UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Metal binding properties of p97 and the trafficking of a novel iron internalization pathway by GPI-anchored… Tiong, Jacqueline Woang Cheing 2001

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

Item Metadata

Download

Media
831-ubc_2001-715469.pdf [ 16.93MB ]
Metadata
JSON: 831-1.0090775.json
JSON-LD: 831-1.0090775-ld.json
RDF/XML (Pretty): 831-1.0090775-rdf.xml
RDF/JSON: 831-1.0090775-rdf.json
Turtle: 831-1.0090775-turtle.txt
N-Triples: 831-1.0090775-rdf-ntriples.txt
Original Record: 831-1.0090775-source.json
Full Text
831-1.0090775-fulltext.txt
Citation
831-1.0090775.ris

Full Text

M E T A L BINDING PROPERTIES O F P97 A N D T H E T R A F F I C K I N G OF A N O V E L I R O N I N T E R N A L I Z A T I O N P A T H W A Y B Y G P I - A N C H O R E D P97 by J A C Q U E L I N E W O A N G C H E I N G T I O N G B . S c , University of British Columbia (Hons), 1994 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E STUDIES Department of Microbiology and Immunology, Biomedical Research Centre, and the Biotechnology Laboratory We accept this thesis as conforming T H E U N I V E R S I T Y O F BRITISH C O L U M B I A September 2001 © Jacqueline Woang Cheing Tiong 2001 UBC Special Collections - Thesis Authorisation Form 30/9/01 11:46 PM In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree a t the U n i v e r s i t y of B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department of M 1 dflfetQU)6r1 j I M^UdO LflfrV The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada Date S&T 3 0 , ZOO | http://www.library.ubc.ca/spcoll/thesauth.html Page 1 of 1 Abstract Iron is an essential for celular metabolism yet excessive iron can lead to deleterious oxidative damages in the cels. Since humans have no physiological pathways for iron excretion, dietary iron absorption in the intestine is tightly regulated. Although the essential nature of iron for normal celular functions is beyond doubt, only in the last two decades has its uptake and regulation been relatively wel understood. The transferrin/ transferrin receptor-independent iron transport pathways play a major role in iron overload diseases such as hemochromatosis. However, very little is known about the molecular details of these pathways. P97, also known as melanotransferin, is an example of iron binding proteins that mediates such pathways. P97 is a member of the transferrin family, which includes serum transferrin, lactoferrin and ovotransferrin. Unlike members of the transferrin family, which are all serum proteins, p97, in addition to being a serum protein, also exists as a glycosylphosphatidylinositol-anchored protein. The primary iron-binding site in the N-terminal lobe is highly conserved among the other transferrin members. However, the iron binding site at the C-terminal lobe of p97 is not. This leads to the question as to whether p97 is a functional iron transport protein. In addition, the iron binding afinity of p97 has never been ascertained. Therefore, an aspect of this thesis was to determine the binding afinity of iron to p97 and to establish whether the C-lobe of p97 can bind iron. The results in this thesis wil demonstrate that the iron binding afinity for the N-lobe of p97 is 2.2 x 1017 M"1 at 25 °C. In addition, the data wil show that C-lobe is able to bind iron albeit very weakly. Metals such as iron, copper, aluminum and zinc have been implicated in the pathology of several neurological disorders. The expression of p97 in the brain capilary endothelium and the elevated level of serum p97 associated with Alzheimer s disease have led to the hypothesis that p97 may play an important role in disease progression of the disease. Therefore, another goal of this study was to establish whether p97 can bind metals such as copper, aluminum and zinc. The results wil show that copper, aluminum and zinc are able to interfere with iron binding to GPI-anchored p97 and soluble p97 is able to bind to these metals as shown by urea polyacrylamide gel electrophoresis. Though the internalization pathways of many GPI-anchored proteins have been studied, debate still exists as to whether caveolae or clathrin-coated vesicles are involved in their uptake. Surprisingly, very little is known about the fate of the ligands internalized by GPI-anchored molecules. Also, there is controversy as to whether iron bound to p97 can be donated to ferritin. Therefore, another aspect of this thesis was to examine the uptake of GPI-anchored p97 and its ligand, iron. The results from confocal immunofluorescence microscopy and sub-celular fractionation demonstrate that GPI-anchored p97 bound with iron is endocytosed into SK-MEL 28 cels via caveolae and trafics to early endosomes. In addition, the data also shows that GPI-anchored p97 is present in the same vesicles as Nramp2, a metal transporter thought to mediate iron transport across the vesicular membrane into the cytoplasm. Furthermore, it is shown that iron taken up by p97 becomes bound by the iron storage molecule, ferritin. The data establishes that iron bound to GPI-anchored p97 is internalized through caveolae and donated to ferritin for celular metabolism and supports an improved understanding of iii both GPI anchor mediated uptake of ligand and the novel transferrin/ transferrin receptor-independent mechanism of iron uptake. Table of Contents Abstract ii Table of contents v List of Figures xi List of Tables xiii List of Abbreviations xiv Acknowledgement xvi Dedication xvii Chapter 1 Introduction 1 I. A brief history of iron in medicine 1 II. Roles of iron in celular metabolism 2 III. Diseases related to iron metabolism 6 A. Hereditary hemochromatosi  7 B. Juvenile hemochromatosi  12 C. Neonatal hemochromatosi  13 D. African dietary iron overload 13 E. Neurodegenerative disease 13 F. Atransferinemia 16 G. Aceruloplasminemia 16 H. Hyperferitnemia 17 I. Disorders related to mitochondrial iron homeostasis 17 v a. Friedreich s ataxia 17 b. X-linked sideroplastic anemia 18 IV. P97 and the transferrin family 19 A. Brief overview of serum transferrin, lactoferrin and ovotransferin 19 B. Overview of p97 35 a. P97 as an iron binding protein 35 b. Regulation of p97 expression 37 c. Orthologues of p97 38 d. Alzheimer s disease and p97 40 e. Other possible functions of p97 41 V. Transferrin/ transferrin receptor independent iron transport 42 A. Transferin-dependent transferrin receptor-independent pathway 43 B. Transferin-independent transferrin receptor-dependent pathway 44 C. Transferrin/ transferrin receptor-independent pathway 44 a. GPI-anchored p97 44 b. Iron absorption in the intestine 45 i. Ferous iron absorption in the intestine 45 ii. Paraferritin mediated iron absorption 49 iii. Heme absorption 49 VI. Hypothesis and General approach 50 Chapter 2 Materials and Methods I. Tissue culture, Cels and Antibodies 54 54 II. Iron-binding and competion experiments 58 III. Dialysis of soluble p97 and iron loading protocols 60 IV. Urea polyacrylamide gel electrophoresis 61 V. UV-Visible spectroscopy study of iron binding to soluble p97 62 VI. Diferential scanning calorimetry 62 VI. Immunoprecipitation of ferritin. 63 VII. Endocytosis inhibition experiment 64 IX. Immunofluorescence staining and confocal laser scanning microscopy 65 X. Cryo-immunoelectron microscopy 67 XI. Sucrose gradient sub-celular fractionation and western blot 68 XI. Transfection of Nramp2 into SK-MEL 28 cels 70 Chapter 3: Metal binding properties of p97 71 I. Other metals can compete with iron for p97 71 A. Rationale 71 B. Results I: Aluminum, copper and zinc can compete with iron C. Results II: Soluble p97 can bind iron, copper and zinc 76 D. Discussion 81 a. Copper, aluminum and zinc can compete with iron for binding to GPI-anchored p97 82 b. The binding of iron, copper, aluminum and zinc to soluble p97 84 vii II. The iron-binding stoichiometry of soluble p97 88 A. Rationale 88 B. Results I: Determination of the peak absorbance wavelength of iron saturated p97 90 C. Results II: Iron-binding titration for soluble p97 90 D. Discussion 93 III. Diferential scanning calorimetry studies on iron binding to soluble p97 94 A. Rationale 94 B. Results: Diferential scanning calorimetry study of iron binding to soluble p97 98 C. Discussion 103 Chapter 4: Endocytosis of GPI-anchored p97 109 I. Brief overview of endocytosis 109 A. Caveolae mediated endocytosis 109 B. Endocytosis of GPI-anchored proteins 116 II. Confocal Immunofluorescence microscopy studies on the internalization pathway of GPI-anchored p97 117 A. Rationale 117 B. Results I: GPI-anchored p97 is internalized through caveolae 118 C. Results II: Cryo-immunoelectron microscopy study on the localization of GPI-anchored p97 126 D. Results III: Iron uptake by GPI-anchored p97 is disrupted by sterol sequestering reagents 127 E. Results IV: GPI-anchored p97 fractionates with caveolae by sub-celular fractionation 129 F. Discussion 132 III. GPI-anchored p97 trafics to endosomes after internalization. 135 A. Rationale 135 B. Results I: GPI-anchored p97 is traficked to the endosomes 136 C. Results II: GPI-anchored p97 trafics to the endosomes more slowly than transferrin receptor 141 D. Discussion 145 IV. The fate of iron transported by GPI-anchored p97 146 A. Rationale 146 B. Results I: Iron internalized by GPI-anchored p97 is fractionated into caveolae fraction 147 C. Results II: GPI-anchored p97 can be localized to vesicles containing Nramp2 151 D. Results III: Iron internalized by GPI-anchored p97 is donated to Ferritin 156 E. Discussion 159 Chapter 5 Appendix I Appendix II Concluding remarks and future perspective Crystallization of soluble p97 Derivation of equation relating thermodynamic parameters to 164 171 binding afinity 173 Bibliography 177 List of Figures Figure 1: The complete protein sequence alignment of members of the transferin family 20 Figure 2: The bi-lobed structure of human transferrin and p97 30 Figure 3: Iron uptake mediated by transferrin/ transferrin receptor 32 Figure 4: The intestinal iron absorption pathways. 46 Figure 5: The efects of competiors metal citrate on the amount of iron bound to GPI-anchored p97 74 Figure 6: Urea polyacrylamide gels of human serum transferrin human soluble p97 77 Figure 7: The three possible interpretations of iron loaded p97 86 Figure 8: UV-visible spectrum of human p97 91 Figure 9: Iron nitrilotriacetate (Fe-NTA) titration of serum transferrin and soluble p97 92 Figure 10: Sketch diagram of the inner chamber of a typical diferential scanning calorimeter and a typical thermogram of hen egg lysozyme 97 Figure 11: Diferential scanning calorimetry apo-p97 100 Figure 12: DSC thermograms of p97 loaded with various ratio iron to p97 102 Figure 13: The three possible unfolding paterns of iron loaded soluble human p97 during a thermoscan 105 Figure 14: The four diferent traficking pathways of endocytosis mediated by caveolae 113 Figure 15: Confocal immunofluorescence staining of GPI-anchored p97 with either clathrin or caveolin. 119 Figure 16: Confocal immunofluorescence staining of transferrin receptor with either clathrin or caveolin 123 Figure 17: Cryo-immunoelectron micrographs of p97 with caveolin and transferrin receptor with (3-adaptin 128 Figure 18: The efects of nystatin and filipin on the iron uptake by GPI-anchored p97 and transferrin receptor 130 Figure 19: Western blot of the sub-celular fractions of SK-MEL 28 133 Figure 20: Confocal immunofluorescence staining of transferrin receptor and p97 with endosomal marker EE A1 137 Figure 21: Confocal immunofluorescence staining of transferrin receptor, GPI-anchored p97 and endosomal marker EEA1 142 Figure 22: Sub-celular fractionation of 5Fe nitrilotriacetate or 5FeNTA loaded human transferrin labeled SK-MEL 28 cels 148 Figure 23: Confocal immunofluorescence staining of transferrin receptor and p97 with Nramp2 in SK-MEL 28 cels 152 Figure 24: Iron bound to GPI-anchored p97 is incorporated into ferritin 157 Figure 25: The proposed traficking pathway of iron bound to GPI-anchored p97 161 Figure 26: Crystal of soluble p97 172 List of Tables Table I: Table II: Table I: Examples of some iron proteins involved in diferent mammalin celular functions 5 Diseases related to iron metabolism 8 Table Ilia: Primary antibodies used in the experiments presented in this thesis 56 Table Illb: Secondary antibodies used in the experiments presented in this thesis 57 Table IV: Percentage of soluble p97 loaded with diferent metals 80 Table V: The parameters of the two-state fits to the unfolding thermogram ofapo-p97 101 List of Abbreviations: Al Aluminum ATP Adenosine triphosphate BSA Bovine serum albumin CHO Chinese hamster ovary Cu Copper Cy5 Cyarine 5 Dcytb Duodenal ferric reductase DMEM Dulbecco modifed eagle s media DMSO Di-methyl-sulfoxide DMT-1 Divalent metal transporter DNA Deoxyribonucleotide DSC Diferential scanning calorimetry ECL Enhanced chemiluminescence EDTA Ethylenediaminetetraacetic acid EEA1 Early endosome associated antigen 1 EGFP Enhanced green fluorescence protein Fe Iron GPI Glycosylphosphatidylinositol HEPES 2-Hydroxyethyl-piperazine-2-ethanesulfonic acid HFE Protein responsible for hereditary hemochromatosi  HLA Human Leukocyte Antigen hpx Hypotransferinemic xiv H2O2 Hydrogen peroxide HO Hydroxyl radical HO" Hydroxyl ion IMDM Iscove's modifed dulbecco media IPvEGl Iron regulated transporter MEM Minimum essential media NEAA Non essential amino acids mRNA messenger ribonucleotide Nramp2 Natural resistance associated macrophage protein 2 NRS Normal rabbit serum NTA Nitrilotriacetate 02»" Superoxide 0 2 Oxygen P97+TRVb Human p97 transfected TRVb PBS Phosphate bufered saline PEG Polyethylene glycol PI-PLC Phosphatidyl-phosphoinositol lipase C TfR Transferin receptor TfR+TRVb Human transferrin receptor transfected TRVb TRVb Chinese hamster ovary cel with defective endogenous transferrin receptor UV Ultraviolet XANES X-ray absorption near edge structure Zn Zinc xv Acknowledgements This thesis could not have been completed without the help of others. I am very grateful for the support and assistance of many people over the years, as their individual contributions have combined to make the completion of this thesis possible. First and foremost, I express my heartfelt gratiude to my research supervisor and mentor, Dr. Wilf Jeferies for the opportunity to work on an exciting project. His advice and "pep talk" during the chalenging times when experiments did not work have propeled me to face the problems. I am also grateful to my commitee members, Drs. Linda Matsuuchi, Pauline Johnson and Rob McMaster, for their valuable guidance and advice over the years. I would also like to thank al my lab mates (past and present): Ian, Gregor, Alex, Judie, Daphne, Gaba, Malcolm, Greg, Maya, Dara, Jamie, Stanley, Kendra, Gene, Mei-Mei, Jason, Jason Xin, Qian-Jing, Francesca, Ray, Andy, Linda, Susan, Laura, KB, Kaan, Joanne, Mat. I thank Brandie Walker and Drs. Maki Ujie, Cheryl Pfeifer and Sarb Ner for critical reading of my thesis and Dr. Jacob Hodgson for our late night discussions on life and science. I thank Dr. Elaine Humphries of the Electron Microscopy facility of UBC. Her enthusiasm for science, her patience in teaching me the various microscopy techniques, and her endless supply of cookies and teas have made learning microscopy fun. I appreciate the help from members of the EM facility: Ying Teen, Ashka, and Kim. Drs. Lacey Samuels and Deidre Mcintosh and Nely Pant have provided valuable technical assistance to me. I acknowledge Dr. Louise Creagh, who shared with me her knowledge and expertise in calorimetry and her tireless patience in answering al my questions. I am xvi grateful to Dr. Mike Murphy and members of his lab, Elena, Helena , Gloria and Marty for their assistance in UV-visible spectroscopy and protein crystalization. I appreciate the help of Sheley Smal, the graduate secretary of Department of Microbiology and Immunology. Her sense of humor about life has made my studies in the department enjoyable. Last but not least, I am eternaly grateful to my parents, brother and sister who have always been there to share my joy when things are going wel and give me a shoulder to cry on when things are not. I was financially supported by the University Graduate Felowship from the University of British Columbia during my studies with the continued subsidy from my parents, without which I would not be able to live comfortably. xvi Dedication This thesis is dedicated to my mother, Jenny S.J. Lam and my father Stanley Y.S. Tiong. Your loves and support have made this journey possible and for this I am eternaly grateful. xvii Chapter 1 Introduction I. A brief history of iron in medicine Mesopotamian civilization long ago regarded iron as the "metal of heaven". The belief in the celestial origin of the ore, which has its inference with the stars, is instiled in its Greek name—sidero. The shower of meteorites, which indeed contained iron, made the Greeks and the Romans believe that iron was a gift from the heavens. Ancient physicians who observed the similarity in the color of blood and iron rust established the association of iron with life. Ebers Papyrus, the oldest Egyptian pharmacopoeia manuscript dated 1500 BC, recorded the use of iron as a remedy for baldness (Ebbel, 1937). Hippocrates, the great Greek physician, believed in using iron salts for relieving gastrointestinal tract ailments. When Arabic civilzations blossomed after the fal  of the Roman Empire, the famous Arabic physician Ibn Sina compiled a massive catalogue of 10th century medical knowledge caled the Al-Qanun, also known as the Canons of Medicine in the West. He cited the use of iron for improving incontinence of urination, restraining menstrual flow and restoring libidinous potency and strength to men. In the 17th century, a mysterious ilness termed chlorosis manifested in menstruating girls and young women. A French physician named Piere Blaud recognized the iron deficiency in these chlorotic girls and prescribed oral iron salts as treatment for the disease. Although iron was not universaly given to treat chlorosis, the disease disappeared before World War II, presumably due to improved diet. Iron deficiency continues to be an important public health concern even today despite the fortification of food supplies and the increased use of iron supplements (Baynes and Bothwel, 1990). Iron deficiency can lead to impaired immune functions, 1 poor cognitve development, and other pathologies (Beard et al., 1996; Dalman, 1986). Paradoxicaly, too much iron is harmful due to its role in the generation of free radicals, which can lead to oxidative injuries such as lipid peroxidation and DNA damage (McCord, 1998). Among the diferent iron overload disorders, hereditary hemochromatosis is the most studied disorder. Hereditary hemochromatosis was first recorded by French physician Armand Trousseau in 1865 (Trousseau, 1865) and the term hemochromatosis, used to describe the disease of iron overload, was coined by von Recklinghausen in 1889 (Recklinghausen, 1889). A breakthrough came in 1935 when an English physician recognized that the inborn eror in iron metabolism is inherited through a single gene defect (Sheldon, 1935). The cloning of the gene responsible for the disorder, hfe, and the identification that mutations in the HFE protein responsible for this disorder, have lead to a breakthrough in the understanding of the pathology of the disease (Federetal., 1996). Advances in techniques and medical research have led to an increased understanding of the role of iron in health and disease. Recently, various genes involved in iron uptake and iron metabolism were identified. Exciting discoveries are continualy being made as novel proteins are identified and researchers are piecing them into the puzzle of iron metabolism. II. Roles of iron in cellular metabolism Iron is an essential element for celular metabolism. Except for some species of Lactobacilli, al organisms require iron for existence and proliferation (Archibald, 1983). However, if the amount of iron exceeds the requirement of the cel, it can be deposited 2 into vital organs and tissues. Excess iron depositon is dangerous due to its role as a catalyst in generating free radicals. Haber and Weiss were the first to recognize that the highly reactive hydroxyl radical (HO) could be generated from the interaction between superoxide (02*") and hydrogen peroxide (H2O2) (Equation 1, also known as Haber-Weiss reaction). They also realized that this reaction is thermodynamicaly unfavorable in a biological system unless a metal ion catalyst is present (Haber and Weiss, 1934). The Haber-Weiss reaction can be broken down into two chemical reactions with iron acting as the catalyst (Equations 1.1 and 1.2). The net reaction (Equation 1) is considered to be the major mechanism by which highly reactive HO is generated in a biological system (Liochev, 1999). 0 2«" + H202 -> 0 2 + HO" + HO« (1) Fe3+ + 0 2»" -> Fe2+ + 0 2 (1.1) Fe2+ + H202 -» Fe3+ +HO" + HO» (1.2) HO* is one of the reactive oxygen species that can potentialy cause serious damage in the cel. For example, guanine bases of DNA are particularly sensitve to oxidation and if this damage is not repaired, it can impair DNA function (Lindahl and Wood, 1999). The double bonds in polyunsaturated faty acids are also susceptible to free radical atack. The resultant lipid peroxyl radical can atack another faty acid thereby seting up a chain reaction. As a result, the normal functions of enzymes or proteins that reside in the plasma membrane are afected. Proteins too are susceptible to oxidative damage. Although most oxidized or damaged proteins are promptly removed by proteolysis (Grune et al., 1997), some can slowly accumulate with time and thereby contribute to plaque depositon as seen in various diseases such as arteriosclerosis and 3 neurodegenerative diseases (Stohs, 1995; Stohs and Bagchi, 1995). Therefore, although iron is an essential element in celular metabolism, the ying side of the metal has to be controled so that the yang side can be fuly exploited for celular functions. The role iron plays in celular metabolism stems from its capacity to accept and donate electrons readily by inter-converting between the ferric and the ferous forms (Beinert, 1976). This property makes iron a cofactor in molecules such as cytochromes, oxygen-binding proteins such as hemoglobin and myoglobin and other essential enzymes. Table I is a list of some of the iron-containing enzymes or proteins involved in mammalin celular functions such as energy generation in the mitochondrial electron transport chain, detoxification of free radical, celular proliferation and regulation of gene expression. The first class of iron-containing proteins, which are involved in electron transfer, usualy contains one or more iron-prosthetic groups such as heme, flavin, and iron-sulfur clusters. One function of proteins containing heme and iron sulfur clusters is electron transport in the respiratory chain of the mitochondria. They can provide orbital overlap pathways for the transfer of electronic charges (Marcus et al., 1980). Some of the mitochondrial iron proteins such as NADH-ubiquinone reductase, ubiquinol: cytochrome c reductase and cytochrome c oxidase also participate in the transfer of protons across the mitochondrial matrix (Rich, 1984). The electrochemical gradient generated by the transfer of protons can then be used for the generation of ATP (Reid et al., 1966). The second class of iron containing proteins interacts directly with oxygen. These proteins, which include dioxygenases, monooxygenases, peroxidases and catalases, are usualy involved in celular detoxification. The dioxygenases catalyze reactions that incorporate 4 Table I: Examples of some iron proteins involved diferent mammalin celular functions Electron transport and celular respiration: Cytochrome C Cytochrome oxidase Ubiquinol: cytochrome c reductase NADH: ubiquinone reductase Succinate: ubiquinone reductase Celular detoxification and oxygen-dependent reactions: Dioxygenases Monooxygenases Peroxidases Catalase Celular proliferations: Ribonucleotide reductase Post translation regulation of gene expression: Iron regulatory protein 1 Iron regulatory protein 2 Other iron containing proteins: Redox enzymes (e.g. Xanthine oxidase) Non-redox enzymes (e.g. Aconitase) Iron transport (e.g. Transferrin, lactoferrin and p97) Iron storage (e.g. Ferritin) Reversible oxygen binding protein (e.g. Hemoglobin and myoglobin) 5 both atoms of molecular oxygen into the substrate (Lange and Que, 1998) while the monooxygenases are required for processes where molecular oxygen is used in hydroxylation reactions, often to break an aromatic ring (Karuzina and Archakov, 1994). The peroxidases and catalases oxidize toxic compounds at the expense of reducing H2O2 to H2O. This reaction provides protection to the cel from the toxic efects of hydrogen peroxide. The third class of iron-containing enzymes includes ribonucleotide reductase, which is essential for synthesis of DNA during celular proliferation (Sahlin and Sjoberg, 2000; Stubbe, 2000). The fourth class of iron containing proteins, which are involved in regulating gene expression, includes the iron regulatory protein 1 and iron regulatory protein 2. They regulate expression of genes that contain the specific iron regulatory element in their messenger RNA (mRNA) (Eisenstein, 2000). Finaly, there is a large category of iron containing proteins, which do not have any functional role in redox reactions. For example, hemoglobin and myoglobin use the heme prosthetic group mainly to bind oxygen reversibly and deliver oxygen to specific sites, and iron transport proteins such as ferritin, transferrin, lactoferrin and p97. III. Diseases related to iron metabolism Due to the toxicity of free iron, the level of iron in an organism is tightly regulated through the gene expression of proteins involved in iron metabolism. Any aberation or disruption to the intricate balance contrived by these proteins would inevitably results in clinical manifestations that usualy cary severe consequences. Micro-molar levels of non-transferin bound iron are detected in organs of patients sufering from iron overload disorder (Batey et al., 1980; Grootveld et al., 1989; 6 Guteridge et al., 1985). For example, in some iron overload disorders such as hereditary hemochromatosis and African dietary iron overload, there is excess iron depositon in vital organs such as liver, pancreas and heart. The increased iron load in these tissues ultimately leads to organ failure if the disease is not treated. The transferrin/ transferrin receptor is thought to be the major pathway for celular iron uptake. The details regarding the iron uptake pathway mediated by transferrin/ transferrin receptor wil be discussed in details later in the chapter. However, it is unclear whether transferrin/ transferrin receptor is responsible for the increased depositon of non-transferin bound iron in the organs of hereditary hemochromatosis patients. Recently, the existence of transferrin/ transferrin receptor independent iron uptake pathways has been described (Conrad and Umbreit, 1993; Conrad et al., 1994; Craven et al, 1987; Egyed, 1988; Kennard et al., 1995; Morgan, 1988; Richardson and Baker, 1991a; Yu et al., 1998). There is increasing evidence to support that these pathways might be responsible for the iron burden in hereditary hemochromatosis patients (McNamara et al., 1999). Table II is a list of diseases related to iron metabolism with an emphasis on diseases relating to iron overload. Each of the diseases wil be discussed briefly. A. Hereditary hemochromatosi  Hereditary hemochromatosis is an autosomal recessive disorder with high prevalence in Causcasians. An estimated 10% of the population is heterozygous for the gene, with 1 in 400 having the disease (Edwards et al., 1988). It is characterized by excessive iron depositon in various vital organs that leads to severe multi-organ failure including diverse pathologies in the liver, pancreas and heart. Liver damage, including 7 Table II: Diseases related to mammalin iron metabolism Hereditary Hemochromatosi  Iron overload in patients carying specific mutations in the HFE gene Non-HFE hemochromatosi  Juvenile hemochromatosi  Iron overload with early onset by the age of 30 with an accelerated rate, the HFE gene is normal Neonatal hemochromatosi  Lethal iron overload resulted in liver failure in neonates African Iron overload Transferin receptor 2 related iron overload Iron overload related to genetic predispositon in 10% of the sub-Saharan African population and is exacerbated by dietary iron intake Non HFE iron overload linked to a Y250X amino acid substiution in the transferrin receptor 2 protein Aceruloplasminemia Neurodegenerative disease Iron accumulation due to aberration in the ceruloplamin protein, iron accumulates in the reticuloendothelial cels and hepatocytes. Celular iron uptake is normal but iron egress is impaired Alzheimer s disease Metals such as iron, copper, aluminum and zinc have been implicated in the disease progression 8 Parkinson s disease Oxidative stress caused by iron has been linked to neuronal death Halervorden Spatz Syndrome Caused by a defect in pantothenate kinase 2 gene which result in increased free radical stress in regions of the brain with high iron accumulation Neuroferinopathy Insertion of adenine in the carboxy terminal of the L chain of ferritin partialy compromises the structure and function of ferritin Hyperferitnemia Point mutation of in the IRE of the L chain ferritin mRNA which leads to increased serum ferritin level with bilateral nuclear cataracts Atransferinemia Disruption of transferrin gene, which ultimately lead to iron depositon in various tissues. Diseases related to mitochondria iron metabolism Friedreich's ataxia Iron depositon in the neuronal cels and cardiac myocytes due to disruption of iron homeostasis in mitochondria X-linked sideroblastic anemia Iron depositon in mitochondria of erythrocyte precursors and neuronal cels as a result of missense mutations in the eALAs gene at Xp 11.21, which afects the biosynthesis of heme 9 hepatocelular carcinoma, is the major cause of mortality. The gene that is