Studies of the Structure and Function of Recombinant Human Hephaestin

Home , Duodenal cytochrome B, Fet3p, Hephaestin, Human iron metabolism, STEAP3

STUDIES OF THE STRUCTURE AND FUNCTION OF RECOMBINANT HUMAN HEPHAESTIN

Ganna Vashchenko

B.Sc., Taras Shevchenko National University of Kyiv, 2007

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

The Faculty of Graduate Studies

(Biochemistry and Molecular Biology)

THE UNIVERSITY OF BRITISH COLUMBIA

(Vancouver)

July 2012

Hephaestin is a multicopper ferroxidase involved in iron absorption in the small intestine.

The ferroxidase activity of hephaestin is thought to play an important role during iron export from intestinal enterocytes and the subsequent iron loading of the blood protein transferrin, which delivers iron to the tissues. Structurally, the ectodomain of hephaestin is predicted to resemble ceruloplasmin, the soluble ferroxidase of blood. In this work I investigated substrate specificity, copper loading and the ferroxidation mechanism of recombinantly expressed human hephaestin.

The hephaestin ectodomain (Fet3Hp) was expressed in Pichia pastoris and purified to electrophoretic homogeneity by immunoaffinity chromatography. Recombinant hephaestin retained ferroxidase activity and showed an average copper content of 4.2 copper atoms per molecule. The Km values of Fet3Hp for such organic substrates as p-phenylenediamine and o- dianisidine were close to values determined for ceruloplasmin. However, in contrast to ceruloplasmin, recombinant hephaestin was incapable of direct oxidation of adrenaline and dopamine implying a difference in biological substrate specificities between these two homologous oxidases.

I also expressed hephaestin ectodomain with the ceruloplasmin signal peptide (CpHp) using BHK cells as an expression system. Ion exchange chromatography of purified CpHp resulted in the production of a hephaestin fraction with improved catalytic and spectroscopic properties. Detailed kinetic analysis of CpHp ferroxidation rates revealed the presence of two types of iron-binding sites with different affinities towards ferrous iron. Michaelis constants for high- and low-affinity ferrous binding sites in CpHp were comparabale to the corresponding Km values in ceruloplasmin suggesting that both paralogs utilize similar amino acid residues for iron binding. To investigate the role of particular residues in iron specificity of hephaestin, mutations of putative iron ligands were introduced into CpHp using site-directed mutagenesis. Kinetic

ii analysis of ferroxidation rates of wild-type CpHp and variants revealed the important roles of residues E960 and H965 in hephaestin ferroxidase activity.

iii PREFACE

The work outlined in chapter 3 has been published (Vashchenko G., Bleackley M.R.,

Griffiths T.A. & MacGillivray R.T. (2011) Oxidation of organic and biogenic amines by recombinant human hephaestin expressed in Pichia pastoris. Arch. Biochem. Biophys. 514, 50-

6). I was responsible for designing and carrying out the experiments. Initial cloning steps required for these experiments were performed by Dr. Mark Bleackley. Sepharose derivative with ethylamine used for purification of human ceruloplasmin was synthesized by Dr. Kay Yu.

The work outlined in Chapter 5 along with the work described in Chapter 4 is being prepared for publication.

I created all of the text, figures and tables in this thesis.

iv TABLE OF CONTENTS

ABSTRACT...... II

PREFACE...... IV

TABLE OF CONTENTS ...... V

LIST OF TABLES...... VII

LIST OF FIGURES...... VIII

LIST OF ABBREVIATIONS...... IX

ACKNOWLEDGEMENTS ...... XI 1. INTRODUCTION ...... 1 1.1 IRON IN BIOLOGY ...... 1 1.2 PROKARYOTIC IRON METABOLISM ...... 2 1.3 YEAST IRON METABOLISM ...... 3 1.3 IRON METABOLISM IN HUMANS ...... 4 1.3.1 Iron absorption in small intestine...... 4 1.3.2 Iron uptake in different cell types...... 8 1.3.3 Regulation of iron homeostasis ...... 10 1.3.4 Inherited disorders of iron metabolism...... 12 1.4 MULTICOPPER OXIDASES ...... 14 1.4.1 Marcus theory of electron transfer...... 15 1.4.2 Type 1 copper sites...... 16 1.4.3 Transfer of electrons to the trinuclear cluster and dioxygen reduction...... 18 1.5 MULTICOPPER METALLOOXIDASES ...... 18 1.5.1 Fet3p...... 20 1.5.2 Ceruloplasmin...... 22 1.5.3 Hephaestin...... 26 1.5.4 Zyklopen ...... 27 1.5.5 Ferroxidases in human iron metabolism...... 28 1.5 STATEMENT OF HYPOTHESES AND OBJECTIVES ...... 29 2. MATERIAL AND METHODS...... 31 2.1 MATERIALS...... 31 2.2 DNA MANIPULATIONS...... 31 2.2.1 Production of the Fet3Hp expression cassette and its integration into the P. pastoris genome ...... 32 2.2.2 Production of stably transfected BHK cells for expression of CpHp and its variants...... 34 2.3 RECOMBINANT PROTEIN EXPRESSION...... 37 2.3.1 Expression of Fet3Hp in P. pastoris...... 37 2.3.2 Expression of CpHp and its variants in BHK cells...... 38 2.4 PROTEIN PURIFICATION AND STORAGE...... 39 2.4.1 Purification of Fet3Hp...... 39 2.4.2 Purification of CpHp and variants...... 40 2.4.3 Purification of human ceruloplasmin...... 40 2.5 ANALYTICAL METHODS ...... 41 2.5.1 Protein quantification...... 41 2.5.2 N-terminal sequence analysis...... 42 2.5.3 Enzymatic deglycosylation...... 42 2.5.4 Far-UV CD spectroscopy ...... 43 2.5.5 UV-visible spectroscopy ...... 43 2.5.6 Inductively coupled plasma mass spectrometry with dynamic reaction cell (ICP-MS with DRC) ...... 43 2.5.7 Biquinoline assay for determination of copper content...... 44 2.6 ACTIVITY ASSAYS...... 45 2.6.1 Fet3Hp ferroxidase assay...... 45 2.6.2 In-gel assay of Fet3Hp pPD-oxidase activity ...... 45 2.6.3 Oxidation of organic substrates by Fet3Hp ...... 46 2.6.4 Measuring specific enzymatic activity of different CpHp samples...... 47 2.6.5 Monitoring CpHp ferroxidase activity by formation of ferric iron (III) oxide hydrate ...... 47 2.6.6 Iron loading of apotransferrin in presence of CpHp ...... 48

v 2.6.7 Discontinuous ferrozine assay for ferroxidase activity of CpHp and its variants ...... 48 2.6.8 pPD-oxidase assay for CpHp and variants...... 49 3. CHARACTERIZATION OF RECOMBINANT HUMAN HEPHAESTIN EXPRESSED IN PICHIA PASTORIS ...... 50 3.1 OVERVIEW AND RATIONALE ...... 50 3.2 RESULTS...... 51 3.2.1 Optimization of Fet3Hp expression...... 51 3.2.2 Purification of Fet3Hp...... 53 3.2.3 Initial characterization of Fet3Hp...... 53 3.2.4 Ferroxidase activity of Fet3Hp...... 55 3.2.5 Total copper content ...... 57 3.2.6 Oxidation of organic substrates ...... 57 3.2.7 Effect of chelators and iron on oxidation of pPD by Fet3Hp...... 60 3.3 DISCUSSION ...... 62 3.4 CONCLUSIONS...... 63 4. ENZYMATIC ACTIVITY AND COPPER CONTENT OF HUMAN HEPHAESTIN ...... 64 4.1 OVERVIEW AND RATIONALE ...... 64 4.2 RESULTS...... 65 4.2.1 Expression, purification and initial characterization of CpHp ...... 65 4.2.2 Correlation between pPD- and ferroxidase activities in different CpHp samples...... 66 4.2.3 Ferroxidase activity of CpHp monitored by product formation...... 68 4.2.4 Far-UV CD spectrum of CpHp...... 71 4.2.5 CpHp separation with ion exchange chromatography ...... 72 4.3 DISCUSSION ...... 75 4.4 CONCLUSIONS...... 76 5. OXIDATION OF FERROUS IRON BY HUMAN HEPHAESTIN ...... 77 5.1 OVERVIEW AND RATIONALE ...... 77 5.2 RESULTS...... 81 5.2.1 Kinetic analysis of iron oxidation by hephaestin and ceruloplasmin...... 81 5.2.2 Strategy for site-directed mutagenesis of putative iron ligands ...... 82 5.2.3 Analysis of enzymatic activity of wild-type CpHp and variants ...... 82 5.3 DISCUSSION ...... 85 5.4 CONCLUSIONS...... 86 6. GENERAL DISCUSSION AND FUTURE DIRECTIONS...... 87 6.1 GENERAL DISCUSSION ...... 87 6.1.1 Recombinant hephaestins ...... 87 6.1.2 Oxidation of organic substrates ...... 89 6.1.3 Oxidation of ferrous iron ...... 90 6.2 SIGNIFICANCE ...... 91 6.3 FUTURE DIRECTIONS...... 91 6.3.1 Size exclusion chromatography of rhHp ...... 91 6.3.2 Ferroxidation mechanism of hephaestin ...... 92 6.3.3 Alternative substrates of hephaestin...... 92 BIBLIOGRAPHY...... 95 APPENDICES...... 120 A. OXIDATION OF ORGANIC AMINES (CHEMICAL EQUATIONS)...... 120 B. OXIDATION OF BIOGENIC AMINES (CHEMICAL EQUATIONS)...... 121 C. KINETIC PARAMETERS FOR DIFFERENT SAMPLES OF CPHP AND VARIANTS...... 122

vi LIST OF TABLES

Table 1.1. Hereditary disorders associated with iron imbalance...... 13 Table 2.1. Oligonucleotides used in this study...... 36 Table 3.1. Kinetic parameters for pPD and o-dianisidine for Fet3Hp and ceruloplasmin...... 57 Table 5.1. Putative iron ligands in hephaestin and ceruloplasmin...... 78 Table 5.2. Kinetic parameters of iron oxidation by hephaestin and ceruloplasmin...... 71 Table 5.3. Kinetic parameters for CpHp and variants...... 83 Table 6.1. Kinetic and spectroscopic properties of different ferroxidases...... 88

vii LIST OF FIGURES

Figure 1.1. Transport of iron in the enterocyte...... 7

Figure 1.2. Mechanism of O2 reduction to water by the MCOs...... 17 Figure 1.3. Iron binding by Fet3p...... 21 Figure 1.4. Iron binding by human ceruloplasmin I...... 23 Figure 1.5. Iron binding by human ceruloplasmin II...... 24 Figure 2.1. Construct used for expression of Fet3Hp (Fet3HppPICZA)...... 33 Figure 2.2. Construct used for expression of CpHp (CpHppNUT)...... 35 Figure 3.1. Western blot analysis of P. pastoris medium using polyclonal anti-Hp antibodies....52 Figure 3.2. Purification and initial characterization of Fet3Hp...... 54 Figure 3.3. Ferroxidase activity of Fet3Hp...... 56 Figure 3.4. Velocity versus substrate analysis of organic amine oxidation by Fet3Hp and ceruloplasmin...... 58 Figure 3.5. Oxidation of pPD, adrenaline and dopamine by Fet3Hp and ceruloplasmin...... 59 Figure 3.6. Iron as a mediator of organic substrates oxidation...... 61 Figure 4.1. SDS-PAGE of CpHp and TfHp...... 66 Figure 4.2. Enzymatic activity of different CpHp samples...... 67 Figure 4.3. Formation iron (III) oxide hydrate in presence of CpHp...... 69 Figure 4.4. Iron loading of apotransferrin in presence of CpHp...... 70 Figure 4.5. CD spectroscopy of CpHp...... 71 Figure 4.6. Fractionation of CpHp by chromatography on DEAE-Sepharose...... 73 Figure 4.7. Spectroscopic properties of CpHp fractions produced by chromatography on DEAE- Sepharose...... 74 Figure 5.1. Partial amino acid sequence alignment between human hephaestin and ceruloplasmin...... 78 Figure 5.2. Ferroxidase activity of recombinant hephaestin...... 79 Figure 5.3. Ferroxidase activity of human ceruloplasmin...... 80 Figure 5.4. Ferroxidase activity of wild-type and variant rhHp...... 84

viii

LIST OF ABBREVIATIONS

BCA – bicinchoninic acid

BHK – baby hamster kidney

BSA – bovine serum albumin

BMGY – buffered glycerol complex medium

BMMY – buffered methanol complex medium

CD – circular dichroism

Cp – ceruloplasmin

CpHp – recombinant human hephaestin with ceruloplasmin signal peptide

DEAE – diethylaminoethyl

DMEM-F12 – Dulbecco's modified Eagle medium-Ham's F12 nutrient mixture

EDTA – ethylenediaminetetraacetic acid

Endo H – endoglycosidase H

DTT – dithiothreitol

Fet3Hp – recombinant human hephaestin with Fet3p signal peptide

Hp – hephaestin

ICP-MS – inductively-coupled plasma mass spectrometry

IEC – ion exchange chromatography

LDL – low-density lipoprotein

MCO – multicopper oxidase mdeg – millidegrees

LDL – low-density lipoprotein

MW – molecular weight

NCS – newborn calf serum

PBST – phosphate-buffered saline Tween

ix PCR – polymerase chain reaction

PNGase F – N-Glycosidase F pPD – p-phenylenediamine

PVDF – polyvinylidenefluoride rhHp – recombinant human hephaestin

SEM – standard error of mean sla – sex-linked anemia

SDS-PAGE – sodium dodecyl sulfate polyacrylamide gel electrophoresis

TfHp – recombinant human hephaestin with transferrin signal peptide

UV – ultraviolet

YNB – yeast nitrogenous base

YPD – yeast extract peptone dextrose medium

YPDS – yeast extract peptone dextrose medium with 1 M sorbitol

x ACKNOWLEDGEMENTS

I would like to acknowledge involvement of a number of people. Each of you made an irreplaceable contribution to this thesis.

First of all, my supervisor Dr. Ross MacGillivray – thank you for taking a risk of taking up a foreign student. I’m so grateful for your patience and wise advice throughout these years.

My committee members, Drs. LeAnn Howe and Grant Mauk – thanks for your guidance and encouraging attitude to the project. Special thanks to Grant for many fruitful discussions we had.

MacGillivray’s lab members, past and present. Special thanks to Dr. Mark Bleackley who introduced me into the lab environment and spent a lot of time teaching me lab techniques. Also

I would like to thank Dr. Tanya Griffiths for helping me with immunoaffinity chromatography and more importantly for all the hephaestin chats.

Dr. Fred Rosell – for answering my never-ending questions while teaching me to use new equipment.

Dr. Kay Yu – thank you for your help with the column for ceruloplasmin purification.

My family members (and especially my mom) – I know you always believed in me.

Alex – for all the support and encouragement; also for helping me to manage the software I was not familiar with:)

1. INTRODUCTION

1.1 Iron in biology

Iron is an essential element in biological systems. Only members of the Lactobacillus and

Bacillus families can sustain life without iron [1]. The ability of iron to redox-cycle between its

Fe(II) and Fe(III) forms is widely utilized in many biological processes. As a functional component of heme, iron participates in oxygen transport by hemoglobin [2] and drug detoxification by cytochrome P450 in liver [3]. When incorporated into iron-sulfur cluster proteins, iron can mediate mitochondrial electron transfer with the subsequent production of

ATP [4]. As a part of a binuclear site in ribonucleotide reductase, iron serves as an important factor in the synthesis of DNA [5]. In addition to function as a protein cofactor, iron has been implicated as playing a role in the immune response [6].

Unfortunately, iron redox activity can also contribute to the production of hydroxyl radicals (Fenton reaction) and superoxide radicals [7]:

2+ 3+ - (1) Fe +O2→Fe +O2

- + (2) 2O2 +2H →O2+H2O2

2+ 3+ - ● (3) Fe +H2O2→Fe +OH +OH

The hydrogen peroxide used in the Fenton reaction (3) is produced by superoxide dismutase. It converts two superoxide molecules into oxygen and H2O2 (2), which reacts with

● - ferrous iron as above. Because the hydroxyl radical (OH ) and superoxide radical (O2 ) have an unpaired electron on the outer orbital, they can be assigned to the group of reactive oxygen species (ROS). ROS can attack lipids, proteins and DNA, sometimes leading to cancer or cell death [8].

In addition to the high toxicity of Fe(II), the low solubility of Fe(III) is another obstacle for incorporation of iron in biological systems. At neutral pH and physiological oxygen tension,

Fe(II) is readily oxidized into Fe(III). Under these conditions Fe(III) tends to hydrolyze and forms the extremely insoluble Fe(OH)3 complex. Due to the low accessibility of this highly abundant metal, 66-88% of the human population is affected by iron deficiency (World Health

Organization Statistics, 2003).

Since both iron overload and iron deficiency cause cell death, levels of biologically available iron must be tightly controlled. This situation led to the development of elaborate mechanisms of iron acquisition, trafficking and storage in a variety of species: from bacteria to multicellular eukaryotic organisms.

1.2 Prokaryotic iron metabolism

Several mechanisms of iron uptake have been revealed in bacteria. For the acquisition of highly insoluble ferric iron, bacteria secrete low molecular weight extracellular chelators, called

-30 siderophores [9]. Siderophores have very high affinities for ferric ion (Kd<10 ) and help to solubilize iron prior to transport. Gram-negative bacteria take up ferri-siderophore complexes via specific outer membrane receptors in a process that is driven by a cytosolic membrane potential and mediated by the energy-transducing TonB-ExbB-ExbD system [9]. Once internalized, the ferri-siderophore complex is dissociated with subsequent release of bioavailable iron.

Under low oxygen conditions, when ferrous iron predominates over ferric iron, the Feo iron transport system is induced. This system includes FeoB, a membrane-bound ferrous iron transporter, as well as other proteins encoded by feo genes [10] [11]. Ferrous iron transport can be facilitated by extracellular ferriductases, identified in many bacteria [12] [13].

Human pathogens needed to develop additional mechanisms of iron uptake as iron availability in the human body is greatly limited due to the presence of iron-binding (transferrin, lactoferrin) and heme-binding (hemopexin) proteins. Many pathogenic bacteria are capable of

2 binding iron-loaded transferrin and lactoferrin along with heme, hemoglobin and heme- hemopexin complex through specific receptors [14].

Bacterial intracellular reserves of iron are deposited within iron storage proteins such as ferritins, bacterioferritins and Dps proteins. Similar to eukaryotic ferritins, bacterial ferritins are comprised of 24 subunits that form a spherical protein shell of high iron-binding capacity.

Bacterioferritins, also consisting of 24 subunits, contain 12 heme groups, which are believed to mediate iron uptake and release from the protein shell [15]. In contrast to the two other classes of iron storage proteins, Dps proteins are 12-meric and participate in the utilization of hydrogen peroxide [16].

Iron metabolism in bacteria is mostly regulated in response to iron availability. In the presence of Fe2+, Fur (ferric uptake regulator protein) represses transcription of iron uptake genes

[17].

1.3 Yeast iron metabolism

For the last decade, the yeast Saccharomyces cerevisiae was studied extensively as it provides an excellent model of many processes conserved between yeast and humans. In this section S. cerevisiae will be used as an example to describe iron metabolism in yeast. Two mechanisms of iron uptake have been described for S. cerevisiae: siderophore iron uptake and reductase-dependent iron uptake. S. cerevisiae does not produce any endogenous siderophores, but it harbors multiple transporters to scavenge siderophores produced by other species [18].

Reductase-dependent iron acquisition depends on the reduction of Fe(III) to Fe(II) by membrane bound reductases Fre1p or Fre2p [19]. Fe(II) can enter the cells through either low-affinity or high-affinity iron transport systems. To date two low affinity iron transporters have been reported for S. cerevisiae: Fet4p and Smf1p [20; 21]. The high affinity iron transport system consists of the multicopper oxidase Fet3p [22] and the high affinity permease Ftr1 [23]. Iron oxidized by Fet3p is directly accepted by Ftr1p and transported into the cell [24].

Synthesis of heme and Fe/S clusters makes mitochondria the major site of iron consumption, while the function of iron storage is conferred to vacuoles. Yeast vacuoles have been suggested to correspond to human lysosomes [25]. Iron uptake by vacuoles is mediated by the transporter Ccc1p [26]. The mechanism of iron export from the vacuole is similar to reductase-dependent iron acquisition by the cell because it also involves a ferric reductase

(Fre6p) [27], a low affinity iron transporter (Smf3p) [28] and a high-affinity iron transport system (Fet5p and Fth1p) [29].

Iron homeostasis in S. cerevisiae is regulated through two distinct mechanisms. First, regulation of iron metabolism is performed at the level of transcription. Under low-iron conditions, transcription factors Aft1p and Aft2p activate expression of genes encoding proteins required for iron uptake [30; 31]. Recently another iron-responsive transcriptional activator,

Yap5p, was identified with a major role in vacuolar iron homeostasis [32]. The second mechanism involves the mRNA-binding proteins, Cth1p and Cth2p. Under iron-limiting conditions, Cth1p and Cth2p can selectively bind mRNAs that encode proteins of iron-dependent pathways [33]. This binding results in degradation of bound mRNA [33].

1.3 Iron metabolism in humans

1.3.1 Iron absorption in small intestine

Absorption of iron occurs in the proximal small intestine and is mediated by specialized epithelial cells called duodenal enterocytes. Iron can be absorbed from the diet as inorganic iron

(iron salts or chelates) or as a part of heme, which is usually released after digestion of hemoglobin and myoglobin in dietary meat. Recent studies also suggested a significant role of plant ferritins as an iron source in humans [34].

Two candidate heme trasporters were identified in the small intestine: heme carrier protein 1 (HCP1) [35] and heme responsive gene-1 (HRG-1) [36]. HCP1 is a member of a large

4 family of proton-coupled transporters known as the major facilitator superfamily. Both HCP1 and HRG-1 expression was detected in the small intestine, while HRG-1 is also expressed in the brain, heart and kidney [35] [36]. It is not clear yet which of these transporters is predominant in dietary heme uptake. Regardless of the permease used for heme to cross the apical membrane of enterocytes, heme must be degraded for the iron to become metabollically available within the cell. Heme degradation is catalyzed by heme oxygenases and results in release of iron.

Interestingly, induction of heme oxygenase 1 also causes an increase in HCP1 expression, suggesting a connection between uptake and degradation of heme [37].

Transport of inorganic iron is mediated by the divalent metal transporter DMT1 (also known as Nramp2, DCT1 and SLC11A2). DMT1 is a H+/divalent metal symporter that also transports other divalent metals (Zn2+, Cd2+, Mn2+, Cu2+, Co2+, Ni2+ and Pb2+) [38]. In addition to its function as a duodenal iron transporter, DMT1 is also responsible for iron release from endosomes in other cell types (see section 1.3.2). Recent studies of SLC11A2 knockout mice have shown that DMT1 plays a significant role in intestinal iron absorption and iron uptake by erythroid cells, but it is dispensable in placenta and liver [39]. Thus, DMT1 may be the primary means for iron transport, but it is not the sole mechanism.

To make iron available for transport by DMT1, Fe3+ must be reduced to Fe2+.

Ferriductase activity on the apical surface of enterocytes has been attributed to duodenal cytochrome b (Dcytb) [40]. Dcytb is a di-heme protein, which is likely to use ascorbate as an electron donor [41]. In addition to this ferriductase function, Dcytb also has cupric reductase activity, providing a link between iron and copper metabolism [42]. Expression of this ferriductase is increased under conditions of iron deficiency, which serves as strong evidence of a role for Dcytb in iron uptake [40]. On the other hand, loss of this protein in Dcytb-/- mice had little or no effect on body iron stores [43], which may imply involvement of some other ferriductase in duodenal iron absorption.

When iron enters the enterocyte, it can be either stored or utilized for local needs or exported into the blood and delivered to the tissues. The main form of iron storage in the humans is a ferritin encapsulated ferric hydroxide mineral. Structurally, ferritin resembles a cage composed of 24 subunits and is capable of storing up to 5000 iron atoms [44]. There are two types of ferritin subunits: H and L. Only the H subunit possesses ferroxidase activity and catalyses the rate-limiting step of iron incorporation into ferritin [44]. The L-subunit assists the ferroxidase activity of the H-subunit by promoting iron nucleation within the ferritin cavity [44].

The ratio between H and L subunits may vary depending on the cell type and physiological conditions [45; 46].

Mitochondria represent another iron-enriched compartment in the cell. Both heme and

Fe-S clusters are synthesized in mitochondria; this process requires high amounts of iron as well as means for its safe handling. To date two mitochondria-specific iron transporters have been reported: mitoferrin, required for efficient heme biosynthesis in erythroid cells, [47] and mitoferrin 2, expressed in nonerythroid cells [48]. Inside the mitochondrion, iron can be captured by frataxin or mitochondrial ferritin. Due to its ability to bind iron, frataxin can function as an iron-storage protein or iron chaperone during production of heme and iron-sulphur clusters [49].

Mitochondrial ferritin (MtF) has properties similar to the H-subunit of cytosolic ferritin. While cytosolic ferritins are ubiquitous, expression of mitochondrial ferritin is mainly restricted to the testis, neuronal cells and islets of Langerhans [50]. The fact that these tissues are highly sensitive to ROS suggests a role of MtF in protecting mitochondria from iron toxicity.

For iron to exit the enterocyte, it has to be transported by the basolateral permease ferroportin 1 (Fpn1, also known as Ireg, MTP, SLC40A1). Ferroportin is the only known human iron exporter. In contrast to the iron uptake systems, which are ubiquitous throughout the body, only certain cell types have an iron export system. These cells play a major role in iron homeostasis (i.e., duodenal enterocytes, macrophages, placental trophoblasts, hepatocytes and erythroblasts along with cells highly sensitive to ROS attacks (neurons, β-cells in pancreas)). The

Figure 1.1. Transport of iron in the enterocyte. Proteins are shown in abbreviated names. Transporters are shows in blue; ferriductases and ferroxidases are depicted in orange; iron storage proteins are shown in green. Fe2+ in the dashed circle represents labile cytosolic iron pool.

7 observed embryonic lethality of Fpn1 null mice indicated that ferroportin is essential early in development [51]. Selective inactivation of Fpn1 in the small intestine, liver and macrophages caused iron accumulation in these tissues, confirming a unique role of Fpn1 as an iron exporter

[51].

Fpn1 exports iron as Fe2+, but transferrin, the major iron transporter protein of blood, can bind only Fe3+ efficiently. This creates a need for a ferroxidase activity at the site of iron export.

In enterocytes, this ferroxidase activity is associated with hephaestin, a putative multicopper oxidase. The hephaestin ectodomain is highly similar to ceruloplasmin, a major ferroxidase of blood. In contrast to ceruloplasmin, hephaestin has a transmembrane domain, which anchors this ferroxidase to the basolateral surface of enterocyte. Interestingly, GPI-linked ceruloplasmin is colocalized with Fpn1 on the surface of glial cells, producing an export system similar to Fpn1-

Hp in the small intestine [52]. De Domenico et al. [53] have shown that ferroxidase activity stabilizes ferroportin in glial cells by preventing Fpn1 ubiquitination and its subsequent degradation. In the absence of ferroxidase activity, Fpn1 remains bound with Fe2+ which makes it accessible for ubiquitination. Oxidation of Fe2+ or use of Fe2+-specific chelators can abolish this effect [54].

