding FMN, electrostatic

interactions in the flavodoxin active site cause a significant negative shift in the midpoint

oxidation-reduction potential of the free flavin semiquinone/hydroquinone couple from a

potential near -0.150 V (pH 7.0) to one that is in the range -0.372 V to -0.518 V (pH 7.0),

significantly increasing the energy required to reduce the flavin with a second electron. The

mechanisms that stabilize the flavosemiquinone largely involve hydrogen bonding with

N(5)H and O(4) of the isoalloxazine ring, while destabilization of the flavohydroquinone

occurs at N(1) and through - interactions at the flavin face [2, 23, 24, 26, 30, 34, 35, 37,

50-54, 57, 58, 60-76]. In addition to these interactions, data also suggest that the semiquinone/hydroquinone couple is particularly susceptible to local charge contributions and interactions at N(3) [53, 77, 78]. Crystal structures of the three oxidation states for

Clostridium flavodoxin (5NLL, 2FOX, 5ULL) demonstrate that a conformational change occurs in a four-residue surface loop between each redox state (Figure 2-3). The loop, positionally referred to as ‘the 50’s loop,’ largely interacts with the si-side of the flavin and stabilizes the flavosemiquinone state by enabling a peptide carbonyl group to form a hydrogen bond with N(5) of the isoalloxazine ring. This loop, along with ‘the 90’s loop’ near N(1), provide most of the contacts with the flavin (Figure 2-3) [2, 23, 24, 57, 58, 60, 61,

67, 69-74, 76, 78]. Residues in several of these positions significantly contribute to modulation of and binding of the flavin cofactor, which has been demonstrated in numerous studies (Table 2-1). Such investigations have significantly contributed to our understanding of protein folding, cofactor binding, and redox potentials in flavoproteins.

WT Variant Residue(s) Interactions Reference Residue(s) Anabaena T12 K PO4 tail (Frago 2007) E16 K PO4 tail (Frago 2007) E20 K surface (Nogués 2004) T56 G, S re-face, ribose tail (Nogués 2004) W57 E, K, R, A, F, L, Y re-face, o -xylene (Frago 2007), (Lostao 1997) N58 C, K surface (Nogués 2004) D65 K surface (Nogués 2004) Y94 A, F, W, L si-face (Lostao 1997) D96 N surface (Nogués 2004) N97 K O(4) (Nogués 2004) I59/I92 A/A, E/E N(5)/surface (Frago 2007) Clostridium M56 L, I, V, A, G re-face, o -xylene (Druhan 1998) G57 A, N, D, T N(5) (Chang 1999), (Ludwig 1997) D58 S, P O(4) (Chang 1999), (Ludwig 1997) E59 Q, D, N, A N(3) (Bradley 1999), (Bradley 2001) N137 A residue D58 (Ludwig 1997) G57/D58 G/G, G/A, A/G, A/A N(5)/O(4) (Kasim 2000) M56/G57/ M/G/A/A, A/G/D/A, re-face, o-xylene/ (Kasim 2001) D58/E59 A/G/A/A N(5)/O(4)/N(3) Anacystis N58 G N(5) (Hoover 1999) D90 N N(1), O(2) (Hoover 1999) D100 N O(2), N(3) (Hoover 1999) Desulfovibrio G61 A, N, V, L N(5) (O'Farrell 1998) D62 N* N(1), surface (Zhou 1995), (Feng 1997) D63 N* N(1), surface (Zhou 1995), (Feng 1997) E66 Q* N(1), surface (Zhou 1995), (Feng 1997) D69 N N(1), surface (Feng 1997) D70 N N(1), surface (Feng 1997) D95 A, N*, E N(1), surface (McCarthy 2002), (Zhou 1995), (Feng 1997) (Swenson 1994), (Stockman 1994), Y98 F, W, H, R, A, M si-face, N(1) (Reynolds 2001), (Chang 1997), (Zhou 1996) E99 Q* N(1), surface (Zhou 1995) D106 N* N(1), surface (Zhou 1995) D127 N N(1), surface (Feng 1997) D129 N N(1), surface (Feng 1997) Megasphaera E60 Q N(1) (Geoghegan 2000)

* Zhou and Feng also investigated these with multiple variants (see references).

Table 2-1. Variant investigations of classical flavodoxins focused in the active sites.

II) III)

IV)

Figure 2-3. Crystal structure of flavodoxin 5NLL. I) Flavodoxin monomer demonstrating the classic fold with FMN bound. II) Oxidized flavodoxin active site with trans O-up at G57. III) Reduced flavodoxin active site with trans O-down at G57. IV) Flavodoxin active site with W90 shown over the isoalloxazine ring.

2.3 Flavodoxin-like Proteins Throughout evolution, compact motifs have emerged within flavodoxin-like proteins, commuting properties such as multimerization, interactions with NAD+/NADH, and incorporation of iron-sulfur clusters. Similarly, flavodoxin genes have been modified and incorporated as functional domains in larger proteins, fusing with other domains to be utilized in more specific roles [25, 38-50, 68, 79-86]. The diverse array of unique proteins that has evolved with these modifications is arguably owed to the usefulness and simplicity of the flavodoxin fold.

2.3.1 Iron-Sulfur Flavoprotein

Iron-Sulfur Flavoprotein (Isf ) is one example protein that has incorporated unique and compact motifs within a flavodoxin fold. Alignments of flavodoxins, Isf proteins, and other flavodoxin-like proteins have demonstrated that Isf is a unique protein family. Genes encoding Isf homologues have been identified in numerous anaerobes from the Bacteria and

Archaea domains, and the first homologue was characterized in 1996 from the anaerobe

Methanosarcina thermophila [82, 87, 88]. Isf is a multimer with one [4Fe:4S] cluster and one

FMN per monomer (Figure 2-4), and the holoprotein reduces O2 and H2O2 at the flavin with water as the only product, concomitant with depletion of the iron-sulfur cluster. Thus Isf may have a role in sensing or reducing reactive oxygen species (ROS), but its global role remains unknown. The 19 Isf homologues in the genome of the M. acetivorans, suggested the protein family could have various cellular activities [80, 81, 87-89].

While sequence alignments delineate Isf as a new protein family, its tertiary structure retains the classic flavodoxin fold, which implies the proteins could have a common ancestor.

A more critical investigation of Isf reveals additional distinguishing characteristics. A novel, compact cysteine motif located towards the N-terminus of the Isf protein is responsible for

binding the [4Fe:4S] cluster [80, 81, 87-89]. The Isf crystal structure shows this metal cofactor is positioned approximately 4 Å from the o-xylene moiety of the flavin cofactor, which is modulated similar to free flavin but with altered protonation states (Figure 2-4) [79].

Biochemical characterizations demonstrate that electron flow proceeds through the low- potential iron-sulfur cluster (-394 mV) to the flavin cofactor (Eox/hq = -277 mV), with a transient flavosemiquinone [89]. The Isf crystal structure suggests dimerization is required for activity, with electron transfer proceeding from the iron-sulfur cluster of one subunit to a methyl group on the flavin of the adjacent subunit. Upon tetramerization, solvent exposure of the flavin is further decreased, though the flavin face remains unshielded [79]. Altogether, the cofactors and multimerization distinguish Isf from flavodoxins and are also examples of properties commuted by smaller motifs that have been incorporated into the flavodoxin fold.

II) III)

Figure 2-4. Crystal structure of Iron-sulfur flavoprotein (Isf) from Methanosarcina thermophila. I) Isf dimer demonstrating the flavodoxin folds with FMN bound near the [4Fe-4S] cluster. II) The Isf active site shown with three monomers coming together (red, blue, and grey). III) The Isf active site from above, illustrating the distance between flavin and the iron-sulfur cluster.

2.3.2 NAD(P)H-Dependent Reductases

Other example proteins where unique and compact motifs have been incorporated into the flavodoxin fold include the multimeric flavodoxin-like reductases, which couple electron transfer from NADH, NADPH, or NRH to a variety of compounds. As with Isf, crystal structures of these reductases are available and the monomeric tertiary structure resembles classical flavodoxin (Figure 2-5). Investigations generally distinguish them from flavodoxins by referencing their multimeric qualities, the unshielded flavin face, and the

NAD(P)H-dependent activity of the reductases. Expression studies, survival models, and in vitro investigations implicate these reductases in oxidative stress and usually suggest they function to couple electron transfer directly to xenobiotic compounds such as quinoid compounds, flavins, metals, and dyes [82, 90-119].

Many studies hypothesize that these reductases participate in two-electron reductions in an effort to bypass the one-electron reduced semiquinones of the cofactors and substrates.

By fully reducing quinoid compounds, the reductases discourage substrates from participating in redox-cycling, and they ultimately protect against the formation of ROS [93,

95, 100-102, 104, 106, 108, 110, 120, 121]. Several structures show that NAD(P)H and quinoid substrates independently bind near the reductase active site, supporting that two electrons are transferred in a ping-pong fashion with the flavin cofactor as a mediator [85,

105]. The preferential cycling between the flavoquinone/ flavohydroquinone is clearly distinct from classical flavodoxins. However, while significant information is available on structure-function relationships in flavodoxins, little is known about the structure-function relationships in flavodoxin-like, NAD(P)H-dependent reductases. These proteins will be explored with more detail in later chapters.

II) T116 III) F80

Figure 2-5. Crystal structure of WrbA from Escherichia coli. I) WrbA dimer demonstrating the flavodoxin folds with FMN bound. II) The WrbA active site, shown with three monomers coming together (red, blue, and grey). III) The WrbA active site from the side, showing residues which physically distinguish WrbA from flavodoxin (F80 and T116).

2.3.3 Cytochrome P450 Reductases

Another way that biology exploits the flavodoxin fold is to incorporate the protein as part of a larger, multidomain protein. By N-terminal or C-terminal fusion to other domains, the fold remains relatively unmodified and is well poised for a specific function in multidomain proteins (Figure 2-6) [39-46, 50, 68, 86, 122].

Cytochrome P450 (CYP) proteins and their reductases are example proteins systems that make use of flavodoxin as a fused, functional domain. These proteins comprise a large and diverse superfamily, and they collectively function in xenobiotic detoxification and steriodogenesis [7, 31, 40, 42, 45, 50, 68, 123-128]. As of 2005, there were 57 CYP genes and 59 pseudogenes identified in humans alone. By 2008, there were 781 CYP families across all domains of life and a report from 2007 suggested 7232 unique sequences, excluding variants and pseudogenes. Not surprisingly, their sheer number makes these proteins difficult to categorize [129, 130]. One trait used to distinguish between classes is their source of electrons: some require ferredoxin and FAD-dependent reductases; others require the CYP reductases, which contain the flavodoxin domain; and still others are made self-sufficient by directly fusing with a flavodoxin-containing domain (Figure 2-6) [40, 44,

45, 86]. Thus, the flavodoxin fold is modular and useful for shuttling electrons to a wide range of proteins.

I) cytochrome P450 (CYP) reductase and CYP proteins:

Flavodoxin reductase flavodoxin P450

Flavodoxin reductase flavodoxin P450 Flavodoxin reductase flavodoxin P450

Ia)

IIa)

II) rubredoxin reductase and rubredoxin:O2 oxidoreductase

flavin domain rubredoxin flavodoxin Lactamase domain flavin domain rubredoxin flavodoxin Lactamase domain

flavin domain rubredoxin flavodoxin Lactamase domain

rubredoxin flavin domain flavodoxin Lactamase domain

flavin domain flavodoxin Lactamase domain

Figure 2-6. Flavodoxin-domain incorporated into other proteins. I) Arrangements of the flavodoxin domain among various CYP systems. Ia) Crystal structure of CYP BM-3 heme domain bound to its flavodoxin domain. II) Structural configurations of ROO electron transfer systems with flavodoxin as a domain. IIa) Crystal structure of ROO. Arrows indicate flavin binding sites.

The flavodoxin domain is primarily useful to these proteins because they stabilize a flavosemiquinone, as occurs in classical flavodoxins. In the reaction cycle, the heme cofactor receives one electron at a time from the flavosemiquinone, which functions as a 1- electron/ 2-electron switch. This enables CYP proteins to perform a wide range of reactions, such as the hydroxylation of unactivated alkenes, and the conversion of alkenes to epoxides, arenes to phenols, sulfides to sulfoxides, and sulfoxides to sulfones. In many species, these activities are utilized in the synthesis of various steroids, hormones, eicosanoids, and vitamins. Additionally, CYP proteins can offer protection against insecticides and herbicides in insects and plants, respectively [7, 31, 32, 40, 42, 45, 123-128]. Thus with little functional modification, the classical flavodoxin appears to be greatly useful as a fused domain to CYP proteins and their reductases, enabling vast functionalities by providing one electron at a time for reduction of the heme.

2.3.4 Rubredoxin:Oxygen Oxidoreductase

As seen in Figure 2-6, rubredoxin:O2 oxidoreductases (ROO) is another protein that uses flavodoxin as a fused, functional domain [38, 47-49]. Although only a few ROO proteins are biochemically characterized, genes encoding the protein can be found in genomes across the Bacteria and Archaea. As described in Chapter 1, several studies indicate that ROO functions in NAD(P)H-dependent electron transport chains that couple electron transfer directly from rubredoxin to compounds such as O2 and NO [131-133]. The protein is one of several found in classically ‘strict’ anaerobes which function to combat oxidative stress. In using reducing equivalents to alleviate oxidative stress, anaerobes are able to remove ROS without regenerating O2, as occurs in aerobes with catalase and dismutase [38, 47-49, 88, 133-143].

As with CYP proteins and their reductases, the flavodoxin domain of ROO is especially useful because it functions as a 1-electron/ 2-electron switch, receiving electrons from a rubredoxin domain and acting in concert with a diiron site to reduce O2 or NO.

Several studies demonstrate that the flavin passes through two 1-electron reductions and can stabilize a flavosemiquinone, similar to classical flavodoxins. Interestingly, the two reduction potentials of the ROO flavin cofactor are significantly more positive than those of flavodoxins, suggesting the flavin environments are different, despite their similar behavior in electron transfer [47-49].

2.4 References

1. Heelis, P.F. (1982). The photophysical and photochemical properties of flavins (isoalloxazines). Chemical Society Reviews. 11: p. 15-39.

2. Müller, F., Chemistry and biochemistry of flavoenzymes. (1991), Boca Raton: CRC Press. v. <1-3 >.

3. Chapman, S.K. and G.A. Reid, Flavoprotein protocols. Methods in molecular biology. (1999), Totowa, N.J.: Humana Press. xii, 256 p.

4. Mathews, F.S., L. Cunane, and R.C. Durley (2000). Flavin electron transfer proteins. Subcell Biochem. 35: p. 29-72.

5. Muller, F. (1983). The flavin redox-system and its biological function. Top Curr Chem. 108: p. 71-107.

6. Blankenhorn, G. (1976). Nicotinamide-dependent one-electron and two-electron (flavin) oxidoreduction: thermodynamics, kinetics, and mechanism. Eur J Biochem. 67(1): p. 67-80.

7. Banerjee, R., Redox biochemistry. (2008), Hoboken, N.J.: Wiley-Interscience. xx, 317 p.

8. Stevenson, K.J., V. Massey, and C.H. Williams, Flavins and Flavoproteins 1996: Proceedings of the Twelfth International Symposium. (1996), Calgary, Canada: University of Calgary. vii, 976 p.

9. Blyth, A. (1879). The composition of cow's milk in health and disease. J Chem Soc, Trans. 35: p. 530-539.

10. Johansson, L., et al. (1979). Electronic Transitions in the Isoalloxazine Ring and Orientation of Flavins in Model Membranes Studied by Polarized Light Spectroscopy. Biochemistry. 18(19): p. 4249-4253.

11. Massey, V. (1995). Introduction: flavoprotein structure and mechanism. FASEB J. 9(7): p. 473-5.

12. Muller, F., S.G. Mayhew, and V. Massey (1973). On the effect of temperature on the absorption spectra of free and protein-bound flavines. Biochemistry. 12(23): p. 4654- 62.

13. Shiga, K., et al. (1979). A study of flavin-protein and flavoprotein-ligand interactions. Binding aspects and spectral properties of D-amino acid oxidase and riboflavin binding protein. J Biochem. 85(4): p. 931-41.

14. Nishimoto, K., Y. Watanabe, and K. Yagi (1978). Hydrogen bonding of flavoprotein. I. Effect of hydrogen bonding on electronic spectra of flavoprotein. Biochim Biophys Acta. 526(1): p. 34-41.

15. Yagi, K., et al. (1980). Effect of hydrogen bonding on electronic spectra and reactivity of flavins. Biochemistry. 19(8): p. 1553-7.

16. Harbury, H.A., et al. (1959). Molecular Interaction of Isoalloxazine Derivatives. Ii. Proc Natl Acad Sci U S A. 45(12): p. 1708-17.

17. Edmondson, D.E. and G. Tollin (1971). Flavin-protein interactions and the redox properties of the Shethna flavoprotein. Biochemistry. 10(1): p. 133-45.

18. Edmondson, D.E. and G. Tollin (1971). Circular Dichroism Studies of the Flavin Chromophore and of the Relation between Redox Properties and Flavin Environment in Oxidases and Dehydrogensases. Biochemistry. 10(1): p. 113-123.

19. Massey, V. and G. Palmer (1966). On the Existance of Spectrally Distinct Classes of Flavoprotein Semiquinones. A New Method for the Quantitative Production of Flavoprotein Semiquinones. Biochemistry. 5(10): p. 3181-3189.

20. Hughes, N.J., et al. (1998). Helicobacter pylori porCDAB and oorDABC genes encode distinct pyruvate:flavodoxin and 2-oxoglutarate:acceptor oxidoreductases which mediate electron transport to NADP. J Bacteriol. 180(5): p. 1119-28.

21. Nogues, I., et al. (2004). Role of the C-terminal tyrosine of ferredoxin-nicotinamide adenine dinucleotide phosphate reductase in the electron transfer processes with its protein partners ferredoxin and flavodoxin. Biochemistry. 43(20): p. 6127-6137.

22. Mulliez, E., et al. (2001). Activation of class III ribonucleotide reductase by flavodoxin: A protein radical-driven electron transfer to the iron-sulfur center. Biochemistry. 40(12): p. 3730-3736.

23. Mayhew, S.G., et al. (1996). Regulation of the redox potentials of flavodoxins: modification of the flavin binding. Biochem Soc Trans. 24(1): p. 122-7.

24. Simondsen, R.P. and G. Tollin (1980). Structure-function relations in flavodoxins. Mol Cell Biochem. 33(1-2): p. 13-24.

25. Sancho, J. (2006). Flavodoxins: sequence, folding, binding, function and beyond. Cell Mol Life Sci. 63(7-8): p. 855-64.

26. Vervoort, J., et al. (1994). Flavodoxins. Methods Enzymol. 243: p. 188-203.

27. Nogues, I., et al. (2005). Anabaena flavodoxin as an electron carrier from photosystem I to ferredoxin-NADP(+) reductase. Role of flavodoxin residues in protein-protein interaction and electron transfer. Biochemistry. 44(1): p. 97-104.

28. Casaus, J.L., et al. (2002). Anabaena sp PCC 7119 flavodoxin as electron carrier from photosystem I to Ferredoxin-NADP(+) reductase - Role of TRP57 and TYR94. Journal of Biological Chemistry. 277(25): p. 22338-22344.

29. Bianchi, V., et al. (1993). Flavodoxin Is Required for the Activation of the Anaerobic Ribonucleotide Reductase. Biochemical and Biophysical Research Communications. 197(2): p. 792-797.

30. Feng, Y.C. and R.P. Swenson (1997). Evaluation of the role of specific acidic amino acid residues in electron transfer between the flavodoxin and cytochrome c(3) from Desulfovibrio vulgaris [Hildenborough]. Biochemistry. 36(44): p. 13617-13628.

31. Lawson, R.J., et al. (2004). Expression and characterization of the two flavodoxin proteins of Bacillus subtilis, YkuN and YkuP: Biophysical properties and interactions with cytochrome p450 BioI. Biochemistry. 43(39): p. 12390-12409.

32. Fujii, K., J.H. Galivan, and F.M. Huennekens (1977). Activation of Methionine Synthase - Further Characterization of Flavoprotein System. Archives of Biochemistry and Biophysics. 178(2): p. 662-670.

33. Kaihovaara, P., et al. (1998). Flavodoxin-dependent pyruvate oxidation, acetate production and metronidazole reduction by Helicobacter pylori. J Antimicrob Chemother. 41(2): p. 171-7.

34. Peleato, M.L., et al. (1994). Isolation and characterization of two different flavodoxins from the eukaryote Chlorella fusca. Biochem J. 302 ( Pt 3): p. 807-11.

35. Bagby, S., et al. (1991). Direct electrochemistry of two genetically distinct flavodoxins isolated from Azotobacter chroococcum grown under nitrogen-fixing conditions. Biochem J. 277 ( Pt 2): p. 313-9.

36. Gaudu, P. and B. Weiss (2000). Flavodoxin mutants of Escherichia coli K-12. J Bacteriol. 182(7): p. 1788-93.

37. Alagaratnam, S., et al. (2005). A crystallographic study of Cys69Ala flavodoxin II from Azotobacter vinelandii: structural determinants of redox potential. Protein Sci. 14(9): p. 2284-95.

38. Vicente, J.B., et al. (2002). Module fusion in an A-type flavoprotein from the cyanobacterium Synechocystis condenses a multiple-component pathway in a single polypeptide chain. Biochem Biophys Res Commun. 294(1): p. 82-7.

39. Wang, M., et al. (1997). Three-dimensional structure of NADPH-cytochrome P450 reductase: prototype for FMN- and FAD-containing enzymes. Proc Natl Acad Sci U S A. 94(16): p. 8411-6.

40. Peterson, J.A., et al. (1997). P450BM-3; a tale of two domains--or is it three? Steroids. 62(1): p. 117-23.

41. Hanley, S.C., T.W. Ost, and S. Daff (2004). The unusual redox properties of flavocytochrome P450 BM3 flavodoxin domain. Biochem Biophys Res Commun. 325(4): p. 1418-23.

42. Lamb, D.C., et al. (2006). A second FMN binding site in yeast NADPH-cytochrome P450 reductase suggests a mechanism of electron transfer by diflavin reductases. Structure. 14(1): p. 51-61.

43. Porter, T.D. and C.B. Kasper (1986). NADPH-cytochrome P-450 oxidoreductase: flavin mononucleotide and flavin adenine dinucleotide domains evolved from different flavoproteins. Biochemistry. 25(7): p. 1682-7.

44. Jenkins, C.M. and M.R. Waterman (1999). Flavodoxin as a model for the P450- interacting domain of NADPH cytochrome P450 reductase. Drug Metabolism Reviews. 31(1): p. 195-203.

45. McLean, K.J., et al. (2005). Biodiversity of cytochrome P450 redox systems. Biochemical Society Transactions. 33: p. 796-801.

46. Cenas, N., et al. (1994). The electron transfer reactions of NADPH: cytochrome P450 reductase with nonphysiological oxidants. Arch Biochem Biophys. 315(2): p. 400-6.

48. Frazao, C., et al. (2000). Structure of a dioxygen reduction enzyme from Desulfovibrio gigas. Nat Struct Biol. 7(11): p. 1041-5.

49. Victor, B.L., et al. (2003). Docking and electron transfer studies between rubredoxin and rubredoxin : oxygen oxidoreductase. Journal of Biological Inorganic Chemistry. 8(4): p. 475-488.

50. Yalloway, G.N., et al. (2003). Letter to the Editor: H-1, C-13 and N-15 assignment of the hydroquinone form of flavodoxin from Desulfovibrio valgaris (Hildenborough) and comparison of the chemical shift differences with respect to the oxidized state. Journal of Biomolecular Nmr. 25(3): p. 257-258.

51. Artali, R., et al. (2005). Structure of S35C flavodoxin mutant from Desulfovibrio vulgaris in the semiquinone state. Acta Crystallogr D Biol Crystallogr. 61(Pt 4): p. 481-4.

52. Zhou, Z. and R.P. Swenson (1996). The cumulative electrostatic effect of aromatic stacking interactions and the negative electrostatic environment of the flavin mononucleotide binding site is a major determinant of the reduction potential for the flavodoxin from Desulfovibrio vulgaris [Hildenborough]. Biochemistry. 35(50): p. 15980-8.

53. Zhou, Z. and R.P. Swenson (1995). Electrostatic effects of surface acidic amino acid residues on the oxidation-reduction potentials of the flavodoxin from Desulfovibrio vulgaris (Hildenborough). Biochemistry. 34(10): p. 3183-92.

54. Curley, G.P., et al. (1991). Redox and Flavin-Binding Properties of Recombinant Flavodoxin from Desulfovibrio vulgaris (Hildenborough). European Journal of Biochemistry. 202(3): p. 1091-1100.

55. Yalloway, G.N., et al. (1999). pH-Dependent spectroscopic changes associated with the hydroquinone of FMN in flavodoxins. Biochemistry. 38(12): p. 3753-3762.

56. Faraggi, M. and M.H. Klapper (1979). One-electron reduction of flavodoxin. A fast kinetic study. J Biol Chem. 254(17): p. 8139-42.

57. Ishikita, H. (2008). Redox potential difference between Desulfovibrio vulgaris and Clostridium beijerinckii flavodoxins. Biochemistry. 47(15): p. 4394-402.

58. Ishikita, H. (2007). Influence of the protein environment on the redox potentials of flavodoxins from Clostridium beijerinckii. J Biol Chem. 282(35): p. 25240-6.

59. Artali, R., et al. (2002). Comparison of the refined crystal structures of wild-type (1.34 A) flavodoxin from Desulfovibrio vulgaris and the S35C mutant (1.44 A) at 100 K. Acta Crystallogr D Biol Crystallogr. 58(Pt 10 Pt 2): p. 1787-92.

60. Hoover, D.M., et al. (1999). Comparisons of wild-type and mutant flavodoxins from Anacystis nidulans. Structural determinants of the redox potentials. J Mol Biol. 294(3): p. 725-43.

61. Chang, F.C. and R.P. Swenson (1999). The midpoint potentials for the oxidized- semiquinone couple for Gly57 mutants of the Clostridium beijerinckii flavodoxin correlate with changes in the hydrogen-bonding interaction with the proton on N(5) of the reduced flavin mononucleotide cofactor as measured by NMR chemical shift temperature dependencies. Biochemistry. 38(22): p. 7168-76.

62. Swenson, R.P. and G.D. Krey (1994). Site-directed mutagenesis of tyrosine-98 in the flavodoxin from Desulfovibrio vulgaris (Hildenborough): regulation of oxidation- reduction properties of the bound FMN cofactor by aromatic, solvent, and electrostatic interactions. Biochemistry. 33(28): p. 8505-14.

63. Stockman, B.J., T.E. Richardson, and R.P. Swenson (1994). Structural changes caused by site-directed mutagenesis of tyrosine-98 in Desulfovibrio vulgaris flavodoxin delineated by 1H and 15N NMR spectroscopy: implications for redox potential modulation. Biochemistry. 33(51): p. 15298-308.

64. Chang, F.C. and R.P. Swenson (1997). Regulation of oxidation-reduction potentials through redox-linked ionization in the Y98H mutant of the Desulfovibrio vulgaris [Hildenborough] flavodoxin: direct proton nuclear magnetic resonance spectroscopic

evidence for the redox-dependent shift in the pKa of Histidine-98. Biochemistry. 36(29): p. 9013-21.

65. Geoghegan, S.M., et al. (2000). Cloning, sequencing and expression of the gene for flavodoxin from Megasphaera elsdenii and the effects of removing the protein negative charge that is closest to N(1) of the bound FMN. Eur J Biochem. 267(14): p. 4434-44.

66. Lostao, A., et al. (1997). Differential stabilization of the three FMN redox forms by tyrosine 94 and tryptophan 57 in flavodoxin from Anabaena and its influence on the redox potentials. Biochemistry. 36(47): p. 14334-44.

67. Ludwig, M.L., et al. (1997). Control of oxidation-reduction potentials in flavodoxin from Clostridium beijerinckii: the role of conformation changes. Biochemistry. 36(6): p. 1259-80.

68. Freigang, J., et al. (2002). Crystal structure of oxidized flavodoxin, an essential protein in Helicobacter pylori. Protein Science. 11(2): p. 253-261.

69. Frago, S., et al. (2007). Tuning of the FMN binding and oxido-reduction properties by neighboring side chains in Anabaena flavodoxin. Arch Biochem Biophys. 467(2): p. 206-17.

70. Kasim, M. and R.P. Swenson (2000). Conformational energetics of a reverse turn in the Clostridium beijerinckii flavodoxin is directly coupled to the modulation of its oxidation-reduction potentials. Biochemistry. 39(50): p. 15322-32.

71. Druhan, L.J. and R.P. Swenson (1998). Role of methionine 56 in the control of the oxidation-reduction potentials of the Clostridium beijerinckii flavodoxin: effects of substitutions by aliphatic amino acids and evidence for a role of sulfur-flavin interactions. Biochemistry. 37(27): p. 9668-78.

72. Nogues, I., et al. (2004). Role of neighboring FMN side chains in the modulation of flavin reduction potentials and in the energetics of the FMN:apoprotein interaction in Anabaena flavodoxin. Biochemistry. 43(48): p. 15111-21.

73. Kasim, M. and R.P. Swenson (2001). Alanine-scanning of the 50's loop in the Clostridium beijerinckii flavodoxin: evaluation of additivity and the importance of interactions provided by the main chain in the modulation of the oxidation-reduction potentials. Biochemistry. 40(45): p. 13548-55.

74. O'Farrell, P.A., et al. (1998). Modulation of the redox potentials of FMN in Desulfovibrio vulgaris flavodoxin: thermodynamic properties and crystal structures of glycine-61 mutants. Biochemistry. 37(23): p. 8405-16.

75. Reynolds, R.A., W. Watt, and K.D. Watenpaugh (2001). Structures and comparison of the Y98H (2.0 A) and Y98W (1.5 A) mutants of flavodoxin (Desulfovibrio vulgaris). Acta Crystallogr D Biol Crystallogr. 57(Pt 4): p. 527-35.

76. McCarthy, A.A., et al. (2002). Crystallographic investigation of the role of aspartate 95 in the modulation of the redox potentials of Desulfovibrio vulgaris flavodoxin. Biochemistry. 41(36): p. 10950-62.

77. Bradley, L.H. and R.P. Swenson (1999). Role of glutamate-59 hydrogen bonded to N(3)H of the flavin mononucleotide cofactor in the modulation of the redox potentials of the Clostridium beijerinckii flavodoxin. Glutamate-59 is not responsible for the pH dependency but contributes to the stabilization of the flavin semiquinone. Biochemistry. 38(38): p. 12377-86.

78. Bradley, L.H. and R.P. Swenson (2001). Role of hydrogen bonding interactions to N(3)H of the flavin mononucleotide cofactor in the modulation of the redox potentials of the Clostridium beijerinckii flavodoxin. Biochemistry. 40(30): p. 8686-95.

79. Andrade, S.L., et al. (2005). Structures of the iron-sulfur flavoproteins from Methanosarcina thermophila and Archaeoglobus fulgidus. J Bacteriol. 187(11): p. 3848-54.

80. Leartsakulpanich, U., M.L. Antonkine, and J.G. Ferry (2000). Site-specific mutational analysis of a novel cysteine motif proposed to ligate the 4Fe-4S cluster in the iron- sulfur flavoprotein of the thermophilic methanoarchaeon Methanosarcina thermophila. J Bacteriol. 182(19): p. 5309-16.

81. Latimer, M.T., M.H. Painter, and J.G. Ferry (1996). Characterization of an iron-sulfur flavoprotein from Methanosarcina thermophila. J Biol Chem. 271(39): p. 24023-8.

82. Patridge, E.V. and J.G. Ferry (2006). WrbA from Escherichia coli and Archaeoglobus fulgidus is an NAD(P)H:quinone oxidoreductase. J Bacteriol. 188(10): p. 3498-506.

83. Grandori, R., et al. (1998). Biochemical characterization of WrbA, founding member of a new family of multimeric flavodoxin-like proteins. J Biol Chem. 273(33): p. 20960-6.

84. Carey, J., et al. (2007). WrbA bridges bacterial flavodoxins and eukaryotic NAD(P)H:quinone oxidoreductases. Protein Sci. 16(10): p. 2301-5.

85. Andrade, S.L., et al. (2007). Crystal structure of the NADH:quinone oxidoreductase WrbA from Escherichia coli. J Bacteriol. 189(24): p. 9101-7.

86. Hu, Y., et al. (2006). Solution structures and backbone dynamics of a flavodoxin MioC from Escherichia coli in both Apo- and Holo-forms: implications for cofactor binding and electron transfer. J Biol Chem. 281(46): p. 35454-66.

87. Zhao, T., F. Cruz, and J.G. Ferry (2001). Iron-sulfur flavoprotein (Isf) from Methanosarcina thermophila is the prototype of a widely distributed family. Journal of Bacteriology. 183(21): p. 6225-6233.

88. Cruz, F. and J.G. Ferry (2006). Interaction of iron-sulfur flavoprotein with oxygen and hydrogen peroxide. Biochim Biophys Acta. 1760(6): p. 858-64.

89. Becker, D.F., et al. (1998). Electrochemical and spectroscopic properties of the iron- sulfur flavoprotein from Methanosarcina thermophila. J Biol Chem. 273(41): p. 26462-9.

90. Brock, B.J. and M.H. Gold (1996). 1,4-Benzoquinone reductase from basidiomycete Phanerochaete chrysosporium: spectral and kinetic analysis. Arch Biochem Biophys. 331(1): p. 31-40.

91. Brock, B.J., S. Rieble, and M.H. Gold (1995). Purification and Characterization of a 1,4-Benzoquinone Reductase from the Basidiomycete Phanerochaete chrysosporium. Appl Environ Microbiol. 61(8): p. 3076-3081.

92. Chen, H., S.L. Hopper, and C.E. Cerniglia (2005). Biochemical and molecular characterization of an azoreductase from Staphylococcus aureus, a tetrameric NADPH-dependent flavoprotein. Microbiology. 151(Pt 5): p. 1433-41.

93. Liu, G., et al. (2007). Azoreductase from Rhodobacter sphaeroides AS1.1737 is a flavodoxin that also functions as nitroreductase and flavin mononucleotide reductase. Appl Microbiol Biotechnol. 76(6): p. 1271-9.

94. Long, D.J., 2nd, et al. (2001). NAD(P)H:quinone oxidoreductase 1 deficiency and increased susceptibility to 7,12-dimethylbenz[a]-anthracene-induced carcinogenesis in mouse skin. J Natl Cancer Inst. 93(15): p. 1166-70.

95. Nebert, D.W., et al. (2002). NAD(P)H:quinone oxidoreductase (NQO1) polymorphism, exposure to benzene, and predisposition to disease: a HuGE review. Genet Med. 4(2): p. 62-70.

96. Wan, J., et al. (2002). Association of genetic polymorphisms in CYP2E1, MPO, NQO1, GSTM1, and GSTT1 genes with benzene poisoning. Environ Health Perspect. 110(12): p. 1213-8.

97. Stiborova, M., et al. (2002). Carcinogenic aristolochic acids upon activation by DT- diaphorase form adducts found in DNA of patients with Chinese herbs nephropathy. Carcinogenesis. 23(4): p. 617-25.

98. Horie, S. (1990). Advances in research on DT-diaphorase--catalytic properties, regulation of activity and significance in the detoxication of foreign compounds. Kitasato Arch Exp Med. 63(1): p. 11-30.

99. Patterson, A.V., et al. (1998). Enzymology of tirapazamine metabolism: a review. Anticancer Drug Des. 13(6): p. 541-73.

100. Jaiswal, A.K. (2000). Regulation of genes encoding NAD(P)H:quinone oxidoreductases. Free Radic Biol Med. 29(3-4): p. 254-62.

101. Ross, D. and D. Siegel (2004). NAD(P)H:quinone oxidoreductase 1 (NQO1, DT- diaphorase), functions and pharmacogenetics. Methods Enzymol. 382: p. 115-44.

102. Ross, D. (2004). Quinone reductases multitasking in the metabolic world. Drug Metab Rev. 36(3-4): p. 639-54.

103. Ernster, L. (1967). DT diaphorase. Methods Enzymol. 10: p. 9.

104. Ross, D., et al. (2000). NAD(P)H:quinone oxidoreductase 1 (NQO1): chemoprotection, bioactivation, gene regulation and genetic polymorphisms. Chem Biol Interact. 129(1-2): p. 77-97.

105. Deller, S., P. Macheroux, and S. Sollner (2008). Flavin-dependent quinone reductases. Cell Mol Life Sci. 65(1): p. 141-60.

106. Long, D.J., 2nd and A.K. Jaiswal (2000). NRH:quinone oxidoreductase2 (NQO2). Chem Biol Interact. 129(1-2): p. 99-112.

107. Huang, M.T., G.T. Miwa, and A.Y. Lu (1979). Rat liver cytosolic azoreductase. Purification and characterization. J Biol Chem. 254(10): p. 3930-4.

108. Sollner, S., et al. (2007). Lot6p from Saccharomyces cerevisiae is a FMN-dependent reductase with a potential role in quinone detoxification. FEBS J. 274(5): p. 1328-39.

109. Liger, D., et al. (2004). Crystal structure and functional characterization of yeast YLR011wp, an enzyme with NAD(P)H-FMN and ferric iron reductase activities. J Biol Chem. 279(33): p. 34890-7.

110. Sugiura, W., et al. (2006). Expression and characterization of the genes encoding azoreductases from Bacillus subtilis and Geobacillus stearothermophilus. Biosci Biotechnol Biochem. 70(7): p. 1655-65.

111. Vorontsov, II, et al. (2007). Crystal structure of an apo form of Shigella flexneri ArsH protein with an NADPH-dependent FMN reductase activity. Protein Sci. 16(11): p. 2483-90.

112. Liu, G., et al. (2008). Enhancing survival of Escherichia coli by expression of azoreductase AZR possessing quinone reductase activity. Appl Microbiol Biotechnol. 80(3): p. 409-16.

113. Hong, Y., G. Wang, and R.J. Maier (2008). The NADPH quinone reductase MdaB confers oxidative stress resistance to Helicobacter hepaticus. Microb Pathog. 44(2): p. 169-74.

114. Ito, K., et al. (2008). Expansion of substrate specificity and catalytic mechanism of azoreductase by X-ray crystallography and site-directed mutagenesis. J Biol Chem. 283(20): p. 13889-96.

115. Punj, S. and G.H. John (2008). Purification and Identification of an FMN-dependent NAD(P)H Azoreductase from Enterococcus faecalis. Curr Issues Mol Biol. 11(1): p. 59-66.

116. Wang, C.J., et al. (2007). Molecular cloning, characterisation and ligand-bound structure of an azoreductase from Pseudomonas aeruginosa. J Mol Biol. 373(5): p. 1213-28.

117. Cohen, R., M.R. Suzuki, and K.E. Hammel (2004). Differential stress-induced regulation of two quinone reductases in the brown rot basidiomycete Gloeophyllum trabeum. Appl Environ Microbiol. 70(1): p. 324-31.

118. Phadtare, S., I. Kato, and M. Inouye (2002). DNA Microarray Analysis of the Expression Profile of Escherichia coli in Response to Treatment with 4,5-Dihydroxy- 2-Cyclopenten-1-One. J. Bacteriol. 184(23): p. 6725-6729.

119. Pomposiello, P.J., M.H. Bennik, and B. Demple (2001). Genome-wide transcriptional profiling of the Escherichia coli responses to superoxide stress and sodium salicylate. J Bacteriol. 183(13): p. 3890-902.

120. Colucci, M.A., C.J. Moody, and G.D. Couch (2008). Natural and synthetic quinones and their reduction by the quinone reductase enzyme NQO1: from synthetic organic chemistry to compounds with anticancer potential. Org Biomol Chem. 6(4): p. 637- 56.

121. Siegel, D., et al. (2004). NAD(P)H:quinone oxidoreductase 1: role as a superoxide scavenger. Mol Pharmacol. 65(5): p. 1238-47.

122. Garnaud, P.E., et al. (2004). Redox properties of the isolated flavin mononucleotide- and flavin adenine dinucleotide-binding domains of neuronal nitric oxide synthase. Biochemistry. 43(34): p. 11035-44.

123. Bachur, N.R., et al. (1979). NADPH cytochrome P-450 reductase activation of quinone anticancer agents to free radicals. Proc Natl Acad Sci U S A. 76(2): p. 954-7.

124. Bistolas, N., et al. (2005). Cytochrome P450 biosensors-a review. Biosens Bioelectron. 20(12): p. 2408-23.

125. Zhang, L., et al. (2002). The B' helix determines cytochrome P450nor specificity for the electron donors NADH and NADPH. J Biol Chem. 277(37): p. 33842-7.

126. Bistolas, N., et al. (2004). Spectroelectrochemistry of cytochrome P450cam. Biochem Biophys Res Commun. 314(3): p. 810-6.

127. Astrom, A. and J.W. DePierre (1986). Rat-liver microsomal cytochrome P-450: purification, characterization, multiplicity and induction. Biochim Biophys Acta. 853(1): p. 1-27.

128. Guengerich, F.P. (1995). Cytochrome P450 proteins and potential utilization in biodegradation. Environ Health Perspect. 103 Suppl 5: p. 25-8.

129. Cupp, M.J. and T.S. Tracy (1998). Cytochrome P450: New nomenclature and clinical implications. American Family Physician. 57(1): p. 107-116.

130. Cytochrome P450 protocols [digital]. 2nd ed. (2005), Totowa, NJ: Humana Press. computer files.

131. Gardner, A.M., R.A. Helmick, and P.R. Gardner (2002). Flavorubredoxin, an inducible catalyst for nitric oxide reduction and detoxification in Escherichia coli. J Biol Chem. 277(10): p. 8172-7.

132. Vicente, J.B. and M. Teixeira (2005). Redox and spectroscopic properties of the Escherichia coli nitric oxide-detoxifying system involving flavorubredoxin and its NADH-oxidizing redox partner. J Biol Chem. 280(41): p. 34599-608.

133. Silaghi-Dumitrescu, R., et al. (2003). A flavodiiron protein and high molecular weight rubredoxin from Moorella thermoacetica with nitric oxide reductase activity. Biochemistry. 42(10): p. 2806-15.

134. Das, A., et al. (2001). Five-gene cluster in Clostridium thermoaceticum consisting of two divergent operons encoding rubredoxin oxidoreductase- rubredoxin and rubrerythrin-type A flavoprotein- high-molecular-weight rubredoxin. J Bacteriol. 183(5): p. 1560-7.

135. Gomes, C.M., et al. (1997). Studies on the redox centers of the terminal oxidase from Desulfovibrio gigas and evidence for its interaction with rubredoxin. J Biol Chem. 272(36): p. 22502-8.

136. Ljungdahl, L.G., Biochemistry and physiology of anaerobic bacteria. (2003), New York: Springer. xvii, 270 p.

137. Chen, C.J., et al. (2003). Preparation and X-ray crystallographic analysis of rubredoxin crystals from Desulfovibrio gigas to beyond ultra-high 0.68 Å resolution. Biochem Biophys Res Commun. 308(4): p. 684-8.

138. Silva, G., et al. (2001). Analysis of the Desulfovibrio gigas transcriptional unit containing rubredoxin (rd) and rubredoxin-oxygen oxidoreductase (roo) genes and upstream ORFs. Biochem Biophys Res Commun. 280(2): p. 491-502.

139. Nolling, J., et al. (1995). Characterization of a 45-kDa flavoprotein and evidence for a rubredoxin, two proteins that could participate in electron transport from H2 to CO2 in

methanogenesis in Methanobacterium thermoautotrophicum. Eur J Biochem. 231(3): p. 628-38.

140. Guedon, E. and H. Petitdemange (2001). Identification of the gene encoding NADH- rubredoxin oxidoreductase in Clostridium acetobutylicum. Biochem Biophys Res Commun. 285(2): p. 496-502.

141. Chen, L., et al. (1993). Purification and characterization of an NADH-rubredoxin oxidoreductase involved in the utilization of oxygen by Desulfovibrio gigas. Eur J Biochem. 216(2): p. 443-8.

142. Sieker, L.C., R.E. Stenkamp, and J. LeGall (1994). Rubredoxin in crystalline state. Methods Enzymol. 243: p. 203-16.

143. Rodrigues, R., et al. (2006). Desulfovibrio gigas flavodiiron protein affords protection against nitrosative stress in vivo. J Bacteriol. 188(8): p. 2745-51.

CHAPTER THREE: WRBA FROM ESCHERICHIA COLI AND ARCHAEOGLOBUS FULGIDUS IS AN NAD(P)H:QUINONE OXIDOREDUCTASE 3.1 Abstract 3.2 Introduction 3.3 Materials and Methods 3.4 Results 3.5 Discussion 3.6 Acknowledgements 3.7 References

Chapter 3

WrbA from Escherichia coli and Archaeoglobus fulgidus is an NAD(P)H:Quinone Oxidoreductase

Eric V. Patridge and James G. Ferry*

Department of Biochemistry and Molecular Biology, Eberly College of Science, The Pennsylvania State University, 205 South Frear Laboratory, University Park, PA 16802-4500.