responsible for causing hereditary hemochromatosis is a human leukocyte antigen (HLA) related protein caled HFE. Like other HLA class-I related proteins, HFE associates with (32-microglobulin on the cel surface (Feder et al., 1996). However, instead of its classical function in peptide binding in its al and a2 domains, crystal structure shows that HFE forms a high afinity complex with transferrin receptor (Lebron et al., 1998). There are several amino acid substiutions in the HFE protein that have been shown to associate with the pathology of hereditary hemochromatosis. The most common one is the C282Y amino acid substiution in the oc3 domain. Patients who are homozygous for this mutation usualy develop iron overload. A subset of heterozygotes is also afected with varying degree of severity (Edwards et al., 1998; Olynyk et al., 1999). In vivo experiments in the HFE knockout mice or mice carying the C282Y HFE transgene, show that these mice exhibit severe iron overload. The change of the cysteine residue at positon 282 prevents the formation of the disulfide bond in the oc3 domain of HFE. This prevents the protein from folding properly and associating with (32-microglobulin, which ultimately eliminates the expression of HFE on the cel surface (Feder et al., 1997; Waheed et al., 1997). There are several hypotheses on how HFE afects iron metabolism. HFE is shown to interact directly with transferrin receptor (Feder et al., 1997; Parkkila et al., 1997). Physical association of HFE with transferrin receptor is required for traficking of transferrin receptor to the cel surface (Gross et al., 1998; Salter-Cid et al., 2000a). However, it is thought that the interaction between HFE and transferrin receptor decreases the binding afinity of transferrin to transferrin receptor (from Kd of 5 nM in the presence of HFE to Kd of 12-75 nM in the presence of HFE) 10 (Feder et al., 1997). Nonetheless, since serum transferrin is present in the micro-molar range, the binding of transferrin to transferrin receptor is saturated even in the presence of HFE (Salter-Cid et al., 2000b). Others have also shown that direct interaction of HFE with transferrin receptor reduces the internalization rate of transferrin/transferrin receptor complex (Salter-Cid et al., 1999; Salter-Cid et al., 2000a). Therefore, when HFE fails to complex with transferrin receptor, such as in the case of the C282Y HFE, the increased rate of transferrin/transferrin receptor uptake can explain the excess iron accumulation in tissues. Although there is a wealth of evidence that the C282Y mutation indeed causes hereditary hemochromatosis, more experiments are needed to define how HFE regulates iron absorption. Other polymorphisms such as the S65C, I105T and G93R and H63D in the HFE protein have been shown to associate with varying degree of iron overload (Barton et al., 1999; Mura et al., 1999). However, the molecular details of how these diferent HFEs result in iron overload are stil  under investigations. Recently, a new homologue of transferrin receptor, transferrin receptor 2, was identified. The human transferrin receptor 2 shares 66% similarity and 46% identiy in protein sequence on the extracelular domain with the human transferrin receptor 1 (In the context of transferrin receptor 2, the first identified human transferrin receptor wil be notified as human transferrin receptor 1) (Kawabata et al., 1999). Transferin receptor 2 can bind to transferrin at a lower afinity and internalize it (West et al., 2000). The transferrin receptor 2 also showed a preferential expression in the liver where the expression of transferrin receptor 1 is lower (Fleming et al., 2000). Interestingly, the transferrin receptor 2 gene has been implicated in another form of hemochromatosis, which is not related to the classical hereditary hemochromatosis. A homozygous 11 nonsense mutation, Y250X, which resulted in a pre-matured stop codon in the transferrin receptor 2 gene was shown in several familes afected with non-classical hereditary hemochromatosi  (Barton et al., 2001; Camaschela et al., 2000). Nevertheless, the exact mechanism causing the iron overload has not been characterized. B. Juvenile hemochromatosi  Juvenile hemochromatosi  is an autosomal recessive iron overload disorder that is clinicaly and geneticaly distinct from hereditary hemochromatosis. Juvenile hemochromatosis has a much earlier onset before the third decade of the life of patients afected. The disease afects both male and female equaly with the same tissue distribution as classical hereditary hemochromatosis, but iron depositon in juvenile hemochromatosis occurs at greatly accelerated rate (Camaschela et al., 1997). These patients also present a more severe case of endocrine dysfunction, joint disease and cardiac abnormalites. The locus for juvenile hemochromatosis has been located at the long arm of chromosome lq (Camaschela et al., 1999; Roeto et al., 1999). C. Neonatal hemochromatosi  Neonatal hemochromatosi  is characterized by abnormal hepatic iron overload in newborn infants. In neonates, if left untreated, the disease often results in an acute, and frequently lethal liver failure. Neonatal hemochromatosis is not a homogenous disorder but rather a colection of disorders because it is often associated with other congenital anomalies (Johal et al., 1998; Vohra et al., 2000). The gene responsible for this disorder has yet to be identified. 12 D. African dietary iron overload African dietary iron overload is associated with increased iron intake from a fermented maize beverage. The disorder occurs mainly in sub-Saharan African population with a genetic predispositon involving a non-HLA-linked gene, discovered through pedigree analysis of afected African familes (Kasvosve et al., 2000; McNamara et al., 1999). The patern of iron overload is diferent from the classical hereditary hemochromatosis. Hepatic Kupfer cels and hepatocytes contain substantial amount of iron. As a result, liver disease is the predominant organ manifestation; cardiomyopathy and diabetes are seen less often (Gangaidzo et al., 1999). E. Neurodegenerative diseases a. Alzheimer s disease Alzheimer s disease is a neurodegenerative disease characterized by the depositon of [3-amyloid plaques in the brain. P-Amyloid deposits in Alzheimer s disease have been shown to contain advanced glycosylation end products, copper and iron ions (Atwood et al., 2000; Lovel et al, 1998; Sasaki et al., 2001; Yan et al., 1996). Recent data have suggested that metals such as copper, zinc and iron can precipitate the formation of p-amyloid plaques (Bush, 2000; Bush et al, 1994; Loske et al., 2000). There are also evidences to support that the over expression of heme degradation enzyme, heme-oxygenase-1 (HO-1) contributes to the pathological iron depositon in Alzheimer s disease (Schipper, 2000; Schipper et al., 2000). In addition, the serum concentrations of one of the transferrin family members, p97, is elevated in Alzheimer s disease patients 13 compared to the age matched control group (Kennard et al., 1996; Kim et al., 2001). The corelation between iron metabolism and Alzheimer s disease is further supported by the fact that the C2 alele of the transferrin gene is more predominant in Alzheimer s disease patients than the control group. The presence of the C2 alele of the transferrin gene has been implicated in the early onset of Alzheimer s disease (van Rensburg et al., 2000). b. Parkinson s disease Another neurodegenerative disorder associated with iron overload is Parkinson s disease. The iron content ofthe brain of Parkinson s disease patients is elevated (Sofic et al., 1988). There is also evidence for increased iron uptake in the dopaminergic nerves of Parkinson s disease patients (Faucheux et al., 1995). The ultimate cause of neuronal death is unclear; nevertheless, a growing body of evidence suggests that increased oxidative stress caused by iron and other factors is one of the main culprits. c. Halervorden Spatz syndrome Halervorden-Spatz syndrome is a rare autosomal recessive disorder clinicaly characterized by Parkinsonism, cognitve impairment and progressive dementia. In additon to the dementia, this disorder is also characterized by iron depositon in the globus palidus, caudate, and substantia nigra areas of the brain (Jankovic et al., 1985; Neumann et al., 2000). However, iron overload is not observed in tissues outside the central nervous system. The symptoms typicaly surface in the first or second decade of life and the patient usualy succumbs to the disease by age 30 (Angelini et al., 1992). Recently, the gene responsible for this disorder has been identified. Missense mutations 14 in the pantothenate kinase 2 result in the metabolic eror in pantothenate metabolism (Zhou et al., 2001). The authors proposed that the defects in the retina and basal ganglia specific pantothenate kinase 2 enzyme gave rise to a secondary excess of metabolites such as cysteine and N-pantothenoyl-cysteine. Cysteine undergoes rapid autooxidation in the presence of iron, which results in free radical production, further stressing the cel with high iron depositon in the afected brain regions (Yoon et al., 2000). Patients with Halervorden-Spatz Syndrome also sufered from retinopathy as a result of the mutations in pantothenate kinase 2 gene. It may be possible to therapeuticaly deliver phosphophopantothenate or an alternative intermediary compound to cels in the afected tissues to bypass the enzymatic defect in the pantothenate metabolism in order to aleviate the disease symptoms. d. Neuroferinopathy Neuroferinopathy was recently identified as a dominantly inherited late onset basal ganglia disease with extrapyramidal features similar to those of Huntington s disease or parkinsonism (Curtis et al., 2001). By linkage analysis, the gene responsible for the disease was identified as ferritin light chain. Brain histochemistry of the neuroferinopathy patients showed abnormal aggregates of ferritin and iron in the basal ganglia region. An adenine insertion in the carboxy terminal of the ferritin light chain may cause a conformational change and partialy compromise the structure integrity of ferritin. As a result, instead of complimenting the ferritin heavy chain in the role of iron storage, irons are released from ferritin and cause toxic free radical reactions in the diseased brain. 15 F. Atransferinemia Atransferinemia is a rare genetic disorder in which plasma transferrin is absent or is present at very low concentrations (Huggenvik et al., 1989). The patients need to be given transferrin infusions or blood transfusions for survival. Even though these patients lack transferrin in the plasma, dietary iron can be absorbed as non-transferin bound iron (Hamil et al., 1991). Most of the non-transferin bound iron is deposited in vital organs such as the liver, pancreas, heart, thyroid and kidney through transferrin independent pathways (Beutler et al., 2000; Hamil et al., 1991; Huggenvik et al., 1989). However, none of this non-transferin bound iron can be used for erythropoiesis and therefore patients sufered from severe microcytic hypochromic anemia (de Silva et al., 1996) G. Aceruloplasminemia Aceruloplasminemia is a rare genetic disorder caused by a lack of ceruloplasmin feroxidase activity due to mutations in the ceruloplasmin gene (Gitlin, 1998; Haris et al., 1998). The feroxidase activity of ceruloplasmin is required for the release of iron from the cels. In the absence of functional ceruloplasmin, there is marked iron accumulation in the liver, pancreas and brain with lesser overload seen in the spleen, heart, kidney and thyroid. Patients with aceruloplasminemia usualy develop diabetes melitus and progressive neurodegeneration of the retina and basal ganglion suggesting that ceruloplasmin is essential for the maintenance of iron homeostasis in the central nervous system. (Harris et al., 1995). A distinction between iron overload in aceruloplasminemia and iron overload in hereditary hemochromatosis and African 16 dietary iron overload is the involvement of the central nervous system in patients with aceruloplasminemia (Klomp et al., 1996). H. Hyperferitnemia Hyperferitnemia is an autosomal dominant disorder characterized by moderately increased serum L-ferritin concentration. This disorder is a result of multiple point mutations in the highly conserved hairpin loop iron response element region of the L-ferritin gene, which afect the interaction of iron regulatory protein with the iron response element(Gireli et al., 1997; Mikulits et al., 1999). Hence, the synthesis of L-ferritin is constiutive and poorly regulated. A surprising phenotype associated with this disorder is bilateral nuclear cataracts. Nevertheless, a direct relationship of the depositon of ferritin in the lens and mutation of the L ferritin iron response element has not been demonstrated (Cazzola et al., 1997). I. Disorders related to mitochondrial iron homeostasis The mitochondrion requires tightly regulated iron transport since it is the site of heme and iron-sulfur cluster synthesis, (Lange et al., 1999; Lill et al., 1999; Tong and Rouault, 2000). Recent discoveries in several diseases with mitochondrial pathology have alowed researchers to identify genes involved in mitochondrial iron homeostasis, a. Friedreich's ataxia One of the genes involved in regulation of iron eflux in the mitochondria is frataxin (Campuzano et al., 1997; Campuzano et al., 1996; Radisky et al., 1999). In patients with an autosomal recessive neurodegenerative disorder caled Friedreich's 17 ataxia, there is an expansion of the GAA trinucleotides in the first intron of the frataxin gene, which results in a non functional protein (Gacy et al, 1998). Data from the yeast frataxin orthologue knockout models as wel as histological and biochemical data from heart biopsies or autopsies of Friedreich's ataxia patients have shown that the frataxin defect causes a specific iron-sulfur protein deficiency and intra-mitochondrial iron accumulation (Babcock et al., 1997; Rotig et al., 1997). The increased mitochondrial iron accumulation results in increased oxidative stress and respiratory deficiency in mitochondria, which ultimately leads to irreversible mitochondrial damage (Babcock et al., 1997; Delatycki et al., 2000; Koutnikova et al., 1997; Wilson and Roof, 1997). However, a second hypothesis suggests that the time-dependent intra-mitochondrial iron accumulation in a frataxin-deficient mammal occurs only after the onset of the pathology (Puccio et al., 2001). This hypothesis implies that mitochondrial iron accumulation is the result of Friedreich s ataxia instead of the cause. Nevertheless, the use of iron chelating reagent, desferoxamine, is efective in both decreasing the iron depositon and inhibiting the progression of neurological symptoms (Palau, 2001; Wong et al., 1999). b. X-linked sideroblastic anemia Another gene associated with human mitochondrial iron homeostasis is the abc7 gene, which encodes an ATP-binding cassete transporter (Kispal et al., 1997; Leighton and Schatz, 1995; Senbongi et al., 1999). A mutation in the human ABC7 gene is responsible for the X-linked sideroblastic anemia ataxia, a non-progressive cerebela ataxia, which manifested in infancy or early childhood. (Alikmets et al., 1999). This gene has an orthologue in yeast caled atml (Csere et al., 1998). Previous studies have 18 shown that ATM IP transports the precursor of the mitochondrial iron-sulfur cluster into the cytosol (Kispal et al., 1997). Mutations in the yeast atml gene result in iron accumulation in the mitochondria leading to oxidative damage (Senbongi et al., 1999). IV. p97 and the transferrin family The transferrin family is made up of serum transferrin, ovotransferin, lactoferrin and p97 (also known as melanotransferin). All members of the family consist of a single polypeptide chain of 80 kDa molecular weight and the crystal structures of several members of the transferrin family have been determined (Bailey et al., 1988; Baker and Lindley, 1992; Kurokawa et al., 1995; Smith et al., 1992). The polypeptide forms two very similar but distinct lobes, each capable of binding one atom of ferric iron together with a synergistic anion such as carbonate or bicarbonate anion. Amino acid sequence alignment of members of the transferrin family shows that they have extensive homology, especialy in the areas where amino acids are involved in coordinating the iron atom (with the exception of p97) (Fig. 1). A detailed alignment of al the known p97 orthologues with members of the transferrin family is shown in Figure 1A. A. Brief overview of serum transferrin, lactoferrin and ovotransferin Human serum transferrin was first identified by Schade and Caroline from the non-hemoglobin fraction of blood plasma in 1946 (Schade and Caroline, 1946). The primary structure of the protein revealed two highly homologous halves with 40% protein sequence identity, which are bridged by a connecting peptide (Fig. 2). Crystalographic studies of rabbit serum transferrin have revealed that the ferric ion is coordinated by the 19 Figure 1: The protein sequence alignment of members of the transferrin family including p97 from diferent species, human serum transferrin (Tf), human lactoferrin (LTf) and chicken ovotransferin (OTf). (A) The complete protein sequence alignment of members ofthe transferrin family. The red colored amino acids are conserved residues whereas the blue colored amino acids are semi-conserved. © denotes the amino acid residues involve in iron coordination and denotes the amino acids involve in synergistic anions (such as carbonate and bicarbonate) coordination. • • • • • are the five amino acids involve in putative zinc binding as proposed by Garet  et a/.(Garrat and Jhoti, 1992) and 4 4 4 denotes the integrin binding RGD motif. (B) Sketch diagram of the amino acids involved iron binding of the members of the transferrin family and the proposed zinc-binding site in diferent orthologues of p97. 20 w i o o 1 0 3 E H fa ft ft i a E - " i 2 J ft 1 fa fa Q < ' < < O i \ I ! \ ! ! 1—1 1 Pi i 1 1 1 1 1 1 1 W i 1 ! ! ! ! \ ! 1—1 1 rt ft w i w 2 i Pi 1 > u c o I 0 0 CO i a Pi E H fa 2 i 2 2 i 5 ft ft W S 1 Q i a ft S ft CD i 2 i fa p O E H i u i u U u * 3 i ft i ft * ft! P C i E H ft >H >H o i a I < w o > - i w i fa j p fa CD ft a i ft ft < c t ; P i Q i a C O a c o l CO l CO CO CO w B I H 2 1 > 1—1 W > i H E H 1 < E H > i u i u • u u i s 5 £ > ft i a ft i a > Pi > i > > I > 1—1 2 q w i w ft I CO > oi 2 s i 2 fa 1 Pi CO E H CD CD i £ > 1 Pi ft E H Q o i u CO i ft 1*3 P ft 1 £ > 1 Pi ft S ft > 1 * > i o 4 . W ft B I H 1 i < < Q O ft 1 Pi i J fa >H > J I J > i u u ft < < I CO ft 1 J > ft > ft I ft i o a u I J J 1 ft < < 2 1 ft 1 4 M z c o 1 5 ft i CD CD < H ft I fa > 1 ft J J CO < 1 E H p 1 fa CO E H ft o I > fa 1 ft ft fa fa CO I CO CO 1 J > U CO ft 1 ft CO 1 > EH J < O 1 P I ft 1 ft U < fa t « i 0 3 2 1 fa J CD J S i 2 1 > i—i > o i 1 ^ p > i i i i i 2 ! i i i i 2 2 > H I i i i 1 1 i fa J i i i 1 1 i > H i i i i 1 1 i 1 i i i 1 1 i c o S I i i i 1 1 i c o O | \ | j 1 1 | r H H r H T—1 r H r H r H r H r H O o Q O C O > > i—i 1—1 fa ft fa fa fa fa w Q fa 2 fa E H J X a a o o c o E H ft C O Q Q Q Q o O w w ft H - : - a > > > > fa > > 2 w w fa H W W w ft H C O WG WG WG DAA VAA VAA WI LIP LVA ft ft ft ft ft ft ft C O H ft ft ft fa i- 1 > CD CD O CD [g 2 ft C O S ft >H ft >H 1 < w fa W ft ft ft 1 3 2 2 W < <C < ! i i i i i i J J h i i i C D CD CD V C D 0 3 < < < < < < fa fa a p fa H < !* fa fa > i— 1—1 M > H < > P < <; < s P fa a P P ft fa C D C D C D < c C D < C O D P P p p P P E H J i - J ft ft ft ft EH E H [ - : c o E - > I—i 1—1 H > > > i—i E H C D : E H < H fa a Q Q P p p ft ft fa fa P fa ft w EH a i O >H p P fa i fa P < c w ft ft i 2 < <C P ft i—i i H - H M 1—I 1—1 i h i 2 1 ft ft a i a a 1 o 2 ft ft > i > E H 1 H H 1—I > i 1 E u U U i 1 o p fa P Q a 1 1—1 p P P i 1 ft >H < ft : i 1 w P E H i fa 1 p > C D i C D C D 1 ft 2 C D ft i a w 1 ft o i—i > i > s 1 !—1 > fa > i u 1 i fa 1 c o E H E H w i 1 > : i—i C O i C O ft 1 ft C O P 3 ft i ft ft I ft i—i J C O o i ft p 1 O ft fa ft 1—1 i 1—1 H 1 ft W > < C D i C D o 1 > o fa fa < 1 < 1 ft o P P r H r H 00 r H r o 1 * Ln « H E H ft <+H O H C N o n ft M-H H i 1 ft ft C E H P r > i p -4-1 a ) 0) e n a - H (U a c ft; ft ft X I CQ u n j * u e - H e B • H >i >> (0 0 ft! ^ ft r H r H p 0 3 2 O P ft u ft fa r - 4 - i r > c n H t - ft W O r H C N c n ft a \ 4 H E H 1 I ft ft a EH P P - r -ft a ) a ) O ) a - H CU ft: a a X ft ft P CO u ffl u B P • H E E • H > 1 > i «fl 0 ft ft r H r H ft ft 2 u ft ft a fa fa 1 5 ? O j J > S3 O S CQ CO CO O J O J - : OS OS O S o s Q ".: u CO CO CO CO >H w Q os p > > > 1 — 1 - p 1 >H 1 O > > > > ! — 1 - OJ > > > > H a 2 a a 2 2 2 s CD CD > > > EH CQ a a a a CD i—i i—i i—i i—i CD r CD CD V EH EH EH EH EH 00 s P U i u 1 U i fj I CO i CO 1 i CO . . . . OS > > > > V -os OS OS " -E H CO O P 3 - -i—i i—i 1 — 1 1 — i EH O > > > 1 — 1 Cu O S a a a o :: : co CO CD Q PH OS OS OS OS W < <: > > > > > < < < < > >H x >H >H >H -0] CO CO CO EH EH EH EH I I I I o i i i i O J L D iri LTI •>* O O ro O o H r H l > r > Cn t> c-on a cn U H Cn os H cu fl - rH (D c ( 1 3 JQ cn CJ rrj e - H B rrj 0 O S o s O S 2 u O S i i i Oi 1 H u Q 1 " Q U O J H > EH D 2 > OS EH OS w CO 2 i—i s > a < o J O J o s o s O J OS o s w OS o O J EH EH " OS EH 1 — 1 1—I OJ OJ > > 1 — 1 01 a a H H s 2 CD CD o OS <c a EH CD 1 os co P OS CD o CD a > >H C 2 EH EH O J O J P OS O J U i u 1 .VC CO CO p EH •S OS « < P O OS OS OS p a p OS [> p > a p o. O j O j OS CO W J H o EH Efl CD CD o - - - OS OS OS a : 1 — 1 < < > > > < < < CO > EH a H OS OS CO a >H OS OS O J O J O J CO CD OS OO t > t > r H O O r H r H r H H r H M H EH M H O r H CN EH 1 I OS f-0) CTl a A S O J On cd V B - H >> >i o> J O r H r H OS u o. O J O J O j O j CO EH : a a OS w u p p p • u > > >^ 1 o o a 1 OS OS X o 1 1 CO CO CO CO 1 CO CO CO 1 p P p p 1 OS OS OS O J rj J OS OS OS o o u u u C J C J p p p CO Oj •fl ] a w w m EH OH CO CO CO O j to >H >H OS >H P CO OS CO CO 1 EH EH < E H m OS OS CO C D C D o O a P EH EH C D CD U O . O j O j O j > > > > u C J C J O : D c V C D C D O < C D C D C D C D CO CO - SU I - - PH P OS Q p CO CO C D CO > hO p P O J w > > > P J O j u u u U OS i ! ! | OS I I C O oo C O l > 1 CM L D C O i_n LT) (Si r H r H r H C^  r> Ch r- cn r - O, o\ D J OH 4 H O J P J C EH 4 J 0) OS • rH 0) A S ES rrj Xi EQ U 0 3 B X ! OS - H £ (Tj 0 OS OS s U OS CO OS co a CO < oo p O j C J C J C J ft OS OS CO • E H C D a OS OS C D a O j O j i OS P I — 1 O j 1 OS 1 I I I ! EH 1 1 i : w < EH >H J s OS : EH w C J U : : p a OS OS C X P O J Cu OS CO >H © h H a OS i 1 p C D 1 C D i 1 H p i 1 1 1 — 1 i 1 I OS i 1 1 P i 1 O J a i 1 a OS i 1 O J i 1 2 P i 1 p O j OS 1 CO > p H p 2 CO O J C D C D >H C D O J > <! > > r U 2 u C J • u 1 CO • K < ri! O J CO zn C D _ Cu Cu OS OS CO OS > > H EH oo < C N < OS OS > i—i 1—1 > O OS OJ CO a W C D i w O j CO i EH I OS a > i 1 i EH i 1 s i 1 i OS CN m vo C O <£> V D r- V O r H H i—i r H MH EH M H O r H CN EH 1 1 P a cu cn cn C OS O J O J 03 CJ E • H > I >> Oi r H r H OS U Cu Cu C O C O CD CD P P ft ft U C J ft ft ft ft ft ft ft Q O ft C O C O ft S ft J < a O ft CD CD c o c o ft ft ft ft 0 CD P P <j 4 ft E H C O C O CD CD P Z ft ft C J U w ^ o ft ft ft E H C O z O O ft ft P P fa ft c o c o 2 ft ft ft c o Q t ft O CD < E H Z -ft U CJ u ft ft p p ft ft ft ft 0 w p p 1 I ft ft ft c o ft ft ft fa ft ft > > E H E H C O C O > > ft ft <; < > > a p CD CD < < CD CD w w <c < ft ft u u ft ft ft ft fll < CD CD O >H >H P P > H > H ft ft ft ft ft ft CD CD P P E H E H ft ft ft ft ft ft > > E H E H C O C O > > ft ft > > p m CD CD < < CD CD ft ft < < ft ft u u ft ft ft ft < < CD CD C O C O P P >H >H > H > H ft o ft w ft a fa fa z H > > E H H E H C O C O E H S w ft > M > fa fa fa < ! <C <C > > > P P P CD CD CD < < W CD CD CD ft J J u u u £d ft^ ft fa fa fa < <c < CD CD CD c o C O c o > H >1 CD C O CD C O ^ H a ft ift z w < ft a z < a ft p ft ft CD CD P P ft ft C J U ft ft ft >H fa C O fa fa E H P P C O ft ft < z co P w < < W ft w a z P P 4 c o a ft p <; p z i i 1 a z < 2 ft fa H W CD C O fa a CD j fa 2 ft fa c o Z ft > > E H P C O P ft ft ft ft ft <c is > p P z CD O fed CD i CD fa C O ft fa fa U U ft >H fa ft < <c a CD w ft >H ft CD CD fa I S ft ft* P P < C E H O r O O O r O O O C N C N f * CPi E H a \ ft 4-1 o H C N ft 4-1 E H il 1 ft ft E H ft r - t > 4J a) cu CJI E l • H CD ft C l ft: ft ft ft! CO CJ m (fl u E ft! • H E E - H > i i d 0 ft ft i -H r H ft ft 2 u ft ft u fa fa r> 4-1 r - CTl E H IT- ft 4-1 o H CN c n ft E R 4-1 H 1 1! ft ft H ft r -ft CD c n d • H CD ft a a ft ft ft m ft! CJ (C (fl CJ E ft! • H E E - H > i > i aS 0 ft ft r H r H ft ft 2 u ft ft u fa fa OJ 0- ft PH ft -. o C O OJ O i OH OJ u 2 2 2 2 J 01 p i—i J p i-l P OS ".' ," . " • C X C X 3 OS OS J OS o s OS E H p P ii, fa PH 3 a a a E H C O C O C O C X p q OS Q OS 0 OH OH OH : C O C O C O < OJ OH C O 0S C X 1 l l I < o 0 Q CD C O C O OS 1 i i i OS fa fa w > OH OS EH P C O 1 i i i OJ X K Q Q " H : 1 i i i u OS OS OH OS LTI LD LO VO O 00 O O CM C O EH a EH CD P C X C X a o s O S 0S < fa C O C O p p PH H co OS C X p c x a c x EH OS >n P P a a p OJ c x C D • 3 a c x p a !—I a a 3 C O VD O I p q p q OJ C O O l E H P P : P OS 2 OS P >H OJ OS ... < p q >H EH i i H OJ > p q co EH OJ P P P P C J u p p p p CD q os os 2 2 | <rj OS C X P P >H >H fa pq OH O CD CD P P rfS <, p q W >H >H H a C X C X i i i i EH EH <C < H H ft ft > > P P p q p q EH P CD U p q C O OS p p < a cx ft 2 H CD <C OS EH p > >H >H pq pq P >H CD CD P P 2 >H < 2 C X S a P C X 2 p q C X co P i i i i OH OS PH PH 3 2 a co a c x M H < <c EH C O P O pq C O fa CD CD P P p a C J C J W p q PS ft H fa CD co >H >H M f=C fa C O a c x H fa < < ft EH >H P CD C X C O C O p p I ft C O ft 2 II p EH > C O C O fa fa p p p p < OS > P P fa OS CD H OS EH m C O C O p p ft 2 fa fa 2 P p p a P c x H I ft ft C O ft ft > o s C O fa CD H < < C O C O 2 < c x p p I p p i—1 < I—1 > a p p C O ft p > fa fa p p p p c x -p p a p 1 1 i i i 1 1 p 1 i 1 1 i i i 1 I > 1 i 1 1 i i i 1 OJ 1 i 1 a C X C X <c 1 1 OS i 1 CD CD co CD 1 C X OS a i 1 >H >H >H CD CD -1 <C k < OS C O CD p EH 1 p q p q 2 EH OJ OJ P OS OS 1 C O C O C O OJ co P 1 C O C O C O P C O CD CD fa P CTl «sj *# 00 LD o LD o C O r H r H «* r H r H 00 r H O o CN OO ro CM 00 00 CM 00 00 00 r - 4 H r> o n EH r > c r , r - O J MH O H CM 00 ft c n MH EH 1 1 ft OH OS EH P ES IT- i r -0) CU c n o n rs • H a j P ES OS X P P rrj X ! co U rrj rrj u £ P i 3 • rH E e • rH > i >* 01 rrj 0 P OS r H r H OS ft 2 CJ OS X u fa fa O MH o n EH t - o n r > O J MH O CM o n O J o n MH EH 1 OH OH ES EH P ES c ~ r -<D CD o n o n ES - H (U AS ES ES AS P P rrj X ! CO U n j rrj U g P - H E E • H >. >H n j 0 P ES OS P r H r H X OS 2 u OS OS u fa fa m ft ft > CD CD O CD ft ' ' rf a C O ft ft CD P ft P : - ft ft ..- ft > EH ft EH < C O C O rf > ft ~ ; : C O rf < rf rf 1—i ft • - • P -> 1—i E H 1—1 rf rf fit CD ft ft T: - B - ft -fa ft ~ r -: ft rf rf ft ft ft ft rf ft a a O a 2 J 2 w ft i—i 1—! 1—i !—1 !—1 > i—i H H ft a O) 2 ft ft ft P W ft w ft rf i—i a P s s S i—i i—i H £ P u u ; K ft ft ft p P p p a a a w ft w ft ft i—i C x a , ft ft ft EH EH EH ft ft C O C O C O EH EH EH ft C O P ft ft ft ft ft C O P ft 2 H re CO CO CO CO CO j > ft > > > 1—1 > C O H ft u u U c u C C J a a O a w EH ft 1—1 o i—i 1—1 i— i i— i M > > ft 1—1 ft ft w w C O P - - p ft ft ft ft 1 1 I ft ft ft ft ft 1 1 1 ft o - . ft . 1 1 1 ft > ft ft ft 1 1 I ft CD a a a 2 1 1 I ft >H ft ft ft C O o o CD rf > ft C O C O ft > ft p ft a ft ft ft ft C O C O C O ft C O j - ft > EH CD EH ft rf a C O C O C O ft a 3 : I S ft Q Q a w ft a ft ft s CD CD CD r j P p p H C O u ' U ft ft - - ft ft a a a ft ft C O P I—1 i — i i—i 1—1 i—i ft ft ft J ft • v ; w w ft a a . ft ft ft ft a Q P H EH EH rf ft ft w ft ft l— l CO . c CQ o a co co ft J ft > > E-« EH - rf rf rf H u U u u u ' • • s ft ft ft ft ft ft p> a H > ft ft J ft > j > i— i i—i >H >H >< C O ft ft ft rf C O ft ft ft m ft rf p ft ft n c n r o C N C O C O c n C O r o C N co r o C N m m r o ^ f ^ ^ r o ^ o o r - c o o o o j i n c j c i ' t H H o n • ^ r o ^ ^ r o - ^ ^ i n ^ [> 4-1 I> 4-1 o cn EH t> cn EH e'- cn r- ft 4H o rH CN e'- cn r> ft 4H O rH CN er* ft cn 4-1 EH 1 1 en ft cn 4-4 EH 1 1 ft ft a EH ft c r- r- ft ft EH J fl e'- r-- U CD <u cn cn j-> CD CD en on fl •H CD ft fl a ft! ft ft G -H CD ft fl fl ft ft ft rd P u rd cd CJ cd JQ CQ 0 cd cd CJ £ P -H E e -H >i >. E P 3 -H E E -H >, >i cd 0 ft ft rH rH cd 0 ft G ft! rH rH ft ft 2 U ft ft U ft fa ft ft 2 CJ ft ft CJ fa fa ft > CD E H - > 1 1 1 i a 1 1 1 ft i 2 U 2 o q ft a a CD - o 2 o ::. C O C O a a w ft p ft 1—1 a < C O C O C O < p i > C O CD CD CD a - P o -> > > > < < <; > ft u u u - u u o - o < < < ft a ft p a a C O oo C O C O EH ft : - o ft : 0 ft ft ft ft ft >H >H >H -a a a ft u D - • C O CD C O C O p P p ft C O > > > > ft ft ft ft ft ft a p E H E H E H o U u U ft C O ft ft ft C O EH > O > > > H a CD CD C O > Q Q p a i—i EH EH P ft i I I i Q p p i I I CD i 2 ft 2 ft 1 1 >H i - - i I 1 ft i ft ft ft a i 1 1 ft i 1—1 H i—i H i I 1 1—1 i ft ft ft i 1 1 i LI i I I ft ft ft ft a ft a E H cn X a ft a a a C O < i—i 1—1 i—i > ft K 1—1 p i—i ft ;> <! C O 2 ft ft p ft H CD < EH > > 1—i H 2 2 2 2 ft p p , ft ft - - o > > > 1—1 i—i M i—i >H > Q P w a a a > EH C O - s u s o : P CD CD c CD o CD CD CD ft • * • ft ft ft E H E H E H E H < < C O C O C O ft ft ft ft EH CD •"" P a O l ft ft P > > > H ft CD O CD < CD C O PH PH : i PH X - ft ft ft ft CO P u u u U U CJ U c • 03 c CO -. ' . EH ft ft ft ft OH ft ft ft - - - - . ro ro ro CM VD VD o o CTl t - L D PO <* Ln ft ; ; ft P ft ft ft ft ft a ft ft a • a a a P o p C D C D : r :- >^ !; . . . ft 2 - C O > J p p ft ft a a ft ft ft ft ft ft ft PH p H ft !