ENHANCING EXPRESION OF RECOMBINANT HEMOPROTEINS: PROGRES TOWARD UNDERSTANDING STRUCTURE/FUNCTION AND THERAPEUTIC APLICATION Except where reference is made to the work of others, the work described in this disertation is my own or was done in collaboration with my advisory commite. This disertation does not include proprietary or clasified information. _________________________________ Cornelius Varnado Certificate of Approval: ___________________________ __________________________ Holly R. Elis Douglas C. Goodwin, Chair Asistant Profesor Asociate Profesor Chemistry and Biochemistry Chemistry and Biochemistry ___________________________ __________________________ James H. Hargis Thomas Albrecht-Schmit Profesor Asociate Profesor Chemistry and Biochemistry Chemistry and Biochemistry __________________________ Stephen L. McFarland Dean Graduate School ENHANCING EXPRESION OF RECOMBINANT HEMOPROTEINS: PROGRES TOWARD UNDERSTANDING STRUCTURE/FUNCTION AND THERAPEUTIC APLICATION Cornelius Varnado A Disertation Submited to the Graduate Faculty of Auburn University in Partial Fulfilment of the Requirements for the Degre of Doctor of Philosophy Auburn, Alabama August 7, 2006 ii ENHANCING EXPRESION OF RECOMBINANT HEMOPROTEINS: PROGRES TOWARD UNDERSTANDING STRUCTURE/FUNCTION AND THERAPEUTIC APLICATION Cornelius Varnado Permision is granted to Auburn University to make copies of this disertation at its discretion, upon request of individuals or institutions and at their expense. The author reserves al publication rights. _____________________________ Signature of Author _____________________________ Date of Graduation iv VITA Cornelius Varnado, son of Alice Winding, was born in McComb, Misisippi on December 6, 1976. He graduated from South Pike High School, Magnolia, Misisippi, in May of 1995. He then atended Alcorn State University in Alcorn State, Misisippi and graduated with a Bachelor of Science in Chemistry in May of 2000. In August 2000, he entered graduate school at Auburn University in the Department of Chemistry and Biochemistry for a Ph.D. degre. v DISERTATION ABSTRACT ENHANCING EXPRESION OF RECOMBINANT HEMOPROTEINS: PROGRES TOWARD UNDERSTANDING STRUCTURE/FUNCTION AND THERAPEUTIC APLICATION Cornelius Varnado Doctor of Philosophy, August 7, 2006 (B.S. Alcorn State University, Alcorn State, MS, May 2000) 209 Typed Pages Directed by Douglas C. Goodwin Nature has developed ways to control the esential need to activate oxygen. Hemoproteins perform this task using iron in a prosthetic group (e.g., iron- protoporphyrin-IX or heme). The environment surrounding the heme iron in hemoproteins dictates their activity and makes them a diverse group of enzymes. This makes hemoproteins good models for exploring the relationship betwen protein structure and function. The techniques necesary to pursue these studies are material-intensive. Commonly available systems for the expresion of recombinant hemoproteins in E. coli are able to produce large quantities of protein. However, in many cases the vast majority of the protein lacks heme and are therefore inactive. A hemoprotein expresion (HPEX) system was developed to resolve this problem. This system, based on the expresion of an outer membrane heme receptor, was tested on two diferent hemoproteins, myoglobin vi and catalase-peroxidase. In both cases, the system was succesful, demonstrating a dramatic increase in the heme content of these proteins. Additional studies were caried out on the catalase-peroxidases. The ability of catalase-peroxidases to catalyze catalase and peroxidase activities using one active site make them ideal for studying the protein structure/function relationship. Moreover, a periplasm-targeted subset of these enzymes have been implicated as virulence factors. The first full characterization of a periplasmic catalase-peroxidase was caried out on KatP, an enzyme from the highly virulent pathogen E. coli O157:H7. Absorption and EPR spectra indicated a high-spin heme enzyme dominated by the hexacoordinate high-spin complex. Apparent k cat values for catalase and peroxidase activities were higher for KatP than with other catalase- peroxidase. However, K M values were also higher for KatP. Feric KatP reacted with peracetic acid to form compound I and with CN ? to form a feri-cyano complex consistent with other catalase?peroxidases. Peroxynitrite also supported compound I formation in this enzyme. This catalytic capability, combined with eficient catalase and peroxidase activities, and a periplasmic location may be advantageous for meting an imune response that generates copious reactive oxygen and reactive nitrogen species. Crystal structures of catalase-peroxidases have revealed the presence of a thre amino acid covalent adduct (Trp-Tyr-Met). It is esential for catalase activity. An SDS-PAGE method to monitor the establishment of the crosslink was developed. Using this method it was demonstrated that crosslink formation requires a redox active porphyrin. Substitution of the Tyr residue with Phe prevent formation of the crosslink and generates a protein that has no catalase activity, enhanced peroxidase activity, and increased susceptibility to peroxide-dependent inactivation. vii ACKNOWLEDGEMENTS The research conducted in this disertation would not have been possible without the help and support of many people whom I wish to thank. First, I would like to thank my research advisor, Dr. Douglas C. Goodwin or ?Doc? as we caled him in the lab. Doc has been a mentor as wel as a friend. His patients and wilingnes to help guide me through my graduate carer has given me a great foundation to build a carer in science. I would like to thank my commite members, Drs. Holly Elis, Thomas Albrecht-Schmit, and James Howard Hargis for their constructive suggestions and al of their profesional asistances. Thanks to al of my lab mates at Auburn University, especialy Ruletha Baker, Yongjiang Li, Carma Cook, Robert Moore, Kristen Hertwig, Kimberley Laband, Robert Thomas, and J. Kenneth Roberts for al of the inteligent discussions, selfles help, and a pleasant lab environment. Special thanks to al of my family and friends for their continuous support, understanding, and encouragement. I especialy want to thank my mom, Alice, whose display of strength, courage and values have made me the person who I am today. Finaly, I would like to thank the Department of Chemistry and Biochemistry, Auburn University, the American Chemistry Society?s Petroleum Research Fund, and COSAM for their funding support for my research. I gratefully acknowledge the Southern Regional Education Board and Auburn University President?s Graduate Opportunity Program for their felowship support during my graduate studies. vii Style Manual Used: Biochemical and Biophysical Research Communications Computer Software Used: Microsoft Word, Microsoft Excel, Prism, ChemDrew, SwisPdb Viewer, MegaPov, Adobe Photoshop ix TABLE OF CONTENTS LIST OF TABLES.................................................... xii LIST OF FIGURES................................................... xii INTRODUCTION..................................................... 1 CHAPTER ONE: LITERATURE REVIEW................................. 7 Oxygen........................................................ 7 Iron........................................................... 10 Iron Cofactors.................................................. 12 Hemoproteins................................................... 19 Summary....................................................... 60 CHAPTER TWO: MATERIALS AND METHODS.......................... 63 Reagents....................................................... 63 Construction of pHPEX Plasmids................................... 64 Expresion of KatG and Myoglobin.................................. 68 Cloning and Expresion of KatP and KatG Y26F ........................ 71 Protein Characterization........................................... 75 CHAPTER THRE: RESULTS.......................................... 83 HPEX System................................................... 83 KatP.......................................................... 105 x Trp-Tyr-Met Covalent Adduct..................................... 123 CHAPTER FOUR: DISCUSION....................................... 141 Expresion of Recombinant Hemoproteins........................... 142 Periplasmic Catalase-Peroxidase................................... 151 Role of Trp-Tyr-Met Covalent Adduct............................... 158 Summary...................................................... 162 REFERENCES....................................................... 166 xi LIST OF TABLES Table 1.1. Reduction Potentials of Some Oxygen Species and Other Compounds.............................................. 8 Table 2.1 Plasmids and strains used for the development of the hemoprotein expresion (HPEX) system.................................. 64 Table 3.1. Catalase Kinetic Parameters for Recombinant KatG from Hemin- Supplemented Cultures of BL-21 (DE3) pHPEX3 and Unsupplemented Cultures of BL-21 (DE3) pLysS................ 96 Table 3.2. Heme Absorption Maxima for Recombinant KatG from pHPEX3- and pLysS-Transformed E. coli BL-21 (DE3)................... 97 Table 3.3. Heme Absorption Maxima for KatP and KatG.................. 110 Table 3.4. Kinetic Parameters for the Catalase and Peroxidase Activities of KatG and KatP........................................... 114 xii LIST OF FIGURES Figure 1.1. Molecular Orbital Diagram of Molecular Oxygen?...??.?...... 9 Figure 1.2. Schematic Diagram of Reactive Oxygen Species (ROS) and their Interactions with Cels..................................... 13 Figure 1.3. Iron-Sulfur Clusters and Mononuclear and Binuclear Non-Heme Iron Centers............................................ 15 Figure 1.4. Biosynthetic Pathway for Heme............................. 16 Figure 1.5. Structure of Common Hemes............................... 18 Figure 1.6. Active Site Representation of Myoglobin...................... 22 Figure 1.7. T ? R Transition in Hemoglobin........................... 24 Figure 1.8. Catalytic Cycle of Cytochrome P450........................ 27 Figure 1.9. Catalytic Cycle of Cytochrome Oxidase...................... 31 xii Figure 1.10. Active Site Comparison of Myoglobin vs. Peroxidase........... 35 Figure 1.11. Catalytic Cycle of Peroxidase............................ 37 Figure 1.12. Catalytic Cycle of Catalase................................ 40 Figure 1.13. Catalytic Cycle of Catalase-peroxidase....................... 45 Figure 1.14. Active Site Comparison of Catalase-peroxidase and Clas I Plant Peroxidases............................................ 47 Figure 1.15. Autoxidation Reaction of Myoglobin and Hemoglobin.......... 52 Figure 1.16. Generation of ROS in Cytochrome P450..................... 54 Figure 1.17. Schematic Representation of the NADPH Oxidase............ 58 Figure 1.18. Catalytic Cycle of Myeloperoxidase........................ 60 Figure 2.1. Oxidation of ABTS..................................... 78 Figure 2.2. Trypsin Digest of Peptide Bond in Protein.................... 80 Figure 3.1. Schematic Representation of the Plasmids pHPEX1 (A), xiv pHPEX2 (B), pHPEX3 (C), and pHPEX-fur(D)............... 84 Figure 3.2. Diagnostic Restriction Digests of pHPEX1 (A), pHPEX2 (B), pHPEX3 (C), and pHPEX-fur (D)......................... 86 Figure. 3.3. Heme Receptor Expresion by Untransformed and pHPEX2- Transformed E. coli BL-21 (DE3).......................... 88 Figure 3.4. Growth of Untransformed BL-21 (DE3) Cels in Minimal Media................................................ 89 Figure 3.5. Growth of pHPEX2-Transformed BL-21 (DE3) Cels in Minimal Media......................................... 91 Figure 3.6. UV-Visible Absorption Spectra of Catalase-Peroxidase Expresed in pHPEX3-Transformed Cels.................... 92 Figure 3.7. UV-Visible Absorption Spectra of Catalase-Peroxidase Expresed in pLysS-Transformed Cels...................... 93 Figure 3.8. Catalase Activity of KatG Expresed in pLysS- and pHPEX- Transformed System..................................... 94 Figure 3.9. Feric Minus Feri-cyano Diference Spectra for Recombinant KatG Expresed in pHPEX3- and pLysS- Transformed Systems... 98 Figure 3.10. UV-Visible Absorption Spectra for Myoglobin Expresed in pHPEX3-Transformed Cels (Lysis cels)................... 99 xv Figure 3.11. Derivative UV-Visible Absorption Spectra for Myoglobin Expresed in pHPEX3-Transformed Cels (Whole cels)......... 101 Figure 3.12. Derivative UV-Visible Absorption Spectra for Myoglobin Expresed in pLysS-Transformed Cels (Whole cels)........... 102 Figure 3.13. Derivative UV-Visible Absorption Spectra for Myoglobin Expresed in pHPEX-fur-Transformed Cels (Whole cels)....... 103 Figure 3.14. Derivative UV-Visible Absorption Spectra for Myoglobin Expresed in BL-21(DE3)-Transformed Cels (Whole cels)...... 104 Figure 3.15. SDS Electrophoretic Separation of Total Celular Protein from BL-21 DE3)pLysS Transformed pKatP3..................... 106 Figure 3.16. Heme Absorption Spectra for Feric, Feri-Cyano, and Ferous Forms of KatP in Soret Region.............................. 108 Figure 3.17. Heme Absorption Spectra for Feric, Feri-Cyano, and Ferous Forms of KatP (480-680 nm)............................... 109 Figure 3.18. EPR Spectrum of KatP and KatG......................... 12 Figure 3.19. Efect of H 2 O 2 Concentration on the Catalase Activity of KatG and KatP.......................................... 113 Figure 3.20. Efect of H 2 O 2 Concentration on the Peroxidase Activity of xvi KatG and KatP.......................................... 115 Figure 3.21. Efect of pH on the Catalase and Peroxidase Activity of KatP..... 117 Figure 3.22. KatP Formation of Compound I by Peracetic Acid.............. 118 Figure 3.23. Absorption Changes Spectra for KatP Formation of Compound I by Peracetic Acid........................................ 119 Figure 3.24. Efect of Peracetic Acid Concentration on the Rate of Formation of Compound I................................. 120 Figure 3.25. Efect of Cyanide Concentration on the Rate of Formation of the Fe III -CN Complex of KatP.............................. 121 Figure 3.26. Absorption Changes Spectra for KatP Formation of Feri-cyano Complex............................................... 122 Figure 3.27. SDS-PAGE of Ni-NTA-Purified KatP ...................... 124 Figure 3.28. Trp-Tyr-Met Covalent Adduct of Catalase-Peroxidases......... 126 Figure 3.29. Absorption Spectra of Purified KatG........................ 128 Figure 3.30. Efect of Heme and Peroxide on Migration of apo-KatG by SDS-PAGE............................................ 129 xvii Figure 3.31. Matrix-Asisted Laser Desorption-Ionization Mas Spectrometry Schematic................................... 132 Figure 3.32. MALDI Spectrum of Tryptic Digest of KatG without Covalent Adduct................................................ 134 Figure 3.33. MALDI Spectrum of Tryptic Digest of KatG with Covalent Adduct................................................ 135 Figure 3.34. SDS Electrophoretic Separation of Total Celular Protein from BL-21 (DE3)pLysS Transformed pKatG Y26F ................. 136 Figure 3.35. Efect of H 2 O 2 Concentration on the Catalase Activity of wtKatG and KatG Y26F .................................... 137 Figure 3.36. Efect of H 2 O 2 Concentration on the Peroxidase Activity of wtKatG and KatG Y26F .................................... 138 Figure 3.37. Efect of Heme and Peroxide on Migration of KatG Y26F by SDS-PAGE............................................. 140 Figure 4.1 Incorporation of Peroxynitrite in the Catalytic Cycle of Catalase-peroxidase........................................ 155 Figure 4.2. KatP Compound I Formation in the present of 200?m Peroxynitrite............................................. 156 xvii Figure 4.3. KatP Compound I Conversion to Compound I in the present of Peroxynitrite............................................. 157 Figure 4.4. Active Site Representation of E. coli Catalase-Peroxidase........ 160 Figure 4.5. Proposed Mechanism for Formation of Trp-Tyr-Met Adduct....... 163 1 INTRODUCTION Hemoproteins are a very important clas of enzymes found in nature and are involved in a wide range of biological proceses including the metabolism of drugs and other xenobiotics, binding and transport of oxygen, respiratory electron transport, and detoxification of reactive oxygen species. As the name implies, hemoproteins are identified by the presence of a heme prosthetic group. This prosthetic group is central to the activity and/or regulation of the hemoproteins. The astounding feature is that this same prosthetic group can be directed to such a wide aray of functions. Clearly, it is the protein structure around the heme group which directs the heme to a prescribed function specificaly and eficiently. Idealy, if we can understand how the structures of proteins do this, we can beter understand how to manipulate or control their many functions or even engineer new heme-based catalysts with new/superior properties. The best models to further investigate hemoprotein structure and function wil be those that provide, first of al, insight into poorly defined aspects of the structure function equation. Second, the best models wil be those for which deeper understanding of their structure and mechanism has direct benefit to unraveling serious biomedical problems. Catalase-peroxidases stand out in both respects. First, the catalase-peroxidases have substantial biomedical importance. They are involved in the activation of the drug isoniazid, which is a front line antibiotic against Mycobacterium tuberculosis. 2 Mutation to Mycobacterium tuberculosis catalase-peroxidase figures prominently in the development of resistance to isoniazid. Inded, over 70% of INH resistant tuberulosis strains cary mutations, which afect the function of catalase-peroxidase. Furthermore, the presence of a periplasmic version of this enzyme has ben identified as unique to virulent pathogenic bacteria such as Escherichia coli O157:H7, Yersinia pestis, and Legionela pneumophila. The distinct absence of this periplasmic enzyme in the non-pathogenic relatives of these organisms implicates it as a virulence factor. The mechanisms by which distant protein structures alter the catalytic properties of the heme prosthetic group are poorly defined. In this respect, the catalase-peroxidases also present an ideal model for study. The catalase-peroxidases have an active site which is virtualy superimposible on those of several monofunctional peroxidases. Yet, only the catalase-peroxidases posses substantial catalase activity. Because of the high similarity of the active sites, influence by structures peripheral to the active site are likely to be the influential factors in this obvious functional disparity. Clearly, catalase-peroxidases provide an ideal system to evaluate a poorly understood aspect of enzyme structure and function and apply this understanding to important biomedical problems. In order to study hemoproteins, large quantities of these proteins are needed for the material-intensive techniques necesary to pursue such studies. To date, the most productive, most eficient, and least expensive methods to produce the quantities of protein required are through the expresion of recombinant forms in Escherichia coli. Unfortunately, for hemoproteins this is often hindered by the fact that a vast majority of the protein produced lacks the heme prosthetic group required for activity [1-6]. Simply 3 stated, the rate of protein synthesis is much greater than the rate of heme biosynthesis. These dificulties have prompted atempts to beter match rates of heme biosynthesis and protein expresion. One approach is to reduce the rate of protein expresion [6] such that heme biosynthetic rates can keep pace. This is efective for producing hemoproteins with the full complement of correctly bound heme, but the decrease in expresion rates may be inconvenient. An alternative approach is to increase the rate of heme biosynthesis. Production of ?-aminolevulinic acid (?-ALA) is the rate-limiting step in the heme biosynthetic pathway [7]. Therefore, the most comon method is to add large quantities of ?-ALA to expresion cultures, stimulating heme production and resulting in higher heme content for the target recombinant hemoproteins [1, 3, 5]. Unfortunately, the amount of ?-ALA added to cultures is quite large, adding considerable expense to protein expresion procedures. The simplest approach would be to add heme to the culture medium during expresion of the target protein. This is hindered by the limited ability of many laboratory strains of E. coli to use exogenously added heme as an iron source for growth [8-10]. Conversely, through expresion of outer membrane-bound heme receptors many pathogenic bacteria efectively overcome this limitation [8, 9, 11-19]. One example is E. coli 0157:H7, which uses a TonB-dependent outer membrane-bound heme receptor for heme acquisition [8, 11, 12, 20]. Most importantly, it has been shown that common laboratory strains of E. coli need only produce a heme receptor in order to gain full ability to import heme into the cel [8-10]. In order to aid the evaluation of the catalase-peroxidase structure/function relationship, an E. coli-based hemoprotein expresion (HPEX) system was developed. This system is based on a series of plasmids (pHPEX) that contain the gene for the heme 4 receptor from E. coli 0157:H7 (chuA). Transformation of our E. coli expresion strains with pHPEX plasmids results in the ability to retrieve heme from the surrounding medium. Heme so collected is then incorporated into the active site of the target recombinant hemoprotein as it is expresed. Thus, this HPEX system is designed to yield target hemoproteins without limiting the rate of protein expresion. The expresion of the heme receptor by pHPEX along with hemin added to the expresion medium resulted in the expresion of KatG protein with full heme content and activity. The addition of the iron uptake regulator domain in pHPEX1-fur and the lacUV5 promoter to pHPEX2 resulted in the control of the heme receptor expresion by iron dependence or IPTG induction. The expresion of T7 lysozyme by pHPEX3 inhibited the leakage expresion of toxic protein under the common lac promoter. The celular locations of many catalase?peroxidases have not been unequivocaly established. Many appear to be cytoplasmic [21, 22], and the vast majority lacks a signal peptide sequence asociated with targeting to the periplasm. However, a smal number of exceptions have recently been identified [23-25]. However, they are exclusively present in highly virulent pathogenic bacteria, including E. coli O157:H7 [12], Y. pestis [25], and L. pneumophila [23]. In these thre cases, the periplasmic catalase?peroxidases are proposed to be virulence factors [23, 25-27]. The mechanisms by which these periplasmic catalase?peroxidases may contribute to virulence have not been determined, necesitating comprehensive studies to evaluate their potentialy important roles. Although one of these enzymes has been isolated [25, 27], the periplasmic catalase?peroxidases remain poorly characterized. This study reports the overexpresion, purification, and characterization of the periplasmic catalase? 5 peroxidase from enterohemorhagic E. coli O157:H7. This enzyme bears the name KatP because it is encoded on a large plasmid (pO157) asociated with virulence [24]. The isolation and characterization of this enzyme provides important additional information and resources to evaluate the mechanisms by which these novel periplasmic catalase? peroxidases may contribute to bacterial virulence. The addition of a histidine tag to KatP enabled the complete isolation of expresed KatP using afinity and hydrophobic interaction chromatography. Absorption spectra of KatP were typical for catalase-peroxidases. KatP showed comparable catalase and peroxidase activities to other catalase-peroxidases. The k cat values for KatP catalase and peroxidase activities were somewhat higher than other enzymes. On the other hand, K M values were considerably higher for KatP. KatP also showed the typical sharp but distinct pH dependence for catalase and peroxidase activities. KatP formation of compound I and CN ? binding rates were also similar. The strong catalase and peroxidase activities of KatP, combined with its unique periplasmic location and its ability to react with peroxynitrite may position it idealy for use by this and other pathogens in meting a host imune response. The crystal structures of catalase-peroxidases each show a thre amino acid covalent adduct [28-30]. The function of this Trp-Tyr-Met covalent adduct is not fully understood. However, its role in the function of catalase-peroxidase is being evaluated [31-34]. Mutation of the Tyr or Trp residue of the covalent adduct results in a loss of catalase activity [31, 33-37]. On the other hand, peroxidase activity is maintained. This indicates that the formation of compound I is not afected by the covalent adduct. The covalent adduct sems to have an efect on the reduction of compound I back to the feric 6 form of the enzyme which is where catalase and peroxidase activities difer. This study looks at the formation of the covalent adduct and its efect on catalase-peroxidase. High-level expresions of KatP appeared to result in production of the enzyme with and without the covalent adduct. Purified KatP evaluated by SDS-PAGE electrophoresis shows two protein bands. Each band was identified as KatP determined by MS. The upper band contained no heme, determined by FPLC, and had no activity, but the lower band had both heme and activity. The results with KatP reflected a general trend observed in the laboratory with KatG. That is, the greater the heme content, the greater the proportion of protein migrating with a lower apparent molecular weight. Reconstitution of apoKatG with heme resulted in the partial conversion of catalase- peroxidase to its cross-linked form. The conversion is enhanced upon the addition of peracetic acid. Conversely, reconstitution of KatG with a redox inactive cofactor (e.g., Zinc-protoporphyrin IX), no cross-link was formed, indicating that formation of the thre-amino acid covalent adduct was heme and peroxidase dependent. In order to further investigate this adduct, the variant Y226F was generated and evaluated by electrophoresis and steady-state kinetics. Purified Y226F showed only one KatG protein band evaluated by SDS-PAGE. The Y226F variant esentialy lost al catalase activity. However, peroxidase activity was retained. This further validates the necesity of the cross-link for catalase activity. Probing of this covalent adduct could very wel prove to be novel in the understanding of the relationship betwen the structure and function of enzymes. 7 CHAPTER ONE LITERATURE REVIEW Oxygen Molecular oxygen (O 2 ) is vital for a range of organisms, serving as a terminal electron aceptor in numerous oxidation reactions, not the least of which is the respiratory electron transport system. In order to obtain as much ATP as possible from a carbon source, the terminal oxidant must be a strong oxidizing agent. The standard potential for reduction of O 2 to H 2 O indicates that O 2 is a very good oxidant (Table 1.1). Thermodynamicaly speaking, O 2 should be able to acept electrons from almost any biological molecule. However, the chemical reactivity of molecular oxygen or dioxygen with most organic, and by extension, biological molecules at ambient temperatures is esentialy nonexistent [38]. The low reactivity of dioxygen is due to the presence of two unpaired electrons in the antibonding orbital (? 2p * ) (Fig.1.1). These unpaired electrons for dioxygen (+1/2, +1/2) give a spin quantum number of one, and a spin multiplicity of thre (2s + 1), which makes it a triplet molecule in its ground state. In order to 8 Reactions E o ? (Volts) Reference ?OH + e - + H + ? H 2 O 2.31 [39] H 2 O 2 + 2 e - + 2H + 2H 2 O 1.35 [40] O 2 ?? + e - + 2H + ? H 2 O 2 0.94 [41] RS ? (cysteine) + e - RS - 0.84 [42] O 2 + 4 e - + 4H + ? 