Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Iron binding and oxidation by Pseudo-nitzschia multiseries ferritin Pfaffen, Stephanie 2014

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

Item Metadata

Download

Media
24-ubc_2014_november_pfaffen_stephanie.pdf [ 6MB ]
Metadata
JSON: 24-1.0167604.json
JSON-LD: 24-1.0167604-ld.json
RDF/XML (Pretty): 24-1.0167604-rdf.xml
RDF/JSON: 24-1.0167604-rdf.json
Turtle: 24-1.0167604-turtle.txt
N-Triples: 24-1.0167604-rdf-ntriples.txt
Original Record: 24-1.0167604-source.json
Full Text
24-1.0167604-fulltext.txt
Citation
24-1.0167604.ris

Full Text

IRON BINDING AND OXIDATION BY PSEUDO-NITZSCHIA MULTISERIES FERRITIN   by  Stephanie Pfaffen  M.Sc., Universität Bern, 2008  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Microbiology and Immunology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   August 2014   © Stephanie Pfaffen, 2014  ii ABSTRACT A novel ferritin was identified in marine pennate diatoms, unicellular photosynthetic organisms that play a major role in global primary production and carbon sequestration in the deep ocean. The expression of the iron storage and detoxifying protein ferritin is thought to facilitate the blooming of pennate diatoms after iron fertilization in the open ocean. X-ray structures of Pseudo-nitzschia multiseries ferritin (PmFTN) from crystals soaked for various durations in ferrous iron and zinc sulfate revealed three distinct metal binding sites; sites A and B comprise the catalytic ferroxidase centre, and site C forms a pathway leading toward the central cavity where iron storage occurs. In contrast, crystal structures derived from anaerobically grown and ferrous iron soaked crystals revealed only one ferrous ion occupying site A. Together with kinetic analysis, these studies suggest a model of stepwise iron binding to the ferroxidase centre of PmFTN followed by a very fast iron oxidation phase and partial mobilization of iron from the ferroxidase centre. Using a combination of rapid reaction kinetics and high resolution crystallography, the function of site C was investigated with site C and site B/C variants. Glu130, a site B/C ligand, functions in stabilizing Fe(III) bound at the ferroxidase centre and as a consequence reducing the rate of mineralization. Furthermore, Glu44, a site C ligand, is shown to be important for limiting the rate of post-oxidation reorganisation of iron coordination. Iron was observed within the B-channels, first identified in prokaryotic ferritins and BFRs, of the E44Q variant of PmFTN and provides the first evidence that these channels are possible routes for Fe(II) entry into the cavity. The anaerobic crystal structure of the bacterioferritin from E. coli (EcBFR) revealed two Fe(II) at the ferroxidase centre sites A and B. In comparison with PmFTN, differences in ferrous iron binding and reaction rates are further evidence that in EcBFR a distinct mechanism is in operation.  iii Clearly, PmFTN shows some characteristics of bacterial ferritins. Moreover, retention of iron at the ferroxidase centre at the expense of mineralization points to a role for this diatom ferritin in facilitating short term rather than long term iron storage.             iv PREFACE The work described in chapters 3 and 4 was mostly drawn from a peer-reviewed publication and a manuscript in preparation. The following contributions were made by collaborators and lab members:    Chapter 3 Chapter 3 was published as: Pfaffen, S., Abdulqadir, R., Le Brun, N. E., and Murphy, M. E. P. (2013) Mechanism of ferrous iron binding and oxidation by ferritin from a pennate diatom. The Journal of Biological Chemistry 288, 14917-25. Angelé Arrieta from Dr. M. E. P. Murphy’s lab provided the wild type PmFTN plasmid construct. I performed the PmFTN protein expression, purification, and crystallization. I also solved all the PmFTN crystal structures and performed the anaerobic work in a glove box. Moreover, I conducted the stopped-flow absorption spectroscopy study at the University of East Anglia with the help of Dr. N. E. Le Brun and R. Abdulqadir. Regeneration experiments were performed by R. Abdulqadir. I wrote the first draft of the manuscript with contributions by Dr. N. E. Le Brun for the kinetics section. The manuscript was edited by Drs. M. E. P. Murphy and N. E. Le Brun.  Chapter 4 Chapter 4 is a manuscript in preparation: Pfaffen, S., Bradley, J. M., Abdulqadir, R., Firme, J. M. R., Moore, G. R., Murphy, M. E. P., and Le Brun, N. E. Glu130 and Glu44 regulate the flux of iron through the ferroxidase centre of a diatom ferritin.   v I performed the site-directed mutagenesis with the help of J. M. R. Firme. I performed the protein expression, purification, and crystallization of all the PmFTN variants and solved all the crystal structures. Moreover, I conducted parts of the stopped-flow measurements with the help of Dr. J. M. Bradley and R. Abdulqadir. Regeneration and mineralization experiments were performed by Dr. J. M. Bradley and R. Abdulqadir. Dr. J. M. Bradley, R. Abdulqadir, Dr. N. E. Le Brun, Dr. G. R. Moore, and I performed stopped-flow data analyses. I wrote the first draft of the manuscript with contributions by Dr. N. E. Le Brun for the kinetics section. Dr. M. E. P. Murphy assisted with analyses of the crystal structures and edited the manuscript.  Chapter 5 Dr. N. E. Le Brun provided the purified E. coli bacterioferritin. I grew crystals, performed ferrous iron soaking experiments in the anaerobic glove box, collected the X-ray data, and solved the structure. Dr. M. E. P. Murphy assisted with data analysis.   vi TABLE OF CONTENTS Abstract ........................................................................................................................................... ii Preface............................................................................................................................................ iv Table of Contents ........................................................................................................................... vi List of Tables ................................................................................................................................. ix List of Figures ..................................................................................................................................x List of Abbreviations .................................................................................................................... xii Acknowledgements ...................................................................................................................... xiv CHAPTER 1: Introduction ............................................................................................................. 1 1.1 Iron in biology .............................................................................................................. 1 1.2 The ferritin family ........................................................................................................ 2 1.2.1 Ferritin..................................................................................................................... 6 1.2.1.1 Vertebrate ferritin............................................................................................ 6 1.2.1.2 Phytoferritin .................................................................................................... 8 1.2.1.3 Bacterial ferritin .............................................................................................. 9 1.2.1.4 Archeal ferritin .............................................................................................. 10 1.2.1.5 Diatom ferritin .............................................................................................. 11 1.2.2 Bacterioferritin ...................................................................................................... 13 1.2.3 Dps ........................................................................................................................ 14 1.3 Mechanism of iron storage ......................................................................................... 16 1.3.1 Iron mineralization in H-chain ferritin .................................................................. 18 1.3.2 Iron mineralization in bacterial ferritin ................................................................. 21 1.3.3 Iron mineralization in BFR ................................................................................... 22 1.3.4 Iron mineralization in archeal ferritin ................................................................... 25 1.3.5 Sequential displacement of Fe(III) by Fe(II) ........................................................ 25 1.4 Ferritin mineral core................................................................................................... 28 1.5 Iron entry and exit ...................................................................................................... 29 1.5.1 The iron entry channels ......................................................................................... 29 1.5.2 Mineral dissolution and iron exit .......................................................................... 32 1.6 Marine diatoms .......................................................................................................... 33  vii 1.7 Objectives................................................................................................................... 37 CHAPTER 2: Material and methods ............................................................................................ 39 2.1 Constructs and site-directed mutagenesis .................................................................. 39 2.1.1. Wild type PmFTN construct ............................................................................. 39 2.1.2. Site-directed mutagenesis ................................................................................. 39 2.2 Recombinant protein expression and purification of wild type PmFTN and  variants .................................................................................................................................. 40 2.3 Crystallization and structure solution ........................................................................ 42 2.3.1 Wild type PmFTN ................................................................................................. 42 2.3.2 PmFTN variants E44Q, E44H, and E130A .......................................................... 44 2.3.3 E. coli bacterioferritin ........................................................................................... 45 2.4 Stopped-flow absorption spectroscopy ...................................................................... 46 CHAPTER 3: Mechanism of ferrous iron binding and oxidation by ferritin from a pennate diatom ........................................................................................................................................... 47 3.1 Introduction ................................................................................................................ 47 3.2 Results ........................................................................................................................ 48 3.2.1 Overall fold ........................................................................................................... 48 3.2.2 Iron binding to PmFTN in the absence of dioxygen ............................................. 51 3.2.3 Iron binding to PmFTN in the presence of dioxygen ........................................... 53 3.2.4 PmFTN binds Zn(II) ............................................................................................. 55 3.2.5 Stopped-flow absorption spectroscopy ................................................................. 56 3.3 Discussion .................................................................................................................. 59 CHAPTER 4: Glu130 and Glu44 regulate the flux of iron through the ferroxidase centre ......... 65 4.1 Introduction ................................................................................................................ 65 4.2 Results ........................................................................................................................ 66 4.2.1 Substitutions of Glu44 and Glu130 disrupt Fe(II) binding cooperativity but not catalytic activity ................................................................................................................ 66 4.2.2 Enhanced rate of post-oxidation reorganization in PmFTN variants .................... 71 4.2.3 E130A PmFTN exhibits significant regeneration of the initial rapid oxidation phase ............................................................................................................................... 74  viii 4.2.4 E130A PmFTN exhibits a ten-fold increase in mineralization rate compared to wild type PmFTN .............................................................................................................. 77 4.2.5 Overall fold of variant crystal structures .............................................................. 78 4.2.6 The ferroxidase centre of site C variant E44Q...................................................... 80 4.2.7 The ferroxidase centre of site C variant E44H...................................................... 83 4.2.8 The ferroxidase centre of site B/C variant E130A ................................................ 84 4.3 Discussion .................................................................................................................. 85 CHAPTER 5: Ferrous iron binding by E. coli bacterioferritin ..................................................... 91 5.1 Introduction ................................................................................................................ 91 5.2 Results ........................................................................................................................ 92 5.3 Discussion .................................................................................................................. 98 CHAPTER 6: Overview and future directions ........................................................................... 101 6.1 PmFTN has characteristics of bacterial ferritins ...................................................... 101 6.2 Ferrous iron binding by EcBFR differs from that by PmFTN ................................. 105 6.3 Future directions ...................................................................................................... 106 6.3.1 Mutagenesis studies ............................................................................................ 107 6.3.2 Detection of the diferric peroxo intermediate in wild type and E44Q PmFTN .. 108 6.3.3 In vivo iron storage property of PmFTN ............................................................. 109 Bibliography ................................................................................................................................110   ix LIST OF TABLES Table 2-1: Primers used in this study ............................................................................................ 40 Table 3-1: Data collection and refinement statistics of wild type PmFTN................................... 50 Table 3-2: Range of iron and zinc ion occupancy observed in binding sites A, B, and C of aerobically Fe(II) and Zn(II) soaked PmFTN crystals .................................................................. 55 Table 4-1: First and second order rate constants of Fe(II) oxidation in PmFTN variants ............ 67 Table 4-2:  Data collection and refinement statistics of PmFTN variants .................................... 79 Table 4-3: Range of iron ion occupancy observed in binding sites A, B, and C of aerobically Fe(II) soaked PmFTN variant crystals .......................................................................................... 81 Table 5-1: Data collection and refinement statistics of EcBFR .................................................... 94 Table 5-2: Average bond lengths in Å between iron ions and ligands of anaerobic and aerobic EcBFR structures .......................................................................................................................... 97   x LIST OF FIGURES Figure 1-1: Crystal structure of EcFtnA (PDB ID 1EUM) ............................................................. 4 Figure 1-2: Amino acid sequence alignment of ferritins and bacterioferritins ............................... 5 Figure 1-3: Ferroxidase centres of ferritin ...................................................................................... 7 Figure 1-4: Ferritin phylogenetic tree ........................................................................................... 13 Figure 1-5: General mechanisms of iron mineralization involving the ferroxidase centre .......... 17 Figure 1-6: Possible mechanisms of iron mineralization by ferritin and bacterioferritin ............. 20 Figure 1-7: A conserved tyrosine acts as a one electron capacitor ............................................... 27 Figure 1-8: Electrostatic surface potential of HuHF and EcBFR ................................................. 31 Figure 1-9: A light micrograph of the marine pennate diatom Pseudo-nitzschia multiseries. ..... 36 Figure 3-1: Crystal structure of PmFTN ....................................................................................... 49 Figure 3-2: Ferroxidase centres of PmFTN .................................................................................. 52 Figure 3-3: Residue Glu130 exhibits flexible coordination to Fe-B and Fe-C ............................. 54 Figure 3-4 (previous page): Kinetic analysis of Fe(II) oxidation catalyzed by PmFTN .............. 59 Figure 4-1: Kinetic analysis of Fe(II) oxidation catalyzed by E44Q PmFTN (A, B), E44H PmFTN (C, D), and E130A PmFTN (E, F) .................................................................................. 68 Figure 4-2: Plots of total amplitude of absorbance changes at 340 nm at 0.5 s as a function of Fe(II) added per PmFTN variant protein ...................................................................................... 70 Figure 4-3 (previous page): Stopped-flow measurements of the second kinetic phase ................ 74 Figure 4-4: Stopped-flow measurements of regeneration of the rapid oxidation phase ............... 76 Figure 4-5: Stopped-flow spectroscopy of iron mineralization in wild type and variant  PmFTN .......................................................................................................................................... 77 Figure 4-6: Ferroxidase centres of PmFTN variants..................................................................... 82 Figure 4-7: B-channel of the E44Q Fe (o.n) structure .................................................................. 83 Figure 4-8: Electrostatic surface potential of E44Q Fe (o.n) ........................................................ 90 Figure 5-1: Crystal structure of EcBFR derived from a crystal soaked for 2 h in ferrous iron under anaerobic conditions ........................................................................................................... 95 Figure 5-2: Ferroxidase centre of EcBFR crystal structure derived from a crystal soaked for 2 h in ferrous iron under anaerobic conditions ................................................................................... 96 Figure 5-3: Superposition of anaerobic and aerobic EcBFR structures ........................................ 97  xi Figure 6-1: Ferroxidase centre and site C of PmFTN ................................................................. 102 Figure 6-2: Stepwise iron and dioxygen binding to the ferroxidase centre of PmFTN .............. 103 Figure 6-3: Detection of the diferric peroxo intermediate .......................................................... 109  xii LIST OF ABBREVIATIONS Å Ångström (1 Å = 0.1 nm) AfFtn Archaeoglobus fulgidus ferritin ASU Asymmetric unit B-factor Crystallographic displacement parameter BfMF Bullfrog M-chain ferritin BFR Bacterioferritin DNase Deoxyribonuclease DdBFR Desulfovibrio desulfuricans bacterioferritin Dps DNA-binding protein from starved cells E. coli  Escherichia coli EDTA  Ethylenediamine tetra-acetic acid EcBFR Escherichia coli bacterioferritin EcFtnA  Escherichia coli ferritin A ESU Estimated standard uncertainty EP Extension peptide Fe (2.5), Fe (5), Fe (45), Fe (65),  Fe (75), Fe (2 h), Fe (3 h), Fe (4 h),  and Fe (o.n)  Structure derived from a crystal soaked in Fe(II) for 2.5 min, 5 min, 45 min, 65 min, 75 min, 2 h, 3 h, 4 h, and overnight, respectively Fe-A Iron bound to site A Fe-B Iron bound to site B  xiii Fe-C Iron bound to site C Ftn Ferritin Fo, Fc Observed and calculated structure factors H Heavy subunit HNLC High nutrient low chlorophyll  HuHF Human H-chain ferritin L Light subunit M Middle subunit MES  2-(N-morpholino)ethanesulfonic acid N/A Not applicable PCR Polymerase chain reaction PDB  Protein data bank PfFtn Pyrococcus furiosus ferritin PmFTN  Pseudo-nitzschia multiseris ferritin R.m.s.d  Root mean square deviation SSRL  Stanford Synchrotron Radiation Lightsource TCEP  Tris(2-carboxyethyl)phosphine TP Transit peptide Tris  Tris(hydroxymethyl)aminomethane Zn (1 h)   Structure derived from a crystal soaked in Zn(II) for 1 h  Zn-A  Zinc bound to site A Zn-B  Zinc bound to site B Zn-C  Zinc bound to site C  xiv ACKNOWLEDGEMENTS Thank you to everyone who in some way contributed to the success of this thesis and who made my time in the Department of Microbiology and Immunology very enjoyable. Especially, I would like to thank:  Dr. Michael Murphy for his supervision of my thesis, his encouragement and support, and for giving me the opportunity to do my PhD in his lab.  My committee members, Dr. Maria Maldonado, Dr. Ross MacGillivray, and Dr. Tom Beatty, for their guidance and advice.  Dr. Nick Le Brun and his lab for the fruitful collaboration and my instructive and interesting stay in Norwich.  Angelé Arrieta, technician extraordinaire, for introducing me to the world of ferritin and teaching me the work in the glove box. Marlo Firme for his help with the mutagenesis.  Current and past lab members, Dr. Anson Chan, Marek Kobylarz, Angelé Arrieta, Amelia Hardjasa, Dr. Slade Loutet, Meghan Verstraete, Catherine Bowden, Mariko Ikehata, Dr. Shinichi Takayama, and Dr. Jason Grigg, for the interesting discussions, the teamwork, the many fun sushi lunches, and the memorable lab outings.  My mum and Karl for the financial support during the last years, and Karl for thoroughly proofreading my thesis.    1 CHAPTER 1: INTRODUCTION  1.1 IRON IN BIOLOGY Iron is the most abundant transition metal in the Earth’s crust and is required by all eukaryotes and most prokaryotes. It can exhibit oxidation states ranging from –II to +VI, although the reduced Fe(II) ferrous form and the oxidized Fe(III) ferric form are the most common in biology. The properties of iron, such as its wide range of reduction potentials (-500 to +300 mV, depending on the iron ligands and environment), electronic configuration, and interconvertible redox states, makes it a versatile prosthetic factor for incorporation into proteins as a biocatalyst or electron carrier. Iron is involved in major biological processes such as cell respiration, photosynthesis, nitrogen fixation, and DNA biosynthesis (1,2). It can be found in proteins as mononuclear or binuclear iron, or in a more complex form such as iron-sulfur clusters or heme groups (3,4).  In early life, when the Earth’s environment was low in dioxygen, Fe(II) was readily available and therefore was incorporated into many biological systems. However, the introduction of dioxygen into the atmosphere by photosynthesis significantly altered the chemistry of iron (5). The predominant form of iron switched from the soluble ferrous state, which has a solubility of ~0.1 M at pH 7.0, to the insoluble ferric state, which has a solubility of ~10-18 M at pH 7.0 (1). This decrease in solubility also decreased bioavailability at physiological pH, and iron became a growth limiting nutrient for many organisms (6,7). Furthermore, the presence of dioxygen increased iron toxicity, and iron became potentially lethal when in excess of that needed for cellular homeostasis. The reduction of dioxygen by Fe(II) leads to the formation of a superoxide radical, which is converted to hydrogen peroxide and water by  2 superoxide dismutase. However, ferrous iron can also react with hydrogen peroxide to generate the reactive hydroxyl radical. This reaction is called the Fenton reaction (Equation 1). Hydroxyl radicals are highly toxic in that they can induce lipid peroxidation, DNA strand breaks, and degrade various other biomolecules (8).   Fe(II) + H2O2  Fe(III) + OH- + ˙OH   (1)      Fe(III) can be reduced back to Fe(II) by cytoplasmic reductants, such as superoxide radicals, resulting in a destructive catalytic cycle (Haber-Weiss reaction) (9,10). Consequently, most organisms have evolved a strategy that includes the use of ferritin-like proteins to protect cells from the harmful effects of excess iron by storing it in a non-reactive form.  1.2 THE FERRITIN FAMILY The ferritin family belongs to the ferritin-like superfamily of proteins and consists of three subfamilies, the classical ferritins (Ftn or ferritin), the bacterioferritins (BFR), and the DNA-binding proteins from starved cells (Dps). Ferritins are the archetypical members of the ferritin family found in all three domains of life. BFRs are similar to ferritins but are restricted to prokaryotes (11). Dps are found in prokaryotes (12); however, they show some major structural and functional differences that are explained in more detail in section 1.2.3. (Bacterio)ferritins are iron storage and detoxifying proteins and typically consist of 24 subunits that form a highly symmetrical, hollow sphere (Figure 1-1A). (Bacterio)ferritins take up soluble ferrous iron, oxidize it at di-iron ferroxidase centres for storage as a ferric mineral within the central cavity. Iron is subsequently released upon demand of the organism’s metabolism (13,14).  3 The ferroxidase centre of ferritin and BFR is located near the centre of each monomer and consists of two iron binding sites, site A and site B. Outside the iron coordinating residues of the ferroxidase centre, only a few residues are absolutely conserved between the two ferritin subfamilies (Figure 1-2), and amino acid sequence identities can be as low as 15%. However, the overall structure is conserved. The monomer is a four helix bundle (helices A, B, C, and D) with a short fifth helix E at the C-terminus. One loop between helices B and C crosses over the length of the bundle (Figure 1-1B). Extensive helix-helix interactions occur between the subunits and loops, resulting in an unusually stable protein complex with octahedral 4/3/2 symmetry. The monomers in the shell are arranged in dimers forming each face of a dodecahedron and containing a 2-fold symmetry axis (15). The outer diameter of the protein shell is ~120 Å, and the inner diameter is ~80 Å. This creates a cavity with a theoretical iron storage capacity of ~4500 iron ions. All three subfamilies have been studied in microorganisms. Ferritin and bacterioferritin both play a role in iron storage but may also have more specialized functions in iron detoxification or iron metabolism (13). However, the precise role varies depending on the particular organism. In Escherichia coli (E. coli), for example, the ferritin FtnA (EcFtnA) functions as the principal iron storage protein (16), whereas a proposed function of E. coli bacterioferritin (EcBFR) in detoxification still remains to be proven (17). BFR is the major iron storage protein in Salmonella enterica sv. typhimurium (18), and in Neiserria gonorrhoeae, BFR serves as an iron source and protects the cell against H2O2. However, the physiological roles of (bacterio)ferritins in microorganisms, in particular the interplay between the different subfamilies where they co-exist in the same cell, are still not well understood (19).    4  Figure 1-1: Crystal structure of EcFtnA (PDB ID 1EUM) (A) The 24-meric, conserved quaternary structure. (B) The 4 helix bundle (helices A, B, C, and D) with a short fifth helix E at the C-terminus. One loop between helices B and C crosses over the length of the bundle. NT, N-terminus; CT, C-terminus.    5  Figure 1-2: Amino acid sequence alignment of ferritins and bacterioferritins E. coli ferritin A (EcFtnA), Pyrococcus furiosus ferritin (PfFtn), Pseudo-nitzschia multiseries ferritin (PmFTN), Human H-chain ferritin (HuHF), Bullfrog M-chain ferritin (BfMF), soybean seed ferritin (FRI1_SOYBN), E. coli bacterioferritin (EcBFR), Desulfovibrio desulfuricans bacterioferritin (DdBFR). Amino acid residues building the ferroxidase centre or binding site C are highlighted in red. Additional highly conserved residues are highlighted in black and gray. The alignment was performed using Clustal Omega (20), and the figure was generated using GeneDoc (21).    