—1 '— EH EH - EH - H > EH EH EH C O a > 0 3 EH p ft PH Q C O ft ft p p a PP' C O - > > > > > M 2 p ft ft ft ft ft ft ft ft ft. ri! ri! ri! ri! ri! ri! ri! ri! > > > > > > > > P P P a p P P ft Q C D C D C D CD • C D ft ft C D o a C D ft P i a i C D 1 ft o;i 0 . ft ii > > > < > > > C O J P P p ft P 1 U u C J C J u u C J u ft ft ft ft ft ft - ft 0 ft ft ft ft ft ft p ft p < ri. < ri! ri! < 0 . : ft 0 3 0 3 a EH EH EH ft P H - • > ft p CD C D C D C O C D C D ->' X ft 1 ft >H >H 1 ft ft ft ft C D ft ft ft 1 ft w - - ft M 1 a a a ft a ft C O 1 a G ri! 1 a a C O C O a a 0 3 P 1 C D C D C D ri! & ft < ft 1 > EH > C O w > > ft 1 \ U ft U 1 ft ft ft ft p P 1 - . ft ft 1 ft ft ft a p w ft 1 • ft C O 1 a a ft a C O a i—i EH 1 i I i i 1 t C D 1 1 w ft ft a 1 w C D 1 1 c - 1 C O 1 o • • LI CD c > > > i—i 2 i—i ft ft c ' ft ' p ft < < < o ft < a P u ft p ft ft j 2 ft co 0 3 < EH 0 3 a ft C O 1 C O .. C O p 1 p ft PH ft 1 II i 1 CO CO CO r> r H r H in o CM If) CM CM CM r H r H CM LD " t f LO LD if) LD VD LD r- 4H 0 0 E B CTl r> PH 4 H O r H CM on PH 0 0 M-l 1 1 PH ft EH P C IS r> 4-) (1) CD 0 0 GO 01 • H 0) P 01 a P ft ft rcj P HQ U rrj rrj 0 B P • H B g • H rrj 0 P 3 P r H r H ft ft 2 u ft ft o ft ft r> 4H 0 0 EH 0 0 r> PH 4-1 U H O r H OJ 0 0 PH CO EH 1 1 PH PH 01 P 01 o r* OJ OJ 0 0 0 0 a -ri CD P 0! 0! P ft ft rrj P CD U rrj rrj (J e P - H B e - H >1 Oi rrj 0 P PS P r H r H ft ft 2 CJ ft K u ft ft ft • H > ft o CD o O Q - ft ft ft > > > > ft i—t ft Eh E-i H < U ft fa fa fa fa < fa H I—i H H fa > ft oo 3 ft H >H >H H < w f - r - r - i x i c o c n r a r H c r i o o H c o c o m h h > i n C N C N C N r H O r O r - r H U ) ^ ^ • ^ • ^ L n r o c N r O r H on r> ft o H CN cn ft cn 4-1 4-1 1 1 a ft a Eh ft r> ft (U CD cn cn fl •H 0) ft d a ft ft ft rd ft: 03 u rd rd u £ ft! •H e e -H >i >i rd 0 ft ft rH rH ft ft 2 u ft ft u fa fa l> 4H ir- on Eh r> on r> ft O rH CN cn ft cn 4H 4-1 1 1 ft ft fl H J r> t> J-) 0) CD cn cn a • r f CD ft; a fl ft ft rd ft! 01 u rd CJ e ft! -H £ e -H >i >i rd 0 ft cd ft rH rH ft ft 2 U ft ft u fa fa 1 i i O l ft i ft ft 1 rf i > CD 1 ft ft ft a 1 ft C O ft ft 1 ft C O ft a 1 ft ft ft 0 1 ft ft CD ft 1 ft1 H 1 rf CD rf H 1 ft > ft ft 1 rf rf E H ft 1 ft ft > 1 ft ft ft ft I " : E H 1 ft ft ft 1—I 1 ft Oi rf > 1 - ; 1 ft ft ft 0 0 ft ft ft rf rf > a ft O O ft a» ft ft ft H E H rf rf > J U ft o ft rf f£ > rf ft ft rf CD CD ft rf rf rf rf ft CD o CD 0 C O u u u u E H a ft o o> C O o a a ft C J C O C O C O ft ft co o ft a « s s s 2 ft o a 0 0 ft w w w w 0 ft ft ft ft > oo < > w ft CTi o n o n C O v o c n C N o n a> o C O C O V O v o £> [> o n r> o n t> ft c n ft o n M-l ft ft ti H ft (1) ti - H 0 ft ti rd £ CQ u rd £ ft! - H £ r d 0 ft Cd ft ft 2 U ft 0 ft ft a 0 ft ft 1 i ft ft 1 i U 1 i rf > 1 i > J 1 i 1 i C O i—i 1 ft 0 1 > ft 0 ft ft ft ft ft ft J ft C O ft ft C D C O ft C O > ft u o 0 0 rf 2 ft rf w a rf C D ft ft ft C O ft H a ft CM Q C O a CO rf ft H ft C D I -C O 53 rf ft C J > H >< ft ft u ft ft Q a ft ft E H E H ft C O 1—1 ft E H C O ft rf 1—1 i—! ft > H a > ft rf ft Q cn C N on C O C O C O r> vo co r» VO 4-1 H O C N 4-1 1 ft ti IS o CD o\ on ti ft ft ft rd U £ • H >i >i cd ft H r H ft u ft ft 0 _Q O I o CO co to CO I I O T T T T TT 5 5 3 5 i 8 iu £ fj iu I 5 I Q O CL _ r-1 CD JD O 11 111 CO C Q CD CO c co i— -*—< c 03 E -J I IN CT) Q . C CO E D I IS O) CL c CD A . o iz O CT) CL Q) CO =3 O CT) CL -t—' JO CO DC f- C M • i N N CL CL > > LL LL 2 9 Transferrin p97 Figure 2: Ske t ch of the b i lobed structure of transferr in a n d p97. Aspar t i c ac id (D), tyros ine (Y) a n d hist idine (H) a re important am ino ac id res idues involved in coord inat ion of iron a tom (Fe3+) in the binding site. T h e aspar t ic ac id in the C - l o b e of p97 is c h a n g e d to a ser ine (S) res idue. 30 carboxylate oxygen of D82, two phenolate oxygens from Y114 and Y207 and the imidazole nitrogen of H268. The hydrogen bonding network to the synergistic carbonate anion is provided by T139, R143 and S144 (Baker and Lindley, 1992). The uptake of iron by transferrin/ transferrin receptor is mediated by clathrin-mediated pathway (Fig. 3). Briefly, circulating holo-transferrin binds to the transferrin receptor on the cel surface at physiological pH. Once bound, the transferrin/ transferrin receptor complex is endocytosed into the cel through the formation of the clathrin-coated pit. The clathrin-coated vesicles either mature into or fuse with endosomes. The decrease in intra-vesicular pH results in a decrease in the afinity of iron for transferrin, causing iron to be released from the transferrin/ transferrin receptor complex. The iron is transported out of the endosomal compartment by the divalent metal transporter Nramp2 and into the cytoplasm. The cytoplasmic iron is transfered either to ferritin for storage or to other organeles such as mitochondria for celular usage (Fleming et al., 1998; Tabuchi et al., 2000). The interaction between transferrin receptor and the hemochromatosis protein, HFE, has required the re-evaluation of iron uptake by this pathway. In vivo studies show that holo-transferrin, transferrin receptor and HFE interact in a 1:2:1 ratio (Gross et al., 1998; Lebron et al., 1998). HFE complexes with transferrin receptor in the endoplasmic reticulum and the complex trafics to the plasma membrane via the Golgi network. The association of HFE and transferrin receptor is required for locating transferrin receptor on the cel surface (Gross et al, 1998; Salter-Cid et al., 2000a) and for decreasing the rate of endocytosis of transferrin/ transferrin receptor complex, which results in a down-regulation of transferrin-dependent iron uptake (Corsi et al., 1999; 31 Figure 3: Iron uptake mediated by transferrin/ transferrin receptor. Circulating holo-transferrin binds to the cels surface transferrin receptor-HFE complex at physiological pH and the complex is endocytosed through formation of the clathrin-coated pit. The clathrin-coated vesicles mature into or fuse with endosomes. The decrease in intra vesicular to approximately pH 5.5 results in the decrease in the afinity of iron for transferrin and hence iron (Fe3+) is released from the complex. Early endosomes gradualy matures into late endosomes in the endocytic pathway. The iron is transported out of the late endosomes or lysosomes by the divalent metal transporter, Nramp2, into the cytoplasm, which is either incorporated into ferritin for storage or transfered to other organeles such as mitochondria for celular usage 32 Transferrin /Transferrin receptor/HFE 33 Riedel et al., 1999; Roy et al., 1999). Hence, HFE, through its binding to the transferrin receptor, appears to function as a negative regulator of transferrin receptor mediated uptake of iron. Lactoferrin shares a 40% protein sequence homology with human serum transferrin. It is an 80 kDa glycoprotein secreted mainly by the glandular epithelial cel of the mammary gland (Campbel et al., 1992). It is also present in the secretion of other exocrine fluids such as saliva, bile, pancreatic fluid and tears (Colomb et al., 1976; Inoue et al., 1993; Janssen and van Bijsterveld, 1983; Jonas et al., 1993; Saito and Nakanuma, 1992). Lactoferrin appears to have additional functions other than iron-binding. Lactoferrin is stored in the secondary granules of polymorphonuclear neutrophils and is released during inflammation (Crouch et al., 1992). It modulates the inflammatory response by blocking the formation of C3 convertase in the classical complement pathway (Britigan et al., 1994; Rainard, 1993). In addition, lactoferrin promotes the rapid migration of polymorphonuclear neutrophils from blood to the site of inflammation by increasing P-selectin mediated binding of the leukocytes to the endothelium (Asako et al., 1994; Kurose et al., 1994). Lactoferrin also plays a role in host defense at mucosal surfaces due to its antibacterial properties (Haversen et al., 2000). In such a role, lactoferrin functions as an iron chelator, exerting a bacteriostatic efect by denying the bacteria access to iron. (Spik and Montreuil, 1983). Another immune modulation efect exerted by lactoferrin is its ability to reduce the secretion of IL-1, IL-6 and tumor necrosis factor-a from monocytes in response to bacterial lipopolysaccaride (Crouch et al., 1992; Cumberbatch et al, 2000; Otsuki et al., 1998). 34 Ovotransferin is one of the components of the egg white protein and is implicated in the transfer of trace elements from the hen oviduct into developing chicken embryos. The high level of amino acid sequence homology between ovotransferin and the rest of the transferrin family members suggests that they share similar biological roles in binding and transporting irons. However, recently the anti-microbial properties of ovotransferin have been investigated. A cationic fragment of ovotransferin appears to interact with bacterial membranes and cause damage to the bacterial cytoplasmic membranes (Ibrahim et al., 1998; Ibrahim et al., 2000). B. Overview of p97 a. p97 as an iron-binding protein P97, also known as melanotransferin, is a 97 kDa protein, which shares a 40% protein sequence identity with human lactoferrin (Rose et al., 1986). It was first identified as a cel surface marker for human skin cancer (Brown et al., 1981; Woodbury et al., 1981; Woodbury et al., 1980). However, the protein was subsequently found expressed at various levels in liver, intestine, umbilcal cord, placenta, sweat gland and more recently in human brain endothelium (Barresi and Tuccari, 1994; Rothenberger et al, 1996; Sciot et al., 1989). The localization of p97 to the capilary endothelium in the brain led to the proposal that p97 may play a role in iron transport to the brain (Rothenberger et al., 1996). Immunohistochemistry and immune electron microscopy studies have also localized p97 to the apical brush border of epithelial cels in the fetal intestine (Alemany et al., 1993). 35 Unlike the other members of the transferrin family, p97 has two diferent forms. It can be found as a membrane protein atached to the cel surface via a glycosylphosphatidylinositol (GPI) anchor or as a soluble form in serum (Alemany et al., 1993; Brown and Rose, 1992; Food, 1992). These forms are thought to originate from alternative splicing of the p97 mRNA (Hsu, 1997; McNagny et al., 1996). Metal binding studies on a cel line transfected with human p97 show GPI-anchored p97 can mediate iron uptake where the process is both temperature dependent and saturable (Kennard et al., 1995). Like other members of the transferrin family, p97 is a bi-lobed metal binding protein (Fig. 2). Protein sequence alignment of p97 with other members of the transferrin family shows that the amino acids involved in coordinating the iron atom in the N-lobe are totaly conserved (Rose et al., 1986) (Fig. 1 and 2). However, there are at least four specific amino acid substiutions in the C-lobe of p97 which were thought to afect the ability of the protein to bind iron (Fig. 1): (i) The aspartic acid at positon 421 is changed to a serine residue (D421S). The D421S substiution is thought to afect the simultaneous hydrogen bonding across the iron binding cleft and results in a failure to lock in the iron atom when it is bound, (ii) The arginine at 482 is changed to a serine residue (R482S). (iii) Threonine at positon 478 is changed to an alanine residue (T478A). (iv) Threonine at position 483 is changed to a proline residue (T483P). Both R482S and T478A substiutions are thought to afect the binding of carbonate anion of the binding site and the T483P substiution is proposed to disrupt the hydrogen bond between threonine and aspartic residue at 421. Nevertheless, the crystal structure of human lactoferrin with a D60S amino acid substiution in the N-lobe, which mirors the D421S substiution in p97, shows that a water molecule coordinate the iron atom and alows the iron to bind to the 36 site. The binding of iron to the modifed pocket in this situation may be weakened (Faber etal., 1996). b. Regulation of p97 expression The binding of the iron regulatory protein to the iron response element is an exquisite system used by cels to regulate the expression of iron proteins involved in maintaining the delicate balance of iron homeostasis (Cairo and Pietrangelo, 2000; Haile, 1999). Most but not al proteins involved in iron metabolism have a conserved RNA stem-loop structure located either on the 3 or 5 untranslated region of the mRNA caled the iron response element. When iron is abundant, the iron regulatory protein does not bind to the iron response element. However, when iron is scarce, the iron regulatory protein binds to the iron response element and confers translational regulation. If the iron response element is located in the 3 untranslated region of the mRNA (such as that of transferrin receptor mRNA), binding of iron regulatory protein to the iron response element of the mRNA stabilzes the transcript and thereby enhances the translation of the protein. However, if the iron response element is located in the 5 untranslated region of the transcript (for example, ferritin mRNA), the binding of the iron regulatory protein to the iron response element disrupts translation (for review please see Eisenstein, 2000). In contrast to the transferrin receptor and ferritin, where the level of protein expression is regulated by the level of iron through the iron response element/ iron regulatory protein system, there is no identifiable iron response element motif in p97 and the level of expression is not regulated by iron (Richardson, 2000). However, a regulatory element has been located 2kb upstream of the p97 promoter region. Deletion of this element 37 severely impairs the expression of p97 (Duchange et al., 1992). This regulatory element was shown to be part of an enhancer composed of two binding sites for AP-1 transcription factor. AP-1 is a transcription factor that is up-regulated after ultra-violet irradiation, a suspected risk factor in melanoma incidence. This may explain how p97 is up-regulated in melanoma cels (Derijard et al., 1994; Devary et al., 1991). c. Orthologues of p97 Several orthologues of p97 have been identified and cloned (Fig. 1). The protein sequences of these diferent p97s show a much higher sequence homology between themselves compared to other transferrin family members. The conservation of all 28 cysteine residues in al the various p97 suggests similar protein folding and tertiary structure with the other transferrins. The identification of the diferent p97 orthologues has implied a variety of functions for the molecules. The chicken p97 (EOS 47) has a 61% protein sequence identiy with human p97 (McNagny et al., 1996). Unlike other p97 orthologues that have conserved iron-binding residues in the N-lobe and non-conserved iron-binding residues in the C-lobe, the chicken p97 has an additonal H278R amino acid substiution that may prevent iron-binding in the N-lobe. Hence, chicken p97, with both atypical N-and C-lobes, appears to have evolved away from high afinity iron-binding. Chicken p97 is highly expressed in epithelial brush borders of smal intestine and kidney and expressed at lower levels in cels lining the sinusoids of the liver. A sub-population of eosinophils in the chicken bone marow also expresses increased level of chicken p97, however, it is undetectable in the matured peripheral eosinophils, suggesting that the expression of chicken p97 ceases shortly 38 before these cels emigrate from the bone marow into the peripheral blood. In addition, chicken p97 appears to be a direct target of the GATA-1 and C/EBPfJ transcription factors, which are responsible for the diferentiation of eosinophils (McNagny et al., 1996). In this regard, chicken p97 seems to be a developmentaly regulated during eosinophil maturation. Folowing the cloning of chicken p97, rabbit p97 was cloned from rabbit chondrocyte cultures. The protein sequence of rabbit p97 shows 85% identiy to human p97. Western blot, Northern blot and reverse-transcription PCR analyses indicated that rabbit p97 is highly expressed in cartilage and chondrocytes. Immunohistochemistry staining of the rabbit chrondrocytes demonstrated that rabbit p97 is expressed on the plasma membrane of the cels. Rabbit p97 transcripts were detected in all zones of the cartilage but the level was relatively low in the hypertrophic zone. These findings suggest that p97 is involved in maintaining the cel surface characteristics of chondrocytes (Kawamoto et al., 1998). The mouse orthologue of p97 was cloned from a mouse embryonic cel line and it shows 62% protein sequence homology to the human p97 (Nakamasu et al., 1999). Sequence alignment shows that the amino acids involved in iron coordination in the N-lobe is conserved, whereas the C-lobe shows a D421R and R482S amino acid substiutions in the iron-binding pocket. It is suggested that the amino acid substiutions can impair iron binding (Nakamasu et al., 1999). Northern blot analysis shows that tissue expression patern of mouse p97 is diferent from human p97. While human p97 is widely expressed, the expression of mouse p97 seems to be restricted to testes and 39 cartilage tissues and its expression seems to be regulated during chondrocyte diferentiation (Nakamasu et al., 1999). The completion of the Drosophila melanogaster genome sequence alowed the identification of two fly orthologues of p97, fly p97-l and fly p97-2. The absence of proteins orthologous to vertebrate transferrin receptor in the fly would suggest that the fly orthologues of p97 could mediate the insect pathway for celular uptake of iron and other metal ligands (Adams et al., 2000). However, some of the critical iron-binding residues in both the fly homologues are not conserved. Therefore, fly p97 might play a role separate from its ability to bind iron. d. Alzheimer s disease and p97 Expression of p97 has been examined in brains from Alzheimer s disease patients Immunohistochemical staining shows that p97 is highly expressed in a subset of reactive microglial cels associated with P-amyloid positve senile plaques in Alzheimer s disease brains (Jeferies et al., 1996). This has been confirmed by positve in situ hybridization of p97 mRNA in reactive microglial cels in Alzheimer s disease brain sections (Yamada et al., 1999). Soluble p97 was also found to be elevated in the cerebral spinal fluid and serum of patients with Alzheimer s disease compared with healthy age matched controls (Kennard et al., 1996; Kim et al., 2001). In addition, serum p97 has been shown to cross the blood-brain barrier and deliver iron across the blood-brain barrier more eficiently than transferrin, if both are injected into the tail vein of the mouse (Moroo et al., 2001). Hence, it was proposed that p97 could play a role in the pathology of Alzheimer s disease. Metals such as iron, zinc and copper have been shown to precipitate (3-amyloid 40 plaque formation (Bush, 2000; Bush et al., 1994). P97 could be involved in the disease by transporting iron or other metals across the blood-brain barrier, which could augment amyloid plaque formation. However, the mechanism of p97 transport across the blood-brain barrier is still being investigated. Alternatively, expression of p97 on reactive microglial cels could be a result of inflammatory response related to Alzheimer s disease. p97 may bind excess amounts of free iron released from damaged tissues and therefore provide a protective efect against iron generated free radicals. Nevertheless, these two hypotheses are still being tested. e. Other possible functions of p97 The identification of several serum transferrin homologues such as saxiphiln in bulfrog, pacifastin in crayfish and porcine inhibitor of carbonic anhydrase in pig, which have physiological roles unrelated to iron transport, leads to the proposal that p97 might have alternate functions (Lenarcic et al., 2000; Liang et al., 1997; Wuebbens et al., 1997). The presence of a proposed zinc- binding motif (346 HEYLH 350) between the N- and the C-lobes, which is identical to the consensus Zn-binding motif in thermolysin, suggests that p97 might have metaloprotease activity (Garat and Jhoti, 1992). However, the H E X X H motif was not conserved in other p97 orthologues (Fig. IB). Therefore, the signifcance of human p97 possessing the zinc binding activity while p97s from other species do not, is in doubt. A recent review proposes a key and lock interaction between GPI-anchored p97 on one cel and transferrin receptor on another cel (Sekyere and Richardson, 2000). It was suggested that the interaction would alow docking of p97 on the cel surface and 41 thus target the metaloprotease action to the site. Recently it was also suggested that p97 could function as an intercelular adhesion molecule (Sekyere and Richardson, 2000). Sequence analysis of the human and mouse p97 shows that the human protein has one arginine-glycine-aspartic acid (RGD) motif whereas the mouse p97 has two. RGD motif is proposed to serve as an atachment site for integrin during cel-cel adhesion (Ruoslahti, 1996). Having a RGD motif could imply that p97 plays a role in celular adhesion by interacting with cel surface integrin molecules (Ruoslahti, 1996). Nevertheless, there is no experimental data to support this claim. While it has been shown that GPI-anchored p97 can mediate celular iron uptake, the function of soluble p97 has not been defined. There are data indicating that soluble p97 can cross the blood-brain barrier, however, the identiy of the receptor that mediates the transcytosis of soluble p97 is still under investigation (Moroo et al., 2001). There is a preliminary report indicating that soluble p97 can bind to transferrin receptor (Walker et al., 1999) but obviously, further investigations are required to define both the metaloprotease activity and the intercelular adhesion mechanism of p97. V. Transferrin/ transferrin receptor independent iron transport The existence of transferrin/ transferrin receptor-independent iron transport pathways is shown from evidence that transferrin is not universaly required for eficient iron uptake. For example, hypotransferinemic (hpx) mice have massive tissue iron overload in al non-hematopoietic tissues even though serum transferrin levels are low (Bernstein, 1987; Craven et al., 1987). The liver iron burden in these animals is 42 100-fold higher than that of normal wild-type mice (Trenor et al., 2000). Clearly, transferin-independent mechanisms are mediating the iron transport in these mice. There are three general categories of transferrin/ transferrin receptor independent iron transport: (A) Transferin-dependent transferrin receptor-independent pathway (e.g. iron transport mediated by binding of transferrin to the newly identified transferrin receptor 2); (B) Transferin-independent transferrin receptor-dependent pathway (e.g. the proposed binding of soluble p97 to transferrin receptor); (C) Transferrin and transferrin receptor-independent pathways (e.g. iron uptake mediated by GPI-anchored p97, iron absorption mediated by heme receptor, Nramp2 and paraferritin complex in the intestine). A. Transferin-dependent transferrin receptor-independent pathway The homologue of transferrin receptor 1, transferrin receptor 2, has recently been identified. The human transferrin receptor 2 is a type II transmembrane glycoprotein that shares 66% protein sequence similarity in its extracelular domain with transferrin receptor (Kawabata et al., 2000; Kawabata et al., 1999). Kawabata et al. showed that transferrin receptor 2 could bind and internalize transferrin, even though transferrin has a lower afinity for transferrin receptor 2 (Kawabata et al., 2000; West et al., 2000). There are also controversies as to whether transferrin receptor 2 can associate with HFE like transferrin receptor 1 and whether HFE has any efect on the transferrin binding to transferin receptor 2 (Lieu et al., 2001; West et al., 2000). Unlike the ubiquitous expression of transferrin receptor 1 expression, transferrin receptor 2 is predominantly expressed in the liver. In addition, transferrin receptor 2 43 does not have the iron response element, which means that its expression is not regulated by celular levels of iron. The lack of the iron regulatory element in transferrin receptor 2 may contribute to iron overloading due to uptake of transferrin bound iron despite down-regulation of transferrin receptor 1 expression by high levels of iron (Fleming et al., 2000). Clearly, further investigations are required to define whether transferrin receptor 1 and transferrin receptor 2 are regulated by distinct pathways and whether they mediate iron uptake by diferent mechanisms. B. Transferin-independent transferrin receptor-dependent pathway The high homology of p97 with serum transferrin suggests that it could bind to the transferrin receptor or even transferrin receptor 2. Experiments have also shown that soluble p97 is able to cross the blood-brain barier into the brain and it is likely that this process is receptor mediated (Moroo et al., 2001). Moreover, there is a preliminary report indicating that soluble p97 can bind to transferrin receptor (Walker et al., 1999). The high sequence homology between transferrin receptor and transferrin receptor 2 suggests that transferrin receptor 2 cannot be ruled out as a possible receptor for p97. C. Transferrin/ transferrin receptor -independent pathways Transferrin/ transferrin receptor independent pathways include examples of iron transport by GPI-anchored p97 and iron absorption in the intestine. a. GPI-anchored p97 44 GPI-anchored p97 is able to transport iron in p97 transfected Chinese hamster ovary cel line and this process is both temperature dependent and saturable (Kennard et al., 1995). In SK-MEL 28 human melanoma cels line, where the level of p97 expression is elevated, a similar activity has been studied, which suggested that iron uptake can be mediated by p97 (Richardson and Baker, 1991a; Richardson and Baker, 1991b). It is interesting to note that GPI-anchored p97 is also found in the apical brush border of fetal intestine, which implies a possible function in gut iron absorption (Alemany et al., 1993). b. Iron absorption in the intestine Iron homeostasis is controled by absorption rather than by excretion because there is no known specific iron excretion mechanism in higher organisms. These cels are responsible for sensing the internal iron load in the serum and responding appropriately by altering iron uptake using one of at least three pathways known to mediate iron absorption on the apical side of the absorptive vilus cel (Fig. 4). i. Ferous iron absorption by Nramp2 The most extensively characterized iron absorption pathway is via a membrane spanning protein caled Nramp2 (also known as the divalent metal transporter, DMT-1 or DCT1). Nramp2 is a metal-proton symporter that transports ferous iron from the intestinal lumen into the vilus cel (Canonne-Hergaux et al., 1999; Fleming et al., 1998; Gunshin et al., 1997). Nramp2 is capable of transporting not only ferous iron but also a broad range of divalent cations including Zn2+, Mn2+, Co2+, Cd2+, Cu2+, Ni2+ and Pb2+ (Gunshin et al., 1997). It is expressed on the apical plasma membrane of the vilus as 45 Figure 4: The intestinal iron absorption pathways. The HFE/ transferrin/ transferrin receptor complex relays the level of plasma iron to the immature crypt cels. The level of iron regulates a concerted action of post-transcriptional regulatory system of iron response element and iron regulatory protein to modulate the expression of proteins involved in iron absorption in the mature vilus cels. There are three proposed pathways for intestinal iron absorption. (1) Dietary iron in the ferric form (Fe3+) can be reduced to ferous iron (Fe2+) by duodenal associated ferric reductase (Dcytb). (2) Fe2+ and Fe3+ can be absorbed by the vilus cel by the paraferritin complex, consisting of (3-integrin, mobilferrin and flavin mono-oxygenase. (3) Heme is released from myoglobin and hemoglobin in the intestinal lumen by gastric enzymes and enters the vilus cel as intact metaloporphyrin through a heme receptor. Iron is released from heme by heme oxygenase-1 (HO-1) and transported across the vesicular membrane by iron dependent ATPase (Fe-ATPase). Intracelular iron can either be stored in ferritin or be transported across the basolateral membrane by IREG1. Hephaestin appears to facilitate iron egress by oxidizing Fe2+ into Fe3+, which is incorporated into serum apo-transfenin. 46 4 7 wel as internaly on the membrane of late endosomes and lysosomes of the cel (Fleming et al., 1998; Tabuchi et al., 2000). The apical Nramp2 protein mediates divalent metal absorption from the intestine whereas the vesicular Nramp2 is responsible for metal egress from the endosomes (Fleming et al., 1999; Grifiths et al., 2000; Tabuchi et al., 2000). The expression of Nramp2 on the apical surface of intestinal vilus cel is regulated by the level of labile iron pool in the cel (Andrews, 1999). Other proteins including the Iron-regulated transporter (IREGl)(also known as feroportin in zebra fish) and the recently cloned apical ferric reductase, Dcytb, are also regulated by this level (Donovan et al., 2000; McKie et al., 2001). The level of labile iron pool is established in the enterocyte precursor caled the crypt cel. In the crypt cel, transferrin receptor bound to the HFE at the basal lateral side of the cel faciltates the uptake of transferrin from the plasma. Iron from transferrin contributes to the labile iron pool inside the cel. This pool of labile iron modulates the activities of the iron regulatory proteins inside the crypt cel. If the level of labile iron pool is low, the translation of Nramp2, IREG1 and Dcytb is initiated when the crypt cel matures into absorptive vilus cel. The apical ferric reductase, Dcytb, functions to reduce dietary ferric iron to ferous form before the iron is transported by Nramp2. Once iron is transported into the vilus cels, it has to exit the vilus cel into the plasma. IREG1, a basal lateral membrane protein of the vilus cel, functions to transport iron into the plasma. Once transported, hephaestin, a copper containing feroxidase, localized to the basal lateral membrane of the vilus cel, oxidizes the ferous iron to ferric irons and promotes the release of iron to the circulating apo-fransferrin in the serum (McKie et al., 2000; Vulpe et al., 1999). 