2H 2 O 0.82 [39] PUFA ? (bis-alylic) + H + ?PUFA-A 0.60 [43] O 2 + 2 e - + 2H + ? H 2 O 2 0.33 [39] Ascorbate ? + H + ascorbate monoanion 0.28 [43] Ubiquinone + H + ? Semiubiquinone ?0.04 [44] NAD + 2 e - + H + NADH ?0.32 [41] Riboflavin ? Riboflavin -. ?0.32 [45] O 2 + e - ? O 2 ?? ?0.33 [40] Fe(II) transferin ? Fe(I) transferin ?0.40 [46] E o ?= standard reduction potential (1atm, pH 7) Table 1.1. Reduction Potentials of Some Oxygen Species and Other Compounds 9 Figure 1.1. Molecular Orbital Diagram of Molecular Oxygen 10 maintain spin conservation during a reaction, oxygen must either react with another unpaired electron or produce a triplet ground state product. Stable triplet states are unusual. Oxygen reaction with most molecules is considered to be spin-forbidden in that most molecules are in a singlet ground state [41]. The reaction of singlet molecules with triplet molecules does not occur at appreciable rates because of this high kinetic barier. On balance, this is an advantageous arangement. Without this kinetic barier, the indiscriminant and rapid oxidation of nearly al biological molecules would occur, making life in an aerobic atmosphere impossible. Nevertheles, to take advantage of the properties of O 2 in biological system, the kinetic barier must be overcome. As it is, the rate of O 2 reaction with singlet molecules is too slow to be of any use to a living organism. The dificulty comes in how to reduce this barier in a controlled and selective manner. Iron In order to lower the high kinetic bariers required for the reaction of triplet oxygen, transition metals are frequently used. Transition metals such as iron (Fe), copper (Cu), and manganese (Mn), which have partialy filed d-orbitals, can exist in several oxidation and spin-states. Transition metals can donate and acept electrons with oxygen forming a metal-dioxygen complex. This complex gives transition metals the ability to act as an eficient catalyst of redox reactions by giving dioxygen an electron environment like singlet oxygen (Fig. 1.1). Bonding of the metal with oxygen alows for electron 11 aceptance and lowers the activation energy of triplet oxygen to overcome the kinetic bariers. The interaction of metals with dioxygen is of great interest and has been wel studied since its discovery in the mid 1800s [47]. Iron, one of the most abundant elements, is an esential component of virtualy al life forms. Iron, under typical physiological conditions, is observed to cycle betwen its two dominant oxidation states ferous (Fe 2+ ) or feric (Fe 3+ ). The utility of iron in the catalytic activation of dioxygen can be sen even in the simplest system. For example, consider the reactions of a low molecular weight Fe complex like Fe-EDTA, where RH is an organic electron donor (reaction 2-5). The final reaction in the sequence, known as the Fe 3+ e - ? Fe 2+ (1) Fe 3+ + RH ? R - + Fe 2+ H + (2) Fe 2+ O 2 ? O 2 ?? + Fe 3+ (3) 2 O 2 ?? + 2H + ? H 2 O 2 + O 2 (4) Fe 2+ H 2 O 2 ? Fe 3+ ?OH + OH ? (5) Fenton reaction, produces hydroxyl radical (?OH) (reaction 5), the most active of al so- caled reactive oxygen or partialy-reduced oxygen species (e.g., O 2 ?? , H 2 O 2 , ?OH) [48- 50]. Hydroxyl radicals participate in hydrogen abstraction, electron transfer, or addition reactions with a truly broad range of compounds, and this at near difusion- limited rates [51, 52]. Clearly, the transition metal-dependent activation of dioxygen can be used for chemical transformations. However, this type of mechanism provides litle if 12 any control, leading to the indiscriminant oxidation of system components. Evidence that such mechanisms of oxygen activation are detrimental to biological molecules and living cels is abundant. Indeed hydroxyl radicals have ben shown to alter the structure and disrupt the function of al clases of biological molecules (e.g., lipid, protein, carbohydrate, and nucleic acid) (Fig. 1.2)[50, 53, 54]. Iron Cofactors Clearly iron can activate molecular oxygen toward many types of oxidative transformations. However, the isue stil remains of controlling its reactivity making it specific and useful rather than indiscriminant and harmful for living organisms. This must be achieved by controlling the environment of the iron. Nature has developed several ways to control iron and its interactions with dioxygen. The first evidence of biological control of iron is in the vanishingly low concentrations of ?fre? or low molecular weight complexes of iron observed within most organisms [55]. The vast majority of iron found in living systems is asociated, in one form or another, with proteins. Clearly, the protein environment of the iron is a key factor in determining its reactivity. The environment of the iron can be evaluated at several levels. The first, and most obvious, is its imediate environment, which is summed up in its nature as a cofactor. In almost al cases, a protein asociated with iron relies on the metal for its activity. The iron, therefore, is participating as a cofactor. The thre most common 13 Figure 1.2. Schematic Diagram of Reactive Oxygen Species (ROS) and their Interactions with Cels. ROS Lipid Protein DNA Lipid peroxidation Cel Death Membrane Damage Protein Oxidation Oxidized nucleic acid Alteration of function Ageing, mutagenicity, and carcinogenesis Cel Components 14 forms of iron-dependent cofactors are heme iron, non-heme iron, and iron-sulfur clusters. Iron-sulfur clusters are a form of non-heme iron cofactor, but they constitute such a widely used catalytic strategy in biology that they can be considered as a separate group. Commonly observed arangements include FeS, Fe 2 S 2 , Fe 3 S 4 , and Fe 4 S 4 (Fig. 1.3). Other non-heme iron centers are typicaly directly ligated by several amino acid side chains from the surrounding protein. Common motifs include mononuclear iron centers observed in proteins like tyrosine hydroxylase and TauD as wel as the binuclear centers observed in enzymes like methane monoxygenase and ribonucleotide reductase (Fig. 1.3). Finaly, hemes are also widely utilized as cofactors in biological systems. Though the other iron-dependent cofactors play very important roles in biological function, the focus of this literature review il be on the heme-type iron cofactors. Heme cofactors are synthesized via a common pathway using the universal tetrapyrrole precursor ?-aminolevulinic acid (?-ALA). The synthesis of ?-ALA is the rate-limiting step in the biosynthetic pathway (Fig. 1.4) [56, 57]. Whether derived from glycine and succinyl-CoA or glutamate (as in some prokaryotes), ?-ALA undergoes a series of six reactions, yielding protoporphyrin IX, the most common biological porphyrin. Protoporphyrin IX is then charged with a single Fe atom by the enzyme ferochelatase to form the most commonly observed heme cofactor, iron-protoporphyrin IX also known as heme b (Fig. 1.4). This synthetic pathway is highly similar across al species with some diferences in the subcelular location of the steps [56, 57]. In prokaryotes, this pathway takes place in the cytosol. However, in eukaryotes, ?-ALA is synthesized in the mitochondrial matrix along with the final thre steps involving coproporphyrinogen II synthesis to heme. Each of the other steps occurs in the cytosol. 15 Figure 1.3. Iron-Sulfur Clusters and Mononuclear and Binuclear Non-Heme Iron Centers 16 Figure 1.4. Biosynthetic Pathway for Heme. NHHN HNNH HOOCCOOH NHHN HNNH HOOCCOOH HOOC COOH NHHN HNNH HOOCCOOH HOOC COOH HOOC COOH HOOC HOOC NHHN HNNH COOH HOOC COOH HOOC COOH HOOC COOH HO HOOC HN H 2 N COOH COOH COOH H 2 N O Succinyl-CoA + Glycine !"aminolevulinate synthesis !-aminolevulinic acid !"aminolevulinate dehydratase Porphobilinogen Porphobilinogen deaminase Hydroxymethylbilane Uroporphyrinogen III cosynthase Uroporphyrinogen III Uroporphyrinogen decarboxylase Coporphyrinogen III Protoporphyrinogen IX Protoporphyrin IX Coproporphyrinogen oxidase Iron-protoporphyrin IX Protoporphyrinogen oxidase Ferrochelatase NHN HNN HOOCCOOH NN NN HOOCCOOH Fe 17 In plants, the mitochondria do not contain the enzyme required for coproporphyrinogen II conversion to protoporphyrinogen IX and requires a precursor form in plastids. Heme cofactors are utilized for a truly diverse range of biological proceses. This requires a complex regulation of the environment of the heme iron, and this regulation occurs at numerous levels. Most simply, there are structural perturbations of the cofactor itself. In al heme structures, iron is complexed to the macrocycle by four equatorial nitrogens from four pyroles. This leaves two additional ligand binding sites. The most commonly observed oxidation states for heme are the feric and ferous state (-200 mV) but fre heme is dominated by the feric state under atmospheric conditions [58]. The ferous form of heme iron in heme proteins is properly refered to as heme. Likewise, when the heme iron is in its feric form, the proper term is hemin. There are thre main types of heme found in proteins, al derived from protoporphyrin IX (Fig. 1.5). Heme b is the most abundant form of heme and is found in hemoglobins, myoglobins, catalases, and peroxidases among others. Heme a is diferent from heme b at two carbon positions of the porphyrin ring: the oxidization of one of its methyl side chains into a formyl group and the replacement of one of the vinyl side chains by an isoprenoid side chain as found in cytochrome-c oxidase (Fig. 1.5). The formyl group is more electron-withdrawing than the methyl group. The long alkyl group makes heme a hydrophobic and is also more electron-withdrawing than the vinyl group [59]. This gives heme a a higher reduction potential. Both heme a and heme b are not covalently bound to the proteins that rely on them. Heme c difers from b and a in that it is covalently atached by two thioether bonds betwen the protein and the vinyl group of the heme by the cysteine sulfurs of a CXCH binding motif (Fig. 1.5) [60]. The 18 Iron-protoporphyrin IX (Heme b) Heme a Heme c Figure 1.5. Structure of Common Hemes Fe N N H 3 C H 3 C C H CH 3 CH 2 CH CH 3 NN CH 2 CH 2 COO CH 2 COO H 3 C Protein CH 3 Protein H 2 C CH 2 CH=CCH 2 CH 3 H 3 Fe N N C H H 3 C CH=CH 2 CH 3 CH 2 CH CH 3 NN CH 2 CH 2 COO CH 2 COO O HO Fe N N H 3 C H 3 C CH=CH 2 CH 3 CH 2 CH=CH 2 CH 3 NN CH 2 CH 2 COO CH 2 COO 19 presence of the thioether bond may add stability to the protein. The loss of the thioether bond in cytochrome c 52 results in destabilization of the protein [61]. However, the reduction potential doesn?t sem to be afected by the presence or absence of the thioether linkage. Hemoproteins Heme is a highly reactive molecule because of its iron center. As with fre iron, heme can be very toxic to cels, tisues, and organs as a result of the open acesibility to the heme iron [57]. The heme structure alone is not sufficient to control and direct the reactivity of the iron-O 2 interaction. For this reason, virtualy al heme is found complexed with proteins as a prosthetic group, hence their designation as hemoproteins. The importance of hemoproteins in biology cannot be overstated. They participate in sensing, storing and transporting of O 2 , metabolism of drugs and other xenobiotics, electron transfer, production and decomposition of reactive oxygen species, and production, sensing, and degradation of nitric oxide. In al of these cases, the heme is central to the activity of the protein in question. Clearly, this diversity of chemical transformations cannot be explained solely by the structure of the heme. In many cases, the proteins in these proceses are using the same type of heme. The protein structure surrounding the heme dictates its function and this occurs at several levels. The most obvious is the imediate coordination environment of the heme moiety. Heme iron has two open coordination sites above and below the plane of the porphyrin 20 ring which ligands can bind. The character of the axial ligands can profoundly change the function of the heme. For instance, the globin proteins, which bind, transport, and store oxygen have a histidine as the fifth ligand bound to the Fe proximal side. The sixth coordination site is open to alow binding of O 2 or H 2 O (when O 2 is absent) as the sixth ligand [62]. O 2 is then transported as a non-reactive species. In the peroxidase proteins, histidine is also the fifth ligand but these proteins do not typicaly bind oxygen. Instead, they bind peroxide and reduce it to H 2 O. On the other hand, cytochrome P450 is a monooxygenase. It has a cysteine ligand bound to the Fe proximal side with oxygen or water as the other ligand [63]. The proximal ligand along with the H-bonding networks asociated with the ligand play a key part in the activity of hemoproteins. The reduction potentials of human myoglobin (+50 mV), cytochrome P450 cam (high spin, ?170 mV; low spin, -270 mV), horseradish peroxidase (-250 mV), and catalase (<-500 mV) are diferent as a result of the nature of the ligands[64]. The reduction potential of the heme iron, of course, dictates the stability of its oxidation states ? a critical consideration in the reactivity of a hemoprotein. Binding of O 2 by the Globins Though the most famous examples of the globins are mamalian hemoglobin and myoglobin, globins have been identified in bacteria, fungi, and plants as wel. The globins are known for their ability to sense, store, and/or transport O 2 . They have also been shown to posses NO dioxygenation activity (reaction 6). In order to achieve al of 21 Hb II + O 2 ? Hb II O 2 + NO ? Hb III ONO - ? Hb III + NO 3 - (6) these functions, the heme iron must be in its reduced form rather than its oxidized form because O 2 only binds to heme in the ferous form not the feric form. The ligation environment of heme in the globins favors the reduced state. As discussed previously, the reduction potential of the heme in these types of proteins is relatively high (~ +50 mV). This is due to the proximal histidine ligand bound to the heme iron and its H- bonding interactions (Fig. 1.6). In the globins, the proximal histidine forms a weak H- bond with a backbone carbonyl. Relatively speaking, the very weak H-bond means that the histidine ligand retains neutral character. This increases the stability of the ferous oxidation state relative to the feric. This is evident in the relatively high reduction potential. However, more is required for efective O 2 binding than heme in its ferous state. The protein environment outside of the iron ligands also plays an important role. A histidine on the distal side of the heme in the globins is positioned close to the heme iron. This histidine forms a strong hydrogen bond to dioxygen (2 ?), when it is bound, stabilizing the Fe-O 2 complex [65]. Amino acids like valine and leucine are also positioned in the distal pocket. These inhibit binding of other ligands and help to protect the active site against side reactions involving the bound O 2 . The ability of hemoglobin and myoglobin to bind oxygen in an un-reactive environment using iron provides an excelent example of how the Fe-O 2 interaction is controlled by protein environment. It is because of the proteins structural features far from the heme group that hemoglobin and myoglobin can be alies in the transport and storage of oxygen. Hemoglobin is found in high concentration in the red blood cels while myoglobin is 22 Figure 1.6. Active Site Representation of Myoglobin [66]. His 93 His 64 Val 68 Leu 69 Ile 107 23 found in muscle tisue. The heme in both hemoglobin and myoglobin is coordinated to the protein through a proximal histidine ligand revealed by X-ray crystalography [67, 68]. Oxygen is left to bind to the one remaining Fe coordination position on the distal side of the heme. X-ray structural analysis has shown that hemoglobin contains two ? and ? chains each resembling myoglobin. The architecture of the heme pocket gives the globin proteins the ability to reversibly bind O 2 . The distal histidine in its hydrophobic environment is a key amino acid in that its hydrogen bonding network helps to stabilize oxyheme against superoxide-releasing autoxidation and also contribute to CO vs. O 2 discrimination [69]. Both globins posses al of these features; however, hemoglobin and myoglobin do not bind to oxygen at the same rate. Binding curve plots show myoglobin binding to oxygen with high afinity acording to a hyperbolic curve [70]. This indicates that myoglobin wil bind oxygen when oxygen partial presure is high and release oxygen when oxygen partial presure is low. On the other hand, hemoglobin produces a sigmoid (S-shaped) curve [71]. This indicates a change in hemoglobin afinity for oxygen or cooperative oxygen binding. Hemoglobin afinity for oxygen rises with increasing oxygen concentration. This cooperative binding is due to a transition betwen its tertiary structures. When oxygen is bound to the iron, a 0.6? shift is observed in helix F (Fig.1.7) [72]. This shift in the iron atom gives rise to a conformational change resulting in a heme environment with an enhanced ligand afinity. This is caled an oxy or relaxed (R) state. When oxygen is not bound, it is caled a deoxy or tense (T) structure (Fig.1.7). It is clear that the high afinity of oxygen binding for myoglobin makes it ideal for oxygen storage. On the other hand, the ability of hemoglobin to bind oxygen more 24 Figure 1.7. T ? R transition in hemoglobin. R state is with O 2 bound. 0.6 A Fe 2+ Fe 2+ O 2 Leu Leu Leu Leu Leu Leu His His Helix F 25 when oxygen is available and release oxygen when it is not makes it ideal for oxygen transport. Cytochrome P450 Cytochrome P450 (P450) is a large family of hemoproteins found in a wide range of eukaryotes and prokaryotes. With over 450 diferent enzymes identified, most P450s are membrane-bound proteins. Cytochromes P450 have been shown to catalyze many reactions including hydroxylations, epoxidations, dealkylations, sulfoxidations, dehalogenations, and oxidative deamination. In humans, P450s play a critical role in celular metabolism, which includes: 1) the conversion of cholesterol to androgen, estrogen, gluco- and mineral-corticoids, 2) the synthesis and degradation of prostaglandins and other unsaturated faty acids, 3) the conversion of vitamins to their active forms, 4) the metabolism of cholesterol to bile acids, and 5) a number of reactions involved in the metabolism of xenobiotics. Generaly speaking, the P450s are mono- oxygenases and are diverse in their ability to incorporate one of the two oxygen atoms of O 2 into a variety of substrates while the other oxygen atom is reduced by two electrons to water. An example hydroxylation is given in reaction 7. R-H + O 2 + 2H + 2e ? ? R-OH + H 2 O (7) 26 The first P450 was identified by Klingenberg in 1958. It exhibited an unusual absorption band at 450 nm when reduced and exposed to CO [73]. This hemoprotein was asigned the name cytochrome P450 as a result. Like many hemoproteins, the P450s use heme b as a prosthetic group. However, cytochromes P450 difer from most hemoproteins in that they have a proximal cysteine ligand. The substantial electron density of the thiolate ligand preferentialy stabilizes the Fe III over the Fe II oxidation state as reflected in the low reduction potential (~ -300 mV for the low-spin form). As a consequence higher oxidation states, like the feryl-oxo porphyrin radical intermediate, are stabilized as wel, contributing to the P450?s ability to heterolyticaly cleave oxygen- oxygen bonds (reaction 8). The globins are not able to achieve this. Heterolytic Bond Cleavage R?OH ? RO + + OH ? (8) Homolytic Bond Cleavage R?OH ? RO ? + ? OH (9) The catalytic cycle of P450 starts with a low-spin, six-coordinate feric heme with water as the distal ligand (Fig. 1.8) [74]. When the substrate binds, water is released and the heme becomes a high-spin, five-coordinate feric species that leaves an open site available for oxygen to bind (a). An external electron donor is also needed. The cofactor NAD(P)H used by P450 reductase serves as an electron delivery system to reduce the P450 heme to a ferous species so that O 2 can bind (b). The binding of oxygen yields an oxygen-P450-substrate complex, which is unstable and undergoes molecular ? 27 Figure 1.8. Catalytic Cycle of Cytochrome P450. Cys S Fe III Cys S Fe III Cys S Fe II O 2 Cys S O Fe II O Cys S O Fe III O . RH e - RH RH e - Cys S O Fe III O 2 Cys S O Fe III OH H + H + ,e - Cys S O Fe IV H 2 O ROH a b c d e f g h RH RHRH RH RH 28 rearangement betwen a ferous-O 2 (c) and a feric-superoxide species (d). The addition of a second electron by NAD(P)H, which is the rate-limiting step, generates a feric peroxide adduct (e) [74]. This intermediate is protonated to a hydroperoxide intermediate (f). Oxygen is then protonated a second time (g). This leads to the heterolytic cleavage of oxygen resulting in the release of water and the generation of an feryl-oxo intermediate, which is similar to the peroxidase enzyme iron-oxy intermediate caled compound I. This intermediate is believed to be responsible for the production of the hydroxylated substrate (h). Cytochrome Oxidase Cytochrome oxidase, also known as cytochrome c oxidase or cytochrome aa 3 , is a complex of metaloproteins that play a fundamental role in oxygen reduction in celular respiration of eukaryotes and prokaryotes. Clearly, cytochrome oxidase is an esential enzyme for most aerobic life forms. Indeed, it is the protein of celular respiration that interacts directly with the terminal electron aceptor, molecular oxygen. This enzyme is an integral membrane protein that facilitates a transmembrane proton gradient by using the driving force of reduction of O 2 to H 2 O to cary out the translocation of protons across the membrane. Consuming more than 90 % of O 2 taken in by living organisms, cytochrome oxidase gives cels the ability to oxidize foodstuff by catalyzing one-electron oxidation of four cytochromes c resulting in the four-electron reduction of O 2 (reaction 10). 29 4 cytochrome c 2+ 4 H + O 2 ? 4 cytochrome c 3+ 2 H 2 O (10) The ability of cytochrome oxidase to reduce O 2 to H 2 O makes it unique compared to other hemoproteins that bind molecular oxygen like the globins. Cytochrome oxidase contains two heme a (a and a 3 ) molecules and two copper (Cu a nd Cu b ) ions at its catalytic core. As established by McMunn, Warburg, and Keilin betwen 1884-1933, this enzyme has spectroscopic properties of cytochromes but binds to oxygen like the oxidases. To acount for both of these properties, it was named cytochrome oxidase [75]. However, it wasn?t until 1938 that Keilin and Hartre discovered the esential role of cytochrome c as an electron donor to the terminal oxidase[75]. They decided to cal the enzyme cytochrome c oxidase. Keilin and Hartre studies of cytochrome oxidases also showed some disimilarity betwen the two hemes. One heme semed to bind oxygen, carbon monoxide, and cyanide while the other one did not [75]. Heme a was obtained as the name for the one that did not bind oxygen and a 3 given to the other. Both heme a and a 3 from cytochrome oxidase have a high reduction potential (+230 and +390 mV, respectively) as a result of the long alkyl group and formyl group which makes heme a more electron-withdrawing than both heme b and c [76]. The crystal structure of cytochrome oxidase shows heme a as a six-coordinate low-spin heme with two histidine residues as axial iron-ligands [77]. Conversely, heme a 3 is a five- coordinate high-spin heme. This diference in coordination environment has much to do with difering roles of these hemes in cytochrome oxidase catalysis. The absence of a sixth ligand to the heme a 3 iron alows for oxygen to bind as a ligand. Whereas heme a 30 simply acts as an electron transfer conduit delivering electrons to the heme a 3 and Cu b as appropriate. The reaction of cytochrome oxidase requires the cooperation of four redox centers. Cytochrome oxidase is composed of as many as thirten subunits with the redox centers located within subunits I and I [78]. Heme a and a 3 and Cu b are located in subunit I, while Cu a is found in subunit I. Heme a 3 and Cu b form a binuclear catalytic center for O 2 reduction to H 2 O. Heme a and Cu a re connected through a series of hydrogen bonds which alows for the transfer of electrons from cytochrome c to the catalytic center. The reduction of O 2 to H 2 O requires eight general steps (Fig. 1.9) [79]. Cu b , coordinated to the catalytic center by the bonding of thre histidines residues, is reduced by one electron from Cu II to Cu I (a). The hexa-coordinate feric heme a 3 is then reduced by one electron to the Fe I state (b). This reduction enables O 2 to bind, forming a dioxygen complex involving Fe I a3 and Cu I b (c). The reduction of the dioxygen complex results in an iron peroxy intermediate (d). Cu b is then reduced by one electron folowed by a proton uptake, giving rise to a Fe II a3 -O - OH intermediate along with Cu I b (e). A second protonation coincides with the heterolytic bond cleavage of O-O, giving a feryl-oxo state for heme a 3 and H 2 O bound to Cu 2 b (f). Heme a 3 is then reduced by one electron initiating a proton rearangement (g). This rearangement gives both iron and coper with hydroxide ligands. Introduction of two more protons yields two H 2 O molecules and a return to the resting state of the enzyme (h). A rapid transfer of electrons from heme a traps O 2 in the iron-copper center alowing for further reduction of oxygen. Inded, the reaction of cytochrome oxidase is 31 O Fe III Cu II e - Fe III Cu I e - Fe II Cu I e - O 2 Fe II Cu I O O Fe III Cu II O O - H + Fe III Cu I O OH Fe IV O 2- Cu II H H - - e - Fe III OH - Cu II OH - 2H + 2H 2 O H + a b c d ef g h Fe II Cu I O O Figure 1.9. Catalytic Cycle of Cytochrome Oxidase. The Fe refers to the iron of heme a 3 and the Cu refers to the Cu b center. 32 one of the fastest in biology with an aparent second order rate constant of about 10 8 M -1 s -1 [80, 81]. This rapid reduction of O 2 to H 2 O decreases the life span of the catalytic intermediates containing partialy reduced oxygen species, which minimizes their release into the surrounding environment. The bi-heme and bi-copper complex gives cytochrome oxidase the ability to efectively and eficiently utilize the terminal oxidant, O 2 , by binding and reducing it by four electrons to H 2 O. However, other redox centers in the electron transport chain may leak electrons to O 2 producing reactive oxygen species (e.g., O 2 ?? ) [82]. Other hemoproteins including superoxide dismutases, catalase, and peroxidase are used as a defense against O 2 ?? . Peroxidases Peroxidases are a large clas of enzymes found among many living organisms. This wide distribution implicates these enzymes as important for most living systems. Most contain heme, and these are divided into two superfamilies: mamalian peroxidases and peroxidases found in plants, microorganisms, and fungi (the so-caled plant peroxidases) [83]. A smal third group includes chloroperoxidase and the di-heme cytochrome c peroxidase. Mamalian peroxidases include lactoperoxidase, myeloperoxidase and prostaglandin H synthase. The second group is further divided into thre clases. Mitochondrial yeast cytochrome c peroxidase, chloroplast and cytosolic ascorbate peroxidases, and the gene-duplicated bacterial catalase-peroxidases make up clas I. 33 Clas I are monomeric glycoproteins with four conserved disulfide bridges and two conserved calcium coordination sites, and are comprised of the secretory fungal peroxidases like lignin and manganese peroxidase [84]. The clasical or secretory plant peroxidases, such as the wel-known enzyme horseradish peroxidase, form clas II. They are also monomeric glycoproteins with four conserved disulfide bridges and two- conserved calcium binding sites. However, the location of the disulfide bridges is diferent from the clas I enzymes. In the overal reaction of most peroxidases, peroxide is reduced to H 2 O using two donated electrons (reaction 11). Peroxidases also participate in a variety of biosynthetic and/or degradative functions using peroxide as an oxidant. On one hand, the consumption of peroxide protects the cel against acumulation of these dangerously reactive compounds. However, the identity of the electron donor also contributes a large part of the particular biological function of each peroxidase. For instance, peroxidases catalyze the dehydrogenation of monolignols, a precursor in the biosynthesis of the cel- wal polymers known as lignin [85]. On the other hand, in white rot fungi, peroxidases help to catalyze the degradation of the plant polymer lignin using H 2 O 2 and veratryl alcohol or Mn II [86]. . ROH + 2AH ? H 2 O + 2A ? + ROH (11) Like the globins and P450s, peroxidases contain the non-covalently bound heme b. As in the globins, a histidine serves as the proximal ligand to the heme iron [87]. 