6 1.2.1 FERRITIN 1.2.1.1 VERTEBRATE FERRITIN Mammalian ferritins are heteropolymers of two homologous monomers, the heavy (H) chain and the light (L) chain (22). The two subunits in humans have ~55% amino acid sequence identity (23). A difference between the two subunit types is that the H-chain ferritin contains the active site, the ferroxidase centre, which is not present in the L-chain subunit. However, the L-chain subunit has more negatively charged residues on the inner surface facing the cavity of the protein shell (24). These residues provide putative nucleation sites for iron core formation. The two subunits are isostructural and can assemble in any proportion in the 24-mer. Although L-chain rich ferritins oxidize Fe(II) more slowly than H-chain ferritins, the presence of some L-chain subunits in H-chain rich ferritin promotes iron nucleation (10,14,25-28). The H:L ratio is tissue specific. For example, ferritin with up to 70% H-chain subunits can be found in tissues that exhibit high ferroxidase activity, such as the heart or the brain. On the other hand, ferritin with up to 90% L-chain subunits can be found in tissues having mainly an iron storage function, such as the liver or spleen (10).  Furthermore, some fish and amphibians produce a homopolymer ferritin constituted of a third subunit type, the middle (M) chain, that harbors the residues forming the ferroxidase centre (29-31).  The ferroxidase centre of human H-chain ferritin (HuHF) is illustrated in Figure 1-3A. There are two iron binding sites, site A and site B, which are not equivalent. Iron in site A (Fe-A) is coordinated by a histidine (His65), a monodentate glutamate (Glu27), and a bridging glutamate (Glu62). On the other hand, iron in site B (Fe-B) is coordinated by Glu62 and two monodentate glutamates (Glu61 and Glu107) (14,32).  7  Figure 1-3: Ferroxidase centres of ferritin (A) Ferroxidase centre of human H-chain ferritin. (B) Ferroxidase centre of E. coli FtnA, including third iron binding site C. (C) Ferroxidase centre of Pseudo-nitzschia multiseries ferritin. No metal ion was found in site A; instead, a water molecule was modeled (33). Metal bonds are drawn as solid lines. Hydrogen bonds are drawn as dotted lines. (D) Ferroxidase centre of E.coli bacterioferritin.   Most known ferritins are cytosolic proteins. However, a ferritin located inside the matrix of human, mouse, and rat mitochondria was identified (34,35). Mitochondrial ferritin is encoded by a chromosomal gene and is targeted to mitochondria, where it is processed to the mature 22 kDa protein that assembles into a homopolymeric ferritin shell (36). The mature human mitochondrial ferritin amino acid sequence shares 79% identity with HuHF, and the residues of  8 the ferroxidase centre are all conserved. The biological role of mitochondrial ferritin is not yet clear. However, it has been proposed that the main biological function of human mitochondrial ferritin is to sequester excess iron as cytosolic ferritins do (37). Since the mitochondria are the sites where heme and Fe/S complexes are synthesized and therefore are exposed to heavy trafficking of iron, iron storage is of particular interest (38). Furthermore, mitochondria are also the major producer of reactive oxygen species, and they require an efficient mechanism to overcome the toxicity of iron. In fact, a protective role of mitochondrial ferritin against iron-induced oxidative damage has been proposed (35,37).   1.2.1.2 PHYTOFERRITIN Ferritins found in plants are called phytoferritins. Their sequences share between 39% and 49% amino acid identity with mammalian ferritin sequences. The amino acid residues of the ferroxidase centre in HuHF are strictly conserved in phytoferritins (Figure 1-2) with the exception of pea seed ferritin, where a histidine is found at position 61 instead of a glutamate. Phytoferritins are localized to plastids: chloroplasts in leaves, amyloplasts in tubers and seeds (39,40). The subunits from phytoferritins are synthesized as 32 kDa precursor proteins that contain a unique N-terminal sequence consisting of a plastid transit peptide (TP) and an extension peptide (EP). The TP is required for the targeting of the precursor to the plastid. The TP is cleaved by an unknown mechanism after transportation of the phytoferritin precursor into the plastid. The mature subunits assemble into a 24-mer ferritin within the plastid. The EP is ~30 amino acids long and partly folds into a helix. The EP is not cleaved, instead, it remains located on the exterior surface of the protein, interacting with neighboring subunits (41). Recent studies of pea seed ferritin suggest that the EP has a role as a second centre or iron binding site that  9 contributes to the iron mineralization at higher iron loading (42). Furthermore, a serine protease like activity has been suggested for the EP that is responsible for auto-degradation during seed germination and therefore for faster iron release to meet the iron demand for rapid seedling growth (39,43). Amino acid sequencing of assembled soybean seed ferritin led to the finding of two subunits, termed H-1 (26.5 kDa) and H-2 (28 kDa) (44,45). Although all characterized subunits of phytoferritins contain the ferroxidase centre, H-1 and H-2 have distinct functions during iron mineralization (46). Their ferroxidase centres exhibit different regeneration and catalytic activities. Under moderate iron flux (up to 200 Fe(II) per ferritin shell), iron oxidation at the di-iron ferroxidase centre is dominated by the H-1 subunit. In H-2, this process is gradually replaced by another mechanism by which the iron oxidation is processed by the EP. At high iron loading, the EP of H-1 also participates in iron oxidation, although its ability to transfer Fe(III) to the cavity is weaker. However, H-1 and H-2 are synergistic for the oxidative deposition of iron in soybean seed ferritin, resulting in greater catalytic activity of heteropolymeric ferritin compared to the homopolymeric ferritin (39,47). In Arabidosis thaliana and cowpeas, four ferritin genes were identified with one being a 28 kDa subunit and three a 26.5 kDa subunit (40,48). Phytoferritins are proposed to be composed of a complex mixture of subunits analogous to mammalian ferritins in different cell types (14,49).  1.2.1.3 BACTERIAL FERRITIN Ferritins have been characterized in several bacteria, including E. coli (50), Helicobacter pylori (51), Campylobacter jejuni (52), and Porphyromonas gingivali (53). Bacterial ferritins are homopolymers, and each subunit contains a ferroxidase centre. A second ferritin, FtnB, that does  10 not contain a typical ferroxidase centre was identified in E. coli. However, heteropolymeric ferritins have not been observed, and the role of FtnB is not yet clear (13).  The best studied bacterial ferritin is EcFtnA (50,54). Structurally, EcFtnA is similar to HuHF, despite a sequence identity of only 22%. The root mean square deviation (r.m.s.d.) for all main chain atoms is only 0.66 Å. The ferroxidase centre of EcFtnA is conserved with that of H-chain ferritin, with the exception that the Fe-B ligand Glu61 is replaced by Glu130 (Figure 1-3B). These two glutamates are derived from different parts of the polypeptide chain of HuHF and EcFtnA. Also, EcFtnA has a third iron binding site, site C. This site lays 7 – 8 Å away from the ferroxidase centre and is located near the inner surface of the protein cavity (50,55). The iron at site C (Fe-C) is coordinated by Glu130, which alternates as a site B or C ligand. Glu126, Glu129, and Glu49 function as additional monodentate ligands (50,56). The side chains of Glu126 and Glu129 are pointing into the central cavity in a metal free crystal structure (50). However, the ferroxidase centre (site A and site B) residues were found to be in a conformation similar to that found in the metal-bound form. The four site C residues are conserved in all known bacterial ferritins (13).    1.2.1.4 ARCHEAL FERRITIN Ferritin homologs were identified in the hyperthermophilic archea Pyrococcus furiosus and Archaeoglobus fulgidus (AfFtn), and they have been structurally characterized (57,58). The ferroxidase centre and site C of these archeal ferritins are identical to those of EcFtnA. However, Archaeoglobus fulgidus ferritin is different in that it has a novel quaternary structure. Despite the typical ferritin subunit structure, subunits assemble into a shell having tetrahedral 2/3 symmetry instead of an octahedral 4/3/2 symmetry as seen in other ferritins. Although the protein shell is somewhat spherical, it is less densely packed, is slightly larger in overall dimensions, and has  11 four large (45 Å) pores (57). The difference in symmetry may be due to amino acid residue differences in the E-helix that prevent subunits to associate with a four-fold symmetry (13). In the absence of an iron core, AfFtn dissociates into dimers, a feature which has not been seen in other ferritins. However, this novel quaternary structure is not a common feature amongst archael ferritins. P. furiosus ferritin (PfFtn) assembles into a 24-mer with the usual octahedral 4/3/2 symmetry (58).  1.2.1.5 DIATOM FERRITIN Relatively little is known about the ferritins of organisms in the marine environment. Marchetti et al. provided evidence that the bloom forming pennate diatoms Pseudo-nitzschia and Fragilariopsis use ferritin to store iron (33). A ferritin homolog was identified in 5 species of pennate diatoms and has not been detected in any other member of the Stramenopiles, a diverse eukaryotic lineage that includes unicellular algae, macroalgae (such as kelp), and plant parasites. A phylogenetic analysis of diatom ferritin sequences with those from prokaryotes and non-marine photosynthetic eukaryotes, suggests that diatom ferritins are clearly distinct from other eukaryotic ferritins but weakly associated with prokaryotic ferritins (Figure 1-4) (33). The crystal structure of recombinant iron-soaked Pseudo-nitzschia multiseries ferritin (PmFTN) was resolved to 1.95 Å. This is the first ferritin structure from any member of the highly divergent Stramenopiles. Like other characterized ferritins, PmFTN is composed of 24 subunits. The structure confirmed the characteristic ferritin ferroxidase centres, monomeric fold, and spherical assembly (33). Amino acid residues that form the ferroxidase centre in other characterized ferritins are highly conserved in PmFTN (59). Furthermore, the sequence of PmFTN shows a transit peptide sequence, and the protein is predicted to be targeted to the  12 chloroplast to control the distribution and storage of iron for proper functioning of the photosynthetic machinery. An iron uptake assay confirmed that PmFTN has ferroxidase activity and is able to store iron in vitro (33).  The crystal structure of PmFTN revealed three iron binding sites; sites A and B form the ferroxidase centre, and a third site C is located close to the ferroxidase centre (33). Three iron ions are observed in and around the ferroxidase centre. One is found in ferroxidase site B, one in site C, and the third is positioned passed site C toward the mineral core. An unexpected finding was that the ferroxidase site A is occupied by a water molecule. Fe-B is coordinated by three conserved glutamate residues (Glu48, Glu94, Glu130) and an unusual fourth ligand (Glu44). Site C is unique in that it is only coordinated by one glutamate, Glu44 (Figure 1-3C). Water molecules complete the coordination sphere. In comparison with other ferritins, Glu44 is unique as a coordinating residue for Fe-C. In EcFtnA a histidine is observed at that position and in PfFtn a glutamine, and they are not directly coordinating Fe-C.   13  Figure 1-4: Ferritin phylogenetic tree Bootstrap consensus tree showing the evolutionary relatedness of ferritins from 26 taxa inferred using a protein distance model. Bootstrap values greater than 50 are indicated at the branch points. Diatoms are in bold. Adapted with permission from (33).   1.2.2 BACTERIOFERRITIN BFRs have been identified in a wide variety of bacteria, including E. coli, Azotobacter vinlandii, and Desulfovibrio desulfuricans (60-63). Typically, BFRs are homopolymers and each subunit contains a catalytic di-iron ferroxidase centre. In some bacteria, such as Pseudomonas aeruginosa, Pseudomonas putida, and the cyanobacterium Synechocystis, genes for more than one BFR were identified; however, heteropolymer assembly has not been detected yet (13,64-66).  14 BFR sequences share about 20% amino acid identity with eukaryotic H-chain ferritins and only 10 – 15% identity with bacterial ferritins. Conserved within both BFRs and ferritins are some of the key residues that build the catalytic ferroxidase centre (Figure 1-2). Like H-chain ferritin, the ferroxidase centre of BFR is a dinuclear site (Figure 1-3D). However, in BFR, the ferroxidase centre is highly symmetrical. Fe-A and Fe-B are each coordinated by a histidine, a glutamate as a capping ligand, and two bridging glutamate residues. The residues involved in iron binding do not change their conformation significantly upon binding of ferrous iron (67). BFRs are unique amongst the ferritin family in that they contain heme groups, which are located at the interfaces between the subunits related by 2-fold symmetry. The heme iron ions are ligated to the protein by two methionine residues one from each adjacent subunit (68). Heme in EcBFR does not play a role in the ferrous iron uptake and oxidation since a heme free variant did not affect the rate of iron mineralization (69). Instead, the heme cofactor plays a role in iron release (70). D. desulfuricans BFR (DdBFR) is unusual in that it contains iron-coproporphyrin III rather than iron-protoporphyrin IX (heme) and is the first example of this porphyrin as a protein cofactor (62,71).  1.2.3 DPS   Dps are the third ferritin subfamily found in prokaryotes. The Dps subunit differs from ferritin and BFR in that the fifth helix E is missing and an extra helix is inserted within the loop connecting the B and C helices in the four helix bundle, which leads to assembly into a 12-mer with 2/3 symmetry (15). Consequently, the central cavity has a diameter of ~40 Å in which an iron core of 400 – 500 iron ions can be laid down, less than in other ferritin subfamilies.  15 The ferroxidase di-iron centre is different in that it is located at the interface of two symmetry related subunits, where residues derived from both subunits ligate the iron ions. This symmetry leads to a total of 12 ferroxidase centres per Dps, two ferroxidase centres at each subunit dimer interface. In addition to the ferroxidase centre of Dps, a third iron-binding site at a similar distance as site C of EcFtnA was found (15,72,73). However, the function of this third iron binding site in Dps remains to be investigated. Dps more efficiently use hydrogen peroxide as an oxidant but may also use dioxygen with much lower turnover rate.  Dps are multifunctional proteins. One of their roles is to maintain the mechanical stability of DNA during stationary phase through the formation of protein-DNA complexes and DNA condensation (74-76). Dps binds non-specifically to DNA, and an extended N-terminus that is positively charged is critical for DNA-Dps interaction (77,78). In the condensed Dps-DNA state, the chromosome or plasmid DNA is DNase resistant as observed in eukaryotes for histone-protected DNA complexes. A model suggests that DNA is wound around the Dps proteins as seen with histones (79); however, such a three dimensional arrangement remains speculative (80). Dps also protect bacteria from radical damage through consumption of hydrogen peroxide for the rapid removal of ferrous iron for storage as insoluble ferric mineral. These functions deprive the cell of substrates for the Fenton reaction, preventing the formation of toxic reactive oxygen species. (80,81). Recently, an additional enzymatic function has been observed for the Dps from Microbacterium arborescens, which is found in the insect gut of the herbivore Spodaptora exigua. This Dps was shown to be an amino acid hydrolase (82) that catalyzes the hydrolysis and less efficiently the synthesis of N-acyl amino acids. The substrates of amino acid hydrolases are N-acyl-glutamines that play an important role in the ability of plants to recognize  16 herbivory and elicit plant defence reactions after introduction into the leaf during feeding (80,83,84).   1.3 MECHANISM OF IRON STORAGE The iron mineralization mechanism in (bacterio)ferritin has been a major research focus over the last 40 years. However, the detailed mechanism for any member of the ferritin family is still not well defined.  First observations of core-surface catalyzed mineralization led to the crystal growth model for core formation (85-87). In this model, Fe(II) enters the empty cavity, is oxidized, and forms small polynuclear clusters, which then act as nucleation centres for mineral growth. At some stage, one of these clusters becomes the dominant nucleation centre and further growth of the mineral predominantly occurs from this cluster. Once the growing mineral comes in contact with other smaller clusters and the inner surface of the cavity, crystal growth is prevented and smaller clusters grow until the cavity is filled. This mechanism is based on observations that the rate of Fe(II) oxidation increases with increasing ferritin iron content until ~50% of the cavity is filled; thereafter, the rate of Fe(II) oxidation decreases as the active core surface area decreases (14,85-87).  However, the crystal growth model does not account for the involvement of the ferroxidase centres in iron mineralization. Crystal structures of several ferritins and BFRs are solved identifying the structure of the iron binding sites and in part leading to models for ferroxidase centre operation. Two general models are well accepted for the mechanism of iron mineralization involving iron oxidation at the ferroxidase centre (Figure 1-5). In one model, the ferroxidase centre functions as a substrate site. Ferrous iron binds to the ferroxidase centre, is  17 oxidized, and subsequently migrates to the mineral core, emptying the ferroxidase centre, which can then accept two more ferrous iron ions (Figure 1-5A). This model is proposed for HuHF (88), EcFtnA (89), and bullfrog M-chain ferritin (BfMF) (90). In a second model, the ferroxidase centre acts as a cofactor, and the bound iron does not migrate to the core. Thus, after binding of two ferrous iron at the ferroxidase centre, the iron ions cycle between the Fe(II) and Fe(III) forms, first oxidized by oxygen and subsequently reduced by electrons transferred from Fe(II) ions that bind within the central cavity (Figure 1-5B). This model is proposed for iron mineralization by EcBFR (67) and PfFtn (91). Ferrous iron oxidation and core formation of the different subfamilies of ferritin are described in more detail below.   Figure 1-5: General mechanisms of iron mineralization involving the ferroxidase centre (A) The ferroxidase centre functions as a substrate site where ferrous iron is oxidized and then moves to the mineral core. (B) The ferroxidase centre functions as a cofactor. Iron ions bound at the ferroxidase centre cycle between the Fe(II) and Fe(III) forms, first oxidized by oxygen and subsequently reduced by electrons transferred from Fe(II) ions that bind within the central cavity.     18 1.3.1 IRON MINERALIZATION IN H-CHAIN FERRITIN The enzymatic activity in mammalian ferritins is associated with the H-chain subunits, where the ferroxidase centres are located. At iron loading below or equal to 2 Fe(II) per subunit, the dominant catalytic reaction takes place at the ferroxidase centre (Equation 2, with z as the net charge of the protein and P as the protein; Figure 1-6A). The two ferrous iron ions bound to the ferroxidase centre are oxidized by dioxygen to form an unstable µ-1,2-peroxodiferric intermediate, which decays to a more stable µ-1,2-oxodiferric species with the release of hydrogen peroxide (Fe(II):O2 stoichiometry of 2:1; Figure 1-6A Phase 2). The formation and decay of the µ-1,2-peroxodiferric intermediate can be detected by its characteristic blue color due to absorbance at 650 nm (92). It is proposed that the µ-1,2-oxodiferric species is already hydrated with the form [Fe2O(OH)2]2+ and is not stable within the ferroxidase centre. A second hydrolysis leads then to the passage of Fe(III) to the mineral core (Equation 3; Figure 1-6A Phase 3) (93,94). Since the apo form of the ferroxidase centre is regenerated, the ferroxidase centre acts as a gated iron pore into the cavity, by which the iron transport is coupled to Fe(II) oxidation.   [Fe(II)2(FC)-P]z+4 + O2 +3H2O  [Fe(III)2O(OH)2(FC)-P]z+2 + H2O2 + 2H+ (2)  [Fe(III)2O(OH)2(FC)-P]z+2 + H2O  2[Fe(III)OOH(core)-P]z + 2H+ (3)  At high iron loading (>800 Fe(II) per 24-mer), the dominant site of catalysis switches from the ferroxidase centre to the growing core surface (93,95,96). The Fe(II):O2 stoichiometry increases to 4:1, implying a complete reduction of dioxygen to water (96) (Equation 4; Figure 1-6B).   4[Fe(II)(cavity)-P]z+2 + O2 + 6H2O  4[Fe(III)OOH(core)-P]z + 8H+ (4)  19 When intermediate amounts of ferrous iron are added (100-500 Fe(II) per24-mer), some of the hydrogen peroxide produced in equation 2 is consumed in a Fe(II) oxidation reaction (Equation 5) (95).    2[Fe(II)(cavity)-P]z+2 + H2O2 +2H2O  2[Fe(III)OOH(core)-P]z + 4H+ (5)  Thus, Fe(II) oxidation in H-chain ferritin is a combination of the three described pathways (Equations 2-5). However, the overall Fe(II):O2 stoichiometry is 4:1, and the products of equations 1-3 and 5 are equivalent to equation 4.  A cluster of negatively charged residues on the cavity surface (Glu107, Glu57, Glu60, Glu61, Glu64, and Glu67, HuHF numbering) is thought to be responsible for the slow incorporation of iron in L-chain ferritin (24,26,31). These residues, constituting the putative nucleation site, are highly conserved in vertebrate H and M-chain ferritins and in phytoferritins and are partly involved in forming the ferroxidase centre (97). The cluster of conserved glutamate residues (Glu61, Glu64, and Glu67) in the vicinity of the ferroxidase centre and on the inner surface of HuHF was proposed to be important for mineral core nucleation (32). Glu61 was proposed to occupy two positions, one toward site B of the ferroxidase centre and one toward the nucleation site at the core surface (32). However, Bou-Abdallah et al. have shown that this putative nucleation site in HuHF is not important for iron uptake or mineralization (98). Thus, the function of the nucleation site in L-chain ferritin has yet to be proven.    20  Figure 1-6: Possible mechanisms of iron mineralization by ferritin and bacterioferritin (A) Iron mineralization mechanism at low iron loading in H-chain ferritin. (B) Mechanism at high iron loading in H-chain ferritin. Essentially equivalent to the crystal growth model, where the ferroxidase centre plays no part in iron mineralization. (C) Iron mineralization in EcBFR, where the ferroxidase centre functions as a cofactor. Adapted with permission from (14).     21 1.3.2 IRON MINERALIZATION IN BACTERIAL FERRITIN The mechanism of iron oxidation in bacterial ferritin is best understood in EcFtnA. The ferroxidase centre of EcFtnA is similar to H-chain ferritin; however, EcFtnA contains a third iron binding site C in the vicinity of the ferroxidase centre (Figure 1-3). Core mineralisation is proposed to be similar to the mechanism proposed for H-chain ferritin, where the ferroxidase centre functions as a substrate site. As in HuHF, a Fe(II) binding and oxidation stoichiometry of 48 Fe(II)/24-mer has been observed at low iron loading. The initial ferrous iron oxidation in EcFtnA produces a coloured intermediate with absorption maxima at 370 nm and 600 nm, likely a diferric peroxo species as seen in H-chain ferritin. This intermediate decays into a µ-oxo-bridged Fe(III) dimer that is proposed to convert spontaneously to a µ-hydroxo-bridged dimer (99,100). However, the presence of a third iron binding site C makes the iron mineralization mechanism more complex. The Fe(II)/O2 stoichiometry at iron loading of 2 Fe(II) per ferroxidase centre for EcFtnA has been reported to be 3 – 4 (101) and ~3 Fe(II) per O2 (55), as compared to 2 Fe(II)/O2 in H-chain ferritin. The increase in this ratio probably results from the binding of iron to site C. At high iron loading, the Fe(II)/O2 stoichiometry rises to 4 Fe(II), and iron oxidation directly on the mineral surface occurs (55). However, site C is not essential for ferroxidase activity. Mutagenesis of site C residues results in only a slight reduction in the initial ferroxidation rate and a change in the initial oxidation stoichiometry to 2 Fe(II) per O2 (55,101). The binding and oxidation of 48 Fe(II)/24-mer with a stoichiometry of ~3 Fe(II)/O2 can be explained by the presence of multiple pathways for iron oxidation in the protein, involving only the partial reduction of O2 to H2O. Mutation of site C ligands of EcFtnA leads to an increase in intensity of the blue differic peroxo intermediate. This suggests that in the site C variants more of the bound iron participates in peroxo complex formation at the sites A and B. Thus, in wild type  22 EcFtnA, some of the initial ferrous iron binds to the site C. Furthermore, Mössbauer data suggested that Fe(III) in site C is only formed simultaneously with the oxidation of iron in sites A and B (99). Thus, the oxidation stoichiometry of 48 Fe(II)/24-mer reflects the existence of some EcFtnA molecules with all three sites (A, B, and C) occupied by iron, whereas others are metal-free or have only A and B sites occupied. Furthermore, Fe(II) at the ferroxidase centre can also be oxidized by H2O2 in a pairwise fashion (55). The regeneration of the apo-ferroxidase centre in EcFtnA is much slower than observed in H-chain ferritin. Addition of iron beyond that required to saturate the ferroxidase centres are oxidized at a much slower rate compared to the initial ferrous iron oxidation. Site C is responsible for this slower rate as the loss of site C results in a ferroxidase centre regeneration rate similar to that seen in H-chain ferritins (55,101). Despite a lower overall rate of Fe(II) oxidation and core formation, site C is proposed to bring EcFtnA the advantage, compared to mammalian ferritins, to avoid the generation of hydrogen peroxide. Furthermore, once oxidized, iron is more readily available to serve the requirements of the cell because it remains at the ligand-accessible ferroxidase centre for a longer period (101).   1.3.3 IRON MINERALIZATION IN BFR EcBFR is the most studied and best understood bacterioferritin, although BFRs have been isolated from a wide range of bacteria. Kinetic studies of ferrous iron uptake and oxidation revealed three distinct phases in EcBFR (Figure 1-6C). The first and fastest phase corresponds to the binding of ferrous iron to the ferroxidase centre and was measured by monitoring a perturbation in heme absorption spectra. The redox state of the heme in BFR does not cycle  23 during iron mineralization, but binding of one or two ferrous iron at the ferroxidase centre leads to a minor structural perturbation in the heme binding pocket (102). The second phase corresponds to the rapid oxidation of Fe(II) to Fe(III) in the presence of dioxygen, leading to the formation of a stable µ-oxo (or hydroxo) bridged dinuclear Fe(III) site and hydrogen peroxide (17,103). This reaction likely proceeds via a peroxo species, but the expected associated absorption spectra has not been detected yet. Thus, it is possible that this intermediate is too short lived or its absorbance is too weak to detect (13). This second phase of the iron oxidation is saturated at 48 Fe(II) per 24-mer, indicating that all 24 ferroxidase centres are involved in the Fe(II) oxidation.  Compared to the ferroxidase centres of ferritins that follow the gated pore model described for H-chain ferritin, the ferroxidase centres of EcBFR are stable in the Fe(III) oxidized form and do not regenerate spontaneously to the apo form. Instead, when in excess of that needed to saturate the ferroxidase centre, Fe(II) enters the central cavity through one or several of the channels through the shell. In the inner cavity, Fe(II) binds to a nucleation site at the inner surface of the protein, is oxidized to Fe(III), and subsequently hydrolyzed to the ferric mineral. This iron oxidation within the ferritin cavity is the third kinetic phase, and it is the slowest phase of all three. The electrons released from ferrous iron oxidation in the central cavity are channeled to the ferroxidase centre, where bound ferric iron is reduced back to ferrous iron. The dinuclear Fe(II) ions are reoxidized to Fe(III) and thus continue to operate as a cofactor in a catalytic cycle throughout core mineralization.  