48 ii Paraferitn-mediated iron absorption A second proposed pathway of iron absorption in the intestine is mediated by a 520 kDa protein complex caled paraferritin, which is made up of (3-integrin, mobilferrin and flavin monooxygenase (Conrad and Umbreit, 1993; Conrad et al., 1993). Ferric iron solubilzed by mucin in the gut lumen, are bound to the cel surface (33-integrin and transfered to mobilferrin, which is found docked on the inner cytoplasmic leaflet of the plasma membrane (Conrad et al., 1994). Flavin monooxygenase becomes associated with the (33 integrin-mobilferrin complex and reduces the iron to ferous form in concert with the oxidation of NADPH. The iron can then be incorporated into heme, ferritin or other molecules. iii. Heme absorption Heme is another important source of dietary iron (Ponka, 1999). Hemoglobin-and myoglobin-derived iron are absorbed more eficiently by the intestine than inorganic iron because the presence of amino acids from proteolytic digestion helps to prevent polymerization and precipitation of heme (Majuri and Grasbeck, 1987; Rafin et al., 1974; Turnbul et al., 1989). The intact metaloporphyrin is liberated from the hemoglobin and myoglobin through enzymatic digestion in the intestinal lumen. There is evidence to suggest the presence of a heme receptor on the luminal side of vilus cels that transports the metaloporphyrin form of heme into the cel (Grasbeck et al., 1982). Once heme is inside the cel, iron is released via heme oxygenase-1 (HO-1), which cleaves the heme ring to generate bilirubin, carbon monoxide, and ferous iron (Maines, 49 1997). In mice deficient in HO-1, tissue iron stores are high while serum iron levels are low, suggesting that the enzyme is involved in the discharge of iron from certain cels (Poss and Tonegawa, 1997a; Poss and Tonegawa, 1997b). The method by which iron is liberated from heme and subsequently routed for use systemicaly is not fuly understood. Recently, a mammalin iron ATPase (Fe-ATPase) associated with the microsomal membrane and co-distributed in tissues with HO-1 was identified (Baranano et al., 2000). There is some evidence to suggest that HO-1 is functionaly coupled to the Fe-ATPase. This novel Fe-ATPase is enriched in spleen and mediates the iron transport upon hydrolyzing nucleotide triphosphate (Ferris et al., 1999; Poss and Tonegawa, 1997a). Bara ano and associates propose a model in which Fe-ATPase and HO-1 co-localize to the endoplasmic reticulum where heme is first degraded by HO-1 and the freed iron is transported by the Fe-ATPase to the luminal side of the endoplasmic reticulum (Baranano et al., 2000). They suggested that the iron in the endoplasmic reticulum could then bind to newly synthesized transferrin and subsequently be recycled back to the cel surface for exocytosis. VI. Hypothesis and General approach The focus of this thesis was to add to the basic understanding of the function of p97. The main hypothesis of this thesis was p97 is an iron binding protein that is actively involved in iron transport into cels. At the start of this research, there were questions regarding the validity of p97 as an iron transport protein for several reasons. First, although sequence analysis showed that p97 shared high sequence homology with other members of the transferrin family, only the amino acids involved in iron coordination in 50 the N-lobe were conserved. The C-lobe of human p97 was proposed not to bind to iron due to two amino acids substiutions at D421S and R482S, in contrast to other members of transferrin family where these amino acids were totaly conserved (Fig. 1). The crystal structure of human lactoferrin, with a D60S substiution in the N-lobe that mirors the D421S substiution in human p97, shows that iron can bind to the site. This suggests that the Asp to Ser substiution would not severely afect the ability of the protein to bind iron (Faber et al., 1996). Second, there were doubts as to whether any iron bound and transported by p97 could be donated to ferritin (Richardson and Baker, 1991a; Richardson and Baker, 1991b). Third, there was evidence showing that metals such as iron, copper, zinc and aluminum play important roles in several neurological disorders such as Alzheimer s disease and Parkinson s diseases (Bush, 2000; Bush et al., 1994; Perl and Pendlebury, 1986; Sayre et al., 2000). However, the route of metal uptake into the brain is not known. Soluble p97 has been shown to cross the blood-brain barrier (Moroo et al., 2001). Therefore, it was hypothesized that p97 could mediate the uptake of metal ions into the brain. However, before establishing whether p97 can mediate the transport if these metals across the blood-brain barier in vivo, it is important to determine whether p97 can bind to these metals in vitro. It is known that transferrin can bind to other metals such as zinc, copper and aluminum at the same iron-binding site. Thus, p97, with identical iron-binding site in the N-lobe, might be able to bind to these metals as wel. The first results chapter therefore addresses whether the p97 can indeed bind two atoms of iron and relatedly whether p97 can bind other metals. Iron binding to p97 was addressed using UV-visible spectroscopy and diferential scanning calorimetry. The binding constant of iron to p97 was ascertained using calorimetric method and the ability 51 of p97 to bind other metals was demonstrated through competion experiment in the presence of iron. Urea polyacrylamide gel analysis was used to coroborate the binding of other metals to p97. In addition, to resolve whether iron transported by p97 is delivered to ferritin, iron loading of ferritin was compared between p97 transfected and non-transfected cel lines. The second results chapter of the thesis examines the endocytosis pathway of GPI-anchored p97 and folows the fate of the iron as it is being taken up by p97. Though the internalization pathways of many GPI-anchored proteins have been studied in detail, debate still exists on whether caveolae or clathrin-coated vesicles are involved in their uptake. For example, there are debates on whether GPI-anchored prion protein is localized to caveolae or clathrin coated pits (Harmey et al., 1995; Shyng et al., 1994). Furthermore, very little is known about the fate of ligands bound to GPI-anchored molecules after they are internalized. The most studied ligand internalized by a GPI-anchored protein is the uptake of folate by GPI-anchored folate receptor (Anderson, 1998; Anderson et al., 1992). However, there are disagreements on whether the folate is traficked to endosomes once it is internalized (Kamen et al., 1988; Rothberg et al., 1990). Confocal immunofluorescence microscopy and sub-celular fractionation was used to address whether GPI-anchored p97 bound with iron is endocytosed into caveolae or clathrin-coated pits. Furthermore, the internalization pathway of GPI-anchored p97 in the cytoplasm was folowed. GPI-anchored p97 was shown to traffic to the endosomes and subsequently to vesicles containing Nramp2. The observations led to a proposed model describing how iron is transported by GPI-anchored p97 into the vesicles and eventualy donated to the ferritin in the cytoplasm. 52 Taken together, the studies on p97 provide new and fundamental information on both transferrin/ transferrin receptor independent uptake of iron and also on GPI anchor mediated uptake of ligands. 53 Chapter 2: Materials and Methods I. Tissue culture, cells and antibodies TRVb cel line, a Chinese Hamster Ovary (CHO) cel line that does not express functional endogenous hamster transferrin receptor, was obtained from Dr. F. Maxfield (New York University, NY) (McGraw et al., 1987). This cel line was co-transfected with pSV2p97a, encoding the ful-length human p97 cDNA driven by the SV40 early promoter, and the geneticin (G418) resistance vector, pSV2neo (Food et al., 1994). The p97 transfected cel line is caled p97+ TRVb in this thesis. TRVb transfected with human transferrin receptor clone pCDTRl and pSV3neo, was also obtained from Dr. F Maxfield. The transferrin receptor positve cel line is named TfR+TRVb in this thesis. All three cel lines were maintained in Ham's-F12 media (Invitrogen Life Technologies Inc., Burlington, ON) supplemented with 10% (v/v) fetal bovine serum (Invitrogen Life Technologies Inc.), 2 mM glutamine (Invitrogen Life Technologies Inc.), 20 mM HEPES (Sigma-Aldrich Canada Ltd), 100 U/ml penicilin (Sigma-Aldrich Canada Ltd, Oakvile ON) and 100 g/ml streptomycin (Sigma-Aldrich Canada Ltd), pH 7.4, at 37 °C in a 5% C02 humidifed incubator. The media for p97+TRVb also included 500 j-ig/ml G418 (Invitrogen Life Technologies Inc.). The human melanoma cel line SK-MEL 28 which expresses high levels of human p97, was obtained from American Tissue Type Culture Colection (ATCC, Manassas, VA). This cel line was maintained in DMEM supplemented with 10% (v/v) fetal bovine serum, 2 mM glutamine, 20 mM HEPES, 100 U/ml penicilin and 100 pig/ml streptomycin, pH 7.4, at 37°C in a 5% CO2 humidifed incubator. 54 Antibodies used in this study are listed in Table Ilia and table Illb. 55 Table Ilia: Primary antibodies used in the experiments presented in this thesis Antibodies Antigen Host IsotvDe Concentration Source L235 Human p97 Mouse IgGl 15 mg/ml ATCC OKT-9 Human TfR Mouse IgGl 12 mg/ml ATCC Caveolin Human caveolin-1 Rabbit 250 jig/ml BD Transduction Laboratories, Franklin Lakes, NJ. Adaptin-(3 Human (3-adaptin Mouse IgGl 250 |ig/ml BD Transduction Laboratories EEA-1 Amino terminal of human EEA-1 Goat 200 u.g/ml Santa Cruz Biotechnology Inc., Santa Cruz, CA. LAMP-1 Amino terminal of human LAMP-1 Goat 200 u:g/ml Santa Cruz Biotechnology Inc Clathrin HC Carboxy terminal of human clathrin heavy chain Rabbit 200|ig/ml Santa Cruz Biotechnology Inc Ferritin Human ferritin L-chain Rabbit 1.3 mg/ml DAKO, Carpinteria, CA. Anti-transferrin receptor Transferin receptor Rabbit Gift from Dr. I. Trowbridge, The Salk Institute, San Diego, CA* * The rabbit anti-transferrin antiserum is used for western bloting 56 Table Illb: Secondary antibodies used in the experiments presented in this thesis Antibody Concentration Dilution used Source Alexa 488 nm conjugated Goat anti-mouse IgG 2 mg/ml 1:1000 Molecular Probes Inc., Eugene, OR Alexa 568 nm conjugated Goat anti-rabbit IgG 2 mg/ml 1:500 Molecular Probes Inc Alexa 568 nm conjugated Rabbit anti-mouse IgG 2 mg/ml 1:500 Molecular Probes Inc Alexa 568 nm conjugated Rabbit anti-goat IgG 2 mg/ml 1:500 Molecular Probes Inc Alexa 568 nm conjugated Goat anti-mouse 2 mg/ml 1:500 Molecular Probes Inc Cy5 conjugated Rabbit anti-mouse IgG 1 mg/ml 1:500 Jackson Laboratories, West Grove, PA. 57 II. Iron-binding and competition experiments In preparation for the iron-binding and metal competion experiments, the cels were cultured in a 35 mm x 10 mm tissue culture dish (Nunc, Rochester, NY). Each dish was seeded with 3 x 105 cels that would grow to 100% confluency in 5 days. On the day of the experiment, each tissue culture dish was visualy checked under the microscope for 100% confluency before commencing the experiment. The cels from two tissue culture dishes were counted separately by trypan blue exclusion method using a hemocytometer and averaged. The number of cels routinely used in each sample average between 1.4-1.6xl06 cels. The method used to study the binding of radioactive iron to p97 is a modifcation of the binding assay published by Richardson and Baker (Richardson and Baker, 1991a; Richardson and Baker, 1991b). Briefly cels were washed 3 times with 1ml of Minimum essential media (MEM) (Invitrogen Life Technologies Inc., Burlington, Ont) supplemented with IX MEM non-essential amino acids (NEAA)(Invitrogen Life Technologies Inc., Burlington, Ont) for 30 minutes at 37°C in a 5% CO2 humidifed environment. This was to ensure that excess iron and transferrin from the growth media were removed. A stock solution of 10 mM of 5Fe-citrate was prepared by mixing 5 5FeCl3 in 0.5M HC1 (Perkin Elmer Life Sciences Inc., Boston, MA) in citric acid at a 1:100 molar ratio to prevent hydrolytic precipitation of the iron complexes, which occur at physiological pH of the incubation media. Varying concentrations of 5Fe-citrate were prepared by serial dilution in MEM supplemented with 20 mM HEPES, 100 U/ml penicilin, 100 p.g/ml streptomycin, 1.25 u,g/ml bovine serum albumin (BSA) (Fisher Scientific Limited, Nepean, ON) and IX MEM NEAA. P97+TRVb and p97"TRVb cel 58 lines were incubated with 1ml of the 5Fe-citrate solution at 4°C. After 4 hours of incubation, the cels were washed 4 times with 1ml Hanks Balanced Salt Solution (Invitrogen Life Technologies Inc.) at 4°C. Phosphate bufered saline (PBS) (1ml) containing 0.25% (w/v) trypsin (Invitrogen Life Technologies Inc.) 1 mM EDTA (Sigma-Aldrich Canada Ltd) was added to the cels and incubated for 30 minutes at 37°C. The cel suspension was transfered into a scintilation vial and the tissue culture dish was rinsed with 0.5 ml of PBS and added to the scintilation vial. Beckman ready safe scintilation fluid (5ml) was added to the scintilation vial and the amount of radioactive decay was measured in the Beckman LS6000IS Liquid Scintilation counter (Beckman Coulter Inc., Fulerton, CA). The radioactivity associated with each of the cel lines was normalized with the number of cels. Amount of radioactivity bound to GPI-anchored p97 was calculated by subtracting the amount of radioactivity associated with TRVb from that of p97+TRVb. Each sample was done in triplicate and the average ploted and standard deviation of the three samples were calculated and ploted as the eror bar of the y-axis in Figure 5. For competion experiment, a stock solution of 10 mM of metal citrate solution was prepared by mixing aluminum chloride, copper chloride and zinc chloride (al from Sigma-Aldrich Canada Ltd) with citric acid in a 1:100 molar ratio. The cels were then incubated for 4 hours at 37°C with l|iM of cold Fe-citrate, Al-citrate, Cu-citrate or Zn-citrate in the presence of increasing concentrations of 5Fe-citrate. At the end of the incubation, the cels were transfered to a 4°C chamber and washed four times with 1 ml of ice cold Hanks Balance Salt Solution. The cels were trypsinized and the radioactivity 59 associated with the cels is counted in the Beckman LS6000IS Liquid Scintilation counter. III. Dialysis of soluble p97 and iron loading protocols Soluble p97, graciously provided by Synapse Technologies Inc, Vancouver, BC, was partialy saturated with iron. In order to load soluble p97 with other metals, iron contained in soluble p97 must be removed. The removal of iron is achieved by the dialysis protocol from Aisen et al (Aisen and Leibman, 1968; Aisen and Leibman, 1972). Briefly, 10 ml of approximately 1 mg/ml of soluble p97 was transfered into the Pierce 10,000 MW dialysis cassete and dialyzed in 0.1 M sodium citrate pH 6.0 in 4°C with constant stirring for an hour before the bufer was changed two more times. The protein was transfered to a new cassete and dialyzed for four hours in 0.1 M sodium perchlorate, 25 mM HEPES pH 7.4 at 4°C, with one change of bufer each hour. The cassete was then transfered to 25 mM HEPES, pH 7.4 dialysis bufer for three hours with one change of bufer each hour. The protein was concentrated in a Centricon-YMIO (Milipore, Bedford, MA) and the concentration of the protein was measured in a spectrophotometer at 280 nm. The concentration of apo-p97 was calculated based on the extinction coeficient of 1.214 (1% p97 solution) at 280 nm (Baker et al, 1992). In order to ensure that the soluble p97 is in the apo-form, the dialyzed sample was electrophoresed in 6M urea polyacrylamide gel before metal loading experiment. Metal nitrilotriacetate (NTA) was prepared by mixing metal chloride with nitrilotriacetic acid-disodium salt (Sigma Chemical Co.) at ratio of 1:100. 1 nmoles of metal-NTA mixture was added to 0.1 nmoles of apo-p97. The mixture was neutralized with the additon of 60 sodium bicarbonate stock solution (10 mM) to pH 7.4. The mixture was alowed to incubate at room temperature for at least one hour before loading onto a 6M urea polyacrylamide gel. IV. Urea polyacrylamide gel electrophoresis The method of urea gel electrophoresis has been shown to resolve the four diferent species expected from partialy iron-saturated transferrin. This method was first described by Makey and Seal (Makey and Seal, 1976). Briefly, a 6 % (w/v) polyacrylamide running gel was cast in a bufer of 0.089 M Tris-Cl, 0.089 M borate (pH 8.4) containing 6 M urea. The stacking gel was a 5% (w/v) polyacrylamide gel (Biorad) cast in 0.07 M Tris-Cl (pH 6.8) and 12% (w/v) sucrose. The running bufer used was 0.09 M Tris, 0.09M Borate and 0.0016 M EDTA pH 8.4. A stock solution of 10 M urea (Sigma Chemical Co) deionized in the AG 501-X8 Mixed Bed Resin (Biorad, Hercules, CA) and stored cold until used. One nanomole of the metal loaded p97 was prepared in 6X sample bufer (10% (w/v) sucrose (Fisher Scientific Limited), 0.002% (w/v) bromophenol blue (Fisher Scientific Limited), 0.09 M Tris (Fisher Scientific Limited), 0.09 M Borate (Fisher Scientific Limited) and 0.016 M EDTA, pH 8.4) and loaded onto the gel. A gradient of 3.5 V/cm was employed for electrophoresis at 4°C for 16-17 hours. The gel was fixed in 30% (v/v) methanol (Fisher Scientific Limited) and 10% (v/v) acetic acid (Fisher Scientific Limited) before staining with Coomassie Blue. The' relative amount of each protein band is quantifed by using densitometry tool provided in the Chemilmager 4000 Low light Imaging System (Alpha Innotech Corporation, San Leandro, CA). 61 V. UV-Visible spectroscopy study of iron binding to soluble p97 Soluble apo-p97 was dialyzed into 25 mM Tris-Cl, 10 mM NaHC03 and 100 mM NaCl pH 7.5. In order to determine the absorbance wavelength of iron saturated p97, 1 ml of soluble apo-p97 at 2 mg/ml, (equivalent to 20.53 nmoles of apo-p97), was transfered into the spectrophotometer cuvete. Iron nitrilotriacetate (Fe-NTA) complex was titrated into the protein solution at 4 nmole increments. After each additon of Fe-NTA, the sample was subjected to a scan in Cary 50 UV-Visible spectrophotometer (Varian Inc., San Diego, CA) from 380 nm to 600 nm in order to detect spectral changes in the absorbance spectrum. To obtain the absorbance profile of p97 binding to iron, 4 nmole increments of iron were titrated into the protein solution. The absorbance at 475 nm was determined and ploted against the amount of iron added As a control, 1ml of 4 mg/ml apo-transferin was titrated with 4 nmoles increment of FeNTA. The absorbance at 465 nm was determined and ploted against the amount of iron added. VI. Differential scanning calorimetry The diferential scanning calorimetry (DSC) studies were performed on a MicroCal VP-DSC (MicroCal Inc., Northampton, MA). All experiments were caried out in 500 mM HEPES, 25 mM NaHC03, pH 7.5 at a scan rate of 60 °C /hr. The sample preparation protocol was obtained from Lin et al (Lin et al., 1993). Samples were degassed for 7 minutes with gentle stirring under vacuum prior to loading into the calorimeter. Protein concentrations were determined spectrophotometricaly (280 nm) 62 using an extinction coeficient of 115,004 M" cm" . Samples containing iron were prepared by mixing appropriate amounts of protein, bufer and Fe-NTA (iron nitrilotriacetate) stock solution. The Fe-NTA stock solution was prepared by mixing 1 volume of 0.3 M FeCl3 (in water) with 2 volumes of 0.3 M NTA, disodium salt (in water) and diluting to the appropriate concentration with bufer. VII. Immunoprecipitation of ferritin. TRVb, p97+TRVb and TfR+TRVb cel lines were separately were grown on 60 mm xl5 mm tissue culture dish to 100% confluency. The cels were washed twice, 30 minutes each time in Iscove's modifed Dulbecco media (IMDM) (Invitrogen Life Technologies Inc.) at 37°C to remove endogenous iron and serum transferrin. They were treated either with or without 300 mU of phosphoinositol-phospholipase C (PI-PLC) (Roche, Basel, Switzerland) in IMDM for 30 minutes at 37°C. PI-PLC is an enzyme that removes GPI-anchor from a subset of GPI-anchored protein, which includes GPI-anchored p97. The cels were then incubated with either 1.5 ml of 1 |iM of 5Fe nitrilotriacetate (NTA) or 2 mg/ml 5FeNTA loaded human transferrin (Sigma-Aldrich Canada Ltd) for 90 minutes at 37°C in the presence of the PI-PLC in a 5% C02 atmosphere. The cels were washed twice with ice cold PBS before lysing them in 1 ml of 1% (v/v) Nonidet-P40 (NP-40)(ICN Pharmaceuticals Inc, Costa Mesa, CA) in lysis bufer (20 mM Tris-Cl pH 7.4, 150 mM NaCl, 2 mM EDTA and 1 tablet of protease inhibitor cocktail (Roche) for 30 minutes on ice. The lysate was colected and centrifuged in Biofuge B (VWR Canlab Mississauga Ont) at 11,000 rpm at 4°C for 10 minutes. The supernatant was colected and pre-cleared at 4°C with 3 |il of normal rabbit 63 serum for 1 hour folowed by another hour with 30 |il of protein G sepharose (50% (v/v) slurry) (Amersham-Pharmacia Biotech, Piscataway, NJ). The lysate (150 ul) was aliquoted into eppendorf tubes and immunoprecipitated with antibodies against human p97 (L235), human transferrin receptor (OKT9) and human ferritin. Protein G sepharose (30 u,l) was used to precipitate the immuno-complexes. The protein G-immuno-complexes were washed twice in bufer B (0.2% (w/v) NP-40, 10 mM Tris-Cl pH 7.5, 150 mM NaCl, 2 mM EDTA), once in bufer C (0.2% (w/v) NP-40, 10 mM Tris-Cl pH 7.5, 500 mM NaCl, 2 mM EDTA) and once in bufer D (10 mM Tris-Cl pH 7.5) before resuspension in 1 ml of PBS. The PBS suspensions were transfered to scintilation vials and 5 ml of ReadySafe scintilation counting fluid (Beckman Coulter Inc., Fulerton, CA) were added. The radioactive counts associated with the immunoprecipitate were using the Beckman LS6000IS Liquid Scintilation counter. VIII. Endocytosis inhibition experiment p97+TRVb, TfR+TRVb and untransfected TRVb cel lines were grown on 35 mm x 10 mm tissue culture plates until 100% confluent. TRVb is a CHO cel line, which has a defective endogenous hamster transferrin receptor. The cels were pre-incubated with the 250 u,g/ml nystatin (Sigma-Aldrich Canada Ltd) and 25 p,g/ml filipin (Sigma-Aldrich Canada Ltd) both prepared in di-methyl-sulfoxide (DMSO) (Sigma-Aldrich Canada Ltd) for 30 minutes before lpM of 5Fe-citrate or 1 mg/ml of 5Fe-transferrin was added to the cels and incubated for 2 hours at 37°C. As a control, the cels were also treated with an equal amount of DMSO as present in the inhibitors. The cels were washed with 1 ml of PBS and harvested with a cel scraper in the presence of 1 ml Versene (2 mM EDTA in 64 PBS) and transfered to scintilation vials. The tissue culture plate was rinsed with 0.5 ml of PBS and the rinse was added into the scintilation vial. Five mililiters of scintilation counting fluid were added and the radioactivity count associated with the immunoprecipitate was determined in Beckman LS6000IS Liquid Scintilation counter. The experiments were performed in triplicate. The radioactive counts associated with each samples were averaged. The radioactivity associated with the TRVb samples was subtracted from the p97+ TRVb and TfR+TRVb samples. IX. Immunofluorescence staining and confocal laser scanning microscopy SK-MEL 28 cels were grown on sterile No.l 18 mm square coverslips (Fisher Scientific Inc.) to 70% confluency. The coverslip was washed twice in PBS before blocking it with 2% (w/v) BSA/ PBS for 1 hour at 4°C. The coverslip was stained with 200 ul of anti-p97 (L235) or anti-transferrin receptor (OKT9) antibodies at 1: 500 dilution in 2% (w/v) BSA/ PBS for 30 minutes on ice. Excess antibodies were washed of five times for 10 minutes with 2% (w/v) BSA/ PBS before transfering the coverslips to pre-warmed 2 % (w/v) BSA/ PBS and incubating them at 37°C for 0, 10, 20 and 30 minutes to alow endocytosis to occur. After the incubation period, the cels were fixed in 2 % (w/v) paraformaldehyde (BDH Laboratory Supplies, Dorset, England) in PBS for 20 minutes and permeabilzed in 0.1% (w/v) saponin in 2% (w/v) BSA/ PBS for 10 minutes at room temperature. A second antibody was used to stain for the various vesicles. Rabbit anti-caveolin antibody (250 (ig/ml) was diluted at 1: 50, goat anti-clathrin antibody (200 p.g/ml) was used at 1: 40, and antibody against Early endosome antigen 1 (EEA1) (200 fig/ml) was used at 1: 40. All antibodies were diluted in 2% (w/v) 65 BSA/ PBS. The cels were stained for 30 minutes at room temperature before five washes for 10 minutes with 0.1% (w/v) saponin, 2% (w/v) BSA/ PBS. Alexa 488, Alexa 568 or Cy5 conjugated secondary antibodies were used to visualize the staining patern (Table Illb). The secondary antibodies are diluted in 2% (w/v) BSA/ PBS. Two hundred microliters of diluted secondary antibodies were used to stain the cels at room temperature for 20 minutes, folowed by five washes for 10 minutes with 0.1% (w/v) saponin, 2% (w/v) BSA/ PBS. The coverslip was mounted with 5 pi of Slow Fade (Molecular Probes, Eugene, OR) before sealing it onto a glass slide with nail polish. For triple antibody labeling, SK-MEL 28 cels grown on coverslips were washed three times 15 minutes in IMDM at 37°C in order to remove serum transferrin. The coverslips were blocked with 2% (w/v) BSA/ PBS for 1 hour on ice and stained with 200 pi of 5 p.g/ml Alexa 488 conjugated transferrin and 200 pi of anti-p97 antibody (L235) diluted at 1:500 for 30 minutes on ice. Excess antibody and Alexa 488 conjugated transferrin were washed of with PBS and the cels were alowed to undergo endocytosis for 0, 10, 20, 30 minutes in 2% (w/v) BSA/ PBS pre-warmed to 37°C. The cels were fixed in 2% (w/v) paraformaldehyde for 15 minutes at room temperature folowed by a 15-minute permeabilzation step in 0.1% (w/v) saponin at room temperature. Goat anti human EEA-1 antibody was used to stain the endosomal marker for 30 minutes folowed by five washes in PBS for a total of 10 minutes. Alexa 568 and Cy-5 conjugated secondary antibodies were used to visualize the staining patern (Table Illb). The secondary antibodies are diluted in 2% (w/v) BSA/ PBS. Excess secondary antibody was removed and the coverslip was mounted onto a glass slide with Slow-Fade. The BioRad 66 Radiance Plus confocal laser scanning microscope was used to capture al the fluorescent images and NIH imaging system was used to analyze the images. X. Cryoimmunoelectron microscopy Two 140 mm x 20 mm plates of SK-MEL 28 cels were grown to 90% confluency and washed twice with 10 ml of pre-warmed (37°C) PBS before fixing in 2% (w/v) paraformaldehyde at 37°C for 30 minutes. The fixed cels were washed twice with PBS and scraped of gently in 1 ml of PBS with a rubber policeman and centrifuged at maximum speed in the Biofuge B centrifuge. The pelet was infused overnight with 25% (w/v) polypyrolidone (Sigma-Aldrich Canada Ltd) in 2.3 M sucrose at 4°C. The next day, the pelet was trimmed with a razor blade and a smal pyramid shaped block was mounted on the aluminum specimen pin (EM Sciences, Hawthorne, NY) with a drop of O.C.T compound (Tissue-Tek, Elkhart, IN). The specimen pin was quickly frozen in liquid nitrogen. Eighty-nanometer sections were cut with a glass knife at -110 °C and sections were picked up with 2.3M sucrose onto a formvar-coated 200 meshed nickel grid (EM Sciences). Before blocking the grid with 0.5% (v/v) fish skin gelatin (Sigma-Aldrich Canada Ltd) in PBS for 10 minutes, the grid was washed in PBS four times within 10 minutes. Rabbit anti-caveolin antibodies (50 pJ) at 1:100 dilution, L235 (50 at 1:100 dilution, goat anti-clathrin antibody (50 ul) at 1: 100 dilution and OKT-9 (50 pJ) at 1:100 dilution were used to stain the grids. All the antibodies were diluted in 0.5% (v/v) fish skin gelatin in PBS. The grids were stained with primary antibody for 20 minutes at room temperature and washed four times in PBS before staining with secondary antibody conjugated to gold. Goat anti-rabbit conjugated to 10 nm gold (50 67 pi), goat anti-mouse conjugated to 5 nm gold (50 |il) and donkey anti-goat conjugated to 10 nm gold (50 ul) antibodies were used at 1:50 dilution in 0.5% (v/v) fish skin gelatin. After staining for 20 minutes in secondary antibody, the grids were rinsed 4 times in PBS and washed 4 times for 15 minutes in PBS folowed by 6 rinses in distiled water for 20 minutes. The grid was then sealed and counter-stained with 2% (w/v) aqueous uranyl acetate (EM Sciences Inc.) in methyl celulose (Sigma-Aldrich Canada Ltd). Excess liquid was bloted of with Whatman filter paper no. 1 (VWR Canlab) and the grid was air-dried before examining it under the Zeiss EM 10C transmission electron microscope at 80 kV (Zeiss, Esslingen, Germany). XI. Sucrose gradient sub-cellular fractionation and western blotting Three confluent 140 mm x 20 mm plates of SK MEL-28 cels with a total of 3 x 107 cels were washed two times with cold PBS. Cels were scraped with a rubber policeman in 5 ml of PBS and centrifuged in Beckman GPR centrifuge (Beckman Coulter Inc.) for 5 minutes at 4°C at 1,200 rpm. The pelet was lysed on ice in 1 ml 1% (v/v) Triton X-100 in 140 mM KC1, 10 mM Tris-Cl, pH 7.5 for 30 minutes. One mililiter of 80% (w/v) sucrose in 140 mM KC1, 10 mM Tris-Cl, pH 7.5 was added to the lysate to make a 40% (w/v) sucrose lysate mixture. The lysate mixture was transfered into a 10 ml polypropylene tube and overlaid with 6 ml of cold 30% (w/v) sucrose in 140 mM KC1, 10 mM Tris-Cl, pH 7.5, folowed by 3.5 ml of 5% (w/v) sucrose in 140 mM KC1, 10 mM Tris-Cl, pH 7.5. This step gradient was then centrifuged in Sorval RC80 ultracentrifuge with a Sorval TH-641 rotor (DuPont Canada Inc., Mississauga ON), at 39,000 rpm for 16 hours at 4°C. A white flocculent band was observed at the interface 68 between the 5% (w/v) and 30% (w/v) sucrose gradient. Eight fractions of 1.5 ml were colected and the pelet was resuspended in lysis bufer. Each fraction (20 u\l) was mixed with 10 |il of 3X reducing sample bufer (250 mM Tris-Cl, 3% (v/v) glycerol, 0.16% (w/v) SDS, 0.002% (w/v) bromophenol blue and 1% (w/v) dithiothreitol (DTT) (Roche). The mixture was heated at 95°C for 5 minutes and alowed to cool down to room temperature before 5 (xl of 0.5 M iodoacetamide (Sigma-Aldrich Canada Ltd) was added to it. The sample was then loaded onto a 10% (w/v) SDS-PAGE and electrophoresed at 150 V for 1 hour. The gel was transfered to Immobilon membranes (Biorad) by electrobloting. The membrane was blocked with 5% skim milk, 0.1% (v/v) Tween-20 (Biorad) in PBS for 2 hours and detected using antibodies against caveolin (1: 100 dilution), p97 (1: 100 dilution), transferrin receptor (1: 100 dilution) and (3-adaptin (1: 100 dilution) in 5% (w/v) skim milk, 0.1% (v/v) Tween-20 in PBS overnight at 4°C. After 4 washes of 30 minutes each in 5% (v/v) skim milk 0.1% (v/v) Tween-20 in PBS, peroxidase conjugated goat anti-mouse and peroxidase conjugated goat anti-rabbit antibodies (1:10,000 dilution) in 5% (w/v) skim milk, 0.1% (v/v) Tween-20 (Biorad) in PBS were used to detect the primary antibodies. The chemiluminescence ECL Western bloting detection system (Amersham Pharmacia Biotech) was used to detected the chemiluminescence. In order to determine whether iron transported by GPI-anchored p97 fractionates into the caveolae fraction. SK-MEL 28 cels were labeled with 1.5 ml of 1 fiM of iron-55 nitrilotriacetate (5Fe-NTA) for 15 minutes at 37°C in a 5% C02 atmosphere before being subjected to sub-celular fractionation. P97 was immunoprecipitated from each fraction. The radioactivity associated with the immunoprecipitate complex was determined using a 69 Beckman LS6000IS Liquid Scintilation counter. The amount of radioactivity represents radioactivity associated with GPI-anchored p97 after subtraction of radioactivity associated with normal rabbit serum (NRS). XII. Transfection of Nramp2 into SK-MEL 28 cells SK-MEL 28 cels were grown to 50% confluency on sterilzed No.l 18mm square glass coverslips. 