*Corresponding author. Tel.: +1 (814) 863-5721; Fax: +1(814) 863-6217; E-mail: [email protected]

30 sequences which indicated that WrbA is the prototype of a distinct family of flavoproteins which exists in a diversity of cell types across all three domains of life and includes documented NAD(P)H:quinone oxidoreductases (NQOs) from the Fungi and Viridiplantae

Kingdoms. Biochemical characterization of the prototypic WrbA from E. coli and WrbA from Archaeoglobus fulgidus, a hyperthermophilic species from the Archaea domain, shows that these enzymes have NQO activity, suggesting this activity is a defining characteristic of the WrbA family that we designate as a new type of NQO: Type IV NQOs. For EcWrbA the

NADH benzoquinone NADH Km was 14 ± 0.43 M and the Km was 5.8 ± 0.12 M. For AfWrbA the Km

benzoquinone was 19 ± 1.7 M and the Km was 37 ± 3.6 M. Both enzymes were found to be homodimeric by gel filtration chromatography, homotetrameric by dynamic light scattering, and to contain 1 FMN per monomer. The NQO activity of each enzyme is retained over a broad pH range, and apparent initial velocities indicate maximal activities are comparable to the optimum growth temperature for the respective organisms. The results are discussed and implicate WrbA in the 2-electron reduction of quinones, protecting against oxidative stress.

3.2 Introduction

The tryptophan (W) repressor binding protein (WrbA) from Escherichia coli

(EcWrbA) was discovered in 1993 when it co-purified with the tryptophan repressor (TrpR)

[1]. Biochemical characterization of EcWrbA showed that the protein binds one FMN per monomer and is multimeric in solution [2]. These results, combined with sequence analysis and homology-based structural modeling, led to the suggestion that EcWrbA is the founding member of a new family of multimeric flavodoxin-like proteins. It was predicted that the

WrbA family contains an  twisted open-sheet fold characteristic of flavodoxins and a conserved insertion after strand 4, forming an additional  unit [3]. The presence of this fold has recently been confirmed in the published crystal structures of the Deinococcus radiodurans and Pseudomonas aeruginosa WrbA homologues [4]. EcWrbA is the only

WrbA purified and biochemically characterized in the literature. No enzyme activity was reported; however, it was reported that EcWrbA enhances the development of non-covalent complexes between the TrpR holo-repressor and operator DNA, and that EcWrbA alone is incompetent to interact with the DNA targets [1]. Thus, it was proposed that EcWrbA is an accessory element which blocks TrpR-specific transcriptional events that are deleterious to cells entering stationary phase. However, it was later concluded that EcWrbA does not specifically affect the TrpR-DNA complex, placing the earlier proposed function in doubt

[2]. Thus, the function of WrbA is unknown.

Several global expression studies show that wrbA in E. coli is under the control of

RpoS (the stress-response sigma factor, s or 38). These studies report that wrbA is upregulated in response to stressors such as acids, salts, H2O2, and diauxie. EcWrbA is also upregulated in the early stages of the stationary phase, indicating that it could play a role in preparing the cell for long-term maintenance under stress conditions [5-12]. Two additional

68 expression studies implicate EcWrbA in oxidative stress. One study shows that E. coli wrbA is repressed when ArcA is phosphorylated, which occurs when the quinone pool is reduced

[13], and another study shows that E. coli wrbA is repressed by FNR (fumarate and nitrate reductase) regulatory protein under anaerobic conditions [14]. Aside from these findings, nothing is known about the role of WrbA in the stress-response.

The most comprehensive sequence analysis with EcWrbA was reported in 1994 by

Grandori (28). Since that time, genomic sequencing has identified over 100 genes annotated as encoding WrbA from metabolically and phylogenetically diverse prokaryotes spanning the

Bacteria and Archaea domains, consistent with a fundamental function for the WrbA family in the physiology of prokaryotes. However, EcWrbA is the only WrbA that has been biochemically characterized. Here we show that several biochemically characterized

NAD(P)H:quinone oxidoreductases (NQOs) from the Fungi and Viridiplantae Kingdoms have significant sequence identity to WrbA and belong to the same family [15-21]. We also report the initial biochemical characterization of WrbA from Archaeoglobus fulgidus

(AfWrbA), a hyperthermophile and a representative of the Archaea domain, and demonstrate that EcWrbA and AfWrbA exhibit robust NQO activity. The results presented are consistent with a role for WrbA in the oxidative stress response of diverse prokaryotes from the

Bacteria and Archaea domains.

3.3 Materials and Methods

Materials and reagents

Primers were obtained through Integrated DNA Technologies (Coralville, IA). E. coli

K-12 MG1655 genomic DNA was a gift from Sue-Jean Hong and Dr. Kenneth Keiler (The

Pennsylvania State University). All materials used to construct E. coli knockouts were gifts from Joe Palladino and Dr. Sarah Ades (The Pennsylvania State University). A. fulgidus genomic DNA was a gift from Dr. Michael W. Adams (University of Georgia). All cell lines and plasmids were obtained through Novagen (Madison, WI). The chromatography resins and equipment were purchased from Amersham Biosciences, (Piscataway, NJ). F420 and 2- hydroxyphenazine were gifts from Dr. Uwe Deppenmeier. All other chemicals were purchased from ICN (MP Biomedical), Sigma, or ACROS. Phenotypic analysis of the E. coli wrbA- strain was contracted to Phenotype MicroArray Services at Biolog, Inc. (Hayward, CA).

Sequence alignments and construction of the phylogenetic tree

Sequences were aligned using ClustalX (v1.83) using a BLOSUM62 matrix and default parameters. The output was edited using the Alignment Editor of MEGA (v3.1) and visualized with BioEdit (v7.0.5.2). The phylogenetic tree was constructed with the MEGA package. Distance matrices were generated in MEGA using the Minimum Evolution method with the Jones-Taylor-Thornton option and the close-neighbor-interchange option with a search level of 1. Gaps were deleted in a pairwise manner. The confidence limits of the nodes were estimated using the bootstrap option. Phylogenetic trees were also constructed using the

Parsimony, Neighbor-Joining, and UPGMA methods available in MEGA, but the Minimum

Evolution method produced the tree with the greatest bootstrap values.

Cloning, expression, purification, and reconstitution

The b1004 ORF was amplified by PCR from E. coli K-12 MG1655 genomic DNA, using sense (ATGGCTAAAGTTCTGGTGC) and anti-sense

(CCTCCTGTTGAAGATTAGCC) primers. The AF0343 ORF was amplified by PCR from

A. fulgidus genomic DNA, using sense (ATGGCCAGGATTCTTGTTATTTTTCATTCC) and anti-sense (CCCTTAGCAGAGCTTTTCAGCCACCTC) primers. Each PCR product was amplified and cloned into pETBlue-1 to obtain the recombinant plasmids: pETecb1004 from b1004 and pETaf0343 from AF0343. E. coli NovaBlue cells were used to amplify each plasmid. DNA sequencing confirmed that E. coli wrbA and A. fulgidus wrbA were intact in pETecb1004 and pETaf0343, respectively. Since the AF0343 ORF contains 12% rare codons with 2 rare-codon repeats, pETecb1004 and pETaf0343 were each transformed into E. coli

RosettaBlue(DE3)pLacI competent cells for expression. The transformed cells were cultured at 37ºC in Terrific Broth containing 50 g/ml ampicillin and 34 g/ml chloramphenicol.

When the OD600 reached 0.6, production of EcWrbA or AfWrbA was induced by the addition of 0.4 mM IPTG at 37ºC. Cells were harvested after 4 hours by centrifugation and stored at -

80ºC.

Initial purification and reconstitution of AfWrbA was carried out anaerobically using an anaerobic chamber (COY Laboratory Products, Grass Lakes, MI) and at room temperature except where indicated. A comparison of the aerobic and anaerobic protocol showed no difference in measured activities of purified protein; thus, all subsequent purifications were carried out aerobically. For AfWrbA, thawed cells (10 g wet weight) were re-suspended in 45 ml of 50 mM sodium phosphate buffer (pH 6.6) and lysed by three passes through a French press at 20,000 lb/in2 (137.9 MPa). Cell debris was removed by centrifugation 65,000 x g for

45 min at 4ºC, and the cleared lysate was diluted to 80 ml with 50 mM sodium phosphate

71 buffer (pH 6.6) and incubated at 75ºC for 30 min while stirring. Denatured proteins were pelleted by centrifugation at 65,000 x g for 45 min at 4ºC. The supernatant was filtered and loaded onto a Q-sepharose column equilibrated with 50 mM sodium phosphate buffer (pH

6.6). The column was developed with a 0.0 to 0.4 M NaCl linear gradient over 500 ml, applied at 2 ml/min. Yellow fractions containing AfWrbA, monitored by SDS-PAGE, were pooled, diluted 4-fold with 50 mM sodium phosphate buffer (pH 7.8), and loaded onto a Q- sepharose column equilibrated with 50 mM sodium phosphate buffer (pH 7.8). The column was developed with a 0.0 to 0.4 M NaCl linear gradient over 500 ml, applied at 2 ml/min.

Yellow fractions containing AfWrbA, monitored by SDS-PAGE, were pooled and concentrated to 1 ml using a Vivacell 70 with a 10,000 MWCO membrane (Vivascience,

Hanover, Germany). The concentrated solution was loaded onto a Sephadex-200 column equilibrated with 150 mM MOPS (pH 7.2). Yellow fractions containing AfWrbA, monitored by SDS-PAGE, were pooled and incubated with 5 mM FMN for 16 hours at 4ºC. This solution was then dialyzed against 50 mM MOPS (pH 7.2) and concentrated to 2.5 ml.

Excess FMN was removed by passage through a PD-10 column and the solution containing

FMN-reconstituted AfWrbA was stored at -80ºC.

The same protocol was used for the purification and reconstitution of EcWrbA, except that heat denaturation was excluded from the protocol.

Cofactor Determination and Purification Analyses

The subunit molecular mass of each protein was estimated by 12% SDS-PAGE using low molecular-weight markers from Bio-Rad. Native molecular mass was estimated from the elution volume from a Sephadex-S200 gel filtration FPLC column calibrated with the following proteins of known molecular masses: bovine serum albumin (66 kDa), ovalbumin

(45 kDa), carbonic anhydrase (31 kDa), chymotrypsinogen (25 kDa), and RNase A (13.7 kDa). The buffer (50 mM MOPS, 150 mM NaCl, pH 7.2) was applied at 0.5 ml min-1.

Further oligomerization was estimated by dynamic light scattering analyses with a Viscotek

802 Dynamic Light Scattering Instrument (Malvern Instruments, UK), which were performed in triplicate with each analysis incorporating at least ten readings.

After reconstitution, the Pierce assay was used to determine protein concentrations. The flavin: monomer ratios were calculated after quantifying the amount of FMN released by acidification (5% trichloroacetic acid) using an extinction coefficient for FMN of 12.2 mM-1 cm-1 at 450 nm.

The flavin cofactor of EcWrbA was previously identified as FMN [2]. To identify the flavin in AfWrbA, the flavin cofactor was extracted from purified protein by acidification of the protein solution with 5% trichloroacetic acid. A Hewlett-Packard model 1050 HPLC system in conjunction with a Hewlett-Packard LiChrosorb C-18 reversed phase column was used to identify the flavin extracted from AfWrbA. Neutralized samples were injected into a mobile phase consisting of 10 mM potassium phosphate, 12.5% acetonitrile, 0.3% triethylamine, and 0.01% sodium azide, at a pH of 6.5. Elution of flavin was monitored at

450 nm.

Kinetic Characterization and Specific Activities

Reactions were monitored with a Varian Cary 50 spectrophotometer, in combination with a Peltier-thermostatted accessory and an anaerobic fluorescence cuvette (2 mm x 1 cm pathlength) from Starna (Atascadero, CA). Although fluorescence was not measured, the cuvette permitted small reaction volumes under anaerobic conditions, and the 2 mm pathlength permitted high concentrations of NADH to be accurately determined spectroscopically. Activities with AfWrbA were obtained at 65ºC and those with EcWrbA 73 were obtained at 37ºC unless where indicated. The reactions were initiated by the addition of enzyme after temperature equilibration of the assay mixture. All activities were performed in triplicate, and all non-enzymatic rates were taken into account.

NADH was used as the electron donor to determine the specific activities of EcWrbA and AfWrbA with a variety of electron acceptors. Reaction mixtures contained 50 mM

MOPS (pH 7.2) with 800 M NADH, 400 M electron acceptor, and variable concentrations of EcWrbA or AfWrbA. With 1,4-benzoquinone and 2,3-dihydroxy-5-methyl-1,4-

-1 -1 benzoquinone, oxidation of NADH was followed at 340 nm (340 = 6.22 mM cm ). With

-1 -1 menadione, NADH oxidation was monitored at 341 nm (341 = 6.22 mM cm ), which is the isosbestic point for menadione at both 37ºC and 65ºC. With naphthoquinone, oxidation was measured at the appropriate isosbestic point of naphthoquinone, which shifted from 340 nm

-1 -1 at 37ºC to 341 nm at 65ºC. With K3Fe(CN)6 (420 = 1.04 mM cm ) and DCPIP (610 = 21 mM-1cm-1), the rates of reduction were followed at the indicated wavelengths. The initial 10 s of each progress curve was taken as the apparent initial velocity. Non-enzymatic reactions represented < 8% of all enzymatic reaction conditions.

Steady-state kinetic studies were conducted in 50 mM MOPS (pH 7.2) at 37ºC with variable enzyme concentrations. The kinetic parameters for 1,4-benzoquinone were determined by monitoring the oxidation of 200 M NADH at 340 nm. The kinetic parameters for NAD(P)H were determined by monitoring the reduction of 200 

K3Fe(CN)6 at 420 nm. Benzoquinone and NAD(P)H concentrations were varied from 0.2 to

10 times the respective Km values. The initial 2 to 5 s of each progress curve was taken as the apparent initial velocity, and the Michaelis-Menton equation was fit to the data using

KaleidaGraph v3.5. Non-enzymatic reactions represented < 2% of all enzymatic reaction conditions.

The pH dependence of apparent initial velocities for electron transfer from NADH to menadione was measured over a wide pH range by monitoring the oxidation of NADH at

341 nm. Reaction mixtures contained 800 M NADH, 400 M menadione, and 0.1 M

EcWrbA or 0.03 M AfWrbA. For EcWrbA, a buffer mix with 50 mM each of Na-

Acetate/MES/MOPS/TAPS was used from pH 4.0 to 8.5. This buffer mix was also used for

AfWrbA from pH 4.0 to 8.0, and a second buffer mix (50 mM TAPS/50 mM glycine) was used from pH 7.5 to 9.5. The initial 10 s of each progress curve was taken as the apparent initial velocity. Non-enzymatic reactions represented < 8% of all enzymatic reaction conditions.

The temperature dependence of apparent initial velocities for electron transfer from

NADH to menadione was determined by monitoring the oxidation of NADH at 341 nm.

Reaction mixtures contained 800 M NADH, 400 M menadione, and 0.1 M EcWrbA or

0.03 M AfWrbA. All reactions were performed in 50 mM sodium phosphate (pH 7.2) at

25ºC. EcWrbA was assayed from 10ºC to 60ºC, and AfWrbA was assayed from 10ºC to

100ºC. The initial 5 s of each progress curve was taken as the apparent initial velocity. Non- enzymatic reactions represented < 8% of all enzymatic reaction conditions.

Construction and Phenotypic Analysis of E. coli wrbA

E. coli K-12 MG1655 (F- lambda- ilvG- rfb-50 rph-1) was used as the wild-type strain, and disruption of wrbA was conducted according to published methods [22]. Primers were constructed using the pKD4 template, and kanamycin resistant genes were eliminated by using the pCP20 FLP helper plasmid. Elimination of temperature sensitive plasmids and thermal induction of FLP synthesis were carried out in Luria Broth at 43ºC. The knockout was confirmed with external and internal primer analyses.

3.4 Results

Homology, Sequence Analysis, and Phylogeny

The most comprehensive sequence analysis of EcWrbA, the only biochemically characterized WrbA to date, was in 1994 when EcWrbA was suggested to be the prototype of a new family of flavoproteins [3]. However, the analysis included only six sequences of putative WrbA homologues; therefore, a BLAST search of all non-redundant databases was performed using the 198-residue protein sequence of EcWrbA, encoded by locus b1004 in E. coli K-12 MG1655, as the query. A survey of the returned 394 sequences revealed 128 predicted WrbA proteins that had between 22% and 99% identity to EcWrbA.

The genes neighboring the loci encoding the WrbA homologues were surveyed, but no consistencies were found to suggest a common function. The returned BLAST results included 20 sequences annotated as fungi and plant quinone reductases, suggesting these enzymes could be related to WrbA. Among the aligned data set were five documented

NAD(P)H:quinone oxidoreductases (NQOs) from the Fungi and Viridiplantae Kingdoms that are biochemically characterized (AtFQR1 from Arabidopsis thaliana, GtQR1 and

GtQR2 from Gloeophyllum trabeum, PcQR from Phanerochaete chrysosporium, and TvQR2 from Triphysaria versicolor) [15-20]. As seen in Figure 3-1, these five documented quinone reductases have between 43% and 48% identity to WrbA from E. coli. Replicate sequences, truncated sequences, and sequences with partial alignments were removed from the BLAST results, and an initial phylogenetic tree was constructed using an alignment of the remaining

208 sequences (not shown). From the collected sequences, 30 were selected to represent the initial tree. These sequences were aligned (Figure 3-1), and a bootstrapped phylogenetic tree was constructed using the Minimum Evolution method with 3 flavodoxin sequences to root the tree (Figure 3-2).

======# # # PJS666WrbA ------MPKVVVLYHSGYGHTQRMAQSVAQGAG--A-ELLAID------ADGNLP------DGG----WET------LNAADAIIMGSPTYM-GSVSWQFKKFADASSKP-WYTQA EcaWrbA ------MSKTVVIYHSGYGHTQRLAAAVAEGAN--A-ELIAID------AEGNIS------DAE----W------EKLNAADAIIFGTPTYM-GGPTWQFKKFADASSKA-WFSRT BcWrbA ------MSNIVIVYHSGYGHTQKLAEAVHAGAQE-AGATVRLL------AVGDVD------DAG----W------ATLDAADAIVFGAPTYM-GGPSAQFKQFADATSKA-WFTQK XfWrbA -----MLIS--IVYDSGYGHTARVAQAVAAGVQEMKGADVRLM------AVADG------SVD----W------EALEKSEAIIFGSPTYN-GLISAKFKQFFEDSTKKAWVSQK NpWrbA ------MPTVAIIYFSGAGHTHLMAQAIAEGATKVEDTTVELL----RVT-GEQIVNGRW--KNDEGL------EKLNQADAIVFGSPTYM-GGVAAQFKAFIDAASEV-WFRHG CaWrbA --- TTPTCR-IFILHDHGEHVTQLAQAIGEGA--AGVTGV---TVKISLP------HHATK----ADLLEADGIIIGTPNWT-GI-KGTLKRWLDTTGDL-WEEGS AfWrbA ------MARILVIFHSITGNTMKLAKAVADGARE-GGAEVAVKRVPETIPAEILEKNPGYVKVRE-ELESFEVAR---PEELQDYDAIIFGSPTRF-GVMSSQMKQFIDMTGRL-WMERR PaeWrbA --GSMSSPYILVLYYSRHGATAEMARQIARGVE-QGGFEARVRTVPAVSTECEAVAPD---IPAEGA---LYATL----EDLKNCAGLALGSPTRF-GNMASPLKYFLDGTSSL-WLTGS ParWrbA --MSQTAPYVLVLYYSNYGTTKTLAYAIAQGIEEAGM-TARIRTVPTVAPETTASKPA---IPDEGD---LYCTM----DDLKNCSGLALGSPTHF-GNMAAPMKYFWDNTVTI-WLAGN MdWrbA ---- MPTPYVLILYYSHKGSTKQLASAIARGVEQAGNIEARLRTVPKVAPVTQVAEPA---IPDDGD---LYCTE----DDLINCAGLALGSPTRF-GNMAAPLKYFLDGTATQ-WVNGQ StWrbA ---MSCKPNILVLFYG-YGSIVELAKEIGKGAEEAGA-EVKIRRVRETLPP-EFQS----RIPFDKVKDIPEVTL----DDMRWADGFAIGSPTRY-GNMAGGLKTFLDTTAIL-WKDNV SsWrbA ---MEC-PKILVLFYG-YGSIVDLAKNVAEGAKEITK-EVKLARVKEYFPQ-EIVNKF--RIPIDTVKDIPEATL----SDLEWADGIVMGSPTRY-GNMTGQLKLFLDQTAEL-WIKGS EcoWrbA ------MAKVLVLYYSMYGHIETMARAVAEGASKVDGAEVVVKRVPETMPPQLFEKAG-----GKT-QTAPVAT----PQELADYDAIIFGTPTRF-GNMSGQMRTFLDQTGGL-WASGA YpWrbA -MEHREMA KILVLYYSMYGHIETLAGAIAEGARKVSGVDVTIKRVPETMPAEAFAKAG-----GKTNQQAPVAT----PHELADYDGIIFGTPTRF-GNMSGQMRTFLDQTGGL-WASGA CcWrbA ------MAKVLVLYYSSYGHLEVMAKAIAEGAREA-GASVDIKRVPETVPLEIAKGAH-----FKLDQDAPVAK----VEDLADYDAIIVGAPTRF-GRMASQMAAFFDAAGGL-WARGA MBNC1WrbA ------MAKILVLYYSSWGHMEAMAMAAARGAGEAGA-NVTIKRVPELVPEEVARKAH-----YKLEQEAQIAT----PLELADYDGFIFGVSTRY-GMMSSQLKNFLDQTGPL-WAAGK MacWrbA ------MVKVNIIFYSMYGHVYRMAEAVAAGAREVEGAEVGIYQVPETLPEEVLEKMGA-IETKKLFAHIPVLTREMNEEVLAGADALIFGTPTRY-GNMTAQMRAVLDGLGGL-WNRDA MmWrbA ------MVKVNIIFHSVHAHIYRMAEAVAAGAREVEGAEVGIYQVPETLPEDVLEKMGA-IETKKLFAHIPVVTRDMYEDVLAGADALIFGTPTRY-GMMTAQMRAVLDGLGKL-WSEDA MbWrbA ------MVKVNVIFHSIHGHTYKMAEAIAEGAREVEGAEVEIYQVPETLPYEVLEKMGA-IETKNLFAHIPVVTRSMYEDVFAGADALIFGTPTRY-GNMTAQMRTVFDGLGGL-WSRDA GtQR1 (35)--AVMSSPKVAIVIYSLYGHIAKLAEAVKSGIESAGG-KAQIFQVPETLSEDILKLLH----APPK--DYPIIT----PEQLATFDAFLIGIPTRY-GNFPAQWKAFWDATGQL-WATGA PcQR ------MPKVAIIIYSMYGHIAKLAEAEKAGIEEAGG-SATIYQIPETLPEEVLAKMH----APPK--EYPVIT----PEKLPEFDAFVFGIPTRY-GNFPGQWKAFWDATGGL-WAQGA GtQR2 (50)TDTTMSSP RLAIVIYTMYGHVAKLAEAIKSGIEGAGG-NASIFQVAETLSPEILNLVK----APPK--DYPVMD----PLDLKNYDGFLFGIPTRY-GNFPVQWKAFWDSTGPL-WASTA SpObr1 ---MSTANTVAIVIYSTYGHVVKLAEAEKAGIEKAGG-KAVIYQFPETLSPEILEKMH----AAPKP-NYPVVT----LDVLTQYDAFLFGYPTRY-GTPPAQFRTFWDSTGGL-WVQGA NcHyp -----MAPKIAIVYYSMYGHIRQLAEAAKAGIEKAGG-TADLYQVPETLSDEVLAKMY----APPKPTDIPVIED---PAILKEYDGFLFGIPTRY-GNFPAQWRAFWDKTGGL-WATGG YlHyp ------MPKIAILIYSTYGHIAELARKEAEGVKAAGG-QVDLYQVEETLSEEILKYMK----APGQPHDFKVLTPAD-VEVLTGYDGFLFGIPSRF-GSFPAQWKTFWDATGGL-WATGG AtFQR1 -----MATKVYIVYYSMYGHVEKLAEEIRKGAASVEGVEAK LWQVPETLHEEALSKMS----APPKS-ESPIIT----PNELAEADGFVFGFPTRF-GMMAAQFKAFLDATGGL-WRAQA TvQR2 -----MATKVYIVYYSTYGHVERLAQEIKKGAESVGNVEVKLWQVPEILSDEVLGKMW----APPKS-DVPVIT----PDELVEADGIIFGFPTRF-GMMAAQFKAFFDSTGGL-WKTQA E255-15WrbA -MSN---VKLAVIYYSSTGTNHQLATWAKEAGEAAGA-EVRLAKIKETAPAEAIASNPLWQKHTDATRDVEEATP----DLLDWADAIIFSVPTRY-GHVPGQFQQFLDTTGGL-WGQGK DrWrbA -MSLTAPVKLAIVFYSSTGTGYAMAQEAAEAGRAAGA-EVRLLKVRETAPQDVIDGQDAWKANIEAMKDVPEATP----ADLEWAEAIVFSSPTRF-GGATSQMRAFIDTLGGL-WSSGK MavWrbA ------MTKLAIIYYSATGHGTTMARRVAAAAESAGA-QVRLRHVAETQDPESFAHNPAWTANYQATKDLPAATG----DDIVWADAVIFGSPTRF-GSPAAQLRAFLDSLGGL-WAQGK ------SSSSSSS----HHHHHHHHHHHHHHHH----SSSSS------SSSSS------HHHHHHHHHH-HHHHHHH- ....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....| 10 20 30 40 50 60 70 80 90 100 110 120 CbFldx ------MK--IVYWSGTGNTEKMAELIAKGIIESGK---DVNTINV--SDVNI------DELLNEDILILGC-SAM-GDEVLEESEFEPFIEEI---STK DgFldx ------MPKALIVYGSTTGNTEGVAEAIAKTLN-SEGMETTVVNVADVTAP------GLAEGYDVVLLGCSTWG-DDEIELQEDFVPLYEDL--DRAG EcoFldx ------MAITGIFFGSDTGNTENIAKMIQKQLG------KDVADVHDIAKSSK------EDLEAYDILLLGIPTWYYGEAQCDWDDFFPTLEEI-----D

# # # PJS666WrbA WKDKLFAGFTNSAALDGDKMSTLNYLFTLAMQH---GGIWVSQGVLPSTSKAAQR-DDANYLGSY-RGAIAQSPSDAGAGE-MFPG-DLETARSLGERVAEVAGKFQR----- 31/47 EcaWrbA WSNKVFGGFTNSASLNGDKQVTLIYLQTLASQH---GGIWVSLNQLPSNAKAAKR-DDLNNLGGS-VGLLAQTPSDASADE-VVAG-DLATGKLYGQRIADIAAKLAN----- 27/44 BcWrbA WKDKIAAGFTNSATMNGDKFSTIQYFVTLSMQH---GMVWVGTSLMPANSKAATR-NDINYLGGS-TGLLAQSPADSTPDEGPLPG-DLETGKAFGRRVAEATARWVAGRG-- 28/44 XfWrbA WRNKIAAGFTNSGAQHGDKLNSLMSMVLFAAQH---GMIWVGLDLMPGNNSSTGSSDDLNRLGSW-LGVMTQSNNDAAPEIAPPAS-DLKTAQHLGLRVAETVRRFQPV---- 30/47 NpWrbA WKDKIAAGFTHSSSPSGDKQGTLLYLATNAGQH---GMIWVNVGDLQSFL-SGL-DDGVNRLGGF-LGVMGQSQLDMSGKEPVLDSGDYLTSVRFGERIAEATKRWNK----- 28/46 CaWrbA LSGKVGAAFTSSAGRHSGTEFTLLSVLHWMLGS---GMIIVGLPWSPL------MEQ-AGSYYG--ATAV-G-----TITEA-DLIQARALGRRVAELALQLRRGAAS- 30/47 AfWrbA LEGKVGAVFTSNEMPHGGKEATLLSMLLPLFAH---GMIIVGLPPAKE------LYR-AGSYYG--AAST-G-----VPKED-DLQVAKMLGKRVAEVAEKLC------43/63 PaeWrbA LVGKPAAVFTSTASLHGGQETTQLSMLLPLLHH---GMLVLGIPYSEP------ALLETRG-GGTPYG--ASHFAGADGKRSLDEH-ELTLCRALGKRLAETAGKLGS----- 40/56 ParWrbA LQNKPAAVFTATGSMHGGQETTLLTMMLPLLHH---GMMIVGVPYAEP------ALNRTTR-GGSPYG--ASHVSGASHDQPVSAD-ERELGIAQGRRLAITASALAQANWSR 39/54 MdWrbA LIGKPAGVFTSTSSMHGGQESTLLSMMVPLLHH---GMVLCGIPYAEQ------ALHTTAT-GGTPYG--PSHLAGSDGTTQLSQE-EKALCVAFGKRLATLAKKLYD----- 36/54 StWrbA LYGKPVTFFTEASTVHGGHETTILTMSTYAYHF---GMIIVPIGYGIP------ELFQTTT-GGGPYG--ATHL-G--SKEELDEM-ERKIARFQGKRITEVAKAIKCCNK-- 40/57 SsWrbA LYGKPVGFFTEASTMHGGHESTILAMANYAYHH---GMIIVPVGYGIK------EVSSTMT-GGSPYG--ASHL-G--NKKELDEN-EINIAKFLGKRVAEVAKKLRC----- 40/59 EcoWrbA LYGKLASVFSSTGT-GGGQEQTITSTWTTLAHH---GMVIVPIAYAAQ----ELFDVSQVR-GGTPYG--ATTIAGGDGSRQPSQE-ELSIARYQGEYVAGLAVKLNG------YpWrbA LYGKVASVFASTGT-GGGQEHTITSTWTTLAHH---GFIIVPIGYGAK----ELFDVSQTR-GGTPYG--ATTIAGGDGSRQPSAE-ELAIARFQGEHVAKITAKLKG----- 81/88 CcWrbA LHGKVAGAFTSTATQHGGQETTLFSIITNMLHF---GTTIVGLDYGHA----GQMTLDEIT-GGSPYG--ATTIAGGDGSRQPSEN-ELTGARYQGRKIAETAIKLHG----- 57/71 MBNC1WrbA LVNKPATVMVSTATQHGGAEIALASTQLALQHH---GMIIVPLSYAYQ----GQSGNDTVR-GGAPYG--MTTTSDTDGSRMSAQ--ELEGARFQGKRLAEITAKLVR----- 54/66 MacWrbA FVGKVGSVFTSSGTQHGGQESTILTFHVTLLHL---GMILVGLPYSEK----RQTRMDEIT-GGSPYG--VSTIAGGDGSRQPSEN-ELAMARYQGRHVTLIAKKIAGK---- 50/68 MmWrbA FVGKVGSVFTSSGTQHGGQESTILSFHVTLLHL---GMVIVGLPYAEK----RQTIMNEIT-GGSPYG--ASTIAGGDGSRQPSEN-ELEMARYQGRHVTQIAKKIAGK---- 50/67 MbWrbA LVGKVGSVFTSSGTQHGGQESTILTTHVTLLHL---GMIIVGLPYSET----RQRRMDEIT-GGSPYG--ASTIAGAEENRQPSEN-ELAMARYQGRHVTQIAKKLIG----- 50/68 GtQR1 LAGKYAGLFVSTASPGGGQESTAIAAMSTFAHH---GLIYVPLGYKHTF--AQLTNLNELR-GGSPWG--AGTFAGGDGSRQPTPL-ELEVATIQGKTFYETVSKVKF----- 43/63 PcQR LAGKYASVFVSTGTPGGGQESTVLNSISTLTHH---GIVFVPLGYSTTF--AQLANLSEVR-GGSPWG--AGTFAGADGSRSPSAL-ELELATAQGKYFWNIIKKVAF----- 48/66 GtQR2 LCGKYAGLFVSTGSPGGGQESTLMAAMSTLVHH---GVIYVPLGYKYTF--AQLANLTEVR-GGSPWG--AGTFANSDGSRQPTPL-ELEIANLQGKSFYEYVARVKW----- 44/61 SpObr1 LHGKYFGQFFSTGTLGGGQESTALTAMTSFVHH---GMIFVPLGYKNTF--SLMANVESIH-GGSSWG--AGSYAGADGSRNVSDD-ELEIARIQGETFFKTVFRK------45/58 NcHyp LYGKAAGLFISTAGLGGGQESTAIAAMSTLAHH---GIIYVPLGYAKVF--GELSDLSAVH-GGSPWG--SGTLSGGDGSRQPSES-ELKVAGIQGEEFYNTLSKLTASQPKP(35) 48/62 YlHyp LHGKYVGQFVSSGTQGGGQEVLPRNTLSIYVHH---GMNYVPLGYKDTF--AEQTNLEEVH-GGSPWG--AGTFADSDGSRTPSPL-EEKVAFTQGKVFTEVLIKAEGGAAGA(147) 42/67 AtFQR1 LAGKPAGIFYSTGSQGGGQETTALTAITQLVHH---GMLFVPIGYTFG---AGMFEMENVK-GGSPYG--AGTFAG-DGSRQPTEL-ELQQAFHQGQYIASITKKLKGSTA-- 48/66 TvQR2 LAGKPAGIFFSTGTQGGGQETTALTAITQLTHH---GMIYVPIGYTFG---ADMFNMEKIK-GGSPYG--AGTFAGADGSRQPSDI-ELKQAFHQGMYIAGITKKIKQTSA-- 28/65 E255-15WrbA LVNKVVSAMSSAQNPHGGQEATVLAVYTSMFHW---GAIVAAPGYSDP------VLFAA-GGNPYG--TTVTVDQD--GNMVES-VEPAVRHQAKRTVDIASRIKG----- 35/50 DrWrbA LANKTFSAMTSAQNVNGGQETTLQTLYMTAMHW---GAVLTPPGYTDEV------IFKS-GGNPYG--ASVTANGQ---PLLEN-DRASIRHQVRRQVELTAKLLEGGS-- 43/56 MavHyp LADKVYAAFTSSNTLHGGQETTLLSLYITLMHF---GGIIVAPGYTDP------SKFVD--GNPYG--ASLVTTHDNIHEFDEP-TANALDHLARRVVETAGRLAS----- 41/55 SSSSSSS------HHHHHHHHHHHHH--- SSSS------SSSS----SSSSSSSSS------HHHHHHHHHHHHHHHHHHH------|  ....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|... 130 140 150 160 170 180 190 200 210 220 230 CbFldx ISGKKVALFGSYGWGDGKWMRDFEERMN---GY---GCVVVETPLIVQ------NEPDEA-EQDCIEF-GKKIANI------ns/ns DgFldx LKDKKVGVFGCGDSSYTYFCGAVDVIEKKAEEL---GATLVASSLKID------GEPDS---AEVLDWAREVLARV------ns/ns EcoFldx FNGKLVALFGCGD-QEDYAEYFCDALGTIRDIIEPRGATIVGHWPTAG------YHFEASKGLADD---DHFVGLAIDEDRQPELTAERVEKWVKQISEELHLDEILNA-- ns/ns

Figure 3-1. Alignment of WrbA, NAD(P)H:quinone oxidoreductase, and flavodoxin sequences. Numerical values at the C-terminus indicate % identity/ % similarity to EcoWrbA. (ns) indicates “not significant.” (#) indicates completely conserved residues. Shaded residues indicate a conservation of similar residues persists across at least 60% of the alignment. (====) indicates the region of the flavodoxin signature motif. JPred at ExPASy’s Proteomic Tools was used to predict (HHHH) -helix or (SSSS) -sheet using the alignment (flavodoxin sequences were excluded). The additional  unit that is discussed is indicated in bold below the secondary structure prediction. (continued on next page)

Figure 3-1 (cont’d). Organisms and ORF designations from the corresponding genomic sequence: Archaeoglobus fulgidus AfWrbA (gi:11497955), Arabidopsis thaliana AtFQR1

(gi:21539481), Burkholderia cepacia BcWrbA (gi:46310790), Chloroflexus aurantiacus

CaWrbA (gi:53798190), Caulobacter crescentus CcWrbA (gi:13422034), Clostridium beijerinckii CbFldx (gi:1941945), Desulfovibrio gigas DgFldx (gi:40801), Deinococcus radiodurans DrWrbA(gi:58177611), Exiguobacterium sp. 255-15 E225-15WrbA (gi:68054352),

Erwinia carotovora EcaWrbA (gi:49613050), Escherichia coli EcoWrbA (gi:148264),

Gloeophyllum trabeum GtQR1 (gi:30027749) and GtQR2 (gi:33668045), Methanosarcina acetivorans MacWrbA (gi:19915027), Mycobacterium avium MavWrbA (gi:41409133),

Methanosarcina barkeri MbWrbA (gi:31793946) and MbFldx (gi:72396729), Mezorhizobium sp. BNC1 MBNC1WrbA(gi:68191386), Microbulbifer degredans MdWrbA (gi:48860599),

Methanosarcina mazei MmWrbA (gi:21228326), Neurospora crassa NcHyp (gi:38566940),

Nostoc punctiforme NpWrbA (gi:23129682), Psychrobacter arcticum ParWrbA(gi:71038102),

Phanerochaete chrysosporium PcQR (gi:4454993), Polaromonas sp. JS666

PJS666WrbA(gi:67847813), Pseudomonas aeruginosa PaeWrbA (gi:9946855),

Schizosaccharomyces pombe SpObr1 (gi:2462689), Sulfolobus solfataricus SsWrbA

(gi:15899861), Sulfolobus todakii StWrbA (gi:15621890), Triphysaria versicolor TvQR2

(gi:12484052), Xylella fastidiosa XfWrbA (gi:22993897), Yarrowia lipolytica YlHyp

(gi:49647887), Yersinia pestis YpWrbA (gi:22126332).

Figure 3-2. Phylogenetic tree of selected WrbA, NAD(P)H:quinone oxidoreductase, and flavodoxin sequences. The full alignment of these sequences is provided as Figure 1. The scale represents the average number of amino acid substitutions per site. EcWrbA and

AfWrbA appear in bold text. (continued on next page)

Figure 3-2 (cont’d). Organisms and ORF designations from the corresponding genomic sequence: Archaeoglobus fulgidus WrbA (gi:11497955), Arabidopsis thaliana FQR1

(gi:21539481), Burkholderia cepacia WrbA (gi:46310790), Chloroflexus aurantiacus WrbA

(gi:53798190), Caulobacter crescentus WrbA (gi:13422034), Clostridium beijerinckii Fldx

(gi:1941945), Desulfovibrio gigas Fldx (gi:40801), Deinococcus radiodurans WrbA (gi:58177611),

Exiguobacterium sp. 255-15 WrbA (gi:68054352), Erwinia carotovora WrbA (gi:49613050),

Escherichia coli WrbA (gi:148264) and Fldx (gi:145986), Gloeophyllum trabeum QR1

(gi:30027749) and QR2 (gi:33668045), Methanosarcina acetivorans WrbA (gi:19915027),

Mycobacterium avium WrbA (gi:41409133), Methanosarcina barkeri WrbA (gi:31793946),

Mezorhizobium sp. BNC1 WrbA (gi:68191386), Microbulbifer degredans WrbA (gi:48860599),

Methanosarcina mazei WrbA (gi:21228326), Neurospora crassa WrbA (gi:38566940), Nostoc punctiforme WrbA (gi:23129682), Psychrobacter arcticum WrbA (gi:71038102), Phanerochaete chrysosporium QR (gi:4454993), Polaromonas sp. JS666 WrbA (gi:67847813), Pseudomonas aeruginosa WrbA (gi:9946855), Schizosaccharomyces pombe Obr1 (gi:2462689), Sulfolobus solfataricus WrbA (gi:15899861), Sulfolobus todakii WrbA (gi:15621890), Triphysaria versicolor

QR2 (gi:12484052), Xylella fastidiosa WrbA (gi:22993897), Yarrowia lipolytica WrbA

(gi:49647887), Yersinia pestis WrbA (gi:22126332).

The clustering of the initial phylogenetic tree indicated that all of the proteins included in the data set diverged from a common ancestor distinct from classical flavodoxins.

Moreover, these results identify NQOs as members of the WrbA family. The tree contained a diverse set of organisms from all three domains of life which included psychrophiles, mesophiles, thermophiles, animal and plant pathogens, and aerobes and anaerobes. The tree also contained 8 NQO sequences from Fungi and Viridiplantae Kingdoms and 15 and 6

WrbA sequences from the Bacteria and Archaea Domains, respectively.

The PROSITE flavodoxin signature sequence [LIV]-[LIVFY]-[FY]-x-[ST]-{V}-x-

[AGC]-x-T-{P}-x-x-A-x-x-[LIV], indicative of the N-terminal region that spans an FMN binding site, was present across all aligned sequences with little deviation from the motif

(Figure 3-1). The  twisted open-sheet fold typical of flavodoxins is present across all sequences of the aligned data set, with an additional  unit previously reported for the

WrbA family [3]. The invariant conservation of this additional  unit (Figure 1) further supports that NQO and WrbA proteins diverged from a common ancestor distinct from classical flavodoxins. These results solidify the earlier proposal [3] for a new family of flavoproteins and further extends the WrbA family to all three domains of life.

In surveying the BLAST results, it was discovered that several sequences annotated as WrbA contain a motif strictly conserved in another FMN-containing protein family called iron-sulfur flavoprotein (Isf) [23-27]. However, the Isf family contains an unusual compact cysteine motif (CX2CX2CX5-7C) that binds a 4Fe-4S cluster not found in the sequences of

WrbA proteins or the NQO enzymes afore mentioned. A sequence alignment was constructed to address the relatedness between the Isf and WrbA families. The alignment included EcWrbA and AfWrbA representing the WrbA family, three Isf sequences including the prototype (MtIsf from Methanosarcina thermophila), and seven misannotated WrbA

81 sequences that contained the compact cysteine motif. Flavodoxin 5NLL from C. beijerinckii was included for comparison to characterized flavodoxins (Figure 3-3). The alignment demonstrates that the misannotated WrbA sequences with the compact cysteine motif belong to the Isf family. While Figure 3 indicates that similarity persists across the alignment, the overall identity between WrbA and Isf sequences suggests significant deviation from a common ancestor. In deviating from the common ancestor, it is possible that either Isf lost the domain containing the 4Fe-4S cluster-binding motif giving rise to the WrbA family, or

WrbA acquired the domain giving rise to the Isf family.