The Fe(II):O2 stoichiometry of EcBFR is 4:1. The oxidation of four Fe(II) per dioxygen together with the stability of the oxidized ferroxidase centre, suggest that two catalytic cycles, likely at two subunits, are involved in reducing one dioxygen (103). At a first ferroxidase centre,  24 two ferrous iron ions are oxidized by dioxygen forming the bridged diferric centre and hydrogen peroxide. Hydrogen peroxide subsequently reacts at another ferroxidase centre, forming another di-Fe(III) oxo-bridge and water. Furthermore, hydrogen peroxide is suggested to be a much better oxidant than dioxygen for EcBFR (17,103). Equations (6) and (7) are the reactions for the oxidation of ferrous iron by O2 and H2O2:    Subunit 1: [Fe(II)2(FC)-P]z + O2 + H2O  [Fe(III)2O(FC)-P]z + H2O2  (6)  Subunit 2: [Fe(II)2(FC)-P]z + H2O2  [Fe(III)2O(FC)-P]z + H2O (7)  Crow et al. identified a binding site for Fe(II) on the inner surface of EcBFR (67). This site is located ~10 Å from the ferroxidase centre and is formed by Asp50 and His46. Variants lacking one or both of these iron binding residues show a decrease in the rate of iron mineralization (phase 3), regardless of the amount of iron EcBFR was previously exposed to. This observation implies that the inner surface site plays a significant role in phase 3 iron mineralization throughout the mineral growth and not only for the nucleation of new mineral clusters. Phase 3 iron mineralization was not completely lost in the variant lacking both iron coordinating residues, indicating the presence of other ferrous iron binding sites or weak nonspecific binding at the inner cavity. Although the inner surface site may be important for core nucleation, a proposed major role is to facilitate electron transfer from Fe(II) in the central cavity to the ferroxidase centre. In the variants, the fastest electron route to the ferroxidase centre is impaired and therefore the iron core mineralization rate decreased (67). The inner surface site of BFR differs from the third iron binding site described for EcFtnA that is 7-8 Å away from the ferroxidase centre and close but not at the inner protein surface.   25 1.3.4 IRON MINERALIZATION IN ARCHEAL FERRITIN The ferroxidase centre of PfFtn is similar to that of EcFtnA. A third iron binding site C, located 7.5 Å and 6.3 Å away from sites A and B, respectively, has also been observed in PfFtn, and its coordinating ligands are identical to Fe-C in EcFtnA (58). However, the iron mineralization mechanism of PfFtn is inconsistent with the substrate site mechanism and is proposed to be similar to the mechanism of EcBFR, including the loss of apo form regeneration (104,105). However, PfFtn is incompatible with the catalytic-core model proposed for EcBFR, and an iron soaked crystal structure does not show fully occupied iron sites in all the subunits. Furthermore, the third iron binding site in PfFtn is not located at the inner cavity as seen in EcBFR. Honarmand Ebrahimi et al. proposed for PfFtn that the dinuclear Fe(III) site in the ferroxidase centre accepts electrons from Fe(II) possibly bound to site C. The oxidized iron at site C is then released into the cavity to form the mineral core (104). However, more recently the same group proposed a revised model for the ferrous iron oxidation mechanism in archeal ferritin and HuHF that is described below in section 1.3.5 (106,107).   1.3.5 SEQUENTIAL DISPLACEMENT OF FE(III) BY FE(II)  A sequential displacement model for the iron mineralization in ferritin combines the proposed mechanisms for mammalian ferritins and prokaryotic ferritins and is supported by studies with HuHF and PfFtn (106,107). This displacement model suggests that after oxidation of two Fe(II) at the ferroxidase centre, the Fe(III) loaded centre remains stable until the arrival of additional Fe(II) triggers sequential displacement of Fe(III) from the ferroxidase centre. The displaced Fe(III) ions presumably move to form the iron core. However, incoming Fe(II) is suggested to be distributed among three binding sites; sites A and B from the ferroxidase centre  26 and a third binding site termed the gateway site, which is suggested to be a transient site with lower affinity for Fe(II) than the ferroxidase centre binding sites. Isothermal titration calorimetry (ITC) experiments demonstrated three ferrous iron binding events for PfFtn consistent with the three iron binding sites observed in crystal structures (58). Three ferrous iron binding events were also shown for HuHF. A transit site involving Glu140 in HuHF has previously been proposed; however, a third iron binding site in HuHF remains to be confirmed crystallographically (41,106).  A blue color intermediate was observed during catalysis by PfFtn and HuHF (92,94,106). The peroxodiferric intermediate is proposed to decay spontaneously to form the metastable oxodiferric intermediate if no Fe(II) is bound to the third site (Figure 1-7, Path 1). However, if the third iron binding site is occupied by Fe(II), this Fe(II) can be oxidized by the peroxodiferric intermediate or by hydrogen peroxide that is formed in the ferroxidase centre (Figure 1-7, Path 2). The gateway site in PfFtn has a higher affinity than the third putative iron binding site in HuHF, and PfFtn might oxidize more Fe(II) by path 2 compared to HuHF (106,107). A fourth electron from a highly conserved tyrosine within 5 Å of site B is suggested to complete the reduction of dioxygen to water, thereby preventing the formation of reactive oxygen species (107). However, the relevance of this unifying model still has to be established. Recent studies on EcFtnA do not support a universal mechanism for iron oxidation, and further studies are necessary to better understand the iron mineralization mechanisms in non-heme ferritins (55).    27  Figure 1-7: A conserved tyrosine acts as a one electron capacitor Incoming Fe(II) is distributed among two iron binding sites that form the ferroxidase centre and the gateway site. As a result, at least two species can be formed; the diferrous filled ferroxidase centre and the diferrous filled ferroxidase centre with a Fe(II) filled gateway site. Upon reaction of dioxygen with the diferrous centre, a peroxodiferric intermediate is formed (blue boxes). If the gateway site is empty, the peroxodiferric intermediate decays spontaneously to form the oxodiferric species (Path 1). If the gateway site is occupied by Fe(II), it may be oxidized by the peroxodiferric intermediate or hydrogen peroxide. A conserved tyrosine residue provides a fourth electron for reduction of oxygen to water. Adapted with permission from (107).   28 1.4 FERRITIN MINERAL CORE The central cavities of ferritins and bacterioferritins accommodate polynuclear iron clusters that constitute the iron core. The central cavity has a volume to accommodate up to 4500 iron ions; however, as isolated (bacterio)ferritin contains significantly less. Isolated BFR usually contains 800 – 1500 iron ions per molecule (108), while as isolated vertebrate ferritin generally contain 1000 – 3000 iron ions per molecule (109). The mean core sizes determined by electron microscopy for BFR and vertebrate ferritin are similar. Thus, the two types of natural cores have different compositions and densities (13,14,110). Structural data for BFR and ferritin cores are lacking due to the poly-disperse nature of the mineral core. The generally accepted composition of the vertebrate ferritin core is a ferrihydrate mineral with the approximate formula Fe10O14(OH)2 (111,112). Furthermore, studies on horse spleen ferritin revealed a multi-phasic mineral composed of ferrihydrite, magnetite, and hematite. All three of these minerals are present and occur at different locations in relation to the protein shell depending on core size. Ferrihydrite dominates as the core size increases (113).  The BFR core is generally described as iron-oxyhydroxide-phosphate and exhibits quite distinct structural and magnetic properties from those of vertebrate ferritin (114-116). The difference between the BFR and the ferritin core is thought to be due to the variation in phosphate content with ferritin having lower phosphate content than BFR. In BFR, phosphate is an integral constituent of the core, forming an amorphous Fe(III)-phosphate complex, whereas in ferritins, the phosphate is largely associated with the surface of the mineral (117).  The native core of bacterial ferritins has not been studied in detail yet. However, their average iron content is 600 – 2300 iron ions per molecule (14). Furthermore, the ferritin from Heliobacter pylori was isolated containing a significant amount of both iron and phosphate and  29 therefore is thought to have a mineral structure similar to that of BFR (13,14,118,119). Similar amounts of phosphate were found in phytoferritins, and the relatively high phosphate concentration found in plastids and the cytoplasm of prokaryotes can explain the phosphate rich iron cores (120).   1.5 IRON ENTRY AND EXIT 1.5.1 THE IRON ENTRY CHANNELS The 24 subunits of the (bacterio)ferritins are tightly packed together, forming eight narrow channels around the 3-fold axes and six channels around the 4-fold axis, each ~4 Å in diameter (10,15). In vertebrate ferritin, the hydrophilic 3-fold channels are believed to be the entry route of ferrous iron. The 3-fold ion channels are constructed by the same helix turn in each of three subunits around the 3-fold symmetry axes of the protein shell. A constriction in the channel created by three conserved glutamates (Glu134 in HuHF) is just large enough for a hydrated Fe(II) ion. Alteration of the conserved residues that line the 3-fold channels decreased ferroxidase activity and reduced iron binding at the ferroxidase centre (121,122). The eight 3-fold channels must distribute incoming Fe(II) to 24 ferroxidase centres. Three aspartate residues (Asp131, HuHF numbering), contributed by each subunit, create a channel proposed to orient the metal ions toward one of three ferroxidase centres (123,124).  NMR studies of BfMF led to the proposal that after oxidation at the ferroxidase centre, multimeric Fe(III) products move through nucleation channels into the cavity (125). If all the subunits are catalytically active (H-chain or M-chain ferritins), four nucleation channels end approximately at the 4-fold axis, thereby enhancing ordered mineral buildup. These proposed nucleation sites near the 4-fold channels differ from a previously proposed nucleation site for  30 HuHF in that they are further away from the ferroxidase centres. L-chain subunits are lacking the ferroxidase active sites and the nucleation channels. Ferritins with a high L-chain content, such as liver ferritin, build a less crystalline mineral (124,126). In EcBFR, DdBFR, and Rhodobacter capsulatus BFR, the ferroxidase centre of each monomer is located at the bottom of a pore, the ferroxidase pore (Figure 1-8C and D) (62,127). This pore has a circular entrance with a radius of ~1.4 Å that is wider at the bottom. Its major axis is roughly aligned with the site A – site B direction, and these iron binding sites are located ~6 Å below the pore entrance. The pores are formed mainly by hydrophilic residues, and ferrous iron is suggested to enter through this pore to form the di-iron ferroxidase centre. In EcFtnA, a similar pore is found; however, this pore is lined with hydrophobic residues and is proposed to provide access for dioxygen and perhaps iron ions shielded by water (50).  B-channels are located where one subunit dimer meets another in a side-on fashion (Figure 1-8). These B-channels extend to the interior of the ferritin cage and are large enough to admit the entry of iron ions into the inner core. However, B-channels are identified in bacterial ferritins and BFRs and not in mammalian ferritins. Calculations of the electrostatic surface potential from EcBFR, DdBFR, EcFtnA revealed clusters of negative potential around B-channels, suggesting a role in iron transport (127). As expected, calculations of electrostatic surface potential revealed a strong negative potentials around the 3-fold channels of mammalian ferritin but not in BFR and EcFtnA. Moreover, positive potential around the 3-fold channels of DdBFR were observed, which would not favour Fe(II) transport. In bacterial ferritins and BFRs, the B-channels are proposed having a role in iron uptake (127).  31  Figure 1-8: Electrostatic surface potential of HuHF and EcBFR (A) View down a 4-fold channel of HuHF. (B) View down a 3-fold channel of HuHF. (C) View down a 4-fold channel of EcBFR. One of four B-channels is marked with B, and one ferroxidase pore is marked with FP. (D) View along the 3-fold channel of EcBFR. Negative and positive surface potentials are colored red and blue, respectively. The contouring value of the potential is in kT/e. Adapted with permission from (127).       32 1.5.2 MINERAL DISSOLUTION AND IRON EXIT   The mechanism of how iron stored in ferritin is released is less clear. One possibility involves the proteolytic degradation of the ferritin shell. In mammalian ferritin, degradation occurs in the lysosome, and the released iron is exported to the cytoplasm (128,129). However, this mechanism implies a poorly controlled release of ferrous iron. In general, the release of Fe(II) from ferritin mineral requires the addition of a reductant (electrons) and protons. Fe(III) is reduced to Fe(II) at the core surface and exits the protein shell to be bound by a metallo-chaperone or Fe(II) chelating metabolites. The path taken by Fe(II) exiting the H-chain ferritin cages is suggested to be the 3-fold channels (130-132). Furthermore, the nature of the reductant in vivo is not yet known, although a dihydroflavin is a possibility (133-135). The discovery of human Poly r(C)-Binding Protein 1 (PCBP1), a metallo-chaperone that docks to ferritin and binds Fe(II), supports this reductant-chelator model. PCBP1 has been shown to deliver iron to ferritin; however, whether it functions as a mediator that releases iron from ferritin when cytosolic iron concentrations decrease, still has to be proven (136). Fe(III) ions from the ferritin mineral may be reduced by direct contact with exterior reductants or by electron transfer through the protein. An electron transfer pathway to mediate electrons across the protein shell of mammalian ferritin has been proposed (132,137,138). Furthermore, the 3-fold channels are proposed to function as gated pores. Open 3-fold channels allow molecules larger than 4 Å to cross the ferritin shell. Therefore, gating can control the rates of Fe(II) exit by regulating the access of biological reductants such as FMNH2 to the ferric iron in the mineral core (130,139). The pores are stabilized by hydrophobic interactions between Leu134 and Leu110 and the formation of an ion pair between Arg72 and Asp122 (133). Localized melting of these residues results in an open pore (140). Increased temperature,  33 chaotropic compounds, and amino acid substitutions are associated with very rapid mineral dissolution and iron release in bullfrog H-chain ferritin (133,139). In vivo molecular regulators, such as small metabolites or proteins, are proposed to recognize the pore gates, and hold them either closed or open, depending on biological iron need (130,141).  How the iron release process works in prokaryotic ferritins and BFRs is mostly unknown. However, kinetic studies of iron release of the heme-containing EcBFR with external reductants were performed by Yasmin et al. (70). Different release rates observed with various reductants indicated that the transfer of electrons from reductant to core Fe(III) is the rate limiting step rather than release of iron. A significant enhancement in rate was observed in the presence of heme; therefore, heme may play a role in catalyzing the electron transfer step. Furthermore, Yasmin et al. suggest that electron transfer occurs, at least in part, through the protein since direct electron transfer at the core surface would not be expected to be influenced by heme (70).   1.6 MARINE DIATOMS Diatoms are a major group of eukaryotic algae and are one of the most common types of marine phytoplankton. They are unicellular, photosynthetic organisms and belong to the Stramenopiles. Diatoms are unique in that their cell wall, called a frustule, is composed of two overlapping valves made of silica. Silica strips, the girdle bands, are used to hold the two valves together. There are four groups of diatoms; the radial centrics, the bi/multipolar centrics, the araphid pennates, and the raphid pennates. The latter are distinguished by a slid (raphe) in their wall that allows them to glide along surfaces (142,143). Centric diatoms are round, whereas the pennate diatoms have an elongated morphology. There are approximately 200,000 different species existing either as single cells or as chains of connected cells, varying in size from a  34 couple of microns to a few millimeters in length, respectively (143). Diatoms are mostly non-motile as their relatively dense cell-walls cause them to readily sink.  Marine diatoms play a major role in global primary production by using CO2, water, and sunlight to produce organic molecules required in cellular metabolism (144). Indeed, they are responsible for as much as 25% of global primary productivity and for an accordingly significant O2 production. Dead phytoplankton cells sink and are consumed by microbes that release inorganic nutrients, including CO2, into the environment. Most diatom degradation happens in the sunlit layer of the ocean, where the CO2 released is available for photosynthesis or is released into the atmosphere. However, more influential to global climate is the particulate organic matter that is exported out of the surface water to deeper depths (1000 m), where it can be sequestered for a few hundred years. About 20% of the diatom production is exported out of the surface water to ocean depth. Since the dissolved gases in the surface ocean are in equilibrium with the atmosphere, increased sequestration of organic C leads to a reduction of atmospheric CO2 (145,146). Larger cells like diatoms sink faster than smaller cells; therefore, diatoms are major contributors to carbon sequestration (147). Diatoms tend to dominate phytoplankton communities in well-mixed coastal and upwelling regions as well as along the sea-ice edge, where sufficient light, nitrogen (N), phosphorus (P), silicon (Si), and trace metals are available for their growth (143,148). In a region like the open Tropical Ocean, where the nutrient concentrations can be extremely low, the phytoplankton community is dominated by nano- or pico-plankton, such as cyanobacteria, with a higher surface to volume ratio (147,149).  Surface waters are generally low in the macronutrients N, P, and Si due to their consumption by phytoplankton. In ~30% of the oceans, including the North East Pacific,  35 Western Subarctic Pacific, the Southern Ocean, and part of the Equatorial Pacific, N and P are high in abundance and primary production is very low (144). In these “high nutrient, low chlorophyll” (HNLC) regions, iron is limiting phytoplankton biomass (150,151). Phytoplankton require iron for proteins involved in fundamental cellular processes, including photosynthesis (152,153).  In the open ocean, about 90% of the iron forms insoluble particulates sequestered in living organisms, adhering to sinking detritus, or forming macromolecular organic colloids. More than 99% of the soluble iron is complexed by organic ligands, such as siderophores. This leaves a concentration of dissolved inorganic iron below 0.2 nM in open ocean surface waters (154). Phytoplankton have several adaptation strategies that allow them to grow in the iron starved waters of the open ocean, although more slowly. In particular, reducing their cell size increases their cell surface area to volume ratio and increases nutrient acquisition (155,156). Furthermore, some diatoms are able to access the iron that is complexed with organic molecules such as siderophores, which are produced by marine bacteria in low iron environments (157-159). In these low iron regions, small pennate and centric diatoms and picoplankton dominate the community (160). However, HNLC regions are sporadically pulsed with new iron inputs from windblown dust, river runoffs, upwelling deep waters, and anthropogenic inputs. The small oceanic phytoplankton do not dominate blooms when iron is added to HNLC regions. Instead, iron addition shifts the composition of the natural community toward a predominance of larger diatoms species. Pennate diatoms readily bloom upon such iron addition (151,161-163). A series of in situ iron fertilization experiments in the Equatorial Pacific, the Southern Ocean, and the Subarctic Pacific all induced a boost of primary productivity and algal biomass and increased photosynthetic efficiency (150,162,164-166). The two pennate diatom genera Pseudo-nitzschia  36 and Fragilariopsis are almost universally responding to the iron inputs in HNLC regions (167-169).  Pseudo-nitzschia and Fragilariopsis are closely related raphid pennate diatoms that are widely distributed in the ocean (170). Pseudo-nitschia includes nearly 30 species, and they are several µm in width and can be 100 µm or longer in length. Pseudo-nitzschia is able to form colonies by overlapping each other in a stepwise fashion at the cell edges (Figure 1-9). During algal blooms, coastal species such as Pseudo-nitzschia multiseries can produce domoic acid, a neurotoxin that can cause shellfish poisoning. Domoic acid is thought to improve the uptake of copper, which is used in high affinity iron uptake systems of diatoms (171). Ferritin homologs were identified in pennate diatoms and thought to facilitate the blooming after iron fertilization in the open ocean by storing iron (33). Thus, pennate diatoms are able to continue to grow and divide after iron concentrations have returned back to a low and ambient level and thereby outcompete the centric diatoms during sporadic iron inputs.    Figure 1-9: A light micrograph of the marine pennate diatom Pseudo-nitzschia multiseries. Shown are one whole cell and two partial cells connected at the cell tips in a chain. The brown components of the cells are the chloroplasts. Scale bar = 5 µm. Image courtesy of K. Holtermann (172).   37 1.7 OBJECTIVES The identification of a novel ferritin from pennate diatoms may explain their success during sporadic iron inputs in low iron regions of the open ocean. Pennate diatoms can take advantage of pulsed iron supplies by using ferritin to safely store large amounts of iron that supports subsequent growth and division well after iron levels return to low concentrations. The initial characterization of ferritin from pennate diatoms supports a primary function as an iron storage protein. Ferritins from many organisms and all known ferritin subfamilies have been characterized structurally and functionally, and multiple iron storage mechanisms have been proposed. The overall objective of this work is to characterize PmFTN structurally and functionally and to provide insight into the iron storage mechanism of the ferritin family. I hypothesize that iron storage by diatom ferritin proceeds by ferrous iron binding to the ferroxidase centre followed by oxidation and mineral core formation. Furthermore, iron storage in pennate diatoms is proposed to be rapid to effectively compete during sporadic iron inputs. To define the ferrous and ferric iron binding sites in PmFTN, several crystal structures were solved from crystals soaked in ferrous iron and zinc sulfate under anaerobic and aerobic conditions. These are the first structures obtained from anaerobically grown and iron soaked crystals of any ferritin or bacterioferritin. Stopped-flow spectroscopy was used to measure the rate of ferrous iron oxidation in PmFTN and to determine the reaction phases of the ferroxidase reaction. These studies gave first insight into the ferrous iron binding and oxidation mechanism of PmFTN and revealed a stepwise iron binding of Fe(II) and dioxygen to the ferroxidase centre followed by a very fast oxidation phase. Furthermore, crystal structures revealed a unique iron binding site C. Fe-C is coordinated by Glu130 and Glu44, which both are suggested to have a  38 role in iron mobilization from site B to site C and toward the core. Therefore, the site C and site B/C variants, E44Q, E44H, and E130A, were constructed to determine the role of these amino acid residues in iron migration and oxidation. X-ray crystallography and kinetic studies were used to investigate these variants, and it was concluded that Glu130 and Glu44 regulate the iron flux through the ferroxidase centre. In particular, Glu130 stabilizes Fe(III) at the ferroxidase centre and thereby reduces the rate of iron mineralization. Moreover, iron was observed within the B-channels of the E44Q variant. This is the first evidence of this channel as an iron entry route in ferritin.  EcBFR is the structurally and functionally best understood bacterioferritin. However, no structure of an anaerobically iron-soaked crystal has been solved for EcBFR or any other ferritin besides PmFTN. For comparison to PmFTN, EcBFR was crystallized and soaked in Fe(II) in an anaerobic environment. The EcBFR structure revealed two Fe(II) bound to sites A and B of the ferroxidase centre.  39 CHAPTER 2: MATERIAL AND METHODS  2.1 CONSTRUCTS AND SITE-DIRECTED MUTAGENESIS 2.1.1. WILD TYPE PMFTN CONSTRUCT The construct used for wild type PmFTN protein expression was a pET28a(+) vector containing the coding region of PmFTN genomic DNA lacking the signal peptide and plastid targeting sequences (33). The expressed protein is missing the proline at the N-terminus and the valine at the C-terminus compared to the sequence found at UniProt entry B6DMH6.  2.1.2. SITE-DIRECTED MUTAGENESIS The site-directed variants E44H and E130A were created by subcloning from the wild type PmFTN construct. The cloning method used for the E130A and E44H variant constructs was a modified whole plasmid polymerase chain reaction (PCR) method (173). To clone the E130A construct, a first PCR reaction using the primers E130A-for and T7 Terminator (Table 2-1) was performed to synthesize megaprimers of ~300 bp. The wild type PmFTN construct was used as a template. To linearize the template, the PmFTN wild type construct was digested with the restriction enzyme NcoI before the PCR reaction. In a second PCR reaction, the whole plasmid was amplified using the megaprimers and the wild type PmFTN construct as a template. For the E44H variant, the megaprimer synthesis and whole plasmid amplification steps were combined into one PCR reaction using the primers E44H-for and T7 Terminator (Table 2-1). The high-fidelity DNA polymerase Phusion was used for all PCR reactions. DpnI was used to selectively degrade the template plasmid methylated DNA before transforming into E. coli dH5α cells.   40 The variant E44Q was synthesized by GenScript (Piscataway, NJ). All clones were verified by sequencing (Agencourt, Beverly, MA).   Table 2-1: Primers used in this study Primer Sequence (5’ to 3’) E130A-for E44H-for T7 Terminator CTTGTCTTCCGCGTTCACTTGTTG TCGCGTTCCTCCGCTGAATGTGCAAGCATGTAGGCGG  GCTAGTTATTGCTCAGCGG   2.2 RECOMBINANT PROTEIN EXPRESSION AND PURIFICATION OF WILD TYPE PMFTN AND VARIANTS E. coli BL21(DE3) cells transformed with the expression vector were inoculated into 2×YT (yeast tryptone) media containing 25 µg/ml kanamycin and grown at 37 °C with shaking to an optical density of ~0.8 at 600 nm. Protein expression was induced with addition of 0.2 mM isopropyl β-D-thiogalactosidase (IPTG). The cells were incubated at 25 °C overnight and afterwards pelleted by centrifugation. The pellet was resuspended in 20 mM tris(hydroxymethyl)aminomethane-hydrochloride (Tris-HCl) pH 8, 1 mM tris(2-carboxyethyl)phosphine (TCEP), and 5% glycerol (v/v), and the cells were lysed at 4 °C using an EmulsiFlex-C5 homogenizer (Avestin). Insoluble cell debris was removed by centrifugation. The supernatant was treated with deoxyribonuclease (DNase) 1 type 2 and filtered through a 0.8 µm syringe filter. Wild type PmFTN and variants were purified using a heat shock method as described by Marchetti et al. (33). The cell extract was aliquoted in 1 ml fractions, heat shocked for 5 min at 60 °C, and put on ice for 4 – 5 min. The precipitated E. coli proteins were removed by  41 centrifugation, and the remaining supernatant was filtered through a 0.22 µm syringe filter. PmFTN was further purified using Source 15Q (GE Healthcare) resin. The buffer used for the Source 15Q purification was 20 mM Tris-HCl pH 8, 5% glycerol (v/v), 1 mM TCEP, and the salt gradient used was 0 – 0.5 M NaCl. Purified PmFTN was dialyzed into 3% sodium dithionite (w/v), 1 M sodium acetate pH 4.8, and 1 mM TCEP to remove bound iron to yield the apoprotein. Apo-PmFTN was further dialyzed into 50 mM 2-(N-morpholino)ethanesulfonic acid (MES) pH 6.5, 100 mM NaCl, and 1 mM TCEP (Buffer A). The absence of iron was verified using the ferene-S assay. Briefly, the protein component is removed with trichloroacetic acid followed by reduction of the released iron ions with freshly prepared ascorbic acid. The colorimetric reagent ferene-S is specific for Fe(II). The cysteine residues were alkylated by first incubating PmFTN in Buffer A supplemented with 2 mM TCEP for 2 h at 37 °C with shaking followed by the addition of 10 mM iodoacetamide and incubation in the dark for 45 min at 37 °C with shaking. Some early preparations retained minor DNA contamination, which did not prevent crystallization but prevented accurate determination of kinetic parameters. Thus, the pellets of subsequent expressions were resuspended in 20 mM Tris-HCl pH 8, 1 mM TCEP, 5% glycerol (v/v), 5 mM ethylenediamine tetra-acetic acid (EDTA), and 0.5 M NaCl. After cell lysis, an alternative DNA precipitation was used by the addition of 10 µl of 10% polyethyleneimine (w/v) per ml of supernatant instead of DNase 1 treatment. The reaction was gently shaken for 10 min on ice, and afterwards, the DNA was pelleted by centrifugation. The supernatant was dialyzed overnight against 20 mM Tris-HCl pH 8, 1 mM TCEP, 5% glycerol (v/v), 5 mM EDTA, the buffer used for heat shock purification and Source 15Q chromatography.  