2 pg of pEGFP-hNramp2 (human Nramp2 subcloned into Enhanced green fluorescence protein vector from Clontech (Clontech, Palo Alto, CA) (gift from Dr. F. Kishi, Center for Gene Research, Yamaguchi University, Ube, Yamaguchi 755-8505 Japan) was transfected into SK-MEL 28 cels using Fugene 6 (Roche) folowing the protocol supplied by the manufacturer. The cels were stained for GPI-anchored p97 or transferrin receptor 24-hour after transfection folowing the immunofluorescence protocol. EGFP was observed directly. 70 Chapter 3: Metal binding properties of p97 I. Other metals can compete with iron for p97 A. Rationale The best-characterized pathway of iron uptake is the transferrin receptor mediated endocytosis of transferrin. However, the in situ distributions of iron and transferrin in the brain are not in the identical regions (Dwork et al., 1988; Hill and Switzer, 1984). Therefore, there is debate on whether transferrin mediates the transport of metals into the brain. Metals such as iron, copper, aluminum and zinc are known to play roles in the pathology of several neurodegenerative disorders such as Alzheimer s and Parkinson s diseases (Bush, 2000; Bush et al., 1994; Perl and Pendlebury, 1986; Sayre et al., 2000). It is yet unknown how these metals are transported into the brain, however, one possible candidate for metal transport into the brain is p97. The level of soluble p97 in the serum and cerebrospinal fluid is elevated in Alzheimer s disease patients (Kennard et al., 1996; Kim et al., 2001). In addition, soluble p97 is also able to cross the blood-brain barrier into the brain more eficiently than transferrin (Moroo et al., 2001). One hypothesis suggests p97 is able to mediate the transport of these metals into the brain. Before establishing p97 as a candidate for metal transport into the brain, it was crucial to determine whether these metals could bind to p97. It is known that metals such as aluminum, copper, zinc and others are able to bind to the same iron-binding site as iron in transferrin. (Congiu-Castelano et al., 1997; Harington et al., 1987; Harris, 1983; Harris and Stenback, 1988; Hutchens and Yip, 1991; Roskams and Connor, 1990). Therefore, due to the high protein sequence homology between transferrin and p97, p97 may also bind these other metals. In addition, there is a consensus zinc binding sequence 71 in between the N-lobe and the C-lobe of p97 (Garat and Jhoti, 1992). It is not known whether zinc could bind to p97 at this motif or whether zinc can bind to the iron-binding site. Furthermore, the close proximity of the zinc-binding site to the two iron-binding sites can afect the binding kinetics of iron to GPI-anchored p97. A recent review suggested that p97 might be a zinc-dependent metaloprotease (Sekyere and Richardson, 2000). However, the protease activity has not been demonstrated. Two sets of experiments to examine whether copper, aluminum and zinc could bind to p97 were carried out. The competion experiment examined whether these metals wil bind to the GPI-anchored p97 and the urea polyacrylamide gel electrophoresis analyzed whether these metals could bind to soluble p97. B. Result I: Aluminum, copper and zinc can compete with iron for GPI-anchored p97 Radioactive copper, zinc and aluminum emit very high-energy (3-particles and are dangerous to work with. An indirect approach to examine whether these metals bind to GPI-anchored p97 is to determine whether they can compete for the iron-binding site of GPI-anchored p97 with radioactive iron. Transferin receptor deficient Chinese hamster ovary cels transfected with or without p97 (p97+ TRVb and p97TRVb) were incubated with varying concentrations of 5Fe-citrate in the presence or absence of 1 uM of cold metal citrate complexes at 37°C for four hours. After the incubation, the cels were washed and the radioactivity associated with p97" and p97+ transfected TRVb cel lines were determined and the diference in radioactivity between the supernatants of both cel 72 lines was ploted (Fig. 5). Therefore, the amount of "Fe-citrate ploted represents "Fe-citrate bound to GPI-anchored p97. Figure 5 shows the amount of 5Fe-citrate bound to GPI-anchored p97 at 37°C in the absence of cold Fe-citrate is approximately 4.2 picomoles per milion cels when the cels were incubated with 2 uM or more 5Fe-citrate. However, the additon of luM of cold iron citrate, copper citrate or aluminum citrate reduced the amount of 5Fe-citrate bound to GPI-anchored p97 to 2.2, 2.5 and 2.4 picomoles per milion cels, respectively after 4 hours (Fig 5A, 5B, 5C). This indicates that cold iron, copper and aluminum compete with 5Fe-citrate for the iron-binding site in GPI-anchored p97. Interestingly, the binding of 5Fe-citrate to GPI-anchored p97 in the presence of 1 pM zinc-citrate was dramaticaly diferent from that of copper citrate and aluminum citrate. In this case, the total amount of 5Fe-citrate bound to GPI-anchored p97 was not afected by the presence of lpM zinc citrate. GPI-anchored p97 was able to bind increasing amount of 5Fe-citrate per milion cels in the presence of lp.M of zinc-citrate, however, higher concentrations of 5Fe-citrate were required for the same number of cels to bind the same amount of 5Fe-citrate (Fig. 5D). This indicates that although the presence of zinc does not afect the overal amount of 5Fe-citrate bound to GPI-anchored p97, the presence of zinc citrate afects the mode of 5Fe-citrate binding to p97. These competion experiments show that copper and aluminum competed for the iron-binding site in GPI-anchored p97, whereas zinc did not afect the total amount of iron bound to p97. 73 Figure 5: The efects of cold metal citrate competiors on the amount of Fe-citrate bound to GPI-anchored p97. (A) Cold Fe-citrate (B) cold Al-citrate and (C) cold Cu-citrate reduces the amount of 5Fe-citrate bound to GPI-anchored p97. (D) Cold Zn-citrate reduces the amount of 5Fe-citrate binding to GPI-anchored p97 at low 5Fe-citrate concentrations but does not reduce the amount of 5Fe-citrate bound to GPI-anchored p97 at high 5Fe-citrate concentrations. The eror bars represent the standard deviations of three samples. 74 75 C. Results II: Soluble p97 can bind iron, copper and zinc Several investigators have shown that partialy iron-saturated human serum transferrin can be resolved into four discrete bands coresponding to: apo-transferrin, C-lobe saturated transferrin, N-lobe saturated transferrin and holo-transferrin, in the order of increasing mobilty when the sample is subjected to 6 M urea polyacrylamide gel electrophoresis (Funk et al., 1990; Makey and Seal, 1976; Penhalow et al., 1991). The presence of 6M urea causes the lobes that do not contain iron to unfold causing diferent mobilty during electrophoresis (Evans and Wilams, 1978). The molecular basis for the electrophoretic separation appears to be related to two observable facts: (1) the increase in protein stabilty as iron is bound to transferrin and (2) the asymmetrical distribution of amino acids between the two lobes of the transferrin molecules (Evans and Wilams, 1978; Makey and Seal, 1976). Since p97 is structuraly similar to transferrin, this technique was applied to examine whether soluble p97 loaded with diferent metals exhibits an altered mobilty on the urea-polyacrylamide gel. As a control, partialy and fuly saturated human serum transferrins were subjected to 6 M urea polyacrylamide gel electrophoresis. Lane 1 in Figure 6A shows that partialy saturated human transferrin can be separated by mobilty into four distinct bands as mentioned in previous observations by Makey and Seal. Fuly saturated human transferrin exists as a single band (Fig. 6A lane 2). Apo-soluble p97 was incubated with excess amount of iron nitrilotriacetate (Fe-NTA), aluminum nitrilotriacetate (Al-NTA), copper nitrilotriacetate (Cu-NTA), or zinc nitrilotriacetate (Zn-NTA). The protein was electrophoresed onto a 6% urea-76 Figure 6: Urea polyacrylamide gels of human serum transferrin and human soluble p97. (A) Partialy iron-saturated (lane 1) and fuly iron-saturated human serum transferrin (lane 2) and (B) dialyzed p97 (lane 1), p97 loaded with either aluminum nitrilotriacetate (Al-NTA) (lane 2), copper nitrilotriacetate (Cu-NTA) (lane 3), iron nitrilotriacetate (Fe-NTA) (lane 4), or zinc nitrilotriacetate (Zn-NTA) (lane 5) and undialyzed p97 (lane 6). This is a representative urea polyacrylamide gel from two separate experiments. 77 po-Tf terminal loaded with iron N-terminal loaded with iron Iron saturated Tf Lane 1 r< TA TA 1A TA TA z 1 z z CD cz Q LL N + + + + o> CT> a> CD 03 C L C L C L C L C L CD CD CD CD CD - Q _Q -C! X 3 =3 =S =3 =3 03 O o O o O CO CO co CO CO T3 TD " D X3 "D CD CD CD CD CD N N N N N >. >> >» >. CO CO CO CO CO T3 T3 X3 -o X5 a > C L _CD jQ _g o CO " D CD N •>. CO X 3 C Apo-so lub le p97 Lobe 1 loaded with metal Lobe 2 loaded with metal Lane 1 78 polyacrylamide gel and the gel was fixed with methanol and stained with Coomassie blue. The relative amount of each band of protein in each lane was quantiated using 1D-Multi (line densitometry) (Fig. 6B). Table IV shows relative amount of protein in each lane coresponding to the protein bands in Figure 6B. Lane 6 represents the protein before dialysis presented as two diferent forms of soluble p97. The upper band represents approximately 60% of the protein and the lower band represents 40% of the protein. After extensive dialysis, only the upper band of the protein exists, as shown in lane 1. Therefore, the upper band seen in lane 6 is the apo-form of soluble p97 whereas the lower band is soluble p97 bound to iron. Lane 4 shows that when the soluble apo-p97 is saturated with iron, a large fraction of the protein (85%) shifted to the lower band and a smal fraction of the protein (15%) shifted to an even lower band. The identiy of these bands wil be discussed later. It is shown in lane 2 that when apo-p97 is loaded with Al-NTA, there is very minimal shift of protein to the lower band. It is shown in lane 3 that soluble apo-p97 binding to Cu-NTA. A slightly stronger binding is observed for Cu-NTA compared with Al-NTA, approximately 29.8 % of the protein shifts to the lower band while 70.2 % of the protein remains in the apo-form. The protein distribution in lane 5 shows that p97 binds to moderate amounts of Zn-NTA. Fifty percent of the protein shifts to the lower band whilst the other 50% remains in the apo-form. In summary, although metals such as copper and zinc show a degree of binding to soluble p97, the binding of iron to the protein is prefered. Aluminum did not appear to bind to soluble p97 whereas zinc is able to bind to soluble p97 as observed in the urea-polyacrylamide gel. 79 Table IV: Relative amount of metal loaded soluble p97 in the urea polyacrylamide gel Lane 1 Lane 2 Lane 3 Lane 4 Lane 5 Lane 6 Metal - Al Cu Fe Zn -Band 1 100% 100% 70.2% - 42.7% 60.2% Band 2 0% 0% 29.8% 85.1% 57.3% 39.8% Band 3 - - - 14.9% - -80 D. Discussion One goal of this thesis is to understand the metal binding properties of p97. The GPI-anchored form has been shown to bind iron and internalize it into the cel (Kennard et al., 1995). The function of the soluble form, however, is not yet established. It is known that the level of soluble p97 is higher in the serum of Alzheimer s disease patients compared to age matched control group (Kennard et al., 1996). Therefore, questions arise as to the role of soluble p97 in Alzheimer s disease. There is increasing evidence to show that metals are important factors in the pathology of several neurological disorders. Iron, zinc, copper and aluminum have al been implicated in the disease progression of Alzheimer s disease (Bush, 2000; Bush et al., 1994; Lynch et al., 2000; Miura et al., 2000; Moir et al., 1999). It is not known how these metals are transported from the serum across the blood-brain barrier into the brain. Initialy, it was thought that transferrin could mediate the transport of these metals into the brain since it has been shown that transferrin can also bind metals other than iron (Congiu-Castelano et al., 1997; Harington et al., 1987; Harris, 1983; Haris and Stenback, 1988; Hutchens and Yip, 1991; Roskams and Connor, 1990), However, based On accumulated evidence, it is highly unlikely that transferrin and transferrin receptor are the only factor that governs the transport of metals across the blood-brain barrier. This prospect is supported by the folowing observations: (1) There is little overlap between the distribution of transferrin receptor and iron in the normal, Alzheimer s disease and Parkinson s disease brains (Dwork et al., 1988; Hill and Switzer, 1984; Schipper, 1999). (2) The concentration of iron exceeds the iron-binding capacity of transferrin in the cerebral spinal fluid. 81 Therefore, there may be other candidates involved in the transport of iron (Beard et al., 1993). (3) Hpx mice have higher than normal iron uptake in the brain (Trenor et al., 2000). Therefore, an alternate route of iron uptake independent of transferrin must exist in these mice since non-transferin bound iron seems to be able to cross the blood-brain barier in these mice (Ueda et al., 1993). p97 is one such candidate. The tissue distribution studies of p97 indicate that it is expressed by the human brain capilary endothelium (Rothenberger et al., 1996) and is produced in reactive microglial associated with senile plaque in Alzheimer s disease (Yamada et al., 1999). In addition, there is increasing evidence suggesting that metals such as iron, copper, zinc and aluminum play a role in precipitating the amyloid plaques in Alzheimer s disease (Bush, 2000; Bush et al., 1994). Therefore, before examining whether p97 can mediate transport of these metals into the brain in vivo, a series of in vitro competion assay between iron and other metals for GPI-anchored p97 and metal loading of soluble p97 were caried out in order to determine whether p97 can bind these metals in vitro. The competion experiment show that copper and aluminum competed for the iron-binding site in GPI-anchored p97, whereas zinc did not afect the total amount of iron bound to p97. Urea polyacrylamide gel electrophoresis of soluble p97 loaded with diferent metals shows that although copper and zinc are able to bind to soluble p97, the binding of iron to the protein is preferred. Aluminum, however, did not appear to bind soluble p97. a. Copper, aluminum and zinc can compete with iron for binding to GPI-anchored p97 82 The competion experiment between 5Fe-citrate and cold metal citrate demonstrated that the presence of copper-citrate and aluminum-citrate decreased the total amount of 5Fe-citrate internalized by GPI-anchored p97 (Fig. 5). This suggests that copper-citrate and aluminum-citrate can compete with 5Fe-citrate for binding to GPI-anchored p97. Crystalographic studies of human copper-lactoferin have shown that the metal binding site for copper is the same as that for iron (Smith et al., 1992). Since the iron-binding site in the N-lobe of human p97 is conserved, it is possible that iron and copper have the same binding site in human p97. Thus, copper is able to compete with iron for p97 and hence the decrease in 5Fe citrate binding to GPI-anchored p97 in the presence of cold copper-citrate. The interaction of aluminum with human serum transferrin and lactoferrin has been examined using X-ray absorption near edge structure (XANES) spectroscopy, circular dichroism, UV-visible, and fluorescence spectrometry (Congiu-Castelano et al., 1997; Tang et al., 1995). These studies ilustrate that aluminum can bind to human serum transferrin and lactoferrin. Figure 5 shows that aluminum-citrate competes with 5Fe-citrate binding to p97 suggesting p97 is able to bind to aluminum. The presence of zinc-citrate also afects the binding of 5Fe-citrate to GPI-anchored p97. Interestingly, the competion profile of zinc-citrate is diferent from the other three metal-citrate complexes. Higher concentrations of 5Fe-citrate are needed for GPI-anchored p97 to bind the same amount of 5Fe-citrate in the presence of zinc-citrate compared to that when zinc-citrate is absent. The total amount of 5Fe-citrate bound to GPI-anchored p97 was not afected by the presence of lpM of zinc-citrate (Fig. 5D). This indicates that zinc may have a diferent binding site in GPI-anchored p97 or that 83 zinc may have a weak binding afinity for p97. An elegant molecular modeling study has identified a potential zinc-binding consensus sequence (HEXXH) present between the Isl-and the C- lobes of p97 (Garat and Jhoti, 1992) (Fig. 2). Therefore, it is possible that zinc does not compete with iron for the binding sites directly but that zinc may afect the binding of iron to GPI-anchored p97 by causing some stereo-spatial distortion due to the close proximity of the zinc-binding site to the two iron-binding sites. The molecular modeling also suggests that the stereochemical orientation of the zinc-binding site is super-imposable with the zinc-binding site of thermolysin. This observation led to a proposal that p97 might have zinc dependent metaloprotease activity (Garat and Jhoti, 1992). However, only the human orthologue of p97 contains this zinc binding consensus sequence. Therefore, it is possible that human p97 has maintained the zinc binding sequence for metaloprotease activity while p97s from other species have somehow lost this function. Experiments to determine whether p97 can function as a metaloprotease are in progress. b. The binding of iron, copper, aluminum and zinc to soluble p97 The availabilty of soluble p97 presented an opportunity to use other methods to study whether metals other than iron, bind to p97. Several investigators have shown that partialy saturated transferrin can be diferentiated into four diferent forms by electrophoresis in a medium containing 6M urea and a Tris/ borate/ EDTA, pH 8.4 bufer (Evans and Wiliams, 1978; Makey and Seal, 1976). This is based on evidence from peptide mapping of transferrin that indicates transferrin is not symmetrical with respect to amino acid compositon between the two lobes and the increase in structural stabilty as 84 the protein binds to iron. The four diferent forms are apo-transferrin, C-lobe saturated transferrin, N-lobe saturated transferrin and holo-transferrin and they show increased mobilty on a 6 M urea polyacrylamide gel as depicted in Figure 6A and Figure 7 (Funk et al., 1990; Makey and Seal, 1976; Penhalow et al., 1991). A similar profile is expected if metals such as iron, copper, aluminum and zinc can bind p97. It is shown in lane 4 of Figure 6 shows that in the presence of iron, majority of the soluble p97 (85%) exists as slower migrating upper band. A smaler but significant fraction of the protein (15%) exists as a faster migrating lower band. There are three possible interpretations as to the identity of these two bands (Fig. 7). In comparison to the transferrin profile, one possibilty is that the major band seen in Figure 6 lane 4 coresponds to the iron bound C-lobe and the minor band coresponds to iron bound to the N-lobe (situation 1 in Fig. 7). However, this is unlikely since the iron-binding site in the N-lobe is conserved in both p97 and transferins but the iron-binding site in the C-lobe is not. Therefore, the upper major band can be assumed to be the iron bound N-lobe and the lower minor band coresponds to the iron bound C-lobe (situation 2 in Fig. 7). There is no additonal band of higher mobilty than the minor band, indicating the absence of the fuly saturated soluble p97. This data would suggest that the C-lobe of soluble p97 is able to bind to iron. Soluble apo-p97 was loaded with excess amount of Fe-NTA before electrophoresis on a 6 M urea-polyacrylamide gel. However, the fact that iron loaded N-lobe was present at a greater proportion than iron loaded C-lobe and there was no fuly saturated soluble p97 suggests that the two lobes bind to iron with very diferent affinity. Perhaps the amino acid substiutions in the C-lobe have greatly afected its ability to bind iron. There is another possible interpretation for the presence of the two diferent bands (situation 3 85 CO CM I Q . CC CD JD O CD CD f> •Q 95 "T CD O u_ CD LL CD JD O 6 CD r> JD CD O Q-- T CD Z LL CD f> o °-• CD O U -I I CD £ • JD 9? • CD Z LL I I \— CD LL CD J Q O OS Q . I I O x I I I I T 3 C CO JD c 'CD o I "O c CO JD _ CO CO E ^ CD D) CD •g "E O CO >. o Q . i CO CD CO CD > O CO CD 1— CO CO r-Q . T 3 C — o JD C O CO c o » CO CD o. t_ CD -«—< _c <_ JD CO CO o Q . CD CD CD JZ CD — D) LL in Fig. 7). Data from the urea polyacrylamide gel electrophoresis of the partialy saturated transferrin suggested that the order of resolution of the diferent forms of transferrin is apo-transferrin, C-lobe loaded transferrin, N-lobe loaded transferrin and fuly saturated transferrin in the order of increasing mobilty. Therefore, an alternative explanation is that the upper major band is the N-lobe loaded with iron and the lower minor band is the fuly saturated p97. Under this interpretation, the band coresponds to iron bound C-lobe alone is not detected. Nevertheless, both scenarios 2 and 3 suggest that the major band is the N-lobe and that the C-lobe can bind iron, albeit weakly. Urea polyacrylamide gel electrophoresis also demonstrates that zinc and copper can bind to the soluble p97 whereas aluminum shows little or no binding to the protein (Fig. 6). The binding of Cu-NTA to soluble p97 as shown by the urea polyacrylamide gel (Fig. 6 lane 3) agrees wel with the data from the competion experiment showing the ability of copper-citrate to compete with 5Fe-citrate for GPI-anchored p97 (Fig. 5C). The competion data suggests that cold aluminum-citrate can interfere with the binding and uptake of 5Fe-citrate by GPI-anchored p97 (Fig. 5). However, the urea polyacrylamide gel electrophoresis data suggests that soluble p97 does not bind aluminum. One explanation for this discrepancy is that there is a significant increase in structural distortion around the metal site when aluminum tries to bind to it. The structural study of aluminum and transferrin shows that the aluminum ion is hexa-coordinated in an octahedral-ike symmetry in the iron coordination site and thus creates a contortion that results in unstable coordination between aluminum ion and transferrin (Congiu-Castelano et al., 1997). Therefore, in a competive binding assay, the presence of cold aluminum-citrate is enough to interfere with the binding of 5Fe-citrate to GPI-anchored 87 p97. However, in a urea gel assay, aluminum-citrate is probably not tightly bound to soluble p97 and therefore when the protein is electrophoresed in the urea gel, the aluminum ion probably detaches from the protein. The competion data demonstrate that zinc-citrate does not afect the total amount of 5Fe-citrate binding to GPI-anchored p97. One possible interpretation given is that zinc does not bind to the iron-binding site in p97. The urea polyacrylamide gel electrophoresis of Zn-NTA loaded soluble p97 shows that zinc is able to bind to soluble p97 and shifts approximately 50% of the apo-p97 to a higher mobilty form of zinc bound soluble p97. There are two possible interpretations of this shift in mobilty. It is possible that in the absence of competing iron atoms, zinc is able to bind to the iron bind site and therefore the lower band shown in Figure 6 lane 5 is zinc binding to N-lobe of p97. However, another interpretation is that Zn-NTA binds to a diferent site. The proposed zinc-binding site of p97 is located in between the N-and the C-lobes (Figs.l and 2). Zinc could bind to the proposed site of soluble p97 and shift the mobilty of the protein to the lower band. Furthermore, the absence of the apo-form of p97 when the protein is loaded with iron compared to the presence of the apo-form of p97 when it is loaded with copper, aluminum or zinc, suggests that p97 binds preferentialy to iron. II. The iron-binding stoichiometry of soluble p97 A. Rationale The members of the transferrin family share a conserved tertiary protein structure. The typical structure is a single chain glycoprotein, which is divided into two similar but 88 not quite identical lobes, the N-lobe and the C-lobe representing the N-terminal and C-terminal halves of the molecule. Each lobe, in turn, comprises two domains surounding a deep hydrophilc cleft enclosing an iron-binding site. The iron-binding site is made up of four conserved amino acid residues: an aspartic acid, two tyrosines and one histidine. They provide the coordination sites for the iron atom. A threonine and an arginine provide a network of electrostatic and hydrogen bonds for the synergistic anion. Sequence alignments of p97 and other members of the transferrin family show that the amino acids involved in coordinating the iron atom of the N-lobe are totaly conserved (Fig. 1). However, there are the four amino acid substiutions (D421S, D482S, T478 and T483P) in the iron and anion binding sites in the C-lobe of human p97. These substiutions have been proposed to afect the iron-binding ability of the C-lobe. A previous study by Baker et al showed that p97 bound iron at 1:1 ratio (Baker et al., 1992). However, the urea-polyacrylamide gel data in the previous section showed that there was a possibilty that the C-lobe of p97 can bind iron (Fig. 6). In addition, X-ray crystalography of the N-lobe D60S human lactoferrin, which mirors the D421S substiution in human p97, suggests that even when the aspartic acid is substiuted, the recombinant human lactoferrin is still able to bind iron (Faber et al., 1996). Furthermore, the binding constant of iron for p97 has never been determined. Therefore, UV-visible spectroscopy was performed on the soluble form of p97 to determine the stoichiometry of iron to p97. 89 B. Results I: Determination of the peak absorbance wavelength of iron saturated p97 p97 and transferrin bound with saturating amounts of iron have similar characteristic absorbance peaks between 460-475nm (Aisen, 1994). This absorbance profile can be used to study the iron-binding characteristics of p97 and to establish whether soluble p97 binds to one or two atoms of iron by observing the change in the absorbance at a specific wavelength as the protein is gradualy saturated with iron. Previously, Baker et al. reported the peak absorbance wavelength of naturaly occuring soluble p97 to be 464 nm (Baker et al., 1992). However, the soluble p97 used in this thesis was a recombinant p97 made by geneticaly deleting the GPI-anchored signal sequence. Therefore, the peak absorbance wavelength of the recombinant protein had to be ascertained by saturating soluble p97 with excess amounts of iron nitrilotriacetate (Fe-NTA). The iron-saturated protein, which has pinkish-orange color, was subjected to a UV-visible wavelength scan from 380 nm to 600 nm. A peak absorbance was observed at 475 nm when soluble p97 was saturated with iron whereas apo-p97 did not exhibit any peak absorbance during the scan (Fig 8). C. Results II: Iron-binding titration for soluble p97 After the peak absorbance of iron saturated soluble p97 was determined, the apo-soluble p97 was titrated with increasing amounts of Fe-NTA. The iron titration curve indicates that the increase in absorbance at 475 nm ceases at a ratio close to 2 moles of Fe-NTA per mole of soluble p97 (Fig. 9B). This is similar to iron titration profile of human serum transferrin for which the increase in absorbance also ceases at the same 90 0.10 0.08 H 400 450 500 550 Wavelength (nm) Figure 8: UV-Vis ib le absorpt ion spect roscopy of soluble human p97. Holo p97 absorbs at 475 nm whereas apo-p97does not have an absorbance peak at 475 nm. This is a representation of three separate exper iments. 91 ( A» 0 .25 . c 0 .204 2 5 0 [ F e - N T A ] (LIM) ( B) 0 .06- , [ F e - N T A ] (LIM) Figure 9: Iron nitrilotriacetate (Fe-NTA) titration of human serum transferrin and p97. (A) 46 moles of apo-human serum transferrin and (B) 20 moles of apo-human soluble p97 was gradually titrated with Fe-NTA. The absorbance of iron-protein complex was measured at 475 nm for human serum transferrin and at 465 nm for human soluble p97. 