UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Structural basis for differential electron flux in human methionine synthase reductase and cytochrome… Meints, Carla Erin 2015

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

Item Metadata

Download

Media
24-ubc_2015_may_meints_carla.pdf [ 8.27MB ]
Metadata
JSON: 24-1.0074430.json
JSON-LD: 24-1.0074430-ld.json
RDF/XML (Pretty): 24-1.0074430-rdf.xml
RDF/JSON: 24-1.0074430-rdf.json
Turtle: 24-1.0074430-turtle.txt
N-Triples: 24-1.0074430-rdf-ntriples.txt
Original Record: 24-1.0074430-source.json
Full Text
24-1.0074430-fulltext.txt
Citation
24-1.0074430.ris

Full Text

Structural basis for differentialelectron flux in human methioninesynthase reductase and cytochromeP450 reductasebyCarla Erin MeintsB.Sc., University of British Columbia, 2010A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE COLLEGE OF GRADUATE STUDIES(Biochemistry and Molecular Biology)THE UNIVERSITY OF BRITISH COLUMBIA(Okanagan)April 2015c© Carla Erin Meints, 2015AbstractHuman diflavin oxidoreductases, methionine synthase reductase (MSR) and cytochromeP450 reductase (CPR), share overall structural organization, substrate specificity, and di-rection of electron flow. However, MSR exhibits repressed catalytic activity and weaksubstrate binding compared to CPR. In this study, we identified structural features thatcontrol the differential kinetic properties of these flavoenzymes. Single point mutations inthe MSR and CPR active sites were studied by steady state and pre-steady state absorbancespectrophotometric techniques to reveal structure-function relationships.In the research first chapter, we investigated the role of a conserved tryptophan inMSR that lies coplanar with the FAD cofactor. Reducing the side chain size resulted ina 1.5- and >400-fold tighter NADPH binding affinity in W697H and W697S and loweredthe preference for NADPH over NADH. W697H also accelerated flavin reduction by 4.6-fold. Thus, the energetic cost of Trp697 displacement by the nicotinamide ring presents anenergy barrier to NADPH binding. Consequently, Trp697 regulates coenzyme affinity, andcoenzyme preference by ensuring initial recognition of the 2’,5’-ADP moiety. The secondresearch chapter evaluated the effect of more conservative mutations, W697Y and W697F,that resulted in improved coenzyme binding affinity and 9-fold accelerated flavin reduction.Equivalent mutations in CPR slowed flavin reduction. Thus, Trp697 gates hydride transferin MSR, but not in CPR. Instead, following hydride transfer, electron flow in CPR islimited by displacement of the oxidized nicotinamide ring. This is supported by the 46-fold faster flavin reduction of the 2’-phosphate-binding site variant K602AV603K whichweakened coenzyme affinity.In the third research chapter, we analyzed the role of FAD proximal His322 which iswithin hydrogen bonding distance to the catalytic triad residue Asp674 in CPR. The equiv-alent residue in MSR is Ala312. Through reciprocal mutagenesis, we found that H322A hadstronger affinity for NADPH but slower flavin reduction, while A312H weakened coenzymeaffinity. His322 is proposed to weaken coenzyme binding affinity by competing with thenicotinamide ring for electrostatic interaction with Asp674, thereby accelerating NADP+release and electron flow through CPR.The final research chapter shifted to the conserved FAD-stacking tryptophan residue(Trp704) in plant CPR from Artemisia annua (AaCPR) for an additional perspective oniiAbstractoverall CPR catalysis. The steady state and pre-steady state data revealed that Trp704also triggers NADP+ release, however this step is not as critical for electron flux in AaCPRcompared to human CPR.iiiPrefaceThe work presented in Chapter 2 has been published: [Carla E. Meints], Frida S. Gustafs-son, Nigel S. Scrutton, and Kirsten R. Wolthers (2010). Tryptophan 697 modulates hydrideand interflavin electron transfer in human methionine synthase reductase. Biochemistry. 50:11131-11142. I designed, expressed, and purified the enzyme variants. I also measured allthe steady state and pre-steady state kinetics for each variant via UV-visible spectropho-tometry and stopped-flow experiments. Frida Gustafsson was an undergraduate studentworking with me on the project. The manuscript was drafted by Dr. Kirsten R. Woltherswith editorial comments by Dr. Nigel Scrutton.A portion of Chapter 3 has been published with the exception of the coenzyme-bindingresidue mutations: [Carla E. Meints], Svetlana Simtchouk, Kirsten R. Wolthers (2013). Aro-matic substitution of the FAD-shielding tryptophan reveals its differential role in regulatingelectron flux in methionine synthase reductase and cytochrome P450 reductase. FEBS Jour-nal. 280: 1460-1474. Svetlana Simtchouk generated and expressed the tryptophan variantsof CPR and conducted some of the kinetic experiments. I carried on the work by designing,expressing and purifying all other variants. I measured all the steady state and pre-steadystate kinetics for each variant via UV-visible spectrophotometry and stopped-flow methods.I also measured the redox potentials of specified CPR variants. I analyzed the results andcontributed to the discussion and overall preparation of the manuscript with Dr. KirstenR. Wolthers.The work of Chapter 4 has been published with the exception of the MSR hinge data.[Carla E. Meints], Sarah M. Parke, and Kirsten R. Wolthers (2014). Proximal FAD histidineresidue influences interflavin electron transfer in cytochrome P450 reductase and methioninesynthase reductase. Archives of Biochemistry and Biophysics. 547: 18-26. Sarah M. Parkewas an undergraduate student that assisted me on the project. I designed, expressed,and purified each variant. I also conducted the steady state and pre-steady state kineticexperiments. With the assistance of Sarah, the redox potentials of the specified variantswere determined. I analyzed the results and contributed to manuscript preparation withDr. Kirsten R. Wolthers.Chapter 5 was based on work conducted by myself in Dr. Kirsten R. Wolthers’ enzy-mology lab. I designed and expressed the variants as well as performed all the steady stateivPrefaceand pre-steady state kinetic experiments through UV-visible and stopped-flow spectropho-tometric methods. I compiled, analyzed the results, and wrote the chapter with editorialcomments from Dr. Kirsten R. Wolthers.I also co-authored the following paper during my PhD study: Svetlana Simtchouk,Jordan L. Eng, [Carla E. Meints], Caitlyn Makins, and Kirsten R. Wolthers (2013). Kineticanalysis of cytochrome P450 reductase from Artemisia annua reveals accelerated rates ofNADPH-dependent flavin reduction. FEBS Journal. 280: 6627-6642. I was involved in thepre-steady state kinetic experimentation and analysis, as well as figure preparation.v vi Table of Contents  Abstract  ..................................................................................  ii Preface  ....................................................................................  iv Table of Contents  ...................................................................  vi    List of Tables  ..........................................................................  x List of Figures  .........................................................................  xii List of Abbreviations ................................................................  xvi Acknowledgments  ...................................................................  xviii Dedication  ...............................................................................  xix Chapter 1: Introduction  ..........................................................  1 1.1 Flavin-dependent Enzymes  ....................................................  1 1.1.1 Biological Prevalence and Diversity  ........................  1 1.1.2 Flavin Discovery  ......................................................  1 1.1.3 Biological Sources of Flavins  ...................................  2 1.1.4 Flavin Structure  ......................................................  4 1.1.5 Common Flavin-Catalyzed Reactions  ......................  7 1.1.6 Unusual Flavin-Catalyzed Reactions  .......................  8 1.1.7 Ferredoxin NADP(H)-dependent Reductase, an Oxidoreductase  ........................................................  10 1.1.8 The Diflavin Oxidoreductase Family  .......................  12 1.2 Cytochrome P450 Reductase  .................................................  15 1.2.1 Structure  ..................................................................  15 1.2.2 Enzymology  .............................................................  19 1.2.3 Physiological Role  ....................................................  21 1.2.4 Health Implications  .................................................  23 1.3 Methionine Synthase Reductase  ............................................  24 1.3.1 Structure  ..................................................................  24 1.3.2 Enzymology  .............................................................  26 1.3.3 Physiological Role  ....................................................  27 1.3.4 Health Implications  .................................................  28 Table of Contents   vii 1.4 Research Aims  .......................................................................  29  Chapter 2: Probing the Regulatory Role of an Active Site               Trp697 in Methionine Synthase Reductase .............  31 2.1 Summary  ...............................................................................  31 2.2 Background  ............................................................................  31 2.3 Results  ...................................................................................  33 2.3.1 Steady state kinetic data  .........................................  33 2.3.2 Uncoupled NADPH oxidation  .................................  35 2.3.3 Coenzyme Preference  ...............................................  35 2.3.4 Multiple wavelength pre-steady state kinetics  .........  36 2.3.5 Single wavelength pre-steady state kinetics  .............  39 2.4 Discussion  ..............................................................................  40 2.4.1 Regulation of coenzyme preference by Trp697  ........  40 2.4.2 Regulation of hydride transfer by Trp697  ...............  43 2.4.3 Regulation of interflavin electron transfer by Trp697 44 2.4.4 Extension of regulatory role of tryptophan to CPR .  45 2.5 Experimental Procedures  .......................................................  46 2.5.1 Materials  ..................................................................  46 2.5.2 Generation and expression of MSR tryptophan  variants  ....................................................................  46 2.5.3 Purification of MSR variants  ...................................  46 2.5.4 Steady state turnover analysis  .................................  48 2.5.5 Pre-steady state kinetic analysis  ..............................  48  Chapter 3: Comparative investigation into the role of coenzyme-coordinating residues and the FAD-shielding tryptophan in coenzyme binding and electron flux in MSR and CPR  ....................................................  50 3.1 Summary  ...............................................................................  50 3.2 Background  ............................................................................  51 3.3 Results  ...................................................................................  53 3.3.1 Steady state kinetic data  .........................................  53 3.3.2 Multiple wavelength pre-steady state kinetics of  CPR variants  ...........................................................  54 3.3.3 Primary kinetic isotope effect with  (R)-[4-2H]-NADPH on CPR reduction  .....................  55 3.3.4 Thermodynamic analysis of CPR variants  ..............  57 3.3.5 Multiple wavelength pre-steady state kinetics of  MSR variants  ..........................................................  60 Table of Contents   viii 3.3.6 Primary kinetic isotope effect with  (R)-[4-2H]-NADPH on MSR reduction  ....................  60 3.3.7 Steady state kinetic data of the coenzyme-binding variants of MSR and CPR  ......................................  62 3.3.8 Multiple and single wavelength pre-steady state  kinetics of coenzyme-binding variants of MSR  and CPR  ..................................................................  64 3.4 Discussion  ..............................................................................  68 3.4.1 Hydride transfer is gated by Trp697 in MSR but not  in CPR  ....................................................................  68 3.4.2 Interflavin ET influenced by residue polarity in MSR  and CPR  ..................................................................  70 3.4.3 Trp676 displaces the oxidized nicotinamide ring and  as such controls electron flow  ..................................  70 3.4.4 Differential influence of coenzyme binding affinity on CPR and MSR flavin reduction kinetics  .................  71 3.5 Experimental Procedures  .......................................................  72 3.5.1 Materials  ..................................................................  72 3.5.2 (R)-[4-2H]-NADPH synthesis and purification  .........  73 3.5.3 Generation and expression of CPR and MSR   tryptophan variants  .................................................  73 3.5.4 Purification of MSR variants  ...................................  74 3.5.5 Purification of CPR variants  ...................................  74 3.5.6 Steady state turnover analysis  .................................  75 3.5.7 Pre-steady state kinetic analysis  ..............................  75 3.5.8 Redox potentiometry  ...............................................  76  Chapter 4: Role of histidine residue at the FAD active site  in regulating intramolecular electron flow in  CPR and MSR .....................................................  78 4.1 Summary  ...............................................................................  78 4.2 Background  ............................................................................  78 4.3 Results  ...................................................................................  81 4.3.1 Flavin characteristics of variants  .............................  81 4.3.2 Steady state kinetic data  .........................................  83 4.3.3 Multiple and single wavelength pre-steady state  kinetics of CPR variants  .........................................  84 4.3.4 Multiple and single wavelength of isolated FAD  domain of wild-type and H322A CPR  ....................  85 4.3.5 Redox Potentiometry of H322A  ..............................  88 Table of Contents   ix 4.3.6 Multiple and single wavelength pre-steady state  kinetics of MSR variants  .........................................  91 4.3.7 Kinetic parameters of the truncated hinge variant of  MSR  ........................................................................  92 4.4 Discussion ...............................................................................  95 4.4.1 Active site His322 weakens coenzyme binding  ........  95 4.4.2 His322 promotes interflavin electron transfer  ..........  97 4.4.3 Differential effect of introducting histidine into MSR active site  ................................................................  98 4.4.4 Role of the extended hinge in MSR  .........................  98 4.5 Experimental Procedures  .......................................................  99 4.5.1 Materials  ..................................................................  99 4.5.2 Generation and expression of MSR and CPR  variants  ....................................................................  99 4.5.3 Generation and expression of the isolated  FAD-NADP(H) domain of CPR  .............................  99 4.5.4 Generation of MSR truncated hinge  ........................  100 4.5.5 Purification of MSR variants  ...................................  102 4.5.6 Purification of CPR variants  ...................................  102 4.5.7 Steady state turnover analysis  .................................  103 4.5.8 Pre-steady state kinetic analysis  ..............................  103 4.5.9 Redox Potentiometry  ..............................................  103  Chapter 5: Conservation of an aromatic FAD-shielding  residue, Trp704, in Artemisia annua  Cytochrome P450 Reductase ................................  104 5.1 Summary  ...............................................................................  104 5.2 Background  ............................................................................  104 5.3 Results  ...................................................................................  106 5.3.1 Steady state kinetic parameters  ...............................  106 5.3.2 Multiple and single wavelength pre-steady state  kinetics  ....................................................................  107 5.4 Discussion  ..............................................................................  108 5.5 Experimental Procedures  .......................................................  115 5.5.1 Materials  ..................................................................  115 5.5.2 Generation and expression of AaCPR variants  .......  115 5.5.3 Steady state turnover analysis  .................................  115 5.5.4 Pre-steady state kinetic analysis  ..............................  115  Chapter 6: Conclusion  .............................................................  117 Bibliography .............................................................................  122	   x	  List of Tables   Table 2.1 Steady state kinetic parameters of wild-type and variant  MSR with NADPH .................................................................  34 Table 2.2 Steady state kinetic parameters of wild-type and W697 MSR variants with NADPH ............................................................  36 Table 2.3 Observed rate constants of the pre-steady state reduction of  MSR variants ..........................................................................  37 Table 2.4 Oligonucleotide primers designed for MSR C-terminal  variants ...................................................................................  47  Table 3.1 Steady state kinetic parameters of MSR W697 and CPR  W676 variants .........................................................................  53 Table 3.2 Observed rate constants of the pre-steady state reduction of  CPR W676 variants ................................................................  55 Table 3.3 Observed rate constants and kinetic isotope effects for the  pre-steady state reduction of CPR W676 variants monitored  at 454 and 600 nm by single-wavelength stopped-flow spectrophotometry ..................................................................  58 Table 3.4 Reduction potentials of the flavin couples in W676Y and  W676F CPR variants .............................................................  59 Table 3.5 Observed rate constants of the pre-steady state reduction of  MSR W697 variants ................................................................  60 Table 3.6 Observed rate constants and kinetics isotope effects for the  pre-steady state reduction of MSR W697Y and W697F  variants monitored at 454 and 600 nm by single-wavelength stopped-flow spectrophotometry .............................................  62 Table 3.7 Steady state kinetic parameters of coenzyme-binding residue variants of CPR and MSR ......................................................  64 Table 3.8 Observed rate constants for flavin reduction in CPR and MSR coenzyme-binding residue variants acquired from multi- wavelength stopped-flow data .................................................  66 Table 3.9 Observed rate constants for the flavin reduction in CPR and  MSR coenzyme-binding residue variants acquired from single wavelength 454 nm stopped-flow data ....................................  67 List of Tables 	   xi	  Table 3.10 Oligonucleotide primers for each specified MSR and CPR  variant ....................................................................................  74  Table 4.1 Steady state kinetic parameters of wild-type, A312 and H322 variants of MSR and CPR ......................................................  83 Table 4.2 Observed rate constants of the pre-steady state reduction of  CPR H322 variants .................................................................  85 Table 4.3 Observed rate constants of the pre-steady state NADPH  reduction of MSR A312 variants by stopped-flow spectrophotometry ..................................................................  92 Table 4.4 Oligonucleotide primers for each specified MSR and CPR  variant ....................................................................................  100 Table 4.5 Recursive PCR oligonucleotide primers designed for synthesis  of wild-type MSR FMN-domain with CPR hinge ...................  101  Table 5.1 Steady state kinetic parameters of AaCPR W704 variants ....  107 Table 5.2 Observed rate constants of the pre-steady state reduction of AaCPR W704 variants at 454 nm ..........................................  109 Table 5.3 The forward (F) and reverse (R) oligonucleotide sequences for W704 variants of AaCPR .......................................................  116	   xii	  List of Figures   Figure 1.1 Diversity of flavin-dependent biochemical processes ............  2 Figure 1.2 Chemical structure of riboflavin, flavin mononucleotide  (FMN) and flavin adenosine dinucleotide (FAD) .................  3 Figure 1.3 Biosynthetic pathway of FMN and FAD from the riboflavin precursor ...............................................................................  5 Figure 1.4 Reductive and oxidative half-reactions of flavin isoalloxazine  ring .......................................................................................  6 Figure 1.5 Spectral profile of the flavoenzyme cytochrome P450  reductase in the oxidized (solid), semiquinone (dotted) and  fully reduced states (dashed) ................................................  6 Figure 1.6 Oxidative and reductive half-reactions of typical  flavoenzymes .........................................................................  7 Figure 1.7 Proposed catalytic mechanism of acyl-CoA dehydrogenase ..  8 Figure 1.8 Proposed catalytic mechanism of D-amino acid oxidase ......  9 Figure 1.9 Proposed catalytic mechanism of p-hydroxybenzoate  hydroxylase ...........................................................................  10 Figure 1.10 Proposed catalytic mechanism of DNA photolyase ..............  11 Figure 1.11 Crystal structure of spinach FNR ........................................  12 Figure 1.12 Sequence of alignment of three diflavin oxidoreductases ......  13 Figure 1.13 Primary sequence domain organization of human methionine synthase reductase and cytochrome P450 reductase ............  14 Figure 1.14 Crystal structure of the prototypical human isoform of cytochrome P450 reductase ..................................................  14 Figure 1.15 Crystal structure of CPR FMN-binding domain and FMN-stabilizing residues ................................................................  16 Figure 1.16 Crystal structure of FAD/NADP(H)-binding domain and  FAD-stabilizing residues .......................................................  17 Figure 1.17 Different orientations of active site residues Trp676 and  Asp634 in CPR .....................................................................  18 Figure 1.18 NADPH stabilizing residues at the coenzyme binding pocket  for CPR ................................................................................  18 Figure 1.19 Proposed mechanism for NADPH-mediated reduction and oxidation of diflavin oxidoreductases ....................................  19  	   List of Figures 	  	   xiii	  Figure 1.20 Stereospecific orientation of the NADPH nicotinamide ring  over the FAD isoalloxazine ring ...........................................  20 Figure 1.21 Flavin redox potentials for diflavin oxidoreductases human  MSR and CPR ......................................................................  21 Figure 1.22 Conformational rearrangements adopted by cytochrome  P450 reductase during catalysis with cytochrome P450 .......  22 Figure 1.23 The mechanism of molecular oxygen activation and two- step rebound radical alkane hydroxylation by microsomal cytochrome P450 and reduction by cytochrome P450  reductase ...............................................................................  23 Figure 1.24 Superimposition of the crystal structure of human MSR and  CPR ......................................................................................  25 Figure 1.25 Key residues around the active site in MSR .........................  26 Figure 1.26 Stabilizing residues at the NADPH-binding cleft for MSR ...  27 Figure 1.27 Catalytic cycle and reactivation mechanism of methionine synthase and methionine synthase reductase .......................  29  Figure 2.1 Crystal structure of MSR (PDB 2QTZ) superimposed with  a NADPH-bound variant rat CPR W676XS677X  (PDB 1JAO) ........................................................................  32 Figure 2.2 The NADPH oxidation activity of MSR variants ................  35 Figure 2.3 Anaerobic reduction of MSR variant S698A with saturating NADPH monitored with multiple wavelength stopped-flow spectrophotometry ................................................................  37 Figure 2.4 Spectral profiles for anaerobic reduction of MSR variants  with saturating NADPH monitored with multiple  wavelength stopped-flow spectrophotometry ........................  38 Figure 2.5 Anaerobic reduction of MSR variants by equimolar NADPH monitored with multiple wavelength stopped-flow spectrophotometry ................................................................  40 Figure 2.6 Anaerobic reduction of MSR variants by saturating NADPH monitored with single wavelength stopped-flow spectrophotometry at 454 nm ...............................................  41 Figure 2.7 Anaerobic reduction of MSR variants by saturating NADPH monitored with single wavelength stopped-flow spectrophotometry at 600 nm ...............................................  42 Figure 2.8 Dependence of MSR variant kobs1 on NADPH concentration  42  Figure 3.1 NADPH-coordinating residues at the coenzyme-binding  pocket for CPR and MSR .....................................................  52 	   List of Figures 	  	   xiv	  Figure 3.2 Spectral changes upon NADPH reduction of CPR W676  variants monitored by multiple-wavelength stopped-flow spectrophotometry ................................................................  56 Figure 3.3 Anaerobic reduction of CPR W676 variants by saturating substrate monitored by single wavelength stopped-flow spectrophotometry ................................................................  57 Figure 3.4 Redox potentiometric data of human CPR W676Y and  W676F variants ....................................................................  59 Figure 3.5 Anaerobic reduction of MSR W697 variants by saturating NADPH monitored with multiple wavelength stopped-flow spectrophotometry ................................................................  61 Figure 3.6 Anaerobic reduction of MSR variants by saturating  substrate monitored with single-wavelength stopped-flow spectrophotometry at 454 and 600 nm .................................  63 Figure 3.7 Spectral changes upon NADPH reduction of CPR  coenzyme-binding variants monitored by multi-wavelength stopped-flow spectrophotometry ...........................................  65 Figure 3.8 Anaerobic reduction of CPR coenzyme-binding variants by saturating NADPH monitored at 454 and 600 nm with  single wavelength stopped-flow spectrophotometry ..............  66 Figure 3.9 Spectral changes upon NADPH reduction of MSR  coenzyme-binding variants monitored by multiple  wavelength stopped-flow spectrophotometry ........................  67 Figure 3.10 Anaerobic reduction of MSR coenzyme-binding variants by saturating NADPH monitored at 454 and 600 nm with  single wavelength stopped-flow spectrophotometry ..............  68  Figure 4.1 Proposed mechanism for NADPH-mediated reduction of  CPR and MSR ......................................................................  79 Figure 4.2 Proximal FAD isoalloxazine ring residues in wild-type CPR  and MSR ..............................................................................  80 Figure 4.3 Visible absorption spectra of CPR H322 and MSR A312  variants under aerobic conditions .........................................  82 Figure 4.4 Uncoupled NADPH oxidation activity of CPR H322 and  MSR A312 variants ..............................................................  84 Figure 4.5 Spectral changes upon NADPH reduction of CPR H322  variants monitored by multi-wavelength stopped-flow spectrophotometry ................................................................  86 Figure 4.6 Anaerobic reduction of H322Q and H322A by saturating  NADPH monitored by single wavelength stopped-flow spectrophotometry at 454 and 600 nm .................................  87 	   List of Figures 	  	   xv	  Figure 4.7 Spectral changes upon equimolar NADPH reduction of  H322Q and H322A monitored by multi-wavelength  stopped-flow spectrophotometry ...........................................  88 Figure 4.8 Spectral changes upon NADPH reduction of wild-type and H322A FAD domain variants monitored by multi- wavelength stopped-flow spectrophotometry ........................  89 Figure 4.9 Anaerobic reduction of WTFAD and H322AFAD by saturating NADPH monitored over 10 s by single wavelength stopped- flow spectrophotometry at 454 and 600 nm .........................  90 Figure 4.10 Redox titrations of H322AFAD ...............................................  91 Figure 4.11 Anaerobic multiple wavelength stopped-flow  spectrophotometry of NADPH-mediated reduction of MSR  A312 variants .......................................................................  93 Figure 4.12 Anaerobic reduction of A312Q and A312H by saturating  NADPH monitored by single wavelength stopped-flow spectrophotometry at 454 and 600 nm .................................  94 Figure 4.13 Single and multiple wavelength absorbance changes during  the anaerobic reduction of the MSR hinge variant ...............  96  Figure 5.1 Visible spectra of AaCPR W704 variants during reduction  by NADPH monitored by multiple wavelength stopped-flow spectrophotometry ................................................................  109 Figure 5.2 Visible spectra of AaCPR W704S variant during reduction  by NADPH monitored by multiple wavelength stopped-flow spectrophotometry ................................................................  110 Figure 5.3 Anaerobic single wavelength stopped-flow traces of AaCPR  W704 variants ......................................................................  111 Figure 5.4 Anaerobic flavin reduction of W704S monitored by single wavelength stopped-flow spectrophotometry ........................  112 List of Abbreviations2’,5’-ADP adenosine 2’,5’-diphosphateAaCPR cytochrome P450 reductase from Artemisia annuaABS Antley-Bixler SyndromeAdoMet S-adenosylmethionineAMP adenosine monophosphateATP adenosine triphosphateCH3THF methyltetrahydrofolateCoA coenzyme ACPR cytochrome P450 reductaseCT charge-transferCYP71AV1 cytochrome P450 71AV1DAAO D-amino acid oxidaseDNA deoxyribonucleic aciddNTP deoxyribonucleotide triphosphateDTT dithiothreitolEDTA ethylenediaminetetraacetic acidETF electron-transferring flavoproteinFAD flavin adenine dinucleotide (quinone)FADH2 flavin adenine dinucleotide, reduced (hydroquinone)FeCN ferricyanide Fe(CN)6 3−FMN flavin mononucleotide (quinone)FMNH2 flavin mononucleotide, reduced (hydroquinone)FNR ferredoxin NADP(H)-dependent reductaseGST glutathione-S-transferaseHepes 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acidIPTG isopropyl β-D-1-thiogalactopyranosideITC isothermal titration calorimetryk cat catalytic turnover numberk cat/Km catalytic efficiencyxviList of AbbreviationsKd dissociation constantkDa kilodaltonsKi inhibition constantKIE kinetic isotope effectKm Michaelis-Menten constantKPi potassium phosphateLB Luria-Bertani brothMetH methionine synthase from Escherichia coliMS methionine synthaseMSR methionine synthase reductaseMTRR gene encoding for MSRNADH nicotinamide adenosine dinucleotide, reducedNAD+ nicotinamide adenosine dinucleotide, oxidizedNADPD (R)-[4-2H]-NADPHNADPH β-nicotinamide adenosine diphosphate, reducedNADP+ β-nicotinamide adenosine diphosphate, oxidizedNMDA N-methyl-D-aspartateNMN nicotinamide mononucleotideNMR nuclear magnetic resonanceNOS nitric oxide synthaseNR1 human novel reductase 1PCR polymerase chain reactionPDA photodiode arrayPDB protein data bankPM photomultiplierPMSF phenylmethylsulfonyl fluoridePOR gene encoding for CPRSVD singular value decompositionTAPS N-tris(hydroxymethyl)methyl-3-aminopropanesulfonic acidTHF tetrahydrofolateTPNH triphosphopyridine nucleotide, reduced (NADPH)Tris tris(hydroxymethyl)aminomethaneUV ultravioletxviiAcknowledgmentsI would first like to full-heartedly thank Dr. Kirsten Wolthers for providing invaluableguidance, advice, and patience throughout my graduate studies and, especially, duringthesis writing. Thank you for giving me this opportunity, my academic career began andcontinues with your support and encouragement.I would also like to thank my committee members: Drs. Al Vaisius, Jim Bailey, andKevin Smith. Thank you all for taking the time to listen to my multiple mutagenesis adven-tures some of which were quite interesting, while others were only moderately interesting.Special thanks to Dr. Jim Bailey for allowing me to use the Ag/Cl and platinum electrodesfor redox potentiometry studies and for helpful troubleshooting discussions. I am gratefulto Dr. Kevin Smith for taking pity on me and letting me into his lab to pilfer liquid nitrogenon numerous occasions. I would also like to thank Dr. Al Vaisius for first getting me hookedon biochemistry and proteins during my undergraduate degree.Finally, I would not have made it this far without the unwavering support of my parentsTim and Blanche, my stepdad Ken, and the other legs to the redheaded tripod Phillip andSarah. Through all the stress, you were all there to encourage me to keep pushing forwards.Through all the success, you were all there to celebrate with me.xviiiDedicationThis thesis is dedicated to the SF-61DX2 Stopped-flow System and glove box. Manynights were spent hand-in-hand with the glove box, attempting to deconvolute diflavoen-zyme spectra...xixChapter 1Introduction1.1 Flavin-dependent Enzymes1.1.1 Biological Prevalence and DiversityFlavin-dependent enzymes are ubiquitous throughout nature and are necessary for main-taining the living condition of all known organisms. Approximately 1-3% of prokaryotic andeukaryotic genomes encode for proteins that use flavins.1,2 Of the estimated 19 000 protein-encoding genes in humans, 89 genes have been identified that encode for flavin-bindingproteins.3–5 The majority of these flavoenzymes catalyze electron transfers between obli-gate two-electron donors (e.g. NADPH, NADH) and one-electron acceptors (e.g. metals,oxygen).5 For this reason, flavoenzymes are common participants in the electron transportchain of major energy metabolic pathways such as oxidative phosphorylation and photosyn-thesis.6 The ability for the flavin cofactor to harvest light energy also enables it to functionin light-dependent reactions such as DNA repair and circadian rhythm regulation.7,8 Flavo-proteins are further utilized in the metabolism of carbohydrates, amino acids, lipids andnucleic acids, as well as a host of other biochemical processes as illustrated in Figure 1.1.9,101.1.2 Flavin DiscoveryIn 1879, an English chemist named Alexander Blyth isolated a yellow compound fromcow’s milk and named it lactochrome.11 Over fifty years later, two independent researchgroups lead by chemists Richard Kuhn and Paul Karrer, solved the chemical structure ofthe yellow compound, and it was renamed riboflavin: ribo- for the ribityl side chain and-flavin from the Latin word for yellow flavius, Figure 1.2.12–14 While the riboflavin chemicalstructure was still being determined, Otto Warburg isolated the first flavin-binding pro-tein, Old Yellow Enzyme, during his work on the mechanism of biological respiration.14,15Following Warburg’s work, Hugo Theorell treated this protein with ammonium sulfate atlow pH, which produced a white protein precipitate and a yellow nonprotein supernatant.Both layers lost the NADPH oxidase activity observed in the untreated protein.14,16 How-ever, mixing the yellow supernatant with the resuspended protein at the physiological pHrestored activity, indicating that both components are required for chemistry. Theorell11.1. Flavin-dependent Enzymes�Figure 1.1: Diversity of flavin-dependent biochemical processes. Figure is adapted fromvan Berkel and Joosten.9later identified the yellow component as a phosphorylated riboflavin, flavin mononucleotide(FMN), shown in Figure 1.2. Three years later, Warburg and Christian