Taking into account the importance of immediate iron oxidation during iron export by ferroportin, it was anticipated that Fpn1 and membrane-anchored ferroxidases physically interact. Indeed, immunocytochemical analysis and immunoprecipitation experiments confirmed an interaction between ferroportin and GPI-linked ceruloplasmin in astrocytes [52] and Fpn1-Hp interaction in enterocytes [55] [56].

1.3.2 Iron uptake in different cell types

Most of the iron in blood plasma is bound by transferrin, a glycoprotein with extremely

3+ -21 high affinity for Fe (KD=10 M) [57]. Under normal conditions, the concentration of transferrin iron-binding sites is greater than the concentration of iron, thereby ensuring a

8 negligible amount of damaging free iron in the blood. Iron binding to transferrin is pH- dependent, which allows efficient iron binding at the neutral pH of plasma and release of iron at the low pH of the endosome, where transferrin is located after internalization [58]. Transferrin endocytosis is mostly mediated by transferrin receptor 1 (TfR1), a ubiquitously expressed protein that binds holotransferrin with an affinity of 109 M-1 [59]. Another transferrin binding protein,

TfR2, is restricted to hepatocytes, duodenal crypt cells and erythroid cells. TfR2 binds transferrin with an affinity 30-fold lower than TfR1 and may play a separate role in the regulation of iron homeostasis [60].

The acidic pH of endosomes stimulates iron release from transferrin with subsequent export of iron into the cytosol by DMT1. Because DMT1 transports only divalent cations and iron released from transferrin is in the Fe(III) form, the existence of an endosomal ferriductase was suggested. In erythroid cells, the main consumers of iron in the human body, this ferriductase function is performed by the protein Steap3 [61]. Ferriductases participating in iron release from endosomes in other cell types have not been reported yet.

Although Tf-dependent iron uptake is probably predominant under normal circumstances, in the case of iron overload (e.g., hereditary hemochromatosis and β-thalassemia), the iron binding capacity of transferrin can be exceeded. This situation results in the appearance of non-

Tf-bound iron (NTBI). Previously known as a zinc transporter, the protein Zip14 was recently found to function as a transporter of NTBI in liver [62].

Megaline and cubilin are multiligand receptors which are primarily expressed in polarized epithelial cells. These proteins are co-expressed in the small intestine, renal proximal tubule and placental cytotrophoblast [63]. Because cubilin does not have any signals for endocytosis, it was proposed that megalin mediates co-internalization of cubilin. Cubilin binds transferrin, while both megalin and cubilin can bind hemoglobin. In the kidney, these binding interactions can be important for minimizing iron losses through the urine. Recently, megaline was suggested to have a new function related to iron homeostasis – binding of lipocalin (also

9 termed neutrophil gelatinase-associated lipocalin, NGAL) [64]. NGAL is capable of binding certain types of bacterial siderophores [65]. By limiting iron availability for pathogenic bacteria,

NGAL works as a bacteriostatic agent [65]. Devereddy et al. also suggested 24p3R as another candidate for the role of a lipocalin receptor [66].

Macrophages play an important role in iron homeostasis by recycling significant amounts of iron through the phagocytosis of old and damaged red blood cells. Furthermore, haptoglobin and hemopexin (blood proteins which show high affinity for hemoglobin and heme, respectively) are endocytosed by macrophages through specialized receptors [67]. Iron recovered after heme degradation inside the macrophage is either held in storage or exported to reload circulating transferrin.

Despite the variety of iron uptake systems described above, biochemical data suggest that additional mechanisms for cellular iron uptake may exist. These mechanisms include iron uptake facilitated by putative ferritin receptors [68; 69; 70] or ceruloplasmin [71].

1.3.3 Regulation of iron homeostasis

In humans, iron metabolism is regulated at both the cellular and systemic levels. At the cellular level, expression of proteins involved in iron homeostasis is modulated by affecting transcription, mRNA stability, translation and posttranslational modifications [72]. Of these processes, posttranscriptional regulation is the best characterized. Iron regulatory proteins 1 and

2 (IRP1 and IRP2) are mammalian proteins that bind iron-binding elements (IRE) in mRNA under iron deplete conditions. IREs in the 5’ untranslated region were identified in mRNAs encoding ferritin chains, erythroid 5-aminolevulinic acid synthase (the first enzyme of heme biosynthesis), mitochondrial aconitase (a citrate cycle enzyme) and one of the ferroportin isoforms [73] [74] [75]. Formation of an IRE/IRP complex in the 5’ UTR inhibits the early steps of translation. On the other hand, binding of IRP at the 3’ UTR of TfR1 mRNA and one isoform of DMT1 stabilizes RNA and enhances translation [76]. The intracellular iron concentration

10 affects the binding of IRP1 and IRP2 through distinct mechanisms. IRP1 senses iron status through an iron-sulfur switch mechanism, alternating between an aconitase form with an iron- sulfur cluster assembled and an apoprotein form that binds IREs. IRP2 activity is regulated primarily by iron-dependent proteosomal degradation in iron-replete cells. Targeted deletions of

IRP1 and IRP2 in animals demonstrated that IRP2 is the main physiologic iron sensor [77]. The central role for IRP-mediated regulation is supported by the early death of mouse embryos lacking both IRP1 and IRP2 [78].

In addition to the intracellular regulation, iron homeostasis has to be coordinated at the organism level. Hepcidin, an iron-regulatory hormone expressed in the liver, is responsible for systemic regulation of iron homeostasis [79]. Upon binding ferroportin, the sole iron exporter in humans, hepcidin induces its internalization and subsequent degradation [80]. Thus, by acting on ferroportin, hepcidin controls the main input of iron to plasma: from duodenal enterocytes absorbing iron, from macrophages involved in the recycling of iron from erythrocytes, and from hepatocytes involved in iron storage.

Hepcidin expression is transcriptionally regulated in response to both intracellular

(hepatic) and extracellular (plasma) signals [81]. Increased concentration of iron inside the hepatocyte leads to increased production of BMP6 protein. The BMP receptor in conjunction with hemojuveline (HJV) binds BMP6 and activates the Smad pathway which results in an increase of hepcidin mRNA level. The concentration of plasma iron is sensed through the interaction of holotransferrin with TfR1 and TfR2. Upon binding of holotransferrin, TfR2 and

HFE enhance the sensitivity of the BMP receptor with a consequent increase of hepcidin production. Other extracellular signals that affect expression of hepcidin are IL-6 and bone marrow-derived factors. The stimulatory effect of IL-6 on hepcidin production results in decreased intestinal iron absorption, iron sequestration in macrophages, and thereby decreases serum iron levels [82]. This has the additional effect of lowering iron availability to pathogens during inflammation. Bone marrow derived-factors are responsible for coordination between

11 hepcidin release from liver and requirements of iron for erythropoiesis [83] [84]. Additional signals that may regulate hepcidin include hypoxia-inducible factors [85].

1.3.4 Inherited disorders of iron metabolism

Numerous mutations in genes encoding proteins of iron metabolism have been reported

(Table 1.1). The resulting dysfunctions of iron homeostasis lead to a variety of human disorders with mild to severe symptoms. While these mutations and associated phenotypes provide valuable insight into the mechanisms of iron homeostasis, they also emphasize the importance of studying iron metabolism for the development of new therapeutics.

Table 1.1. Hereditary disorders associated with iron imbalance.

Gene Function of Disorder Phenotype References the protein DMT1 Ferrous iron Multiple missense Iron deficiency [175; 176; transporter mutations anaemia 177] H-ferritin Iron storage Mutation in 5’UTR Iron loading [178] L-ferritin Iron storage Neuroferritinopathia Brain iron overload [179; 180] Hyperferritinaemia Cataract [181] Frataxin Iron chaperone Freidreich ataxia Mitochondrial iron [182] overloading Ferroportin Ferrous iron Hemochromatosis Plasma [183] exporter type 4 hypoferraemia with tissue iron loading Ceruloplasmin Systemic iron Aceruloplasminaemia Plasma [150] oxidase hypoferraemia with tissue iron loading Transferrin Plasma iron Atransferrinaemia Anaemia refractory to [184; 185] transport iron therapy protein TfR2 Uptake of Hemochromatosis Iron loading [186] transferrin type 3 Regulator of iron homeostasis HFE Regulator of Hemochromatosis Iron loading [187] iron type 1 homeostasis Hemojuveline Regulator of Juvenile Iron loading [188] iron hemochromatosis homeostasis (type 2A) Hepcidin Regulator of Juvenile Iron loading [188] iron hemochromatosis homeostasis (type 2B)

1.4 Multicopper oxidases

Multicopper oxidases are enzymes that oxidize their substrate with the concomitant reduction of dioxygen to two water molecules. Among other copper proteins, the unique feature of multicopper oxidases (MCO) is the presence of at least one of each of the three types of copper sites, type 1, type 2 and binuclear type 3 [86]. This classification of protein copper sites is based on their spectroscopic and magnetic features that reflect the geometric and electronic structure of the active site. A type 1 copper site shows intense absorption at around 600 nm and narrow hyperfine splitting in the electron paramagnetic resonance (EPR) spectrum. A type 2 copper site exhibits no maxima in the visible region of the spectrum and exhibits hyperfine splittings of normal magnitude in the EPR spectroscopy. Unlike type 1 and type 2 copper sites, a type 3 copper site is EPR-silent owing to the strong antiferromagnetic coupling. In UV-visible spectrum, a type 3 copper site exhibits a maximum at 330 nm.

MCOs contain two, three or six cupredoxin domains, which consist of a mixture of antiparallel and parallel β-strands [87]. Three and six-domain MCOs can function as a monomer while two-domain MCOs possess oxidase activity only when assembled as a homotrimer [88;

89]. Most MCOs are comprised of three domains with type 1 copper in domain 3 and a trinuclear cluster at the interface of domains 1 and 3. Type 1 copper serves as an acceptor of electrons from the substrate while the trinuclear cluster, comprised of a type 2 and binuclear type 3 copper atoms, operates as a site of dioxygen reduction to water. Six-domain MCOs, such as ceruloplasmin and as predicted for hephaestin, contain type 1 copper atoms in domains 2, 4 and 6 and a trinuclear cluster at the interface of domains 1 and 6 [90].

Generally, MCOs are promiscuous in regards to their reducing substrate. Aromatic amines and phenols represent substrates of laccases [91] while ascorbic acid oxidase shows specificity towards ascorbic acid [92]. A small group of MCOs designated as metallooxidases exhibit an additional reactivity towards transition metals – Fe2+, Cu+, Mn2+ [106; 107; 108].

1.4.1 Marcus theory of electron transfer

Marcus theory was developed to explain the rate of electron transfer reactions, in which electron is transferred from one chemical species (an electron donor) to another (an electron acceptor) [93]. As confirmed by accumulating experimental data, this theory is applicable towards electron transfer reactions in proteins [94].

The rate of electron transfer depends on four factors, which are represented as terms of the Marcus equation:

3 0 2 4π 2 ⎡− (ΔG + λ) ⎤ k ET = SK A 2 (H DA ) exp⎢ ⎥ h λk BT ⎣ 4λk BT ⎦

These factors and corresponding terms in Marcus equation include [87]:

1. Interaction between the donor and acceptor with probability and lifetime

sufficient for electron transfer; this factor is expressed as the SKA term – the effective

substrate association constant from a productive Michaelis complex.

2. Efficient electronic coupling through the protein, which results in a large

HDA factor (electronic matrix coupling element) – a measure of relative conductivity of

the path followed by an electron on its way from the donor to an acceptor.

3. The change in coordinates of the nuclei (overall bond lengths and angle

change) in the donor and acceptor as a result of their change in redox state. The energy

absorbed or released during these changes is termed the reorganization energy (λ) (also

known as Frank-Condon effect).

4. The thermodynamic driving force which is given as ∆G0 and is directly

related to the difference in reduction potentials of donor and acceptor.

Protein surroundings in MCOs beneficially affect all these parameters and provide efficient pathways for electron transfer.

1.4.2 Type 1 copper sites

Amino acid ligands normally found in the coordination sphere of type 1 copper sites of

MCOs are two histidine and cysteine residues as equatorial ligands and methionyl as an axial ligand; the methionine residue may be substituted by non-coordinating leucyl/phenylalanyl [95].

Multiple studies have shown that the nature of the axial ligand in the type 1 copper center is a strong modulator of the copper reduction potential [96; 97; 98]. Copper coordination by methionine residue results in a relatively low potential, while substitution of methionyl with a non-coordinating residue leads to a significant increase in copper potential [98]. Second sphere ligands have also been suggested to affect the potential of type 1 copper in MCOs [99].

Due to the restraints imposed by a protein on type 1 copper, the geometric properties of this copper site differ significantly from those of inorganic copper complexes. A type 1 copper sites exhibits distorted tetrahedral coordination, which does not correspond to either the preferred geometry of Cu(II) (planar) or of Cu(I) (tetrahedral). Therefore it has been proposed that Cu(I) geometry is imposed on the oxidized site in order to minimize energy expenses of structural rearrangements required for electron transfer (reorganization energy) [100]. The distorted tetrahedral orientation of the type 1 copper centre is commonly referred as an example of an entatic state, characteristic of high energy geometry of a metal site required for efficient electron transfer [100].

Coordination by protein ligands also affects the electronic structure of type 1 copper.

Cu(II) harbors 9 5d-electrons with an unpaired dx2-y2 electron. Significant overlap between the

- dx2-y2 orbital of copper and the Sp orbital of a coordinating cysteine allows for a charge-transfer,

- in which a large fraction of electronic charge of the electronic donor (Cys(Sp )) is transferred to

the electron acceptor (dx2-y2 orbital of type 1 copper). This charge-transfer results in a band of high intensity (ε~5000 M-1cm-1) visible in the absorption spectrum at 600 nm that is responsible for the intense blue color of MCOs [95; 101].

Figure 1.2. Mechanism of O2 reduction to water by the MCOs. Broad arrows indicate the steps that take place in the catalytic cycle of the MCO. Thin arrows indicate steps that can be experimentally observed but are not part of the catalytic cycle.

1.4.3 Transfer of electrons to the trinuclear cluster and dioxygen reduction

Electron transfer from the type 1 copper (Cu1) center to the trinuclear cluster passes through the histidyl-cysteinyl-histidyl (H-C-H) triad, where cysteinyl is a ligand of Cu1 and the histidine residues coordinate Cu3a and Cu3b (binuclear type 3 copper atoms). On its way from

Cu1 to the trinuclear site, an electron has to pass a distance of 13 Å using a through-bond mechanism. An alternative electron path was suggested for ascorbate oxidase that involves a hydrogen bond between the O of C507 and the Nδ of H506 which may serve as a shortcut [102].

Copper atoms of the trinuclear cluster are arranged in a triangular fashion with six histidine residues coordinating the Cu3 pair and two histidine residues coordinate the type 2 copper (Cu2). Cu3a and Cu3b possess inequivalent second sphere ligands. The H-bond network created by a conserved aspartic acid residue lowers of the potentials of Cu2 and Cu3b [103]. This effect allows reduction of a dioxygen in two sequential two-electron steps (Fig. 1.2). First, the fully reduced MCO transfers two electrons to O2 to form a peroxy intermediate [104]. At this stage, the copper atoms with the higher potential (Cu1 and Cu3a) remain reduced. The remaining two electrons are then delivered to the peroxy intermediate to form the native intermediate.

Decay of the native intermediate to H2O proceeds via successive proton assisted steps [105].

1.5 Multicopper metallooxidases

Metallooxidases comprise a group of multicopper oxidases which, in addition to oxidation of organic substrates, can oxidize metal ions. Members of this group include ferroxidases (ceruloplasmin, Fet3p, algal ferroxidase Fox1), cuprous oxidases (CueO) and manganese oxidases (mnxG) [106; 107; 108]. Oxidation of Fe2+ by human ferroxidases is important for the efficient export of iron from cells [109; 110], while fungal and algal ferroxidases were reported to serve as a part of high-affinity iron uptake systems [22; 106].

Bacterial cuprous oxidase has been implicated to have a role in copper detoxification because

Cu2+ is considered to be less toxic than Cu+ [107]. The physiological role of manganese oxidase mnxG, expressed in marine Bacillus spores, is less clear [108].

The domain organization of metallooxidases is highly diverse, suggesting their relative evolutionary remoteness. Fet3p from S. cerevisiae and its homologs from other yeast species are comprised of three cupredoxin domains that possess a trinuclear cluster at the interface of domains 1 and 3, thus resembling the structure of laccases [111; 112]. All known human multicopper ferroxidases contain six domains with the trinuclear cluster coordinated by ligands from domains 1 and 6 [113; 114; 115].The recently identified metallooxidases, Fox1 and mnxG, while containing six domains, have a peculiar feature in regards to their copper-ligands distribution. In contrast to the canonical MCOs, where the trinuclear cluster is located between the N-terminal and C-terminal domains, these metallooxidases have a trinuclear center formed between two internal domains [116].

Binding sites for metal substrates in metallooxidases have many common features which are distinct from those of laccases (MCOs with broad specificity towards various organic substrates). The substrate pocket in laccases is more “open” and has a few steric constraints for substrate binding. In metallooxidases this site is more structurally defined. Metal binding is accomplished by 3 or 4 ligands, which are located in a narrow cleft, thus preventing access to bulky substrates. Acidic amino acid residues are commonly used as metal ligands. These residues create a negative charge at the substrate binding site and facilitate complex formation between metallooxidase and metal ion. The coordination by protein ligands imposes strong restraints on the bound metal, which results in lowering of the potential and decreased reorganization energy of the metal-binding site [117]. Another difference related to the substrate binding to metallooxidases and laccases is the mechanism of transfer of electrons from the the substrate-binding site to the type 1 copper. In Trametes versicolor laccase, this electron transfer can be mediated through H458, which serves both as a substrate-binding residue and a ligand of type 1 copper [118]. In metallooxidases, this process is indirect: the electron is first accepted by

19 metal ligands and then transferred through hydrogen bond(s) to the histidine residue(s) of the type 1 copper coordination sphere [119; 120]. Thus, the substrate-binding site in metallooxidases facilitates electron transfer by 1) effective binding of the substrate (low Km) 2) decreasing the potential of the bound metal which makes it more “oxidizable” 3) decreasing the reorganization energy, and 4) providing an efficient path for electron transfer to type 1 copper (efficient coupling).

1.5.1 Fet3p

Fet3p is a multicopper ferroxidase that comprises a functional part of the high affinity iron uptake system in S. cerevisiae. Fet3p oxidises Fe(II) to Fe(III) thus making iron accessible for further transport into the cell by the ferric transporter Ftr1p. Fet3p has a molecular weight of

62 kDa and is heavily glycosylated with mannose-enriched glycosyl residues [121]. While glycosylation was found to be dispensable for the ferroxidase activity of Fet3p [122], it has an important role in Fet3p trafficking [123].

Iron binding to Fet3p is mediated through a single iron-binding site comprised of residues

E185, D283 and D409 [119; 120] (Fig. 1.3 A). These acidic amino acid residues create a negatively charged cavity on the surface of Fet3p [111] (Fig. 1.3 B). Because of their hydrogen bonding to type 1 copper ligands, H489 and H413, iron ligands E185 and D409 are predicted to participate in electron transfer from Fe2+ to type 1 copper [119] (Fig. 1.3 C). Replacing the iron

2+ ligands of Fet3p with alanine results in significant increase of Km towards Fe [119] and suppression of high affinity iron trafficking [124].

While discovered as a ferroxidase essential for high affinity iron uptake, Fet3p has recently been shown to possess cuprous oxidase activity [125]. All known iron ligands in Fet3p also participate in binding Cu+ although M345 serves as a copper-specific ligand [126]. Similar to Fe2+, Cu+ can support a one-electron reduction processes that result in the production of ROS

[127]. As Cu2+ is regarded as the less toxic form of copper, the cuprous oxidase activity of Fet3p

Figure 1.3. Iron binding by Fet3p. (A) Surface model of Fet3p; predicted ligands of high-affinity iron-binding site (E185, D283 and D409) are shown as sticks. (B) Surface charge distribution on Fet3p. The negative and positive electrostatic potential regions are scaled from red for -80 to blue for +80 kT/e. (C) The iron binding site in Fet3p. Residues E185, D283 and D409 represent iron ligands, residues H413 and H489 are type 1 copper ligands. Arrow shows the putative electron transfer path from the iron atom to the type 1 copper atom. Figures were generated using Pymol software (PDB ID 1ZPU); the electrostatic map was obtained using an APBS plug-in.

21 has been suggested to play an important role in copper detoxification. Thus, oxidation of Cu+ by

Fet3p can compensate for cupric reductase activity of Fre1p on the surface of the yeast cell

[128]. A similar function of copper detoxification was proposed for ceruloplasmin, whose cuprous oxidase activity was revealed recently [125].

1.5.2 Ceruloplasmin

Ceruloplasmin was first purified by Holmberg and Laurell in 1948 [129]. The name

“ceruloplasmin” literally means “a blue substance from plasma”. After discovery of ceruloplasmin’s enzymatic activity, some authors proposed to change its name to “ferroxidase”

[130]. Ceruloplasmin is an abundant glycoprotein in human plasma and is mainly produced in liver [131]. In addition to its soluble form, GPI-anchored ceruloplasmin has been found in glial cells (CNS and retina) and Sertoli cells (testis) [132; 133; 134].

Ceruloplasmin contains six cupredoxin domains and has a high molecular weight of 134 kDa [135]. Type 1 copper centers are located in domains 2, 4 and 6 and a trinuclear cluster is formed between domains 1 and 6. The three-copper cluster is critical not only to the catalytic activity of ceruloplasmin, but also to the structural stability of the protein because it holds together the N- and C-terminal domains of holoceruloplasmin conferring a globular shape to this protein [136]. As revealed by crystal soaking experiments, ferrous binding sites are located in the vicinity of the type 1 copper atoms in domains 4 and 6 [137](Fig. 1.4). The putative iron ligands of ceruloplasmin are buried ~10 Å beneath the protein surface at the bottom of a narrow channel that limits access of bulky organic substrates. Due to the abundance of acidic amino acid residues, these predicted iron-binding sites and the surrounding protein surface possess significant negative charge (Fig. 1.5). Both putative iron-binding sites are comprised of two glutamyl, one aspartyl and one histidyl residues. As shown by near-infrared magnetic circular dichroism (near-IR-MCD), Fe2+ bound by ceruloplasmin is six-coordinated, suggesting two

Figure 1.4. Iron binding by human ceruloplasmin I. (A) The ribbon diagram of human ceruloplasmin. Top view of the molecule along the pseudo-3-fold axis. (B) Side view of ceruloplasmin almost perpendicular to the pseudo-3-fold axis with the putative iron binding site. (C) Iron-binding site in domain 6 of ceruloplasmin. Residues E272, E935, H940 and D1025 represent iron ligands, residue H1026 is a ligand of type 1 copper in domain 6. Arrow shows putative electron transfer path from the iron atom to the adjacent type 1 copper atom. Figures were generated using Pymol software (PDB ID 1KCW).

Figure 1.5. Iron binding by human ceruloplasmin II. (A) Surface model of ceruloplasmin, top view of the molecule; predicted ligands of high-affinity iron-binding sites in domain 4 (E597, H602, D684, E971) and in domain 6 (E935, H940, D1025, E272) are shown as sticks. (B) Same as A; protein loops that cover iron-binding sites are shown as ribbons; groups of residues at the top and at the bottom represent iron-binding sites in domains 6 and 4 respectively. (C) Surface charge distribution on human ceruloplasmin. Top view of the molecule along the pseudo-3-fold axis. (D) Surface charge distribution on human ceruloplasmin. Bottom view of the molecule along the pseudo-3-fold axis. The negative and positive electrostatic potential regions are scaled from red for -76 to blue for +76 kT/e. The crystal structure of ceruloplasmin (PDB code 2J5W) was visualized with Pymol; the electrostatic map was obtained using APBS plug-in.

24 water molecules as additional iron ligands [120].The iron-binding site in domain 6 of ceruloplasmin is comprised of E272, E935, H940 and D1025 with the last three residues contributed by domain 6 and the first one supplied by domain 2. Due to its hydrogen-bonding with H1026, which coordinates the type 1 copper in domain 6, E272 was predicted to participate in electron transfer between the iron-binding site and the adjacent type 1 copper site [120](Fig.

1.4 C). In domain 2 of human ceruloplasmin, the residues that correspond to iron ligands in domains 4 and 6 are two glutamyl, one aspartyl and one tyrosyl residues; these residues are not expected to form a ferrous binding site. In addition, the type 1 copper in domain 2 has a sufficiently high reduction potential that it can not be oxidized without damaging the protein

[138]. While involvement of domain 2 in ferroxidase activity of ceruloplasmin remains unconfirmed, the functionality of iron-binding sites in domains 4 and 6 was recently supported by experimental data [139].

In addition to its role as a ferroxidase, ceruloplasmin exhibits several other catalytic activities. For example, ceruloplasmin was reported to have both NO-oxidase and glutathione- peroxidase activities [140; 141]. The prooxidant site of domain 2 of ceruloplasmin has been implicated in the oxidation of low-density lipoprotein (LDL) [142]. Furthermore, ceruloplasmin is capable of oxidizing an extensive group of organic substrates that includes both xenobiotics

(organic amines) and physiologically relevant substrates (biogenic amines) [143; 144]. The latter group includes hormones (adrenaline, noradrenaline) and neurotransmitters (serotonin, dopamin). Crystal soaking experiments revealed separate binding sites for these two groups of organic substrates. Organic substrates bind ceruloplasmin at domain 4 while the binding site for biogenic amines is located in domain 6 [145].

Although ceruloplasmin has been suggested to possess multiple physiological functions, including roles in copper transport and oxidation of biogenic amines, studies involving aceruloplasminaemia patients revealed the major role of ceruloplasmin in iron metabolism.

Aceruloplasminaemia is an autosomal recessive disease caused by mutations in the

25 ceruloplasmin gene [109]. Most reported mutations result in premature termination of ceruloplasmin mRNA translation [146; 147], while recently found missense mutations affect ceruloplasmin trafficking and copper loading [148; 149]. Overall, the critical physiologic defect in aceruloplasminaemia is the absence of enzymatically-active holoceruloplasmin. Confirming the role of ceruloplasmin in iron export, aceruloplasminaemic patients develop massive accumulation of iron in various tissues, including the liver, pancreas and brain [109; 150]. Long- term iron accumulation leads to diabetes, retinal degeneration and neurologic symptoms in affected individuals [109; 150]. These symptoms can be explained by iron toxicity, which results in free radical damage through the Fenton chemistry [151].

1.5.3 Hephaestin

Hephaestin was first discovered by Vulpe et al. while studying the sex-linked anemia

(sla) mouse [152]. Sla mice develop microcytic hypochromic anemia with iron accumulation in the intestinal epithelium [110], suggesting that while apical iron intake is not impaired, iron export from enterocytes into the blood is blocked. By using positional cloning, the sla candidate gene was identified and named hephaestin after the Greek God of metalworking, Hephaestus

[152].