92 ratio (Fig. 9A). Therefore, the UV-visible spectrum shows that under iron saturating conditons, soluble p97 could bind to two atoms of iron. D. Discussion The four amino acid substiutions in the C-lobe of human p97 raise the question as to whether the C-lobe of the protein is able to bind iron. Therefore, UV-Visible spectroscopy study of iron binding to p97 was caried out. The absorption spectroscopy study reveals that soluble p97 has a peak absorption wavelength at 475 nm (Figure 8). This transition is due to the charge transfer from tyrosyl oxygen of the two tyrosines to bound ferric iron (Aisen, 1994) and gives holo-soluble p97 a characteristic orange-pink color in solution rather than the colorless appearance of apo-soluble p97. The Xmax previously reported for naturaly secreted soluble p97 is 465nm, which is a 10 nm shift in Timax compared to 475 nm (Baker et al., 1992). A possible explanation for the diference between the two Xmax is that the soluble p97 used in Baker et al. is a naturaly secreted protein from SK-MEL 28 cels, whereas the soluble p97 used in this study is a recombinant p97, produced by eliminating the GPI-anchored signal, and expressed the protein in baby hamster kidney cels (Yang, 2000). Iron nitrilotriacetate (FeNTA) titration of soluble p97 at 475 nm shows that two atoms of can be bound to soluble p97 (Figure 9). Thus, the interpretation that the minor lower band in lane 6 of Figure 6 can be p97 loaded with 2 atoms of iron would agree with the data from Figure 8. It is possible that the iron binds to the C-lobe of soluble p97 with low affinity. An interesting observation from the absorption spectra analysis of the D63S human serum transferrin (a mutation which miror the D421S in the C-lobe of p97) 93 shows that the recombinant protein is able to bind iron and results in a peak in the absorption spectra of the proteins (He et al., 1997). A possible explanation for this observation is that serine is able to form a hydrogen bond with a water molecule, which is able to provide the coordination for iron (Faber et al., 1996; He et al, 1997). However, the absence of the aspartic acid in the recombinant serum transferrin may result in a failure to lock in the iron-pocket in the C-lobe (He et al., 1997). Therefore, the binding afinity of iron for the recombinant human serum transferrin is expected to be weak. Similarly, titration of iron to soluble p97 as monitored by UV-visible spectroscopy suggests that two atoms of iron can bind to the soluble p97. This would suggest that the C-lobe of soluble p97 is able to bind iron although D421S substiution may have resulted in a weakened binding affinity. III. Differential scanning calorimetry studies on iron binding to soluble p97 A. Rationale The binding of ligands to macromolecules such as proteins, nucleic acids and polymers can lead to conformational change in the macromolecule structures. It has been shown that the equilbrium between the folded and unfolded state of the macromolecule is related to the diference in free energy of ligand binding between the two states (Schelman, 1975). Calorimetry has been used to measure the energy of macromolecule-ligand interaction (Brandts and Lin, 1990). Isothermal titration calorimetry and diferential scanning calorimetry are the two diferent calorimetric techniques used to determine the energy of the interactions between the macromolecule and the ligand. The isothermal calorimetry method is used to characterize the binding interactions of a 94 macromolecule to a ligand at a constant temperature, whereas the diferential scanning calorimetry method measures the conformational energy of a molecule over a range of temperatures. Direct measurement of the heat of reaction is one of the best ways to characterize the thermodynamics of binding. Data obtained from isothermal titration calorimetry alows rapid determination of binding affinity. However, the largest binding constant that can be measured reliably by isothermal titration calorimetry is about 109M_1 (Haynie, 2001). This poses a problem if the binding constant is large under physiological conditons such as in the case of iron binding to the members of the transferrin family. An alternative method to measure large binding constant is diferential scanning calorimetry. The parameters measured in the diferential scanning calorimetry method alow the investigator to calculate the binding constant of a ligand to a macromolecule by taking advantage of a series of equations derived by Brandts and Lin (Brandts and Lin, 1990). Appendix II contains a simplified version of the derivation of the equation used to measure the binding constant of ligand to macromolecule. The conventional method for measuring the binding constant of a ligand to a macromolecule is by equilbrium dialysis. Diferential scanning calorimetry has two important advantages over the conventional equilbrium method. First, diferential scanning calorimetry alows the determination of the binding constants of an individual domain in a multi-domain molecule. Secondly, although diferential scanning calorimetry may be criticized as an indirect method of measuring the binding constant, it measures large binding constants (Brandts and Lin, 1990; Lin et al., 1994). Direct equilbrium methods often fail because they lack the sensitivity if the binding constant is 95 too large due to limitation imposed on concentrations of both the ligand and macromolecule. In addition, in an ultra-tight binding situation, the on rate is often faster than the of rate . Therefore, thermodynamic equilbrium between the ligand bound and unbound state may take months to achieve. Diferential scanning calorimetry, on the other hand, avoids the need to lower the concentration of macromolecule and ligand as the interaction becomes stronger or the need to wait until thermodynamic equilbration is reached. Brandts and Lin have derived a series of thermodynamic equations to determine the binding constants of macromolecule-ligand interactions using the parameters measured from diferential scanning calorimetry (Brandts and Lin, 1990). The binding constant and the heat capacity of binding determined from the diferential scanning calorimetry methods agree quantiatively with the coresponding estimates obtained from the classical equilbrium method extrapolated to the same temperature (Brandts and Lin, 1990). After weighing out the advantages and disadvantages, diferential scanning calorimetry was used in this thesis to determine the binding constant of iron to soluble p97. During a diferential scanning calorimetry experiment, several parameters are measured as the protein is subjected to a temperature scan. Figure 10A depicts a simple diagram of the inside of a diferential scanning calorimeter. The sample cel is filed with a protein solution and the reference cel is filed with an identical volume of solvent. During a diferential scanning calorimetry experiment, the sample and the reference cels are heated over a range of temperature. As the protein solution undergoes the unfolding process (and dissociation process, if ligand is bound), energy is either absorbed or 96 (A) sample cell (protein solution) reference cel l (solvent) Heater Heater • — Computer to monitor temperature and regulate heat flow (B) 5 0 0 0 0 ^ 4 0 0 0 0 E 3 0 0 0 0 Q. O 2 0 0 0 0 1 0 0 0 0 0 3 0 0 3 1 0 3 2 0 3 3 0 3 4 0 3 5 0 3 6 0 T e m p (K) Figure 10: (A) Ske tch d iagram of the inner chamber of a typical differential scann ing calorimeter. Samp le cel l containing the protein solution and reference cel l containing the buffer respectively. During a thermoscan, power is suppl ied to the heaters at a s teady rate and a computer is l inked to the calorimeter to monitor the temperature and regulate heat flow. (B) A typical thermogram of hen egg lysozyme. The heat capaci ty (Cp) is plotted against the temperature (Temp). The difference between the basel ine of the folded and unfolded protein is the difference in heat 97 released. A thermogram is a plot of the heat capacity (Cp) of the protein solution as a function of temperature. Figure 10B is an example of a thermogram of hen egg lysozyme. The melting temperature (Tm) is the temperature at which there is equal concentration of unfolded protein and folded protein species. The area under the curve represents the total heat or enthalpy change (A H) for the entire process. In another words, A H=I Cp dT, where Cp is the heat capacity. In some cases, two separate peaks are seen in the Cp versus temperature plot. The two peaks represent independent unfolding of two domains within the proteins, which can be studied separately. For p97 to be a physiologicaly functional iron binding protein, it must have a high afinity for iron in order to compete for iron with serum transferrin, which binds iron at a very high binding constant. T binding afinities of iron for both lobes of the ovotransferin and serum transferrin had been determined by diferential scanning calorimetry (Lin et al., 1994). Therefore, this method was employed to study to determine the binding afinity of iron for soluble human p97. B. Results I: Diferential scanning calorimetry study of iron binding to soluble p97 Isothermal calorimetry was initialy atempted to determine the binding constant of iron to soluble p97. Since the binding constant of iron to p97 is expected to be much larger that \09M'\ the efort was abandoned in favor of the diferential scanning calorimetry. The diferential scanning calorimetry thermogram for apo-p97 shows a peak heat capacity at around 52°C and a trailing heat capacity plot at scanning temperature greater than 52°C suggesting that there are at least two transitons during the unfolding of p97 98 (Fig 11). Using the equations derived from Brandts and Lin, the heat capacity measured agrees with two possible unfolding paterns. The upper trace of the Figure 11 assumes that the apo-p97 unfolds into two domains, whereas the lower trace assumes the protein unfolds in three domains. Using the equations from Brandts and Lin, the fited Tm and enthalpies of each domain is listed in Table V. The diference in heat capacity (A Cp) between the folded and unfolded apo-p97 is 6 kcal/mol/°C. The A Cp for each domain is estimated by taking the experimentaly observed A Cp and dividing it by the number of domain. Therefore, in a two-domain fit, the A Cp for each domain is fixed at 3 kcal/mol/°C, whereas in a three-domain fit, the A Cp for each domain is fixed at 2 kcal/mol/°C. The total calorimetric heat of unfolding of apo-p97, obtained by integration of the experimental thermogram, is 367.5 kcal/mol. From Table V, the experimental calorimetric heat of unfolding agrees wel with the sum of the A H obtained from the three-domain fit. Therefore, according to diferential scanning calorimetry data, apo-p97 unfolds into three domains. Figure 12 shows the diferential scanning calorimetry thermograms of apo-p97 compared with several ratios of soluble p97 loaded with iron. It appeared that only one domain was shifted during the thermoscan. The Tm of this domain shift is about 30 °C, from 52.1 °C for apo-p97 to 82.4 °C for holo-p97. The shifted domain is presumed to be the conserved N-lobe of p97. Therefore, according to the diferential scanning calorimetry method, only one domain of soluble p97 binds to iron, presumably to the N-lobe. There is a minimal shift in the Tm for the other two domains. Hence, these domains do not appear to bind iron. 99 Tempera ture ( °C) Figure 11: Differential s c a n n i n g calor imetry of the two transit ion unfolding s ta tes of h u m a n a p o - s o l u b l e p97. (A )The uppe r t race dep i c t s the assumpt i on that the protein unfo lds into two d o m a i n s w h e r e a s (B) the lower t race dep ic ts the assumpt i on that the protein unfolds into three doma ins . Th i s is a representat ion of three sepa ra te exper iments . 1 0 0 Table V: Results of the two-state fits for apo-p97 Model Tm (°C) A H (kcal/mol) A C P (kcal/mol/ °C) 2-domain fit: peak 1 52.0 205 3 peak 2 55.9 122 3 3-domain fit: peak 1 52.1 188 2 peak 2 52.4 101 2 peak 3 57.3 97.5 2 101 \ F e : p 9 7 10 20 30 40 50 60 70 80 90 100 Temperature (°C) Figure 12: DSC t he rmograms of p97 loaded with var ious ratio iron to p 9 7 (Fe :p97 ) . T h i s is a represen ta t ion of th ree exper iments . 1 0 2 In summary, the diferential scanning calorimetry study reveals that soluble p97 has an iron-binding constant of 2 x 1017 M"1 at 25 °C and 2.6 x 1017 M"1 at 37 °C, which are similar to the binding constant of iron for transferrin measured with the same method (1.1 x 102 M" 1 (25°C) for the C-lobe and 8 x 1013 M ' 1 (25°C) for the N-lobe). Furthermore, the diferential scanning calorimetry study suggests only one region of the soluble p97, the N-lobe, binds iron under the conditons used. Therefore, the issue whether p97 binds two atoms of iron remains unclear and wil be discussed. C. Discussion In order for p97 to be a functional iron binding protein, it must have a high afinity for iron to be able to compete with serum transferrin, which binds iron at a very high binding constant. Diferential scanning calorimetry was used to determine the binding constant of iron to p97. Apo-p97 was subjected to a thermoscan in the calorimeter to determine the number of domains that could potentialy be involved in iron-binding. The thermogram of the unfolding of apo-p97 suggests that the protein unfolds in two transition states. The mathematical fitting of the total enthalpy and the Tm of unfolding associated with apo-p97 indicates that there are two possible unfolding paterns. Figure 11A shows that the fitting traces that suggest apo-p97 can unfold into two domains while the fitting traces in Figure 1 IB suggest that apo-p97 unfold into three domains. Table V gives the results of the Tm and enthalpy associated with each domain in the two- or three-domain fit. The sum of the enthalpies associated with individual domains of the three domain-fit agrees beter with the total experimental enthalpy measured. The reason for this could be that there are two more adjustable parameters in 103 the three-domain fit compared to the two-domain fit (additional Tm and enthalpy parameters associated with the third domain). The two-domain unfolding patern suggests that N- and the C- lobes of the apo-p97 unfold into two individual domains (Figure 13A). There are two possible interpretations for the three-domain fit. Figure 13B shows that apo-p97 unfolds into the N-lobe and the open or unlocked C-lobe (due to the D421S substiution) is seen as two separate domains and Figure 13C suggests that apo-p97 can unfold into three domains, the zinc binding domain in additon to the N- and the C-lobes. The shift in Tm when apo-p97 is loaded with iron is due to the increased stabilty of the protein when ligand is bound and therefore requires higher energy to unfold. The thermogram of p97 loaded with iron shows that only one domain shifts to a higher Tm during a thermoscan. The identiy of the shifted domain is presumed to be the N-lobe. One way to confirm the N-lobe s identity is to perform the thermoscan with a recombinant N-lobe p97. However, since the amino acids for iron-binding in the N-lobe is conserved and the Tm of the domain shifts to a higher Tm at an iron to p97 ratio of 0.5: 1, the assumption is reasonably valid. The binding constant of iron to the N-lobe of p97 was determined to be 2.2 x lO17!^ "1. However, since the other lobe did not shift when iron was present during the thermoscan, it is likely that the C-lobe of p97 does not bind iron efficiently. Even at 100:1 iron to p97 ratio, the apparently iron loaded C-lobe showed the same thermogram profile as the C-lobe of apo-p97 (Figure 12). To put the binding constant of iron to human p97 in perspective, the binding constant of iron for transferrin measured by diferential scanning calorimetry method is reported to be 1.1 x 102 M"1 for the C-lobe and 8 x 1013 M"1 for the N-lobe (25°C). Lin 104 Figure 13: The three possible unfolding paterns of iron loaded soluble human p97 during a thermoscan. (A) The two lobes of p97 unfold into the N-and the C-domains. (B) The N-lobe of p97 unfolds into one domain whereas the C-lobe unfolds into two-sub-domains. (C) The N-and the C-lobes unfold into two domains and the proposed zinc-binding site unfolds into a separate domain. 105 et al. suggested that these binding constants were too large compared to the values obtained from the conventional equilbrium binding method because the Tm values observed for the holo-transferrin are too high (Lin et al., 1994). This is due to the fact that the heat scan rate is faster than the rate of unfolding of the protein and as a result, the unfolding process had not progressed to completion at a specific temperature before the heat scan proceeded. Similarly, the binding constant of iron to p97 could be an overestimation as wel for the same reason. However, relative to the binding constant of iron for transferrin, p97 has a very high afinity for iron. Although the binding constant of iron to p97 is five log scale lower than that of iron for transferrin, in situations where transferrin is saturated such as in Hereditary hemochromatosis, p97 can bind to iron very tightly and mediates uptake of iron into cels. Diferential scanning calorimetry studies of the D63S N-terminal half molecule of human serum transferrin (the D63S substiution in human serum transferrin is the same substiution as the D421S substiution in human p97) also show that there is little if any shift between the Tm of iron free and supposedly iron-loaded form of the molecule (Lin et al., 1993). Interestingly, UV-Visible spectroscopy of the D63S N-lobe human transferrin showed a peak absorbance at 426nm, indicating that iron was bound to it (He et al., 1997). The failure of the iron bound D63S human transferrin to exhibit a Tm shift during a thermoscan could be due to a failure to lock in the C-lobe as a result of the amino acid substiution. Therefore, when subjected to a thermoscan, there is no observable change in the iron bound D63S N-lobe human transferrin. Likewise, for human p97, the presence of serine instead of aspartic acid in the C-lobe can result in the failure to lock up the domain when iron binds. The binding of iron to the C-lobe did not change the conformation of 107 the lobe and therefore, there was no diference observed in the conformation between the iron bound and iron free C-lobe of p97. Hence, the presence of aspartic acid at positon 421 is likely crucial for the tight binding of iron to the C-lobe of p97. 108 Chapter 4: Endocytosis of GPI-anchored p97 I. Brief overview of endocytosis Endocytosis is an essential celular process through which the cel regulates the uptake of extracelular nutrients, and the level of receptors or membrane proteins on the cel surface. This process is very dynamic and responds to numerous stimuli or membrane changes that occur during cel growth. During endocytosis, smal invaginations of the membrane occur at specialized sites that are then pinched of into smal vesicles, which are subsequently transported into the cels. The contents of the vesicles are either sorted to the degradative compartment or recycled back to the cel surface. There are two general forms of endocytosis: phagocytosis (celular eating), which is the uptake of large particles including microorganisms and celular debris, and pinocytosis (celular drinking), which refers to the ingestion of fluids and solutes. Pinocytosis can be furthered sub-divided into clathrin-dependent and clathrin-independent pathways. An example of clathrin-mediated pathway is the transferrin receptor mediated uptake of transferrin (Fig. 3). The clathrin-independent pathway includes caveolae-mediated endocytosis. A. Caveolae mediated endocytosis The name caveolae was proposed by Yamada to define the celular morphology he observed in gal bladder endothelial cels as a smal pocket, vesicle, cave or recess communicating with the outside of the cel (Yamada, 1955). Palade had earlier described similar invaginations in endothelial cels that shutle molecules across the cel. 109 He named them plasmalemmal vesicles (Palade, 1953). Since then caveolae have been implicated in many diferent celular phenomena. The study of caveolae has been greatly faciltated by the identification of the marker protein, caveolin (Rothberg et al., 1992). The 21-24 kDa integral membrane proteins are thought to be the principal components of caveolae and provide curvature to the shape of the invaginations. There are three distinct caveolin genes, caveolin-1, caveolin-2 and caveolin-3 (Monier et al., 1995; Rothberg et al., 1992; Scherer et al., 1996; Tang et al., 1996). While caveolin-1 arid -2 are highly expressed in adipocytes, endothelial cels and fibroblasts, caveolin-3 is specificaly expressed in striated muscle cels (Mastick and Saltiel, 1997; Scherer et al., 1997; Song et al., 1996). (Unless specified, caveolin in this thesis refers to caveolin-1) Caveolin is thought to anchor the membrane through its 33 amino acids hydrophobic domain, leaving the C- and the N-termini free in the cytoplasm (Kurzchalia et al., 1994). Caveolin appears to be a cholesterol-binding protein and the integrity of cholesterol is required to maintain the oligomerization of caveolin (Li et al., 1996; Monier et al., 1995; Murata et al., 1995). Sterol binding reagent such as filipin and nystatin can be used to sequester cholesterol in membranes and thereby to modulate the endocytosis of certain ligands (e.g., cholera toxin) by the caveolae-like pathway of endocytosis (Deckert et al., 1996; Orlandi and Fishman, 1998; Rothberg et al., 1992). Caveolae have specific lipid compositon and are enriched in proteins with either GPI or faty acid modifcation (Zurzolo et al., 1994). These proteins and lipids appear to be concentrated in caveolae relative to the surounding membrane. Lipids such as glycosphingolipids, sphingomyelins and cholesterol form the core of caveolae and play 110 an important role in recruiting modifed proteins (Fielding and Fielding, 2000; Hooper, 1999; Ilangumaran and Hoessli, 1998). The biogenesis of caveolae is a multi-steps process. The specialty lipids such as the glycosphingolipids, sphingomyelins and cholesterol that make up the lipid core of caveolae originate in the transitional region of the Golgi apparatus. GPI-anchored proteins and caveolin, which are synthesized in the endoplasmic reticulum and matured in the Golgi, are incorporated into these lipid cores and are shipped to the cel surface (Dupree et al., 1993; Lisanti et al., 1993). This assembly process of caveolae is diferent from the clathrin-coated vesicles. While caveolae are assembled in the Golgi apparatus and delivered to the plasma membrane, the clathrin coat is synthesized de novo at the sites of vesicle formation. Maintenance of caveolae on the plasma membrane requires the constant shutling of cholesterol and other lipid from the endoplasmic reticulum to the caveolae domain (Fielding and Fielding, 1996). Depletion of cholesterol by sterol binding reagents such as filipin and nystatin causes the disassembly of caveolin and eventualy leads to the cessation of caveolae internalization (Smart et al., 1996). Recently, the study of the exocytic pathway of caveolin and caveolin-2 in MDCK cel revealed that the two caveolins can form homo- and hetero-oligorheric complexes in the endoplasmic reticulum, and that these complexes are modifed during transport to the plasma membrane. Caveolin homo-oligomeric complexes are present in apical transport vesicles whereas caveolin and caveolin-2 hetero-oligomers are routed to the basolateral membrane (Scheifele et al., 1998). This work provides the first glimpse into the possible existence of functionaly specialized complexes of caveolins. I l l Early morphological studies suggest a role for caveolae in the transport of molecules across the endothelium (Simionescu and Simionescu, 1991). Other studies have also demonstrated that caveolae can endocytose specific ligands such as cholera toxin, bacteria and virus in various cels (Pelkmans et al., 2001; Shin et al., 2000,Orlandi, 1998 #387). The use of filipin and nystatin disrupts the structure of caveolae and inhibits the transcytosis of insulin and native albumin in endothelial cel (Schnitzer et al., 1996). These findings support the concept that caveolae plays an active role in celular ligand transport. The study on the internalization process of caveolae was further facilitated by GPI-anchored proteins with known ligands. One of them is GPI-anchored folate receptor, which provided the first biochemical data that caveolae could mediate the uptake of molecules (Anderson et al., 1992). GPI-anchored p97 is another model for such study because the ligand is iron. Ligands and receptors internalized by caveolae can have one of four fates as depicted by Figure 14 (for review please see (Anderson, 1998). (1) The receptor is recycled back to the cel surface where as the ligand is transported into the cytoplasm, for example, folate and iron. (2) The receptor recycles back to the surface while the ligand is transported to the endoplasmic reticulum. Examples are cholesterol, faty acids and viruses such as simian virus 40 (SV-40). (3) The ligand is transported across the cel while the receptor is recycled back to the plasma membrane such as the transcytosis of albumin and various plasma proteins. (4) Both receptor and ligand are sequestered in the vesicular caveolae compartment, for example alkaline phosphatase, endothelin and bradykinin and their receptors. 112 113 The formation of detached caveolae vesicles from the plasma membrane requires dynamin protein. When anti-dynamin antibody is injected into hepatocytes, numerous non-clathrin-coated flask-shaped structures resembling caveolae accumulate at the plasma membrane (Henley et al., 1998). The internalization of fluorescein-labeled cholera toxin B, which is normaly mediated by caveolae, is inhibited in anti-dynamin antibody-injected cels. Electron microscopy shows that the cholera toxin remains concentrated in plasmalemmal caveolae in these inhibited cels and cannot gain access to cytoplasmic organeles (Henley et al., 1999). These data support the model that dynamin participates in the fission of caveolae from the plasma membrane. Caveolae also play an important role in signal transduction. Caveolin itself was initialy discovered as a phosphorylation target of the v-src kinase encoded by the Rous sarcoma virus (Glenney, 1989; Glenney and Zokas, 1989). A variety of signaling molecules are found concentrated in caveolae. Receptor and non-receptor tyrosine kinases such as protein kinase C, MAP kinases, PI3 kinases and others are found to be highly concentrated in caveolae as shown either by biochemical purification or by morphological localization methods (Lisanti et al., 1994; Liu et al., 1997; Liu et al., 1996; Smart et al., 1996). G proteins are also found in most caveolar preparations. There is now considerable evidence showing that caveolae are the sites of calcium storage and entry (Isshiki and Anderson, 1999; Lohn et al., 2000). Lipids such as ceramides, inositol triphosphates and inositol-phosphoglycans are produced in the caveolae in response to specific stimuli such as IL-lp% bradykinin and insulin respectively (Liu and Anderson, 1995; Pike and Casey, 1996). With so many diferent signaling molecules congregated in 114 one location, caveolae are considered to be the place for signal integration, alowing cross talk between the diferent pathways and the plasma membrane. There are confusions about the relationship between caveolae and lipid rafts. Lipid rafts are dynamic assemblies of cholesterol and sphingolipids in the exoplasmic leaflet of the lipid bilayer of the plasma membrane. The similarity in lipid compositon between lipid rafts and caveolae has led to the assumption that the two entiies are the same. The problem has arisen primarily due to the dificulty in isolating pure caveolae and the variation of techniques used by diferent investigator to purify and study lipid rafts and caveolae. The most frequently used method for isolating caveolae is based on density gradient centrifugation (Smart et al., 1995). This technique alows the separation of celular membrane according to their density by centrifugation. Cholesterol, glycosphingolipids and sphingomyelins are widely believed to be enriched in or around caveolae (Fielding and Fielding, 2000; Hooper, 1999; Ilangumaran and Hoessli, 1998). This unique lipid compositon of caveolae results in a higher lipid protein ratio and hence alows the lighter density caveolae to be separated from the other celular membranes. However, in cels that lack caveolin such as neurons, these light density membranes are also present (Gorodinsky and Harris, 1995). Obviously, caveolin is not required for the formation of light density lipid rafts and cannot be used to defined lipid rafts. In order to resolve the confusion between caveolae and lipid rafts, the term caveolae is reserved strictly for morphologicaly defined cel surface invaginations which contain the marker protein caveolin (Brown and London, 1998; Simons and Toomre, 2000). The localization of a protein to caveolae must be supported by double labeled immunofluorescence and immunoelectron microscopy. 