42 Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was used to verify protein purity and showed a single band at ~19 kDa, the molecular weight of the PmFTN monomer.  Protein concentrations were determined with a NanoDrop 2000c spectrophotometer (Thermo Scientific), using the molecular mass of the mature PmFTN and an extinction coefficient of 24980 M-1 cm-1 at 280 nm, which was calculated from the primary sequence with the ProtParam tool at ExPASy (http://web.expasy.org/protparam/, May 24, 2014).  2.3 CRYSTALLIZATION AND STRUCTURE SOLUTION All structural figures were generated with the program PyMol (Version 1.2.0.0, Schrödinger, LLC).   2.3.1 WILD TYPE PMFTN Diffraction quality crystals of wild type PmFTN grew by hanging drop vapour diffusion in a 1:1 mixture of ~20 mg/ml protein in Buffer A supplemented with 2 mM TCEP and 10 mM iodoacetamide with reservoir solutions of 0.1 M sodium acetate pH 5.5, 1.1 – 1.4 M ammonium sulfate, and 0.9 – 1.4 M sodium chloride. The crystals were soaked in mother liquor supplemented with freshly prepared 2 mM ammonium ferrous sulfate hexahydrate for a time period of 5 min, 45 min, 4 h, or overnight or with 2 mM zinc sulfate for 1 h. The crystals were transferred to a cryoprotectant consisting of mother liquor supplemented with 30% glycerol (v/v) before flash freezing in liquid nitrogen.  To obtain crystal structures from anaerobic crystals, a PmFTN protein solution was brought into a glove box (Belle Technology, UK) containing a dinitrogen atmosphere maintained  43 at less than 35 ppm dioxygen. Anaerobic diffraction quality crystals of PmFTN grew by sitting drop vapor diffusion at room temperature at a 1:1 ratio of protein to well solution (0.1 M sodium acetate pH 5.5, 1.3 – 1.5 M ammonium sulfate, and 0.9 – 1 M NaCl). The crystals were soaked anaerobically in mother liquor supplemented with 2 mM ammonium ferrous sulfate hexahydrate for time periods of 75 min and 2 h. Thereafter, crystals were transferred into mother liquor supplemented with 30% glycerol (v/v) and immersed anaerobically in liquid nitrogen through a specialized port in the glove box.  PmFTN data sets were collected at the Stanford Synchrotron Radiation Lightsource (SSRL; Palo Alto, CA) on beamline 7.1 at 1 Å wavelength. The data was processed using HKL2000 (174) to resolutions of 1.65 – 2.2 Å. Phases were determined using MolRep (175) with a previously determined PmFTN crystal structure (Protein Data Bank (PDB) ID 3E6S) as the search model after removal of the iron ions and solvent molecules (33). The initial model was edited in Coot (176) and refined with Refmac5 (177). Waters were added by running coot:findwaters in Refmac5. Anomalous dispersion data were used to identify metal sites. Anomalous maps were obtained with the program fft using the model phases. Final occupancies for metal sites were set such that the B-factors were similar to that of the coordinating residues. In more detail, a range of iron occupancies for each metal was tested in 5% increments and narrowed to the B-factor most similar to that of the coordinating residues. When B-factors of the coordinating residues clustered over a wide range, metal occupancies were refined such that the B-factor was closest to that of the residues with the shortest ligand bonds. Furthermore, occupancies were also estimated using anomalous dispersion maps, in which a higher peaks corresponded to a higher metal occupancy. Lastly, (Fo – Fc) difference maps were inspected to optimise the fit of the occupancy to the electron density. A cif file containing atom composition,  44 bond length, bond angles, and torsion angles was generated for refinement of the alkylated cysteine (s-(2-amino-2-oxoethyl)-l-cysteine) at position 77.  2.3.2 PMFTN VARIANTS E44Q, E44H, AND E130A Crystals of PmFTN variants were grown by hanging drop vapor diffusion at room temperature at a 1:1 ratio of protein to well solution (0.1 M sodium acetate pH 5.5, 1 – 1.2 M ammonium sulfate, and 0.9 – 1.2 M sodium chloride). The protein was concentrated to ~20 mg/ml in Buffer A supplemented with 2 mM TCEP and 10 mM iodoacetamide. The crystals were soaked in mother liquor supplemented with freshly prepared 2 mM ammonium ferrous sulfate hexahydrate for a time period of 5 min, 45 min, and overnight for the E44Q variant, 45 min and 3 h for the E44H variant, and 5 min and overnight for the E130A variant. The crystals were transferred to a cryoprotectant consisting of mother liquor supplemented with 30% glycerol (v/v) before flash freezing in liquid nitrogen.  PmFTN data sets were collected at the SSRL on beamline 7.1 and at the Canadian Macromolecular Crystallography Facility of the Canadian Lightsource (CLS; Saskatoon, SK) on beamline 081D-1 at 1 Å wavelength. Data were processed using Mosflm (178). The data were merged using Scala (179) in CCP4, and the resolution cut-off of 1.8 – 2.0 Å was determined by Mean((I)/sd(I)) ≥ 2 and Mn(I) half set correlation CC(1/2) > 50%. Phases were determined using MolRep (175) with a previously determined wild type PmFTN crystal structure (PDB ID 4IWJ) as the search model after removal of the iron ions and solvent molecules. The initial model was edited in Coot (176) and refined with Refmac5 (177). Waters were added by running findwaters in Coot and refining in Refmac5. Anomalous maps to identify metal sites were obtained with the  45 program fft using the model phases. As previously described, final occupancies for metal sites were set such that the B-factors were similar to that of the coordinating residues.  2.3.3 E. COLI BACTERIOFERRITIN EcBFR was expressed and purified by Dr. Nick E. Le Brun and his lab at the University of East Anglia as described by Crow et al. (67). EcBFR protein solution was brought into a glove box (Belle Technology, UK) containing a dinitrogen atmosphere maintained at less than 35 ppm dioxygen. The protein was in 100 mM MES pH 6.5 and concentrated to 10 mg/ml. Diffraction quality crystals grew by sitting drop vapor diffusion at room temperature at a 1:1 ratio of protein to well solution (0.15 M sodium cacodylate, 2.4 M ammonium sulfate, and 0.2 M NaCl). A crystal was soaked anaerobically in mother liquor supplemented with 2 mM ammonium ferrous sulfate hexahydrate for a time period of 2 h. The crystal was transferred into mother liquor supplemented with 30% glycerol (v/v) before flash freezing anaerobically in liquid nitrogen.  An EcBFR data set was collected at SSRL on beamline 7.1 at 1 Å wavelength. Data were processed in the space group P1 using XDS (180). The space group P42212 was determined using Pointless in CCP4 (181), and the data were merged using Scala in CCP4 (179). The resolution cut-off was determined by Mean((I)/sd(I)) ≥ 2 and Mn(I) half set correlation CC(1/2) > 50%. Phases were determined using MolRep (175) with a previously determined aerobic crystal structure (PDB ID 3E1M) as the search model after removal of the iron ions and solvent molecules. Refinement of the model and determining metal sites was performed as described for PmFTN variants.   46 2.4 STOPPED-FLOW ABSORPTION SPECTROSCOPY  Rapid kinetic experiments were carried out using a stopped-flow instrument (Applied Photophysics DX17MV). Changes in absorption on addition of ferrous iron to apo-PmFTN wild type and variants were measured. Prior to each experimental run, 10 – 400 µM ferrous iron working solutions were freshly prepared using a 50 mM stock solution of ferrous ammonium sulfate prepared in deoxygenated water, which was bubbled with argon gas for 2 h prior to use. The stock solution was acidified with 1 ml 37% HCl per 100 ml solution. For wild type PmFTN and each variant, 1 µM apoprotein in 100 mM MES, pH 6.5 and 200 mM NaCl was mixed 1:1 by the stopped-flow instrument with the various ferrous iron working solutions, resulting in a protein concentration of 0.5 µM during data acquisition. All stopped-flow experiments were performed at 25 °C. Resulting data were fit to a single, double, or tri-exponential function as required using the program Origin (v8, OriginLab). Regeneration experiments were carried out using 1 µM wild type and variant PmFTN containing 48 iron ions added under aerobic conditions a fixed period (30 min or 20 h) prior to the stopped-flow experiment. Resulting data were fit to an appropriate exponential function as above. For additions of 400 Fe(II) per PmFTN, rates of Fe(II) oxidation were calculated from initial, linear increases in A340 nm per unit time. Because 340 nm values could not be assumed to be constant between proteins, values of ΔA340 nm min-1 were converted to Fe(II) oxidized per min (μM min-1) (182). All the stopped-flow spectroscopy was performed in Dr. N. E. Le Brun’s lab at the University of East Anglia (UK).   47 CHAPTER 3: MECHANISM OF FERROUS IRON BINDING AND OXIDATION BY FERRITIN FROM A PENNATE DIATOM   3.1 INTRODUCTION Diatoms are unicellular photosynthetic organisms that play a major role in global primary production and carbon sequestration into the deep ocean (144-147). In HNLC regions of the open ocean, primary productivity and therefore CO2 uptake from the atmosphere is limited due to iron availability. These regions are sporadically pulsed with new iron inputs from dust or upwelling deep waters. Pennate diatoms readily bloom upon such iron additions and continue to grow and divide after iron levels return to a low and ambient level (151,161). The expression of ferritin is thought to facilitate the blooming of pennate diatoms after iron fertilization in the open ocean. A crystal structure of recombinant, iron-soaked ferritin derived from the pennate diatom Pseudo-nitzschia multiseries (PmFTN) revealed three iron biding sites: sites A and B forming the ferroxidase centre and a third site C at the vicinity of the ferroxidase centre. Site C is unique in that Fe-C is coordinated by one glutamate residue (Glu44) (33). A glutamate is found at position 44 only in diatoms and cyanobacteria; moreover, no third iron binding site is found in human H chain ferritin or other eukaryotic ferritins. To better understand the ferroxidase reaction and iron binding in PmFTN, the X-ray structures of several PmFTN crystals soaked for various durations in ferrous iron and zinc sulfate under aerobic and anaerobic conditions were solved. Furthermore, stopped-flow kinetic analysis was applied to determine reaction phases of the ferroxidase reaction. These studies suggest a  48 model of stepwise iron binding to the ferroxidase centre of PmFTN followed by a very fast iron oxidation phase and partial mobilization of iron from the ferroxidase centre.  3.2 RESULTS 3.2.1 OVERALL FOLD Protein crystals of recombinant PmFTN soaked in Fe(II) or Zn(II) under various conditions diffracted to at least 2.2 Å resolution. Analysis of Ramachandran plots showed that in all structures more than 96% of the residues were in the preferred regions. All but one of the crystals analyzed were of space group P23 with eight ferritin monomers in the asymmetric unit (Table 3-1). The refined structures are nearly complete with at most 12 residues absent from the N or C terminus. As shown previously, the structures confirm the typical ferritin arrangement (Figure 3-1): 24 subunits assemble to form a hollow spherical shell (33). The monomers adopt a four-helix bundle plus a shorter C-terminal α-helix. No significant changes in the fold of the ferritin monomer were observed upon metal treatment. Superposition of the monomers from each structure resulted in a 0.28 Å r.m.s.d for all Cα atoms.  49  Figure 3-1: Crystal structure of PmFTN (A) Crystal structure of the recombinant, aerobic iron-soaked (overnight) P. multiseries ferritin multimer shows the typical ferritin arrangement, in which the 24 subunits form a spherical shell. (B) Monomer showing the ferroxidase centre side chains and bound iron ions. Magenta, A helix; green, B helix; blue, C helix; yellow, D helix; red, E helix; orange spheres, iron ions.   50 Table 3-1: Data collection and refinement statistics of wild type PmFTN 1Soaking time for each metal is indicated in parentheses in minutes unless otherwise indicated.  2Values in parenthesis for the data collection are for the highest resolution shell indicated.  Anaerobic Fe (75)1 Anaerobic Fe (2 h) Fe (5) Fe (45) Fe (4 h) Fe (o.n) Zn (1h) Data collection    Resolution range (Å)     Space group    Unit cell dimensions (Å)       No. subunits in ASU    Unique reflections    Completeness (%)    Redundancy    Average I/σI    R-merge    Wilson B (Å2) Refinement    R-work (R-free)    Avg. B (Å2)    No. water    No. iron    No. zinc    R.m.s.d. bond length    ESU from maximum     likelihood (Å)        Ramachandran plot (%)    In most-favourable    In allowed PDB ID   48.52 - 2.00  (2.05 - 2.00)2 P23 a = b = c =  174.77 8 119135 99.8 14.3 23.8 (4.8) 0.115 (0.658) 29.8  0.187 (0.235) 31.5 853 8 N/A 0.020 0.099   96.6 3.4 4ITW  48.40 - 2.10  (2.15 - 2.10) P23 a = b = c = 174.32 8 102117 99.5 14.6 32.3 (5.4) 0.084 (0.597) 30.3  0.201 (0.247) 32.9 524 8 N/A 0.019 0.116   96.0 4.0 4IXK  42.55 - 2.10  (2.16 - 2.10) P23 a = b = c = 175.28 8 103862 99.6 13.0 35.5 (7.1) 0.075 (0.371) 24.2  0.176 (0.225) 27.9 821 6 N/A 0.020 0.102   96.5 3.6 4ITT  41.36 - 1.95 (2.00 - 1.95) P23 a = b = c = 175.32 8 128778 99.2 11.6 33.4 (3.8) 0.064 (0.528) 24.8  0.173 (0.213) 28.5 1006 16 N/A 0.024 0.081   96.3 3.7 4IWJ  42.59 - 2.20  (2.26 - 2.20) P23 a = b = c = 175.45 8 90434 99.7 14.0 21.9 (4.7) 0.122 (0.553) 26.2  0.187 (0.242) 28.3 538 35 N/A 0.019 0.120   96.9 3.1 4ISP  47.95 - 1.65 (1.69 - 1.65) P4212 a = b = 126.25 c = 170.30 6 163306 99.2 10.7 32.1 (3.7) 0.065 (0.507) 22.4  0.176 (0.205) 25.5 872 24 N/A 0.028 0.051   96.7 3.3 4IWK  42.51 - 2.00  (2.05 - 2.00) P23 a = b = c =  175.11 8 117223 97.5 14.7 29.5 (7.4) 0.086 (0.452) 26.9  0.184 (0.232) 30.1 912 N/A 32 0.021 0.096   96.0 4.0 4ISM  51 3.2.2 IRON BINDING TO PMFTN IN THE ABSENCE OF DIOXYGEN The first step in the ferritin reaction is binding of ferrous iron to the ferroxidase centre. To observe the binding of ferrous iron, apo-PmFTN crystals were grown in an anaerobic environment, and the octahedral crystals were soaked in 2 mM ferrous iron. Two crystal structures were obtained with crystals soaked in ferrous iron for 75 min and 2 h to resolutions of 2 Å and 2.1 Å, respectively. The ferroxidase centre of the higher resolution structure obtained from the crystal soaked for 75 min is shown in Figure 3-2A. Ferrous iron ions are exclusively found in the ferroxidase centre of each monomer. Furthermore, ferrous iron is observed solely in site A and is refined with an occupancy of ∼50% and B-factors similar to the coordinating residues. Fe-A is coordinated in an approximate tetrahedral arrangement by the side chains of Glu15 (∼2.2 Å) and Glu48 (∼2.2 Å). His51 forms a weaker interaction with ferrous iron (2.5 – 2.9 Å). The coordination sphere is completed by one or two solvent molecules (∼2.1 Å and ∼2.5 Å) depending on the monomer in the asymmetric unit. Ferroxidase site B is occupied by a solvent molecule.     52  Figure 3-2: Ferroxidase centres of PmFTN Monomer A of the asymmetric unit is shown. Iron and zinc ions are drawn as orange and gray spheres, respectively, and cyan spheres represent water molecules. Side chains of selected residues are drawn in sticks with carbon, nitrogen, and oxygen atoms in the backbone color, blue, and red, respectively. Solid lines are metal ligand bonds, and dashed lines are selected hydrogen bonds. (A) Crystal soaked for 75 min in ferrous sulfate under anaerobic conditions. One iron ion is bound at ferroxidase site A at 50% occupancy, and site B is occupied by a water molecule. The black mesh represents the anomalous dispersion map contoured at 5 σ. (B) Crystal soaked overnight in ferrous sulfate under aerobic conditions. Three iron ions are bound at sites A, B, and C, and one further iron ion was found beyond site C. (C) Crystal soaked for 1 h in zinc sulfate. Three zinc ions are bound at the ferroxidase sites A, B, and C. (D) A 2Fo – Fc electron density map contoured at 1 σ around site A and site B of PmFTN soaked overnight in ferrous sulfate is shown as a gray mesh.     53 3.2.3 IRON BINDING TO PMFTN IN THE PRESENCE OF DIOXYGEN In previous work, apo-PmFTN crystals were soaked for 10 min in ferrous sulfate in the presence of dioxygen (33). The resulting structure revealed iron bound at two sites: ferroxidase site B and a new site C nearby. Site A was occupied by solvent. To determine whether occupation of the ferroxidase centre and site C is time-dependent, apo-PmFTN crystals were soaked for 5 min (Fe (5)), 45 min (Fe (45)), 4 h (Fe (4 h)), and overnight (Fe (o.n)),  in 2 mM ferrous sulfate. Structures of crystals exposed to ferrous iron for 4 h or longer revealed iron bound in all three sites as well as up to two additional sites at the inner surface of the ferritin sphere (Figures 3-2B and 3-3A). The latter two sites may serve as the nucleation site for formation of the mineral core. Fe-A is coordinated by the conserved residues Glu15, Glu48, and His51. Coordination of Fe-B is by 2 – 3 glutamate residues, depending on the subunit. A bridging oxygen atom is modeled between sites A and B and between sites B and C (Figure 3-2B). A second solvent oxygen atom coordinates Fe-A. The bridging solvent molecule between sites A and B might be a bridging oxo group rather than a water molecule, building a diferric-oxo bridge as seen in other ferritins (183). Sample electron density for the overnight-soaked structure at the ferroxidase site including the bridging ligand is presented in Figure 3-2D. In all subunits, the side chain carboxylate of Glu48 is a bridging ligand to sites A and B, whereas Glu94 coordinates to Fe-B only. Glu130 adopts multiple conformations depending on the soaking time and is observed coordinated to Fe-B and/or Fe-C. In the Fe (o.n) structure, Glu130 is tilted toward Fe-B, whereas in the Fe (4 h) structure, it is predominately coordinated to Fe-C (Figures 3-3, A and B). In addition to Glu130, Fe-C is coordinated by Glu47 and Glu44. Iron occupancy of the sites A, B, and C varies depending on the incubation time of the crystals in ferrous iron in the presence of dioxygen (Table 3-2). Iron is observed in site A only  54 after soaking for 45 min, whereas after 5 min, iron partially occupies site B in six of eight crystallographic subunits. In the other two subunits, a water molecule is modeled at the equivalent position. The average distance between the iron ions in sites A and B is 3.78 ± 0.09 Å in Fe (45), which decreases to 3.59 ± 0.01 Å upon overnight soak. In general, site C, which is situated between the ferroxidase centre (sites A and B) and the mineral core, has the lowest overall iron occupancy (40 – 60%). Only in crystals incubated in ferrous iron for 4 h and longer was iron bound to site C. In the subunits with water bound to site C, the side chain of Glu130 adopts an alternative rotamer such that it is directed away from the site and toward the core.   Figure 3-3: Residue Glu130 exhibits flexible coordination to Fe-B and Fe-C (A) Monomer A of PmFTN Fe (4 h). Glu130 is coordinating to Fe-C and may also interact with Fe-B. (B) Monomer B of PmFTN Fe (o.n). Glu130 is coordinating to Fe-B and may also interact with Fe-C. (C) Monomer C of PmFTN Zn (1 h). Glu130 is coordinating to Zn-B or Zn-C. Molecular representations and the color scheme are the same as in Figure 3-2. Bond lengths are in Å.        55 Table 3-2: Range of iron and zinc ion occupancy observed in binding sites A, B, and C of aerobically Fe(II) and Zn(II) soaked PmFTN crystals   1Soaking time for each metal is indicated in parentheses in minutes unless otherwise indicated. 2The metal occupancy was determined with an estimated error of ± 10%.   3.2.4 PMFTN BINDS ZN(II) Zinc ions are competitive inhibitors of ferritin and bacterioferritin (103). A crystal structure of Zn(II) inhibited ferritin confirms that this metal binds to the ferroxidase centre and site C. The zinc ions are coordinated by the same residues as seen in the structures from aerobically iron soaked crystals (Figure 3-2C). Glu130 is predominately coordinated to the zinc ion in site C (Zn-C), although in three of the monomers in the asymmetric unit, Glu130 adopts two equal occupancy conformations, one that coordinates to the zinc ion in site B (Zn-B), and the other that coordinates to Zn-C (Figure 3-3C). Furthermore, the zinc ion in site A (Zn-A) is modeled as coordinated by two water molecules, one of which is bridging to site B. This arrangement of coordinated solvent is similar to that observed in the iron structures. In the case of iron, the modeled oxygen atom may be an oxo-bridge rather than water. An overlay of the zinc structure with the iron structures showed that the Zn-B is slightly shifted toward site A and that Zn-C is closer toward the core as compared with the iron.   Fe (5)1 Fe (45) Fe (4h) Fe (o.n) Zn (1h) Site A H2O Fe (75-90 %) Fe (75-90 %) Fe (85-95 %) Zn (100 %) Site B Fe (0-50 %)2 Fe (40-50 %) Fe (75-80 %) Fe (80-85 %) Zn (85-90 %) Site C  H2O Fe (50-60 %) Fe (40-50 %) Zn (60-75 %) Inner surface metal ions   Fe (50-60 %) Fe (0-50 %) Fe (30-35 %)   56 3.2.5 STOPPED-FLOW ABSORPTION SPECTROSCOPY Stopped-flow experiments were performed to monitor kinetics of iron oxidation after ferrous iron addition to apo-PmFTN. Absorption was measured at 340 nm, which reports on the oxidation of Fe(II) to Fe(III). Figure 3-4A shows the absorption changes following the mixing of 1 μM apo-PmFTN with a variable amount of ferrous iron as a function of time. Fe(II) oxidation occurred extremely rapidly with completion for all additions within 0.5 s. For the addition of ∼50 Fe(II) ions per protein (two Fe(II) ions per subunit), the t½ was <50 ms, demonstrating a rate of reaction that is an order of magnitude faster than those measured for EcFtnA and human H-chain ferritin under comparable conditions (101,184). Each trace fitted well to a single exponential function (Figure 3-4A), giving an observed (pseudo-first order) rate constant, k0, for each addition. Figure 3-4B shows a plot of k0 as a function of Fe(II) concentration. Remarkably, there is a linear relationship demonstrating a first order dependence of the rate of oxidation on the concentration of Fe(II). This observation is significant because it indicates that Fe(II) binding, rather than Fe(II) oxidation, is the rate-determining step of the reaction, consistent with the unprecedented overall rate of Fe(II) oxidation. The slope of the line gives an apparent second order rate constant, k, of 5.1 ± 0.10 × 105 M−1 s−1. A plot of ΔA340 nm at 0.5 s versus Fe(II) added per PmFTN complex (Figure 3-4C) reveals that the rapid oxidation phase is saturated at ∼50 iron ions per protein or two per subunit. Such behavior is characteristic of ferritins in which the ferroxidase centre is the initial site of rapid Fe(II) oxidation. Clearly, Fe(II) added in excess of that required to saturate the ferroxidase centre is not oxidized at a rate close to that of the initial oxidation phase, and we conclude that Fe(II) does not undergo rapid oxidation at the third site (site C). This is demonstrated in Figure 3-4D, which shows changes in A340 nm over 1000 s following the addition of 100 Fe(II) ions per protein.  57 Although the first ∼50 Fe(II) ions are oxidized extremely rapidly, the subsequent 50 are not; in fact, oxidation is not complete at 1000 s. Thus, the kinetic distinction between these phases is unusually clear. The partial occupancy of iron in the ferroxidase sites A and B after prolonged exposure to Fe(II) under aerobic conditions (Table 3-2) indicates that iron is likely to be mobile following oxidation at the ferroxidase centre. Thus, A340 nm was monitored over a longer time period (Figure 3-4E). The data show that a small increase in absorption occurs during the few seconds following the rapid oxidation phase. This increase was observed at all levels of Fe(II) additions, including those that are substoichiometric, and so these changes are not due to oxidation of Fe(II) bound elsewhere. Thus, these absorption changes may be attributed to reorganization of iron in the ferroxidase centre and possibly site C following the initial rapid oxidation. Mobility of iron following oxidation of Fe(II) at the ferroxidase centre could result in complete loss of Fe(III) from the centre, leading to regeneration of the apo form. This form would be expected to exhibit the rapid oxidation phase observed above (Figure 3-4A). To determine whether the rapid oxidation phase is recovered, PmFTN was loaded with 48 Fe(II) ions per 24-mer and then subsequently incubated for 30 min or 20 h. Then, a further aliquot of Fe(II) (either 40 or 80 Fe(II) ions per protein) was added, and changes in A340 nm were measured. The data (for 30 min; Figure 3-4F) show that no rapid phase of Fe(II) oxidation is observed. Two distinct phases are observed, although it is not yet clear what these correspond to. Data for 20 h are similar (not shown). Clearly, there is no full regeneration of the apo form of PmFTN following oxidation of the initial addition of Fe(II).  58          59 Figure 3-4 (previous page): Kinetic analysis of Fe(II) oxidation catalyzed by PmFTN (A) Stopped-flow measurements of ΔA340 nm as a function of time following additions of variable Fe(II) (as indicated) to apo-PmFTN. Each of the A340 nm traces were fit to a single exponential function (solid lines). (B) Plot of observed (pseudo-first order) rate constants obtained from fitting the data in A as a function of Fe(II) concentration. Standard errors (error bars) are shown, and a linear fit of the data is drawn. (C) Plot of total amplitude of absorbance changes at 340 nm at 0.5 s (data from A) as a function of Fe(II) added per PmFTN protein. The two clear phases are highlighted, intersecting at ∼50 Fe(II) ions per 24mer. (D) Stopped-flow A340 nm measurement as a function of time following addition of 100 Fe(II) ions per apo-PmFTN. The initial very rapid oxidation of 48 Fe(II) ions per protein is not captured well on this extended time scale, but the subsequent, much slower oxidation of Fe(II) in excess of that needed to saturate the ferroxidase centre sites is observed clearly. (E) Stopped-flow measurements of A340 nm over an extended time period revealing small increases in absorbance following the rapid oxidation phase. The number of Fe(II) ions per apo-PmFTN is indicated. (F) Stopped-flow measurements of absorbance changes at 340 nm following the addition of 40 or 80 Fe(II) ions per protein to a sample of PmFTN previously treated with 48 Fe(II) ions per protein under aerobic conditions. The incubation time between Fe(II) additions was 30 min. For all experiments, PmFTN (1 μM) was in 0.1 mM MES, pH 6.5.   3.3 DISCUSSION The ferritin family is part of the extensive ferritin-like superfamily of proteins, which includes the enzymes ribonucleotide reductase and methane monooxygenase. All members of the superfamily are believed to share the same characteristic four-helix bundle (or part thereof) structural motif. The iron storage ferritin family proteins use catalytic di-iron centres to oxidize Fe(II) and deposit the resulting Fe(III) as a mineral in the central cavity. A key question in ferritin function is the mechanism of iron mobilization in and out of the ferroxidase centre and the core. Iron was visualized in the ferroxidase centre (sites A and B) and a third site C of PmFTN. The residues interacting with Fe-A (Glu15, Glu48, and His51) are conserved with those of all characterized eukaryotic and prokaryotic ferritins. In addition to Glu48, Glu94 coordinates Fe-B and is also conserved. A third iron binding site is generally not associated with ferritins from vertebrates; however, a site C has been observed in some ferritins from bacteria and archea, for example in   60 EcFtnA (50,101), PfFtn (58), and AfFtn (57). Nevertheless, these sites differ from that of PmFTN in terms of the number and origin of glutamate residues. Of these residues, Glu130 is in common, is observed bridging the iron ions of sites B and C (Figure 3-3), and is a conserved residue in prokaryotic ferritins. In the PmFTN structures Fe (4 h) and Fe (o.n), Glu130 is a ligand to Fe-B and Fe-C. In contrast, in the earlier PmFTN structure derived from a crystal soaked in ferrous iron for 10 min by Marchetti et al. (33), Fe-C is coordinated by only Glu44. In all known prokaryotic ferritins, Glu44 is substituted by a glutamine or a histidine. The equivalent His46 in EcFtnA is proposed to orient Glu130 so it can bind iron ions in sites B and C as well as gating the passage of the metal through these sites (50). Glu44 and Glu130 in PmFTN may have a similar function of gating the passage of iron from site B to site C and toward the core.  