Hephaestin is predicted to be a transmembrane protein with a molecular weight of approximately 130 kDa [113], and it was first detected in the small intestine [152; 153]. The predicted amino acid sequence of human hephaestin is 50% identical and 68% similar to the sequence of human ceruloplasmin [113]. In contrast to its soluble serum homolog, hephaestin also contains a predicted transmembrane domain at the C-terminus. Based on the known crystal structure of ceruloplasmin, comparative structural modeling of the hephaestin ectodomain revealed that, with the exception of the axial type 1 copper ligand in domain 2, all residues involved in copper binding as well as all cysteinyl residues involved in disulfide bond formation in ceruloplasmin are conserved in hephaestin [113]. Unfortunately, coordinates for this

26 hephaestin model [113] are not available from the Protein Data Base, and the direct request for the PDB file from the corresponding author did not result in any response. In view of the similarity between ceruloplasmin and hephaestin, hephaestin was expected to have ferroxidase activity. This ferroxidase activity of murine and human hephaestin has been confirmed by both in vivo and in vitro experiments [154; 155]. In conjunction with iron transporter ferroportin, hephaestin mediates iron efflux from enterocytes into the blood. By oxidizing ferrous ions, hephaestin promotes iron binding by transferrin and ensures efficient delivery of this metal to the tissues. The physiological importance of hephaestin-catalyzed ferroxidation is illustrated by the sla mice phenotype in which iron export from intestinal epithelium to the circulation is significantly impaired [110]. While hephaestin is mainly expressed in the intestine, this protein has recently been found in the placenta, heart, brain and pancreas [152; 153; 156; 157; 158].

1.5.4 Zyklopen

Zyklopen is another human six-domain multicopper ferroxidase [115]. This protein was detected in multiple tissues with the major site of expression in the placenta [115]. Structurally, zyklopen is expected to be most closely related to hephaestin because both possess a putative transmembrane region at the C-terminus and have identical copper ligands [115; 159]. In domain

6, zyklopen harbors a putative iron-binding site that is comprised of highly conserved residues – one histidine and three acidic residues [115]. The same set of putative iron ligands occurs in domain 6 of hephaestin and ceruloplasmin. As a reflection of its similarity to hephaestin, zyklopen was named after Zyklops, the mythical one-eyed iron workers who helped Hephaestus in the forge of the gods.

1.5.5 Ferroxidases in human iron metabolism

Since the discovery of hephaestin and the more recent identification of zyklopen, ceruloplasmin is not considered to be the unique ferroxidase that facilitates iron export from the cells. The presence of several genes encoding proteins with ferroxidase activity emphasizes the importance of ferroxidation in iron metabolism but also raises the question about the particular function of each ferroxidase.

The distribution of MCOs in the human body is overlapping, with two ferroxidases present at most sites of expression, and all multicopper ferroxidases are expressed in retina [115;

133]. This observation suggests that compensatory relationships exist between the human ferroxidases. Indeed, in Cp-/- mice, hephaestin compensates for the lack of cerulopasmin ferroxidase function in defined regions of the brain [160]. This region-specific compensation may explain iron accumulation associated with certain parts of the brain in aceruloplasminaemia patients [160]. Double knockout mice (Cp-/-Hpsla/Y) represent another useful tool for studying functional cooperation between these ferroxidases. While single knockout mice had a very mild iron loading phenotype [161; 162], a cumulative effect was clearly observed in double knockouts. Cp-/-Hpsla/Y mice developed severe iron overload in the pancreas, heart, brain and retina, suggesting cooperation of ceruloplasmin and hephaestin in these tissues [133; 150]. In contrast, these studies showed that iron efflux from the liver is facilitated solely by ceruloplasmin [150]. The age-dependent changes in phenotype of sla mice may provide another illustration of a compensatory link between ceruloplasmin and hephaestin. Young sla mice have severe anemia with symptoms decreasing with age [162]. Anemia of newborn mice may be explained by insufficient iron feeding of the fetus due to decreased ferroxidase activity of hephaestin in the placenta [163]. While ceruloplasmin is unable to compensate for hephaestin function in placental iron efflux [164], it can promote iron export from enterocytes [165]. This compensatory effect of ceruloplasmin in enterocytes can explain the weakening of anemic symptoms in adult sla mice.

Based on current knowledge, ceruloplasmin is the most multifunctional member of the human MCO family. Ceruloplasmin is also the only ferroxidase that is expressed in a soluble form. Thus, as an abundant protein in blood, ceruloplasmin can perform systemic functions. As a result of its important role in iron export and detoxification, ceruloplasmin transcription is upregulated under conditions of iron deficiency and oxidative stress [166; 167]. Due to ceruloplasmin function as an acute phase protein, its expression is also affected by cytokines such as interferon and interleukin 1β [168; 169]. On the other hand, the expression of hephaestin is regulated by iron in the intestine and in certain parts of the brain [157; 170]; the effect of iron on hephaestin levels in the heart was found to be negligible [156]. In the intestine hephaestin expression is regulated by CDX2, a transcription factor with a key role in intestinal development and differentiation [171]. Expression of both hephaestin and zyklopen is modulated by copper

[115; 172; 173], whereas plasma ceruloplasmin content remains unchanged in copper-deficient rats [174].

In conclusion, ceruloplasmin, hephaestin and zyklopen show distinctive expression patterns and have unique mechanisms for regulating their expression. These features of human multicopper ferroxidases can serve as a basis for precise control of iron efflux in various tissues.

1.5 Statement of hypotheses and objectives

Hephaestin is one of the three known human ferroxidases. Expression patterns of hephaestin are distinct from other ferroxidases, suggesting that this protein plays a unique role in human iron homeostasis. The decreased hephaestin ferroxidase activity of sla mice results in anemic symptoms confirming that hephaestin serves as a key component in iron metabolism.

Despite this important physiological function, our knowledge of the biochemical and biophysical properties of hephestin is very limited. Due to low expression levels, biochemical studies of hephaestin purified from human tissues have never been performed. Previous studies with recombinant hephaestin confirmed its ferroxidase activity, while the molar absorptivity at 600

29 nm that is characteristic of type 1 copper and the total copper content were significantly lower than predicted [155]. To investigate further the structure and functions of human hephaestin, I proposed to test the following three hypotheses:

1. Recombinant human hephaestin expressed in yeast will exhibit enzymatic activity towards ferrous iron and organic substrates.

2. The use of an additional purification step will allow isolation of hephaestin with predicted biochemical properties.

3. Conserved amino acid residues that are predicted to serve as iron ligands play important roles in the ferroxidase activity of human hephaestin.

To investigate the first hypothesis, I expressed the ectodomain of human hephaestin in the yeast Pichia pastoris. The resulting enzymatic activity of recombinant hephaestin was studied using established methods as described in Chapter 3. To test Hypothesis 2, I used ion exchange chromatography for fractionation of recombinant hephaestin that had been immunoaffinity-purified from the BHK cells medium. The resulting hephaestin fractions were analyzed for enzymatic and spectroscopic properties and the copper content of these fractions was determined by ICP-MS. This work is outlined in Chapter 4. Hypothesis 3 was investigated by using site-directed mutagenesis of putative iron-binding residues in human hephaestin. Wild- type recombinant hephaestin and variants were analyzed for ferroxidase and pPD-oxidase activities. This work is described in Chapter 5.

2. MATERIAL AND METHODS

2.1 Materials

Oligonucleotide synthesis was performed by Integrated DNA Technologies (San-Diego,

CA). Enzymes for DNA manipulation and glycosidases were from New England Biolabs

(Beverly, MA). Casamino acids and yeast extract were from Fisher BD (Ottawa, ON). Peptone was from Bioshop (Burlington, ON). Yeast nitrogenous base was purchased from Difco

(Mississauga, ON). The Pichia Expression Kit was from Invitrogen (Burlington, ON). All other reagents were purchased from Sigma-Aldrich (Oakville, ON) unless otherwise noted.

2.2 DNA manipulations

For expression of recombinant human hephaestin (rhHp), I utilized two constructs: one contained the coding sequence for hephaestin with the Fet3p signal peptide (Fet3Hp) and was used for hephaestin expression in the yeast P. pastoris. The other construct was used for production of hephaestin in mammalian cells (BHK cells). This construct encoded hephaestin with the ceruloplasmin signal peptide (CpHp). Plasmid pBSSK- with the modified hephaestin cDNA incorporated via NotI restriction sites [155] was used as an initial construct for DNA manipulation. Modifications previously introduced to the hephaestin cDNA include replacement of the native signal peptide by the transferrin signal sequence and substitution of the predicted transmembrane domain with a factor Xa cleavage site and a 1D4 epitope [189] at the C-terminus.

Due to the presence of the transferrin signal sequence, the initial construct was designated as

TfHppBSSK-.

DNA manipulations employed established protocols of molecular cloning. For propagating DNA, E. coli Mach1 cells were grown in Luria-Bertani medium supplemented with antibiotic. Plasmid DNA was purified with a QIAprep spin miniprep kit (Qiagen, Mississauga,

ON). Automated DNA sequence analysis of all constructs was performed at the B.C. Centre for

Excellence in HIV/AIDS at St. Paul's Hospital, Vancouver, BC using an ABI 3700 DNA sequencer (Applied Biosystems, Streetsville, ON).

2.2.1 Production of the Fet3Hp expression cassette and its integration into the P. pastoris genome

The Fet3 signal sequence was amplified with primers FET3PRO-F and FET3PRO-R (see

Table 2.1) with S. cerevisiae genomic DNA as a template. Using TfHppBSSK- as a template, the

5’ region of the hephaestin sequence was amplified with the primers FETHPLINK and Hp5

(Table 2.1). The resulting PCR products, which both contained 22 bp complementary to the 5’ hephaestin sequence and 25 bp complementary to the 3’ Fet3 signal sequence, were denatured and allowed to re-anneal. This product was used as a template in a subsequent PCR reaction with the primers FET3PRO-F and Hp5 (Table 2.1) resulting in the 5’ hephaestin sequence preceded by the Fet3 signal sequence. This PCR product was used to replace the transferrin signal sequence of the TfHppBSSK- construct to yield a final expression cassette encoding the Fet3 signal peptide, hephaestin ectodomain and a 1D4 epitope. To introduce two different restriction sites for further cloning, the XhoI restriction site within the hephaestin cDNA was mutated with a

Quick Change Mutagenesis Kit (Stratagene, La Jolla, CA) and primers HP-XHO-F and HP-

XHO-R (Table 2.1). The coding region for soluble hephaestin with a Fet3 signal peptide was subcloned into the expression vector pPICZA via XhoI and NotI restriction sites. The resulting construct (Fet3HppPICZA) (Fig. 2.1) was used for yeast transformation. Before transforming yeast cells, the final construct was linearized with PmeI. Cells were plated on YPDS with varying concentrations of zeocin, and transformants from the plate with the highest concentration of zeocin (500 μg/mL) were used for protein expression.

Figure 2.1. Construct used for expression of Fet3Hp (Fet3HppPICZA). Fet3Hp – region encoding encoding the Fet3p signal peptide, hephaestin ectodomain (from A25 to S1070), factor Xa cleavage site and 1D4 epitope on the C-terminus; AOX T – AOX1 transcription termination region; TEF1 P – fragment containing transcription elongation factor 1 promoter; Zeocin – zeocin resistance gene; ori – pUC origin of replication; AOX1 P - AOX1 promoter region.

2.2.2 Production of stably transfected BHK cells for expression of CpHp and its variants

The transferrin signal sequence of the TfHp pBSSK- construct was replaced with the ceruloplasmin signal sequence using the Quick Change Mutagenesis kit (Stratagene, La Jolla,

CA) and primers CpSignal-F and CpSignal-R (Table 2.1). The sequence encoding the hephaestin ectodomain with the ceruloplasmin signal peptide and the 1D4 epitope at the C-terminus was subcloned into a pNUT vector [190] using NotI restriction sites. The resulting construct

(CpHppNUT) (Fig. 2.2) was used for expression of CpHp in baby hamster kidney (BHK) cells.

Prior to transfection with CpHppNUT, BHK cells were grown to confluence in six-well tissue culture plates in Dulbecco’s modified Eagle’s medium-Ham F12 nutrient mixture (DMEM-F12) containing 5% newborn calf serum (NCS) (Invitrogen, Burlington, ON) in a humidified 5% CO2 atmosphere at 37°C. Transfection of BHK cells with recombinant DNA was performed with

FuGene 6 Transfection Reagent according to the manufacturer’s instructions (Roche,

Indianopolis, IN). Briefly, the plasmid and the transfection reagent were combined in serum-free

DMEM-F12 medium. After incubation for 20 min, the resulting lipoplexes were added in a dropwise fashion directly to the cell medium that was bathing the cells. After transfection (24 h), methotrexate was added to the medium to a final concentration of 200 mg/L. Colonies of methotrexate resistant cells were observed after 14 days of growth on selective medium. At confluence, transfected cells were transferred into T25 flasks. The same transfection protocol was used for expression of hephaestin variants. Mutations in the hephaestin sequence were introduced using the megaprimer technique [191] with primers E264A/H269A-R,

E616A/H621A-F, E960A/H965A-F, pNUT5’ and pNUT3’ (Table 2.1). The DNA sequences of all mutants were verified with an ABI 3700 DNA sequencer (Applied Biosystems, Streetsville,

ON).

Figure 2.1. Construct used for expression of CpHp (CpHppNUT). Fet3Hp – region encoding encoding the Fet3p signal peptide, hephaestin ectodomain (from A25 to S1070), factor Xa cleavage site and 1D4 epitope on the C-terminus; GH T – growth hormone transcription termination region; ori – pBR322_ origin of replication; Amp – ampicillin resistance gene; HBV T – hepatitis B virus termination signal; DHFR – dihydrofolate reductase cDNA; SV40 P – simian virus 40 promoter; MT-I P – meallothionein I promoter.

Table 2.1. Oligonucleotides used in this study. FET3PRO-F 5’ GGTACCGGGCCCCCCCTCGAGGATGACTAACGCTTTGCTCT CTATA 3’ FET3PRO-R 5’ CCGGATGCCCAGGTAGTAGACTCGAGTGGCCGCTTGTGCT AGCGAGAGCATC 3’ FETHPNLINK 5’ CTCGATGCTCTCGCTAGCACAAGCGGCCACTCGAGTCTAC TACCTGGGCA 3’ Hp5 5’ ATTGCATGCATCCTATTGCTCTCCTGA 3’ HP-XHO-F 5’ CAAGCGGCCACTAGAGTCTACTACC 3’ HP-XHO-R 5’ GGTAGTAGACTCTAGTGGCCGCTT 3’ CpSignal-F 5’ GTACGCGGCCGCCCCTCGAGGATGAAGATTTTGATACTTGG TATTTTTCTGTTTTTATGTAGTACCCCAGCCTGGGCGGCCACT AGAGTCTACTAC 3’ CpSignal-R 5’ GTAGTAGACTCTAGTGGCCGCCCAGGCTGGGGTACTACAT AAAAACAGAAAAATACCAAGTATCAAAATCTTCATCCTCGAG GGGCGGCCGCGTAC 3’ E264A/H269A-R 5’ GCCATTGATTGCAGCCATCCTATTGCTCGCCTGAAATGTC 3’ D616A/H621A-F 5’ GCTTCCAAGCCTCCAATCGGATGGCTGCCATTAATGGG 3’ E960A/H965A-F 5’ GGATGAAACTTTCTTGGCGAGCAATAAAATGGCTGCAATC AATGGG 3’ pNUT5’ 5’ ACTATAAAGAGGGCAGGCTG 3’ pNUT3’ 5’ AATTTTATTAGGACAAGGCT 3’

2.3 Recombinant protein expression

2.3.1 Expression of Fet3Hp in P. pastoris

In this expression system, the gene of interest was placed under control of a methanol inducible promoter of the yeast AOX1 gene which encodes alcohol oxidase [192]. Stable expression of the recombinant protein was achieved by integration of the expression cassette into the yeast genome via homologous recombination. For efficient selection of transformed cells, the expression cassette contains a zeocin resistance gene controlled by a constitutive promoter. By using increased concentrations of zeocin, clones with high gene copy number can be selected.

Cells grown on solid YPD were used to inoculate BMGY (1% yeast extract, 2% peptone,

100mM potassium phosphate, pH 6.5, 1.34% YNB, 4x10-5% biotin, 1% glycerol). Baffled flasks

(flask volume 500 mL, culture volume 100 mL) were shaken at 30°C (250 rpm) until cells reached an A600 of ~6. Appropriate dilutions were prepared prior to A600 measurements to ensure that absorption values do not exceed the detection limit of spectrophotometer. At this point, cells were harvested by centrifugation with subsequent resuspension in BMMY (1% yeast extract, 2% peptone, 100mM potassium phosphate, pH 6.5, 1.34% YNB, 4x10-5% biotin, 0.5% methanol) containing 300 μM CuSO4, 1% sorbitol, 1% casamino acids (A600 after resuspension of ~1).

Cells were grown in non-baffled flasks (flask volume 2 L, culture volume 0.5 L) at 30°C (200 rpm) for 3 days. To maintain these induction conditions, methanol was added to the yeast cultures to a final concentration of 0.5% every 24 h.

To study the effect of additives and determine the time required for expression of significant levels of Fet3Hp, samples (15 mL) of conditioned P. pastoris medium were collected and concentrated to 300 μL with Amicon Ultra 15 centrifugal ultrafiltration units (30 kDa cut off; Millipore, Billerica, MA). Samples of concentrated supernatant solution (25 μL per well) were subjected to SDS-PAGE (10% acrylamide) and transferred to polyvinylidine difluoride

(PVDF) membranes. PVDF membranes were blocked in 4% skim milk powder in 50 mM

37 phosphate buffer (pH 7.4), 150 mM NaCl, 0.05% Tween-20 (PBST) at room temperature for 40 min. The blocked membranes were incubated for 1 hour with either 1:5000 monoclonal mouse anti-1D4 antibodies [193] or 1:20 000 polyclonal rabbit anti-hephaestin antibodies [158] in 0.5% skim milk in PBST. The membranes were then washed with PBST and incubated with corresponding horseradish peroxidase conjugated secondary antibodies (1:20 000, 0.5% skim milk, PBST) for 60 min. The washed membranes were subjected to detection with Amersham

ECL-Plus chemiluminescent substrate (GE Healthcare, Piscataway, New Jersey) with

ChemiGenius imaging equipment (PerkinElmer). The specificity of antibodies was confirmed by the absence of signal on Western blotting with concentrated medium of cells that were transformed with empty vector.

2.3.2 Expression of CpHp and its variants in BHK cells

The current study uses an established Baby Hamster Kidney (BHK) cell expression system. In this system, recombinant protein is stably expressed by BHK cells transfected with the pNUT plasmid [190]. Expression of the cloned cDNA is controlled by the metallothionein I promoter and growth hormone transcription termination signals. pNUT contains a mutated dihydrofolate reductase cDNA under transcriptional control of the SV40 promoter and hepatitis

B virus termination signals, allowing for methotrexate selection of cells stably expressing recombinant protein.

After transfection, BHK cells were expanded into flasks. For storage, BHK cells were collected by trypsinization followed by centrifugation, transferred into DMEM-F12 with 5%

DMSO and frozen at -80 ˚C. Expanded surface roller bottles (1700 cm2, Fisher Scientific,

Ottawa, ON) were used for large scale production of recombinant proteins. When cells in the roller bottles reached confluence, the DMEM-F12-NCS-methotrexate medium was replaced with

DMEM-F12, 2% Ultroser G (BioSepra, Marlborough, MA) containing 10 µM CuSO4.

Subsequent batches contained DMEM-F12, 1% Ultrocer G, 10 µM CuSO4 (200 ml per roller

38 bottle). The culture medium containing Ultrocer G was collected every 2 days and stored at 4 ˚C prior to purification (storage time did not exceed 4 days).

2.4 Protein purification and storage

This section describes purification protocols utilized for purification of rhHp and ceruloplasmin. Both Fet3Hp and CpHp were purified with immunoaffinity chromatography whereas ceruloplasmin was purified from human plasma using a single-step purification with ethylamine-derivatized Sepharose [194].

2.4.1 Purification of Fet3Hp

Monoclonal antibodies to the 1D4 epitope [193] were coupled to Sepharose 2B by the

CNBr activation method described by Cuatrecasas [195]. Antibodies (~5 mg) were coupled to 5 mL of Sepharose 2B. The column was equilibrated with 20mM Tris-HCl buffer containing 150 mM NaCl, pH 7.4 (equilibration buffer). Yeast medium was clarified by centrifugation (30 min at 8000 rpm using a Beckman JLA-8.100 rotor) and passed through the column with a flow rate of 18 mL per min at 4°C. The column was then thoroughly washed with 20 mM Tris-HCl buffer containing 2 M NaCl, pH 7.4 and equilibrated with equilibration buffer. Elution was performed by repetitive application of 0.1 mg/mL of 1D4 peptide (N-acetylated-TETSQVAPA) (BioBasics,

Markham, ON) [189] diluted in equilibration buffer. The purity of the eluted protein was confirmed by 10% SDS-PAGE followed by staining with 0.1% Coomassie Blue. All Fet3Hp- containing fractions were concentrated using Amicon Ultra-15 centrifugal ultrafiltration units (30 kDa cut off) (Millipore, Billerica, MA) and stored at -20°C until use. As Fet3Hp appeared to be very susceptible to proteolysis, its enzymatic activity and copper content were assayed within 2 weeks of purification.

2.4.2 Purification of CpHp and variants

The immunoaffinity matrix (anti-1D4 antibodies coupled to Sepharose 2B) was synthesized by the method of Cuatrecasas [195]. Tissue culture medium containing CpHp

(freshly collected or stored at 4 ˚C) was centrifuged to remove any cell debris. Protein purification was performed at room temperature. Clarified medium was applied onto the immunoaffinity column equilibrated with 20 mM Tris-HCl, 150 mM NaCl, pH 7.4. The column was then washed with 10 bed volumes of equilibration buffer. CpHp was eluted with 0.1 mg/mL of 1D4 peptide (N-acetylated-TETSQVAPA) diluted in equilibration buffer. When necessary, the eluted CpHp was concentrated with Amicon Ultra-15 centrifugal ultra filtration units (30 kDa cut off) (Millipore, Billerica, MA). As shown by SDS-PAGE, CpHp purified by this procedure was at least 90% pure. For a short period of time (within 1 week), purified CpHp was stored in elution buffer at 4 ˚C. For long-term storage, CpHp solutions with protein concentration of 0.3-

2.5 mg/mL were stored at -20 ˚C. As prolonged storage appeared to adversely affect the enzymatic activity of CpHp, only freshly purified hephaestin was used for activity assays.

The same procedures for purification and storage were used to all CpHp variants. To avoid cross-contamination between different types of hephaestin, separate immunoaffinity columns were used for purification of wild-type CpHp and variants.

2.4.3 Purification of human ceruloplasmin

A sepharose CL-6B derivative with covalently bound ethylamine was synthesized as described [196]. Human blood was collected in a plastic syringe containing a few crystals of

EDTA. The plasma collected after centrifugation was diluted 10-fold with a solution of ε- aminocaproic acid (final concentration 1 g/L) and applied to a Sepharose CL-6B ethylamine affinity column. The column was washed with potassium phosphate buffer (20 mM, pH 7.2) and ceruloplasmin was eluted with 50 mM phosphate buffer, pH 7.2. All purification steps were performed at room temperature. Eluted ceruloplasmin was buffer exchanged quickly against 20

40 mM Tris-HCl, 150 mM NaCl, pH 7.4 using Amicon Ultra-15 centrifugal ultrafiltration units (30 kDa cut off) (Millipore, Billerica, MA) at 4 ˚C. Only freshly purified ceruloplasmin was used for assessment of its catalytic and spectroscopic properties. Based on SDS-PAGE, the purified ceruloplasmin was at least 90% pure. The concentration of ceruloplasmin was calculated from the published extinction coefficient for the absorbance at 280 nm (205 000 M-1cm-1) [135]. The

UV-visible spectrum of ceruloplasmin exhibited a distinct peak at 600 nm with the molar absorption coefficient of 3020 M-1cm-1, which is somewhat lower than reported data (10 000 M-

1cm-1 [135]). Addition of oxidant did not result in an increase of the absorbance, suggesting that all type 1 copper atoms were in their oxidized state. The low ε600nm of purified ceruloplasmin may be caused by high content of apoceruloplasmin, so the catalytic rates of ceruloplasmin determined in this study were adjusted using the following coefficient:

reported ε 600nm 10,000 a = observed = = 3.31 ε 600nm 3,020

The final catalytic rates for ceruloplasmin were calculated from the following relationship:

final observed kcat = a * kcat

2.5 Analytical methods

2.5.1 Protein quantification

Fet3Hp quantification was performed with the Micro BCA Protein Assay Kit (Thermo

Scientific, Rockford, IL) using BSA standards. The concentration of Fet3Hp was calculated based on the mean value of triplicate measurements. When determined by BCA assay, the

Fet3Hp molar absorption coefficient for the absorbance at 280 nm was ~200 000 M-1cm-1, which is consistent with the previously reported value of 215 474 M-1cm-1 for TfHp [155].

Concentrations of CpHp and variants were determined, using this molar absorption coefficient for TfHp. The concentration of ceruloplasmin was calculated from the published extinction coefficient for the absorbance at 280 nm (205 000 M-1cm-1) [135]. A Varian Cary 4000 spectrophotometer (University of British Columbia Laboratory for Molecular Biophysics, www.lmb.ubc.ca) was used for recording UV-visible absorbance spectra.

2.5.2 N-terminal sequence analysis

Prior to electrophoresis, Fet3Hp was deglycosylated with endoglycosidase H (Endo H).

CpHp and deglycosylated Fet3Hp (3 μg of each) were electrophoresed on a 10% SDS-PAGE and transferred to a polyvinylidene difluoride membrane (Biorad, Missisauga, ON) in N-cyclohexyl-

3-aminopropanesulfonic acid buffer, pH 11. The membrane was stained with 0.1% Coomassie

Blue. The bands representing Fet3Hp and CpHp were excised and analysed with an ABI 492

Procise cLC sequencer at the Hospital for Sick Children in Toronto (www.sickkids.ca).

2.5.3 Enzymatic deglycosylation

Prior to deglycosylation, CpHp or Fet3Hp (1 μg of protein) were denatured by incubation in 0.5% SDS, 40 mM DTT at 95 ˚C for 10 min. Deglycosylation with PNGase F (New England

Biolabs, Beverly, MA) was performed in 50 mM Na3PO4, pH 7.5 containing 1% NP-40, 0.25%

SDS and 20 mM DTT. Reaction with PNGase F (final volume 20 μL) contained 1 μg of recombinant hephaestin and 250 U of PNGase F. Deglycosylation with Endo H (New England

Biolabs, Beverly, MA) was performed in 50 mM sodium citrate, pH 5.5 containing 0.25% SDS and 20 mM DTT. Reaction with Endo H (final volume 20 μL) contained 1 μg of recombinant hephaestin and 250 U of Endo H.Both reaction mixtures were incubated at 37 ˚C for 1 hour. The resulting reaction products were visualized by 10% SDS-PAGE after staining with 0.1%

Coomassie Blue.