115 B. Endocytosis of GPI-anchored proteins The endocytotic pathway of GPI-anchored protein has been studied with various techniques. A sizable amount of information regarding the endocytotic pathway of the GPI-anchored protein was derived from studies of the uptake of folate or folic acid by GPI-anchored folate receptor. Anderson et al. had shown that in MA-104 monkey cels, GPI-anchored folate receptor entered the cel through a process termed potocytosis (Anderson et al., 1992). Folate receptors are clustered in caveolae, where the acidification of the caveolar content causes the dissociation of folate from its receptor and folate is translocated into the cytoplasm by a putative carrier (Anderson, 1994; Anderson et al., 1992; Rothberg et al., 1992; Rothberg et al., 1990). During this process, the caveolae is thought to remain atached to the plasma membrane. However, there is evidence showing that caveolae are not stationary flask-shaped invaginations underneath the plasma membrane. They are shown to participate in intracelular traficking (Schnitzer et al., 1996). There are examples of GPI-anchored proteins, which are internalized into the cels through clathrin-coated vesicles. Rijnbout  et al. proposed that GPI-anchored folate receptors are internalized into the cels via the clathrin-coated vesicles and reach the endosomal compartments. In additon to GPI-anchored folate receptor, other GPI-anchored proteins, such as prion protein and placental alkaline phosphatase, have been shown to cluster in clathrin-coated vesicles (Harmey et al., 1995; Makiya et al., 1992; Shyng et al., 1994). Therefore, the route of endocytosis of GPI-anchored protein remains to be established. There are debates on whether GPI-anchored proteins naturaly reside in caveolae or that the clustering of GPI-anchored molecules in caveolae is induced by 116 specific antibodies (Mayor et al., 1994). Data from electron microscopy studies clearly show that GPI-anchored are clustered on the cel surface and within the caveolae (Ide and Saito, 1980). However, immunohistochemistry, by contrast, shows that some GPI-anchored protein are either difusely distributed and are clustered into caveolae depending on either the use of primary antibodies, secondary antibodies or fixatives (Cerneus et al., 1993; Jemmerson et al., 1985; Jemmerson and Low, 1987; Mayor et al., 1994; Parton et al., 1994). However, majority of the morphological and functional data support the conclusion that GPI-anchored proteins tend to cluster and associate with caveolae. II. Confocal Immunofluorescence microscopy studies on the internalization pathway of GPI-anchored p97 A. Rationale The study of the endocytosis pathway of GPI anchored p97 presents a unique opportunity to determine the route of endocytosis of GPI anchored protein and to folow the fate of its ligand, iron, in the process. GPI-anchored p97 has been shown previously to be able to bind and internalize iron (Kennard et al., 1995). The internalization of GPI-anchored p97 is both energy and temperature dependent. Since there are still debates on whether GPI-anchored proteins are internalized through caveolae or clathrin-coated vesicles, the internalization route of GPI-anchored p97 was examined. Three independent methods were caried out to examine the endocytic pathway of GPI-anchored p97. The microscopy methods would provide a visual guide to the route of internalization. The sub-celular fractionation method would circumvent the argument 117 that clustering of GPI-anchored proteins into caveolae is due to antibody cross-linking. Finaly, sterol-binding reagents were used to examine whether iron uptake by GPI-anchored p97 is inhibited if the caveolae structure is disrupted (Chen and Norkin, 1999; Fielding and Fielding, 2000; Orlandi and Fishman, 1998; Rothberg et al., 1992). The results of this study reveal a new iron uptake pathway separate from those described previously. B. Results I: GPI-anchored p97 is internalized through caveolae In order to determine whether p97 co-localizes with caveolae or clathrin coated pits, a double-labeled immunofluorescence microscopy study of SK-MEL28 cels was performed. SK-MEL-28 is a human melanoma cel line, which expresses transferrin receptor, caveolin, clathrin and elevated levels of p97. Cels were grown on coverslips to 70% confluency and stained with monoclonal antibodies against human p97 and transferrin receptor. Excess antibodies were removed before transfering the cels into the 37 °C chamber to alow for endocytosis for diferent periods of time. The cels were then fixed, permeabilzed and visualized with secondary antibodies conjugated to either Alexa-488 or Alexa-546 fluorophores. Figure 15A shows that GPI-anchored p97 did not endocytose through clathrin vesicles during the endocytic process. Instead, a large amount of GPI-anchored p97 co-localized with caveolin during the endocytic process (Fig. 15B). The majority of p97 was localized to the membrane patches that contained caveolin at the 0-minute time point. At 10 minutes, a smal number of p97 molecules were internalized into caveolae and subsequently at 20 minutes, most of the p97 was observed inside caveolae. As a control, the co-localization of 118 Figure 15: Confocal immunofluorescence micrographs of the localization of GPI-anchored p97 with clathrin or caveolin on SK-MEL 28 cels. SK-MEL 28 cels were labeled with antibodies against GPI-anchored p97 at 4°C. Excess antibodies were removed before internalization was alowed at 37 °C for various periods. The cels were fixed and permeabilzed before incubating with antibodies against either (A) clathrin-coated vesicles or (B) caveolae. Labeled GPI-anchored p97 were visualized with Alexa 488-conjugated goat anti-mouse antibodies (green) and labeled caveolae or clathrin vesicles were visualized with Alexa 568-conjugated goat anti-rabbit antibodies (red). Merged images are shown on the right. Co-localization of Alexa 488 and Alexa 568 is indicated by the yelow color in the merged panels. (C) The enlarged image of the 20 minutes merged panel of the co-localization of GPI-anchored p97 and caveolae. The scale bar represents lu,m. These confocal micrographs are representatives of three separate experiments. 119 (A) Clathrin p97 Merged 120 Caveolae p 9 7 Merged 121 122 Figure 16: Confocal immunofluorescence micrographs of the localization of transferrin receptor (TfR) with clathrin or caveolin on SK-MEL 28 cels. SK-MEL 28 cels were labeled with antibodies against TfR at 4°C. Excess antibodies were removed before internalization was alowed at 37 °C for various periods. The cels were fixed and permeabilzed before incubating with antibodies against either (A) clathrin-coated vesicles or (B) caveolae. Labeled TfRs were visualized with Alexa 488-conjugated goat anti-mouse antibodies (green) and labeled caveolae or clathrin vesicles were visualized with Alexa 568-conjugated goat anti-rabbit antibodies (red). Merged images are shown on the right. Co-localization of Alexa 488 and Alexa 568 is indicated by the yelow color in the merged panels. The scale bar represents lLim. These confocal micrographs are representatives of three separate experiments. 123 Clathrin T f R Merged 124 Caveolae T f R Merged 125 transferrin receptor with either clathrin-coated pits or caveolae was examined. Unlike p97, there was a significant co-localization of transferrin receptor to clathrin-coated vesicles during the 10- and 20-minute time points of internalization. However, after 30 minutes, most of the transferrin receptors did not co-localize with clathrin-coated vesicles (Fig. 16A). It is possible that at 30 minutes after endocytosis, the transferrin receptor has either recycled to the cel surface or has moved to diferent compartments such as endosomes. In any case, most of the transferrin receptors did not co-localize with caveolin at any time point folowing endocytosis, demonstrating that transferrin receptor enters the cel by clathrin-mediated endocytosis (Fig. 16B). In summary, the immunostaining and co-localization data show that GPI-anchored p97 is preferentialy internalized through caveolae. C. Results II: Cryo immunoelectron microscopy study on the localization of GPI-anchored p97 The immunofluorescence data suggest caveolae and p97 are in similar regions of the cel. However, the data does not unambiguously show that caveolae and p97 are absolutely co-localized since the size of caveolae are smaler than the spatial resolution of a confocal microscope. Therefore, cryo immunoelectron microscopy data was obtained to support the confocal immunofluorescence study described in the previous section. Due to the sensitivity of the caveolin epitope to paraformaldehyde and glutaldehyde, various conventional electron microscopy fixation methods could not be used. Therefore, cryo-electron microscopy was employed. SK-MEL 28 cels was infused with 25% polypyrolidone in 2.3M sucrose over-night before being mounted on 126 the aluminum sample pin and cryo-fixed in liquid nitrogen. Thin sections were cut at -110 °C and double stained for caveolin and p97. Caveolae were detected with anti caveolin antibodies and protein A conjugated to 5 nm coloidal gold particle and p97 was detected with anti p97 antibodies and a goat-anti-mouse secondary antibody conjugated to 10 nm coloidal gold particles. The staining of secondary antibodies conjugated to gold particles alone to the cryo-sections showed that the blocking agent used, 0.5% fish skin gelatin, was adequate in preventing non specific binding of secondary antibodies to the sections. The electron micrograph in Figure 17A shows that p97 is found in vesicular structures, which also contain caveolin (see arows). As a control, SK-MEL 28 cryo sections were also stained for transferrin receptor and P-adaptin, a component of the clathrin coat. Figure 17B shows that some transferrin receptor co-localizes with p-adaptin (see arows). The structural preservation of the cels subjected to this technique is poor. In Figure 17A, the structural remnant resembles a mitochondrion is observed (see leter M). However, the membrane structures of the vesicular components of the cels are poorly preserved. Therefore, alternative methods were employed to further support the observations from the confocal microscopy. D. Results III: Iron uptake by GPI-anchored p97 is disrupted by sterol sequestering reagents In order to determine whether sterol-binding reagent can disrupt iron uptake by GPI-anchored p97 mediated by caveolae, both human p97 and human transferrin receptor transfected TRVb cel lines were pre-incubated with either filipin or nystatin for 30 127 Figure 17: Cryo immunoelectron micrograph of SK-MEL 28 cels. (A) GPI-anchored p97 labeled with 5 nm gold particles co-localizes with caveolin labeled with lOnm gold particles (see arows). (B) Transferin receptor labeled with 5 nm gold co-localizes with (3-adaptin labeled with lOnm gold, a subunit of the clathrin coat (see arows). Mitochondrion is labeled with M. These micrographs are representative of two separate experiments. 128 minutes before Fe-citrate or Fe-transferrin was added to the cels respectively. Figure 18 shows the efect of the sterol binding reagents on iron uptake by transferrin/ transferrin receptor and GPI-anchored p97 pathways. Both nystatin and filipin significantly decreased the internalization of iron uptake mediated by GPI-anchored p97. Filipin reduces the uptake of 5Fe-citrate by 60% at 25 p.g/ml while nystatin reduces the uptake of 5Fe-citrate by 70% at 250 plg/ml (Fig. 18). Interestingly, filipin, was able to reduce the uptake of 5Fe-transferrin by 20% whereas nystatin has little or no efect in the uptake of 5Fe-transferrin. Therefore, the presence of cholesterol perturbing agents clearly afects the iron uptake by GPI-anchored p97. This demonstrates that GPI-anchored p97 uptake is mediated by cholesterol/ glycolipid-enriched microdomains, which include caveolae. E. Results IV: GPI-anchored p97 fractionates with caveolae by sub-celular fractionation GPI-anchored proteins, cholesterol, glycosphingolipids and sphingomyelins can form a detergent-insoluble membrane domain in light buoyant density folowing a sub-celular fractionation employing a sucrose step gradient (Brown and Rose, 1992; Smart et al., 1996). Therefore, in order to provide independent biochemical evidence that p97 co-fractionate with caveolae and to exclude the possibilty that the clustering of GPI-anchored proteins in caveolae is due to receptor cross-linking by antibodies, SK-MEL-28 cels were subjected to biochemical fractionation. Celular lysates separated on a sucrose gradients yielded a white flocculent band, coresponding to the low density lipid complex, between the third and the fourth fractions 129 Figure 18: The efect of nystatin on the iron uptake by GPI-anchored p97 and transferrin receptor. P97+TRVb and TfR+TRVb cels were treated with either nystatin, filipin or the solvent DMSO before iron uptake experiment. The presence of nystatin and filipin reduces the uptake of iron by GPI-anchored p97. In comparison, filipin and nystatin have little or no efect on iron uptake through the transferrin receptor respectively. The eror bars represent the standard deviations of three separate experiments. 130 of the gradient as previously reported (Brown and Rose, 1992; Smart et al., 1996). Protein components in the fractions were identified by Western blot analysis. Figure 19 shows that caveolin and p97 are found in fraction 3, whereas p-adaptin, which is a protein subunit of the clathrin complex, is found in fraction 8. Transferin receptor, on the other hand, is found in fractions 7 and 8. The presence of p97 and transferrin receptor in completely unrelated and separate fractions is in agreement with the confocal immunofluorescence data and demonstrates that p97 is found in vesicles containing caveolin whereas transferrin receptor fractionates with clathrin. The result clearly shows that GPI-anchored p97 and transferrin receptor fractionates into diferent density fraction. GPI-anchored p97 fractionates with caveolin, whereas transferrin receptor is detected in the same fraction as P-adaptin. F. Discussion A second aspect of this thesis is to establish how GPI-anchored p97 is internalized and to determine whether iron caried by GPI-anchored p97 is donated to the cel. Studies on the GPI-anchored folate receptor provided the first biochemical clue that caveolae could mediate uptake of molecules in a variety of cels. In MA 106 cels, folate receptor internalizes bound folate through caveolae (Rothberg et al., 1990). However, there are controversies on whether the endocytosis of al GPI-anchored proteins is mediated by caveolae. Some reports have suggested that the clustering of GPI-anchored proteins in caveolae depends on the use of antibodies and fixatives. Others have also suggested that some GPI-anchored proteins such as prion protein and placental alkaline phosphatase enter the cels through clathrin-coated vesicles (Makiya et al., 1992; Shyng et al., 1994). 132 0 % sucrose gradient 4 0 % Adapt in Transferrin receptor Caveo l in p97 Fraction number 1 2 3 4 5 6 7 8 Figure 19: Western blots of subcel lu lar fractionation of SK -MEL 28 cell lysate in a 0-40% sucrose step gradient. Eight fractions were col lected from the sucrose gradient and each fraction w a s subjected to SDS-PAGE and transferred onto Immobilon membrane. E a c h membrane w a s probed separately for adaptin (a component of the clathrin coat complex) , transferrin receptor, caveol in and p97. 133 Therefore, several independent experimental methods were caried out in this study to determine the endocytic pathway of GPI-anchored p97. The immunofluorescence experiments show that in SK-MEL 28 cels, GPI-anchored p97 co-localize with caveolae (Fig. 15A). This result was also supported by cryo-immunoelectron microscopy (Figure 17) and sub-celular fractionation of SK-MEL 28 cels on sucrose gradients (Fig. 19). In addition, the iron uptake by GPI-anchored p97 was also sensitve to the presence of two sterol binding reagents, filipin and nystatin, indicating that caveolae was involved in the process. Electron microscopy was atempted initialy in order to provide definitive high-resolution electron micrograph to coroborate the co-localization of GPI-anchored p97 with caveolae from the confocal immunofluorescence experiments. However, due to the sensitivity of caveolin to conventional fixatives, cryo-immunoelectron microscopy methods were used to obtain thin sections for immuno-staining. SK-MEL 28 cels were infused with cryo-preservatives, polypyrolidone and sucrose, before freezing in liquid nitrogen. This method of preservation is barely adequate to preserve celular structures. In addition, due to the sensitivity of the caveolin to aldehydes, the cels were fixed with only 2% (w/v) paraformaldehyde. As a result, the cels were poorly fixed and the electron micrograph ilustrated in Figure 17 does not show intact celular structures such as Golgi apparatus and nucleus. Therefore, the technique was not pursued any further. Exposure of the p97 transfected CHO cels to sterol-binding agents such as filipin and nystatin, which are known to disrupt caveolae structure, reduces the uptake of 5Fe-citrate by GPI-anchored p97 into the cels (Fig. 18). The experiments also showed that 5Fe-transferrin uptake was reduced in the presence of filipin but not nystatin. There are 134 reports suggesting that cholesterol is required for clathrin-mediated endocytosis. Subtil et al. and Rodal et al. showed that the exposure of cels to an acute cholesterol-depleting drug such as P-methyl-cyclodextrin perturbed the formation of clathrin-coated vesicles (Rodal et al., 1999; Subtil et al., 1999). However, others have shown that the use of filipin and nystatin does not afect the formation of clathrin-coated pits in the cels (Deckert et al., 1996). However, Figure 18 shows that filipin reduced the uptake of 5Fe-transferrin by transferrin receptor and nystatin has little efects on 5Fe-transferrin uptake by transferrin receptor. One possible explanation for this apparent dichotomy of the efect of cholesterol on clathrin-mediated pathway is that the sequestration of cholesterol by filipin and nystatin has diferent efects than the removal of cholesterol by P-methyl-cyclodextrin (Subtil et al., 1999). The diferent efect of nystatin and filipin on 5Fe-transferrin by transferrin receptor is more likely due to the diference in the mechanism of actions of the two sterol binding reagents. III. GPI-anchored p97 traffics to endosomes after internalization. A. Rationale There is evidence that some GPI-anchored proteins such as GPI-anchored diphtheria toxin receptors and GPI-anchored CD 14 are localized to endosomes after they are internalized (Poussin et al., 1998; Skreting et al., 1999). However, there are also contrary data suggesting that GPI-anchored proteins are not found in the endosomes (Rothberg et al., 1990). Therefore, the study of the endocytic pathway of GPI-anchored p97 is needed to determine whether GPI-anchored p97 is traficked to the endosomes. 135 The localization of p97 to the endosomes wil provide additonal support for the model that proteins inside caveolae can be traficked into the endocytic pathway of the cel. B. Result I: GPI-anchored p97 is traficked to the endosomes In order to determine whether vesicles carying GPI-anchored p97 mature into endosomes, SK-MEL 28 cels were co-stained for p97 and the early endosome marker EEA1. Similar to the previous experiment in Figure 15, SK-MEL 28 cels were incubated with antibodies against p97 for 30 minutes at 4°C. Excess anti-p97 antibodies were removed before the cels were alowed to undergo endocytosis at 37°C for various periods of time. The cels were fixed, permeabilzed and incubated with anti-EEAl antibodies. Since it is known that transferrin receptor can be found in early endosomes after it is internalized, transferrin receptor was again used as a positve control for the experiment. In the merged panels of Figure 20A, a significant number of transferrin receptors is shown to localize to endosomes 10 minutes after internalization and remains localized to the endosomes up to 30 minute after internalization. This is demonstrated by the yelow color when both EEA-1 and p97 co-localized. A smal number of GPI-anchored p97 co-localizes with endosomes approximately 10 minutes after p97 is internalized. At 20 minutes after internalization, a large proportion of p97 co-localizes with EEA1 containing vesicles. This co-localization patern of p97 with EEA1 persists to 30 minutes after internalization (Fig. 20B). Therefore, the result clearly shows that GPI-anchored p97 is traficked to the endosomes after internalization. 136 Figure 20: Confocal immunofluorescence micrographs of the localization of the endosome antigen 1 (EEA1) transferrin receptor (TfR) and GPI-anchored p97 in SK-MEL 28 cels. SK-MEL 28 cels were labeled with antibodies against either (A) TfR or (B) GPI-anchored p97 at 4°C. Excess antibodies were removed before internalization was alowed at 37°C for various periods. The cels were fixed and permeabilzed before incubating with antibodies against EEA1 Labeled GPI-anchored p97 and TfR were visualized with Alexa 488-conjugated rabbit anti-mouse antibodies (green) and labeled EEA1 were visualized with Alexa 568-conjugated donkey anti-goat antibodies (red). Merged images are shown on the right. Co-localization of Alexa 488 and Alexa 568 is indicated by the yelow color in the merged panels. (C) The enlarged image of the merged panel of the 20 min co-localization of GPI-anchored p97 and EEA1. The scale bar represents lum. These confocal micrographs are representatives of three separate experiments. 137 138 139 1 4 0 C. Result II: GPI-anchored p97 trafics to the endosomes more slowly than transferrin receptor } The results of the co-localization of p97 and EEA1 (Fig. 20) suggest that the majority of GPI-anchored p97 trafics to endosomes 20 minutes after internalization, whereas, transferrin receptor trafics to endosomes 10 minutes after internalization. In order to determine whether GPI-anchored p97 and transferrin localized to early endosomes at the same time, a triple labeling experiment was performed. The experiment alowed the location of EEA-1 antigen, the internalization of transferrin and GPI-anchored p97 to be folowed. Serum transferrin was removed from SK-MEL 28 cels before they were incubated with Alexa 488 conjugated transferrin and antibodies against GPI-anchored p97 at 4°C. Excess antibodies and Alexa 488 conjugated transferrin were removed and the cels were alowed to undergo endocytosis at 37°C before they are fixed and permeabilzed. Anti-EEAl antibodies were used to localize the early endosomes in the cels. At 0 minute, both p97 and transferrin receptor were present at the cel surface of SK-MEL 28 whereas EEA1 exhibited a typical punctate endosomal staining (Fig. 21). At 10 minutes, most of the transferrin receptor was endocytosed into the cels and co-localized with endosomes, whereas most of p97 was still present on the cel surface. A smal number of GPI-anchored p97 molecules were seen in the endosomes containing transferrin receptor. The co-localization of three fluorescence dyes (Alexa 488, Alexa 568 and Cy5) was observed as a white punctate staining in the merged image (arows). At 20 minutes after endocytosis, the majority of transferrin receptors was no longer present in the endosomes, however, there is an increase in the number of p97 molecules associated with endosomes as shown by the pink color (when Alexa 568 co-localized 141 Figure 21: Confocal immunofluorescence micrographs of the localization of the endosome antigen 1 (EEA1) transferrin receptor (TfR) and GPI-anchored p97 in SK-MEL 28 cels. (A) SK-MEL 28 cels were labeled with Alexa 488 conjugated human transferrin (Tf) and antibodies against GPI-anchored p97 at 4°C. Excess Alexa 488 conjugated Tf and antibodies were removed before internalization was alowed at 37°C for various periods. The cels were fixed and permeabilzed before incubating with antibodies against EEA1 Labeled GPI-anchored p97 were visualized with Cy5-conjugated rabbit anti-mouse antibodies (blue) and labeled EEA1 were visualized with Alexa 568-conjugated donkey anti-goat antibodies (red). Alexa 488 conjugated Tf was visualized directly as green color. Merged images are shown on the right. Co-localization of Cy5- and Alexa 568-conjugated antibodies is indicated by the pink color in the merged panels and co-localization of Cy5-conjugated antibodies and Alexa 488-conjugated Tf is visualized as green color in the merged panels. The co-localization of Alexa 488, Alexa 568-conjuagted antibodies and Cy5-conjugated antibodies is visualized as white color in the merged panel (see arows). (B) The enlarged image of a cel from the 10-minutes time point merged panels. The arows point to the co-localization of al three fluorochromes. (C) The enlarged image of a cel from the 30-minutes time point merged panel. Co-localization of EEA1 and p97 is indicated by he pink punctate staining. The scale bar represents 1 fim. These confocal micrographs are representatives of three separate experiments. 142 0 m i n 143 with Cy5). At 30 minutes, there were a large number of endosomal vesicles containing p97. Therefore, GPI-anchored p97 was internalized more slowly than transferrin receptor. D. Discussion GPI-anchored p97 represents the second only example where the delivery of a ligand by GPI-anchored protein is folowed during the endocytic process. The internalization of GPI-anchored p97 into caveolae occured between 10-20 minutes after the initiation of the 37°C endocytosis period. At 30 minutes after internalization, p97 co-localized with endosomes, which were labeled with EEA1 (Fig. 20). Therefore, the confocal micrograph revealed that after p97 is internalized via caveolae and traficked to the early endosomes, thus implying the intersection of caveolae and endocytosis pathway. Specificaly, most of the transferrin is shown to move into endosomes after 10 minutes at 37°C, whereas the bulk of GPI-anchored p97 trafics to the endosomes after 20 minutes. At 30 minutes, while most of the transferrin is not seen in the endosomes any longer, the majority of GPI-anchored p97 is still localized into the endosomes (Fig. 21). This would suggest that the traficking of transferrin/ transferrin receptor complex is faster than GPI-anchored p97. In contrast, there is evidence to support the notion that endosomes do not play a role in the traficking of some GPI-anchored proteins into the intracelular compartments of the cels (Nichols et al., 2001). Early endosome function requires the GTPase rab5, and expression of the GDP-bound form of rab5 reduces the uptake of transferrin by the transferrin receptor. Nichols et al. expressed the GDP-bound form of rab5 and showed 145 that the absence of rab5 activity did not impair the traficking of GPI-anchored cholera toxin from the plasma membrane to the Golgi (Nichols et al., 2001). It is also possible that iron uptake by GPI-anchored p97 is not afected by the GDP-bound form of rab5. This can be tested by transfecting the GDP-bound form of rab5 construct into SK-MEL 28 cels and determine whether rab5 is required for iron uptake by GPI-anchored p97. Recently, a new compartment along the caveolae internalization pathway has been described in Simian virus-40 (SV-40) infected cels. The SV-40 virus was shown to infect the cels through caveolae and was traficked to the endoplasmic reticulum. The virus entered a newly identified intermediate organele (termed the caveosome) instead of early endosomes before proceeding to the endoplasmic reticulum (Pelkmans et al., 2001). Caveosome is defined as a pre-existing , caveolin rich intracelular organeles with a pH of ~7.0; it seems to be unique to the caveolar uptake pathway . This raises the question as to whether GPI-anchored protein is targeted to this new compartment along the endocytic pathway. However, since the internalization pathways mediated by caveolae have several fates (Fig. 14). It is possible that the circumvention of the endosomal compartments only applies to a subset of caveolae-mediated pathways. Viral entry and protein internalization are separate events and may have diferent targeting pathways inside the cels. IV. The fate of iron transported by GPI-anchored p97 A. Rationale The study of iron transported by GPI-anchored p97 represents a unique opportunity to folow the fate of the ligand transported by a GPI-anchored protein. As mentioned previously, the best-characterized ligand of a GPI-anchored protein is folate. 146 However, even in the GPI-anchored folate receptor system, it is not known how the folate escapes the caveolae, and whether endosomal compartments are involved in the uptake process. Although it has been shown that GPI-anchored p97 can internalize iron in a time and energy dependent manner (Kennard et al., 1995), others have suggested that p97 does not donate its iron to ferritin (Richardson and Baker, 1991a; Richardson, 2000; Richardson and Baker, 1991b). In the next set of experiments, the fate iron internalized by GPI-anchored p97 wil be examined. The sub-celular fractionation data wil show that the ligand, iron, is indeed found in the caveolae fraction in 5Fe-NTA labeled cels and using immunofluorescence GPI-anchored p97 is shown to traffic to the Nramp2 containing vesicles. Immunoprecipitation of ferritin from 5Fe-NTA labeled cels wil demonstrate that iron bound to p97 is subsequently donated to ferritin. B. Results I: Iron internalized by GPI-anchored p97 is fractionated into caveolae fraction In order to determine whether iron can be found in the caveolae fraction, SK-MEL 28 cels were labeled with either 5Fe-nitrilotriacetic acid (5Fe-NTA) or 5Fe-transferrin for 15 minutes. The cels were then dounce homogenized and subjected to sub-celular fractionation employing a sucrose density gradient as previously described in Figure 19. Each fraction was immunoprecipitated with antibodies against p97 or transferrin receptor and the radioactivity associated with respective proteins was determined. The amount of iron associates with GPI-anchored p97 in each fraction is shown in Figure 22. As ilustrated in the previous sub-celular fractionation experiment 147 Figure 22: Iron transported by GPI-anchored p97 fractionates into caveolae fraction. SK-MEL 28 cels were incubated with either (A) 5Fe nitrilotriacetate (5Fe-NTA) or (B) 5Fe-NTA loaded human transferrin. The cel lysates were subjected to sub-celular fractionation on sucrose gradient. Each fraction was immunoprecipitated for either p97 in (A) or transferrin receptor (TfR) in (B). As a control, normal rabbit serum (NRS) is used to immunoprecipitated each fraction. The radioactivity associated with each fraction is ploted. These graphs are representatives of two separate experiments. 