34 Water may serve as the sixth ligand, but this varies from peroxidase to peroxidase. This, of course, raises the very important question: What about the structures of these proteins acount for their diferent activities and functions? First, although both groups of proteins share a common proximal histidine ligand, the environment of that ligand is quite diferent in peroxidases as compared to the globins (Fig. 1.10). In peroxidases, there is a strong hydrogen bond betwen the -N of the histidine ligand and a strictly conserved aspartate residue. The strong H-bond imparts a substantial anionic character (i.e., electron density) to the histidine ligand. Thus, the higher oxidation states of the heme iron tend to be favored. Indeed, the reduction potential for a typical peroxidase is ~ -250 mV [64]. Conversely, the histidine ligand of the globins participates in an H-bond with a backbone carbonyl oxygen. The considerably weaker H-bond ensures that this proximal ligand retains a more neutral character. Thus, the reduction potential of the heme iron is considerably higher ~ +50 mV[64]. In its resting state, peroxidases are a feric heme protein unable to bind oxygen, whereas the globins are found in the ferous state, and therefore able to acept O 2 as a sixth ligand. A common distal histidine is also found in both groups. In peroxidases, the distal histidine serves as a general acid-base catalyst to transfer a proton from the peroxide oxygen-1, bound to the heme, to oxygen-2 to promote the heterolytic cleavage of the peroxide O-O bond. However, the distal histidine in myoglobin forms an H-bond with the O 2 to help stabilize oxyheme. Another unique feature of the distal cavity of peroxidases is the presence of an arginine and tryptophan. The positive charge arginine stabilizes the developing negative charge on oxygen-2. 35 Figure 1.10. Active Site Comparison of Myoglobin vs. Peroxidase [66, 88]. His 93 Ile 107 Leu 69 Val 68 His 64 Phe 46 Asp 235 His 175 Trp 191 Asn 82 Trp 51 His 52 Arg 48 Myoglbin Peroxidase 36 While the peroxidases are involved in a variety of functions, the catalytic cycle of many of them are highly similar. The resting state of the enzyme is found as a high-spin feric heme. The heme iron is oxidized by two electrons by H 2 O 2 , resulting in the heterolytic cleavage of the O-O bond (Fig. 1.11)[89, 90]. This leads to the release of one equivalent of H 2 O and the generation of the intermediate known as compound I, which has a formal oxidation state of +5. Compound I is a feryl-oxo porphyrin ?- cation radical intermediate with an oxygen from hydrogen peroxide stil bound. In some cases, like cytochrome c peroxidase, the radical is transfered to a near-by amino acid side chain [91]. Compound I is then reduced by one electron by an exogenous electron donor to yield a feryl-oxo intermediate caled compound I and one equivalent of the substrate radical. Although similar to compound I, compound I does not contain the ?-cation radical and has a formal oxidation state of +4 [89]. Compound I is finaly reduced back to the feric or resting state of the enzyme by a second equivalent of exogenous electron donor. Thus, a second equivalent of substrate radical is generated. The peroxidases are known to undergo H 2 O 2 -dependent inactivation. In the presence of exces H 2 O 2 , compound I is oxidized, resulting in the formation of compound II (formal oxidation state of +6) and water (reaction 12) [92]. Because the structure and oxidation state of compound II is similar to those of oxymyoglobin or oxyhemoglobin, it is often caled ?oxyperoxidase? [92, 93]. Though the reaction of compound I with H 2 O 2 is the most commonly observed route to compound II, the production of compound II has also been shown to occur by two additional pathways [94]. Ferous peroxidase can react with O 2 (reaction 13), and feric peroxidase may react 37 Figure 1.11. Catalytic Cycle of Peroxidase. 38 Fe +4 =O (compound I) + H 2 O 2 ? Fe +3 -O 2 ?? (compound II) + H 2 O (12) Fe +2 (ferous) + O 2 ? Fe +3 -O 2 ?? (compound II) (13) Fe +3 (feric) + O 2 ?? ? Fe +3 -O 2 ?? (compound II) (14) with O 2 ?? (reaction 14). The formation of compound II from the reaction of compound I is slow and has an estimated second-order rate constant of 25 M ?1 s ?1 [95]. While compound II is an inactivated intermediate of peroxidases, compound II spontaneously decomposes to the feric or resting form of the enzyme with a first- order rate constant of 10 ?3 s ?1 [95]. Nevertheles, in the presence of high H 2 O 2 concentrations, this inactive compound II intermediate tends to acumulate. Unfortunately, in the presence of exces H 2 O 2 , compound II wil undergo aditional reactions that lead to ireversible inactivation [93, 96]. The mechanism of this ireversible inactivation is unclear. The presence of the reducing substrate may play an important role in the inhibition of this inactivating pathway by limiting the buildup of compound I. Catalases Like the peroxidases, catalases are a clas of hemoproteins that degrade hydrogen peroxide. Although catalases were named by Loew in 1901, it wasn?t until 1923 that Warburg discovered the presence of iron at the active site [97]. Stern later identified heme b as the prosthetic group of al then known catalases [98]. The overal reaction of 39 catalase involved the degradation of two molecules of hydrogen peroxide to water and oxygen (reaction 15). 2H 2 O 2 ? 2H 2 O + O 2 (15) The primary role of catalase is proposed to be the protection of organisms from H 2 O 2 and/or the reactive oxygen species generated from H 2 O 2 . The catalytic cycle of catalases is similar in many respects to peroxidases. However, in the case of catalase, H 2 O 2 is used both as an oxidant and reductant. The cycle is initiated by the oxidation of the feric heme by one molecule of H 2 O 2 to an oxyferyl porphyrin cation radical intermediate or compound I (Fig. 1.12) [99]. This results in the heterolytic O-O cleavage and the release of one equivalent of H 2 O. Compound I is reduced by to the feric form of the enzyme by a second molecule of H 2 O 2 , releasing a second equivalent of H 2 O and a molecule of O 2 . Catalases, also caled hydroperoxidases, are categorized into thre main groups. The most widespread and extensively characterized group is the monofunctional catalases. The second group is the bifunctional catalase-peroxidases. This clas of enzyme is closely related by sequence and structure to plant peroxidase and wil be discussed in more detail in a later section. The last group includes the nonheme or Mn- containing catalase. Other heme-containing proteins also exhibit low levels of catalase activity including chloroperoxidases, plant peroxidases, and myoglobins [100]. This is likely due to the presence of heme. However, at best, these enzymes show 1000 fold 40 Figure 1.12. Catalytic Cycle of Catalase. 41 lower catalase activity than monofunctional catalases. Phylogenetic analysis of monofunctional catalases has revealed a division into thre clases. Clade I are mostly plant enzymes with one branch of bacterial catalases [101]. Clade I contains the large subunit catalases. Al Clade I enzymes are of bacterial or fungal origin with the exception of one archaebacterial enzyme. Clade II contains the smal subunit catalases. These are widely distributed in bacteria, archaebacteria, fungiand other eukaryotes including mamals. The most noticeable diference betwen smal and large catalases is the presence in large catalases of an extra carboxyl-terminal domain, of about 150 residues, that resemble flavodoxin along with the absence of NADPH [102]. Like the peroxidases, most catalases contain heme b. However, the large subunit catalase or clade I contains a chlorin as the prosthetic group rather then heme b [103]. Heme b is believed to be bound first to the enzyme followed by a self-catalyzed reaction using the first H 2 O 2 molecules bound to the enzyme [104]. Ring II of heme b is converted to cis-hydroxy ?-spirolactone. This modified heme is termed heme d. As with most hemoproteins, catalases have a distal histidine. However, the distal histidine is coplanar with the heme [99]. This orientation favored intensive ?-? interactions betwen the esential histidine and the porphyrin [100]. This favors catalase activity by decreasing the reactivity of compound I with other substrates. The heterolytic O-O cleavage by hemoproteins requires the heme iron to have a low reduction potential in order to stabilize the higher oxidation states of iron. In the peroxidases and P450s, this is acomplished by the proximal anionic histidine and thiolate (cysteine) heme iron ligand respectively as previous discused. Catalases have 42 a tyrosine proximal iron ligand diferent from the peroxidases and the P450s. The proximal tyrosine is presented as a phenolate anion with the asistance of an adjacent arginine side chain. The anionic character of the phenolate ligand contributes to the very low reduction potential (<-500 mV) of catalases [105]. Most catalases exist as homotetramers with a characteristic globule for each subunit. Each subunit is comprised of four domains: The N-terminal arm, the ?-barel domain, the domain connection, sometimes refered to as `wrapping domain' and the ?- helical domain [103, 106, 107]. As previously mentioned, a smal group contains an extra domain on the C-terminus that resembles a flavodoxin. While the ?-barel is wel conserved, the ?-helical domain has more variability. The N- and C-terminal domains have the most divergences. The heme of catalase is deeply buried in the core of the subunits. The N-terminal arm extends for more than 50 residues and is the first region of catalases. It contains a 20-residue helix, which is the first secondary structure common to al catalases [108]. The esential distal histidine is located at the C-terminal end of the N- terminal arm. The second region is the ?-barel domain, which is an eight-stranded antiparalel ?-barel with six ?-helical insertions [99]. It represents the core of each subunit and forms the distal side of the heme pocket. It also participates in the binding of NADPH in smal subunit enzymes [109]. NADPH serves to protect catalase from inactivation by transfering electrons to the heme. Catalases become inactive when compound I undergoes a one-electron reduction to compound I. Compound I is reduced, or its formation is prevented when NADPH is bound to the enzyme. 43 The third region is the wrapping domain. This domain is a link betwen the ?- barel and ?-helical domain. It contains some of the residues of the proximal side of the heme pocket. The fourth region is the ?-helical domain. This domain interacts with the ?-barel and helps to stabilize its structure. It is also involved in the binding of NADPH. In the large subunit enzymes, a fifth domain links the ?-helical domain to the C-terminal domain. This domain structure resembles flavodoxin and is caled the flavodoxin domain. A flavodoxin domain may indicate a potential point of binding for flavin mononucleotide to serve as an electron transfer source as with NADPH. However, at this time its role in catalase function is not understood. Catalase-Peroxidases Catalase-peroxidases are a member of the Clas I plant peroxidases and posses the unique ability to use one active site to catalyze two distinct activities (catalase and peroxidase). This is compared to monofunctional catalases, which have virtualy no peroxidase activity, and typical peroxidases, which have virtualy no catalase activity. With either activity, H 2 O 2 is catalyticaly removed. These enzymes have been identified in a wide range of bacteria and a few fungi. It is suggested that catalase?peroxidase forms part of an esential defense against the deleterious efects of H 2 O 2 and other reactive hydroperoxides [21, 23, 110]. In fact, the loss of catalase-peroxidase in Mycobacterium tuberculosis has ben shown to decrease its ability to deal with reactive species generated by the host oxidative burst defense resulting in death [112]. 44 Like most catalases and al plant peroxidases, catalase-peroxidases have heme b as a prosthetic group in a feric high spin form. Histidine is the fifth ligand. The catalytic cycle for both catalase and peroxidase is initiated by the reduction of H 2 O 2 by two electrons to H 2 O with the five-coordinate feric heme forming the iron-oxygen complex of the enzyme caled compound I (Fig. 1.13). The two activities difer in their ability to reduce compound I back to its original feric form. For catalase, compound I is reduced by two electrons to return to the feric state by a second equivalent of H 2 O 2 . This results in the production of O 2 and a second equivalent of H 2 O. In the case of peroxidase, two one-electron steps reduce the enzyme back to the feric form. In the first step, compound I is reduced by one electron by an exogenous electron donor to produce compound I. Compound I is then reduced by one electron to produce the feric form of the enzyme by a second equivalent of the exogenous electron donor. This two-step reduction results in the production of a second equivalent of H 2 O and two equivalents of the electron donor radical. Compared to catalases and peroxidases, catalase-peroxidases are a relative new enzyme clas and have not been as extensively studied. Although the first catalase- peroxidase was discovered in 1979, a full sequence was not known until 1988 [111, 112]. These enzymes are now clasified as clas I plant peroxidases because of the similarity of their sequence and structure. Crystal structures of the enzyme started appearing in the early 2000s [28-30]. Catalase-peroxidases are found as a dimer or a tetramer with each subunit composed of 20 ?-helices joined by linkers and thre or four ?-strands [113]. This gives catalase-peroxidases a very diferent tertiary/quaternary structure from monofunctional catalases, which have four distinct regions as previously mentioned. 45 Figure 1.13. Catalytic Cycle of Catalase-peroxidase. 46 Indeed, there is no sequence or structural similarity betwen the catalase-peroxidases and monofunctional catalases. Clearly, the catalase activity of these bifunctional enzymes arises from a diferent structural strategy. On the other hand, comparison of these bifunctional enzymes to monofuncional peroxidases reveals some striking similarities. In addition to the same heme group and fifth ligand (histidine), the other key residues in the active sites of catalase-peroxidases are virtualy superimposable on those of monofunctional peroxidases (Fig. 1.14). This indicates that structural components external to this sphere of active site functional groups modulates the active site to impart catalytic abilities that do not typicaly come with peroxidase enzymes. Importantly, this opens a doorway to evaluating how active site capabilities are influenced by distant protein structures, a poorly understood aspect of enzyme structure/function. In this regard, catalase-peroxidases have structural components that are distinctly absent from monofunctional peroxidases. Many of these unique features are external to the active site. It is believed that catalase-peroxidases arose through gene duplication, which resulted in a gene with two sequence-related domains (N-terminal and C-terminal) [114]. This is diferent from most peroxidases, which only have a single domain. Both domains are similar to monofunctional clas I plant peroxidases. The active site is located in the N-terminal domain. However, the C-terminal domain doesn?t have an active site and has undergone a greater evolutionary drift. The active site is more deply buried than peroxidase but is acesed through a chanel lateraly like peroxidase rather then perpendicularly as in the monofunctional catalases. The C- terminal domain helps support the architecture of the active site by preventing the 47 Figure 1.14. Active Site Comparison of Catalase-peroxidase and Clas I Plant Peroxidases [29, 88, 115]. Arg 38 His 42 Trp 41 Asn71 His 163 Asp 208 Trp 179 Ascorbate Peroxidase Cytochrome c Peroxidase Arg 48 His 52 Trp51 Asn 82 His 175 Asp 235 Trp 191 Catalase-Peroxidase Arg 102 His 106 Trp 105 Asn 136 His 267 Asp 377 Trp 318 48 esential distal histidine from cordinating to the heme iron. Removal of the C- terminal domain results in the colapse of the active site and a complete los of activity [116]. In lignin and manganese peroxidases, a Ca 2+ ion helps support the architecture of the active site [117, 118]. However, the C-terminal domain likely has additional roles. Another diference from plant peroxidases revealed by sequence and structure analysis is the presence of two interhelical insertions (~35 amino acids) found in the N-terminal domain of catalase-peroxidases (an FG insertion found in the conector of the F and G helices, and an DE insertion found in connector of the D and E helices) [19]. Los of the FG insertion results in the selective decrease of catalase activity [120, 121]. The FG insertion is believed to help peroxides aces the active site and its los interferes with the acesibility of the active site and the H-bonding network. The DE insertion has more of an influence on catalase-peroxidases. Our laboratory has also observed that the los of the DE insertion results in a change of the active site environment and complete los of catalase activity [120, 121]. One role of the DE insertion was revealed by crystal structures and electron density maps of catalase-peroxidases [28-30]. The presence of a thre amino acid covalent adduct (Trp- Tyr-Met) is found in the distal heme pocket. The central amino acid Tyr is located on the DE insertion. Removal of the DE insertion interupts this covalent adduct. The activity of the enzyme is afected by the presence of this adduct and requires the presence of heme for its formation. Interuption of this thre amino acid covalent adduct results in the loss of catalase activity indicating one role of this adduct is to protect the catalase activity of the enzyme. The presence of this adduct also sems to be a characteristic common to al catalase-peroxidases. The key to understanding what makes catalase- 49 peroxidases unique in their function is held in these structures and how they afect the active site. The study of catalase-peroxidases wil provide valuable insight to the general question of how an entire enzyme structure important for determining its catalytic capabilities, even those structures that are distant from the active site. In addition to the knowledge of enzyme structure/function that can be gained from the study of catalase-peroxidases, these enzyme have roles in important biomedical proceses that further increase the benefits to be gained from understanding the link betwen their structure and catalytic abilities. Tuberculosis is one of the most deadly infectious diseases in developing countries and is estimated to have infected one-third of the world?s population [122]. Isoniazid (INH) has been a first-line antibiotic used against Mycobacterium tuberculosis infections since its discovery in 1952 [123]. It is known that INH is a prodrug which is converted to an active form by M. tuberculosis catalase- peroxidase (KatG) [124]. Activated INH inhibits the pathway for mycolic acid biosynthesis leading to the death of the organism. However, there are some key questions that need to be answered: 1) How does KatG activate INH? 2) What is its active form? 3) How do alterations of KatG gene by mutations lead to interupted INH activation? These are important biomedical questions that relate to the structure/function relationship in catalase-peroxidases. 50 Catalase-peroxidases and Reactive Oxygen Species The need to direct the reactivity of oxygen is esential for survival in an aerobic environment. While nature has gone to great lengths to ensure the interaction of Fe with O 2 is controlled and specific by the storage of Fe in complexes and proteins as pointed out earlier, ROS are stil inevitably generated. A key role for catalases, peroxidases, and catalase-peroxidases is to remove reactive oxygen species (particularly but not exclusively H 2 O 2 ) to prevent the damaging consequences of ROS. The partial reduction of O 2 , starting with O 2 ?? , can lead to the production of H 2 O 2 and eventualy to ?OH if it goes unchecked (reactions 16-19). In this regard, there are two general sources for ROS. The first is ?misfiring? that occur in electron transfer proceses in biology, particularly where Fe O 2 + e - ? O 2 ?? (16) O 2 ?? + H + ? HO 2 ? (17) HO 2 ? + O 2 ?? + H + ? H 2 O 2 + O 2 (18) H 2 O 2 + e - ? OH - + ?OH (19) and O 2 interactions occur. In this respect, hemoproteins provide some examples of these types of ?misfiring?. In some instances, much higher levels of ROS in addition to reactive nitrogen species are produced as part of a host defense against infection by microorganisms. The biological context of catalase-peroxidases can be understood to some degre in terms of these situations where ROS are generated. 51 Misfires in Hemoproteins Reaction The globins are known to generate ROS through an autoxidation reaction (Fig. 1.15) [125]. Autoxidation reactions are found in nature in al oxygen binding hemoproteins [126]. However, there is no clear mechanism for this reaction. The reversible binding of O 2 to the globins first results in a ferous-heme-oxygen complex. This oxygenated form of the protein can form a resonance hybrid with the feric superoxo state. The release of superoxide anion results in the feric form of the globin proteins, which are caled met (e.g., metmyoglobin and methemoglobin) (Fig. 1.15). The heme can eventualy be reduced back to the ferous state. On the other hand, the feric enzyme cannot bind O 2 and is physiologicaly inactive. The rate of autoxidation is influenced by thre factors: partial presure of O 2 , pH, and the diferent globin proteins. Studies on the efect of partial presure of O 2 studies on autoxidation indicate an increase in rate up to a maximum when half the protein is saturated with O 2 [127, 128]. A 2-3 fold decrease in the rate is observed at higher partial presures. High O 2 presure may serve to protect the protein from autoxidation. Changes in pH are also known to afect the rate of autoxidation. At high concentration of both hydrogen and hydroxyl ions, an increase in the rate of autoxidation is observed [127, 129]. Although it is not understood why the increase occurs at high hydroxyl ions, Weis theorizes that at low pH, a lost of hydrogen bonding betwen the distal histidine and Fe II ?O 2 results in the equilibrium protonation of Fe II ?O 2 to Fe II ?O 2 H + [130]. This leads to 52 Figure 1.15. Autoxidation Reaction of Myoglobin and Hemoglobin. Fe II + O 2 Fe II_ O 2 Fe III_ O 2 e - Fe III O 2 53 the formation of the met species. The rate of autoxidation wil also vary betwen diferent globin proteins [131, 132]. The cytochrome P450s catalyze the insertion of an oxygen atom into diferent substrates. Electron transfer to oxygen, which ultimately leads to substrate oxidation, controls this reaction. Uncoupling of the catalytic cycle can result in the production of ROS rather than the target product [133, 134]. One type of uncoupling outcome is the autoxidation decay of the one-electron reduced ternary complex to release superoxide (Fig. 1.16 (i). This is reminiscent of globin autoxidation and comes as a result of slow delivery of the second electron to the oxygen. Superoxide subsequently produces hydrogen peroxide. Alternatively, the protonation of the feri-peroxo complex leads to the release of hydrogen peroxide (j). The presence of a substrate has a substantial efect on the production of ROS by cytochrome P450. Indeed, it has been estimated that in liver microsomal cytochrome P450 as much as 55% of the consumed oxygen appears as hydrogen peroxide in the presence of substrate and 100% in its absence [135]. The presence of substrate serves to increase the rate of the first electron transfer. Cytochrome b 5 may also play a role in the coupling of cytochrome P450. Although not esential for activity, cytochrome b 5 increases the activity of cytochrome P450, which decreases the production of ROS [136, 137]. However, it is not clear whether cytochrome b 5 functions as an electron carier or helps stabilize a particular cytochrome P450 conformation. In any case, cytochromes P450 and others like them can be a source of partialy reduced oxygen species. Autoxidation of cytochromes P450 and the globins are two examples among many of generation of ROS by ?misfiring?. Living organisms must have ways to remove 54 Figure 1.16. Generation of ROS in Cytochrome P450. Cys S Fe III Cys S Fe III Cys S Fe II O 2 Cys S O Fe II O Cys S O Fe III O . RH e - RH RH e - Cys S O Fe III O 2 Cys S O Fe III OH H + H + ,e - Cys S O Fe IV H 2 O RO H a b c d e f g h RH RHRH RH RH H 2 O 2 H 2 O 2 2H + 2H + O 2 . i j 55 these ROS when they are generated. Systems to prevent formation of ?OH are ubiquitous. Superoxide dismutase catalyzes removal of O 2 ? ? to produce O 2 and H 2 O 2 . However, the presence of H 2 O 2 can easily be reduced to ?OH, posing enormous risks. Therefore, catalase or peroxidase is esential for safe removal of partialy reduced oxygen species. Catalase-peroxidases appear to fil this role in numerous bacteria. Although most catalase-peroxidases are found in the cytoplasm, some bacterial have a second catalase-peroxidase located in the periplasmic space [22-24]. This sems to be overkil for bacterial protection against H 2 O 2 . Why would a second catalase- peroxidase be necesary? In each case, this periplasmic catalase-peroxidase is found in a highly pathogenic bacterial species, including Escherichia coli O157:H7, Yersina pestis and Legionela pneumophila. Importantly, the non-pathogenic counterparts to these organisms do not have a periplasmic catalase-peroxidase, indicating that the periplasmic form ay be a virulence factor. Sequence analysis has revealed a signal peptide present in the periplasmic enzymes, which was not found with the cytoplasmic catalase- peroxidases [24, 25, 27]. Periplasmic catalase-peroxidases also show a higher sequence homology with each other than with the cytoplasmic enzymes. The potential role of the periplasmic catalase-peroxidases from (KatP) as a virulence factor is further supported by the following: In E. coli O157:H7, catalase-peroxidase is encoded on the pO157 plasmid - a large plasmid asociated with virulence [24]. In Y. pestis, KatY is expresed corresponding with a temperature shift from 26 ?C to 37 ?C [138]. This corresponds to a transfer of the bacterium from flea to mamalian host. In L. pneumophila, KatA sems to impart characteristics that help it survive and grow inside the macrophage host [23]. The mechanism by which these periplasmic catalase-peroxidases contribute to the 56 virulence of E. coli O157:H7, Y. pestis and L. pneumophila is not understood. It is proposed that periplasmic catalase-peroxidases may serve as the first line of defense against reactive oxygen species (ROS) and reactive nitrogen species (RNS) generated by the imune response [27]. Imune Response Generation of ROS and RNS are a fundamental part of the imune response, in which the body defends itself against microorganisms, viruses, and other foreign substances [139]. ROS are also found to be generated in plants as a defensive response to pathogen atack [140]. In an imune response, neutrophil cels initiate respiratory burst, using hemoproteins to generate ROS and RNS. NADPH oxidase and myeloperoxidase are two hemoproteins used by the imune system. NADPH oxidase, also known as respiratory burst oxidase, is responsible for this proces. NADPH oxidase is a multi-enzyme complex membrane protein that catalyzes the one electron reduction of oxygen to superoxide (O 2 ?? ) [141]. The overal reaction is: Oxidase NADPH + 2 O 2 NADP + + H + + 2 O 2 ?? (20) The cofactor NADPH is required as a source of electrons. The NADPH oxidase transfers an electron from NADPH to O 2 . As a result, O 2 ?? is produced along with 57 oxidized NADP + . O 2 ?? is the first step in the generation of a wide aray of reactive oxidants. The NADPH oxidase complex is comprised of two membrane components: p22 and gp91 (Fig. 1.17). These two components form cytochrome b 58 . Cytochrome b 58 caries its name because in its reduced form, it has a characteristic absorption peak at 558 nm. Cytochrome b 58 is a heterodimer with two hemes. One heme is located within gp91 where O 2 binds while the other heme is shared betwen p22 and gp91 [142]. In anaerobic conditions, the hemes low midpoint potential of ?245mV prevents the direct reduction of O 2 to O 2 ?