2.5.4 Far-UV CD spectroscopy

A solution of CpHp was buffer exchanged into 20 mM Tris-HCl, 150 mM NaCl, pH 7.4 using Amicon Ultra-15 centrifugal ultrafiltration units (30 kDa cut off) (Millipore, Billerica,

MA). Far- and near-UV CD spectra of 1 μM CpHp samples were recorded with a Jasco J-810 spectropolarimeter (1 nm slit width and a four second averaging time) (University of British

Columbia Laboratory for Molecular Biophysics, www.lmb.ubc.ca). The cell path length was 2 mm, and the spectrum was scanned from 190 nm to 300 nm. An average of 10 scans was acquired for each sample. The temperature was maintained at 20 ˚C with a Pelletier device. The molar ellipticity was calculated as:

θ MRW []θ = λ λ 10lc

2 -1 where [θ]λ is the molar ellipticity (in mdeg cm dmol ), θλ is the measured ellipticity (in mdeg), MRW is the mean residue molecular weight calculated from protein sequence (for CpHp

MRW is 112.4), l is the cell path length (in cm), and c is the protein concentration (in mg/mL).

2.5.5 UV-visible spectroscopy

Solutions of CpHp or human ceruloplasmin in 20 mM Tris-HCl, 150 mM NaCl, pH 7.4 were loaded into a masked quartz cuvette (light path 10 mm) (Hellma, Concord, ON). UV- visible absorbance spectra were recorded with a Varian Cary 4000 spectrophotometer (25 ˚C) in the 240-900 nm wavelength range.

2.5.6 Inductively coupled plasma mass spectrometry with dynamic reaction cell

(ICP-MS with DRC)

ICP-MS with a dynamic reaction cell was performed by Applied Speciation and

Consulting, LLC (Bothel, WA) according to their established procedures

(www.appliedspeciation.com). Briefly, aliquots of each sample were diluted using an acidic

43 diluent and introduced into a radio frequency plasma where energy transfer process cause desolvation, atomization, and ionization. The ions extracted from the plasma travel through a pressurized chamber (dynamic reaction cell) containing a specific reactive gas that preferentially reacts with interfering ions of the same mass to charge ratios. A solid-state detector detects ions transmitted through the mass analyzer on the basis of their mass-to-charge ratio, and the resulting current is processed by a data handling system. All sample analyses were preceded by a minimum of a five-point calibration curve spanning the entire concentration range of interest.

Ongoing instrument performance was identified by the analysis of continuing calibration verification standards and continuing calibration blanks at a minimal interval of every ten analytical runs. As a negative control, aliquots of the buffer (20 mM Tris-HCl, 150 mM NaCl, pH 7.4) were submitted for analysis together with protein solutions. The copper content of these blank solutions was at least two orders lower than the copper content of the protein samples, confirming insignificant copper contamination of the buffer.

2.5.7 Biquinoline assay for determination of copper content

The copper content of Fet3Hp was determined with a biquinoline assay as described

[197]. Briefly, 75 μL of the sample solution was mixed with 25 μL of 0.8% sodium ascorbate in a 0.6 mL tube. Biquinoline solution (200 μL) (0.04% biquinoline, 0.2% Tritone X-100 in 95% ethanol) was then added. The reaction mixture was thoroughly mixed and 200 μL sample was transferred into the 96-well plate. The absorbance at 535 nm was recorded with a Spectramax190 microplate reader (Molecular Devices). The concentration of copper in the protein sample was determined using a standard curve prepared with CuSO4 solutions. To ensure the complete lability of bound copper, Fet3Hp was heat-denatured prior to copper determination.

2.6 Activity assays

2.6.1 Fet3Hp ferroxidase assay

The ferroxidase assays were performed in 96-well plates with 5 µg of Fet3Hp per well at room temperature. Reactions were performed in triplicate in 75 mM sodium acetate buffer, pH 5

(total reaction volume of 200 µL). A defined amount of ferrous ammonium sulfate was added to each reaction mixture (final concentration 2-12.5 µM). After 20 min incubation with either protein solution or elution buffer, reactions were quenched with 50 µL of 4 mM ferrozine, an

Fe(II)-specific chelator. The ferrozine-Fe(II) complex has an absorbance maximum at 562 nm

-1 -1 (ε562nm=27.9 mM cm [198]) that was measured with a Spectramax190 microplate reader

(Molecular Devices). The concentration of Fe(II) before and after incubation was determined from a calibration curve that was linear in the concentration range of interest. For determination of kinetic parameters, initial kinetic data were analyzed by Eadie-Hofstee plots and linear least squares fit to the data was performed with gnuplot software (www.gnuplot.info).

2.6.2 In-gel assay of Fet3Hp pPD-oxidase activity

An in-gel oxidase assay was performed as described by Chen et al. [154]. Samples of a

Fet3Hp solution containing 30 and 5 μg of purified hephaestin were subjected to native PAGE using a 10% polyacrylamide gel. A constant current of 25 mA was applied to the gel for 45 min.

The part of the gel with 5 μg of Fet3Hp was stained with Coomassie Blue, while the rest of the gel was incubated in 0.1% pPD, 100 mM sodium acetate, pH 5 for 30 min at 37 ˚C.

2.6.3 Oxidation of organic substrates by Fet3Hp

Oxidation of organic substrates was assayed in 250 µL of 80 mM sodium acetate, pH 5 using a 96-well plate. Reactions with organic substrates containing defined amount of Fet3Hp (5 or 10 µg per reaction) were incubated for 24 hours at room temperature. As an autooxidation control, equivalent amounts of elution buffer were used instead of protein solution. Because oxidation of organic substrates used in this study results in the formation of colored products, oxidation was monitored at the wavelength characteristic for absorbance of the product.

Oxidation of p-phenylenediamine dihydrochloride (pPD) was monitored at 562 nm, while oxidation of o-dianisidine dihydrochloride (o-dianisidine) was measured by the increase in absorbance at 535nm. Reaction rates were calculated using extinction coefficients for pPD (1910

M-1cm-1) and o-dianisidine (9600 M-1cm-1) oxidation products [199, 139]. Formation of these oxidation products involves three pPD molecules and two o-dianisidine molecules, respectively.

Thus, the absorbance-based oxidation rates were multiplied by three for pPD and by two for o- dianisidine. The resulting catalytic rates were expressed as the number of molecules of organic substrate oxidized by one molecule of enzyme per minute (min-1). For determination of kinetic parameters, oxidation rates were measured at 7 substrate concentrations ranging from 0.25 to 2 mM for pPD and from 0.05 to 1 mM for o-dianisidine. Gnuplot software (www.gnuplot.info) was employed to fit initial reaction rates into the Michaelis-Menten equation using non-linear regression.

Oxidation of adrenaline and dopamine was monitored at 535 and 450 nm, respectively.

Extinction coefficients for the corresponding oxidation products, adrenochrome (4470 M-1cm-1) and aminochrome (3058 M-1 cm-1), were used to determine the oxidation rates [200]. As a positive control, ceruloplasmin purified from human plasma was used.

2.6.4 Measuring specific enzymatic activity of different CpHp samples

Ferroxidase activity was assayed in 0.75 mM sodium acetate, 10 µM ferrous ammonium sulfate, pH 5, in a total reaction volume 200 µL. CpHp (0.5-5 μg) was added to the reaction mix and reaction was allowed to proceed for 15 min at room temperature. Reactions were quenched with 50 μL of 3 mM ferrozine. The amount of iron oxidized by CpHp was calculated based on the difference in A562 between reactions containing CpHp and elution buffer. One unit of ferroxidase activity is defined as 1 nM of Fe2+ oxidized per min by 1 µg of rhHp.

pPD-oxidase activity was assayed in 0.8 mM sodium acetate, 80 µM EDTA, 1.5 mM pPD, pH 5 (reaction volume 250 µL). CpHp (2.5-5 μg) was added to the reaction mix and reaction was allowed to proceed for 24 hours at room temperature. pPD oxidation rates were calculated based on the absorbance of oxidation product, Bandrowski’s base (1910 M-1cm-1)

[199]. Elution buffer was used as a control to account for substrate autooxidation. One unit of pPD-oxidase activity is defined as 1 nM of pPD oxidized per min by 1 µg of rhHp.

Reactions were performed in 96-well plates and absorbance was recorded with a

Spectramax190 microplate reader (Molecular Devices). All reactions were performed in triplicate and standard error values did not exceed 17% for ferroxidase assay or 20% for pPD- oxidase assay.

2.6.5 Monitoring CpHp ferroxidase activity by formation of ferric iron (III) oxide hydrate

CpHp (final concentration of 1.6 μM) was mixed with 1 mM ferrous ammonium sulfate in 50 mM sodium acetate buffer, pH 5. The reaction mixture (total volume 100 μL) was then loaded into a quartz cuvette. Spectra were recorded every 5 minutes during the first 30 minutes of the reaction at room temperature. As a control to correct for autooxidation, spectra of the reaction mixture without CpHp were recorded at corresponding intervals.

2.6.6 Iron loading of apotransferrin in presence of CpHp

The reaction mixture contained 80 nM CpHp, 10 μM ferrous ammonium sulfate and 55

μM human apotransferrin in 0.2 M acetate buffer, pH 6.0. Reactions were incubated at room temperature. The concentration of apotransferrin was determined from the reported extinction

-1 -1 coefficient for the absorbance at 280 nm (ε280nm=93,000 M cm ). After 4 minutes of incubation, the spectra were recorded and repeated every 2 minutes. As a control to correct for substrate autooxidation, spectra of the reaction mixture without CpHp were recorded at the corresponding intervals.

2.6.7 Discontinuous ferrozine assay for ferroxidase activity of CpHp and its variants

The ferroxidase activity assay was performed in 75 mM sodium acetate buffer, pH 5 with a total reaction volume of 200 µL. A defined amount of ferrous ammonium sulfate was added into each reaction mixture. After incubation with either protein solution or elution buffer, reactions were quenched with 50 µL of 4 mM ferrozine, an Fe(II)-specific chelator. The

-1 -1 ferrozine-Fe(II) complex has an absorbance maximum at 562nm (ε562nm=27.9 mM cm [198]) that was measured with a Spectramax190 microplate reader (Molecular Devices). The concentration of Fe(II) before and after incubation was determined from a calibration curve that was linear over a Fe(II) concentration range of 2-100 µM. For iron concentrations greater than

100 µM, a sample of the reaction mixture was taken and diluted 10-fold prior to quenching with ferrozine. This dilution decreased the absorbance to be within the detection limit of the microplate reader. For determination of kinetic parameters, initial kinetic data was analyzed as

Eadie-Hofstee plots, and a linear least squares fit to the data was performed with gnuplot software (www.gnuplot.info).

2.6.8 pPD-oxidase assay for CpHp and variants

Oxidation of pPD (p-phenylenediamine dihydrochloride) was assayed in a 96-well plate in 80 mM sodium acetate buffer containing 80 µM EDTA, pH 5 (total reaction volume 250 µL).

EDTA was added to the reaction mixture to prevent iron-mediated oxidation of pPD [201; 202].

Reactions were incubated for 24-48 hours at room temperature. Enzyme concentrations was varied over the range of 4-400 nM. As an autooxidation control, equivalent amounts of elution buffer were used instead of the protein solution. The absorption coefficient for Bandrowski’s

-1 -1 base (ε535nm=1910 M cm ), the product of pPD oxidation [199], was used to calculate pPD- oxidation rates. Because production of a single molecule of Bandrowski’s base involves oxidation of three pPD molecules [145], the absorbance-based oxidation rates were multiplied by three, and the resultant catalytic rates were expressed as the number of pPD molecules oxidized

-1 by one molecule of enzyme per minute (min ). For determination of Km and kcat, pPD oxidation rates were measured at a minimum of 8 substrate concentrations ranging from 0.05 to 7.5 mM.

Initial reaction rates were fitted directly into the Michaelis-Menten equation using non-linear regression.

3. CHARACTERIZATION OF RECOMBINANT HUMAN

HEPHAESTIN EXPRESSED IN PICHIA PASTORIS

3.1 Overview and rationale

Despite the implied function of hephaestin in iron metabolism little is known about the biochemical properties of this protein. Previously, the hephaestin ectodomain was expressed in this laboratory using mammalian cells as an expression system. Because of a low copper content

(3.13 copper atoms per molecule instead of the predicted 6) and a low molar absorptivity at 600 nm, this recombinant human hephaestin (rhHp) was proposed to be incompletely loaded with copper [155]. In an attempt to produce fully copper-loaded hephaestin and achieve higher yields of recombinant protein, the use of an alternative expression system was evaluated. Yeast are generally regarded as a robust expression system capable of producing high amounts of post- translationally modified proteins [192]. Furthermore, full-length hephaestin expressed in S. cerevisiae can complement the low-iron growth defect of a ∆fet3 strain that suggests the presence of ferroxidase activity in rhHp [203]. Recently, the hephaestin homolog ceruloplasmin was successfully expressed in fully copper-loaded form in P. pastoris [204]. In addition to the benefits of a yeast expression systems mentioned above, P. pastoris is also known for its high secretion capacity and for not being prone to hyperglycosylate of recombinant proteins as in the case for other yeast [192]. Because of these advantageous features, P. pastoris was selected for the production of soluble hephaestin (the hephaestin ectodomain).

Since its discovery in 1999 [152], most studies of hephaestin as an enzyme were focused on its ferroxidase activity. On the other hand, other members of the MCO family, such as laccases, are well known for their ability to oxidize a wide range of organic substrates. This enzymatic activity is also conserved in multicopper metallooxidases [121; 205]. The hephaestin paralog ceruloplasmin can oxidize both xenobiotic (organic amines) and physiological (biogenic amines) organic substrates [143; 144]. Hephaestin itself was reported to oxidize an artificial 50 organic substrate p-phenylenediamine (pPD) [154], but kinetic constants were not determined.

The oxidation of biogenic amines by hephaestin has never been evaluated.

This chapter describes studies performed with a purified hephaestin ectodomain expressed in P. pastoris. I determined the kinetic parameters of ferroxidase activity and total copper content of recombinant hephaestin. I also studied the specificity of hephaestin towards organic substrates, including physiologically relevant biogenic amines.

3.2 Results

3.2.1 Optimization of Fet3Hp expression

Modifications of the previously reported hephaestin construct [155] resulted in the production of a cDNA encoding the Fet3p signal peptide, hephaestin ectodomain (from A25 to

S1070), factor Xa cleavage site and 1D4 epitope on the C-terminus. This cDNA was subcloned into the pPICZA vector, and the resulting construct was used for transformation of P. pastoris cells. To reach the desired cell density, yeast transformants were grown in buffered glycerol complex medium (BMGY) in the presence of a non-repressing carbon source and then transferred to induction medium containing methanol (buffered methanol complex medium,

BMMY). Because rhHp expressed with this construct contained a Fet3p signal peptide, it was designated as Fet3Hp.

Initially, I used BMMY culture medium for expression of a secreted form of hephaestin.

Detection of the recombinant hephaestin in yeast medium resulted in the identification of two protein bands of MW 35 and 60 kDa on western blotting (Fig. 3.1, lane 1); however, the predicted MW for Fet3Hp is 119 kDa. As this difference could have been the result of proteolysis occurring after secretion into the P. pastoris medium, I attempted to stabilize recombinant hephaestin with casamino acids, which are known to be competitive inhibitors

Figure 3.1. Western blot analysis of P. pastoris medium using polyclonal anti-Hp antibodies. Lane 1, standard BMMY medium; lane 2, BMMY with casamino acids added; lane 3, BMMY with sorbitol added; lane 4, BMMY with both casamino acids and sorbitol added.

52 of proteolysis [192]. Use of casamino acids allowed the detection of a high MW band (Fig. 3.1, lane 2) corresponding to the predicted size of the full-length Fet3Hp. In a further attempt to increase the yield of Fet3Hp, I utilized sorbitol – a non-repressing carbon source in this expression system [206]. Addition of sorbitol to the culture medium led to detection of much stronger signals on Western blots (Fig. 3.1, lane 3), indicating an increased production of

Fet3Hp. As a final medium composition, I used both sorbitol and casamino acids to increase protein production and minimize proteolysis (Fig. 3.1, lane 4). These conditions provided an approximate yield of 50-75 µg of rhHp per liter of yeast medium.

3.2.2 Purification of Fet3Hp

Because of the 1D4 epitope introduced on its C-terminus, Fet3Hp could be readily purified by affinity chromatography as used previously for the hephaestin expressed in mammalian cells [155]. The conditioned medium was applied to an immunoaffinity column containing anti-1D4 antibodies. After the column was washed with equilibration buffer (20 mM

Tris-HCl, 150 mM NaCl, pH 7.4), a 70kDa product was detected in the eluate in addition to a protein consistent with the size of full-length hephaestin (Fig. 3.2 A, lane 1). The 70 kDa protein was also recognized by anti-Hp antibodies (Fig. 3.2 A, lane 2), suggesting that it was a Fet3Hp degradation product. A high salt wash (20 mM Tris-HCl, 2 M NaCl, pH 7.4) allowed the removal of contaminating proteins of lower molecular weight. The bound protein was eluted by repetitive application of 1D4 peptide solution and resulted in the appearance of electrophoretically pure Fet3Hp (Fig. 3.2 A, lane 3).

3.2.3 Initial characterization of Fet3Hp

N-terminal sequence analysis of purified Fet3Hp resulted in the following sequence:

(A/S/G)TRVYY. Despite the ambiguous identity of the first amino acid, the remainder of the

53 sequence (as well as the first residue when alanine is included) corresponds to the predicted N- terminal amino acid sequence of rhHp.

Figure 3.2. Purification and initial characterization of Fet3Hp. (A) Immunoaffinity purification and deglycosylation of Fet3Hp. Lane 1, Protein eluted after column wash with equilibrium buffer, Coomassie stained SDS-PAGE; lane 2, Western blot of lane 1 with anti-Hp antibodies; lane 3, electrophoretically pure Fet3Hp eluted after high salt wash, Coomassie stained SDS-PAGE; lane 4, purified Fet3Hp after treatment with Endo H, Coomassie stained SDS-PAGE. (B) In-gel assay using pPD as a substrate. Fet3Hp (5 and 30 µg) was electrophoresed on native PAGE. Part of the gel containing 5 µg of protein was stained with Coomassie Blue. The remaining part with 30 µg of Fet3Hp was incubated in 0.1% pPD, 0.1 M sodium acetate, pH 5 at 37°C for 30 min. Lane 1, Activity stain with 30 µg of rhHp loaded onto the gel; lane 2, Coomassie stained gel with 5 µg of Fet3Hp. → shows the border between stacking and separating gels.

When analyzed by SDS-PAGE, purified hephaestin electrophoresed as a smear (Fig. 3.2

A, lane 3) suggesting that the protein was heterogeneously glycosylated. It has also been commonly reported that yeast tend to hyperglycosylate recombinant proteins, particularly by producing high mannose structures [206]. To deglycosylate rhHp, I used endoglycosidase H

(Endo H), which is specific for high mannose oligosaccharides from N-linked glycoproteins.

After incubation with Endo H, the smear of potentially hyperglycosylated hephaestin was converted to a single band of the molecular weight predicted for Fet3Hp (119 kDa – see Fig. 3.2

A, lane 4). These results suggest that P. pastoris-expressed recombinant hephaestin contains mannose-enriched glycosyl residues.

To confirm the enzymatic activity of Fet3Hp, I used an in-gel assay with the organic substrate pPD. After known amounts of protein were electrophoresed on a native PAGE, one part of the gel was incubated with pPD and another part of the gel was stained with Coomassie

Blue. As shown in Fig. 3.2 B, the band of amine oxidase activity clearly corresponds to the

Coomassie-stained band of Fet3Hp. This result indicates the absence of contaminating Fet3p activity co-purifying with the Fet3Hp. Interestingly, Fet3Hp activity could only be detected on native PAGE. In the presence of SDS, which is commonly used for assaying in-gel activity of ceruloplasmin and Fet3p [204] [121], the same quantity of hephaestin as used under native conditions did not produce any enzymatically-active band of protein.

3.2.4 Ferroxidase activity of Fet3Hp

The ferroxidase activity of recombinant hephaestin was measured by a discontinuous ferrozine assay. Initial ferroxidation rates of Fet3Hp were measured at substrate concentrations ranging from 2 to 12.5 µM. Under these conditions, Fet3Hp exhibited Michaelis-Menten kinetics for the oxidation of ferrous iron (Fig. 3.3 A). Using an Eadie-Hofstee plot (Fig. 3.3 B), I

2+ determined the Km for Fe as 3.0 ± 0.6 µM and the Vmax as 146 ± 12 nM/min, which corresponds

-1 to a kcat of 0.73 ± 0.06 min .

Figure 3.3. Ferroxidase activity of Fet3Hp. (A) Kinetics of Fet3Hp ferroxidase activity as a function of iron concentration. Ferroxidase activity was assayed using a discontinous ferrozine assay. Reactions were performed in triplicate. Error bars represent SEM. (B) Eadie- Hofstee plot used to determine Km and Vmax for the data shown in A.

3.2.5 Total copper content

The copper content of Fet3Hp was determined by using a biquinoline assay – this assay is based on the formation of a colored complex between 2,2’-biquinoline and Cu(I) that can be detected spectrophotometrically at 535nm. Consequently, this allows the measurement of the total copper content of the sample under denaturing and reducing conditions. The copper content was measured on three independent preparations of Fet3Hp, resulting in the average value of 4.2

± 0.2 copper atoms per molecule. To confirm the validity of this method, one of the Fet3Hp samples was also analyzed for copper content by ICP-MS resulting in a total copper content of

4.8 copper atoms per molecule of Fet3Hp; this agreed favorably with the copper content of this particular sample (4.5 copper atoms per molecule) determined by the biquinoline assay.

3.2.6 Oxidation of organic substrates

Despite the fact that there is a body of evidence regarding hephaestin ferroxidase activity

[154] [155] being essential for iron uptake [152] [110], our knowledge about the hephaestin amine oxidase activity is very limited. Ceruloplasmin is well known for its ability to oxidize a wide range of organic substrates, which include various aromatic amines and phenols [143]. The two substrates most commonly used for detection of ceruloplasmin activity are pPD and o- dianisidine. As shown in Table 3.1, Km values determined for Fet3Hp for both pPD and o- dianisidine are similar to the corresponding values for ceruloplasmin. For kinetic curves see Fig.

3.4; chemical equations of the organic amines oxidation are shown in Appendix A.

Table 3.1. Kinetic parameters for pPD and o-dianisidine for Fet3Hp and ceruloplasmin. Substrates Km for Fet3Hp Km for Cp kcat for Fet3Hp kcat for Cp (mM) (mM) (min-1) (min-1) pPD 1.0 ± 0.2 0.5 ± 0.2 1.10 ± 0.07 81 ± 8 o-dianisidine 0.08 ± 0.03 0.09 ± 0.02 0.10 ± 0.01 8.5 ± 0.5

Figure 3.4. Velocity versus substrate analysis of organic amine oxidation by Fet3Hp and ceruloplasmin. (A) Oxidation of pPD. (B) Oxidation of o-dianisidine. Open circles, Fet3Hp; closed circles, ceruloplasmin. Lines (dashes lines for hephaestin, solid lines for ceruloplasmin) represent best-fit curves for determination of Km and Vmax. All reactions were performed in triplicate. Error bars represent SEM. Reaction rates are presented as nM of substrate oxidized per min.

Figure 3.5. Oxidation of pPD, adrenaline and dopamine by Fet3Hp and ceruloplasmin. Activity of ceruloplasmin (squares, 0.2 µg per reaction) and Fet3Hp (circles; 10 µg per reaction) towards different organic substrates was determined spectrophotometrically after 24 hours of incubation. Solid symbols correspond to reactions performed in absence of EDTA; reactions in presence of 100 µM EDTA are represented by open circles. Rates of oxidation are expressed as nM of oxidation product produced per min. All reactions were performed in triplicate. Error bars represent SEM.

While oxidation of pPD, o-dianisidine and their structural analogs is of questionable physiological significance, biogenic amines such as adrenaline and dopamine represent more significant biological substrates. I analyzed the oxidation of adrenaline and dopamine by Fet3Hp in a range of substrate concentrations. As a control for Fet3Hp enzymatic activity, the same preparation of hephaestin was analyzed for pPD-oxidase activity. To minimize the effect of iron- mediated oxidation, oxidation of biogenic amines was also assayed in the presence of EDTA (see section 3.2.8). In addition, the same analysis was performed with human ceruloplasmin. I have measured the adrenaline oxidase activity of hephaestin (Fig. 3.5 D), while no activity towards dopamine was found (Fig. 3.5 F). Interestingly, the activity of hephaestin towards adrenaline was completely abolished in the presence of EDTA (Fig. 3.5 D), whereas ceruloplasmin retained activity against adrenaline even when EDTA was added to the reaction (Fig. 3.5 C). For chemical equations of biogenic amines oxidation see Appendix B.

3.2.7 Effect of chelators and iron on oxidation of pPD by Fet3Hp

In addition to the direct oxidation of various organic substrates, ceruloplasmin can also utilize an iron-coupled mechanism, where iron serves as an electron mediator between molecules of organic substrate and ferroxidase (Fig. 3.6 A) [201]. To verify the role of iron in the oxidation of organic substrates by Fet3Hp, I compared the oxidation rate for pPD in the presence of EDTA,

1,10-phenanthroline and various concentrations of Fe2+ (Fig. 3.6 B). Chelators inhibited the oxidation of pPD by recombinant hephaestin, while addition of iron enhanced the pPD oxidation rate in a concentration-dependent manner. Both of these effects can be explained by the involvement of iron-coupled oxidation: thus, inhibition by EDTA and 1,10-phenanthroline is caused by chelating trace amounts of iron, and the stimulatory effect of iron is due to its mediator function in the reaction of pPD oxidation by the ferroxidase.

Even in the presence of chelators, Fet3Hp showed detectable pPD-oxidase activity as compared to the autooxidation control (Fig. 3.6 B). The fact that oxidation of pPD by Fet3Hp

Figure 3.6. Iron as a mediator of organic substrates oxidation. (A) Iron-coupled oxidation of organic substrates (modified from [201]). Oxidation of some organic substrates can be significantly enhanced in the presence of iron and ferroxidase. In this system, electron transfer is mediated by Fe3+. Ferroxidase provides fast regeneration of Fe3+ for the next oxidation round. (B) Hephaestin-mediated oxidation of pPD. Reaction mixture contained 0.2 µM Fet3Hp, 1.5 mM pPD and various additives in 100 mM sodium acetate buffer, pH 5.0. Despite the fact that at pH 4- 2- 5.0 the concentration of EDTA is very low, iron can still be chelated by EDTA-H2 , an abundant form of EDTA at this pH. Column 1, in the presence of 100 µM 1,10-phenanthroline; column 2, in the presence of 100 µM EDTA; column 3, no EDTA or Fe2+ added; column 4, in the presence of 1 µM Fe2+; column 5, in the presence of 10 µM Fe2+; column 6, in presence of 100 µM Fe2+. All reactions were performed in triplicate. Error bars represent SEM.

61 was not abolished in the presence of EDTA suggests a direct oxidation mechanism for this substrate and therefore implies the presence of a binding site for pPD within the hephaestin molecule. This is consistent with previous reports that showed that EDTA was unable to inhibit completely the oxidation of substrates oxidized by ceruloplasmin directly (pPD, dopamine, adrenaline), but could abolish the oxidation of iron-dependent pseudosubstrates (such as ascorbic acid) [144].