As the aerobic iron soaking time increases from minutes to hours, iron is observed first in site B followed by site A and eventually occupies all three sites (Table 3-2). In contrast, only site A is occupied by Fe(II) in crystals of PmFTN under anaerobic conditions, even though the crystals were exposed to 2 mM Fe(II) for over an hour (Figure 3-2A). This observation is in contrast to short (∼1 min) aerobic Fe(II) soaking experiments with BfMF in which both sites A and B are occupied (183). Interestingly, the inter-iron distance in some subunits of BfMF is comparable with that observed with Cu(II) as a proxy for Fe(II) (∼4.3 Å). Longer exposure to ferrous iron results in a shortening of the di-iron interatomic iron distance to ∼3.1 Å. In PmFTN, iron occupancy at both sites is only observed after prolonged iron exposure, and the Fe-A and Fe-B intermetal distance of less than 3.8 Å decreases only slightly with time to 3.6 Å, suggesting that in the structures where iron is bound to both sites, it is in the Fe(III) state. Note that with the data presented here, we are not able to directly determine the oxidation state of iron bound to the crystals under aerobic conditions, including Fe-B in Fe (5). A single high affinity and two low  61 affinity Fe(II) binding sites were identified in PfFtn by calorimetry in the absence of dioxygen (106). Site-directed mutagenesis was used to propose assignment of the high affinity site to site A, consistent with our anaerobic crystallographic observations in PmFTN. Kinetic measurements of PmFTN iron oxidation revealed an extremely rapid initial oxidation phase involving the binding and oxidation of two ferrous iron ions. The first order dependence of the rate of ferroxidase centre oxidation on the concentration of Fe(II) demonstrates a close link between binding and oxidation events such that they cannot be distinguished. Thus, oxidation occurs immediately upon Fe(II) binding to PmFTN, and the binding event can be viewed as the slow step of the reaction. This is in contrast to previous reports of ferritins in which binding and oxidation are considered to be kinetically distinct events. Measurement of Fe(II) binding kinetics is not generally straightforward, although it was possible for EcBFR because Fe(II) binding caused a perturbation of absorbance due to the heme groups. In that case, Fe(II) binding occurred on a much shorter time scale than the subsequent Fe(II) oxidation. Interestingly, Fe(II) binding to EcBFR occurred with a second order rate constant of 2.5 × 105 M−1 s−1 (at 30 °C) (184), a value similar to that measured here (at 25 °C) for PmFTN-catalyzed Fe(II) oxidation. An oximetric assay previously showed that the ferroxidase reaction of PmFTN is associated with consumption of dioxygen in a ratio of 1.9 ± 0.2 Fe(II):O2 (33). Furthermore, addition of catalase to the assay solution resulted in the regeneration of O2, indicating production of H2O2 by the ferroxidase reaction as seen with ferritins (93) but not BFRs (103) or Dps (185). In contrast, the oxidation stoichiometry of EcFtnA is 3 – 4 Fe(II) ions per O2, which is suggested to be a consequence of the binding of three iron ions (one at site C), leading to reduction of O2 to  62 water rather than hydrogen peroxide, or a mix of the proportions of each product generated (55,101).  The ferroxidase reaction in EcFtnA is similar to that observed in human H chain ferritin, although it is more complex due to the third iron in site C (50). Site C, however, is not essential for ferroxidase activity in EcFtnA as site C variants showed only a slight decrease in the overall oxidation rate but the expected stoichiometry of two ferrous iron per dioxygen (101). In contrast, although a third iron site is present in PmFTN, a 2:1 Fe:O2 stoichiometry is retained. The kinetic data reported here support the conclusion that only two Fe(II) ions are initially oxidized per subunit. A key structural difference is that site C in PmFTN is only 3.5–3.7 Å from site B, whereas site C in EcFtnA is 7 – 8 Å from the A/B pair. A third iron binding site was observed in EcBFR as well as human mitochondrial ferritin (36,67). However, these iron sites were observed to be at the core surface and are more likely involved in the nucleation/mineralization process rather than in ferroxidase centre-catalyzed iron oxidation (67,186). Only ferroxidase site A is occupied by ferrous iron in the anaerobic crystal; however, stopped-flow data show saturation of the rapid phase 2 after binding of 2 ferrous iron equivalents per monomer of PmFTN. Together these results point to stepwise binding of the ferrous iron and dioxygen to the ferroxidase site. A model can be proposed in which one ferrous iron binds to site A followed by the binding of the oxidant. Only when the latter is bound can the second ferrous iron bind to site B. Thus, at the moderate iron concentrations (2 mM) used for soaking experiments, a second Fe(II) ion is not observed at the centre in the absence of the oxidant (dioxygen). We note that a similar model was proposed for the two Dps from Bacillus anthracis (187). These are 12-mer ferritin-like proteins that contain intersubunit dinuclear ferroxidase centres that are distinct from those of the 24-mer ferritins but nevertheless share some common  63 features. For PmFTN, such a model accounts for the observed rate dependence on Fe(II) because once the second Fe(II) binds, oxidation can proceed immediately. Thus, ferrous iron binding to site B of the ferroxidase centre is proposed to be the rate-determining step. Zn(II) is an inhibitor of EcFtnA and is proposed to compete with ferrous iron for the dinuclear centre and consequently inhibit oxidation at these sites (188). In the crystal structure, Zn(II) does bind to sites A and B but is not observed in site C (50). Nonetheless, from the Fe:O2 stoichiometry of the EcFtnA reaction, all three sites were proposed to bind Fe(II) during catalysis. Crystal structures of Zn(II) complexes of EcBFR and human mitochondrial ferritin have Zn(II) bound at sites A and B of the ferroxidase centre, consistent with a proposed model of Fe(II) binding (187,189). However, anaerobic Fe(II) complexes for these systems are not available in the literature. We have directly compared Zn(II) and Fe(II) binding in PmFTN (Figure 3-2). Zn(II) bound to sites A, B, and C in contrast to the two sites observed in EcFtnA. Furthermore, overall Zn(II) occupancy of the three metal sites resembles iron bound in the presence of dioxygen in PmFTN rather than mimicking Fe(II) binding. Thus, the use of Zn(II) and likely other metal ions as analogs of Fe(II)/Fe(III) may not identify the correct binding sites in other ferritins. Two models have been proposed for the mechanism of ferroxidation by ferritins and bacterioferritins. In one model, the ferroxidase centre functions as a substrate site as seen in HuHF (88), EcFtnA (89), and BfMF (90). Ferrous iron binds to the ferroxidase centre, and after oxidation, ferric iron rapidly migrates to the mineral core. In a second model, first described for EcBFR (67) and PfFtn (91), the ferroxidase centre is a stable di-iron site that functions as a cofactor after the binding of 2 equivalents of ferrous iron per subunit. Additional ferrous ion ions are then added directly to the mineral core, and the ferroxidase centre functions solely in oxygen  64 or peroxide reduction. Recently, Honarmand et al. (106) proposed a unifying mechanism in which the Fe(III) product at the ferroxidase centre remains bound to the ferroxidase centre but is rapidly displaced by incoming Fe(II). A prediction of this revised model is the observation of a fully Fe(III)-loaded ferroxidase centre in crystals after prolonged soaking in Fe(II). The ferroxidase centre of PmFTN was not fully occupied after soaking aerobic crystals in ferrous iron for 45 min, suggesting that iron movement occurred at the ferroxidase centre during iron loading with 2 mM ferrous iron over an extended time period. Small absorbance changes immediately following oxidation of Fe(II) at the ferroxidase centre are consistent with this conclusion. Nevertheless, for PmFTN, the rapid oxidation of ferrous iron was not regenerated upon up to 20 h incubation, indicating that iron remains present at least in part at the ferroxidase centre. Hence, if Fe(III) is displaced, the subsequent iron oxidation is much slower than the initial oxidation. Thus, the mechanism of mineralization in PmFTN appears to be more complex with partial iron migration to the core. Our data indicate that site A of the ferroxidase centre has a higher affinity than site B for Fe(II) under anaerobic conditions. The two distinct kinetic phases observed after the second addition of 48 Fe(II) ions may also be related to slow iron migration to the core likely involving the third iron binding site (site C) and perhaps other sites along a path to the cavity. In BfMF, the transit of Fe(III) from the ferroxidase centre to the cavity has been shown to occur via a pathway through the subunit toward a 4-fold channel (125). Site C may function to direct Fe(III) along a different path to the mineral core. In EcBFR, the two ferroxidase centre sites were fully occupied after 2.5 min of soaking, and occupancy was not affected following oxidation, suggesting that a distinct mechanism is in operation (67).   65 CHAPTER 4: GLU130 AND GLU44 REGULATE THE FLUX OF IRON THROUGH THE FERROXIDASE CENTRE  4.1 INTRODUCTION Ferrous iron binding by PmFTN and the rapid kinetics of initial iron oxidation are described in chapter 3 and fit a stepwise model. Sustained ferroxidase activity by PmFTN is much slower than the initial oxidation and is proposed to involve iron migration through the ferroxidase sites. The mechanism of sustained ferroxidase site turnover and transport of iron to the mineral core in PmFTN remains unknown. Kinetic studies involving sequential Fe(II) additions showed that ferric iron does not completely vacate the ferroxidase centre following oxidation, consistent with time-resolved structural studies that revealed a partial mobilization of Fe(III) from the ferroxidase centre to site C and sites further toward the central cavity. Thus, a complex iron transport mechanism may exist that likely involves site C. In the PmFTN crystal structures, Glu130 can coordinate to Fe-B and Fe-C, while Glu44 coordinates iron ion both at site C and on the inner surface of the cavity. Glu130 and Glu44 may function to shuttle metal ions between these sites. Glu44 is not conserved in ferritins, and the equivalent residue is a histidine in EcFtnA and a glutamine in PfFtn and HuHF.  To better define the function of site C in iron storage of PmFTN, variants of Glu130 (E130A) and Glu44 (E44H, E44Q) were characterized both functionally and structurally. The data reveal that Glu130 is not required for rapid Fe(II) oxidation but functions to stabilize Fe(III) at the ferroxidase centre and as a consequence reduces the rate of mineralization. Glu44 is shown to be important for regulating post-oxidation reorganization of iron coordination. Retention of  66 iron at the ferroxidase centre in wildtype and variants of PmFTN at the expense of mineralization points to a role for this ferritin in facilitating iron-sparing rather than long term iron storage. Finally, the observation of iron within the B-channels of the E44Q variant of PmFTN provides the first clear evidence that these channels, first identified in prokaryotic ferritins, are important routes for Fe(II) entry into the protein.   4.2 RESULTS 4.2.1 SUBSTITUTIONS OF GLU44 AND GLU130 DISRUPT FE(II) BINDING COOPERATIVITY BUT NOT CATALYTIC ACTIVITY To explore the role of site C in iron mineralization by diatom ferritin, Glu44 was substituted with glutamine and histidine, residues that occur naturally at this position in other characterized ferritins. Glutamine has a similar structure to Glutamate but is not charged and does not usually coordinate to iron ions. Histidine can also be non-charged, but is commonly found as a ligand for iron in metalloproteins. Stopped-flow experiments were performed to monitor the kinetics of iron oxidation after Fe(II) additions to apo-PmFTN variants. Absorption changes at 340 nm indicate the oxidation of Fe(II) to Fe(III) (Figure 4-1A and C). For both Glu44 variants, initial oxidation occurred very rapidly. In each case, data fitting required a double exponential function. A plot of observed (pseudo-first order) rate constants corresponding to the initial, rapid reaction as a function of Fe(II) concentration revealed a linear relationship demonstrating a first order dependence of the rate of oxidation on the concentration of Fe(II) (Figure 4-1B and D), as previously observed for the wild type protein. Apparent second order rate constants were derived from the slope of a best fit line. The second order rate constants for the initial fast oxidation for the two Glu44 variants revealed significant differences. E44Q  67 PmFTN has a rate constant approximately six-fold higher than that of wildtype (Table 4-1), while the rate constant for E44H is about 3-fold lower than that of wild type PmFTN (Table 4-1). However, in both variants, the ferroxidase centre is clearly catalytically functional.  Table 4-1: First and second order rate constants of Fe(II) oxidation in PmFTN variants   Wild type E44Q E44H E130A 1st phase (2nd order rate constant, M-1 s-1)  5.1 ± 0.10 × 105  2.7 ± 0.25 × 106 1.67 ± 0.19 × 105  3.14 ± 0.01 x 105   2nd phase (1st order rate constant, s-1)  0.13 ± 0.04  9 ± 1.4  0.50 ± 0.09  0.32 ± 0.07  3rd phase (1st order rate constant, s-1) N/A 0.25 ± 0.1  N/A N/A  68  Figure 4-1: Kinetic analysis of Fe(II) oxidation catalyzed by E44Q PmFTN (A, B), E44H PmFTN (C, D), and E130A PmFTN (E, F) (A, C, E) Fe(II) was added to final concentrations at increasing ratios as indicated to PmFTN variants (0.5 μM final concentration) in MES buffer (0.1 M, pH 6.5, 25 °C). Each of the ΔA340 nm traces were fitted (solid line) to a double exponential function for A and C and to a single exponential function for E. (B, D, F) Plot of observed (pseudo-first order) rate constants for the initial oxidation reaction as a function of Fe(II) concentration. A linear fit of the data, giving the second order rate constant, is drawn in. Error bars represent the standard errors and for some data points lie within the circles.  69 A plot of ΔA340 nm as a function of molar equivalent of Fe(II) added for the site C variant E44Q PmFTN (Figure 4-2A) revealed saturation of the rapid oxidation phase at a level of ~2 Fe(II) per subunit, indicating cooperativity of Fe(II) binding/oxidation, similar to that of the wild type protein. Distinct behaviour was observed for E44H PmFTN. A plot of ΔA340 nm as a function of Fe(II) shows that absorption changes are small up to a level of ~1 Fe(II) per subunit, after which they significantly increase, levelling off when all 24 subunits have bound ~2 equivalents of Fe(II), consistent with binding and oxidation of two Fe(II) ions per ferroxidase centre (Figure 4-2B). The initial shallow slope is indicative of either negative cooperativity, with binding of one Fe(II) at each ferroxidase centre favoured over double occupancy, or altered affinities of the two ferroxidase sites such that one is preferentially occupied first.  Glu130, a ligand that can coordinate iron at both sites B and C, was also substituted with the non-coordinating residue alanine. Fe(II) oxidation kinetic data for E130A PmFTN fitted well to a single exponential function (Figure 4-1E), with the same first order dependence on Fe(II) (Figure 4-1F) as observed for wild type protein and the Glu44 variants, giving an apparent second order rate constant, k, of 3.14 ± 0.01 x 105 M-1 s-1, slightly lower than that of wild type PmFTN. A plot of ΔA340 nm as a function of Fe(II) revealed a similar loss of cooperative Fe(II) binding as observed for E44H (Figure 4-2C).         70   Figure 4-2: Plots of total amplitude of absorbance changes at 340 nm at 0.5 s as a function of Fe(II) added per PmFTN variant protein (A) E44Q PmFTN. A change in absorbance can be observed after oxidation of up to ~2 Fe(II) per E44Q PmFTN. (B) E44H PmFTN. The incremental changes in absorption are small up to a level of ~1 Fe(II) per subunit, increase up to ~2 Fe(II) per subunit, and decrease again after ~2 Fe(II) per subunit. (C) E130A PmFTN. A similar absorption profile as for E44H PmFTN is observed.   71 4.2.2 ENHANCED RATE OF POST-OXIDATION REORGANIZATION IN PMFTN VARIANTS For E44H and E44Q PmFTN a second kinetic phase was clearly observed following the initial rapid oxidation and was investigated further for all the variants over a longer time period (5 s). Figures 4-3C, E, and G show kinetic traces following the addition of sub-stoichiometric, stoichiometric, and excess levels of Fe(II) to E44Q, E44H, and E130A variants. For E44H and E130A PmFTN, the data fitted well to a double exponential function, whereas for the E44Q, a tri-exponential function was required. Rate constants due to the second, slower phase were plotted as a function of Fe(II) (Figures 4-3D, F, and H). For each variant, the second phase was essentially independent of the Fe(II) concentration, indicating that this phase results from a process that occurs subsequent to the oxidation of Fe(II). Compared to E44H PmFTN, this second phase in the E130A variant is about two-fold slower (Table 4-1). Given that the initial phase was more rapid in the E130A variant, this accounts for the clearer kinetic separation of the phases in E130A compared to E44H PmFTN. The second phase rate constant of the E44Q variant was about 30-fold higher than for E130A PmFTN.  A second phase was previously observed for wild type PmFTN as it was noted that changes in absorbance occur following the completion of the initial oxidation process. In that case, because the rate of the phase was slow, only the first part was captured during the 5 s time-based acquisition, and it could not be fitted reliably. Measurements of the wild type protein over 10 s revealed this phase in more detail (Figure 4-3A). The rapid initial oxidation phase was followed by a lag (more obvious at high iron loadings). Fitting of absorbance changes that follow the initial rapid oxidation to an exponential function gave an Fe(II)-independent rate constant of k = 0.13 ± 0.04 s-1 (Figure 4-3B). This rate is slower than that observed in the three variants, especially when compared to E44Q PmFTN (~75 fold). Thus, perturbation of Fe(III)- 72 coordination following initial Fe(II) oxidation occurs more rapidly in these site C and site B/C variants. The kinetics for E44Q PmFTN revealed the presence of a third slower phase (k = 0.25 ± 0.1 s-1) that was only apparent at Fe(II) loadings above 33 Fe(II) per 24-mer (Figure 4-3C, data not shown). This third phase suggests that further re-organization occurs over a longer time period.   73   74 Figure 4-3 (previous page): Stopped-flow measurements of the second kinetic phase (A) Fe(II) was added at increasing ratios to wild type PmFTN (0.6 μM final concentration) in MES buffer (0.1 M, pH 6.5, 25 °C), and absorbance changes were followed over 10 s. Absorbance changes following initial oxidation were fitted (from 3 s, solid line) to a single exponential function. (B) Plot of the first order rate constant corresponding to the second, slower phase as a function of Fe(II) concentration. (C) Absorbance measurements over a 5 s time period following addition of Fe(II) to E44Q PmFTN (0.5 μM) at the ratios indicated. Data were fitted (solid line) to a tri-exponential function. (D) Plot of the first order rate constant corresponding to the second, slower phase as a function of Fe(II) concentration. Data from (C) and equivalent experiments at other iron loadings. (E) Absorbance measurements over a 5 s time period following addition of Fe(II) to E44H PmFTN (0.6 μM) at the ratios indicated.  Data were fitted (solid line) to a double exponential function. (F) Plot of the first order rate constant corresponding to the second, slower phase as a function of Fe(II) concentration. Data from (E) and equivalent experiments at other iron loadings. (G) Absorbance measurements over a 5 s time period following addition of Fe(II) to E130A PmFTN (0.6 μM) at the ratios indicated. Data were fitted (solid line) to a double exponential function. (H) Plot of the first order rate constant corresponding to the second, slower phase as a function of Fe(II) concentration. Data from (G) and equivalent experiments at other iron loadings.     4.2.3 E130A PMFTN EXHIBITS SIGNIFICANT REGENERATION OF THE INITIAL RAPID OXIDATION PHASE PmFTN variants were also examined to determine whether the rapid oxidation of Fe(II) observed upon addition of 48 Fe(II) per 24-mer can be regenerated. PmFTN variants were loaded with 48 Fe(II) per 24-mer and subsequently incubated for 30 min or 20 h. Then, a further aliquot of Fe(II) (40 or 80 Fe(II) per E130A or E44Q; 41 or 83 Fe(II) per E44H) was added and changes in A340 nm were measured. Figures 4-4A, 4-4B, and 4-4C show changes over 100 s after a 30 min incubation. As previously shown for the wild type protein (see Chapter 3, Figure 3-4F), rapid oxidation was not observed in the Glu44 variants. Remarkably, significant oxidation occurred in this time frame in the E130A variant, and a tri-exponential function was required to fit the data. The amplitude of the initial phase was low, however, corresponding to ~15% of that measured for apo-E130A. This indicates that ~15% of ferroxidase centres of E130A PmFTN regenerated  75 their apo-form following the first round of Fe(II) oxidation. The second phase occurred with a rate constant (~0.11 s-1) similar to the slower, iron-independent phase observed for the apoprotein. The third, slowest phase had a rate constant of k = ~0.01 s-1. Incubation over 20 h showed similar behaviour to the 30 min incubation (data not shown).         76  Figure 4-4: Stopped-flow measurements of regeneration of the rapid oxidation phase (A) Measurement of absorbance changes at 340 nm following the addition of 40 or 80 Fe(II) ions per protein to a sample of E44Q PmFTN (0.5 μM after mixing) previously treated with 48 Fe(II) per protein under aerobic conditions. The incubation time between Fe(II) additions was 30 min. (B) Measurement of absorbance changes at 340 nm following the addition of 41 or 83 Fe(II) ions per protein to a sample of E44H PmFTN (0.6 μM after mixing) previously treated with 48 Fe(II) per protein under aerobic conditions. The incubation time between Fe(II) additions was 30 min. (C) Measurement of absorbance changes at 340 nm following the addition of 40 or 80 Fe(II) ions per protein to a sample of E130A PmFTN (0.5 μM after mixing) previously treated with 48 Fe(II) per protein under aerobic conditions. The incubation time between Fe(II) additions was 30 min. The inset shows the changes over the first 0.5 s. Fits to the data are shown as solid lines.     77 4.2.4 E130A PMFTN EXHIBITS A TEN-FOLD INCREASE IN MINERALIZATION RATE COMPARED TO WILD TYPE PMFTN  Given the differences observed in the kinetics of post-Fe(II) oxidation and in the regeneration of initial rapid oxidation activity, the ability of the variants to mineralize iron was investigated. Iron core formation following addition of 400 Fe(II) ions per apo-PmFTN (wild type and variant proteins) was followed by monitoring absorbance changes at 340 nm for 1000 s (Figure 4-5), and initial rates of mineralization (i.e. post the rapid oxidation of two Fe(II) per ferroxidase centre) were calculated. Mineralization in E44Q PmFTN was similar to wild type PmFTN (initial rates of 6.95 ± 0.9 µM and 4.14 ± 0.8 µM Fe(II) min-1, respectively), whereas mineralization in E44H PmFTN was significantly slower (1.74 ± 0.5 µM Fe(II) min-1). Remarkably, mineralization in the E130A variant occurred more rapidly, with an initial rate of 40.2 ± 1.2 µM Fe(II) min-1, and was ~10-fold faster than in wild type PmFTN.   Figure 4-5: Stopped-flow spectroscopy of iron mineralization in wild type and variant PmFTN Absorbance changes at 340 nm showing Fe(II) oxidation following addition of 400 Fe(II)/PmFTN to wild type and variant PmFTNs (0.5 μM) in 0.1 MES pH 6.5, at 25 °C.   78 4.2.5 OVERALL FOLD OF VARIANT CRYSTAL STRUCTURES Crystals of each of the PmFTN variants E44Q, E44H, and E130A were grown, subsequently soaked in an aerobic solution of Fe(II), and data were collected to between 1.8 Å and 2.0 Å resolution (Table 4-2). The variant protein crystals were isomorphous with crystals of wild type PmFTN, and the space group was either P4212 or P23, with 6 or 8 monomers in the asymmetric unit, respectively. Neither the amino acid substitutions nor the metal treatment altered the overall fold or the formation of the spherical structure. The refined structures had at most 2 or 11 residues absent from the N or C terminus, respectively. Ramachandran plot analysis showed that in all structures, more than 98% of the residues were in the preferred regions. Superposition of the variant monomers with wild type PmFTN (PDB ID 4IWK) resulted in a r.m.s.d. of 0.27 Å or less for all Cα atoms.      79 Table 4-2:  Data collection and refinement statistics of PmFTN variants  E44Q Fe (5)1 E44Q Fe (45) E44Q Fe (o.n) E44H Fe (45) E44H Fe (3 h) E130A Fe (5) E130A Fe (o.n) Data collection    Resolution range (Å)     Space group    Unit cell dimensions (Å)       No. subunits in ASU    Unique reflections    Completeness (%)    Redundancy    Average I/σI    R-merge    Wilson B (Å2)    CC_Imean Refinement    R-work (R-free)    Avg. B (Å2)    No. water    No. iron    R.m.s.d. bond length    ESU from maximum     Likelihood (Å)      Ramachandran plot (%)    In most-favourable    In allowed  48.56 – 1.80  (1.90 – 1.80)2 P23 a = b = c =  175.07 8 164331 100 (100) 14.3 (13.3) 20.6 (2.9) 0.085 (0.866) 23.1 0.826  0.177 (0.212) 25.7 986 8 0.025 0.065   98.5 1.3  48.51 – 1.80 (1.90 –  1.80) P23 a = b = c = 174.92 8 163899 100 (100) 14.0 (12.4) 20.1 (2.9) 0.082 (0.868) 24.3 0.791  0.178 (0.216) 27.1 913 16 0.024 0.067   98.2 1.2  48.61 – 1.90  (2.00 – 1.90) P23 a = b = c =  175.25 8 140366 100 (100) 13.4 (12.1) 15.9 (2.5) 0.110 (0.977) 22.6 0.779  0.163 (0.200) 27.8 880 31 0.020 0.073   98.5 0.9  47.93 – 1.85  (1.95 – 1.85) P4212 a = b = 126.46 c = 170.31 6 117178 99.6 (99.2) 6.7 (6.3) 12.0 (2.1) 0.083 (0.809) 23.0 0.646  0.178 (0.218) 28.3 728 24 0.024 0.076   98.8 0.9  42.28 – 1.90 (2.00 – 1.90) P4212 a = b = 126.72 c = 170.31 6 108379 99.3 (96.7) 8.6 (4.8) 16.5 (2.5) 0.077 (0.504) 17.6 0.756  0.174 (0.211) 27.1 725 24 0.024 0.074   98.4 1.3  48.53 – 1.90  (2.00 – 1.90) P23 a = b = c =  174.98 8 139649 100 (100) 14.3 (14.0) 16.8 (2.5) 0.110 (1.124) 25.3 0.710  0.176 (0.218) 29.8 885 7 0.023 0.080   98.6 1.3  48.55 –  2.00 (2.11 –  2.00) P23 a = b = c = 175.05 8 120022 100 (100) 13.9 (13.7) 14.9 (2.4) 0.107 (1.135) 30.9 0.670  0.188 (0.229) 37.5 533 42 0.020 0.103   98.0 1.3 1Soaking time is indicated in parentheses in minutes unless otherwise indicated.  2Values in parenthesis for the data collection are for the highest resolution shell indicated.  80 4.2.6 THE FERROXIDASE CENTRE OF SITE C VARIANT E44Q Three crystal structures were obtained from PmFTN E44Q variant crystals soaked for 5 min (E44Q Fe (5)), 45 min (E44Q Fe (45)), and overnight (E44Q Fe (o.n)) in an aerobic Fe(II) solution. The structure of the ferroxidase centre of the latter is shown in Figures 4-6A and B. Iron was observed bound at sites A and B with near full occupancy (Table 4-3), whereas site C was empty. As previously reported for wild type PmFTN, Fe-A is coordinated by Glu15, Glu48, His51, and Fe-B is coordinated by Glu48, Glu94, and Glu130. Three solvent molecules are modeled as iron ligands, each of the Fe-A and Fe-B is coordinated by one solvent molecule and the third solvent is bridging Fe-A and Fe-B. The bridging atom could be an oxo/hydroxo group, forming a diferric-oxo/hydroxo bridge (183). The solvent coordinated to Fe-B forms a hydrogen bond to Gln44. Residue Glu47 is part of site C in wild type PmFTN; however, in the E44Q variant structure, the side-chain of this residue is pointing into the mineral core, and in 5 of 8 monomers, up to 2 iron ions are observed coordinating to Glu47 (Figure 4-6A). These iron binding sites are not at the same position as the binding sites observed beyond site C of wild type PmFTN and are 25 – 35% occupied. Furthermore, an additional iron ion was modeled within some of the B-channels that are formed at monomer interfaces and connect the core to the outer surface (Figure 4-7). The B-channel is lined with aspartate and glutamate residues. The iron ion is close to the inner surface and is coordinated by Glu35, weakly by Asp30, and by up to two solvent molecules modeled with occupancies of 25 – 35%. Glu35 adopts two conformations, one coordinated to the iron ion and the other pointing away from the channel.  