conducted a similarexperiment on D-amino acid oxidase and determined the yellow nonprotein component to bethe condensation product of FMN and adenosine monophosphate (AMP), flavin adenosinedinucleotide (FAD), as shown in Figure 1.2.14,171.1.3 Biological Sources of FlavinsFlavin cofactors, FMN and FAD, are derived from the water-soluble dietary precursorriboflavin, an essential growth factor also known as vitamin B2. Plants and most mi-croorganisms can synthesize riboflavin. However animals, rare prokaryotic and eukaryoticmicroorganisms cannot and must obtain it from exogenous sources.18,19 Dietary sources in-clude milk and dairy products, meats, fatty fish, and green vegetables. A secondary sourceof riboflavin is provided by natural microflora of the large intestine.18 In foods, most ofvitamin B2 is derived from flavin-bound enzymes.20 As the enzymes are digested, the flavincofactors are hydrolyzed by nonspecific diphosphatases and liberated as free riboflavin inthe intestinal lumen.20,21 Free riboflavin is absorbed by small intestine enterocytes whilecolonocytes absorb riboflavin produced by large intestine microflora.21,22 Once across thecellular membrane, riboflavin is phosphorylated by riboflavin kinase to yield FMN in thereaction shown in Figure 1.3. Phosphorylation serves as a metabolic trap to prevent dif-21.1. Flavin-dependent EnzymesNHNNN O OOOHHO OHRiboflavin IsoalloxazineRibityl chain P-O OO FMNP-O OO NNNN NH2O OHOH Adenosine FADFigure 1.2: Chemical structure of riboflavin, flavin mononucleotide (FMN) and flavinadenosine dinucleotide (FAD). The isoalloxazine ring, the ribityl chain, phosphate groups,and adenosine are coloured in yellow, green, blue, and purple, respectively.31.1. Flavin-dependent Enzymesfusion of free riboflavin out of the cell, which can be quickly excreted through the urinarysystem.20,21 Available FMN can be converted to FAD by an adenylylation catalyzed byFMN adenylyltransferase as shown in Figure 1.3. Blood plasma proteins, albumin andcertain immunoglobins bind and protect riboflavin and its derivatives for transport withinthe bloodstream to target tissues.21,23 Upon arrival, the flavins are incorporated into newlysynthesized flavin-dependent enzymes.1.1.4 Flavin StructureThe versatile chemistry of flavins arises from the chemically active isoalloxazine ring:7,8-dimethylbenzo[g]pteridine-2,4(3H,10H)-dione. It can exist in three redox states: fullyoxidized quinone, semiquinone (one-electron reduced), and hydroquinone (two-electron re-duced), see Figure 1.4. Each of these states can exist in a pH-dependent protonated,deprotonated, or ionic form.24 The oxidized quinone has a highly conjugated ring systemthat is responsible for its visible absorbance properties. It is a brilliant yellow colour withmaximal absorbance peaks at 360 and 450 nm as shown in the spectral profile of a flavoen-zyme in the oxidized state (Figure 1.5). The neutral semiquinone is blue with a broadabsorbance maximum from 580-625 nm. The unpaired electron on the semiquinone is par-tially stabilized through delocalization in the conjugated ring.25 Noncovalent or covalentinteractions between the flavin and protein can act to destabilize or further stabilize thesemiquinone form.14 In the hydroquinone state, conjugation is interrupted and creates theleuco (bleached) form which causes an absorbance loss at 450 nm. This electron-rich flavincan participate as a one-electron donor, a two-electron donor via hydride transfer, or act asa nucleophile at N5 or C4a atoms (see Figure 1.4 for atom numbering).2In most flavoenzymes, the flavin is bound through noncovalent interactions with lo-cal amino acids.26 The binding affinity is typically in the nanomolar range or lower, sothe flavin is considered a nondissociable prosthetic group rather than a dissociable coen-zyme.2 Catalytic function, substrate specificity, substrate stereochemistry and flavin redoxpotential have all been found to be influenced by the protein environment surrounding thecofactor.2,14,26 For example, local protein interactions can finely tune the reactivity of theoxidized isoalloxazine ring, given that it is a highly polarizable molecule.A few flavoenzymes form a covalent linkage with the C8a or C6 positions on the isoal-loxazine ring. The side chains of histidine, cysteine or tyrosine residues participate in thecovalent linkage.27 Covalent linkage has been proposed to be involved in protein stability,oxygen reactivity, and controlling the cofactor redox potential. It has also been shown tosuppress deleterious modifications of the flavin during catalysis. For example, the cova-lent linkage at C6 of FMN in trimethylamine dehydrogenase prevents formation of inactive6-hydroxy FMN.2741.1. Flavin-dependent EnzymesNHNNN O OCH2H OHH OHH OHCH2OHATPADP NHNNN O OCH2H OHH OHH OHCH2ORiboflavin kinase RiboflavinFlavin MononucleotideP NHNNN O OCH2H OHH OHH OHCH2O P O P ON N NNNH2 O OHOHFlavin Adenine DinucleotideATPPPiFMN adenylyltransferase OHOHOO OHOH OFigure 1.3: Biosynthetic pathway of FMN and FAD from the riboflavin precursor.51.1. Flavin-dependent EnzymesNHNNN O OH3CH3C R NHNNNH O OH3CH3C RNHHNNNH O OH3CH3C R e  , H+e  , H+2e  , 2H+ Oxidized Neutral Semiquinone (one-electron reduced)Hydroquinone (two-electron reduced)78 9 106 5 4 321Figure 1.4: Reductive and oxidative half-reactions of flavin isoalloxazine ring. Red illus-trates the bonds across which the electrons are redistributed.350 400 450 500 550 600 650 7000.00.10.20.30.40.50.60.7  Absorbance Wavelength (nm)Figure 1.5: Spectral profile of the flavoenzyme cytochrome P450 reductase in the oxidized(solid), semiquinone (dotted) and fully reduced states (dashed).61.1. Flavin-dependent Enzymes1.1.5 Common Flavin-Catalyzed ReactionsFlavoenzyme catalysis typically consists of two separate half-reactions where the boundflavin alternates between oxidized and reduced states, shown in Figure 1.6. In the reduc-tive half-reaction, an electron donor is oxidized and the flavin is reduced whereas in theoxidative half-reaction, an electron acceptor is reduced and the flavin is oxidized. In theoxidoreductase class of flavoenzymes, both the oxidative and reductive half-reactions in-volve one- or two-electron transfers between redox active groups, such as metal centersand pyridine nucleotides. An example of an oxidoreductase is ferredoxin-NADP(H) reduc-tase, which be discussed in greater detail later in this chapter. Below are examples of howother classes of flavoenzymes, dehydrogenase, oxidase and monooxygenase, employ differentmechanisms in their oxidative and reductive half-reactions to catalyze a great diversity ofchemical reactions. SredPox FloxFlred S'oxP'red   Oxidative Half-Reaction   Reductive Half-ReactionFigure 1.6: Oxidative and reductive half-reactions of typical flavoenzymes.Mammalian acyl-Coenzyme A (CoA) dehydrogenase catalyzes the first step in fattyacid β-oxidation.28 The enzyme introduces a trans double bond into short, medium, longand very long chain acyl-CoA thioester substrates. The reductive half-reaction illustratedin Figure 1.7 shows the concerted breakage of the αC-H and βC-H bonds where the firsthydrogen is abstracted by an active site base and the second is transferred as a hydride tothe flavin N5 atom to generate the reduced flavin and the trans-∆2-enoyl-CoA product. Inthe oxidative half-reaction, the flavoenzyme is regenerated by two consecutive transfers ofreducing equivalents to electron transferring flavoproteins (ETF), which shuttle electronsinto the respiratory chain.Mammalian D-amino acid oxidase (DAAO) catalyzes the oxidative deamination of D-α-amino acids to the corresponding α-imino acids.29 Neuronal DAAO has been implicated inregulating signaling in the brain through its action on D-serine, a ligand for N-methyl-D-aspartate (NMDA) receptors.30 Figure 1.8 shows that binding of the appropriate D-α-aminoacid substrate is followed by direct hydride transfer to the N5 atom of the oxidized flavinisoalloxazine ring in the reductive half-reaction.29,31 The imino product is then released andnonenzymatically hydrolyzed to the corresponding keto acid. The reduced flavin is then71.1. Flavin-dependent EnzymesN ONHONNRR' SCoAOHHEnzB HN ONHONHNRR' SCoAO2x H-Product SubstrateHHH EnzAFigure 1.7: Proposed catalytic mechanism of acyl-CoA dehydrogenase.oxidized in the oxidative half-reaction by molecular oxygen. Oxygen is released as hydrogenperoxide.Bacterial p-hydroxybenzoate hydroxylase acts in the catabolism of aromatic compoundssuch as p-hydroxybenzoate that are liberated during the biodegradation of lignin, a ma-jor component of wood.32 First, the oxidized enzyme binds both p-hydroxybenzoate andNADPH. A hydrogen-bond network then deprotonates p-hydroxybenzoate to the pheno-late anion. This triggers protein rearrangements that lead to a pendulum-like movementof the isoalloxazine ring from a buried ’in’ conformation to a more solvent-exposed ’out’conformation. The latter enables NADPH-FAD complex formation and reduction of theisoalloxazine ring. Following NADP+ release, the ring returns to the ’in’ conformation.32,33In the oxidative half-reaction, the reduced FAD transfers a single electron to molecularoxygen to form a superoxide anion, which then rapidly reacts with the C4a atom to forma C4a-hydroperoxoflavin intermediate. The electrophilic properties of this intermediate en-ables transfer of a hydroxyl to the substrate, which generates 3,4-dihydroxybenzoate. Theenzyme is restored following release of water and product.1.1.6 Unusual Flavin-Catalyzed ReactionsDNA photolyases catalyze the reduction of cyclobutane pyrimidine dimers that areformed in UV irradiated DNA.34 They are present in many organisms, including bacte-ria, plants and some animals such as the goldfish (Carassius auratus) and the marsupial ratkangaroo (Potorous tridactylis).35 The flavoenzyme contains FAD and a light-harvestingchromophore, either 5,10-methenyltetrahydrofolate or 8-hydroxy-deazaflavin. In the firststep of catalysis, the positively charged groove of the enzyme binds to the damaged DNAphosphodiester backbone and the pyrimidine dimer flips out of the helix into the enzyme81.1. Flavin-dependent EnzymesN ONHONRNR' HNH2 OO N ONHONHRNN ONHONHRN O2OOH N ONHONRN D-amino acidH2O2 H2O NH3 OR' NH2OOR' NH2OOR' O OFigure 1.8: Proposed catalytic mechanism of D-amino acid oxidase.91.1. Flavin-dependent EnzymesEox NADPHpOHB Eox-NADPH Ered-pO BN ONHONHNR-OO OOHN ONHONHNR -OO OH OHOH-OO OHOH O2H2O3,4-diOHB NADP+-OO O BH OFigure 1.9: Proposed catalytic mechanism of p-hydroxybenzoate hydroxylase.cavity.36 Light is then absorbed by the chromophore and the excitation energy is transferredto the reduced flavin which is located near the interior of the cavity. Electron transfer fromthe reduced flavin to the pyrimidine dimer forms a ketyl radical, which decomposes stepwiseto split the cyclobutane into a pyrimidine-pyrimidine anion radical pair.37 Final electrontransfer back onto the flavin semiquinone regenerates the reduced flavin and the repairedDNA is released.Iodotyrosine deiodinase is the first identified mammalian member of the NADH oxi-dase/flavin reductase superfamily.38 This enzyme relies on FMN to catalyze the reductivedehalogenation of byproducts of thyroid hormone synthesis to generate iodide and tyro-sine.39 These liberated products are then reused to produce more thyroid hormone. In theabsence of this flavoenzyme, precious iodine sources are excreted and an iodine deficiencyensues.39 Iodotyrosine deiodinase is an unusual enzyme since reductive dehalogenation israrely observed in aerobic organisms, even less so in mammals.38 Only one other reduc-tive dehalogenase is known in humans, a nonflavoenzyme that involves the participationof thiol reductants and cysteine residues.40 Although the mechanistic details are still notfully understood, iodotyrosine deiodinase catalysis has been determined to proceed withoutthe use of thiols and cysteines.38,41 Thus, this flavoenzyme is an excellent example of theever-expanding repetoire of known flavin-mediated reactions.1.1.7 Ferredoxin NADP(H)-dependent Reductase, an OxidoreductaseIn a recent review of over 300 flavoenzymes, it was determined that ∼ 90% of flavin-dependent enzymes belong to the catalytic class of oxidoreductase, which encompasses101.1. Flavin-dependent EnzymesN ONHONHN N ONHONHHN*R NHNO OHN NOO dR dRPi NHNO OHN NOO dR dRPi N ONHONHN NHNO OHN NOO dR dRPiRN ONHONHN NHNO OHN NOO dR dRPiREE-DNAChromophore Excitation   transferRepaired DNADamaged DNA hvFigure 1.10: Proposed catalytic mechanism of DNA photolyase.oxidation-reduction catalysis.5 Ferredoxin NADP(H)-dependent reductases (FNR) form oneof the largest families of oxidoreductases. They are monomeric, hydrophilic FAD-boundenzymes. The first FNR was isolated from spinach chloroplasts in 1956 by Avron andJagendorf and given the name ”TPNH diaphorase” for its ability to transfer electrons to orfrom NADP(H).42 In the following decade, Arnon and colleagues confirmed the physiologicalrole for FNR.43 The enzyme catalyzes the final step of photosynthesis where electrons aretransferred from ferredoxin, a small iron-sulphur protein, to NADP+ to yield NADPH.In 1991, the X-ray crystal structure of spinach FNR was solved.44 Figure 1.11 shows thestructure is comprised of a two-domain motif. The FAD binds to an αβα-motif with acompact six-stranded anti-parallel flattened β-barrel. The NADP(H)-domain consists of afive-stranded β-sheet sandwiched by six α-helices with Rossman topology. Together, thesedomains are referred to as the FNR module.44,45Many more enzymes in the FNR family have since been discovered with a myriad ofphysiological roles. For example, bacterial FNR transfers reducing equivalents from NADPHto flavodoxin, an FMN-bound structural analogue of ferredoxin. Flavodoxin has numerousredox partners that makes FNR indispensible for nitrogen fixation, steroid metabolism,terpenoid biosynthesis, oxidative-stress response, and iron-sulphur cluster biogenesis in het-erotrophic bacteria and fungi.46111.1. Flavin-dependent EnzymesFigure 1.11: Crystal structure of spinach FNR with bound FAD (yellow) and 2’,5’-adenosinediphosphate substrate moiety (grey). PDB 1FND.1.1.8 The Diflavin Oxidoreductase FamilyMembers of the diflavin oxidoreductase family are multidomain enzymes that containnoncovalently bound molecules of FAD and FMN. The enzymes transfer reducing equiva-lents from NADPH to FAD to FMN and onto a physiological electron acceptor. Prokaryoticmembers include the α-subunit of sulphite reductase of Escherichia coli and the reductasedomain of a fatty acid hydroxylase called flavocytochrome P450 BM3 from Bacillus mega-terium.47,48 Known eukaryotic members are cytochrome P450 reductase (CPR), the reduc-tase domain of nitric oxide synthase (NOS), methionine synthase reductase (MSR), andhuman novel reductase 1 (NR1).49–53 This family likely emerged from the early evolution-ary fusion of ancestral genes of prokaryotic FNR and flavodoxin.54 All members share thefollowing structural arrangement from N- to C-terminal: i) a FMN-binding domain homolo-gous to bacterial flavodoxin, ii) a flexible hinge region that joins with a bridging connectingdomain, and iii) a FAD/NADPH-binding domain homologous to bacterial FNR. The over-all domain organization of diflavin oxidoreductases, shown in Figure 1.14, is representedby the structure for the prototypic family member CPR. CPR and MSR share the great-est sequence similarity among the known diflavin oxidoreductases; a sequence alignmentof human MSR, human CPR, and plant CPR are shown in Figure 1.12 and distinguishingstructural features between CPR and MSR are highlighted in Figure 1.13.121.1. Flavin-dependent EnzymesFigure 1.12: Sequence alignment of three diflavin oxidoreductases: human MSR, humanCPR, and plant CPR derived from Artemisia annua. An asterisk denotes an identicalresidue, a colon denotes a conserved residue, and a period denotes a semiconserved residue.The FMN-binding residues are boxed in blue, the FAD-binding residues are boxed in red,and the NADPH-binding residues are boxed in purple.131.1. Flavin-dependent EnzymesFigure 1.13: Primary sequence domain organization of human methionine synthase re-ductase and cytochrome P450 reductase. The FAD/NADPH-binding domain in purple,connecting domain (CD) in red, hinge region in green, FMN-binding domain in teal, andtransmembrane (TM) domain in peach.Figure 1.14: Crystal structure of the prototypical human isoform of cytochrome P450reductase [PDB 3QE2]. Structures are shown in a cartoon model with the FAD/NADPH-binding domain, connecting domain, and FMN-domain coloured in purple, blue, and teal,respectively. FAD and FMN flavins are represented in yellow stick model and the NADPHmolecule is in grey with the nitrogen, oxygen and phosphorous atoms coloured as blue, red,and orange, respectively.141.2. Cytochrome P450 Reductase1.2 Cytochrome P450 Reductase1.2.1 StructureCPR is a 78 kDa monomeric enzyme that is localized to the cytosolic side of the en-doplasmic reticulum.55 It is anchored to the endoplasmic reticulum membrane by a 6 kDahydrophobic α-helical domain. To date, crystallization of full-length native human CPR hasbeen unsuccessful. However, removal of the N-terminal membrane-spanning helix enabledexpression and purification of a soluble form of CPR which formed crystals that diffracted inthe X-ray beam.49,56,57 Soluble CPR is a poor reductant for membrane-bound cytochromeP450s, but it can reduce the non-physiological electron acceptor cytochrome c.58The protein structure of the FMN-binding domain of CPR is presented in Figure 1.15. Itcomprises a five-stranded parallel β-sheet sandwiched between five α-helices.49 The planarFMN isoalloxazine ring is stacked between two aromatic residues, Tyr115 and Tyr77. Thelatter is tilted 40◦with respect to the cofactor.56 The crystal structure further reveals thatthe solvent-exposed dimethyl benzene edge of FMN is located within 4 A˚ of that of FAD.The FMN mononucleotide moiety is also stabilized by electrostatic interaction with a helixdipole formed at the N-terminal end of an α-helix.A short 12 amino acid (Gly231-Arg242) hydrophilic hinge links the FMN domain tothe connecting domain which tethers to the FAD/NADPH-binding domain.49 Hamdane etal. observed that shortening the hinge length progressively impeded electron transfer fromFAD to FMN and reduced enzyme activity.59 The hinge provides the flexibility required forthe FMN domain to form mutually exclusive complexes with i) the FAD/NADPH domain,and ii) the terminal electron acceptor.By definition of the diflavin oxidoreductase family, the FAD/NADPH-binding domainis homologous to the FNR module; see Section 1.1.7.49,57 The FAD isoalloxazine ring issurrounded by the ’catalytic triad’ (residues Ser457, Cys629 and Asp675 - human CPRnumbering), which facilitate the transfer of a hydride ion from the NADPH nicotinamidering to the isoalloxazine ring.60 The two conserved aromatic residues that stack againstthe FAD isoalloxazine ring in Figure 1.16 are Trp676 and Tyr455. With NADP+ bound,Trp676 is partially stacked over the isoalloxazine ring. Due to steric clash, the indole sidechain blocks the NADPH nicotinamide ring from moving into the FAD active site, whichis necessary for direct hydride transfer. Deletion of the C-terminal residues Trp676 andSer677 enabled the nicotinamide ring to stack over the isoalloxazine ring, as observed inthe variant crystal structure.61 Thus, conformational rearrangement of Trp676 is requiredto accommodate the incoming nicotinamide ring.Interestingly, Figure 1.17 shows a second orientation of Trp676 that was captured in theNADP+-free form of CPR where a disulfide bond was engineered between the FMN and151.2. Cytochrome P450 ReductaseFigure 1.15: Crystal structure of CPR FMN-binding domain and FMN-stabilizing residues[PDB 3QE2]. Left : The FMN-binding domain represented in cartoon model in teal and theFMN cofactor shown in yellow stick model. Right : FMN-stabilizing residues. The FMNcofactor in yellow with the standard atom-coloured stick model, shown stacked againstTyr77 and 115 on the re- and si -sides, respectively. Hydrogen bonds are shown in blackdashed between the flavin and hydroxyl groups of Ser23, Thr25, Thr27 and Tyr77. Possiblecontacts between the Gly78 and the isoalloxazine ring are also shown.161.2. Cytochrome P450 ReductaseFigure 1.16: Crystal structure of CPR FAD/NADP(H)-binding domain and FAD-stabilizing residues. Left Human CPR FAD/NADP(H)-binding domain crystal structure[PDB 3QE2]. The FAD-binding domain is represented in blue cartoon model, NADPH-binding domain in purple, the FAD cofactor is shown in yellow stick model, and the NADPHis shown in grey with the standard atom colouring. Right FAD-stabilizing residues. TheFAD cofactor is in yellow stick model with the standard atom colouring and potential hy-drogen bonds are shown in black dashes.FAD/NADPH domains (NADP+-free CPR without the disulfide link is not available).57In this structure the Trp676 indole side chain is flipped and rotated by 90◦, such thatthe entire indole ring lies planar to the isoalloxazine ring. This second structure suggeststhat Trp676 adopts multiple conformations during catalysis.57 Trp676 is first fully stackedwith the isoalloxazine ring, then undergoes a NADPH-induced conformational change to thepartially stacked position, followed by displacement from the active site by the nicotinamidering.Further scrutiny of the active site reveals a loop adjacent to the FAD isoalloxazine ringwith an aspartic acid (Asp634). Figure 1.17 displays the different orientations captured forthis residue. In substrate-bound CPR, the side chain of Asp634 is rotated away from theactive site.49 However in rat variants, with an engineered disulfide link or deleted Trp676,the aspartic acid is projected between the isoalloxazine ring and the coenzyme-bindingsite, therefore this residue must also undergoes NADPH-induced conformational change.57Additional residues that form the coenzyme-binding site are shown in Figure 1.18. Theresidues (Tyr604, Lys602, Ser596, and Arg597) are positioned for hydrogen bonding withthe 2’-phosphate of NADPH. Further stabilizing interactions are formed between Arg298and Thr535 and the NADP+ pyrophosphate.171.2. Cytochrome P450 ReductaseFigure 1.17: Different orientations of active site residues Trp676 and Asp634 in CPR. Wild-type human CPR is shown in purple while the disulfide linked rat variant is in cyan [PDB3QE2 and 3OJW]. The 2’,5’-ADP moiety of NADPH is shown in grey with the standardatom-coloured stick model. FAD isoalloxazine ring is in yellow with the ribityl adenosinetail removed for clarity.Figure 1.18: NADPH stabilizing residues at the coenzyme binding pocket for CPR [PDB3EQ2]. Right Residues that interact with the NADPH 2’-phosphate. Left Residues thatinteract with the NADPH pyrophosphate. NADPH is shown in grey with the standardatom-coloured stick model. Potential hydrogen bonding interactions are shown in blackdashes.181.2. Cytochrome P450 Reductase1.2.2 EnzymologyA general reaction scheme for the CPR reductive half-reaction has been proposed, andit is shown in Figure 1.19.57,60,61 First, NADPH binds to the fully oxidized enzyme (StepI). In Step II, a hyride ion is transferred from NADPH to the FAD isoalloxazine to formthe FAD hydroquinone. A single electron is then transferred from FADH2 to FMN toyield the disemiquinone (III; FMNH• and FADH•)). In the fourth step (IV), the FMN isfully reduced to the hydroquinone by a second electron transfer. Dissociation of oxidizedNADP+ allows for a second molecule of NADPH to bind and further reduce the enzymeto the four-electron reduced dihydroquinone (V). Release of NADP+ and transfer of tworeducing equivalents to the appropriate acceptor completes the catalytic cycle.E FADFMN E FADFMNNADPH NADPH E FADH2FMNNADP+E FADHFMNH(NADP+)E FAD(NADP+)FMNH2E FADH2(NADP+)FMNH2 NADPHNADP+4H+, 4e- I II IIIIVVVINADP+ H++ H+Figure 1.19: Proposed mechanism for NADPH-mediated reduction and oxidation of diflavinoxidoreductases. I NADPH binds. II Donation of hydride ion from NADPH to FAD. IIIFormation of the disemiquinone. IV Both electrons are transferred to the FMN to form thehydroquinone. V Full four-electron reduction to the dihydroquinone by a second moleculeof NADPH. VI Transfer of reducing equivalents from the enzyme to terminal electronacceptors regenerates the fully oxidized enzyme. The presence/absence of the oxidizedcoenzyme is not absolutely known and depicted as parentheseses around NADP+.Early experiments with spinach FNR established a clear bipartite binding model forNADPH binding.62,63 Bipartite binding is defined as the initial anchoring of the 2’,5’-ADPmoiety to the enzyme, followed by displacement of the conserved FAD-stacking aromaticresidue by the nicotinamide mononucleotide (NMN) moiety of NADPH. This productiveplacement orients the pro-R-hydrogen on the C4 atom of the nicotinamide ring to the N5atom of the FAD isoallaoxazine ring shown in Figure 1.20. Deng et al. determined that forFNR, the binding energy for the coenzyme is derived from polar interactions between the2’,5’-ADP moiety and that initial docking of this moiety ensures that the enzyme favours191.2. Cytochrome P450 Reductasebinding of NADPH over NADH.62 Evidence of this bipartite binding model has also beenfound in diflavin oxidoreductases, including CPR and MSR.64,65Figure 1.20: Stereospecific orientation of the NADPH nicotinamide ring over the FADisoalloxazine ring. For clarity, only partial NADPH is shown in grey and the isoalloxazinering is shown in yellow with standard atom colouring.Following productive placement of the nicotinamide ring, a hydride ion is donated to theisoalloxazine ring to form the FADH2 hydroquinone.60 The midpoint potentials of the redoxcouples NADPH/NADP+ (-320 mV) and FADox/hq (-333 mV) are similar, indicating thatreaction is not spontaneous (see Figure 1.21 for redox potentials).66 However, the highlyelectropositive FMNox/sq couple (-66 mV) ensures that the next electron transfer step isthermodynamically favourable. Thus, the FMNox/sq couple provides the driving force tofavour the forward flow of electrons in CPR. Kinetic analysis has shown that the hydrideand interflavin electron transfer steps (II and III) are tightly coupled.66,67Following the first hydride transfer, the two-electron reduced CPR forms a quasi-equilibrium mixture of disemiquinone (FADH•-FMNH•) and FMN hydroquinone (FAD-FMNH2) due to the near isopotential values between the FADox/sq and FMNsq/hq cou-ples.66–68 Since the FMNH• has extremely low reactivity towards cytochrome P450 reduc-tion, FMNH2 is envisioned to serve as the electron donor.66,69 The final electron transferfrom FMN to the heme center of cytochrome P450 (∼ -300 mV) is a thermodynamicallyunfavourable step. However, recent studies have shown that, when bound with substrate,the heme center becomes +10-80 mV more electropositive than the FMNsq/hq.70,71 In addi-201.2. Cytochrome P450 Reductase0-100-200-300-400 NADP(H)FADox/sqFADox/sqFADsq/hq FADsq/hqFMNox/sq FMNox/sqFMNsq/hq FMNsq/hqMSR CPRFigure 1.21: Flavin redox potentials for diflavin oxidoreductases human MSR andCPR. The oxidized/semiquinone redox couples are shown in white boxes, and thesemiquinone/hydroquinone redox couples are in grey boxes.tion, the redox potentials of membrane-embedded CPR may shift to a less electronegativevalue compared to soluble CPR.72For intra- and intermolecular electron transfer, CPR undergoes large-scale conforma-tional changes. Global domain motion in CPR has been extensively investigated through anumber of techniques including small-angle X-ray scattering, small-angle neutron scattering,NMR and Forster fluorescence resonance energy transfer.73–76 All solved crystal structuresof native CPR are in a compact closed conformation where the interflavin distance is short-ened to favour intramolecular electron transfer. However, since the FMN flavin is buriedin the closed state, it does not allow for electron transfer to an external protein partner.57Therefore, the FMN domain must dissociate from the FAD/NADPH domain into an openconformation that exposes the FMN cofactor for contact with and electron transfer to theappropriate protein partner. Figure 1.22 shows the conformational dynamics of CPR withrespect to cytochrome P450 using available crystal structures.1.2.3 Physiological RoleCPR catalyzes electron transfers to numerous physiological acceptors, including cy-tochrome b5, squalene monooxygenase, and heme oxygenase.77–79 It also supplies electronsto all microsomal cytochrome P450s, a large heme monooxygenase superfamily.67,80,81 In211.2. Cytochrome P450 ReductaseFMNFADNADPH FMNFADNADPH HemeCPR CPR P450Figure 1.22: Conformational rearrangements adopted by cytochrome P450 reductase duringcatalysis with cytochrome P450. Crystal structures are in cartoon coloured from N-terminal(blue) to C-terminal (red) [PDB 3QE2, 3ES9 and 2CPP]. NOTE: the ’open’ structure is ofa CPR variant with 4 amino acids deleted from the hinge region and does not reflect theexact in vivo conformation.humans, fifty seven cytochrome P450 genes have been identified.82 These P450 enzymescatalyze reactions as diverse as hydroxylations, N-, O- and S-dealkylations, peroxidations,and epoxidations. Collectively, they act on a broad repertoire of substrates and are in-volved in the metabolism of lipids, steroids, and fat-soluble vitamins.82,83 Additionally,P450s metabolize xenobiotics such as drugs, polycyclic biphenols and the carcinogeneticbenzo[α]pyrene found in tobacco smoke.81 In general, the P450-catalyzed hydroxylation ofmetabolites increases their solubility for downstream metabolism and excretion.Microsomal P450 activity requires two consecutive electron transfers onto the P450 hemeto catalyze oxygen activation and substrate hydroxylation as presented in Figure 1.23. Theheme is an iron-protoporphyrin IX coordinated to a cysteine thiolate.84–87 In the restingstate, the ferric heme is bound to water. The incoming substrate displaces the water fromthe center and shifts the redox potential of the heme (from -330 mV to -173 mV) to favourelectron transfer from CPR.84 A single electron is shuttled from the FMN hydroquinone ofCPR to the P450 heme. The reduced ferrous heme readily binds molecular oxygen to yielda ferrous-dioxygen complex, which rapidly converts to a ferric superoxide complex. It isreduced by a second electron and protonated to generate the ferric hydroperoxide complex.This unstable complex rapidly undergoes a second protonation and heterolytic cleavageof the O-O bond. Water is then released and the FeIV -oxo radical is formed.84,88 It is apowerful oxidant and abstracts a substrate hydrogen to generate a short-lived carbon radicaland ferryl hydroxo intermediates.88 Finally, through a rebound mechanism the substrate ishydroxylated and released, and the heme returns to the resting state.221.2. Cytochrome P450 ReductaseN N N NFeIIN N N NFeIIIR H R H N N N NFeIIIR H O ON N N NFeIIIR H O OHR HN N N NFeIVR OHN N N NFeIIIN N N NFeIIIOH H R H NADPHNADP+ FADFADH FMNFMNH FMN O2 FMNFMNH2OR OHH2O N N N NFeVOCPRO HRFigure 1.23: The mechanism of molecular oxygen activation and two-step rebound radicalalkane hydroxylation by microsomal cytochrome P450 and reduction by cytochrome P450reductase. The heme porphyrin ring is abbreviated to the pyrrole nitrogens coordinated toiron. See text for details.All microsomal P450s share similar overall topology: the conserved motif around theheme-binding pocket, a proline-rich region, and an N-terminal hydrophobic transmembranedomain. The most structural variation is around the substrate binding site.89,90 This struc-tural variation accounts for the ability of the P450 family to utilize such a varied collectionof substrates. Although only 57 genes were found in humans, many more can be found inflora. For example, 272 are identified in Arabidopsis thaliana and 457 in rice Oryza sativagenomes.82,91,92 In plants, more P450s are required for the synthesis of pigments, growthregulators, and plant toxins.931.2.4 Health ImplicationsCPR acts to maintain the activity of all microsomal cytochrome P450s, therefore de-rangements in CPR function result in a broad spectrum of clinical phenotypes. These aredependent on the affected P450s and the degree of CPR impairment. There are over 40naturally occurring mutations identified in the gene encoding for CPR, POR.94,95 A poly-morphic variation in one allele may not elicit any adverse effects, while eliciting subtleeffects on the physiological activity of P450s. For example, three polymorphisms in PORhave been identified that influence the P450-mediated metabolism of warfarin.96 Warfarinis a potent anticoagulant with a very narrow therapeutic range, thus there is a potentialfor thrombolic complications and haemorrhage if the dosage is overestimated. A recent231.3. Methionine Synthase Reductasemethod proposed to ensure safe dosage entails pharmacogenetic testing to determine thespecific POR polymorphisms in the patient to predict that individual’s ability to metabolizethe drug.96,97Severe CPR deficiency arises from missense mutations in functionally important struc-tural regions. This deficiency is a rare autosomal recessive disease with clinical phenotypessuch as congenital adrenal hyperplasia, disordered sexual development, and Antley-BixlerSyndrome (ABS).95,98 ABS is a rare disease that causes craniofacial malformations andadditional skeletal anomalies. A comprehensive study of patients with apparant ABS re-vealed two distinct diseases.98 First, ’classic’ ABS with normal steroidogenesis that is causedby mutations in the fibroblast growth factor receptor 2 gene. Second, an ABS-phenotypewith steroid abnormalities and ambiguous genitalia that is caused by missense mutationsin POR.98,99 Clearly, normal CPR function is necessary for normal development. Inter-estingly, while complete POR gene knockout mice is embryonically lethal, liver-specificknockout mice develop normally with reproductive capabilities despite impaired bile acidproduction and steroid metabolism.100–102 Therefore, hepatic CPR activity is not essentialfor survival and nonhepatic CPR dysfunction must be the origin of the dysmorphology seenin CPR deficiency.1.3 Methionine Synthase Reductase1.3.1 StructureMSR and CPR share a diflavin topology with 42% amino acid sequence similarity.Figure 1.24 displays the superimposition of MSR and CPR crystal structures to illustratetheir shared structural motifs. Unlike CPR, MSR is a 78 kDa monomeric soluble enzymethat is localized in the cytosol.103,104 Although crystallization of full-length native humanMSR has not yet been successful, the structure of the enzyme without the FMN domainhas been solved.65 Based on the sequence alignment with CPR, the FMN domain of MSR isexpected to have the conserved FMN-binding fold, as described in Section 1.2.1. A notablestructural difference is the 82 amino acid hinge region in MSR, which is 5 times longerthan the corresponding region in CPR. The extended hinge is partially captured in theMSR crystal structure. However, since the hinge is located at the N-terminal end of thetruncated protein, the observed extended position is likely nonphysiological.Like CPR, the MSR FAD domain contains the ’catalytic triad’, a trio of residues(Asp695, Cys650 and Ser454), and a conserved aromatic residue, Trp697, that lies pla-nar to the re-face of the FAD isoalloxazine ring, shown in Figure 1.25.65 Interestingly, theorientation of the indole side chain of Trp697 in MSR - both with and without NADP+bound to the active site - is similar to that found in NADP+-free CPR where the entire241.3. Methionine Synthase ReductaseFigure 1.24: Superimposition of the crystal structure of human MSR and CPR (with theFMN domain of CPR removed) [PDB 3QE2 and 2QTZ]. Crystal structures are representedin cartoon model. The connecting domain of CPR is red and light pink for MSR, FAD-domain is in blue for CPR and light blue for MSR, NADPH-domain is in purple for CPRand light purple for MSR. FAD bound flavin and NADPH are represented stick model asyellow and grey, respectively.indole side chain lies planar to the FAD. Moreover, the MSR aspartic acid residue (Asp652)which is in the adjacent loop, projects towards the pyrophosphate in both NADP+-boundand NADP+-free MSR. Again, while the position of the Asp652 carboxylate is similar tothat observed in the substrate-free engineered rat CPR, it is in stark contrast with that ofthe substrate-bound human CPR.57,65 Thus, it seems that the binding of NADP+ to CPRcan induce stable structural changes in the coenzyme-binding cleft. These changes includeflipping of Trp676 and repositioning of the Asp loop which enables docking of the NADP+pyrophosphate closer to the N-terminal end of the helix.57 In contrast, coenzyme binding toMSR is unable to induce the same stable structural changes. Inspection of the 2’-phosphatebinding pocket, see Figure 1.26, reveals a similar network of polar residues that includeTyr624, Ser610, and Arg611. A notable difference is Lys623, which is shifted by one aminoacid position in the primary sequence. This shift means that it is unable to form a directhydrogen bond with the 2’-phosphate.65 Figure 1.26 shows the coenzyme binding cleft ofMSR around the NADP(H) pyrophosphate. Compared to CPR, the cleft is more negatively251.3. Methionine Synthase ReductaseFigure 1.25: Key residues around the active site in MSR [PDB 2QTZ]. The FAD isoallox-azine