3.3 Discussion

The hephaestin ectodomain (Fet3Hp) was expressed in P. pastoris in a secreted form and purified from yeast medium. Fet3Hp retained ferroxidase activity confirming the functional integrity of the recombinant protein. As shown by a biquinoline assay, the total copper content of

Fet3Hp produced in P. pastoris is 4.2 copper atoms per molecule. This value is consistent with the hypothesis that hephaestin utilizes a catalytic mechanism similar to Fet3p with one type 1 copper and the rest of the copper atoms forming a trinuclear cluster. On the other hand, I can not exclude the possibility that complete copper loading of Fet3Hp (6 copper atoms per molecule as predicted by computational modeling) has not yet been achieved. [113]. The last hypothesis is supported by the 80-fold lower oxidation rates observed for Fet3Hp for both pPD and o- dianisidine compared to ceruloplasmin (see Table 3.1).

In the second part of this study, I investigated the oxidation of organic substrates by

Fet3Hp. The Km values for hephaestin using both pPD and o-dianisidine as substrates are similar to the corresponding values for ceruloplasmin. These results suggest similarities of binding sites for organic amines between these two homologous proteins.

As shown by crystal soaking experiments [145], ceruloplasmin has a separate binding site for biogenic amines. In contrast to the organic amine binding site located in the vicinity of the type 1 copper in domain 4, the binding site for biogenic amines was positioned in domain 6.

Remarkably, while Fet3Hp was still capable of oxidizing pPD in presence of EDTA, its activity

62 towards adrenaline was completely abolished. This suggests that hephaestin can only oxidize adrenaline indirectly. Together with the absence of activity against another biogenic amine dopamine, this last observation raises two possibilities. Firstly, the type 1 copper that accepts electrons from biogenic amines may be absent in Fet3Hp. Secondly, hephaestin may be missing part of the substrate binding site for biogenic amines. The second possibility may be related to the physiological distribution of hephaestin and ceruloplasmin. Hephaestin is only known to be expressed as a transmembrane protein and thus fulfills its functions locally, while as an abundant protein in serum, ceruloplasmin is able to modulate the levels of biogenic amines in the blood.

3.4 Conclusions

I expressed the hephaestin ectodomain (Fet3Hp) in P. pastoris and purified it to electrophoretic homogeneity by immunoaffinity chromatography. The identity of the purified protein was confirmed by N-terminal sequence analysis. Enzymatic deglycosylation revealed the presence of N-linked high-mannose structures in Fet3Hp. As shown by a biquinoline assay, recombinant hephaestin contained an average copper content of 4.2 copper atoms per molecule.

When analyzed at the range of ferrous concentrations from 2 to 12.5 µM, ferroxidase activity of

-1 Fet3Hp exhibited Michaelis-Menten kinetics (Km=3 µM; kcat=0.73 min ). The Km values of

Fet3Hp for such organic substrates as p-phenylenediamine and o-dianisidine were close to values determined for ceruloplasmin. However, in contrast to ceruloplasmin, recombinant hephaestin was incapable of direct oxidation of adrenaline and dopamine implying a difference in biological substrate specificities between these two homologous oxidases.

4. ENZYMATIC ACTIVITY AND COPPER CONTENT OF

HUMAN HEPHAESTIN

4.1 Overview and rationale

Previously hephaestin was expressed in baby hamster kidney (BHK) cells [155]. The cDNA used for expression of rhHp in BHK cells contained the transferrin signal peptide, the hephaestin ectodomain (including amino acid residues A25 to S1070), a factor Xa cleavage site and a 1D4 epitope. Hephaestin ectodomain with the transferrin signal peptide (TfHp) showed lower ferroxidase activity compared to other multicopper ferroxidases [155]. As determined by inductively coupled mass spectrometry (ICP-MS), the copper content of TfHp was 3.13 copper atoms per molecule instead of the anticipated 6 [155]. Despite the predicted three type 1 copper

-1 centers, the low molar absoptivity of TfHp at 600 nm (ε607nm=2010 M ) implied an average of approximately 0.5 Type 1 copper site per molecule of hephaestin [155]. All these facts indicated incomplete copper loading of TfHp. In this study, I produced the hephaestin ectodomain using a ceruloplasmin signal peptide (CpHp) using BHK cells as an expression system. After immunoaffinity purification, CpHp was analyzed for both pPD- and ferroxidase activity; far-UV

CD spectra of purified CpHp were also recorded.

The distinct biophysical properties of apo- and holo-ceruloplasmin allow separation of apo- and holo-forms of ceruloplasmin with ion exchange chromatography (IEC). During elution with increasing salt concentration, holoceruloplasmin was eluted first, while the apo-component was recovered at the higher ionic strength [207]. In attempt to isolate a fully copper-loaded form of hephaestin, I applied a similar procedure to the purification of CpHp. Fractions of CpHp eluting at different salt concentrations were analyzed for enzymatic activity, spectroscopic properties and copper content.

4.2 Results

4.2.1 Expression, purification and initial characterization of CpHp

The hephaestin ectodomain with a ceruloplasmin signal peptide (CpHp) was expressed in

BHK cells as described previously [155]. Recombinant hephaestin was secreted into the tissue culture medium with the yield of 0.5-3 mg of CpHp per liter of the medium. The 1D4 epitope at the C-terminus of CpHp allowed the purification of recombinant hephaestin in a single immunoaffinity chromatographic step. To confirm the purity of eluted hephaestin, 1 µg of freshly purified CpHp was electrophoresed with 10% SDS-PAGE. As a control, equal amount of

TfHp was loaded on the same gel. Protein amounts were quantified based on absorbance at 280 nm using the extinction coefficient of 215 000 (as determined previously for TfHp [155]). After

Coomassie staining of the gel, both CpHp and TfHp showed the same apparent MW and intensity of protein bands (Fig. 4.1 A). These results confirm that ε280 for TfHp is applicable to

CpHp.

As shown by SDS-PAGE, treatment of CpHp with PNGase F resulted in the shift of apparent MW from ~ 150 kDa to ~120 kDa, while the use of Endo H did not result in a mobility shift (Fig. 4.1 B). PNGase F cleaves a wide range of N-linked glycosyl residues from glycoproteins, whereas Endo H has specificity towards asparagine-linked mannose rich oligosaccharides. Thus, similar to TfHp [155], CpHp is glycosylated with hybrid or complex oligosaccharides.

N-terminal sequence analysis of CpHp revealed the following sequence:

(A/S/G)TRVYY. Despite the ambiguous identity of the first amino acid, the remainder of the sequence (along with the first residue when alanine is included) corresponds to the amino acid sequence of human hephaestin. In contrast to TfHp, which contained additional amino acids from the transferrin signal peptide [155], the hephaestin used in this study had an authentic N- terminal sequence.

Figure 4.1. SDS-PAGE of CpHp and TfHp. After running 10% SDS-PAGE, gels were stained with Coomassie Blue. (A) CpHp and TfHp purified with immunoaffinity chromatography. Lane 1, protein MW marker; lane 2, 1 µg of CpHp; lane 3, 1 µg of TfHp. (B) Enzymatic deglycosylation of CpHp. Each lane contains 1 µg of CpHp. Lane 1, CpHp prior to deglycosylation; lane 2, CpHp after treatment with Endo H; lane 3, CpHp after treatment with PNGase F.

4.2.2 Correlation between pPD- and ferroxidase activities in different CpHp samples

When the ferroxidase and pPD-oxidase activities of CpHp were analyzed, it appeared that different samples of hephaestin (i.e. hephaestin purified from different batches of the medium) had significant variability in specific activity. Correlation between pPD-oxidase and ferroxidase activity of CpHp samples was observed (Fig. 4.2 A), suggesting that both types of enzymatic activity originate from the same protein species. Kinetic analysis of several CpHp samples revealed that kcat for pPD varied 4-fold while Km values were consistent throughout the samples

(Fig. 4.2 B). These results can be explained by variation in the amount of enzymatically active hephaestin in the different samples.

Figure 4.2. Enzymatic activity of different CpHp samples. Specific ferroxidase and pPD-oxidase activity of the samples is expressed as the number of activity units (U) (for unit definition and assay conditions see section 2.6.3). (A) Correlation between ferroxidase and pPD- oxidase activity in different samples of CpHp. All activity assays were performed in triplicate with the mean values shown on the plot. Error bars represent SEM. (B) Kinetic analysis of pPD oxidation. CpHp samples subjected to analysis are shown in black; the rest of the samples are in grey.

4.2.3 Ferroxidase activity of CpHp monitored by product formation

As an alternative to the ferrozine assay, iron oxidation by CpHp was assayed by monitoring the formation of iron (III) oxide hydrate spectrophotometrically [208]. A sample of

CpHp (343 U of ferroxidase activity, 228 U of pPD-oxidase activity) at a final concentration of

1.6 μM was incubated with 1 mM ferrous ammonium sulfate in 50 mM sodium acetate buffer, pH 5 at room temperature. Spectra were recorded every 5 minutes during the first 30 minutes of the reaction. Corresponding autooxidation controls (in the absence of CpHp) were subtracted; under the assay conditions autooxidation of Fe2+ was negligible. Plotting absorbance at 315 nm versus time showed that CpHp ferroxidation rates remained constant for the time period studied

(Fig. 4.3).

Using urea-PAGE it has been shown that recombinant hephaestin (TfHp) can facilitate iron loading of apotransferrin [155]. In the current study, I utilized a spectrophotometric approach to measure the extent of holotransferrin formation in the presence of CpHp. Upon binding of Fe3+ transferrin develops a distinguishable peak in the visible spectrum (maximum absorption at 460 nm) [209]. The reaction mixture contained 80 nM CpHp (280 U of ferroxidase activity, 174 U of pPD-oxidase activity), 10 μM ferrous ammonium sulfate and 55 μM human apotransferrin in 0.2 M acetate buffer, pH 6.0. After 4 minutes of incubation the spectra were recorded; then spectra measurments were repeated every 2 minutes. As a control of autooxidation, spectra of the reaction mixture without CpHp were recorded at the corresponding time points. Despite the significant autooxidation of ferrous iron observed under assay conditions, iron loading of apotransferrin was noticeably enhanced in the presence of CpHp.

Upon depletion of ferrous iron in the reaction, rates of hephaestin-mediated holotransferrin formation reduced dramatically (Fig. 4.4).

Figure 4.3. The formation of iron (III) oxide hydrate in presence of CpHp. (A) Spectra recorded at different time points. ··········· - 5 minutes; · · · · · · - 10 minutes; ------15 minutes; – ·–·–·–·– - 20 minutes; – – – – - 25 minutes; –––––– - 30 minutes. Corresponding autooxidation controls were subtracted. (B) A315 versus time plot. Accumulation of ferric oxide hydrate was monitored at 315 nm at different time points.

Figure 4.4. Iron loading of apotransferrin in presence of CpHp. (A) Spectra recorded at different time points. ··········· - 4 minutes; · · · · · · - 6 minutes; ------8 minutes; –·–·–·–·– - 10 minutes; – – – – - 12 minutes; –––––– - 14 minutes. Corresponding autooxidation controls were subtracted. (B) A460 versus time plot. Iron loading of transferrin was monitored at 460 nm at different time points.

Figure 4.5. CD spectroscopy of CpHp. –––––– - CD spectrum of native CpHp; – – – – - CD spectrum of heat denatured CpHp.

4.2.4 Far-UV CD spectrum of CpHp

A solution of CpHp (42 U of pPD-oxidase activity) was dialyzed into 20 mM Tris-HCl,

150 mM NaCl, pH 7.4. The far-UV CD spectrum of native CpHp (1 μM solution) was recorded at 20°C. CpHp was then incubated at 95°C for 10 minutes. Centrifugation of the heat-denatured

CpHp sample did not result in the formation of a visible pellet suggesting that the protein did not precipitate as a result of the heating step. After the solution of heat-denatured hephaestin cooled down to 20 ˚C, the CD spectrum was recorded. Remarkably even after exposure to 95°C CpHp retained a significant quantity of structured motifs suggesting presence of a stable protein core in hephaestin molecule (Fig. 4.5).

4.2.5 CpHp separation with ion exchange chromatography

Purified hephaestin was loaded onto a column of DEAE-Sepharose in a low salt buffer (5 mM Tris-HCl, 15 mM NaCl, pH 7.4). After unbound protein was removed with the same low salt buffer, the salt concentration in the elution buffer was increased in a stepwise mode.

Fractions of hephaestin eluting at different salt concentrations were dialyzed against 20 mM

Tris-HCl, 150 mM NaCl, pH 7.4 and concentrated. SDS-PAGE confirmed that only full-size hephaestin was eluted (Fig. 4.6 A).

Analysis of specific pPD-oxidase activity has shown that the fraction of hephaestin eluting at 150 mM NaCl had higher activity compared to the CpHp sample prior to IEC and other fractions eluting at higher salt concentrations (Fig. 4.6 A). I performed a kinetic analysis of this highly active fraction. Initial ferroxidation rates were measured at iron concentrations ranging from 2 to 12.5 µM. Kinetic analysis revealed the following parameters for IEC-purified

2+ -1 CpHp: 1) for Fe : Km = 3.8 ± 1.0 µM and kcat = 21 ± 2 min 2) for pPD: Km = 1.5 ± 0.2 mM and

-1 kcat = 29 ± 1 min . IEC of CpHp was performed in two independent experiments, and in both cases specific activity of hephaestin fraction eluting at 150 mM was significantly higher than activity of unfractionated CpHp samples (Fig. 4.6 B).

To detect type 1 copper sites, I measured the UV-visible absorption spectra of hephaestin fractions eluting at different salt concentrations. The absorption spectra of protein solutions were determined against a blank comprised of dialysis buffer (20 mM Tris-HCl, 150 mM NaCl, pH

7.4). All fractions showed a distinguishable peak at 600 nm characteristic for type 1 copper, while the hephaestin fraction eluted at 150 mM NaCl had the highest absorption coefficient

-1 -1 ε600nm=5500 M cm (Fig. 4.7). As only Cu(II) is responsible for absorption at 600 nm, all type 1 copper atoms have to be fully oxidized for the accurate determination of ε600nm. There are two sets of evidence supporting complete oxidation of CpHp type 1 copper atoms. First, type 1 copper atoms in ceruloplasmin achieve full oxidation at a neutral pH in presence of 150 mM chloride anion [210]. Thus, conditions used during recording CpHp UV-visible spectra should

Figure 4.6. Fractionation of CpHp by chromatography on DEAE-Sepharose. (A) The plot (at the top) shows specific pPD-oxidase activity of hephaestin fractions eluted at different salt concentrations (Error bars represent SEM). SDS-PAGE confirmed the electrophoretic homogeneity of CpHp fractions. Coomassie staining of the gel (at the bottom) revealed only single protein bands with apparent MW of 150 kDa. Lane 1 - CpHp prior to chromatography; lane 2 - CpHp fraction eluted at 150 mM NaCl; lane 3 - CpHp fraction eluted at 200 mM NaCl; 4 - CpHp fraction eluted at 250 mM NaCl. (B) Correlation between the specific pPD- and Fe2+- oxidase activities of different CpHp samples. ● - samples of unfractionated CpHp; ♦ - CpHp samples eluting during IEC at 150 mM NaCl (represent two independent experiments).

73 support the fully oxidized state of type 1 copper. Second, addition of oxidant (ferricyanide) to

CpHp did not result in the increase of absorbance at 600 nm confirming the absence of any reduced type 1 copper atoms under conditions tested.

The copper content of hephaestin fractions was determined by ICP-MS analysis, which resulted in the following values (expressed as the number of copper atoms per molecule): 2.43 ±

0.02 for CpHp prior to IEC, 3.5 ± 0.1 for CpHp eluting at 150 mM NaCl, 3.5 ± 0.1 for CpHp eluting at 200 mM NaCl, 3.2 ± 0.1 for CpHp eluting at 250 mM NaCl. Thus, while differing in enzymatic activity and spectroscopic properties, hephaestin fractions did not show a significant difference in copper content. This could be caused by the adventitious binding of non-functional copper to CpHp.

Figure 4.7. Spectroscopic properties of CpHp fractions produced by chromatography on DEAE-Sepharose. Absorption spectra of protein solutions were determined against a blank comprised of dialysis buffer (20 mM Tris-HCl, 150 mM NaCl, pH 7.4). Absorbance is given as the molar absorptivity calculated from spectra normalized for protein concentration. ─── - CpHp fraction eluted at 150 mM NaCl; − − − - CpHp fraction eluted at 200 mM NaCl; · · · · · - CpHp fraction eluted at 250 mM NaCl.

4.3 Discussion

Long-term observations showed that CpHp samples purified from different batches of medium exhibited significant difference in specific enzymatic activity. It was demonstrated that the ferroxidase activity of the samples was directly proportional to pPD-oxidase activity.

Furthermore, the Michaelis constant for pPD was consistent for CpHp samples with different kcat.

These results suggest that both types of activity originate from enzymatically active form(s) of

CpHp, and variations in kcat result from the variations in the amount of this functional hephaestin in different CpHp samples. This means that, in addition to enzymatically active rhHp, BHK cells also produce CpHp with impaired enzymatic function, which may result from the incomplete copper loading or misfolding of hephaestin.

To isolate the enzymatically most-active form of hephaestin from the gross CpHp produced by BHK cells, I utilized IEC. After elution with increasing salt concentration, I obtained several hephaestin fractions. CpHp with the highest pPD-oxidase activity and molar absorptivity at 600 nm eluted at the lowest salt concentration; as the salt concentration in elution buffer was increased, both the enzymatic activity and ε600 of the eluting hephaestin were gradually decreased. Based on these observations, the CpHp fractions were anticipated to contain different amounts of copper-loaded hephaestin. Surprisingly, the fractions of CpHp produced during IEC did not show significant differences in copper content. The last fact may be explained by the presence of non-prosthetic copper, which has been commonly reported for ceruloplasmin [211; 212].

Using separation by IEC I produced recombinant hephaestin with improved catalytic and

2+ spectroscopic properties. While determined under similar conditions, the kcat for Fe of IEC- purified CpHp (18 min-1) was higher than the ferroxidation rates reported for Fet3Hp (0.74 min-1

-1 -1 -1 [202]) and TfHp (2.5 min [155]) CpHp absorbance at 600 nm (ε600nm=5500 M cm ) suggests the presence of one type 1 copper per molecule of hephaestin. These data disagree with the structural model of hephaestin ectodomain based upon ceruloplasmin. According to this model,

75 hephaestin contains three putative type 1 copper sites. Furthermore, the total copper content of hephaestin isolated in this study is 3.5 instead of predicted 6. One explanation for these results is incomplete copper loading of the hephaestin preparation. Otherwise, hephaestin may contain only 4 copper atoms arranged as a single type 1 copper and a trinuclear cluster.

4.4 Conclusions

I expressed the hephaestin ectodomain with ceruloplasmin signal peptide (CpHp) using

BHK cells as an expression system. The N-terminal sequence of CpHp corresponded to the amino acid sequence of human hephaestin. Enzymatic deglycosylation revealed that CpHp contained highly processed N-linked oligosaccharides. Ion exchange chromatography of purified

CpHp resulted in the production of hephaestin fractions with distinct catalytic and spectroscopic properties. The fraction of CpHp with the highest enzymatic activity also showed enhanced molar absorptivity at 600 nm, characteristic for type 1 copper. Despite the differences in specific enzymatic activity and molar absorptivity of type 1 copper sites, all hephaestin fractions had similar copper content ranging from 3.2 to 3.5 copper atoms per molecule.

5. OXIDATION OF FERROUS IRON BY HUMAN HEPHAESTIN

5.1 Overview and rationale

For a long time, ceruloplasmin has been known for its unusual ferroxidation kinetics.

When the ferroxidase activity of ceruloplasmin was analysed using a wide range of substrate concentrations, an Eadie-Hofstee plot of these kinetic data corresponded to a biphasic curve rather than a line predicted by the Michaelis-Menten’s equation. This effect was observed in human ceruloplasmin using different activity assays [213; 214]; later, similar results were obtained while characterizing the ferroxidase activity of rat ceruloplasmin [215]. The biphasic shape of the Eadie-Hofstee plot suggests that ceruloplasmin harbors two types of iron-binding sites with different affinities for ferrous iron. The presence of high- and low-affinity ferrous binding sites is consistent with the results of equilibrium dialysis studies, which revealed a biphasic profile of Eadie-Scatchard plot for iron binding by ceruloplasmin [216]. Taking into account the high level of amino acid identity between ceruloplasmin and hephaestin, human hephaestin was expected to show similar kinetics of iron oxidation. Previously, hephaestin rates of ferrous oxidation were determined only at low iron concentrations (at 2-12.5 μM range of Fe2+ concentration) [155; 202]. Under these conditions, hephaestin exhibited Michaelis-Menten type kinetics for the oxidation of Fe2+. In attempt to uncover additional kinetic features of hephaestin,

I analyzed the ferroxidase activity of CpHp using wide range of substrate concentrations (2-750

μM).

Studies of ceruloplasmin metal-ion soaked crystals identified two putative cation-binding sites in the vicinity of type 1 copper centers in domains 4 and 6 [114; 137]. Binding of Fe2+ at these sites has been proposed as the first step in the ferroxidase catalytic mechanism [137]. Due to their structural similarity with the high-affinity ferrous binding site in Fet3p [120], these putative iron-binding sites of ceruloplasmin were expected to show high affinity for ferrous iron.

To further study the involvement of putative iron ligands in the ferroxidase activity of

77 ceruloplasmin, residues proposed to be involved in iron binding were altered with site-directed mutagenesis [139]. Both ferroxidase and amine oxidase activities of recombinant wild-type ceruloplasmin and variants were assayed using a single substrate concentration [139]. A double variant of putative iron ligands in domain 4 (E597A/H602A) exhibited a significant reduction in both activities compared to recombinant wild-type ceruloplasmin, whereas a domain 6 iron ligands variant (E935A/H940A) had unaltered amine oxidase activity with ferroxidase activity reduced by 50% [139]. The last fact confirms the important role of domain 6 iron-binding site in the ferroxidation mechanism of ceruloplasmin. Hephaestin retained the conserved iron-binding site in domain 6 (comprised of three acidic residues and histidine residue), while corresponding residues in domains 2 and 4 of human hephaestin are represented by histidyl, serinyl and two acidic residues (Table 5.1, Fig. 5.1).

Table 5.1. Putative iron ligands in hephaestin and ceruloplasmin Putative iron ligands Domain 2 Hp E264 H269 S351 E652 Cp E236 Y241 N323 E633 Domain 4 Hp D616 H621 S703 D996 Cp E597 H602 D684 E971 Domain 6 Hp E960 H965 D1050 E300 Cp E935 H940 D1025 E272

Figure 5.1. Partial amino acid sequence alignment between human hephaestin and ceruloplasmin. Only the putative iron ligands and adjacent amino acids are shown. Predicted iron binding residues are shown in red, conserved residues are shown in green.

Hp: LFGMGN E300 IDVH...DETFL E960 SNKM H965 AING...HCHVT D1050 HVHAGM Cp: LFGMGN E272 VDVH...DEEFL E935 SNKM H940 AING...HCHVT D1025 HIHAGM

To investigate the involvement of putative iron ligands in the ferroxidase activity of human hephaestin, I mutated conserved amino acid residues of CpHp in each of the three predicted iron-binding sites. I then performed kinetic analyses of ferroxidase and amine oxidase activities for wild-type and variant CpHp.

Figure 5.2. Ferroxidase activity of recombinant hephaestin. (A) Velocity versus substrate plot for CpHp. Initial ferroxidation rates were determined using a discontinuous ferrozine assay. The specific activity of CpHp used in this assay was estimated as 95 U of pPD- oxidase activity and 173 U of ferroxidase activity. (B) Kinetic data from A plotted in Eadie- Hofstee coordinates. The Eadie-Hofstee plot was used to determine kinetic parameters for high- and low-affinity ferrous binding sites. – – – – kinetic curve and line on Eadie-Hofstee plot modeled for the high-affinity site; · ― · ― · kinetic curve and line on Eadie-Hofstee plot modeled for the low-affinity site. All reactions were performed in sextuplicate. Error bars represent SEM.

Figure 5.3. Ferroxidase activity of human ceruloplasmin. (A) Velocity versus substrate plot for ceruloplasmin. Initial ferroxidation rates were determined using a discontinuous ferrozine assay. (B) Kinetic data from A plotted in Eadie-Hofstee coordinates. The Eadie- Hofstee plot was used to determine kinetic parameters for high- and low-affinity iron-binding sites of ceruloplasmin. – – – – kinetic curve and line on Eadie-Hofstee plot modeled for the high- affinity site; · ― · ― · kinetic curve and line on Eadie-Hofstee plot modeled for the low-affinity site. All reactions were performed in sextuplicate. Error bars represent SEM.

5.2 Results

5.2.1 Kinetic analysis of iron oxidation by hephaestin and ceruloplasmin

Previously, the ferroxidase activity of hephaestin was analyzed only at low substrate concentrations [155; 202]. Under these conditions, hephaestin-catalyzed ferroxidation exhibited

Michaelis-Menten type kinetics. Using a discontinuous ferrozine assay, I determined hephaestin ferroxidation rates at a wide range of iron concentrations. As a control, I analyzed the ferroxidase activity of ceruloplasmin under the same assay conditions. Both ferroxidases exhibited non-

Michaelis-Menten kinetics, which is especially noticeable when the kinetic data are plotted in

Eadie-Hofstee coordinates (Fig. 5.2 and 5.3). For both hephaestin and ceruloplasmin, Eadie-

Hofstee plots of initial kinetic data corresponded to a biphasic curve suggesting the presence of two types of iron-binding sites (high-affinity and low-affinity). Based on the Eadie-Hofstee plots, kinetic parameters of iron oxidation were calculated for each type of ferrous binding sites

(Table 5.2).

Table 5.2. Kinetic parameters of iron oxidation by hephaestin and ceruloplasmin. Ferroxidase Km for high- Km for low- Vmax for Vmax for Ratio between affinity sites, affinity high-affinity low-affinity Vmax for high- µM sites, µM sites, sites, and low-affinity nM/min nM/min sites CpHp 3.5 ± 0.2 107 ± 13 67 ± 2 286 ± 16 4.3 Ceruloplasmin 4 ± 0.5 94 ± 11 81 ± 4 339 ± 17 4.2

The Michaelis constants for the corresponding iron-binding sites in CpHp and ceruloplasmin did not show significant difference. The Km values for high-affinity sites were determined as 3.5 ± 0.2 μM for hephaestin and 4 ± 0.5 μM for ceruloplasmin. These values are consistent with the previously reported Michaelis constant for Fet3p (2 μM [121]), which harbors a single high-affinity iron binding site. For the low-affinity sites, CpHp showed a Km of 107 ± 13

μM while the corresponding Km in ceruloplasmin was 94 ± 11 μM. In addition to the similarity of Km values, both ferroxidases showed comparable ratios between the Vmax for low- and high- affinity ferrous binding sites (4.3 for hephaestin and 4.2 for ceruloplasmin).