Reducing the Fe(II) exposure time of crystals to 5 min and 45 min resulted in iron bound at site A and site B, with the same coordination sphere for Fe-A as above but with lower metal occupancy for both Fe-A and Fe-B. Site C remained unoccupied (Table 4-3). The coordination  81 sphere of Fe-B is less well ordered in the E44Q Fe (45) structure. His51 in this structure is weakly coordinated to Fe-A (ligand bond length of ~3 Å) in contrast to the Fe (o.n) structure (~2.3 Å). The side chain of Glu130 is directed away from the ferroxidase centre and is not coordinated to Fe-B. For the structure obtained from a crystal with the shortest iron exposure, no iron was bound at site B. For some monomers of these structures, two side chain conformers for Glu48 were modeled. In one conformation, Glu48 is coordinated to Fe-A, and in the other, the side chain is more strongly coordinating Fe-B (in E44Q Fe (45)) or is directed toward the vacant site B (in E44Q Fe (5)). Furthermore, in some monomers of the E44Q Fe (45) structure, the low occupancy Fe-B is weakly coordinated to Asn97, a residue in the vicinity of the ferroxidase centre but which previously has not been observed to coordinate iron. In some monomers, Asn97 adopts two conformations, one toward Fe-B and coordinating Fe-B and one toward Fe-A but not interacting with the iron ion. No iron was found within the B-channels in these structures.  Table 4-3: Range of iron ion occupancy observed in binding sites A, B, and C of aerobically Fe(II) soaked PmFTN variant crystals 1Fe(II) soaking time in parentheses in minutes unless otherwise indicated.  2The metal occupancy was determined with an estimated error of ± 10%.   E44Q Fe (5)1 E44Q Fe (45) E44Q  Fe (o.n) E44H  Fe (45) E44H  Fe (3h) E130A  Fe (5) E130A  Fe (o.n) Site A 20-25%2 45-55% 100% 55-65% 60-65% 0-35% 50-60% Site B - 20% 95-100% 85-100% 95-100% - 0-35% Site C - - - 30-40% 40-45% - 40-65% Inner surface iron ions - - 0-35% - - - 35-55%  82  Figure 4-6: Ferroxidase centres of PmFTN variants (A) Monomer B of the asymmetric unit of the E44Q Fe (o.n) structure. Iron ions are bound to sites A and B, and two iron ions are bound at the inner core surface. (B) A 2Fo-Fc electron density map contoured at 1 σ around site A and site B of the E44Q Fe (o.n) structure is shown as a gray mesh. (C) Monomer A of the asymmetric unit of the E44H Fe (3 h) structure. Iron ions are bound to site A, site B, and site C. (D) Monomer D of the asymmetric unit of the E130A Fe (o.n) structure. Iron ions are bound to sites A, B, and C. Two additional iron ions are bound past site C toward the mineral core. Iron ions are drawn as orange spheres and cyan spheres represent water molecules. Side chains of selected residues are drawn in sticks with carbon, nitrogen, and oxygen atoms in the backbone color, blue, and red, respectively. Solid lines are metal ligand bonds, and dashed lines are selected hydrogen bonds.    83  Figure 4-7: B-channel of the E44Q Fe (o.n) structure (A) B-channel built by subunits A (purple), D (blue), and E (yellow). (B) Iron bound at the B channel as viewed from the core surface toward the outside. Side chains of selected residues are drawn in sticks with carbon, and oxygen atoms in backbone colour and red, respectively. Solid lines are metal bonds. The iron ion and water molecule are shown as orange and cyan spheres, respectively. An anomalous dispersion map contoured at 3σ is shown as a grey mesh.    4.2.7 THE FERROXIDASE CENTRE OF SITE C VARIANT E44H Two structures were obtained from PmFTN E44H variant crystals soaked for 45 min (E44H Fe (45)) and 3 h (E44H Fe (3h) in aerobic Fe(II) solution. In contrast to the E44Q variant structures, all three sites are occupied by iron ions with modest difference in metal occupancy regardless of the crystal soaking time (Figure 4-6C). Site B is almost fully occupied, whereas site A ranges between 55 – 65% and Fe-C has an occupancy of 30 – 45%. However, the iron coordination spheres of the iron ions in sites A, B, and C are markedly altered in the E44H variant in comparison to the wild type PmFTN structure. Also, Fe-C is displaced toward site A by ~3.5 Å as compared to wild type PmFTN and the E44Q variant. The side chain conformation of His51 is altered such that this residue is now coordinated to Fe-C rather than Fe-A, which is thus coordinated only by Glu15, Glu48, a water molecule, and the bridging water or oxo group.  84 His44 and Glu130 are ligands of Fe-B. Fe-C is coordinated by His51 and Glu130 and forms weaker interactions with Gln126 and Glu47. A water molecule is observed bridging Fe-C and Fe-B. In both E44H structures, peaks in an anomalous map were observed at the outer surface of monomers B and E near the 2-fold symmetry axis at the interface between two PmFTN spheres. Up to three iron atoms have been modeled with 100% or 50% occupancy.  4.2.8 THE FERROXIDASE CENTRE OF SITE B/C VARIANT E130A Two structures were obtained from PmFTN E130A variant crystals soaked for 5 min (E130A Fe (5)) and overnight (E130A Fe (o.n)) in aerobic Fe(II) solution. In the E130A Fe (o.n) structure, iron was found in all three binding sites A, B, and C (Figure 4-6D). Iron occupancy at site A is 50 – 60%, whereas iron in site B was observed in only 6 of 8 monomers with an occupancy of 15 – 35%. Fe-C has an occupancy of 40 – 65% (Table 4-3). The iron coordination spheres of the Fe-A and Fe-C are unchanged, except for the absence of the Glu130 side chain. As seen in the E44Q Fe (45) structure, Asn97 is coordinated to site B in the E130A Fe (o.n) structure. Asn97 is observed coordinating Fe-B or a water molecule modelled between the ferroxidase centre sites A and B or at site B, depending on the monomer in the asymmetric unit. The electron density observed between the ferroxidase centre iron sites is not a well-defined oval shape indicative of a putative oxo-bridged diferric species. In addition to these sites, up to two iron ions with occupancies of 35 – 55% were found beyond site C toward the mineral core, as seen in wild type PmFTN. One iron ion is coordinated by Glu47 and the other by Glu44 and solvent molecules. The electron density of Glu47 is weak, suggesting greater flexibility of this residue. The structure obtained from E130A Fe (5) has iron bound solely in site A. As in the  85 E44H variants, anomalous signal was observed in maps at the interface of the ferritin spheres. However, the anomalous map was not defined well enough to model iron ions.  4.3 DISCUSSION Ferroxidase centres from bacterial and archeal ferritins are distinct from those of H-chain ferritin and BFR. In particular, like PmFTN, bacterial ferritins typically contain an additional iron binding site, site C, coordinated by glutamate residues and located 6 – 7 Å away from Fe-B. One of the glutamate residues, Glu130 (EcFtnA and PmFTN numbering), can ligate iron at either site B or site C, or bridge the iron ions at these sites (50,56,58). Site C appears to function somewhat differently depending on the specific ferritin. In EcFtnA and PfFtn, the Fe(II)/O2 ratio for the initial oxidation reaction is 3 – 4 and 2 – 3, respectively, suggesting that in these proteins, site C can participate in Fe(II) oxidation, with either H2O or a mixture of H2O and H2O2 as the final product of O2 reduction (55,101,104,107). In EcFtnA, site-directed mutagenesis of site C ligands Glu49 or Glu130 resulted in a decrease in oxidation rate and a drop in the Fe(II)/O2 stoichiometry from 3 – 4 to ~2 (55,101). Furthermore, these variants exhibited faster regeneration of the initial rapid oxidation phase, suggesting that the site C is important for controlling iron flux through the centre (101). In PfFtn, substitution of site C ligands led to complete loss of Fe(II) oxidation, indicating that the capacity of sites A and B to catalyze Fe(II) oxidation is dependent on site C (106). In PmFTN, the Fe(II):O2 ratio is ~2:1, suggesting that site C does not function as a site of Fe(II) oxidation (33). Structural and kinetic data indicate that instead site C functions as a transit site for iron from the ferroxidase centre to the central cavity. The structure of site C of PmFTN is distinct from those of bacterial and archael ferritins in that  86 the coordinating glutamate residues are not identical. To explore the function of site C in PmFTN, amino acid substitutions were made at site C (Glu44) and site B/C (Glu130).  The initial ferroxidase rate is a measure of Fe(II) binding to site A and site B followed by oxidation to diferric iron. In wild type PmFTN, this rate is exceptionally fast (Table 4-1), and none of the three variants examined here were greatly diminished in this initial rate of ferroxidation. All exhibited the same first order dependence on Fe(II) concentration with saturation of this phase at ~2 Fe(II) per ferroxidase centre. This demonstrates that rapid oxidation of Fe(II) at the ferroxidase centre is not dependent on Glu130, nor on site C in general. Structural data for E44Q and E130A are consistent in that neither site A nor site B was significantly perturbed in these variants. These sites in the E44H structure exhibited some alterations of coordinating residues to bound Fe(III), yet this variant was not more affected in its initial Fe(II) oxidation. Possibly, the E44H substitution does not alter Fe(II) binding or the subsequent catalytic oxidation. For two variants, E44H and E130A, plots of ΔA340 nm as a function of added Fe(II) were distinct from that of wild type PmFTN, indicating that Fe(II) bound to one of the ferroxidase centre sites preferentially, leading to the observation of significant Fe(II) oxidation only above one Fe(II) per subunit. This suggests either the presence of negative cooperativity or a perturbation in site A or site B substrate affinity. From the structural data, the latter is the simplest explanation as in both cases, iron at one of the sites (Fe-B in E130A and Fe-A in E44H) has fewer ligands than in the wild type structure. In terms of overall mineralization, the most striking observations were made for the site B/C variant E130A. Glu130 plays a key role in regulating the rate of mineralization in the wild type protein as substitution with a non-coordinating residue led to an unprecedented 10-fold  87 increase in mineralization activity; an unusual observation in that active site mutations rarely enhance activity. Iron was observed bound to site B and site C in E130A PmFTN, showing that Glu130 is not essential for iron binding at either site. However, the occupancy of Fe-B was significantly lower, indicating that the loss of Glu130 reduces the affinity and/or the stability of Fe(III) at site B. Note that the occupancy of site C in the overnight soak structure is not significantly different from the equivalent wild type structure.  In ferritins, such as HuHF, in which the ferroxidase centre functions as a gated site for the transfer of iron into the central cavity, the mineralization rate can be interpreted as a measure of the flux of Fe(III) through the ferroxidase centre to the inner cavity, in particular at lower iron loading. Consistent with this model, partial regeneration of the initial rapid oxidation phase was observed in E130A, with ~15% of centres in their apo-form (or in a form capable of catalyzing oxidation at the rate of apoprotein) after 30 min incubation following an initial round of Fe(II) oxidation. The half-life for oxidation of the complete second addition of Fe(II) was ~5 s (compared to >100 s for wild type PmFTN). Regeneration was not dependent on the incubation time between additions, indicating that the extent of regeneration observed here is under thermodynamic control. Structural data on wild type PmFTN showed that, as well as being a site C ligand, at longer soaking times, Glu44 also coordinates iron ions on the inner surface and so appears to play a role in guiding iron toward nucleation sites for mineralization. Substitution of Glu44 with glutamine resulted in the loss of iron binding at site C in iron-soak experiments. In this variant and after soaking overnight in Fe(II), Glu47, which is a site C ligand in wild type PmFTN, was pointing away from the site, coordinating iron on the inner surface site in a manner distinct from that observed in the wild type protein.  88 The kinetic data revealed significant differences following the initial rapid oxidation of Fe(II) at the ferroxidase centre. Immediately following the completion of oxidation, further changes in absorbance occurred which were previously interpreted as post-oxidation rearrangements of iron (see Chapter 3). For all of the variants as well as the wild type PmFTN, the rate of this subsequent phase was independent of the initial Fe(II) concentration, consistent with the interpretation of a rearrangement of or at iron sites. In the E44Q variant, this rearrangement occurs ~75-fold more rapidly than in the wild type protein (and ~30 fold more rapidly than in E130A). Thus, in the absence of a fully functional site C, post-oxidation rearrangement is significantly enhanced. The overall mineralization kinetics for E44Q showed that this enhancement does not translate into a large increase in the rate of overall core formation. Glu130, which is likely key for controlling the flux of iron through the ferroxidase centre, is still present; therefore, the rate of iron transfer remains limited. The nature of the post-oxidation rearrangement is not clear, but it appears that it is not a rearrangement of iron location because the structural data showed full occupancy of Fe-A and Fe-B in E44Q following overnight iron soak. Since absorbance at 340 nm is due to Fe(III)-O charge transfer transitions, post-oxidation absorbance changes may reflect changes in iron coordination. If iron itself is not moving, then these absorption changes are likely to be attributed to a reorganization of coordinating residues at the iron sites immediately following oxidation.  The absence of a site C and the nucleation sites beyond site C toward the mineral core would suggest limited core nucleation in E44Q PmFTN and therefore a lower rate of mineralization compared to the wild type protein. Since this is not the case in E44Q PmFTN, other nucleation sites must be involved in iron mineralization, for example the iron binding site at the inner surface that has been found in the overnight soaked E44Q structure. The fact that the  89 ferroxidase centre in E44Q Fe (o.n) is fully occupied suggests that iron movement into the core is limited. Iron oxidized after the initial oxidation at the ferroxidase centre may largely take place directly at the core surface.  Prior to structural studies of PmFTN, B-channels had only been observed in prokaryotic ferritins and bacterioferritins. The conserved polar characteristics of these channels led to the suggestion that they might be involved in transporting Fe(II) in or out of the central cavity (127). Though from a eukaryotic diatom, PmFTN possesses a cholorplast targeting sequence, and the encoding gene may originally be of prokaryotic origin. Indeed, PmFTN shares some important features with prokaryotic ferritins, such as the presence of site C and the existence of B-channels. Figure 4-8 shows the electrostatic surface potential of the E44Q Fe (o.n). Clearly, negative potential is visible at the entrance to the B-channels. The overnight soaked structure of E44Q provides the first evidence that these channels can accommodate iron ions coordinated by Glu35, Asp30, and by up to two water molecules. Interestingly, iron ions at the B-channels were not observed in the wild type protein or in any of the other variants, suggesting that iron entry through the B-channels into the core is enhanced in the E44Q variant.  A remaining question is why PmFTN has evolved to oxidise Fe(II) at its ferroxidase centres so rapidly but to form an iron mineral so slowly. The result of this is that the protein is optimized to oxidize small amounts of iron extremely rapidly and to hold it at the ferroxidase centre. This could indicate that for some photosynthetic organisms, iron sparing and iron buffering functions may be more critical than long term iron storage. For example, in the unicellular green algae Chlamydomonas reinhardtii, ferritin is not upregulated under excess iron but rather under iron limitation, where it is involved in buffering iron as it is released from photosystem I by degradation (190-192). It may be that such organisms require the ability to  90 rapidly remodel their iron proteome in response to limitation and that ferritin plays a key role in this. Pseudo-nitzschia ferritin was shown to be important for the organism’s ability to utilize transiently available iron in an otherwise iron-limited marine environment (33). Our data suggest that the protein’s ability to hold iron at ferroxidase centres, rather than to mineralize it, may be key to this role.    Figure 4-8: Electrostatic surface potential of E44Q Fe (o.n) (A) View down a 3-fold channel. One of three visible B-channels is marked with B. (B) View down a 4-fold channel. One of four visible B-channels is marked with B. Negative and positive surface potentials are colored red and blue, respectively. The contouring value of the potential is in kT/e. Surface potentials were made using the PyMol APBS tool.     91 CHAPTER 5: FERROUS IRON BINDING BY E. COLI BACTERIOFERRITIN  5.1 INTRODUCTION The first step of iron uptake by (bacterio)ferritin is the binding of ferrous iron to the di-iron centre. The binding of Fe(II) at the ferroxidase centre of most ferritins is not readily measurable using electronic spectroscopy because bound Fe(II) ions have weak or no absorption bands in the visible region. The crystal structure of PmFTN soaked in ferrous iron under anaerobic conditions was the first ferritin structure revealing the product of this first step, which is a single ferrous iron bound to ferroxidase site A. Furthermore, stopped-flow data showed that the fast initial oxidation of ferrous iron is saturated after addition of two ferrous iron ions per monomer. Together, these data suggest a stepwise initial binding of ferrous iron to the ferroxidase centre; after binding of one ferrous iron to site A, a second iron can only bind if an dioxygen is present (Chapter 3). Such a stepwise iron binding has not been reported previously nor have anaerobic crystal structures from other (bacterio)ferritins been solved.   BFRs are unique among the ferritin family in that they contain up to 12 b-type heme groups. The rapid binding of Fe(II) in bacterioferritin can be measured indirectly through an effect on the heme absorbance (102). Spectroscopy and kinetic studies support a model that Fe(II) binds pairwise to the ferroxidase centre prior to dioxygen binding and oxidation (102). The oxidized form of the ferroxidase centre of EcBFR is stable and acts as a cofactor cycling between the reduced and oxidized states throughout core formation. A third iron binding site has been located at the inner core surface with Asp50 and His46 as iron coordinating residues (67).  92 This site is proposed to facilitate electron transfer from Fe(II) in the inner cavity to the ferroxidase centre.  A crystal structure was solved of EcBFR in the presence of Fe(II) under anaerobic conditions to characterize the Fe(II) binding sites of the ferroxidase centre prior to dioxygen binding and oxidation. The crystal structure supports diferrous iron binding and a mechanism of mineralization that varies among (bacterio)ferritins.  5.2  RESULTS Recombinant EcBFR was crystallized in an anaerobic environment. An octahedral crystal was soaked for 2 h in 2 mM ferrous iron, and X-ray diffraction data were collected to 1.6 Å resolution. The crystal was of the space group P42212 with 12 EcBFR subunits in the asymmetric unit (Table 5-1). The refined structure is nearly complete with one glycine residue missing from the C terminus of each monomer. A Ramachandran plot showed that 95.9% of the residues are in the most favoured regions and no residues are in the disallowed regions. As reported previously for EcBFR in the presence of dioxygen, the anaerobic EcBFR forms the typical ferritin four helix bundle with a shorter fifth helix at the C-terminus (67,193,194). The monomers assemble into a 24-mer to form the ferritin sphere (Figure 5-1A). The anaerobic EcBFR structure was superimposed with the highest resolution (2.4 Å) aerobic structure, solved by Crow et al.,  derived from a Zn(II) treated crystal soaked in Fe(II) (67). Superposition of the monomers of these structures resulted in a 0.12 Å r.m.s.d. for all Cα atoms. Two Fe(II) ions are observed in ferroxidase centre sites A and B with an occupancy of 100% and 90 – 95%, respectively (Figure 5-2A). The average inter-iron distance is 3.95 ± 0.01 Å (Table 5-2). The coordination of the two ferroxidase centre iron ions is pseudo 2-fold  93 symmetrical with Glu51 and Glu127 as bridging ligands between Fe-A and Fe-B. Fe-A is further coordinated by His54 and Glu18, and Fe-B is coordinated by Glu94 and His130. In addition, Fe-B is coordinated by a water molecule. However, the electron density for the water molecule is poorly defined, and positive density in a difference map suggests that there may be additional coordinating solvent molecules present. Furthermore, the electron density clearly does not show any bridging density as seen for an oxo bridge (Figure 5-2B) (67).  Asp50 and His46 were reported to bind iron in aerobically grown crystals (67); however, in the anaerobic EcBFR structure, this inner surface site is empty (Figure 5-2A). A water molecule is H-bonded to Asp50, and in most subunits, His46 adopts multiple conformations; one pointing toward the binding site and one pointing away from the site (Figure 5-3B).  Heme was modeled between each EcBFR dimer with Met52 from each monomer coordinated to the iron ion (Figure 5-1B). An anomalous dispersion map and a (Fo – Fc) difference map were used to determine the heme occupancy of 25 – 30%. The B-factors of the iron ions are similar to those of the coordinated methionine residues. At this low occupancy, the electron density of the porphyrin ring is poorly defined. Residual positive density in a difference map suggests that solvent molecules may occupy the interface between the two monomers when heme is not present.         94 Table 5-1: Data collection and refinement statistics of EcBFR                   1Values in parenthesis for the data collection are for the highest resolution shell indicated.     Anaerobic EcBFR Fe (2 h) Data collection    Resolution range (Å)     Space group    Unit cell dimensions (Å)       No. subunits in ASU    Unique reflections    Completeness (%)    Multiplicity    Average I/σI    R-merge    Wilson B (Å2)    CC_Imean Refinement    R-work (R-free)    Avg. B (Å2)    No. water    No. iron (non-heme)    No. heme    No. sulfate    R.m.s.d. bond length (Å)    ESU from maximum     likelihood (Å)  Ramachandran plot (%)    In most-favourable    In allowed  39.16 – 1.60 (1.69 – 1.60)1 P42212 a = b = 207.64 c = 142.50 12 392619 97.4 (83.9) 8.3 (3.6) 17.7 (3.7) 0.060 (0.251) 8.0 0.999 (0.970)  0.155 (0.178) 11.4 2395 24 6 4 0.025 0.036   95.9 4.1  95  Figure 5-1: Crystal structure of EcBFR derived from a crystal soaked for 2 h in ferrous iron under anaerobic conditions (A) Crystal structure derived from an anaerobic Fe(II) soaked crystal of recombinant EcBFR shows the typical spherical 24-mer arrangement. (B) Two monomers are shown that form the typical four helix bundles. A heme group is shown between the two monomers. Orange spheres are iron ions of the ferroxidase centres. In stick are the heme and ferroxidase centre iron coordinating residues.        96  Figure 5-2: Ferroxidase centre of EcBFR crystal structure derived from a crystal soaked for 2 h in ferrous iron under anaerobic conditions Monomer A of the asymmetric unit is shown. Iron ions are drawn as orange spheres and cyan spheres represent water molecules. Side chains of selected residues are drawn in sticks with carbon, nitrogen, and oxygen atoms in the backbone color, blue, and red, respectively. Solid lines are metal ligand bonds, and dashed lines are hydrogen bonds. (A) Two ferrous iron ions are bound to the ferroxidase centre sites A and B. The coordination of the iron ions is pseudo 2-fold symmetrical. In addition Fe-A coordinates to a water molecule. (B) A 2Fo – Fc electron density map contoured at 1 σ around site A and site B and the water molecule.             97 Table 5-2: Average bond lengths in Å between iron ions and ligands of anaerobic and aerobic EcBFR structures    1Ferrous iron soaking time is indicated in parentheses in minutes unless otherwise indicated. 2Bond lengths were measured with the help of Amelia Hardjasa.    Figure 5-3: Superposition of anaerobic and aerobic EcBFR structures (A) Superposition of the ferroxidase centre of the anaerobic EcBFR structure (green) and the aerobic EcBFR Fe (65) structure (pink). Orange and red spheres are iron ions in the anaerobic and aerobic structure, respectively. Water molecules are drawn as cyan (anaerobic EcBFR) and blue (EcBFR Fe (65)) spheres. (B) Superposition of the inner surface site of the anaerobic EcBFR structure (green) and the aerobic EcBFR Fe (2.5) structure (pink). An iron ion is drawn as an orange sphere; a water molecule is drawn as a cyan sphere.   Metal ligand bond Anaerobic EcBFR (2 h)1 EcBFR Fe (2.5) EcBFR Fe (65) Glu18 OE1 – Fe-A 2.152 ± 0.03 2.15 ± 0.02 2.02 ± 0.02 Glu18 OE2 –  Fe-A 2.21 ± 0.02 2.42 ± 0.02 2.52 ± 0.03 His54 ND1 – Fe-A 2.17 ± 0.02 2.32 ± 0.02 2.39 ± 0.03 Glu51 OE2 – Fe-A 2.01 ± 0.03 2.18 ± 0.02 2.04 ± 0.03 Glu127 OE1 – Fe-A 2.00 ± 0.02 2.05 ± 0.02 1.91 ± 0.02 Glu127 OE2 – Fe-B 1.97 ± 0.02 1.95 ± 0.02 2.00 ± 0.03 Glu51 OE1 – Fe-B 2.02 ± 0.02 2.08 ± 0.02 2.01 ± 0.02 His130 ND1 – Fe-B 2.16 ± 0.02 2.33 ± 0.01 2.41 ± 0.02 Glu94 OE1 – Fe-B 2.41 ± 0.04 2.48 ± 0.02 2.35 ± 0.02 Glu94 OE2 – Fe-B 2.08 ± 0.02 2.27 ± 0.02 2.20 ± 0.03 H2O – Fe-A 2.38 ± 0.07 N/A 2.87 ± 0.27 H2O – Fe-B N/A N/A 2.84 ± 0.13 Fe-A – Fe-B 3.95 ± 0.01 3.69 ± 0.03 3.59 ± 0.03  98 5.3 DISCUSSION The structure derived from an anaerobic grown and Fe(II) soaked EcBFR crystal showed two Fe(II) ions bound to the ferroxidase centre sites A and B. This structure is in agreement with kinetic and spectroscopic data that fit a model of pairwise Fe(II) binding to the ferroxidase centre of EcBFR prior to dioxygen binding (102). The inner surface site is not occupied with Fe(II), suggesting that the initial Fe(II) binding occurs at sites A and B only. The inner surface site is proposed to be a Fe(II)/Fe(III) binding site with a role of facilitating electron transfer from Fe(II) ions in the central cavity to the ferroxidase centre (67). The absence of Fe(II) at this site in the anaerobic EcBFR structure suggests that the inner surface site is not involved in the initial fast Fe(II) oxidation but may have a role in subsequent iron mineralization as previously suggested (67). Oxidative turnover may be necessary for the inner surface site to be populated.  Stopped-flow kinetics showed that Fe(II) binding to the ferroxidase centre in EcBFR occurs much faster than the subsequent iron oxidation, which is the rate-limiting step. In contrast, the PmFTN Fe(II) oxidation rate is similar to the Fe(II) binding rate of EcBFR, and iron binding has been proposed to be the rate-limiting step in PmFTN (Chapter 3). In particular, the presence of a stepwise iron binding suggested the binding of Fe(II) to site B to be rate-determining. The absence of a stepwise binding in favour of a pairwise Fe(II) binding prior to dioxygen binding in EcBFR is supported by the different initial ferroxidase reaction rates of these two proteins of the ferritin family.  Aerobic crystal structures of EcBFR were previously solved (193,194) and allow comparison of the ferroxidase centre in the reduced and oxidized states. Two aerobic Fe(II) soaked structures, 2.5 min (EcBFR Fe (2.5)) and 65 min (EcBFR Fe (65)), were solved by Crow et al. (67). In both structures, the ferroxidase centre is fully occupied by two iron ions. A  99 bridging small ligand such as water or oxygen is modeled between sites A and B of the ferroxidase centre of EcBFR Fe (65), suggesting the formation of a diferric oxo-bridge. Due to the lack of such a bridging ligand observed in EcBFR Fe (2.5), Crow et al. suggested that the iron ions present are in the Fe(II) state, whereas the iron ions in EcBFR Fe (65) are in the Fe(III) state. Superposition of the ferroxidase centre of the anaerobic EcBFR structure with that of the ferroxidase centre of EcBFR Fe (65) (PDB ID 3E1N) showed that the ligand geometry of iron in the ferrous versus ferric state (Figure 5-3A) are similar. The average bond lengths to specific ligands do vary modestly upon iron oxidation (Table 5-2). The inter-iron distance decreases from ~3.95 Å in the anaerobic structure to ~3.59 Å in the diferric EcBFR Fe (65) structure. However, the inter-iron distance in EcBFR Fe (2.5) is in between the distances of the oxidized and fully reduced form, suggesting that the iron in EcBFR Fe (2.5) is in a mixed state. In BfMF, the shortening of the inter-iron distance upon iron oxidation was inferred from a comparison of the diferric and dicupric forms (183).  Histidine preferably coordinates to Fe(II) rather than Fe(III). This is reflected in the metal ligand bond lengths of the different EcBFR structures (Table 5-2). The bond length between His54 and Fe-A increased from ~2.17 Å to ~2.39 Å upon iron oxidation. Similarly, the bond length between His130A and Fe-B increased from ~2.16 Å to ~2.41 Å. However, His54 and His130 do not significantly move upon iron oxidation. Instead, the iron ions move toward the water molecule, thereby shortening the inter-iron distance and lengthening the bonds with the histidine residues. The coordinating glutamates do slightly move with the iron ions (Figure 5-3A).  