======# # # ▲ ▲ ▲ ▲ # AfWrbA MARILVIFHSIT--GNTMKLAKAVADGA-REGGAEVAVKRVPETIPAEILEKNPGYVKVREELESFEVARPEELQDYDAI EcWrbA MAKVLVLYYSMY--GHIETMARAVAEGASKVDGAEVVVKRVPETMPPQLFEKAGG------KTQTAPVATPQELADYDAI MtIsf -MKITGISGSPRKGQNCEKIIGAALEVAKER-GFETDTVFISNEEVAPCKACGACRDQDFCVIDDDMDEIYEKMRAADGI AfIsf -MKLLAINGSPN-KRNTLFLLEVIAEEVKKL-GHEAEIIHLKDYEIKECKGCDACLK-GDCSQKDDIYKVLEKMQEADAI MjIsf -MKVIGISGSPRPEGNTTLLVREALNAIAEE-GIETEFISLADKELNPCIGCNMCKEEGKCPIIDDVDEILKKMKEADGI MkWrbA --MILGISGSPR-EGNTE YLVRIALEAAEEVSGEETEFITVRDLDISPCEACGECLETGECAIDDDMQDVYELMRECDGM CaWrbA MSKIVIFKGSPRKNGYTAKLLEQVAKGAKSK-GAEVIEFDLNNPGIRGCQGCFYCRTHDGCAVNDYLQPMYEDIKEADAI ....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....| 10 20 30 40 50 60 70 80 CbFldx -MKIVYWSGTGN----TEKMAELIAKGIIES-GKDVNTINVSDVNI------DELLNE-DIL

# # AfWrbA IFGSPTRFGVMSSQMKQFIDMTGRLW-MERRLEGKVGAVFTSNEMPHGGKEATLLSMLLPLFAHGMIIVGLP-PAKELY- EcWrbA IFGTPTRFGNMSGQMRTFLDQTGGLW-ASGALYGKLASVFSSTGTG-GGQEQTITSTWTTLAHHGMVIVPIAYAAQELFD MtIsf IVAAPVYMGNYPAQLKALFDRSVLLRRKNFALKNKVGAALSVGGSRNGGQEKTIQSIHDWMHIHGMIVVGDN---SHFG- AfIsf VIGTPTYFGNVTGIVKNLIDRSRMARMGNYRLRNRVFAPVVTSGLRNGGAEYAAMSLIVYALGQAMLPVSIVENPITTGT MjIsf ILGSPVYFGGVSAQLKMLMDRSRPLR-IGFQLRNKVGGAVAVGASRNGGQETTIQQIHNFFLIHSMIVVGDNDPTAHYGG MkWrbA IVGSPVYYGGVSAQLKALIDRTRPLR-INWELKDKVGGAIAIGGARNGGQEHTLRDIQNFFMIHAMIVVGDSDPTAHFGG CaWrbA VFGSPIYYYKITGQSKIWFDRTFPMTGNDYKPKYPGKKLITVFAQGNPDPKIGAEGVKFANDMLEELGWKLEDSINYCG- ....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....| 90 100 110 120 130 140 150 160 CbFldx ILG------CSAMGDEVLEESEFEPFIEEISTKISGKKVALFGSYGWGDGKWMRDFEERMNGYGCVVVETPLIVQNEP-

AfWrbA ----RAGSYYGAASTG------VPKEDDLQVAKMLGKRVAEVAEKLC------EcWrbA VSQVRGGTPYGATTIAGGDGSRQPSQEELSIARYQGEYVAGLAVKLNG---- 43/63 MtIsf ------GITWNP------AEEDTVGMQTVSETAKKLCDVLELIQKNRDK 29/49 AfIsf ------FPVGVIQGDAGWRSVKKDEIAINSAKALAKRIVEVAEATKNLRES 31/50 MjIsf ------TGVGKAPGD-----CKNDDIGLETARNLGKKVAEVVKLIKK---- 30/47 MkWrbA ------AGVGLEPGD-----VEEDETGIETARNTGRRVGEVVKLIKG---- 28/45 CaWrbA ------TSHDPDLAMFDELSLRAFKDGENLVG---- 30/45 ....|....|....|....|....|....|....|....|....|....|.. 170 180 190 200 210 CbFldx ------DEAEQDCIEFGKKIANI------ns/ns

Figure 3-3. Sequence alignment with WrbA sequences, Isf sequences, and misannotated WrbA sequences that have a compact cysteine motif. Numerical values at the C-terminus indicate % identity/ % similarity to EcWrbA. (ns) indicates “not significant.” (#) indicates completely conserved residues. () indicates the region of the flavodoxin signature motif. (▲) indicates compact cysteine motif. Organisms and ORF designations from the corresponding genomic sequence: Archaeoglobus fulgidus AfIsf (gi:11499480) and AfWrbA (gi:11497955), Clostridium acetotrophicum CaWrbA (gi:15896732), Escherichia coli EcWrbA (gi:148264), Clostridium beijerinckii CbFldx (gi:1941945), Methanosarcina jannaschii MjIsf (gi:15669271), Methanopyrus kandleri MkWrbA (gi:20094374), Methanosarcina thermophila MtIsf (gi:2246438).

Biochemical Characterization of EcWrbA and AfWrbA

EcWrbA is the only prokaryotic WrbA biochemically characterized; thus, AfWrbA from the Archaea domain was heterologously produced in E. coli and characterized for comparison to EcWrbA from the Bacteria domain. EcWrbA has 43% identity and 62% similarity to AfWrbA, encoded by locus AF0343 in A. fulgidus. EcWrbA and AfWrbA were each over-expressed in E. coli and purified to homogeneity as determined by SDS-PAGE.

The identity of each protein was confirmed by N-terminal sequencing of the first five residues: AKVLV for EcWrbA and ARILV for AfWrbA. SDS-PAGE revealed that the subunit molecular mass for AfWrbA was 22.1 kDa and that for EcWrbA was 21.2 kDa. Size- exclusion chromatography showed that AfWrbA had a native molecular mass of 52 kDa, suggesting that AfWrbA is able to form a dimer as previously reported for EcWrbA [2].

Dynamic light scattering analyses support that both EcWrbA and AfWrbA participate in tetramerization. Purified EcWrbA had a hydrodynamic radius of 3.9 ± 0.12 nm, corresponding to a molecular mass of 79 ± 6.7 kDa, while purified AfWrbA had a hydrodynamic radius for 4.1 ± 0.12 nm, corresponding to a molecular mass of 89 ± 8.4 kDa.

Each sample was found to be monodisperse with EcWrbA or AfWrbA representing over

99.5% of the mass (Figure 3-4 and Table 3-1).

% Intensity

Radius (nm)

Figure 3-4. Example of raw data for Dynamic Light Scattering (DLS) Analysis of WrbA Proteins. NaCl.

Radius (nm) % polydispersity apparent kDa % mass Run 1 3.8 21.2 75 100 Run 2 3.8 21.1 76 100 Run 3 4 20.7 87 100

Table 3-1. DLS data from initial analyses of WrbA. Each reported data set was an average of 10 readings and was collected from a sample of 75 M

WrbA with 400 M (Run 1), 800 M (Run 2), or 1600 M (Run 3) NaCl.

During purification, EcWrbA retained FMN during elution from the first Q-sepharose column developed at pH 6.6, as indicated by the yellow color of the fractions containing

EcWrbA. However, EcWrbA that eluted from the second Q-sepharose column developed at pH 7.8 did not contain FMN, indicating that binding of FMN in EcWrbA is pH dependent.

After reconstitution with FMN, the ratio of 1.04 ± 0.03 FMN/monomer was obtained suggesting 1 FMN per monomer, consistent with a previous report [2]. The reconstituted protein exhibited a spectrum characteristic of oxidized flavoproteins (Figure 3-5). Other than an initial heat denaturation step, AfWrbA was purified as EcWrbA, although AfWrbA retained FMN through all steps of the purification, with a ratio of 0.87 ± 0.05 FMN/monomer upon elution from the second Q-sepharose column developed at pH 7.8. After reconstitution, a ratio of 1.10 ± 0.04 FMN/monomer was obtained suggesting 1 FMN per monomer, and the reconstituted protein exhibited a spectrum characteristic of oxidized flavoproteins.

Differences in the UV-visible spectra at the peak centered on 450nm suggest the FMN environment is somewhat different in EcWrbA than in AfWrbA (Figure 3-5). The ratio of absorbance for EcWrbA at 274 nm / 450 nm was 4.7 with an extinction coefficient of 450 =

11.6 mM-1cm-1. The ratio of absorbance for AfWrbA at 274 nm / 457 nm was 3.7 with an

-1 -1 extinction coefficient of 457 = 14.0 mM cm .

Hydrophobicity plots were produced using TMpred (available at ExPASy’s

Proteomic Tools) to clarify whether WrbA is localized at the membrane. The plots indicated two hydrophobic regions. The first of these overlaps with the flavodoxin signature motif, suggesting this region binds the flavin and contributes to stabilization of the  twisted open- sheet fold typical of flavodoxin-like proteins. The second hydrophobic span, predicted to be a transmembrane region, overlaps with the unique additional  unit (Figure 3-1). Gorman and

Shapiro found that this unique hydrophobic region significantly contributes to tetramerization

86 and is located at the core of the tetramer [4]. Thus, WrbA proteins do not appear to be integral membrane proteins, although it is unclear whether WrbA proteins are membrane- associated.

Since sequence comparisons indicated EcWrbA and AfWrbA are related to fungi and plant NQOs, this activity was determined for both EcWrbA and AfWrbA. As shown in Table

3-2, both EcWrbA and AfWrbA were able to couple electron transfer from NADH to several quinones. There was significantly less activity with Cyt c (< 1%), and no activity with benzyl viologen, methyl viologen (paraquat), FMN, FAD, F420, 2-hydroxyphenazine,

0.9

0.45 Absorbance

0 300 375 450 525 Wavelength (nm)

Figure 3-5. UV-visible spectra of reconstituted, oxidized EcWrbA and

AfWrbA. Inset: Comparison of WrbA-bound FMN versus free-FMN.

Symbols: AfWrbA (▬), EcWrbA (▬), free-FMN (―).

Electron Acceptor EcWrbA AfWrbA U mg-1 U mg-1 1,4-benzoquinone 990 ± 30 25000 ± 850

1,4-naphthoquinone 560 ± 30 8900 ± 260 menadione 430 ± 10 1300 ± 10 2,3-dihydroxy-5-methyl-1,4-benzoquinone 930 ± 60 6700 ± 140 potassium ferricyanide 50 ± 1.6 1400 ± 80 dichloroindolphenol 160 ± 3.0 1400 ± 70

U = mol NADH oxidized min-1

Table 3-2. Specific activities for EcWrbA and AfWrbA

conducted at 37°C and 65°C, respectively. NADH was present

at 800 M. Electron acceptor was present at 400 M.

EcWrbA AfWrbA k app/K app (M -1s-1) app -1 app k app/K app (M -1s-1) app -1 app cat m k cat (s ) K m (M) cat m k cat (s ) K m (M) NADHa 6.4 x 105 ± 0.20 x 105 8.9 ± 0.064 14 ± 0.43 6.3 x 106 ± 0.59 x 106 120 ± 2.9 19 ± 1.7 a 4 4 6 6 NADPH 3.5 x 10 ± 0.26 x 10 6.0 ± 0.15 170 ± 12 5.5 x 10 ± 0.33 x 10 170 ± 2.4 31 ± 1.8

1,4-benzoquinoneb 6.4 x 107 ± 0.14 x 107 370 ± 1.7 5.8 ± 0.12 8.6 x 107 ± 0.090 x 107 3200 ± 130 37 ± 3.6

a K3Fe(CN)6 was the electron acceptor bNADH was the electron donor

Table 3-3. Apparent kinetic parameters for EcWrbA and AfWrbA conducted at 37°C and 65°C, respectively. Saturating substrate was present at 200 M. A fit of raw data can be seen in Figure 3-6.

-1 4

2 micromoles min micromoles

0 0 500 1000 1500 [NADPH] (M)

Figure 3-6. Example Michaelis-Menton plot of raw data for kinetic parameters of E. coli WrbA with NADPH and 200 M app K3Fe(CN)6. Using GraphPad Prism to fit the data: Vmax = 6.0 app ± 0.15M, Km = 170 ± 11.7M. All kinetic parameters were determined with a fit of triplicate data.

2-hydroxynaphthoquinone, or 9,10-anthroquinone-2,6-disulfonate (data not shown). The apparent kinetic parameters determined for EcWrbA or AfWrbA catalyzing electron transfer from NAD(P)H to 1,4-benzoquinone or to K3Fe(CN)6 (Table 3-3) support that the activities are physiological. Both EcWrbA and AfWrbA preferred NADH over NADPH, which suggests the enzyme does not play a role in biosynthesis since NADPH is the preferred electron donor in biosynthetic pathways of prokaryotes.

The pH dependence of electron transfer from NADH to menadione was measured for both EcWrbA and AfWrbA (Figure 3-7). Apparent initial velocities showed that EcWrbA retained nearly 95% activity over the range of pH 6.0 to 8.0 and AfWrbA retained at least

95% activity from pH 5.0 to 8.5. The temperature dependence of electron transfer from

NADH to menadione was measured for both EcWrbA and AfWrbA. The apparent initial velocities reflect the optimal growth temperature for the respective species. In the linear range of highest activity, Arrhenius plots of the apparent initial velocities show that the Ea for

AfWrbA is 3.63 kcal/mol and that for EcWrbA is 1.88 kcal/mol (Figure 3-8). The Arrhenius plots also suggest that the kinetic system undergoes a temperature-dependent shift that increases the activation energy of the rate-limiting steps when the temperature dips below

30ºC.

Under aerobic conditions, auto-oxidation took about 20 minutes for 10 M reduced

EcWrbA or AfWrbA (not shown). Despite this low activity with O2, all activity assays were conducted anaerobically to prohibit electron acceptors from interacting with O2 to produce

ROS, which might affect the rates.

) )

1800 600 ○ ▲

) ) 1

500 - 1 - 1400 mg mg 1 1 - - 400

1000 mol min mol mol min mol 300   ( (

600 200 Specific Activity for EcWrbA ( EcWrbA for Activity Specific Specific Activity for AfWrbA ( AfWrbA for Activity Specific 3.5 5 6.5 8 9.5 pH

Figure 3-7. pH dependence of apparent initial velocities. Activities were obtained at 37ºC with EcWrbA and at 65ºC with AfWrbA.

Symbols: AfWrbA (▲), EcWrbA (○).

3.4 y = -3.63x + 5.53 ) 1 - 3 ) mg y = -1.88x + 3.96 1 app - 0 2.6 L og (v

mol mi n 2.2  (

1.8 0.55 0.6 0.65 0.7 0.75 0.8

1/(2.3*T (K)*1.98E-3 (kcal K-1 mol-1))

Figure 3-8. Arrhenius plot of apparent initial velocities. Activities were obtained at pH 7.2. Symbols: AfWrbA (▲), EcWrbA (○). Fitted range was 30° to 80°C for AfWrbA and 30° to 40°C for EcWrbA.

Phenotypic Screening at Biolog, Inc.

A wrbA- knockout was constructed from E. coli K-12 MG1655, and the construct was confirmed with external and internal primer analyses as before (Figure 3-9). The knockout and wild-type (WT) strains were submitted to Biolog, Inc. for phenotypic analyses. Growth of the wrbA- strain was compared to growth of the WT strain at 37ºC in Luria Broth with nearly 2000 phenotypes tested. (Phenotypes tested are available with Biolog, Inc. at http://www.biolog.com/phenoMicro.html.) No phenotypes were gained; however, N- trichloromethyl-mercapto-4-cyclohexene-1,2-dicarboximide and 8-hydroxyquinoline significantly inhibited growth of the wrbA- knockout relative to WT, which is consistent with a role for WrbA in protecting against environmental stressors through its quinone reductase activity.

WT KO WT KO ABC ABC ABC ABC

2.0 Kbp 2.0 Kbp 1.5 Kbp 1.5 Kbp

1.0 Kbp 2.0 Kbp

500 bp 500 bp

Figure 3-9. Construct analysis of E. coli WT and E. coli wrbA- kan+ by internal and external primers. (WT) amplifications for wild-type E. coli. (KO) amplifications for E. coli wrbA-kan+. (A) amplification using upstream and internal kan primer – presence of kan results in a product near 550 bp. (B) amplification using downstream and internal kan primer – presence of kan results in a product near 1000 bp. (C) amplification using upstream and downstream. Presence of wrbA results in a product near 700 bp, while presence of kan results in a product near 1.55 Kbp. In wrbA-kan- constructs, the external primers results in a product near 200 bp.

3.5 Discussion

EcWrbA was previously proposed to play a role as an accessory element in blocking

TrpR-specific transcriptional processes [1]; however, this role has been questioned [2] and the function is unknown. Further, EcWrbA from E. coli was the only biochemically characterized member of the WrbA family and no enzyme activity has been reported. Here, we demonstrate that EcWrbA has NAD(P)H-dependent redox activity and reduces quinones.

We also report the characterization of AfWrbA, a homologue from a hyperthermophilic archaeon, and that show it has the same redox activity. Phylogenetic analyses indicate that several characterized fungi and plant NQOs belong to the WrbA family, which suggests this enzyme activity is a defining characteristic common to the WrbA family. Further, the activities and kinetic constants of EcWrbA and AfWrbA reported here correlate well with those reported for NQO homologues in Gloeophyllum trabeum [18] and Phanerochaete chrysosporium [16].

At least six other protein families with NQO activity are known, three of which have been designated Type I through Type III. Other than activity, the WrbA family bears no sequence similarity has unique structural characteristics not present in any of the six other protein families; in addition to the unique  unit, WrbA has 1 FMN per monomer and participates in a dimer-tetramer equilibrium. Thus, we adopt the nomenclature initiated by

Type I through Type III and designate the WrbA family Type IV. Types I-III are more commonly referred to as NADH dehydrogenases. Perhaps the best characterized Type I is

NADH-dependent Complex I, the H+/Na+-transporting integral membrane complex comprised of up to 46 subunits that participates in electron transfer in prokaryotes from the

Bacteria domain during respiration [28, 29]. Type II NADH dehydrogenase is usually a single polypeptide and does not generate a chemical gradient [28]. The Na+-transporting

(Type III) NAD(P)H dehydrogenase is comprised of six or seven subunits and is responsible for generating a Na+ potential that can be used for flagella movement or for substrate transfer

[28, 30]. There is no standard nomenclature for the remaining three protein families with

NQO activity. The first of these three families is eukaryotic DT-Diaphorase/NQO1, a dimeric

FAD-containing protein that reduces anti-tumor quinones with NAD(P)H [31, 32]. The second family is FAD-containing NQO2, which has sequence similarity to NQO1, uses dihydronicotinamide riboside as the electron donor and is resistant to typical inhibitors of

DT-diaphorases [31]. The third family is -crystalline, and is a major protein present in some mammalian ocular lenses which participates in the one-electron reduction of quinones

[33].

Possible physiological roles for Type IV NQOs

Several NQOs from the Fungi and Viridiplantae Kingdoms identified here as belonging to the WrbA family have been suggested to function in reducing quinones to the hydroquinone state to prevent interaction of the semiquinone with O2 and production of superoxide. Quinoid compounds are essential to organisms from all domains of life.

Quinones are generally tethered to the plasma membrane or freely traverse the lipid bilayer and function in electron transport chains, in cell-signaling, and in protection from environmental oxidizers through direct reduction [34, 35]. Although they usually serve as electron mediators by cycling between the oxidized (hydroquinone) state and the two- electron reduced (hydroquinol) state, quinones can also participate in deleterious redox- cycling through direct interactions with single electron acceptors such as O2. This one- electron redox-cycling leads to the accumulation of reactive oxygen species (ROS) such as superoxide, hydrogen peroxide, and the hydroxyl radical [15, 16, 18, 20, 21, 36-45].

Intracellular production of ROS can result in the peroxidation of lipids, the destruction of cofactors, and the hydroxylation of proteins and nucleic acids [36, 43, 46-51]. Several reports suggest that in order to guard against the production of ROS from one-electron redox- cycling, a wide diversity of cells have evolved NQOs to maintain quinones in the fully- reduced state [15, 16, 18, 21, 31, 36, 44, 52-56].

A role for the Type IV NQO family in alleviating and recovering from oxidative stress through quinone reduction fits with the available data, and both EcWrbA and AfWrbA derive from organisms that synthesize quinones; E. coli synthesizes ubiquinone-8 and menaquinone-8 while A. fulgidus produces a menaquinone with a fully saturated heptaprenyl side chain [57]. However, NQO activity could also suggest a role in cell-signaling as has been proposed for a Type IV NQO found in yeast (SpObr1 in Figure 3-2) [58-61]. Numerous studies show that the redox status of the quinone pool is tightly coupled to cell-signaling and cell growth of diverse cell types [62-65]. More specifically, the ArcA/ArcB two-component system of E. coli, which regulates cellular transcription in response to external electron acceptors, is one method by which E. coli controls expression in response to redox changes

[66, 67]. As previously reported, E. coli wrbA is repressed by phosphorylated ArcA, which is phosphorylated when the quinone pool is reduced [13]. Further, WrbA is upregulated when

E. coli enters the stationary phase and in the presence of a variety of stressors such as acids,

s salts, H2O2, and diauxie, under the control of RpoS (the stress response sigma factor,  or

38) [5-12]. Another study demonstrated that under anaerobic conditions, E. coli wrbA is also repressed by FNR (fumarate and nitrate reductase) regulatory protein, suggesting EcWrbA does not function during anaerobiosis [14]. Cumulatively, these expression studies suggest that EcWrbA functions in response to environmental stress when various electron-transfer chains are affected or when the environment is highly oxidizing. Although speculative,

EcWrbA could function in cell signaling by reducing the quinone pool, which would subsequently signal redox sensors such as the ArcA modulon. Chang and co-workers speculate a similar mechanism for a -crystalline type NQO that is expressed under stress and reduces the quinone pool, ultimately phosphorylating ArcA and inducing a shift from respiratory metabolism to fermentative metabolism [5].

3.6 Acknowledgements

This work was supported by DOE grant N.O. DE-FG02-95ER20198, by a grant from the NASA Astrobiology Institute, and in part by a grant from the Integrative Biosciences

Program at the Pennsylvania State University. We would like to thank Sarah Ades and Ming

Tien of the Pennsylvania State University for valuable direction, discussion, and support.

100

3.7 References

1. Yang, W., L. Ni, and R. Somerville (1993). A Stationary-Phase Protein of Escherichia coli that Affects the Mode of Association Between the trp Repressor Protein and Operator-Bearing DNA. PNAS. 90(12): p. 5796-20966.

2. Grandori, R., et al. (1998). Biochemical Characterization of WrbA, Founding Member of a New Family of Multimeric Flavodoxin-like Proteins. J. Biol. Chem. 273(33): p. 20960-20966.

3. Grandori, R. and J. Carey (1994). Six new candidate members of the / twisted open-sheet family detected by sequence similarity to flavodoxin. Protein Sci. 3(12): p. 2185-20966.

4. Gorman, J. and L. Shapiro (2005). Crystal structures of the tryptophan repressor binding protein WrbA and complexes with flavin mononucleotide. Protein Sci. 14(12): p. 3004-12.

5. Chang, D.E., D.J. Smalley, and T. Conway (2002). Gene expression profiling of Escherichia coli growth transitions: an expanded stringent response model. Mol Microbiol. 45(2): p. 289-306.

6. Cheung, K.J., et al. (2003). A Microarray-Based Antibiotic Screen Identifies a Regulatory Role for Supercoiling in the Osmotic Stress Response of Escherichia coli. Genome Res. 13(2): p. 206-215.

7. Ishihama, A. (2000). Functional Modulation of Escherichia coli RNA Polymerase. Annu. Rev. Micrbiol. 54(1): p. 499-518.

8. Kirkpatrick, C., et al. (2001). Acetate and Formate Stress: Opposite Responses in the Proteome of Escherichia coli. J. Bacteriol. 183(21): p. 6466-6477.

9. Phadtare, S., I. Kato, and M. Inouye (2002). DNA Microarray Analysis of the Expression Profile of Escherichia coli in Response to Treatment with 4,5-Dihydroxy- 2-Cyclopenten-1-One. J. Bacteriol. 184(23): p. 6725-6729.

10. Pomposiello, P.J., M.H. Bennik, and B. Demple (2001). Genome-wide transcriptional profiling of the Escherichia coli responses to superoxide stress and sodium salicylate. J Bacteriol. 183(13): p. 3890-902.

11. Tao, H., et al. (1999). Functional Genomics: Expression Analysis of Escherichia coli Growing on Minimal and Rich Media. J. Bacteriol. 181(20): p. 6425-411.

12. Tucker, D.L., N. Tucker, and T. Conway (2002). Gene Expression Profiling of the pH Response in Escherichia coli. J. Bacteriol. 184(23): p. 6551-1100.

101

13. Liu, X. and P. De Wulf (2004). Probing the ArcA-P modulon of Escherichia coli by whole genome transcriptional analysis and sequence recognition profiling. J Biol Chem. 279(13): p. 12588-97.

14. Kang, Y., et al. (2005). Genome-wide expression analysis indicates that FNR of Escherichia coli K-12 regulates a large number of genes of unknown function. J Bacteriol. 187(3): p. 1135-60.

15. Brock, B.J. (1995). Purification and characterization of a 1,4-benzoquinone reductase from the basidiomycete Phanerochaete chrysosporium. Appl Environ Microbiol. 61(8): p. 3076-3081.

16. Brock, B.J. and M.H. Gold (1996). 1,4-Benzoquinone reductase from basidiomycete Phanerochaete chrysosporium: spectral and kinetic analysis. Arch Biochem Biophys. 331(1): p. 31-40.

17. Cohen, R., M.R. Suzuki, and K.E. Hammel (2004). Differential stress-induced regulation of two quinone reductases in the brown rot basidiomycete Gloeophyllum trabeum. Appl Environ Microbiol. 70(1): p. 324-31.

18. Jensen Jr, K.A., Jr., et al. (2002). An NADH:quinone oxidoreductase active during biodegradation by the brown-rot basidiomycete Gloeophyllum trabeum. Appl Environ Microbiol. 68(6): p. 2699-703.

19. Laskowski, M.J., et al. (2002). FQR1, a Novel Primary Auxin-Response Gene, Encodes a Flavin Mononucleotide-Binding Quinone Reductase. Plant Physiology. 128(2): p. 578-20966.

20. Matvienko, M., et al. (2001). Quinone oxidoreductase message levels are differentially regulated in parasitic and non-parasitic plants exposed to allelopathic quinones. Plant J. 25(4): p. 375-87.

21. Wrobel, R.L. (2002). Heterologous expression and biochemical characterization of an NAD(P)H: quinone oxitoreductase from the hemiparasitic plant Triphysaria versicolor. Plant Physiol Biochem. 40: p. 265-272.

22. Datsenko, K.A. and B.L. Wanner (2000). One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A. 97(12): p. 6640-5.

23. Zhao, T., F. Cruz, and J.G. Ferry (2001). Iron-sulfur flavoprotein (Isf) from Methanosarcina thermophila is the prototype of a widely distributed family. J Bacteriol. 183(21): p. 6225-33.

24. Andrade, S.L., et al. (2005). Structures of the iron-sulfur flavoproteins from Methanosarcina thermophila and Archaeoglobus fulgidus. J Bacteriol. 187(11): p. 3848-54.

102

25. Leartsakulpanich, U., M.L. Antonkine, and J.G. Ferry (2000). Site-specific mutational analysis of a novel cysteine motif proposed to ligate the 4Fe-4S cluster in the iron- sulfur flavoprotein of the thermophilic methanoarchaeon Methanosarcina thermophila. J Bacteriol. 182(19): p. 5309-16.

26. Becker, D.F., et al. (1998). Electrochemical and spectroscopic properties of the iron- sulfur flavoprotein from Methanosarcina thermophila. J Biol Chem. 273(41): p. 26462-9.

27. Latimer, M.T., M.H. Painter, and J.G. Ferry (1996). Characterization of an iron-sulfur flavoprotein from Methanosarcina thermophila. J Biol Chem. 271(39): p. 24023-8.

28. Melo, A.M., T.M. Bandeiras, and M. Teixeira (2004). New insights into type II NAD(P)H:quinone oxidoreductases. Microbiol Mol Biol Rev. 68(4): p. 603-16.

29. Yagi, T. and A. Matsuno-Yagi (2003). The proton-translocating NADH-quinone oxidoreductase in the respiratory chain: the secret unlocked. Biochemistry. 42(8): p. 2266-74.

30. Bogachev, A.V. and M.I. Verkhovsky (2005). Na(+)-Translocating NADH:quinone oxidoreductase: progress achieved and prospects of investigations. Biochemistry (Mosc). 70(2): p. 143-9.

31. Chen, S., K. Wu, and R. Knox (2000). Structure-function studies of DT-diaphorase (NQO1) and NRH: quinone oxidoreductase (NQO2). Free Radic Biol Med. 29(3-4): p. 276-84.

32. Siegel, D., et al. (2004). NAD(P)H:quinone oxidoreductase 1: role as a superoxide scavenger. Mol Pharmacol. 65(5): p. 1238-47.

33. Rao, P.V., C.M. Krishna, and J.S. Zigler, Jr. (1992). Identification and characterization of the enzymatic activity of -crystallin from guinea pig lens. A novel NADPH:quinone oxidoreductase. J Biol Chem. 267(1): p. 96-102.

34. Berry, S. (2002). The chemical basis of membrane bioenergetics. J Mol Evol. 54(5): p. 595-613.

35. Meganathan, R. (2001). Ubiquinone biosynthesis in microorganisms. FEMS Microbiol Lett. 203(2): p. 131-9.

36. Adams, M.A. and Z. Jia (2005). Structural and biochemical evidence for an enzymatic quinone redox cycle in Escherichia coli: identification of a novel quinol monooxygenase. J Biol Chem. 280(9): p. 8358-63.

37. Arroyo, A., et al. (2000). Interactions between ascorbyl free radical and coenzyme Q at the plasma membrane. J Bioenerg Biomembr. 32(2): p. 199-210.

103

38. Boersma, M.G., et al. (1994). NADPH-cytochrome reductase catalysed redox cycling of 1,4-benzoquinone; hampered at physiological conditions, initiated at increased pH values. Biochem Pharmacol. 47(11): p. 1949-55.

39. Bond, D.R. and D.R. Lovley (2002). Reduction of Fe(III) oxide by methanogens in the presence and absence of extracellular quinones. Environ Microbiol. 4(2): p. 115- 24.

40. Cervantes, F., et al. (2000). Quinones as terminal electron acceptors for anaerobic microbial oxidation of phenolic compounds. Biodegradation. 11(5): p. 313-21.

41. Cervantes, F.J., et al. (2002). Reduction of humic substances by halorespiring, sulphate-reducing and methanogenic microorganisms. Environ Microbiol. 4(1): p. 51- 7.

42. Finneran, K.T., et al. (2002). Desulfitobacterium metallireducens sp. nov., an anaerobic bacterium that couples growth to the reduction of metals and humic acids as well as chlorinated compounds. Int J Syst Evol Microbiol. 52(Pt 6): p. 1929-35.

.- 43. Fridovich, I. (1997). Superoxide anion radical (O2 ), superoxide dismutases, and related matters. J Biol Chem. 272(30): p. 18515-7.

44. Jaiswal, A.K. (2000). Regulation of genes encoding NAD(P)H: quinone oxidoreductases. Free Radical Biology and Medicine. 29(3/4): p. 254-262.

45. Roginsky, V., et al. (1998). Kinetics of redox interaction between substituted 1,4- benzoquinones and ascorbate under aerobic conditions: critical phenomena. Free Radic Res. 29(2): p. 115-25.

46. Chen, S.X. and P. Schopfer (1999). Hydroxyl-radical production in physiological reactions. A novel function of peroxidase. Eur J Biochem. 260(3): p. 726-35.

47. Fang, Y.Z., S. Yang, and G. Wu (2002). Free radicals, antioxidants, and nutrition. Nutrition. 18(10): p. 872-9.

48. Fridovich, I. (1998). Oxygen toxicity: a radical explanation. J Exp Biol. 201 ( Pt 8): p. 1203-9.

49. Lim, P., et al. (2004). Peroxyl radical mediated oxidative DNA base damage: implications for lipid peroxidation induced mutagenesis. Biochemistry. 43(49): p. 15339-48.

50. Schopfer, P., et al. (2002). Evidence that hydroxyl radicals mediate auxin-induced extension growth. Planta. 214(6): p. 821-8.

51. Vranova, E., D. Inze, and F. Van Breusegem (2002). Signal transduction during oxidative stress. J Exp Bot. 53(372): p. 1227-36.

104

52. Hayashi, M., et al. (1990). Characterization of FMN-dependent NADH-quinone reductase induced by menadione in Escherichia coli. Biochim Biophys Acta. 1035(2): p. 230-6.

53. Long, D.J., 2nd and A.K. Jaiswal (2000). NRH:quinone oxidoreductase2 (NQO2). Chem Biol Interact. 129(1-2): p. 99-112.

54. Morre, D.J. (2004). Quinone oxidoreductases of the plasma membrane. Methods Enzymol. 378: p. 179-99.

55. Ross, D., et al. (2000). NAD(P)H:quinone oxidoreductase 1 (NQO1): chemoprotection, bioactivation, gene regulation and genetic polymorphisms. Chem Biol Interact. 129(1-2): p. 77-97.

56. Sparla, F., et al. (1999). Cloning and heterologous expression of NAD(P)H:quinone reductase of Arabidopsis thaliana, a functional homologue of animal DT-diaphorase. FEBS Lett. 463(3): p. 382-6.

57. Tindall, B., K. Stetter, and M. Collins (1989). A novel, fully saturated menaquinone from the thermophilic, sulphate-reducing archaebacterium Archaeoglobus fulgidus. J. Gen. Microbiol. 135: p. 693–696.

58. Daher, B.S., et al. (2005). The highly expressed yeast gene pby20 from Paracoccidioides brasiliensis encodes a flavodoxin-like protein. Fungal Genet Biol. 42(5): p. 434-43.

59. Naresh, A., S. Saini, and J. Singh (2003). Identification of Uhp1, a ubiquitinated histone-like protein, as a target/mediator of Rhp6 in mating-type silencing in fission yeast. J Biol Chem. 278(11): p. 9185-94.

60. Shimanuki, M., et al. (1995). A novel essential fission yeast gene pad1+ positively regulates pap1(+)-dependent transcription and is implicated in the maintenance of chromosome structure. J. Cell Sci. 108(2): p. 569-579.

61. Turi, T.G., P. Webster, and J.K. Rose (1994). Brefeldin A sensitivity and resistance in Schizosaccharomyces pombe. Isolation of multiple genes conferring resistance. J Biol Chem. 269(39): p. 24229-36.

62. Bauer, C.E., S. Elsen, and T.H. Bird (1999). Mechanisms for redox control of gene expression. Annu Rev Microbiol. 53: p. 495-523.

63. Grabbe, R. and R.A. Schmitz (2003). Oxygen control of nif gene expression in Klebsiella pneumoniae depends on NifL reduction at the cytoplasmic membrane by electrons derived from the reduced quinone pool. Eur J Biochem. 270(7): p. 1555-66.

64. Kowaltowski, A.J., R.F. Castilho, and A.E. Vercesi (2001). Mitochondrial permeability transition and oxidative stress. FEBS Lett. 495(1-2): p. 12-5.

105

65. Oh, J.I. and S. Kaplan (2000). Redox signaling: globalization of gene expression. Embo J. 19(16): p. 4237-47.

66. Georgellis, D., O. Kwon, and E.C. Lin (2001). Quinones as the redox signal for the arc two-component system of bacteria. Science. 292(5525): p. 2314-6.

67. Malpica, R., et al. (2004). Identification of a quinone-sensitive redox switch in the ArcB sensor kinase. Proc Natl Acad Sci U S A. 101(36): p. 13318-23.

106

CHAPTER FOUR: REVIEW: FLAVODOXIN-LIKE NAD(P)H:QUINONE REDUCTASES FUNCTIONING IN DETOXIFICATION AND REDOX-LINKED PROTEOLYSIS 4.1 Introduction 4.1.1 NQO1 Proteins (Class I) 4.1.2 FMN Reductase Proteins (Class II) 4.1.3 Azoreductase Proteins (Class III) 4.1.4 WrbA Proteins (Class IV) 4.2 Bioinformatics Analysis of Type IV NQOs (Classes I-IV) 4.3 Discussion 4.3.1 Xenobiotic Detoxification 4.3.2 Redox-Associated Proteasomal Degradation 4.4 Future Directions 4.5 References

Chapter 4

Review: Flavodoxin-like NAD(P)H:Quinone Reductases Functioning in Detoxification and Redox-linked Proteolysis

Eric V. Patridge and James G. Ferry*

Department of Biochemistry and Molecular Biology, Eberly College of Science, The Pennsylvania State University, 205 South Frear Laboratory, University Park, PA 16802-4500.

*Corresponding author. Tel.: +1 (814) 863-5721; Fax: +1(814) 863-6217; E-mail: [email protected]

4. 1 Introduction

First discovered in 1958, DT-diaphorase is an enzyme that has activity with electrophiles and is strongly induced in cancer cells [1-7]. Now called NAD(P)H:quinone oxidoreductase (NQO1), it is well established that the enzyme has activity with various mutagenic xenobiotics and anti-cancer therapies [5, 6, 8-12]. In related activity, through reduction of menaquinone (Vitamin K2), NQO1 contributes to blood coagulation via - carboxylation of glutamate residues [9, 13, 14]. Reviews frequently highlight that the enzyme preferentially functions in protective, two-electron reductions; this curtails the production of damaging reactive oxygen species (ROS) that result from one-electron reductions [5-7, 9, 10, 14-25]. It is not surprising that NQO1 garners significant attention; its protective activity, its accumulation in cancer cells, and its interactions with anti-cancer compounds make the protein an ideal target for chemotherapy [5, 6, 8, 9].

Mammalian NQO1 was the first NAD(P)H:quinone oxidoreductase (NQO) implicated in protection from environmental stressors, but a growing number of proteins show similar NQO activity [17, 26-35]. In the last few decades, proteins from all domains of life were found to have NAD(P)H-dependent activity with electrophilic compounds. Among these are tryptophan repressor binding protein A (WrbA), azoreductase (AzoR), arsenic reductase (ArsH), chromate reductase (ChrR), flavin reductase (FMNred), and modulator of drug activity B (MdaB). There is little sequence similarity to establish a firm relationship among these proteins, but all of them (including NQO1) share a common flavodoxin-like structure. Numerous publications utilize this structural similarity to establish a relationship to these proteins, and mounting evidence suggests they could be more closely related than suggested by sequence alignments [17, 31, 32, 34, 36-41].

108

Here we have taken a broad approach to highlight the widespread flavodoxin-like

NQOs and to delineate between them. An exhaustive evaluation of the literature offers significant evidence to suggest these proteins belong to a larger family that functions in cell survival during environmental stress. These NQOs are distinct from the NADH

Dehydrogenases (Types I-III) and from ζ-crystalline protein [16, 35]. Phylogenetic analysis shows that the identified flavodoxin-like NQOs (Type IV NQOs) fall into to four subclasses.

For discussion, we provide background for the proteins and refer to each subclass using a well-investigated, representative protein: NQO1 proteins (Class I), FMN reductase proteins

(Class II), azoreductase proteins (Class III), and WrbA proteins (Class IV). Additional subclasses may exist, but we are focusing on the established NAD(P)H:quinone oxidoreductases (also NRH:quinone oxidoreductases) in hopes of aiding research of these proteins.

109

4.1.1 NQO1 Proteins (Class I)

Considerable research has been performed with Class I NQOs. In eukaryotes, the model Class I protein (NQO1) is a phase II enzyme (as is the related, NRH-dependent

NQO2), and it is responsible for detoxifying and escorting xenobiotic compounds from an organism. Crystal structures are available for members of Class I, including several structures with antitumor xenobiotics bound in the active site [42-47]. In vitro studies show that wild type NQO1 couples electron transfer from NAD(P)H to various compounds. As seen in Table 4-1, in vivo studies indicate that the protein protects organisms from quinones and poly aromatic hydrocarbons which contribute to cellular damage and cancers or other physiological conditions.

Aside from the functional wild type NQO1*1 allele, there are two known variant alleles: NQO1*2 (C609T / Pro187Ser) and the less frequent NQO1*3 (C465T / Arg139Trp), which result in abrogated or reduced NQO1 activity, respectively. Notably, cancer relapse and related mortality are significantly increased in NQO1 *2/*2 homozygotes, which have no

NQO1 activity. This increased susceptibility is associated more closely with chemotherapy rather than radiation treatment. In different studies (and with varying results), NQO1 variant alleles or knockouts have been associated with cancers of the kidney, colon or rectum, breast, urinary system, skin, myeloid line of white blood cells, bone marrow, lung, and nose or pharynx [5, 6, 8, 11, 17, 19, 23, 48-52]. Fortunately, the increased expression of wild type

NQO1 in cancer cells makes it an opportune target for anti-cancer treatments. Such treatments involve quinoid-like compounds that can be activated by wild type NQO1 (Table

4-1). Upon activation these compounds cross-link DNA, which interrupts cellular processes and induces apoptosis. As expected, the reduced or null activity of NQO1 variant alleles is prohibitive of these treatments.

110

Implicated Xenobiotic Protein Compound Effect of Xenobiotic Compound References

In vivo studies affecting cell survival Affects transcription/translation via DNA adducts and DNA WrbA captan (Patridge 2006; Snyder 2006) cross-linking. 2+ 2+ (1) Affects DNA/RNA structure by chelating Mg and Mn . (Patridge 2006; Mills 1978; WrbA 8-hydroxyquinoline* (2) Increases DNA strand breaks by chelating metals. Leanderson 1996) MdaB tetracycline Inhibits translation by interacting with 30S ribosome. (Adams 2006; Epe 1984) Inhibits topo II by intercalating/cross-linking with DNA. (Adams 2006; Chatterjee 1995; MdaB adriamycin Participates in redox-cycling. Tewey 1984) MdaB bisnaphthalimide‡ Inhibits topo II by intercalating/cross-linking with DNA. (Chatterjee 1995; Nitiss 1998) MdaB etoposide Induces DNA cleavage by interacting with topo II. (Chatterjee 1995; Burden 1996) Inhibits topo II by intercalating/cross-linking with DNA. (Zwelling 1991; NQO1 anthracyclines* Participates in redox-cycling. Fagerholm 2008) (1) Inhibits growth in pancreatic cells by increasing superoxide NQO1 dicumarol production. (Lewis 2005; Anwar 2003) (2) Inhibits ability of NQO1 to protect proteins from proteolysis. aziridinylquinones Upon activation by NQO1: NQO1 mitosenes (Winski 2001; Ross 1990) Form DNA adducts, cross-link DNA, and increase cell death. reactive quinones Lot6p 1,4-naphthoquinone* Intercalates DNA and causes damage while redox-cycling. (Sollner 2007; Hakura 1994) Other in vivo studies NQO1 (1) Induces detoxifying Phase 2 enzymes such as NQO1. oltipraz (2) Inhibits P450 epoxidation of aflotoxin, which forms (Beglieter 2004; Lagouët 1995) P450 DNA adducts. (1) Induces detoxifying Phase 2 enzymes such as NQO1, NQO1 (Hsu 2005; Zenzes 1999; benzopyrene, PAHs protecting against related cancer. Pandey 2006) P450 (2) Epoxidation by P450 leads to DNA-adducts. NQO1 (1) Induce detoxifying Phase 2 enzymes such as NQO1. (Angeloni 2008; Strick 2000; (2) Inhibit topo II, as evidence by increased DNA topo II bioflavonoids Thangasamy 2007; endoreplication and translocations. Bandele 2007) --- (3) Sequester toxic metals, sequester ROS, and intercalate DNA. NQO1 (1) Induce detoxifying Phase 2 enzymes such as NQO1. (2) Inhibit topo II, as evidence by increased DNA (Mas 2003; Lovett 2001; topo II quinones, catechols* endoreplication and translocations. Keller 1983) --- (3) Intercalate DNA and cause damage while redox-cycling. NQO1 aristolochic acids Upon activation by NQO1: Increase cancers in humans. (Stiborova 2002) In vitro assays WrbA (1) Inhibitor of quinone reductase activity with WrbA. Lot6p (2) Inhibitor of quinone reductase activity with Lot6p. (Ito 2008; Brock 1996; dicumarol (3) Inhibitor of quinone reductase activity with NQO1. NQO1 Anwar 2003; Sollner 2007) (4) Inhibits ability of NQO1 to protect proteins from proteolysis. AzoR (5) Binds to the active site in azoreductase.

Table 4-1. Literature evidence for Type IV NQOs interacting with or conferring protection from xenobiotics.

111

Importantly, researchers in 2001 demonstrated a second role for NQO1 in cancer protection. In addition to its redox quinone reductase activity, NQO1 stabilizes the labile p53 protein and protects it from degradation by the 20S proteasome. The p53 protein has a half life of minutes, but plays a key role in cell-cycle regulation and can signal cell death, apoptosis, and cell senescence. (With such a key role in cell-cycle regulation, it is perhaps not surprising that nearly half of all cancers involve a mutation of p53.) By stabilizing and protecting p53, NQO1 encourages p53-dependent apoptosis [53-62].

Since the discovery of this association, NQO1 has also been shown to stabilize and protect several other proteins from proteasomal degradation, including: ornithine decarboxylase (ODC), heat shock protein 70 (Hsp70/p73/DnaK), and heat shock protein 40

(Hsp40/DnaJ) [57, 59, 61-67]. It was also demonstrated that NQO1 associates with the 20S proteasome itself and has been nominated as a gate-keeping protein of ubiquitin-independent,

20S proteasomal degradation. Interestingly, unlike its association with the 20S proteasome, the interactions of NQO1 with p53, ODC, DnaK, and DnaJ are NADH-dependent and can be inhibited by dicumarol, which is also a potent inhibitor of its quinone reductase activity [54,

57-59, 64-67]. Thus, mounting evidence suggests that NQO1 may function in ubiquitin- independent, redox-linked proteolysis, serving as gate-keeping control between the redox state of the cell and proteasomal degradation of enzymes key to protein folding, DNA processes, and cellular regulation.

112

4.1.2 FMN REDUCTASE Proteins (Class II)

Our phylogenetic analysis identified another subclass of NAD(P)H:quinone oxidoreductases which encompasses a broad number of proteins, including flavin reductases, nitroreductases, azoreductases, arsenic reductases, and chromate reductases. Flavodoxin-like

YhdA proteins [68] also cluster with Class II NQOs, but these are distinct from stomatin-like

YhdA [69]. One azoreductase from Bacillus sp. OY1-2 clusters with a Class II NQOs and has been suggested to be flavin-free, but this proposal is likely an oversight due to purification techniques, since flavin will elute during purification [31, 70-73].