- by reduced cytochrome [143-145]. However, cytochrome is rapidly oxidized in aerobic conditions [146]. Although a thre-dimensional structure of the cytochrome has yet to be solved, sequence comparisons indicates the presence of a flavin binding site [147]. The NADPH complex is also comprised of thre cytosolic components and a G protein: p67, p47, p40, and rac1 or rac2 (Fig 1.17) [148]. When activated, the cytosolic proteins move to the membrane to form the active NADPH oxidase (Fig 1.17). While gp91 serves as the electron transporter of NADPH oxidase, the other components serve to regulate the transfer of electrons The oxidative burst is only the first step in the antibacterial defense. Indeed, O 2 ?- is a weak microbicidal alone [149]. If O 2 ?- is reduced by a second electron, H 2 O 2 wil be generated (reaction 3). In the imune response, this is acomplished by superoxide dismutase (SOD) (reaction 21). Along with O 2 ?- , H 2 O 2 also has a relatively weak SOD 2 O 2 ?- + 2 H + O 2 + H 2 O 2 (21) 58 Figure 1.17. Schematic Representation of the NADPH oxidase. Inactive Active Membrane rac P67 P40 P47 gp91 p22 O 2 O 2 gp91 P47 P67 rac P40 p22 2O 2 2O 2 ?- Bacteria 59 microbicidal efect and needs to be activated to a more reactive oxidant [150]. One possibility is the generation of ?OH by the 1 e ? reduction of H 2 O 2 , but this is not the principal strategy of neutrophils. Instead, H 2 O 2 is used as an oxidant to generate HOCl. Central to this proces in neutrophils is myeloperoxidase. Myeloperoxidase is a heme- containing peroxidase that reacts with H 2 O 2 (Fig. 1.18). As with other peroxidases, this reaction results in the formation of the feryl-oxo porphyrin/protein radical intermediate compound I. Compound I is then reduced by a halide ion, Cl ? in particular, to generate the corresponding hypohalous acid (i.e., HOCl) and the feric enzyme. Hypochlorous acid is a strong oxidizing agent and can react with many biomolecules leading to the death of the invading cels [150]. Macrophages also generate large quantities of O 2 ?? by way of an oxidative burst mechanism. In these cels, however, there is very litle production of HOCl. Instead, NO? is rapidly generated by inducible nitric oxide synthase (iNOS). The O 2 ?? generated by NADPH oxidase and NO? generate by iNOS react at difusion controlled rates to generate peroxynitrite (ONO ? ) [151, 152]. Peroxynitrite is a highly bactericidal agent, many times more potent than H 2 O 2 [153]. Peroxynitrite is a peroxide and has been shown to interact with peroxidases-type enzymes in a fashion similar to H 2 O 2 . In the proces, ONO ? is reduced to the much les harmful NO 2 ? anion [151, 154]. The placement of an enzyme in the periplasmic space that rapidly consumes H 2 O 2 and ONO ? may provide a distinct selective advantage for pathogens who must encounter activated macrophages and neutrophils from a host organism. The key to survival for microorganisms in host organisms is to have a good defense against the 60 Figure 1.18. Catalytic Cycle of Myeloperoxidase. 61 imune response. The reduction of O 2 ?? by superoxide dismutase is counter productive in producing the more powerful oxidant H 2 O 2 , which regulates the production of HOCl. HOCl is too powerful of an oxidant to control, making the most eficient way to counteract the imune response by neutrophil cels the decomposition of H 2 O 2 . Summary Hemoproteins are important in the esential need to activate oxygen. The presence of heme is critical to the function of hemoproteins. However, there are many factors that contribute to the diversity of hemoprotein activities including the structure of heme, heme iron ligands, and the environment surrounding the heme. The diversity of hemoproteins makes them a good model in the evaluation of the relationship betwen protein structure and function. In this disertation a development of an expresion system for the enhancement of recombinant hemoproteins expresion is used as a tool to aid in the study of that relationship in these important proteins. Catalase-peroxidases are a group of hemoproteins that give bacteria an enhanced capacity to combat ROS. The ability of catalase-peroxidases to use one active site, which is almost identical to monofunctional peroxidases, for two distinct activities also make them an ideal system to study protein structure and function relationship. In this disertation the presence of a periplasmic catalase-peroxidase found in pathogenic bacteria is evaluated with its potential role in bacterial virulence in mind. The role of a 62 thre amino acid covalent adduct is also evaluated so as to beter understand the protein structure and function relationship. 63 CHAPTER TWO MATERIALS AND METHODS Reagents The following compounds or materials, cataloged by source, were obtained. Hydrogen peroxide (30%), peracetic acid, ?-amino levulinic acid, hemin, imidazole, Sephacryl 300 HR, phenyl-Sepharose resin, 2,2 ? -bipyridyl, phenylmethylsulfonyl fluoride (PMSF), ampicilin, chloramphenicol, ferous amonium sulfate, sodium hydrosulfite, trifluroacetic acid, and 2,2 ? -azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) were purchased from Sigma (St. Louis, MO). Isopropyl-?-D-thiogalactopyranoside (IPTG), acetonitrile, kanamycin monosulfate, potasium fericyanide, sodium hydroxide, and tetracycline hydrochloride were obtained from Fisher (Pitsburgh, PA). Bugbuster and benzonase were purchased from Novagen (Madison, WI). Pyridine was purchased from ACROS (New Jersey). Zinc (I) protoporphyrin IX (ZnPPIX) was purchased from 64 Frontier Scientific (Logan, UT). Al restriction enzymes were purchased from New England Biolabs (Beverly, MA) and al oligonucleotide primers were purchased from Invitrogen (Carlsbad, CA). The E. coli strains BL-21 (DE3) pLysS, and XL-1 Blue were obtained from Stratagene (La Jolla, CA) as was Pfu polymerase. Nickel?nitrilotriacetic acid (Ni?NTA) resin was obtained from Qiagen (Valencia, CA). Al buffers and media were prepared using water purified through a Milipore Q-PakII or a Barnstead EASY pure I system (18.2 M?/ cm resistivity). Construction of pHPEX Plasmids A description and abbreviations of plasmids and E. coli strains described in this work are summaries in Table 2.1. The pHPEX1 plasmid was constructed by first amplifying the heme receptor gene (chuA) from E. coli O157:H7 by PCR. This was acomplished using primers HPEX a01 (TA CA AC AT GC AT TGC AG GA TA CGC) and HPEX a02 (GTG TGT AG AT TCA TG TCT CTC TC T CA G) with internal restriction endonuclease recognition sequences (italics) for NcoI and EcoRI, respectively. The PCR product included the entire coding sequence plus 295 bp upstream of the ATG start codon and 50 bp downstream of the TA stop codon. The chuA PCR product and the low-copy number plasmid, pACYC184, were digested with NcoI and EcoRI and isolated by agarose gel electrophoresis. Digested products were excised and extracted from the gel by Qiagen QIAquick gel extraction kit protocol. In 65 Table 2.1 Plasmids and strains used for the development of the hemoprotein expresion (HPEX) system. Genotype Strain Size ( bp ) E.coli B F - dcm ompT hsdS ( r B- m b- ) gal (DE3) BL-21 (DE3) [ pLysS Cam r ] BL-21 (DE3) [pHPEX2 Tet r ] BL-21 (DE3) [pHPEX3 Tet r ] BL-21 (DE3) [ pHPEX-fur Tet r ] BL-21(DE3) BL-21 (DE3) pLysS BL-21 (DE3) pHPEX2 BL-21 (DE3) pHPEX3 BL-21 (DE3) pHPEX-fur 624 547 540 6243 5765 chuA inserted, Cam r interrupted lac UV5 inserted 5 ?to chuA T7 lyzozyme (lysS ) inserted fur inserted 5 ?to chuA E. coli KatG inserted (C-term 6 His-tag) pACYC184 pHPEX1 pHPEX2 pHPEX1 pET20b(+) pHPEX1 pHPEX2 pHPEX3 pHPEX-fur pKatG1 Modifications Starting construct Plasmid 66 order to ligate DNA together, T4 DNA ligase was added to extracted DNA and incubated at 37?C for 1 hr. acording to standard procedures [155]. The ligation products were then used to transform E. coli (XL-1 Blue). Transformants were selected on the basis of tetracycline resistance. Plasmid isolated from candidate colonies was analyzed by AvaI and BsHI restriction digest to ensure the correct construction of pHPEX1. The entire chuA gene was sequenced to ensure that no acidental mutations occurred during the procedure. The pHPEX2 plasmid was constructed from pHPEX1. The natural chuA upstream regulatory elements present in pHPEX1 were replaced by the lacUV5 promoter using the Seamles Cloning TM procedure (Stratagene). This protocol incorporates the restriction enzyme recognition sequence for Eam1104 I (5?-CTCTC-3?). Eam1104 I is a type IS restriction enzyme that cuts outside its recognition sequence as ilustrated below. It cuts one nucleotide downstream of the recognition sequence on one strand and four nucleotides on the complementary strand, leaving a thre nucleotide overhang that can be designed by the user for ligation. During amplification, 5-methyldeoxycytosine is included so that it is incorporated into the newly synthesized DNA. This serves to protect already-existing Eam1104 I sites against cleavage and ensure that only the restriction sites in the primers are hydrolyzed by Eam1104 I. The portion of pHPEX1 to be retained for pHPEX2 was amplified using primers HPEX a05 (TG AGA CTC TC ATG TCA CGT CG CA TT AC TCG) 5?-CTC T C N N N-3? 3?-GAG AG N N N-5? 67 (Eam1104 I recognition site in italics) and HPEX a06 (CGT GT CTC TC GA GG TC AC TCT CTG TG C). The lacUV5 promoter was amplified from the pBT plasmid using primers lacUV5 a00 (GT TC TT CTC TC AC ATA CGC TGT TC TG TGT GA A) and lacUV5 a01 (GT AC TCT CT CAC TCA GT GT AGC TCA GAG AC ). Both PCR products were hydrolyzed with Eam1104 I and ligated together with T4 DNA ligase. The ligation products were then used to transform E. coli (XL-1 Blue). Transformants were selected on the basis of tetracycline resistance. Plasmid isolated from candidate colonies was analyzed by ClaI restriction digest to ensure the correct construction of pHPEX2. The entire chuA gene and lacUV5 promoter were sequenced to ensure no acidental mutations occurred during PCR amplification. The pHPEX3 plasmid was constructed by incorporating the gene for T7 lysozyme into the pHPEX2 plasmid. This was also acomplished by Seamles Cloning TM . The portion of pHPEX2 to be retained for pHPEX3 was amplified by the primers HPEX a07 (GA CG GC GA CTC TC A AA GAT GC AG AG ATA C) and HPEX a08 (CAC AT CT GCT CT CTG G AT GCT CAT CG A TC). The T7 lysozyme gene was amplified from the pLysS plasmid using the primers T7 LYS a01 (GTC TG TCT CT CGC C AT GC TGC TC CA CA) and T7 LYS a00 (GAG AT CG CTC TC T TGA TAG AT AA AG A AG AG). Both PCR products were hydrolyzed with Eam1104 I and ligated together with T4 DNA ligase. The ligation products were then used to transform E. coli (XL-1 Blue). Transformants were selected on the basis of tetracycline resistance. Plasmid isolated from candidate colonies was analyzed by BsaI and BsHI restriction digest to ensure the correct construction of pHPEX3. 68 Finaly, the pHPEX-fur plasmid was constructed by incorporating the iron uptake regulator into the pHPEX1 plasmid. This was acomplished using primers FURINS 1(CAT GA TA TA TC TCA TA TA) and FURINS 2 (CAT GTA TA TGA GA TA TA TC), which contain the fur box and an NcoI overhang. The FURINS primers were annealed together by first heating equal quantities together at 90 ?C for 5 min. and then cooling to 37 ?C for 5 min. The duplex was then stored at 4 ?C until used. The pHPEX1 plasmid was digested with NcoI. Both products were ligated together with T4 DNA ligase. The ligation products were then used to transform E. coli (XL-1 Blue). Transformants were selected on the basis of tetracycline resistance. Plasmid isolated from candidate colonies was analyzed by Bst BI and Nco I/Kpn I restriction digest to ensure the correct construction of pHPEX-fur. DNA sequencing was performed to ensure that the fur box was inserted once and in the proper orientation. Expresion of KatG and Myoglobin Expresion of KatG for Heme Evaluation The pET-based plasmid pKatG1 for the expresion of recombinant histidine- tagged E. coli catalase?peroxidase was used to transform two E. coli strains, BL-21 DE3 pLysS and BL-21 DE3 pHPEX3, and transformants were selected based on ampicilin or tetracycline resistance. Expresion of KatG was caried out in Luria?Bertani broth (2 L). Cels were grown to mid-log phase (OD 60 =0.5) and expresion of KatG was induced by 69 addition of IPTG (1 mM). At the time of induction, KatG cultures were supplemented with 8 ?M hemin. Cultures were grown at 37 ?C with constant shaking for four hours. At four hours post-induction, cels were harvested by centrifugation (13,000g), and the cel pelets were stored at ?80 ?C until purification. To evaluate KatG expresion, a quantity of cels suficient to yield a 0.05 OD 60 reading (when diluted to 1 ml) was treated with an equal volume of 10% trichloroacetic acid (4 ?C) and centrifuged in a microcentrifuge at maximum rcf for 5 min. The supernatant was discarded and the pelet was washed with 1 ml acetone. The pelets were then dried and resuspended in SDS?PAGE loading buffer, adjusting the pH as appropriate with trizma base. Samples were then separated by SDS?PAGE using a 7.6% acrylamide resolving gel. Cel pelets were resuspended in Bugbuster (Novagen) reagent supplemented with benzonase (250 U) and PMSF (0.1 mM). The cel lysate was then centrifuged at 16,000g. The supernatant was loaded onto a Ni?NTA column by recirculating the solution at 1 ml/min through the column bed overnight. The column was then washed with 50 mM phosphate buffer, pH 8.0/200 mM NaCl/2 mM imidazole. Protein was eluted with 50 mM phosphate buffer, pH 8.0/200 mM NaCl/100 mM imidazole. Expresion of Myoglobin for Heme Evaluation The pET-based plasmid pMb1 for the expresion of recombinant sperm whale myoglobin (pMb1) was generously provided by the laboratory of John S. Olson at Rice 70 University and was used to transform four E. coli strains (BL-21 DE3, BL-21 DE3 pLysS, BL-21 DE3 pHPEX3, and BL-21 DE3 pHPEX-fur). Transformants were selected based on kanamycin or tetracycline resistance as appropriate. Expresion of Mb was caried out in Luria?Bertani broth (10 mL). Cels were grown to mid-log phase (OD 60 =0.5), and expresion of Mb was induced by addition of IPTG (1 mM) or 2,2?- bipyridine (0.15 mM). At the time of induction, Mb cultures were supplemented with 8 ?M hemin. Cultures were grown at 37 ?C with constant shaking for four hours. At four hours post-induction, cels were harvested by centrifugation (13,000g), and the cel pelets were stored at 4 ?C. To evaluate Mb expresion, a quantity of cels sufficient to yield a 0.05 OD 60 reading (when diluted to 1 ml) was treated with an equal volume of 10% trichloroacetic acid (4 ?C) and centrifuged in a microcentrifuge at maximum rcf for 5 min. The supernatant was discarded and the pelet was washed with 1 ml acetone. The pelets were then dried and resuspended in SDS?PAGE loading buffer, adjusting the pH as appropriate with trizma base. Samples were then separated by SDS?PAGE using a 12% acrylamide resolving gel. 71 Cloning and Expresion of KatP and KatG Y26F Cloning of KatP Construction of the KatP plasmids was performed in our lab by Kristen Hertwig with the asistance of Dr. Douglas Goodwin. The katP gene was amplified from pO157 plasmid isolated from E. coli O157:H7 using primers O157a02 (CT CC TCA GT CTC GAG TT AT GT TA ATC AA CG ATC) and O157a03 (CC TA TC CG AGA TCT CA TAT GAT AA A AC TCT TC). The resulting PCR product and pET20b(+) were digested with BglI and XhoI, and the fragments were separated by agarose gel electrophoresis. The bands were excised from the gel and extracted acording to standard procedures [155]. The DNA fragments were then ligated and used to transform E. coli (XL-1 Blue). Transformants were selected based on ampicilin resistance and screned by restriction digests. The resulting construct (pKatP1) was used as a template for site-directed mutagenesis by the Quik-Change TM procedure (Stratagene, La Jolla, CA). The primers O157m01(+) (CG AC ATA CAG GAC TA TGA TG CG G) and O157m01(?) (CC GC AT CAT AG TC TGT ATG TC G) were used to make a silent mutation to the codon for Thr 110 and thus remove an internal NdeI restriction site. The resulting PCR product was treated with DpnI and used to transform E. coli (XL-1 Blue) by electroporation. Transformants were selected by ampicilin resistance and candidate plasmids were screned by restriction digest with NdeI. The resulting construct (pKatP2) and pET20b(+) were digested with NdeI and XhoI and the fragments were separated by agarose gel electrophoresis. The 72 bands corresponding to the katP gene and the pET plasmid were excised from the gel, extracted, and ligated together as described above. The resulting construct (pKatP3) was used to transform E coli (XL-1 Blue). Transformants were screned by diagnostic restriction digests and candidate plasmids were subjected to DNA sequence analysis to ensure that no acidental mutations had acumulated during the cloning procedure. The final plasmid (pKatP3) was constructed such that the protein would be expresed with a C-terminal six-histidine tag. Expresion of KatP The pKatP3 plasmid was used to transform E. coli (BL-21 DE3 pLysS), and transformants were selected based on ampicilin resistance. Expresion of KatP was caried out in Luria?Bertani broth (2L). Cels were grown to mid-log phase (OD 60 =0.5) and expresion of KatP was induced by addition of IPTG (1 mM). At the time of induction, KatP cultures were also supplemented with ?-amino levulinic acid (0.5 mM) and ferous amonium sulfate (0.5 mM). Cultures were grown at 37 ?C with constant shaking for four hours. At four hours, post-induction cels were harvested by centrifugation (13,000g), and the cel pelets were stored at ?80 ?C until purification. To evaluate KatP expresion, a quantity of cels suficient to yield a 0.05 OD 60 reading (when diluted to 1 ml) was treated with an equal volume of 10% trichloroacetic acid (4 ?C) and centrifuged in a microcentrifuge at maximum rcf for 5 min. The supernatant was discarded and the pelet was washed with 1 ml acetone. The pelets were 73 then dried and resuspended in SDS?PAGE loading buffer, adjusting the pH as appropriate with trizma base. Samples were then separated by SDS?PAGE using a 7.6% acrylamide resolving gel. Purification of KatP Cel pelets were resuspended in Bugbuster (Novagen) reagent supplemented with benzonase (250 U) and PMSF (0.1 mM). The cel lysate was then centrifuged at 16,000g. The supernatant was loaded onto a Ni?NTA column by recirculating the solution at 1 ml/min through the column bed overnight. For KatP, the column was then washed with Buffer A (50 mM phosphate buffer, pH 8.0; 200 mM NaCl) supplemented with 2 mM imidazole. A second wash was then performed with Buffer A supplemented with 20 mM imidazole. Finaly, protein was eluted off the column with Buffer A supplemented with 200 mM imidazole. Exces imidazole was then removed either by dialysis against Buffer A lacking imidazole or by gel filtration chromatography. This initial purification procedure resulted in two-KatP proteins one of which contained heme and displayed peroxidase and catalase activities. The other lacked heme and showed no activity. We were able to isolate the active fraction using either a phenyl- sepharose column or FPLC anion exchange chromatography. For anion exchange, protein from the initial purification of KatP was loaded onto a UNO Q1 column (Bio- Rad, Hercules, CA). Separation was acomplished using a two buffer system: 50 mM Tris, pH 8.1 (Buffer I), and 50 mM Tris, pH 8.1/375 mM, NaCl (Buffer I). The initial 74 buffer condition of the column was 100% Buffer I/0% Buffer I. Following a 0.45 min isocratic period, bufer conditions were ramped to 67% Buffer I/33% Buffer I over 0.58 min. The buffer composition was then changed to 50% Buffer I/50% Buffer I over 8 min and then up to 40% Buffer I/60% Buffer I over the next 1.25 min. Finaly, 0% Buffer I/100% Buffer I was pased over the column for 0.75 min, followed by 100% Buffer I/0% Buffer I for 2 min. For separation using phenyl-sepharose, amonium sulfate was added to partialy purified KatP up to 20% saturation. This solution was used to load a phenyl-sepharose column. Following a wash step, active KatP was eluted from the column by a linear gradient from buffer 20% saturated with amonium sulfate to bufer without amonium sulfate. Fractions were evaluated for catalase activity and active fractions were analyzed for purity by SDS?PAGE. Pure and active fractions were collected and concentrated. Cloning and Expresion of KatG Y26F KatG Tyr 226 was mutated to Phe using a ?Quik Change? procedure from Stratagene. Primers ECPm 23(+) (GAG ATG GT CTG ATC TC GT AC C) and ECPm 23(-) (GG TA CG AG ATC AGA CC ATC TC) were used to produce the Y226F substitution. Nucleotide substitutions designed to produce the mutation are underlined. The resulting PCR product was digested with Dpn I to remove the starting template and then used to transform E. coli (XL-1 Blue) by electroporation. Transformants were selected by ampicilin resistance and candidate plasmids were 75 evaluated by Hpa I digest. Primers were designed to make a unique restriction site making the digest possible. Positive candidates were subjected to DNA sequence analysis to ensure that no unintended mutations had acumulated during the site-directed mutagenesis procedure. Plasmids verified to cary the correct mutation were used to transform E. coli (BL-21 [DE3] pLysS). Expresion of Y226F was acomplished by the same method as wtKatG except that some of the protein was expresed in inclusion bodies. Expresion of some proteins at high rates can alow for aggregation due to hydrophobic efects before proper folding leading to formation of inclusion bodies. Expresion was caried out at lower temperature (18 ?C) in order to slow down expresion and alow more protein to be produced in their native conformation. Protein was expresed for six hours following induction with IPTG. Purification of Y226F was caried by the same procedures as wtKatG. Protein Characterization UV?Visible Absorption Spectra Spectral recordings of catalase-peroxidase preparations were caried out by scanning betwen 700 and 250 nm using Gilford Response-I or Shimadzu UV 1601 UV? V is spectrophotometer. Al spectral measurements were caried out at room temperature in 100 mM phosphate buffer, pH 7.0. The molar absorptivity of catalase-peroxidase (143 mM ?1 cm ?1 ) at 280 nm was estimated acording to the method of Gil and von Hippel 76 [156]. The heme concentration was estimated by the pyridine hemichrome asay [157]. Briefy, to ensure that al the heme iron was in its oxidized (feric) state, 50 mM potasium fericyanide was added to heme/hemoprotein in 20% pyridine/ 0.1M NaOH solution. Feric solution was used for baseline. The heme iron was then reduced to its ferous form by addition of sodium hydrosulfite (~5mg). A spectrum was then recorded from 560-530 nm. The maximum absorption diference from the spectrum (typicaly Abs at 555 nm ? Abs at 540 nm) was divided by the appropriate molar absorptivity (??20.7 mM ?1 cm ?1 ) to calculate heme concentration. Feri-cyano catalase-peroxidase was prepared by addition of 2 mM NaCN to feric catalase-peroxidase. Ferous catalase- peroxidase was obtained by addition of dithionite to feric catalase-peroxidase. The so- caled Reinheitszahl (RZ) values were calculated with feric catalase-peroxidase. This entails measuring the ratio of absorbance due to the heme (408 nm) over that of the protein (280 nm). Catalase and Peroxidase Activity Catalase activity was evaluated by monitoring the decrease in H 2 O 2 concentration with time at 240 nm. The molar absorptivity of H 2 O 2 at 240 nm is 39.4 M ?1 cm ?1 [158]. Unles otherwise indicated, al asays were caried out at 23 ?C in 100 mM phosphate buffer, pH 7.0. Al activities were normalized on the basis of heme content as determined by the pyridine hemichrome asay [157]. 77 Peroxidase activity was evaluated by monitoring the production of 2,2 ? -azino- bis(3-ethylbenzthiazoline-6-sulfonic acid) radical (ABTS ?+ ) over time at 417 nm. The peroxidase activity of catalase-peroxidases generates the one-electron oxidation product of ABTS (ABTS ?+ ) (Fig. 2.1). The molar absorptivity of ABTS ?+ at 417 nm is 34.7 mM ?1 cm ?1 [159]. Al asays were caried out at 23 ?C in 50 mM acetate buffer, pH 5.0. pH-Dependence of Catalase-Peroxidase Activity The pH profile for catalase activity of KatP was measured from pH 6.0 to 8.0 by the decrease in H 2 O 2 concentration with time at 240 nm. The efect of pH on peroxidase activity for KatP was measured by monitoring the production of ABTS +? over time at 417 nm betwen pH 4.0 and 6.0. Al activities were normalized based upon heme content as determined by the pyridine hemichrome asay [157]. Transient-State Kinetics Al stopped-flow measurements were performed using an SX.18MV Rapid Reaction Analyzer from Applied Photophysics (Surey, UK) in single-mixing mode. Full spectral recordings of feric KatP reaction with peracetic acid or cyanide were obtained by diode aray. Rate constants for either reaction were determined by single-wavelength 78 Figure 2.1. Oxidation of ABTS. N S N N S N SO 3 O 3 S N S N + N S N SO 3 O 3 S 79 measurements at 403 nm. In either case, one syringe contained 2 ?M holo KatP as determined by heme content [2, 157]. The second syringe contained varying concentrations of peracetic acid or potasium cyanide. Reactions were caried out in 100 mM phosphate buffer, pH 7.0. The temperature for al reactions was maintained at 25 ?C using a circulating water bath. In Gel Digestion of KatG Apo and holo KatG protein was separated using sodium dodecyl sulfate (SDS)- polyacrylamide gel electrophoresis (PAGE) with 7.6% (w/v) acrylamide gel. In gel digestion of protein was achieved by adapting methods described by Helman and Rosenfeld [160, 161]. Protein was excised from the gel, and gel pieces were diced in a 1.7 ?l microfuge tube. To gel pieces, 200 ?l of 25mM NH 4 HCO 3 / 50% acetonitile was added and vortexed for 10 min to remove Coomasie stain from protein. This was repeated twice each time discarding the supernatant. Gel pieces were dried completely in a Savant SC110 Speed Vac. Trypsin (1?g) was added to gel pieces followed by incubation on ice to rehydrate. Trypsin is a serine protease, which cleaves peptide bonds on the carboxyl side of basic amino acid residues like lysine or arginine in proteins (Fig. 2.2). Trypsin was alowed to digest the protein in gel by incubating at 37 ?C for 4- 8 hrs. Digested protein was then extracted from gel by adding 50% acetonitrile/ 5% formic acid and vortexing for 20-30min. This mixture was sonicated, and the proces was repeated. Extracted digest volume was reduced to the appropriate volume (10-50?l). 