ring is in yellow with the ribityl adenosine moiety removed for clarity. The ribose andnicotinamide ring moieties of NADPH are disordered in this crystal structure, the 2’,5’-ADPmoiety is in grey with the standard atom colouring.charged, but the electrostatic interactions from Thr547 and Lys291 with the coenzyme areconserved.1.3.2 EnzymologyThe reductive half-reaction of MSR is expected to follow the same general reactionscheme offered earlier for CPR in Section 1.2.2. Previous isothermal titration experimentshave also revealed the two-step binding mode of NADPH as discussed earlier for FNR andCPR.65 Coenzyme binding in MSR is much weaker than in CPR, by >700-fold. Further-more, while in CPR the dissociation constants (Kd) for NADP+ and 2’,5’-ADP are the same(∼ 50 nM), in MSR they are not.64,65 The Kd for NADP+ is 37 µM and 2.4 µM for 2’,5’-ADP, suggesting that the NMN moiety contributes unfavourably to the overall coenzymebinding energy, as has been described for FNR.63,65Like in CPR, hydride transfer is dependent on the displacement of Trp697 (Step I inFigure 1.19). In contrast to CPR, hydride transfer is thermodynamically favourable, becausethe FADox/hq is more electropositive (-272 mV). Moreover, hydride transfer is not coupledto subsequent interflavin electron transfer (II and III).105 Interflavin electron transfer isa much slower kinetic event in MSR. While the redox potentials of MSR are similar toCPR, they span a much narrower range as shown in Figure 1.21. More compressed redox261.3. Methionine Synthase ReductaseFigure 1.26: Stabilizing residues at the NADPH-binding cleft for MSR [PDB 2QTZ].NADPH is shown in grey with the standard atom-coloured stick model. Potential hy-drogen bonding interactions are shown in black dashes and water molecules are representedas red spheres.potentials may be a factor in the limited and slower formation of the disemiquinone inMSR. In Step IV, the FMN hydroquinone is formed by a second electron transfer. Bindingof another NADPH generates the four-electron reduced state in Step V. Final electrondelivery to the physiological electron acceptor cobalamin-dependent methionine synthaseis thermodynamically unfavourable. The midpoint potential of the FMNsq/hq couple is -260 mV, while that for the cob(I)alamin/cob(II)alamin couple is -490 mV.106 However, thisendergonic step is overcome by coupling electron transfer to the highly exogonic methylationof cob(I)alamin by S-adenosylmethionine.1.3.3 Physiological RoleMethionine synthase reductase is critical in one-carbon and folate metabolism by main-taining the activity of methionine synthase (MS).52,103,104 MS, a vitamin B12(cobalamin)-dependent enzyme, is at the junction of two major metabolic pathways. It catalyzes themethyl group transfer from methyltetrahydrofolate (CH3THF) to homocysteine to producetetrahydrofolate (THF) and methionine.52 In the catalytic cycle shown in Figure 1.27, thesupernucleophilic cob(I)alamin (cobalt in 1+ oxidation state) abstracts the methyl groupfrom CH3THF to yield methylcobalamin and THF. Methylcobalamin donates the methyl tothe thiol group of homocysteine to generate methionine and regenerate cob(I)alamin. Ap-proximately every 2000 catalytic turnovers, cob(I)alamin undergoes one electron oxidation,producing cob(II)alamin.52 Cob(II)alamin is unable to carry on with catalysis, renderingMS inactive. To restore MS to its active state, the MSR FMN domain forms a complex with271.3. Methionine Synthase Reductasethe MS activation domain and transfers an electron to cob(II)alamin to yield cob(I)alamin.This reductive reaction is coupled with a methyl donation by S-adenosylmethionine.106In addition, MSR is also a special molecular chaperone for the conversion of MS apoen-zyme to holoenzyme. The presence of MSR not only stabilizes the apoenzyme but itsreductase activity on aquacobalamin also produces cob(II)alamin. This is a cobalamin formthat is more easily integrated into the apoenzyme.107MS is important for cell homeostasis for several reasons.108 First, CH3THF is the cir-culating form of folic acid and MS is the only enzyme that can convert it to tetrahydrofo-late. Tetrahydrofolate is an important biological cofactor for numerous pathways, includ-ing purine and pyrimidine biosynthesis. Second, ATP-dependent conversion of methionineyields S-adenosylmethionine (AdoMet), which is involved in all mammalian methylationsincluding nucleic acid, amino acid, protein, lipid, and secondary metabolite methylations.109Enzymatic hydrolysis of the byproduct of AdoMet methylation, S-adenosylhomocysteine,generates homocysteine, which is toxic to vascular endothelial cells in high levels.110 MS isvital for recycling homocysteine to methionine which, in turn, is needed in protein synthesisand reformation of AdoMet.MS is a dynamic multidomain enzyme consisting of four distinct modules. The N-terminal module comprises two tightly packed α/β barrel domains that form respectivesubstrate binding pockets for homocysteine and CH3THF (PDB 2CCZ).111 The centralmodule binds to cobalamin, a cobalt-containing cofactor. In the crystal structure of Es-cherichia coli methionine synthase (MetH), the cobalamin lies at the interface of a Rossman-like fold (βαβαβ motif) and a four helix bundle cap.112 The cap shields the cobalamin whensubstrates are not bound. Lastly, the C-terminal activation domain is a C-shaped mixed αand β domain with a central, bent antiparallel β-sheet with an AdoMet binding pocket.1131.3.4 Health ImplicationsThe importance of MSR in preserving the activity of MS is underscored by the clinicalconsequences of derangements in folate and homocysteine metabolism. Two highly prevalentpolymorphisms have been identified in the MSR-encoding gene MTRR in humans: c.66A>G(p.Ile22Met) and c.524C>T (p.Ser175Leu).103,114 Both polymorphisms result in MSR witha 4-fold reduced ability to reactivate MS.115 Women that have the c.66A>G polymorphismon both alleles have an increased risk of bearing children with neural tube closure defects ortrisomy 21 (Down’s Syndrome).116–119 Pregnant women with low plasma cobalamin levelsexasperate this risk.116,120Severe MSR dysfunction arises from the autosomal recessive inheritance of mutationsin the MTRR gene.103 Impaired MSR leads to a decrease in MS activity that resultsin hyperhomocysteinemia and hypomethioninemia.103,121 These metabolic conditions can281.4. Research AimsN N N NCoI N N N NCoIIN N N NCoIIICH3Homocysteine MethionineH4folate CH3H4folateCoCo Co FADFMNO2O2-NADPHNADP+AdoMetAdoHcye- FADFMNe-Figure 1.27: Catalytic cycle and reactivation mechanism of methionine synthase and me-thionine synthase reductase. For the modular methionine synthase, the homocysteine-,CH3THF-, cobalamin-, and activation domain are shown in orange, green, blue, and pur-ple, respectively. MSR is shown in purple.lead to megaloblastic anemia, severe neurological deficits, developmental delay, and blind-ness.122,123 Elevated plasma homocysteine levels are linked to an increase in risk for cardio-vascular disease and Alzheimer’s disease.122,124–126 Neurological disorders may be causedby a depletion of folate pools, thereby disrupting protein and DNA synthesis and cellulardivision.126 The absence of any MSR activity is embryonically lethal, as shown in MTRRgene knockout studies in mice.1231.4 Research AimsThe aim of this research was to investigate the structural origins of the differential ki-netics of electron transfer in two diflavin oxidoreductases, CPR and MSR. Superimpositionof primary and tertiary structures of CPR and MSR allowed for the design of structuralvariants based on conserved and nonconserved amino acid residues at the FAD/NADPH-binding domain. First of all, I sought out the role of the conserved FAD-stacking tryptophanresidue in the coenzyme binding and electron transfer properties in MSR. Through compar-ative mutagenesis studies with CPR, I aimed to determine if the homologous tryptophanretained the same functional role. Through each experiment, I strove to shed insight intothe regulation of electron flux in these enzymes. Secondly, I targeted potential active site291.4. Research Aimsresidues involved in coenzyme binding and examined their influence on catalysis. Finally,for an additional perspective of diflavin oxidoreductase catalysis, I extended my research toCPR from Artemisia annua, a plant species.In an effort to meet these research goals, conspicuous and subtle structural variantsof CPR and MSR were expressed and isolated. The overall catalytic performance of thesevariants was assessed by the steady state reduction of an artificial terminal electron acceptor.Furthermore, the strength of coenzyme binding in enzyme variants was measured by productand dead-end inhibition studies. Rapid enzymatic reduction kinetics were characterized bystopped-flow spectrophotometry. A comprehensive analysis of various enzyme variants bythese techniques revealed the functional role of specific structural elements in the catalyticmechanism of CPR and MSR.30Chapter 2Probing the Regulatory Role of anActive Site Trp697 in MethionineSynthase Reductase2.1 SummaryIn this chapter, the role of the FAD isoalloxazine ring-shielding tryptophan residue(Trp697) in MSR is investigated through various strategic amino acid exchanges. Twomutations are made to the penultimate Trp697 to reduce and remove the pi-pi-stacking in-teraction with the isoalloxazine ring: W697H and W697S. The C-terminal residue Ser698is also targeted for mutagenesis: S698A and S698∆. Steady state and pre-steady statekinetic properties of each MSR variant are characterized through the use of aerobic spec-trophotometry and anaerobic stopped-flow spectrophotometry. Through these experimentaltechniques, the large indole side chain of Trp697 is found to play an important regulatoryrole in coenzyme recognition and intramolecular electron flow in MSR. Interflavin electrontransfer is enhanced by the presence of Trp697. Bipartite binding of the coenzyme is sup-ported by shielding the FAD isoalloxazine ring, thereby ensuring substrate selectivity basedon the 2’-phosphate group.2.2 BackgroundFNR enzymes, including diflavin oxidoreductases, all have a conserved aromatic residuethat stacks against the re-face of the FAD isoalloxazine ring. The conserved aromatic is atyrosine in FNR, phenylalanine in nitric oxide synthase, and a tryptophan in cytochromeP450 reductase and methionine synthase reductase.49,65,127 X-ray crystallography of plantFNRs show that the tyrosine lies coplanar to maximize pi orbital overlap with the FADisoalloxazine ring.44 For direct hydride transfer, the nicotinamide ring must position theC4 pro-R hydrogen within 4A˚ of the N5 nitrogen of the isoalloxazine ring.61,128 In thisconformation (see Figure 2.1), the nicotinamide ring occupies the same space as the aromatic312.2. Backgroundresidue, and therefore productive substrate binding requires a structural rearrangementthat, at a minimun, involves residue displacement.44Early X-ray crystallography and mutagenesis studies in FNR have provided insight intothe role of the aromatic residue in coenzyme binding. Productively-bound nicotinamide hasonly been captured in FNR crystals upon deletion or mutation of the aromatic to a smallerside chain, suggesting that the residue does obstruct nicotinamide ring placement.62 Bymutating Tyr308 to a nonaromatic serine, the FNR variant has a much greater affinity forNADP+ and NAD+.63 Thus, Tyr308 destabilizes the nicotinamide ring such that the ener-getic cost of residue displacement outweighs the energetic gain of nicotinamide ring bind-ing.62,63,129 Moreover, the Tyr308Ser variant also significantly reduced the preference forNADPH over NADH. Equivalent mutations in NOS and CPR were shown to produce similareffects on substrate binding.61,130–132 By obstructing the nicotinamide ring of NADPH andNADH, the residue ensures the coenzyme binds according to the bipartite binding modelfirst described for FNR.44,62 Despite enhanced coenzyme binding, nonaromatic mutationsto the FAD active site residue generate reduced catalytic activities in both FNR and CPRvariants. The reduced catalytic activity is attributed to slow NADP+ release.61,63,132Figure 2.1: Crystal structure of MSR (PDB 2QTZ) superimposed with a NADPH-boundvariant rat CPR W676XS677X (PDB 1JAO) to illustrate the inevitable steric clash betweenTrp and the nicotinamide ring. The flavin is shown in yellow with the ribityl adenosine tailremoved for clarity. NADPH is in grey and the MSR C-terminal residues, Trp697 andSer698, are shown in light purple with the standard elemental colouring.For MSR, the re-face of the FAD isoalloxazine ring forms pi-pi interactions with the indolering of a tryptophan residue (Trp697). X-ray crystallography shows that the entire indolering of Trp697 overlaps with the isoalloxazine ring.65 In NADP+-bound MSR, only electron322.3. Resultsdensity for the bound 2’,5’-ADP moiety is observed, which indicates that the nicotinamidering moiety is delocalized. Isothermal titration calorimetry revealed that 2’,5’-ADP bindsto MSR with a 14-fold stronger affinity than NADP+.65 As in FNR and other diflavin oxi-doreductases, the coenzyme binds to MSR in a bipartite manner which promotes a strongpreference for NADPH over NADH. Thus, the nicotinamide ring energetically disfavourscoenzyme binding which, by analogy with FNR, is attributed to the energetic cost of aro-matic residue displacement. To determine if the penultimate Trp697 has a differential rolein coenzyme binding in MSR, the effects of mutations to this residue are examined. Fur-thermore, since coenzyme binding precedes hydride transfer and subsequent electron flowthrough the enzyme, the effects of mutations to Trp697 on the catalytic activities of humanMSR are also investigated.The functional role of C-terminal residues, Trp697 and Ser698, in regulating coenzymebinding and electron flux in MSR was determined by mutagenesis. Two mutations ofTrp697 were made: W697H and W697S. The imidazole side chain of W697H reduces thepi-pi stacking interaction with the FAD isoalloxazine ring. This variant is designed to reducethe energetic cost of residue displacement by the coenzyme. The W697S variant abolishesany pi orbital overlap with the isoalloxazine ring of FAD and is anticipated to significantlyimprove coenzyme binding by favouring the productive nicotinamide ring conformation.Structural rearrangement of Trp697 is expected to confer movement to the adjacent C-terminal Ser698, and therefore the S698∆ and S698A mutations were constructed. TheS698∆ variant was designed to examine the effect of the additional conformational changeof the C-terminal serine backbone on coenzyme binding, while the S698A variant assessedthe potential contribution of the hydroxyl side chain. The steady state kinetic propertiesof each variant were measured by spectrophotometric analysis of cytochrome c reduction.Coenzyme binding was evaluated through product and dead-end inhibition studies withNADP+ and 2’,5’-ADP, respectively. Finally, stopped-flow spectrophotometry allowed fordetection of the rapid reduction of each variant with saturating NADPH.2.3 Results2.3.1 Steady state kinetic dataMSR-catalyzed reduction of cytochrome c3+ was used to determine the steady statekinetic parameters for the four variants. The data are summarized in Table 2.1. Cytochromec3+ is widely used as a nonphysiological terminal electron acceptor for the kinetic analysis ofdiflavin oxidoreductases, making it an ideal substrate for comparative purposes. It also onlyaccepts an electron from the FMN cofactor, therefore catalytic turnover, k cat, encompassesboth the hydride transfer to FAD and the FAD to FMN electron transfer steps. For wild-332.3. ResultsTable 2.1: Steady state kinetic parameters of wild-type and variant MSR with NADPH.Conditions: 1 mL reaction volume, 8 µM cytochrome c3+, 2 pmole enzyme, varied [NADPH]and [inhibitor], 50 mM Tris-HCl pH 7.5 at 25 ◦C. Assays were done in triplicate and fit tothe Michaelis-Menten equation. Inhibition data from four inhibitor concentrations were fitto a competitive inhibition equation.a Data acquired from Wolthers et al.65MSR Inhibitor k cat Km Ki k cat/Km(s−1) (M × 10−6) (M × 10−6) (s−1M−1 × 10+6)aMSRWT NADP+ 7.23 ± 0.14 2.37 ± 0.17 37 ± 3 3.1 ± 0.32’,5’-ADP 6.62 ± 0.13 1.95 ± 0.14 1.44 ± 0.11 3.4 ± 0.3S698∆ NADP+ 12.2 ± 0.4 1.95 ± 0.18 2.2 ± 0.3 6.2 ± 0.82’,5’-ADP 13.9 ± 0.5 2.7 ± 0.3 0.45 ± 0.05 5.2 ± 0.8S698A NADP+ 3.2 ± 0.2 4.7 ± 1.1 9 ± 2 0.7 ± 0.22’,5’-ADP 2.56 ± 0.13 5.6 ± 0.8 0.43 ± 0.06 0.45 ± 0.09W697S NADP+ 0.64 ± 0.02 0.25 ± 0.06 0.09 ± 0.02 2.6 ± 0.72’,5’-ADP 0.69 ± 0.01 0.44 ± 0.06 2.4 ± 0.5 1.6 ± 0.2W697H NADP+ 15.3 ± 0.6 4.0 ± 0.5 26 ± 5 3.8 ± 0.62’,5’-ADP 18.3 ± 0.5 3.8 ± 0.3 0.98 ± 0.08 4.8 ± 0.5type MSR, the catalytic turnover, k cat, for cytochrome c3+ reduction is 3-7 s−1.65 Anenhanced k cat of 15 s−1 and 12 s−1 is observed for W697H and S698∆, respectively. Therate of reduction for W697S is reduced by 12-fold to 0.6 s−1 while that for S698A (3s−1) remains similar to wild-type. The catalytic efficiency (k cat/Km) with NADPH for allvariants is also comparable to wild-type, with the exception of a 6-fold less efficient S698A.Steady state inhibition studies were conducted to measure the affect of the amino acidvariations on the binding of NADP+ and 2’,5’-ADP. The results are compiled in Table2.1. Inhibition constants, Ki, for 2’,5’-ADP binding in the variants range from 0.4 µMto 2.4 µM, similar to wild-type. Across all variants, the 2’,5’-ADP moiety elicits a tighterbinding affinity over NADP+, as observed for wild-type MSR, with the exception of W697S.Stronger binding affinity for the oxidized coenzyme was measured in all variants, with thehighest binding affinity observed for W697S (Ki = 0.09 µM). Notably, the W697S variantalso elicited the slowest catalytic turnover. Next strongest is S698∆ with Ki of 2 µM,followed by S698A (Ki = 9 µM) and then W697H (Ki = 25 µM).342.3. Results0 . 00 . 51 . 01 . 52 . 02 . 53 . 03 . 54 . 0   NADPH oxidation (min-1)W i l d - t y p e S 6 9 8 ∆ S 6 9 8 A W 6 9 7 S W 6 9 7 HFigure 2.2: The NADPH oxidation activity of MSR variants. Conditions: 1 mL reactionvolume, 2 pmole enzyme, 1 mM NADPH, 50 mM Tris-HCl pH 7.5 at 25 ◦C. The averageof three replicates and the associated error are listed above for each variant.2.3.2 Uncoupled NADPH oxidationTo dissect the individual steps of the catalytic mechanism that are affected by the activesite substitutions, uncoupled NADPH oxidase activity was evaluated. The results are shownin Figure 2.2. Uncoupled NADPH oxidase activity is defined as flavin reduction by NADPHand the subsequent flavin re-oxidation by O2. Therefore, the observed rates of NADPHoxidation are not dependent on interflavin electron transfer. W697H and S698∆ exhibitedenhanced NADPH oxidase activity of 8- and 10-fold, respectively. S698A oxidase activitywas modestly slower than wild-type, while that of W697S was reduced by approximatelyhalf. As the interflavin electron transfer step is excluded from these activity studies, theenhanced steady state turnover rates observed are accredited to improved hydride transferand/or coenzyme binding/dissociation.2.3.3 Coenzyme PreferenceWild-type MSR possesses a strong >19 000-fold preference for 2’-phosphorylatedNADPH over NADH.65 To test if this coenzyme specificity is preserved in the W697variants, the kinetic properties using NADH as an electron donor were measured, andthe results are summarized in Table 2.2. Coenzyme specificity is reported by the ra-tio of [(k cat/KNADPH)/(k cat/KNADH)]; W697H preference for NADPH dropped to 3473-fold, while W697S almost lost coenzyme discrimination with only a 20-fold preference for352.3. ResultsTable 2.2: Steady state kinetic parameters of wild-type and W697 MSR variants withNADH. Conditions: 1 mL reaction volume, 2 pmole enzyme, 8 µM cytochrome c3+, var-ied [NADH], 50 mM Tris-HCl pH 7.5 at 25 ◦C. Assays were done in triplicate and fit toMichaelis-Menten equation. a Data acquired from Wolthers et al.65MSR k cat Km k cat/Km (k cat/KNADPH)/(s−1) (M ×10−3) (s−1M−1 ×10+3) (k cat/KNADH)aMSRWT 0.24 ± 0.03 3.5 ± 0.6 0.068 ± 0.014 19400W697S 8.1 ± 0.4 0.06 ± 0.01 131 ± 30 20W697H 4.7 ± 0.9 4.5 ± 0.4 1.1 ± 0.3 3473NADPH. Aromatic stacking with the isoalloxazine ring clearly imparts regulatory controlin the proclivity of NADPH use versus NADH.2.3.4 Multiple wavelength pre-steady state kineticsPre-steady state reduction of MSR variants with saturating NADPH was followed bystopped-flow spectrophotometry in an oxygen-free environment. Photodiode array detectionallowed for the recording of UV-visible spectra from 380 to 700 nm over a select time domain.Figure 2.3 shows the full spectra collected for S698A over 200 s by this method. These spec-tral data are then resolved by the singular value decomposition (SVD) algorithm to give anapproximate number of spectral intermediates. This analysis aids in determining the bestmodel for use in global analysis, see Figure 2.4 and Table 2.3. As for wild-type MSR, theS698 variants were best fit to a four species, triphasic irreversible model (a→b→c→d).133W697 variants fit best to a three species, biphasic model (a→b→c). The intermediatespectral profiles and their respective observed rate constants of formation/decay representan equilibrium distribution of enzyme intermediates that form over the course of the re-action. As such, the spectral intermediates (a, b, c, d) do not represent distinct enzymeintermediates.The first kinetic phase is assigned to the reduction of fully oxidized enzyme, spectralspecies a, to a partially reduced species b. Across all variants, this kinetic phase (kobs1)shows only a partial loss of absorbance at the flavin maxima 454 nm. Based off of this ab-sorbance change, the first kinetic phase represents hydride transfer to FAD to form FADH2.The fastest observed rate constants for hydride transfer were observed for W697H andS698∆ (115 s−1 and 82 s−1, respectively), which are 4- and 3-fold faster than wild-typeMSR. The rate constant for S698A is comparable to native MSR. The W697S variant,however, is 8-fold slower (3 s−1).362.3. Results4 0 0 5 0 0 6 0 0 7 0 00 . 0 00 . 0 50 . 1 0  Absorbance W a v e l e n g t h  ( n m )Figure 2.3: Anaerobic reduction of MSR variant S698A with saturating NADPH monitoredwith multiple wavelength stopped-flow spectrophotometry. Conditions: 10 µM enzyme, 100µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Full spectral reduction of S698A,300 scans over 200 s, is reduced to ∼20 traces for clarity.Table 2.3: Observed rate constants of the pre-steady state reduction of MSR variants. Con-ditions: 10 µM enzyme, 100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. S698variant reduction was best fit to a three-step kinetic model while W697 variant reductionwas best fit to a two-step kinetic model.a Data acquired from Wolthers et al.65MSR kobs1 (s−1) kobs2 (s−1) kobs3 (s−1)aMSRWT 24.9 ± 0.1 0.18 ± 0.01 0016 ± 0.003S698∆ 82.1 ± 0.7 2.01 ± 0.04 0.019 ± 0.001S698A 27.2 ± 0.3 2.33 ± 0.01 0.025 ± 0.001MSR kobs1 (s−1) kobs2 (s−1)W697S 3.4 ± 0.1 0.011 ± 0.001W697H 115.3 ± 0.6 0.008 ± 0.001372.3. Results01 02 03 04 0 0 4 5 0 5 0 0 5 5 0 6 0 0 6 5 001 02 03 04 0 0 4 5 0 5 0 0 5 5 0 6 0 0 6 5 0 A  Molar Absorbtivity (mM-1 cm-1 ) abcd   B  abcdCcbaMolar Absorptivity (mM-1 cm-1 )W a v e l e n g t h  ( n m )   W a v e l e n g t h  ( n m )a bcDFigure 2.4: Spectral profiles for anaerobic reduction of MSR variants by saturating NADPHmonitored with multiple wavelength stopped-flow spectrophotometry. Conditions: 10µMenzyme, 100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Time-resolvedspectral intermediates for S698A (A), S698∆ (B), W697H (C ), and W697S (D). S698variants follow a three-step kinetic model with four discrete spectral species (a→b→c→d)and W697 variants follow a two-step kinetic model with three discrete spectral species(a→b→c), associated rate constants are summarized in Table 2.3.382.3. ResultsThe second kinetic phase (kobs2) for the S698 variants is attributed to interflavin elec-tron transfer, and this is represented by a further decrease in absorbance at 454 nm andan increase in absorbance at 600 nm. The latter absorbance change is indicative of theformation of the disemiquinone species E-FADH•-FMNH•. Both S698A and S698∆ elicita kobs2 of ∼2 s−1 for the conversion of spectral species b to c. Over 100 s, the formationand decay of the disemiquinone is observed at 600 nm, see Figure 2.7. The third phase,conversion of species c to d, represents the binding of a second molecule of NADPH andreduction to the four-electron reduced state E-FADH2-FMNH2. This final phase, kobs3, isequivalent to wild-type MSR for both S698 variants (∼0.02 s−1).The kinetic profile for the W697 variants are notably different. For these variants,reduction occurs in two kinetic phases without an observable disemiquinone intermediate.As stated above, the first kinetic phase occurs with the absorbance loss at 454 nm. For bothvariants, the associated amplitude change for the first kinetic phase is less than that observedin wild-type and S698 variants. Furthermore, unlike the native enzyme, the second kineticphase is not accompanied by an increase in absorbance at 600 nm. Inspection of absorbancetraces at 600 nm over 100 s (Figure 2.7) reveals a lack of any detectable disemiquinone.Thus, the second kinetic step encompasses the conversion of two-electron reduced species bto the full four-electron reduced species c by a second NADPH molecule. The kobs2 for thissecond kinetic phase is ∼0.01 s−1 for both W697 variants.The fact that W697H and W697S are reduced to the full four-electron reduced state(evident by the final absorbance spectra), indicates that interflavin electron transfer occursin these enzyme variants. This is further confirmed by the ability of the two variantsto reduce cytochrome c3+, which accepts electrons from the FMN cofactor. To furthertest that the W697 variants are capable of interflavin electron transfer and formation ofthe disemiquinone species, they were rapidly mixed with an equimolar concentration ofNADPH to produce a two-electron reduced species. As predicted, a decrease in absorbanceat 454 nm with an increase at 600 nm marked the slow interflavin exchange of electronsas equilibrium is established over 750 s (Figure 2.5). These results suggest that the lack ofdetectable disemiquinone is a consequence of attenuated interflavin electron transfer, ratherthan the inability of the enzymes to form the disemiquinone species.2.3.5 Single wavelength pre-steady state kineticsEnzymatic reduction was followed at the flavin and disemiquinone absorption max-ima 454 nm and 600 nm, respectively, by anaerobic stopped-flow spectrophotometry. Thecollected traces are shown in Figure 2.6 and Figure 2.7. Data acquisition over a singlewavelength generates stopped-flow traces which facilitate determination of individual, fastrate constants. Fitting traces at 454 nm to a double-exponential equation produced two392.4. Discussion4 0 0 5 0 0 6 0 00 . 0 00 . 0 50 . 1 00 . 1 50 . 2 04 0 0 5 0 0 6 0 0 7 0 0W a v e l e n g t h  ( n m )AAbsorbanceW a v e l e n g t h  ( n m )BFigure 2.5: Anaerobic reduction of MSR variants by equimolar NADPH monitored withmultiple wavelength stopped-flow spectrophotometry. Conditions: 10µM enzyme, 10 µMNADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. W697S (Panel A) and W697H (PanelB) were rapidly mixed with NADPH and monitored over 750 s.observed rate constants for each variant over a 10 s time frame. As in the PDA data anal-yses, kobs1 is fastest for W697H (136 s−1) followed by S698∆ (106 s−1), S698A (23 s−1)and W697S (3.5 s−1). The second kinetic phase for S698∆ and S698A occurs with a rateconstant kobs2 of 3.7 s−1 and 1.6 s−1, respectively. An extended time frame of 500 s capturesthe complete reduction of W697S and W697H and yields a kobs2 of 0.005 s−1 for both whichare comparable to the PDA acquired rate constants.The dependence of kobs1 on the concentration of NADPH for each variant was also ana-lyzed by single wavelength absorption studies at 454 nm. In wild-type MSR, the hyperbolicdependence of kobs1 on NADPH concentration allows for the extraction of a dissociationconstant, Kd.133 Figure 2.8 shows the independence of kobs1 on [NADPH] observed for allMSR variants, which suggests that the binding of NADPH is much tighter for these variants.2.4 Discussion2.4.1 Regulation of coenzyme preference by Trp697Coenzyme preference is clearly established in native MSR with a strong 19 400-foldpreference for NADPH over NADH.65 The two coenzymes only differ by a 2’-phosphategroup. At the coenzyme binding cleft, the 2’-phosphate group sits in a pocket of well-conserved polar residues. The strong affinity for the 2’,5’-ADP moiety is attributed to402.4. DiscussionFigure 2.6: Anaerobic reduction of MSR variants by saturating NADPH monitored withsingle wavelength stopped-flow spectrophotometry at 454 nm. Conditions: 10 µM enzyme,100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. S698A (Panel A), S698∆(Panel B), W697H (PanelC ), and W697S (Panel D) were rapidly mixed with NADPH andmonitored over 10 s at 454 nm. Insets for Panel C and D are the linear absorbance tracesat 454 nm over 500 s to show completion.412.4. Discussion0 2 0 4 0 6 0 8 00 . 0 0 40 . 0 0 80 2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 00 2 0 4 0 6 0 8 00 . 0 0 40 . 0 0 80 2 0 4 0 6 0 8 0 1 0 0  Absorbance 600nm T i m e  ( s )ACB   T i m e  ( s )  Absorbance 600nm T i m e  ( s )D   T i m e  ( s )Figure 2.7: Anaerobic reduction of MSR variants by saturating NADPH monitored withsingle wavelength stopped-flow spectrophotometry at 600 nm. Conditions: 10 µM enzyme,100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. S698A (Panel A), S698∆(Panel B), W697S (PanelC ), and W697H (Panel D) were rapidly mixed with NADPH andmonitored over 100 - 150 s at 600 nm.3 0 4 0 5 0 6 0 7 0 8 0 9 0 1 0 002 04 06 08 01 0 01 2 01 4 0   k obs1 (s-1 ) [ N A D P H ]  ( µM )Figure 2.8: Dependence of kobs1 on NADPH concentration. S698A (), S698∆(•), W697H(N) and W697S (H).422.4. Discussionpolar interactions with these residues. While the 2’,5’-ADP moiety binds tightly, the bindingof full-length NADP+ is 15-fold weaker, indicating that the nicotinamide mononucleotide(NMN) moiety does not contribute favourably to the coenzyme binding energy.By exchanging the bulky indole side chain of Trp697 for a smaller side chain, a loss inNADPH specificity over NADH was observed. The mutations did not affect the 2’,5’-ADPbinding site, as the inhibition constants (Ki) for 2’,5’-ADP remained similar to wild-type.However, the resulting Ki for NADP+ revealed a more pronounced effect on the binding ofthe NMN moiety among the variants. In W697H, the affinity for NADP+ is only 1.5-foldstronger, indicating that the aromatic imidazole can partially substitute for the indole indisrupting NMN binding. For this variant and both S698 variants, the two-step bindingof NADPH is still preserved where the 2’,5’-ADP first binds tightly and then the NMNmoiety moves into the FAD active site. The tighter binding of NADP+ reflects the reducedenergetic cost of productive nicotinamide binding in these variants. In contrast, bipartitebinding is not maintained in W697S since NADP+ binds with a greater affinity than 2’,5’-ADP. The energetic cost of NMN binding is substantially reduced as removal of the aromaticresidue likely exposes the re-face isoalloxazine ring to solvent, which would favour bindingof the nicotinamide ring. Therefore, both binding sites are immediately accessible. As aconsequence, the initial selection for the 2’-phosphorylated substrate by the polar pocketis lost, hence the weak 64-fold preference for NADPH over NADH in W697S. Thus, thepresence of Trp697 controls substrate preference for NADPH over NADH.2.4.2 Regulation of hydride transfer by Trp697Hydride ion transfer to the isoalloxazine ring occurs directly from the nicotinamide ring,therefore variants that reduce the energetic barrier for nicotinamide stacking against theflavin are expected to accelerate this first chemical step. The first observed rate constant,kobs1, is assigned to hydride transfer. For both W697H and S698∆, the rate constant isindeed accelerated by 5- and 6-fold respectively, while the rate constant was unchanged forS698A. The accelerated rate constant for W697H is likely attributed to the reduced pi-pistacking between the FAD isoalloxazine ring and the imidazole side chain, which presumablylowers the thermodynamic barrier for residue displacement. Likewise, the increase in kobs1for S698∆ suggests that the C-terminal residue may also contribute to the thermodynamicbarrier to Trp697 conformational change, possibly through steric constraint. Thus, removalof S698 would alleviate some conformational restriction on Trp697 displacement. Since theequivalent rate constant for S698A is the same as wild-type, the steric constraint likelyoriginates from the C-terminal residue backbone rather than the Cβ-hydroxyl side chain.Although W697S is expected to ameliorate productive NADPH binding by removing thethermodynamic barrier for residue displacement, the variant elicited a 8-fold slower kobs1.432.4. DiscussionSince it exhibits strong coenzyme affinity, the W697S variant may be limited by NADP+release. However, no evidence of a charge-transfer complex between the nicotinamide ringand the FAD cofactor was observed in the stopped-flow data. The bipartite binding modeof NADPH is impaired in W697S and may be the origin of the slower kobs1. Withoutthe initial docking of the 2’,5’-ADP moiety, the nicotinamide ring may adopt a number ofnonproductive binding states before attaining a catalytically relevant orientation. Moreover,while the FAD active site is more accessible to the nicotinamide ring in W697S, as serineoccupies less space than the bulky tryptophan, the distance between the nicotinamide andisoalloxazine rings may not be optimal for hydride transfer.2.4.3 Regulation of interflavin electron transfer by Trp697During the MSR reductive half-reaction, the disemiquinone species forms as an interme-diate. This is evident by the appearance and subsequent disappearance of an absorbanceband centered at 600 nm. For the W697 variants, the disemiquinone intermediate doesnot accumulate to a detectable amount. However, interflavin electron transmission doesstill take place, as both W697 variants are able to catalyze cytochrome c3+ reduction,fully reduce to the four-electron level under saturating NADPH conditions, and achieve thedisemiquinone state upon equilibration with equimolar NADPH. Therefore, while electrontransfer between the flavins is still possible in the W697 variants, the presence of Trp697in wild-type and the S698 variants accelerates this step and allows for buildup of the dis-emiquinone.The mechanism by which Trp697 accelerates interflavin electron transfer is still un-clear, however given that the Trp697 is positioned at the FAD/FMN domain interface, itis conceivable that the residue influences the conformational equilibria of the enzyme. Theenzyme is envisioned to alternate between a closed conformation for intramolecular electrontransfer and an open conformation for intermolecular electron transfer. The open state in-creases the distance between flavins, thereby disfavouring interflavin electron transfer. Theclosed state, captured in the crystal structure of rat CPR, brings the flavins less than 4 A˚together in an arrangement that is anticipated to allow for rapid electron exchange.49 SinceTrp697 lies at the FAD/FMN domain interface, it is proposed that the residue stabilizesthe interdomain contact to favour the closed state in which the interflavin distance is mini-mized and interflavin electron transfer is favoured. Thus, mutation of this residue may shiftthe conformational equilbiria to the open state which would attenuate interflavin electrontransfer.Despite the enhanced rate constants associated with hydride transfer and uncoupledNADPH oxidation in W697H and S698∆, the variants only modestly increased the rate ofsteady state cytochrome c3+ reduction. Cytochrome c3+ receives a single electron from the442.4. DiscussionFMN cofactor, thus reduction of cytochrome c3+ reflects the sum of all the forward andreverse rate constants of the catalytic mechanism, including NADP(H) binding and release,hydride transfer, and interflavin electron transfer. For W697H, the faster rate constant forhydride transfer is offset by slower disemiquinone formation, so turnover is only modestlyaffected. S698∆ elicited a faster rate constant for hydride transfer, interflavin electrontransfer, and a strong NADPH binding affinity. However, the turnover for this variant maybe limited by the slower dissociation of the tightly bound oxidized coenzyme for the nextround of catalysis.2.4.4 Extension of regulatory role of tryptophan to CPRWild-type human CPR evokes an over 4 300-fold preference for NADPH over NADH.Previous studies have shown that by exchanging the Trp676 residue (CPR equivalent ofTrp697) for a histidine or alanine, the substrate affinity is greater and the preference forNADPH drops. A dramatic decrease was measured for W676A whose preference for NADPHover NADH is only 4-fold. The shift in coenzyme selectivity is attributed to the loss ofbipartite binding due to a lack of steric control by the active site Trp676.132For full-length CPR, hydride transfer is tightly coupled to interflavin electron transfer,and this is