As the collection of names suggests, Class II NQOs have demonstrated enzymatic activity with a wide range of electrophilic, toxic, or mutagenic compounds, including flavins, quinones, quinone-like compounds, dyes, and metal compounds [31-33, 71, 73-77]. Some of these proteins may prefer NADPH over NADH [68]. As with Class I, members of Class II are inhibited by dicumarol, and they have several structures which appear in the literature

[31, 41, 68, 74, 75]. In vivo investigations support a role in xenobiotic protection [27, 31,

78-81]. The structural topology and activities of Class II NQOs are similar to Class I proteins and are generally ascribed a role in protection from toxic environmental compounds.

Interestingly, a recent publication has demonstrated a second function for these flavoproteins; a Class II NQO from yeast (Lot6p/YLR011wp) protects proteins from degradation by the 20S proteasome in a redox-linked manner [66]. While the flavoprotein- proteasome complex is not redox-linked, its association with the Yap4 transcription factor

(yeast activator protein/Cin5p) is redox-dependent, which is remarkably similar to Class I

NQOs. The reduced flavoprotein dimer recruits Yap4 to the flavoprotein-proteasome complex and prohibits ubiquitin-independent degradation of the protein, concomitantly

113 localizing Yap4 to the cytosol. This mechanism implicates Class II NQOs in cellular regulation by redox-linked protein degradation/accumulation.

4.1.3 AZOREDUCTASE Proteins (Class III)

The Class III NQOs also have enzymatic activity with numerous compounds, but they are generally ascribed a role in reducing azo dyes, which are organic compounds containing azo groups (-N=N-), used in printing, food, and cosmetics [38, 82-85]. Large quantities of azo dyes can be toxic and mutagenic, and they are classically resistant to biodegradation processes

[37, 70, 86-91]. As with Classes I and II, several crystal structures of Class III NQOs are available, and one of these shows dicumarol bound in the active site [38, 39, 82, 92, 93].

Despite features that distinguish each NQO subclass, complications in nomenclature for azoreductases still persist. In general, any observed activity in reducing azo-compounds is identified as azoreductase activity [26, 87, 89-91, 94]. For example, since they are active with azo dyes, Class I and II NQOs are frequently referred to as azoreductases [33, 37, 71, 73, 95-

97]. However, the term also more specifically refers to NAD(P)H-dependent enzymes considered biochemically similar to Class I, II, and IV NQOs. Additionally, other flavin-free proteins, such as those purified from Pigmentiaphaga kullae K24 and Xenophilus azovorans

KF46F are reported to have “aerobic” azoreductase activity. These proteins are distinct from

Class III NQOs and appear similar to NmrA transcriptional regulators [72, 87, 88, 98, 99].

Further, some azoreductases were briefly misannotated as acyl carrier protein (ACP) phosphodiesterases, but they have since been shown to lack such activity [84, 100]. For clarity, the more formal azoreductases we describe comprise a unique subclass (Class III) of flavodoxin-like, FMN-containing proteins that retain NAD(P)H activity with various compounds such as azo dyes.

114

4.1.4 WRBA Proteins (Class IV)

The fourth subclass of multimeric, flavodoxin-like NQO proteins includes the tryptophan repressor binding protein (WrbA), first identified in 1993 when it purified with the tryptophan repressor protein (TrpR) [101]. Proteome and expression studies implicate

WrbA proteins in protection from environmental stressors including oxidative stress, acids, salts, and diauxie [102-110]. We previously demonstrated the E. coli protein has NQO activity [35]. As with the Classes I-III, Class IV NQOs couple electron transfer from

NAD(P)H to a variety of electrophiles, and the activity can be inhibited with dicumarol [35,

111-117]. In addition to activity characterizations, several studies report structures for Class

IV NQOs and have demonstrated that the loosely-bound FMN-cofactor encourages multimerization and imparts significant thermostability to the holo-tetramer [34, 35, 118-

120].

Similar to Classes I and II, evidence suggests that Class IV NQOs may associate with

DNA-binding proteins. Early characterizations of WrbA proteins suggested the protein binds

TrpR and stabilizes the TrpR-DNA complex but this role was later questioned [101, 121]. In addition to this disputed interaction, another Class IV NQO member from yeast

(Uhp1/obr1/p22/p25) binds histone H2B and associates with chromatin, contributing to gene silencing [122]. Thus, evidence suggests that these NQO proteins may be able to form protein-protein complexes with DNA-binding proteins, as has been demonstrated for Class I and Class II NQOs.

115

4.2 Bioinformatic Analyses of Type IV NQOs (Classes I-IV)

Our previous investigation demonstrated that WrbA proteins are distinct from flavodoxins and are widely present across all domains of life [35]. In reviewing the literature, we found numerous publications which highlight sequence conservation across

Type IV NQO subclasses, despite the general lack of significant similarity between the sequences [17, 27, 36, 40, 77, 82, 84]. In an effort to enhance discussion and investigation of these proteins, we present a phylogenetic analysis with a wide representation of Type IV

NQOs across all domains of life. To our knowledge, this is the broadest and most inclusive analysis of these flavodoxin-like proteins to appear in the literature.

Genes encoding Type IV NQOs are abundant across all domains of life. To investigate the relatedness of these proteins, we collected sequences of Type IV NQOs and related flavodoxin-like NAD(P)H:quinone oxidoreductases that appear in the literature. The sequences included in our phylogenetic analysis fall into four subclasses (Figure 4-1). The crystal structures beside the tree are published members of each clade and offer a qualitative comparison of tertiary and quaternary characteristics. There is no significant sequence similarity in protein-protein alignments of these structures, so the multiple alignment has been manually cured using secondary structure information (Figure 4-2).

116

ArsH

AzoR 2q62 2z9d

MdaB 2b3d s

1t0i FMN Reductase 0.2 Flavodoxins NQO1 1d4a WrbA 3b6i

Figure 4-1. Phylogram of Type IV NQOs, cured for representative sampling and for illustration. Crystal structures of the NQOs are located near their respective branches and demonstrate general topologies of NQOs. Flavodoxins were used as the reference.

117

10 20 30 40 50 60 70 80 90 3b6i --AK-VLVLYYS------MYG----HIETMARA-VAEGA--SKVD----GAEVVVKRVPETMP-PQLFE--KAGGKTQTA----PVAT 2z9d --SK-VLVLKSSI------LAGYSQS N-Q-LSDYFV-E-QW-REK---HSADEITVRDLAA-NPIPVL--DGELVGALRPS--DAPL-T 2q62 [32]HRPR-ILILYGS---LRT------VSYSRLLAEEARRLLEFF----GAEVKVFDPSG-LPLPDAAPVSH------1t0i --MK-VGIIMGSVRAKR------VCP-E-IAAY-VKRTIENSEELIDQK-LKIQVVDLQQI-ALPLYEDDDELI----PAQI----KS 1d4a VGRRALIVLAHS---ERT------SFN-YAMKEAAAAA--L-KKK-----GWEVVESDLYAMNFNPIISRKDITGKLKDPANFQYPAES 2b3d -SSN-ILIINGA---KKFAHSNGQ--LN-DTLTE--VADGTL-RD------GHDVRIVR--A------DSDYD------SsSSSSSs------hhhhhHHhHHHHHHhHHHHHHhHHHh------SSSSSSs------hhh-hhhh------hhhhh

100 110 120 130 140 150 160 170 180 3b6i ------PQELADYDAIIFGTPTRFGNMSGQMRTFLDQTG----G-LWASG------ALY-GKLASVFSSTGT 2z9d PRQQEALALS---DELIAELKAHDVIVIAAPMYNFNISTQLKNYFDLVAR--AGVTFRYT------ENGPEGLVTGKKAIVITSRGG 2q62 ------PKVQELRELSIWSEGQVWVSPERHGAMTGIMKAQIDWIPLS----TGSIR------PTQGKTLAVMQVSGG 1t0i -VDEYADSKTRSWSRIVNAL---DIIVFVTPQYNWGYPAALKNAIDRL------YHEWH------GKPALVV-SYGG 1d4a VLAYKEGHLSPDIVAEQKKLEAADLVIFQFPLQWFGVPAILKGWFERVFI---G-EFAYTYA------AMYDKGPFRSKKAVLSITTGG 2b3d ------VKAEVQNFLWA-DVVIWQMPGWWMGAPWTVKKYIDDVFTEGHGTLYASDGRTRKDPSKKYGSGGLVQGKKYMLSLTWNA

hhhhhhhhhhhHHHHHHHHHHHh-sSSSSSs------hHHHHHHHHHHHH---ssssss------SSSSSSSSs-

190 200 210 220 230 240 250 260 270 3b6i -GGG---QEQTITST---WTTL-AHH------GM-----VIV--PIGYAAQELFDVSQVRGGTPYGATTIAGGDGSRQP-----SQEE 2z9d IH-----KDG-PTDLVT--PYLSTFL-----GFIGI-TDVKFVFAE------GIAYG------PEMAAKAQSD 2q62 SQSFNAVNQMRILGRWMRMITI-PNQSSVAKAFQEFDANGRMKP------SSYYDRVVDVMEELVKFTLLTRDCSAYLTD 1t0i -HGGSKCNDQ-LQEVLH---GLK-MN------VIGG-----VAVKIP------VGTIPLP------EDIV--PQLSV-HNEE 1d4a --SGSMYSLQGIHGDM--NVILWPIQSGI-LHFCGFQVLEPQLTY------SIGHTPADARIQILE 2b3d PMEAFTEKDQFFHGVGVDGVYL-PFHKA--NQFLGMEPLPTFIAN------DVIKMPDVPR-YTEE ---hhhhhhhhhHHHHHHHHHHhhhhhhhhhHHh-----sssSSss------ssssss------hHHHHhHHHH

280 290 300 310 320 330 3b6i LSIARYQGEYVAGLAVKLNG------2z9d -AKAAIDSIVSA------2q62 RYSERKESAAELEHRVTLKSV------1t0i ILQLLASCI------1d4a GWKKRLENIWDETPLYFAPSSLFDLNFQAGFLMKKEVQDEEKNKKFGLSVGHHLGKSIPTDNQIKARK 2b3d -YRKHLVEIFG------HHHHHHHHHHhhhhhhhhh-hhhhhhhhhhhhhhhhhhhhh-----ssssss------

Figure 4-2. Alignment of NQOs using sequences of the crystal structures in Figure 4-1.

118

We performed extensive BLAST searches and significantly expanded each clade in an effort to produce a phylogram that adequately represents the widespread Type IV NQOs

(Figure 4-1). The sequence alignments were performed in ClustalX v1.83 and cured in

BioEdit v7.0.5.3 and the phylogram was generated with MEGA v4.0. In the consensus tree, we included many of the proteins appearing in the literature. Finally, the consensus phylogram was pruned to illustrate a diverse and representative set of Type IV NQOs from all domains of life.

Each clade includes several homologues that appear in the literature. The NQO1 proteins (Class I) consist of dimeric proteins with FAD cofactors, including NQO1, NQO2, and MdaB (modulator of drug activity B) homologues. The FMN reductases (Class II) represent the broadest and most loosely-clustered clade of NQOs, and they consist of dimeric and tetrameric FMN-containing proteins, including ArsH (arsenic reductase) homologues,

ChrR (chromate reductase) homologues, flavin reductases, nitroreductases, and proteins with azoreductase activity. The formal azoreductases (Class III) are a tightly-clustered clade and contain dimeric FMN-containing proteins, including AzoR (azoreductase) and ACP (acyl carrier protein) phosophodiesterase. The WrbA proteins (Class IV) are a tightly-clustered clade and consist of tetrameric FMN-containing proteins, including WrbA (tryptophan repressor binding protein), the Uhp1/obr1/p22/p25 protein from yeast, and many proteins from all domains of life which are being re-annotated as NAD(P)H:quinone oxidoreductases.

The remaining, loosely-clustered clade included in Figure 4-1 encompasses classic flavodoxins, which serve as a reference point for the multimeric, flavodoxin-like proteins.

Figure 4-2 features the sequences of the six structures displayed in Figure 4-1. They were aligned with ClustalX v1.83 and cured with BioEdit v7.0.5.3 using crystal structure information and secondary structure predictions made by JPRED 3. Most of the sequence

119 conservation in this alignment (Figure 4-2) is constrained to the central sheet motifs, which contact the FMN-cofactor in several places. The first sheet-turn motif, near the N-terminus, retains part of the PROSITE flavodoxin signature

(LIV)(LIVFY)(FY)X(ST)(V)X(AGC)XT(P) XXAXX(LIV) and contacts the FMN ribotyl- tail binding site. The third sheet-turn motif contacts the ribotyl-tail, the si-face of the isoalloxazine ring, and N(5) of the isoalloxazine ring. The fifth sheet-turn motif contacts the isoalloxazine ring near N(1) and O(2). In addition to these core sheet-turn motifs, it should be also noted that immediately following the third sheet-turn motif is a helix which has substantial sequence conservation and which contributes to the dimer interface. Altogether, these motifs construct the core of the protein monomer and dimer interface and make significant contacts with the redox-active FMN-cofactor, which has been found to impart thermostability to the protein and augment multimerization.

120

4.3 Discussion

There is little sequence similarity among the Type IV NQO subclasses to substantiate a firm relationship. As a result, identifying the physiological role of Type IV NQOs has proven difficult; there is a large volume of information available for NQOs, but it is divided among proteins that are seemingly unrelated.

Importantly, while the human genome appears to encode only NQO1 and NQO2, several genomes code for all subclasses of NQOs. For example, the genome of E. coli K-12

MG1655 encodes all of the NQO subclasses: Class I MdaB, Class II ChrR, Class III AzoR, and Class IV WrbA [36, 39, 123, 124]. Interestingly, Class II ArsH was also encoded in a few E. coli genomes but not that of E. coli K-12 MG1655 Consistent with our phylogenetic analysis, there is no primary sequence similarity among these proteins and an alignment (not shown) resembles Figure 4-2. The presence of all NQO subclasses in a single genome suggests that there is either a multiplicity in function for these proteins or that their redox- activities are linked with interactions that are unique to each subclass. Moreover, the presence of multiple NQOs in a single genome may occlude in vivo efforts to elucidate the physiological role [78].

Our phylogenetic analysis and evaluation of Type IV NQOs offers new perspectives for these proteins, demonstrating a relationship among them and highlighting possible physiological functions in protection from environmental xenobiotics.

121

4.3.1 Xenobiotic Detoxification

A role for the larger family of Type IV NQOs in detoxification is consistent with previous investigations. NQOs could contribute to alleviating damage related to the toxic or bioactive xenobiotic compounds that pervade most environments.

Among the most prevalent bioactive xenobiotics is molecular oxygen (O2). Reduced forms of O2 have increased toxicity as a result of the increased electrophilicity of reactive oxygen species (ROS) (Figure 4-3): molecular oxygen (Em,7 = -330 mV), superoxide (Em,7 =

940 mV), hydrogen peroxide (Em,7 = 320 mV), hydroperoxy radicals (Em,7 = 1060 mV), and the hydroxyl radical (Em,7 = 2310 mV). Accumulation of these highly energetic ROS and other electrophilic compounds leads to destruction of cofactors, modification of peptides, nucleotides, and lipids, and breakage of DNA strands. Under certain circumstances, this damage is further propelled by compounds like quinones, anthracyclines, and heavy metals, which participate in redox-cycling (Figure 4-3).

Xenobiotics affecting DNA structure, such as those listed in Table 4-1, are of particular concern because they are mutagenic and induce structural changes in DNA through intercalation and chemical modification (Figure 4-4). Several crystal structures demonstrate that xenobiotics, such as redox-cycling quinones and anti-cancer drugs, can localize and bind to DNA [125-135]. Exposure to these compounds, concomitant with the occurrence of ROS, can escalate damaging effects and increase the local concentration of chemical modifications.

These changes subsequently alter super-coiling properties, bonding energies, and protein-

DNA interactions, which affect transcription, translation, and replication; thus, they also increase probabilities of mutations, cell death, or diseases such as cancers [12, 23, 48, 49,

136-152].

122

In vivo protection from toxic or mutagenic xenobiotic stressors has been widely demonstrated for both NQO1 and NQO2 and for several bacterial Type IV NQOs, including several Class II homologues [27, 31, 78, 79, 81] and MdaB [28-30]. At least one publication demonstrates protection can be conferred by two azoreductases, though one of the sequences clusters with Class II NQOs [97]. In vitro investigations indicate that these NQOs are capable of coupling electron transfer from NAD(P)H to a wide range of compounds, and all of them are inhibited by dicumarol. Collectively, these studies implicate Type IV NQOs in alleviating environmental stress, and they support a redox-associated function such as reductive detoxification.

123

I) II) 2500 HO• -O O• 2000 O2 1500 1e- • O2 - HO • O O mV 1000 • 2 O2 - 2H+, 2e- H2O2 500 O2 H2O2 HO OH 0

O2 -500

Figure 4-3. I) Redox potentials (pH 7.0) for each reactive oxygen species. II) Scheme for quinone redox-cycling.

124

Figure 4-4. DNA and modifications caused by xenobiotics. A) Unmodified DNA. B) Intercalating xenobiotic displacing a base. C) Intercalation and cross-link to a nucleotide. D) Cross-link between strands. E) Intercalation and elongation of DNA structure. F) Combination of elongation due to intercalation and nucleotide adduct.

125

4.3.2 Redox-Associated Proteasomal Degradation

In addition to participating in enzymatic reductions, several publications indicate that

Type IV NQOs demonstrate redox-associated interactions with other proteins [54, 58, 59, 64-

66]. While reductive detoxification is a possible function for NQOs, the protein-protein interactions and in vivo studies more specifically link their redox state to metabolic processes, perhaps related to environmental stress.

Organisms necessarily acquired mechanisms to defend against and to recuperate from encounters with toxic or mutagenic xenobiotic compounds. One of the primary concerns for organisms is to limit exposure to such compounds; research indicates that organisms employ an array of proteins to directly metabolize or extrude xenobiotic compounds (such as phase II enzymes, drug efflux transporters, and heavy metal transporters). Significant research has established that organisms have developed mechanisms to repair or replace nucleotides, to repair broken DNA strands, and to condense or coil DNA for regulation and protection.

Further, organisms have also developed measures to ensure physical protection and metabolic security by selectively regulating the processes of protein production, protein folding, and constitutive or specific proteolysis. Altogether, these coincident resources leave cells adept at surviving many environmental factors.

Evidence suggests Type IV NQOs may function in regulation of cellular process and confer xenobiotic protection, dependent on the redox state of the cell. Two proteins from different subclasses – NQO1 (Class I) and Lot6p (Class II) – have demonstrated redox- independent associations with proteasomes, and redox-dependent associations with various proteins important for cellular regulation and protein turnover. Reduced, dimeric NQOs can bind to these proteins and confer protection from proteasomal degradation, implicating these proteins in metabolic regulation by controlling specific protein levels (Figure 4-5).

126

In addition to NQO1 and Lot6p, the Uhp1/obr1/p22/p25 protein (Class IV) from yeast also associates with a DNA-binding protein in vivo and contributes to gene silencing; however, this association has not yet been investigated for redox-dependence. Therefore, while several subclasses of these proteins appear to function as redox-related gates for proteasomal degradation, additional research with Type IV NQOs needs to be performed in order to fully elucidate their physiological roles; such research could be more explicitly directed towards DNA-protection, protein-protein interactions, protein degradation, and redox-linked processes.

127

Figure 4-5. Equilibria that have been demonstrated for Type IV NQO proteins. A) Holo- and apo- protein interactions. B) Multimerization. C) Redox- dependent protection from proteasomal degradation.

128

4.4 Future Directions

In reviewing the available data for flavodoxin-like NQOs, a noteworthy trend appears; the enhanced protection from xenobiotic compounds demonstrated for these proteins occurs with compounds that specifically affect DNA and DNA-binding proteins (Table 4-1).

This trend could suggest an association between xenobiotic protection and proteasomal degradation of DNA-binding proteins. Although speculative, this consideration could provide a measurable phenotype to aid research. Studies will need to address whether the enhanced survival is dependent on protective DNA structures (i.e. DNA condensation) or if it is dependent on the concentration of DNA-binding proteins.

Future investigations will require intimate knowledge of Type IV NQOs. It is apparent that sequence similarity among these proteins is not sufficient for identifying new

NQO subfamilies, while secondary structure alignments and biochemical activities are essential. Defining characteristics of these NQO proteins include redox-dependent activity

(with NADH, NADPH, or NRH), a flavodoxin-like fold, FAD or FMN cofactors, and an ability to form multimers. Lastly, recent investigations indicate that reduced, dimeric holo- proteins are necessary to confer protection from proteolysis (Figure 4-5). Apart from this knowledge, elucidation and definition of the physiological functions of Type IV NQOs will require a broad range of novel investigations across each subclass. Prediction of any other undiscovered subclasses may prove difficult, because there is little sequence similarity among these proteins.

129

4.5 References

1. Ernster, L. and F. Navazio (1958). Soluble Diaphorase in Animal Tissues. Acta Chemica Scandinavica. 12(3): p. 595-595.

2. Ernster, L., L. Danielson, and M. Ljunggren (1962). DT diaphorase. I. Purification from the soluble fraction of rat-liver cytoplasm, and properties. Biochim Biophys Acta. 58: p. 171-88.

3. Williams-Ashman, H.G. and C. Huggins (1961). Oxydation of reduced pyridine nucleotides in mammary gland and adipose tissue following treatment with polynuclear hydrocarbons. Med Exp Int J Exp Med. 4: p. 223-6.

4. Ernster, L., M. Ljunggren, and L. Danielson (1960). Purification and Some Properties of a Highly Dicumarol-Sensitive Liver Diaphorase. Biochemical and Biophysical Research Communications. 2(2): p. 88-92.

5. Ross, D. (2004). Quinone reductases multitasking in the metabolic world. Drug Metab Rev. 36(3-4): p. 639-54.

6. Ross, D., et al. (2000). NAD(P)H:quinone oxidoreductase 1 (NQO1): chemoprotection, bioactivation, gene regulation and genetic polymorphisms. Chem Biol Interact. 129(1-2): p. 77-97.

7. Jaiswal, A.K. (2000). Regulation of genes encoding NAD(P)H:quinone oxidoreductases. Free Radic Biol Med. 29(3-4): p. 254-62.

8. Ross, D. and D. Siegel (2004). NAD(P)H:quinone oxidoreductase 1 (NQO1, DT- diaphorase), functions and pharmacogenetics. Methods Enzymol. 382: p. 115-44.

9. Chen, S., K. Wu, and R. Knox (2000). Structure-function studies of DT-diaphorase (NQO1) and NRH: quinone oxidoreductase (NQO2). Free Radic Biol Med. 29(3-4): p. 276-84.

10. Phillips, R.M. and S. Basu (2004). Biological and Clinical Significance of Polymorphisms in NAD(P)H:Quinone Oxidoreductase 1 (NQO1). Current Pharmacogenomics. 2(1): p. 75-82.

11. Patterson, A.V., et al. (1998). Enzymology of tirapazamine metabolism: a review. Anticancer Drug Des. 13(6): p. 541-73.

12. Stiborova, M., et al. (2002). Carcinogenic aristolochic acids upon activation by DT- diaphorase form adducts found in DNA of patients with Chinese herbs nephropathy. Carcinogenesis. 23(4): p. 617-25.

13. Litwack, G. (2008). Vitamin K. Preface. Vitam Horm. 78: p. XIX-XX.

130

14. Oppermann, U. (2007). Carbonyl reductases: the complex relationships of mammalian carbonyl- and quinone-reducing enzymes and their role in physiology. Annu Rev Pharmacol Toxicol. 47: p. 293-322.

15. Tedeschi, G., S. Chen, and V. Massey (1995). DT-diaphorase. Redox potential, steady-state, and rapid reaction studies. J Biol Chem. 270(3): p. 1198-204.

16. Iyanagi, T. and I. Yamazaki (1970). One-electron-transfer reactions in biochemical systems. V. Difference in the mechanism of quinone reduction by the NADH dehydrogenase and the NAD(P)H dehydrogenase (DT-diaphorase). Biochim Biophys Acta. 216(2): p. 282-94.

17. Deller, S., P. Macheroux, and S. Sollner (2008). Flavin-dependent quinone reductases. Cell Mol Life Sci. 65(1): p. 141-60.

18. Lind, C., et al. (1990). DT-diaphorase: purification, properties, and function. Methods Enzymol. 186: p. 287-301.

19. Nebert, D.W., et al. (2002). NAD(P)H:quinone oxidoreductase (NQO1) polymorphism, exposure to benzene, and predisposition to disease: a HuGE review. Genet Med. 4(2): p. 62-70.

20. Bianchet, M.A., M. Faig, and L.M. Amzel (2004). Structure and mechanism of NAD(P)H:quinone acceptor oxidoreductases (NQO). Methods Enzymol. 382: p. 144- 74.

21. Ernster, L. (1967). DT Diaphorase. Methods Enzymol. 10: p. 309-317.

22. Horie, S. (1990). Advances in research on DT-diaphorase--catalytic properties, regulation of activity and significance in the detoxication of foreign compounds. Kitasato Arch Exp Med. 63(1): p. 11-30.

23. Long, D.J., 2nd and A.K. Jaiswal (2000). NRH:quinone oxidoreductase2 (NQO2). Chem Biol Interact. 129(1-2): p. 99-112.

24. Iyanagi, T. (1990). On the mechanism of one-electron reduction of quinones by microsomal flavin enzymes: the kinetic analysis between cytochrome B5 and menadione. Free Radic Res Commun. 8(4-6): p. 259-68.

25. Chen, S., et al. (1999). Molecular characterization of binding of substrates and inhibitors to DT-diaphorase: combined approach involving site-directed mutagenesis, inhibitor-binding analysis, and computer modeling. Mol Pharmacol. 56(2): p. 272-8.

26. Russ, R., J. Rau, and A. Stolz (2000). The function of cytoplasmic flavin reductases in the reduction of azo dyes by bacteria. Appl Environ Microbiol. 66(4): p. 1429-34.

131

27. Liu, G., et al. (2008). Enhancing survival of Escherichia coli by expression of azoreductase AZR possessing quinone reductase activity. Appl Microbiol Biotechnol. 80(3): p. 409-16.

28. Wang, G. and R.J. Maier (2004). An NADPH quinone reductase of Helicobacter pylori plays an important role in oxidative stress resistance and host colonization. Infect Immun. 72(3): p. 1391-6.

29. Hong, Y., G. Wang, and R.J. Maier (2008). The NADPH quinone reductase MdaB confers oxidative stress resistance to Helicobacter hepaticus. Microb Pathog. 44(2): p. 169-74.

30. Chatterjee, P.K. and N.L. Sternberg (1995). A general genetic approach in Escherichia coli for determining the mechanism(s) of action of tumoricidal agents: application to DMP 840, a tumoricidal agent. Proc Natl Acad Sci U S A. 92(19): p. 8950-4.

31. Sollner, S., et al. (2007). Lot6p from Saccharomyces cerevisiae is a FMN-dependent reductase with a potential role in quinone detoxification. FEBS J. 274(5): p. 1328-39.

32. Liu, G., et al. (2008). Site-directed mutagenesis of substrate binding sites of azoreductase from Rhodobacter sphaeroides. Biotechnol Lett. 30(5): p. 869-75.

33. Liu, G., et al. (2007). Azoreductase from Rhodobacter sphaeroides AS1.1737 is a flavodoxin that also functions as nitroreductase and flavin mononucleotide reductase. Appl Microbiol Biotechnol. 76(6): p. 1271-9.

34. Carey, J., et al. (2007). WrbA bridges bacterial flavodoxins and eukaryotic NAD(P)H:quinone oxidoreductases. Protein Sci. 16(10): p. 2301-5.

35. Patridge, E.V. and J.G. Ferry (2006). WrbA from Escherichia coli and Archaeoglobus fulgidus is an NAD(P)H:quinone oxidoreductase. J Bacteriol. 188(10): p. 3498-506.

36. Adams, M.A. and Z. Jia (2006). Modulator of drug activity B from Escherichia coli: crystal structure of a prokaryotic homologue of DT-diaphorase. J Mol Biol. 359(2): p. 455-65.

37. Abraham, K.J. and G.H. John (2007). Development of a Classification Scheme Using a Secondary and Tertiary Amino Acid Analysis of Azoreductase Gene. Journal of Medical and Biological Sciences. 1(2).

38. Ito, K., et al. (2008). Expansion of substrate specificity and catalytic mechanism of azoreductase by X-ray crystallography and site-directed mutagenesis. J Biol Chem. 283(20): p. 13889-96.

39. Ito, K., et al. (2006). Three-dimensional structure of AzoR from Escherichia coli. An oxidereductase conserved in microorganisms. J Biol Chem. 281(29): p. 20567-76.

132

40. Nissen, M.S., et al. (2008). Crystal structures of NADH:FMN oxidoreductase (EmoB) at different stages of catalysis. J Biol Chem. 283(42): p. 28710-20.

41. Agarwal, R., et al. (2006). Structure determination of an FMN reductase from Pseudomonas aeruginosa PA01 using sulfur anomalous signal. Acta Crystallogr D Biol Crystallogr. 62(Pt 4): p. 383-91.

42. Faig, M., et al. (2000). Structures of recombinant human and mouse NAD(P)H:quinone oxidoreductases: species comparison and structural changes with substrate binding and release. Proc Natl Acad Sci U S A. 97(7): p. 3177-82.

43. Foster, C.E., et al. (1999). Crystal structure of human quinone reductase type 2, a metalloflavoprotein. Biochemistry. 38(31): p. 9881-6.

44. Winski, S.L., et al. (2001). Characterization of a mechanism-based inhibitor of NAD(P)H:quinone oxidoreductase 1 by biochemical, X-ray crystallographic, and mass spectrometric approaches. Biochemistry. 40(50): p. 15135-42.

45. Asher, G., et al. (2006). The crystal structure of NAD(P)H quinone oxidoreductase 1 in complex with its potent inhibitor dicoumarol. Biochemistry. 45(20): p. 6372-8.

46. Li, R., et al. (1995). The three-dimensional structure of NAD(P)H:quinone reductase, a flavoprotein involved in cancer chemoprotection and chemotherapy: mechanism of the two-electron reduction. Proc Natl Acad Sci U S A. 92(19): p. 8846-50.

47. Faig, M., et al. (2001). Structure-based development of anticancer drugs: complexes of NAD(P)H:quinone oxidoreductase 1 with chemotherapeutic quinones. Structure. 9(8): p. 659-67.

48. Shen, Y.M., et al. (2006). Comparison of p53 mutations induced by PAH o-quinones with those caused by anti-benzo[a]pyrene diol epoxide in vitro: role of reactive oxygen and biological selection. Chem Res Toxicol. 19(11): p. 1441-50.

49. Long, D.J., 2nd, et al. (2001). NAD(P)H:quinone oxidoreductase 1 deficiency and increased susceptibility to 7,12-dimethylbenz[a]-anthracene-induced carcinogenesis in mouse skin. J Natl Cancer Inst. 93(15): p. 1166-70.

50. Cullen, J.J., et al. (2003). Dicumarol inhibition of NADPH:quinone oxidoreductase induces growth inhibition of pancreatic cancer via a superoxide-mediated mechanism. Cancer Res. 63(17): p. 5513-20.

51. Fagerholm, R., et al. (2008). NAD(P)H:quinone oxidoreductase 1 NQO1*2 genotype (P187S) is a strong prognostic and predictive factor in breast cancer. Nat Genet. 40(7): p. 844-53.

133

52. Wan, J., et al. (2002). Association of genetic polymorphisms in CYP2E1, MPO, NQO1, GSTM1, and GSTT1 genes with benzene poisoning. Environ Health Perspect. 110(12): p. 1213-8.

53. Asher, G., et al. (2001). Regulation of p53 stability and p53-dependent apoptosis by NADH quinone oxidoreductase 1. Proc Natl Acad Sci U S A. 98(3): p. 1188-93.

54. Asher, G., et al. (2005). A mechanism of ubiquitin-independent proteasomal degradation of the tumor suppressors p53 and p73. Genes Dev. 19(3): p. 316-21.

55. Chene, P. (2001). The role of tetramerization in p53 function. Oncogene. 20(21): p. 2611-7.

56. Wang, Y., et al. (1995). Interaction of p53 with its consensus DNA-binding site. Mol Cell Biol. 15(4): p. 2157-65.

57. Asher, G., et al. (2003). P53 hot-spot mutants are resistant to ubiquitin-independent degradation by increased binding to NAD(P)H:quinone oxidoreductase 1. Proc Natl Acad Sci U S A. 100(25): p. 15065-70.

58. Asher, G. and Y. Shaul (2005). p53 proteasomal degradation: poly-ubiquitination is not the whole story. Cell Cycle. 4(8): p. 1015-8.

59. Lavin, M.F. and N. Gueven (2006). The complexity of p53 stabilization and activation. Cell Death Differ. 13(6): p. 941-50.

60. Gong, X., et al. (2007). NRH:quinone oxidoreductase 2 and NAD(P)H:quinone oxidoreductase 1 protect tumor suppressor p53 against 20s proteasomal degradation leading to stabilization and activation of p53. Cancer Res. 67(11): p. 5380-8.

61. Asher, G., et al. (2002). Mdm-2 and ubiquitin-independent p53 proteasomal degradation regulated by NQO1. Proc Natl Acad Sci U S A. 99(20): p. 13125-30.

62. Anwar, A., et al. (2003). Interaction of human NAD(P)H:quinone oxidoreductase 1 (NQO1) with the tumor suppressor protein p53 in cells and cell-free systems. J Biol Chem. 278(12): p. 10368-73.

63. Anwar, A., et al. (2002). Interaction of the molecular chaperone Hsp70 with human NAD(P)H:quinone oxidoreductase 1. J Biol Chem. 277(16): p. 14060-7.

64. Asher, G., et al. (2005). 20S proteasomal degradation of ornithine decarboxylase is regulated by NQO1. Mol Cell. 17(5): p. 645-55.

65. Asher, G., N. Reuven, and Y. Shaul (2006). 20S proteasomes and protein degradation "by default". Bioessays. 28(8): p. 844-9.

66. Sollner, S., et al. (2009). Quinone reductase acts as a redox switch of the 20S yeast proteasome. EMBO Rep. 10(1): p. 65-70.

134

67. Horn, H.F. and K.H. Vousden (2007). Coping with stress: multiple ways to activate p53. Oncogene. 26(9): p. 1306-16.

68. Deller, S., et al. (2006). Characterization of a thermostable NADPH : FMN oxidoreductase from the mesophilic bacterium Bacillus subtilis. Biochemistry. 45(23): p. 7083-7091.

69. Jonas, K., et al. (2006). Identification of YhdA as a regulator of the Escherichia coli carbon storage regulation system. Fems Microbiology Letters. 264(2): p. 232-237.

70. Bafana, A., T. Chakrabarti, and S.S. Devi (2008). Azoreductase and dye detoxification activities of Bacillus velezensis strain AB. Applied Microbiology and Biotechnology. 77(5): p. 1139-1144.

71. Suzuki, Y., et al. (2001). Molecular cloning and characterization of the gene coding for azoreductase from Bacillus sp. OY1-2 isolated from soil. J Biol Chem. 276(12): p. 9059-65.

72. Blumel, S. and A. Stolz (2003). Cloning and characterization of the gene coding for the aerobic azoreductase from Pigmentiphaga kullae K24. Appl Microbiol Biotechnol. 62(2-3): p. 186-90.

73. Chen, H., S.L. Hopper, and C.E. Cerniglia (2005). Biochemical and molecular characterization of an azoreductase from Staphylococcus aureus, a tetrameric NADPH-dependent flavoprotein. Microbiology. 151(Pt 5): p. 1433-41.

74. Liger, D., et al. (2004). Crystal structure and functional characterization of yeast YLR011wp, an enzyme with NAD(P)H-FMN and ferric iron reductase activities. J Biol Chem. 279(33): p. 34890-7.

75. Vorontsov, II, et al. (2007). Crystal structure of an apo form of Shigella flexneri ArsH protein with an NADPH-dependent FMN reductase activity. Protein Sci. 16(11): p. 2483-90.

76. Sugiura, W., et al. (2006). Expression and characterization of the genes encoding azoreductases from Bacillus subtilis and Geobacillus stearothermophilus. Biosci Biotechnol Biochem. 70(7): p. 1655-65.

77. Sparla, F., et al. (1999). Cloning and heterologous expression of NAD(P)H : quinone reductase of Arabidopsis thaliana, a functional homologue of animal DT-diaphorase. Febs Letters. 463(3): p. 382-386.

78. Butcher, B.G., S.M. Deane, and D.E. Rawlings (2000). The chromosomal arsenic resistance genes of Thiobacillus ferrooxidans have an unusual arrangement and confer increased arsenic and antimony resistance to Escherichia coli. Appl Environ Microbiol. 66(5): p. 1826-33.

135

79. Gonzalez, C.F., et al. (2005). ChrR, a soluble quinone reductase of Pseudomonas putida that defends against H2O2. J Biol Chem. 280(24): p. 22590-5.

80. Ramirez-Diaz, M.I., et al. (2008). Mechanisms of bacterial resistance to chromium compounds. Biometals. 21(3): p. 321-32.

81. Neyt, C., et al. (1997). Virulence and arsenic resistance in Yersiniae. J Bacteriol. 179(3): p. 612-9.

82. Chen, H., et al. (2008). Functional role of Trp-105 of Enterococcus faecalis azoreductase (AzoA) as resolved by structural and mutational analysis. Microbiology. 154(Pt 9): p. 2659-67.

83. Punj, S. and G.H. John (2008). Purification and Identification of an FMN-dependent NAD(P)H Azoreductase from Enterococcus faecalis. Curr Issues Mol Biol. 11(1): p. 59-66.

84. Nakanishi, M., et al. (2001). Putative ACP phosphodiesterase gene (acpD) encodes an azoreductase. J Biol Chem. 276(49): p. 46394-9.

85. Wang, C.J., et al. (2007). Molecular cloning, characterisation and ligand-bound structure of an azoreductase from Pseudomonas aeruginosa. J Mol Biol. 373(5): p. 1213-28.

86. Platzek, T., et al. (1999). Formation of a carcinogenic aromatic amine from an azo dye by human skin bacteria in vitro. Human & Experimental Toxicology. 18(9): p. 552-559.

87. Blumel, S., H.J. Knackmuss, and A. Stolz (2002). Molecular cloning and characterization of the gene coding for the aerobic azoreductase from Xenophilus azovorans KF46F. Appl Environ Microbiol. 68(8): p. 3948-55.

88. Maier, J., et al. (2004). A new alkali-thermostable azoreductase from Bacillus sp. strain SF. Appl Environ Microbiol. 70(2): p. 837-44.

89. Pricelius, S., et al. (2007). Enzymatic reduction of azo and indigoid compounds. Applied Microbiology and Biotechnology. 77(2): p. 321-327.

90. Rafii, F., J.D. Hall, and C.E. Cerniglia (1997). Mutagenicity of azo dyes used in foods, drugs and cosmetics before and after reduction by Clostridium species from the human intestinal tract. Food and Chemical Toxicology. 35(9): p. 897-901.

91. Rafii, F. and C.E. Cerniglia (1995). Reduction of azo dyes and nitroaromatic compounds by bacterial enzymes from the human intestinal tract. Environ Health Perspect. 103 Suppl 5: p. 17-9.

92. Liu, Z.J., et al. (2007). Crystal structure of an aerobic FMN-dependent azoreductase (AzoA) from Enterococcus faecalis. Arch Biochem Biophys. 463(1): p. 68-77.

136

93. Ito, K., et al. (2005). Crystallization and preliminary X-ray analysis of AzoR (azoreductase) from Escherichia coli. Acta Crystallogr Sect F Struct Biol Cryst Commun. 61(Pt 4): p. 399-402.

94. Bafana, A. and T. Chakrabarti (2008). Lateral gene transfer in phylogeny of azoreductase enzyme. Comput Biol Chem. 32(3): p. 191-7.

95. Huang, M.T., et al. (1979). Rat-Liver Cytosolic Azoreductase - Electron-Transport Properties and the Mechanism of Dicumarol Inhibition of the Purified Enzyme. Journal of Biological Chemistry. 254(22): p. 1223-1227.

96. Huang, M.T., G.T. Miwa, and A.Y.H. Lu (1979). Rat-Liver Cytosolic Azoreductase - Purification and Characterization. Journal of Biological Chemistry. 254(10): p. 3930- 3934.

97. Kim, S., et al. (2009). Comparison of the effects of NADH- and NADPH-perturbation stresses on the growth of Escherichia coli. Curr Microbiol. 58(2): p. 159-63.

98. Stammers, D.K., et al. (2001). The structure of the negative transcriptional regulator NmrA reveals a structural superfamily which includes the short-chain dehydrogenase/reductases. EMBO J. 20(23): p. 6619-26.

99. Zheng, X., et al. (2007). Restructuring of the dinucleotide-binding fold in an NADP(H) sensor protein. Proc Natl Acad Sci U S A. 104(21): p. 8809-14.

100. Fischl, A.S. and E.P. Kennedy (1990). Isolation and properties of acyl carrier protein phosphodiesterase of Escherichia coli. J Bacteriol. 172(9): p. 5445-9.

101. Yang, W., L. Ni, and R.L. Somerville (1993). A stationary-phase protein of Escherichia coli that affects the mode of association between the trp repressor protein and operator-bearing DNA. Proc Natl Acad Sci U S A. 90(12): p. 5796-800.

102. Pomposiello, P.J., M.H. Bennik, and B. Demple (2001). Genome-wide transcriptional profiling of the Escherichia coli responses to superoxide stress and sodium salicylate. J Bacteriol. 183(13): p. 3890-902.

103. Chang, D.E., D.J. Smalley, and T. Conway (2002). Gene expression profiling of Escherichia coli growth transitions: an expanded stringent response model. Mol Microbiol. 45(2): p. 289-306.

104. Cheung, K.J., et al. (2003). A microarray-based antibiotic screen identifies a regulatory role for supercoiling in the osmotic stress response of Escherichia coli. Genome Res. 13(2): p. 206-15.

105. Ishihama, A. (2000). Functional modulation of Escherichia coli RNA polymerase. Annu Rev Microbiol. 54: p. 499-518.

137

106. Kirkpatrick, C., et al. (2001). Acetate and formate stress: opposite responses in the proteome of Escherichia coli. J Bacteriol. 183(21): p. 6466-77.

107. Tao, H., et al. (1999). Functional genomics: expression analysis of Escherichia coli growing on minimal and rich media. J Bacteriol. 181(20): p. 6425-40.

108. Tucker, D.L., N. Tucker, and T. Conway (2002). Gene expression profiling of the pH response in Escherichia coli. J Bacteriol. 184(23): p. 6551-8.

109. Liu, X. and P. De Wulf (2004). Probing the ArcA-P modulon of Escherichia coli by whole genome transcriptional analysis and sequence recognition profiling. J Biol Chem. 279(13): p. 12588-97.

110. Kang, Y.S., et al. (2005). Genome-wide expression analysis indicates that FNR of Escherichia coli K-12 regulates a large number of genes of unknown function. Journal of Bacteriology. 187(3): p. 1135-1160.

111. Brock, B.J. and M.H. Gold (1996). 1,4-Benzoquinone reductase from basidiomycete Phanerochaete chrysosporium: spectral and kinetic analysis. Arch Biochem Biophys. 331(1): p. 31-40.

112. Brock, B.J., S. Rieble, and M.H. Gold (1995). Purification and Characterization of a 1,4-Benzoquinone Reductase from the Basidiomycete Phanerochaete chrysosporium. Appl Environ Microbiol. 61(8): p. 3076-3081.

113. Cohen, R., M.R. Suzuki, and K.E. Hammel (2004). Differential stress-induced regulation of two quinone reductases in the brown rot basidiomycete Gloeophyllum trabeum. Appl Environ Microbiol. 70(1): p. 324-31.

114. Jensen Jr, K.A., Jr., et al. (2002). An NADH:quinone oxidoreductase active during biodegradation by the brown-rot basidiomycete Gloeophyllum trabeum. Appl Environ Microbiol. 68(6): p. 2699-703.

115. Laskowski, M.J., et al. (2002). FQR1, a novel primary auxin-response gene, encodes a flavin mononucleotide-binding quinone reductase. Plant Physiol. 128(2): p. 578-90.

116. Matvienko, M., et al. (2001). Quinone oxidoreductase message levels are differentially regulated in parasitic and non-parasitic plants exposed to allelopathic quinones. Plant J. 25(4): p. 375-87.

117. Wrobel, R.L., M. Matvienko, and J.I. Yoder (2002). Heterologous expression and biochemical characterization of an NAD(P)H : quinone oxidoreductase from the hemiparasitic plant Triphysaria versicolor. Plant Physiology and Biochemistry. 40(3): p. 265-272.

118. Invernizzi, G., et al. (2007). Protein-protein and protein-ligand interactions studied by electrospray-ionization mass spectrometry. Protein and Peptide Letters. 14(9): p. 894-902.

138

119. Gorman, J. and L. Shapiro (2005). Crystal structures of the tryptophan repressor binding protein WrbA and complexes with flavin mononucleotide. Protein Science. 14(12): p. 3004-3012.