Iron at the inner surface site is observed in the EcBFR Fe (2.5) structure; however, the occupancy is partial (~40%). The oxidation state of the iron in this site is not known (67). In  100 contrast, no iron was found in the inner surface site of either the anaerobic structure or EcBFR Fe (65), suggesting that this iron bound transiently. Zn(II) was shown to be a potent inhibitor of BFR due to its capacity to tightly bind at the ferroxidase centre (103,182,195). Crow et al. showed that soaking an apo-EcBFR crystal in Zn(II) leads to a fully metal occupied ferroxidase centre but to no occupancy of the inner surface site. However, soaking of Zn-treated crystals in Fe(II) did lead to the observation of fully occupied iron binding at the inner surface site. Thus, the presence of Zn(II) at the ferroxidase centre favoured Fe(II) binding at the inner surface site. The absence of Fe(II) at the surface site of the anaerobic soaked structure, despite nearly fully occupied ferroxidase centre sites, supports the hypothesis that iron binding at the surface site requires the oxidative turnover at the ferroxidase centre. Moreover, Zn(II) soaked crystals of PmFTN revealed that Zn(II) binding does not correspond to Fe(II) binding. In EcBFR, Zn(II) in the ferroxidase centre may resemble Fe(III) coordination in terms of its effect on occupancy of the inner surface site. A superposition of the anaerobic inner surface site with the iron occupied surface site of the 2.5 min aerobic structure revealed no significant changes in the position of Asp50 and the iron binding conformation of His46 (Figure 5-3B). However, the water molecule coordinated to Asp50 in the anaerobic structure is not in the same position as the iron ion.  In conclusion, the anaerobic EcBFR structure supports the previously proposed pairwise Fe(II) binding prior to oxygen binding and oxidation. The inner surface site is not involved in the initial iron oxidation. However, the occupancy of the inner surface site is directly correlated with the status of the ferroxidase centre, and oxidative iron turnover is necessary for this site to be populated. Furthermore, the comparison of EcBFR with PmFTN revealed differences in ferrous iron binding and reaction rates and is further evidence that structural differences between the ferroxidase centres of BFRs and ferritins result in large functional differences.  101 CHAPTER 6: OVERVIEW AND FUTURE DIRECTIONS  6.1 PMFTN HAS CHARACTERISTICS OF BACTERIAL FERRITINS  The mechanisms of iron oxidation by ferritins from several organisms have been studied, including those from EcFtnA and HuHF (55,93-96,101). A general model is proposed for ferritins, where the ferroxidase centre functions as a substrate site, and the oxidized ferric iron leaves the ferroxidase centre to form the mineral core (88-90). However, despite the highly conserved overall fold and assembly of ferritins, detailed mechanistic studies revealed differences between ferritins from a variety of organisms. These include differences in the structure of the ferroxidase centre, iron oxidation stoichiometry, and the presence of a third iron binding site (site C) in prokaryotic ferritins. Many ferritins from eukaryotes have been studied, including human, fish, and plant ferritins; however, ferritin from the stramenopile, PmFTN, has several unique characteristics. In Chapter 3, wild type PmFTN is characterized both structurally and functionally. An initial crystal structure of wild type PmFTN revealed three iron binding sites as seen in prokaryotic ferritins (33). However, site C is unique in that the iron ion is coordinated only by residue Glu44 (Figure 1-3D). The crystal structures reported in Chapter 3 were derived from crystals soaked for various durations in ferrous iron and zinc sulfate. In these structures, Fe-C is also coordinated by Glu47 and Glu130 in addition to Glu44. An updated figure of the ferroxidase centre and site C of PmFTN is shown in Figure 6-1.    102  Figure 6-1: Ferroxidase centre and site C of PmFTN Metal ligand bonds are drawn as solid lines. Hydrogen bonds are drawn as dotted lines.    Mechanistic studies of ferritin have been mainly focused on the initial fast iron oxidation phase due to technical limitations of measuring the first step of ferrous iron binding to the ferroxidase centre. A crystal structure derived from an anaerobically grown and ferrous iron soaked crystal revealed a single ferrous iron in site A of the ferroxidase centre of PmFTN. This result together with the observation that the rate of Fe(II) oxidation exhibited a first order dependence on iron concentration suggested a model in which Fe(II) binds stepwise in a dioxygen dependent manner, and the binding of the second iron is the trigger for oxidation to occur (Figure 6-2). Kinetic studies involving sequential Fe(II) additions showed that, upon oxidation, ferric iron does not completely vacate the ferroxidase centre. Structural studies from crystals soaked for various durations revealed a partial mobilization of Fe(III) from the ferroxidase centre to site C and sites further toward the central cavity.   103  Figure 6-2: Stepwise iron and dioxygen binding to the ferroxidase centre of PmFTN (A) Fe(II) binds to site A followed by the binding of dioxygen (B). (C) Only when the dioxygen is bound can the second Fe(II) bind to site B. (D) Fe(II) is oxidized and forms an oxo-diferric intermediate.   Chapter 4 describes the characterization of the PmFTN variants E44Q, E44H, and E130A. Both of the targeted residues, Glu130 and Glu44, are proposed to be important for the iron transport from site B to site C and onward to the mineral core. Kinetic and crystallographic studies revealed that Glu130 is important for retaining Fe(III) at the ferroxidase centre and thereby reducing the rate of iron mineralization. The function of the residue Glu44 is less apparent; however, Glu44 may be important for regulating post-oxidation reorganization of iron coordination. These studies gave insight into the iron oxidation mechanism and role of amino acid residues involved. It can be concluded that the ferroxidase centre of PmFTN functions as a substrate site as seen in other ferritins rather than a cofactor. However, the mechanism of ferroxidase centre turnover and transport of iron to the mineral core in PmFTN is not fully described.  Eukaryotic ferritins do not typically have a third iron binding site C, which is unique to prokaryotic ferritins. However, a phylogenetic analysis suggested that diatom ferritins are more  104 closely related to prokaryotic than eukaryotic ferritins (33). In fact, PmFTN has a chloroplast targeting sequence and the encoding gene may have arisen through a lateral gene transfer and then diverged to the needs of diatoms. PmFTN can be proposed to be a member of the prokaryotic ferritin family; however, there are structural and functional differences between PmFTN and the well-studied bacterial ferritin EcFtnA. Site C in PmFTN is not essential for ferroxidase activity and does not serve as an additional site for initial Fe(II) oxidation. In contrast, the site C found in EcFtnA is, although not essential for ferroxidase activity, a third ferrous iron binding site (55,101). Furthermore, the ligands of Fe-C from EcFtnA are different in origin and number, and site C of PmFTN is found to be closer to the ferroxidase centre sites. Moreover, site C found in the archeal ferritin PfFtn is identical to site C of EcFtnA with respect to position and ligand residues. However; substitution of site C ligands led to complete loss of Fe(II) oxidation, indicating that the capacity of sites A and B to catalyze Fe(II) oxidation is dependent on site C (106).  In H-chain ferritin, Fe(II):O2 stoichiometry has been shown to be ~2:1 at low iron loading and increases to ~4:1 at high iron loading (96). A similar increase in stoichiometry has been observed for EcFtnA. At low iron loading (equal or below 2 Fe per subunit) a Fe:O2 stoichiometry of ~3:1 has been observed which can be explained by the presence of site C. However, increasing iron concentrations, increased the Fe(II):O2 ratio to ~4:1 as seen in HuHF. This increased stoichiometry indicates that more than one reaction is occurring and can be ascribed to an increasing fraction of the iron being oxidized by H2O2 as well as increasing involvement of the mineral surface reaction (55). An oxometric assay of PmFTN showed that the initial ferroxidase reaction consumes 1.9 ± 0.2 Fe(II):O2, identical to the value generally determined for HuHF and horse spleen ferritin (33). Further studies are necessary to clarify if the  105 stoichiometry in PmFTN increases upon iron loading and if a similar mechanism of autocatalysis at the core surface, as proposed for H-chain ferritin and EcFtnA, also exists in PmFTN. Furthermore, the consumption of H2O2 by PmFTN has to be proven.  Both PmFTN and EcFtnA do not show regeneration of the ferroxidase activity after addition of a minimum of 40 Fe(II)/24-mer to pre-loaded ferritin after 20h. However, site C EcFtnA variants E49A, E126A, and E130A show regeneration of most of their original ferroxidase activity in a few hours as seen in HuHF (55,101). Similarly, the E130A variant of PmFTN shows 15% recovery of the ferroxidase activity after 30 min. Although a lower recovery was observed for PmFTN, both ferritins seem to retain iron at the ferroxidase centre. This function is proposed to be due to the presence of site C in EcFtnA; however, Glu44 variants of PmFTN do not show any regeneration of the ferroxidase activity. Iron retained at the ferroxidase centre may be more readily available to the cell than core iron during iron mobilization.  Although Honarmand et al. suggested a common iron oxidation and storage mechanism for ferritins, with the data now available, this is difficult to justify (106). H-chain ferritin and bacterial ferritin do function in a similar way; however, the presence of a site C in both bacterial ferritin and diatom ferritin result in the iron mineralization mechanism being more complex. It is more likely that ferritins in different organisms have evolved in a way that they fine-tuned their iron storage mechanism to meet the organism’s needs and adapt to their environment.  6.2 FERROUS IRON BINDING BY ECBFR DIFFERS FROM THAT BY PMFTN Chapter 5 focuses on the bacterioferritin subfamiliy and in particular the anaerobic crystal structure of EcBFR. Two ferrous iron ions are found in the ferroxidase centre sites A and B, in agreement with the proposed mechanism of pairwise ferrous iron binding prior to oxygen  106 binding and subsequent iron oxidation. The inner surface site is not occupied with ferrous iron; therefore, it is not involved in the initial fast iron oxidation. It is suggested that oxidative turnover is necessary for the inner surface site to be populated. Zn(II) occupancy of the ferroxidase centre in a structure derived from an aerobic zinc soaked crystal does not lead to Zn(II) binding at the inner surface site but does allow Fe(II) binding upon soaking in Fe(II) (67). However, further studies are necessary to answer the question of ferrous iron binding at the inner surface site.  The comparison of EcBFR with PmFTN reveals differences in ferrous iron binding and reaction rates and is further evidence that the ferroxidase centre of PmFTN does not function as a cofactor as proposed for EcBFR. However, anaerobic crystal structures have not been solved for ferritins and BFRs from other species, and it remains to be investigated if a stepwise iron binding is unique to PmFTN.   6.3 FUTURE DIRECTIONS Although studies on ferritin and bacterioferritin have been performed for many years, there are still many unanswered questions about the function of these ferritin family proteins. Studies about Fe(II) oxidation have been performed and give good insight into the mechanism of iron oxidation and storage in various ferritins and BFRs. However, the detailed mechanism still remains to be investigated. Research has mainly focused on the initial Fe(II) oxidation, the formation of intermediates during the ferroxidase centre reaction, the involvement of a third iron binding site, and the iron entry pathways into H-chain ferritin. However, little is known about iron entry pathways of prokaryotic ferritins and BFRs. Even less well studied is how iron stored as a mineral is reduced, and the exit path iron ions take when released from the mineral core.  107 Furthermore, the physiological role of (bacterio)ferritins in microorganisms, and their individual role when they co-exist in the same cell remains to be investigated.   6.3.1 MUTAGENESIS STUDIES Further mutagenesis studies are necessary to understand the iron oxidation and storage mechanism of PmFTN in more detail. In the suggested sequential displacement mechanisms for HuHF and PfFtn, a highly conserved tyrosine within 5 Å of site B completes the reduction of dioxygen to water after oxidation of two Fe(II) at the ferroxidase centre and one Fe(II) at the gateway site (107), thereby preventing the formation of reactive oxygen species. This tyrosine is conserved in all studied ferritins and BFRs. EPR studies showed the occurrence of a tyrosine radical in HuHF, horse spleen ferritin, and EcFtnA (55,196). However, studies of EcFtnA variants showed that fully functional A and B sites, but not the C site, are required for its generation (55). Furthermore, Tyr24 in EcFtnA is not required for rapid iron oxidation. In wild type PmFTN, Tyr22 is ~4.3 Å from Fe-B, whereas Tyr24 of EcFtnA is only ~2.5Å away from Fe-B. Substitution of Tyr22 would give insight if this residue is required for the rapid iron oxidation in PmFTN.  Asn97 is weakly coordinated to Fe-B in the PmFTN E44Q Fe (45) structure. Substitution of Asn97 would give information as to the role of this residue in the function of PmFTN. As it is located just behind the ferroxidase centre, a function in guiding iron to the centre might be possible; however no iron binding and no movement of this residue was observed in wild type PmFTN. Furthermore, an asparagine can be found at this position in EcFtnA; however, in other characterized ferritins, this residue is a lysine.  108 How iron is transported from the ferroxidase centre to the mineral core and what residues are involved in the ferroxidase centre turnover is still not fully explained. Further substitution of ferroxidase centre sites and site C ligands to alanine might give more detailed information about their function in iron oxidation and transport and also about the nature of the post-oxidation rearrangement.   6.3.2 DETECTION OF THE DIFERRIC PEROXO INTERMEDIATE IN WILD TYPE AND E44Q PMFTN In H-chain ferritin, the two Fe(II) bound to the ferroxidase centre are oxidized by dioxygen to form an unstable µ-1,2-peroxodiferric intermediate, which decays to a more stable µ-1,2-oxodiferric species. The peroxo species could not be detected in BFR; however, it is possible that this intermediate is too short lived or its absorption is too weak to detect. Interestingly, preliminary results by Dr. N. E. Le Brun and his lab at the University of East Anglia showed the occurrence of such a diferric peroxo intermediate in PmFTN (Figure 6-3; unpublished data). Absorption was measured at 650 nm upon mixing of 4.17 µM PmFTN with 48 equivalents Fe(II) at 25 °C and 12 °C for E44Q and wild type PmFTN, respectively. The formation of the peroxo species could not be observed for wild type PmFTN at 25°C, presumably since the formation and decay of the intermediate occurs too rapidly as suggested for BFR. This absorption experiment has not been performed for other variants (E44H, E130A). Thus, further experiments will involve measuring the absorption at 650 nm after Fe(II) addition to the variants E44H and E130A PmFTN. Also, whole spectrum rapid kinetic experiments would provide more information on how the kinetics of the formation of the peroxo intermediate relates to kinetics observed at 340 nm. Furthermore, the temperature dependence of the formation of the peroxo intermediate will give insight in to the energetics of the reaction.    109  Figure 6-3: Detection of the diferric peroxo intermediate  (A) Time dependence of the absorbance at 650 nm upon mixing of wild type PmFTN with 48 equivalents ferrous iron, showing formation of a transient diferric peroxo species. (B) Time dependence of the absorbance at 650 nm upon mixing of 4.17 µM E44Q PmFTN (final concentration) with 48 equivalents (200 µM) ferrous iron, showing formation of a transient diferric peroxo species. The concentration used is 4 times greater than for the Fe(II) oxidation experiments to ensure reasonable signal to noise.   6.3.3 IN VIVO IRON STORAGE PROPERTY OF PMFTN  Ferritin in pennate diatoms probably contributes to their success in chronically low-iron regions that receive intermittent iron inputs. Measuring the iron content of PmFTN isolated from P. multiseries grown under iron replete and limiting conditions would show if PmFTN can store significant quantities of iron in vivo. Antisera to recombinant PmFTN was raised and can be used for affinity purification. Using immunoaffinity chromatography, a native ferritin sample could be isolated. The iron content of isolated PmFTN can be assayed by inductively-coupled-plasma mass-spectrometry (ICP-MS). Iron content of ferritin is expected to be low in cultures grown under iron limitation and increases upon iron fertilization. High iron content (>1000 Fe ions per 24-mer) will prove that PmFTN is an iron storage protein in P. multiseries.   110 BIBLIOGRAPHY 1. Crichton, R. R., and Pierre, J. L. (2001) Old iron, young copper: from Mars to Venus. Biometals 14, 99-112 2. Crichton, R. R., and Ward, R. J. (1992) Iron metabolism--new perspectives in view. Biochemistry 31, 11255-11264 3. Beinert, H., Holm, R. H., and Munck, E. (1997) Iron-sulfur clusters: nature's modular, multipurpose structures. Science 277, 653-659 4. Lange, S. J., and Que, L., Jr. (1998) Oxygen activating nonheme iron enzymes. Curr. Opin. Chem. Biol. 2, 159-172 5. Williams, R. J. P., and Da Silva, J. J. R. F. (2003) Evolution was chemically constrained. J. Theor. Biol. 220, 323-343 6. Andrews, S. C., Robinson, A. K., and Rodriguez-Quinones, F. (2003) Bacterial iron homeostasis. FEMS Microbiol. Rev. 27, 215-237 7. Braun, V., and Killmann, H. (1999) Bacterial solutions to the iron-supply problem. Trends Biochem. Sci. 24, 104-109 8. Gutteridge, J. M., Rowley, D. A., and Halliwell, B. (1981) Superoxide-dependent formation of hydroxyl radicals in the presence of iron salts. Detection of 'free' iron in biological systems by using bleomycin-dependent degradation of DNA. Biochem. J. 199, 263-265 9. Wardman, P., and Candeias, L. P. (1996) Fenton chemistry: an introduction. Radiat. Res. 145, 523-531 10. Bou-Abdallah, F. (2010) The iron redox and hydrolysis chemistry of the ferritins. Biochim. Biophys. Acta 1800, 719-731 11. Andrews, S. C. (2010) The Ferritin-like superfamily: Evolution of the biological iron storeman from a rubrerythrin-like ancestor. Biochim. Biophys. Acta 1800, 691-705 12. Nair, S., and Finkel, S. E. (2004) Dps protects cells against multiple stresses during stationary phase. J. Bacteriol. 186, 4192-4198 13. Le Brun, N. E., Crow, A., Murphy, M. E., Mauk, A. G., and Moore, G. R. (2010) Iron core mineralisation in prokaryotic ferritins. Biochim. Biophys. Acta 1800, 732-744 14. Lewin, A., Moore, G. R., and Le Brun, N. E. (2005) Formation of protein-coated iron minerals. Dalton T., 3597-3610 15. Crichton, R. R., and Declercq, J. P. (2010) X-ray structures of ferritins and related proteins. Biochim. Biophys. Acta 1800, 706-718 16. Abdul-Tehrani, H., Hudson, A. J., Chang, Y. S., Timms, A. R., Hawkins, C., Williams, J. M., Harrison, P. M., Guest, J. R., and Andrews, S. C. (1999) Ferritin mutants of Escherichia coli are iron deficient and growth impaired, and fur mutants are iron deficient. J. Bacteriol. 181, 1415-1428 17. Bou-Abdallah, F., Lewin, A. C., Le Brun, N. E., Moore, G. R., and Chasteen, N. D. (2002) Iron detoxification properties of Escherichia coli bacterioferritin. Attenuation of oxyradical chemistry. J. Biol. Chem. 277, 37064-37069 18. Velayudhan, J., Castor, M., Richardson, A., Main-Hester, K. L., and Fang, F. C. (2007) The role of ferritins in the physiology of Salmonella enterica sv. Typhimurium: a unique role for ferritin B in iron-sulphur cluster repair and virulence. Mol. Microbiol. 63, 1495-1507  111 19. Smith, J. L. (2004) The physiological role of ferritin-like compounds in bacteria. Crit. Rev. Microbiol. 30, 173-185 20. Sievers, F., Wilm, A., Dineen, D., Gibson, T. J., Karplus, K., Li, W. Z., Lopez, R., McWilliam, H., Remmert, M., Soding, J., Thompson, J. D., and Higgins, D. G. (2011) Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7 21. Nicholas, K. B., Nicholas, H. B. Jr., and Deerfield, D. W. II. (1997) GeneDoc: Analysis and Visualization of Genetic Variation. EMBNEW.NEWS  22. Hempstead, P. D., Yewdall, S. J., Fernie, A. R., Lawson, D. M., Artymiuk, P. J., Rice, D. W., Ford, G. C., and Harrison, P. M. (1997) Comparison of the three-dimensional structures of recombinant human H and horse L ferritins at high resolution. J. Mol. Biol. 268, 424-448 23. Boyd, D., Vecoli, C., Belcher, D. M., Jain, S. K., and Drysdale, J. W. (1985) Structural and functional relationships of human ferritin H and L chains deduced from cDNA clones. J. Biol. Chem. 260, 11755-11761 24. Crichton, R. R., Herbas, A., ChavezAlba, O., and Roland, F. (1996) Identification of catalytic residues involved in iron uptake by L-chain ferritins. J. Biol. Inorg. Chem. 1, 567-574 25. Chasteen, N. D., and Harrison, P. M. (1999) Mineralization in ferritin: an efficient means of iron storage. J. Struct. Biol. 126, 182-194 26. Santambrogio, P., Levi, S., Cozzi, A., Corsi, B., and Arosio, P. (1996) Evidence that the specificity of iron incorporation into homopolymers of human ferritin L- and H-chains is conferred by the nucleation and ferroxidase centres. Biochem. J. 314 ( Pt 1), 139-144 27. Levi, S., Yewdall, S. J., Harrison, P. M., Santambrogio, P., Cozzi, A., Rovida, E., Albertini, A., and Arosio, P. (1992) Evidence That H-Chains and L-Chains Have Cooperative Roles in the Iron-Uptake Mechanism of Human Ferritin. Biochem. J. 288, 591-596 28. Zhao, G., Su, M., and Chasteen, N. D. (2005) mu-1,2-peroxo diferric complex formation in horse spleen ferritin. A mixed H/L-subunit heteropolymer. J. Mol. Biol. 352, 467-477 29. Scudiero, R., Trinchella, F., and Parisi, E. (2008) Iron metabolism genes in Antarctic notothenioids: A review. Mar. Genomics 1, 79-85 30. Ha, Y., Shi, D., Small, G. W., Theil, E. C., and Allewell, N. M. (1999) Crystal structure of bullfrog M ferritin at 2.8 A resolution: analysis of subunit interactions and the binuclear metal center. J. Biol. Inorg. Chem. 4, 243-256 31. Giorgi, A., Mignogna, G., Bellapadrona, G., Gattoni, M., Chiaraluce, R., Consalvi, V., Chiancone, E., and Stefanini, S. (2008) The unusual co-assembly of H- and M-chains in the ferritin molecule from the Antarctic teleosts Trematomus bernacchii and Trematomus newnesi. Arch. Biochem. Biophys. 478, 69-74 32. Lawson, D. M., Artymiuk, P. J., Yewdall, S. J., Smith, J. M., Livingstone, J. C., Treffry, A., Luzzago, A., Levi, S., Arosio, P., Cesareni, G., and et al. (1991) Solving the structure of human H ferritin by genetically engineering intermolecular crystal contacts. Nature 349, 541-544 33. Marchetti, A., Parker, M.S., Moccia, L.P., Lin, E.O., Arrieta, A.L., Ribalet, F., Murphy M.E.P., Maldonado, M.T., Armbrust E.V. (2009) Ferritin used for iron storage in bloom-forming marine pennate diatoms. Nature 457, 467-470  112 34. Levi, S., Corsi, B., Bosisio, M., Invernizzi, R., Volz, A., Sanford, D., Arosio, P., and Drysdale, J. (2001) A human mitochondrial ferritin encoded by an intronless gene. J. Biol. Chem. 276, 24437-24440 35. Drysdale, J., Arosio, P., Invernizzi, R., Cazzola, M., Volz, A., Corsi, B., Biasiotto, G., and Levi, S. (2002) Mitochondrial ferritin: a new player in iron metabolism. Blood Cells Mol. Dis. 29, 376-383 36. Langlois d'Estaintot, B., Santambrogio, P., Granier, T., Gallois, B., Chevalier, J. M., Precigoux, G., Levi, S., and Arosio, P. (2004) Crystal structure and biochemical properties of the human mitochondrial ferritin and its mutant Ser144Ala. J. Mol. Biol. 340, 277-293 37. Levi, S., and Arosio, P. (2004) Mitochondrial ferritin. Int. J. Biochem. Cell Biol. 36, 1887-1889 38. Lill, R., and Kispal, G. (2000) Maturation of cellular Fe-S proteins: an essential function of mitochondria. Trends Biochem. Sci. 25, 352-356 39. Liao, X., Yun, S., and Zhao, G. (2014) Structure, function, and nutrition of phytoferritin: a newly functional factor for iron supplement. Crit. Rev. Food Sci. Nutr. 54, 1342-1352 40. Briat, J. F., Duc, C., Ravet, K., and Gaymard, F. (2010) Ferritins and iron storage in plants. Biochim. Biophys. Acta 1800, 806-814 41. Masuda, T., Goto, F., Yoshihara, T., and Mikami, B. (2010) Crystal structure of plant ferritin reveals a novel metal binding site that functions as a transit site for metal transfer in ferritin. J. Biol. Chem. 285, 4049-4059 42. Li, C., Fu, X., Qi, X., Hu, X., Chasteen, N. D., and Zhao, G. (2009) Protein association and dissociation regulated by ferric ion: a novel pathway for oxidative deposition of iron in pea seed ferritin. J. Biol. Chem. 284, 16743-16751 43. Fu, X., Deng, J., Yang, H., Masuda, T., Goto, F., Yoshihara, T., and Zhao, G. (2010) A novel EP-involved pathway for iron release from soya bean seed ferritin. Biochem. J. 427, 313-321 44. Masuda, T., Goto, F., and Yoshihara, T. (2001) A novel plant ferritin subunit from soybean that is related to a mechanism in iron release. J. Biol. Chem. 276, 19575-19579 45. Li, C., Hu, X., and Zhao, G. (2009) Two different H-type subunits from pea seed (Pisum sativum) ferritin that are responsible for fast Fe(II) oxidation. Biochimie 91, 230-239 46. Masuda, T., Goto, F., Yoshihara, T., Ezure, T., Suzuki, T., Kobayashi, S., Shikata, M., and Utsumi, S. (2007) Construction of homo- and heteropolymers of plant ferritin subunits using an in vitro protein expression system. Protein Expr. Purif. 56, 237-246 47. Deng, J., Liao, X., Yang, H., Zhang, X., Hua, Z., Masuda, T., Goto, F., Yoshihara, T., and Zhao, G. (2010) Role of H-1 and H-2 subunits of soybean seed ferritin in oxidative deposition of iron in protein. J. Biol. Chem. 285, 32075-32086 48. AGI (Arabidopsis Genome Initiative) (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, 796-815 49. Wardrop, A. J., Wicks, R. E., and Entsch, B. (1999) Occurrence and expression of members of the ferritin gene family in cowpeas. Biochem. J. 337 ( Pt 3), 523-530 50. Stillman, T. J., Hempstead, P. D., Artymiuk, P. J., Andrews, S. C., Hudson, A. J., Treffry, A., Guest, J. R., and Harrison, P. M. (2001) The high-resolution X-ray crystallographic structure of the ferritin (EcFtnA) of Escherichia coli; comparison with human H ferritin (HuHF) and the structures of the Fe(3+) and Zn(2+) derivatives. J. Mol. Biol. 307, 587-603  113 51. Waidner, B., Greiner, S., Odenbreit, S., Kavermann, H., Velayudhan, J., Stahler, F., Guhl, J., Bisse, E., van Vliet, A. H., Andrews, S. C., Kusters, J. G., Kelly, D. J., Haas, R., Kist, M., and Bereswill, S. (2002) Essential role of ferritin Pfr in Helicobacter pylori iron metabolism and gastric colonization. Infect. Immun. 70, 3923-3929 52. Clerte, S., Dautant, A., Langlois d'Estaintot, B., Gallois, B., Mizunoe, Y., Wai, S. N., and Precigoux, G. (1999) Expression, purification, crystallization and preliminary X-ray diffraction results from Campylobacter jejuni ferritin. Acta Crystallogr. D Biol. Crystallogr. 55, 299-301 53. Ratnayake, D. B., Wai, S. N., Shi, Y., Amako, K., Nakayama, H., and Nakayama, K. (2000) Ferritin from the obligate anaerobe Porphyromonas gingivalis: purification, gene cloning and mutant studies. Microbiology 146 ( Pt 5), 1119-1127 54. Hudson, A. J., Andrews, S. C., Hawkins, C., Williams, J. M., Izuhara, M., Meldrum, F. C., Mann, S., Harrison, P. M., and Guest, J. R. (1993) Overproduction, purification and characterization of the Escherichia coli ferritin. Eur. J. Biochem. 218, 985-995 55. Bou-Abdallah, F., Yang, H., Awomolo, A., Cooper, B., Woodhall, M. R., Andrews, S. C., and Chasteen, N. D. (2014) Functionality of the three-site ferroxidase center of Escherichia coli bacterial ferritin (EcFtnA). Biochemistry 53, 483-495 56. Hempstead, P. D., Hudson, A. J., Artymiuk, P. J., Andrews, S. C., Banfield, M. J., Guest, J. R., and Harrison, P. M. (1994) Direct observation of the iron binding sites in a ferritin. FEBS Lett. 350, 258-262 57. Johnson, E., Cascio, D., Sawaya, M. R., Gingery, M., and Schroder, I. (2005) Crystal structures of a tetrahedral open pore ferritin from the hyperthermophilic archaeon Archaeoglobus fulgidus. Structure 13, 637-648 58. Tatur, J., Hagen, W. R., and Matias, P. M. (2007) Crystal structure of the ferritin from the hyperthermophilic archaeal anaerobe Pyrococcus furiosus. J. Biol. Inorg. Chem. 12, 615-630 59. Andrews, S. C., Smith, J. M., Yewdall, S. J., Guest, J. R., and Harrison, P. M. (1991) Bacterioferritins and ferritins are distantly related in evolution. Conservation of ferroxidase-centre residues. FEBS Lett. 293, 164-168 60. Stiefel, E. I., and Watt, G. D. (1979) Azotobacter cytochrome b557.5 is a bacterioferritin. Nature 279, 81-83 61. Romao, C. V., da Costa, P. N., Macedo, S., Mitchell, E., Matias, P. M., Melo, E., Liu, M. Y., Xavier, A. V., Lindley, P., LeGall, J., Carrondo, M. A., Saraiva, L. M., and Teixeira, M. (2001) A bacterioferritin from the sulphate reducing bacterium Desulfovibrio desulfuricans ATCC 27774, with a novel haem cofactor. J. Inorg. Biochem. 