observed in one kinetic phase with a kobs1 of 20 s−1.67 For CPR W676H, thisrate constant decreases to 3 s−1 and the enzyme only reduces to the two-electron level.Interestingly, the kobs1 for W676H reduction is the same as that measured for the wild-typeFAD domain. In full-length native CPR, the hydride transfer step is not thermodynamicallyfavourable since the midpoint potential of the hydride-receiving FADox/hq redox couple (-340mV) is more electronegative than that of NADP(H) (-320 mV).66,67 However, the presenceof the very electropositive FMNox/sq couple (-66 mV) shifts the equilibria towards reducedFMN. In the isolated FAD domain, the ∼ 7-fold slower kobs1 is a consequence of the lostthermodynamic pull by FMN, and this is reflected by an attenuated rate constant. Althoughthe FMN domain is still present in W676H, the lack of reduction past the two-electron level,along with the reduced kobs1 and slow overall catalytic turnover, indicates that interflavinelectron transfer is impeded in this variant.If the role proposed for Trp697 in accelerating interflavin electron transfer in MSR isapplied to CPR, then the histidine variant would be expected to slow down the kobs1 inCPR but not in MSR. In W697H of MSR, the kobs1 for hydride transfer is not negativelyaffected by the reduced interflavin electron transfer because the two events are kineticallydistinct. In contrast, the two kinetic events are coupled in CPR. As such, the slow kobs1 ofCPR W676H is due to disruption of electron transfer from FAD to the more electropositiveFMN, thereby adversely affecting the thermodynamic drive of electron flux through the452.5. Experimental Proceduresenzyme. Thus, the role of the conserved active site tryptophan in accelerating interflavinelectron transfer is conserved in CPR and MSR.2.5 Experimental Procedures2.5.1 MaterialsUnless otherwise stated, all chemical reagents were purchased from Fisher. The reagentsNADPH, NADH, NADP+, 2’,5’-ADP, and cytochrome c3+ were obtained from SigmaAldrich (Oakville, ON, Canada). Pfu Turbo DNA polymerase, Taq DNA polymerase andXl1 Blue cell lines were purchased from Agilent Technologies (Mississauga, ON, Canada).Rosetta(DE3)pLysS competent cells were obtained from EMD Biosciences. Protein purifi-cation supplies, Resource Q column and glutathionine sepharose 4B resin, were purchasedfrom GE Biosciences.2.5.2 Generation and expression of MSR tryptophan variantsThe W697S, W697H, S698∆ and S698A MSR variants were generated from the wild-typeMSR plasmid template pGEX-4T1-MSR using the QuikChange Site-Directed MutagenesisKit (Agilent Technologies). Oligonucleotide primers were designed based on the publishedsequence (Accession number AF121214) and purchased from Integrated DNA Technolo-gies (Coralville, Iowa, USA). The primers used are tabulated in Table 2.4. The NAPSDNA Sequencing Laboratory of the University of British Columbia (Vancouver, Canada),confirmed the desired amino acid mutations and the absence of additional PCR-inducederrors. Successfully designed mutant plasmids were transformed into the Escherichia colistrain Rosetta2(DE3)pLysS using the heat shock method. Recombinant GST-fusion pro-teins were grown in 200 mL cultures of Luria-Bertani medium, containing 100 µg mL−1of ampicillin and 34 µg mL−1 chloramphenicol, overnight at 37 ◦C and 225 rpm using aInnova44 incubator shaker. With 10 mL of the 200 mL cultures, 0.5 L cultures of TerrificBroth containing 100 µg mL−1 ampicillin and 34 µg mL−1 chloramphenicol were inoculatedand grown at 30 ◦C to an Abs600 of 0.8. Cultures were then induced with 0.1 mM IPTGand the temperature was lowered to 25 ◦C overnight. The cells were harvested at 6,360 ×g for 10 min and stored at −80 ◦C.2.5.3 Purification of MSR variantsAll purification steps were performed on ice or at 4 ◦C. Cells overexpressing the recom-binant enzyme were resuspended in 200 mL of 1×GST bind/wash buffer (10×GST = 43mM Na2HPO4, 14.7 mM KH2PO4, 1.37 M NaCl, 27 mM KCl, pH 7.3 with 1 mM EDTA,462.5. Experimental ProceduresTable 2.4: The forward (F) and reverse (R) oligonucleotide primers designed for each MSRC-terminal variant. Mutation is in bold.MSR Variant Oligonucleotide SequenceS698∆ F 5’ CTTCAGGATATTTGG∆TAGCGGCCGCATCGTGAC 3’R 5’ GTCACGATGCGGCCGCTA∆CCAAATATCCTGAAG 3’S698A F 5’ CTTCAGGATATTTGGGCATAGCGGCCGCATCGTGAC 3’R 5’ GTCACGATGCGGCCGCTATGCCCAAATATCCTGAAG 3’W697H F 5’ CTTCAGGATATTCATTCATAGCGGCCGCATCGTGAC 3’R 5’ GTCACGATGCGGCCGCTATGAATGAATATCCTGAAG 3’W697S F 5’ CTTCAGGATATTTCGTCATAGCGGCCGCATCGTGAC 3’R 5’ GTCACGATGCGGCCGCTATGACGAAATATCCTGAAG 3’1 mM DTT) and protease inhibitors benzamidine (2 mM) and PMSF (1 mM). The cellmixture was sonicated with the Sonicator S-4000 (Misonix Inc.) at 22% power amplitudewith alternating 8 s pulses and 1 min pauses for a total of 45 min. Cell lysate was thencentrifuged at 39,120 × g for 45 minutes to separate the soluble protein from cellular debris.The supernatant was collected and loaded onto a Glutathione-Sepharose 4B (GE Health-care) column equilibrated with 1×GST bind/wash buffer and was washed with 5 columnvolumes of 1×GST bind/wash buffer. The GST-tagged protein was eluted with the elutionbuffer (50 mM Tris-HCl, 1 mM EDTA and DTT, and 10 mM glutathione). The eluate wasthen dialyzed overnight with 20 U/mg thrombin (GE Healthcare) at 4 ◦C in dialysis buffer(4 L of 1×GST bind/wash buffer, 1 mM EDTA and 1 mM DTT). The dialyzed sample wasagain applied to a 1×GST-equilibrated column and the cleaved protein was collected fromthe flow-through. The sample was further purified by anion exchange chromatography on a1 mL Resource Q column (GE Healthcare) using an AKTApurifier system (GE Healthcare)with a linear gradient from 50 to 500 mM NaCl in 50 mM Tris-HCl pH 7.5 at a flow rate of 2mL/min. The desired fractions were collected based on the absorbance readings at 280 nmand the yellow colour that is characteristic of oxidized flavoproteins. The final concentrationof purified MSR was determined by the absorption at 454 nm and the Beer-Lambert Law,Equation 3.1.Abs = εlc (2.1)where c is the concentration of sample, Abs is the absorbance at 454 nm, ε is the molarextinction coefficient of MSR (25.6 mM−1cm−1), and l is the light path length (1 cm).472.5. Experimental ProceduresThe purified MSR mutants were flash-frozen in liquid nitrogen and stored in 20% glycerolat −80 ◦C for later steady-state assays and pre-steady state stopped-flow analysis.2.5.4 Steady state turnover analysisThe rate of cytochrome c3+ reduction was assessed for each variant at 25 ◦C by thechange in absorbance at 550 nm (∆ ε = 21.1 mM−1cm−1) on a Lambda 25 UV/Vis Spec-trometer (Perkin Elmer). The 1 mL reaction mixture contained 50 mM Tris-HCl buffer,pH 7.5, 8 µM cytochrome c3+, and varied NADPH concentrations from 0.25 µM - 50 µM,and the reaction was initiated by 2 pmoles of the MSR variant. Reaction velocites for eachNADPH concentration were collected in triplicate and the data were fit to the Michaelis-Menten equation by non-linear least-squares regression analysis using Origin 8.5 software(OriginLab Co.). For the inhibition assays, the reaction mixtures contained 50 mM Tris-HCl, pH 7.5, 8 µM cytochrome c3+, and varying concentrations of substrate, NADPH, andinhibitor (NADP+, 2’,5’-ADP); the inhibition reactions were initiated by 2.0 × 10−12 moleof MSR. Product and dead-end inhibition assays were triplicated and fit with non-linearleast-squares regression analysis to the competitive inhibition equation, Equation 2.2 in theOrigin 8.5 software (OriginLab Co.)νi =V AKm(1 + IKi ) +A(2.2)where νi is the initial velocity, V is the maximal velocity, A is the varied substrateconcentration, Km is the apparent Michaelis constant, I is the inhibitor concentration, andKi is the inhibition constant.Uncoupled NADPH oxidation 1 mL reactions were carried out at 25 ◦C in 50 mM Tris-HCl, pH 7.5 with saturating 1 mM NADPH. Consumption of NADPH was moniteredspectrally at 340 nm over 90 s. The assays were performed in triplicate. Initial veloc-ity (M/min−1) was calculated from the slope at 340 nm using the extinction coefficient of6.22 mM−1cm−1.2.5.5 Pre-steady state kinetic analysisStopped-flow studies were performed under anaerobic conditions using the SF-61DX2Stopped-flow apparatus from TgK Scientific in a Belle Technology glove box. Kinetic exper-iments were conducted in 50 mM Tris-HCl pH 7.5 that was degassed by extensive bubblingwith nitrogen gas and allowed to equilibriate in the glove box overnight. Protein sampleswere brought into the glove box and oxygen was removed via gel filtration over Bio-RadEcono-Pac 10 DG column equilibrated with anaerobic 50 mM Tris-HCl pH 7.5. Substratewas dissolved in anaerobic 50 mM Tris-HCl pH 7.5 buffer. A diluted sample was taken out482.5. Experimental Proceduresof the glove box to determine the concentration spectrally at 340 nm ( = 6.22 mM−1cm−1).For single and multiple wavelength studies, pseudo-first order conditions were establishedwith a 10-fold saturating concentration of substrate. Two syringes were filled with an initialconcentration of 20 µM enzyme and 200 µM substrate, respectively.Multiple wavelength absorption changes over substrate-mediated reduction were moni-tored with a photodiode array detector (PDA) (TgK Scientific). Substrate and MSR variantsamples were both diluted 2-fold after mixing. The spectra were measured over 200 s withthe monochromator auto-shutter on to reduce the likelihood of photoreduction. Spectralspecies were resolved by applying the singular value decomposition (SVD) algorithm tothe spectral data in ReactLab Kinetics (Jplus Consulting Pty Ltd., Karawara, Australia).SVD is an unbiased algorithm that decomposes the original data matrix into a reduced setof output data that is defined in terms of the linearly independent components and theirweighted significance. This allows for the determination of the number of spectral species ina reaction model. Global analysis of the reduced SVD data were best fitted to a three-step(a→b→c→d) or two-step (a→b→c) model with four or three discrete spectral species.For single wavelength studies, the reduction of NADPH was followed at 454 nm witha photomultiplier detector at 25 ◦C with varying NADPH concentrations. The substrateand MSR sample were both diluted 2-fold after mixing. An average of 3 traces of thesingle-wavelength data were fitted to a double-exponential equation (Equation 2.3)A = C1e−kobs1t + C2e−kobs2t + b (2.3)where kobs1 and kobs2 are the observed rate constants for the fast and slow phases,respectively. C1 and C2 are their relative amplitudes, and b is the final absorbance. Thedata were extracted from an Excel file generated from ReactLab and imported into Originfor graphical representation.49Chapter 3Comparative investigation into therole of coenzyme-coordinatingresidues and the FAD-shieldingtryptophan in coenzyme bindingand electron flux in MSR and CPR3.1 SummaryIn this chapter, the functional role of the FAD isoalloxazine ring-stacking tryptophan isfurther investigated in MSR with a comparative study in CPR. In addition, the influenceof stabilizing interactions between active site residues and the coenzyme on MSR and CPRcatalytic properties is examined. First, conservative amino acid mutagenesis of the activesite tryptophan to phenylalanine and tyrosine assessed the effect of residue side chain sizeand polarity on catalytic properties in MSR (W697F and W697Y) and in CPR (W676Fand W676Y). Second, strategic mutations were made to coenzyme-coordinating residuesto strengthen or weaken the interaction with NADPH in MSR (K291R and A622K) andin CPR (R298A and K602AV603K). Aerobic and anaerobic spectrophotometric techniqueswere employed to characterize the steady state and pre-steady state kinetics of each variant.Through these studies, I determined a differential regulatory role for the FAD-shieldingtryptophan in electron flux in MSR and CPR. The hydride transfer step in MSR is gatedby tryptophan displacement, as the free energy cost of disrupting the FAD-Trp interactionoutweighs the gain of nicotinamide placement. In contrast, electron flow through CPR is notgated by tryptophan displacement but rather by displacement of the oxidized nicotinamidering by the bulky residue following hydride transfer. Alterations at the coenzyme-bindingsites of MSR and CPR also elicited differential results. As anticipated, the removal oraddition of potential hydrogen bonding interactions with the coenzyme resulted in weakenedor strengthened coenzyme binding affinity, respectively. The accelerated flavin reduction503.2. Backgroundin CPR variants with weaker coenzyme binding affinity was attributed to faster releaseof oxidized coenzyme. In contrast, the accelerated flavin reduction in MSR variants withstronger coenzyme binding affinity was ascribed to improved coenzyme binding.3.2 BackgroundOf the diflavin oxidoreductase family members, MSR and CPR are the most structurallyrelated with 41 % amino acid sequence similarity. Despite the high level of sequence sim-ilarity, the human forms of MSR and CPR elicit different rates of electron transfer andcoenzyme binding affinity. Both 2’,5’-ADP and NADP+ bind tightly to CPR with a disso-ciation constant (Kd) of ∼50 nM.64 The nearly equal binding affinities indicates that thenicotinamide ring moiety does not contribute to the coenzyme binding energy of NADP+.In contrast, for MSR and also for FNR, the Kd values for NADP+ and 2’,5’-ADP are notequivalent with values of 37 µM and 2µM for MSR and 14 µM and 2µM for FNR.65,134Thus, the nicotinamide ring moiety contributes unfavourably (7 kJ/mol MSR and 5 kJ/molFNR) to the NADP+ binding energy of these enzymes.63,65 The Kd values also reveal thatthe binding of 2’,5’-ADP and NADP+ is 40- and 700-fold weaker in MSR compared to CPR.For the electron transfer kinetics, the differences between these two enzymes are evidentin their overall turnover rates for the non-physiological electron acceptor cytochrome c invitro. The reductase activity of CPR is >7-fold faster than that of MSR (in varied bufferconditions).65,132 Cytochrome c reduction by CPR was performed in 300 mM phosphatebuffer, while 50 mM Tris was used for MSR assays. This distinction is made since inorganicphosphate competitively inhibits NADPH at the 2’-phosphate-binding site and may influ-ence the measured rate of cytochrome c reduction at high concentrations.135 Analogousmutagenesis studies on the FAD-shielding tryptophan residue in MSR and CPR emphasizetheir differing kinetic properties. While in the W676H variant of CPR the rates of FADreduction and cytochrome c turnover are attenuated, the W697H variant of MSR elicits anenhancement of both kinetic events.132,136,137 In light of these differences, the specific role ofthe FAD-isoalloxazine ring-shielding tryptophan in MSR and CPR was further scrutinizedby generating more conservative substitutions. In addition to providing insight into thefunctional role of this active site tryptophan, these substitutions also enabled evaluationof the clear preference for the bulky tryptophan over smaller aromatic residues in diflavinoxidoreductases.Furthermore, the structural origin of the contrasting coenzyme binding affinities in MSRand CPR are investigated. Inspection of the available crystal structures for MSR and CPRshows that the coenzyme binds to the NADPH/FAD-binding domain primarily througha number of polar interactions and hydrophobic contacts with the 2’,5’-ADP moiety of513.2. BackgroundFigure 3.1: NADPH-coordinating residues at the coenzyme-binding pocket for CPR (PDB3QE2) Right and MSR (2QTZ) Left. NADPH is shown in grey with the standard atom-coloured stick model. Potential hydrogen bonding interactions are shown in black dashes.the substrate. Most of these residues are conserved between MSR and CPR, however twonotable differences are highlighted in Figure 3.1. At the pocket lined with polar residuesthat coordinate to the 2’-phosphate group of the substrate, a lysine residue (Lys602) ofCPR is clearly within hydrogen bonding distance to the 2’-phosphate group. In MSR, awater molecule instead hydrogen bonds with the 2’-phosphate as the equivalent residue isan alanine (Ala622). A lysine residue (Lys623) is present in the polar pocket, however itis not in a position for direct hydrogen bonding to the phosphate group. At the NADPHpyrophosphate-coordinating site, CPR has an additional interaction with the substratethrough an arginine (Arg298) as opposed to a lysine (Lys291) in MSR. Thus, amino acidsubstitutions to these designated active site residues are made in an effort to identify theorigin of the respective coenzyme binding affinities of MSR and CPR, and also to examinehow substrate binding influences intramolecular electron flux.The specific nature of the functional role of the penultimate C-terminal residue, Trp697and Trp676, in regulating coenzyme binding and electron flux in MSR and CPR was deter-mined through mutagenesis. The FAD isoalloxazine ring-shielding tryptophan was mutatedto a tyrosine and a phenylalanine: W697Y and W697F (MSR) and W676Y and W676F(CPR). These conservative residues were chosen to preserve residue aromaticity and pi-pi in-teractions with the FAD isoalloxazine ring, while testing the effect of residue side chain sizeand polarity. To study the potential benefit of a bidentate over a monodentate interactionwith the substrate pyrophosphate group, CPR Arg298 was mutated to an alanine and MSRLys291 was mutated to an arginine. In an effort to mirror the polar pocket active site ofCPR, the Ala622 of MSR was exchanged for a lysine. The active site of CPR was made moreMSR-like by a double mutation of Lys602Ala Val603Lys. Steady state kinetic propertiesof each variant were measured by spectrophotometric analysis of cytochrome c reduction.523.3. ResultsTable 3.1: Steady state kinetic parameters of MSR W697 and CPR W676 variants. Condi-tions: 1 mL reaction volume, 8 µM cytochrome c3+, 0.2-2 pmole enzyme, varied [NADPH]and [inhibitor] at 25 ◦C. Assay buffer for MSR is 50 mM Tris-HCl pH 7.5 and 50 mMKPi pH 7.5 for CPR. Assays were done in triplicate and fit to the Michaelis-Menten equa-tion. Inhibition data from four inhibitor concentrations were fit to a competitive inhibitionequation.a Data acquired from Wolthers et al.65k cat Km Ki k cat/KmEnzyme (s−1) (M × 10−6) (M × 10−6) (s−1M−1 × 10+6)NADP+ 2’,5’-ADPWild-typeMSRa 7.2 ± 0.1 2.4 ± 0.2 37 ± 3 1.4 ± 0.1 3.1 ± 0.3W697FMSR 17.0 ± 0.3 1.2 ± 0.1 3.0 ± 0.3 0.10 ± 0.01 15 ± 2W697YMSR 24.3 ± 0.4 1.2 ± 0.1 4.2 ± 0.5 0.7 ± 0.1 21 ± 2Wild-typeCPR 20.0 ± 0.2 0.7 ± 0.1 1.0 ± 0.1 0.6 ± 0.1 28 ± 2W676FCPR 3.8 ± 0.1 0.3 ± 0.1 0.4 ± 0.1 1.9 ± 0.3 12 ± 1W676YCPR 13.1 ± 0.2 0.6 ± 0.1 0.7 ± 0.1 1.2 ± 0.1 24 ± 2Product and dead-end inhibition studies with NADP+ and 2’,5’-ADP, respectively, wereperformed to assess coenzyme binding. The pre-steady state kinetics of flavin reductionwere measured by stopped-flow spectrophotometry. Lastly, to determine if the amino acidsubstitutions affected the redox potentials of the flavin centers, redox potentiometry wasperformed for the CPR W676F and W676Y variants.3.3 Results3.3.1 Steady state kinetic dataThe steady state kinetic parameters for cytochrome c3+ reduction by MSR and CPRvariants are summarized in Table 3.1. For the MSR variants, a 2.4- and 3.4-fold fasterk cat was measured for W697F and W697Y. They were also more catalytically efficient withk cat/Km values 4.8- and 6.7-fold greater than wild-type. In contrast, the same mutationsin CPR resulted in slower and less efficient catalytic turnover. The k cat decreased by 6.4-and 1.5-fold for W676F and W676Y, with a 2.4- and 1.2-fold reduced catalytic efficiency.The effect of amino acid substitutions on substrate binding was examined by steady stateinhibition studies with NADP+ and 2’,5’-ADP; the results are summarized in Table 3.1.Previous isothermal titration calorimetry (ITC) experiments with MSR revealed a Kd of34.5 µM for NADP+ which is in agreement with the inhibition constant of 37 µM determinedfrom steady state inhibition assays.65 For CPR, a Kd of 50 nM was determined from ITC533.3. Resultsexperiments in phosphate-free buffer.64 This Kd represents an over 700-fold tighter NADP+binding affinity in CPR compared to MSR. In the present studies, the NADP+ Ki in CPRis only ∼40-fold tighter. The discrepancy is accredited to the buffer conditions used. Aphosphate-free buffer is used for MSR assays, while a potassium phosphate (KPi) bufferis used for CPR assays. Binding studies in housefly CPR reveal that the free inorganicorthophosphate anion in the KPi buffer will competitively bind to the NADPH bindingsite. This competitive inhibition leads to an increase in the observed Ki for NADP+ and2’,5’-ADP and in the apparant Km for NADPH.135 Indeed, ITC experiments conducted inphosphate buffer demonstrated a 15-fold increase in the Ki for NADP+ binding in CPR.64Presumably, due to competition of the phosphate anion with the substrate at the 2’,5’-ADPbinding site. Here, the CPR assays are in 50 mM KPi buffer because the true Km is too lowfor the detection limits of a 1 cm path length cuvette. In the analyses to follow, I assumethat the orthophosphate anion binds to each of the CPR variants with the same affinityand, thus, perturbs the apparant Ki and Km values to the same extent.In MSR, a 12.3- and 8.8-fold tighter binding affinity for NADP+ was observed for W697Fand W697Y, respectively. Slightly tighter binding of the substrate analogue 2’,5’-ADP wasalso observed in W697Y (Ki = 0.1 µM) and W697F (Ki = 0.7 µM). A modest increase inNADP+ binding affinity was measured for CPR W676Y (Ki = 0.4 µM) and W676F (Ki =0.7 µM) compared to the native enzyme (Ki = 1.0 µM). In addition, a slight decrease isobserved for 2’,5’-ADP.3.3.2 Multiple wavelength pre-steady state kinetics of CPR variantsPre-steady state reduction of the CPR variants by saturating NADPH was monitored bystopped-flow spectrophotometry in an anaerobic environment. The spectral changes uponreduction with NADPH, shown in Figure 3.2, were resolved by SVD and fitted globallyto a best fit model. The observed rate constants are tabulated in Table 3.2. For wild-type CPR, reduction occurs in two kinetic phases. The first kinetic phase involves thereduction of oxidized enzyme to a two-electron reduced species (a → b). This relativelyfast phase has a rate constant of 19 s−1. The spectral changes show flavin bleaching at454 nm accompanied with the appearance of a broad absorbance band centered at 600 nm.These spectral features indicate that the hydride anion transfer from NADPH to FAD iskinetically coupled to interflavin electron transfer. The second kinetic phase entails furtherreduction of the enzyme population by another molecule of NADPH to spectral species cwith an observed rate constant of 2.7 s−1. The spectral changes associated with this kineticphase include additional bleaching of absorbance at 454 and 600 nm. Unlike wild type CPR,the reduction of W676 variants occurs in only a single kinetic phase. Conversion of oxidizedspectral species a to spectral species b shows loss of absorbance at 454 nm concomitant543.3. ResultsTable 3.2: Observed rate constants of the pre-steady state reduction of CPR W676 variants.Conditions: 22 µM wild-type CPR, 16 µM W676F, 36 µM W676Y, 10-fold excess NADPH,anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Wild-type CPR was best fit to a two-stepmodel with three spectral intermediates (a → b → c). W676 variant spectral data werebest fit to a single-step model with two spectral intermediates (a → b).CPR kobs1 kobs2(s−1) (s−1)Wild-type 19.00 ± 0.01 2.72 ± 0.01W676F 0.64 ± 0.01 –W676Y 6.12 ± 0.03 –with the emergence of a flat absorbance band in the longer wavelengths (550-700 nm). Therate constant for this kinetic step is 0.64 s−1 for W676F and 6.12 s−1 for W676Y.In addition, unique spectral features are observed for these variants. In wild-type CPR,the absorbance maxima is at 454 nm with a shoulder at 474 nm, Figure 3.2A,D. The ab-sorbance maxima at 454 nm is attributed to both flavins, while the shoulder at 474 nmoriginates from the FMN cofactor. For native CPR, the shoulder at 474 nm becomes lessprominent in the conversion of spectral species a to b, which indicates that the FMN isreduced over this kinetic phase. The spectra of W676Y is similar to that of wild-type CPR(Figure 3.2 E,F). On the other hand, the flavin absorbance maxima of W676F shifts to 474nm with a 456 nm shoulder. Upon NADPH reduction of both CPR variants, the absorbancemaxima decreases and shifts to reveal a pronounced 474 nm shoulder (more so for W676F).Given that the 474 nm shoulder is a diagnostic of the oxidized FMN flavin, the stopped-flow data suggests that the FMN is not significantly reduced to the semiquinone in the firstkinetic phase. Instead, this phase involves the reduction of FAD by a hydride ion trans-fer from NADPH, followed by formation of the FADH2-NADP+ charge-transfer complex.Evidence of the charge-transfer complex is provided by the broad, flat absorbance bandover the longer wavelengths (>550 nm). The lack of appreciable disemiquinone absorbancemaxima centered at 600 nm supports the formation of the charge-transfer complex ratherthan the disemiquinone species. Lastly, the greater amplitude change at 454 nm observedfor W676Y over W676F, indicates that W676Y is further shifted from the oxidized state tothe two-electron reduced form (E-FADH2-FMN).3.3.3 Primary kinetic isotope effect with (R)-[4-2H]-NADPH on CPRreductionRapid reduction of the CPR variants with NADPH and (R)-[4-2H] NADPH was followedat 454 and 600 nm by stopped-flow spectrophotometry. The absorbance traces collected553.3. ResultsFigure 3.2: Spectral changes upon NADPH reduction of CPR W676 variants monitored bymultiple-wavelength stopped-flow spectrophotometry. Conditions: 22 µM wild-type CPR,16 µM W676F, 36 µM W676Y, 10-fold excess NADPH, anaerobic 50 mM Tris-HCl pH 7.5at 25 ◦C. Time-dependent spectral changes over a 200 s time frame are shown for wild-typeCPR (A), W676F (B) and W676Y (C). Global analysis of SVD-resolved spectra generateddeconvoluted spectral intermediates for wild-type CPR (D), W676F (E) and W676Y (F).Wild-type CPR was best fit to a two-step model with three spectral intermediates (a → b→ c). W676 variant spectral data were best fit to a single-step model with two spectralintermediates (a → b).563.3. ResultsFigure 3.3: Anaerobic reduction of CPR variants by saturating substrate monitored at454 and 600 nm by single-wavelength stopped-flow spectrophotometry. Conditions: 10 µMenzyme, 100 µM NADPH, 100 µM (R)-[4-2H] NADPH, anaerobic 50 mM Tris-HCl pH 7.5at 25 ◦C. Absorbance traces at 454 nm over 10 s for W676F (A) and W676Y (B) reductionwith NADPH (black line) and (R)-[4-2H]-NADPH (grey line). Absorbance traces at 600nm over 10 s for NADPH-mediated reduction of W676F (C) and W676Y (D).at 454 and at 600 nm over 10 s are shown in Figure 3.3 and were best fit to a singleexponential equation. The equivalent rate constants for the monophasic spectral changesat 454 and 600 nm represent the same kinetic phase and are assigned to substrate binding,hydride transfer, and charge-transfer formation. The resulting observed rate constants aresummarized in Table 3.3. In W676F and W676Y, reduction with (R)-[4-2H]-NADPH yieldsa slight increase in the observed rate constant for flavin reduction, and the primary kineticisotope effect is lost with KIE of 0.90 for W676F and 0.94 for W676Y. In these variantsthere is more of a commitment to the reverse reaction (e.g. hydride transfer from FADH2to NADP+) than there is in native enzyme.3.3.4 Thermodynamic analysis of CPR variantsAnaerobic redox potentiometry was performed with each CPR variant to determine ifthe kinetic behavior observed is a consequence of altered flavin redox potentials. Spec-tral data were collected during gradual titration of the fully oxidized enzyme with 0.25573.3. ResultsTable 3.3: Observed rate constants and kinetic isotope effects for the pre-steady state reduc-tion of CPR W676 variants monitored at 454 and 600 nm by single-wavelength stopped-flowspectrophotometry over 10 s. Conditions: 10 µM enzyme, 100 µM NADPH, 100 µM (R)-[4-2H] NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Absorbance traces were averaged(3-5) then fit to a single exponential equation.Wavelength CPR Substrate kobs1 kobs2 isotope effect(s−1) (s−1) kobs1H/kobs1D454 nm Wild-type NADPH 15.8 ± 0.2 4.12 ± 0.06 1.08 ± 0.02NADPD 14.63 ± 0.07 3.29 ± 0.05W676F NADPH 0.66 ± 0.01 – 0.90 ± 0.03NADPD 0.73 ± 0.01 –W676Y NADPH 5.60 ± 0.04 – 0.94 ± 0.02NADPD 6.0 ± 0.1 –600 nm W676F NADPH 0.70 ± 0.01W676Y NADPH 7.5 ± 0.1µL aliquots of sodium dithionite until full reduction was reached. With each addition ofdithionite, the potential value and the absorbance spectrum were recorded. A plot of thesummed absorbance values from 590 to 605 nm (the semiquinone absorbance maxima) ver-sus the reduction potential (normalized to the standard hydrogen electrode) allowed for theextraction of midpoint potentials. For W676F, the midpoint potentials of the FMNox/sq(-99 mV) and FMNsq/hq (-215 mV) couples are comparable to those of wild-type enzyme.On the other hand, the FADox/sq (-252 mV) and FADsq/hq (-275 mV) couples are ∼30and 110 mV more electropositive. However, this shift in potential would be anticipated tofavour hydride transfer from NADPH, which is not reflected in the first rate constant. ForW676Y, the redox potentials of FADox/sq (-263 mV) and FADsq/hq (-260 mV) are also moreelectropositive than native enzyme and similar to those values determined for W676F albeitwith greater error. Surprisingly, the tyrosine substitution has a more pronounced effect onthe FMN redox couples with midpoint potentials of -166 mV and -198 mV for FMNox/sqand FMNsq/hq, respectively. It is unclear as to why W676Y elicits such an effect on theFMN redox center, however it may be due to an altered electronic environment surroundingW676Y since tyrosine is a polar residue and the isoalloxazine rings are highly polarizable.Although the midpoint potentials are more compressed in these variants compared to na-tive CPR, the electron flow is still thermodynamically favoured through NADPH to FADto FMN. Therefore, the dampened rates of flavin reduction observed are not a product ofless favourable thermodynamics.583.3. ResultsFigure 3.4: Redox potentiometric data of human CPR W676Y and W676F variants. Spec-tral properties of W676F (A) and W676Y (C) during redox titration. Spectra were recordedafter each addition of the reductant dithionite. Plots of the summed absorbance values be-tween 590 and 605 nm versus the normalized reduction potential for W676F (B) and W676Y(D). Fits of both data sets were made with the four-electron Nernst equation, as describedin Experimental Procedures, and results are summarized in Table 3.4.Table 3.4: Reduction potentials of the flavin couples of W676Y and W676F CPR variants.aData acquired from Munro et al.66reduction potential (mV) vs standard hydrogen electrodeFAD cofactor FMN cofactorEnzyme Ox/Sq Sq/Hq Ox/Hq Ox/Sq Sq/Hq Ox/HqWild-type CPRa -283 ± 5 -382 ± 8 -333 ± 7 -66 ± 8 -269 ± 10 -168 ± 9W676F -252 ± 10 -275 ± 8 -263 ± 9 -99 ± 1 -215 ± 5 -157 ± 5W676Y -263 ± 56 -260 ± 25 -262 ± 40 -166 ± 3 -198 ± 4 -182 ± 4593.3. ResultsTable 3.5: Observed rate constants of the pre-steady state reduction of MSR W697 variants.Conditions: 10 µM enzyme, 100 µM NADPH, anaerobic 50 mM Tris pH 7.5 at 25 ◦C. W697variant reduction was best fit to a three-step kinetic model.a Data acquired from Woltherset al.65.MSR kobs1 kobs2 kobs3(s−1) (s−1) (s−1)Wild typea 24.9 ± 0.1 0.18 ± 0.01 0.016 ± 0.003W697F 215.1 ± 0.1 1.15 ± 0.01 0.016 ± 0.001W697Y 214.5 ± 0.1 1.32 ± 0.01 0.015 ± 0.0013.3.5 Multiple wavelength pre-steady state kinetics of MSR variantsThe anaerobic NADPH-mediated reduction of MSR variants was followed by stopped-flow photodiode array detection. The spectral changes over the course of the reductivehalf-reaction are shown in Figure 3.5. Bleaching at the flavin absorbance maxima 454 nmis not accompanied by significant formation of an absorbance band at 600 nm. Inspectionof the single absorbance trace at 454 nm reveals three resolvable kinetic phases, shown inFigure 3.6. The first extracted rate constant (kobs1) is 215 s−1 and 213 s−1 for W697F andW697Y, respectively. Since this first kinetic phase occurs in less than 6 ms and photodiodearray detector scans every 1.5 ms, this initial rate constant could not be extracted from thePDA data. However, the time-resolved spectral data were globally fit to a three-step kineticmodel with the first rate constant fixed at the value determined from single wavelengthanalysis, refer to Section 3.2.6.Similar rate constants were extracted for kobs2 (1.2 - 1.3 s−1) and kobs3 (0.015 s−1) forboth MSR variants. The second rate constant is 5-fold higher than native MSR but similarto S698A and S698∆. Full reduction of the enzyme by a second equivalent of NADPH iscomparable in W697F, W697F and native enzyme. Neither variant shows accumulationof the semiquinone absorbance band at 600 nm from data acquired from single or multi-wavelength mode. The stopped-flow data reveal that reduction in the size of the aromaticside chain significantly increases the rate of FAD reduction in MSR, while suppressingelectron transfer from FAD to FMN.3.3.6 Primary kinetic isotope effect with (R)-[4-2H]-NADPH on MSRreductionReduction of MSR variants using NADPH and (R)-[4-2H] NADPH was monitored at454 nm using stopped-flow spectrophotometry. The absorbance traces were biphasic over 5s with a third slow phase observed over 200 s. The absorbance traces were fit from 0.001603.3. ResultsFigure 3.5: Anaerobic reduction of MSR W697 variants by saturating NADPH monitoredby multiple-wavelength stopped-flow spectrophotometry. Conditions: 10 µM enzyme, 100µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Full spectra of W697F (A) andW697Y (B) NADPH-mediated reduction over 200 s with ∼ 20 traces shown for clarity.Time-resolved spectral intermediates for W697F (C ) and W697Y (D) variants follow athree-step kinetic model with four discrete spectral species (a→b→c→d), associated rateconstants are summarized in Table 3.5.613.3. ResultsTable 3.6: Observed rate constants and kinetic isotope effects for the pre-steady statereduction of MSR W697Y and W697F variants monitored at 454 and 600 nm by single-wavelength stopped-flow spectrophotometry.Wavelength MSR Substrate kobs1 kobs2 isotope effect(s−1) (s−1) kobs1H/kobs1D454 nm W697F NADPH 215 ± 13 0.91 ± 0.12 2.02 ± 0.12NADPD 106 ± 10 1.36 ± 0.03W697Y NADPH 213 ± 12 1.11 ± 0.01 2.05 ± 0.14NADPD 104 ± 1 1.3 ± 0.2600 nm W697F NADPH 0.20 ± 0.01 –W697Y NADPH 0.207 ± 0.006 –to 5 s to a double exponential generating an initial rate constant kobs1 of 215 and 213 s−1and a kobs2 of 0.91 and 1.1 s−1 for W697F and W697Y, respectively. The kobs1 is 10-foldfaster than the corresponding rate in native MSR, indicating that the indole side chain oftryptophan greatly attenuates step(s) in the mechanism leading to the first hydride transferevent. When the same experiments were performed with deuterated substrate, the kobs1decreased to 108 and 106 s−1 for W697F and W697Y. As a result, these variants both elicita primary kinetic isotope effect of 2.0. Compared to wild-type MSR (KIE = 1.7), the KIEis slightly greater which indicates that hydride transfer becomes less rate-determining inthese variants.Unlike the W697H and W697S variants described in Chapter 2, W697F and, to a greaterdegree, W697Y exhibited a very small signal for the disemiquinone at 600 nm. The am-plitude change of the formation and decay of the signal is very minor but shows that theW697Y can partially preserve the role of the Trp697 in facilitating disemiquinone formation.3.3.7 Steady state kinetic data of the coenzyme-binding variants ofMSR and CPRMSR and CPR variant-catalyzed reduction of cytochrome c3+ was monitored by spec-trophotometric methods. The determined steady state kinetic parameters are summarizedin Table 3.7. In CPR, substituting the coenzyme pyrophosphate-coordinating Arg298 toan alanine only moderately affected the overall catalytic turnover, with a slower k cat of 12s−1 for cytochrome c reduction and a moderate decrease in catalytic efficiency comparedto native CPR. The binding affinity for NADP+, as measured by the inhibition constantKi, was 2-fold weaker for R298A. Likewise, a reduced k cat of 13 s−1 was determined for thedouble variant at the coenzyme 2’-phosphate-binding site, K602AV603K. A significant 39-fold reduction in the catalytic efficiency was also measured. K602AV603K elicited a much623.3. ResultsFigure 3.6: Anaerobic reduction of MSR W697 variants by saturating substrate monitoredwith single-wavelength stopped-flow spectrophotometry at 454 and 600 nm. Conditions:12.5 µM enzyme, 250 µM NADPH, 250 µM (R)-[4-2-H] NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. W697F (Panel A and C ) and W697Y (Panel B and D) were rapidlymixed with NADPH (black line) and (R)-[4-2-H] NADPH (grey line) and monitored over10 and 200 s at 454 nm and 600 nm, respectively.633.3. ResultsTable 3.7: Steady state kinetic parameters of coenzyme-binding residue variants of CPRand MSR. Conditions: 1 mL reaction volume, 8 µM cytochrome c3+, 2-12 pmole enzyme,varied [NADPH] and [inhibitor], 50 mM KPi pH 7.5(CPR), 50 mM Tris-HCl pH 7.5(MSR) at25 ◦C. Assays were done in triplicate and fit to the Michaelis-Menten equation. Inhibitiondata from four inhibitor concentrations were fit to a competitive inhibition