120. Natalello, A., et al. (2007). Role of flavin mononucleotide in the thermostability and oligomerization of Escherichia coli stress-defense protein WrbA. Biochemistry. 46(2): p. 543-553.

121. Grandori, R., et al. (1998). Biochemical characterization of WrbA, founding member of a new family of multimeric flavodoxin-like proteins. J Biol Chem. 273(33): p. 20960-6.

122. Naresh, A., S. Saini, and J. Singh (2003). Identification of Uhp1, a ubiquitinated histone-like protein, as a target/mediator of Rhp6 in mating-type silencing in fission yeast. J Biol Chem. 278(11): p. 9185-94.

123. Andrade, S.L., et al. (2007). Crystal structure of the NADH:quinone oxidoreductase WrbA from Escherichia coli. J Bacteriol. 189(24): p. 9101-7.

124. Ackerley, D.F., et al. (2004). Chromate-reducing properties of soluble flavoproteins from Pseudomonas putida and Escherichia coli. Appl Environ Microbiol. 70(2): p. 873-82.

125. Schwartz, J.L., et al. (1997). Solution structure of the minor conformer of a DNA duplex containing a dG mismatch opposite a benzo[a]pyrene diol epoxide/dA adduct: glycosidic rotation from syn to anti at the modified deoxyadenosine. Biochemistry. 36(37): p. 11069-76.

126. Williams, L.D., et al. (1990). Ternary interactions of spermine with DNA: 4'- epiadriamycin and other DNA: anthracycline complexes. Nucleic Acids Res. 18(18): p. 5533-41.

127. Gallego, J. and B.R. Reid (1999). Solution structure and dynamics of a complex between DNA and the antitumor bisnaphthalimide LU-79553: intercalated ring flipping on the millisecond time scale. Biochemistry. 38(46): p. 15104-15.

128. Leonard, G.A., et al. (1993). Anthracycline-DNA interactions at unfavourable base- pair triplet-binding sites: structures of d(CGGCCG)/daunomycin and d(TGGCCA)/adriamycin complexes. Acta Crystallogr D Biol Crystallogr. 49(Pt 5): p. 458-67.

129. Frederick, C.A., et al. (1990). Structural comparison of anticancer drug-DNA complexes: adriamycin and daunomycin. Biochemistry. 29(10): p. 2538-49.

130. Searle, M.S., A.J. Maynard, and H.E. Williams (2003). DNA recognition by the anthracycline antibiotic respinomycin D: NMR structure of the intercalation complex with d(AGACGTCT)2. Org Biomol Chem. 1(1): p. 60-6.

139

131. Lipscomb, L.A., et al. (1994). Water ring structure at DNA interfaces: hydration and dynamics of DNA-anthracycline complexes. Biochemistry. 33(12): p. 3649-59.

132. Leonard, G.A., T. Brown, and W.N. Hunter (1992). Anthracycline binding to DNA. High-resolution structure of d(TGTACA) complexed with 4'-epiadriamycin. Eur J Biochem. 204(1): p. 69-74.

133. Gao, Y.G. and A.H. Wang (1991). Influence of aglycone modifications on the binding of anthracycline drugs to DNA: the molecular structure of idarubicin and 4- O-demethyl-11-deoxydoxorubicin complexed to d(CGATCG). Anticancer Drug Des. 6(3): p. 137-49.

134. Yeh, H.J., et al. (1995). NMR solution structure of a nonanucleotide duplex with a dG mismatch opposite a 10S adduct derived from trans addition of a deoxyadenosine N6- amino group to (+)-(7R,8S,9S,10R)-7,8-dihydroxy-9,10-epoxy-7,8,9,10- tetrahydrobenzo[a]pyrene: an unusual syn glycosidic torsion angle at the modified dA. Biochemistry. 34(41): p. 13570-81.

135. Zegar, I.S., et al. (1996). Adduction of the human N-ras codon 61 sequence with (-)- (7S,8R,9R,10S)-7,8-dihydroxy-9,10-epoxy-7,8,9,10-tetrahydrobenzo[a] pyrene: structural refinement of the intercalated SRSR(61,2) (-)-(7S,8R,9S,10R)-N6-[10- (7,8,9,10- tetrahydrobenzo[a]pyrenyl)]-2'-deoxyadenosyl adduct from 1H NMR. Biochemistry. 35(20): p. 6212-24.

136. Ling, H., et al. (2004). Crystal structure of a benzo[a]pyrene diol epoxide adduct in a ternary complex with a DNA polymerase. Proc Natl Acad Sci U S A. 101(8): p. 2265- 9.

137. Szyf, M. (2007). The dynamic epigenome and its implications in toxicology. Toxicol Sci. 100(1): p. 7-23.

138. Park, S.H., Y.J. Ko, and C.K. Kini (2001). Toxic effects of catechol and 4- chlorobenzoate stresses on bacterial cells. Journal of Microbiology. 39(3): p. 206- 212.

139. Vranova, E., D. Inze, and F. Van Breusegem (2002). Signal transduction during oxidative stress. J Exp Bot. 53(372): p. 1227-36.

140. Long, D.J., 2nd, et al. (2000). NAD(P)H:quinone oxidoreductase 1 deficiency increases susceptibility to benzo(a)pyrene-induced mouse skin carcinogenesis. Cancer Res. 60(21): p. 5913-5.

141. Simpkins, H., L.F. Pearlman, and L.M. Thompson (1984). Effects of adriamycin on supercoiled DNA and calf thymus nucleosomes studied with fluorescent probes. Cancer Res. 44(2): p. 613-8.

140

142. Guengerich, F.P. (2000). Metabolism of chemical carcinogens. Carcinogenesis. 21(3): p. 345-51.

143. Rabbani, A., R.M. Finn, and J. Ausio (2005). The anthracycline antibiotics: antitumor drugs that alter chromatin structure. Bioessays. 27(1): p. 50-6.

144. Fang, Y.Z., S. Yang, and G.Y. Wu (2002). Free radicals, antioxidants, and nutrition. Nutrition. 18(10): p. 872-879.

145. Al-Maghrebi, M.A. and L.T. Benov (2001). Polyphosphate accumulation and oxidative DNA damage in superoxide dismutase-deficient Escherichia coli. Free Radic Biol Med. 31(11): p. 1352-9.

146. Flowers, L., S.T. Ohnishi, and T.M. Penning (1997). DNA strand scission by polycyclic aromatic hydrocarbon o-quinones: role of reactive oxygen species, Cu(II)/Cu(I) redox cycling, and o-semiquinone anion radicals. Biochemistry. 36(28): p. 8640-8.

147. Sjolin, A.M. and D.R. Livingstone (1997). Redox cycling of aromatic hydrocarbon quinones catalysed by digestive gland microsomes of the common mussel (Mytilus edulis L). Aquatic Toxicology. 38(1-3): p. 83-99.

148. Lim, P., et al. (2004). Peroxyl radical mediated oxidative DNA base damage: Implications for lipid peroxidation induced mutagenesis. Biochemistry. 43(49): p. 15339-15348.

149. Varanasi, U., et al. (1987). Chemical carcinogenesis in feral fish: uptake, activation, and detoxication of organic xenobiotics. Environ Health Perspect. 71: p. 155-70.

150. Pivonkova, H., et al. (2006). DNA modification with cisplatin affects sequence- specific DNA binding of p53 and p73 proteins in a target site-dependent manner. FEBS J. 273(20): p. 4693-706.

151. Silvian, L.F., J. Wang, and T.A. Steitz (1999). Insights into editing from an ile-tRNA synthetase structure with tRNAile and mupirocin. Science. 285(5430): p. 1074-7.

152. Hsu, G.W., et al. (2005). Structure of a high fidelity DNA polymerase bound to a benzo[a]pyrene adduct that blocks replication. J Biol Chem. 280(5): p. 3764-70.

141

CHAPTER FIVE: CO-FACTOR BINDING AND REDOX-POTENTIAL MODULATION IN WRBA, FOUNDING MEMBER OF THE CLASS IV NAD(P)H:QUINONE OXIDOREDUCTASES 5.1 Abstract 5.2 Introduction 5.3 Materials and Methods 5.4 Results 5.5 Equilibria Derivations 5.6 Discussion 5.7 Acknowledgements 5.8 References

Chapter 5

Co-Factor Binding and Redox-Potential Modulation in WrbA, Founding Member of the Class IV NAD(P)H:Quinone Oxidoreductases

Eric V. Patridge and James G. Ferry*

Department of Biochemistry and Molecular Biology, Eberly College of Science, The Pennsylvania State University, 205 South Frear Laboratory, University Park, PA 16802-4500.

*Corresponding author. Tel.: +1 (814) 863-5721; Fax: +1(814) 863-6217; E-mail: [email protected]

5.1 Abstract

We recently reviewed the flavodoxin-like NAD(P)H:Quinone Oxidoreductases

(NQOs) which function in redox-linked processes and protection from environmental stressors (Chapter 4). The reduced NQO dimer is required to protect protein targets from proteasomal degradation. This study is the first comprehensive biophysical investigation of any flavodoxin-like NQO protein. It focuses on the peptide interactions at the perimeter of the flavin isoalloxazine ring in the prototype WrbA protein from Escherichia coli and its hyperthermophilic homologue from Archaeoglobus fulgidus. The specific role of WrbA is elusive, but the protein participates in monomer-dimer-tetramer equilibria with one loosely- bound FMN cofactor per monomer. Investigations with WrbA proteins were based on studies performed with the distinct monomeric flavodoxin cofactor which stabilizes a one- electron semiquinone. Biochemical and biophysical characterizations were performed with wild-type proteins and alanine variants of T78, F80, W98, T116, H133, and Y143 in E. coli

WrbA. Fluorescence quenching studies demonstrated cooperativity in flavin-binding and that flavin-binding is linked to multimerization in E. coli WrbA; monomer WrbA does not bind flavin. The T78, F80, and W98 residues function in flavin-dependent multimerization equilibria and cooperative behavior is lost in the F80A variant. We also determined redox potentials for each of the alanine variants with a gold-capillary spectroelectrochemical cell modified from previous investigations [76-80]. The H133 residue functions in electron accumulation on the E. coli WrbA isoalloxazine ring. Wild-type protein participates in two one-electron transfers while H133A participates in one two-electron transfer. Specific activities indicate that WrbA proteins retain NAD(P)H-dependent activity with smaller electron acceptors and that of the residues investigated by alanine-scanning are also important for enzyme activity.

143

5.2 Introduction

In 1993, WrbA co-immunoprecipitated during purification of TrpR [1]. Gel mobility shift assays probed whether WrbA enhanced interactions between TrpR and its DNA target but they had conflicting results [1, 2]. Several publications demonstrate that WrbA expression is under the control of the FNR (fumarate and nitrate reductase) regulatory protein, ArcA/ArcB (anoxic redox control), and RpoS (the stationary phase sigma factor, s or 38) [1, 3, 4]. Expression of WrbA is repressed in the absence of oxygen and when the quinone pool is reduced, but it is induced upon entering the stationary phase. Several articles indicate that expression is up-regulated in response to stressors such as acids, salts, H2O2, and diauxie, implicating WrbA in environmental protection [5-12]. Nothing else is known about the role of WrbA in a stress response.

We previously reported on flavodoxin-like NAD(P)H:quinone oxidoreductases

(NQOs) and delineated a phylogeny of the proteins (in process). The flavodoxin-like NQOs are widespread and resolve into four clades (Class I-IV). Investigations indicate NQOs confer protection from xenobiotic stressors [13-21]. The role of the NQO1 protein in xenobiotic protection and in activating anti-cancer drugs has been investigated for nearly a half-century. Additionally, members from Classes I (NQO1), III (Lot6p), and IV

(Uhp1/obr1) demonstrate interactions with various chaperones and DNA-binding proteins.

NQO1 and Lot6p dimers protect several proteins from proteasomal degradation in a redox- linked manner [22-25]. NQO1 also protects ornithine decarboxylase, the first and rate- controlling enzyme in polyamine biosynthesis [26]. Thus, through redox-linked processes,

NQOs are thought to confer protection from xenobiotics and regulate cellular processes according the cellular redox state.

144

NQOs share the /, twisted open-sheet fold that is typical of flavodoxins [27].

Classical flavodoxin is a small protein and has long served as an ideal model for examining the thermodynamic characteristics of strong peptide-flavin interactions. Thus, binding and modulation of the flavodoxin co-factor is well investigated [28-30]. Significant research has also delineated mechanisms that stabilize the flavodoxin semiquinone [31-64]. Many of these studies significantly augment our comprehension of protein folding, cofactor binding, and redox potentials in flavoproteins.

In contrast to flavodoxins, NQOs form multimers and are active with NADH,

NADPH, or NRH. Classes I-III retain the FMN co-factor of classical flavodoxins, while

Class I NQOs (including NQO1, NQO2, and MdaB) have FAD co-factors [65-67]. Peptide- flavin interactions are not well characterized for these proteins. Unlike the flavodoxin cofactor, reports indicate that NQO flavin is loosely associated and is highly affected by pH and ionic strength [15, 68-72]. Further, NQOs prefer two-electron transfers, which is also distinct from flavodoxins. Importantly, published investigations with WrbA indicate the flavin encourages multimerization and imparts significant thermostability [2, 68, 73]. Given that the observed interactions with NQOs require dimer formation and are redox-linked, elucidating the mechanisms that bind and modulate flavin will enhance our ability to investigate Type IV NQOs.

Here, we present the first analysis of a flavodoxin-like NQO using alanine-scanning of the flavin active site. Only one study has examined any peptide interaction with a NQO isoalloxazine ring (Class II AzoA). The published investigation focused on an aromatic residue at the re-face of the flavin, which is not present in WrbA [74]. Here, we focus on the peptide-flavin interactions at the isoalloxazine perimeter of WrbA proteins. The experiments presented offer substantial insight into co-factor binding and modulation in Type IV NQOs.

145

5.3 Materials and Methods

Materials and reagents

Primers were obtained through Integrated DNA Technologies (Coralville, IA).

Plasmids encoding wild type proteins were previously constructed and used for mutagenesis reactions. RosettaBlue cell lines were obtained through Novagen (Madison, WI). Other materials used to construct E. coli (RosettaBlue) knockouts were gifts from Joe Palladino,

Jennifer Hayden, Mary Laubacher, Sarah Barchinger, and Sarah Ades (The Pennsylvania

State University). The chromatography resins and equipment were purchased from GE

Healthcare. All other chemicals were purchased from ICN (MP Biomedical), Sigma, VWR,

ACROS, Alfa Aesar, or AccuStandard (New Haven, CT).

Construction and Analysis of E. coli wrbA

RosettaBlue(DE3)pLacI was used as the wild-type strain, and disruption of wrbA was constructed according to published methods [75]. Primers were designed with the pKD4 template, and kanamycin resistant genes were spliced from the knockout with a FLP helper plasmid (pCP20). Thermal induction of FLP synthesis was carried out in Luria Broth at

43ºC, which also eliminated the temperature sensitive plasmid. The knockout was confirmed with external and internal primer analyses as before (Figure 3-8). Cell stocks were flash frozen in 20% glycerol and stored at -80ºC.

Cloning, expression, purification, and reconstitution

Genes encoding wild type WrbA proteins were previously constructed [68]. Variants of E. coli WrbA were constructed from wild type plasmid using site-directed mutagenesis.

Final plasmids were sequenced and transformed into electrocompetent RosettaBlue wrbA- cells for expression. Transformed cells were cultured at 37ºC in Terrific Broth containing 50 146

g/ml ampicillin and 34 g/ml chloramphenicol. When the OD600 reached 0.8, protein expression was induced by the addition of 1 mM IPTG at 37ºC. Cells were harvested after 4 hours by centrifugation and stored at -80ºC.

Using an AKTA Explorer 100 FPLC (GE Healthcare), aerobic purifications were performed as previously reported, using 50 mM MOPS instead of 50 mM sodium phosphate

[68]. Apo-protein was obtained by extensively washing the protein while it was bound to the second Q-sepharose column (pH 7.8). Fractions containing apo-protein were monitored by

SDS-PAGE, pooled, and dialyzed into 50 mM MOPS pH 7.0. After dialysis, proteins were passed through a 0.2 m syringe filter. After 16 hour dialysis into 50 mM MOPS pH 7.0, proteins were concentrated to 2.5 ml. Proteins were flash-frozen and stored at -80ºC.

Fluorescence Quenching

Titrations of 150 M and 5 M FMN with apo-protein were measured using a F-2000 fluorimeter (Hitachi) and a fluorimeter cuvette (2 mm x 1 cm pathlength) from Starna

(Atascadero, CA). Spectra collection was automated through the fluorimeter’s RS232 port using the GW-BASIC command-line environment. For all proteins with 150 M FMN, emission spectra (480 nm to 680 nm) were collected at excitation = 459 nm. With 5 M FMN, emission spectra (580 nm to 750 nm) were collected at excitation = 459 nm for all proteins except F80A (which required emission spectra from 480 nm to 680 nm).

Apo-proteins were thawed and twice dialyzed at 4°C for 8 hours against 4 L of 100 mM MOPS pH 6.5, 100 mM NaCl, 15% glycerol (Buffer A). Protein concentrations were determined from frozen stocks using extinction coefficients (ε280) calculated with ExPASy’s

ProtParam tool. Titrations were incubated in the dark and at room temperature for an hour prior to measurement. Binding assays were conducted in triplicate.

147

Raw titration curves were fitted in GraphPad Prism v5.0 with non-linear fitting. Data with an apparent cooperativity of 4 were fit with:

4 푎푝푝 2 4 4 푌 = 푌0 + 퐵푚푎푥 × [ 푋 /((퐾푑1 ) + 푋 )] + (푁푆 × 푋 )

Data with an apparent cooperativity of 2 were fit with:

2 푎푝푝 2 2 2 푌 = 푌0 + 퐵푚푎푥 × [ 푋 /((퐾푑1 ) + 푋 )] + (푁푆 × 푋 )

Data with no cooperativity were fit with:

푎푝푝 푌 = 푌0 + 퐵푚푎푥 × [ 푋 /(퐾푑1 + 푋)] + (푁푆 × 푋)

Data with two binding events, were fit with:

푎푝푝 4 푎푝푝 2 4 4 푌 = 푌0 + 퐵푚푎푥 × [ 푋 /(퐾푑1 + 푋)] + 퐵푚푎푥2 × [ 푋 /((퐾푑2 ) + 푋 ] + 푁푆 × 푋

Data with limited flavin and an apparent cooperativity of 2 were fit with:

2 푎푝푝 2 2 푌 = 푌0 + 퐵푚푎푥 × [ 푋 /((퐾푑1 ) + 푋 )]

Derivations of these equations are described in section 5-5. In the equations Bmax was the maximum quenching observed, NS represented nonspecific binding, and Y0 was the Y-axis offset. Kd values were the apparent binding constants and n was the observed cooperativity.

Fitted curves were scaled for comparison in Figures 5-7 and 5-8. Data were also fit using the

Hill equation with transformed data, and these correlated well with reported cooperativities.

Potentiometric Titrations

Potentiometric titrations were conducted with a custom-engineered, fiber-optic spectroelectrochemical cell, built as previously described [76-80]. Engineering schematics are available in Appendix A. The cell was housed in a custom anaerobic box purged with argon. The cell potential was controlled using an EG&G PAR 273 or 273A potentiostat with platinum wire electrodes, a gold capillary electrode, and a calomel double-junction reference electrode purchased from Hach (Loveland, CO). Membranes were constructed with dialysis membrane from 3500 MWCO Slide-A-Lyzer cassettes (Pierce Rockford, IL) and with

148 custom carbonate flanges. These were epoxied together and cured at room temperature for

48 hours, using a mixture of EPON Resin 828: EPI-CURE Curing Agent 3140: water (1.5: 1:

0.02). Resin and Curing Agent were gifts from Hexion (Columbus, OH). Spectra were measured through a Kevlar-jacketed, 960m-core fiber (Varian Inc, Palo Alto, CA), using a

USB4000-UV-VIS spectrometer, a DH-2000 light source, and the SpectraSuite spectroscopy platform from Ocean Optics (Dunedin, FL).

Apo-protein concentrations were determined from frozen stocks using extinction coefficients (ε280) calculated with ExPASy’s ProtParam tool. All proteins were diluted to 5X their respective KL2 values (Table 2) as determined in this study. After addition of excess

FMN, proteins were twice dialyzed at 4°C for 8 hours in 4 L of Buffer A. Final concentrations were checked by flavin absorbance at max near 450 nm.

Concentrated redox slurry was made in Buffer A with 50% ethanol, degassed, and stored under nitrogen in the dark. Anaerobic redox slurry was diluted into Buffer A or protein samples immediately before equilibrating the cell. The diluted slurry contained 1,4- hydroquinone (1 M), 1,4-naphthoquinone (1 M), menadione (1 M), phenosafranine (0.5

M), benzyl viologen (0.5 M), and methyl viologen (0.5 M). Neomycin dissolved in

Buffer A was then added to the samples (final concentration 2 mM) as an electrochemical promoter and to discourage protein adsorption. The system was equilibrated at length before experiments. First the cell was equilibrated overnight in Buffer A, and then it was equilibrated for several hours with protein samples. For the final equilibration, the cell was flushed with protein several times in 10-minute intervals. The redox cell was then cycled to -

400 mV for 10 minutes and back to +200 mV where it was permitted to reach equilibrium.

Potentiometric titrations were measured in triplicate and were followed to 300 mV beyond both sides of the redox potentials. All transitions were reversible, but oxidation was

149 slower than reduction, so all data were obtained when the cell reached equilibrium. For each measurement, the cell was considered to be at equilibrium when absorbance spectra were stable for 20 minutes. Only one transition was observed for each protein. Both oxidation and reduction data were utilized to calculate the redox potentials and electron transfer properties from Nernst plots (Figures 5-4 and 5-5).

Specific Activities

Reactions were monitored with a Varian Cary 50 spectrophotometer and an anaerobic fluorescence cuvette (2 mm x 1 cm pathlength) from Starna (Atascadero, CA). Although fluorescence was not measured, the cuvette permitted small reaction volumes under anaerobic conditions, and the 2 mm pathlength permitted high concentrations of NADH to be accurately determined spectroscopically. All activities were obtained at 24ºC, and the reactions were initiated by the addition of enzyme after brief equilibration of the assay mixture. All activities were performed in triplicate, and all non-enzymatic rates were negligible.

NADH was used as the electron donor to determine the specific activities of each protein with a variety of electron acceptors. Reaction mixtures contained 50 mM MOPS (pH

7.0) with 800 M NADH, 400 M electron acceptor, and variable concentrations of protein.

1,4-anthraquinone and 9,10-anthraquinone were dissolved in 50% acetonitrile and assayed in

25% acetonitrile. In all assays except with K3Fe(CN)6, oxidation of NADH was followed at

-1 -1 340 nm (340 = 6.22 mM cm ). Reduction of K3Fe(CN)6 was followed at 420 nm (420 =

1.04 mM-1cm-1). The linear data of each progress curve was taken as the apparent initial velocity. Non-enzymatic reactions represented < 2% of all enzymatic reaction conditions.

150

5.4 Results

Design of variants

This is the first comprehensive, biochemical investigation of cofactor binding and modulation for flavodoxin-like NQOs. In preparation for this study, numerous articles and crystal structures of classical flavodoxins and flavodoxin-like proteins were surveyed and compared. Peptide elements that are central to classical flavodoxin function, specifically the features that bind and modulate the FMN cofactor, were implemented in the design of variants to investigate their portability and effect in a similar protein. Only one study mechanistically investigates peptide-flavin interactions of any flavodoxin-like NQO, and it specifically focuses on an aromatic contribution at the re-face. This is the first investigation of the NQO isoalloxazine perimeter.

Two intersecting sets of E. coli WrbA variants were investigated and compared to wild type. The primary focus of this study utilized a set of alanine variants (T78A, F80A,

W98A, T116A, H133A, Y143A) to identify residues interacting with the isoalloxazine ring that contribute to flavin-affinity and co-factor modulation (Figure 5-1). Wild type WrbA from Archaeoglobus fulgidus was also investigated for comparison with an archaeal homologue. A second set of variants, in the vicinity of N(1) and N(5), qualitatively investigated the local environment surrounding the cofactor (F80A, F80G, F80GG, F80V,

F80Y and T116A, T116M, T116F, T116Y, T116W).

151

I) W98 I) W98

F80 F80

T78 T78

II) Y143 II) Y143

T116 T116 H133 H133

Figure 5-1. Stereo images displayed in PyMOL, illustrating the active site and locations of each variant residue. I) Dimeric E. coli WrbA, highlighting the re-face/N(5) loop which includes T78, F80, and W98. II) Tetrameric E. coli WrbA, highlighting T116, H133, and

Y143 at the periphery of the isoalloxazine’s pyrimidine moiety.

152

Purification and characterization of wild type and variant NQO proteins

All proteins were produced and purified as described in Materials and Methods and expression levels varied between proteins. Variants with weaker cofactor affinities generally showed decreased expression, and T116W showed significantly increased expression over all expressed proteins. During heterologous expression, all proteins enhanced cellular FMN content as visualized during cell lysis. Apo-proteins were prepared by extensively washing them before elution from the second Q-sepharose column (pH 7.8) or by overnight dialysis in

4 L of 300 mM sodium phosphate buffer (pH 8.0) with 300 mM NaCl. Purification profiles demonstrated little variance between proteins (except during flavin elution), and contaminate proteins increasingly persisted throughout purifications when heterologous expression was allowed to continue for longer than 4 hours. Protein instabilities were effectively addressed by including 100 mM NaCl and 15% glycerol in experimental and storage buffers.

UV-visible spectra of wild type proteins have been previously reported. Spectra of reconstituted, oxidized variants revealed subtle differences in flavin environments. The absorbance profiles of F80A and T116A showed the greatest perturbations among the alanine variants (Figure 5-2): the F80A flavin environment was less-ordered than wild type, indicated by the decreased fine structure and the hypochromic effect near the 450 nm peak; the T116A flavin environment was more non-polar than wild type, as indicated by a slight bathochromic shift near the 380 nm peak. These features were also present (to a lesser extent) among the remaining N(1)/N(5) variants, except that holo-protein could not be obtained for either F80GG or F80V due to extremely low affinities for FMN. Lastly, T116Y and T116W showed an extended absorbance shoulder above 500 nm, suggesting charge-transfer interactions at the flavin re-face (Figure 5-3). In general, the oxidized spectra were typical of flavoproteins, and the noted perturbations were consistent with previously investigated flavin environments.

153

EcWT F80A T116A

Absorbance

300 400 500

Wavelength (nm)

Figure 5-2. UV-visible absorbance spectra of reconstituted, oxidized wild type E. coli WrbA ( ), F80A ( ), and T116A ( ).

154

EcWT T116Y

T116W

Absorbance

300 400 500 600 Wavelength (nm)

Figure 5-3. UV-visible absorbance spectra of reconstituted, oxidized wild type E. coli WrbA ( ), T116Y ( ), and T116W ( ).

155

Reduction of wild type NQO and alanine variants

Several studies have identified residues affecting redox potential modulation and electron transfer in flavodoxins. This is the first investigation to report redox potentials for flavodoxin-like NQO variants. In this study, potentiometric titrations examined flavin reduction and electron transfer in Type IV NQOs using an alanine variant set derived from E. coli WrbA. Reductive titrations with sodium dithionite were also conducted to qualitatively examine the N(1)/N(5) variants. No visible flavosemiquinone was stabilized in this study under any tested condition, but redox potential and electron transfer perturbations were observed with the alanine variants.

Crystal structures show the active sites of WrbA homologues are composed of residues from three monomers (Figure 5-1). Accordingly, the redox potentials and electron transfer characteristics are expected to vary between the monomer, dimer, and tetramer states. To simplify binding studies, the protein concentrations were held 5-fold greater than

FMN the respective Kd measured at 5 M flavin (KL2 in Table 5-3); by this, the holo-NQO proteins were forced into a tetrameric state during all fluorescence quenching assays, as confirmed by dynamic light scattering (data was similar to Figure 3-4 and Table 3-1). The flavin to monomer ratio for each protein was 1:1, and potentiometric titrations were followed with a fiber-optic spectrometer and variable light path. Additional considerations are described in Materials and Methods. Data was transformed and analyzed as in Figures 5-4 and 5-5.

Potentiometric titrations of the wild type WrbA proteins (E. coli and A. fulgidus) yielded redox potentials similar to those of other flavodoxin-like NQOs (Table 5-1). Redox potentials for the alanine variants of E. coli WrbA spanned a range of nearly 35 mV (Table 5-

2). F80A had the most negative redox potential of all variants characterized (-181 ± 2 mV).

156

The remaining alanine variants had redox potentials more positive than wild type but demonstrated no specific trend. H133A had the most positive redox potential (-148 ± 3 mV).

Lastly, Nernst plots of the data indicated that H133A proceeded through a single two- electron reduction, while all remaining proteins appeared to participate in two sequential, one-electron reductions.

Notably, the electrochemically-reduced proteins differed in absorbance spectra, suggesting variation in molecular orbital configuration between variants. Reduced F80A and

W98A had spectra that resembled fully reduced flavin, and the remaining proteins had absorbance spectra that resembled partially reduced flavin (Figure 5-6). The data were all generally consistent with flavoproteins that do not accumulate the flavin semiquinone.

157

1 I) Reduced 0.8 Oxidized 0.6

0.4

0.2 Fraction Oxidized Oxidized Fraction

0 -400 -300 -200 -100 0 100 mV

II) 1 1

0.5 0.5

0 0 -400 -300 -200 -100 0 100 -0.5 -0.5 for oxidized direction oxidized for for reduced directionreduced for

reduced) % oxidized/ (% Log Log (% reduced/ % oxidized) oxidized) % reduced/ (% Log -1 -1 mV

Reduced Direction y = 0.0139x + 2.3778 R² = 0.9998 Oxidized Direction y = -0.0132x - 2.027 R² = 0.9837

III) Redox Potential: -0.0132x – 2.027 = 0.0139x + 2.3778 -0.0271x = 4.4048 x = -162.5 (mV)

Electrons transferred (from the Nernst equation):

[풐풙풊풅풊풛풆풅] 퐋퐨퐠 n = 58.1 mV • [풓풆풅풖풄풆풅] 푬

0.0271 * 58.1 / 2 = 0.787 (electrons)

Figure 5-4. Data handling for EcWT electrochemical titration. I) Raw data displayed in terms of Fraction Oxidized, II) Transformation and linear fits of the data, and III)

Example calculations for the redox potentials and electrons transferred in wild-type and variant WrbA proteins. 158

I) Reduced 0.8 Oxidized 0.6

0.4

0.2

Oxidized Fraction 0 -400 -300 -200 -100 0 100 mV

II) 1 1

0.5 0.5

0 0 -400 -300 -200 -100 0 100 -0.5 -0.5 for oxidized direction oxidized for for reduceddirectionfor

Log (% oxidized/ % reduced) % oxidized/ (% Log Log (% reduced/ % oxidized) % reduced/(% Log -1 -1 mV

Reduced Direction y = 0.0436x + 6.4856 R² = 0.994

Oxidized Direction y = -0.0447x - 6.3606 R² = 0.9904

III) Redox Potential: 0.0436x – 6.4856 = -0.0447x - 6.3606 0.0883x = -12.8462 x = -145.5 (mV)

Electrons transferred (from the Nernst equation):

[풐풙풊풅풊풛풆풅] 퐋퐨퐠 n = 58.1 mV • [풓풆풅풖풄풆풅] 푬

0.0883 * 58.1 / 2 = 2.57 (electrons)

Figure 5-5. Data handling for H133A electrochemical titration. I) Raw data displayed in terms of Fraction Oxidized, II) Transformation and linear fits of the data, and III) Example calculations for the redox potentials and electrons transferred in wild-type and variant WrbA proteins.

159

E m Species Ref Free FMN -195 ± 3 N/A Lot6p -172 Saccharomyces cerevisiae (Sollner 2006) EcWrbA -172 ± 5 Escherichia coli NQO1 -159 ± 3 Rattus norvegicus (Tedeschi 1994) AfWrbA -158 ± 5 Archaeoglobus fulgidus

Table 5-1. Redox potentials of wild type NQO proteins determined here and in previously published experiments. Em = midpoint redox potentials (mV) at pH 6.5 to 7.0. Unreferenced values were determined here.

160

E m,6.5 G h G FMN n h F80A -181 ± 2 17.5 -1.4 1.19 ± 0.13 EcWrbA -172 ± 5 16.6 -2.2 1.01 ± 0.13 T116A -159 ± 2 15.3 -3.5 0.96 ± 0.07 Y143A -156 ± 4 15.1 -3.8 0.97 ± 0.18 T78A -155 ± 4 14.9 -3.9 0.96 ± 0.07 W98A -152 ± 2 14.7 -4.1 0.84 ± 0.02 H133A -148 ± 3 14.3 -4.5 2.30 ± 0.26

Table 5-2. Redox potentials of the alanine variant NQO proteins, determined by potentiometric titrations performed in triplicate as described in Materials and Methods, Figure 5-4, and Figure 5-5. Em,6.5 = midpoint redox potentials at pH 6.5 (mV). G = kJ/mol. G = kJ/mol with Em,6.5 = -195 mV as the reference. n = number of electrons transferred, as determined by the slope of a Nernst plot.

161

I) II)

Absorbance

380 430 480 530 380 430 480 530 Wavelength (nm) Wavelength (nm)

Figure 5-6. UV-visible absorbance spectra of several variant proteins, oxidized and reduced. I) UV-visible spectra of E. coli WrbA, oxidized ( ) and reduced ( ); and spectra of A. fulgidus WrbA, oxidized ( ) and reduced ( ). II) UV-visible spectra of F80A, oxidized ( ) and reduced ( ); and spectra of W98A, oxidized ( ) and reduced ( ).

162

Flavin affinity and cooperativity of apo-NQO

Numerous articles discuss the tight binding of apo-flavodoxin with its FMN cofactor.

Peptide-flavin interactions that bind NQO cofactor are not well understood, and there is no published investigation of peptide-flavin interactions at the isoalloxazine perimeter. In contrast to classical flavodoxins, FMN is loosely-bound in E. coli WrbA, and its affinity is strongly dependent on ionic strength and pH [2, 68]. Binding of the flavin cofactor imparts significant thermostability and enhances tetramer population, suggesting that there is a relationship between multimerization and flavin content in E. coli WrbA [73, 81]. In an effort to further characterize the apo-holo equilibria for NQO proteins, cofactor binding was investigated with the alanine variant set.

Fluorescence quenching experiments with either 150 M FMN (saturated) or 5 M

FMN (limited) were conducted as described in Materials and Methods. Binding assays indicate the mechanisms of cofactor binding differ between E. coli WrbA and A. fulgidus

WrbA (Figure 5-7). Cofactor binding with apo-WrbA from E. coli was dependent on flavin concentration and demonstrated tighter affinity with saturated flavin. One binding event was observed with either saturated or limited flavin, and the event is protein-dependent and appears to correlate with multimerization. Binding occurred with an apparent cooperativity of

4 for saturated flavin (150 M FMN) and with an apparent cooperativity of 2 for limited flavin (5M FMN). Cofactor binding with apo-WrbA from A. fulgidus had two transitions.

Binding occurred with no cooperativity in the first transition and with an apparent cooperativity of 4 in the second transition, and both transitions were independent of flavin concentration under the tested conditions. A full account of all binding parameters measured is available in Table 5-3. Equations used to fit the data were derived as explained in section

5-5.

163

EcWrbA AfWrbA

Relative Fluorescence Relative

0.01 0.1 1 10

[protein]

Figure 5-7. Fluorescence quenching behavior of E. coli WrbA (EcWrbA) and A. fulgidus WrbA (AfWrbA). x-axis units = M apo-protein added. Equations used to fit the data are described in section 5-5.

164

Saturated FMN Limited FMN Saturated FMN Limited FMN

K S1 K S2 K L1 K L2 G S1 G S2 G L1 G L2 AfWrbA 1 0.10 ± 0.03 4 13.3 ± 0.6 1 0.76 ± 0.08 4 13.6 ± 0.1 -40.0 -27.8 -34.9 -27.8 EcWrbA 4 0.82 ± 0.09 2 16.8 ± 0.4 -34.7 -27.3 T116A 4 0.66 ± 0.08 2 14.8 ± 0.3 -35.3 -27.6 H133A 4 0.62 ± 0.08 2 18.6 ± 0.5 -35.4 -27.0 T78A 1 0.11 ± 0.03 4 0.43 ± 0.04 2 22.3 ± 0.4 -39.7 -36.3 -26.6 Y143A 4 0.25 ± 0.05 2 26.1 ± 0.5 -37.7 -26.2 W98A 2 0.44 ± 0.05 2 28.0 ± 0.4 -36.3 -26.0 F80A 1 0.13 ± 0.03 1 39.4 ± 1.0 -39.3 -25.1

Table 5-3. Binding parameters for alanine variants and wild type proteins, A. fulgidus WrbA

(AfWrbA) and E. coli WrbA (EcWrbA). Superscripts to the left of K values indicate

cooperativity. Where values are blank, no event was observed for the associated transition. All K values are M, all G values are kJ/mol. Saturated FMN = 150 M, limited FMN = 5 M. Equations used to fit the data are described in section 5-5.

165

Importantly, the alanine variants differed in binding affinities and in apparent cooperativities at saturated flavin (150 M FMN). Specifically, T78A, F80A, and W98A showed various departures from wild type cofactor binding (Figure 5-8): F80A demonstrated no cooperativity in binding; W98A demonstrated binding with an apparent cooperativity of

2; and T78A demonstrated binding behavior similar to wild type A. fulgidus WrbA (one transition with no cooperativity and a second transition with an apparent cooperativity of 4).

The remaining variants each had increased affinity over wild type, with an apparent cooperativity of 4.

Binding behavior differed with limited flavin (5 M FMN). For E. coli proteins, addition of apo-protein augmented a new fluorescence shoulder above 600 nm, except with

F80A. This characteristic was not observed at 150 M FMN due to a masking effect by free flavin (Figure 5-9). At 5 M FMN, the new fluorescence shoulder was the only signal associated with binding, except with F80A where quenching assays were performed as with saturated flavin. F80A demonstrated no cooperativity and had the lowest cofactor affinity.

The remaining proteins each demonstrated an apparent cooperativity of 2 and had decreased affinity over wild type (Table 5-3). Binding in A. fulgidus WrbA with limited flavin demonstrated behavior similar to saturated flavin and did not reveal a new fluorescence spectrum.

There were no other identifiable binding events, and all data were consistent with previous reports for wild type NQO proteins. Overall, fluorescence quenching with alanine variants indicate the cooperativity in flavin binding for E. coli WrbA is strong and is dependent on the ligand concentration for all proteins except F80A, which appears to only bind flavin non-cooperatively. Apparent cooperativity in flavin binding for A. fulgidus

WrbA is also strong, but shows no dependence on ligand concentration.

166

EcWrbA W98A T78A F80A

Fluorescence Relative

0.01 0.1 1 [protein]

Figure 5-8. Fluorescence quenching behavior of E. coli

WrbA (EcWrbA), and alanine variants W98A, T78A, and F80A. x-axis units = M apo-protein added. Equations used to fit the data are described in section 5-5.

167

Fluorescence

480 580 680

Wavelength

Figure 5-9. Fluorescence emission spectra of E. coli WrbA (EcWrbA) with limited flavin. This shoulder was also observed with the E. coli variants, except for F80A. INSET: Increase in emission spectra above 600nm. Shoulder begins to emerge at 5 M EcWrbA. FMN concentration is 5 M.

168

Specific Activities of Wild-type and Variant Proteins

The activities of redox-active enzymes are dictated through a variety of mechanisms, including modulation of the cofactor’s redox potential, affinity for the substrates, and specific peptide-substrate interactions. Significant research has examined other Class IV NQOs (for example NQO1) to identify active-site residues that affect activity through substrate interaction. However as already indicated, these studies examined residues that do not directly interact with the isoalloxazine ring. In contrast, the active-site residues of E. coli

WrbA that we investigate directly interact with the isoalloxazine ring (T78, F80, W98, T116,

H133, and Y143). To examine how each of these residues contributes to enzyme activity, we measured specific activities for each alanine variant with various electron acceptors as described in Materials and Methods and as illustrated in Figure 5-10.

As indicated in Table 5-4, all of the alanine variants demonstrate decreased activity with the electron acceptors as compared to wild-type. Enzyme activities with smaller electron acceptors (1,4-benzoquinone and potassium ferricyanide) were slightly less affected among the variants. The general trend of the data indicates that F80A retained the least activity with the electron acceptors, further highlighting the importance of this residue. The general retention of activity with 1,4-anthraquinone over 9,10-anthraquinones suggests that steric hindrance may function in substrate specificity with WrbA proteins. Therefore while

WrbA proteins retain activity with a variety of electron acceptors, the activity appears to be limited by the substrate size and the functional group accessibility.

169

1.3 AfWT1 y = -0.9735x + 1.2549 AfWT2 y = -0.9662x + 1.2514 AfWT3 y = -0.9517x + 1.2243 1.2

1.1

0.9 (340nm) Absorbance

0.8

0.7 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Minutes

Figure 5-10. Example of raw data collected for specific activities with each WrbA protein and each electron acceptor. Activities with varying amounts of A. fulgidus WrbA (AfWT), E. coli WrbA (EcWT), and the E. coli WrbA variants were each performed in triplicate. Oxidation of NADH was measured at 340 nm, and plotted versus time (minutes). Between 0.15 min and 1 min of data were included in a linear fit. These slopes were transformed into specific activities, accounting for volume (400 M), mg protein, pathlength (0.2 mm), and extinction coefficient of

-1 -1 NADH (340 = 6.22 mM cm ). Data was collected aerobically at room temperature (24°C) in 50 mM MOPS pH 7.0, 100 mM NaCl. Data shown was collected across the 0.2 mm pathlength of a quartz fluorimeter cuvette. 800 M NADH and 400 M electron acceptor were incubated and catalytic amount of enzyme was added to initiate the reaction.

170

AfWTa EcWT H133A T78A T116A Y143A W98A F80A 1,4-benzoquinone 4940 (270)b 660 (10) 650 (10) 390 (10) 150 (6) 270 (10) 180 (10) 50 (2) 1,4-naphthoquinone 3000 (30) 510 (10) 80 (2) 70 (1) 70 (3) 60 (2) 60 (1) 10 (0.1) menadione 630 (20) 320 (10) 2.00 (0.04) 20 (0.01) 80 (3) 8.50 (0.2) 7.90 (0.1) 10 (1) potasisum ferricyanide 3970 (60) 180 (10) 180 (4) 150 (10) 140 (10) 140 (7) 110 (4) 59 (2) 2-hydroxy-1,4-naphthoquinone 2.90 (0.2) 3.80 (0.03) N.D. 0.60 (0.01) 1.20 (0.02) 0.10 (0.01) 0.20 (0.02) 1.40 (0.1) 1,4-anthraquinone 380 (10) 7.90 (0.5) 0.80 (0.01) 0.70 (0.1) 0.50 (0.01) 0.70 (0.01) 0.80 (0.01) 1.00 (0.01) 8-hydroxyquinoline 0.30 (0.03) 0.10 (0.01) 0.10 (0.01) 0.10 (0.01) 0.10 (0.01) 0.10 (0.01) 0.20 (0.02) 0.20 (0.02) 9,10-anthraquinone 3.30 (0.4) N.D. 0.20 (0.02) precip.d 0.20 (0.01) 0.20 (0.05) N.D. 0.20 (0.01) 9,10-anthraquinone-2-sulfonate 0.20 (0.01) N.D. N.D. precip. N.D. 0.80 (0.1) N.D. N.D. 9,10-anthraquinone-2,6-disulfonate 0.60 (0.03) 0.60 (0.01) 0.10 (0.01) 0.20 (0.01) 1.60 (0.03) 0.20 (0.01) 0.40 (0.02) 0.10 (0.02) c FMN N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. a Average specific activities (micromoles ∙ min-1 ∙ mg-1) b Standard Error c N.D. = Not Detectable d precip. = precipitation formed on reduction

Table 5-4. Specific activities for alanine variants and wild type proteins, A. fulgidus WrbA (AfWrbA) and E. coli WrbA (EcWrbA). Standard errors are shown in parentheses. All values are micromoles • min-1 mg-1. In all cases, NADH was included at 800 M and electron acceptor was included at 400 M. Data was collected as described in Materials and Methods and Figure 5-10.

171

5.5 Equilibria Derivations

WrbA is a tetrameric protein at saturated receptor concentrations. The protein has been described as having one FMN cofactor per monomer, and it participates in monomer- dimer-tetramer equilibria. Previous investigations observed a multimerization event for apo- protein with a midpoint near 30 M, while the same midpoint for holo-protein shifted to below 5 M. In addition to augmenting multimer population, cofactor binding confers significant thermostability [73]. This is the first study to propose mechanisms of cofactor binding for WrbA proteins.