80 Figure 2.2. Trypsin digest of peptide bond in protein. H 2 N Trypsin CH C CH 2 N H O OH N H CH C CH 2 N H O C O O CH COO- CH 2 SH H 2 N CH C CH 2 N H O OH H 2 N CH C CH 2 N H O C O O CH COO- CH 2 SH Ser Arg CysAsp Ser Asp Cys CH C CH 2 OH O CH 2 CH 2 NH C NH 2 NH Arg CH C CH 2 O CH 2 CH 2 NH C NH 2 NH 81 Digested samples were prepared for MALDI using 2, 5-dihydroxybenzoic acid (DHBA) as the matrix. For sample preparation, 1.0 ?l of digested solution was mixed with 1.0 ?l of matrix (160 mg/ml DHBA in water: acetonitrile 3:1, 2% formic acid). The samples (1 ?l) were loaded onto a Bruker stainles stel Micro SCOUT plate. Samples were air dried at room temperature. Peptide Mass Identification Using MALDI-TOF MS Mas spectrometric measurements of KatG after trypsin digestion were caried out on a Bruker Microflex Matrix-Asisted Laser Desorption-Ionization Time of Flight mas spectrometer (MALDI TOF-MS). The spectrometer was equipped with a microSCOUT ion source, a clas II B ultraviolet laser, a 2-GHz digitizer, and a multichannel plate detector. Measurements were caried out in a positive ionization mode using a reflector voltage of 24 kV. Mas spectra were asembled as a sum of ion signals acquired from 50-100 laser pulses. Spectra were calibrated using Bruker peptide calibration standard as a reference. Peptide mases from mas spectra were searched against SwisPort 45.5 protein database. Electron Paramagnetic Resonance Spectroscopy 82 EPR spectra of KatP were recorded on a Bruker EMX spectrometer equipped with an Oxford ESR 900 cryostat and ITC temperature controller. Spectrometer setings were as follows: temperature, 10 K; microwave frequency, 9.38 GHz; microwave power, 0.10 mW; receiver gain, 6.32? 10 4 ; modulation frequency, 100 kHz; modulation amplitude, 10 G; conversion time, 163.84 ms; and time constant, 163.84 ms. 83 CHAPTER THRE RESULTS HPEX System The pHPEX Plasmids A low-copy number plasmid, pACYC184, was selected as the basis for the pHPEX constructs to avoid competition for replication machinery by the plasmids most commonly used for high-level protein expresion. These expresion plasmids often cary the Col E1 replicon; pACYC184 caries the p15A replicon. The plasmid pHPEX1 (Fig. 3.1 A) was constructed to contain the gene for a heme receptor (chuA). The chuA gene was inserted into pACYC184 betwen the NcoI and EcoRI restriction sites, efectively interupting the chloramphenicol-resistance gene. The pHPEX1 plasmid imparts tetracycline resistance as do al of the pHPEX plasmids constructed to date. The pHPEX2 plasmid was constructed from pHPEX1 by the addition of the lacUV5 promoter 84 Figure 3.1. Schematic representation of the plasmids pHPEX1 (A), pHPEX2 (B), pHPEX3 (C), and pHPEX-fur (D). 85 upstream of chuA (Fig. 3.1 B), alowing for the control of chuA expresion levels by the addition of lactose analogs (e.g., IPTG). The pHPEX3 plasmid (Fig. 3.1 C) was constructed by inserting the gene for T7 lysozyme into pHPEX2. The T7-based expresion system, which uses T7 RNA polymerase to transcribe genes downstream of the T7 promoter, is commonly used to expres recombinant proteins in E. coli [162]. The expresion of a smal amount of T7 lysozyme inhibits low levels of T7 RNA polymerase produced by leakage expresion under the lacUV5 promoter [162]. This prevents the premature expresion of the target recombinant protein and is helpful for the T7 RNA polymerase-based expresion of proteins otherwise toxic to E. coli. The pHPEX-fur plasmid (Fig. 3.1 D) was also constructed from pHPEX1 by the addition of a iron uptake regulator sequence upstream of chuA. The fur box alows for the control of chuA expresion by iron limitation. Diagnostic restriction endonuclease cleavage sites are denoted in the exterior of the plasmids (Fig. 3.1) as is the origin of replication (ori) and the position of the tetracycline resistance marker (tet r ). Diagnostic restriction digests and DNA sequence analysis confirmed the correct construction of al pHPEX plasmids. The plasmid pHPEX1 was digested with restriction endonuclease Eco RI/Nco I, Eco RI alone, and Ava I/Bs HI (Fig. 3.2 A). The plasmid pHPEX2 was digested with Cla I (Fig. 3.2 B). Plasmid pHPEX3 was digested with Cla I, Ava I/Bs HI, and Eco RI (Fig. 3.2 C). Plasmid pHPEX-fur was digested with Bst BI (Fig. 3.2 D). 86 Figure 3.2. Diagnostic restriction digests of pHPEX1 (A), pHPEX2 (B), pHPEX3 (C), and pHPEX-fur (D). Incomplete plasmid digestion products denoted by arows and marker denoted by M. 87 Production of ChuA by HPEX Cels To verify that transformation with an HPEX plasmid led to synthesis of ChuA (the heme receptor), sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS- PAGE) was used to compare proteins produced by untransformed and pHPEX2- transformed BL-21 (DE3) cels under various culture conditions (Fig. 3.3). A protein corresponding to the monomeric molecular weight of ChuA in crude lysates of BL-21 (DE3) pHPEX2 cels grown in LB supplemented with 0.15 mM 2,2?-bipyridine and 8 ?M hemin was observed (Fig. 3.3, Lane 1). Modest increases in the intensity of this band were observed with the inclusion of IPTG in the culture medium (Fig. 3.3, Lane 2). Conversely, this protein was never detected in the cel lysates of untransformed BL-21 DE3 cels grown in LB supplemented with 0.15 mM 2,2?-bipyridine and 8 ?M hemin, regardles of the addition of 0.2 mM IPTG (Fig. 3.3, Lane 3 and 4). Function of ChuA by HPEX Cels To determine if the protein observed by SDS-PAGE was functional as a heme receptor, the abilities of untransformed and pHPEX2-transformed BL-21 (DE3) cels to grow in iron-deficient media supplemented with hemin were compared. BL-21 DE3 was unable to grow in iron-deficient media either in the presence or absence of added hemin (Fig. 3.4). Conversely, addition of hemin to iron-deficient BL-21 (DE3) pHPEX2 cultures resulted in a substantial increase in growth rate, demonstrating the expresion 88 Figure. 3.3. Heme receptor expresion by untransformed (lanes 3 and 4) and pHPEX2- transformed (lanes 1 and 2) E. coli BL-21 (DE3). Cultures were grown in the presence (lanes 2 and 4) and absence of IPTG (lanes 1 and 3). 89 Figure 3.4. Growth of untransformed BL-21 (DE3) cels in minimal media supplemented with 2,2?-bipyridine alone (squares), 2,2?-bipyridine and hemin (circles), 2,2?-bipyridine, hemin, and IPTG (triangles). 468101214 0. 0.2 0.4 0.6 0.8 1.0 1.2 OD 60 Time (h) 90 of an active ChuA protein (Fig. 3.5). On the other hand, supplementing cultures of BL- 21 (DE3) pHPEX2 with IPTG only resulted in slight increases in growth. The results of SDS?PAGE suggest that although addition of IPTG results in modest increases in expresion of chuA, a substantial quantity of the heme receptor sufficient to support growth is produced in the absence of added inducing agents. Efect of HPEX System on KatG In order to evaluate if hemin internalized using chuA was incorporated into a target recombinant protein, KatG a heme dependent catalase-peroxidase was expresed using the HPEX system. KatG expresed with BL-21 (DE3) pHPEX3 showed a large increase in heme content when hemin was added to the medium compared to cultures where hemin was withheld from the media as evaluated by UV-Visible absorption spectra (Fig. 3.6). KatG expresed with BL-21 (DE3) pLysS resulted in only a minor increase in heme incorporated (Fig. 3.7). The efect of expresion strain and culture conditions on the catalase activity of isolated KatG was also evaluated. Consistent with the UV?Visible spectra, the highest activity was observed with protein isolated from BL-21 (DE3) pHPEX3 cels where hemin was added to culture at the time of induction (Fig. 3.8). Very litle activity was observed for KatG isolated from these cels if hemin was not added. For KatG isolated from BL-21 (DE3) pLysS, addition of hemin at time of induction did result in a modest increase in the active of KatG (Fig. 3.8) commensurate with the observed increase in 91 Figure 3.5. Growth of pHPEX2-transformed BL-21 (DE3) cels in minimal media supplemented with 2,2?-bipyridine alone (squares), 2,2?-bipyridine and hemin (circles), 2,2?-bipyridine, hemin, and IPTG (triangles). 468101214 0. 0.2 0.4 0.6 0.8 1.0 1.2 OD 60 Time (h) 92 Figure 3.6. UV-Visible absorption spectra of catalase-peroxidase expresed in pHPEX3- transformed cels. Cultures were grown in the presence (a) and absence (b) of hemin. 30350404505050 . 0.2 0.4 0.6 0.8 1.0 1.2 Rel Abs a b Wavelngth (nm) 93 Figure 3.7. UV-Visible absorption spectra of catalase-peroxidase expresed in pLysS- transformed cels. Cultures were grown in the presence (a) and absence (b) of hemin. 30350404505050 . 0.2 0.4 0.6 0.8 1.0 1.2 Rel Abs a b Wavelngth (nm) 94 Figure 3.8. Catalase activity of KatG expresed in pLysS- (circles) and pHPEX3- transformed (squares) cels. Cels were grown in the presence (closed symbols) and absence (open symbols) of hemin. 0.0.40.80.120.160.2 2 4 6 8 10 [H 2 O 2 ] consumed (M) Time (in) 95 heme content (Fig. 3.7). Without added hemin, very litle catalase activity was detected regardles of the strain used for KatG expresion. KatG showed similar k cat values when expresed with both pHPEX3 and pLysS. Likewise, the K M was comparable for both (Table 3.1). Heme absorption spectra at diferent iron states was examined to identify at diferences in KatG expresed with pHPEX3 versus pLysS (Table 3.2). Only very minor diferences in absorption characteristics were identified. Similarly, diference spectra (feric minus feri-cyano) for KatG from the two expresion systems were virtualy superimposable (Fig. 3.9). These comparisons were done on a per-heme basis to ensure that hemin supplementation produced KatG with the same properties as those obtained when heme was synthesized de novo by the expresion strain. Expresion of Recombinant Myoglobin using the HPEX System. Sperm whale myoglobin was expresed with the HPEX system to evaluate the ability of the system for enhancing expresion of other recombinant hemoproteins in their holo form. UV-visible absorption spectra for expresed myoglobin with the HPEX system showed a large holomyoglobin as determined from lysed cels (Fig.3.10). In order to insure the heme content was a result of heme incorporation during expresion and not adventitious uptake after cels were lysed in the presence of exces heme, UV- visible absorption spectra of whole cels containing expresed myoglobin were performed. Derivatives of the raw UV-visible absorption spectra were performed to 96 Table 3.1. Catalase kinetic parameters for recombinant KatG from hemin- supplemented cultures of BL-21 (DE3) pHPEX3 and unsupplemented cultures of BL-21 (DE3) pLysS. 6.8 2.6 x 10 6 2560 7.0 2.2 x 10 6 422 pHPEX (+ hemin) pLysS (no hemin) K M k cat /K M Sp. Act. (mM) (M -1 s -1 ) (U/mg) Culture condition 97 Table 3.2. Heme absorption maxima for recombinant KatG from pHPEX3- and pLysS-transformed E. coli BL-21 (DE3). pHPEX 406 ? ? 497 640 pLysS 406 ? ? 497 637 pHPEX 424 542 ? ? ? pLysS 424 540 ? ? ? pHPEX 440 560 590 ? ? pLysS 440 559 590 ? ? Feric Feric-CN Ferous Cel type Absorption band maxima (nm) Soret ? ? CT2 CT1 Heme State 98 Figure 3.9. Feric minus feri-cyano diference spectra for recombinant KatG expresed in pHPEX3- (solid line) and pLysS-( dashed line) transformed systems. 30350404500506065070 -1.2 -0.9 -0.6 -0.3 0. 0.3 0.6 0.9 1.2 Wavelngth (nm) 99 Figure 3.10. UV-Visible absorption spectra for myoglobin expresed in pHPEX3-transformed cels (following lysis). 400500600 0.0 0.05 0.10 Wavelngth (nm) 100 acent the signals due to holomyoglobin. Myoglobin expresed with BL-21 (DE3) pHPEX3 induced with IPTG showed a strong signal corresponding to holo myoglobin provided that hemin was added to culture medium. If hemin was withheld from the media a much weaker signal was observed (Fig. 3.11). Myoglobin expresed in BL-21 (DE3) pLysS generated litle or no holomyoglobin even when hemin was added (Fig. 3.12). Likewise, the expresion of holomyoglobin was enhanced in BL-21 (DE3) pHPEX-fur cels in comparison to a standard E. coli expresion strain [BL-21 (DE3)] (Fig. 3.13 and Fig. 3.14). Using dipyridyl, to simulate iron-limitation conditions and induce chuA expresion, and using IPTG to induce expresion of myoglobin, the pHPEX- fur-transformed cels showed a strong derivative spectrum corresponding to holo myoglobin when hemin was present in the growth medium. When hemin was excluded from the growth medium, a much weaker derivative spectrum corresponding to holo myoglobin was observed. To insure that the heme signal resulted only from heme incorporated into Mb, spectra of non-induced cultures with hemin added were also compared for each cel line. 101 Figure 3.11. Derivative UV-Visible absorption spectra for myoglobin expresed in pHPEX3-transformed cels (whole cels). 350400450500 -0.05 0.0 0.5 0.10 IPTG alone IPT + Hem Hem alone Wavelngth(nm) 102 Figure 3.12. Derivative UV-Visible absorption spectra for myoglobin expresed in pLysS-transformed cels (whole cels). 350400450500 -0.05 0.0 0.5 0.10 IPTG alone IT + Hem Hem alone Wavelngth(nm) 103 Figure 3.13. Derivative UV-Visible absorption spectra for myoglobin expresed in pHPEX-fur-transformed cels (whole cels). 350400450500 -0.05 0.0 0.5 0.10 IPTG alone IPT + Hem Hem alone Wavelngth(nm) 104 Figure 3.14. Derivative UV-Visible absorption spectra for myoglobin expresed in BL-21(DE3)-transformed cels (whole cels). 350400450500 -0.05 0.0 0.5 0.10 IPTG alone IT + Hem Hem alone Wavelngth(nm) 105 KatP Expresion and Purification of KatP The katP gene was cloned into the pET20b(+) vector to produce the pKatP3 expresion plasmid, and an E. coli expresion strain (BL-21 DE3 pLysS) was transformed with the construct as described in Methods. Addition of IPTG to these cels at mid-log phase resulted in the production of a protein at high levels corresponding to the expected molecular weight of mature KatP (Fig. 3.15). However, the expresion of another protein was also observed at comparable levels, and the apparent diference in molecular weight betwen the two was about 3 kDa. Both expresion products co-purified when nickel afinity procedures were used, suggesting that both represented some form of recombinant KatP (Fig. 3.15 B, Lane 1). This was confirmed by tryptic digests and mas spectrometric analysis of each band from SDS?PAGE. Each band was determined to be a single protein and each one was identified as KatP. These initial purification products were subjected to anion exchange FPLC. Catalase activity measurements, UV?Vis absorption spectra, and SDS?PAGE analysis of the fractions from FPLC revealed that heme and catalase activities were asociated with the protein of lower apparent molecular weight (Fig. 3.15 B). The complete isolation of KatP corresponding to the lower band was also acomplished using phenyl-Sepharose- based chromatography (Fig.3.15 B, Lane 2). Our analysis of the amino acid sequence of KatP by SignalP 2.0 and that of another group using a diferent program suggested the presence of an N-terminal 106 Figure 3.15. SDS-PAGE electrophoretic separation of total celular protein (A) from BL- 21 (DE3)pLysS transformed pKatP3 at time of induction with 1 mM IPTG (lane 0), 1 h post-induction (lane 1), 2 h post-induction (lane 2), 3 h post- induction (lane 3), and 4 h post-induction (lane 4). (B) Purification of KatP by Ni-NTA chromatography (lane 1) and phenyl-Sepharose chromatography (lane 2). No heme or activity asociated Heme and activity asociated M 0 1 2 3 4 1 2 A B 107 sequence targeting KatP to the periplasmic space [24, 163]. The point of cleavage for the N-terminal signal peptide was predicted to be betwen residues 23 and 24. N-terminal peptide sequence analysis of our FPLC-isolated KatP was A 24 D 25 K 26 K 27 E 28 T 29 Q 30 N 31 F 32 Y 3 Y 34 P 35 E 36 T 37 L 38 , demonstrating that the signal peptide was cleaved as predicted. Spectroscopic and Kinetic properties of KatP vs KatG Absorption spectra recorded for KatP revealed an RZ value of 0.67, consistent with the high aromatic amino acid content of the enzyme. The spectrum for feric KatP showed a Soret band at 406 nm along with charge transfer transitions at 502 nm and 629 nm indicating a mixture of pentacoordinate and hexacoordinate high-spin heme iron (Fig. 3.16 and 3.17). There was also evidence of minute quantities of hexacoordinate low-spin heme in feric KatP indicated by a smal shoulder at 540 nm. The spectrum for ferous KatP (Fig. 3.16 and 3.17) suggested predominantly pentacoordinate high-spin heme iron with a Soret band at 439 nm and a ?- and ?-band at 561 and 581 respectively. Comparisons of the absorption maxima with those of the intracelular E. coli catalase? peroxidase (KatG) reveal that the heme environments of the two enzymes are highly similar (Table 3.3). These results and those of others show that E. coli KatG has spectral characteristics that are highly similar to those of Mycobacterium tuberculosis KatG [37, 164, 165], and M. tuberculosis KatG contains a large proportion of hexacoordinate high- spin heme following purification and storage [165]. 108 Figure 3.16. Heme absorption spectra for the Soret band of feric, feri-cyano, and ferous forms of KatP. 34038042046050 20 40 60 80 10 120 Fe I Fe I Fe I -CN Wavelngth (nm) 109 . Figure 3.17. Heme absorption spectra for feric, feri-cyano, and ferous forms of KatP (480-680 nm). 48052056060640680 4 8 12 16 Fe I Fe I Fe I -CN Wavelngth (nm) 110 Absorption Band Maxima (nm) Heme State Protein Soret ? ? CT2 CT1 Fe III KatG 408 _ _ 502 629 KatP 406 _ _ 495 634 Fe II -CN KatG 423 542 _ N N KatP 422 543 _ N N Fe II KatG 439 561 581 _ _ KatP 438 560 590 _ _ Table 3.3. Heme Absorption Maxima for KatP and KatG. Absorption bands that were too weak to make unequivocal asignments of wavelength maxima are indicated by a dash. Transitions that were not expected for a particular species are indicated by N. 111 The EPR spectrum of KatP is consistent with other catalase-peroxidases in many, but not al, respects [116]. An axial signal [g = 5.94 (A ?) and 1.99 (A ?)] dominates the spectrum of KatP. The very weak rhombic signal [g = 6.6, 5.06, and 1.95] is also evident (Fig. 3.18). These signals correspond to hexacoordinate and pentacoordinate high-spin hemes, respectively. A spectrum of KatG (intracelular catalase-peroxidase) is also shown for comparison. A larger contribution from rhombic signal in this protein suggests that KatP has a larger proportion of heme in the hexacoordinate high-spin state than other catalase-peroxidases. However, it is important to bear in mind that the presence or absence of the sixth ligand (likely H 2 O) is highly sensitive to minute changes in conditions and storage time [166]. KatP showed strong catalase activity consistent with its intracelular relative, E. coli KatG (Fig. 3.19). The k cat determined for KatP was modestly higher than that determined for KatG (Table 3.4). However, the apparent K M for H 2 O 2 was considerably higher for KatP than that for KatG (Table 3.4). This contributed to an apparent second- order rate constant for the catalase activity of KatP (6.4 x 10 5 M ?1 s ?1 ) that was fivefold lower than KatG (3.5 x 10 6 M ?1 s ?1 ). The peroxidase activities of KatP and KatG were also compared (Fig. 3.20). The apparent k cat value with respect to H 2 O 2 for KatP-catalyzed ABTS oxidation was comparable to that observed for KatG (Table 3.4). The peroxidase activity of KatP gave an apparent K M with respect to H 2 O 2 that was threfold higher than that determined for KatG (Table 3.4). The apparent second-order rate constant for the peroxidase activity was about threfold lower for KatP (2.6 x 10 4 M -1 s -1 ) than for KatG (7.1 x 10 4 M -1 s -1 ). 112 Figure 3.18. EPR Spectrum of feric KatP (A) and feric KatG (B). 100200300400500 5.94 6.56 5.06 1.9 2.01 Magnetic Field (G) A B 1020304050 6. 6.0 1.95 5.1 Magnetic Field (G) 113 Figure 3.19. Efect of H 2 O 2 Concentration on the Catalase Activity of KatG and KatP. 0102030405060 250 50 750 100 1250 150 KatG tP [H 2 O 2 ] mM 114 Table 3.4. Kinetic Parameters for the Catalase and Peroxidase Activities of KatG and KatP 7 3.0 2.6 ? 10 4 58 0.83 7.1 ? 10 4 k cat (s -1 ) K M (mM) K ap ( -1 s -1 ) Peroxidase 1.8 ? 10 4 27 6.4 ? 10 5 1.4 ? 10 4 4 3.5 ? 10 6 k cat (s -1 ) K M (mM) K ap ( -1 s -1 ) Catalase KatP KatG Enzyme Parameter Reaction 115 Figure 3.20. Efect of H 2 O 2 Concentration on the Peroxidase Activity of KatG and KatP. 0 2 4 6 8 10 12 0 10 20 30 40 50 60 70 [H 2 O 2 ]mM KatP KatG 116 Both the catalase and peroxidase activities of KatP showed a sharp pH dependence that is consistent with the family of catalase-peroxidases. Maximal peroxidase activity was observed at a pH of ~4.7 and maximal catalase activity was observed at a pH of ~7.2 (Fig. 3.21). Because of the rapid reaction of catalase?peroxidase compound I with H 2 O 2 , other hydroperoxides (e.g., peracetic acid) must be used to monitor compound I formation in transient-state kinetic studies (Fig. 3.22). Heme absorption spectral changes were monitored from the reaction betwen feric KatP and peracetic acid by a decrease in absorbance at 403 nm and were consistent with the formation of a compound I-like intermediate (Fig. 3.23). Increases in peracetic acid concentration, produced a linear increase in k obs for compound I formation (Fig. 3.24). The apparent second-order rate constant for the reaction was 8.8 ? 10 3 M ?1 s ?1 . By comparison, the rate constant for the same reaction for KatG from Synechocystis PC 6803 is 3.9 ? 10 4 M ?1 s ?1 [35] while that measured for KatG from M. tuberculosis is ~6 ? 10 3 M ?1 s ?1 [166]. Stopped-flow kinetic analysis of CN ? binding to feric KatP revealed two- exponential behavior at al cyanide concentrations tested (Fig. 3.25). The exponential for the most rapid reaction acounted for 75% of the signal amplitude and was linearly dependent upon cyanide concentration (Fig. 3.26). From the cyanide concentration apparent second-order rate constant of 3.9 ? 10 5 M ?1 s ?1 was determined. This value is very similar to cyanide binding rate constants determined for E. coli KatG (5 ? 10 5 M ?1 s ?1 ), Synechocystis PC 6803 KatG (4.5 ? 10 5 M ?1 s ?1 ), and monofunctional heme peroxidases [167-169]. The exponential for the second much 117 Figure 3.21. Efect of pH on the Catalase and Peroxidase Activity of KatP. 3456789 10 20 30 40 50 50 10 150 20 250 Peroxidase Catlase pH 118 Figure 3.22. KatP Formation of Compound I by Hydrogen Peroxide (A) and Peracetic Acid (B). 119 Figure 3.23. Absorption Changes Spectra for KatP Formation of Compound I by Peracetic Acid. 0.0.40.81.21.62.0 0.7 0.8 0.9 0.1 Time (s) 120 Figure 3.24. Efect of Peracetic Acid Concentration on the Rate of Formation of Compound I. 0.0.51.01.52.02.53.03.5 5 10 15 20 25 30 35 [peracetic aid] mM 121 Figure 3.25. Efect of Cyanide Concentration on the Rate of Formation of the Fe III -CN Complex of KatP. 0.0.10.20.30.40.50.6 50 10 150 20 250 [KCN] mM 122 Figure 3.26. Absorption Changes Spectra for KatP Formation of Feri-cyano Complex. 0. 0.50.1 .7 0.8 0.9 0.1 0.1 510 Time (s) 123 slower reaction (~0.3 s ?1 ) acounted for ~25% of the signal amplitude and appeared to be independent of cyanide concentration (Fig 3.26). Trp-Tyr-Met Covalent Adduct Covalent Adduct Formation Evaluation by Denaturing Electrophoresis In our studies of catalase-peroxidases (KatG, KatP and mutants thereof), we have consistently observed the presence of two proteins that are expresed and co-purified. There is also evidence to suggest that others have observed a similar phenomenon [170]. The relative levels of these two proteins in our catalase-peroxidases studies correlate very wel with the heme content of the final purified protein. Low heme content correlates with one protein; high heme content correlates with the other. Typical results with KatP provide an excelent example. Following expresion and Ni-NTA purification, two bands of roughly equal intensity are observed (Fig. 3.27). Mas spectrometric analysis of the two bands shows that both are KatP. This raised the question: What is the basis for the diference in migration of these two KatP proteins? Our initial hypothesis was that KatP, a protein targeted to the periplasmic space, was expresed more rapidly than it was procesed, yielding a larger unprocesed protein and the smaler mature protein. However, repeated atempts to isolate each protein based upon subcelular fractionation techniques were unsuccesful. Furthermore, we produced a construct for the expresion of KatP eliminating the codons corresponding to the signal peptide (Ile 2 - Ala 23). This 124 Figure 3.27. SDS-PAGE of Ni-NTA-Purified KatP (lane 1). 125 construct produced exactly the same results in expresion and isolation as observed for the full-length gene. As described previously, the RZ values calculated from absorption spectra of purified KatP revealed that ~50% of the protein produced contained no heme. This correlated to about half the expected activity of the protein. Further separation of the two proteins by FPLC showed that the presence of heme and activity was asociated with the protein lower apparent molecular weight. The crystal structures of catalase-peroxidase from M. tuberculosis, H. marismortui and B. pseudomallei revealed the presence of a thre amino acid covalent adduct (Fig. 3.28) [28-30, 171]. The presence of this covalent adduct in catalase- peroxidases might explain this production of the two proteins observed by SDS-PAGE in our typical KatG and KatP preparations. The nature of the covalent adduct is consistent with formation as a result of protein oxidation, and heme is observed in only one of the two proteins. SDS is an anionic detergent, which binds to and denatures protein. SDS interupts the protein interaction involved in tertiary folding and places a negative charge on the protein in proportion to the number of amino acids. This uniformity in charge alows the proteins to be separated by mas. The covalent adduct found in catalase- peroxidases should not be disrupted by SDS. The presence of the covalent adduct would inhibit the ability of SDS to completely denature the protein. Protein not completely linearized would be expected to migrate as a more compact structure and at a lower apparent molecular weight as compared to its counterpart. In other words KatG or KatP protein with and without the covalent adduct should migrate with two diferent apparent molecular weights. Protein that was expresed containing heme would be expected to 126 Figure 3.28. Trp-Tyr-Met Covalent Adduct of Catalase-Peroxidases. S Met Fe Tyr OH N H Trp 127 migrate as a lower molecular weight protein while protein that never contained heme would migrate at the expected molecular weight. Therefore, the next important step was to determine if the addition of heme to catalase-peroxidase lacking the covalent adduct may result in its formation. Given the discussion above, SDS-PAGE presents a good system to evaluate the role of heme in the formation of the covalent adduct. In our lab, expresion of KatG typicaly results in the production of a large amount of apo-KatG. KatG expresed in this manner can be reconstituted with heme to produce an enzyme with full catalase and peroxidase activities [120]. If the covalent adduct is esential for catalase activity, then reconstitution would sem to alow for adduct formation. Taken together, this makes KatG ideal for evaluating the need for heme in the formation of the covalent adduct found in al catalase-peroxidases. The KatG protein was produced in E. coli as previously described in the methods section. The RZ values calculated from absorption spectra recorded for purified KatG showed that ~95% of the protein produced contained no heme (Fig. 3.29). As with KatP, SDS-PAGE analysis showed the production of two KatG proteins, but the band of higher apparent weight dominated over that with a lower apparent molecular weight (Fig. 3.30). To evaluate the afect of heme on the formation of the covalent adduct hemin was added to KatG and its migration by SDS-PAGE was compared to apo-KatG. Upon the addition of 1 eq. of hemin, a migration shift for KatG was observed (Fig. 3.30, lane 2). This migration shift indicated that ~50% of the protein had undergone covalent adduct formation. As mentioned above, the structure of the covalent adduct suggests formation by oxidative chemistry. It is consistent with the catalytic abilities of peroxidases to suggest that such chemistry could be supported by an Fe-containing porphyrin. To 128 Figure 3.29. Absorption Spectrum of KatG purified from cultures with no supplementation of ?