equation. aAssays were conducted in 50 mM Tris pH 7.5 to compare this CPR variant with wild-typeMSR.b Data acquired from Wolthers et al.65Enzyme k cat Km Ki k cat/Km(s−1) (M × 10−6) (M × 10−6) (s−1M−1 × 10+6)CPRWT 20.0 ± 0.2 0.7 ± 0.1 1.0 ± 0.1 28.2 ± 1.8R298A 12.4 ± 0.1 0.64 ± 0.02 2.0 ± 0.1 19.5 ± 0.9K602AV603Ka 13.1 ± 0.1 18.1 ± 1.1 46.4 ± 6.5 0.72 ± 0.05MSRWT b 7.23 ± 0.14 2.37 ± 0.17 36.89 ± 2.70 3.05 ± 0.28K291R 3.73 ± 0.08 2.92 ± 0.29 16.08 ± 1.05 1.28 ± 0.15A622K 1.90 ± 0.02 0.84 ± 0.07 13.07 ± 1.88 2.26 ± 0.20greater effect on coenzyme binding, with a 46-fold weaker binding affinity for NADP+ (Ki= 46 µM).For MSR, exchanging the coenzyme pyrophosphate-coordinating Lys291 for an arginine,produced no notable effect on the catalytic turnover with only an approximate 2-fold de-crease in catalytic efficiency. However, the NADP+ binding affinity was 2.3-fold strongerfor K291R. Mutation at the 2’-phosphate-coordinating polar pocket in the A622K variantelicited a moderately slower reduction of cytochrome c with slightly reduced catalytic effi-ciency. As with K291R, A622K exhibited a stronger binding affinity for NADP+ comparedto native MSR, by 2.8-fold.3.3.8 Multiple and single wavelength pre-steady state kinetics ofcoenzyme-binding variants of MSR and CPRThe pre-steady state kinetics of flavin reduction in the coenzyme-binding variants ofMSR and CPR were measured by stopped-flow spectrophotometry. The spectral changesoccurring over the reductive half-reaction were resolved and fit to the appropriate kineticmodel. CPR variant spectra are featured in Figure 3.7 and the observed rate constants arein Table 3.8. Both R298A and K602AV603K elicited biphasic reduction and were best fit toa two-step kinetic model with three discrete spectral species, as described for wild-type CPRin Section 3.3.2. The observed rate constants for the first (kobs1) and second (kobs2) kineticphases are 14 s−1 and 3 s−1 for R298A, which are comparable to those determined for wild-type CPR reduction. See Table 3.9 and Figure 3.8 for rate constants derived from double643.3. ResultsFigure 3.7: Spectral changes upon NADPH reduction of CPR coenzyme-binding variantsmonitored by multi-wavelength stopped-flow spectrophotometry. Conditions: 10 µM en-zyme, 100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Full spectra of R298A(A) and K602AV603K (B) NADPH-mediated reduction over 100 s with ∼ 20 traces shownfor clarity. Time-resolved spectral intermediates for R298A (C ) and K602AV603K (D)variants follow a two-step kinetic model with three discrete spectral species (a→b→c), as-sociated rate constants are summarized in Table 3.8.exponential fits of single wavelength absorbance data at 454 nm. These are in agreementwith the multiple wavelength data. No significant differences were observed in the singleand multiple wavelength spectra of R298A. In contrast, K602AV603K elicited observed rateconstants that were both accelerated with a kobs1 of 57 s−1 and a kobs2 of 17 s−1. Theserates are moderately faster than those measured from single wavelength absorbance traces.Spectral data from single wavelength traces show considerable differences in the amplitudechange at 454 nm and particularly at 600 nm. The small amplitude change at 600 nmsuggests that the disemiquinone does not accumulate in this variant to the same extentas in wild-type. Furthermore, variant reduction by (R)-[4-2H] NADPH yielded a kobs1 of34.76 ± 0.18 s−1, resulting in a KIE of 1.2 that suggests the hydride transfer step is lessrate-determining.653.3. ResultsTable 3.8: Observed rate constants for flavin reduction in CPR and MSR coenzyme-bindingresidue variants acquired from multi-wavelength stopped-flow data. Conditions: 10 µMenzyme, 100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. CPR variants werebest fit to a biphasic kinetic model while MSR variants were best fit to a triphasic kineticmodel.a Data acquired from Wolthers et al.65Enzyme kobs1 kobs2 kobs3(s−1) (s−1) (s−1)Wild-typeCPR 19.00 ± 0.01 2.72 ± 0.01 –R298A 13.58 ± 0.03 3.37 ± 0.01 –K602AV603K 57.34 ± 0.01 17.44 ± 0.06 –Wild-typeMSRa 24.9 ± 0.1 0.18 ± 0.01 0.016 ± 0.003K291R 28.64 ± 0.01 3.12 ± 0.01 0.033 ± 0.001A622K 35.00 ± 0.01 0.372 ± 0.002 0.017 ± 0.001Figure 3.8: Anaerobic reduction of CPR coenzyme-binding variants by saturating NADPHmonitored at 454 and 600 nm with single wavelength stopped-flow spectrophotometry. Con-ditions: 10 µM enzyme, 100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. PanelA shows the absorbance traces over 1 s at 454 nm of native CPR (black line), R298A (grayline) and K602AV603K (light gray line). Panel B shows the absorbance traces over 10 s at600 nm of the CPR variants in the same colours detailed for Panel A.663.3. ResultsTable 3.9: Observed rate constants for flavin reduction in CPR and MSR coenzyme-bindingresidue variants acquired from single wavelength 454 nm stopped-flow data. Conditions: 10µM enzyme, 100 µM NADPH, anaerobic 50 mM Tris pH 7.5 at 25 ◦C. CPR variants werebest fit to a biphasic kinetic model while MSR variants were best fit to a triphasic kineticmodel.a Data acquired from Wolthers et al.65Enzyme kobs1 kobs2(s−1) (s−1)R298A 12.52 ± 0.05 1.95 ± 0.09K602AV603K 40.9 ± 0.4 10.9 ± 0.4K291R 25 ± 2 2.00 ± 0.01A622K 26.9 ± 0.4 1.76 ± 0.14Figure 3.9: Spectral changes upon NADPH reduction of MSR coenzyme-binding variantsmonitored by multiple wavelength stopped-flow spectrophotometry. Conditions: 10 µMenzyme, 100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Full spectra ofK291R (A) and A622K (B) NADPH-mediated reduction over 150 s with ∼ 20 traces shownfor clarity. Time-resolved spectral intermediates for K291R (C ) and A622K (D) variantsfollow a three-step kinetic model with four discrete spectral species (a→b→c→d), associatedrate constants are summarized in Table 3.8.673.4. DiscussionFigure 3.10: Anaerobic reduction of MSR coenzyme-binding variants by saturating NADPHmonitored at 454 and 600 nm with single wavelength stopped-flow spectrophotometry. Con-ditions: 10 µM enzyme, 100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. PanelA shows the absorbance traces over 10 s at 454 nm of native MSR (black line), K291R (grayline) and A622K (light gray line). Panel B shows the absorbance traces over 200 s at 600nm of the MSR variants in the same colours detailed for Panel A.For NADPH-mediated reduction of MSR variants, the spectral changes occurring overthe course of the reaction are similar to native MSR. Figure 3.9 shows the progressiveloss of absorbance at the flavin maxima at 454 nm along with the appearance and partialdisappearance of the semiquinone absorbance signal centered at 600 nm. Global analysisof the photodiode array-derived data revealed a kobs1 of 29 s−1 for K291R, which is onlyslightly improved from wild-type, and a kobs1 of 35 s−1 for A622K. Similar rate constantswere extracted for kobs2 (3.7 and 0.1 s−1) and kobs3 (0.03 and 0.02 s−1) of K291R andA622K, respectively. Rate constants determined from global analysis are consistent withthose obtained from double-exponential fits of 454 nm absorbance traces (Table 3.9). Figure3.10 shows the absorbance changes occurring over the course of the flavin reduction inK291R and A622K. At both 454 and 600 nm, K291R exhibits similar absorbance changesas wild-type MSR. However, for A622K at 454 nm, a faster kinetic phase is clearly observedfollowed by a slower multiphasic reduction from 0.1 - 10 s. Furthermore, a greater amplitudechange at 600 nm is observed for A622K, indicating that a greater fraction of the enzymepopulation is forming the disemiquinone compared to wild-type.3.4 Discussion3.4.1 Hydride transfer is gated by Trp697 in MSR but not in CPRSubstitution of the FAD isoalloxazine ring-stacked Trp residue had opposite effects onMSR and CPR catalysis. In MSR, mutation of Trp697 to a smaller aromatic residue re-683.4. Discussionsulted in variants with an enhanced rate of cytochrome c reduction with greater catalyticefficiency. These catalytic improvements originate from the 10-fold more rapid initial rateconstant, kobs1, for flavin reduction measured for both variants. In MSR, kobs1 is assignedto the hydride transfer step and also all the physical preceding steps including coenzymebinding and displacement of Trp697. Since hydride transfer is expected to be rapid oncethe nicotinamide ring is correctly positioned over the FAD isoalloxazine ring, the rate en-hancement is likely due to a reduction in the energy barriers for the physical steps that leadto productive NADPH binding. As Trp697 forms extensive pi-pi stacking interactions withthe FAD isoalloxazine ring, disruption of these dispersion forces presents an energetic bar-rier for productive nicotinamide ring placement and subsequent hydride transfer136. Thus,substitution of Trp697 with a smaller aromatic is expected to reduce the energy barrier.Moreover, the slight increase in the kinetic isotope effect of W697Y and W697F for kobs1suggests that the isotopically insensitive steps (e.g. structural rearrangement of the aro-matic residue) are less rate-determining in these variants and the intrinsic isotope effectassociated with breakage of a C-H bond is partially unmasked. As an additional note, thepolarity of the residue does not affect the rate of displacement since both variants have,within error, similar values for kobs1 and KIE.Swapping Trp676 for a tyrosine or phenylalanine in CPR did not result in as dramaticeffects as in MSR. Instead of enhancing flavin reduction, the same variants elicited slightlyslower rates of flavin reduction. The size of the aromatic residue does not influence pro-ductive substrate binding in the same way as in MSR, as the CPR variants only elicitedmoderately tighter NADP+ binding affinity. Subtle conformational changes at the activesite upon coenzyme binding may account for these results.As previously proposed for CPR, two separate conformational rearrangements of theactive site tryptophan are envisioned to take place in MSR prior to hydride transfer.57In substrate-free and NADP+-bound forms of MSR, the indole ring is orientated over there-side of the FAD isoalloxazine ring to maximize pi orbital overlap with the FAD. In CPR,however, substrate binding may induce structural rearrangements in the active site that leadto a flip and rotation of the indole where only partial pi-pi interaction is maintained. Thismay be the stable position of Trp676, as it is consistently captured in all wild-type crystalstructures of substrate-bound CPR. This second conformation may weaken the residue-FADinteraction such that the energetic cost of tryptophan displacement is overcome by the en-ergetic gain of nicotinamide ring placement. Since CPR appears to adopt this conformationmore readily than MSR, it may be the origin of the differential coenzyme binding propertiesbetween these two enzymes. Thus, the weaker interaction of the smaller aromatic residueswith the FAD isoalloxazine in MSR would be expected to lower the energetic cost of therequired conformational changes and may account for the dramatic increase in kobs1 for693.4. Discussionthe MSR variants. For the CPR variants, a similar shift to the second conformation as innative CPR is expected. Thus, the weakened interaction between the isoalloxazine ring andresidue is potentially very similar among the variants and wild-type CPR, thereby havinglittle effect on the overall rate of flavin reduction.3.4.2 Interflavin ET influenced by residue polarity in MSR and CPRMSR and CPR adopt closed and open conformational states over the course of catalysis.In the closed state, the interflavin distance is shortened to facilitate electron transfer. Ac-cording to Dutton’s electron transfer rate ruler and the 4 A˚ interflavin distance (obtainedfrom CPR crystal structure), the internal electron transfer rate should be approximately 3x 1010 s−1.138 However, the measured rates in wild-type MSR and CPR are substantiallylower.133,138 In MSR, although the rate of electron transfer from FAD to FMN is rela-tively slow, we have previously shown that the active site Trp697 accelerates this step.136By exchanging Trp697 for histidine or serine, the transient absorbance signal at 600 nmis lost indicating a lack of any detectable semiquinone buildup. However, interflavin elec-tron transfer does still take place (at an attenuated rate) since these variants achieve fullfour-electron reduction by two NADPH molecules and reduce cytochrome c, which receiveselectrons directly from FMN. In the W697Y variant there remained an absorbance signalat 600 nm albeit with a smaller amplitude than in wild type MSR. This indicates that,to a certain degree, tyrosine can act as trytophan in accelerating FAD to FMN electrontransfer. The modest increase in internal electron transfer in W697Y may be the origin ofthe slightly faster cytochrome c turnover compared to W697F. This may also be the case forthe CPR variants since the W676Y variant has a greater rate of flavin reduction with a largeamplitude change compared to W676F as well as faster cytochrome c turnover. The precisemechanism of how Trp, and to a lesser extent Tyr, enhance internal electron transfer is stillunclear. However, it may arise from the role of the bulky tryptophan residue in displacingthe oxidized nicotinamide ring which is discussed further in the following section.3.4.3 Trp676 displaces the oxidized nicotinamide ring and as suchcontrols electron flowThe hydride transfer and interflavin electron transfer steps are tightly coupled in onekinetic phase in CPR. Let us consider the reduction potentials of the CPR flavins. Basedoff of the -320 mV midpoint potential of NADP(H) and -333 mV of FADox/hq, the transferof a hydride ion from NADPH to the FAD isoalloxazine ring is not thermodynamicallyfavoured. However, the FMNox/sq couple is much more electropositive (-66 mV) and pro-vides the thermodynamic driving force for the forward electron flow through the enzyme.703.4. DiscussionBy exchanging Trp676 for a smaller aromatic the transfer of an electron from FADH2 toFMN is hindered, as evident by the lack of a disemiquinone absorbance signal at 600 nmthat typically accompanies the loss of absorbance at 454 nm. Instead, the first phase offlavin reduction by NADPH results in the formation of a broad charge-transfer band thatis assigned to a distribution of the NADPH-FAD-FMN and NADP+-FADH2-FMN com-plexes. Since the electrons remain on the FAD cofactor and the nicotinamide ring remainsassociated in the active site, there is a shift in favour of the reverse reaction (hydride trans-fer back to NADP+). Moreover, the charge-transfer complex is not observed in wild-typeCPR, indicating that NADP+ dissociation is faster than in the variants. Thus, followinghydride transfer to FAD, the bulky tryptophan is required for displacement of the oxidizednicotinamide ring.Moreover, early stopped-flow studies on rabbit CPR first hypothesized that NADP+release was rate-limiting in the rapid flavin reduction kinetics.68 Through mutagenesis andX-ray crystallographic methods, Hubbard et al. identified residues that form a hydrogenbond network around the FAD isoalloxazine ring, and the involvement of these residues inproton release and FAD semiquinone stabilization. They proposed that NADP+ release wasrequired prior to interflavin electron transfer.61 Therefore, the impeded rate of flavin reduc-tion observed in the W676 variants may be a consequence of a NADP+ release becomingmore rate-determining.3.4.4 Differential influence of coenzyme binding affinity on CPR andMSR flavin reduction kineticsAdditional electrostatic interactions with the coenzyme improves substrate binding affin-ity, however tighter coenzyme binding has differential effects on flavin reduction in CPR andMSR. In CPR, removal of the Arg298 bidentate interaction with the coenzyme pyrophos-phate group did not generate dramatic changes to CPR catalytic turnover. As expected,binding of NADP+ was weakened by this mutation but only by 2-fold. The relatively smallchange in Ki may be due to the presence of other stabilizing active site residues, includ-ing a second arginine (Arg567) that coordinates to the pyrophosphate group (also presentin MSR - Arg581). A much greater effect on coenzyme binding and flavin reduction wasmeasured for K602AV603K. Removing the direct hydrogen bonding interaction of Lys602with the 2’-phosphate group of NADPH resulted in a 46-fold weaker binding affinity forNADP+. Since the binding energy for the coenzyme is largely derived from hydrogen bondinteractions between the 2’,5’-ADP moiety and polar residues at the coenzyme binding cleft,the significantly weaker binding affinity for NADP+ in K602AV603K was anticipated.64Weakening the interaction between the enzyme and substrate in K602AV603K produceda 3- and 6-fold enhancement in the kobs1 and kobs2 of flavin reduction, respectively. These713.5. Experimental Proceduresaccelerated rate constants are attributed to faster dissociation of the oxidized coenzyme asNADP+ release becomes less rate-determining. Indeed, a slight increase in the KIE to 1.2from that of wild-type (KIE = 1.1), indicates that hydride transfer becomes slightly morerate-determining in K602AV603K.As anticipated for MSR, coenzyme binding is improved by additional electrostatic inter-actions at the coenzyme-binding site. Replacing a monodentate with a bidentate interactionwith the coenzyme pyrophosphate group, the K291R variant elicited tighter binding affinityfor NADP+ with modestly improved flavin reduction. Furthermore, an additional hydrogenbond interaction with the coenzyme 2’-phosphate group also increased the enzyme’s affinityfor NADP+, as evident by the 2.8-fold lower Ki for A622K. This variant also elicited fasterrates of flavin reduction with a kobs1 of 35 s−1. As kobs1 encompasses hydride transfer aswell as the preceding physical steps such as coenzyme binding, the accelerated kobs1 forthese MSR variants is attributed to improved coenzyme binding affinity.Despite the accelerated rates of flavin reduction in the K602AV603K and A622K vari-ants, the overall k cat of cytochrome c reduction was modestly slower than the wild-typecounterparts. Although the origin of this effect is not immediately clear, it may be dueto alterations in the active residues that confer subtle structural changes upon coenzymebinding. Previous studies with CPR have shown that binding of the coenzyme and the2’,5’-ADP moiety alone enhance the interflavin electron transfer rate in CPR.73,75 Thisrate enhancement was attributed to a shift in the conformational equilibria to a closedstate that brings the flavin cofactors together to facilitate interflavin electron communica-tion. For CPR, and by extension for MSR, it is possible that the mutations introducedat the 2-phosphate-binding site may influence the residues involved in this conformationalequilibrium. These variants may adopt the closed conformation more readily than the openconformation, which would disfavour electron transfer from FMNH2 to the external electronacceptor.3.5 Experimental Procedures3.5.1 MaterialsReagents NADPH, NADH, NADP+, 2’,5’-ADP, and cytochrome c3+, ethanol-d6 and al-cohol dehydrogenase were obtained from Sigma Aldrich (Oakville, ON, Canada). Pfu TurboDNA polymerase, Taq DNA polymerase and Xl1 Blue cell lines were from Agilent Tech-nologies (Mississauga, ON, Canada). Rosetta(DE3)pLysS and BL21(DE3)pLysS competentcells and yeast alcohol dehydrogenase were purchased from EMD Biosciences. Protein pu-rification supplies, Resource Q column and glutathionine sepharose 4B resin, were obtainedfrom GE Biosciences. All other chemical reagents were purchased from Fisher.723.5. Experimental Procedures3.5.2 (R)-[4-2H]-NADPH synthesis and purification(R)-[4-2H]-NADPH was synthesized enzymatically using a procedure adapted from Violaand colleagues.139 In a 30 mL beaker wrapped in foil, a 20 mL reaction mixture was madeof 20 mM TAPS buffer pH 9, 330 mg NADP+, 33 units of alcohol dehydrogenase, 100 unitsaldehyde dehydrogenase and 0.343 mL of ethanol d6. At room temperature, the reaction isstirred and the pH was monitored continuously to maintain a pH of 9. Throughout the day,small samples were analyzed at 340 nm using a Lambda 25 UV/Vis Spectrometer (PerkinElmer). At a concentration of 5.37 mM, 100 µL of chloroform was added and the solutionwas mixed thoroughly. It was then spun at 27,170 × g for 30 minutes. The supernatantwas lyophilized overnight.The first purification method was by ethanol precipitation. In the morning, the NADPDpreparation was resuspended in 2 mL of 20 mM TAPS pH 9 then 24 mL of −20 ◦C ethanolwas added. The solution was spun at 27 170 × g for 30 minutes. The supernatant wasdiscarded and the resuspension was repeated. After a second round of centrifugation, thesupernatant was discarded and the pellet was resuspended in 2 mL 20 mM TAPS thentransferred to a round bottom flask. Contents were lyophilized overnight.A further purification step was performed with a Q-Sepharose column. The columnwas equilibrated with 10 mM ammonium hydrogen carbonate pH 9 (buffer A) and a 2 mLsolution of the previously dried NADPD sample was applied to the column. The columnwas washed with two column volumes of buffer A then a gradient was applied with 400 mMammonium hydrogen carbonate pH 9 over 3 column volumes (buffer B). Two additionalcolumn volumes were run at 100 % buffer B. The resulting peaks were collected and analyzedspectrophotoscopically to determine the 260/340 nm ratio. Fractions with a ratio of 2.5 orless were pooled together and lyophilized until completely dry. Samples were wrapped infoil and stored at −80 ◦C.3.5.3 Generation and expression of CPR and MSR tryptophan variantsThe CPR W676F and W676Y variants and the MSR W697F and W697Y variants weregenerated from soluble (transmembrane domain cleaved) wild-type CPR plasmid pET-15band wild-type MSR plasmid pGEX-4T1 using the QuikChange Site-Directed MutagenesisKit (Agilent Technologies). Oligonucleotide primers were designed based on the publishedsequence (Accession number: NM000941 CPR and AF121214 MSR) and purchased fromIntegrated DNA Technologies (Coralville, Iowa, USA). The primers are tabulated in Table3.10. Successful mutagenesis was confirmed by the NAPS DNA Sequencing Laboratory ofthe University of British Columbia (Vancouver, Canada). Mutant plasmids were then trans-formed into the Escherichia coli strain Rosetta2(DE3)pLysS for MSR and BL21(DE3)pLysS733.5. Experimental ProceduresTable 3.10: The forward (F) and reverse (R) oligonucleotide primers designed for eachspecified MSR and CPR variant. Mutation is in bold.Enzyme Variant Oligonucleotide SequenceMSRW697F F 5’ CTTCAGGATATTTTTTCATAGCGGCCGCATCG 3’R 5’ CGATGCGGCCGCTATGAAAAAATATCCTGAAG 3’W697Y F 5’ CTTCAGGATATTTACTCATAGCGGCCGCATCG 3’R 5’ CGATGCGGCCGCTATGAGTAAATATCCTGAAG 3’K291R F 5’ CGAATGATGCCATAAGAACCACTCTGCTG 3’R 5’ CAGCAGAGTGGTTCTTATGGCATCATTCG 3’A622K F 5’ GAGGAGGAAGCCCCAAAAAAGTATGTACAAGAC 3’R 5’ GTCTTGTACATACTTTTTTGGGGCTTCCTCCTC 3’CPRW676F F 5’ GCCTACTCCTGGACGTGTTTAGCTAGGATCCGAATTCGG 3’R 5’ CCGAATTCGGATCCTAGCTAAACACGTCCAGGAGTAGC 3’W676Y F 5’ GCTACTCCTGGACGTGTACAGCTAGGATCCGAATTCGG 3’R 5’ CCGAATTCGGATCCTAGCTGTACACGTCCAGGAGTAGC 3’R298K F 5’ CCAGGGAACCGAGCGCCCACCTCATGCACCTGG 3’R 5’ CCAGGTGCATGAGGTGGGCGCTCGGTTCCCTGG 3’K602AV603K F 5’ CGGGAGCAGTCCCACGCTAAGTACGTCCAGCAC 3’R 5’ GTGCTGGACGTACTTAGCGTGGGACTGCTCCCG 3’for CPR using the heat shock method. The MSR recombinant GST-fusion proteins wereexpressed and harvested as previously described in Section 2.5.2. The CPR recombinantHis-tagged proteins were grown in 200 mL cultures of Luria-Bertani medium, containing100 µg mL−1 of ampicillin and 34 µg mL−1 chloramphenicol, overnight at 37 ◦C and 225rpm. 0.5 L cultures of Terrific Broth containing 100 µg mL−1 ampicillin and 34 µg mL−1chloramphenicol were inoculated with 10 mL of the 200 mL cultures and grown at 30 ◦C toan Abs600 of 0.8. Cultures were then induced with 0.1 mM IPTG and the temperature waslowered to 20 ◦C to grow overnight. In the morning, cells were harvested at 6,360 × g for10 min and stored at −80 ◦C.3.5.4 Purification of MSR variantsPurification of MSR variants followed the procedure previously described in Section2.5.4.3.5.5 Purification of CPR variantsAll purification steps were performed on ice or at 4 ◦C. Cells overexpressing the recom-binant His-tagged enzyme were resuspended in 300 mL of 50 mM Tris/HCl pH 7.5 withprotease inhibitors benzamidine (2 mM) and PMSF (1 mM). The resuspension was soni-743.5. Experimental Procedurescated with the Sonicator S-4000 (Misonix Inc.) at 22% power amplitude with alternating8 s pulses and 1 min pauses for a total of 45 min. Cell lysate was then centrifuged at 39120 × g for 45 minutes to separate the soluble protein from cellular debris. 0.5 M NaCland 20 mM imidazole was added to the collected supernatant. The crude extract was thenloaded onto a HisTrap FF crude 5 mL affinity column (GE Healthcare) equilibrated with 10column volumes of 50 mM Tris/HCl pH 7.5, 0.5 M NaCl and 20 mM imidazole. Followedby a 5 column volume wash, the column was attached to the AKTApurifier system (GEHealthcare). Buffer A: 50 mM Tris/HCl pH 7.5 and 0.5 M NaCl; Buffer B: 50 mM Tris/HClpH 7.5, 0.5 M NaCl and 0.3 M imidazole. Nonspecifically bound protein was eluted with 10% buffer B for 2 column volumes. His-tagged enzyme was eluted by a linear gradient from10% to 100% buffer B over 8 column volumes. Fractions were pooled together and dialyzedovernight in 4 L 50 mM Tris/HCl pH 7.5, 1 mM EDTA and 1.7 mL β-mercaptoethanolat 4 ◦C to remove excess salt. Dialysate is concentrated down to 20 mL, loaded onto theAKTApurifier system and applied to an equilibrated 55 mL Q Sepharose anion exchangecolumn. Buffer A: 50 mM Tris/HCl pH 7.5; Buffer B: 50 mM Tris/HCl pH 7.5 and 0.5 MNaCl. The enzyme was eluted with a linear gradiant from 0 % to 100 % buffer B over eightcolumn volumes at a flow rate of 2 mL/min. The desired fractions were collected based onthe absorbance readings at 280 nm and the yellow colour that is characteristic of oxidizedCPR. The final concentration of purified CPR (∆ε = 21.6 mM−1cm−1) was determined bythe absorption at 454 nm and the Beer-Lambert Law, Equation 3.1.Abs = εlc (3.1)where c is the concentration of sample, Abs is the absorbance at 454 nm, ε is the molarextinction coefficient of CPR, and l is the light path length (1 cm).3.5.6 Steady state turnover analysisSteady state turnover assays for MSR variants were conducted as previously describedin Section 2.5.5. CPR variant assays followed the same protocol but were performed in 50mM KPi pH 7.5 buffer.3.5.7 Pre-steady state kinetic analysisPre-steady state kinetic assays for MSR and CPR variants were conducted as previouslydescribed in Section 2.5.6.753.5. Experimental Procedures3.5.8 Redox potentiometryAll potentiometric titrations were performed in an anaerobic nitrogen environment atroom temperature in 50 mM Hepes/KOH pH 7.0 buffer. Through extensive bubbling withnitrogen, the buffer was made anaerobic and left to equilibrate in the glove box for over 16hours. Enzyme samples W676Y and W676F were fully oxidized with potassium ferricyanide,brought into the glove box then gel-filtered over a 50 mM Hepes/KOH pH 7.0 equilibrated 10mL size-exclusion column (Bio-Rad Econo-Pac 10 DG column, Mississauga, ON, Canada).Small fractions of the eluted protein were removed from the glove box for spectrophotometricdetermination of the sample concentration. In the anaerobic environment, the enzymewas diluted to 14-28 µM in a total volume of 3 mL. Redox mediators benzyl viologen(1 µM), methyl viologen (0.2 µM), 2-hydroxy-1,4-napththoquinone (5 µM) and phenazinemethosulfate (2 µM) were added to the glass cuvette with the enzyme solution. A speciallydesigned cuvette holder held the fiber optic cables of the stopped-flow to allow for sampleillumination by the xenon lamp and the spectra were recorded by the photodiode arraydetector. Absorbance spectra (280-703 nm) were recorded on a SF-61DX2 Stopped-flowapparatus from TgK Scientific and the electrochemical potential was monitored using aMettler Toledo FiveEasy voltmeter coupled to a platinum-Ag/AgCl electrode. The enzymesolutions were titrated electrochemically with small aliquot volumes of sodium dithionite asdescribed by Dutton et al.140 After each addition, the sample was mixed thoroughly, allowedto equilibrate, and the spectrum was recorded when the potential stabilized. This step wasrepeated until the enzyme was fully reduced. The observed potentials were normalized tothe standard hydrogen electrode by adding 197 mV. The spectral data were adjusted to thesame absorbance baseline at 703 nm to account for drift observed during the redox titration.Data manipulation and analysis was performed using Origin 8.5 software (OriginLabCo.). Absorbance values from 590 to 605 nm, the semiquinone absorbance maxima, weresummed and plotted against the normalized potential to determine the midpoint potentials.Data were fitted to Equation 3.1, which represents a sum of two two-electron redox processesderived by the Beer-Lambert Law and the Nernst equation.A =a10(E−E′1)/59 + b + c10(E′2−E)/591 + 10(E−E′1)/59 + 10(E′2−E)/59)+d10(E−E′3)/59 + e + f10(E′4−E)/591 + 10(E−E′3)/59 + 10(E′4−E)/59)(3.2)In this equation, E is the observed potential, E1’ and E2’ are the oxidized/semiquinoneand semiquinone/hydroquinone midpoint potentials of one flavin. E3’ and E4’ are the corre-sponding midpoint potentials for the second flavin. A is the total absorbance, a, b and c arethe component absorbance values contributed by one flavin in the oxidized, semiquinone,and reduced states, respectively. The d, e and f are the corresponding absorbance com-763.5. Experimental Proceduresponents for the second flavin. The absorbance contribution of oxidized and reduced formsof FAD and FMN are assumed to be equal (a = d and c = f ). The initial values for En’(where n = 1, 2, 3, 4) were based on the previously determined midpoint potentials forwild-type CPR.66 With only one variable fixed and all others floating, data fitting resultedin decent flavin absorbance values (a-f ). Several iterations of data fitting with varied com-binations of absorbance values fixed (a and d, b and e, and c and f ) with the potentialsfloating generated reasonable estimates of the individual midpoint potentials. This fittingwas followed by a final iteration where the potential values and absorbance values a, c, dand f were allowed to vary. To prevent overparameterization, b and e were fixed in thefinal iteration.77Chapter 4Role of histidine residue at theFAD active site in regulatingintramolecular electron flow inCPR and MSR4.1 SummaryIn this chapter the functional role of a proximal FAD histidine residue in interflavinelectron transfer is investigated in human CPR and MSR. Hydride transfer and interflavinelectron transfer in CPR are tightly coupled in one kinetic step, while the two catalyticevents are two kinetically distinct phases in MSR. The FAD proximal catalytic triad is well-conserved in CPR and MSR with an established role in stabilizing steps of the catalyticmechanisms. In CPR, Asp674 of the triad forms a hydrogen bond with His322. Theequivalent residue in MSR is Ala312. Consequently, the potential role of this residue inregulating electron flux in CPR and MSR is evaluated by generating reciprocal substitutions:H322A and A312H. The steady state and pre-steady state kinetic properties of these variantswere determined through aerobic and anaerobic spectrophotometric techniques. Throughthese studies, the hydride and interflavin electron transfer steps in MSR were successfullycoupled in one kinetic phase. In CPR, I proposed that His322 weakens coenzyme bindingaffinity by competing with the nicotinamide ring for interaction with Asp674. By weakeningthe coenzyme-enzyme interaction, coenzyme release is promoted which, in turn, accelerateselectron flow through CPR.4.2 BackgroundMSR and CPR are envisioned to switch from a closed to an extended conformationduring the transmission of electrons from NADPH to external electron accepting pro-teins.73,74,113,141 The compact, closed conformation brings FAD and FMN flavins together,such that they are poised for rapid interflavin electron transfer. In this conformation, the784.2. BackgroundFMN cofactor is buried and cannot interact with and transfer an electron to the terminalelectron acceptor. To accomplish this half of the reaction, the FMN domain undergoeslarge-scale conformational motion to adopt an extended state that leads to solvent expo-sure of the FMN.73,74,113 Several studies using different methodologies including NMR andsmall-angle X-ray crystallography in CPR have provided strong evidence that these con-formational transitions take place during catalysis.73–76,142 The 12-amino acid hinge thattethers the FMN domain to the connecting domain of CPR has been shown to be critical tothe conformational dynamics.57 Interestingly, the corresponding hinge in MSR is an extra82 amino acids longer suggesting an increased mobility of the FMN domain.Spectroscopic studies of the rapid, single turnover kinetics of CPR and MSR reductionwith NADPH reveal differential kinetic behaviors. Scheme 4.1 shows the mechanistic stepsinvolved in MSR and CPR catalysis. In CPR, the initial event of hydride transfer fromNADPH to FAD (II) and electron transfer from FAD to FMN (III) are tightly coupled suchthat there is no accumulation of the FAD hydroquinone intermediate. In MSR, however,these two catalytic events are represented by two discrete kinetic phases such that there istransient accumulation of E-FADH2-FMN. This species accumulates due to the slow rateof interflavin electron transfer. To further investigate the origin of these different kineticbehaviors, the amino acids surrounding the FAD isoalloxazine ring were compared for CPRand MSR. E FADFMN E FADFMNNADPH NADPH E FADH2FMNNADP+E FADHFMNH(NADP+)E FAD(NADP+)FMNH2E FADH2(NADP+)FMNH2 NADPHNADP+2H+, 2e- I II IIIIVVVINADP+Figure 4.1: Proposed mechanism for NADPH-mediated reduction of CPR and MSR. INADPH binds. II Donation of hydride ion from NADPH to FAD. III Formation of thedisemiquinone. IV Both electrons are transferred to the FMN to form the hydroquinone.V Full four-electron reduction to the dihydroquinone by a second molecule of NADPH. VITransfer of reducing equivalents from the enzyme to terminal electron acceptors regeneratesthe fully oxidized enzyme. The presence/absence of the oxidized coenzyme is not absolutelyknown and depicted as parentheseses around NADP+.794.2. BackgroundFigure 4.2: Proximal FAD isoalloxazine ring residues in wild-type MSR (Right) and CPR(Left). PDB 2QTZ and 3Q2EThree well-conserved residues referred to as the catalytic triad, Asp674, Ser457 andCys629 (CPR numbering), are positioned in close proximity to the FAD isoalloxazine ringactive site, see Figure 4.2. Ser457 is positioned within hydrogen bonding distance to thecatalytically important N5 atom of FAD. Asp674 forms a hydrogen bond interaction withSer457; however, this interaction is disrupted upon coenzyme binding as Asp674 forms newcontacts with the C4 atom of the nicotinamide ring. Cys629, lying above the isoalloxazinering, is orientated to also interact with the incoming nicotinamide ring. These residuesstabilize the transition state for hydride transfer from the C4 atom of the nicotinamide ringto the N5 atom of the FAD isoalloxazine ring. Upon nicotinamide ring release, the Ser457and Asp674 residues reform a hydrogen bond network to the N5 atom of the reduced FADand stabilize the semiquinone state.44,60,61The catalytic triad is conserved in plant-type ferredoxin NADP+-reductases, althoughthe aspartate residue is replaced for a glutamate.129 Recall that FNR operates in the reversedirection from MSR and CPR, by transferring reducing equivalents from reduced ferredoxinonto NADP+. Theoretical studies on FNR by Dumit et al. have proposed that, duringcatalysis, the glutamate residue functions as a proton donor to the N5 nitrogen of the FADisoalloxazine ring through the serine side chain.143 Catalysis of MSR and CPR functions inthe reverse of that of FNR, and the aspartate residue has been suggested to play a similarrole in catalysis but as a proton acceptor. Notably, the FAD active site of CPR depicts ahistidine residue (His322 - numbering based off of ratCPR crystal structure and not human804.3. ResultsCPR sequence (His319)) that lies within hydrogen bond distance of Asp674 as shown inFigure 4.2. This histidine residue is conserved among CPR homologues and those of nitricoxide synthase, which also couple hydride and interflavin electron transfer. However, asshown in Figure 4.2, the equivalent residue in MSR is an alanine (Ala312). The presenceof histidine in CPR and NOS but not in MSR may influence the ability of Asp to act as ageneral base and may be the origin of the differences in kinetic behaviors in CPR and MSR.In an effort to determine the structural basis for the differential kinetic behavior ofCPR and MSR, mutations were made to His322 in CPR and the equivalent residue Ala312in MSR. Reciprocal mutations H322A and A312H were made in an attempt to replicatethe local amino acid environment of MSR in CPR and vice versa. Both residues were alsomutated to a glutamine, which can still participate in hydrogen bonding to aspartate but notin acid/base chemistry. The steady state kinetic properties of each variant were measuredby spectrophotometric analysis of cytochrome c or ferricyanide (Fe(CN)63−, abbreviatedFeCN) reduction. Uncoupled NADPH oxidase activity was also determined for each variant.Product inhibition assays with NADP+ reported on the relative coenzyme binding affinitiesof each variant. Lastly, stopped-flow spectrophotometric studies of flavin reduction underpseudo-first order conditions were utilized to evaluate the effect of the mutation on therapid turnover kinetics for these two enzyme variants.In addition, the effect of the hinge length on the kinetic properties of MSR was examined.The much longer hinge of MSR compared to CPR may impart greater conformationalfreedom to the FMN domain, which could limit electron transfer between FAD and FMN.Therefore, the MSR hinge was truncated to 12 amino acids (the equivalent length in CPR)to determine the potential consequence of a longer, more flexible hinge on electron flux inMSR.4.3 Results4.3.1 Flavin characteristics of variantsThe visible absorption spectrum of each variant was measured under aerobic conditionsusing a UV-Vis spectrophotometer. Figure 4.3 shows the reduction of the enzyme variantsby equimolar and saturating concentrations of NADPH. Under equimolar conditions, a lossof absorbance at the flavin absorbance maxima 454 nm was observed with the appearanceof a large disemiquinone absorbance signal centered around 585-605 nm. In the presenceof saturating amounts of NADPH, the absorbance peaks at both 454 nm and 600 nm werereduced. These spectral properties are similar to those observed for wild-type CPR andMSR, indicating that mutation at the His322 (CPR) and Ala312 (MSR) do not affectNADPH-mediated flavin disemiquinone formation in an aerobic environment.814.3. ResultsFigure 4.3: Visible absorption spectra of CPR and MSR variants H322Q (A), H322A (B),A312Q (C), and A312H (D) under aerobic conditions. The recorded spectral propertiesof fully oxidized enzyme (solid black line), after a 5 min incubation with an equimolarconcentration of NADPH (dashed line), and after 10 min incubation with a saturatingconcentration of NADPH (dotted line).824.3. ResultsTable 4.1: Steady state kinetic parameters of wild-type, A312 and H322 variants of MSRand CPR. Conditions: 1 mL reaction volume, 8 µM cytochrome c3+, 0.5 mM FeCN, 2-10pmole enzyme, varied [NADPH] and [NADP+] at 25 ◦C. Assay buffer for MSR is 50 mMTris-HCl pH 7.5 and 50 mM KPi pH 7.5 for CPR. Assays were done in triplicate and fit tothe Michaelis-Menten equation. Inhibition data from four inhibitor concentrations were fitto a competitive inhibition equation.a Data acquired from Wolthers et al.Enzyme k catcytc Km NADPH Ki NADP+ k cat/Km k catFeCN(s−1) (M × 10−6) (M × 10−6) (s−1M−1 × 10+6) (s−1)WTMSRa 7.2 ± 0.1 2.4 ± 0.1 36.9 ± 2.7 3.1 ± 0.3 7.9 ± 0.1A312Q 7.8 ± 0.1 6.2 ± 0.2 29.1 ± 1.9 1.3 ± 0.1 7.1 ± 0.1A312H 3.6 ± 0.1 15.0 ± 0.7 72.9 ± 7.8 0.24 ± 0.01 7.1 ± 0.1WTCPR 20.0 ± 0.2 0.71 ± 0.04 0.95 ± 0.05 28.3 ± 1.8 43.5 ± 0.1H322Q 11.3 ± 0.2 0.58 ± 0.04 0.65 ± 0.05 19.4 ± 1.7 26.1 ± 0.1H322A 5.4 ± 0.1 0.31 ± 0.04 0.30 ± 0.04 17.4 ± 2.5 18.1 ± 0.14.3.2 Steady state kinetic dataSteady state kinetics of each variant were analyzed using two different artificial electronacceptors, ferricyanide (FeCN) and cytochrome c3+. FeCN receives an electron from eitherFAD or FMN cofactors, while cytochrome c only receives an electron directly from theFMN cofactor. The data collected from all steady state assays are summarized in Table4.1. The H322A CPR variant elicited a 4-fold decrease in cytochrome c+3 reduction and a1.5-fold decrease in FeCN reduction. H322Q showed an approximate 50% reduction in thesteady state turnover of cytochrome c+3 and FeCN. Both variants had moderately tighterbinding affinity for NADP+ and a loss of catalytic efficiency attributed to the reduced rateof catalytic turnover with cytochrome c+3. As shown in Figure 4.4, uncoupled NADPHoxidation assays revealed a modest decrease in k cat (1.5-fold slower) for H322A, while k catwas unchanged for H322Q compared to wild-type.Exchanging alanine for histidine in MSR (A312H) weakened the binding of NADP(H),evident by the increase in the apparent Km and Ki for the binding of NADP(H). Thisvariant also elicited slower rates of FeCN and cytochrome c+3 turnover and, together withweaker substrate binding, has 13-fold lower catalytic efficiency. A312Q elicited a similark cat for cytochrome c+3 and FeCN reduction, with a modest increase in substrate bindingaffinity. Catalytic efficiency of this variant decreased by about half due to a higher apparentKm value. The uncoupled NADPH oxidase activity of the MSR variants did not changesignificantly from that of the native enzyme.834.3. ResultsW T C P R H 3 2 2 Q H 3 2 2 A W T M S R A 3 1 2 Q A 3 1 2 H0 . 