86, 405-405 62. Macedo, S., Romao, C. V., Mitchell, E., Matias, P. M., Liu, M. Y., Xavier, A. V., LeGall, J., Teixeira, M., Lindley, P., and Carrondo, M. A. (2003) The nature of the di-iron site in the bacterioferritin from Desulfovibrio desulfuricans. Nat. Struct. Biol. 10, 285-290 63. Romao, C. V., Regalla, M., Xavier, A. V., Teixeira, M., Liu, M. Y., and Le Gall, J. (2000) A bacterioferritin from the strict anaerobe Desulfovibrio desulfuricans ATCC 27774. Biochemistry 39, 6841-6849 64. Moore, G. R., Kadir, F. H., al-Massad, F. K., Le Brun, N. E., Thomson, A. J., Greenwood, C., Keen, J. N., and Findlay, J. B. (1994) Structural heterogeneity of Pseudomonas aeruginosa bacterioferritin. Biochem. J. 304 ( Pt 2), 493-497  114 65. Miller, C. D., Kim, Y. C., Walsh, M. K., and Anderson, A. J. (2000) Characterization and expression of the pseudomonas putida bacterioferritin alpha subunit gene. Gene 247, 199-207 66. Keren, N., Aurora, R., and Pakrasi, H. B. (2004) Critical roles of bacterioferritins in iron storage and proliferation of cyanobacteria. Plant Physiol. 135, 1666-1673 67. Crow, A., Lawson, T. L., Lewin, A., Moore, G. R., and Le Brun, N. E. (2009) Structural basis for iron mineralization by bacterioferritin. J. Am. Chem. Soc. 131, 6808-6813 68. Cheesman, M. R., Thomson, A. J., Greenwood, C., Moore, G. R., and Kadir, F. (1990) Bis-methionine axial ligation of haem in bacterioferritin from Pseudomonas aeruginosa. Nature 346, 771-773 69. Andrews, S. C., Le Brun, N. E., Barynin, V., Thomson, A. J., Moore, G. R., Guest, J. R., and Harrison, P. M. (1995) Site-directed replacement of the coaxial heme ligands of bacterioferritin generates heme-free variants. J. Biol. Chem. 270, 23268-23274 70. Yasmin, S., Andrews, S. C., Moore, G. R., and Le Brun, N. E. (2011) A new role for heme, facilitating release of iron from the bacterioferritin iron biomineral. J. Biol. Chem. 286, 3473-3483 71. Romao, C. V., Louro, R., Timkovich, R., Lubben, M., Liu, M. Y., LeGall, J., Xavier, A. V., and Teixeira, M. (2000) Iron-coproporphyrin III is a natural cofactor in bacterioferritin from the anaerobic bacterium Desulfovibrio desulfuricans. FEBS Lett. 480, 213-216 72. Zeth, K., Offermann, S., Essen, L. O., and Oesterhelt, D. (2004) Iron-oxo clusters biomineralizing on protein surfaces: structural analysis of Halobacterium salinarum DpsA in its low- and high-iron states. Proc. Natl. Acad. Sci. U. S. A. 101, 13780-13785 73. Ilari, A., Stefanini, S., Chiancone, E., and Tsernoglou, D. (2000) The dodecameric ferritin from Listeria innocua contains a novel intersubunit iron-binding site. Nat. Struct. Biol. 7, 38-43 74. Wolf, S. G., Frenkiel, D., Arad, T., Finkel, S. E., Kolter, R., and Minsky, A. (1999) DNA protection by stress-induced biocrystallization. Nature 400, 83-85 75. Frenkiel-Krispin, D., Ben-Avraham, I., Englander, J., Shimoni, E., Wolf, S. G., and Minsky, A. (2004) Nucleoid restructuring in stationary-state bacteria. Mol. Microbiol. 51, 395-405 76. Frenkiel-Krispin, D., Levin-Zaidman, S., Shimoni, E., Wolf, S. G., Wachtel, E. J., Arad, T., Finkel, S. E., Kolter, R., and Minsky, A. (2001) Regulated phase transitions of bacterial chromatin: a non-enzymatic pathway for generic DNA protection. EMBO J. 20, 1184-1191 77. Ceci, P., Cellai, S., Falvo, E., Rivetti, C., Rossi, G. L., and Chiancone, E. (2004) DNA condensation and self-aggregation of Escherichia coli Dps are coupled phenomena related to the properties of the N-terminus. Nucleic Acids Res. 32, 5935-5944 78. Calhoun, L. N., and Kwon, Y. M. (2011) Structure, function and regulation of the DNA-binding protein Dps and its role in acid and oxidative stress resistance in Escherichia coli: a review. J. Appl. Microbiol. 110, 375-386 79. Tan, S., and Davey, C. A. (2011) Nucleosome structural studies. Curr. Opin. Struct. Biol. 21, 128-136 80. Zeth, K. (2012) Dps biomineralizing proteins: multifunctional architects of nature. Biochem. J. 445, 297-311  115 81. Wiedenheft, B., Mosolf, J., Willits, D., Yeager, M., Dryden, K. A., Young, M., and Douglas, T. (2005) An archaeal antioxidant: characterization of a Dps-like protein from Sulfolobus solfataricus. Proc. Natl. Acad. Sci. U. S. A. 102, 10551-10556 82. Martic, M., Jakab-Simon, I. N., Haahr, L. T., Hagen, W. R., and Christensen, H. E. (2013) Heterometallic [AgFe(3)S (4)] ferredoxin variants: synthesis, characterization, and the first crystal structure of an engineered heterometallic iron-sulfur protein. J. Biol. Inorg. Chem. 18, 261-276 83. Ping, L., Buchler, R., Mithofer, A., Svatos, A., Spiteller, D., Dettner, K., Gmeiner, S., Piel, J., Schlott, B., and Boland, W. (2007) A novel Dps-type protein from insect gut bacteria catalyses hydrolysis and synthesis of N-acyl amino acids. Environ. Microbiol. 9, 1572-1583 84. Pesek, J., Buchler, R., Albrecht, R., Boland, W., and Zeth, K. (2011) Structure and mechanism of iron translocation by a Dps protein from Microbacterium arborescens. J. Biol. Chem. 286, 34872-34882 85. Macara, I. G., Hoy, T. G., and Harrison, P. M. (1973) The formation of ferritin from apoferritin. Catalytic action of apoferritin. Biochem. J. 135, 343-348 86. Macara, I. G., Hoy, T. G., and Harrison, P. M. (1972) The formation of ferritin from apoferritin. Kinetics and mechanism of iron uptake. Biochem. J. 126, 151-162 87. Clegg, G. A., Fitton, J. E., Harrison, P. M., and Treffry, A. (1980) Ferritin: molecular structure and iron-storage mechanisms. Prog. Biophys. Mol. Biol. 36, 56-86 88. Treffry, A., Hirzmann, J., Yewdall, S. J., and Harrison, P. M. (1992) Mechanism of catalysis of Fe(II) oxidation by ferritin H chains. FEBS Lett. 302, 108-112 89. Bauminger, E. R., Harrison, P. M., Hechel, D., Nowik, I., and Treffry, A. (1991) Mossbauer spectroscopic investigation of structure-function relations in ferritins. Biochim. Biophys. Acta 1118, 48-58 90. Hwang, J., Krebs, C., Huynh, B. H., Edmondson, D. E., Theil, E. C., and Penner-Hahn, J. E. (2000) A short Fe-Fe distance in peroxodiferric ferritin: control of Fe substrate versus cofactor decay? Science 287, 122-125 91. Tatur, J., and Hagen, W. R. (2005) The dinuclear iron-oxo ferroxidase center of Pyrococcus furiosus ferritin is a stable prosthetic group with unexpectedly high reduction potentials. FEBS Lett. 579, 4729-4732 92. Pereira, A. S., Small, W., Krebs, C., Tavares, P., Edmondson, D. E., Theil, E. C., and Huynh, B. H. (1998) Direct spectroscopic and kinetic evidence for the involvement of a peroxodiferric intermediate during the ferroxidase reaction in fast ferritin mineralization. Biochemistry 37, 9871-9876 93. Yang, X., Chen-Barrett, Y., Arosio, P., and Chasteen, N. D. (1998) Reaction paths of iron oxidation and hydrolysis in horse spleen and recombinant human ferritins. Biochemistry 37, 9743-9750 94. Bou-Abdallah, F., Zhao, G., Mayne, H. R., Arosio, P., and Chasteen, N. D. (2005) Origin of the unusual kinetics of iron deposition in human H-chain ferritin. J. Am. Chem. Soc. 127, 3885-3893 95. Zhao, G., Bou-Abdallah, F., Arosio, P., Levi, S., Janus-Chandler, C., and Chasteen, N. D. (2003) Multiple pathways for mineral core formation in mammalian apoferritin. The role of hydrogen peroxide. Biochemistry 42, 3142-3150 96. Xu, B., and Chasteen, N. D. (1991) Iron oxidation chemistry in ferritin. Increasing Fe/O2 stoichiometry during core formation. J. Biol. Chem. 266, 19965-19970  116 97. Harrison, P. M., and Arosio, P. (1996) The ferritins: molecular properties, iron storage function and cellular regulation. Biochim. Biophys. Acta 1275, 161-203 98. Bou-Abdallah, F., Biasiotto, G., Arosio, P., and Chasteen, N. D. (2004) The putative "nucleation site" in human H-chain ferritin is not required for mineralization of the iron core. Biochemistry 43, 4332-4337 99. Bauminger, E. R., Treffry, A., Quail, M. A., Zhao, Z., Nowik, I., and Harrison, P. M. (1999) Stages in iron storage in the ferritin of Escherichia coli (EcFtnA): analysis of Mossbauer spectra reveals a new intermediate. Biochemistry 38, 7791-7802 100. Bauminger, E. R., Treffry, A., Quail, M. A., Zhao, Z. W., Nowik, I., and Harrison, P. M. (2000) Metal binding at the active centre of the ferritin of Escherichia coli (EcFtnA). A Mossbauer spectroscopic study. Inorg. Chim. Acta 297, 171-180 101. Treffry, A., Zhao, Z., Quail, M. A., Guest, J. R., and Harrison, P. M. (1998) How the presence of three iron binding sites affects the iron storage function of the ferritin (EcFtnA) of Escherichia coli. FEBS Lett. 432, 213-218 102. Le Brun, N. E., Wilson, M. T., Andrews, S. C., Guest, J. R., Harrison, P. M., Thomson, A. J., and Moore, G. R. (1993) Kinetic and structural characterization of an intermediate in the biomineralization of bacterioferritin. FEBS Lett. 333, 197-202 103. Yang, X., Le Brun, N. E., Thomson, A. J., Moore, G. R., and Chasteen, N. D. (2000) The iron oxidation and hydrolysis chemistry of Escherichia coli bacterioferritin. Biochemistry 39, 4915-4923 104. Honarmand Ebrahimi, K., Hagedoorn, P. L., Jongejan, J. A., and Hagen, W. R. (2009) Catalysis of iron core formation in Pyrococcus furiosus ferritin. J. Biol. Inorg. Chem. 14, 1265-1274 105. Honarmand Ebrahimi, K., Hagedoorn, P. L., and Hagen, W. R. (2010) Inhibition and stimulation of formation of the ferroxidase center and the iron core in Pyrococcus furiosus ferritin. J. Biol. Inorg. Chem. 15, 1243-1253 106. Honarmand Ebrahimi, K., Bill, E., Hagedoorn, P. L., and Hagen, W. R. (2012) The catalytic center of ferritin regulates iron storage via Fe(II)-Fe(III) displacement. Nat. Chem. Biol. 8, 941-948 107. Ebrahimi, K. H., Hagedoorn, P. L., and Hagen, W. R. (2013) A conserved tyrosine in ferritin is a molecular capacitor. Chembiochem 14, 1123-1133 108. LeBrun, N. E., Thomson, A. J., and Moore, G. R. (1997) Metal centres of bacterioferritins or non-haem-iron-containing cytochromes b(557). Metal Sites in Proteins and Models 88, 103-138 109. Lindsay, S., Brosnahan, D., Lowery, T. J., Jr., Crawford, K., and Watt, G. D. (2003) Kinetic studies of iron deposition in horse spleen ferritin using O2 as oxidant. Biochim. Biophys. Acta 1621, 57-66 110. Moore, G. R., Mann, S., and Bannister, J. V. (1986) Isolation and properties of the complex nonheme-iron-containing cytochrome b557 (bacterioferritin) from Pseudomonas aeruginosa. J. Inorg. Biochem. 28, 329-336 111. Michel, F. M., Ehm, L., Antao, S. M., Lee, P. L., Chupas, P. J., Liu, G., Strongin, D. R., Schoonen, M. A., Phillips, B. L., and Parise, J. B. (2007) The structure of ferrihydrite, a nanocrystalline material. Science 316, 1726-1729 112. Towe, K. M., and Bradley, W. F. (1967) Mineralogical Constitution of Colloidal Hydrous Ferric Oxides. J. Colloid Interface Sci. 24, 384-392  117 113. Galvez, N., Fernandez, B., Sanchez, P., Cuesta, R., Ceolin, M., Clemente-Leon, M., Trasobares, S., Lopez-Haro, M., Calvino, J. J., Stephan, O., and Dominguez-Vera, J. M. (2008) Comparative structural and chemical studies of ferritin cores with gradual removal of their iron contents. J. Am. Chem. Soc. 130, 8062-8068 114. St Pierre, T. G., Bell, S. H., Dickson, D. P., Mann, S., Webb, J., Moore, G. R., and Williams, R. J. (1986) Mossbauer spectroscopic studies of the cores of human, limpet and bacterial ferritins. Biochim. Biophys. Acta 870, 127-134 115. Mann, S., Williams, J. M., Treffry, A., and Harrison, P. M. (1987) Reconstituted and native iron-cores of bacterioferritin and ferritin. J. Mol. Biol. 198, 405-416 116. Reid, N. M. K., Dickson, D. P. E., Greenwood, C., Thompson, A., Kadir, F. H. A., and Moore, G. R. (1990) Evidence for Mossbauer-Spectroscopy for Different Forms of Iron Core in Pseudomonas-Aeruginosa Bacterial Ferritin. Biochem. J. 272, 263-264 117. Rohrer, J. S., Islam, Q. T., Watt, G. D., Sayers, D. E., and Theil, E. C. (1990) Iron environment in ferritin with large amounts of phosphate, from Azotobacter vinelandii and horse spleen, analyzed using extended X-ray absorption fine structure (EXAFS). Biochemistry 29, 259-264 118. Doig, P., Austin, J. W., and Trust, T. J. (1993) The Helicobacter pylori 19.6-kilodalton protein is an iron-containing protein resembling ferritin. J. Bacteriol. 175, 557-560 119. Aitken-Rogers, H., Singleton, C., Lewin, A., Taylor-Gee, A., Moore, G. R., and Le Brun, N. E. (2004) Effect of phosphate on bacterioferritin-catalysed iron(II) oxidation. J. Biol. Inorg. Chem. 9, 161-170 120. Waldo, G. S., Wright, E., Whang, Z. H., Briat, J. F., Theil, E. C., and Sayers, D. E. (1995) Formation of the ferritin iron mineral occurs in plastids. Plant Physiol. 109, 797-802 121. Levi, S., Santambrogio, P., Corsi, B., Cozzi, A., and Arosio, P. (1996) Evidence that residues exposed on the three-fold channels have active roles in the mechanism of ferritin iron incorporation. Biochem. J. 317 ( Pt 2), 467-473 122. Treffry, A., Bauminger, E. R., Hechel, D., Hodson, N. W., Nowik, I., Yewdall, S. J., and Harrison, P. M. (1993) Defining the roles of the threefold channels in iron uptake, iron oxidation and iron-core formation in ferritin: a study aided by site-directed mutagenesis. Biochem. J. 296 ( Pt 3), 721-728 123. Tosha, T., Ng, H. L., Bhattasali, O., Alber, T., and Theil, E. C. (2010) Moving metal ions through ferritin-protein nanocages from three-fold pores to catalytic sites. J. Am. Chem. Soc. 132, 14562-14569 124. Theil, E. C. (2011) Ferritin protein nanocages use ion channels, catalytic sites, and nucleation channels to manage iron/oxygen chemistry. Curr. Opin. Chem. Biol. 15, 304-311 125. Turano, P., Lalli, D., Felli, I. C., Theil, E. C., and Bertini, I. (2010) NMR reveals pathway for ferric mineral precursors to the central cavity of ferritin. Proc. Natl. Acad. Sci. U. S. A. 107, 545-550 126. St Pierre, T. G., Tran, K. C., Webb, J., Macey, D. J., Heywood, B. R., Sparks, N. H., Wade, V. J., Mann, S., and Pootrakul, P. (1991) Organ-specific crystalline structures of ferritin cores in beta-thalassemia/hemoglobin E. Biol. Met. 4, 162-165 127. Carrondo, M. A. (2003) Ferritins, iron uptake and storage from the bacterioferritin viewpoint. EMBO J. 22, 1959-1968  118 128. Radisky, D. C., and Kaplan, J. (1998) Iron in cytosolic ferritin can be recycled through lysosomal degradation in human fibroblasts. Biochem. J. 336 ( Pt 1), 201-205 129. Liu, X., and Theil, E. C. (2005) Ferritins: dynamic management of biological iron and oxygen chemistry. Accounts Chem. Res. 38, 167-175 130. Theil, E. C., Liu, X. S., and Tosha, T. (2008) Gated Pores in the Ferritin Protein Nanocage. Inorg. Chim. Acta 361, 868-874 131. Wardeska, J. G., Viglione, B., and Chasteen, N. D. (1986) Metal-Ion Complexes of Apoferritin - Evidence for Initial Binding in the Hydrophilic Channels. J. Biol. Chem. 261, 6677-6683 132. Watt, G. D., Jacobs, D., and Frankel, R. B. (1988) Redox reactivity of bacterial and mammalian ferritin: is reductant entry into the ferritin interior a necessary step for iron release? Proc. Natl. Acad. Sci. U. S. A. 85, 7457-7461 133. Jin, W., Takagi, H., Pancorbo, B., and Theil, E. C. (2001) "Opening" the ferritin pore for iron release by mutation of conserved amino acids at interhelix and loop sites. Biochemistry 40, 7525-7532 134. Chasteen, N. D., Ritchie, I. M., and Webb, J. (1991) Stepped potential microcoulometry of ferritin. Anal. Biochem. 195, 296-302 135. Jacobs, D. L., Watt, G. D., Frankel, R. B., and Papaefthymiou, G. C. (1989) Redox reactions associated with iron release from mammalian ferritin. Biochemistry 28, 1650-1655 136. Shi, H. F., Bencze, K. Z., Stemmler, T. L., and Philpott, C. C. (2008) A cytosolic iron chaperone that delivers iron to ferritin. Science 320, 1207-1210 137. Zhang, B., and Watt, G. D. (2007) Anaerobic iron deposition into horse spleen, recombinant human heavy and light and bacteria ferritins by large oxidants. J. Inorg. Biochem. 101, 1676-1685 138. Watt, R. K., Hilton, R. J., and Graff, D. M. (2010) Oxido-reduction is not the only mechanism allowing ions to traverse the ferritin protein shell. Biochim. Biophys. Acta 1800, 745-759 139. Liu, X., Jin, W., and Theil, E. C. (2003) Opening protein pores with chaotropes enhances Fe reduction and chelation of Fe from the ferritin biomineral. Proc. Natl. Acad. Sci. U. S. A. 100, 3653-3658 140. Takagi, H., Shi, D., Ha, Y., Allewell, N. M., and Theil, E. C. (1998) Localized unfolding at the junction of three ferritin subunits. A mechanism for iron release? J. Biol. Chem. 273, 18685-18688 141. Liu, X. S., Patterson, L. D., Miller, M. J., and Theil, E. C. (2007) Peptides selected for the protein nanocage pores change the rate of iron recovery from the ferritin mineral. J. Biol. Chem. 282, 31821-31825 142. Sims, P. A., Mann, D. G., and Medlin, L. K. (2006) Evolution of the diatoms: insights from fossil, biological and molecular data. Phycologia 45, 361-402 143. Armbrust, E. V. (2009) The life of diatoms in the world's oceans. Nature 459, 185-192 144. Falkowski, P. G., Barber, R. T., and Smetacek, V. V. (1998) Biogeochemical controls and feedbacks on ocean primary production. Science 281, 200-207 145. Orr J.C., M.-R., E., Mikolajewicz, U., Monfray, P., Sarmiento, J.L., Toggweiler, J.R., Taylor, N.K., Palmer, J., Gruber, N., Sabine, C.L., Le Quéré, C., Key, R.M., Boutin, J. (2001) Estimates of anthropogenic carbon uptake from four three-dimensional global ocean models. Global Biogeochem. Cy. 15, 43-60  119 146. Falkowski, P. G. (2002) The ocean's invisible forest. Sci. Am. 287, 54-61 147. Sarthou, G., Timmermans, K. R., Blain, S., and Tréguer, P. (2005) Growth physiology and fate of diatoms in the ocean: a review. J. Sea Res. 53, 25-42 148. Morel, F. M., and Price, N. M. (2003) The biogeochemical cycles of trace metals in the oceans. Science 300, 944-947 149. Falkowski, P. G., and Oliver, M. J. (2007) Mix and match: how climate selects phytoplankton. Nat. Rev. Microbiol. 5, 813-819 150. Boyd, P. W., Jickells, T., Law, C. S., Blain, S., Boyle, E. A., Buesseler, K. O., Coale, K. H., Cullen, J. J., de Baar, H. J., Follows, M., Harvey, M., Lancelot, C., Levasseur, M., Owens, N. P., Pollard, R., Rivkin, R. B., Sarmiento, J., Schoemann, V., Smetacek, V., Takeda, S., Tsuda, A., Turner, S., and Watson, A. J. (2007) Mesoscale iron enrichment experiments 1993-2005: synthesis and future directions. Science 315, 612-617 151. Martin, J. H., Fitzwater, S.E. (1988) Iron deficiency limits phytoplankton growth in the north-east pacific subarctic. Nature 331, 341-343 152. Raven, J. A., Evans., M.C.W., Korb, R.E. (1999) The role of trace metals in photosynthetic electron transport in O2-evolving organisms. Photosynth. Res. 60, 111-149 153. Castruita, M., Shaked, Y., Elmegreen, L.A., Stiefel, E.I. (2008) Availability of iron from iron-storage proteins to marine phytoplankton. Limnology and Oceanography 53 890-899 154. Gledhill, M., and Buck, K. N. (2012) The organic complexation of iron in the marine environment: a review. Front. Microbiol. 3, 69 155. Brand, L. E., Sunda, W. G., and Guillard, R. R. L. (1983) Limitation of Marine-Phytoplankton Reproductive Rates by Zinc, Manganese, and Iron. Limnology and Oceanography 28, 1182-1198 156. Marchetti, A., and Cassar, N. (2009) Diatom elemental and morphological changes in response to iron limitation: a brief review with potential paleoceanographic applications. Geobiology 7, 419-431 157. Maldonado, M. T., and Price, N. M. (1999) Utilization of iron bound to strong organic ligands by plankton communities in the subarctic Pacific Ocean. Deep-Sea Res. Pt. II 46, 2447-2473 158. Maldonado, M. T., and Price, N. M. (2001) Reduction and transport of organically bound iron by Thalassiosira oceanica (Bacillariophyceae). J. Phycol. 37, 298-309 159. Hutchins, D. A., Witter, A. E., Butler, A., and Luther, G. W. (1999) Competition among marine phytoplankton for different chelated iron species. Nature 400, 858-861 160. Ditullio, G. R., Hutchins, D. A., and Bruland, K. W. (1993) Interaction of Iron and Major Nutrients Controls Phytoplankton Growth and Species Composition in the Tropical North Pacific-Ocean. Limnology and Oceanography 38, 495-508 161. Martin, J. H. (1990) Glacial-Interglacial Co2 Change: The Iron Hypothesis. Paleoceanography 5, 1-13 162. Boyd, P. W., Law, C. S., Wong, C. S., Nojiri, Y., Tsuda, A., Levasseur, M., Takeda, S., Rivkin, R., Harrison, P. J., Strzepek, R., Gower, J., McKay, M., Abraham, E., Arychuk, M., Barwell-Clarke, J., Crawford, W., Crawford, D., Hale, M., Harada, K., Johnson, K., Kiyosawa, H., Kudo, I., Marchetti, A., Miller, W., Needoba, J., Nishioka, J., Ogawa, H., Page, J., Robert, M., Saito, H., Sastri, A., Sherry, N., Soutar, T., Sutherland, N., Taira, Y., Whitney, F., Wong, S. K., and Yoshimura, T. (2004) The decline and fate of an iron-induced subarctic phytoplankton bloom. Nature 428, 549-553  120 163. Gall, M. P., Boyd, P. W., Hall, J., Safi, K. A., and Chang, H. (2001) Phytoplankton processes. Part 1: Community structure during the Southern Ocean Iron RElease Experiment (SOIREE). Deep-Sea Research Part II-Topical Studies in Oceanography 48, 2551-2570 164. Boyd, P. W., Watson, A. J., Law, C. S., Abraham, E. R., Trull, T., Murdoch, R., Bakker, D. C., Bowie, A. R., Buesseler, K. O., Chang, H., Charette, M., Croot, P., Downing, K., Frew, R., Gall, M., Hadfield, M., Hall, J., Harvey, M., Jameson, G., LaRoche, J., Liddicoat, M., Ling, R., Maldonado, M. T., McKay, R. M., Nodder, S., Pickmere, S., Pridmore, R., Rintoul, S., Safi, K., Sutton, P., Strzepek, R., Tanneberger, K., Turner, S., Waite, A., and Zeldis, J. (2000) A mesoscale phytoplankton bloom in the polar Southern Ocean stimulated by iron fertilization. Nature 407, 695-702 165. Martin, J. H., Coale, K. H., Johnson, K. S., Fitzwater, S. E., Gordon, R. M., Tanner, S. J., Hunter, C. N., Elrod, V. A., Nowicki, J. L., Coley, T. L., Barber, R. T., Lindley, S., Watson, A. J., Vanscoy, K., Law, C. S., Liddicoat, M. I., Ling, R., Stanton, T., Stockel, J., Collins, C., Anderson, A., Bidigare, R., Ondrusek, M., Latasa, M., Millero, F. J., Lee, K., Yao, W., Zhang, J. Z., Friederich, G., Sakamoto, C., Chavez, F., Buck, K., Kolber, Z., Greene, R., Falkowski, P., Chisholm, S. W., Hoge, F., Swift, R., Yungel, J., Turner, S., Nightingale, P., Hatton, A., Liss, P., and Tindale, N. W. (1994) Testing the Iron Hypothesis in Ecosystems of the Equatorial Pacific-Ocean. Nature 371, 123-129 166. de Baar, H. J. W., Boyd, P. W., Coale, K. H., Landry, M. R., Tsuda, A., Assmy, P., Bakker, D. C. E., Bozec, Y., Barber, R. T., Brzezinski, M. A., Buesseler, K. O., Boye, M., Croot, P. L., Gervais, F., Gorbunov, M. Y., Harrison, P. J., Hiscock, W. T., Laan, P., Lancelot, C., Law, C. S., Levasseur, M., Marchetti, A., Millero, F. J., Nishioka, J., Nojiri, Y., van Oijen, T., Riebesell, U., Rijkenberg, M. J. A., Saito, H., Takeda, S., Timmermans, K. R., Veldhuis, M. J. W., Waite, A. M., and Wong, C. S. (2005) Synthesis of iron fertilization experiments: From the iron age in the age of enlightenment. J. Geophys. Res.-Oceans 110 167. Bollens, G. C. R., and Landry, M. R. (2000) Biological response to iron fertilization in the eastern equatorial Pacific (IronEx II). II. Mesozooplankton abundance, biomass, depth distribution and grazing. Marine Ecology Progress Series 201, 43-56 168. Gervais, F., Riebesell, U., and Gorbunov, M. Y. (2002) Changes in primary productivity and chlorophyll a in response to iron fertilization in the Southern Polar Frontal Zone. Limnology and Oceanography 47, 1324-1335 169. Marchetti, A., Sherry, N. D., Kiyosawa, H., Tsuda, A., and Harrison, P. J. (2006) Phytoplankton processes during a mesoscale iron enrichment in the NE subarctic Pacific: Part I - Biomass and assemblage. Deep-Sea Res. Pt. II 53, 2095-2113 170. Cefarelli, A. O., Ferrario, M. E., Almandoz, G. O., Atencio, A. G., Akselman, R., and Vernet, M. (2010) Diversity of the diatom genus Fragilariopsis in the Argentine Sea and Antarctic waters: morphology, distribution and abundance. Polar Biol. 33, 1463-1484 171. Wells, M. L., Trick, C. G., Cochlan, W. P., Hughes, M. P., and Trainer, V. L. (2005) Domoic acid: The synergy of iron, copper, and the toxicity of diatoms. Limnology and Oceanography 50, 1908-1917 172. Murphy, M., E., P. (2009) Marine diatoms survive iron droughts in the ocean by storing iron in ferritin. SSRL Science Highlight   121 173. MacPherson, I. S., Rosell, F. I., Scofield, M., Mauk, A. G., and Murphy, M. E. (2010) Directed evolution of copper nitrite reductase to a chromogenic reductant. Protein Eng. Des. Sel. 23, 137-145 174. Otwinowski, Z., and Minor, W. (1997) Processing of X-ray diffraction data collected in oscillation mode. Method. Enzymol. 276, 307-326 175. Vagin, A., Teplyakov, A. (1997) MOLREP: an automated program for molecular replacement. J. Appl. Cryst. 30, 1022-1025 176. Emsley, P., and Cowtan, K. (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 60, 2126-2132 177. Murshudov, G. N., Vagin, A. A., and Dodson, E. J. (1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr. D Biol. Crystallogr. 53, 240-255 178. Leslie, A. G. W. (2006) The integration of macromolecular diffraction data. Acta Crystallogr. D Biol. Crystallogr. 62, 48-57 179. Evans, P. (2006) Scaling and assessment of data quality. Acta Crystallogr. D Biol. Crystallogr. 62, 72-82 180. Kabsch, W. (2010) Xds. Acta Crystallogr. D Biol. Crystallogr. 66, 125-132 181. Evans, P. R. (2011) An introduction to data reduction: space-group determination, scaling and intensity statistics. Acta Crystallogr. D Biol. Crystallogr. 67, 282-292 182. Baaghil, S., Lewin, A., Moore, G. R., and Le Brun, N. E. (2003) Core formation in Escherichia coli bacterioferritin requires a functional ferroxidase center. Biochemistry 42, 14047-14056 183. Bertini, I., Lalli, D., Mangani, S., Pozzi, C., Rosa, C., Theil, E. C., and Turano, P. (2012) Structural insights into the ferroxidase site of ferritins from higher eukaryotes. J. Am. Chem. Soc. 134, 6169-6176 184. Treffry, A., Zhao, Z., Quail, M. A., Guest, J. R., and Harrison, P. M. (1997) Dinuclear center of ferritin: studies of iron binding and oxidation show differences in the two iron sites. Biochemistry 36, 432-441 185. Yang, X., Chiancone, E., Stefanini, S., Ilari, A., and Chasteen, N. D. (2000) Iron oxidation and hydrolysis reactions of a novel ferritin from Listeria innocua. Biochem. J. 349 Pt 3, 783-786 186. Wong, S. G., Tom-Yew, S. A., Lewin, A., Le Brun, N. E., Moore, G. R., Murphy, M. E., and Mauk, A. G. (2009) Structural and mechanistic studies of a stabilized subunit dimer variant of Escherichia coli bacterioferritin identify residues required for core formation. J. Biol. Chem. 284, 18873-18881 187. Schwartz, J. K., Liu, X. S., Tosha, T., Diebold, A., Theil, E. C., and Solomon, E. I. (2010) CD and MCD spectroscopic studies of the two Dps miniferritin proteins from Bacillus anthracis: role of O2 and H2O2 substrates in reactivity of the diiron catalytic centers. Biochemistry 49, 10516-10525 188. Treffry, A., Zhao, Z., Quail, M. A., Guest, J. R.,  Harrison, P. M. (1998) The use of zinc(II) to probe iron binding and oxidatoin by the ferritin (EcFtnA) of Escherichia coli. J. Biol. Inorg. Chem. 3, 682-688 189. Willies, S. C., Isupov, M. N., Garman, E. F., and Littlechild, J. A. (2009) The binding of haem and zinc in the 1.9 A X-ray structure of Escherichia coli bacterioferritin. J. Biol. Inorg. Chem. 14, 201-207  122 190. Glaesener, A. G., Merchant, S. S., and Blaby-Haas, C. E. (2013) Iron economy in. Front Plant Sci 4, 337 191. Long, J. C., and Merchant, S. S. (2008) Photo-oxidative stress impacts the expression of genes encoding iron metabolism components in Chlamydomonas. Photochem. Photobiol. 84, 1395-1403 192. Busch, A., Rimbauld, B., Naumann, B., Rensch, S., and Hippler, M. (2008) Ferritin is required for rapid remodeling of the photosynthetic apparatus and minimizes photo-oxidative stress in response to iron availability in Chlamydomonas reinhardtii. Plant J. 55, 201-211 193. Frolow, F., Kalb, A. J., and Yariv, J. (1994) Structure of a unique twofold symmetric haem-binding site. Nat. Struct. Biol. 1, 453-460 194. van Eerde, A., Wolterink-van Loo, S., van der Oost, J., and Dijkstra, B. W. (2006) Fortuitous structure determination of 'as-isolated' Escherichia coli bacterioferritin in a novel crystal form. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 62, 1061-1066 195. Le Brun, N. E., Andrews, S. C., Guest, J. R., Harrison, P. M., Moore, G. R., and Thomson, A. J. (1995) Identification of the ferroxidase centre of Escherichia coli bacterioferritin. Biochem. J. 312 ( Pt 2), 385-392 196. Chenbarrett, Y., Harrison, P. M., Treffry, A., Quail, M. A., Arosio, P., Santambrogio, P., and Chasteen, N. D. (1995) Tyrosyl Radical Formation during the Oxidative Deposition of Iron in Human Apoferritin. Biochemistry 34, 7847-7853   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0167604/manifest

Comment

Related Items