-ALA of ferous amonium sulfate. 2030405060 0. 0.25 0.50 0.75 0.10 0.125 0.150 0.175 0.20 Abs Wavelngth(nm) 129 Figure 3.30. Efect of Heme and Peroxide on Migration of apo-KatG by SDS-PAGE. Apo-KatG (lane 1) with hemin (lanes 2 and 5), Zn-PIX (lanes 3 and 6), and peracetic acid (lanes 4-6) added. M 1 2 3 4 5 6 130 explore this further, apo KatG was reconstituted with the redox-inactive porphyrin, Zn- protoporphyrin IX. Upon the addition of Zn-protoporphyrin IX, no migration shift was observed (Fig. 3.30, lane 3). An oxidant may also be required in addition to the heme group. Bacteria are constantly exposed to oxidants like H 2 O 2 and other hydroperoxides (i.e., misfires in aerobic metabolism, etc). The presence of contaminating oxidants could explain why the covalent adduct could form with only the addition of heme. To further examine the need of an oxidant with heme on the formation of the covalent adduct, peracetic acid (2 eq) was added to KatG along with hemin. SDS-PAGE revealed a near complete migration shift of KatG protein (Fig. 3.30, lane 5). This shift indicated the formation of the covalent adduct in ~90-95% of KatG. However, the presence of peracetic acid alone or with Zn- protoporphyrin IX and KatG resulted in no migration shift (Fig. 3.30, lane 4 and lane 6). The formation of this adduct also requires a redox active protoporphyrin. Mass Spectrometry Mas spectrometry (MS) has long been used to identify the mas of chemical compounds, and relies on succesful ionization of the compound to be analyzed. The ionized compounds wil travel at a certain velocity through an electrostatic field in acordance with their mas-to-charge ratios (m/Z). Consequently, the time of flight of these ions wil be reflective of their mas. However, problems arise when trying to get large molecules into the gas phase. The development of the Matrix-Asisted Laser 131 Desorption-Ionization (MALDI) mas spectrometry (MS) is a highly efective approach and has become one of the leading tools used to acurately determine the mas of large molecules (e.g., proteins, peptides). With MALDI, the molecules are diluted into a matrix. Using a short intense laser pulse, the matrix is excited, leading to the desorption and ionization of the target molecules (Fig. 3.31). Because the typical mas of the matrix is much les than the molecules to be evaluated, their time of flight are dramaticaly diferent, making it relatively easy to ignore signals due to the matrix itself. Covalent Adduct Evaluation by MALDI The presence of a Met-Tyr-Trp covalent adduct in KatG was evaluated by MALDI-MS. The presence of a thre amino acid covalent adduct was first discovered with the publication of the first crystal structures of catalase-peroxidase in M. tuberculosis, H. marismortui and B. pseudomallei [28-30, 171, 172]. This was later confirmed by mas spectrometry of tryptic digest products of catalase-peroxidases [33, 173]. Catalase-peroxidases is expected to be cut into various peptide fragments by trypsin. However, trypsin does not cleave the covalent bonds of the Met-Tyr-Trp adduct. This wil alter the expected fragments obtained from tryptic digests of these proteins. In gel digestion using trypsin was preformed on co-purified KatG containing both proteins with and without the covalent adduct. Monoisotopic mases of peptides from SDS-PAGE extracted and digested protein were identified by MALDI-MS. Fragment mases were run through a sequence database to predict fragment sequence using Biotool 132 Figure 3.31. Matrix-Asisted Laser Desorption-Ionization Mas Spectrometry Schematic. Time of Flight Mas Analyzer Sample Matrix Ion Ion Laser Beam 133 Mascot Sequence Identifier. Digestion of KatG without the covalent adduct resulted in thre separated peptide fragment peaks corresponding to each of the thre peptides containing a residue involved in the covalent adduct (Fig. 3.32). However, digestion of KatG already containing the covalent adduct lacked the corresponding individual peptide fragments (Fig. 3.33), indicating that the thre fragments were linked together. Efects of Y226F on KatG In order to interupt formation of the thre amino acid adduct, site-specific substitution of the Tyr participant (Tyr 226 in KatG) with Phe was caried out. This Tyr is conserved in al catalase-peroxidases. The mutation of Tyr 226 to Phe would prevent the coupling of radicals betwen Trp and Tyr and should stop the formation of the covalent adduct in KatG. Expresion of KatG Y26F resulted in the production of one protein band as evaluated by SDS-PAGE (Fig. 3.34). The molecular weight of this band corresponded to wtKatG lacking the covalent adduct. Catalase and peroxidase activities of KatG Y26F were compared to wtKatG. Catalase activity for KatG Y26F was esentialy eliminated (Fig. 3.35). Conversely, KatG Y26F showed an increase in peroxidase activity relative to wtKatG at low concentrations of hydrogen peroxide (Fig. 3.36). However, at high concentrations of hydrogen peroxide, a loss of peroxidase activity was observed. This loss of activity may be due to inactivation of peroxidase by hydrogen peroxide. This phenomenon is commonly observed in monofunctional peroxidases (e.g., horseradish peroxidase) 134 Figure 3.32. MALDI Spectrum of Tryptic Digest of KatG without Covalent Adduct. Trp peptide fragment designated by arow. 149 150 151 0 20 40 60 m/Z 135 Figure 3.33. MALDI Spectrum of Tryptic Digest of KatG with Covalent Adduct. Trp peptide fragment designated by arow. 1149 1501151 0 200 400 600 m/Z 136 Figure 3.34. SDS electrophoretic separation of total celular protein from BL-21 (DE3)pLysS transformed pKatG Y26F at time of induction with 1 mM IPTG (lane 0), 1 h post-induction (lane 1), 2 h post-induction (lane 2), 3 h post- induction (lane 3), and 4 h post-induction (lane 4). M 0 1 2 3 4 137 Figure 3.35. Efect of H 2 O 2 Concentration on the Catalase Activity of wtKatG and KatG Y26F . 0102030405060 250 50 750 100 1250 150 wt KatG Kat Y26F [H 2 O 2 ] mM 138 Figure 3.36. Efect of H 2 O 2 Concentration on the Peroxidase Activity of wtKatG and KatG Y26F . 012345 10 20 30 40 50 60 70 KatG Y26F wt at [H 2 O 2 ] mM 139 [174-177]. In catalase-peroxidases, the loss of catalase activity makes it susceptible to inactivation by hydrogen peroxide [178]. In the peroxidase catalytic cycle, the reduction of Compound I is the rate-limiting step. The acumulation of Compound I alows it to react with exces hydrogen peroxide to produce compound II (reaction 12). Compound II is an inactive form of the enzyme and results in the loss of activity. These results and those of others indicate that the covalent adduct appears to be vital for catalase activity. Conversely, this adduct in not required for peroxidase activity [31, 33, 34, 173]. SDS-PAGE was also used to evaluate the formation of the covalent adduct in KatG Y26F as with wtKatG (above). In order to evaluate the afect of heme on the formation of covalent adduct in KatG Y26F , hemin was added to KatG Y26F and compared to apo-KatG Y26F . The addition of 1 eq. of hemin resulted in no apparent migration shift of KatG Y26F (Fig. 3.37, lane 2). This indicated that none of the protein had formed the covalent adduct and the substitution of Tyr with Phe disrupted its formation. The need of an oxidant in the formation of the covalent adduct in KatG Y26F was also evaluated. Upon the addition of 2 eq. of peracetic acid to KatG Y26F along with hemin, there was no apparent migration shift (Fig. 3.37, lane 5). This further indicated the complete disruption of the covalent adduct formation. As with wtKatG, peracetic acid alone did not result in a migration shift (Fig. 3.37, lane 4). The non-redox active Zn- PIX also did not have any afect on the protein migration with or without the presence of peracetic acid (Fig.3.37, lane 3 and lane 6). 140 Figure 3.37. Efect of Heme and Peroxide on Migration of KatG Y26F by SDS-PAGE. Hemin (lanes 2 and 5), Zn-PIX (lanes 3 and 6), and peracetic acid (lanes 4- 6) were added to Apo-KatG Y26F (lane1) M 1 2 3 4 5 6 141 CHAPTER FOUR DISCUSION Hemoproteins are a widely distributed in nature and compose an important clas of biologicaly important metaloproteins that have a common active site cofactor, an iron-porphyrin complex also caled heme. These proteins are involved in a wide range of functions such as the transportation of dioxygen (hemoglobin and myoglobin), catalysis of the disproportionation and reduction of hydrogen peroxide (catalases and peroxidase), oxidation of organic substrates (cytochrome P450), production of reactive oxygen species (NADPH oxidase and myeloperoxidase), and electron transport (cytochrome oxidase). Each hemoprotein, unique in its function, difers in the specific form of the heme complex, which undergoes diferent transformations in the oxidation state, ligands, and spin states controlled by the protein. This diversity of function makes hemoproteins a robust model for studying the important relationship betwen the structure of proteins and their functions. The research in this disertation addreses hemoprotein structure and function in thre ways. It addreses a long-standing impediment to progres in hemoprotein 142 structure/function studies. That is, the dificulty of expresing large quantities of these enzymes and proteins in their holo state. Secondly, it is the first to give a complete description of the spectral and catalytic properties of a periplasmic catalase-peroxidase uniquely produced by highly virulent pathogenic bacteria. The catalase-peroxidases provide an excelent opportunity to addres a poorly understood aspect of the structure/function equation. That is, mechanisms by which distant structural components fine tune an active site for a specific catalytic purpose. Along these lines, the research described in this disertation has addreses the requirement for formation of a unique Trp-Tyr-Met covalent adduct and its role in catalase-peroxidase catalysis. Expresion of Recombinant Hemoproteins A commonly observed problem in the expresion of recombinant hemoproteins is that large quantities of the protein are expresed but very litle of it actualy contains heme. A major goal of the research described in this disertation was to develop an expresion system that resolves this problem. The idea was to produce an E. coli strain for hemoprotein expresion that could draw exogenously added heme directly from its growth medium to incorporate into the target protein as it was expresed. The development of the HPEX system acomplishes this by giving E. coli strains a heme receptor. The HPEX plasmid (e.g., pHPEX2) containing the heme receptor gene, chuA, was transformed into E. coli expresion strain BL-21 (DE3) and expresed. This produced a protein corresponding to the molecular weight of ChuA. While ChuA semed 143 to be expresed with the HPEX plasmids, the question of its ability to function as a heme transporter remained unclear. The function of ChuA was evaluated by the ability of HPEX-transformed cels to grow in iron-deficient media using exogenous heme as the only iron source. E. coli cels transformed with pHPEX2 were able to use exogenous heme for growth in iron-deficient media. This indicated that ChuA was functional and could be used to draw in heme into the E. coli cels from the media. The question about whether or not hemin brought in by ChuA would be incorporated into target hemoproteins was also examined. The expresion of KatG in E. coli cels transformed with pHPEX3, showed a 10 fold increase in heme content. Thus, KatG exhibited ful activity. The HPEX system was shown to give standard E. coli expresion strains the ability to produce a functional heme receptor, which, as we have demonstrated, can be used to give recombinant hemoproteins a full capacity of heme. The benefits to be realized from enhancing expresion of recombinant hemoproteins in their holo state are far-reaching. These benefits can be divided into two broad categories: 1) therapeutic application of hemoproteins and 2) advancing the study of hemoprotein structure/function relationships. This disertation demonstrates that the HPEX system is likely to be useful in both. Application for Therapeutic use of Hemoproteins Due to the constant occurrence of medical emergencies, a significant population with congenital blood diseases like hemophila, and the ever-present danger of natural 144 disasters, there is always a high demand for stored blood to use for life-saving transfusions. The short supply of stored blood is exacerbated by several chalenges. These chalenges have fueled interest in the development of efective blood substitutes. One of the most promising possibilities is cel-fre hemoglobin. Where blood, even under refrigeration, has a short shelf-life (30-45 days), hemoglobin can be stored at room temperature for months and even years [179]. This opens the possibility for transportation and imediate use by emergency personnel at the scene of need. Blood cels, due to their antigenic properties, must be typed and cross-matched to ensure safe transfusion. Cel-fre hemoglobin is not antigenic, alowing for the elimination of these time consuming procedures. In spite of the many advantages for using cel-fre hemoglobin as a blood substitute, there are two major drawbacks that prevent its application in this role. The first is that Hb is a potent scavenger of NO. NO binds to Hb 8000 times faster than O 2 [180]. One important role of NO is the mediation of smooth muscle relaxation. The uptake of NO by extracelular Hb causes vasoconstriction and increase in blood presure [180, 181]. Substitution of amino acids in the distal pocket of the globin proteins with large aromatic or aliphatic amino acid has been shown to inhibit the non-covalent capture of NO [182]. The substitution of Leu in the ? subunit and Val in the ? subunit of hemoglobin with Trp decrease NO binding and inhibits NO dioxygenation reaction. This resolves the hypertensive efect of extracelular hemoglobin as a blood substitute. The second problem with using cel-fre hemoglobin as a blood substitute is that the hemoglobin tetramer is susceptible to disociation into ?/? dimers in an extracelular environment. In this form, the protein becomes much more susceptible to autoxidation 145 leading to greater oxidative stres [179]. Furthermore, kidney function is interupted by these disociated ?/? dimers. This is resolved by generating cross-linked or polymeric forms of hemoglobin on one hand, or by generating a fusion construct to expres two ? subunits in tandem with a glycine linker [182]. The later, of course, requires manipulation of the hemoglobin gene by molecular cloning techniques. Interestingly, the resolution of both of these major problems by molecular cloning techniques (i.e., by producing unnatural hemoglobins) makes the use of natural sources of hemoglobin as blood substitutes impossible. They must be produced through the expresion of recombinant forms. In terms of expresion levels, cost of materials, ease of use, and ease of manipulation, E. coli expresion systems are by far the most advantageous for producing large quantities of recombinant proteins. Unfortunately, with it comes to the expresion of recombinant hemoproteins in E. coli, a major drawback is the fact that a vast majority of the protein lacks the heme prosthetic group esential for activity. Clearly, hemoglobin without heme is useles as a blood substitute, even if it is produced at very high levels. Part of the research described in this disertation involved the development of a hemoprotein expresion (HPEX) system designed to increase the heme content of recombinant hemoproteins expresed in E. coli. The HPEX strains described expres a heme receptor, increasing their capacity to withdraw hemin directly from the extracelular environment. This system eliminates the need to supplement media with relatively high concentrations of heme precursors (e.g., 0.5 mM ?-ALA) instead only requiring supplementation with low-hemin concentrations (e.g., 8 ?M). The data presented in this disertation strongly suggest that 146 the HPEX system wil be useful for the expresion of recombinant hemoglobin to be used as a blood substitute. Hemoglobin and myoglobin are alies in the transport and storage of oxygen. The ? and ? subunits of hemoglobins are analogous to myoglobin. Here myoglobin is used as a model for hemoglobin. The globins are an ideal group to use with the HPEX system. The expresion of sperm whale myoglobin with the HPEX system confirmed the production of a functional heme receptor. Mb expresed with BL-21 (DE3) pHPEX3 and BL-21 (DE3) pHPEX-fur cels showed an increase in heme content when hemin was added to the cultures at the time of induction. When hemin was excluded from the culture medium, absorption derivative spectra of whole cels showed a low heme content. When Mb was expresed with BL-21 (DE3) and BL-21 (DE3) pLysS cels, absorption derivative spectra showed no increase in heme content even if hemin was added to the culture medium. Application to Studies of Heme Enzyme Structure and Function The wide range of biochemical proceses mediated by hemoproteins is nothing short of astonishing. The fact that these proteins al use esentialy the same iron porphyrin cofactor to acomplish these tasks makes the diversity even more impresive. Clearly, the structure of the protein around the heme cofactor is the factor that dictates the function of one hemoprotein in comparison to another. Indeed, hemoproteins are a robust 147 group for studying the relationship betwen the structure of a protein and its catalytic abilities. The study of structure/function relationships in hemoproteins requires large amounts of active protein and the ability to produce large quantities of variants (produced by site-directed or other mutagenesis procedures). Indeed, techniques like stopped-flow kinetic analysis, EPR, NMR, X-ray crystalography, magnetic circular dichroism, etc. are very protein intensive. Almost always, high-level expresion of recombinant forms is required, and as mentioned previously, E. coli expresion systems are the most productive, least expensive, and least complicated. Unfortunately, it is also very common that a significant portion of the expresed protein lacks heme. Indeed, the purified enzymes often contain les than 10% of the heme expected. Addition of heme biosynthetic precursors such as ?-ALA and ferous amonium sulfate aleviates this problem to some degre. However, large quantities of these materials must be added (0.5 mM). Catalase-peroxidases are ideal enzymes to examine a poorly understood aspect of hemoprotein structure and function. They utilize a single active site to catalyze at least two diferent reactions. Interestingly, this active site is almost identical to active sites found in monofunctional peroxidases. As the name implies, monofunctional peroxidases have virtualy no catalase activity. Given their nearly identical active sites, it appears that unique structures external to the catalase-peroxidase active site modulate its catalytic properties even from distances of 30 ? and more. There are very few systems so idealy set up to evaluate long-range influences on active site structure and function. 148 Although, expresion of catalase-peroxidases in standard E. coli strains results in large quantities of protein, it is almost al in its apo-state. Even addition of ?-ALA and ferous amonium sulfate do litle to improve expresion of the holoenzyme. These enzymes would appear to be the perfect candidates for the HPEX system. In our laboratory, we use the T7-RNA polymerase-based pET system for the expresion of catalase?peroxidases. The best results has been observed using BL-21 (DE3) pLysS. Consequently, BL-21 (DE3) pHPEX3 strain was the most appropriate strain to use as a basis for comparison. Using BL-21 (DE3) pHPEX3 cels, KatG was expresed at levels comparable to those observed in the pLysS-transformed system. When hemin was excluded from the culture medium, absorption spectra of the partialy purified protein showed a very low- heme content. Addition of hemin to the culture at the time of induction resulted in the same yield of isolated KatG protein, but the enzyme had a 10-fold greater heme content. Conversely, when BL-21 (DE3) pLysS was used in place of our pHPEX3-transformed strain, addition of hemin to the culture at induction resulted in only a 3.5-fold increase in heme content. The expresion of KatG with the HPEX system alowed for substantial improvement in its heme content by simply adding hemin to the growth medium at time of induction. The moderate increase in heme content and activity of KatG expresed in BL-21 (DE3) pLysS cultures containing hemin is noteworthy. Many E. coli strains (including this one) are apparently unable to use extracelular heme as an iron source for growth only because they lack a heme receptor [8-10]. Nevertheles, addition of hemin to expresion cultures can improve the heme content of recombinant hemoproteins [167]. 149 Our mechanism by which this might occur is the transmembrane movement of hemin through the membrane bilayer [183]. The rate-determining step in this proces is almost certainly ?heme flipping? whereby the propionate groups go from an orientation into the extracelular environment to pointing into the periplasmic space. As expected, this proces is slow and not sufficient to supply the demands of over-expresed hemoproteins [184, 185]. The spectral and kinetic properties of KatG expresed in BL-21 (DE3) pHPEX3 grown in hemin-supplemented media were compared to those of KatG expresed in BL- 21 (DE3) pLysS grown in unsupplemented media. Spectraly, the absorption maxima for the feric, feri-cyano, and ferous forms of KatG derived from each system were consistent with one another and other catalase?peroxidases [35, 37, 165, 167]. Indeed, the feric minus feri-cyano diference spectra were nearly identical to one another (Fig. 3.9). Kineticaly, the K M of KatG for H 2 O 2 was unchanged by expresion conditions and was consistent with values reported for KatG [37, 186] and other catalase?peroxidases [167]. Likewise, on a per-heme basis, k cat /K M values for KatG obtained from either expresion system were consistent with one another and with values reported for other catalase?peroxidases [37, 167]. On the other hand, on the basis of protein concentration (i.e., specific activity), KatG isolated from hemin-supplemented BL-21 (DE3) pHPEX3 cultures had much greater activity than unsupplemented BL-21 (DE3) pLysS cultures. This indicates that holo KatG expresed in the HPEX system has the same properties as holo KatG produced by other means. However, the yield of holo KatG is much greater in the HPEX system than other E. coli expresion strains. 150 The study suggests that the HPEX system wil be a versatile system for hemoprotein expresion. As already noted, it has enhanced Mb expresion as wel as KatG, two completely diferent proteins. It has also enhanced the expresion of variants of KatG, such as KatG ?FG , KatG Y26F , KatG R117A , and KatG D597A . A potential benefit derived from using a receptor-based heme scavenging system is that other sources of heme or heme derivatives could be used in place of hemin. The outer membrane-bound heme receptors like ChuA produced by Gram-negative pathogens alow for heme uptake from hemoproteins like hemoglobin or myoglobin [9, 10]. Likewise, these receptors have also been shown to support uptake of tetrapyrroles other than hemin. Substitutions of the metal (e.g., zincprotoporphyrin IX) as wel as substitutions of the side chains on the macrocycle (e.g., feriporphyrin) are wel acomodated by these heme receptors [187]. Consequently, it is possible that the HPEX system ay be used to expres recombinant hemoproteins bearing unusual natural heme derivatives or even unnatural synthetic hemes. One limitation sen thus far with the HPEX system is that it has not been able to resolve the expresion of some hemoproteins in inclusion bodies. However, it is feasible that the HPEX system ay be used in conjunction with other systems (e.g., chaperones) to resolve this isue. The chaperone GroEL, which is responsible for asisting the correct folding of many diferent proteins, has been shown to increase expresion of correctly folded cytochrome P450 when expresed with the enzyme [188]. It is likely that the presence of heme is also required for the correct folding. The HPEX system ay be ideal for increasing the production of the correctly folded cytochrome P450 along with GroEL. 151 Periplasmic Catalase-Peroxidase Escherichia coli O157:H7 is a highly virulent and deadly food-borne pathogen. Indeed, O157:H7 is recognized as one of the most virulent even among the shiga-toxin- producing E. coli strains [12, 20]. This strain produces several factors that are absent from les pathogenic or non-pathogenic E. coli. Among these is a unique periplasm- targeted catalase-peroxidase (KatP). KatP along with KatA (from L. pneumophila) and KatY (from Y. pestis) are part of a unique subgroup among the family of the bifunctional catalase?peroxidases. This is evident from available sequence data showing these enzymes to be more closely related to one another than they are to their cytoplasmic counterparts [189, 190]. For example, KatP shares greater sequence identity with KatY and KatA than it does with KatG, the intracelular catalase?peroxidase common to al E. coli strains [189, 190]. Indeed, it has been suggested that these periplasmic catalase? peroxidases were the result of lateral gene transfer from an archeaon [17]. Analysis of numerous new catalase?peroxidase gene sequences (recently available due to intensive genome sequencing eforts) has cast doubt on this origin for the periplasmic catalase? peroxidases [28]. However, it is clear that the periplasmic enzymes form a cluster that is distinct from other catalase?peroxidases [28]. Interestingly, it has been suggested that KatP represents one of the most recently acquired virulence factors by E. coli O157:H7 [191]. Clearly, the routes by which such organisms acquire new potential virulence factors is an important problem whose solution wil be aided by additional genomic and proteomic information. 