00 . 20 . 40 . 60 . 81 . 0  Uncoupled NADPH oxidase activity (s-1 )Figure 4.4: Uncoupled NADPH oxidation activity of CPR H322 and MSR A312 variants.Conditions: 1 mL reaction volume, 2 pmole enzyme, 1 mM NADPH, 50 mM Tris pH 7.5at 25 ◦C. For wild-type CPR, H322Q and H322A, the rate constant of uncoupled NADPHoxidase activity generated from linear fits of triplicated data is 1.1 ± 0.1 s−1, 1.0 ± 0.1 s−1and 0.71 ± 0.02 s−1, respectively. The rate constants measured for wild-type MSR, A312Qand A312H are 0.85 ± 0.02 s−1, 0.85 ± 0.03 s−1 and 0.77 ± 0.003 s−1.4.3.3 Multiple and single wavelength pre-steady state kinetics of CPRvariantsThe pre-steady state rates of NADPH-dependent reduction of CPR variants were moni-tored by anaerobic stopped-flow spectrophotometry. Spectral changes upon reduction withNADPH were collected over 1 s and resolved by SVD. Spectral data are shown in Figure4.5. Resolved data were then fitted to a two-step biphasic kinetic model for H322Q and aone-step monophasic kinetic model for H322A; the generated observed rate constants aresummarized in Table 4.2. The kinetic profile of H322Q reduction was comparable to thatof wild-type CPR. The initial fast rate constant (kobs1 = 17 s−1) occurred with a 77% ab-sorbance loss at 454 nm with a simultaneous appearance of a broad absorbance band at 600nm representing formation of the disemiquinone. The second phase was slower (kobs2 = 5s−1) with only a 23% amplitude change at 454 nm and a slight loss of absorbance in thelonger wavelengths. This second phase is assigned to a small portion of the two-electronreduced enzyme population being further reduced to the four-electron level by a secondequivalent of NADPH.The H322A variant exhibited monophasic reduction to the two-electron reduced state.This phase change occurred with a slower rate constant of 8 s−1, 2.5-fold less than kobs1 ofwild-type CPR. Unlike H322Q and wild-type CPR, the absorbance loss at 454 nm was notaccompanied by the formation of an absorbance signal at 600 nm, see Figure 4.5. Indeed,844.3. ResultsTable 4.2: Observed rate constants of the pre-steady state reduction of CPR H322 variants.For details see Figure 4.5CPR kobs1 (s−1) kobs2 (s−1)Wild-type 19.00 ± 0.01 2.72 ± 0.01H322Q 16.6 ± 0.1 5.2 ± 0.1H322A 8.0 ± 0.1 -single wavelength traces for H322A at 600 nm do not show an absorbance change, confirmingthe absence of the disemiquinone species or charge-transfer (CT) complex (Figure 4.6)during the reductive half-reaction. These stopped-flow data indicate that the two-electronreduced enzyme population is either in the FAD hydroquinone (E-FADH2-FMN), the FMNhydroquinone (E-FAD-FMNH2), or perhaps a mixture of the two redox states. Over alonger time frame of 150 s with the autoshutter on to avoid light-induced flavin bleaching,there is a small, gradual buildup of the broad absorbance band at 600 nm indicative of thedisemiquinone species(Figure 4.6). This signal is only a fraction of the equivalent signal inwild-type CPR and H322Q when allowed to reach equilibrium, suggesting that interflavinelectron transfer is not as well supported in the H322A variant. Spectral data for theequimolar reduction of this variant over a longer time frame, in Figure 4.7, also show onlya minor disemiquinone signal. As H322A does not reduce beyond the two-electron levelduring single turnover assays, the enzyme likely populates the FAD hydroquinone state (E-FADH2-FMN) rather than the FMN hydroquinone (E-FAD-FMNH2), as the latter wouldpresumably allow for reduction by a second equivalent of NADPH, which is not observedin the stopped-flow data.4.3.4 Multiple and single wavelength of isolated FAD domain ofwild-type and H322A CPRNADPH-mediated flavin reduction of the isolated FAD domain of wild-type (WTFAD)and H322A (H322AFAD) CPR were analyzed by stopped-flow spectrophotometry. Ab-sorbance changes over multiple wavelengths (375-700 nm) over 10 s were collected andsubjected to SVD analysis, show in Figure 4.8. Both WTFAD and H322AFAD elicitedbiphasic reduction and were best fitted to a two-step kinetic model. The resulting rateconstants were comparable for the two enzymes with a kobs1 of ∼ 32 s−1 and a kobs2 of1 s−1. These rates are also comparable to those obtained from single wavelength tracesobtained at 454 nm, see Figure 4.9. Spectral changes associated with kobs1 include the lossof absorbance at 454 nm coupled with an appearance of a small flat absorbance band inthe long wavelengths (>550 nm). This flat absorbance band represents the formation of a854.3. ResultsFigure 4.5: Spectral changes upon NADPH reduction of CPR H322 variants monitoredby multi-wavelength stopped-flow spectrophotometry. Conditions: 20 µM enzyme, 10-foldexcess NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Time-dependent spectralchanges over a 1 s time frame are shown for H322Q (A) and H322A (B). Global analysisof SVD-resolved spectra generated deconvoluted spectral intermediates for H322Q (C) andH322A (D). H322Q spectral data were best fit to a two-step model with three spectralintermediates (a → b → c). H322A spectral data were best fit to a single-step model withtwo spectral intermediates (a → b).864.3. ResultsFigure 4.6: Anaerobic reduction of H322Q and H322A by saturating [NADPH] monitoredby single wavelength stopped-flow spectrophotometry at 454 and 600 nm. Conditions: 20µM enzyme, 200 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Wild-type CPR(a), H322Q (b) and H322A (c). Rapid absorbance changes over 10 s at 454 nm (Panel A)and 600 nm (Panel B). Panel C shows the absorbance changes occuring at 600 nm over 150s.874.3. ResultsFigure 4.7: Spectral changes upon equimolar NADPH reduction of H322Q and H322A mon-itored by multi-wavelength stopped-flow spectrophotometry. Conditions: 20 µM enzyme,20 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C.charge-transfer complex between the FAD isoalloxazine ring and NADP+. Figure 4.9 showsthe 454 nm single wavelength traces and double exponential fits of these spectra revealedthat the first rate constant of flavin reduction at 454 nm is equal to that of the rate constantfor the ’up’ phase at 600 nm, while decay of the CT complex occurs with the same rateconstant as the kobs2 for flavin reduction.In full-length CPR, the flat absorbance band is not observed, instead the absorbancepeak at 600 nm is attributed to the appearance and decay of the disemiquinone species.The lack of the CT signal in full-length CPR indicates the E-FADH2-NADP+ complex doesnot accumulate as an intermediate, presumably due to rapid interflavin electron transfer.For full-length H322Q, and to a much greater degree for H322A, the diminished absorbancesignal at 600 nm observed in Figure 4.6 indicates impeded interflavin electron transfer inthese variants.4.3.5 Redox Potentiometry of H322APotentiometric analysis of H322A was performed to determine if the removal of theimidazole side chain alters the midpoint potentials of the FAD redox couples. Previous redoxtitrations with wild-type CPR found that the FAD couples were similar in the full-lengthenzyme and the isolated NADPH/FAD domain.66 Thus, the fitting process was simplified byusing the H322AFAD alone in the titrations. Under anaerobic conditions, gradual additionof dithionite led to formation of the FAD semiquinone and complete reduction to the FADhydroquinone. A plot of the absorbance values at 600 nm against the measured potentialvalues that were normalized to the standard hydrogen electron is presented in Figure 4.10.884.3. ResultsFigure 4.8: Spectral changes upon NADPH reduction of wild-type and H322A FAD domainvariants monitored by multi-wavelength stopped-flow spectrophotometry. Conditions: 20µM enzyme, 10-fold excess NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Time-dependent spectral changes over a 1 s time frame are shown for wild-type FAD domain (A)and H322A FAD domain (B). Global analysis of SVD-resolved spectra generated deconvo-luted spectral intermediates for WTFAD (C) and H322AFAD (D). WTFAD and H322AFADwere best fit to a two-step model with three spectral intermediates (a → b → c).894.3. ResultsFigure 4.9: Anaerobic reduction of WTFAD (black line) and H322AFAD (gray line) bysaturating [NADPH] monitored over 10 s by single wavelength stopped-flow spectropho-tometry at 454 and 600 nm. Conditions: 20 µM enzyme, 200 µM NADPH, anaerobic 50mM Tris-HCl pH 7.5 at 25 ◦C. The 454 nm spectral data were fit to a double exponentialgenerating a kobs1 of 31.9 ± 0.1 s−1 and 31.6 ± 0.3 s−1 and a kobs2 of 1.0 ± 0.1 s−1 and 1.2± 0.1 s−1 for WTFAD and H322AFAD.904.3. ResultsA fit of the data in the plot to the Nerst equation generated midpoint potentials of -273 ±6 mV for the FADox/sq couple and -398 ± 20 mV for FADsq/hq. These values compare wellto those obtained for WTFAD, indicating that the imidazole side chain does not influencethe midpoint potentials of the FAD redox center.Figure 4.10: Redox titrations of H322AFAD. Panel A: Spectral features of H322AFAD aftereach addition of reductant. Panel B: Plot of absorbance at 600 nm againsts the normalizedredox potential. These data were fit to the Nernst equation which yielded a midpointpotential of -273 ± 6 mV for FADox]sq and -398 ± 20 mV for FADsq/hq.4.3.6 Multiple and single wavelength pre-steady state kinetics of MSRvariantsRapid turnover of the MSR variants with saturating amounts of NADPH were followedby anaerobic stopped-flow spectrophotometry. After rapid mixing with NADPH, the spec-tral changes over 150 s were recorded from 375 to 700 nm. The spectra obtained for A312Qresembles that of wild-type MSR where reduction follows a three-step kinetic model withfour distinct spectral species. The first kinetic phase (a → b) represents reduction of thefully oxidized enzyme to the FAD hydroquinone species (E-FADH2-FMN). This is evidentby the absorbance bleaching at 454 nm and lack of absorbance change at 600 nm. The firstphase occurs with a kobs1 that is 1.9-fold slower than that of wild-type MSR but with a sim-ilar amplitude change. Conversion of spectral species b to c involves absorbance loss at 454nm and the emergence of an absorbance peak at 600 nm. Thus, the second phase reportson interflavin electron transfer and formation of the disemiquinone. Complete loss of an ab-sorbance signal across the detected wavelengths corresponds to the enzyme reduction by asecond reducing equivalent from NADPH to form the four-electron reduced dihydroquinone(E-FADH2-FMNH2).914.3. ResultsTable 4.3: Observed rate constants of the pre-steady state NADPH reduction of MSR A312variants by stopped-flow spectrophotometry. See Figure 4.11 for details.MSR kobs1 (s−1) kobs2 (s−1) kobs3 (s−1)Wild-type 24.9 ± 0.1 0.18 ± 0.01 0.016 ± 0.003A312Q 13.0 ± 0.1 1.6 ± 0.01 0.10 ± 0.01A312H 0.76 ± 0.01 0.032 ± 0.001 -In contrast to wild-type MSR and A312Q, the kinetic profile of A312H reduction underpseudo-first order conditions fits best to a two-step kinetic model. The conversion of spectralspecies a to b occurs with the loss of absorbance at 454 nm and the appearance of anabsorbance signal at 600 nm, as is observed in wild-type CPR. These spectral changesindicate that the hydride and interflavin electron transfer steps are coupled in A312H.However, the rate constant associated with this step is 30-fold slower than the equivalentstep in wild-type MSR. To determine if the reduced rate of flavin reduction measured forA312H is due to slower hydride transfer, kinetic isotope effect studies were performed with(R)-[4-2H] NADPH. A KIE of 1.3 was determined for the MSR variant, which is slightlyless than the KIE of 1.7 measured for native MSR.65 Therefore, the hydride transfer stepbecomes less rate-determining in A312H than in wild-type MSR.Single wavelength absorption spectrophotometry conducted at 454 and 600 nm allowedfor the determination of initial rate constants for A312Q and A312H reduction. Bothvariants exhibited bleaching of the flavin absorbance maxima at 454 nm and were fittedto a double and single exponential equation for A312Q and A312H. The resulting rateconstants were in agreement with those previously determined by global analysis of multiplewavelength data. Both variants also exhibited an ’up’ and ’down’ phase at 600 nm thatreached a maxima at ∼5 - 10 s then decayed over 100 s during the reductive half-reaction.These two kinetic phases were assigned to the formation and decay of the disemiquinone(E-FADH•-FMNH•). Reliable fits of the 600 nm absorbance changes were not possible.Despite a smaller amplitude change of the 600 nm peak in A312Q and A312H, the variantsform the disemiquinone over a shorter time frame than wild-type MSR, suggesting thatthese variants are more efficient at FADH2 to FMN interflavin electron transfer.4.3.7 Kinetic parameters of the truncated hinge variant of MSRA variant was constructed with only the first 12 amino acids of the 82 amino acid longhinge region for wild-type MSR. Steady state assays on the MSR hinge variant revealedonly a moderate decrease in the rate of cytochrome c3+ reduction (k cat = 2.01 ± 0.05s−1). The ability of the variant to catalyze the reduction of cytochrome c3+ indicates924.3. ResultsFigure 4.11: Anaerobic multiple wavelength stopped-flow spectrophotometry of NADPH-mediated reduction of MSR A312 variants. Conditions: 20 µM enzyme, 200 µM NADPH,anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Absorbance changes from 375 to 700 nm duringthe course of A312Q (A) and A312H (B) reduction. Global analysis of spectral data revealeda three-step kinetic model for A312Q (C) and a two-step kinetic model for A312H (D).934.3. ResultsFigure 4.12: Anaerobic reduction of A312Q and A312H by saturating NADPH monitoredby single wavelength stopped-flow spectrophotometry at 454 and 600 nm. Conditions: 20µM enzyme, 200 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 25 ◦C. Wild-type MSR(a), A312Q (b) and A312H (c). Rapid absorbance changes over 10 s at 454 nm (Panel A)and 600 nm (Panel B). Panel C shows the absorbance changes occuring at 600 nm over 150s.944.4. Discussionthat electron transfer through the variant is still possible. Pre-steady state kinetics of theNADPH-mediated reduction of the MSR hinge variant was monitored by photomultiplierand photodiode array detection, shown in Figure 4.13. Due to low purification yields, a finalconcentration of only 6 µM of the hinge variant was possible. Pseudo-first order conditionswith NADPH were still maintained, however the smaller amplitude changes are attributedto the lower concentration. As expected, the mutation did not affect the rate of FADreduction (kobs1). However, the second and third rate constant decreased by 10- and 2-foldfrom wild-type, reporting on hindered interflavin electron.4.4 Discussion4.4.1 Active site His322 weakens coenzyme bindingThe presence of His322 in the FAD isoalloxazine active site of CPR weakens the bind-ing affinity for the coenzyme. Steady state inhibition assays with NADP+ revealed thatexchanging the histidine for an alanine in CPR resulted in tighter coenzyme binding whilesubstituting the equivalent residue in MSR with a histidine resulted in weaker coenzymebinding. To understand the origin of this effect on coenzyme binding by the active sitehistidine, the crystal structure of the W676X variant is analyzed. This variant, in whichthe C-terminal residues Trp676 and Ser677 are removed, enables productive binding of thenicotinamide ring which is stacked at a 30◦tilt over the FAD isoalloxazine ring. In thisposition, the carboxamide group of the coenzyme nicotinamide ring forms a strong hydro-gen bond (∼2.5 A˚) to the side chain of Asp674. This polar interaction with the coenzymerepositions Asp674, such that the hydrogen bonds with the side chain of Ser457 is severed(over 4 A˚ away). However, His322 remains hydrogen-bonded to Asp674 and potentiallycompetes with the carboxamide group for polar interaction with Asp674. As a consequenceof this competitive interaction, the histidine may indirectly weaken NADP(H) binding andpromote NADP+ release.Previous mutagenesis studies by Hubbard et al. in rat CPR proposed that the releaseof oxidized coenzyme is required for interflavin electron transfer and disemiquinone for-mation.61 They suggested that disruption of the hydrogen bonding network among thecatalytic triad, due to stacking of the nicotinamide ring, prevents deprotonation of FADH2.Supporting this hypothesis are my results for the W676Y and W676F variants. In thesevariants, interflavin electron transfer was impeded because the nicotinamide ring remainedstacked against the FADH2, forming a stable CT species.144 Since the oxidized coenzymeremains tightly bound in the flavin active site, the hydrogen-bonding network presumablyremains disrupted and prevents FAD hydroquinone deprotonation.954.4. DiscussionFigure 4.13: Single and multiple wavelength absorbance changes during the anaerobicreduction of the MSR hinge variant. Conditions: 6 µM enzyme, 60 µM NADPH, anaerobic50 mM Tris-HCl pH 7.5 at 25 ◦C. Panel A and B show the 454 and 600 nm absorbancechanges over the 10 s rapid reduction of the hinge variant, respectively. Hinge variant isshown in grey and wild-type MSR is in black. Panel C shows full spectral changes over 500s (300 spectral scans reduced to 10 for clarity). Panel D shows the deconvoluted spectralprofile of the hinge variant following a three-step kinetic model with four spectral species(a→b→c→d) with observed rate constants of kobs1 = 25.78 ± 0.01 s−1, kobs2 = 0.0200 ±0.0004 s−1 and kobs3 = 0.00711 ± 0.0008 s−1.964.4. Discussion4.4.2 His322 promotes interflavin electron transferMy results show that participation of the His322 imidazole side chain in the hydrogen-bonding network around the FAD isoalloxazine ring favours interflavin electron transfer. InH322Q and to a much greater degree in H322A, the interflavin electron transfer is impeded,as evident by the loss of the disemiquinone absorbance signal at 600 nm. Reduction ofH322A did not involve any significant buildup of a NADP+-FADH2 CT signal. Thus,unlike the W676 variants, the impeded intramolecular electron transfer in H322A is notdue to persistent stacking of the oxidized coenzyme at the FAD active site. Despite thetighter binding affinity observed in H322A, the bulky Trp676 still functions to displace theoxidized nicotinamide ring. Moreover, the equivalent rate constants measured for wild-typeand H322A FAD domain variants indicate that hydride transfer is not influenced by theimidazole side chain and is not the origin of the diminished disemiquinone signal. Instead,the 2.5-fold decrease in the kobs1 of H322A is likely due to disrupted electron transfer fromFADH2 to FMN. Disrupted electron transfer is also evident in the steady state kinetic datafor reduction of cytochrome c and FeCN, where a larger decrease was observed for theformer oxidant compared to the latter. Given that FeCN receives an electron from FADor FMN and cytochrome c receives an electron only from FMN, a mutation that wouldhinder interflavin electron transfer would be expected to affect cytochrome c turnover moresignificantly than that of FeCN. This was observed for the H322A variant.The stopped-flow data did however show that, under anaerobic conditions, interflavinelectron transfer is possible in H322A since the variant can form the disemiquinone. How-ever, it is only over much longer time domains and only at a fraction of the absorbancesignal observed in wild-type. Under aerobic conditions, the disemiquinone absorbance bandin H322A is comparable to that of wild-type. Formation of the disemiquinone under theseconditions is likely attributed to single electron transfer from FADH2 to O2, to form FADH•and a superoxide radical.The definitive mechanism of how His322 promotes interflavin electron transfer is stillunclear. However, due to its position at the active site, the influence of His322 on catalysislikely originates from its role in the hydrogen-bonding network around the FAD cofactor.Through the network, His322 contributes to the correct positioning of Asp674 to optimizethe hydrogen bond contact with Ser457. A stronger interaction between Asp674 and Ser475would favour proton abstraction and formation of the FAD semiquinone. It is also possiblethat the pKa of Asp674 is affected by the non-covalent interaction with His322, therebyinfluencing the ability of Asp674 to act as a general base.974.4. Discussion4.4.3 Differential effect of introducing histidine into MSR active siteIn an effort to reconstruct the hydrogen-bonding network observed for CPR in MSR,the A312H and A312Q variants were expressed and analyzed. Substituting an alanine fora glutamine only elicited modest effects on the kinetic and coenzyme binding properties ofMSR. Notably, disemiquinone formation did peak at an earlier time in A312Q comparedto wild-type MSR. It is possible that the polar side chain of glutamine forms weak polarinteractions with Asp695 that partially mimics the interactions of the water molecules foundin this position in MSR. Interestingly, reduction of the A312H variant was more similar towild-type CPR than MSR. The first observable phase involved the absorbance loss at 454nm along with the emergence of an absorbance band at 600 nm. Like in CPR, the hydridetransfer and interflavin electron transfer steps are tightly coupled into one kinetic phase inA312H. However, this kinetic coupling may be a consequence of the dramatically reducedrate of FAD reduction as well as the long delay that precedes flavin reduction observed at454 nm. Thus, the FAD hydroquinone does not accumulate in this variant. The earlierappearance of the disemiquinone in A312 variants suggests that the more solvent-exposedAsp695 in wild-type MSR is less efficient at deprotonating the N5 nitrogen of the FADisoalloxazine ring.4.4.4 Role of the extended hinge in MSRIn an effort to determine if the 82 amino acid extended hinge of MSR adversely af-fected the rate of intramolecular electron transfer, a variant with a shortened hinge wasconstructed. The selected amino acid sequence was homologous to the first 12 amino acidsof the CPR hinge. MSR and CPR require large-scale conformational motion to alternatebetween a closed state for intramolecular electron transfer and an open state for intermolec-ular electron transfer. Since MSR has a much longer hinge region that tethers the FMNdomain to the FAD/NADPH domain compared to CPR, the enzyme is expected to have amuch greater entropic cost going from the open to closed conformation. Therefore, short-ening the hinge was hypothesized to reduce the entropic cost of conformational change andenhance interflavin electron transfer. Kinetic analysis of the hinge variant revealed thatthe first rate constant associated with hydride transfer was not altered by the mutationhowever, the second and third rate constants of flavin reduction were significantly slower.A possible explanation for these effects are that the residues specifically involved in thecoordinated motion of the FMN domain were removed, thereby disrupting the formationof the closed enzyme state. Furthermore, truncating such a large segment of the primaryamino acid sequence may have led to a degree of misfolding in the secondary and tertiarystructures of MSR.984.5. Experimental Procedure4.5 Experimental Procedure4.5.1 MaterialsUnless otherwise stated, all chemical reagents were purchased from Fisher Scientific.NADPH, NADP+, 2’,5’-ADP, and cytochrome c3+ were obtained from Sigma Aldrich(Oakville, ON, Canada). [4(R)-2H]NADPH (A-side NADPD) was synthesized and iso-lated as previously described.144 Pfu Turbo DNA polymerase, Taq DNA polymerase andXl1 Blue cell lines were purchased from Agilent Technologies (Mississauga, ON, Canada).Rosetta(DE3)pLysS competent cells were obtained from EMD Biosciences. Protein purifi-cation supplies, Resource Q column and glutathionine sepharose 4B resin, were purchasedfrom GE Biosciences.4.5.2 Generation and expression of MSR and CPR variantsThe MSR A312H and A312Q variants and the CPR H322A and H322Q variants weregenerated from soluble (transmembrane domain cleaved) wild-type CPR plasmid pET-15band wild-type MSR plasmid pGEX-4T1 using the QuikChange Site-Directed MutagenesisKit (Agilent Technologies). Oligonucleotide primers were designed based on the publishedsequence (Accession number: NM000941 CPR and AF121214 MSR) and purchased fromIntegrated DNA Technologies (Coralville, Iowa, USA). The primers are tabulated in Table4.4. NAPS DNA Sequencing Laboratory of the University of British Columbia (Vancouver,Canada) confirmed the desired mutations were made without additional errors. Successfulplasmids of both MSR and CPR were then transformed into the Escherichia coli strainRosetta2(DE3)pLysS using the heat shock method. The MSR recombinant GST-fusionproteins were expressed and harvested as previously described in Section 2.5.2. The CPRrecombinant His-tagged proteins were expressed and harvested as described in Section 3.4.3.4.5.3 Generation and expression of the isolated FAD/NADP(H) domainof CPRThe gene encoding for the isolated FAD/NADP(H)-binding and connecting domain ofCPR was PCR amplified from the cDNA of wild-type CPR. The primers were designedbased on the published sequence (Accession number: NM000941) with flanking NdeI andBamHI restriction sites, Table 4.4. Both the PCR amplified product and the pET15bvector were digested with NdeI and BamHI then ligated together using the T7 DNA ligaseand allowed to ligate overnight at 16 ◦C. Ligation products were then transformed into theXL1-Blue cell strain (Novagen). The plasmid construct (pCPR-FAD) was isolated using theE.Z.N.A Plasmid Mini Kit I from Omega Bio-tek then sent off for sequencing at the NAPS994.5. Experimental ProcedureTable 4.4: The forward (F) and reverse (R) oligonucleotide primers designed for each MSRand CPR variant. Mutation is in bold.Enzyme Variant Oligonucleotide SequenceA312HMSR F 5’ CCTATCAGCCTGGAGATCACTTCAGCGTGATCTGC 3’A312HMSR R 5’ GCAGATCACGCTGAAGTGATCTCCAGGCTGATAGG 3’A312QMSR F 5’ CTATCAGCCTGGAGATCAGTTCAGCGTGATCTGC 3’A312QMSR R 5’ GCAGATCACGCTGAACTGATCTCCAGGCTGATAG 3’H322ACPR F 5’ GTATGAATCTGGGGACGCCGTGGCTGTGTACC 3’H322ACPR R 5’ GGTACACAGCCACGGCGTCCCCAGATTCATAC 3’H322QCPR F 5’ GTATGAATCTGGGGACCAAGTGGCTGTGTACC 3’H322QCPR R 5’ GGTACACAGCCACTTGGTCCCCAGATTCATAC 3’FAD-domain F 5’ GTAGCTCATATGCGTCAGTACGAGCTTGT 3’R 5’ CAAGTCGGATCCCTAGCTCCACACGTCC 3’DNA Sequencing Laboratory at the University of British Columbia (Vancouver, Canada).Successful clones of pCPR-FAD were kept and stored at −20 ◦C. The H322A mutation wasmade in pCPR-FAD by site-directed mutagenesis then isolated and sent for sequencing.Successfully mutated plasmids were transformed into Rosetta2(DE3)pLysS and expressedas described in Section 3.4.3.4.5.4 Generation of MSR truncated hingeTo avoid adding more mutations than necessary to the sequence of MSR, recursive PCRwas used to synthesize the wild-type FMN-domain with the new hinge sequence. First, anXbal restriction enzyme site was introduced into the C-terminal end of the hinge region inwild-type MSR using the following primers: Forward 5’ GCC TCT CTG AAT ATT CTAGAT TTA CCC CCA GAA TAT TTA C 3’, and Reverse 5’ G TAA ATA TTC TGG GGGTAA ATC TAG AAT ATT CAG AGA GGC 3’. The mutation was introduced using therecommended reaction mixture concentrations and cycling conditions from the QuikChangeSite-Directed Mutagenesis Kit (Agilent Technologies). The mutation was confirmed by theNAPS DNA Sequencing Laboratory of the University of British Columbia (Vancouver,Canada).1004.5.ExperimentalProcedureTable 4.5: Recursive PCR forward (F) and reverse (R) oligonucleotide primers designed for wild-type MSR FMN-domain withCPR hinge synthesis.Name Oligonucleotide SequenceFlanking1 (EcoR1 Site) F 5’ GTGCCGCGCGGCGAATTCATGCGCCGCTTTCTGCTGCTGTATGCTACCCAGCAGGG 3’2 (Xba1 Site) R 5’ CCTGTAAATATTCTGGGGGTAAACCAGGAATATCTAGACGAATGCTGGACTCC 3’Internal1 R 5’ GCTCACATATTTCTTCTGCGATGGCCTTTGCCTGTCCCTGCTGGGTAGCATACAG 3’2 F 5’ CGCAGAAGAAATATGTGAGCAAGCTGTGGTACATGGATTTTCTGCAGATCTTC 3’3 R 5’ TCGGTTTTTAGGTCATACTTATCGGATTCACTAATACAGTGAAGATCTGCAGAAAATCCA 3’4 F 5’ GACCTAAAAACCGAAACAGCTCCTCTTGTTGTTGTGGTTTCTACCACGGGCACCGGAG 3’5 R 5’ CTGTATTTCCTTAACAAACTTGCGGGCTGTGTCGGGTGGGTCTCCGGTGCCCGTG 3’6 F 5’ GCAAGTTTGTTAAGGAAATACAGAACCAAACACTGCCGGTTGATTTCTTTGCTCACCTG 3’7 R 5’ GTATTCTGAATCACCGAGACCCAGTAACCCATACCGCAGGTGAGCAAAGAAATCAAC 3’8 F 5’ GTCTCGGTGATTCAGAATACACCTACTTTTGCAATGGGGGGAAGATAATTGATAAACG 3’9 R 5’ GTGTCATAGAAATGCCGGGCTCCAAGCTCTTGAAGTCGTTTATCAATTATCTTCCCCC 3’10 F 5’ CCGGCATTTCTATGACACTGGACATGCAGATGACTGTGTAGGTTTAGAACTTGTGGTTG 3’11 R 5’ CTTTCTGAGGGCTGGCCAGAGTCCAGCAATCCACGGCTCAACCACAAGTTCTAAACC 3’12 F 5’ GCCAGCCCTCAGAAAGCATTTTGGGGTGGAAGCCACTGGCGAGGAGTCCAGCATTCGT 3’1014.5. Experimental ProcedureDe novo synthesis of the wild-type MSR FMN-domain with the CPR hinge was accom-plished through ”one pot” recursive PCR.145. Fourteen individual oligonuleotide strandswith lengths ranging 53-60 bases were designed based on the coding and non-coding nu-cleotide sequence to form ∼20 bp overlapping complimentary ends for extension duringthe PCR cycles, see Table 4.5. The existing N-terminal EcoR1 site and the original FMNdomain sequence were preserved while the hinge sequence (5’463AGGT...TATT726 3’) wasexchanged for that of CPR with a terminal Xba1 site (underlined) (5’ GGG GTG GAAGCC ACT GGC GAG GAG TCC AGC ATT CGT CTA GA 3’).105 The 50 µL reactionmixture consisted of 5 µL 10× Taq reaction buffer, 1.5 µL dNTP (10 mM each), 0.2 µM ofeach internal oligo, 2 µM of each flanking oligo, and 2 units of Taq DNA polymerase. PCRcycling conditions were as follows: 30 cycles of 1 min at 95 ◦C, 1 min at 64 ◦C, and 1 min at72 ◦C with a final extension time of 5 min at 72 ◦C. The resulting PCR product was thenamplified using the flanking oligos. Successful, complete gene synthesis was confirmed bythe NAPS DNA Sequencing Laboratory using the flanking oligos as primers.Synthesized gene insert and full length MSR plasmid with introduced Xba1 site vectorwere purified using a Cycle Pure Kit from Qiagen (Valencia,CA), and digested with EcoR1and Xba1 restricting enzymes for two hours, or overnight at 37 ◦C, respectively. The vectorwas then dephosphorylated with antarctic phosphatase (37 ◦C for 30 min, followed by 65 ◦Cfor 30 min). The insert and vector were again purified using a Cycle Pure Kit from Qiagen.Cut plasmid was isolated using a Gel Extraction Kit (Qiagen). One hundred nanograms ofvector and 21 ng of insert were then ligated in a total volume of 20 µL, containing 1 unit ofT4 DNA ligase and 2 µL 10× buffer at 16 ◦C overnight. Ligation products were transformedinto E.coli Xl1 Blue cell strain onto LB plates containing 100 µg mL−1 ampicillin and grownovernight at 37 ◦C. Successful mutagenesis and ligation without any PCR-induced errorswas confirmed through sequencing.4.5.5 Purification of MSR variantsPurification of the MSR variants followed the procedure previously described in Section2.5.4.4.5.6 Purification of CPR variantsPurification of the CPR variants followed the procedure previously described in Section3.4.5.1024.5. Experimental Procedure4.5.7 Steady state turnover analysisSteady state turnover assays for the MSR variants were conducted as previously de-scribed in Section 2.5.5. CPR variant assays followed the same protocol but were performedin 50 mM KPi pH 7.5 buffer. Steady state reduction of FeCN was also monitored at 420nm by spectrophotometry (extinction coefficient of 1 020 M−1cm−1) with 0.5 mM FeCN,100 µM NADPH and 10 nM of enzyme.4.5.8 Pre-steady state kinetic analysisPre-steady state kinetic assays for the MSR and CPR variants were conducted as pre-viously described in Section 2.5.6.4.5.9 Redox PotentiometryPre-steady state kinetic assays for the MSR and CPR variants were conducted as pre-viously described in Section 3.4.8. Potentiometric titrations on the isolated FAD/NADPH-and connecting domain of CPR were conducted in 50 mM potassium phosphate pH 7.0 with20 % glycerol to discourage precipitation.103Chapter 5Conservation of an aromaticFAD-shielding residue, Trp704, inArtemisia annua CytochromeP450 Reductase5.1 SummaryIn this chapter, the role of the FAD isoalloxazine ring-shielding tryptophan residue(Trp704) in plant CPR derived from Artemisia annua is investigated through site-directedmutagenesis. Four mutations are made to the C-terminal Trp704 to reduce and remove thepi-pi-stacking interaction with the isoalloxazine ring: W704Y, W704F, W704H, and W704S.Aerobic and anaerobic spectrophotometry allowed for the steady state and pre-steady statekinetic characterization of each AaCPR. Through these studies, the Trp704 in AaCPR,as in human CPR, was determined to trigger NADP+ release which is a partially rate-determining step in AaCPR catalysis. However, dissociation of the coenzyme is not asrate-determining in AaCPR as it is in human CPR. The