5.2.2 Strategy for site-directed mutagenesis of putative iron ligands

Hephaestin has three putative iron-binding sites in domains 2, 4 and 6. To study the role of each binding site in the ferroxidase activity of hephaestin, I produced the following variants:

IB (E264A/H269A; D616A/H621A; E960A/H965A), IB6 (E264A/H269A; D616A/H621A), IB4

(E264A/H269A; E960A/H965A) and IB2 (D616A/H621A; E960A/H965A). All three iron- binding sites are affected in IB variant, while variants IB6, IB4 and IB2 have a single unaffected binding site in domains 6, 4 and 2 respectively. By mutating residues in two iron-binding sites at once (variants IB6, IB4 and IB2), the impact of the unaffected site on the ferroxidase activity of

CpHp can be studied.

A commonly utilized strategy of site-directed mutagenesis involves mutating the residue of interest with further assessment of the critical parameters of the variant and the wild-type proteins. This strategy does not take into consideration potential compensatory mechanisms – for example, in a variant, the role of a mutated amino acid residue may be fulfilled by a different residue [217]. In this case, the functional role of the residue of interest may remain obscured. In contrast, the mutagenesis strategy used in this study is twofold, and conclusions can be drawn from two sets of observations: (1) monitoring the impairment of ferroxidase activity while comparing IB2, IB4, IB6 and IB variants to wild-type CpHp; (2) monitoring the gain of ferroxidase function while comparing IB2, IB4 and IB6 variants to IB variant.

5.2.3 Analysis of enzymatic activity of wild-type CpHp and variants

Similar to CpHp, all variants were expressed in BHK cells and purified by immunoaffinity chromatography. Taking into account the variability in specific activity of different samples of CpHp, four independent samples of wild-type CpHp and each of the variants were analyzed. Similar to CpHp, kcat values in different samples of the same hephaestin variant were variable, while the shapes of kinetic curves (represented by Km) were consistent for each of the variants. Similar to the observed correlation of oxidation rates of pPD and Fe2+ by different

82 samples of CpHp, wild-type hephaestin and each of the variants exhibited constant ratios between pPD- and Fe2+-oxidation rates (Table 5.3, Appendix C). Thus, two unchanging

2+ parameters (Km and the ratio between pPD- and Fe -oxidation rates) were closely monitored while analyzing the enzymatic activity of CpHp and its variants of the putative iron ligands.

To ensure that the effect of mutations on the ferroxidase activity of hephaestin was not the result of protein misfolding, I performed kinetic analysis of pPD-oxidation rates for wild- type and variant CpHp. All variants retained pPD-oxidase activity and showed lower the Km values compared to Km for pPD of wild-type CpHp (Table 5.3). While wild-type hephaestin and the IB6 variant exhibited essentially identical ratios between pPD- and Fe2+-oxidation rates, this value was significantly increased in the IB, IB2 and IB4 variants (Table 5.3 column 6), suggesting impairment of ferroxidase activity in these variants.

Table 5.3. Kinetic parameters for CpHp and variants.1 1. CpHp 2. Km for 3. Range of 4. Km for high 5. Range of 6.kcat for 2+ and pPD (mM) kcat for pPD affinity kcat` for Fe pPD to -1 -1 3 2+ variants (min ) oxidation of (min ) kcat` for Fe Fe2+ (µM)2 ratio CpHp 1.0 ± 0.1 2.3-8.7 4.3 ± 0.7 4.5-8.5 0.75 ± 0.11 IB6 0.41 ± 0.05 0.48-0.50 2.2 ± 0.1 0.47-0.67 0.84 ± 0.06 IB4 0.37 ± 0.01 4.2-5.9 n/a 0.39-0.66 9.5 ± 1.0 IB2 0.35 ± 0.01 1.3-3.0 n/a 0.11-0.28 10.7 ± 1.0 IB 0.22 ± 0.01 0.33-0.46 n/a 0.06-0.08 5.5 ± 0.1 1 – Data were obtained after analysis of multiple protein samples (4 different samples for CpHp and each of the variants). Columns 2, 4 and 6 show the average values with error representing SEM. Because variability of kcat can be caused not only by method error but also by significant differences in the internal properties of the samples, catalytic rates of different samples were not averaged but are represented as a range of values (columns 3 and 5). For kinetic data of individual samples see Appendix C. 2+ 2 - Km for Fe was calculated using an Eadie-Hofstee plot; the initial ferroxidation rates were measured at six substrate concentrations ranging from 2 to 10 µM of iron. 2+ 3 - kcat` for Fe is the number of ferrous ions oxidized per minute by one molecule of enzyme at the iron concentration of 87.5 µM.

Ferroxidation rates of wild-type CpHp and the variants were studied in the range of 2-100

µM of iron. While the kinetic curves of IB, IB4 and IB2 were linear in this range of iron concentrations, initial velocity versus substrate plots for wild-type hephaestin and IB6 variant showed a distinguishable hyperbola at low iron concentrations (Fig. 5.4 inserts). This

Figure 5.4. Ferroxidase activity of wild-type and variant rhHp. Representative Eadie- Hofstee plots are shown with the corresponding kinetic curves on the insert. (A) rhHp; (B) IB6; (C) IB4; (D) IB2; (E) IB. Fitted lines/curves are not shown for the low affinity component on A and B because of the lack of kinetic data at higher substrate concentrations.

84 observation is also supported by the Eadie-Hofstee plot which resembles a characteristic biphasic curve only for CpHp and IB6; the remaining variants produce linear Eadie-Hofstee plots with only the low-affinity component present (Fig. 5.4). Thus, the only variant which retained high- affinity iron oxidation was IB6 - the variant with unaffected ferrous binding site in domain 6.

5.3 Discussion

Kinetic analysis of CpHp ferroxidation rates revealed non-Michaelis-Menten kinetics.

When the initial kinetic data were plotted in Eadie-Hofstee coordinates, the plot displayed a biphasic profile, suggesting the presence of both low- and high-affinity iron-binding sites (Fig.

5.2 and 5.3). Similar biphasic Eadie-Hofstee plots were generated while analyzing ferroxidase activity of ceruloplasmin [213; 214; 215]. In ceruloplasmin, the predicted high-affinity ferrous binding sites are located underneath large protein loops [120; 137; 139](Fig. 1.4 B). Due to the abundance of acidic amino acid residues, the high-affinity iron-binding sites and the surrounding protein surface harbor significant negative charges [137] (Fig. 1.5 C). The overall structure of the molecule as well as the charge distribution in ceruloplasmin is predicted to be conserved in hephaestin [113]. Together with the results of kinetic analysis, these data suggest that hephaestin and ceruloplasmin utilize similar mechanisms for iron oxidation.

To study the role of the predicted iron ligands in the ferroxidase activity of CpHp, I produced a variant with all three putative iron-binding sites affected (IB) and variants that retained a single predicted site for ferrous binding (IB2, IB4, IB6). Mutating the amino acid residues in all three iron-binding sites resulted in a loss of high-affinity ferrous oxidation along with a 7-fold increase in the ratio between pPD- and Fe2+-oxidation rates (Table 5.3, column 6).

Similar results were obtained for the IB2 and IB4 variants. In contrast, the IB6 variant showed high-affinity iron oxidation with a Km comparable to the one of wild-type CpHp (Table 5.3, column 4); the ratios between pPD-and Fe2+-oxidation rates for IB6 and CpHp were also equivalent (Table 5.3, column 6). These results confirmed that residues E960/H965 serve as iron

85 ligands of the high-affinity binding site in the domain 6 of CpHp. On the contrary, residues

E264/H269 in domain 2 and D616/H621 in domain 4 appear to be dispensable for hephaestin ferroxidase activity; supporting this observation, mutations of these residues in IB6 variant did not result in any detectable impairment of ferroxidase function. There is a possibility that the role of residues E264/H269 and D616/H621 in iron oxidation was not revealed due to the absence of copper at type 1 sites in domains 2 and 4 of CpHp. On the other hand, the iron-binding sites in domains 2 and 4 of hephaestin are not comprised of a canonical set of ligands. Conserved ferrous binding sites in domain 6 of hephaestin and in domain 4 and 6 of ceruloplasmin are formed by three acidic residues and a histidine residue (Table 5.1). In contrast, the putative sites for iron binding in domains 2 and 4 of hephaestin have one of the acidic residues substituted with serinyl.

The unusual arrangement of ligands may have a detrimental effect on iron oxidation at these sites.

5.4 Conclusions

Detailed kinetic analysis of CpHp ferroxidation rates revealed the biphasic shape of an

Eadie-Hofstee plot that indicates the presence of two types of iron-binding sites. The Michaelis constants for high- and low-affinity ferrous binding sites in CpHp were comparable to the corresponding Km values in ceruloplasmin, suggesting that both paralogs utilize similar amino acid residues for iron binding. To investigate the role of particular residues in the iron specificity of hephaestin, mutations of putative iron ligands were introduced into CpHp using site-directed mutagenesis. All variants retained pPD-oxidase activity suggesting that the overall protein structures were preserved and the resulting differences in ferroxidase activity were reflective of the mutations that had been introduced into the predicted iron-binding sites. Kinetic analysis of ferroxidation rates of wild-type CpHp and variants confirmed the important roles of residues

E960 and H965 in hephaestin ferroxidase activity.

6. GENERAL DISCUSSION AND FUTURE DIRECTIONS

6.1 General discussion

6.1.1 Recombinant hephaestins

Even in the small intestine, the major site of hephaestin expression, the expression level of this protein is too low to isolate the amounts of hephaestin sufficient for biochemical studies.

Thus, I used recombinant hephaestin to study the enzymatic and biophysical properties of human hephaestin. Previously, the hephaestin ectodomain with a transferrin signal peptide (TfHp) was expressed in BHK cells [155]. In this study I expressed a soluble form of hephaestin with a ceruloplasmin signal peptide (CpHp) using BHK cells as an expression system; after the initial immunoaffinity purification, CpHp was further fractionated using ion exchange chromatography.

The last step resulted in isolation of CpHp enriched in enzymatically active protein. As an alternative to expression in mammalian cells, I also produced recombinant hephaestin in the yeast P. pastoris; as this yeast hephaestin contained the Fet3p signal peptide, it was designated as

Fet3Hp. Here I would like to summarize the similarities and differences in biochemical properties of these three recombinant forms of hephaestin.

All three recombinant forms of hephaestin had a similar N-terminal sequence (except for a few extra amino acids from the transferrin signal peptide at the N-terminus of TfHp [155]). As shown by enzymatic deglycosylation, TfHp [155] and CpHp contained highly processed (hybrid or complex) N-linked oligosaccharides, while N-glycosyl residues in yeast-derived hephaestin were represented by mannose-enriched structures. Although Fet3Hp was probably more heavily glycosylated than rhHp expressed in BHK cells, this glycosylation did not interfere with the ferroxidase activity of yeast hephaestin. This was supported by the observation that hyperglycosylated ceruloplasmin expressed in P. pastoris showed the same enzymatic activity as native (blood plasma derived) ceruloplasmin [218].

TfHp, CpHp and Fet3Hp exhibited similar Km values towards ferrous iron, while the kcat values differed significantly (Table 6.1). After further chromatography on DEAE-Sepharose,

CpHp showed the highest kcat for iron oxidation compared to other recombinant hephaestin; this value was also comparable to ferroxidation rates of Fet3p and ceruloplasmin. The low ferroxidation rate of Fet3Hp may have been caused by the use of a heterogeneous expression system.

Table 6.1. Kinetic and spectroscopic properties of different ferroxidases. Ferroxidase Absorption Km towards kcat towards coefficient at Fe2+, µM Fe2+, min-1 600nm Fet3Hp ND 3.0 ± 0.6 0.73 ± 0.06 TfHp 2010 [155] 2.1 [155] 2.5 [155] CpHp1 5000 3.8 ± 1 21 ± 2 Fet3p 5500 [122] 2 [121] 9 [121] Ceruloplasmin 10,000 [135] 4.7 ± 0.6 34 ± 2 1 - parameters shown in this table correspond to the CpHp after fractionation with IEC (fraction eluted with 150 mM NaCl).

Despite the significant differences in ferroxidation rates, the average copper content of recombinant hephaestins ranged from 3.13 in TfHp [155] to 4.2 for Fet3Hp, while the copper content of CpHp was determined as 3.5 copper atoms per molecule. This copper content was lower than the 6 copper atoms per molecule of hephaestin as predicted from amino acid sequence homology with ceruloplasmin. While a copper content of 3 copper atoms per molecule is not compatible with ferroxidase activity, 4 copper atoms are sufficient for oxidation by a MCO [218;

219]. This minimal requirement was essentially met in Fet3Hp and CpHp, suggesting that these proteins can function with only 4 copper atoms arranged as a single type 1 copper and a trinuclear cluster. In this case, hephaestin was most likely to retain the type 1 copper atom in domain 6 that is adjacent to the trinuclear cluster. This suggestion was supported by the presence of a functional iron-binding site in domain 6 of human hephaestin (see Section 6.1.3). Domains 2 and 4 of hephaestin, which also contain predicted type 1 copper ligands, may still play important

88 roles in maintaining the overall structure of the molecule or may participate in the interactions of hephaestin with other proteins.

Importantly, total copper analysis does not provide information on the nature of the detected copper atoms, so the stoichiometry between type 1, type 2 and type 3 copper atoms could not be determined. It is possible that complete copper-loading of rhHp was not achieved even after additional refinement on DEAE-Sepharose. In that case CpHp may have represented a heterogeneous mixture of molecules comprised of six copper-loaded and incompletely copper- loaded forms. Interestingly, while Fet3Hp and CpHp had very close copper contents, their ferroxidation kcat differed more than ten-fold. This implied that some of the detected copper atoms in Fet3Hp were not properly bound at prosthetic sites, which rendered them non- functional.

6.1.2 Oxidation of organic substrates

Using rhHp expressed in yeast, I investigated hephaestin specificity towards different organic substrates. Similar to ceruloplasmin, Fet3Hp retained activity towards organic amines such as pPD and o-dianisidine. EDTA did not abolish the oxidation of organic amines by Fet3Hp suggesting that hephaestin contains the binding site for this group of substrates. Km values towards pPD and o-dianisidine in Fet3Hp were consistent with corresponding values in ceruloplasmin, implying a similar structure of the organic amine binding site. Due to the low resolution of protein crystals, crystal soaking experiments did not reveal the amino acid residues participating in the binding of organic amines by ceruloplasmin. This makes difficult to predict amino acids comprising the binding site for organic amines in hephaestin. In laccase from

Trametes versicolor, the binding pocket for organic amines is comprised of multiple phenylalanine residues that interact with the aromatic ring of the substrate, while an acidic residue at the bottom of this pocket interacts with the positively charged amino group [118]. A

89 similar arrangement of amino acid residues may be utilized for binding of organic amines by hephaestin and ceruloplasmin.

In contrast to ceruloplasmin, Fet3Hp did not exhibit direct enzymatic activity against biogenic amines. These results can be related to the significant structural differences between hephaestin and ceruloplasmin in the structure of the binding sites for biogenic amines. On the other hand, the absence of activity towards biogenic amines in Fet3Hp may have resulted from incomplete copper loading of type 1 copper adjacent to the binding site for these substrates.

Experiments with hephaestin containing a full set of copper atoms can clarify this ambiguity.

6.1.3 Oxidation of ferrous iron

Kinetic studies revealed that, similar to ceruloplasmin, hephaestin has two types of iron- binding sites with different affinities towards ferrous iron. Studies involving site-directed mutagenesis confirmed that residues E960/H965 serve as iron ligands of a high-affinity binding site located in domain 6 of hephaestin. Based on homology with ceruloplasmin, the remaining ligands of this high-affinity iron-binding site are residues E300 and D996. Thus, the high-affinity iron-binding site in domain 6 of hephaestin is likely to be comprised of a canonical set of ligands

– three acidic residues and one histidine residue. The same arrangement of ligands is found in the iron-binding sites located in domains 4 and 6 of ceruloplasmin [137].

The nature of the low-affinity iron-binding site in both hephaestin and ceruloplasmin is less clear. At the top of the molecule, ceruloplasmin has a negatively charged patch that hosts two high-affinity binding sites [137] and may also accommodate the low-affinity binding site(s)

(Fig. 1.5 C). Acidic residues of this negatively charged area are contributed by all six domains of ceruloplasmin. The high structural homology with ceruloplasmin along with similar kinetic behavior predicts similar structure of the low-affinity binding site in hephaestin.

As observed in both ceruloplasmin and hephaestin, binding of iron at the low-affinity binding site results in a four-fold increase in Vmax of the ferroxidation reaction, suggesting an

allosteric effect. Interestingly, the Km for iron in the low-affinity site is approximately two orders higher than the physiological concentration of Fe2+ in blood (less than 1 µM) [220]. In thalassemia patients, the concentration of free ferrous iron in blood may be as high as 16 µM

[220]. Therefore, it is unlikely that the low-affinity binding site is involved in iron oxidation under non-pathological conditions. On the other hand, activation of this ferroxidase function may be important while dealing with excessive iron loading.

6.2 Significance

To date, at least three multicopper ferroxidases have been detected in humans. Close amino acid sequence homology between these proteins predicts similar biochemical properties; on the other hand, specific functions for each of the paralogs may be determined by their structural differences. Hephaestin, a multicopper ferroxidase expressed predominantly in the small intestine, represents the object of the current study. This work significantly advanced our current knowledge of the substrate specificity, copper loading and ferroxidation mechanism in human hephaestin. Furthermore, the knowledge acquired from this study will facilitate research focused on the structure and catalytic mechanism of hephaestin paralogs.

6.3 Future directions

6.3.1 Size exclusion chromatography of rhHp

Despite the fact that IEC allowed isolation of rhHp with improved catalytic and spectroscopic properties, the copper content of this hephaestin fraction was 3.5 atoms per molecule instead of the predicted 6. Due to the substantial differences in the biophysical properties of apo- and holo-ceruloplasmin, these two ceruloplasmin forms can be separated using size exclusion chromatography [136]. This method may be useful for obtaining fully copper-

91 loaded hephaestin. Using this fully-copper loaded rhHp, kinetic and spectroscopic properties of human hephaestin could be studied.

6.3.2 Ferroxidation mechanism of hephaestin

Experiments described in Chapter 5 suggest the important role of residues E960/H965 in high-affinity iron binding. In addition to the above-mentioned residues, the predicted iron- binding site in domain 6 of hephaestin contains residues E300 and D1050 [113; 137]. It would be interesting to investigate the role of each of these residues in the ferroxidation mechanism of hephaestin. This could be accomplished by studying the ferroxidase activity of E960A, H965A,

E300A and D1050A variants of CpHp. The pPD-oxidase activity of the variants can be monitored to control for protein misfolding. Mutation of E300A is expected to result in the most prominent effect on hephaestin ferroxidation rates as the carboxylate group of E300 is predicted to provide an electron transfer path from iron to the type 1 copper in domain 6. Experiments with an E300A hephaestin variant are currently in progress.

In addition to the ferrous binding site, hephaestin also contains a predicted ferric binding site (known as “holding” site) that is responsible for the release of the oxidized iron [113; 137].

Located in the vicinity of high-affinity ferrous binding site in domain 6, this holding site is predicted to be comprised of E258, E956 and E960 residues [113]. The last residue also functions as a ligand of the ferrous binding site and may be involved in transfer of oxidized iron to the holding site. By studying the enzymatic activity of variants E258A and E956A, the role of these residues in the ferroxidation mechanism of human hephaestin could be revealed.

6.3.3 Alternative substrates of hephaestin

Ceruloplasmin is well-known for its wide substrate specificity. As a close ceruloplasmin homolog, hephaestin may also possess activity towards substrates other than iron and organic

92 amines. Homology with ceruloplasmin predicts a list of potential substrates for hephaestin including Cu(I), LDL, glutathione and NO.

Xu et al. [150] suggested that cuprous oxidase activity of ceruloplasmin plays an important role in protecting the brain against toxicity of Cu(I). As hephaestin is expressed in brain, its putative cuprous oxidase activity may also have physiological significance.

Another candidate substrate for hephaestin is LDL (low-density lipoprotein). In the shallow cavity in the domain 2, ceruloplasmin harbors a binding site for a labile copper atom that mediates LDL oxidation [142]. As the oxidation of LDL plays an important role in the formation of atherosclerotic lesions [221], this enzymatic activity may be involved in the development of atherosclerosis. As shown by site-directed-mutagenesis, the H426 residue of ceruloplasmin serves as a ligand for the labile copper [142]. The corresponding residue in hephaestin is H452

[113]. If the LDL-oxidase function is preserved in hephaestin, mutagenesis of this residue can confirm its role in the oxidation of LDL.

The presence of glutathione-peroxidase activity in ceruloplasmin suggests that this MCO can function as a GSH-linked antioxidant enzyme [141]. This antioxidant activity has been associated with the redox-active C699 residue [141]. Conservation of this residue between hephaestin and ceruloplasmin implies conservation of glutathione-peroxidase activity between these homologous enzymes.

Administration of nitric oxide after ischemia-reperfusion has been shown to be cardioprotective [222]. This effect was attributed to S-nitrosation – reversible modification that protects protein cysteinyl thiols against unwanted and possibly irreversible oxidation into intra/intermolecular disulfide bonds [222]. While nitric oxide is a poor S-nitrosating agent, its conversion into the NO+ form results in significantly enhanced rates of S-nitrosation. With the expression detected in the heart [156], hephaestin may be responsible for cardioprotection through its predicted NO-oxidase activity.

The activity of hephaestin towards the putative substrates can be tested using previously described activity assays [125; 141; 142; 223].

BIBLIOGRAPHY

1. Crichton, R. R. & Pierre, J. L. (2001). Old iron, young copper: from Mars to Venus.

Biometals 14, 99-112.

2. Ponka, P. (1999). Cell biology of heme. Am. J. Med. Sci. 318, 241-56.

3. Danielson, P. B. (2002). The cytochrome P450 superfamily: biochemistry, evolution and

drug metabolism in humans. Curr. Drug Metab. 3, 561-97.

4. Lill, R. & Muhlenhoff, U. (2006). Iron-sulfur protein biogenesis in eukaryotes:

components and mechanisms. Annu. Rev. Cell Dev. Biol. 22, 457-86.

5. Eklund, H., Uhlin, U., Farnegardh, M., Logan, D. T. & Nordlund, P. (2001). Structure

and function of the radical enzyme ribonucleotide reductase. Prog. Biophys. Mol. Biol.

77, 177-268.

6. Wink, D. A., Hines, H. B., Cheng, R. Y., Switzer, C. H., Flores-Santana, W., Vitek, M.

P., Ridnour, L. A. & Colton, C. A. (2011). Nitric oxide and redox mechanisms in the

immune response. J. Leukoc. Biol. 89, 873-91.

7. Pierre, J. L. & Fontecave, M. (1999). Iron and activated oxygen species in biology: the

basic chemistry. Biometals 12, 195-9.

8. Taketani, S. (2005). Aquisition, mobilization and utilization of cellular iron and heme:

endless findings and growing evidence of tight regulation. Tohoku J. Exp. Med. 205, 297-

318.

9. Andrews, S. C., Robinson, A. K. & Rodriguez-Quinones, F. (2003). Bacterial iron

homeostasis. FEMS Microbiol. Rev. 27, 215-37.

10. Kammler, M., Schon, C. & Hantke, K. (1993). Characterization of the ferrous iron uptake

system of Escherichia coli. J. Bacteriol. 175, 6212-9.

11. Velayudhan, J., Hughes, N. J., McColm, A. A., Bagshaw, J., Clayton, C. L., Andrews, S.

C. & Kelly, D. J. (2000). Iron acquisition and virulence in Helicobacter pylori: a major

role for FeoB, a high-affinity ferrous iron transporter. Mol. Microbiol. 37, 274-86. 95

12. Worst, D. J., Gerrits, M. M., Vandenbroucke-Grauls, C. M. & Kusters, J. G. (1998).

Helicobacter pylori ribBA-mediated riboflavin production is involved in iron acquisition.

J. Bacteriol. 180, 1473-9.

13. Cowart, R. E. (2002). Reduction of iron by extracellular iron reductases: implications for

microbial iron acquisition. Arch. Biochem. Biophys. 400, 273-81.

14. Braun, V. (2001). Iron uptake mechanisms and their regulation in pathogenic bacteria.

Int. J. Med. Microbiol. 291, 67-79.

15. Andrews, S. C., Le Brun, N. E., Barynin, V., Thomson, A. J., Moore, G. R., Guest, J. R.

& Harrison, P. M. (1995). Site-directed replacement of the coaxial heme ligands of

bacterioferritin generates heme-free variants. J. Biol. Chem. 270, 23268-74.

16. Zhao, G., Ceci, P., Ilari, A., Giangiacomo, L., Laue, T. M., Chiancone, E. & Chasteen, N.

D. (2002). Iron and hydrogen peroxide detoxification properties of DNA-binding protein

from starved cells. A ferritin-like DNA-binding protein of Escherichia coli. J. Biol.

Chem. 277, 27689-96.

17. Escolar, L., Perez-Martin, J. & de Lorenzo, V. (1999). Opening the iron box:

transcriptional metalloregulation by the Fur protein. J. Bacteriol. 181, 6223-9.

18. Kosman, D. J. (2003). Molecular mechanisms of iron uptake in fungi. Mol. Microbiol.

47, 1185-97.

19. Georgatsou, E. & Alexandraki, D. (1994). Two distinctly regulated genes are required for

ferric reduction, the first step of iron uptake in Saccharomyces cerevisiae. Mol. Cell.

Biol. 14, 3065-73.

20. Dix, D. R., Bridgham, J. T., Broderius, M. A., Byersdorfer, C. A. & Eide, D. J. (1994).

The FET4 gene encodes the low affinity Fe(II) transport protein of Saccharomyces

cerevisiae. J. Biol. Chem. 269, 26092-9.

21. Chen, X. Z., Peng, J. B., Cohen, A., Nelson, H., Nelson, N. & Hediger, M. A. (1999).

Yeast SMF1 mediates H(+)-coupled iron uptake with concomitant uncoupled cation

currents. J. Biol. Chem. 274, 35089-94.

22. Askwith, C., Eide, D., Van Ho, A., Bernard, P. S., Li, L., Davis-Kaplan, S., Sipe, D. M.

& Kaplan, J. (1994). The FET3 gene of S. cerevisiae encodes a multicopper oxidase

required for ferrous iron uptake. Cell 76, 403-10.

23. Stearman, R., Yuan, D. S., Yamaguchi-Iwai, Y., Klausner, R. D. & Dancis, A. (1996). A

permease-oxidase complex involved in high-affinity iron uptake in yeast. Science 271,

1552-7.

24. Kwok, E. Y., Severance, S. & Kosman, D. J. (2006). Evidence for iron channeling in the

Fet3p-Ftr1p high-affinity iron uptake complex in the yeast plasma membrane.

Biochemistry 45, 6317-27.

25. Matile, P. & Wiemken, A. (1967). The vacuole as the lysosome of the yeast cell. Arch.

Mikrobiol. 56, 148-55.

26. Li, L., Chen, O. S., McVey Ward, D. & Kaplan, J. (2001). CCC1 is a transporter that

mediates vacuolar iron storage in yeast. J. Biol. Chem. 276, 29515-9.

27. Singh, A., Kaur, N. & Kosman, D. J. (2007). The metalloreductase Fre6p in Fe-efflux

from the yeast vacuole. J. Biol. Chem. 282, 28619-26.