152 Among the periplasmic catalase?peroxidases, KatA, KatY, and KatP are produced by highly virulent bacterial pathogens but not by their les pathogenic or non-pathogenic relatives, suggesting that these enzymes may be used as virulence factors. This hypothesis is supported by the fact that expresion of KatY by pathogenic Yersinia species corresponds with a shift from 26 to 37 ?C, a feature that coincides with transfer of the bacterium from the flea to the mamalian host [25, 26]. Similarly, in L. pneumophila, KatA has been shown to impart characteristics that support its survival within the macrophage host [23, 192]. Despite the potential role of these enzymes in the virulence of these pathogens, only one has been previously isolated [31] and none of the thre has received a careful evaluation of its spectral or kinetic properties. The cloning, expresion, isolation, and characterization of KatP, a periplasmic catalase?peroxidase recently identified in highly virulent E. coli O157:H7 is described in this disertation. KatP was post-translationaly procesed acording to the putative periplasm- targeting sequence on its N-terminal end. The mature enzyme displayed heme content consistent with other catalase?peroxidases as judged by its 0.67 RZ value. Absorption spectra of feric and ferous KatP are suggestive of a mixed population of high-spin hexacoordinate and pentacoordinate hemes, as wel as some of the hexacoordinate low- spin complex. EPR spectra confirm the dominance of the high-spin species and suggest a higher than usual proportion of hexacoordiante high-spin heme. Consistent with its heme content, KatP displays strong catalase and peroxidase activities. The k cat values calculated for each activity are comparable to, if not greater than, those determined for the cytosolic catalase?peroxidases (Table 4). KatP shows 153 sharp pH dependence in both catalase and peroxidase activities. The optima for catalase ~pH 7 and peroxidase ~ pH 5 are similar to catalase?peroxidases. One striking diference betwen KatP and other catalase?peroxidases evaluated to date is the relatively high K M for H 2 O 2 . At 27 mM, the K M of KatP for H 2 O 2 as determined from its catalase activity is the highest yet reported for any of the bifunctional enzymes by at least fivefold [167, 193]. The K M for H 2 O 2 as determined from the peroxidase activity of KatP (3 mM) is also larger than that determined for KatG (0.83 mM). Though the same kinetic treatment has not been undertaken for the other periplasmic enzymes, it has been reported that KatY has a low-catalase activity [25, 26]. Supposing that low concentrations of H 2 O 2 (i.e., 0.05?K M ) were used in these studies, a low apparent catalase activity would be expected. One possible explanation for this observation is the presence of a sixth ligand (probably H 2 O) in the resting form of the enzyme. The EPR spectrum indicates a greater proportion of hexacoordinate high-spin heme. An inflated K M for H 2 O 2 would be an expected outcome if the displacement of this sixth ligand was more dificult in KatP than in other catalase-peroxidases. On the other hand, rate constants determined for the reactions of KatP with other hydroperoxides (e.g., peracetic acid) and heme ligands (e.g., CN ? ) are wel within the range of those determined for other catalase?peroxidases. Together these data suggest that high apparent K M values of KatP for H 2 O 2 are due to quite subtle diferences in its active site compared with other catalase?peroxidases. Though the periplasmic catalase?peroxidases have been implicated as virulence factors, the mechanisms by which they contribute to the growth and survival of these pathogens have not been elucidated. It has been suggested that the primary benefit to 154 these pathogens for producing a second catalase?peroxidase is derived from the celular location of the enzyme [27]. By placing a highly active H 2 O 2 decomposition catalyst within the periplasmic space, the ability to remove H 2 O 2 before it is alowed to damage critical inner-membrane components is realized. Indeed, there is precedent for strong periplasmic catalase activity (from monofunctional catalases) in Brucela abortus (a mamalian pathogen) [194, 195], Vibrio fischeri (a light-generating symbiont of Euprymna scolopes) [196, 197], and Pseudomonas syringae (a plant pathogen) [198, 199]. Al of these organisms appear to use catalase to fend off host responses that involve production of abundant reactive oxygen species and oxidants derived from them (e.g., HOCl). In a similar manner, the organisms that produce these unique periplasmic bifunctional catalase?peroxidases encounter similar host responses. Y. pestis and L. pneumophila are both intracelular pathogens, and E. coli O157:H7 induces the migration of neutrophils to the site of infection in the intestinal lumen [200]. The peroxidase activity of the periplasmic catalase?peroxidases may provide another esential function by reducing reactive peroxides likely to be produced as a result of the imune response, including peroxynitrite (ONO ? ) and its conjugate peroxynitrous acid (ONOH). Catalase-peroxidases also have peroxynitritase activity to decompose ONO ? generated by macrophages (Fig. 4.1). KatG in M. tuberculosis has been shown to decompose ONO ? at a similar rate to myeloperoxidase, 10 5 M -1 s -1 and 10 6 M -1 s -1 respectively [151]. In our laboratory, we have observed that ONO ? reacts rapidly with KatP to form compound I as sen by a decrease in absorbance at 404nm (Fig. 4.2). Following this, the slow conversion of compound I to compound I is also observed indicated by the increase in absorbance 420nm (Fig. 4.3). It is important to bear 155 Figure 4.1 Incorporation of Peroxynitrite in the Catalytic Cycle of Catalase-peroxidase. 156 Figure 4.2. KatP Compound I Formation in the present of 200?m Peroxynitrite. 0.0.40.81.21.62.0 0.92 0.94 0.96 0.98 0.10 0.12 Time (s) 157 Figure 4.3. KatP Compound I Conversion to Compound I in the present of Peroxynitrite. 048121620 0.75 0.80 0.85 0.90 0.95 0.10 Time (s) 158 . in mind that compound I can also be reduced back to the feric state by H 2 O 2 (Fig. 4.1), another peroxide generated at high levels by macrophages and neutrophils. The ability of catalase-peroxidases to simultaneously decompose ONO ? and H 2 O 2 would be expected to greatly enhance bacterial ability to survive. In the face of agents such as ONO ? , the advantages for producing a periplasmic peroxide decomposition catalyst become even more pronounced. The pK a for peroxynitrite is around 7, thus, under physiological circumstances a significant proportion of peroxynitrite is expected to exist in its ionized form, and thus, be unable to cross the inner-membrane where it might be dealt with by cytosolic catalases and catalase? peroxidase. The presence of a periplasmic catalase?peroxidase might be particularly efective for preventing such host-derived damage. It has also been suggested that the peroxidase activity may serve additional functions contributing to virulence. Indeed, others have observed that the katA gene is esential for the survival of L. pneumophila in the stationary phase, and they suggest that this benefit most likely results from the peroxidase activity of KatA rather than its catalase activity [32]. Role of Trp-Tyr-Met Covalent Adduct Since the publication of the crystal structure of catalase-peroxidases, more insight has been given to the function of this enzyme. The crystal structures of KatG from M. tuberculosis, H. marismortui and B. pseudomallei reveals an active site almost identical to clas I peroxidases (e.g., cytochrome c peroxidase). However, the electron density 159 maps do suggest the presence of a Trp-Tyr-Met covalent adduct peripheral to the active site (Fig. 4.4) [30, 171]. Al thre amino acids found in this adduct are conserved in al catalase-peroxidases. The expresion of recombinant hemoproteins results in the vast majority of the protein lacking the heme prosthetic group [1-6]. This comes as a consequence of the expresion of the protein out pacing the production of the heme prosthetic group. SDS- PAGE evaluation of recombinant KatP shows two proteins with molecular weights around that of KatP. Mas spectrometric analysis revealed both proteins to be KatP. Characterization of both showed one to have heme and activity while the other had neither. The discovery of the Trp-Tyr-Met covalent adduct in catalase-peroxidases may help to explain this phenomenon. The lack of heme in a similar system (e.g., cytochrome c peroxidase) results in protein without the covalent adduct which requires the presence of heme in the active site for its formation [201]. The presence of heme in catalase- peroxidases would yield a protein unable to be completely denatured by SDS as a result of its inability to disrupt the covalent adduct. Incompletely denatured catalase- peroxidases would migrate at a diferent apparent molecular weight than their completely denatured counterparts. The ability to distinguish KatP with and without this covalent adduct by SDS-PAGE makes it a good way to monitor the presence of the cross-link in catalase-peroxidases. The expresion of KatG results in ~ 95 % of apo-protein, and is ideal for monitoring cross-link formation. The addition of heme to purified apo-KatG showed a migration shift of ~ 50% of the protein indicating the formation of the covalent adduct. In similar systems, the presence of heme has also been shown to be required in the active site for the formation of this covalent adduct [201]. An interesting aspect about 160 Figure 4.4. Active Site Representation of E. coli Catalase-Peroxidase [29]. Arg102 Trp105 His106 Tyr226 His267 Trp318 Asp377 Met252 161 the formation of this covalent adduct is that it appears to involves some redox chemistry. This was confirmed when the non-redox active Zn-protoporphyrin IX was added to apo- KatG, no formation of cross-link was observed. The formation of the covalent adduct in only half of KatG protein indicated that heme alone was not sufficient enough to facilitate the cross-link. The redox chemistry of the cross-link formation implicated perhaps the need of an exogenous oxidant. The addition of peracetic acid along with heme to apo- KatG showed formation of the covalent adduct in nearly 100% of the protein. Others have also shown that the formation of the covalent adduct involves a redox-linked reaction and requires the presence of peroxide or other oxidants alone with heme in the active site [201, 202]. The substitution of the central amino acid, Tyr to Phe, results in the disruption of the covalent adduct revealed by SDS-PAGE. The addition of heme or peroxide to KatG Y26F showed no migration shift indicating no formation of covalent adduct. This adduct appears to be esential for the full function of the enzyme. Indeed, substitution of either the tyrosine or the tryptophan residue with phenylalanine has been shown to interupt the formation of the adduct and completely eliminate catalase activity [203, 204]. Interestingly, peroxidase activity is retained or even enhanced in these variants. On the other hand, mutation of methionine does not interupt the cross-link betwen tyrosine and tryptophan [202]. Like the Trp and Tyr substitutions, mutation of Met also eliminates catalase activity but maintains peroxidase activity. This indicates that the formation of the Trp-Tyr link is the first step in adduct formation and suggests that Trp-Tyr cross-linking is not sufficient to generate catalase activity. 162 The formation of the Trp-Tyr-Met adduct is proposed by Ortiz de Montelano to be an autocatalytic proces that uses compound I as the oxidizing species (Fig. 4.5) [202]. Formation of compound I leads to oxidation of both Trp and Tyr (Fig. 4.5a-b). Coupling of the two radicals results in the formation of the Trp-Tyr link (c). Formation of a second compound I intermediate further oxidizes the Trp-Tyr link (d-e). A nucleophilic atack of the sulfur atom of Met results in the formation of the Tyr-Met link, yielding the Trp-Tyr- Met adduct (f-g). The role this covalent adduct plays in catalase-peroxidases activity is not fully understood. The Trp-Tyr-Met covalent adduct may help provide the correct architecture of the active site. It also may be involved in the electron tunneling that is esential for catalysis. What is obvious is that this covalent adduct is required for the two-electron reduction of compound I involved in catalase activity but does not sem to be required for the two-electron reduction for peroxidase activity. Summary The activation of dioxygen is important to life in an aerobic environment. Transition metals are the key to this activation. However, active oxygen is highly reactive and can react with most macromolecules with damaging even life-threatening consequences. The control of this double-edged sword is esential. Nature has developed ways to control active oxygen by controling the environment in which oxygen is activated. The control of iron during transport and storage and its placement in structures like porphyrins to generate a prosthetic group is part of the control of oxygen 163 Figure 4.5. Proposed Mechanism for Formation of the Trp-Tyr-Met Adduct of Catalase- peroxidase. H 2 O 2 H 2 O a Fe IV+ 2H + Fe IV+ H 2 O 2 H 2 O O O H 2 O H 2 O 2H + b c d e f g S Met Fe Tyr OH N H Trp S Met Fe Tyr OH N H Trp S Met Fe Tyr O N Trp S Met Tyr OH N H Trp S Met Fe Tyr OH N H Trp S Met Fe Tyr O N Trp S Met Fe Tyr OH N H Trp S Met Tyr OH N H Trp 164 activation. This alone is not sufficient. The protein environment around the prosthetic group (e.g., heme) must provide pinpoint control in order to achieve the desired function. Furthermore, there must be mechanisms in place to dispose of reactive oxygen species when they are unintentionaly released. In both respects, hemoproteins are an intriguing group of proteins. They demonstrate exquisite catalytic control, and some of them are used for rapid removal of reactive oxygen species. The versatility of function demonstrated by proteins that al use the same prosthetic group is ensured by the structure of the protein itself. This, of course, occurs at many levels. The first being the ligand(s) to the heme iron which are usualy supplied by the protein. The second being the imediate environment of the heme and its ligand, and the third being the global protein structure, which may include features some distance from the active site. A complete understanding of how protein structure dictates heme function at al levels wil have invaluable benefits to science and medicine. This disertation describes the development and testing of a protein expresion system that aleviates a common dificulty in succesful production of hemoproteins in their holo state. This HPEX system is designed to increase heme content of a protein of interest by simply supplementing the expresion medium with hemin. The HPEX system was shown to expres a functional heme receptor. The system has been shown to enhance expresion of very diferent proteins in their holo state. The catalase-peroxidases are heme-containing enzymes that provide a unique opportunity to evaluate how protein structures distant from the active site actualy serve to fine-tune its catalytic capabilities. 165 Interestingly, there is a set of periplasmic catalase-peroxidases which are only produced by highly pathogenic bacteria. These periplasmic enzymes may be virulence factors. This disertation describes the first complete characterization of one of these enzymes, KatP from E. coli O157:H7. Absorption and EPR spectra of KatP were consistent with other catalase-peroxidase in that they indicated a dominance of high-spin heme. However, KatP showed a greater proportion of the hexacoordinate high-spin state. Apparent k cat values for both catalase and peroxidase were somewhat higher consistent with most other catalase-peroxidases. On the other hand, K M values were higher for KatP. KatP formation of compound I and CN ? binding were also consistent with other catalase-peroxidase. Along with its location in the periplasm, KatP reacts with peroxynitrite to form compound I. This may make KatP idealy suited to addres the reactive species produced by the imune response to invading bacteria. The presence of a Trp-Tyr-Met covalent adduct was shown to afect the migration of KatG protein by SDS-PAGE. The formation of the covalent adduct required the presence of the heme. The non-redox active Zn-protoporphyrin IX did not result in the formation of the covalent adduct, indicating that redox activity was required. The formation of the covalent adduct was enhance by a peroxide only if a redox active porphyrin was also present. The covalent adduct was required for catalase activity but not peroxidase activity. Interestingly, KatG became more sensitive to higher concentrations of hydrogen peroxide if it was unable to establish the covalent adduct. This is typical of monofunctional peroxidases, which also lack catalase activity. 166 REFERENCES [1] Chouchane, S., Lippai, I., and Magliozzo, R. S., Catalase-peroxidase (mycobacterium tuberculosis katg) catalysis and isoniazid activation, Biochemistry 39 (2000) 9975-9983. [2] Nie, G., Reading, N. S., and Aust, S. D., Expresion of the lignin peroxidase h2 gene fromphanerochaete chrysosporiuminescherichia coli, Biochemical and Biophysical Research Communications 249 (1998) 146-150. [3] Nishimoto, M., Clark, J. E., and Masters, B. S., Cytochrome p450 4a4: Expresion in escherichia coli, purification, and characterization of catalytic properties, Biochemistry 32 (1993) 8863-8870. [4] Smith, M. A., Napier, J. A., Stymne, S., Tatham, A. S., Shewry, P. R., and Stobart, A. K., Expresion of a biologicaly active plant cytochrome b5 in escherichia coli, Biochemical Journal 303 (1994) 73-79. [5] Sinha, N., and Ferguson, S. J., An escherichia coli cm (cytochrome c maturation) deletion strain substantialy expreses hydrogenobacter thermophilus cytochrome c552 in the cytoplasm: Availability of haem influences cytochrome c552 maturation, FEMS Microbiology Leters 161 (1998) 1-6. [6] Jung, Y., Kwak, J., and Le, Y., High-level production of heme-containing holoproteins in escherichia coli, Applied Microbiology and Biotechnology 55 (2001) 187-191. 167 [7] Woodard S. I., and Dailey H. A., Regulation of heme biosynthesis in escherichia coli, Archives of Biochemistry and Biophysics 316 (1995) 110-115. [8] Mils, M., and Payne, S. M., Genetics and regulation of heme iron transport in shigela dysenteriae and detection of an analogous system in escherichia coli o157:H7, Journal of Bacteriology 177 (1995) 3004-3009. [9] Wandersman, C., and Stojiljkovic, I., Bacterial heme sources: The role of heme, hemoprotein receptors and hemophores, Current Opinion in Microbiology 3 (2000) 215- 220. [10] Bracken, C. S., Baer, M. T., Abdur-Rashid, A., Helms, W., and Stojiljkovic, I., Use of heme-protein complexes by the yersinia enterocolitica hemr receptor: Histidine residues are esential for receptor function, Journal of Bacteriology 181 (1999) 6063- 6072. [11] Torres, A. G., and Payne, S. M., Haem iron-transport system in enterohaemorrhagic escherichia coli o157:H7, Molecular Microbiology 23 (1997) 825-833. [12] Law, D., Virulence factors of escherichia coli o157 and other shiga toxin- producing e. Coli, Journal of Applied Microbiology 88 (2000) 729-745. [13] Law, D., and Kely, J., Use of heme and hemoglobin by escherichia coli o157 and other shiga-like- toxin-producing e. Coli serogroups, Infection and Imunity 63 (1995) 700-702. [14] Mils, M., and Payne, S., Identification of shua, the gene encoding the heme receptor of shigela dysenteriae, and analysis of invasion and intracelular multiplication of a shua mutant, Infect. Imun. 65 (1997) 5358-5363. 168 [15] Thompson, J. M., Jones, H. A., and Pery, R. D., Molecular characterization of the hemin uptake locus (hmu) from yersinia pestis and analysis of hmu mutants for hemin and hemoprotein utilization, Infection and Imunity 67 (1999) 3879-3892. [16] Schryvers, A. B., and Stojiljkovic, I., Iron acquisition systems in the pathogenic neiseria, Molecular Microbiology 32 (1999) 1117-1123. [17] Litwin, C. M., and Byrne, B. L., Cloning and characterization of an outer membrane protein of vibrio vulnificus required for heme utilization: Regulation of expresion and determination of the gene sequence, Infection and Imunity 66 (1998) 3134-3141. [18] Stojiljkovic, I., Larson, J., Hwa, V., Anic, S., and So, M., Hmbr outer membrane receptors of pathogenic neiseria spp.: Iron- regulated, hemoglobin-binding proteins with a high level of primary structure conservation, Journal of Bacteriology 178 (1996) 4670- 4678. [19] Stojiljkovic, I., and Hantke, K., Transport of haemin across the cytoplasmic membrane through a haemin-specific periplasmic binding-protein-dependent transport system in yersinia enterocolitica., Mol. Microbiol. 13 (1994) 719-732. [20] Law, D., and Kely, J., Use of heme and hemoglobin by escherichia-coli o157 and other shiga-like-toxin-producing escherichia-coli serogroups, Infection and Imunity 63 (1995) 700-702. [21] Hilar, A., Van Caesele, L., and Loewen, P. C., Intracelular location of catalase- peroxidase hydroperoxidase i of escherichia coli, FEMS Microbiology Leters 170 (1999) 307-312. 169 [22] Bandyopadhyay, P., and Steinman, H. M., Legionela pneumophila catalase- peroxidases: Cloning of the katb gene and studies of katb function, Journal of Bacteriology 180 (1998) 5369-5374. [23] Bandyopadhyay, P., and Steinman, H. M., Catalase-peroxidases of legionela pneumophila: Cloning of the kata gene and studies of kata function, Journal of Bacteriology 182 (2000) 6679-6686. [24] Brunder, W., Schmidt, H., and Karch, H., Katp, a novel catalase-peroxidase encoded by the large plasmid of enterohaemorrhagic escherichia coli o157:H7, Microbiology 142 (1996) 3305-3315. [25] Garcia, E., Nedialkov, Y. A., Eliott, J., Motin, V. L., and Brubaker, R. R., Molecular characterization of katy (antigen 5), a thermoregulated chromosomaly encoded catalase-peroxidase of yersinia pestis, Journal of Bacteriology 181 (1999) 3114- 3122. [26] Mehigh, R. J., and Brubaker, R. R., Major stable peptides of yersinia pestis synthesized during the low- calcium response, Infection and Imunity 61 (1993) 13-22. [27] Amemura-Maekawa, J., Mishima-Abe, S., Kura, F., Takahashi, T., and Watanabe, H., Identification of a novel periplasmic catalase-peroxidase kata of legionela pneumophila, FEMS Microbiology Leters 176 (1999) 339-344. [28] Bertrand, T., Eady, N. A. J., Jones, J. N., Nagy, J. M., Jamart-Gregoire, B., Raven, E. L., and Brown, K. A., Crystal structure of mycobacterium tuberculosis catalase- peroxidase, Journal of Biological Chemistry 279 (2004) 38991-38999. 170 [29] Yamada, Y., Fujiwara, T., Sato, T., Igarashi, N., and Tanaka, N., The 2.0 angstrom crystal structure of catalase-peroxidase from haloarcula marismortui, Nature Structural Biology 9 (2002) 691-695. [30] Carpena, X., Loprasert, S., Mongkolsuk, S., Switala, J., Loewen, P. C., and Fita, I., Catalase-peroxidase katg of burkholderia pseudomalei at 1.7 a resolution, Journal of Molecular Biology 327 (2003) 475-489. [31] Yu, S., Girotto, S., Zhao, X., and Magliozzo, R. S., Rapid formation of compound i and a tyrosyl radical in the y229f mutant of mycobacterium tuberculosis catalase- peroxidase disrupts catalase but not peroxidase function, J. Biol. Chem. 278 (2003) 44121-44127. [32] Santoni, E., Jakopitsch, C., Obinger, C., and Smulevich, G., Manipulating the covalent link betwen distal side tryptophan, tyrosine, and methionine in catalase- peroxidases: An electronic absorption and resonance raman study, Biopolymers 74 (2004) 46-50. [33] Jakopitsch, C., Kolarich, D., Petutschnig, G., Furtmuller, P. G., and Obinger, C., Distal side tryptophan, tyrosine and methionine in catalase-peroxidases are covalently linked in solution, FEBS Leters 552 (2003) 135-140. [34] Jakopitsch, C., Auer, M., Ivancich, A., Ruker, F., Furtmuller, P. G., and Obinger, C., Total conversion of bifunctional catalase-peroxidase (katg) to monofunctional peroxidase by exchange of a conserved distal side tyrosine, J. Biol. Chem. 278 (2003) 20185-20191. [35] Regelsberger, G., Jakopitsch, C., Ruker, F., Krois, D., Peschek, G. A., and Obinger, C., Efect of distal cavity mutations on the formation of compound i in catalase- peroxidases, Journal of Biological Chemistry 275 (2000) 22854-22861. 171 [36] Regelsberger, G., Jakopitsch, C., Furtmuller, P. G., Rueker, F., Switala, J., Loewen, P. C., and Obinger, C., The role of distal tryptophan in the bifunctional activity of catalase-peroxidases., Biochemical Society Transactions 29 (2001) 99-105. [37] Hilar, A., Peters, B., Pauls, R., Loboda, A., Zhang, H., Mauk, A. G., and Loewen, P. C., Modulation of the activities of catalase-peroxidase hpi of escherichia coli by site- directed mutagenesis, Biochemistry 39 (2000) 5868-5875. [38] Haliwel, B., and Gutteridge, J. M. C., Oxygen-toxicity, oxygen radicals, transition- metals and disease, Biochemical Journal 219 (1984) 1-14. [39] Koppenol, W. H., Thermodynamics of reactions involving oxyradicals and hydrogen-peroxide, Bioelectrochemistry and Bioenergetics 18 (1987) 3-11. [40] Wood, P. M., The potential diagram for oxygen at ph-7, Biochemical Journal 253 (1988) 287-289. [41] Miler, D. M., Buetner, G. R., and Aust, S. D., Transition-metals as catalysts of autoxidation reactions, Fre Radical Biology and Medicine 8 (1990) 95-108. [42] Surdhar, P. S., and Armstrong, D. A., Redox potentials of some sulfur-containing radicals, Journal of Physical Chemistry 90 (1986) 5915-5917. [43] Buetner, G. R., The pecking order of fre-radicals and antioxidants - lipid- peroxidation, alpha-tocopherol, and ascorbate, Archives of Biochemistry and Biophysics 300 (1993) 535-543. [44] Devries, S., Berden, J. A., and Slater, E. C., Properties of a semi-quinone anion located in the qh2-cytochrome-c oxidoreductase segment of the mitochondrial respiratory-chain, Febs Leters 122 (1980) 143-148. 172 [45] Anderson, R. F., Energetics of the one-electron reduction steps of riboflavin, fmn and fad to their fully reduced forms, Biochimica Et Biophysica Acta 722 (1983) 158-162. [46] Haris, D. C., Rinehart, A. L., Hereld, D., Schwartz, R. W., Burke, F. P., and Salvador, A. P., Reduction potential of iron in transferin, Biochimica Et Biophysica Acta 838 (1985) 295-301. [47] Klotz, I. M., and Kurtz, D. M., Metal-dioxygen complexes - a perspective, Chemical Reviews 94 (1994) 567-568. [48] Toyokuni, S., Iron and carcinogenesis: From fenton reaction to target genes, Redox Report 7 (2002) 189-197. [49] Lloyd, R. V., Hanna, P. M., and Mason, R. P., The origin of the hydroxyl radical oxygen in the fenton reaction, Fre Radical Biology and Medicine 22 (1997) 885-888. [50] Raleigh, J. A., and Shum, F. Y., Hydroxyl radical scavengers and membrane damage - supplementary role for alpha-tocopherol in scavenging secondary radicals, Radiation Research 94 (1983) 664-665. [51] Anbar, M., Meyerste.D, and Neta, P., Reactivity of aliphatic compounds towards hydroxyl radicals, Journal of the Chemical Society B-Physical Organic (1966) 742. [52] Pryor, W. A., Why is the hydroxyl radical the only radical that commonly adds to DNA? Hypothesis: It has a rare combination of high electrophilicity, high thermochemical reactivity, and a mode of production that can occur near DNA, Fre Radical Biology and Medicine 4 (1988) 219-223. [53] Jaruga, P. W., Rodriguez, H., and Dizdaroglu, M., Measurement of 8-hydroxy-2 '- deoxyadenosine in DNA by liquid chromatography/mas spectrometry, Fre Radical Biology and Medicine 31 (2001) 336-344. 173 [54] Zastawny, T. H., Altman, S. A., Randerseichhorn, L., Madurawe, R., Lumpkin, J. A., Dizdaroglu, M., and Rao, G., DNA-base modifications and membrane damage in cultured-mamalian-cels treated with iron ions, Fre Radical Biology and Medicine 18 (1995) 1013-1022. [55] Kumar, S., and Bandyopadhyay, U., Fre heme toxicity and its detoxification systems in human, Toxicology Leters 157 (2005) 175-188. [56] O'Brian, M., Heme synthesis in the rhizobium-legume symbiosis: A palete for bacterial and eukaryotic pigments, J. Bacteriol. 