flavin reduction kinetics of AaCPRvariants differ notably from those of human CPR and are discussed herein.5.2 BackgroundArtemisia annua is a Chinese herb that is harvested for the medicinal chemicalartemisinin. Artemisinin, an endoperoxidized sesquiterpene, is a highly effective antimalar-ial, and artemisinin-based combination therapies are recommended by the World HealthOrganization as the first-line treatment for the disease.146 To increase drug affordabil-ity and availability, metabolic engineering of microbes has been employed for large-scalesemi-synthesis of artemisinin.147–149 Keaslings and coworkers transformed Saccharomycescerevisiae and Escherischia coli with three genes from A. annua that encode for enzymesin the artemisinin biosynthetic pathway: amorpha-4,11-diene synthase, CYP71AV1, andcytochrome P450 reductase (AaCPR).148,149 Amorpha-4,11-diene synthase catalyzes the1045.2. Backgroundfirst committed step in artemisinin synthesis, the conversion of farnesyl pyrophosphate toamorpha-4,11-diene. The cytochrome P450 monooxygenase CYP71AV1 catalyzes the three-step oxidation of amorpha-4,11-diene to produce the artemisinin precursor, artemisic acid.CYP71AV1 activity requires the sequential transfer of two electrons from AaCPR.CPR from Artemisia annua shares ∼ 50% amino acid sequence similarity with humanCPR, and the overall catalytic mechanism of the plant enzyme is expected to be similarto that formerly described for human CPR. In an effort to further optimize the engineeredbiosynthetic pathway for artemisinin production, Simtchouk et al. investigated the kineticand thermodynamic properties of AaCPR.150 Overall, the catalytic performance of AaCPRis improved compared to human CPR as the plant enzyme reduces cytochrome c with a3-fold faster k cat. Interestingly, it does so with a 30-fold weaker binding affinity for NADP+.Most striking of all is the pre-steady state flavin reduction of AaCPR over the reductivehalf-reaction. Even with the temperature lowered from 25 ◦C to 6 ◦C, flavin reduction ismuch faster in AaCPR with reduction of the enzyme by the first and second NADPHequivalent occurring at rate constants of >500 and 18 s−1. The improved rates are notattributed to the flavin center redox potentials, as they are similar to those determined inhuman CPR. However, unlike human CPR, the spectral changes associated with the firstkinetic phase include formation of a large absorbance signal in the long wavelengths thatis characteristic of a charge-transfer species. This charge-transfer species is assigned to theinteraction between FAD and the nicotinamide ring of the coenzyme. A greater propensityfor the Trp704-displaced conformation in AaCPR may be the origin of the acceleratedobserved rates of flavin reduction.Alternatively, Simtchouk et al. also proposed that the accelerated flavin reduction maybe due to weaker binding of the oxidized coenzyme. Previous studies with mammalianCPR have found that NADP+ release is necessary for formation of the disemiquinone in-termediate and binding of a second NADPH for further flavin reduction; as such NADP+release has been proposed as the rate-determining step in the reductive half-reaction inmammalian CPR.60,68,144 These studies revealed an increase in the primary kinetic isotopeeffect associated with hydride transfer under high ionic strength conditions. These condi-tions weaken coenzyme binding, such that product release is less rate-determining. This, inturn, unmasks the isotope effect associated with breakage of a C-H/D bond. In low ionicstrength conditions (conditions that promote tight coenzyme binding), the intrinsic KIE forhydride transfer is diminished to 1.0, which indicates that other steps (i.e. NADP+ release)are more rate-determining.151 Furthermore, an elevated primary KIE measured for AaCPR(1.7) in comparison with human CPR (1.1) under the same assay conditions reveal a shiftin the rate-determining step from NADP+ release to hydride transfer in AaCPR.1501055.3. ResultsDespite an exceptional difference in coenzyme binding affinity in AaCPR, an amino acidsequence alignment with human CPR shows that the key residues identified for coenzymebinding are conserved in AaCPR. Furthermore, the FAD isoalloxazine ring-stacking aro-matic trytophan (Trp704) is also conserved. In light of the regulatory role of the conservedaromatic residue in coenzyme binding and intramolecular electron flow in human CPR andother diflavin oxidoreductases, the potential differential regulatory role of Trp704 in AaCPRwas investigated.The subtle differential regulatory role of the C-terminal Trp704 residue in coenzymebinding and electron flow in AaCPR was determined by mutagenesis. For direct compari-son with human CPR, the following mutations were made: W704S, W704H, W704Y, andW704F. Mutation to histidine (W704H), tyrosine (W704Y) and phenylalanine (W704F)were chosen to maintain residue aromaticity, while investigating the role of side chain sizeand polarity. The effect of reducing side chain size and aromaticity is evaluated from mu-tation of Trp704 to a serine (W704S). Spectroscopic analysis cytochrome c reduction witheach variant was used to determine their respective steady state kinetic properties. Pre-steady state flavin reduction under anaerobic conditions with saturating NADPH for eachvariant was measured by stopped-flow spectrophotometry.5.3 Results5.3.1 Steady state kinetic parametersCPR-catalyzed reduction of cytochrome c was used to determine the steady state kineticproperties of each W704 variant. The data are summarized in Table 5.1. Compared to wild-type AaCPR, all W704 variants elicited slower enzymatic turnover with k cat values rangingfrom 0.11 to 9.21 s−1. The slowest variant is W704S followed by W704F, W704H, and thefastest is W704Y. Under the same assay conditions as wild-type AaCPR, it was not possibleto extract an inhibition constant for NADP+ as the necessary [NADPH] values were toolow for the detection limits of a 1 cm path length cuvette. The lack of detectable Ki forW704Y and W704F, together with the Ki determined for W704H and W704S, indicatesthat the coenzyme binding affinity is much higher in the W704 variants.1065.3. ResultsTable 5.1: Steady state kinetic parameters of AaCPR W704 variants. Conditions: 1 mL re-action volume, 8 µM cytochrome c3+, 2 pmole enzyme, varied [NADPH] and [NADP+], 50mM Tris-HCl pH 7.5 at 25 ◦C. Assays were done in triplicate and fit to the Michaelis-Mentenequation. Inhibition data from four inhibitor concentrations were fit to a competitive inhi-bition equation.Enzyme k cat Km Ki k cat/Km(s−1) (M × 10−6) (M × 10−6) (s−1M−1 × 10+6)WT 66 ± 1 1.22 ± 0.11 1.59 ± 0.18 54 ± 5W704Y 22.5 ± 0.4 0.07 ± 0.01 – 132 ± 20W704F 4.3 ± 0.1 0.15 ± 0.02 – 29 ± 4W704H 5.8 ± 0.1 0.42 ± 0.04 0.38 ± 0.04 14 ± 2W704S 0.110 ± 0.001 0.11 ± 0.01 0.24 ± 0.03 1.0 ± 0.15.3.2 Multiple and single wavelength pre-steady state kineticsPre-steady state reduction of AaCPR variants with saturating NADPH was monitoredby anaerobic stopped-flow spectrophotometry. With wild-type AaCPR, initial studies at25 ◦C revealed a large loss of spectral data within the dead-time of the stopped-flow.150 Thetemperature was then reduced to 6 ◦C in an effort to capture the first rate of flavin reduction.Thus, for a direct comparison with wild-type AaCPR, all stopped-fow experiments withW704 variants were also performed at 6 ◦C. Figure 5.1 displays the multiple wavelengthspectra of the oxidized variant combined with the spectra recorded over 15 s after rapidmixing with excess NADPH. Like in wild-type AaCPR, some spectral data of W704F andW704Y are lost in the dead-time (1.5 ms), indicating the relatively rapid (>600 s−1) rateof flavin reduction in these variants. An approximate 30 % amplitude change at the flavinabsorbance maxima is observed for W704Y, while an amplitude change of ∼ 14% is observedfor W704F. Despite the lost spectral data in W704Y and W704F, double exponential fits ofsingle wavelength traces at 454 nm (Figure 5.3) enabled the extraction of a fast kobs1 andslow kobs2. The data are summarized in Table 5.2. The kobs1 for both W704Y and W704Fis > 700 s−1, while the corresponding rate constant in wild-type AaCPR is 500 s−1. Thesevalues underestimate the true rate constant, as there was a small loss of absorbance withinthe dead-time of the stopped-flow.150 The smaller initial amplitude change associated withkobs1 of W704Y and W704F may indicate that a smaller fraction of the population is beinginitially reduced. The 4- and 6-fold slower kobs2 of W704Y and W704F indicates a slowerreduction in these variants. No spectral data are lost within the dead-time for W704H andW704S reduction, indicating that flavin reduction is relatively slow in these variants withkobs1 of 4 s−1 and 0.018 s−1, respectively. The initial reduction of W704S is significantlydelayed, as shown in Figure 5.3 and 5.4 A.1075.4. DiscussionLike native AaCPR, the spectral profile of the NADPH-dependent reduction of W704Yshows the loss of the flavin absorbance maxima with the simultaneous appearance of abroad absorbance band from 530-700 nm. This broad flat absorbance band is assignedto the accumulation of a charge-transfer species between the nicotinamide and the FADisoalloxazine rings. The band is fully formed in the dead-time and Figure 5.3 shows thatthe concentration of the species is constant over the course of the reductive half-reaction,even over a 150 s extended time domain. In Figure 4.5 panel B, the flavin reduction ofW704F at 454 nm is also accompanied by a build up of a flat absorbance band at >500nm that is indicative of a charge-transfer species. Unlike wild-type AaCPR and W704Y,flavin reduction of W704F and formation of the CT species at 600 nm occurs gradually andcompletes within 15 s. As equilibrium is established over 150 s, the disemiquinone signal isslightly unmasked as a slight peak centered at 600 nm is observed. The overlapping signalsof the disemiquinone and the charge-transfer complex complicate the analysis of spectraldata at 600 nm. Attempts have been made in Anabaena FNR to assign specific spectralfeatures to different charge-transfer complexes, however these analyses are conducted atlonger wavelengths (700-1000 nm) which are beyond the limits of the photodiode arrayused here.152Reduction of W704H is well within the detection limits of the stopped-flow instrument.Like W704F, a gradual decrease in absorbance at 454 nm occurs with the gradual increasein absorbance in the longer wavelengths. While W704H forms the charge-transfer species, itdoes so at approximately half the amplitude observed in W704F (see 5.1 and 5.3). Substi-tution of Trp704 with a serine severely disrupted flavin reduction as evident by the minimalspectral changes at 454 and 600 nm over the same time domain as the other W704 variants,shown in Figure 5.3. Figures 5.1 and 5.4 show that even after an extended time domain of150-300 s, the variant is only partially reduced with a small absorbance signal assigned tothe disemiquinone species.The early ’up’ and ’down’ phases at 600 nm observed in wild-type AaCPR are not presentin any of the variants. In wild-type AaCPR, this peak was assigned to the formation anddecay of the disemiquinone intermediate. In the W704 variants, the disemiquinone speciesmay not accummulate and/or the spectral signal may be masked by the greater absorbancesignal for the CT species.5.4 DiscussionA comparison of the Trp704 variants of AaCPR with the Trp676 variants of human CPRwas made to assess the potential differential role of the FAD-stacking aromatic residuein CPR catalysis. Recall from Chapter 3 that mutation of W676 to a smaller aromatic1085.4. DiscussionFigure 5.1: Visible spectra of AaCPR W704 variants during reduction by NADPH, moni-tored by multiple wavelength stopped-flow spectrophotometry. Conditions: 10 µM enzyme,10-fold excess NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 6 ◦C. Spectral scans of wild-type AaCPR (A), W704Y (B), W704F(C ) and W704H(D) show the progressive reductionof oxidized enzyme (black) to reduced (light gray) over a 15 s time domain. After 150 s,another trace was recorded (black dashed line).Table 5.2: Observed rate constants of the pre-steady state reduction of AaCPR variants at454 nm. Conditions: 10µM enzyme, 100 µM NADPH, anaerobic 50 mM Tris-HCl pH 7.5at 6 ◦C. An average of 3-5 traces was taken and the resulting trace was fitted to a doubleexponential equation.Enzyme kobs1 (s−1) kobs2 (s−1)WT 499 ± 4 18.10 ± 0.04W704Y >700 4.76 ± 0.04W704F >700 2.78 ± 0.02W704H 4.4 ± 0.1 0.82 ± 0.01W704S 0.018 ± 0.007 0.006 ± 0.0011095.4. DiscussionFigure 5.2: Visible spectra of W704S variant during reduction by NADPH, monitored bymultiple wavelength stopped-flow spectrophotometry. Conditions: 10 µM enzyme, 10-foldexcess NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at 6 ◦C. Spectral scans of W704S showthe oxidized enzyme (black) mixed with NADPH over 15 s (gray). After 150 s, anothertrace was recorded (black dashed line).residue led to tighter binding of NADP(H).144 This was evident in the lower Ki values forNADP+ in the W676F and W676Y variants. By decreasing the size of the isoalloxazinering-stacking residue, the energetic cost of nicotinamide displacement is presumably loweredand coenzyme binding becomes more energetically favourable. Tighter coenzyme bindingwas also evident in stopped-flow studies, where a stable NADP+-FADH2 charge-transfer(CT) complex was generated upon reduction of the fully-oxidized enzyme with NADPH.The inability of the smaller aromatic side chains to effectively displace the nicotinamide ringaway from the FAD likely leads to the stable formation of this CT complex. As a result,NADPH-mediated reduction of human CPR W676Y and W676F is monophasic (as opposedto biphasic for the native enzyme) leading to a two-electron reduced form of the enzyme.At the completion of the reductive half-reaction, there is a distribution of the NADPH-FAD-FMN and NADP+-FADH2-FMN complexes. The rate constants for flavin reductionin human CPR W676Y and W676F are reduced to 6 and 0.7 s−1, respectively, presumablybecause tight binding of NADP+ prevents disemiquinone formation, which would otherwisedrive the forward flux of electrons to the higher potential FMN cofactor.Likewise, the W676H variant, reported by Gutierrez and coworkers, also exhibitsmonophasic flavin reduction to the two-electron level with a slower kobs of 3 s−1.137 W676Hflavin reduction is also accompanied by formation of the CT complex (NADP+-FADH2-1105.4. DiscussionFigure 5.3: Anaerobic single wavelength stopped-flow traces of AaCPR W704 variants.Conditions: 10 µM enzyme, 10-fold excess NADPH, anaerobic 50 mM Tris-HCl pH 7.5 at6 ◦C. Panel A: Averaged traces taken at 454 nm over 1 s. Panel B: Averaged traces takenat 600 nm over 10 s. W704Y (dark gray), W704F (gray), W704H (light gray), and W704S(black).1115.4. DiscussionFigure 5.4: Anaerobic flavin reduction of W704S monitored by single wavelength stopped-flow spectrophotometry. Conditions: 10 µM enzyme, 10-fold excess NADPH, anaerobic 50mM Tris-HCl pH 7.5 at 6 ◦C. Panel A: an average of 3 traces of W704S reduction at 454nm over 300 s. Panel B : an average of 3 traces of W704S reduction at 600 nm over 150 s.1125.4. DiscussionFMN). For W676H, W676Y and W676F, the oxidized coenzyme is essentially ’locked’ inthe active site, however the CT complex does not form irreversibly since all three variantsare capable of reducing cytochrome c under steady state conditions, albeit at a slower ratethan that of wild-type CPR.Like the W676Y and W676F variants of human CPR, the corresponding variants ofAaCPR elicited a reduction in k cat for cytochrome c3+ and a tighter binding affinity forthe coenzyme as evident by the lower Km for the substrate. These data suggest that,as for the human enzyme, the bulky W704 triggers release of NADP+ to make this stepless rate-determining in the overall mechanism. That being said, W676 is perhaps morecritical for this step of catalysis compared to the AaCPR W704. Since coenzyme bindingaffinity is much stronger in human CPR, the oxidized coenzyme remains bound in the activesite and requires disruptive repulsive interactions to promote its release. For AaCPR, thecoenzyme binding affinity is weaker compared to human CPR, therefore release of theoxidized coenzyme may be less rate-determining in the plant species.This hypothesis is based on stopped-flow data collected for the W704F and W704Yvariants of AaCPR, where in contrast to the W676 variants of human CPR, the W704Yvariants of AaCPR exhibit biphasic reduction kinetics. Like native AaCPR, there was aninitial rapid phase of flavin reductin (kobs1) at >700 s−1, for both W704Y and W704F,followed by a slower kobs2 of 5 and 3 s−1, respectively. For W704Y, approximately half ofthe first kinetic phase is complete within the dead-time of the stopped-flow as there wasan ∼58% absorbance loss at 454 nm. This absorbance loss is approximately 38% greaterthan that observed for native AaCPR, suggesting that reduction by the first NADPH isfaster with the W704Y substitution. Also like native AaCPR, a charge-transfer species alsofully forms within the dead-time of the stopped-flow, but unlike in human CPR W676Yand W676F variants, this species does not appear to greatly affect further reduction ofthe enzyme by a second equivalent of NADPH. This is evident by the biphasic nature offlavin reduction in W704Y and W704F. Thus, further reduction suggests that the smalleraromatic side chains are more proficient at displacement of the nicotinamide ring in theAaCPR variants compared to the human CPR variants.It is unclear as to why the appearance of the charge-transfer complex in AaCPR does notlead to inhibition of further reduction of the enzyme by a second equivalent of NADPH. It ispossible that, in AaCPR, more FAD is initially reduced by NADPH and that a fraction of theenzyme population will be converted to the disemiquinone and FMN hydroquinone statesupon NADP+ release while the rest remains bound to the coenzyme as NADP+-FADH2-FMN. Further reduction of the FMN hydroquinone (FAD-FMNH2) population by a secondmolecule of NADPH then forms the four-electron reduced state FADH2-FMNH2 which mayalso be bound to oxidized coenzyme. Thus, the large CT signal is attributed to a mixture of1135.4. Discussiontwo-electron reduced (NADP+-FADH2-FMN) and four-electron reduced (NADP+-FADH2-FMNH2) enzyme species, accounting for the constant high concentration of CT signal inW704Y. Alternatively, the charge-transfer complex may arise from the unproductive bindingof NADPH against the various forms of two- and four-electron reduced flavins (NADPH-FADH•-FMNH•, NADPH-FAD-FMNH2, or NADPH-FADH2-FMNH2). The identity of thecharge-transfer complexes that form over the course of the reductive half-reaction of AaCPRis key to understanding the catalytic behavior of this enzyme. Recently, some success inthis area has been acheived in Anabaena FNR however a photodiode array with a widerwavelength range (up to 900-1000 nm) is necessary.152For W704F, the amplitude change associated with kobs1 is significantly smaller comparedto native AaCPR and W704Y, but like these two forms, the rate constant is >700 s−1 anda significant portion of this phase of the reaction occurs in <1 ms. The smaller amplitudechange indicates that a smaller fraction of the enzyme is reduced in this first kinetic phase.It is unclear as to why this is, but it may be due to changes in the electronic environmentaround the FAD cofactor or the greater accumulation of the CT absorbance band in thisvariant. Further reduction of W704F spans a greater time domain than W704Y and is alsomultiphasic.In W704H, the biphasic flavin reduction was slower with kobs1 of 4 s−1 and kobs2 of 0.8s−1, and the coenzyme binding affinity is moderately stronger in this variant, as evidentby the Ki for NADP+. Unlike W704Y and W704F, the initial kinetic phase of W704Hreduction is fully captured by stopped-flow spectrophotometry. The reason behind thissignificantly slower flavin reduction is not immediately clear. Interestingly, the W704Hvariant also does not form a CT band within the dead-time of the stopped-flow, meaningthat placement of the nicotinamide ring against the FAD is inhibited in this variant. Thismay account for the reduced kobs1.Exchanging the Trp704 for a serine renders the AaCPR variant catalytically inactive.This is evident by the severely reduced rates for the steady state reduction of cytochromec. Moreover, the severely delayed flavin reduction is no longer under pre-steady stateconditions and is likely nonphysiological. Since the serine side chain does not form anypi-pi-stacking interactions with the FAD isoalloxazine ring, the energetic cost of residuedisplacement is expected to be reduced significantly. Therefore, nicotinamide ring placementover the hydrophobic isoalloxazine should be favourable. However, there is no spectralevidence of a rapidly formed charge-transfer complex in W704S. Interestingly, mutation ofthe equivalent Trp704 residue in pea FNR to a serine (Y308) resulted in a much greateractive site occupancy of the nicotinamide ring as determined from spectral data, and wascaptured in Y308S crystal structures.62 Later theoretical modeling studies on Y303S inAnabaena FNR suggest that the active site aromatic Tyr303 residue contributes to the1145.5. Experimental Proceduresoptimal orientation of the nicotinamide ring to the isoalloxazine ring, thereby influencingthe efficiency and mechanism of the reaction.152–154 Therefore, it is possible that in W704S,the nicotinamide ring may be binding to the active site in a nonproductive way that doesnot favour formation of the charge-transfer complex or hydride transfer.5.5 Experimental Procedures5.5.1 MaterialsThe reagents NADPH, NADP+, 2’,5’-ADP, and cytochrome c3+ were ordered fromSigma Aldrich (Oakville, ON, Canada). Agilent Technologies (Mississauga, ON, Canada)provided the Pfu Turbo DNA polymerase, Taq DNA polymerase and Xl1 Blue cell lines.Rosetta(DE3)pLysS competent cells were purchased from EMD Biosciences. Protein purifi-cation supplies, nickel-nitriloacetetic acid, Resource Q column and glutathionine sepharose4B resin, were obtained from GE Biosciences. All other chemical reagents were purchasedfrom Fisher Scientific.5.5.2 Generation and expression of AaCPR variantsThe W704 variants of Artemisia annua CPR were generated from soluble (trans-membrane domain cleaved) wild-type AaCPR plasmid pET-15b (pETAaCPR) using theQuikChange Site-Directed Mutagenesis Kit (Agilent Technologies).150 Oligonucleotideprimers were designed from the DNE sequencing results of pETAaCPR and ordered from In-tegrated DNA Technologies (Coralville, Iowa, USA). See Table 5.3 for the primer sequences.The mutations were confirmed by sequencing from the NAPS DNA Sequencing Laboratoryof the University of British Columbia (Vancouver, Canada). We transformed the plasmidsusing the heat shock method into the Escherichia coli strain Rosetta2(DE3)pLysS. The His-tagged recombinant AaCPR variants were expressed and harvested as previously describedin Section 3.4.3.5.5.3 Steady state turnover analysisSteady state turnover assays for the AaCPR variants were performed as previouslydescribed in Section 2.5.5. for MSR, including the same 50 mM Tris-HCl pH 7.5 bufferconditions.5.5.4 Pre-steady state kinetic analysisPre-steady state kinetic assays for the AaCPR variants were performed as previouslydescribed in Section 2.5.6.1155.5. Experimental ProceduresTable 5.3: The forward (F) and reverse (R) oligonucleotide sequences for W704 variants ofAaCPR. Mutations are highlighted in bold.CPR Variant Oligonucleotide SequenceW704S F 5’ CTACCTGCGTGATGTCTCGTAAGGATCCGGCTGC 3’R 5’ GCAGCCGGATCCTTACGAGACATCACGCAGGTAG 3’W704F F 5’ CTACCTGCGTGATGTCTTCTAAGGATCCGGCTGC 3’R 5’ GCAGCCGGATCCTTAGAAGACATCACGCCAGTAG 3’W704Y F 5’ CTACCTGCGTGATGTCTACTAAGGATCCGGCTGC 3’R 5’ CTACCTGCGTGATGTCTACTAAGGATCCGGCTGC 3’W704H F 5’ CTACCTGCGTGATGTCCACTAAGGATCCGGCTGC 3’R 5’ GCAGCCGGATCCTTAGTGGACATCACGCAGGTAG 3’116Chapter 6ConclusionMSR and CPR share considerable similarities in structure and function, including 41%amino acid similarity. Large-scale conformational motion allows these dynamic multidomainflavoenzymes to adopt a closed and open conformation to support intra- and intermolecularelectron transfer. Catalysis involves transfer of a hydride ion from the obligate two-electrondonor NADPH to the bound FAD isoalloxazine ring, followed by the consecutive transfer ofsingle electrons to the FMN isoalloxazine ring, and final one-electron delivery from FMN toan external acceptor. While these general characteristics are shared, their kinetic propertiesand coenzyme affinity differ. These dissimilarities spurred investigation into the inherentregulatory structures of MSR and CPR using site-directed mutagenesis. Evaluation of thesteady state kinetic properties of each generated variant was performed by steady-statekinetic analysis of cytochrome c reduction. Dead-end and product inhibition experimentsdetermined the relative substrate binding affinities. Stopped-flow spectroscopy was used toinvestigate the pre-steady state kinetic properties of the variants. Through these methods,the structural nuances that lead to the distinct catalytic properties of MSR and CPR wereuncovered.In MSR, a large aromatic residue Trp697 forms pi-pi stacking interactions with the FADisoalloxazine ring. Steric clash between Trp697 and the nicotinamide ring weakens thebinding of the coenzyme.65 Swapping the tryptophan indole ring for a smaller aromaticside chain improved the rate of cytochrome c reduction and accelerated the pre-steady staterate of NADPH-dependent reduction of FAD. An increase in the KIE from 1.7 in nativeMSR to 2.0 in W697Y and W697F variants suggest that isotopically insensitive steps, suchas conformational movement of the aromatic residue, becomes less rate-determining in thevariants. Reducing the size of the aromatic side chain weakens the pi-pi stacking interactionwith the isoalloxazine ring and presumably lowers the energetic cost for displacement of theside chain, which, in turn, accelerates hydride transfer. These combined data suggest thatdisplacement of Trp697 by the nicotinamide ring partially gates hydride transfer in MSR.NADPH initially binds to a polar pocket of MSR through the 2’,5’-ADP half of themolecule. The majority of the favourable coenzyme binding energy arises from interactionsbetween the 2’-phosphate group and conserved active site residues. The nicotinamide ring,however, does not contribute favourably to the binding energy because of steric clash be-117Chapter 6. Conclusiontween the nicotinamide ring and Trp697.65 As expected, aromatic substitution of Trp697 didnot affect the 2’,5’-ADP binding affinity. However, the substitutions did increase the bind-ing affinity for both NADPH and NADH such that the variants elicited a lower substratepreference for NADPH over NADH compared to the native enzyme. Presumably, reducedsteric clash between the nicotinamide ring and the smaller side chains in the variants madethe binding of this portion of the nicotinamide ring more favourable, which reduced the dif-ferences in binding affinities between the phosphorylated and unphosphorylated coenzyme.Thus, Trp697 contributes to the strong coenzyme preference observed in MSR.In addition to Trp697, other residues in the FAD/NADPH binding domain have beenfound to influence the catalytic properties of the enzyme. Weaker coenzyme binding wasobserved in the A312H variant of MSR. A312 is located within hydrogen bond distance of thecatalytic triad that surrounds the FAD isoalloxazine ring. In human CPR, the correspondingresidue is a histidine (His322). The histidine residue was proposed to weaken coenzymeassociation by competing with the nicotinamide ring for hydrogen bond contact with thecatalytic triad residue Asp695 (MSR numbering). Flavin reduction was significantly reduced(33-fold slower rate) in the A312H variant. Weaker coenzyme binding affinity potentiallycontributes to the slow flavin reduction of A312H. Alternatively, given the location of theamino acid variation near the FAD isoalloxazine ring, the substitution may also affect theFAD redox potential.Mutagenesis studies targeting residues that form electrostatic interactions with the py-rophosphate and 2’phosphate group of NADPH generated variants that improve coenzymebinding affinity and accelerated FAD reduction. Not surprisingly, a substitution at the 2’-phosphate-binding pocket (A622K) elicited the greatest effect, as amino acid interactionswith this moiety provides the majority of the coenzyme binding energy. Given that thesesubstitutions were not near the FAD isoalloxazine ring, they were not expected to affectthe reduction potential of the cofactor. Thus, increases in flavin reduction are attributedto improved coenzyme binding affinity.In contrast to MSR, the FAD-shielding Trp676 of human CPR does not pose a thermo-dynamic barrier to coenzyme binding and hydride transfer. The binding of the nicotinamidering is not significantly disrupted by Trp676, as evident by the equivalent dissociation con-stants of 2’,5’-ADP and NADP+.64 As a consequence, reducing the size of the aromaticside chain does not dramatically enhance the coenzyme binding affinity or accelerate flavinreduction, as observed for MSR. CPR catalysis is not limited by Trp676 displacement. In-stead, Trp676 acts to displace the oxidized nicotinamide ring. This conclusion is supportedby stopped-flow data of Trp676 variants that show the stable formation of the NADP+-FADH2-FMN complex. Since the nicotinamide ring remains planar with the isoalloxazinering, interflavin electron transfer is hindered. The prolonged interaction between the FAD118Chapter 6. Conclusionhydroquinone and the oxidized nicotinamide ring also leads a shift towards the reverse re-action (hydride transfer to NADP+); this is evident by the inverse KIE of 0.9 measured forW676Y and W676F. Slow flavin reduction is also a product of a greater shift towards the re-verse reaction as well as impaired interflavin electron transfer. Thus, the large bulky indolering of Trp676 is more proficient than other aromatic residues at displacing the oxidizednicotinamide ring from the FAD active site.Parallel alignment of the nicotinamide ring against the FAD disrupts the hydrogen bondthat forms between the Asp674 and Ser457 residues of the catalytic triad (CPR number-ing). Asp674 repositions itself to interact with the carboxamide group of the nicotinamidering. This interaction is likely responsible for optimizing the orientation of the C4 atom ofNADPH for hydride transfer. It has been proposed that electron transfer from FADH2 toFMN requires the reformation of the hydrogen bond between Asp674 and Ser457, whichenables Asp674 to act as a general base and abstract a proton from the N5 of FADH2via Ser467.61,155 Previously, only the three residues of the catalytic triad (Asp674, Ser457,Cys629) were assigned to this hydrogen bond network, however in this work, a neighbouringhistidine residue that is within hydrogen bond distance to Asp674 was identified. Substitut-ing the histidine for an alanine revealed that the imidazole weakens coenzyme binding andpromotes interflavin electron transfer. Competition between the histidine and the nicoti-namide ring for hydrogen bonding with Asp674 likely weakens coenzyme affinity, therebyfavouring NADP+ release. Restoration of the hydrogen bond network provides a protonrelay pathway for the hydrogen atom of the N5 nitrogen of the isoalloxazine ring and in-teraction with Ser457 stabilizes the resulting FAD semiquinone, thus enhancing interflavinelectron transfer.NADP+ release is the rate-determining step in CPR catalysis. This conclusion is sup-ported by the K602AV603K variant, which effectively removes a direct salt-bridge interac-tion with the coenzyme 2’-phosphate group. Coenzyme binding affinity was dramaticallyweakened and a faster rate constant of flavin reduction was observed. Thus, promotingNADP+ release (by weak coenzyme binding or by bulky Trp676) is important in CPRcatalysis.The different catalytic role of the FAD-stacking tryptophan in MSR and CPR may arisefrom how the tryptophan is orientated with respect to the FAD in both enzymes. In bothsubstrate-free and substrate-bound MSR, the entire indole ring is positioned over the FAD,maximizing pi-pi interaction. For CPR, the only available substrate-free crystal structurehas an engineered disulfide bond between the FAD and FMN domains.57 In this structure,Trp676 adopts the same position as that observed in MSR, while in substrate-bound CPR,Trp676 is flipped by 180 degrees and rotated so that only the phenyl group is in van derWaals contact with the isoalloxazine ring. Therefore, a conformational change in Trp676 to119Chapter 6. Conclusiona presumably more stable state upon coenzyme binding is apparently more favourable inCPR than in MSR. Thus, it is possible that the partially stacked position of Trp676 in CPRweakens the flavin-indole contact such that it is easily displaced by the nicotinamide ring.Moreover, previous studies on the conformational dynamics of CPR have shown that bindingof 2’,5’-ADP and NADP+ confer a shift towards the closed conformation, which shortensthe interflavin distance. In the presence of 2’,5’-ADP alone, the rate constant associatedwith interflavin electron transfer is enhanced.73,75 Recent studies on the conformationalbehavior of MSR indicate that it exists primarily in the open state, and the conformationaldistribution is not influenced by the presence of 2’,5’-ADP.156,157 However, in the presenceof NADP+, the conformational equilibria is shifted to the closed state. Thus, I propose thatsubtle interactions at the 2’,5’-ADP binding site confer conformational changes to Trp676to facilitate displacement by the nicotinamide ring in CPR, but not in MSR.The structural nuances involved in regulating the differential catalytic performance ofMSR and CPR have been investigated in this work. Examining these regulatory structurescontributes to the overall understanding of diflavin oxidoreductase catalysis and provide astepping stone for future work on these flavoenzymes. Through mutagenesis studies, I havesuccessfully enhanced MSR catalysis by reducing the size of the FAD-shielding aromaticside chain. In an effort to improve interflavin electron communication as well as FAD re-duction, a combination of the W697Y and the A312H mutation would be of interest. ForCPR, the presence of the Trp676 and His322 appears to be necessary for rapid flavin ki-netics with NADPH. Instead, further mutations aimed to weaken coenzyme binding at thecoenzyme-binding cleft may prove to be effective in accelerating flavin reduction. In fact,the double mutant K602AV603K did elicit much faster flavin reduction, however the overallrate of cytochrome c reduction remained similar to native CPR. This may be a consequenceof disruption to the subtle conformational changes that take place upon 2’,5’-ADP bindingthat influence the conformational dynamics of CPR. Thus, investigation into the partici-pating residues and the mechanism of this relationship would be of interest. Moreover, theaspartic acid loop that extends into the coenzyme-binding site ought to be analyzed for itspotential role in regulating coenzyme binding through structural rearrangement. Furtheranalysis of single-amino acid mutations in the isolated FAD/NADPH domain in comparisonwith full-length MSR and CPR would aid in pinpointing the effect elicited from local ver-sus large-scale structural changes. If the intention is to engineer a NADH-dependent MSRor CPR, then reducing the size of the FAD-shielding aromatic side chain consistently re-duced the preference for NADPH over NADH.132 Use of NADH is favourable for engineeredsystems because it is relatively inexpensive compared to the phosphorylated counterpart.Smaller aromatic substitutions coupled with mutations at the coenzyme-binding site aimed120Chapter 6. Conclusionto weaken coenzyme affinity, may enhance the overall catalytic performance of CPR or MSRwith NADH.Lastly, there is some ambiguity in the charge-transfer complex assignment among di-flavin oxidoreductases. In AaCPR, appropriate identification of the charge-transfer complexthat is formed over the reductive half-reaction is important in understanding the catalyticbehavior of this plant CPR species. In human CPR, substitution of Trp676 for a smalleraromatic results in the persistent stacking of the oxidized nicotinamide ring over the FADisoalloxazine ring, thereby impeding further reduction. In contrast, reduction of AaCPR ismuch more rapid and occurs with the formation of a charge-transfer complex. With muta-tions that mirror those of human CPR, the charge-transfer complex still forms and does notappear to impede further reduction in the same way as human CPR. Thus, the ability toidentify which charge-transfer species are formed or favoured in AaCPR compared to humanCPR would be paramount to characterizing the catalytic behavior of these enzymes. Thisproblem has been explored in FNR, and studies have shown some success at distinguishingcharge-transfer complexes over longer wavelengths (600 - 1000 nm).152,158,159In the future this research may contribute to improving MSR activity in MS deficientindividuals through gene therapy. A small study where patient cell lines were transfectedwith wild-type MTRR minigene constructs in vitro resulted in a significant