28. Portnoy, M. E., Jensen, L. T. & Culotta, V. C. (2002). The distinct methods by which

manganese and iron regulate the Nramp transporters in yeast. Biochem. J. 362, 119-24.

29. Urbanowski, J. L. & Piper, R. C. (1999). The iron transporter Fth1p forms a complex

with the Fet5 iron oxidase and resides on the vacuolar membrane. J. Biol. Chem. 274,

38061-70.

30. Shakoury-Elizeh, M., Tiedeman, J., Rashford, J., Ferea, T., Demeter, J., Garcia, E.,

Rolfes, R., Brown, P. O., Botstein, D. & Philpott, C. C. (2004). Transcriptional

remodeling in response to iron deprivation in Saccharomyces cerevisiae. Mol. Biol. Cell

15, 1233-43.

31. Rutherford, J. C., Jaron, S., Ray, E., Brown, P. O. & Winge, D. R. (2001). A second iron-

regulatory system in yeast independent of Aft1p. Proc. Natl. Acad. Sci. U. S. A. 98,

14322-7.

32. Li, L., Bagley, D., Ward, D. M. & Kaplan, J. (2008). Yap5 is an iron-responsive

transcriptional activator that regulates vacuolar iron storage in yeast. Mol. Cell. Biol. 28,

1326-37.

33. Puig, S., Vergara, S. V. & Thiele, D. J. (2008). Cooperation of two mRNA-binding

proteins drives metabolic adaptation to iron deficiency. Cell Metab. 7, 555-64.

34. Theil, E. C. (2011). Iron homeostasis and nutritional iron deficiency. J. Nutr. 141, 724S-

728S.

35. Shayeghi, M., Latunde-Dada, G. O., Oakhill, J. S., Laftah, A. H., Takeuchi, K., Halliday,

N., Khan, Y., Warley, A., McCann, F. E., Hider, R. C., Frazer, D. M., Anderson, G. J.,

Vulpe, C. D., Simpson, R. J. & McKie, A. T. (2005). Identification of an intestinal heme

transporter. Cell 122, 789-801.

36. Rajagopal, A., Rao, A. U., Amigo, J., Tian, M., Upadhyay, S. K., Hall, C., Uhm, S.,

Mathew, M. K., Fleming, M. D., Paw, B. H., Krause, M. & Hamza, I. (2008). Haem

homeostasis is regulated by the conserved and concerted functions of HRG-1 proteins.

Nature 453, 1127-31.

37. Latunde-Dada, G. O., Takeuchi, K., Simpson, R. J. & McKie, A. T. (2006). Haem carrier

protein 1 (HCP1): Expression and functional studies in cultured cells. FEBS Lett. 580,

6865-70.

38. Gunshin, H., Mackenzie, B., Berger, U. V., Gunshin, Y., Romero, M. F., Boron, W. F.,

Nussberger, S., Gollan, J. L. & Hediger, M. A. (1997). Cloning and characterization of a

mammalian proton-coupled metal-ion transporter. Nature 388, 482-8.

39. Gunshin, H., Fujiwara, Y., Custodio, A. O., Direnzo, C., Robine, S. & Andrews, N. C.

(2005). Slc11a2 is required for intestinal iron absorption and erythropoiesis but

dispensable in placenta and liver. J. Clin. Invest. 115, 1258-66.

40. McKie, A. T., Barrow, D., Latunde-Dada, G. O., Rolfs, A., Sager, G., Mudaly, E.,

Mudaly, M., Richardson, C., Barlow, D., Bomford, A., Peters, T. J., Raja, K. B., Shirali,

S., Hediger, M. A., Farzaneh, F. & Simpson, R. J. (2001). An iron-regulated ferric

reductase associated with the absorption of dietary iron. Science 291, 1755-9.

41. McKie, A. T. (2008). The role of Dcytb in iron metabolism: an update. Biochem. Soc.

Trans. 36, 1239-41.

42. Wyman, S., Simpson, R. J., McKie, A. T. & Sharp, P. A. (2008). Dcytb (Cybrd1)

functions as both a ferric and a cupric reductase in vitro. FEBS Lett. 582, 1901-6.

43. Gunshin, H., Starr, C. N., Direnzo, C., Fleming, M. D., Jin, J., Greer, E. L., Sellers, V.

M., Galica, S. M. & Andrews, N. C. (2005). Cybrd1 (duodenal cytochrome b) is not

necessary for dietary iron absorption in mice. Blood 106, 2879-83.

44. Chasteen, N. D. & Harrison, P. M. (1999). Mineralization in ferritin: an efficient means

of iron storage. J. Struct. Biol. 126, 182-94.

45. Miller, L. L., Miller, S. C., Torti, S. V., Tsuji, Y. & Torti, F. M. (1991). Iron-independent

induction of ferritin H chain by tumor necrosis factor. Proc. Natl. Acad. Sci. U. S. A. 88,

4946-50.

46. Leggett, B. A., Fletcher, L. M., Ramm, G. A., Powell, L. W. & Halliday, J. W. (1993).

Differential regulation of ferritin H and L subunit mRNA during inflammation and long-

term iron overload. J. Gastroenterol. Hepatol. 8, 21-7.

47. Shaw, G. C., Cope, J. J., Li, L., Corson, K., Hersey, C., Ackermann, G. E., Gwynn, B.,

Lambert, A. J., Wingert, R. A., Traver, D., Trede, N. S., Barut, B. A., Zhou, Y., Minet,

E., Donovan, A., Brownlie, A., Balzan, R., Weiss, M. J., Peters, L. L., Kaplan, J., Zon, L.

I. & Paw, B. H. (2006). Mitoferrin is essential for erythroid iron assimilation. Nature 440,

96-100.

48. Li, F. Y., Nikali, K., Gregan, J., Leibiger, I., Leibiger, B., Schweyen, R., Larsson, C. &

Suomalainen, A. (2001). Characterization of a novel human putative mitochondrial

transporter homologous to the yeast mitochondrial RNA splicing proteins 3 and 4. FEBS

Lett. 494, 79-84.

49. Bencze, K. Z., Kondapalli, K. C., Cook, J. D., McMahon, S., Millan-Pacheco, C., Pastor,

N. & Stemmler, T. L. (2006). The structure and function of frataxin. Crit. Rev. Biochem.

Mol. Biol. 41, 269-91.

50. Levi, S. & Arosio, P. (2004). Mitochondrial ferritin. Int. J. Biochem. Cell Biol. 36, 1887-

51. Donovan, A., Lima, C. A., Pinkus, J. L., Pinkus, G. S., Zon, L. I., Robine, S. & Andrews,

N. C. (2005). The iron exporter ferroportin/Slc40a1 is essential for iron homeostasis. Cell

Metab. 1, 191-200.

52. Jeong, S. Y. & David, S. (2003). Glycosylphosphatidylinositol-anchored ceruloplasmin is

required for iron efflux from cells in the central nervous system. J. Biol. Chem. 278,

27144-8.

53. De Domenico, I., Ward, D. M., di Patti, M. C., Jeong, S. Y., David, S., Musci, G. &

Kaplan, J. (2007). Ferroxidase activity is required for the stability of cell surface

ferroportin in cells expressing GPI-ceruloplasmin. EMBO J. 26, 2823-31.

54. Kono, S., Yoshida, K., Tomosugi, N., Terada, T., Hamaya, Y., Kanaoka, S. & Miyajima,

H. (2010). Biological effects of mutant ceruloplasmin on hepcidin-mediated

internalization of ferroportin. Biochim. Biophys. Acta 1802, 968-75.

55. Han, O. & Kim, E. Y. (2007). Colocalization of ferroportin-1 with hephaestin on the

basolateral membrane of human intestinal absorptive cells. J. Cell. Biochem. 101, 1000-

10.

100

56. Yeh, K. Y., Yeh, M., Mims, L. & Glass, J. (2009). Iron feeding induces ferroportin 1 and

hephaestin migration and interaction in rat duodenal epithelium. Am. J. Physiol.

Gastrointest. Liver Physiol. 296, G55-65.

57. Aisen, P., Leibman, A. & Zweier, J. (1978). Stoichiometric and site characteristics of the

binding of iron to human transferrin. J. Biol. Chem. 253, 1930-7.

58. Dautry-Varsat, A., Ciechanover, A. & Lodish, H. F. (1983). pH and the recycling of

transferrin during receptor-mediated endocytosis. Proc. Natl. Acad. Sci. U. S. A. 80,

2258-62.

59. Kawabata, H., Germain, R. S., Vuong, P. T., Nakamaki, T., Said, J. W. & Koeffler, H. P.

(2000). Transferrin receptor 2-alpha supports cell growth both in iron-chelated cultured

cells and in vivo. J. Biol. Chem. 275, 16618-25.

60. Trinder, D. & Baker, E. (2003). Transferrin receptor 2: a new molecule in iron

metabolism. Int. J. Biochem. Cell Biol. 35, 292-6.

61. Ohgami, R. S., Campagna, D. R., Greer, E. L., Antiochos, B., McDonald, A., Chen, J.,

Sharp, J. J., Fujiwara, Y., Barker, J. E. & Fleming, M. D. (2005). Identification of a

ferrireductase required for efficient transferrin-dependent iron uptake in erythroid cells.

Nat. Genet. 37, 1264-9.

62. Liuzzi, J. P., Aydemir, F., Nam, H., Knutson, M. D. & Cousins, R. J. (2006). Zip14

(Slc39a14) mediates non-transferrin-bound iron uptake into cells. Proc. Natl. Acad. Sci.

U. S. A. 103, 13612-7.

63. Christensen, E. I. & Birn, H. (2002). Megalin and cubilin: multifunctional endocytic

receptors. Nat. Rev. Mol. Cell. Biol. 3, 256-66.

64. Hvidberg, V., Jacobsen, C., Strong, R. K., Cowland, J. B., Moestrup, S. K. & Borregaard,

N. (2005). The endocytic receptor megalin binds the iron transporting neutrophil-

gelatinase-associated lipocalin with high affinity and mediates its cellular uptake. FEBS

Lett. 579, 773-7.

101

65. Smith, K. D. (2007). Iron metabolism at the host pathogen interface: lipocalin 2 and the

pathogen-associated iroA gene cluster. Int. J. Biochem. Cell Biol. 39, 1776-80.

66. Devireddy, L. R., Gazin, C., Zhu, X. & Green, M. R. (2005). A cell-surface receptor for

lipocalin 24p3 selectively mediates apoptosis and iron uptake. Cell 123, 1293-305.

67. Nielsen, M. J., Moller, H. J. & Moestrup, S. K. (2010). Hemoglobin and heme scavenger

receptors. Antioxid. Redox. Signal. 12, 261-73.

68. Troadec, M. B., Ward, D. M. & Kaplan, J. (2009). A Tf-independent iron transport

system required for organogenesis. Dev. Cell 16, 3-4.

69. Chen, T. T., Li, L., Chung, D. H., Allen, C. D., Torti, S. V., Torti, F. M., Cyster, J. G.,

Chen, C. Y., Brodsky, F. M., Niemi, E. C., Nakamura, M. C., Seaman, W. E. & Daws,

M. R. (2005). TIM-2 is expressed on B cells and in liver and kidney and is a receptor for

H-ferritin endocytosis. J. Exp. Med. 202, 955-65.

70. Li, L., Fang, C. J., Ryan, J. C., Niemi, E. C., Lebron, J. A., Bjorkman, P. J., Arase, H.,

Torti, F. M., Torti, S. V., Nakamura, M. C. & Seaman, W. E. (2010). Binding and uptake

of H-ferritin are mediated by human transferrin receptor-1. Proc. Natl. Acad. Sci. U. S. A.

107, 3505-10.

71. Mukhopadhyay, C. K., Attieh, Z. K. & Fox, P. L. (1998). Role of ceruloplasmin in

cellular iron uptake. Science 279, 714-7.

72. Hentze, M. W., Muckenthaler, M. U. & Andrews, N. C. (2004). Balancing acts:

molecular control of mammalian iron metabolism. Cell 117, 285-97.

73. Eisenstein, R. S. & Ross, K. L. (2003). Novel roles for iron regulatory proteins in the

adaptive response to iron deficiency. J. Nutr. 133, 1510S-6S.

74. McKie, A. T., Marciani, P., Rolfs, A., Brennan, K., Wehr, K., Barrow, D., Miret, S.,

Bomford, A., Peters, T. J., Farzaneh, F., Hediger, M. A., Hentze, M. W. & Simpson, R. J.

(2000). A novel duodenal iron-regulated transporter, IREG1, implicated in the basolateral

transfer of iron to the circulation. Mol. Cell 5, 299-309.

102

75. Zhang, D. L., Hughes, R. M., Ollivierre-Wilson, H., Ghosh, M. C. & Rouault, T. A.

(2009). A ferroportin transcript that lacks an iron-responsive element enables duodenal

and erythroid precursor cells to evade translational repression. Cell Metab. 9, 461-73.

76. Lee, P. L., Gelbart, T., West, C., Halloran, C. & Beutler, E. (1998). The human Nramp2

gene: characterization of the gene structure, alternative splicing, promoter region and

polymorphisms. Blood Cells Mol. Dis. 24, 199-215.

77. Meyron-Holtz, E. G., Ghosh, M. C., Iwai, K., LaVaute, T., Brazzolotto, X., Berger, U.

V., Land, W., Ollivierre-Wilson, H., Grinberg, A., Love, P. & Rouault, T. A. (2004).

Genetic ablations of iron regulatory proteins 1 and 2 reveal why iron regulatory protein 2

dominates iron homeostasis. EMBO J. 23, 386-95.

78. Smith, S. R., Ghosh, M. C., Ollivierre-Wilson, H., Hang Tong, W. & Rouault, T. A.

(2006). Complete loss of iron regulatory proteins 1 and 2 prevents viability of murine

zygotes beyond the blastocyst stage of embryonic development. Blood Cells Mol. Dis.

36, 283-7.

79. Fleming, R. E. & Sly, W. S. (2001). Hepcidin: a putative iron-regulatory hormone

relevant to hereditary hemochromatosis and the anemia of chronic disease. Proc. Natl.

Acad. Sci. U. S. A. 98, 8160-2.

80. Nemeth, E., Tuttle, M. S., Powelson, J., Vaughn, M. B., Donovan, A., Ward, D. M.,

Ganz, T. & Kaplan, J. (2004). Hepcidin regulates cellular iron efflux by binding to

ferroportin and inducing its internalization. Science 306, 2090-3.

81. Ganz, T. (2011). Hepcidin and iron regulation, 10 years later. Blood 117, 4425-33.

82. Nemeth, E., Rivera, S., Gabayan, V., Keller, C., Taudorf, S., Pedersen, B. K. & Ganz, T.

(2004). IL-6 mediates hypoferremia of inflammation by inducing the synthesis of the iron

regulatory hormone hepcidin. J. Clin. Invest. 113, 1271-6.

83. Vokurka, M., Krijt, J., Sulc, K. & Necas, E. (2006). Hepcidin mRNA levels in mouse

liver respond to inhibition of erythropoiesis. Physiol. Res. 55, 667-74.

103

84. Pak, M., Lopez, M. A., Gabayan, V., Ganz, T. & Rivera, S. (2006). Suppression of

hepcidin during anemia requires erythropoietic activity. Blood 108, 3730-5.

85. Peyssonnaux, C., Nizet, V. & Johnson, R. S. (2008). Role of the hypoxia inducible

factors HIF in iron metabolism. Cell Cycle 7, 28-32.

86. Sakurai, T. & Kataoka, K. (2007). Basic and applied features of multicopper oxidases,

CueO, bilirubin oxidase, and laccase. Chem. Rec. 7, 220-9.

87. Kosman, D. J. (2010). Multicopper oxidases: a workshop on copper coordination

chemistry, electron transfer, and metallophysiology. J. Biol. Inorg. Chem. 15, 15-28.

88. Nakamura, K., Kawabata, T., Yura, K. & Go, N. (2003). Novel types of two-domain

multi-copper oxidases: possible missing links in the evolution. FEBS Lett. 553, 239-44.

89. Skalova, T., Dohnalek, J., Ostergaard, L. H., Ostergaard, P. R., Kolenko, P., Duskova, J.,

Stepankova, A. & Hasek, J. (2009). The structure of the small laccase from Streptomyces

coelicolor reveals a link between laccases and nitrite reductases. J. Mol. Biol. 385, 1165-

78.

90. Nakamura, K. & Go, N. (2005). Function and molecular evolution of multicopper blue

proteins. Cell. Mol. Life Sci. 62, 2050-66.

91. Giardina, P., Faraco, V., Pezzella, C., Piscitelli, A., Vanhulle, S. & Sannia, G. (2009).

Laccases: a never-ending story. Cell. Mol. Life Sci. 67, 369-85.

92. Hirose, J., Sakurai, T., Imamura, K., Watanabe, H., Iwamoto, H., Hiromi, K., Itoh, H.,

Shin, T. & Murao, S. (1994). Characterization of ascorbate oxidase from Acremonium sp.

HI-25. J. Biochem. 115, 811-3.

93. Marcus, R. A. & Sutin, N. (1985). Electron transfers in chemistry and biology. Biochim.

Biophys. Acta 811, 265-322.

94. Gray, H. B. & Winkler, J. R. (1996). Electron transfer in proteins. Annu. Rev. Biochem.

65, 537-61.

104

95. Sakurai, T. & Kataoka, K. (2007). Structure and function of type I copper in multicopper

oxidases. Cell. Mol. Life Sci. 64, 2642-56.

96. Palmer, A. E., Szilagyi, R. K., Cherry, J. R., Jones, A., Xu, F. & Solomon, E. I. (2003).

Spectroscopic characterization of the Leu513His variant of fungal laccase: effect of

increased axial ligand interaction on the geometric and electronic structure of the type 1

Cu site. Inorg. Chem. 42, 4006-17.

97. Xu, F., Palmer, A. E., Yaver, D. S., Berka, R. M., Gambetta, G. A., Brown, S. H. &

Solomon, E. I. (1999). Targeted mutations in a Trametes villosa laccase. Axial

perturbations of the T1 copper. J. Biol. Chem. 274, 12372-5.

98. Solomon, E. I., Szilagyi, R. K., DeBeer George, S. & Basumallick, L. (2004). Electronic

structures of metal sites in proteins and models: contributions to function in blue copper

proteins. Chem. Rev. 104, 419-58.

99. Quintanar, L., Stoj, C., Taylor, A. B., Hart, P. J., Kosman, D. J. & Solomon, E. I. (2007).

Shall we dance? How a multicopper oxidase chooses its electron transfer partner. Acc.

Chem. Res. 40, 445-52.

100. Holm, R. H., Kennepohl, P. & Solomon, E. I. (1996). Structural and Functional Aspects

of Metal Sites in Biology. Chem. Rev. 96, 2239-2314.

101. Solomon, E. I. (2006). Spectroscopic methods in bioinorganic chemistry: blue to green to

red copper sites. Inorg. Chem. 45, 8012-25.

102. Kyritsis, P., Messerschmidt, A., Huber, R., Salmon, G. A. & Sykes, A. G. (1993). Pulse-

radiolysis studies on the oxidised form of the mulicopper enzyme ascorbate oxidase:

evidence for two intramolecular electron-transfer steps. J. Chem. Soc., Dalton Trans.,

731-735.

103. Quintanar, L., Stoj, C., Wang, T. P., Kosman, D. J. & Solomon, E. I. (2005). Role of

aspartate 94 in the decay of the peroxide intermediate in the multicopper oxidase Fet3p.

Biochemistry 44, 6081-91.

105

104. Yoon, J. & Solomon, E. I. (2007). Electronic structure of the peroxy intermediate and its

correlation to the native intermediate in the multicopper oxidases: insights into the

reductive cleavage of the o-o bond. J. Am. Chem. Soc. 129, 13127-36.

105. Yoon, J., Liboiron, B. D., Sarangi, R., Hodgson, K. O., Hedman, B. & Solomon, E. I.

(2007). The two oxidized forms of the trinuclear Cu cluster in the multicopper oxidases

and mechanism for the decay of the native intermediate. Proc. Natl. Acad. Sci. U. S. A.

104, 13609-14.

106. Terzulli, A. & Kosman, D. J. (2010). Analysis of the high-affinity iron uptake system at

the Chlamydomonas reinhardtii plasma membrane. Eukaryot. Cell 9, 815-26.

107. Roberts, S. A., Weichsel, A., Grass, G., Thakali, K., Hazzard, J. T., Tollin, G., Rensing,

C. & Montfort, W. R. (2002). Crystal structure and electron transfer kinetics of CueO, a

multicopper oxidase required for copper homeostasis in Escherichia coli. Proc. Natl.

Acad. Sci. U. S. A. 99, 2766-71.

108. Dick, G. J., Torpey, J. W., Beveridge, T. J. & Tebo, B. M. (2008). Direct identification of

a bacterial manganese(II) oxidase, the multicopper oxidase MnxG, from spores of several

different marine Bacillus species. Appl. Environ. Microbiol. 74, 1527-34.

109. Harris, Z. L., Takahashi, Y., Miyajima, H., Serizawa, M., MacGillivray, R. T. & Gitlin, J.

D. (1995). Aceruloplasminemia: molecular characterization of this disorder of iron

metabolism. Proc. Natl. Acad. Sci. U. S. A. 92, 2539-43.

110. Edwards, J. A. & Bannerman, R. M. (1970). Hereditary defect of intestinal iron transport

in mice with sex-linked anemia. J. Clin. Invest. 49, 1869-71.

111. Taylor, A. B., Stoj, C. S., Ziegler, L., Kosman, D. J. & Hart, P. J. (2005). The copper-iron

connection in biology: structure of the metallo-oxidase Fet3p. Proc. Natl. Acad. Sci. U. S.

A. 102, 15459-64.

106

112. Ziegler, L., Terzulli, A., Gaur, R., McCarthy, R. & Kosman, D. J. (2011). Functional

characterization of the ferroxidase, permease high-affinity iron transport complex from

Candida albicans. Mol. Microbiol. 81, 473-85.

113. Syed, B. A., Beaumont, N. J., Patel, A., Naylor, C. E., Bayele, H. K., Joannou, C. L.,

Rowe, P. S., Evans, R. W. & Srai, S. K. (2002). Analysis of the human hephaestin gene

and protein: comparative modelling of the N-terminus ecto-domain based upon

ceruloplasmin. Protein Eng. 15, 205-14.

114. Zaitseva, I., Zaitsev, V., Card, G., Moshkov, K., Bax, B., Ralph, A. & Lindley, P. (1996).

The X-ray structure of human serum ceruloplasmin at 3.1Ǻ: nature of the copper centres.

J. Biol. Inorg. Chem. 1, 15-23.

115. Chen, H., Attieh, Z. K., Syed, B. A., Kuo, Y. M., Stevens, V., Fuqua, B. K., Andersen, H.

S., Naylor, C. E., Evans, R. W., Gambling, L., Danzeisen, R., Bacouri-Haidar, M., Usta,

J., Vulpe, C. D. & McArdle, H. J. (2010). Identification of zyklopen, a new member of

the vertebrate multicopper ferroxidase family, and characterization in rodents and human

cells. J. Nutr. 140,1728-35.

116. Terzulli, A. J. & Kosman, D. J. (2009). The Fox1 ferroxidase of Chlamydomonas

reinhardtii: a new multicopper oxidase structural paradigm. J. Biol. Inorg. Chem. 14,

315-25.

117. Kosman, D. J. (2008). Substrate entasis and electronic coupling elements in electron

transfer from Fe in a multicopper ferroxidase. Inorganica Chim. Acta 361, 844-849.

118. Bertrand, T., Jolivalt, C., Briozzo, P., Caminade, E., Joly, N., Madzak, C. & Mougin, C.

(2002). Crystal structure of a four-copper laccase complexed with an arylamine: insights

into substrate recognition and correlation with kinetics. Biochemistry 41, 7325-33.

119. Stoj, C. S., Augustine, A. J., Zeigler, L., Solomon, E. I. & Kosman, D. J. (2006).

Structural basis of the ferrous iron specificity of the yeast ferroxidase, Fet3p.

Biochemistry 45, 12741-9.

107

120. Quintanar, L., Gebhard, M., Wang, T. P., Kosman, D. J. & Solomon, E. I. (2004).

Ferrous binding to the multicopper oxidases Saccharomyces cerevisiae Fet3p and human

ceruloplasmin: contributions to ferroxidase activity. J. Am. Chem. Soc. 126, 6579-89.

121. de Silva, D., Davis-Kaplan, S., Fergestad, J. & Kaplan, J. (1997). Purification and

characterization of Fet3 protein, a yeast homologue of ceruloplasmin. J. Biol. Chem. 272,

14208-13.

122. Hassett, R. F., Yuan, D. S. & Kosman, D. J. (1998). Spectral and kinetic properties of the

Fet3 protein from Saccharomyces cerevisiae, a multinuclear copper ferroxidase enzyme.

J. Biol. Chem. 273, 23274-82.

123. Ziegler, L., Terzulli, A., Sedlak, E. & Kosman, D. J. (2010). Core glycan in the yeast

multicopper ferroxidase, Fet3p: a case study of N-linked glycosylation, protein

maturation, and stability. Protein Sci. 19, 1739-50.

124. Wang, T. P., Quintanar, L., Severance, S., Solomon, E. I. & Kosman, D. J. (2003).

Targeted suppression of the ferroxidase and iron trafficking activities of the multicopper

oxidase Fet3p from Saccharomyces cerevisiae. J. Biol. Inorg. Chem. 8, 611-20.

125. Stoj, C. & Kosman, D. J. (2003). Cuprous oxidase activity of yeast Fet3p and human

ceruloplasmin: implication for function. FEBS Lett. 554, 422-6.

126. Stoj, C. S., Augustine, A. J., Solomon, E. I. & Kosman, D. J. (2007). Structure-function

analysis of the cuprous oxidase activity in Fet3p from Saccharomyces cerevisiae. J. Biol.

Chem. 282, 7862-8.

127. Multhaup, G., Schlicksupp, A., Hesse, L., Beher, D., Ruppert, T., Masters, C. L. &

Beyreuther, K. (1996). The amyloid precursor protein of Alzheimer's disease in the

reduction of copper(II) to copper(I). Science 271, 1406-9.

128. Shi, X., Stoj, C., Romeo, A., Kosman, D. J. & Zhu, Z. (2003). Fre1p Cu2+ reduction and

Fet3p Cu1+ oxidation modulate copper toxicity in Saccharomyces cerevisiae. J. Biol.

Chem. 278, 50309-15.

108

129. Holmberg, C. G. & Laurell, C. B. (1948). Investigation serum copper. Isolation of the

copper containing protein, and a description of some its properties. Acta. Chem. Scand. 2,

550-556.

130. Osaki, S., Johnson, D. A. & Frieden, E. (1966). The possible significance of the ferrous

oxidase activity of ceruloplasmin in normal human serum. J. Biol. Chem. 241, 2746-51.

131. Healy, J. & Tipton, K. (2007). Ceruloplasmin and what it might do. J. Neural Transm.

114, 777-81.

132. Patel, B. N. & David, S. (1997). A novel glycosylphosphatidylinositol-anchored form of

ceruloplasmin is expressed by mammalian astrocytes. J. Biol. Chem. 272, 20185-90.