In an attempt to produce the simplest models for cofactor binding, all of the data were carefully considered and analyzed for construction of proposed equilibria. Cofactor binding in WrbA proteins is complex, with several competing mechanisms that include overlapping apo-holo and monomer-dimer-tetramer equilibria. The techniques used in this study separate ligand-dependence from ligand-binding events by measuring receptor-dependent transitions at saturated and limited flavin concentrations. The resulting data is a substantial step towards understanding the overlapping equilibria and provides significant base for future studies to further develop the multimerization and binding models for WrbA proteins.

The fluorescence of 150 M FMN (ligand) was monitored during titration with apo- proteins (receptor) as presented in Figures 5-7 and 5-8. Increased fluorescence quenching was correlated with increased population of (receptor-ligand) holo-protein. There was no apparent quenching of ligand early in the titrations with wild-type protein, suggesting that monomer doesn’t bind ligand. The only transition for E. coli wild-type WrbA appears centered near 0.9 M and is consistent with a single multimerization event similar to:

4푅 ⇌ 푅4

4 [푅4] 푅 퐾푎 = 4 퐾푑 = 푅 [푅4] 172

However, the data dictate that ligand is incorporated during multimerization. The E. coli wild-type data are consistent with cofactor being bound by multimeric WrbA with high affinity and appears concomitant with the apparent multimerization event. Other equilibria for A. fulgidus wild-type and the E. coli variant proteins (T78A, F80A, and W98A) demonstrate binding events with distinct characteristics.

Under saturated conditions, ligand binding in wild-type E. coli WrbA appears to occur in a single transition, concurrent with tetramerization (bold-faced type indicates the observable transition):

푅 + 푅 ⇌ 퐷; ퟐ푫 + ퟒ푳 ⇌ 푻푳ퟒ

By this, apo-protein dimerizes and affords the opportunity to bind ligand, which encourages tetramerization without accumulation of dimer. As Figure 5-8 indicates, T78A variant appears to permit binding to the monomer, but the variant also enables binding behavior similar to wild-type. Thus, the data are consistent with two competing mechanisms of cofactor binding, one of which becomes cooperative at increased protein concentration:

푹 + 푳 ⇌ 푹푳; 푅퐿 + 푅퐿 ⇌ 퐷퐿2; 2퐷퐿2 ⇌ 푇퐿4

푅 + 푅 ⇌ 퐷; ퟐ푫 + ퟒ푳 ⇌ 푻푳ퟒ

In this case, the monomer binds ligand in an uncooperative manner, but at sufficient protein concentration, the cooperative mechanism out-competes binding by the monomer. Unlike

E. coli wild-type WrbA and T78A, theW98A variant appears to proceed through a binding mechanism involving a protein-dependent transition, concurrent with dimerization (rather than tetramer formation):

푅 + 푅 ⇌ 퐷; 푫 + ퟐ푳 ⇌ 푫푳ퟐ; 2퐷퐿2 ⇌ 푇퐿4

This suggests that the W98 residue is responsible for communicating cooperativity in binding ligand during tetramer formation. Finally, the F80A variant doesn’t appear to bind cofactor with any cooperativity despite the observed transition from monomer to tetramer.

173

Accordingly, the observed transition suggests independent binding sites which are distinct from multimerization:

푹 + 푳 ⇌ 푹푳; 푅퐿 + 푅퐿 ⇌ 퐷퐿2; 2퐷퐿2 ⇌ 푇퐿4 푅 + 푅 ⇌ 퐷; 2퐷 ⇌ 푇

Under conditions of limited flavin, binding occurs via a different route for all proteins except F80A, which proceeds as with saturated flavin. For the remaining proteins, the observed transition incurs a new fluorescence that peak appears during the dimer-tetramer transition, with all flavin bound in a dimer near the beginning of the transition. The models at saturated ligand incorporate a ligand-independent multimerization, so dimerization of apo- protein occurs before the observed transition. Accordingly, the proposed model for limited flavin incorporates a “labeled” dimer (with one cofactor per dimer) from the start of the transition, coupling with an apo-dimer during the observed transition:

∗ ∗ ∗ ∗ 푅 + 푅 ⇌ 푅2; 푅 + 푅 ⇌ 퐷 ; 푫 + 푹ퟐ ⇌ 푹ퟒ

In this model, the observed fluorescence peak appears during the dimer-tetramer transition either by direct interaction to the cofactor or by flavin-peptide rearrangements through the dimer.

As discussed, the proposed equilibria offer simple models which all have similar modes of multimerization and cofactor binding. We feel that these equilibria present reasonable models for the WrbA proteins; however it is also possible that ligand binds concurrent with dimer formation. The models we present offer a macroscopic view of the possible apo-holo and monomer-dimer-tetramer equilibria, but significant research is necessary to further delineate other models, the microscopic ligand effects on multimerization, and additional subtleties to the mechanisms.

174

Derivation for wild-type E. coli WrbA:

푅 + 푅 ⇌ 퐷; ퟐ푫 + ퟒ푳 ⇌ 푻푳ퟒ

[퐷] 퐾 = 푎1 [푅]2

[푇퐿4] 퐾 = 푎2 [퐷]2[퐿]4

[푇퐿4] 퐵 = 퐿 퐿 + [푇퐿 ] 4 2 4 퐾푎2[퐷] [퐿] 퐵 = 퐿 퐿 + 퐾 [퐷]2[퐿]4 푎2 4 4 퐾푎2퐾푎1[푅] [퐿] 퐵 = 퐿 퐿 + 퐾 퐾 [푅]4[퐿]4 푎2 푎1 4 3 퐾푎2퐾푎1[푅] [퐿] 퐵 = 퐿 1 + 퐾 퐾 [푅]4[퐿]3 푎2 푎1 [푅]4[퐿]3 퐵 = 퐿 (퐾푎푝푝 )2 + [푅]4[퐿]3 푑 [푅]4[퐿]3 퐵 = 퐿 (퐾푎푝푝 )2 + [푅]4[퐿]3 푑 [푅]4 퐵 = 퐿 푎푝푝 2 4 (퐾푑 ) + [푅]

Explanation: The proposed equation describes the fraction of ligand incorporated during tetramer formation from apo-dimers. Other models were considered, but this was the most descriptive model offering the fewest variables. Future studies will need to delineate the order of ligand binding. At saturated ligand, the binding of ligand becomes independent of ligand concentration, and the model enables the assumption that apparent ligand binding is only receptor dependent. In Figures 5-7 and 5-8, the data are presented in terms of unquenched ligand versus receptor. Since the model doesn’t account for quenched ligand after multimerization, an additional term NS ∙ 푅 4 is included in the fit to incorporate ligand quenched beyond the dimer-tetramer transition.

175

Derivation for the T78A variant:

푹 + 푳 ⇌ 푹푳; 푅퐿 + 푅퐿 ⇌ 퐷퐿2; 2퐷퐿2 ⇌ 푇퐿4 푅 + 푅 ⇌ 퐷; ퟐ푫 + ퟒ푳 ⇌ 푻푳ퟒ

[푅퐿] 퐾 = 푎1 [푅][퐿]

[퐷] 퐾 = 푎2 [푅]2

[푇퐿4] 퐾 = 푎3 [퐷]2[퐿]4

[푅퐿] [푇퐿4] 퐵 = + 퐿 퐿 + [푅퐿] 퐿 + [푇퐿 ] 4 2 4 퐾푎1 푅 퐿 퐾푎3 퐷 퐿 퐵 = + 퐿 퐿 + 퐾 푅 퐿 퐿 + 퐾 퐷 2 퐿 4 푎1 푎3 4 4 퐾푎1 푅 퐿 퐾푎3퐾푎2[푅] 퐿 퐵 = + 퐿 퐿 + 퐾 푅 퐿 퐿 + 퐾 퐾 [푅]4 퐿 4 푎1 푎3 푎2 4 3 퐾푎1[푅] 퐾푎3퐾푎2[푅] 퐿 퐵 = + 퐿 1 + 퐾 [푅] 1 + 퐾 퐾 [푅]4 퐿 3 푎1 푎3 푎2 [푅] [푅]4 퐿 3 퐵 = + 퐿 푎푝푝 2 4 3 퐾푑1 + [푅] (퐾 ) + [푅] 퐿 푑 [푅] [푅]4 퐿 3 퐵 = + 퐿 푎푝푝 2 4 3 퐾푑1 + [푅] (퐾 ) + [푅] 퐿 푑 [푅] [푅]4 퐵 = + 퐿 푎푝푝 2 4 퐾푑1 + [푅] (퐾푑 ) + [푅]

Explanation: The proposed equation describes the fraction of ligand incorporated by monomer and then during tetramer formation. Other models were considered, but this was the most descriptive model offering the fewest variables. Future studies will need to delineate the order of ligand binding. At saturated ligand, the model enables the assumption that apparent ligand binding is only receptor dependent. In Figure 5-8, the data are presented in terms of unquenched ligand versus receptor. Since the model doesn’t account for quenched ligand after multimerization, an additional term NS ∙ 푅 4 is included in the fit to incorporate ligand quenched beyond the dimer-tetramer transition.

176

Derivation for the W98A variant:

푅 + 푅 ⇌ 퐷; 푫 + ퟐ푳 ⇌ 푫푳ퟐ; 2퐷퐿2 ⇌ 푇퐿4

[퐷] 퐾 = 푎1 [푅]2

[퐷퐿2] 퐾 = 푎2 [퐷][퐿]2

[퐷퐿2] 퐵 = 퐿 퐿 + [퐷퐿 ] 2 2 퐾푎2[퐷][퐿] 퐵 = 퐿 퐿 + 퐾 [퐷][퐿]2 푎2 2 2 퐾푎2퐾푎1[푅] [퐿] 퐵 = 퐿 퐿 + 퐾 퐾 [푅]2[퐿]2 푎2 푎1 2 퐾푎2퐾푎1[푅] [퐿] 퐵 = 퐿 1 + 퐾 퐾 [푅]2[퐿] 푎2 푎1 [푅]2[퐿] 퐵 = 퐿 (퐾푎푝푝 )2 + [푅]2[퐿] 푑 [푅]2 퐵 = 퐿 푎푝푝 2 2 (퐾푑 ) + [푅]

Explanation: The proposed equation describes the fraction of ligand incorporated during multimerization. At saturated ligand, the model enables the assumption that apparent ligand binding is only receptor dependent. The model doesn’t account for tetrameric binding since the transition apparently doesn’t involve ligand, thus further studies will need to examine the

W98A dimer-tetramer transition in more detail. In Figure 5-8, the data are presented in terms of unquenched ligand versus receptor. Since the model doesn’t account for ligand quenched by tetrameric W98A, an additional term NS ∙ 푅 2 is included in the fit to incorporate ligand quenched beyond the dimer transition. Replacing this term with NS ∙ 푅 4 gave a similar result.

177

Derivation for the F80A variant:

푹 + 푳 ⇌ 푹푳; 푅퐿 + 푅퐿 ⇌ 퐷퐿2; 2퐷퐿2 ⇌ 푇퐿4

[푅퐿] 퐾 = 푎1 [푅][퐿]

[푅퐿] 퐵 = 퐿 퐿 + [푅퐿]

퐾푎1 푅 퐿 퐵 = 퐿 퐿 + 퐾 푅 퐿 푎1 퐾푎1 푅 퐿 퐵 = 퐿 퐿 + 퐾 푅 퐿 푎1 퐾푎1 푅 퐵 = 퐿 1 + 퐾 푅 푎1 푅 퐵 = 퐿 퐾 + 푅 푑1 푅 퐵 = 퐿 퐾 + 푅 푑1 [푅] 퐵 = 퐿 퐾 + [푅] 푑1

Explanation: The proposed equation describes the fraction of ligand incorporated by monomeric receptor. The model doesn’t account for dimeric binding since there was no apparent transition observable under the reported conditions, although apparently dimer begins to accumulate over the range reported in Figure 5-8. Further studies will need to examine F80A dimer formation, its ligand affinities, and several possible binding transitions.

In Figure 5-8, the data are presented in terms of unquenched ligand versus receptor. Since the model doesn’t account for quenched ligand by F80A multimers, an additional term

NS ∙ 푅 is included in the fit to account for ligand quenched by the multimers which appear

app to accumulate. Replacing NS ∙ 푅 with a term for dimer accumulation (Kd ~5 M) gave similar results.

178

Derivation for the apparent tetramerization with one “labeled” dimer and one apo-dimer:

∗ ∗ 푅 + 푅 ⇌ 푅2; 푫 + 푹ퟐ ⇌ 푹ퟒ

푅2 퐾 = 푎1 푅 2

∗ [푅4] 퐾 = 푎2 [푅 ][퐷∗] 2 ∗ [푅4] 퐵 ∗ = 퐷 퐷∗ + [푅∗] 4 ∗ 퐾푎2 푅2 [퐷 ] 퐵 ∗ = 퐷 퐷∗ + 퐾 푅 [퐷∗] 푎2 2 2 ∗ 퐾푎2퐾푎1 푅 [퐷 ] 퐵 ∗ = 퐷 퐷∗ + 퐾 퐾 푅 2[퐷∗] 푎2 푎1 2 퐾푎2퐾푎1 푅 퐵 ∗ = 퐷 1 + 퐾 퐾 푅 2 푎2 푎1 푅 2 퐵 ∗ = 퐷 푎푝푝 2 2 (퐾푑 ) + 푅

Explanation: The proposed equation describes the fraction of ligand incorporated during the dimer-tetramer transition. Other models were considered, but this was the most descriptive model offering the fewest variables. Future studies will need to delineate the order of ligand binding. At the limited ligand concentration used in this study, the model enables the assumption that all ligand is bound at the start of the experiment, so the measured transition is only receptor dependent. Since this experiment measured the appearance of a peak over tetramer formation, no additional term was included as was done for the other data fits.

179

5.6 Discussion

The experiments presented here are the first to explicate a relationship between cofactor binding and multimerization in flavodoxin-like NAD(P)H reductases (NQOs).

Further, this is also the first investigation to detail the modulation of redox potentials and electron transfer within NQOs. Collectively, the results offer significant insight regarding residues interacting with the flavin cofactor.

The “re-face/N(5) loop”

FMN The flavodoxin cofactor is tightly bound (Kd = <1 M) and the binding mechanisms are well investigated. The isoalloxazine interactions in oxidized flavodoxin largely involve aromatic residues which sandwich the ring. In flavodoxin-like E. coli WrbA, there are no aromatic interactions at the isoalloxazine faces, suggesting the loosely-bound cofactor has distinct binding mechanisms. As with flavodoxins, the isoalloxazine interactions in the re-face/N(5) loop are important, but in WrbA they confer unique cooperativity in cofactor binding. Further, the novel fluorescence spectrum we observed suggests the architecture of the E. coli WrbA active site has unique peptide-flavin interactions that depend on the re-face/N(5) loop.

Our experiments indicate that residue F80, which interacts with N(5), is essential for flavin binding. F80A was the only variant for which the apparent cooperativity was abrogated. This variant enables monomeric WrbA to bind flavin, as evidence by a non- cooperative binding event with increased affinity. It could be reasoned the variant flavin- binding loop permits a novel arrangement that allows the monomer to bind flavin but prohibits apparent cooperative effects near N(5). Potentiometric data show that F80A was

180 also the only variant with a redox potential more negative than wild type, which suggests the

F80A flavin environment more closely resembles free-flavin.

According to data in Table 5-3, F80 appears to collaborate with nearby residues to coordinate flavin. The binding studies suggest that W98 plays a role ancillary to F80. The

W98A variant has apparent decreased cooperativity in flavin binding (Figure 5-8), while F80 is essential for commuting cooperativity. However, W98 is contributed to the active site from an opposing monomer and has no direct interaction with the isoalloxazine ring. Therefore,

F80 appears to provide the primary cooperativity and communicate with W98 through the dimer interface in order to commute the additional cooperativity. A reasonable mechanism to explain this cooperativity is that they collaboratively orient the conserved helix which links them in primary structure and which contributes to the dimer interface (Figure 5-11).

In addition to binding, both of these residues affect reduction of the isoalloxazine ring. Upon reduction, absorbance spectra of W98A and F80A resemble fully reduced flavin, which was demonstrated for neither wild type nor the remaining variant proteins under the tested conditions (Figure 5-6). These spectra suggest the cofactors W98A and F80A have distinct molecular orbital arrangements, and their redox potentials indicate that this reduced state is energetically more favorable for W98A than for F80A.

Another residue that affects flavin binding is T78. The T78A variant was specifically designed to mimic interactions with the re-face of the flavodoxin cofactor, which binds to monomeric flavodoxin with significant affinity. As depicted in Figure 5-8 and Table 5-3, interactions between T78 and the flavin direct an early binding event (KS1) without significantly affecting the primary binding event (KS2). This supports that T78A introduces structural arrangements allowing monomeric WrbA to bind flavin without affecting

181 cooperative communications between F80 and W98, whereas wild type protein altogether discourages the early binding event.

A comparison of crystal structures from wild type E. coli WrbA and flavodoxin

5NLL offers an explanation for the early binding event (Figure 5-12). In E. coli WrbA, the hydroxyl group of T78 interacts with the peptide backbone at G81. This interaction appears to orient T78 different from the equivalent residue (A55) in flavodoxin 5NLL. The orientation of T78 could discourage peptide-flavin interactions in the monomer. Introducing volume at the re-face of the flavin in T78A may direct the flavin to adopt conformation that has increased affinity. Given that T78A did not significantly affect the second binding event, it is unlikely that T78A significantly affects the overall topology of the loop.

In summary, binding studies with T78, F80, and W98 were particularly instructive about flavin binding in NQO proteins. Each of these residues affects binding behavior, and they are all located in the third “sheet-turn-helix” motif that was previously identified as a highly conserved region across the Type IV NQO proteins (Chapter 4). However, given that the binding mechanisms differ between wild type E. coli WrbA and A. fulgidus WrbA, future studies may elucidate various binding mechanisms across Type IV NQOs. The detailed binding mechanism in A. fulgidus WrbA remains to be explained.

182

Monomer A

Monomer B

Figure 5-11. Crystal structure of E. coli WrbA dimer. The orange and blue turn-helix motifs have significant sequence conservation across NQOs, contribute to the dimer interface, and contain T78, F80, and W98.

183

I) T78 II) A55

Gly118

Figure 5-12. Crystal structures of I) the E. coli WrbA active site and of II) the 5NLL active site.

184

Position O(4) of the isoalloxazine ring

Flavodoxin modulates its flavin cofactor to preferentially stabilize the semiquinone intermediate. In contrast to this, Type IV (flavodoxin-like) NQOs cycle between the quinone and hydroquinone states. A two-electron mechanism in these proteins is thought to protect cells from redox-cycling quinones and from reactive oxygen species that are destructive to proteins, lipids, and DNA. Additionally, several NQOs demonstrate redox-linked interactions, and their reduced flavin is essential for protecting their protein targets from proteasomal degradation. Elucidating the reduction mechanisms will enhance to our ability to investigate possible xenobiotic or protein-protein interactions with the reduced proteins.

In this investigation, we demonstrate that wild type WrbA is reduced via two sequential, one-electron reductions without accumulation of a visible semiquinone. This is in agreement with previous studies which demonstrate that K3Fe(CN)6 oxidizes WrbA homologues. The evidence supports that wild type WrbA cycles through a semiquinone intermediate rather than being obligated to participate in a hydride transfer. It should be noted that characterized flavodoxin-like NQOs prefer two-electron processes, but this does not exclude one-electron processes. Proteins from several Type IV NQO subclasses

(including NQO1) are reported to participate in one-, two-, and even four-electron processes through serial reduction [82-84]. In fact, while NQO1 doesn’t accumulate its flavin semiquinone at pH 6.5, the protein was observed to form a transient red semiquinone at pH

7.0 and it accumulates at pH 8.0 [85, 86]. It remains to be seen whether WrbA exhibits pH- dependence in accumulating the semiquinone.

Peptide interactions with N(5) of the isoalloxazine ring are essential for stabilizing the one-electron, reduced flavodoxin cofactor. Parallel to this is the interaction between O(4) and H133 in E. coli WrbA, which directs electron accumulation on its flavin cofactor. While

185 wild type protein cycles through two sequential one-electron reductions, potentiometric titrations suggest that H133A is reduced via a single two-electron reduction. This behavior was not observed for any other variant, highlighting that the interaction between H133 and

O(4) is essential for modulating the WrbA cofactor. This interaction appears in all available holo-WrbA crystal structures. A similar histidine-flavin interaction appears in crystal structures for other NQOs but not for Class I NQOs, suggesting there are multiple mechanisms employed for modulating one-electron transfers in Type IV NQOs.

186

Implications for flavodoxin-like NQOs

Accumulating evidence suggests that the protein-protein interactions observed with

NQOs require reduced holo-dimers. The data we present indicate WrbA proteins (and perhaps other NQOs) prefer two sequential one-electron reductions. However, the essential interaction with H133 is not uniformly present across Type IV NQOs, signifying there are different mechanisms for flavin modulation. In addition to these findings, the apparent cooperativity in flavin-binding elucidated here will be useful for investigating protein-protein interactions with NQOs, although binding mechanisms may differ across the NQO subclasses

(as they do between E. coli WrbA and A. fulgidus WrbA). Accordingly, conditions that permit protein-protein interactions with NQO proteins may vary significantly.

It is important to note that the binding constants presented here suggest the WrbA-

TrpR interaction first presented by Yang, Ni, and Somerville (1993) should be re-examined.

Previous investigations of this interaction likely did not contain WrbA concentrations sufficient for holo-protein formation. Further, the putative protein-protein interactions with

WrbA need to be investigated for redox-dependence.

Finally, it is well established that NQO1 proteins function in cancers contributes to blood coagulation, but the specific mechanisms are not well understood. Articles highlight that these proteins prefer two-electron transfers but are also able to participate in one-electron transfers, and anti-cancer compounds can be activated by one-electron or two-electron mechanisms. More detailed investigations of NQO proteins could provide significant opportunity to enhance in vivo treatments, either through uniquely designed anti-cancer and coagulation compounds or through more efficient NQO variants.

187

5.7 Acknowledgements

We would like to thank Sarah Ades, Joe Palladino, Jennifer Hayden, Mary

Laubacher, Sarah Barchinger for providing materials for E. coli knockouts (Pennsylvania

State University). Andreas Christenson, Gilbert Nöll, and Tobias Gustavsson were very helpful and provided training and direction in techniques and design of the electrochemical cell (Kemicentrum, Lund University). Ken Biddle, Bill Diehl, and Bill Genet were essential for construction of the electrochemical cell (Earth and Mineral Sciences Machine Shop,

Pennsylvania State University). Christine Keating, Stacey Dean, Sarah Brunker, Barbara

Shaw, Rob Kreuter, Robert Crable, James Miller, Howard Pickering, and Eric Sagmuller assisted greatly with providing or fixing potentiostat equipment (Pennsylvania State

University).

188

5.8 References

1. Yang, W., L. Ni, and R.L. Somerville (1993). A stationary-phase protein of Escherichia coli that affects the mode of association between the trp repressor protein and operator-bearing DNA. Proc Natl Acad Sci U S A. 90(12): p. 5796-800.

2. Grandori, R., et al. (1998). Biochemical characterization of WrbA, founding member of a new family of multimeric flavodoxin-like proteins. J Biol Chem. 273(33): p. 20960-6.

3. Kang, Y.S., et al. (2005). Genome-wide expression analysis indicates that FNR of Escherichia coli K-12 regulates a large number of genes of unknown function. Journal of Bacteriology. 187(3): p. 1135-1160.

4. Liu, X. and P. De Wulf (2004). Probing the ArcA-P modulon of Escherichia coli by whole genome transcriptional analysis and sequence recognition profiling. J Biol Chem. 279(13): p. 12588-97.

5. Chang, D.E., D.J. Smalley, and T. Conway (2002). Gene expression profiling of Escherichia coli growth transitions: an expanded stringent response model. Mol Microbiol. 45(2): p. 289-306.

6. Cheung, K.J., et al. (2003). A Microarray-Based Antibiotic Screen Identifies a Regulatory Role for Supercoiling in the Osmotic Stress Response of Escherichia coli. Genome Res. 13(2): p. 206-215.

7. Ishihama, A. (2000). Functional Modulation of Escherichia coli RNA Polymerase. Annu. Rev. Micrbiol. 54(1): p. 499-518.

8. Kirkpatrick, C., et al. (2001). Acetate and Formate Stress: Opposite Responses in the Proteome of Escherichia coli. J. Bacteriol. 183(21): p. 6466-6477.

11. Tao, H., et al. (1999). Functional Genomics: Expression Analysis of Escherichia coli Growing on Minimal and Rich Media. J. Bacteriol. 181(20): p. 6425-411.

12. Tucker, D.L., N. Tucker, and T. Conway (2002). Gene Expression Profiling of the pH Response in Escherichia coli. J. Bacteriol. 184(23): p. 6551-1100.

13. Gonzalez, C.F., et al. (2005). ChrR, a soluble quinone reductase of Pseudomonas putida that defends against H2O2. J Biol Chem. 280(24): p. 22590-5.

189

14. Liu, G., et al. (2008). Enhancing survival of Escherichia coli by expression of azoreductase AZR possessing quinone reductase activity. Appl Microbiol Biotechnol. 80(3): p. 409-16.

15. Sollner, S., et al. (2007). Lot6p from Saccharomyces cerevisiae is a FMN-dependent reductase with a potential role in quinone detoxification. FEBS J. 274(5): p. 1328-39.

16. Butcher, B.G., S.M. Deane, and D.E. Rawlings (2000). The chromosomal arsenic resistance genes of Thiobacillus ferrooxidans have an unusual arrangement and confer increased arsenic and antimony resistance to Escherichia coli. Appl Environ Microbiol. 66(5): p. 1826-33.

17. Neyt, C., et al. (1997). Virulence and arsenic resistance in Yersiniae. J Bacteriol. 179(3): p. 612-9.

18. Hong, Y., G. Wang, and R.J. Maier (2008). The NADPH quinone reductase MdaB confers oxidative stress resistance to Helicobacter hepaticus. Microb Pathog. 44(2): p. 169-74.

19. Wang, G. and R.J. Maier (2004). An NADPH quinone reductase of Helicobacter pylori plays an important role in oxidative stress resistance and host colonization. Infect Immun. 72(3): p. 1391-6.

20. Chatterjee, P.K. and N.L. Sternberg (1995). A general genetic approach in Escherichia coli for determining the mechanism(s) of action of tumoricidal agents: application to DMP 840, a tumoricidal agent. Proc Natl Acad Sci U S A. 92(19): p. 8950-4.

21. Kim, S., et al. (2009). Comparison of the effects of NADH- and NADPH-perturbation stresses on the growth of Escherichia coli. Curr Microbiol. 58(2): p. 159-63.

22. Anwar, A., et al. (2003). Interaction of human NAD(P)H:quinone oxidoreductase 1 (NQO1) with the tumor suppressor protein p53 in cells and cell-free systems. J Biol Chem. 278(12): p. 10368-73.

23. Asher, G., et al. (2002). Mdm-2 and ubiquitin-independent p53 proteasomal degradation regulated by NQO1. Proc Natl Acad Sci U S A. 99(20): p. 13125-30.

24. Sollner, S., et al. (2009). Quinone reductase acts as a redox switch of the 20S yeast proteasome. EMBO Rep. 10(1): p. 65-70.

25. Naresh, A., S. Saini, and J. Singh (2003). Identification of Uhp1, a ubiquitinated histone-like protein, as a target/mediator of Rhp6 in mating-type silencing in fission yeast. J Biol Chem. 278(11): p. 9185-94.

26. Asher, G., et al. (2005). 20S proteasomal degradation of ornithine decarboxylase is regulated by NQO1. Mol Cell. 17(5): p. 645-55.

190

27. Grandori, R. and J. Carey (1994). Six new candidate members of the alpha/beta twisted open-sheet family detected by sequence similarity to flavodoxin. Protein Sci. 3(12): p. 2185-93.

28. Maldonado, S., et al. (1998). Apoflavodoxin: structure, stability, and FMN binding. Biochimie. 80(10): p. 813-20.

29. Lostao, A., et al. (2000). Dissecting the energetics of the apoflavodoxin-FMN complex. J Biol Chem. 275(13): p. 9518-26.

30. Lostao, A., et al. (2003). How FMN binds to Anabaena apoflavodoxin - A hydrophobic encounter at an open binding site. Journal of Biological Chemistry. 278(26): p. 24053-24061.

31. Ludwig, M.L., et al. (1997). Control of oxidation-reduction potentials in flavodoxin from Clostridium beijerinckii: the role of conformation changes. Biochemistry. 36(6): p. 1259-80.

32. Müller, F., Chemistry and biochemistry of flavoenzymes. (1991), Boca Raton: CRC Press. v. <1-3 >.

33. Simondsen, R.P. and G. Tollin (1980). Structure-function relations in flavodoxins. Mol Cell Biochem. 33(1-2): p. 13-24.

34. Mayhew, S.G., et al. (1996). Regulation of the redox potentials of flavodoxins: modification of the flavin binding. Biochem Soc Trans. 24(1): p. 122-7.

35. Curley, G.P., et al. (1991). Redox and Flavin-Binding Properties of Recombinant Flavodoxin from Desulfovibrio vulgaris (Hildenborough). European Journal of Biochemistry. 202(3): p. 1091-1100.

36. Yalloway, G.N., et al. (2003). Letter to the Editor: H-1, C-13 and N-15 assignment of the hydroquinone form of flavodoxin from Desulfovibrio vulgaris (Hildenborough) and comparison of the chemical shift differences with respect to the oxidized state. Journal of Biomolecular Nmr. 25(3): p. 257-258.

37. Freigang, J., et al. (2002). Crystal structure of oxidized flavodoxin, an essential protein in Helicobacter pylori. Protein Science. 11(2): p. 253-261.

38. Ishikita, H. (2008). Redox potential difference between Desulfovibrio vulgaris and Clostridium beijerinckii flavodoxins. Biochemistry. 47(15): p. 4394-402.

39. Alagaratnam, S., et al. (2005). A crystallographic study of Cys69Ala flavodoxin II from Azotobacter vinelandii: structural determinants of redox potential. Protein Sci. 14(9): p. 2284-95.

191

40. Ishikita, H. (2007). Influence of the protein environment on the redox potentials of flavodoxins from Clostridium beijerinckii. J Biol Chem. 282(35): p. 25240-6.

41. Artali, R., et al. (2005). Structure of S35C flavodoxin mutant from Desulfovibrio vulgaris in the semiquinone state. Acta Crystallogr D Biol Crystallogr. 61(Pt 4): p. 481-4.

42. Frago, S., et al. (2007). Tuning of the FMN binding and oxido-reduction properties by neighboring side chains in Anabaena flavodoxin. Arch Biochem Biophys. 467(2): p. 206-17.

43. Hoover, D.M., et al. (1999). Comparisons of wild-type and mutant flavodoxins from Anacystis nidulans. Structural determinants of the redox potentials. J Mol Biol. 294(3): p. 725-43.

44. Chang, F.C. and R.P. Swenson (1999). The midpoint potentials for the oxidized- semiquinone couple for Gly57 mutants of the Clostridium beijerinckii flavodoxin correlate with changes in the hydrogen-bonding interaction with the proton on N(5) of the reduced flavin mononucleotide cofactor as measured by NMR chemical shift temperature dependencies. Biochemistry. 38(22): p. 7168-76.

45. Bagby, S., et al. (1991). Direct electrochemistry of two genetically distinct flavodoxins isolated from Azotobacter chroococcum grown under nitrogen-fixing conditions. Biochem J. 277 ( Pt 2): p. 313-9.

46. Feng, Y.C. and R.P. Swenson (1997). Evaluation of the role of specific acidic amino acid residues in electron transfer between the flavodoxin and cytochrome c(3) from Desulfovibrio vulgaris [Hildenborough]. Biochemistry. 36(44): p. 13617-13628.

47. Peleato, M.L., et al. (1994). Isolation and characterization of two different flavodoxins from the eukaryote Chlorella fusca. Biochem J. 302 ( Pt 3): p. 807-11.

48. Vervoort, J., et al. (1994). Flavodoxins. Methods Enzymol. 243: p. 188-203.

49. Zhou, Z. and R.P. Swenson (1995). Electrostatic effects of surface acidic amino acid residues on the oxidation-reduction potentials of the flavodoxin from Desulfovibrio vulgaris (Hildenborough). Biochemistry. 34(10): p. 3183-92.

50. Kasim, M. and R.P. Swenson (2000). Conformational energetics of a reverse turn in the Clostridium beijerinckii flavodoxin is directly coupled to the modulation of its oxidation-reduction potentials. Biochemistry. 39(50): p. 15322-32.

51. Druhan, L.J. and R.P. Swenson (1998). Role of methionine 56 in the control of the oxidation-reduction potentials of the Clostridium beijerinckii flavodoxin: effects of substitutions by aliphatic amino acids and evidence for a role of sulfur-flavin interactions. Biochemistry. 37(27): p. 9668-78.

192

52. Nogues, I., et al. (2004). Role of neighboring FMN side chains in the modulation of flavin reduction potentials and in the energetics of the FMN:apoprotein interaction in Anabaena flavodoxin. Biochemistry. 43(48): p. 15111-21.

53. Kasim, M. and R.P. Swenson (2001). Alanine-scanning of the 50's loop in the Clostridium beijerinckii flavodoxin: evaluation of additivity and the importance of interactions provided by the main chain in the modulation of the oxidation-reduction potentials. Biochemistry. 40(45): p. 13548-55.

54. O'Farrell, P.A., et al. (1998). Modulation of the redox potentials of FMN in Desulfovibrio vulgaris flavodoxin: thermodynamic properties and crystal structures of glycine-61 mutants. Biochemistry. 37(23): p. 8405-16.

55. Swenson, R.P. and G.D. Krey (1994). Site-directed mutagenesis of tyrosine-98 in the flavodoxin from Desulfovibrio vulgaris (Hildenborough): regulation of oxidation- reduction properties of the bound FMN cofactor by aromatic, solvent, and electrostatic interactions. Biochemistry. 33(28): p. 8505-14.

56. Stockman, B.J., T.E. Richardson, and R.P. Swenson (1994). Structural changes caused by site-directed mutagenesis of tyrosine-98 in Desulfovibrio vulgaris flavodoxin delineated by 1H and 15N NMR spectroscopy: implications for redox potential modulation. Biochemistry. 33(51): p. 15298-308.

57. Reynolds, R.A., W. Watt, and K.D. Watenpaugh (2001). Structures and comparison of the Y98H (2.0 A) and Y98W (1.5 A) mutants of flavodoxin (Desulfovibrio vulgaris). Acta Crystallogr D Biol Crystallogr. 57(Pt 4): p. 527-35.

58. Chang, F.C. and R.P. Swenson (1997). Regulation of oxidation-reduction potentials through redox-linked ionization in the Y98H mutant of the Desulfovibrio vulgaris [Hildenborough] flavodoxin: direct proton nuclear magnetic resonance spectroscopic evidence for the redox-dependent shift in the pKa of Histidine-98. Biochemistry. 36(29): p. 9013-21.

59. McCarthy, A.A., et al. (2002). Crystallographic investigation of the role of aspartate 95 in the modulation of the redox potentials of Desulfovibrio vulgaris flavodoxin. Biochemistry. 41(36): p. 10950-62.

60. Zhou, Z. and R.P. Swenson (1996). The cumulative electrostatic effect of aromatic stacking interactions and the negative electrostatic environment of the flavin mononucleotide binding site is a major determinant of the reduction potential for the flavodoxin from Desulfovibrio vulgaris [Hildenborough]. Biochemistry. 35(50): p. 15980-8.

61. Geoghegan, S.M., et al. (2000). Cloning, sequencing and expression of the gene for flavodoxin from Megasphaera elsdenii and the effects of removing the protein negative charge that is closest to N(1) of the bound FMN. Eur J Biochem. 267(14): p. 4434-44.

193

62. Lostao, A., et al. (1997). Differential stabilization of the three FMN redox forms by tyrosine 94 and tryptophan 57 in flavodoxin from Anabaena and its influence on the redox potentials. Biochemistry. 36(47): p. 14334-44.

63. Bradley, L.H. and R.P. Swenson (1999). Role of glutamate-59 hydrogen bonded to N(3)H of the flavin mononucleotide cofactor in the modulation of the redox potentials of the Clostridium beijerinckii flavodoxin. Glutamate-59 is not responsible for the pH dependency but contributes to the stabilization of the flavin semiquinone. Biochemistry. 38(38): p. 12377-86.

64. Bradley, L.H. and R.P. Swenson (2001). Role of hydrogen bonding interactions to N(3)H of the flavin mononucleotide cofactor in the modulation of the redox potentials of the Clostridium beijerinckii flavodoxin. Biochemistry. 40(30): p. 8686-95.

65. Asher, G., et al. (2006). The crystal structure of NAD(P)H quinone oxidoreductase 1 in complex with its potent inhibitor dicoumarol. Biochemistry. 45(20): p. 6372-8.

66. Adams, M.A. and Z. Jia (2006). Modulator of drug activity B from Escherichia coli: crystal structure of a prokaryotic homologue of DT-diaphorase. J Mol Biol. 359(2): p. 455-65.

67. Li, R., et al. (1995). The three-dimensional structure of NAD(P)H:quinone reductase, a flavoprotein involved in cancer chemoprotection and chemotherapy: mechanism of the two-electron reduction. Proc Natl Acad Sci U S A. 92(19): p. 8846-50.

68. Patridge, E.V. and J.G. Ferry (2006). WrbA from Escherichia coli and Archaeoglobus fulgidus is an NAD(P)H:quinone oxidoreductase. J Bacteriol. 188(10): p. 3498-506.

69. Bafana, A. and T. Chakrabarti (2008). Lateral gene transfer in phylogeny of azoreductase enzyme. Comput Biol Chem. 32(3): p. 191-7.

70. Suzuki, Y., et al. (2001). Molecular cloning and characterization of the gene coding for azoreductase from Bacillus sp. OY1-2 isolated from soil. J Biol Chem. 276(12): p. 9059-65.

71. Blumel, S. and A. Stolz (2003). Cloning and characterization of the gene coding for the aerobic azoreductase from Pigmentiphaga kullae K24. Appl Microbiol Biotechnol. 62(2-3): p. 186-90.

72. Chen, H., S.L. Hopper, and C.E. Cerniglia (2005). Biochemical and molecular characterization of an azoreductase from Staphylococcus aureus, a tetrameric NADPH-dependent flavoprotein. Microbiology. 151(Pt 5): p. 1433-41.

73. Natalello, A., et al. (2007). Role of flavin mononucleotide in the thermostability and oligomerization of Escherichia coli stress-defense protein WrbA. Biochemistry. 46(2): p. 543-553.

194

74. Chen, H., et al. (2008). Functional role of Trp-105 of Enterococcus faecalis azoreductase (AzoA) as resolved by structural and mutational analysis. Microbiology. 154(Pt 9): p. 2659-67.

75. Datsenko, K.A. and B.L. Wanner (2000). One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A. 97(12): p. 6640-5.

76. Christenson, A., Spectroelectrochemistry of Redox Enzymes, in Department of Analytical Chemistry. 2006, Lund University: Lund, SE. p. 169.

77. Bistolas, N., et al. (2004). Spectroelectrochemistry of cytochrome P450cam. Biochem Biophys Res Commun. 314(3): p. 810-6.

78. Christenson, A., et al. (2008). Direct and mediated electron transfer between intact succinate : quinone oxidoreductase from Bacillus subtilis and a surface modified gold electrode reveals redox state-dependent conformational changes. Biochimica Et Biophysica Acta-Bioenergetics. 1777(9): p. 1203-1210.

79. Christenson, A., et al. (2004). Direct heterogeneous electron transfer of theophylline oxidase. Biosensors & Bioelectronics. 20(2): p. 176-183.

80. Larsson, T., A. Lindgren, and T. Ruzgas (2001). Spectroelectrochemical study of cellobiose dehydrogenase and diaphorase in a thiol-modified gold capillary in the absence of mediators. Bioelectrochemistry. 53(2): p. 243-249.

81. Invernizzi, G., et al. (2007). Protein-protein and protein-ligand interactions studied by electrospray-ionization mass spectrometry. Protein and Peptide Letters. 14(9): p. 894-902.

82. Patterson, A.V., et al. (1998). Enzymology of tirapazamine metabolism: a review. Anticancer Drug Des. 13(6): p. 541-73.

83. Cui, K., A.Y. Lu, and C.S. Yang (1995). Subunit functional studies of NAD(P)H:quinone oxidoreductase with a heterodimer approach. Proc Natl Acad Sci U S A. 92(4): p. 1043-7.

84. Walton, M.I. and P. Workman (1990). Enzymology of the reductive bioactivation of SR 4233. A novel benzotriazine di-N-oxide hypoxic cell cytotoxin. Biochem Pharmacol. 39(11): p. 1735-42.

85. Tedeschi, G., S. Chen, and V. Massey (1995). DT-diaphorase. Redox potential, steady-state, and rapid reaction studies. J Biol Chem. 270(3): p. 1198-204.

86. Iyanagi, T. and I. Yamazaki (1970). One-electron-transfer reactions in biochemical systems. V. Difference in the mechanism of quinone reduction by the NADH dehydrogenase and the NAD(P)H dehydrogenase (DT-diaphorase). Biochim Biophys Acta. 216(2): p. 282-94.

195

APPENDIX A: UV-VISIBLE SPECTROELECTROCHEMISTRY OF REDOX-ACTIVE PROTEINS A.1 Introduction to Electrochemistry A.2 Classical Electrochemical Techniques in Research A.3 Design of a Fiber-Optic Gold-Capillary Electrochemical Cell A.4 References

Appendix A

UV-Visible Spectroelectrochemistry of Redox-Active Proteins

A.1 Introduction to Electrochemistry

Benjamin Franklin was among the first to enhance our understanding of electricity, though interestingly, the field of electrochemistry largely unfolded from biological origins shortly after his research (in the 1700’s). Luigi Galvani, who knew of Franklin’s experiments with lightening, observed paroxysms in frogs after connecting their spinal nerves to metals and other sources of electrical currents. Later, Alessandro Volta reproduced

Galvani’s research and invented various devices such as the Voltaic “pile” (battery), which is often credited as the first electrochemical cell. Volta’s “pile” then became an essential tool for Humphry Davy, who used the invention to isolate a variety of elements through electrolysis, including: potassium, sodium, barium, calcium, strontium, magnesium, boron, and silicon. Lastly, after years of studying under Davy, Michael Faraday provided his Laws of Electrochemistry which offer a relationship between applied electrical charge and amount of chemical change produced [1-5]. Thus, electrochemistry has evolved from several fields over the last few centuries, and it has grown to encompass a huge array of phenomena (i.e. electrophoresis and corrosion), technologies (i.e. electroplating and large-scale chemical reductions), and devices (i.e. batteries and electroanalytical sensors).

196

The applications of electricity and electrochemical processes have groomed our society’s dependence on energy. For example, electro-refining of elemental aluminum (from molten sodium hexafluoroaluminate) is a highly energetic process that requires nearly 5% of all electricity generated in North America (and nearly 10% in other countries like Australia).

Other high-energy applications which are vital to our economy include heating and travel. In order to supply energy for these, we rely on many resources, including electricity, natural gas, petroleum, and coal. As a result of our energy dependence, the availability of non- renewable resources is becoming a more significant concern for our society. Thus in recent years there has been increased interest in renewable energy and in energy efficiency. Some of the related research that has gained significant attention includes: energy-yielding metabolic processes, enzyme engineering for faster catalysis, transfer reactions between enzymes or microbes and electrode surfaces, surface chemistry, and novel electrode and fuel cell designs . Electrochemical systems bring together several of these energy research areas and specifically incorporate technologies applicable to inexpensive, renewable energy [1, 6].

Electrochemical cell

While electrochemical systems are being developed to serve as alternative sources of energy, specialized systems are also being employed in various biological and chemical applications such as bioremediation, industrial manufacturing, health monitoring, and medical treatments. There are two main components to consider when designing electrochemical systems: the chemical phase (or an electrode) and the ionic conductor (or an electrolyte) – and they function in tandem to create an energy gradient between two electrodes. Electrodes composed of noble metals (i.e. gold, silver, platinum, copper, mercury) or of carbon are popularly employed in electrochemical research, while less

197 expensive elements (i.e. lithium, zinc, nickel, and lead) are used in common batteries. The second component (the electrolyte) is popularly composed of a liquid ionic solution, although there are also fused salts, conductive polymers, and solids with ionic components that can serve as electrolytes. With a wide range of materials available for specialized electrodes and electrolytes, and with numerous applications, an intimate understanding of electrochemical applications is essential in order to develop novel electrochemical systems [4, 6].