enhancementin MS activity, marked by methionine production.160 Although this may seem far-fetchedat the moment, it may prove to be an attractive treatment option for early neonatal treat-ment or for individuals that do not benefit from the available therapies today.161,162 Cer-tainly, the co-expression of mammalian CPR in cytochrome P450 gene-directed enzymetherapy has already been employed in preclinical trials to support cytochrome P450 ac-tivity on anticancer prodrugs within target tumor cells.163–165 Mammalian CPR has alsobeen co-expressed with cytochrome P450s for detoxification of water and soil in transgenicplants, however the system is limited by CPR activity.166 Cytochrome P450s are also widelyemployed as biocatalysts for industrial synthesis of numerous chemicals, food ingredients,and pharmaceuticals.167,168 Thus, strategies to maximize the catalytic efficiency of the cy-tochrome P450-CPR system are desired to further develop these biocatalysts for commercialand medicinal use. This research thesis provides the structural platform on which to mod-ulate the catalytic performance of MSR and CPR.121Bibliography1. Colibus, L. D., and Mattevi, A. (2006) New Frontiers in Structural Flavoenzymology.Current Opinion in Structural Biology 16, 722–728.2. Fagan, R. L., and Palfey, B. A. In Flavin-Dependent Enzymes; Begley, T. P., Ed.;Comprehensive Natural Products II: Volume 7; Elsevier, 2010; pp 37–114.3. Ezkurdia, I., Juan, D., Rodriguez, J. M., Frankish, A., Diekhans, M., Harrow, J.,Vazquez, J., Valencia, A., and Tress, M. L. (2014) Multiple Evidence Strands SuggestThat There May be as Few as 19 000 Human Protein-Coding Genes. HumanMolecular Genetics 23, 5866–78.4. Lienhart, W. D., Gudipati, V., and Macheroux, P. (2013) The HumanFlavoproteome. Archives of Biochemistry and Biophysics 535, 150–162.5. Macheroux, P., Kappes, B., and Ealick, S. E. (2011) Flavogenomics–a Genomic andStructural View of Flavin-Dependent Proteins. Federation of European BiochemicalSocieties Journal 278, 2625–2634.6. Tomiki, T., and Saitou, N. (2004) Phylogenetic Analysis of Proteins Associated inthe Four Major Energy Metabolism Systems: Photosynthesis, Aerobic Respiration,Denitrification, and Sulfur Respiration. Journal of Molecular Evolution 59, 158–176.7. Jorns, M. S., Wang, B., and Jordan, S. (1987) DNA Repair Catalyzed by Escherichiacoli DNA Photolyase Containing Only Reduced Flavin: Elimination of the Enzyme’sSecond chromophore by Reduction with Sodium Borohydride. Biochemistry 26,6810–6816.8. Sancar, A. (2004) Regulation of the Mammalian Circadian Clock by Cryptochrome.Journal of Biological Chemistry 279, 34079–34082.9. van Berkel, W. J. H., and Joosten, W. (2007) Flavoenzymes. Current Opinion inChemical Biology 11, 195–202.10. Gropper, S., and Smith, J. L. Advanced Nutrition and Human Metabolism, sixth ed.;Wadsworth Publishing: Belmont, California, 2012.122Chapter 6. Bibliography11. Blyth, A. W. A Manual of Practical Chemistry: The Analysis of Foods and theDetection of Poisons; Charles Griffin and Company, 1879.12. Karrer, P., Salomon, H., Schopp, K., Benz, F., and Becker, B. (1935) Synthesen VonDrei Weiteren Stereoisomeren Des Lactoflavins (German)[Syntheses of ThreeStereoisomers of Lactoflavins]. Helvetica Chimica Acta 18, 908–910.13. Kuhn, R., Reinemund, K., and Weygand, F. (1934) Synthese des Lumi-Lactoflavins(German)[Synthesis of Lumi-lactoflavins]. Berichte der Deutschen ChemischenGesellschaft (A and B Series) 67, 1460–1462.14. Massey, V. (2000) The Chemical and Biological Versatility of Riboflavin. BiochemicalSociety Transactions 28, 283–296.15. Warburg, O., and Christian, W. (1933) Uber Das Gelbe Ferment und SeineWirkungen (German) [About the Yellow Enzyme and its Effects]. BiochemischeZeitschrift 266, 377–411.16. Theorell, H. (1935) Preparation in Pure State of the Effect Group of YellowEnzymes. Biochemische Zeitschrift 275, 344–346.17. Warburg, O., and Christian, W. (1938) Isolierung Der Prosthetischen Gruppe DerD-Aminosaureoxydase (German)[Isolation of the Prosthetic Group ofD-Aminosaureoxydase]. Biochemische Zeitschrift 298, 81.18. Powers, H. J. (2003) Riboflavin (Vitamin B2) and Health. The American Journal ofClinical Nutrition 77, 1352–1360.19. Koser, S. A. Vitamin Requirements of Bacteria and Yeasts.; Charles C. Thomas:Springfield, IL, 1968.20. Bender, D. A. Nutritional Biochemistry of the Vitamins, second edition ed.; PressSyndicate of the University of Cambridge: Cambridge, United Kingdom, 2003.21. Said, H. M., and Mohammed, Z. M. (2006) Intestinal Absorption of Water-SolubleVitamins: An Update. Current Opinion in Gastroenterology 22, 140–146.22. Moriyama, Y. (2011) Riboflavin Transporter is Finally Identified. Journal ofBiochemistry 150, 341–343.23. White, H., and Merrill, A. (1988) Riboflavin-Binding Proteins. Annual Review ofNutrition 8, 279–299.123Chapter 6. Bibliography24. Miura, R. (2001) Versatility and Specificity in Flavoenzymes: Control Mechanisms ofFlavin Reactivity. Chemical Record (New York, N.Y.) 1, 183–194.25. Muller, F., Grande, H. J., Harding, L. J., Dunham, W. R., Visser, A. J.,Reinders, J. H., Hemmerich, P., and Ehrenberg, A. (1981) On the Molecular andSubmolecular Structure of the Semiquinone Cations of Alloxazines and Isoalloxazinesas Revealed by Electron-Paramagnetic-Resonance Spectroscopy. European Journal ofBiochemistry 116, 17–25.26. Hefti, M. H., Vervoort, J., and van Berkel, W. J. H. (2003) Deflavination andReconstitution of Flavoproteins. Federation of European Biochemical SocietiesJournal 270, 4227–4242.27. Mewies, M., McIntire, W. S., and Scrutton, N. S. (1998) Covalent Attachment ofFlavin Adenine Dinucleotide (FAD) and Flavin Mononucleotide (FMN) to Enzymes:The Current State of Affairs. Protein Science 7, 7–20.28. Thorpe, C., and Kim, J. J. (1995) Structure and Mechanism of Action of theAcyl-CoA Dehydrogenases. Federation of American Societies for ExperimentalBiochemistry Journal 9, 718–725.29. Tishkov, V. I., and Khoronenkova, S. V. (2005) D-Amino Acid Oxidase: Structure,Catalytic Mechanism, and Practical Application. Biochemistry 70, 40–54.30. Khoronenkova, S., and Tishkov, V. (2008) D-Amino Acid Oxidase: PhysiologicalRole and Applications. Biochemistry (Mosc) 73, 1511–1518.31. Ghisla, S., Pollegioni, L., and Molla, G. (2011) Revisitation of the βCl-EliminationReaction of D-Amino Acid Oxidase: New Interpretation of the Reaction ThatSparked Flavoprotein Dehydrogenation Mechanisms. Journal of Biological Chemistry286, 40987–40998.32. Entsch, B., Cole, L. J., and Ballou, D. P. (2005) Protein Dynamics and Electrostaticsin the Function of p-Hydroxybenzoate Hydroxylase. Archives of Biochemistry andBiophysics 433, 297–311.33. Gatti, D., Palfey, B., Lah, M., Entsch, B., Massey, V., Ballou, D., and Ludwig, M.(1994) The Mobile Flavin of 4-OH Benzoate Hydroxylase. Science 266, 110–114.34. Carell, T., Burgdorf, L. T., Kundu, L. M., and Cichon, M. (2001) The Mechanism ofAction of DNA Photolyases. Current Opinion in Chemical Biology 5, 491–498.124Chapter 6. Bibliography35. Lucas-Lledo, J. I., and Lynch, M. (2009) Evolution of Mutation Rates: PhylogenomicAnalysis of the Photolyase/Cryptochrome Family. Journal of Biology and Evolution26, 1143–1153.36. Park, H., Kim, S., Sancar, A., and Deisenhofer, J. (1995) Crystal Structure of DNAPhotolyase From Escherichia coli. Science 268, 1866–1872.37. Weber, S. (2005) Light-Driven Enzymatic Catalysis of DNA Repair: A Review ofRecent Biophysical Studies on Photolyase. Biochimica et Biophysica Acta -Bioenergetics 1707, 1–23.38. Watson, J. A., McTamney, P. M., Adler, J. M., and Rokita, S. E. (2008)Flavoprotein Iodotyrosine Deiodinase Functions Without Cysteine Residues.Chembiochem 9, 504–506.39. Stanbury, J. B. (1960) Deiodination of the Iodinated Amino Acids. Annals of theNew York Academy of Sciences 86, 417–439.40. Bianco, A. C., Salvatore, D., Gereben, B., Berry, M. J., and Larsen, P. R. (2002)Biochemistry, Cellular and Molecular Biology, and Physiological Roles of theIodothyronine Selenodeiodinases. Endocrine Reviews 23, 38–89.41. Hu, J., Chuenchor, W., and Rokita, S. E. (2014) A Switch Between One- andTwo-Electron Chemistry of the Human Flavoprotein Iodotyrosine Deiodinase isControlled by Substrate. Journal of Biological Chemistry42. Avron, M., and Jagendorf, A. T. (1956) A TPNH Diaphorase from Chloroplasts.Archives of Biochemistry and Biophysics 65, 475–490.43. Arnon, D. I. (1965) Ferredoxin and Photosynthesis. Science 149, 1460–1470.44. Karplus, P., Daniels, M., and Herriott, J. (1991) Atomic Structure ofFerredoxin-NADP+ Reductase: Prototype for a Structurally Novel FlavoenzymeFamily. Science 251, 60–66.45. Correll, C. C., Ludwig, M. L., Bruns, C. M., and Karplus, P. A. (1993) StructuralPrototypes for an Extended Family of Flavoprotein Reductases: Comparison ofPhthalate Dioxygenase Reductase with Ferredoxin Reductase and Ferredoxin.Protein Science 2, 2112–2133.46. Aliverti, A., Pandini, V., Pennati, A., de Rosa, M., and Zanetti, G. (2008) Structuraland Functional Diversity of Ferredoxin-NADP+ Reductases. Archives ofBiochemistry and Biophysics 474, 283–291.125Chapter 6. Bibliography47. Zeghouf, M., Fontecave, M., Macherel, D., and Coves, J. (1998) The FlavoproteinComponent of the Escherichia coli Sulfite Reductase: Expression, Purification, andSpectral and Catalytic Properties of a Monomeric Form Containing Both the FlavinAdenine Dinucleotide and the Flavin Mononucleotide Cofactors. Biochemistry 37,6114–6123.48. Narhi, L. O., and Fulco, A. J. (1987) Identification and Characterization of TwoFunctional Domains in Cytochrome P450BM3, a Catalytically Self-SufficientMonooxygenase Induced by Barbiturates in Bacillus megaterium. Journal ofBiological Chemistry 262, 6683–6690.49. Wang, M., Roberts, D. L., Paschke, R., Shea, T. M., Masters, B. S., and Kim, J. J.(1997) Three-Dimensional Structure of NADPH-Cytochrome P450 Reductase:Prototype for FMN- and FAD-Containing Enzymes. Proceedings of the NationalAcademy of Sciences of the United States of America 94, 8411–8416.50. Masters, B. S., McMillan, K., Sheta, E. A., Nishimura, J. S., Roman, L. J., andMartasek, P. (1996) Neuronal Nitric Oxide Synthase, a Modular Enzyme Formed byConvergent Evolution: Structure Studies of a Cysteine Thiolate-Liganded HemeProtein That Hydroxylates L-Arginine to Produce NO as a Cellular Signal.Federation of American Societies for Experimental Biology Journal 10, 552–558.51. Stuehr, D. J. (1997) Structure-Function Aspects in the Nitric Oxide Synthases.Annual Review of Pharmacology and Toxicology 37, 339–359.52. Olteanu, H., and Banerjee, R. (2001) Human Methionine Synthase Reductase, aSoluble P450 Reductase-like Dual Flavoprotein, is Sufficient for NADPH-DependentMethionine Synthase Activation. Journal of Biological Chemistry 276, 35558–35563.53. Paine, M. J. I., Garner, A. P., Powell, D., Sibbald, J., Sales, M., Pratt, N., Smith, T.,Tew, D. G., and Wolf, C. R. (2000) Cloning and Characterization of a Novel HumanDual Flavin Reductase. Journal of Biological Chemistry 275, 1471–1478.54. Porter, T. D., and Kasper, C. B. (1986) NADPH-Cytochrome P450 Oxidoreductase:Flavin Mononucleotide and Flavin Adenine Dinucleotide Domains Evolved FromDifferent Flavoproteins. Biochemistry 25, 1682–1687.55. Phillips, A. H., and Langdon, R. G. (1962) Hepatic TriphosphopyridineNucleotide-Cytochrome c Reductase: Isolation, Characterization, and KineticStudies. Journal of Biological Chemistry 237, 2652–2660.126Chapter 6. Bibliography56. Zhao, Q., Modi, S., Smith, G., Paine, M., Mcdonagh, P. D., Wolf, C. R., Tew, D.,Lian, L.-Y., Roberts, G. C. K., and Driessen, H. P. C. (1999) Crystal Structure ofthe FMN-Binding Domain of Human Cytochrome P450 Reductase at 1.93 AngstromResolution. Protein Science 8, 298–306.57. Xia, C., Hamdane, D., Shen, A. L., Choi, V., Kasper, C. B., Pearl, N. M., Zhang, H.,Im, S. C., Waskell, L., and Kim, J. J. (2011) Conformational Changes ofNADPH-Cytochrome P450 Oxidoreductase are Essential for Catalysis and CofactorBinding. Journal of Biological Chemistry 286, 16246–16260.58. Andersen, J. F., Utermohlen, J. G., and Feyereisen, R. (1994) Expression of HouseflyCYP6A1 and NADPH-Cytochrome P450 Reductase in Escherichia coli andReconstitution of an Insecticide-Metabolizing P450 System. Biochemistry 33,2171–2177.59. Hamdane, D., Xia, C., Im, S. C., Zhang, H., Kim, J. J., and Waskell, L. (2009)Structure and Function of an NADPH-Cytochrome P450 Oxidoreductase in an OpenConformation Capable of Reducing Cytochrome P450. Journal of BiologicalChemistry 284, 11374–11384.60. Shen, A. L., Sem, D. S., and Kasper, C. B. (1999) Mechanistic Studies on theReductive Half-Reaction of NADPH-Cytochrome P450 Oxidoreductase. Journal ofBiological Chemistry 274, 5391–5398.61. Hubbard, P. A., Shen, A. L., Paschke, R., Kasper, C. B., and Kim, J.-J. P. (2001)NADPH-Cytochrome P450 Oxidoreductase: Structural Basis for Hydride andElectron Transfer. Journal of Biological Chemistry 276, 29163–29170.62. Deng, Z., Aliverti, A., Zanetti, G., Arakaki, A. K., Ottado, J., Orellano, E. G.,Calcaterra, N. B., Ceccarelli, E. A., Carrillo, N., and Karplus, P. A. (1999) AProductive NADP+ Binding Mode of Rerredoxin-NADP+ Reductase Revealed byProtein Engineering and Crystallographic Studies. Nature Structural Biology 6,847–853.63. Piubelli, L., Aliverti, A., Arakaki, A. K., Carrillo, N., Ceccarelli, E. A.,Karplus, P. A., and Zanetti, G. (2000) Competition Between C-Terminal Tyrosineand Nicotinamide Modulates Pyridine Nucleotide Affinity and Specificity in PlantFerredoxin-NADP+ Reductase. Journal of Biological Chemistry 275, 10472–10476.64. Grunau, A., Paine, M. J., Ladbury, J. E., and Gutierrez, A. (2006) Global Effects ofthe Energetics of Coenzyme Binding:NADPH Controls the Protein InteractionProperties of Human Cytochrome P450 Reductase. Biochemistry 45, 1421–1434.127Chapter 6. Bibliography65. Wolthers, K. R., Lou, X., Toogood, H. S., Leys, D., and Scrutton, N. S. (2007)Mechanism of Coenzyme Binding to Human Methionine Synthase ReductaseRevealed Through the Crystal Structure of the FNR-like Module and IsothermalTitration Calorimetry. Biochemistry 46, 11833–11844.66. Munro, A. W., Noble, M. A., Robledo, L., Daff, S. N., and Chapman, S. K. (2001)Determination of the Redox Properties of Human NADPH-Cytochrome P450Reductase. Biochemistry 40, 1956–1963.67. Gutierrez, A., Lian, L.-Y., Wolf, C. R., Scrutton, N. S., and Roberts, G. C. K. (2001)Stopped-Flow Kinetic Studies of Flavin Reduction in Human Cytochrome P450Reductase and its Component Domains. Biochemistry 40, 1964–1975.68. Oprian, D. D., and Coon, M. J. (1982) Oxidation-Reduction States of FMN andFAD in NADPH-Cytochrome P450 Reductase During Reduction by NADPH.Journal of Biological Chemistry 257, 8935–8944.69. Iyanagi, T., Makino, N., and Mason, H. (1974) Redox Properties of the ReducedNicotinamide Adenine Dinucleotide Phosphate-Cytochrome P450 and ReducedNicotinamide Adenine Dinucleotide-Cytochrome b5 Reductases. Biochemistry 13,1701–1710.70. Sadeghi, S. J., Fantuzzi, A., and Gilardi, G. (2011) Breakthrough in P450Bioelectrochemistry and Future Perspectives. Biochimica et Biophysica Acta 1814,237–248.71. Das, A., Grinkova, Y. V., and Sligar, S. G. (2007) Redox Potential Control by DrugBinding to Cytochrome P450 3A4. Journal of the American Chemical Society 129,13778–13779.72. Das, A., and Sligar, S. G. (2009) Modulation of the Cytochrome P450 ReductaseRedox Potential by the Phospholipid Bilayer. Biochemistry 48, 12104–12112.73. Ellis, J., Gutierrez, A., Barsukov, I. L., Huang, W.-C., Grossmann, J. G., andRoberts, G. C. K. (2009) Domain Motion in Cytochrome P450 Reductase:Conformation Equilibria Revealed by NMR and Small-Angle X-Ray Scattering.Journal of Biological Chemistry 284, 36628–36637.74. Vincent, B., Morellet, N., Fatemi, F., Aigrain, L., Truan, G., Guittet, E., andLescop, E. (2012) The Closed and Compact Domain Organization of the 70-kDaHuman Cytochrome P450 Reductase in its Oxidized State as Revealed by NMR.Journal of Molecular Biology 420, 296–309.128Chapter 6. Bibliography75. Pudney, C., Khara, B., Johannissen, L., and Scrutton, N. (2011) Coupled MotionsDirect Electrons Along Human Microsomal P450 Chains. Public Library of ScienceBiology 9 .76. Huang, W., Ellis, J., Moody, P., Raven, E., and Roberts, G. (2013) Redox-LinkedDomain Movements in the Catalytic Cycle of Cytochrome P450 Reductase. Structure21, 1581–1589.77. Enoch, H. G., and Strittmatter, P. (1979) Cytochrome b5 Reduction byNADPH-Cytochrome P450 Reductase. Journal of Biological Chemistry 254,8976–8981.78. Ono, T., and Bloch, K. (1975) Solubilization and Partial Characterization of RatLiver Squalene Epoxidase. Journal of Biological Chemistry 250, 1571–1579.79. Schacter, B. A., Nelson, E. B., Marver, H. S., and Masters, B. S. (1972)Immunochemical Evidence for an Association of Heme Oxygenase With theMicrosomal Electron Transport System. Journal of Biological Chemistry 247,3601–3607.80. Omura, T., and Sato, R. (1962) A New Cytochrome in Liver Microsomes. Journal ofBiological Chemistry 237, PC1375–PC1376.81. Strobel, H., Hodgson, A., and Shen, S. Cytochrome P450: Structure, Mechanism andBiochemistry, second edition ed.; Plenum Press: New York, 1995.82. Nelson, D. R., Schuler, M. A., Paquette, S. M., Werck-Reichhart, D., and Bak, S.(2004) Comparative Genomics of Rice and Arabidopsis. Analysis of 727 CytochromeP450 genes and Pseudogenes from a Monocot and a Dicot. Plant Physiology 135,756–772.83. Wilkinson, G. R. (2005) Drug Metabolism and Variability Among Patients in DrugResponse. New England Journal of Medicine 352, 2211–2221.84. Meunier, B., Visser, S. P. D., and Shaik, S. (2004) Mechanism of Oxidation ReactionsCatalyzed by Cytochrome P450 Enzymes. Chemical Reviews 104, 3947–3980.85. Sono, M., Roach, M. P., Coulter, E. D., and Dawson, J. H. (1996) Heme-ContainingOxygenases. Chemical Reviews 96, 2841–2888.86. Guengerich, F. P. (1992) Characterization of Human Cytochrome P450 Enzymes.Federation of American Societies for Experimental Biology Journal 6, 745–748.129Chapter 6. Bibliography87. De Montellano, P. R. O. Cytochrome P450: Structure, Mechanism, andBiochemistry ; Springer, 2005.88. Groves, J. T., and McClusky, G. A. (1976) Aliphatic Hydroxylation Via OxygenRebound. Oxygen Transfer Catalyzed by Iron. Journal of the American ChemicalSociety 98, 859–861.89. Hasemann, C. A., Kurumbail, R. G., Boddupalli, S. S., Peterson, J. A., andDeisenhofer, J. (1995) Structure and Function of Cytochromes P450:A ComparativeAnalysis of Three Crystal Structures. Structure 3, 41–62.90. Williams, P. A., Cosme, J., Sridhar, V., Johnson, E. F., and McRee, D. E. (2000)Mammalian Microsomal Cytochrome P450 Monooxygenase: Structural Adaptationsfor Membrane Binding and Functional Diversity. Molecular Cell 5, 121–131.91. Nelson, D. R. (2009) The Cytochrome P450 Homepage. Human Genomics 4, 59–65.92. Mao, G., Seebeck, T., Schrenker, D., and Yu, O. (2013) CYP709B3, a CytochromeP450 Monooxygenase Gene Involved in Salt Tolerance in Arabidopsis thaliana.BioMedCentral Plant Biology 13, 169.93. Anzenbacher, P., and Anzenbacherova, E. (2001) Cytochromes P450 and Metabolismof Xenobiotics. Cellular and Molecular Life Sciences 58, 737–747.94. Scott, R. R., and Miller, W. L. (2008) Genetic and Clinical Features of P450Oxidoreductase Deficiency. Hormone Research in Pediatrics 69, 266–275.95. Pandey, A. V., and Fluck, C. E. (2013) NADPH P450 Oxidoreductase: Structure,Function, and Pathology of Diseases. Pharmacology & Therapeutics 138, 229–254.96. Zhang, X., Li, L., Ding, X., and Kaminsky, L. S. (2011) Identification of CytochromeP450 Oxidoreductase Gene Variants That are Significantly Associated with theInterindividual Variations in Warfarin Maintenance Dose. Drug Metabolism andDisposition 39, 1433–1439.97. Hart, S. N., Wang, S., Nakamoto, K., Wesselman, C., Li, Y., and Zhong, X. (2008)Genetic Polymorphisms in Cytochrome P450 Oxidoreductase Influence MicrosomalP450-Catalyzed Drug Metabolism. Pharmacogenetics and Genomics 18, 11–24.98. Huang, N., Pandey, A. V., Reardon, W., Lapunzina, P. D., Mowat, D., Jabs, E. W.,Vliet, G. V., Sack, J., and Fluck, C. E. (2005) Diversity and Function of Mutationsin P450 Oxidoreductase in Patients with Antley-Bixler Syndrome and DisorderedSteroidogenesis. The American Journal of Human Genetics 76, 729–749.130Chapter 6. Bibliography99. Miller, W. L. (2004) P450 Oxidoreductase Deficiency: A New Disorder ofSteroidogenesis With Multiple Clinical Manifestations. Trends in Endocrinology andMetabolism 15, 311–315.100. Henderson, C. J., Otto, D. M., Carrie, D., Magnuson, M. A., McLaren, A. W.,Rosewell, I., and Wolf, C. R. (2003) Inactivation of the Hepatic Cytochrome P450System by Conditional Deletion of Hepatic Cytochrome P450 Reductase. Journal ofBiological Chemistry 278, 13480–13486.101. Shen, A. L., O’Leary, K. A., and Kasper, C. B. (2002) Association of MultipleDevelopmental Defects and Embryonic Lethality with Loss of MicrosomalNADPH-Cytochrome P450 Oxidoreductase. Journal of Biological Chemistry 277,6536–6541.102. Gu, J., Weng, Y., Zhang, Q.-Y., Cui, H., Behr, M., Wu, L., Yang, W., Zhang, L., andDing, X. (2003) Liver-Specific Deletion of the NADPH-Cytochrome P450 ReductaseGene Impact on Plasma Cholesterol Homeostasis and the Function and Regulation ofMicrosomal Cytochrome P450 and Heme Oxygenase. Journal of Biological Chemistry278, 25895–25901.103. Leclerc, D., Wilson, A., Dumas, R., Gafuik, C., Song, D., Watkins, D., Heng, H. H.,Rommens, J. M., Scherer, S. W., Rosenblatt, D. S., and Gravel, R. A. (1998) Cloningand Mapping of a cDNA for Methionine Synthase Reductase, a FlavoproteinDefective in Patients With Homocystinuria. Proceedings of the National Academy ofSciences of the United States of America 95, 3059–3064.104. Froese, D. S., Wu, X., Zhang, J., Dumas, R., Schoel, W. M., Amrein, M., andGravel, R. A. (2008) Restricted Role for Methionine Synthase Reductase Defined bySubcellular Localization. Molecular Genetics and Metabolism 94, 68–77.105. Wolthers, K. R., Basran, J., Munro, A. W., and Scrutton, N. S. (2003) MolecularDissection of Human Methionine Synthase Reductase: Determination of the FlavinRedox Potentials in Full-Length Enzyme and Isolated Flavin-Binding Domains.Biochemistry 42, 3911–3920.106. Jarrett, J. T., Hoover, D. M., Ludwig, M. L., and Matthews, R. G. (1998) TheMechanism of Adenosylmethionine-Dependent Activation of Methionine Synthase: ARapid Kinetic Analysis of Intermediates in Reductive Methylation of Cob(II)alaminEnzyme. Biochemistry 37, 12649–12658.107. Yamada, K., Gravel, R. A., Toraya, T., and Matthews, R. G. (2006) HumanMethionine Synthase Reductase is a Molecular Chaperone for Human Methionine131Chapter 6. BibliographySynthase. Proceedings of the National Academy of Sciences of the United States ofAmerica 103, 9476–9481.108. Banerjee, R. V., and Matthews, R. G. (1990) Cobalamin-Dependent MethionineSynthase. Federation of American Societies for Experimental Biology 4, 1450–1459.109. Fujimori, D. (2013) Radical SAM-Mediated Methylation Reactions. Current Opinionin Chemical Biology 17, 597–604.110. Boldyrev, A. A. (2009) Molecular Mechanisms of Homocysteine Toxicity.Biochemistry (Mosc) 74, 589–598.111. Evans, J. C., Huddler, D. P., Hilgers, M. T., Romanchuk, G., Matthews, R. G., andLudwig, M. L. (2004) Structures of the N-Terminal Modules Imply Large DomainMotions During Catalysis by Methionine Synthase. Proceedings of the NationalAcademy of Sciences of the United States of America 101, 3729–3736.112. Drennan, C. L., Huang, S., Drummond, J. T., Matthews, R. G., and Ludwig, M.(1994) How a Protein Binds B12: A 3.0 Angstrom X-Ray Structure of B12-BindingDomains of Methionine Synthase. Science 266, 1669–1674.113. Wolthers, K. R., and Scrutton, N. S. (2007) Protein Interactions in the HumanMethionine Synthase-Methionine Synthase Reductase Complex and Implications forthe Mechanism of Enzyme Reactivation. Biochemistry 46, 6696–6709.114. Olteanu, H., Munson, T., and Banerjee, R. (2002) Differences in the Efficiency ofReductive Activation of Methionine Synthase and Exogenous Electron AcceptorsBetween the Common Polymorphic Variants of Human Methionine SynthaseReductase. Biochemistry 41, 13378–13385.115. Olteanu, H., Wolthers, K. R., Munro, A. W., Scrutton, N. S., and Banerjee, R.(2004) Kinetic and Thermodynamic Characterization of the Common PolymorphicVariants of Human Methionine Synthase Reductase. Biochemistry 43, 1988–1997.116. van der Linden, I. J., den Heijer, M., Afman, L. A., Gellekink, H., Vermeulen, S. H.H. M., Kluijtmans, L. A. J., and Blom, H. J. (2006) The Methionine SynthaseReductase 66AG Polymorphism is a Maternal Risk Factor for Spina Bifida. Journalof Molecular Medicine 84, 1047–1054.117. Coppede, F., Bosco, P., Lorenzoni, V., Denaro, M., Anello, G., Antonucci, I.,Barone, C., Stuppia, L., Romano, C., and Migliore, L. (2014) The MTRR 66AGPolymorphism and Maternal Risk of Birth of a Child with Down Syndrome in132Chapter 6. BibliographyCaucasian Women: A CaseControl Study and a Meta-Analysis. Molecular BiologyReports 41, 5571–5583.118. Ouyang, S., Li, Y., Liu, Z., Chang, H., and Wu, J. (2013) Association Between MTRA2756G and MTRR A66G Polymorphisms and Maternal Risk for Neural TubeDefects: A Meta-Analysis. Gene 515, 308–312.119. Bosco, P., Gueant-Rodriguez, R.-M., Anello, G., Barone, C., Namour, F., Caraci, F.,Romano, A., Romano, C., and Gueant, J.-L. (2003) Methionine Synthase (MTR)(AG) Polymorphism, Double Heterozygosity Methionine Synthase AG MethionineSynthase Reductase (MTRR) 66 AG, and Elevated Homocysteinemia are Three RiskFactors for Having a Child with Down Syndrome. American Journal of MedicalGenetics Part A 121A, 219–224.120. Wilson, A., Platt, R., Wu, Q., Leclerc, D., Christensen, B., Yang, H., Gravel, R. A.,and Rozen, R. (1999) A Common Variant in Methionine Synthase ReductaseCombined with Low Cobalamin (Vitamin B12) Increases Risk for Spina Bifida.Molecular Genetics and Metabolism 67, 317–323.121. Watkins, D., and Rosenblatt, D. S. (1988) Genetic Heterogeneity Among Patientswith Methylcobalamin Deficiency. Definition of Two Complementation Groups, cblEand cblG. The Journal of Clinical Investigation 81, 1690–1694.122. Wilson, A., Leclerc, D., Rosenblatt, D. S., and Gravel, R. A. (1999) Molecular Basisfor Methionine Synthase Reductase Deficiency in Patients Belonging to the cblEComplementation Group of Disorders in Folate/Cobalamin Metabolism. HumanMolecular Genetics 8, 2009–2016.123. Elmore, C. L., Wu, X., Leclerc, D., Watson, E. D., Bottiglieri, T., Krupenko, N. I.,Krupenko, S. A., Cross, J. C., Rozen, R., Gravel, R. A., and Matthews, R. G. (2007)Metabolic Derangement of Methionine and Folate Metabolism in Mice Deficient inMethionine Synthase Reductase. Molecular genetics and metabolism 91, 85–97.124. Wald, D. S., Law, M., and Morris, J. K. (2002) Homocysteine and CardiovascularDisease: Evidence on Causality from a Meta-Analysis. British Medical Journal 325,1202.125. Castro, R., Rivera, I., Blom, H. J., Jakobs, C., and de Almeida, I. T. (2006)Homocysteine Metabolism, Hyperhomocysteinaemia and Vascular Disease: AnOverview. Journal of Inherited Metabolic Disease 29, 3–20.133Chapter 6. Bibliography126. van der Put, N. M., van Straaten, H. W. M., Trijbels, F. J. M., and Blom, H. J.(2001) Folate, Homocysteine and Neural Tube Defects: An Overview. ExperimentalBiology and Medicine 226, 243–270.127. Zhang, J., Martsek, P., Paschke, R., Shea, T., Masters, B. S. S., and Kim, J.-J. P.(2001) Crystal Structure of the FAD/NADPH-Binding Domain of Rat NeuronalNitric-Oxide Synthase: Comparisons With NADPH-cytochrome P450Oxidoreductase. Journal of Biological Chemistry 276, 37506–37513.128. Sem, D. S., and Kasper, C. B. (1992) Geometric Relationship Between theNicotinamide and Isoalloxazine Rings in NADPH-Cytochrome P450 Oxidoreductase:Implications for the Classification of Evolutionarily and Functionally RelatedFlavoproteins. Biochemistry 31, 3391–3398.129. Orellano, E. G., Calcaterra, N. B., Carrillo, N., and Ceccarelli, E. A. (1993) Probingthe Role of the Carboxyl-Terminal Region of Ferredoxin-NADP+ Reductase bySite-Directed Mutagenesis and Deletion Analysis. Journal of Biological Chemistry268, 19267–19273.130. Adak, S., Sharma, M., Meade, A. L., and Stuehr, D. J. (2002) A ConservedFlavin-Shielding Residue Regulates NO Synthase Electron Transfer andNicotinamide Coenzyme Specificity. Proceedings of the National Academy of Sciencesof the United States of America 99, 13516–13521.131. Konas, D. W., Zhu, K., Sharma, M., Aulak, K. S., Brudvig, G. W., and Stuehr, D. J.(2004) The FAD-Shielding Residue Phe1395 Regulates Neuronal Nitric OxideSynthase Catalysis by Controlling NADP+ Affinity and a ConformationalEquilibrium Within the Flavoprotein Domain. Journal of Biological Chemistry 279,35412–35425.132. Dohr, O., Paine, M. J. I., Friedberg, T., Roberts, G. C. K., and Wolf, C. R. (2001)Engineering of a Functional Human NADH-Dependent Cytochrome P450 System.Proceedings of the National Academy of Sciences of the United States of America 98,81–86.133. Wolthers, K. R., and Scrutton, N. S. (2004) Electron Transfer in Human MethionineSynthase Reductase Studied by Stopped-Flow Spectrophotometry. Biochemistry 43,490–500.134. Batie, C. J., and Kamin, H. (1986) Association of Ferredoxin-NADP+ Reductasewith NADP(H) Specificity and Oxidation-Reduction Properties. Journal of BiologicalChemistry 261, 11214–11223.134Chapter 6. Bibliography135. Murataliev, M. B., and Feyereisen, R. (2000) Interaction of NADP(H) With Oxidizedand Reduced P450 Reductase During Catalysis. Studies With Nucleotide Analogues.Biochemistry 39, 5066–5074.136. Meints, C. E., Gustafsson, F. S., Scrutton, N. S., and Wolthers, K. R. (2011)Tryptophan 697 Modulates Hydride and Interflavin Electron Transfer in HumanMethionine Synthase Reductase. Biochemistry 50, 11131–11142.137. Gutierrez, A., Doehr, O., Paine, M., Wolf, C. R., Scrutton, N. S., and Roberts, G.C. K. (2000) Trp-676 Facilitates Nicotinamide Coenzyme Exchange in the ReductiveHalf-Reaction of Human Cytochrome P450 Reductase:Properties of the SolubleW676H and W676A Mutant Reductases. Biochemistry 39, 15990–15999.138. Gutierrez, A., Paine, M., Wolf, C. R., Scrutton, N. S., and Roberts, G. C. K. (2002)Relaxation Kinetics of Cytochrome P450 Reductase:Internal Electron Transfer isLimited by Conformational Change and Regulated by Coenzyme Binding.Biochemistry 41, 4626–4637.139. Viola, R., Cook, P., and Cleland, W. (1979) Stereoselective Preparation ofDeuterated Reduced Nicotinamide Adenine Nucleotides and Substrates byEnzymatic Synthesis. Analytical Biochemistry 96, 334–340.140. Dutton, P. (1978) Redox Potentiometry: Determination of Midpoint Potentials ofOxidation-Reduction Components of Biological Electron Transfer Systems. Methodsin Enzymology 54, 411–435.141. Haque, M., Bayachou, M. A., Fadlalla, D. D., and Stuehr, D. J. (2013)Charge-Pairing Interactions Control the Conformational Setpoint and Motions of theFMN Domain in Neuronal Nitric Oxide Synthase. Biochemical Journal 450, 607–617.142. Jenner, M., Ellis, J., Huang, W.-C., LloydRaven, E., Roberts, G. C. K., andOldham, N. J. (2011) Detection of a Protein Conformational Equilibrium byElectrospray Ionisation-Ion Mobility-Mass Spectrometry. Angewandte ChemieInternational Edition 50, 8291–8294.143. Dumit, V. I., Essigke, T., Cortez, N., and Ullmann, G. M. (2010) MechanisticInsights Into FerredoxinNADP(H) Reductase Catalysis Involving the ConservedGlutamate in the Active Site. Journal of Molecular Biology 397, 814–825.144. Meints, C. E., Simtchouk, S., and Wolthers, K. R. (2013) Aromatic Substitution ofthe FAD-Shielding Tryptophan Reveals its Differential Role in Regulating Electron135Chapter 6. BibliographyFlux in Methionine Synthase Reductase and Cytochrome P450 Reductase.Federation of European Biological Societies Journal 280, 1460–1474.145. Prodromou, C., and Pearl, L. H. (1992) Recursive PCR: A Novel Technique ForTotal Gene Synthesis. Protein Engineering 5, 827–829.146. World Health Organization, World Malaria Report 2008 ; Nonserial PublicationSeries; World Health Organization, 2008.147. Zeng, Q., Qiu, F., and Yuan, L. (2008) Production of Artermisinin byGenetically-Modified Microbes. Biotechnology Letters 30, 581–592.148. Ro, D. K., Paradise, E. M., Ouellet, M., Fisher, K. J., Newman, K. L.,Ndungu, J. M., Ho, K. A., Eachus, R. A., Ham, T. S., Kirby, J., Chang, M. C.,Withers, S. T., Shiba, Y., Sarpong, R., and Keasling, J. D. (2006) Production of theAntimalarial Drug Precursor Artemisinic Acid in Engineered Yeast. Nature 440,940–943.149. Chang, M. C., Eachus, R. A., Trieu, W., Ro, D. K., and Keasling, J. D. (2007)Engineering Escherichia coli for Production of Functionalized Terpenoids usingPlant P450s. Nature Chemical Biology 3, 274–277.150. Simtchouk, S., Eng, J. L., Meints, C. E., Makins, C., and Wolthers, K. R. (2013)Kinetic Analysis of Cytochrome P450 Reductase from Artemisia annua RevealsAccelerated Rates of NADPH-Dependent Flavin Reduction. Federation of EuropeanBiochemical Societies Journal 280, 6627–6642.151. Sem, D. S., and Kasper, C. B. (1995) Effect of Ionic Strength on the KineticMechanism and Relative Rate Limitation of Steps in the Model NADPH-CytochromeP450 Oxidoreductase Reaction with Cytochrome C. Biochemistry 34, 12768–74.152. Peregrina, J., Sanchez-Azqueta, A., Herguedas, B., Martinez-Julvez, M., andMedina, M. (2010) Role of Specific Residues in Coenzyme Binding, Charge-TransferComplex Formation, and Catalysis in Anabaena Ferredoxin NADP+-Reductase.Biochimica et Biophysica Acta 1797, 1638–46.153. Lans, I., Medina, M., Rosta, E., Hummer, G., Garcia-Viloca, M., Lluch, J., andGonzalez-Lafont, A. (2012) Theoretical Study of the Mechanism of the HydrideTransfer Between Ferredoxin-NADP+ Reductase and NADP+: The Role of Tyr303.Journal of the American Chemical Society 134, 20544–53.136Chapter 6. Bibliography154. Lans, I., Peregrina, J., Medina, M., Garcia-Viloca, M., Gonzalez-Lafont, A., andLluch, J. (2010) Mechanism of the Hydride Transfer Between Anabaena Tyr303SerFNR(rd)/FNR(ox) and NADP+/H. A Combined Pre-Steady StateKinetic/Ensemble Averaged Transition-State Theory With MultidimensionalTunneling Study. Journal of Physical Chemistry 114, 3368–79.155. Meints, C. E., Parke, S. M., and Wolthers, K. R. (2014) Proximal FAD HistidineResidue Influences Interflavin Electron Transfer in Cytochrome P450 Reductase andMethionine Synthase Reductase. Archives of Biochemistry and Biophysics 547,18–26.156. Haque, M., Bayachou, M., Tejero, J., Kenney, C., Pearl, N., Im, S., Waskell, L., andStuehr, D. (2014) Distinct Conformational Behaviors of Four MammalianDual-Flavin Reductases (Cytochrome P450 Reductase, Methionine SynthaseReductase, Neuronal Nitric Oxide Synthase, Endothelial Nitric Oxide Synthase)Determine Their Unique Catalytic Profiles. Federation of European BiologicalSocieties Journal 281, 5325–5340.157. Rigby, S. E., Lou, X., Toogood, H. S., Wolthers, K. R., and Scrutton, N. S. (2011)ELDOR Spectroscopy Reveals That Energy Landscapes in Human MethionineSynthase Reductase are Extensively Remodelled Following Ligand and PartnerProtein Binding. Chembiochem 12, 863–867.158. Tejero, J., Martinez-Julvez, M., Mayoral, T., Luquita, A., Sanz-Aparicio, J.,Hermoso, J. A., Hurley, J. K., Tollin, G., Gmez-Moreno, C., and Medina, M. (2003)Involvement of the Pyrophosphate and the 2’-Phosphate Binding Regions ofFerredoxin-NADP+ Reductase in Coenzyme Specificity. Journal of BiologicalChemistry 278, 49203–14.159. Tejero, J., Perez-Dorado, I., Maya, C., Martinez-Julvez, M., Sanz-Aparicio, J.,Gomez-Moreno, C., Hermoso, J., and Medina, M. (2005) C-Terminal Tyrosine ofFerredoxin-NADP+ Reductase in Hydride Transfer Processes with NAD(P)+/H.Biochemistry 44, 13477–90.160. Zavadakova, P., Fowler, B., Suormala, T., Novotna, Z., Mueller, P., Hennermann, J.,Zeman, J., Vilaseca, M., Vilarinho, L., Gutsche, S., Wilichowski, E., Horneff, G., andKozich, V. (2005) cblE Type of Homocystinuria Due to Methionine SynthaseReductase Deficiency: Functional Correction by Minigene Expression. HumanMutation 25, 239–47.137Chapter 6. Bibliography161. Watkins, D., and Rosenblatt, D. S. (2011) Inborn Errors of Cobalamin Absorptionand Metabolism. American Journal of Medical Genetics Seminars in MedicalGenetics 157, 33–44.162. Waddington, S., Kennea, N., Buckley, S., Gregory, L., Themis, M., and Coutelle, C.(2004) Fetal and Neonatal Gene Therapy: Benefits and Pitfalls. Gene Therapy 11,S92–7.163. Jounaidi, Y., and Waxman, D. (2001) Frequent, Moderate-Dose CyclophosphamideAdministration Improves the Efficacy of Cytochrome P450/Cytochrome P450Reductase-Based Cancer Gene Therapy. Cancer Research 61, 4437–44.164. Chen, L., and Waxman, D. (2002) Cytochrome P450 Gene-Directed Enzyme ProdrugTherapy (GDEPT) for Cancer. Current Pharmaceutical Design 8, 1405–16.165. Roy, P., and Waxman, D. (2006) Activation of Oxazaphosphorines by CytochromeP450: Application to Gene-Directed Enzyme Prodrug Therapy For Cancer.Toxicology In Vitro 20, 176–86.166. Kumar, S., Jin, M., and Weemhoff, J. (2012) Cytochrome P450-MediatedPhytoremediation Using Transgenic Plants: A Need for Engineered CytochromeP450 Enzymes. Journal of Petroleum and Environmental Biotechnology 3, 1000127.167. Guengerich, F. (2002) Cytochrome P450 Enzymes in the Generation of CommercialProducts. Nature Reviews Drug Discovery 1, 359–66.168. Urlacher, V., Lutz-Wahl, S., and Schmid, R. (2004) Microbial P450 enzymes inBiotechnology. Applied Microbiology and Biotechnology 64, 317–25.138

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0074430/manifest

Comment

Related Items