133. Hahn, P., Qian, Y., Dentchev, T., Chen, L., Beard, J., Harris, Z. L. & Dunaief, J. L.

(2004). Disruption of ceruloplasmin and hephaestin in mice causes retinal iron overload

and retinal degeneration with features of age-related macular degeneration. Proc. Natl.

Acad. Sci. U. S. A. 101, 13850-5.

134. Fortna, R. R., Watson, H. A. & Nyquist, S. E. (1999). Glycosyl phosphatidylinositol-

anchored ceruloplasmin is expressed by rat Sertoli cells and is concentrated in detergent-

insoluble membrane fractions. Biol. Reprod. 61, 1042-9.

135. Ryden, L. & Bjork, I. (1976). Reinvestigation of some physicochemical and chemical

properties of human ceruloplasmin (ferroxidase). Biochemistry 15, 3411-7.

136. Vachette, P., Dainese, E., Vasyliev, V. B., Di Muro, P., Beltramini, M., Svergun, D. I.,

De Filippis, V. & Salvato, B. (2002). A key structural role for active site type 3 copper

ions in human ceruloplasmin. J. Biol. Chem. 277, 40823-31.

137. Lindley, P. F., Card, G., Zaitseva, I., Zaitsev, V., Reinhammar, B., Selin-Lindgren, E. &

Yoshida, K. (1997). An X-ray structural study of human ceruloplasmin in relation to

ferroxidase activity. J. Biol. Inorg. Chem. 2, 454-463. .

109

138. Machonkin, T. E., Zhang, H. H., Hedman, B., Hodgson, K. O. & Solomon, E. I. (1998).

Spectroscopic and magnetic studies of human ceruloplasmin: identification of a redox-

inactive reduced Type 1 copper site. Biochemistry 37, 9570-8.

139. Brown, M. A., Stenberg, L. M. & Mauk, A. G. (2002). Identification of catalytically

important amino acids in human ceruloplasmin by site-directed mutagenesis. FEBS Lett.

520, 8-12.

140. Inoue, K., Akaike, T., Miyamoto, Y., Okamoto, T., Sawa, T., Otagiri, M., Suzuki, S.,

Yoshimura, T. & Maeda, H. (1999). Nitrosothiol formation catalyzed by ceruloplasmin.

Implication for cytoprotective mechanism in vivo. J. Biol. Chem. 274, 27069-75.

141. Cha, M. K. & Kim, I. H. (1999). Ceruloplasmin has a distinct active site for the

catalyzing glutathione-dependent reduction of alkyl hydroperoxide. Biochemistry 38,

12104-10.

142. Mukhopadhyay, C. K., Mazumder, B., Lindley, P. F. & Fox, P. L. (1997). Identification

of the prooxidant site of human ceruloplasmin: a model for oxidative damage by copper

bound to protein surfaces. Proc. Natl. Acad. Sci. U. S. A. 94, 11546-51.

143. Young, S. N. & Curzon, G. (1972). A method for obtaining linear reciprocal plots with

caeruloplasmin and its application in a study of the kinetic parameters of caeruloplasmin

substrates. Biochem. J. 129, 273-83.

144. McDermott, J. A., Huber, C. T., Osaki, S. & Frieden, E. (1968). Role of iron in the

oxidase activity of ceruloplasmin. Biochim. Biophys. Acta 151, 541-57.

145. Zaitsev, V. N., Zaitseva, I., Papiz, M. & Lindley, P. F. (1999). An X-ray crystallographic

study of the binding sites of the azide inhibitor and organic substrates to ceruloplasmin, a

multi-copper oxidase in the plasma. J. Biol. Inorg. Chem. 4, 579-87.

146. Yazaki, M., Yoshida, K., Nakamura, A., Furihata, K., Yonekawa, M., Okabe, T.,

Yamashita, N., Ohta, M. & Ikeda, S. (1998). A novel splicing mutation in the

110

ceruloplasmin gene responsible for hereditary ceruloplasmin deficiency with

hemosiderosis. J. Neurol. Sci. 156, 30-4.

147. Kono, S., Suzuki, H., Oda, T., Shirakawa, K., Takahashi, Y., Kitagawa, M. & Miyajima,

H. (2007). Cys-881 is essential for the trafficking and secretion of truncated mutant

ceruloplasmin in aceruloplasminemia. J. Hepatol. 47, 844-50.

148. Hellman, N. E., Kono, S., Miyajima, H. & Gitlin, J. D. (2002). Biochemical analysis of a

missense mutation in aceruloplasminemia. J. Biol. Chem. 277, 1375-80.

149. Kono, S., Suzuki, H., Takahashi, K., Takahashi, Y., Shirakawa, K., Murakawa, Y.,

Yamaguchi, S. & Miyajima, H. (2006). Hepatic iron overload associated with a decreased

serum ceruloplasmin level in a novel clinical type of aceruloplasminemia.

Gastroenterology 131, 240-5.

150. Xu, X., Pin, S., Gathinji, M., Fuchs, R. & Harris, Z. L. (2004). Aceruloplasminemia: an

inherited neurodegenerative disease with impairment of iron homeostasis. Ann. N. Y.

Acad. Sci. 1012, 299-305.

151. Patel, B. N., Dunn, R. J., Jeong, S. Y., Zhu, Q., Julien, J. P. & David, S. (2002).

Ceruloplasmin regulates iron levels in the CNS and prevents free radical injury. J.

Neurosci. 22, 6578-86.

152. Vulpe, C. D., Kuo, Y. M., Murphy, T. L., Cowley, L., Askwith, C., Libina, N., Gitschier,

J. & Anderson, G. J. (1999). Hephaestin, a ceruloplasmin homologue implicated in

intestinal iron transport, is defective in the sla mouse. Nat. Genet. 21, 195-9.

153. Frazer, D. M., Vulpe, C. D., McKie, A. T., Wilkins, S. J., Trinder, D., Cleghorn, G. J. &

Anderson, G. J. (2001). Cloning and gastrointestinal expression of rat hephaestin:

relationship to other iron transport proteins. Am. J. Physiol. Gastrointest. Liver Physiol.

281, G931-9.

154. Chen, H., Attieh, Z. K., Su, T., Syed, B. A., Gao, H., Alaeddine, R. M., Fox, T. C., Usta,

J., Naylor, C. E., Evans, R. W., McKie, A. T., Anderson, G. J. & Vulpe, C. D. (2004).

111

Hephaestin is a ferroxidase that maintains partial activity in sex-linked anemia mice.

Blood 103, 3933-9.

155. Griffiths, T. A., Mauk, A. G. & MacGillivray, R. T. (2005). Recombinant expression and

functional characterization of human hephaestin: a multicopper oxidase with ferroxidase

activity. Biochemistry. 44, 14725-31.

156. Qian, Z. M., Chang, Y. Z., Leung, G., Du, J. R., Zhu, L., Wang, Q., Niu, L., Xu, Y. J.,

Yang, L., Ho, K. P. & Ke, Y. (2007). Expression of ferroportin1, hephaestin and

ceruloplasmin in rat heart. Biochim. Biophys. Acta 1772, 527-32.

157. Qian, Z. M., Chang, Y. Z., Zhu, L., Yang, L., Du, J. R., Ho, K. P., Wang, Q., Li, L. Z.,

Wang, C. Y., Ge, X., Jing, N. L., Li, L. & Ke, Y. (2007). Development and iron-

dependent expression of hephaestin in different brain regions of rats. J. Cell. Biochem.

102, 1225-33.

158. Hudson, D. M., Curtis, S. B., Smith, V. C., Griffiths, T. A., Wong, A. Y., Scudamore, C.

H., Buchan, A. M. & MacGillivray, R. T. (2010). Human hephaestin expression is not

limited to enterocytes of the gastrointestinal tract but is also found in the antrum, the

enteric nervous system, and pancreatic {beta}-cells. Am. J. Physiol. Gastrointest. Liver

Physiol. 298, G425-32.

159. Danzeisen, R., Fosset, C., Chariana, Z., Page, K., David, S. & McArdle, H. J. (2002).

Placental ceruloplasmin homolog is regulated by iron and copper and is implicated in

iron metabolism. Am. J. Physiol. Cell Physiol. 282, C472-8.

160. Cui, R., Duan, X. L., Anderson, G. J., Qiao, Y. T., Yu, P., Qian, Z. M., Yoshida, K.,

Takeda, S., Guo, P., Yang, Z. L. & Chang, Y. Z. (2009). Age-dependent expression of

hephaestin in the brain of ceruloplasmin-deficient mice. J. Trace Elem. Med. Biol. 23,

290-9.

112

161. Harris, Z. L., Durley, A. P., Man, T. K. & Gitlin, J. D. (1999). Targeted gene disruption

reveals an essential role for ceruloplasmin in cellular iron efflux. Proc. Natl. Acad. Sci.

U. S. A. 96, 10812-7.

162. Bannerman, R. M. & Cooper, R. G. (1966). Sex-linked anemia: a hypochromic anemia of

mice. Science 151, 581-2.

163. Kingston, P. J., Bannerman, C. E. & Bannerman, R. M. (1978). Iron deficiency anaemia

in newborn sla mice: a genetic defect of placental iron transport. Br. J. Haematol. 40,

265-76.

164. Danzeisen, R., Ponnambalam, S., Lea, R. G., Page, K., Gambling, L. & McArdle, H. J.

(2000). The effect of ceruloplasmin on iron release from placental (BeWo) cells;

evidence for an endogenous Cu oxidase. Placenta 21, 805-12.

165. Cherukuri, S., Potla, R., Sarkar, J., Nurko, S., Harris, Z. L. & Fox, P. L. (2005).

Unexpected role of ceruloplasmin in intestinal iron absorption. Cell Metab. 2, 309-19.

166. Mukhopadhyay, C. K., Mazumder, B. & Fox, P. L. (2000). Role of hypoxia-inducible

factor-1 in transcriptional activation of ceruloplasmin by iron deficiency. J. Biol. Chem.

275, 21048-54.

167. Tapryal, N., Mukhopadhyay, C., Das, D., Fox, P. L. & Mukhopadhyay, C. K. (2009).

Reactive oxygen species regulate ceruloplasmin by a novel mRNA decay mechanism

involving its 3'-untranslated region: implications in neurodegenerative diseases. J. Biol.

Chem. 284, 1873-83.

168. Sampath, P., Mazumder, B., Seshadri, V. & Fox, P. L. (2003). Transcript-selective

translational silencing by gamma interferon is directed by a novel structural element in

the ceruloplasmin mRNA 3' untranslated region. Mol. Cell. Biol. 23, 1509-19.

169. Persichini, T., Maio, N., di Patti, M. C., Rizzo, G., Colasanti, M. & Musci, G. (2010).

Interleukin-1beta induces ceruloplasmin and ferroportin-1 gene expression via MAP

kinases and C/EBPbeta, AP-1, and NF-kappaB activation. Neurosci. Lett. 484, 133-8.

113

170. Chen, H., Su, T., Attieh, Z. K., Fox, T. C., McKie, A. T., Anderson, G. J. & Vulpe, C. D.

(2003). Systemic regulation of Hephaestin and Ireg1 revealed in studies of genetic and

nutritional iron deficiency. Blood 102, 1893-9.

171. Hinoi, T., Gesina, G., Akyol, A., Kuick, R., Hanash, S., Giordano, T. J., Gruber, S. B. &

Fearon, E. R. (2005). CDX2-regulated expression of iron transport protein hephaestin in

intestinal and colonic epithelium. Gastroenterology 128, 946-61.

172. Nittis, T. & Gitlin, J. D. (2004). Role of copper in the proteosome-mediated degradation

of the multicopper oxidase hephaestin. J. Biol. Chem. 279, 25696-702.

173. Chen, H., Huang, G., Su, T., Gao, H., Attieh, Z. K., McKie, A. T., Anderson, G. J. &

Vulpe, C. D. (2006). Decreased hephaestin activity in the intestine of copper-deficient

mice causes systemic iron deficiency. J. Nutr. 136, 1236-41.

174. Gitlin, J. D., Schroeder, J. J., Lee-Ambrose, L. M. & Cousins, R. J. (1992). Mechanisms

of caeruloplasmin biosynthesis in normal and copper-deficient rats. Biochem. J. 282, 835-

175. Bardou-Jacquet, E., Island, M. L., Jouanolle, A. M., Detivaud, L., Fatih, N., Ropert, M.,

Brissot, E., Mosser, A., Maisonneuve, H., Brissot, P. & Loreal, O. (2011). A novel

N491S mutation in the human SLC11A2 gene impairs protein trafficking and in

association with the G212V mutation leads to microcytic anemia and liver iron overload.

Blood Cells Mol. Dis.

176. Beaumont, C., Delaunay, J., Hetet, G., Grandchamp, B., de Montalembert, M. &

Tchernia, G. (2006). Two new human DMT1 gene mutations in a patient with microcytic

anemia, low ferritinemia, and liver iron overload. Blood 107, 4168-70.

177. Blanco, E., Kannengiesser, C., Grandchamp, B., Tasso, M. & Beaumont, C. (2009). Not

all DMT1 mutations lead to iron overload. Blood Cells Mol. Dis. 43, 199-201.

178. Kato, J., Fujikawa, K., Kanda, M., Fukuda, N., Sasaki, K., Takayama, T., Kobune, M.,

Takada, K., Takimoto, R., Hamada, H., Ikeda, T. & Niitsu, Y. (2001). A mutation, in the

114

iron-responsive element of H ferritin mRNA, causing autosomal dominant iron overload.

Am. J. Hum. Genet. 69, 191-7.

179. Curtis, A. R., Fey, C., Morris, C. M., Bindoff, L. A., Ince, P. G., Chinnery, P. F.,

Coulthard, A., Jackson, M. J., Jackson, A. P., McHale, D. P., Hay, D., Barker, W. A.,

Markham, A. F., Bates, D., Curtis, A. & Burn, J. (2001). Mutation in the gene encoding

ferritin light polypeptide causes dominant adult-onset basal ganglia disease. Nat. Genet.

28, 350-4.

180. Levi, S., Cozzi, A. & Arosio, P. (2005). Neuroferritinopathy: a neurodegenerative

disorder associated with L-ferritin mutation. Best Pract. Res. Clin. Haematol. 18, 265-76.

181. Cazzola, M., Bergamaschi, G., Tonon, L., Arbustini, E., Grasso, M., Vercesi, E., Barosi,

G., Bianchi, P. E., Cairo, G. & Arosio, P. (1997). Hereditary hyperferritinemia-cataract

syndrome: relationship between phenotypes and specific mutations in the iron-responsive

element of ferritin light-chain mRNA. Blood 90, 814-21.

182. Koeppen, A. H. (2011). Friedreich's ataxia: pathology, pathogenesis, and molecular

genetics. J. Neurol. Sci. 303, 1-12.

183. De Domenico, I., Ward, D. M., Nemeth, E., Vaughn, M. B., Musci, G., Ganz, T. &

Kaplan, J. (2005). The molecular basis of ferroportin-linked hemochromatosis. Proc.

Natl. Acad. Sci. U. S. A. 102, 8955-60.

184. Hamill, R. L., Woods, J. C. & Cook, B. A. (1991). Congenital atransferrinemia. A case

report and review of the literature. Am. J. Clin. Pathol. 96, 215-8.

185. Shamsian, B. S., Rezaei, N., Arzanian, M. T., Alavi, S., Khojasteh, O. & Eghbali, A.

(2009). Severe hypochromic microcytic anemia in a patient with congenital

atransferrinemia. Pediatr. Hematol. Oncol. 26, 356-62.

186. Chen, J. & Enns, C. A. (2011). Hereditary hemochromatosis and transferrin receptor 2.

Biochim. Biophys. Acta. 1820, 256-63.

115

187. Rochette, J., Le Gac, G., Lassoued, K., Ferec, C. & Robson, K. J. (2010). Factors

influencing disease phenotype and penetrance in HFE haemochromatosis. Hum. Genet.

128, 233-48.

188. Goldberg, Y. P. Juvenile Hereditary Hemochromatosis. 2005 Feb 17 (Updated 2011 Aug

11). In: Pagon, R.A., Bird, T.D., Dolan, C.R., at al., editors. GeneReviewsTM [Internet].

PMID: 20301349.

189. Hodges, R. S., Heaton, R. J., Parker, J. M., Molday, L. & Molday, R. S. (1988). Antigen-

antibody interaction. Synthetic peptides define linear antigenic determinants recognized

by monoclonal antibodies directed to the cytoplasmic carboxyl terminus of rhodopsin. J.

Biol. Chem. 263, 11768-75.

190. Palmiter, R. D., Behringer, R. R., Quaife, C. J., Maxwell, F., Maxwell, I. H. & Brinster,

R. L. (1987). Cell lineage ablation in transgenic mice by cell-specific expression of a

toxin gene. Cell 50, 435-43.

191. Kammann, M., Laufs, J., Schell, J. & Gronenborn, B. (1989). Rapid insertional

mutagenesis of DNA by polymerase chain reaction (PCR). Nucleic. Acids Res. 17, 5404.

192. Macauley-Patrick, S., Fazenda, M. L., McNeil, B. & Harvey, L. M. (2005). Heterologous

protein production using the Pichia pastoris expression system. Yeast 22, 249-70.

193. Molday, R. S. & MacKenzie, D. (1983). Monoclonal antibodies to rhodopsin:

characterization, cross-reactivity, and application as structural probes. Biochemistry 22,

653-60.

194. Calabrese, L., Mateescu, M. A., Carbonaro, M. & Mondovi, B. (1988). Reexamination of

spectroscopic properties of ceruloplasmin freshly isolated with a novel very rapid single-

step procedure. Biochem. Int. 16, 199-208.

195. Cuatrecasas, P. (1970). Protein purification by affinity chromatography. Derivatizations

of agarose and polyacrylamide beads. J. Biol. Chem. 245, 3059-65.

116

196. Stern, R. V. & Frieden, E. (1990). Detection of rat ceruloplasmin receptors using

fluorescence microscopy and microdensitometry. Anal. Biochem. 190, 48-56.

197. do Nascimento Rocha, S. A., Dantas, A. F., Jaeger, H. V., Costa, A. C., Leao Edos, S. &

Goncalves, M. R. (2008). Spectrofotometric determination of copper in sugar cane spirit

using biquinoline in the presence of ethanol and Triton X-100. Spectrochim. Acta. A.

Mol. Biomol. Spectrosc. 71, 1414-8.

198. Stookey, L. L. (1970). Ferrozine - a new spectrophotometric reagent for iron. Anal.

Chem. 42, 779-781.

199. Rice, E. W. (1962). Standardization of ceruloplasmin activity in terms of International

Enzyme Units. Oxidative formation of "Bandrowski's base" from p-phenylenediamine by

ceruloplasmin. Anal. Biochem. 3, 452-6.

200. Baez, S., Segura-Aguilar, J., Widersten, M., Johansson, A. S. & Mannervik, B. (1997).

Glutathione transferases catalyse the detoxication of oxidized metabolites (o-quinones) of

catecholamines and may serve as an antioxidant system preventing degenerative cellular

processes. Biochem. J. 324, 25-8.

201. Curzon, G. (1961). Some properties of coupled iron-caeruloplasmin oxidation systems.

Biochem. J. 79, 656-63.

202. Vashchenko, G., Bleackley, M. R., Griffiths, T. A. & MacGillivray, R. T. (2011).

Oxidation of organic and biogenic amines by recombinant human hephaestin expressed

in Pichia pastoris. Arch. Biochem. Biophys. 514, 50-6.

203. Li, L., Vulpe, C. D. & Kaplan, J. (2003). Functional studies of hephaestin in yeast:

evidence for multicopper oxidase activity in the endocytic pathway. Biochem. J. 375,

793-8.

204. Bielli, P., Bellenchi, G. C. & Calabrese, L. (2001). Site-directed mutagenesis of human

ceruloplasmin. production of a proteolytically stable protein and structure-activity

relationships of type 1 sites. J. Biol. Chem. 276, 2678-85.

117

205. Kim, C., Lorenz, W. W., Hoopes, J. T. & Dean, J. F. (2001). Oxidation of phenolate

siderophores by the multicopper oxidase encoded by the Escherichia coli yacK gene. J.

Bacteriol. 183, 4866-75.

206. Sreekrishna, K., Brankamp, R. G., Kropp, K. E., Blankenship, D. T., Tsay, J. T., Smith,

P. L., Wierschke, J. D., Subramaniam, A. & Birkenberger, L. A. (1997). Strategies for

optimal synthesis and secretion of heterologous proteins in the methylotrophic yeast

Pichia pastoris. Gene 190, 55-62.

207. Musci, G., Di Marco, S., Bellenchi, G. C. & Calabrese, L. (1996). Reconstitution of

ceruloplasmin by the Cu(I)-glutathione complex. Evidence for a role of Mg2+ and ATP.

J. Biol. Chem. 271, 1972-8.

208. Bonomi, F., Kurtz, D. M., Jr. & Cui, X. (1996). Ferroxidase activity of recombinant

Desulfovibrio vulgaris rubrerythrin. J. Biol. Inorg. Chem. 1, 67-72.

209. Aasa, R., Malmstroem, B. G. & Saltman, P. (1963). The Specific Binding Of Iron(III)

And Copper(II) To Transferrin And Conalbumin. Biochim. Biophys. Acta 75, 203-22.

210. Musci, G., Bonaccorsi di Patti, M. C. & Calabrese, L. (1995). Modulation of the redox

state of the copper sites of human ceruloplasmin by chloride. J. Protein Chem. 14, 611-9.

211. Ma, Y., Yeh, M., Yeh, K. Y. & Glass, J. (2006). Iron Imports. V. Transport of iron

through the intestinal epithelium. Am. J. Physiol. Gastrointest. Liver Physiol. 290, G417-

22.

212. Zgirski, A. & Frieden, E. (1990). Binding of Cu(II) to non-prosthetic sites in

ceruloplasmin and bovine serum albumin. J. Inorg. Biochem. 39, 137-48.

213. Osaki, S. (1966). Kinetic studies of ferrous ion oxidation with crystalline human

ferroxidase (ceruloplasmin). J. Biol. Chem. 241, 5053-9.

214. Huber, C. T. & Frieden, E. (1970). Substrate activation and the kinetics of ferroxidase. J.

Biol. Chem. 245, 3973-8.

118

215. Ryan, T. P., Grover, T. A. & Aust, S. D. (1992). Rat ceruloplasmin: resistance to

proteolysis and kinetic comparison with human ceruloplasmin. Arch. Biochem. Biophys.

293, 1-8.

216. McKee, D. J. & Frieden, E. (1971). Binding of transition metal ions by ceruloplasmin

(ferroxidase). Biochemistry. 10, 3880-3.

217. Kataoka, K., Tsukamoto, K., Kitagawa, R., Ito, T. & Sakurai, T. (2008). Compensatory

binding of an asparagine residue to the coordination-unsaturated type I Cu center in

bilirubin oxidase mutants. Biochem. Biophys. Res. Commun. 371, 416-9.

218. Babcock, M., de Silva, D., Oaks, R., Davis-Kaplan, S., Jiralerspong, S., Montermini, L.,

Pandolfo, M. & Kaplan, J. (1997). Regulation of mitochondrial iron accumulation by

Yfh1p, a putative homolog of frataxin. Science 276, 1709-12.

219. Zoppellaro, G., Sakurai, T. & Huang, H. (2001). A novel mixed valence form of Rhus

vernicifera laccase and its reaction with dioxygen to give a peroxide intermediate bound

to the trinuclear center. J. Biochem. 129, 949-53.

220. Esposito, B. P., Breuer, W., Sirankapracha, P., Pootrakul, P., Hershko, C. & Cabantchik,

Z. I. (2003). Labile plasma iron in iron overload: redox activity and susceptibility to

chelation. Blood 102, 2670-7.

221. Kita, T., Kume, N., Minami, M., Hayashida, K., Murayama, T., Sano, H., Moriwaki, H.,

Kataoka, H., Nishi, E., Horiuchi, H., Arai, H. & Yokode, M. (2001). Role of oxidized

LDL in atherosclerosis. Ann. N. Y. Acad. Sci. 947, 199-205; discussion 205-6.

222. Sun, J., Steenbergen, C. & Murphy, E. (2006). S-nitrosylation: NO-related redox

signaling to protect against oxidative stress. Antioxid. Redox. Signal. 8, 1693-705.

223. Paradis, M., Gagne, J., Mateescu, M. A. & Paquin, J. (2010). The effects of nitric oxide-

oxidase and putative glutathione-peroxidase activities of ceruloplasmin on the viability of

cardiomyocytes exposed to hydrogen peroxide. Free Radic. Biol. Med. 49, 2019-27.

119

APPENDICES

A. Oxidation of organic amines (chemical equations).

120

B. Oxidation of biogenic amines (chemical equations).

121

C. Kinetic parameters for different samples of CpHp and variants.

2+ CpHp and Number of the Km for pPD kcat for pPD Km for high- kcat` for Fe kcat for pPD to 1 -1 -1 3 2+ variants sample (mM) (min ) affinity oxidation (min ) kcat` for Fe of Fe2+ (µM)2 ratio CpHp 1 1.1 ± 0.1 2.3 ± 0.1 6.1 ± 1.8 4.5 0.50 2 0.9 ± 0.1 3.4 ± 0.1 4.3 ± 2.1 4.5 0.74 3 0.85 ± 0.10 8.7 ± 0.3 3.1 ± 1.2 6.5 1.03 4 1.3 ± 0.2 4.7 ± 0.3 3.5 ± 0.2 8.5 0.72 IB6 1 0.36 ± 0.06 0.50 ± 0.03 2.0 ± 0.7 0.67 0.75 2 0.48 ± 0.10 0.50 ± 0.04 2.1 ± 0.4 0.67 0.75 3 0.48 ± 0.10 0.50 ± 0.04 2.2 ± 0.3 0.58 0.85 4 0.30 ± 0.06 0.48 ± 0.03 2.4 ± 0.7 0.47 1.02 IB4 1 0.37 ± 0.03 4.2 ± 0.1 n/a 0.39 10.8 2 0.40 ± 0.07 5.2 ± 0.3 0.66 7.8 3 0.34 ± 0.06 4.3 ± 0.3 0.55 7.8 4 0.38 ± 0.05 5.9 ± 0.3 0.52 11.4 IB2 1 0.38 ± 0.04 3.0 ± 0.1 n/a 0.29 10.6 2 0.33 ± 0.07 1.3 ± 0.1 0.12 10.9 3 0.35 ± 0.06 1.3 ± 0.1 0.16 8.3 4 0.32 ± 0.07 1.6 ± 0.1 0.13 13.0 IB 1 0.24 ± 0.07 0.46 ± 0.04 n/a 0.079 5.9 2 0.23 ± 0.07 0.45 ± 0.03 0.085 5.3 3 0.19 ± 0.05 0.33 ± 0.03 0.059 5.6 4 0.21 ± 0.05 0.34 ± 0.03 0.064 5.3 1 – This table shows data for individual samples of CpHp and variants. Errors represent goodness of the fit to Michaelis-Menten equation. 2+ 2 - Km for Fe was calculated using Eadie-Hofstee plot; the initial ferroxidation rates were measured at six substrate concentrations ranging from 2 to 10 µM of iron. 2+ 3 - kcat` for Fe is the number of ferrous ions oxidized per minute by one molecule of enzyme at the iron concentration of 87.5 µM.

122