Besides the materials used in an electrochemical application, the processes and the cell geometries employed should also be considered. Processes occurring at electrodes are either faradaic or non-faradaic. Faradaic processes are usually the primary interest, and they

,involve charge-transfer between the electrode and the solution interface, following Faraday’s laws in a quantitative manner. Non-faradaic processes include adsorption and desorption of solutes or conformational reorganizations, which can change the electrode-solution interface and affect associated reactions. As a result of these processes and related surface chemistry

(i.e. electrode modification), an electrical double layer persists at the electrode-solution interface: at the electrode surface, solutes are adsorbed and charge carriers are reduced – while further from the surface, mass transfer occurs across the bulk of the electrolyte. The electric current across the electrolyte is dependent on cell geometry, and important considerations include electrode surface area, electrode shape, volume of electrolyte, flow channels for electrolyte, and membranes used in the apparatus. Functional requirements, such as the need to collect UV-visible spectra, could also further restrict options for possible electrochemical cell geometries. When considering an electrochemical cell design, it becomes apparent that significant time, planning, and testing is required to engineer an effective cell for specialized applications [4, 6].

198

A.2 Classical Electrochemical Techniques in Research

Electrochemical research typically involves measuring observable signal changes while applying electrical perturbations such as increasing/decreasing voltage steps, voltage ramps, pulse voltammetry, or current controlled methods. Understanding these techniques is vital for designing specialized electrochemical systems and for investigating phenomena or technologies such as corrosion, electroplating, and oxidation-reduction potentials of chemicals and proteins. Each technique has advantages and disadvantages and these should be carefully weighed for each application.

Two electrodes are required for an electrochemical system, although some use a third electrode as a reference. At the cathode, electrons are passed to the electrolyte, according the general reaction:

kf O + 푛푒 ⇌ R kb

Where O and R represent oxidized and reduced electrolyte, respectively, as a number (n) of electrons (e) are passed at forward (kf) and reverse (kb) rates of reaction. If the system is ideal and reversible, then the reverse reaction happens at the anode. The Nernst equation offers a reliable description of the chemical system with respect to the cell potential Eh:

0 푅푇 퐶O 퐸ℎ = 퐸 + ∗ 푙푛 푛퐹 퐶R

Electrolysis rates are dependent on the current, the rate of electron transfer at the electrode surface, any chemical reactions related to electron transfer, and the non-faradaic reactions. The simplest model involves a reactant being transported to the cathode surface, faradaic electron transfer to the reactant, and the product being transported away from the cathode [4, 5, 7].

Potential step methods are frequently utilized in electrochemistry. These methods employ a potentiostat to control the voltage applied across the cell. In potential steps, current

199 flow decreases in an equilateral hyperbolic fashion, slowing as the cell voltage nears the applied voltage.

퐸 𝑖 = 푒−푡/푅푠퐶푑 푅푠

In such experiments, time and experimental stability can be limiting factors. Thus, larger

(and carefully calculated) potential steps can increase the coverage over an observed transition, while also increasing accuracy of the apparent equilibrium in an electrochemical cell. With special consideration, various chemicals can be incorporated into the electrolyte in order to enhance the electrode reactions and current across the electrochemical cell. Further modifications or conditions can enhance measurements with potential step techniques such as by employing rotating electrodes, which reduce the diffusion-controlled limitations of many electrochemical cells [4, 5, 7].

Potential ramps methods are similar to the potential step methods. One popular potential ramp technique is cyclic voltammetry, typically used to characterize electron transfer reactions. This is generally a much faster technique than potential step techniques, since it essentially applies a continuous force to reduce the electrochemical system, according to the equations:

퐸 = 푣푡

−푡/푅푠퐶푑 𝑖 = 푣퐶푑 [1 − 푒 ]

By nature, potential ramp techniques are limited in accuracy. For example, slower redox transitions measured by ramp techniques are significantly more inaccurate than those determined by step techniques. Redox potentials are estimated from the apparent midpoint of the transition, and ramp methods produce data intrinsically far from equilibrium with regards to the cell potential. Such inaccuracy is amplified when transitions are not ideally reversible

[4, 5, 7].

200

A.3 Design of a Fiber-Optic Gold-Capillary Electrochemical Cell

There are a number of spectroelectrochemical cell constructions available to researchers. Currently, the most common scheme is an optically transparent, thin layer electrochemical cell [8, 9]. The small volumes and geometries of these cells are superior for certain systems, since they incorporate small diffusion distances from the electrode.

However, the optical path length in these cells yields a low optical sensitivity, limiting the system’s utility. A longer optical path length significantly enhances the usability electrochemical cells, and further modifications can enhance data collection (such as implementation of a micro stirrer) [9-11].

For the research performed in this dissertation, an electrochemical cell with a long

(and variable) optical pathlength was constructed, and a gold capillary electrode was employed similar to other cell constructions [11, 12]. This construction afforded reasonable speed and accuracy in data collection, and the water-tight construction provided stability of the system longer than 24 hours. Several modifications from previous constructions were considered and implemented.

In the early stages of development, a LS-1 tungsten-halogen light source from Ocean

Optics (Dunedin, FL) was utilized in conjunction with a broad spectrum optical glass filter

(BG3 Schott Filter Glass). Unfortunately, the design was limited in usability since the light path needed to be delicately handled during data collection. While conducting research over a 24-hour period in a busy laboratory, it was difficult to insure the light path was not disrupted. The latest design incorporated a DH-2000 deuterium-halogen light source from

Ocean Optics, which had more secure connections and brought significant stability to the light path. The new light source was also significantly brighter and covered a wider range of the spectrum, affording more sensitive and effective data collection.

201

All data was collected between 310 to 650 nm using a Kevlar-jacketed, 0.9-micron fiber. Considerations for choosing the fiber include physical flexibility, wavelengths employed in data collection, and light intensity (luminous flux through the fiber). Low hydroxyl content is generally utilized in Vis-NIR spectroscopy, high hydroxyl content is utilized in UV-Vis spectroscopy, and solarization-resistant fibers are more useful in far and near UV spectroscopy. The intrinsic properties of fibers designed for UV or shorter visible wavelengths decrease flexibility of the fiber and thus the fibers become much more fragile.

Thicker fibers (over 400 microns) have increased luminous flux, but they are more costly and can be difficult to find. The jacketed fibers used here were obtained from Varian Inc. and designed for general use in UV-Vis spectroscopy. For connections within the apparatus, fibers were fitted (and epoxied) with SMA-905 connectors.

The voltage across the electrochemical cell was controlled with an EG&G PAR 273 potentiostat for data collection. A schematic of the instrument’s relevant circuitry is presented in Figure A-1. This particular potentiostat model has been widely used in electrochemistry and has been recognized as the benchmark for comparison with other potentiostats [13]. It employs a three-electrode electrochemical cell with a differential amplifier and has a high input impedance. The potentiostat has been described as “reliable and rugged” and its design provides extremely stable current and voltage measurements.

202

- +

C.E. +

Ref

W.E.

+ Rm - i·Rm

Figure A-1. Schematic of the EG&G PAR 273 Potentiostat. The circutry illustrates how the current/voltage is controlled and measured with respect to the Voltage Source (E), Reference Electrode (Ref), Working Electrode (W.E.), Counter Electrode (C.E.), and matching resistor (Rm).

203

The gold-capillary electrode and electrochemical cell serve to couple the fiber optic light source to a USB4000-UV-Vis fiber optic diode-array spectrometer from Ocean Optics, as depicted in Figure A-2. The cell construction employed here alleviated the physical stress that was applied to the optical fiber in previous constructions. The cell is essentially a clamp that ensures a water-tight seal to a gold-capillary electrode between two modified HPLC parts. An exploded view of the cell illustrates the basic construction (Figure A-3). The gold rod and HPLC parts are presented in Figures A-4 and A-5, respectively. An engineering schematic for the plastic clamp is presented in Figure A-6 and is essentially a complete design that could be submitted for reproduction. Connections from the cell to the counter electrodes are custom plastic connectors with platinum wire epoxied within plastic stoppers.

The reference electrode is connected to the cell through a modified glass port, sized for most commercially-available electrodes. Finally, the membranes that isolate the protein within the cell were constructed from 3500 MWCO dialysis membrane and custom plastic flanges

(Figure A-7). Circular membrane “dots” were epoxied to the flanges and cured at room temperature for 48 hours, using a mixture of EPON Resin 828: EPI-CURE Curing Agent

3140: water (1.5: 1: 0.02). Different ratios of resin and curing agent resulted in membranes that were brittle or defective. Resin and Curing Agent were gifts from Hexion (Columbus,

OH). Modification directions are delineated at the end of this chapter.

204

C.E.

Light source

Spectrometer

Ref

W.E.

Figure A-2. Views of the spectroelectrochemical cell. (Top) Functional view of the electrochemical cell in connection with the potentiostat, light source, and spectrometer. Abbreviations: Counter Electrode (C.E.), Working Electrode (W.E.), and Reference Electrode (Ref). (Bottom) Function view of the electrochemical cell with sample injection ports indicated by arrows.

205

Figure A-3. Exploded top-view of the electrochemical cell and clamp.

206

Figure A-4. MicroCross. (Left) Unmodified MicroCross Assembly. (Right) Modified MicroCross Assembly.

Figure A-5. Gold Rod. (Left) Unmodified Gold Rod. (Right) A ferrule from the MicroCross Assembly that was used as a template to modify the ends of the gold rod.

207

4 mm 12 mm 26 mm Counterbore: 11 mm x 5 mm

7 mm

20 mm 14 mm 4 mm 12 mm 26 mm

7 mm 10 mm 40 mm 10 mm 25 mm 20 mm 70 mm

26 mm

12 mm

2 mm

11 mm 20 mm 3 mm

9/16 -32 10 mm 10 mm 4 mm thread

4 mm 17 mm thru 14 mm 1 mm 3 mm 45 mm

25 mm

50 mm 70 mm

Delvin Carbonate

Figure A-6. Top, side, and front schematics for the clamp of the electrochemical cell. The holder was made of carbonate, and the screw press was made of delvin.

208

2 mm

3 mm (with clearance at bottom)

3 mm

4 mm (snug, not friction fit)

6 mm

Figure A-7. Side and top schematics for the flange bushing (made of carbonate).

209

Custom design of HPLC connectors, Gold Capillary Electrode, and Membrane

Modified Parts: - (x2) Scivex (formerly Upchurch Scientific), P-777 MicroCross Assembly (Oak Harbor, WA, USA) - (x1) Goodfellow, Gold Rod # 088-438-67 (10mm L x 3mm D) (Cambridge, UK)

- Pierce, Slide-A-Lyzer® Dialysis Cassettes (3,500 MWCO) (Rockford, IL, USA)

Modifications for MicroCross (Figure A-#): - Cleave one port off the four-port MicroCross (left), as seen on the right. - Drill 1mm through all ports for the entire length of the assembly. - Drill a 1mm hole from the top center of the MicroCross to meet the inner space. - Friction-fit (and epoxy) PEEK tubing to the new hole for sample injection port. - Repeat this for the second, opposing MicroCross assembly.

Modifications for Gold Rod (Figure A-#): - Modify the gold rod to make a gold capillary of 1cm L x 2mm inner diameter. - Shape each end of the gold capillary to precisely mimic the appropriate ferrule.

Modifications for Slide-A-Lyzer® Dialysis Cassette (Figure A-#): - Cut a single membrane layer out of the cassette. - Using a circular tool, cut one 4mm-diameter membrane “dot” per flange bushing. - Mix an epoxy EPON Resin 828: EPI-CURE Curing Agent 3140: water (1.5: 1: 0.02) - Dab the flange’s edge with epoxy and affix membrane “dot.” Cure 48 hours at 22°C.

210

A.4 References

1. History of electrochemistry, in Wikipedia. April 27, 2009.

2. Henney, K. (1965). Alessandro Volta and Electric Battery. Ieee Spectrum. 2(9): p. 176-&.

3. RM, W., The Electric Current. (1894), London: Cassell and Company, Limited.

4. Bard, A.J. and L.R. Faulkner, Electrochemical methods : fundamentals and applications. 2nd ed. (2001), New York: Wiley. xxi, 833 p.

5. Prentice, G., Electrochemical engineering principles. Prentice-Hall international series in the physical and chemical engineering sciences. (1991), Englewood Cliffs, N.J.: Prentice Hall. xxii, 296 p.

6. O'Hayre, R.P., Fuel cell fundamentals. 2nd ed. (2008), Hoboken, NJ: J. Wiley & Sons.

7. Hamann, C.H., A. Hamnett, and W. Vielstich, Electrochemistry. (1998), Weinheim ; New York: Wiley-VCH. xvii, 423 p.

8. Murray, R.W., W.R. Heineman, and G.W. Odom (1967). An Optically Transparent Thin Layer Electrochemical Cell. Analytical Chemistry. 39(13): p. 1666-&.

9. Niu, J.J. and S.J. Dong (1995). A New Simple Long-Path-Length Thin-Layer Spectroelectrochemical Cell in Various Uses. Electroanalysis. 7(11): p. 1059-1062.

10. Gui, J.Y., G.W. Hance, and T. Kuwana (1991). Long Optical Pathlength Thin-Layer Spectroelectrochemistry - Study of Homogeneous Chemical-Reactions. Journal of Electroanalytical Chemistry. 309(1-2): p. 73-89.

11. Christenson, A., Spectroelectrochemistry of Redox Enzymes, in Department of Analytical Chemistry. 2006, Lund University: Lund, SE. p. 169.

12. Larsson, T., A. Lindgren, and T. Ruzgas (2001). Spectroelectrochemical study of cellobiose dehydrogenase and diaphorase in a thiol-modified gold capillary in the absence of mediators. Bioelectrochemistry. 53(2): p. 243-9.

13. Research, P.A., 273A Potentiostate/Galvanostat. 2009: Oak Ridge, TN.

211

APPENDIX B: Crystal Structure of the NADH:Quinone Oxidoreductase WrbA from Escherichia coli B.1 Personal Contribution to the Manuscript B.2 Article as Published

Appendix B

Crystal Structure of the NADH:Quinone Oxidoreductase WrbA from Escherichia coli

B.1 Personal Contribution to the Manuscript

The paper reproduced in this appendix describes a collaborative research effort between the research groups of Dr. James G. Ferry (Pennsylvania State University, USA),

Dr. Oliver Einsle (Lund University, Germany). The paper was published in The Journal of

Bacteriology1. My contributions to this manuscript included the preparation of protein for crystallization, direction for investigation of cofactor binding, and concepts that significantly enhanced the manuscript.

1Reproduced with permission from Susana L. A. Andrade, Eric V. Patridge, James G. Ferry, and Oliver Einsle. 2007. Journal of Bacteriology. 189:9101-9107. Copyright 2007 American Society for Microbiology. 212

Crystal Structure of the NADH:Quinone Oxidoreductase WrbA from Escherichia coliᰔ Susana L. A. Andrade,1* Eric V. Patridge,2 James G. Ferry,2 and Oliver Einsle1 Institute for Microbiology and Genetics, Georg August University Go¨ttingen, Justus von Liebig Weg 11, 37077 Go¨ttingen, Germany,1 and Department of Biochemistry and Molecular Biology, Eberly College of Science, The Pennsylvania State University, 205 South Frear Laboratory, University Park, Pennsylvania 168022

Received 17 August 2007/Accepted 5 October 2007

The flavoprotein WrbA, originally described as a tryptophan (W) repressor-binding protein in Escherichia coli, has recently been shown to exhibit the enzymatic activity of a NADH:quinone oxidoreductase. This finding points toward a possible role in stress response and in the maintenance of a supply of reduced quinone. We have determined the three-dimensional structure of the WrbA holoprotein from E. coli at high resolution (1.66 Å), and we observed a characteristic, tetrameric quaternary structure highly similar to the one found in the WrbA homologs of Deinococcus radiodurans and Pseudomonas aeruginosa. A similar tetramer was originally observed in an iron-sulfur flavoprotein involved in the reduction of reactive oxygen species. Together with other, recently characterized proteins such as YhdA or YLR011wp (Lot6p), these tetrameric flavoproteins may constitute a large family with diverse functions in redox catalysis. WrbA binds substrates at an active site that provides an ideal stacking environment for aromatic moieties, while providing a pocket that is structured to stabilize the ADP part of an NADH molecule in its immediate vicinity. Structures of WrbA in complex with benzoquinone and NADH suggest a sequential binding mechanism for both molecules in the catalytic cycle.

In 1993, a previously unknown protein was reported to co- 22, 35). Based on subsequent biochemical assays, NQO activity purify with the tryptophan repressor TrpR of Escherichia coli was demonstrated for the WrbA proteins of both E. coli and and was, due to this assumed interaction, termed tryptophan Archaeoglobus fulgidus, and this family of flavoproteins was (W) repressor binding protein, WrbA (36). At closer inspec- therefore described as a new class of type IV NQOs (29). The Ϯ ␮ tion, WrbA was found to bind a flavin mononucleotide (FMN), E. coli protein bound NADH with a Km of 14 0.43 M and Ϯ ␮ making it a founding member of a novel family of flavodoxins benzoquinone with a Km of 5.8 0.12 M. The affinities for (12). Based on sequence analyses, WrbA was predicted to WrbA from A. fulgidus were 19 Ϯ 1.70 ␮M for NADH and show the typical ␤-␣-␤-fold of flavodoxins, with a twisted, five- 37 Ϯ 3.60 ␮M for benzoquinone, and an analysis of apparent stranded, parallel ␤-sheet and a binding site for the FMN initial velocities indicated that the maximal activities of the cofactor at the carboxy terminal end of this ␤-sheet. Distinct enzymes were consistent with the optimal growth temperatures from the basic flavodoxin fold (30), a conserved insertion was of the mesophilic E. coli and the hyperthermophilic A. fulgidus, found in sheet ␤5 (residues 140 to 160) and was predicted to respectively (29). Along with cyclic voltammetry measurements form an additional ␣␤-unit (11). Proteins with this type of of WrbA (27), these results implicate WrbA proteins in the insertion have been classified as long-chain flavodoxins (23). two-electron reduction of quinones as a measure of protection The functional role of WrbA remained controversial, as initial against oxidative stress. work indicated that E. coli WrbA promotes complex formation We have crystallized WrbA from E. coli in its native state between the repressor TrpR and its operator DNA. WrbA and with both the substrate benzoquinone and the cosubstrate alone failed to bind to DNA, such that it was consequently NADH bound. Our structures point toward mutually exclusive proposed to be an accessory element that could enhance TrpR- binding of quinones and NADH to the FMN cofactor, indicat- dependent repression of genes upon transition to stationary ing an ordered, sequential mechanism of catalysis. Further- phase (36). Subsequent experiments addressed the influence of more, the structures underline the obvious homologies of WrbA on the TrpR-DNA complex, and as these did not show WrbA with the iron-sulfur flavoproteins (ISF) from Methano- any effect, the involvement of WrbA in transcription regulation sarcina thermophila and A. fulgidus that form the same char- was revoked (12). acteristic homotetrameric structures but differ significantly in While the function of WrbA was entirely enigmatic, se- functionally relevant, structural details (1). quence database searches hinted that the WrbA family of fla- We note that at the time of submission of the manuscript, voproteins was homologous to a group of several biochemically crystallization of E. coli WrbA and preliminary evaluation of characterized NAD(P)H:quinone oxidoreductases (NQO) data had been presented independently in two publications from various species of fungi and green plants (2, 3, 5, 13, 17, (33, 34).

* Corresponding author. Mailing address: Institute for Microbiology MATERIALS AND METHODS and Genetics, Georg August University Go¨ttingen, Justus von Liebig Crystallization and data collection. Crystallization experiments for WrbA Weg 11, 37077 Go¨ttingen, Germany. Phone: 49 551 391 4088. Fax: 49 from E. coli were set up at 20°C for sitting-drop vapor diffusion using a 1:1 drop 551 391 4082. E-mail: [email protected]. ratio composed of 10 to 50 mg/ml of protein in 20 mM Tris-HCl buffer at pH 7.8 ᰔ Published ahead of print on 19 October 2007. and a suitable reservoir solution. The reservoir was composed solely of 15 to 35%

9101 9102 ANDRADE ET AL. J. BACTERIOL.

TABLE 2. Reﬁnement statistics

Value for the indicated crystal type Data seta Native NADH Benzoquinone Small cell

Resolution range (Å) 50.00–1.66 50.00–2.05 50.00–1.95 50.00–1.84 Rcryst 0.169 0.168 0.172 0.187 Rfree 0.186 0.199 0.208 0.223 RMSD in bond lengths (Å) 0.012 0.018 0.017 0.019 RMSD in bond angles (°) 1.328 2.046 1.686 1.671 Avg B-factor for protein (Å2) 26.5 30.7 29.9 35.2 Avg B-factor for waters (Å2) 47.7 49.1 45.5 47.3 Cruickshank’s DPI (Å) 0.047 0.087 0.087 0.096

a DPI, diffraction-component precision index. FIG. 1. Crystals of E. coli WrbA. The tetragonal bipyramidal shape of the crystals reflects their P422 symmetry. between the protruding helix ␣3 from one monomer and the C-terminal parts of helices ␣2 and ␣7 of another monomer from a different WrbA tetramer. The (wt/vol) polyethylene glycol (PEG) 1500, and single crystals appeared after 1 to replacement solutions were initially refined using the REFMAC program (25), 2 days and continued to grow for 2 to 3 weeks (Fig. 1). For flash freezing prior which provided the first electron density maps. The electron density for the FMN to data collection, crystals of E. coli WrbA were washed for 1 to 5 min in cofactor that was absent in the search model was readily identified in the first Ϫ solutions containing increasing amounts of PEG 1500 in steps of 5% until a final Fo Fc (the observed and calculated structure factor amplitudes, respectively) concentration of 50% (wt/vol) was reached. Diffraction data sets to a maximum electron density maps. All model building was carried out in the programs O (14) resolution of 1.66 Å were collected using synchrotron radiation at EMBL/DESY, and COOT (9). The model for E. coli WrbA comprises the entire polypeptide chain with the exception of the first residue (Met 1) that might have been Hamburg, Germany. Crystals belonged to the tetragonal space group P43212, but occurred in two distinct crystal packing arrangements with different cell axes. The naturally cleaved off. The initially planar conformation of the isoalloxazine ring smaller cell (a ϭ b ϭ 62.80 Å; c ϭ 201.53 Å) yielded a typical crystal packing with of FMN did not fit the observed electron density. In order to allow the refine- 3 ment programs to introduce a slight bend, the estimated standard deviations of a Matthews parameter VM (21) of 2.26 Å /Da, corresponding to a solvent content of 46%, but the larger cell (a ϭ b ϭ 94.36 Å; c ϭ 175.36 Å) resulted in a very the planarity restraints were increased in the library files. 3 With continuing improvement of the models, additional electron density that sparse crystal packing with a solvent content of 74% (VM of 4.39 Å /Da), which is unusually high for soluble proteins. Nevertheless, crystals with both cells was at first modeled as water molecules became clearly continuous. It was then diffracted well, with the more frequently occurring, large cell giving the highest modeled with fragments of PEG of variable lengths. The final atomic model of resolution data set. Data were indexed, integrated, and scaled using the HKL native E. coli WrbA at a resolution of 1.66 Å comprises 198 amino acid residues suite (28) and analyzed using XPREP (Bruker). Table 1 gives the data collection plus one FMN molecule per monomer and a total of 463 water molecules and and refinement statistics. four PEG fragments of various lengths. Stereochemical analyses of the structures Substrate soaks and refinement. For all soaking experiments, substrate and were carried out with PROCHECK (17). In a Ramachandran plot, all residues cosubstrate molecules were prepared as stock solutions of 200 mM NADH and were found in the most favored or additionally allowed regions. Refinement 100 mM 1,4-benzoquinone. These were then added to the cryoprotectant buffer statistics are given in Table 2. All figures were made with MOLSCRIPT (16) and containing 50% (wt/vol) PEG 1500 to a final substrate concentration of 1 mM. Raster3D (24) or PyMOL (7). Equivalent crystallization and soaking experiments were also carried out under Protein Data Bank accession codes. Structure factors and coordinates have anaerobic conditions, either with protein crystallized in a glove box containing been deposited with the Protein Data Bank at http://www.pdb.org. Accession less than 1 ppm of oxygen or by the addition of 2 mM sodium dithionite solution, codes are 3B61 for the native crystal, 3B6J for the NADH complex, 3B6K for the pH 8, to all cryoprotectant solutions. benzoquinone complex, and 3B6M for the second crystal form (small cell). Structure solution and refinement. The structure was solved by molecular replacement with programs from the CCP4 suite (6), using the homologous apo-WrbA structure of Deinococcus radiodurans (Protein Data Bank identifier RESULTS [PDB ID] 1YDG) as a search model (10). For refinement, 5% of the total Crystal structure of E. coli WrbA. WrbA from E. coli shows reflections were chosen at random and used as a test set for cross validation (4). In both unit cells, the replacement solution contained two monomers in the high homologies to the previously described WrbA proteins asymmetric unit. In the larger cell, crystal contacts were exclusively mediated from D. radiodurans (PDB ID 1YRH; root mean square devi-

TABLE 1. Data collection statisticsa

Value for the indicated crystal type Data set Native NADH Benzoquinone Small cell

Space group P 43212 P43212 P43212 P43212 Unit cell constants (Å) a 94.36 94.20 94.24 62.80 b 94.36 94.20 94.24 62.80 c 175.36 173.64 175.55 201.53 Resolution range (outer 50.00–1.66 (1.69–1.66) 50.00–2.05 (2.15–2.05) 50.00–1.95 (2.05–1.95) 50.00–1.84 (1.94–1.84) shell [Å]) No. of unique reﬂections 91,774 48,152 59,061 35,588 Completeness (%) 98.7 (97.3) 96.9 (93.7) 99.5 (97.3) 99.4 (97.0) Redundancy 6.3 (5.3) 4.1 (3.4) 5.9 (5.2) 6.7 (6.3) Mean I/␴ (I) 12.1 (2.2) 5.9 (1.9) 6.2 (1.8) 10.3 (2.6)

Rsym 0.12 (0.68) 0.12 (0.50) 0.11 (0.58) 0.06 (0.40) Rpim 0.06 (0.34) 0.07 (0.36) 0.03 (0.29) a Values in brackets represent the last resolution shells. VOL. 189, 2007 STRUCTURE OF E. COLI WrbA 9103

FIG. 2. Cartoon representation of WrbA from E. coli. The stereo image of the protein chain shows the N terminus in blue and the C terminus in red. Strands of the central, parallel ␤-sheet and the surrounding ␣-helices are numbered according to their occurrence in the protein sequence. The FMN cofactor is bound peripherally at the C-terminal end of the ␤-sheet.

ation [RMSD] for all C␣ atoms, 1.53 Å) and P. aeruginosa region is not involved in crystal lattice contact formation in (PDB ID 1ZWL; RMSD, 1.20 Å) (10). However, the E. coli either of the two unit cells. A possible functional significance of protein was always crystallized in its holo-form, i.e., with the this conformational change can therefore presently not be ex- cofactor FMN tightly bound to the protein. The structural cluded. No significant differences were observed in data sets hallmarks of the fold of WrbA family proteins are five parallel obtained from crystals reduced with dithionite. ␤-strands forming a twisted ␤-sheet, with each strand, ␤1to␤5, The FMN cofactor is bound on the periphery of the mono- followed by an ␣-helix, ␣1to␣5 (Fig. 2). Three insertions add mer, in a hydrogen-bonding network formed by the loop re- to this topology: a loop inserted between strand ␤2 and helix gions at the C termini of the ␤-sheets. It packs against the ␣ 2b (residues 38 to 64; loop 1) that contains a further helical protein with its re face, whereby the catalytically relevant N5 segment, ␣2a; the aforementioned insertion within strand ␤5 atom of the isoalloxazine ring forms a long (3.2 Å) hydrogen (loop 2); and a short insertion before helix ␣5 (loop 3). In all bond to the backbone nitrogen of Phe 80. Below the FMN, an structures, the entire protein chain is clearly defined, and the arginine residue, Arg 79, stabilizes the cofactor through a absence of residue Met 1 is most likely due to posttranslational stacking interaction. While Arg 79 is conserved among WrbA processing, as the amino terminus at Ala 2 and the terminal proteins, it seems to be consistently replaced by a tyrosine in carboxy group at Gly 198 form a direct salt bridge of 2.6 Å. the ISF family of flavoproteins (see below) (29). Contacts Disorder was observed, however, in the second crystal form between FMN and WrbA predominantly occur through the with the smaller unit cell. Here, the loop 2 region showed oxygen or nitrogen atoms of the backbone amide bonds, with significant disorder, and residues 146 to 156 were not visible in the notable exception of His 14 that forms a hydrogen bond to electron density maps. In this structure, residue Tyr 143, mod- an oxygen atom of the FMN phosphate. erately conserved among the WrbA family (Fig. 3) was flipped WrbA has been found to behave like a dimer in size exclusion outward, now facing away from the active site. This disorder chromatography and like a tetramer in a dynamic light scattering cannot be explained by intermolecular interactions within the experiment (29). The crystal structures of all WrbA proteins crystal lattice, as the protein surface surrounding the loop 2 known today strongly support a homotetrameric quaternary structure with four independent active sites. The FMN cofactor becomes buried in a distinct active-site cavity upon multimerization, and residues from three different monomers participate in composing the structural features of this cavity. The environment of the active site itself is largely hydrophobic. Its rear wall is formed by the side chain of Phe 80, and its roof is formed by Trp 98 from a neighboring subunit, ideally suited to stack aromatic moieties—such as quinones or nicotinamide—between the indole side chain and the FMN. A second residue involved from the same subunit as Trp 98 is His 133, forming a hydrogen bond to

oxygen O4 of FMN. From the third subunit involved in forming

the active site, Tyr 143 forms a long (3.25 Å) hydrogen bond to N3 of the FMN, and Phe 149 shields this residue against the solvent (Fig. 3). Residues His 133 and Tyr 143, although close in protein sequence, in this way form parts of two different active sites within the tetramer (Fig. 4A). FIG. 3. Close-up of the active site of E. coli WrbA. Residues from E. coli WrbA was crystallized from a precipitant solution three monomers combine to create a speciﬁc, hydrophobic active-site pocket. Aromatic substrates can be stacked in between the FMN moi- containing exclusively PEG 1500, and in the electron density ety and the side chain of Trp 98. All residues are colored by chain. maps, several ordered, continuous features were detected that 9104 ANDRADE ET AL. J. BACTERIOL.

FIG. 4. Tetrameric arrangement of WrbA and binding of NADH. (A) Cartoon representation of the tetramer. A single monomer is shown, colored as described in the legend of Fig. 1. Note the extended loop 2 region (Fig. 2) interacting with a neighboring monomer. (B) Surface representation of the tetramer, centered on the active site of one monomer. The picture shows a PEG molecule bound at the active site, an NADH molecule in close proximity, and several further PEG molecules along a dimer interface. (C) Stereo representation of a close-up of the active site in the same orientation as in panel B. The left and right images are for the left and right eyes, respectively. The nicotinamide ring of NADH stacks onto the side chain of Tyr 12.

were modeled with PEG chains (Fig. 4B). Most of these are hydroxo group hydrogen bonded to His 133. At the pH of 7.5 wrapped around one of the dimer interfaces, with the longest used in the crystallization setup, this histidine is likely to be segment involving the extended side chain of Lys 195. Also, an deprotonated. Since crystallization and stabilization of crystals undefined electron density feature was located on top of the were critically dependent on the presence of PEG 1500, we did FMN cofactor in a putative binding position for substrate and not succeed in replacing the putative PEG molecule in soaks cosubstrate. This density was most prominent in the NADH with NADH. As a consequence, the nicotinamide-ribityl part soak, where it was modeled as a PEG chain. In all other of NADH was not modeled on the flavin ring but, rather, in a structures, this electron density was weaker and was conse- pocket next to it, with the nicotinamide ring in a ␲-stacking quently modeled with only two or three water molecules, but interaction with the side chain of Tyr 12, a residue that is not we cannot exclude at least partial presence of a PEG molecule. conserved among either WrbA or ISF family members. This NADH binding to WrbA. In soaks of E. coli WrbA with observed position of the nicotinamide can be considered non- NADH, the dinucleotide was found to bind in close proximity functional (Fig. 4C). However, manual removal of the PEG to the FMN cofactor to a specific binding pocket that is pre- molecule and rotation of the less-well-ordered part of the dominantly composed of the extended loop 1 region but is NADH molecule show that the specific binding of the ADP situated at the interface of two monomers in the tetramer (Fig. moiety places the remainder of the dinucleotide in a perfect 4B and C). In this binding mode, excellent electron density and position to interact with the flavin and form an electron trans- low B-factors were observed for the ADP-part of NADH, fer-competent complex. while the remaining nicotinamide-ribityl moiety was less or- Complex of WrbA with benzoquinone. Binding of the known dered. Strikingly, contacts between NADH and protein are substrate benzoquinone to the FMN site of WrbA was ob- almost exclusively mediated by water molecules, with only a served in a complex structure at a resolution of 1.95 Å. The single hydrogen bond being observed between O␩ of Tyr 12 aromatic ring system of the substrate was found to stack be- and a phosphate oxygen of NADH. In the complex structure, tween the isoalloxazine ring of FMN and the indole side chain the putative substrate binding pocket between the FMN cofac- of Trp 98, placing it in an ideal position for electron transfer tor and the side chain of Trp 98 was occupied by what was (Fig. 5). In the benzoquinone soaks, the quality of the observed interpreted to be the end of a PEG chain, with its terminal electron density map varied consistently between the two VOL. 189, 2007 STRUCTURE OF E. COLI WrbA 9105

parison between WrbA and ISF underlines the high homology of both classes of proteins, particularly in the region of the central ␤-sheet (residues 70 to 150), where the structures of M. thermophila ISF and E. coli WrbA align with an

RMSD of their C␣ atoms of 0.74 Å (Fig. 6). The three loop regions inserted into the WrbA sequence with respect to regular flavodoxins are also found in ISF family proteins. The first of them, loop 1, holds the [4Fe:4S] cluster in ISF, but its conformation varies drastically between ISF and WrbA, where it plays a crucial role in forming the binding cleft for the cosubstrate NADH (Fig. 4C and 6). In contrast to the position of the cluster in ISF, this binding cleft allows for electron transfer from NADH bound predominantly by residues of the same subunit that holds the corresponding FMN cofactor. FIG. 5. Binding of benzoquinone to the FMN site of WrbA. The ␤ ␤ picture shows the amino acid residues surrounding the active site and The two remaining insertions, loop 2 within -strand 5 Ϫ ␴ an Fo Fc density map, contoured at 3.5 , from which the benzo- from residues 140 to 160 and loop 3 from residues 165 to 174, quinone molecule was omitted. have been found to be more extensive in the structure of ISF-3 from A. fulgidus than in ISF of M. thermophila (1). They have been hypothesized to increase the stability of the tetrameric monomers of WrbA present in the asymmetric unit of the arrangement of subunits, and indeed loop 2 of WrbA forms crystal. Consequently, after refinement higher B-factors were extensive interactions with neighboring subunits (Fig. 4A). As observed for the benzoquinone molecule on the less ordered highlighted by the architecture of the active site and by the side. It can be assumed that full occupancy of the binding sites location of the NADH binding site at the interface of two with the substrate was not achieved, possibly due to the partial subunits, the tetramer seems to be essential for the function- binding of PEG molecules at this location, as observed in the ality of WrbA, and its additional stabilization by the extended NADH soak. However, comparisons of different electron den- loop 2 region supports this hypothesis. sity features with native crystals clearly confirmed the presence At the FMN site, the N atom is shielded by a phenylalanine of the planar benzoquinone molecule (Fig. 5). 5 A. Structural comparison with the ISF of M. thermophila. The residue, Phe 80, and a similar situation is found in ISF-3 of tetrameric quaternary structure of WrbA of E. coli, as well fulgidus, where Phe 84 is found at this position; in the ISF of M. as of its homologs from D. radiodurans and P. aeruginosa, thermophila a methionine, Met 86 provides a potentially more was first observed in a different family of flavoproteins, the reactive environment (1). In all cases, the roof of the active site ISFs from M. thermophila and A. fulgidus (1, 37). These cavity is set to interact with aromatics in a stacking interaction proteins show the same, typical flavodoxin fold with a five- between the FMN moiety and Trp 98 in E. coli WrbA, Arg 104 stranded ␤-␣-␤-topology, but they are characterized by an in M. thermophila ISF, and Arg 102 in A. fulgidus ISF-3. How- additional loop region between strand ␤2 and helix ␣2 that ever, the exact environment of this site, the proximity of puta- contains a compact Cys-X2-Cys-X2-Cys-X5-Cys motif. A cu- tive donor and acceptor sites for hydrogen-bonding interaction bane-type [4Fe:4S] cluster is bound to this motif, and the with a substrate, and the electrostatic potential are very differ- loop attains a conformation that brings this cluster in close ent. Similar to A. fulgidus ISF-3, E. coli WrbA shows a strongly proximity to an FMN cofactor not of the same, but of a negative electrostatic surface potential (data not shown), while neighboring, subunit in the tetramer (1). A structural com- that of M. thermophila ISF is strongly positive (1).

FIG. 6. Superposition of the monomers of E. coli WrbA (red) and M. thermophila ISF (black). Based on a common ﬂavodoxin-like fold, the structures differ in three characteristic loop regions. Strongest variations are visible in the loop region 1, where ISF binds a [4Fe:4S] cluster while WrbA forms a speciﬁc binding cleft for NADH at a very different position in the monomer. However, multimer formation then places both putative electron donor sites at very similar positions in respect to the FMN cofactor. 9106 ANDRADE ET AL. J. BACTERIOL.

DISCUSSION cluster that serves as the initial electron acceptor. In ISF, a flavin semiquinone was not detected during redox titrations Reaction mechanism of WrbA. The structure of E. coli (18), meaning that the first electron delivered may be stored in WrbA in complex with an NADH molecule outlines a clear the cluster while the second electron to arrive then triggers a reaction pathway. The two-electron carrier NADH binds to its two-electron reduction of the flavin to its hydroquinone form. high-affinity binding site in immediate proximity to the oxi- In WrbA the electron donor system has been changed to a dized FMN cofactor. With the nicotinamide ring stacking on specific binding cleft for the two-electron donor NADH, thus the isoalloxazine, two electrons and a proton can be trans- ϩ removing the need for a separate one-electron storage site ferred, and the remaining NAD then leaves the active site, while retaining a possibly similar mechanism at the FMN. making room for the binding of a substrate molecule such as Further support for the hypothesis of a widespread family of benzoquinone. Subsequent two-electron reduction of the sub- FMN reductases comes from another candidate member, the strate returns the FMN cofactor back to its oxidized state and NADPH:FMN reductase YhdA from Bacillus subtilis (8, 32). closes the reaction cycle. A ping-pong mechanism of this kind, This structure has been independently deposited in the PDB by with overlapping sites for the hydride donor and acceptor, has the Midwest Structural Genomics Consortium in 2003 (PDB been proposed for other oxidoreductases such as NADPH: ID 1NNI) and by the Northeast Structural Genomics Consor- quinone reductase from rat liver (19). The significance of the tium in 2006 (PDB ID 2GSW). However, no publication fur- conserved tetrameric arrangement of this class of proteins lies ther describing the structure is available to date. YhdA shows in providing residues required to construct an intricate active- a fold with high homology to ISF and WrbA but with a very site cavity. As the WrbA monomer is evolutionarily derived open and accessible FMN site. The region equivalent to loop 1 from flavodoxins, it binds the FMN cofactor at the very pe- does not harbor a [4Fe:4S] cluster as in ISF (Fig. 6), but it riphery of the protein, with no possibility or only very limited adopts a conformation close to that of ISF and distinct from possibilities to arrange residues around the exposed si face of the one observed in WrbA. This implies that the NADH bind- the flavin. In WrbA, multimer formation brings two additional ing groove of WrbA is absent, and thus (5) YhdA must employ regions in close proximity to a neighboring FMN: the C termini a very different binding mode for NADPH. YhdA also fully ␣ ␣ of helices 3 (residues Gly 95 and Trp 98) and 4 (His 133) as retains the tetrameric quaternary structure typical for this pro- ␤ well as the characteristic insertion in strand 5, the loop 2 tein family, which has been suggested to contribute to the region between residues 140 and 150 (Tyr 143 and Phe 149) significant thermostability of both WrbA (26) and YhdA (8). (Fig. 2 and 4A). However, although its FMN and the putative substrate azo- Moreover, the arrangement of the loop 1 region in WrbA benzene are aromatic, the structure of YhdA does not show creates a specific binding cleft for NADH, and although our any residues placed to engage in a ␲-stacking such as Trp 98 in structure shows a flexible binding of the nicotinamide-ribityl E. coli WrbA or in cation-␲-stacking such as Arg 104 in M. part of the NADH molecule, the positioning of the binding thermophila ISF. sites clearly allows for an interaction in which the nicotinamide In contrast, one of the few structures of flavodoxin-like pro- ring stacks right above the N5 position of FMN. Very similar teins from eukaryotic organisms, the NADPH:FMN reductase arrangements are commonly observed in structures of nicotin- YLR011wp (Lot6p) from Saccharomyces cerevisiae, was purified amide dinucleotides bound to flavoproteins, as for example in and crystallized as a dimer, and although a similar tetramer is ϩ ferredoxin:NADP reductase at the end of the photosynthetic generated by crystal packing, the interaction of its two monomers electron transport chain (15, 31). is loose, putting them at a much larger distance than in any WrbA WrbA and ISF as reductases under oxidative stress. Similar or ISF structure (20). Accordingly, YLR011wp lacks the extended to WrbA, ISF proteins were not discovered in conjunction with loop 2 region that we presume to act in stabilizing the tetramer a defined enzymatic function, and their exact role in the bac- and is therefore to be classified as a short-chain flavodoxin (23). terial cell remains to be elucidated. They are widespread in As an increasing number of primary sequences and three- bacteria and archaea and are able to accept electrons from dimensional structures become available through genomics ferredoxins, such that they were originally proposed to func- and structural genomics efforts, the tetrameric oxidoreductases tion as one-electron/two-electron switches in electron transfer of the WrbA and the ISF families turn out to be a widespread chains (18). However, the discovery of a large family of these class of redox proteins with a wide range of possible substrates. proteins soon suggested a broader functionality (37). Recently, In spite of the wealth of information gained from this broad- a reducing activity toward O2 and H2O2 was shown for M. ening base of data, however, the tools of protein biochemistry thermophila ISF, suggesting a protective role toward oxidative and enzymology will remain crucial for achieving a compre- stress, similar to the one assumed for WrbA (29). hensive understanding of the physiological function of the in- In this light, the obvious, evolutionary kinship of ISF and dividual enzymatic machineries in question. WrbA proteins outlines a class of modular, tetrameric proteins that act as putative reductases of aromatic substrates. Their active sites show substantial differences, but a ␲-stacking in- ACKNOWLEDGMENTS teraction involving both the FMN isoalloxazine and an amino This work was supported by a Marie Curie Intra-European Fellow- acid residue from a neighboring subunit seems to be a common ship within the 6th Framework Programme (S.L.A.A.), by the EMBO theme. At the same time, the proteins show considerable flex- Young Investigator Programme (O.E.), by Deutsche Forschungsge- meinschaft (S.L.A.A. and O.E.), and by grant DE-FG02-95ER20198 ibility in their choice of electron donor: the ISF family exem- from the U.S. Department of Energy (J.G.F.). plifies a system optimized to interact with soluble electron Diffraction data were collected at EMBL/DESY, Hamburg, Ger- donors such as ferredoxins. To this end they possess a [4Fe:4S] many. We thank Ralf Ficner for continuous support. VOL. 189, 2007 STRUCTURE OF E. COLI WrbA 9107