178 (1996) 2471-2478. [57] Ryter, S. W., and Tyrel, R. M., The heme synthesis and degradation pathways: Role in oxidant sensitivity - heme oxygenase has both pro- and antioxidant properties, Fre Radical Biology and Medicine 28 (2000) 289-309. [58] Shifman, J. M., Gibney, B. R., Sharp, R. E., and Dutton, P. L., Heme redox potential control in de novo designed four-alpha-helix bundle proteins, Biochemistry 39 (2000) 14813-14821. [59] Caughey, W. S., Smythe, G. A., Okeefe, D. H., Maskasky, J. E., and Smith, M. L., Heme-a of cytochrome-c oxidase - structure and properties - comparisons with heme-b, heme-c and heme-s and derivatives, Journal of Biological Chemistry 250 (1975) 7602- 7622. [60] Alen, J. W. A., Leach, N., and Ferguson, S. J., The histidine of the c-type cytochrome cxxch haem-binding motif esential for haem atachment by the escherichia coli cytochrome maturation (cm) apparatus, Biochemical Journal 389 (2005) 587-592. [61] Tomlinson, E. J., and Ferguson, S. J., Conversion of a c type cytochrome to a b type that spontaneously forms in vitro from apo protein and heme: Implications for c type 174 cytochrome biogenesis and folding, Procedings of the National Academy of Sciences of the United States of America 97 (2000) 5156-5160. [62] Antonini, E., and Brunori, M., in, 1971, North-Holland, Amsterdam. [63] White, R. E., and Coon, M. J., Oxygen activation by cytochrome-p-450, Annual Review of Biochemistry 49 (1980) 315-356. [64] Adachi, S., Nagano, S., Ishimori, K., Watanabe, Y., Morishima, I., Egawa, T., Kitagawa, T., and Makino, R., Roles of proximal ligand in heme-proteins - replacement of proximal histidine of human myoglobin with cysteine and tyrosine by site-directed mutagenesis as models for p-450, chloroperoxidase, and catalase, Biochemistry 32 (1993) 241-252. [65] Quilin, M. L., Arduini, R. M., Olson, J. S., and Philips, G. N., High-resolution crystal-structures of distal histidine mutants of sperm whale myoglobin, Journal of Molecular Biology 234 (1993) 140-155. [66] Krzywda, S., Murshudov, G. N., Brzozowski, A. M., Jaskolski, M., Scott, E. E., Klizas, S. A., Gibson, Q. H., Olson, J. S., and Wilkinson, A. J., Stabilizing bound o-2 in myoglobin by valine(68) (e11) to asparagine substitution, Biochemistry 37 (1998) 15896- 15907. [67] Perutz, M. F., Rossmann, M. G., Cullis, A. F., Muirhead, H., Wil, G., and North, A. C. T., Structure of hemoglobin. A thre-dimensional fourier synthesis at 5.5-a. Resolution, obtained by x-ray analysis, Nature 185 (1960) 416-422. [68] Kendrew, J. C. D., R. E.; Strandberg, B. E.; Hart, R. G.; Davies, D. R.; Philips, D. C.; Shore, V. C., Structure of myoglobin. Thre-dimensional fourier synthesis at 2 a. Resolution, Nature 185 (1960) 422-427. 175 [69] Collman, J. P., Boulatov, R., Sunderland, C. J., and Fu, L., Functional analogues of cytochrome c oxidase, myoglobin, and hemoglobin, Chemical Reviews 104 (2004) 561- 588. [70] Theorel, H., Crystaline myoglobin. V. Oxygen-binding curve of myoglobin, Biochemische Zeitschrift 268 (1934) 73-82. [71] Finch, J. T., Perutz, M. F., Bertles, J. F., and Doebler, J., Structure of sickled erythrocytes and of sickle-cel hemoglobin fibers, Procedings of the National Academy of Sciences of the United States of America 70 (1973) 718-722. [72] Gelin, B. R., Le, A. W. M., and Karplus, M., Hemoglobin tertiary structural-change on ligand-binding - its role in the co-operative mechanism, Journal of Molecular Biology 171 (1983) 489-559. [73] Klingenberg, M., Pigments of rat liver microsomes, Archives of Biochemistry and Biophysics 75 (1958) 376-386. [74] Sono, M., Roach, M. P., Coulter, E. D., and Dawson, J. H., Heme-containing oxygenases, Chemical Reviews 96 (1996) 2841-2887. [75] Lemberg, M. R., Cytochrome oxidase, Physiological Reviews 49 (1969) 48. [76] Wikstrom, M., Krab, K., and Saraste, M., Cytochrome oxidase a synthesis, ed., Academic Pres Inc., New York 1981. [77] Iwata, S., Ostermeier, C., Ludwig, B., and Michel, H., Structure at 2.8-angstrom resolution of cytochrome-c-oxidase from paracoccus-denitrificans, Nature 376 (1995) 660-669. 176 [78] Antalik, M., Jancura, D., Palmer, G., and Fabian, M., A role for the protein in internal electron transfer to the catalytic center of cytochrome c oxidase, Biochemistry 44 (2005) 14881-14889. [79] Varotsis, C., Zhang, Y., Appelman, E. H., and Babcock, G. T., Resolution of the reaction sequence during the reduction of o2 by cytochrome-oxidase, Procedings of the National Academy of Sciences of the United States of America 90 (1993) 237-241. [80] Chance, B., and Erecinsk.M, Flow flash kinetics of cytochrome a3-oxygen reaction in coupled and uncoupled mitochondria using liquid dye laser, Archives of Biochemistry and Biophysics 143 (1971) 675. [81] Gibson, Q. H., and Grenwood, C., Reactions of cytochrome oxidase with oxygen and carbon monoxide, Biochemical Journal 86 (1963) 541. [82] Turrens, J. F., Mitochondrial formation of reactive oxygen species, Journal of Physiology-London 552 (2003) 335-344. [83] Welinder, K. G., Superfamily of plant, fungal and bacterial peroxidases, Current Opinion in Structural Biology 2 (1992) 388-393. [84] Schuller, D. J., Ban, N., vanHuyste, R. B., McPherson, A., and Poulos, T. L., The crystal structure of peanut peroxidase, Structure 4 (1996) 311-321. [85] Boerjan, W., Ralph, J., and Baucher, M., Lignin biosynthesis, Annual Review of Plant Biology 54 (2003) 519-546. [86] Hilden, K. S., Makela, M. R., Hakala, T. K., Hatakka, A., and Lundel, T., Expresion on wood, molecular cloning and characterization of thre lignin peroxidase (lip) encoding genes of the white rot fungus phlebia radiata, Current Genetics 49 (2006) 97-105. 177 [87] Loew, G., Structure, spectra, and function of heme sites, International Journal of Quantum Chemistry 77 (2000) 54-70. [88] Wang, J. M., Mauro, J. M., Edwards, S. L., Oatley, S. J., Fishel, L. A., Ashford, V. A., Xuong, N. H., and Kraut, J., X-ray structures of recombinant yeast cytochrome-c peroxidase and 3 heme-cleft mutants prepared by site-directed mutagenesis, Biochemistry 29 (1990) 7160-7173. [89] Pennerhahn, J. E., Eble, K. S., McMurry, T. J., Renner, M., Balch, A. L., Groves, J. T., Dawson, J. H., and Hodgson, K. O., Structural characterization of horseradish- peroxidase using exafs spectroscopy - evidence for fe=o ligation in compound-i and compound-i, Journal of the American Chemical Society 108 (1986) 7819-7825. [90] Moss, T. H., Ehrenber.A, and Bearden, A. J., Mossbauer spectroscopic evidence for electronic configuration of iron in horseradish peroxidase and its peroxide derivatives, Biochemistry 8 (1969) 4159. [91] Houseman, A. L. P., Doan, P. E., Goodin, D. B., and Hofman, B. M., Comprehensive explanation of the anomalous epr-spectra of wild-type and mutant cytochrome-c peroxidase compound-es, Biochemistry 32 (1993) 4430-4443. [92] Witenbe, J. B., Noble, R. W., Witenbe.Ba, Antonini, E., Brunori, M., and Wyman, J., Studies on equilibria and kinetics of reactions of peroxidase with ligands .2. Reaction of feroperoxidase with oxygen, Journal of Biological Chemistry 242 (1967) 626. [93] Adediran, S. A., Kinetics of the formation of p-670 and of the decay of compound ii of horseradish peroxidase, Archives of Biochemistry and Biophysics 327 (1996) 279-284. [94] Yamazaki, I., and Piete, L. H., The mechanism of aerobic oxidase reaction catalyzed by peroxidase, Biochimica Et Biophysica Acta 77 (1963) 47-64. 178 [95] Nakajima, R., and Yamazaki, I., The mechanism of oxyperoxidase formation from feryl peroxidase and hydrogen-peroxide, Journal of Biological Chemistry 262 (1987) 2576-2581. [96] Arnao, M. B., Acosta, M., Delrio, J. A., and Garciacanovas, F., Inactivation of peroxidase by hydrogen-peroxide and its protection by a reductant agent, Biochimica Et Biophysica Acta 1038 (1990) 85-89. [97] Warburg, O., Versuche und uberledbeudem carcinomgewebe (methoden). Biochem Z. 142 (1923) 317-333. [98] Stern, K. G., The constitution of the prosthetic group of catalase, J. Biol. Chem. 112 (1936) 661-669. [99] Fita, I., and Rossmann, M. G., The active center of catalase, Journal of Molecular Biology 185 (1985) 21-37. [100] Zamocky, M., and Koller, F., Understanding the structure and function of catalases: Clues from molecular evolution and in vitro mutagenesis, Progres in Biophysics & Molecular Biology 72 (1999) 19-66. [101] Chelikani, P., Fita, I., and Loewen, P. C., Diversity of structures and properties among catalases, Celular and Molecular Life Sciences 61 (2004) 192-208. [102] Melik-Adamyan, W. R., Barynin, V. V., Vagin, A. A., Borisov, V. V., Vainshtein, B. K., Fita, I., Murthy, M. R. N., and Rossmann, M. G., Comparison of beef liver and penicilium vitale catalases, Journal of Molecular Biology 188 (1986) 63-72. [103] Chaga, G. S., Medin, A. S., Chaga, S. G., and Porath, J. O., Isolation and characterization of catalase from penicilium-chrysogenum, Journal of Chromatography 604 (1992) 177-183. 179 [104] Murshudov, G. N., Grebenko, A. I., Barynin, V., Dauter, Z., Wilson, K. S., Vainshtein, B. K., MelikAdamyan, W., Bravo, J., Feran, J. M., Ferer, J. C., Switala, J., Loewen, P. C., and Fita, I., Structure of the heme d of penicilium vitale and escherichia coli catalases, Journal of Biological Chemistry 271 (1996) 8863-8868. [105] Paoli, M., Marles-Wright, J., and Smith, A., Structure-function relationships in heme-proteins, DNA and Cel Biology 21 (2002) 271-280. [106] Vainshtein, B. K., Melikadamyan, W. R., Barynin, V. V., Vagin, A. A., and Grebenko, A. I., 3-dimensional structure of the enzyme catalase, Nature 293 (1981) 411- 412. [107] Bravo, J., Mate, M. J., Schneider, T., Switala, J., Wilson, K., Loewen, P. C., and Fita, I., Structure of catalase hpii from escherichia coli at 1.9 angstrom resolution, Proteins-Structure Function and Genetics 34 (1999) 155-166. [108] Nicholis, P., Fita, I., and Loewen, P. C., Enzymology and structure of catalase, in Advances in inorganic chemistry, in Sykes, A., (Ed.), Academic Pres, San Diego, 2001, pp. 51-106. [109] Kirkman, H., Galiano, S., and Gaetani, G., The function of catalase-bound nadph, J. Biol. Chem. 262 (1987) 660-666. [110] Ng, V. H., Cox, J. S., Sousa, A. O., MacMicking, J. D., and McKinney, J. D., Role of katg catalase-peroxidase in mycobacterial pathogenesis: Countering the phagocyte oxidative burst, Molecular Microbiology 52 (2004) 1291-1302. [111] Triggsraine, B. L., Doble, B. W., Mulvey, M. R., Sorby, P. A., and Loewen, P. C., Nucleotide-sequence of katg, encoding catalase hpi of escherichia-coli, Journal of Bacteriology 170 (1988) 4415-4419. 180 [112] Claiborne, A., and Fridovich, I., Purification of the ortho-dianisidine peroxidase from escherichia-coli-b - physicochemical characterization and analysis of its dual catalytic and peroxidatic activities, Journal of Biological Chemistry 254 (1979) 4245- 4252. [113] Cannac-Cafrey, V., Hudry-Clergeon, G., Petilot, Y., Gagnon, J., Zacai, G., and Franzeti, B., The protein sequence of an archaeal catalase-peroxidase, Biochimie 80 (1998) 1003-1011. [114] Welinder, K. G., Bacterial catalase-peroxidases are gene duplicated members of the plant peroxidase superfamily, Biochimica Et Biophysica Acta 1080 (1991) 215-220. [115] Paterson, W. R., and Poulos, T. L., Crystal-structure of recombinant pea cytosolic ascorbate peroxidase, Biochemistry 34 (1995) 4331-4341. [116] Baker, R. D., Cook, C. O., and Goodwin, D. C., Properties of catalase-peroxidase lacking its c-terminal domain, Biochemical and Biophysical Research Communications 320 (2004) 833-839. [117] Sundaramoorthy, M., Kishi, K., Gold, M. H., and Poulos, T. L., The crystal structure of manganese peroxidase from phanerochaete chrysosporium at 2.06-a resolution, Journal of Biological Chemistry 269 (1994) 32759-32767. [118] Sutherland, G. R. J., and Aust, S. D., Thermodynamics of binding of the distal calcium to manganese peroxidase, Biochemistry 36 (1997) 8567-8573. [119] Zamocky, M., Regelsberger, G., Jakopitsch, C., and Obinger, C., The molecular peculiarities of catalase-peroxidases, Febs Leters 492 (2001) 177-182. [120] Li, Y., in Chemistry and Biochemistry, 2005 p. 190 Auburn University, Auburn,AL. 181 [121] Li, Y. J., and Goodwin, D. C., Vital roles of an interhelical insertion in catalase- peroxidase bifunctionality, Biochemical and Biophysical Research Communications 318 (2004) 970-976. [122] Murray, C. J. L., World tuberculosis burden, Lancet 335 (1990) 1043-1044. [123] Bernstein, J., Lott, W. A., Steinberg, B. A., and Yale, H. L., Chemotherapy of experimental tuberculosis .5. Isonicotinic acid hydrazide (nydrazid) and related compounds, American Review of Tuberculosis 65 (1952) 357-364. [124] Zhang, Y., Heym, B., Alen, B., Young, D., and Cole, S., The catalase peroxidase gene and isoniazid resistance of mycobacterium-tuberculosis, Nature 358 (1992) 591- 593. [125] Shikama, K., A controversy on the mechanism of autoxidation of oxymyoglobin and oxyhemoglobin - oxidation, disociation, or displacement, Biochemical Journal 223 (1984) 279-280. [126] Shikama, K., The molecular mechanism of autoxidation for myoglobin and hemoglobin: A venerable puzzle, Chemical Reviews 98 (1998) 1357-1373. [127] Walace, W. J., Houtchens, R. A., Maxwel, J. C., and Caughey, W. S., Mechanism of autoxidation for hemoglobins and myoglobins - promotion of superoxide production by protons and anions, Journal of Biological Chemistry 257 (1982) 4966-4977. [128] Brantley, R. E., Smerdon, S. J., Wilkinson, A. J., Singleton, E. W., and Olson, J. S., The mechanism of autooxidation of myoglobin, Journal of Biological Chemistry 268 (1993) 6995-7010. [129] George, P., and Stratmann, C. J., The oxidation of myoglobin to metmyoglobin by oxygen .3. Kinetic studies in the presence of carbon monoxide, and at diferent hydrogen- 182 ion concentrations with considerations regarding the stability of oxymyoglobin, Biochemical Journal 57 (1954) 568-573. [130] Weis, J. J., Nature of iron-oxygen bond in oxyhaemoglobin, Nature 202 (1964) 83. [131] Sugawara, Y., Matsuoka, A., Kaino, A., and Shikama, K., Role of globin moiety in the autoxidation reaction of oxymyoglobin - efect of 8 m urea, Biophysical Journal 69 (1995) 583-592. [132] Tsubamoto, Y., Matsuoka, A., Yusa, K., and Shikama, K., Protozoan myoglobin from paramecium-caudatum - its autoxidation reaction and hemichrome formation, European Journal of Biochemistry 193 (1990) 55-59. [133] Gorsky, L., Koop, D., and Coon, M., On the stoichiometry of the oxidase and monooxygenase reactions catalyzed by liver microsomal cytochrome p-450. Products of oxygen reduction, J. Biol. Chem. 259 (1984) 6812-6817. [134] Zangar, R. C., Davydov, D. R., and Verma, S., Mechanisms that regulate production of reactive oxygen species by cytochrome p450, Toxicology and Applied Pharmacology 199 (2004) 316-331. [135] Nordblom, G. D., and Coon, M. J., Hydrogen-peroxide formation and stoichiometry of hydroxylation reactions catalyzed by highly purified liver microsomal cytochrome-p-450, Archives of Biochemistry and Biophysics 180 (1977) 343-347. [136] Hlavica, P., and Lewis, D. F. V., Alosteric phenomena in cytochrome p450- catalyzed monooxygenations, European Journal of Biochemistry 268 (2001) 4817-4832. [137] Gruenke, L. D., Konopka, K., Cadieu, M., and Waskel, L., The stoichiometry of the cytochrome p-450-catalyzed metabolism of methoxyflurane and benzphetamine in the 183 presence and absence of cytochrome b(5), Journal of Biological Chemistry 270 (1995) 24707-24718. [138] Mehigh, R. J., and Brubaker, R. R., Major stable peptides of yersinia-pestis synthesized during the low-calcium response, Infection and Imunity 61 (1993) 13-22. [139] Demircay, Z., Eksioglu-Demiralp, E., Ergunt, T., and Akoglut, T., Phagocytosis and oxidative burst by neutrophils in patients with recurrent furunculosis, British Journal of Dermatology 138 (1998) 1036-1038. [140] Bolwel, G. P., and Wojtaszek, P., Mechanisms for the generation of reactive oxygen species in plant defence - a broad perspective, Physiological and Molecular Plant Pathology 51 (1997) 347-366. [141] Babior, B. M., Nadph oxidase, Current Opinion in Imunology 16 (2004) 42-47. [142] Jesaitis, A. J., Structure of human phagocyte cytochrome-b and its relationship to microbicidal superoxide production, Journal of Imunology 155 (1995) 3286-3288. [143] Cross, A. R., Rae, J., and Curnutte, J. T., Cytochrome-b(-245) of the neutrophil superoxide-generating system contains 2 nonidentical hemes - potentiometric studies of a mutant form of gp91(phox), Journal of Biological Chemistry 270 (1995) 17075-17077. [144] Fuji, H., Finnegan, M. G., Miki, T., Crouse, B. R., Kakinuma, K., and Johnson, M. K., Spectroscopic identification of the heme axial ligation of cytochrome b(558) in the nadph oxidase of porcine neutrophils, Febs Leters 377 (1995) 345-348. [145] Fuji, H., Finnegan, M. G., and Johnson, M. K., The active form of the feric heme in neutrophil cytochrome b(558) is low-spin in the reconstituted cel-fre system in the presence of amphophil, Journal of Biochemistry 126 (1999) 708-714. 184 [146] Cross, A. R., Parkinson, J. F., and Jones, O. T. G., Mechanism of the superoxide- producing oxidase of neutrophils - o-2 is necesary for the fast reduction of cytochrome b-245 by nadph, Biochemical Journal 226 (1985) 881-884. [147] Rotrosen, D., Yeung, C. L., Leto, T. L., Malech, H. L., and Kwong, C. H., Cytochrome-b558 - the flavin-binding component of the phagocyte nadph oxidase, Science 256 (1992) 1459-1462. [148] Babior, B. M., Nadph oxidase: An update, Blood 93 (1999) 1464-1476. [149] Rosen, H., and Klebanoff, S. J., Bactericidal activity of a superoxide anion- generating system - model for the polymorphonuclear leukocyte, Journal of Experimental Medicine 149 (1979) 27-39. [150] Eberhardt, M. K., Reactive oxygen metabolites, ed., CRC Pres LC, Boca Raton, FL 2001. [151] Wengenack, N. L., Jensen, M. P., Rusnak, F., and Stern, M. K., Mycobacterium tuberculosiskatg is a peroxynitritase, Biochemical and Biophysical Research Communications 256 (1999) 485-487. [152] Beckman, J. S., -oono: Rebounding from nitric oxide, Circ Res 89 (2001) 295-297. [153] Bruneli, L., Crow, J. P., and Beckman, J. S., The comparative toxicity of nitric- oxide and peroxynitrite to escherichia-coli, Archives of Biochemistry and Biophysics 316 (1995) 327-334. [154] Goodwin, D. C., Landino, L. M., and Marnet, L. J., Reactions of prostaglandin endoperoxide synthase with nitric oxide and peroxynitrite, Drug Metabolism Reviews 31 (1999) 273-294. 185 [155] Sambrook, J., Maniatis, T., and Fritsch, E. F., Molecular cloning: A laboratory manual, ed., Cold Spring Harbor 1989. [156] Gil, S. C., and von Hippel, P. H., Calculation of protein extinction coeficients from amino acid sequence data, Analytical Biochemistry 182 (1989) 319-326. [157] Falk., J. E., Porphyrins and metaloporphyrins, ed., Elsevier Publishing, New York 1964. [158] Nelson, D. P., and Kiesow, L. A., Enthalpy of decomposition of hydrogen-peroxide by catalase at 25 o c (with molar extinction coeficients of h 2 o 2 solutions in uv), Analytical Biochemistry 49 (1972) 474. [159] S.L. Scott, W.-J. C., A. Bakac and J.H. Espenson, J. Phys. Chem. 97 (1993) 6710? 6714. [160] Helman U., Wernstedt C., Gonez J., and Heldin C. H., Improvement of an in-gel digestion procedure for the micropreparation of internal protein fragments for amino acid sequencing, Analytical Biochemistry 224 (1995) 451-455. [161] Rosenfeld, J., Capdeviele, J., Guilemot, J. C., and Ferara, P., In-gel digestion of proteins for internal sequence analysis after one- or two-dimensional gel electrophoresis, Analytical Biochemistry 203 (1992) 173-179. [162] Studier, F. W., Use of bacteriophage t7 lysozyme to improve an inducible t7 expresion system, Journal of Molecular Biology 219 (1991) 37-44. [163] Nielsen, H., Engelbrecht, J., Brunak, S., and Von Heijne, G., Identification of prokaryotic and enkaryotic signal peptides and prediction of their cleavage sites, Protein Engineering 10 (1997) 1-6. 186 [164] Hilar, A., and Loewen, P. C., Comparison of isoniazid oxidation catalyzed by bacterial catalase-peroxidases and horseradish peroxidase, Archives of Biochemistry and Biophysics 323 (1995) 438-446. [165] Chouchane, S., Giroto, S., Kapetanaki, S., Schelvis, J. P. M., Yu, S., and Magliozzo, R. S., Analysis of heme structural heterogeneity in mycobacterium tuberculosis catalase-peroxidase (katg), Journal of Biological Chemistry 278 (2003) 8154-8162. [166] Chouchane, S., Giroto, S., Kapetanaki, S., Schelvis, J. P. M., Yu, S. W., and Magliozzo, R. S., Analysis of heme structural heterogeneity in mycobacterium tuberculosis catalase-peroxidase (katg), Journal of Biological Chemistry 278 (2003) 8154-8162. [167] Jakopitsch, C., Ruker, F., Regelsberger, G., Dockal, M., Peschek, G. A., and Obinger, C., Catalase-peroxidase from the cyanobacterium synechocystis pcc 6803: Cloning, overexpresion in escherichia coli, and kinetic characterization, Biological Chemistry 380 (1999) 1087-1096. [168] Erman, J. E., Kinetic studies of fluoride binding by cytochrome-c peroxidase, Biochemistry 13 (1974) 34-39. [169] Dunford, H. B., Hewson, W. D., and Steiner, H., Horseradish-peroxidase .29. Reactions in water and deuterium-oxide - cyanide binding, compound-i formation, and reactions of compound-i and compound-i with ferocyanide, Canadian Journal of Chemistry-Revue Canadienne De Chimie 56 (1978) 2844-2852. 187 [170] Johnsson, K., Froland, W. A., and Schultz, P. G., Overexpresion, purification, and characterization of the catalase-peroxidase katg from ycobacterium tuberculosis, Journal of Biological Chemistry 272 (1997) 2834-2840. [171] Yamada, Y., Fujiwara, T., Sato, T., Igarashi, N., and Tanaka, N., The 2.0 ? crystal structure of catalase-peroxidase from haloarcula marismortui, Nature Structural Biology 9 (2002) 691-695. [172] Bertrand, T., Eady, N. A. J., Jones, J. N., Jesmin, Nagy, J. M., Jamart-Gregoire, B., Raven, E. L., and Brown, K. A., Crystal structure of mycobacterium tuberculosis catalase-peroxidase, J. Biol. Chem. 279 (2004) 38991-38999. [173] Donald, L. J., Krokhin, O. V., Duckworth, H. W., Wiseman, B., Demagarn, T., Singh, R., Switala, J., Carpena, X., Fita, I., and Loewen, P. C., Characterization of the catalase-peroxidase katg from burkholderia pseudomalei by mas spectrometry, J. Biol. Chem. 278 (2003) 35687-35692. [174] Arnao, M. B. A., M.; Del Rio, J. A.; Varon, R.; Garcia-Canovas, F, A kinetic study on the suicide inactivation of peroxidase by hydrogen peroxide., Biochimica et Biophysica Acta (BA) - Bioenergetics 1041 (1990) 43-47. [175] Arnao, M. B. A., M.; Del Rio, J. A.; Garcia-Canovas, F, Inactivation of peroxidase by hydrogen peroxide and its protection by a reductant agent., Biochimica et Biophysica Acta (BA) - Bioenergetics 1038 (1990) 85-89. [176] Baynton, K. J. B., Jatinder K.; Biswas, Nihar; Taylor, Keith E., Inactivation of horseradish peroxidase by phenol and hydrogen peroxide: A kinetic investigation., Biochimica et Biophysica Acta (BA) - Bioenergetics 1206 (1994) 272-278. 188 [177] Nakajima, R., and Yamazaki, I., The mechanism of oxyperoxidase formation from feryl peroxidase and hydrogen peroxide, J. Biol. Chem. 262 (1987) 2576-2581. [178] Jakopitsch, C., Wanasinghe, A., Jantschko, W., Furtmuller, P. G., and Obinger, C., Kinetics of interconversion of ferous enzymes, compound i and compound ii, of wild- type synechocystis catalase-peroxidase and y249f, Journal of Biological Chemistry 280 (2005) 9037-9042. [179] Cohn, S., Alternatives to blood in the 21st century, Critical Care 8 (2004) S15 - S17. [180] Fitzpatrick, C. M., Savage, S. A., Kerby, J. D., Clouse, W. D., and Kashyap, V. S., Resuscitation with a blood substitute causes vasoconstriction without nitric oxide scavenging in a model of arterial hemorrhage, Journal of the American College of Surgeons 199 (2004) 693-701. [181] Eich, R. F., Li, T. S., Lemon, D. D., Doherty, D. H., Curry, S. R., Aitken, J. F., Mathews, A. J., Johnson, K. A., Smith, R. D., Philips, G. N., and Olson, J. S., Mechanism of no-induced oxidation of myoglobin and hemoglobin, Biochemistry 35 (1996) 6976-6983. [182] Olson, J. S., and Mailet, D. H., Designing recombinant hemoglobin for use as a blood subtitute, in Blood substitutes, in Winslow, R. M., (Ed.), Academic Pres, New York, NY, 2005, pp. 354-374. [183] Light, W., 3d, and Olson, J., Transmembrane movement of heme, J. Biol. Chem. 265 (1990) 15623-15631. [184] Light, W., 3d, and Olson, J., The efects of lipid composition on the rate and extent of heme binding to membranes, J. Biol. Chem. 265 (1990) 15632-15637. 189 [185] Cannon, J. B., Kuo, F. S., Pasternack, R. F., Wong, N. M., and Mullereberhard, U., Kinetics of the interaction of hemin liposomes with heme binding-proteins, Biochemistry 23 (1984) 3715-3721. [186] Varnado, C. L., Hertwig, K. M., Thomas, R., Roberts, J. K., and Goodwin, D. C., Properties of a novel periplasmic catalase-peroxidase from escherichia coli o157:H7, Archives of Biochemistry and Biophysics 421 (2004) 166-174. [187] Stojiljkovic, I., Kumar, V., and Srinivasan, N., Non-iron metaloporphyrins: Potent antibacterial compounds that exploit haem/hb uptake systems of pathogenic bacteria, Molecular Microbiology 31 (1999) 429-442. [188] Inoue, E., Takahashi, Y., Imai, Y., and Kamataki, T., Development of bacterial expresion system with high yield of cyp3a7, a human fetus- specific form of cytochrome p450, Biochemical and Biophysical Research Communications 269 (2000) 623-627. [189] Faguy, D. M., and Doolitle, W. F., Horizontal transfer of catalase-peroxidase genes betwen archaea and pathogenic bacteria, Trends in Genetics 16 (2000) 196-197. [190] Klotz, M. G., and Loewen, P. C., The molecular evolution of catalatic hydroperoxidases: Evidence for multiple lateral transfer of genes betwen prokaryota and from bacteria into eukaryota, Molecular Biology and Evolution 20 (2003) 1098-1112. [191] Lathem, W. W., Bergsbaken, T., Witowski, S. E., Perna, N. T., and Welch, R. A., Acquisition of stce, a c1 esterase inhibitor-specific metaloprotease, during the evolution of escherichia coli o157:H7, Journal of Infectious Diseases 187 (2003) 1907-1914. [192] Bandyopadhyay, P., Byrne, B., Chan, Y., Swanson, M. S., and Steinman, H. M., Legionela pneumophila catalase-peroxidases are required for proper traficking and growth in primary macrophages, Infection and Imunity 71 (2003) 4526-4535. 190 [193] Kengen, S. W., Bikker, F. J., Hagen, W. R., de Vos, W. M., and van der Oost, J., Characterization of a catalase-peroxidase from the hyperthermophilic archaeon archaeoglobus fulgidus, Extremophiles: Life Under Extreme Conditions 5 (2001) 323- 332. [194] Kim, J.-A., Sha, Z., and Mayfield, J. E., Regulation of brucela abortus catalase, Infection and Imunity 68 (2000) 3861-3866. [195] Sha, Z., Stabel, T. J., and Mayfield, J. E., Brucela abortus catalase is a periplasmic protein lacking a standard signal sequence, Journal of Bacteriology 176 (1994) 7375- 7377. [196] Ruby, E. G., and McFal-Ngai, M. J., Oxygen-utilizing reactions and symbiotic colonization of the squid light organ by vibrio fischeri, Trends in Microbiology 7 (1999) 414-420. [197] Visick, K. L., and Ruby, E. G., The periplasmic, group ii catalase of vibrio fischeri is required for normal symbiotic competence and is induced both by oxidative stres and by approach to stationary phase, Journal of Bacteriology 180 (1998) 2087-2092. [198] Klotz, M. G., Kim, Y. C., Katsuwon, J., and Anderson, A. J., Cloning, characterization and phenotypic-expresion in escherichia-coli of catf, which encodes the catalytic subunit of catalase isozyme catf of pseudomonas-syringae, Applied Microbiology and Biotechnology 43 (1995) 656-666. [199] Klotz, M. G., and Hutcheson, S. W., Multiple periplasmic catalases in phytopathogenic strains of pseudomonas-syringae, Applied and Environmental Microbiology 58 (1992) 2468-2473. 191 [200] Slutsker, L., Ries, A. A., Grene, K. D., Wels, J. G., Hutwagner, L., and Grifin, P. M., Escherichia coli o157:H7 diarhea in the united states: Clinical and epidemiologic features, Annals of Internal Medicine 126 (1997) 505-513. [201] Bhaskar, B., Imoos, C. E., Shimizu, H., Sulc, F., Farmer, P. J., and Poulos, T. L., A novel heme and peroxide-dependent tryptophan-tyrosine cross-link in a mutant of cytochrome c peroxidase, Journal of Molecular Biology 328 (2003) 157-166. [202] Ghiladi, R. A., Medzihradszky, K. F., and de Montelano, P. R. O., Role of the met- tyr-trp cross-link in mycobacterium tuberculosis catalase-peroxidase (katg) as revealed by katg(m255i), Biochemistry 4 (2005) 15093-15105. [203] Donald, L. J., Krokhin, O. V., Duckworth, H. W., Wiseman, B., Demagarn, T., Singh, R., Switala, J., Carpena, X., Fita, I., and Loewen, P. C., Characterization of the catalase-peroxidase katg from burkholderia pseudomalei by mas spectrometry, Journal of Biological Chemistry 278 (2003) 35687-35692. [204] Jakopitsch, C., Auer, M., Ivancich, A., Ruker, F., Furtmuller, P. G., and Obinger, C., Total conversion of bifunctional catalase-peroxidase (katg) to monofunctional peroxidase by exchange of a conserved distal side tyrosine, Journal of Biological Chemistry 278 (2003) 20185-20191.