ADDENDUM IN PROOF 18. Latimer, M. T., M. H. Painter, and J. G. Ferry. 1996. Characterization of an iron-sulfur flavoprotein from Methanosarcina thermophila. J. Biol. Chem. While this paper was under review, a description of the structure was 271:24023–24028. published (J. Carey, J. Brynda, J. Wolfova´, R. Grandori, T. Gustavs- 19. Li, R., M. A. Bianchet, P. Talalay, and L. M. Amzel. 1994. Structural studies son, R. Ettrich, and I. K. Smatanova´, Protein Sci. 16:2301–2305, 2007). of quinone reductase by X-ray crystallography. Biophys. J. 66:A265. It is in accordance with our data but does not show any ligand binding 20. Liger, D., M. Graille, C. Z. Zhou, N. Leulliot, S. Quevillon-Cheruel, K. to the FMN. Blondeau, J. Janin, and H. van Tilbeurgh. 2004. Crystal structure and functional characterization of yeast YLR011wp, an enzyme with NAD(P)H- REFERENCES FMN and ferric iron reductase activities. J. Biol. Chem. 279:34890–34897. 21. Matthews, B. W. 1968. Solvent content of protein crystals. J. Mol. Biol. 1. Andrade, S. L. A., F. Cruz, C. L. Drennan, V. Ramakrishnan, D. C. Rees, 33:491–497. J. G. Ferry, and O. Einsle. 2005. Structures of the iron-sulfur flavoproteins 22. Matvienko, M., A. Wojtowicz, R. Wrobel, D. Jamison, Y. Goldwasser, and from Methanosarcina thermophila and Archaeoglobus fulgidus. J. Bacteriol. J. I. Yoder. 2001. Quinone oxidoreductase message levels are differentially 187: 3848–3854. regulated in parasitic and non-parasitic plants exposed to allelopathic qui- 2. Brock, B. J., and M. H. Gold. 1996. 1,4-benzoquinone reductase from the nones. Plant J. 25:375–387. basidiomycete Phanerochaete chrysosporium: spectral and kinetic analysis. 23. Mayhew, S. G., and M. L. Ludwig. 1975. Flavodoxins and electron-transfer- Arch. Biochem. Biophys. 331:31–40. ring flavoproteins. Academic Press, New York, NY. 3. Brock, B. J., S. Rieble, and M. H. Gold. 1995. Purification and characterization of a 1,4-benzoquinone reductase from the basidiomycete Phanerochaete 24. Merritt, E. A., and D. J. Bacon. 1997. Raster3D: photorealistic molecular chrysosporium. Appl. Environ. Microbiol. 61:3076–3081. graphics. Methods Enzymol. 277:505–524. 4. Bru¨nger, A. T. 1993. Assessment of phase accuracy by cross validation: the 25. Murshudov, G. N., A. A. Vagin, and E. J. Dodson. 1997. Refinement of free R-value. Methods and applications. Acta Crystallogr. D 49:24–36. macromolecular structures by the maximum-likelihood method. Acta Crys- 5. Cohen, R., M. R. Suzuki, and K. E. Hammel. 2004. Differential stress- tallogr. D 53:240–255. induced regulation of two quinone reductases in the brown rot basidiomy- 26. Natalello, A., S. M. Doglia, J. Carey, and R. Grandori. 2007. Role of flavin cete Gloeophyllum trabeum. Appl. Environ. Microbiol. 70:324–331. mononucleotide in the thermostability and oligomerization of Escherichia 6. Collaborative Computational Project No. 4. 1994. The CCP4 suite: programs coli stress-defense protein WrbA. Biochemistry 46:543–553. for protein crystallography. Acta Crystallogr. D 50:760–763. 27. No¨ll,G., E. Kozma, R. Grandori, J. Carey, T. Schodl, G. Hauska, and J. 7. DeLano, W. L. 2002. The PyMOL molecular graphic system. DeLano Sci- Daub. 2006. Spectroelectrochemical investigation of a flavoprotein with a entific, Palo Alto, CA. flavin-modified gold electrode. Langmuir 22:2378–2383. 8. Deller, S., S. Sollner, R. Trenker-El-Toukhy, I. Jelesarov, G. M. Gubitz, and 28. Otwinowski, Z., and W. Minor. 1996. Processing of X-ray diffraction data P. Macheroux. 2006. Characterization of a thermostable NADPH:FMN ox- collected in oscillation mode. Methods Enzymol. 276:307–326. idoreductase from the mesophilic bacterium Bacillus subtilis. Biochemistry 29. Patridge, E. V., and J. G. Ferry. 2006. WrbA from Escherichia coli and 45:7083–7091. Archaeoglobus fulgidus is an NAD(P)H:quinone oxidoreductase. J. Bacteriol. 9. Emsley, P., and K. Cowtan. 2004. Coot: model-building tools for molecular 188:3498–3506. graphics. Acta Crystallogr. D 60:2126–2132. 30. Sancho, J. 2006. Flavodoxins: sequence, folding, binding, function and be- 10. Gorman, J., and L. Shapiro. 2005. Crystal structures of the tryptophan yond. Cell Mol. Life Sci. 63:855–864. repressor binding protein WrbA and complexes with flavin mononucleotide. 31. Serre, L., F. Vellieux, M. Medina, C. Gomez Moreno, J. C. Fontecilla- Protein Sci. 14:3004–3012. Camps, and M. Frey. 1996. Crystal structures of a ferredoxin:NADPϩ re- 11. Grandori, R., and J. Carey. 1994. 6 New candidate members of the alpha/ ductase and of a complex with NADPϩ. Biochem. Soc. Trans. 24:10S. beta twisted open-sheet family detected by sequence similarity to flavodoxin. 32. Suzuki, Y., T. Yoda, A. Ruhul, and W. Sugiura. 2001. Molecular cloning and 3: Protein Sci. 2185–2193. characterization of the gene coding for azoreductase from Bacillus sp OY1-2 12. Grandori, R., P. Khalifah, J. A. Boice, R. Fairman, K. Giovanielli, and isolated from soil. J. Biol. Chem. 276:9059–9065. J. Carey. 1998. Biochemical characterization of WrbA, founding member of 33. Wolfova, J., R. Grandori, E. Kozma, N. Chatterjee, J. Carey, and I. K. a new family of multimeric flavodoxin-like proteins. J. Biol. Chem. 273: Smatanova. 2005. Crystallization of the flavoprotein WrbA optimized by 20960–20966. using additives and gels. J. Cryst. Growth 284:502–505. 13. Jensen, K. A., Z. C. Ryan, A. V. Wymelenberg, D. Cullen, and K. E. Hammel. 2002. An NADH:quinone oxidoreductase active during biodegradation by 34. Wolfova, J., J. R. Mesters, J. Brynda, R. Grandori, A. Natalello, J. Carey, the brown-rot basidiomycete Gloeophyllum trabeum. Appl. Environ. Micro- and I. Kuta Smatanova. 2007. Crystallization and preliminary diffraction biol. 68:2699–2703. analysis of Escherichia coli WrbA in complex with its cofactor flavin mono- 14. Jones, T. A., J.-Y. Zou, S. W. Cowan, and M. Kjelgaard. 1991. Improved nucleotide. Acta Crystallogr. F 63:571–575. methods for building proteins in electron density maps and location of errors 35. Wrobel, R. L., M. Matvienko, and J. I. Yoder. 2002. Heterologous expression in these models. Acta Crystallogr. A 47:110–119. and biochemical characterization of an NAD(P)H: quinone oxidoreductase 15. Karplus, P. A., M. J. Daniels, and J. R. Herriott. 1991. Atomic-structure of from the hemiparasitic plant Triphysaria versicolor. Plant Physiol. Biochem. ferredoxin-NADPϩ reductase: prototype for a structurally novel flavoen- 40:265–272. zyme family. Science 251:60–66. 36. Yang, W. P., L. Y. Ni, and R. L. Somerville. 1993. A stationary-phase protein 16. Kraulis, P. 1991. MOLSCRIPT: a program to produce both detailed and of Escherichia coli that affects the mode of association between the Trp schematic plots of proteins. J. Appl. Crystallogr. 24:946–950. repressor protein and operator-bearing DNA. Proc. Natl. Acad. Sci. USA 17. Laskowski, M. J., K. A. Dreher, M. A. Gehring, S. Abel, A. L. Gensler, and 90:5796–5800. I. M. Sussex. 2002. FQR1, a novel primary auxin-response gene, encodes a 37. Zhao, T., F. Cruz, and J. G. Ferry. 2001. Iron-sulfur flavoprotein (Isf) from flavin mononucleotide-binding quinone reductase. Plant Physiol. 128:578– Methanosarcina thermophila is the prototype of a widely distributed family. J. 590. Bacteriol. 183:6225–6233. APPENDIX C: TRYPTOPHAN REPRESSOR-BINDING PROTEINS FROM ESCHERICHIA COLI AND ARCHAEOGLOBUS FULGIDUS AS NEW CATALYSTS FOR 1,4-DIHYDRONICOTINAMIDE ADENINE DINUCLEOTIDE-DEPENDENT AMPEROMETRIC BIOSENSORS AND BIOFUEL CELLS C.1 Personal Contribution to the Manuscript C.2 Article as Published

Appendix C

Tryptophan Repressor-Binding Proteins from Escherichia coli and Archaeoglobus fulgidus as New Catalysts for 1,4-Dihydronicotinamide Adenine Dinucleotide-Dependent Amperometric Biosensors and Biofuel Cells

C.1 Personal Contributions to the Manuscript

The paper reproduced in this appendix describes a collaborative research effort between the research groups of Dr. James G. Ferry (Pennsylvania State University, USA),

Dr. Lo Gorton (Lund University, Sweden), and Gilbert Nöll (University of Siegen,

Germany). The paper was published in Analytical Chemistry1. My contributions to this manuscript included the preparation of protein utilized to couple electron transfer from

NADH to the electrode and assisted with development in the use of WrbA proteins with the gold capillary spectroelectrochemical cell.

1Reproduced with permission from Muhammad Nadeem Zafar, Federico Tasca, Lo Gorton, Eric V. Patridge, James G. Ferry, and Gilbert Nöll. 2009. Analytical Chemistry. doi:10.1021/ac900365n. Copyright 2009 American Chemical Society. 220

Anal. Chem. XXXX, xxx, 000–000

Muhammad Nadeem Zafar,† Federico Tasca,† Lo Gorton, Eric V. Patridge,‡ James G. Ferry,‡ and Gilbert No¨ll*,†,§

Department of Analytical Chemistry/Biochemistry, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden, Department of Biochemistry and Molecular Biology, Eberly College of Science, Pennsylvania State University, 205 South Frear Laboratory, University Park, Pennsylvania, 16802-4500, and University of Siegen, Organic Chemistry 1, Adolf-Reichwein-Strasse 2, D-57068 Siegen, Germany

The tryptophan (W) repressor-binding proteins (WrbA) of NAD+ to NADH, constitute the largest family of redox from Echerichia coli (EcWrbA) and Archaeoglobus fulgi- enzymes known today.1 In principle, a variety of amperometric dus (AfWrbA) were investigated for possible use in 1,4- biosensors can be realized by employing these enzymes. The dihydronicotinamide adenine dinucleotide (NADH) de- electrons gained from substrate oxidation are transferred to pendent amperometric biosensors and biofuel cells. NAD+ and have to be detected at the electrode in terms of EcWrbA and AfWrbA are oligomeric flavoproteins binding NADH. Unfortunately, the direct oxidation of NADH at bare one flavin mononucleotide (FMN) per monomer and electrodes requires a large overpotential and suffers from belonging to a new family of NAD(P)H:quinone oxi- pronounced irreversibility.1-3 Therefore a second catalytic doreductases (NQOs). The enzymes were covalently linked process becomes necessary. One strategy to mediate the electron to a low potential Os redox polymer onto graphite in the transfer (ET) from NAD+-dependent enzymes to electrodes presence of single-walled carbon nanotube (SWCNT) involves the covalent attachment of pyrroloquinoline quinone preparations of varying average lengths. The performance (PQQ) to a cystamine monolayer on gold followed by a covalent of the enzyme modified electrodes for NADH oxidation linkage of N6-(2-aminoethyl)-NAD+.2 Within this monolayer, the was strongly depending on the average length of the ET from the attached NADH to the gold surface is mediated applied SWCNTs. By blending the Os redox polymer with by PQQ. NAD+-dependent dehydrogenases such as lactate SWCNTs, the electrocatalytic current could be increased dehydrogenase can be assembled on top of the NAD+ contain- up to a factor of 5. Results obtained for AfWrbA modified ing layer.2 An advantage of this method is that NAD+ is electrodes were better than those for EcWrbA. For NADH immobilized at the surface, and a constant level of the NAD+/ detection, a linear range between 5 µM and 1 mM, a lower NADH redox couple in the bulk solution is not required. On limit of detection of 3 µM, and a sensitivity of 56.5 nA the other hand, this strategy is limited to enzyme monolayers µM-1 cm-2 could be reached. Additionally spectroelec- at the electrode. Alternatively, nitrocompounds immobilized on trochemical measurements were carried out in order nanostructured electrodes have been shown to catalyze the to determine the midpoint potentials of the enzymes oxidation of NADH at low overpotentials.4,5 Furthermore, (-115 mV vs NHE for EcWrbA and -100 mV vs NHE different types of carbon nanotubes (sometimes after further for AfWrbA pH 7.0). Furthermore, an AfWrbA modi- modification or combination with redox mediators) were found fied electrode was used as an anode in combination to catalyze the oxidation of NADH quite efficiently.6-19 Another with a Pt black cathode as a biofuel cell prototype. possibility is the combination of NAD+-dependent dehydroge-

(1) Gorton, L.; Domı´nguez, E. In Encyclopedia of Electrochemistry, Vol. 9, To catalyze the oxidation of 1,4-dihydronicotinamide adenine Biochemistry; Wilson, G. S., Ed.; Wiley-VCH: Weinheim, 2002; pp 67- dinucleotide (NADH) is an important task in bioanalytical chem- 143. (2) Bardea, A.; Katz, E.; Bu¨ckmann, A. F.; Willner, I. J. Am. Chem. Soc. 1997, + istry, because NAD -dependent dehydrogenases, which catalyze 119, 9114–9119. the oxidation of speciﬁc substrates with concomitant reduction (3) Barton, S. C.; Gallaway, J.; Atanassov, P. Chem. Rev. 2004, 104, 4867– 4886. * To whom correspondence should be addressed. Gilbert No¨ll, University of (4) Mano, N.; Kuhn, A. Biosens. Bioelectron. 2001, 16, 653–660. Siegen, Organic Chemistry 1, Adolf-Reichwein-Str. 2, D-57068 Siegen, Germany. (5) Mano, N.; Thienpont, A.; Kuhn, A. Electrochem. Commun. 2001, 3, 585– Phone: +49 (0)271 740 4360. Fax: +49 (0)271 740 3270. E-mail: [email protected] 589. siegen.de. (6) Li, X.; Zhou, H.; Yu, P.; Su, L.; Ohsaka, T.; Mao, L. Electrochem. Commun. † Lund University. 2008, 10, 851–854. ‡ Pennsylvania State University. (7) Luz, R. d. C. S.; Damos, F. S.; Tanaka, A. A.; Kubota, L. T.; Gushikem, Y. § University of Siegen. Electrochim. Acta 2008, 53, 4706–4714.

10.1021/ac900365n CCC: $40.75  XXXX American Chemical Society Analytical Chemistry, Vol. xxx, No. xx, Month XX, XXXX A nases with NAD(P)H:quinone oxidoreductases (NQOs) such EcWrbA and AfWrbA catalytic activity toward oxidation of NADH as diaphorase, a dimeric flavin adenin dinucleotide (FAD) has been reported,26 it is of interest to study these enzymes with containing protein, which oxidize NAD(P)H with concomitant respect to applications in NAD+/NADH-dependent amperomet- reduction of quinones or other redox mediators.20-23 On the ric biosensors and biofuel cells. Similar to diaphorase, EcWrbA basis of glucose dehydrogenase and diaphorase, a glucose and AfWrbA can be covalently linked to a low potential Os biosensor was developed,20 and by combination of diaphorase with redox polymer. The performance of both enzymes in the alcohol, aldehyde, and formate dehydrogenase, a biofuel cell presence of different types of SWCNTs was compared. Fur- capable of oxidizing methanol completely to CO2 and H2O has thermore, spectroelectrochemical measurements were carried been presented.21 In the latter case, benzylviologen was used out in order to determine the redox potentials of the enzymes. as a redox mediator in order to catalyze ET from diaphorase to the anode. In our previous work we have shown that EXPERIMENTAL PART diaphorase can be covalently linked to a low potential Os redox Chemicals and Materials. EcWrbA and AfWrbA were prepared polymer, which transfers the electrons gained from NADH as described elsewhere.26 Water was purified in a Milli-Q water oxidation to the electrode.24,25 When the Os polymer was purification system (Millipore, Bedford, MA). Poly(ethylene glycol) blended with oxidatively shortened and length-separated single- (400) diglycidyl ether (PEGDGE) was obtained from Aldrich (http:// walled carbon nanotubes (SWCNTs), the catalytic current for www.sigmaaldrich.com). Poly(vinylpyridine)-[osmium-(N,N′-methy- 24 NADH oxidation could be increased by a factor of 5. 2+/3+ lated-2,2′-biimidalzole)3] was synthesized as reported else- In this contribution, we report on the performance of the where.32 Single-walled carbon nanotubes (SWCNTs) were tryptophan (W) repressor-binding protein (WrbA) from Echerichia purchased from Nanocyl, Sambreville, Belgium. Triton X-100 and coli (EcWrbA) and Archaeoglobus fulgidus (AfWrbA) with respect controlled pore glass (CPG 3000 Å) were both from Fluka (Buchs, to applications in biosensors and biofuel cells.26-29 WrbA is an Switzerland). Spectrographic graphite electrodes and homemade oligomeric flavoprotein that binds one flavin mononucleotide FMN pyrolytic graphite electrodes were used. Pyrolytic graphite (PG) per monomer. The molecular mass of monomeric EcWrbA is 21 was obtained as a gift from Mr. Robert Pulley, Minerals Technolo- kDa and that of AfWrbA is 22 kDa.26 EcWrbA was discovered in gies (mineralstech.com). Spectrographic graphite electrodes were 1993, when it was copurified with the tryptophan repressor from Ringsdorff Werke GmbH, Bonn, Germany, (type RW001, (TrpR).30 For a long time, the function of WrbA was not known. 3.05 mm diameter and 13% porosity http://www.sglcarbon.com). With respect to its redox properties, a role in oxidative stress All solutions used for immobilization were prepared in Milli-Q 31 defense was implicated. In line with these findings, WrbA has water (Millipore, Bedford, MA), and the NADH used as a 26,28 been recently identified as a new family of NQOs. Since for substrate was dissolved in 0.1 M MOPS buffer solutions. For flow injection measurements, the working buffer solutions were de- (8) Manso, J.; Mena, M. L.; Yanez-Sedeno, P.; Pingarron, J. M. Electrochim. gassed before use to avoid air bubbles in the flow system. For Acta 2008, 53, 4007–4012. (9) Radoi, A.; Compagnone, D.; Valcarcel, M. A.; Placidi, P.; Materazzi, S.; voltammetric measurements (unless otherwise stated), argon was Moscone, D.; Palleschi, G. Electrochim. Acta 2008, 53, 2161–2169. purged through the solutions for some minutes prior to the (10) Du, P.; Liu, S.; Wu, P.; Cai, C. Electrochim. Acta 2007, 53, 1811–1823. experiments. Flow injection measurements were performed with (11) Chakraborty, S.; Retna Raj, C. Electrochem. Commun. 2007, 9, 1323–1330. 33 (12) Huang, M.; Jiang, H.; Qu, X.; Xu, Z.; Wang, Y.; Dong, S. Chem. Commun. a flow-through amperometric cell of the wall-jet type at an applied 2005, 5560–5562. potential of +290 mV vs NHE. The carrier flow was maintained (13) Zhang, M.; Gorski, W. J. Am. Chem. Soc. 2005, 127, 2058–2059. at a constant flow rate of 1 mL min-1 by a peristaltic pump. The (14) Valentini, F.; Salis, A.; Curulli, A.; Palleschi, G. Anal. Chem. 2004, 76, 3244– 34 3248. injection loop volume was 50 µL. The dispersion factor of the (15) Zhu, L.; Zhai, J.; Yang, R.; Tian, C.; Guo, L. Biosens. Bioelectron. 2007, 22, system was 1.04 at this flow rate. Voltammetric measurements 2768–2773. were performed with an EG&G potentiostat/galvanostat model (16) Raj, C. R.; Chakraborty, S. Biosens. Bioelectron. 2006, 22, 700–706. (17) Wang, J. Electroanalysis 2005, 17, 7–14. 273 A or an Autolab potentiostat/galvanostat PGSTAT30 (Eco (18) Yan, Y.-M.; Yehezkeli, O.; Willner, I. Chem.sEur. J. 2007, 13, 10168–10175. Chemie, Utrecht, The Netherlands) using modified electrodes as (19) Willner, I.; Katz, E. Angew. Chem., Int. Ed. 2000, 39, 1181–1218. the working electrode, a saturated calomel reference electrode (20) Antiochia, R.; Gorton, L. Biosens. Bioelectron. 2007, 22, 2611–2617. (21) Palmore, G. T. R.; Bertschy, H.; Bergens, S. H.; Whitesides, G. M. J. (SCE), and a platinum foil counter electrode. All potentials Electroanal. Chem. 1998, 443, 155–161. discussed in the main part are referred to the normal hydrogen (22) Gros, P.; Comtat, M. Biosens. Bioelectron. 2004, 20, 204–210. electrode (NHE). The current densities were calculated with (23) Montagne, M.; Durliat, H.; Comtat, M. Anal. Chim. Acta 1993, 278, 25– 33. respect to the geometric electrode area. (24) Tasca, F.; Gorton, L.; Wagner, J. B.; No¨ll, G. Biosens. Bioelectron. 2008, The spectroelectrochemical setup used in this work has been 24, 272–278. described elsewhere.35 As working electrode, a gold capillary cell (25) Nikitina, O.; Shleev, S.; Gayda, G.; Demkiv, O.; Gonchar, M.; Gorton, L.; Csoeregi, E.; Nistor, M. Sens. Actuators, B 2007, B125, 1–9. with an optical path length of 1 cm was applied. The extinction (26) Patridge, E. V.; Ferry, J. G. J. Bacteriol. 2006, 188, 3498–3506 -1 -1 . coefficients of the enzymes were ε450 ) 11.6 mM cm for (27) Andrade, S. L. A.; Patridge, E. V.; Ferry, J. G.; Einsle, O. J. Bacteriol. 2007, EcWrbA and ε ) 14.0 mM-1 cm-1 for AfWrbA.26 As redox 189, 9101–9107. 457 (28) Carey, J.; Brynda, J.; Wolfova, J.; Grandori, R.; Gustavsson, T.; Ettrich, R.; Smatanova, I. K. Protein Sci. 2007, 16, 2301–2305. (32) Mao, F.; Mano, N.; Heller, A. J. Am. Chem. Soc. 2003, 125, 4951–4957. (29) Wolfova, J.; Brynda, J.; Mesters, J. R.; Carey, J.; Grandori, R.; Smatanova, (33) Appelqvist, R.; Marko-Varga, G.; Gorton, L.; Torstensson, A.; Johansson, I. K. Mater. Struct. Chem., Biol., Phys. Technol. 2008, 15, 55–57. G. Anal. Chim. Acta 1985, 169, 237–247. (30) Yang, W.; Ni, L.; Somerville, R. L. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, (34) Ruzicka, J.; Hansen, E. H. In Flow Injection Analysis, 2nd ed.; Winefordner, 5796–5800. J. D., Ed.; John Wiley & Sons: New York, 1988; pp 23-26. (31) Natalello, A.; Doglia, S. M.; Carey, J.; Grandori, R. Biochemistry 2007, 46, (35) Bistolas, N.; Christenson, A.; Ruzgas, T.; Jung, C.; Scheller, F. W.; 543–553. Wollenberger, U. Biochem. Biophys. Res. Commun. 2004, 314, 810–816.

B Analytical Chemistry, Vol. xxx, No. xx, Month XX, XXXX mediators,36-38 2,6-dichloroindophenol, phenazine methosulfate, RESULTS AND DISCUSSION methylene blue, resazurin, 2-OH-1,4-naphthoquinone, anthraquino- Flow Injection Analysis and Evaluation of SWCNTs. The ne 2,6-disulfonate, safranine T, diquat, 1,1′-bis(hydroxyethyl)-4,4′- enzymes EcWrbA and AfWrbA were covalently linked to a low bipyridyl dichloride, and 1,1′-propylene-2,2′-bipyridylium dibromide potential Os redox polymer onto graphite electrodes and inves- were applied. tigated with respect to their ability to catalyze the oxidation of Electrode Preparation and Equipment. Spectrographic NADH. Similar to a previous study, the Os polymer was blended graphite electrodes were polished as reported previously.39 In this with oxidatively shortened and length-separated SWCNTs in order work, 0.7 mg of the different SWCNT preparations was dissolved to increase the electrocatalytic current. The SWCNTs were in 1 mL of Milli-Q water and sonicated overnight. Then 5 µLof shortened by acid treatment for a period of 12 h and subsequently dispersion was placed on the top of the polished electrode and separated by size exclusion chromatography with respect to their spread evenly using a microsyringe tip. Next, 2 µL of the osmium different average lengths. A total of 60 fractions of equal volume redox polymer (10 mg/mL in Milli-Q water) was mixed with 5 (15 mL) were collected. For flow injection analysis (FIA), the µL of SWCNT solution. Following this, 5 µL of enzyme (0.1 mg/ WrbA/Os redox polymer/SWCNT coated spectrographic graphite mL in Milli-Q water) was added to the mixture. Finally 1 µLof electrodes were placed in a flow through electrochemical cell and PEGDGE (1 mg/mL) was added. The electrode was then allowed examined in the flow injection mode. A constant potential of +290 to dry and placed overnight at 4 °C in a water saturated mV vs NHE24 was applied, and the NADH oxidation current was atmosphere for the complete cross-linking reaction to occur. measured at NADH concentrations of 0.5 and 1 mM (see Figure Electrodes not including SWCNTs were prepared as described 1A). above but without any addition of the SWCNT dispersions. As shown in Figure 1A, the highest catalytic current was Oxidative Shortening and Length Separation of SWCNTs. detected for AfWrbA/Os redox polymer modified electrodes In order to perform the oxidative shortening, a suspension of 200 blended with SWCNTs of fraction 10. The maximum current was mg of SWCNTs and a mixture of 6 mL of sulfuric acid and 2 mL 5 times higher than measured for electrodes prepared in the of nitric acid (98% and 70%, respectively) was sonicated for 12 h absence of SWCNTs (not shown). Similar results were obtained at 40 °C. The solution was then adjusted to pH 7 by adding a for EcWrbA/Os redox polymer modified electrodes. As reported solution of NaOH. Thereafter the solvent was removed by previously for diaphorase/Os redox polymer modified electrodes centrifugation. To remove amorphous carbon, a suspension of the blended with SWCNTs,24 the average length of the SWCNTs has SWCNTs in Piranha solution with a 4:1 ratio of sulfuric acid (98%) a strong influence on the magnitude of the electrocatalytic current. and hydrogen peroxide (30 wt %) (note, this solution has to be When AfWrbA and EcWrbA modified electrodes were compared ° treated with great care) was sonicated at 70 Cfor2h.The (see Figure 1B), the better performance (i.e., higher electrocata- solution was then again neutralized with NaOH and centrifuged. lytic current and higher linear current range) were found for Next, the SWCNTs were dissolved ina1wt%Triton X-100 Milli-Q AfWrbA. As shown in Figure 1C, for an AfWrbA/Os redox water solution and stabilized by sonication overnight. The disper- polymer/SWCNTs fraction 10 modified electrode, the linear range sion was length separated by size exclusion chromatography using for NADH detection was between concentrations of 5 µM and 1 a column filled with controlled pore glass (CPG 3000 Å). Milli-Q mM. The limit of detection (LOD) was at 3 µM, and a sensitivity water was used as the eluent. A total of 60 fractions of equal of 56.5 nA µM-1 cm-2 was calculated. Hence, the performance volume (15 mL) containing SWCNTs were collected. of the AfWrbA/Os redox polymer/SWCNT fraction 10 modified Cryogenic Transmission Electron Microscopy (CryoTEM). electrode is similar to that of a diaphorase/Os redox polymer/ Specimens for electron microscopy were prepared in a controlled SWCNT modified electrode (linear range, 5 µM-7 mM; LOD, environment vitrification system (CEVS) to ensure stable tem- - - 1 µM; sensitivity, 47.4 nA µM 1 cm 2), which has been perature and to avoid loss of solution during sample preparation. published previously.24 The specimens were prepared as thin liquid films, <300 nm thick, Cryo-Transmission Electron Microscopy. In order to obtain on lacey carbon filmed copper grids and plunged into liquid ethane some information about the morphology of the SWCNTs of at -180 °C. This leads to vitrified specimens, avoiding component fraction 10, cryo-transmission electron microscopy (cryoTEM) segmentation and rearrangement, and water crystallization, thereby measurements were carried out. CryoTEM was chosen as the preserving original microstructures. The vitrified specimens were appropriate method, because prior to conventional TEM measure- stored under liquid nitrogen until measured. An Oxford CT3500 ments, the solvent has to evaporate, and during this process, the cryoholder and its workstation were used to transfer the specimen SWCNTs show a strong tendency to agglomerate. In Figure 2, into the electron microscope (Philips CM120 BioTWIN Cryo, two cryoTEM images of SWCNTs of fraction 10 are presented. Amsterdam, The Netherlands) equipped with a postcolumn energy As obvious from Figure 2, the average length of the carbon filter (Gatan GIF100, Warrendal, PA). The acceleration voltage nanotubes is beyond 100 nm. Even though some SWCNTs are was 120 kV. The images were recorded digitally with a CCD separated, the majority of SWCNTs is arranged in small bundles. camera under low electron dose conditions. Therefore the determination of the average length is not possible. However, besides the presence of ice due to freezing of the (36) Dutton, P. L. Methods Enzymol. 1978, 54, 411–435. (37) Wilson, G. S. Methods Enzymol. 1978, 54, 396–410. samples, the SWCNT sample seems to be of high purity. No (38) No¨ll, G.; Hauska, G.; Hegemann, P.; Lanzl, K.; No¨ll, T.; von Sanden-Flohe, amorphous carbon or transition metal catalyst particles, which M.; Dick, B. ChemBioChem 2007, 8, 2256–2264. (39) Tasca, F.; Timur, S.; Ludwig, R.; Haltrich, D.; Volc, J.; Antiochia, R.; Gorton, are required for the manufacturing process of the SWCNTs, could L. Electroanalysis 2007, 19, 294–302. be traced.

Analytical Chemistry, Vol. xxx, No. xx, Month XX, XXXX C Figure 2. Two representative cryoTEM images of SWCNTs of fraction 10 at different magniﬁcations

redox centers with a midpoint potential of about 40 mV can be seen. At an NADH concentration of 1 mM, the current density reached its maximum at potential values of less than 100 mV (at this potential, the vast majority of the Os redox centers are in their oxidized state). This observation implies that the specific analytical detection of NADH should be possible even at a potential Figure 1. (A) Response of AfWrbA/Os redox polymer modified as low as 100 mV. Nevertheless, a potential of 290 mV as applied electrodes blended with different fractions of SWCNTs (fr 1, longest previously24 is sufficiently low to avoid nonspecific oxidation of average length; fr 50, shortest) after injection of 0.5 and 1 mM NADH. interfering compounds. An important difference in the perfor- (B) Comparison of the response of an AfWrbA and an EcWrbA/Os mance of EcWrbA and AfWrbA becomes obvious, when the CVs redox polymer modified electrode, blended with SWCNTs of fraction 10. (C) NADH calibration curves for an AfWrbA/Os redox polymer measured at NADH concentrations of 10 mM are compared. For modified electrode blended with SWCNTs of fraction 10 in the EcWrbA, the catalytic current density at 100 mV is only moderate concentration range from 5 µM to 10 mM NADH. Inset: Concentration (J ) 70 µAcm-2) but increases further when the potential is range from 5 to 100 µM. Experiments were performed in 0.1 M MOPS raised to more positive values (see Figure 3A). In contrast, for buffer at pH 7.5. AfWrbA at a potential slightly below 100 mV, the catalytic current density has reached a maximum of about 240 µAcm-2 (deter- Cyclic Voltammetry. NADH is an expensive fuel for any mined by subtracting the capacitive current) and remains at biofuel cell application, but it can be recovered by the oxidation this high level when the potential is further increased (see of cheaper substrates such as alcohols.21 While for biosensors Figure 3B). A similar tendency was observed when the electrode usually high sensitivity and substrate specificity, fast response material was changed from spectrographic graphite to pyrolytic time, and a low limit of detection are desired, for biofuel cell graphite. The difference in behavior between EcWrbA and applications high current density and good long-term stability of AfWrbA, as seen in parts A and B of Figure 3, could be ascribed the enzyme modified electrodes are required. In order to study to a suggested difference in flavin affinities. It has been noted the EcWrbA and AfWrbA/Os redox polymer/SWCNTs fraction that the flavin in EcWrbA is lost during purification while that of 10 modified electrodes with respect to biofuel cell applications, AfWrbA is not.26 Perhaps the behavior of EcWrbA in Figure 3A cyclic voltammograms (CVs) were measured at low scan rate (v is indicative of partial cofactor dissociation.26,40 EcWrbA and ) 0.5 mV s-1) for different concentrations of NADH and in the absence of substrate (see Figure 3). In the CVs measured in (40) Ji, H.-F.; Shen, L.; Carey, J.; Grandori, R.; Zhang, H.-Y. THEOCHEM 2006, the absence of substrate, the oxidation and reduction of the Os 764, 155–160.

D Analytical Chemistry, Vol. xxx, No. xx, Month XX, XXXX Figure 3. CVs of an EcWrbA/Os redox polymer/SWCNTs fraction Figure 4. Spectroelectochemical titration curves of EcWrbA (A) and 10 modified spectrographic graphite electrode (A) and an AfWrbA/ AfWrbA (B) measured in the presence of a mixture of redox mediators. Os redox polymer/SWCNTs fraction 10 modified spectrographic Depending on their redox states, the mediators are contributing to graphite electrode (B) measured at a scan rate of v ) 0.5 mV s-1 in the optical spectra to some extent (the change in absorption between the presence of different concentrations of NADH and in the absence 550 and 700 nm is mainly caused by the mediator mixture). For both of substrate. Experiments were performed in 0.1 M MOPS buffer at enzymes, the complete set of spectra collected during ongoing pH 7.5. reduction is shown. The missing absorption in the spectra around 665 nm is an error of the instrument caused by the deuterium light source. Experiments were performed at 0.1 M KCl in 0.1 M MOPS AfWrbA exist as oligomers, and also structural differences between buffer, pH 7.0. both enzymes in the oligomeric state might affect the catalytic activity. Mass spectrometry has shown that FMN promotes stable negatively charged flavosemiquinone radical anion or a EcWrbA association into tetramers, which are more stable than neutral flavosemiquinone radical within the time scale of the 31 38,41 dimers or monomers. High NADH concentration might have a experiment. With dependence on the pKa of the doubly similar or a conformational effect on EcWrbA. reduced flavohydroquinone, a second protonation step may In order to check the long-term stability of an AfWrbA/Os follow the second ET leading to the flavohydroquinone in its redox polymer/SWCNTs fraction 10 modified electrode, multi- neutral form. Any differences in the redox behavior of both - cycle CVs at a scan rate of 0.1 mV s 1 were measured over a enzymes, which could explain the increasing catalytic activity period of 20 h. Within 20 h, the decrease in electrocatalytic for EcWrbA with increasing redox potential at high NADH current density was about 10%. concentration, could not be observed. Spectroelectrochemistry. In order to gain more information Biofuel Cell Performance. As model for a membraneless about the redox properties of EcWrbA and AfWrbA and to biofuel cell, an AfWrbA/Os redox polymer/SWCNTs fraction 10 determine the midpoint potentials of both enzymes, spectroelec- modified electrode was applied as the anode together with a Pt trochemical measurements were carried out. In Figure 4, the black electrode as the cathode in a solution of NADH (8 mM) in spectral changes during the reduction of EcWrbA and AfWrbA MOPS buffer (0.1 M) at pH 7.5. Oxygen was gently purged around are shown. There is a minor experimental error (±20 mV), mainly the cathode. Since the area of the cathode was much larger than due to the presence of redox mediators, which contribute to the that of the anode, the current density of this cell was limited by 36-38 absorption spectra depending on their redox states. The the anode. After an equilibration time of 500 s, a polarization curve presence of mediators was required in order to establish electro- was collected using linear sweep voltammetry (v ) 0.1 mV s-1) chemical equilibrium. Within the experimental error, the same connecting the anode as the working, and the cathode as the midpoint potentials were measured during reduction and reoxi- reference and counter electrodes. The polarization curve and dation. As midpoint potential, values of -115 mV for EcWrbA and the dependence of the power density on the operating voltage -100 mV for AfWrbA at pH 7.0 were determined as an average are presented in parts A and B of Figure 5. The cell exhibited of three independent experiments performed at varying redox a maximum voltage (Vmax) of 300 mV, a maximum current mediator concentrations. The reduction of both types of WrbA was a two electron one proton reduction without formation of a (41) No¨ll, G. J. Photochem. Photobiol., A 2008, 200, 34–38.

Analytical Chemistry, Vol. xxx, No. xx, Month XX, XXXX E -150 mV simulating a variable load (see Figure 5C).42 In this period, the current density decreased about 65%. Possibly, some material was desorbed from the electrode surface due to the oxygen purging/nonquiescent conditions. Similar to other types of biofuel cell anodes,42 also for the AfWrbA/Os redox polymer/ SWCNTs fraction 10 modified electrode, the strongest decrease in activity was observed during the first hours of performance. In another stability experiment, the Pt electrode was cleaned and the NADH/buffer solution was replaced by a freshly prepared NADH/buffer solution, after the current density had decreased within the first hours of performance. This did not increase the catalytic current density. Hence, the maximum current density is limited by the AfWrbA/Os redox polymer/SWCNTs fraction 10 modified anode but not by limited stability of NADH43 leading to degradation products poisoning the Pt electrode.

CONCLUSIONS In this work, the NQOs EcWrbA and AfWrbA were investigated with respect to applications in biosensors and biofuel cells. For this purpose, the enzymes were covalently linked to a low potential Os redox polymer in the presence of oxidatively shortened and length-separated SWCNTs. By blending the Os redox polymer with SWCNTs, the electrocatalytic current could be increased up to a factor of 5. In line with previous studies on the NQO diaphorase,24 in the current study a dependence of the electrocatalytic current on the average length of the SWCNTs was also found. The highest current was obtained with SWCNTs of fraction 10. When AfWrbA and EcWrbA/Os redox polymer/SWCNT fraction 10 modified electrodes were compared, the better results were obtained with AfWrbA. Also during cyclic voltammetry experiments, a higher catalytic current could be obtained, and the maximum current was reached at lower potential for AfWrbA than for EcWrbA. The AfWrbA based NADH biosensor exhibited excellent performance. The linear range for NADH detection by an AfWrbA/Os redox polymer/SWCNT fraction 10 modified electrode was between concentrations of 5 µM and 1 mM, the Figure 5. Polarization curve (A) (measured with linear sweep LOD was at 3 µM, and a sensitivity of 56.5 nA µM-1 cm-2 was voltammetry by scanning from 0 to -350 mV with a scan rate of 0.1 calculated. These parameters are in the same range as reported -1 mV s after an equilibration time of 500 s) and dependence of the previously for other types of enzymatic NADH biosensors.24 power density on the operating voltage (B) for a membraneless biofuel For a diaphorase/Os redox polymer/SWCNT modified elec- cell consisting of an AfWrbA/Os redox polymer/SWCNTs fraction 10 modified electrode (basal oriented) as the anode and a Pt black trode (linear range, 5 µM-7 mM), a slightly lower LOD (1 electrode as the cathode. As fuel, a solution of NADH (8 mM) in µM) but also a somewhat lower sensitivity (47.4 nA µM-1 cm-2) MOPS buffer (0.1 M, pH 7.5) was used. For stability measurements, has been reported.24 Hence, AfWrbA turned out to be equally ) multicycle CVs (39 cycles) were collected at a scan rate of v 0.1 suited for the development of an NADH sensor as diaphorase mV s-1 in the potential range between -50 and -150 mV in order to (from Bacillus stearothermophilus). Possibly, future types of simulate a variable load. WrbA may be even more suited to catalyze the electrochemical oxidation of NADH than AfWrbA. -2 density (J max)of105µAcm , and a maximum power density In order to study EcWrbA and AfWrbA more in detail, -2 (Pmax)of12µWcm at an operating voltage of 165 mV (under spectroelectrochemical measurements were performed. Within the oxygen purging/nonquiescent conditions). A fill factor of 0.38 experimental error, the same midpoint potentials at pH 7.0 (-115 was calculated by dividing Pmax by the product of Vmax and Jmax. mV for EcWrbA and -100 mV for AfWrbA) were measured during The fill factor is the ratio of the experimentally determined reduction and reoxidation. For both types of WrbA, a two electron maximum power density divided by the maximum power one proton reduction without formation of a stable negatively density, which can be reached theoretically for a biofuel cell charged flavosemiquinone radical anion or a neutral flavosemi- with a given Vmax and Jmax under ideal conditions. In order to prove the stability, the biofuel cell was investigated (42) Tasca, F.; Gorton, L.; Harreither, W.; Haltrich, D.; Ludwig, R.; No¨ll, G. J. Phys. Chem. C 2008, 112, 13668–13673. for a period of 21 h and 40 min by multicycle CVs at low scan (43) Chenault, H. K.; Whitesides, G. M. Appl. Biochem. Biotechnol. 1987, 14, - rate (v ) 0.1 mV s 1) in a potential range between -50 and 147–197.

F Analytical Chemistry, Vol. xxx, No. xx, Month XX, XXXX quinone radical was observed. Furthermore, an AfWrbA/Os redox ACKNOWLEDGMENT polymer/SWCNTs fraction 10 modiﬁed electrode was applied as M.N.Z. and F.T. contributed equally to this work. This work the anode together with a Pt black electrode as the cathode as a was supported by the Deutsche Forschungsgemeinschaft DFG model for a membraneless biofuel cell using NADH as the (DFG Postdoctoral Fellowship NO 740/1-1 and Ru¨ckkehrstipen- substrate. This cell exhibited a maximum voltage of 300 mV, a dium NO 740/3-1), the Swedish Research Council (Projects 621- maximum current density of 105 µA · cm-2, and a maximum 2004-4476 and 621-2007-4124), and the “Higher Education Com- power density of 12 µWcm-2 at an operating voltage of 165 missionofPakistan”.WethankGunnelKarlssonattheBiomicroscopy mV (under oxygen purging/nonquiescent conditions). In Unit, Polymer and Materials Chemistry, Institute of Chemistry, contrast to multicycle CV experiments (with a decrease of about Lund University, Lund, Sweden, for performing the cryoTEM 10% in catalytic current density within 20 h), there was a work. stronger decrease in activity (65% of decrease in 21 h and 40 min) during the biofuel cell experiments. Possibly, this was due to limited mechanical stability of the electrocatalytically Received for review February 17, 2009. Accepted March active layer at the electrode surface under oxygen purging/ 24, 2009. nonquiescent conditions. AC900365N

Analytical Chemistry, Vol. xxx, No. xx, Month XX, XXXX G Eric Vincent Patridge Curriculum Vitae

EDUCATION Ph.D. Integrative Biosciences (Chemical Biology Option) May 2009 The Pennsylvania State University. Pennsylvania, USA.

Bachelor of Arts, cum laude May 2001 Dual Major: Chemistry & Biology (Molecular and Cellular Concentration) Skidmore College. New York, USA.

PUBLICATIONS Patridge EV, Ferry JG. Co-factor Binding and Redox Potential Modulation in WrbA Proteins, Class IV NAD(P)H:Quinone Oxidoreductases. (in process) Patridge EV, Ferry JG. Review: Flavodoxin-like NAD(P)H:Quinone Oxidoreductases Functioning in Detoxification and Redox-Linked Proteolysis. (in process) Nadeem MZ, Tasca F, Gorton L, Patridge EV, Ferry JG, Nöll G. WrbA from Escherichia coli and Archaeoglobus fulgidus as New Catalyst for NADH-Dependent Amperometric Biosensors and Biofuel Cells. (accepted to Analytical Chemistry) Andrade SL, Patridge EV, Ferry JG, Einsel O. The Crystal Structure of NAD(P)H:Quinone Oxidoreductase WrbA from Escherichia coli. Journal of Bacteriology. (2007) 189, 9101-7. Patridge EV, Ferry JG. WrbA from Escherichia coli and Archaeoglobus fulgidus is an NAD(P)H: Quinone Oxidoreductase. Journal of Bacteriology. (2006) 188, 3498-506.

TEACHING EXPERIENCE The Pennsylvania State University Biology: Function and Development of Organisms (BIOL 240W) 2003 Biology: Molecules and Cells (BIOL 230W) 2002

HONORS AND DISTINCTIONS Susan R. Rankin Award, Penn State University 2007 Outstanding Service Award, Penn State University 2006 Lambda Alumni Outstanding Student Award, Penn State University 2006 Life Science Consortium Graduate Fellowship, Penn State University 2001 Honors in Biology, Skidmore College 2001 Periclean Honors Society, Skidmore College 2001 Eagle Scout, Boy Scouts of America 1995

PROFESSIONAL MEMBERSHIP AND LEADERSHIP EXPERIENCE Member: American Association for Advancement of Science (AAAS) Member: Natl Org of GL Scientists and Technical Professionals (NOGLSTP) Member: Eberly College of Science Climate & Diversity Committee 2009 Member: Biochemistry & Molecular Biology Climate & Diversity Committee 2009 Co-Chair: Presidential Commission on LGBT Equity, Penn State University 2008 Co-Director: Coalition of LGBTA Graduate Students, Penn State University 2006 Founder: Out in Science, Technology, Engineering & Mathematics (oSTEM) 2005 Head Resident: Office of Residential Life, Skidmore College 2001 Founder: Students, Staff, and Faculty for Equality, Skidmore College 2000

GRANTS AND FUNDING Graduate Researcher: Natural Materials, Systems & Extremophiles 2007-2009