148 149 150 (Fig. 19), caveolae is detected in fraction 3 of the sucrose density gradient. The majority of the 5Fe-NTA associated with GPI-anchored p97 is also found in fraction 3 (Fig.22A). Therefore, caveolae appear to be involved in the transportation of iron internalized by GPI-anchored p97. As a control, 5Fe-transferrin labeled SK-MEL 28 cel lysates were subjected to sub-celular fractionation and the results show that 5Fe-transferrin does not appear in any specific compartment (Fig. 22B). C. Results II: GPI-anchored p97 can be localized to vesicles containing Nramp2 Nramp2 is the metal ion transporter found in late endosomes or early lysosomes where iron is thought to be transported across the membrane into the cytoplasm and be donated to ferritin (Tabuchi et al., 2000). It is also a wel-described marker for the late endosomes and the lysosomes (Grifiths et al., 2000; Tabuchi et al., 2000). Using immunofluorescence microscopy, I have observed that GPI-anchored p97 is traficked to the early endosomes (Figs. 20 and 21). It is possible that GPI-anchored p97 is subsequently traficked to the late endosomes, where Nramp2 is localized. If that is the case, then Nramp2 could mediate the egress of iron into the cytoplasm. In order to determine the possible mechanism of how iron transported by GPI-anchored p97 can be transported into the cytoplasm, SK-MEL 28 cels were transfected with a plasmid, which expresses human Nramp2 fused with enhanced green fluorescence protein (pEGFP-hNramp2). The cels expressing the fusion protein were immuno-stained with anti-p97 antibodies to detect p97. A time-course profile of GPI-anchored p97 internalization pathway co-localizing with Nramp2 is shown in Figure 23 A. Transferin receptor has previously been shown to co-localize with Nramp2 and hence was used as a 151 Figure 23: Confocal immunofluorescence micrographs of the localization of GPI-anchored p97 or transferrin receptor (TfR) with Nramp2 on SK-MEL 28 cels. SK-MEL 28 cels were transfected with Enhanced green fluorescence protein conjugated human Nramp2 (pEGFP-hNramp2) 24 hours before incubating with antibodies against either (A) TfR or (B) GPI-anchored p97 at 4°C. Excess antibodies were removed before internalization was alowed at 37°C for various periods. The cels were fixed and permeabilzed before incubating with Alexa 568-conjugated goat anti-mouse antibodies (red). The EGFP-conjugated Nramp2 is visualized as green color. Merged images are shown on the right. Co-localization of EGFP-conjugated hNramp2 and Alexa 568 is indicated by the yelow color in the merged panels. (C) The enlarged image of the 20 minutes merged panel of the co-localization of GPI-anchored p97 and EGFP-hNramp2. The scale bar represents 1 u,m. These confocal micrographs are representatives of two separate experiments. 152 E G F P - h N r a m p 2 TfR M e r g e d 153 155 control in this experiment (Fig. 23B) (Tabuchi et al., 2000) . Co-localization of GPI-anchored p97 with Nramp2 occured at 20 minutes after internalization (as shown by the yelow merged color) and continues for another 10 minutes at 37°C (arows). The data shows that GPI-anchored p97 is in the same vesicles as Nramp2 after being taken up by the cels. Co-localization of the two proteins suggests that Nramp2 could serve as the exporter for iron taken up by GPI-anchored p97, and also that p97 likely appears in the late endosomes or lysosomes. D. Results III: Iron internalized by GPI-anchored p97 is donated to ferritin In order to address whether the iron internalized through GPI-anchored p97 is incorporated into ferritin, ferritin from 5FeNTA labeled p97+ or p97" transferrin receptor deficient TRVb cel line was immunoprecipitated in order to determine whether 5FeNTA is associated with the protein. These cels were also treated either with or without phosphoinositol-phospholipase C (PI-PLC), an enzyme that cleaves GPI-anchored protein to ensure that the iron loading of ferritin is dependent on the intact GPI-anchored p97. Figure 24A shows that when p97+TRVb cels were treated with PI-PLC, the amount of 5Fe associated with GPI-anchored p97 decreased. This coresponds to a decrease in 5Fe incorporated into ferritin. As a control, transferrin receptor deficient TRVb cels transfected with human transferrin receptor (TfR+TRVb) were labeled with 5Fe-transferrin. The cels, treated with or without PI-PLC, were immunoprecipitated with antibodies against transferrin receptor and ferritin (Fig. 24B). There was no significant diference in 5Fe incorporation of ferritin between the PI-PLC treated or untreated cels since transferrin receptor is not sensitve to PI-PLC treatment. 156 Figure 24: GPI-anchored p97 is able to donate iron to ferritin. Endogenous transferrin receptor negative TRVb cel, and those transfected with human p97 (p97+TRVb) and human transferrin receptor (TfR+TRVb) were treated either with or without phosphoinositol-phospholipase C (PI-PLC) before they were incubated with (A) 5Fe nitrilotriacetate (5FeNTA) for the p97+TRVb cels or (B) 5FeNTA loaded human transferrin for the TfR+TRVb cels. The cel lysates were pre-cleared and immunoprecipitated with antibodies against normal rabbit serum (NRS), ferritin and p97 (L235) (for the p97+TRVb cels) or transferrin receptor (OKT9) (for the TfR+TRVb cels). The radioactivity count associated with the immunoprecipitate was ploted. The eror bars represent the standard deviations of three separate experiments. 157 ( A ) 3 5 0 0 1 3000-2 5 0 0 -^ 2 0 0 0 -fx Q 1500-1000-cell line p97+ p97+ p97- p97-PI-PLC - + - + • NRS • L235 (anti-human p97) o Ferritin (B) Q_ Q 4 0 0 1 300-^ 2 0 0 , 100H 0 cell line Tf R + PI-PLC • NRS • OKT9 (anti-human TfR) n Ferritin T f R + T f R - T f R -158 Interestingly, in Fe labeled p97 TRVb cels, there is a significant expression of an endogenous protein, likely hamster p97, which could bind 5Fe and was sensitve to PI-PLC treatment. When these TRVb cels were treated with PI-PLC, the amount of 5Fe associated with this protein decreased, which coresponded to a decrease in 5Fe incorporation into ferritin. In summary, the results demonstrate that iron bound to GPI-anchored p97 fractionates into caveolae fraction and is subsequently incorporated to ferritin. GPI-anchored p97 is in the same vesicular compartments as Nramp2, which would suggest a possible route for iron egress into the cytoplasm. E. Discussion It has been proposed by several investigators that Nramp2 expressed on endosomal membrane is the channel for iron taken up by transferrin/ transferrin receptor to egress into the cytoplasm and ultimately to be incorporated into ferritin (Burdo et al., 1999; Fleming et al., 1998). Using confocal immunofluorescence techniques, the route of endocytosis of GPI-anchored p97 was folowed through to the endosomal compartments and vesicles containing the Nramp2 protein. Therefore, it is possible that iron taken up by p97 can be released into the cytoplasm by Nramp2. This provides a new understanding of a novel route of iron uptake through caveolae and ligand internalization by GPI-anchored protein. Previous report showed that Nramp2 is localized to the recycling endosomes and lysosomes (Tabuchi et al., 2000). Hence, a significant number of p97 appears to traffic from the early endosomes to the recycling endosomes or the lysosomesafter it is internalized. A model of iron internalization pathway by GPI-159 anchored p97 is ilustrated in Figure 25. The iron bound to GPI-anchored p97 is internalized through caveolae and trafics through from early endosomes to recycling endosomes or lysosomes where Nramp2 resides. Since GPI-anchored p97 endocytose more slowly than transferrin receptor, only a smaler number of GPI-anchored p97 associates with transferrin receptor in the endosomes compartment. A majority of the GPI-anchored p97 enters the endosomes at a later stage. Iron dissociates from p97 and is transported by Nramp2 into the cytoplasm where it is incorporated into ferritin. Recently, it was pointed out that p97 does not have an iron response element (Richardson, 2000). Therefore, its expression is likely not regulated by the level of intracelular iron, similar to the transferrin receptor 2, which does not possess an iron response element (Kawabata et al., 2000). Other researchers have also suggested that p97 does not play a substantial role in iron uptake in HeLa cels (Kriegerbeckova and Kovar, 2000). It has been shown that HeLa cels have few if any caveolae (Skreting et al., 1999). Therefore, studies of the internalization of p97 and any other GPI-anchored molecules in these cels might be impeded. In previous studies using SK-MEL 28 cels, it was suggested that while a membrane component identified on SK-MEL 28 consistent with p97 could bind iron, it did not appear to donate iron to the cel (Richardson and Baker, 1991a; Richardson and Baker, 1991b). The experiments were performed on the SK-MEL 28 cel line without a p97 negative control cel line. The results in Figure 22A clearly indicate that irons internalized by GPI-anchored p97 can be localized to caveolae fractions and they are subsequently donated to ferritin (Figure 24). When human p97 transfected transferrin receptor deficient TRVb cels were treated with PI-PLC, the amount of iron donated by 160 Figure 25: The proposed model of iron uptake pathway by GPI-anchored p97. (A) GPI-anchored p97 bound with Fe3+ is internalized via caveolae. As a comparison, transferrin receptor (TfR) mediated iron-transferrin uptake is shown. Holo-transferrin bound to transferrin receptor is endocytosed through clathrin-coated pits. The caveolae and the clathrin-coated vesicles either mature into or fuse with endosomes separately. Iron released from p97 in the endosomes escapes the late endosome through the divalent metal transporter Nramp2 and incorporated into ferritin in the cytoplasm. (B) GPI-anchored p97 and TfR are endocytosed into the same endosomal compartments. Iron released from transferrin/ transferrin receptor complex and GPI-anchored p97 escapes the late endosomes through the divalent metal transporter Nramp2 and incorporated into ferritin in the cytoplasm. 161 GPI-anchored p97 Caveolae H F E Transferrin / - N ^ - X Receptor \ /{ 0\ complex Clathrin coated pits 10 minutes Early endosomes o Fe Nramp 2 0 Caveolin EEA1 • Clathrin complex R e c y c l i n g e n d o s o m e s r l y s o s o m e s p97 to ferritin decreased in corespondence to the decrease in iron uptake due to the removal of GPI-anchored p97 by PI-PLC. Interestingly, p97" transferrin receptor deficient CHO cel line, which does not express a functional transferrin receptor, is able to accumulate iron in ferritin. Presumably, this cel line can take up iron by either internalizing non-specificaly bound 5Fe-NTA or by a specific membrane bound channel or carrier (Egyed, 1988). The data also indicated that this specific component was sensitve to PI-PLC treatment. It has been suggested that this PI-PLC sensitve component could be the GPI-anchored hamster p97 (Kennard et al., 1995). Experiments to determine whether this component is hamster p97 are in progress. In summary, the internalization pathway of iron by GPI-anchored p97 was folowed by a series of time-course confocal immunofluorescence and sub-celular fractionation experiments. From these experiments, it was determined that GPI-anchored p97 is internalized through caveolae in SK-MEL 28 cels and is subsequently traficked to endosomes and vesicles containing Nramp2. Iron internalized by GPI-anchored p97 is found in caveolae containing fractions of the sucrose gradients and is incorporated into ferritin. These results establish iron taken up by GPI-anchored p97 as a new pathway for celular iron acquisition. 163 Chapter 5 Concluding remarks and future perspectives There is renewed interest in the study of iron metabolism. The recent cloning of several proteins involved in iron metabolism including basal lateral iron transporter (IREG1), transferrin receptor 2, duodenum ferric reductase (Dcytb) and HFE, has created a major breakthrough in the understanding of the mechanisms of iron absorption, iron transport and celular regulation of iron metabolism. Iron is an essential nutrient for celular metabolism. Deficiency in iron results anemic symptoms, which includes palor, fatigue and decreased work performance. Data colected from multiple countries by the World Health Organization estimated that 50% of children and women and 25% of men in developing countries, and 7% to 12% of the population in developed countries are iron deficient (Freire, 1997). However, excess amount of iron is deleterious due to iron s role in the generation of free radicals, which can lead to tissue damage. The best-characterized pathway for iron uptake is the transferrin receptor mediated holo-transferin internalization through the clathrin-coated pit. Holo-transferin binds to transferrin receptor on the cel surface and is internalized within the endosomal compartments. The hemochromatosis protein, HFE, appears to play the role of modulating the afinity of transferrin receptor for transferrin during this process. The acidification of the endosomes prompts the release of iron from the transferrin/ transferrin receptor complex. Iron is transported across the membrane into the cytoplasm by Nramp2 and is subsequently utilzed for celular metabolism or is sequestered within the iron-storage protein, ferritin. Although the interaction between HFE and transferrin receptor seems to lower the receptor afinity for transferrin by approximately 10-fold (Feder et al., 1997; Lebron et al., 1998), the precise role of HFE remains obscure. 164 The pathology of several iron metabolism disorders has provided hints on the existence of transferrin/ transferrin receptor independent mechanisms of iron uptake. The studies of hypotransferinemic (hpx) mouse and hereditary hemochromatosis in humans, where circulating non-transferrin bound iron can reach micro-molar levels, have propeled the researchers to identify transferrin/ transferrin receptor-independent iron transport mechanisms such as iron uptake by p97 (Batey et al., 1980; Bernstein, 1987; Buys et al., 1991; Craven et al., 1987; Guteridge et al, 1985; McNamara et al., 1999). The physiological functions of p97 are just coming to light at the start of this study. GPI-anchored p97 is shown to internalize iron in p97 transfected CHO cels (Kennard et al., 1995). The association of elevated levels of serum p97 with the progression of Alzheimer s disease and the pathological roles diferent metals play in the disease have led to the hypothesis that p97 may play a role in the pathology of the disease by transporting metals related the into the brain (Kennard et al., 1996). Before establishing p97 as a candidate for metal transport into the brain, it was crucial to determine whether these metals could bind to p97. The identification of the orthologues of p97 in diferent species suggested that the function of the gene is evolutionary preserved. However, the amino acid substiutions in one of the iron binding sites (in some species, both sites are afected) has cast doubts as to whether p97 does function as an iron binding protein. In order to confirm p97 as an iron binding protein, it is important to determine whether the protein bind iron at a high affinity. Furthermore, others have suggested that p97 is not a functional iron transport protein and does not donate its iron to ferritin (Richardson and Baker, 1991a; Richardson, 2000; Richardson and Baker, 1991b). 165 The increased controversies regarding the physiological roles of p97 has led to the studies outlined in this thesis. Aluminum, copper and zinc were shown to be able to interfere with iron binding to GPI-anchored p97 by the competion experiments. In addition, iron-loading experiments with soluble p97 and these diferent metals also demonstrated that p97 could bind to metals other than iron. However, the urea polyacrylamide gel electrophoresis of soluble p97 loaded with diferent metals showed that p97 bound preferentialy to iron. Indeed the high binding afinity of iron for p97 determined from diferential scanning calorimetry demonstrated that at least one atom of iron can bind to the N-lobe of p97 very tightly at 2.2 x lO^M"1. Data from urea polyacrylamide gel electrophoresis revealed that the C-lobe of p97 appears to bind iron very weakly. The titration of iron to p97 monitored by UV-visible spectroscopy also ilustrated that the molar ratio of iron to p97 is 2:1. The failure to detect iron binding in the C-lobe using diferential scanning calorimetry could be due to the fact that the conformation of iron bound C-lobe is no diferent from iron-free C-lobe. The D421S amino acid substiution in the C-lobe of p97 may have resulted in an inefective domain closure required for tight iron binding. The observation with the D63S recombinant N-lobe transferrin also suggests that the aspartic acid to serine substiution appears to result in a failure to lock in in the iron when it is bound to the protein (He et al., 1997). Therefore, alternate method is needed to determine the iron binding constant of the C-lobe of p97. The process of crystalizing soluble p97 is underway (see Appendix I). The crystal structure of p97 wil show that whether or not the C-lobe contains iron. In 166 addition, by crystalizing zinc-bound soluble p97, it is possible to determine whether zinc binds to the proposed zinc-binding site (Garat and Jhoti, 1992). The internalization of iron bound to GPI-anchored p97 represents a unique opportunity to study another example of the uptake of a ligand bound to a GPI-anchored protein. Iron bound to GPI-anchored p97 is internalized through caveolae and is subsequently traficked through the endosomes and the Nramp2 containing vesicles before it is incorporated into ferritin in the cytoplasm. There were doubts as to whether caveolae were able to mediate internalization of any substances. Caveolae were initialy thought to form flask shape invaginations underneath the plasma membrane. These invaginations were thought to not detach from the plasma membrane and therefore some investigators suggest that caveolae are not involved in uptake of extracelular substances. However, the study of folate uptake by GPI-anchored folate receptor has altered the perception of the static nature of caveolae. The recent observations that caveolae are involved in the entry of bacteria and virus into cel have also widen the possible role of caveolae in mediating endocytosis of particles from extracelular space (Pelkmans et al., 2001; Shin et al., 2000). There are now mounting evidence suggesting caveolae, in additon to being the sites of signal integration in the celular signaling pathways, are also the sites of calcium storage and entry (Isshiki and Anderson, 1999; Lohn et al., 2000). Furthermore, SV40 viral entry and GPI-anchored mediated bacterial internalization have also been linked to caveolae (Parton and Lindsay, 1999; Pelkmans et al., 2001; Shin and Abraham, 2001a; Shin and Abraham, 2001b; Shin et al., 2000). The uptake of iron by GPI-anchored p97 through caveolae presented in this thesis represents a novel pathway of iron acquisiton and iron bound to the GPI-anchored p97 can be incorporated into ferritin. 167 The increased understanding of the role of caveolae play in celular processes has over turned the previous perception that caveolae are static vesicles reside underneath the plasma membrane. The existence of several transferrin/ transferrin receptor independent iron uptake pathways in organisms, of which p97 is one of them, raises the question as to the purpose of these pathways. The eficiency of transferrin/ transferrin receptor to deliver iron into the cels also questions the necessity of these alternate routes of iron internalization. One of the purposes of having transferrin/ transferrin receptor independent iron uptake pathway may lie with the regulation of iron uptake. Transferrin/ transferrin receptor is very eficient in amassing iron in the cels. However, this may be necessary only in red blood cels where iron is required for heme assembly or in rapidly dividing cels such as cancer cels (for review please see (Laufer, 1992). In cels that do not have high requirement for iron, transferrin/ transferrin receptor may not be the prefered route for iron uptake. For example, in brain tissue, where very few cels undergo proliferation, iron is not a major element required. The expression of p97 on brain capilary endothelium implies that it may have functional role in mediating brain iron uptake (Rothenberger et al., 1996). In addition, the amino acid substiution in human p97, which results in only one high afinity iron-binding lobe, may serve as way to limit the amount of iron internalized by certain cels. Therefore, depending on the celular need for iron, diferent iron uptake pathway may be employed and p97 may be the prefered pathway for iron uptake. The presence of transferrin/ transferrin receptor independent iron uptake pathways in specific cels suggests specialized functions of these pathways. For example, the 168 presence of Nramp2 on the apical plasma membrane of intestinal epithelial cels is necessary for ferous iron uptake. The expression of GPI-anchored p97 on the fetal pig intestinal cels implies that p97 may be involved in iron uptake during fetal development (Alemany et al., 1993). The presence of these two independent iron absorption pathways can infer diferential iron uptake. Dietary iron is in ferric form and has to be reduced by a ferric reductase (Dcytb) to the ferous form before transported by Nramp2. GPI-anchored p97, however, can bind and transport ferric iron. In addition, while Nramp2 seems to be a non-specifc divalent metal transporter, p97 binds iron preferentialy. Therefore, p97 may play a unique role in celular iron uptake in specialized cels such as polarized cels. Future studies should include the investigations of the role of p97 in intestinal iron uptake. Furthermore, p97 has been shown to be expressed in the human epithelial cels of the sweat gland. One could hypothesize that p97 could mediate iron transport in specific tissue where the environmental salinity is diferent from the rest of the body. It is clear that there are several directions this project could take in the future. For example: studies to understand whether p97 plays a role in intestinal iron uptake and iron depositon in several iron-overload disorders using mouse deficient in mouse p97 (p97 knock-out mouse) and mouse that over expresses p97. By culturing brain endothelial cel layer in a transwel apparatus, an in vitro model of the blood brain barrier can be set up to study the role of p97 in transcytosis metals across the monolayer. The identification of a potential receptor for soluble p97 wil also improve the understanding of the physiological role of p97. Although these ideas were not discussed in details, they are important in understanding the physiological role of p97. In conclusion, the results of this 169 thesis show that p97 is able to bind at least one atom of iron at a very high affinity. The C-lobe of p97 is able to bind iron as wel although the binding afinity is expected to be low. A novel route of iron uptake independent of transferrin/ transferrin receptor is also demonstrated. GPI-anchored p97 can serve as a functional iron transport protein, releasing iron into the cel for incorporation into ferritin. This data sheds significant new light on transferrin/transferrin receptor independent pathways and also on the fate of ligands internalized by GPI-anchored proteins. 170 Appendix I: Crystallization of soluble p97 Deciphering the crystal structure of p97 wil help visualizing the spatial interaction of iron with the iron-binding site of the protein. However, to solve the crystal structure of p97 is an ambitous task. The initial conditon for crystal growth has been determined and optimization of the conditons is in progress. Analysis of the X-ray difraction patern of the protein wil  be carried out by a group of expertise by colaboration. Experimental design and results: The most common setup to grow protein crystals is by the hanging drop technique. In order to pin point the exact conditon required for crystal growth, two pre-made kits were used to screen for optimal conditon (Hampton research Crystal screen 1 and Crystal screen 2 and Emerald Biostructures Wizard 1 screen and Wizard 2 screen). 3 1 of soluble p97 (17mg/ml) was mixed with 3 1 of reservoir solution containing the precipitants. This mixture was put on a glass cover slip and sealed over the reservoir. This method is based on the idea that as the protein/precipitant mixture in the drop is less concentrated than the reservoir solution, water evaporates from the drop into the reservoir. The concentration of both protein and precipitant in the drop slowly increases, and crystals may form. After 4 days at 2013C, a rectangular shape crystal was observed in the reservoir containing 20% PEG-8000, 200 mM NaCl, 100 mM CAPS pH 10.5. The crystal was orange in color, which is indicative of an iron binding protein. Measurement of the crystal showed approximate dimensions of 0.9 mm x 0.06 mm x 0.05 mm (Fig. 26). In order to optimize the conditon for crystal growth, various combinations of bufer at diferent pH and salt concentrations wil  be examined. 171 Appendix II: Derivation of the equation to calculate the binding constant in terms of enthalpy, heat capacity and temperature. The transpositon of any equilbrium constant Ki and K2 at temperature Ti to another temperature T2 is derived from Gibb s free energy formula. AG= AH(T)-TAS(T) where H(enthalpy) and S(entropy) is a variable of temperature T AG= -RTlnK where K is the equilbrium constant and R is the gas constant -RTlnK= AH(T)-TAS(T) lnK= -AH(T)/RT + TAS(T)/RT lnK= -AH(T)/R(1/T) + AS(T)/R Diferentiate with respect to T dlnK= AH(T)/R(1/T2) + 0 dT Integrate from T2 to Ti T2 T2 1 dlnK = | AH(T)/R(1/T2) dT T l T l T l AH(T)= AH(Ti)+ ACp(T-Ti) from AH= j Cp dT, where is relatively constant in a T Tl constant pressure condition.^.e. if Cp is relatively constant, then AH= J Cp dT can be T integrated to give an approximate expression for temperature dependence of AH with respect to an arbitrary reference temperature Ti) 173 T2 T2 J dlnK = J [AH(Ti)+ ACP(T-Ti)] /R(l/T2) dT Tl T l T2 T2 T2 T2 I dlnK = j [AH(Ti)/R](l/T2)dT + J ACpT/RT2dT-j ACpTi/RT2dT Tl T l T l T l lnK2/K! =-[A HCTO/R] (1/T) | \ ACp/R(lnT) I ^ + ACpT,/R (1/T) I ^ Tl Tl Tl lnK2/Ki = -[A H(Ti)/R] (1/T2-1/Ti) + ACp/R(lnT2/Ti) + ACpTi/R (1/T2-1/Ti) lnK2/Ki =-[A H(TO/R] (1/T2-1/Ti) + ACP/R (lnT2/Ti+Ti/T2-l) v l v _ -[A H(T,)/R] (1/T2-1/Ti) + ACp/R (lnT2/Ti+Ti/T2-l) lv 2 / tv l —C v _ v -[AH(T,)/R] (I/T2-I/TO + ACP/R(hT^ i+Ti^ -l) (1) 1V2— Jvic The folowing is a summary of the derivation of an equation to calculate the binding constant (Kjof a ligand L (in this thesis L is iron) to a protein P (soluble p97). Since only one binding site is observed for iron to soluble p97 binding, the binding constant K L is calculated using equation that fits 1:1 binding stoichiometry in a single unfolding transition (Brandts and Lin) The folowing abbreviations are the same as those used in Brandts and Lin. Pt = Total protein Lt = Total ligand P = Folded protein L = Free ligand P = Unfolded protein [L] Tm= Free ligand concentration at Tm 174 KL=Binding constant K = Unfolding constant Tm = temperature at which there are equal concentration of folded and unfolded in the presence of ligand. T0 = Temperature at which there are equal concentration of folded and unfolded protein in the absence of ligand. The reactions involved at equilbrium for the protein in the absence and presence of ligand are as folowed. 1 P = P K = [ P ]/[P] (2) P + L = P L K L = [ P L ] / [ P ] [ L ] (3) At equilbrium, the concentrations of folded and unfolded protein are equal. Therefore Keq(To) and Keq(Tm) are equal to 1. In the presence of ligand, at equilbrium, P =[P] + [PL] [ P ] Keq(T m ) — = 1 [P] + [PL] Substiute for [P ] and [PL] using equations (2) and (3), 175 K(Tm)[P] K(Tm) Keq(Tm)= = =1 (4) [P] + KL(Tm) [P][L] T m 1 + KL(Tm) [L]Tm KL(Tm)=(K(Tm)-l)/[L]Tm (5) According to equation (1), K(Tm) = K ( T 0 ) exp [-AH(T0)/ R(l/Tm-1/T0) + A Cp /R(LnTm/T0 + To/Tm-l)] (6) At equilbrium K(T0)=1 (i.e. at equilbrium P =P of the protein in the absence of ligand), when the folded and unfolded proteins are. in equal concentrations in the absence of ligand. Therefore, K(Tm) = exp [-AH(To)/ R(l/Tm-1/T0) + A Cp /R(lnTm/T0 + TJTm-l)] (7) Substiute (10) into (8) KL(Tm)= [exp [-AH(To)/ R(l/Tm-1/T0) + A Cp /R(lnTm/T0 + T0/Tm-1)]-1]/ [L]Tm (8) 176 Bibliography Adams, M. D., Celniker, S. E., Holt, R. A., Evans, C. A., Gocayne, J. D., Amanatides, P. G., Scherer, S. E., Li, P. W., Hoskins, R. A., Gale, R. F., et al. (2000). The genome sequence of Drosophila melanogaster, Science 287, 2185-95. Aisen, P. (1994). Iron metabolism: An evolutionary perspective. In Iron metabolism in health and disease, J. H. Brock, J. W. Haliday, M. J. Pippard, and L. W. Powel, eds. (London ; Philadelphia, Saunders), pp. x, 485. Aisen, P., and Leibman, A. (1968). The stabilty constants of the Fe3+ conalbumin complexes, Biochem Biophys Res Commun 30, 407-13. Aisen, P., and Leibman, A. (1972). Lactoferrin and transferrin: a comparative study, Biochim Biophys Acta 257, 314-23. Alemany, R., Vila, M. R., Franci, C, Egea, G., Real, F. X., and Thomson, T. M. (1993). Glycosyl phosphatidylinositol membrane anchoring of melanotransferin (p97): apical compartmentalization in intestinal epithelial cels, J Cel Sci 104, 1155-62. Alikmets, R., Raskind, W. H., Hutchinson, A., Schueck, N. D., Dean, M., and Koeler, D. M. (1999). Mutation of a putative mitochondrial iron transporter gene (ABC7) in X-linked sideroblastic anemia and ataxia (XLSA/A), Hum Moi Genet 8, 743-9. Anderson, R. G. (1994). Functional specialization of the glycosylphosphatidylinositol membrane anchor, Semin Immunol 6, 89-95. Anderson, R. G. (1998). The caveolae membrane system, Annu Rev Biochem 67, 199-225. Anderson, R. G., Kamen, B. A., Rothberg, K. G., and Lacey, S. W. (1992). Potocytosis: sequestration and transport of smal molecules by caveolae, Science 255, 410-1. 177 Andrews, N. C. (1999). Disorders of iron metabolism, N Engl J Med 341, 1986-95. Angelini, L., Nardocci, N., Rumi, V., Zorzi, C, Strada, L., and Savoiardo, M. (1992). Halervorden-Spatz disease: clinical and MRI study of 11 cases diagnosed in life, J Neurol 239,417-25. Archibald, F. (1983). Lactobacilus plantarum, and organism not requiring iron, FEMS Microbiology Leters 19, 29-32. Asako, H., Kurose, I., Wolf, R. E., and Granger, D. N. (1994). Mechanisms of lactoferrin-induced leukocyte-endothelial cel adhesion in postcapilary venules, Microcirculation 1, 27-34. Atwood, C. S., Huang, X., Khatri, A., Scarpa, R. C., Kim, Y. S., Moir, R. D., Tanzi, R. E., Roher, A. E., and Bush, A. I. (2000). Copper catalyzed oxidation of Alzheimer Abeta, Cel Mol Biol (Noisy-le-grand) 46, 777-83. Babcock, M., de Silva, D., Oaks, R, Davis-Kaplan, S., Jiralerspong, S., Montermini, L., Pandolfo, M., and Kaplan, J. (1997). Regulation of mitochondrial iron accumulation by Yfhlp, a putative homolog of frataxin, Science 276,1709-12. Bailey, S., Evans, R. W., Garrat, R. C, Gorinsky, B., Hasnain, S., Horsburgh, C, Jhoti, H, Lindley, P. F., Mydin, A., Sarra, R, and et al. (1988). Molecular structure of serum fransferrin at 3.3-A resolution, Biochemistry 27, 5804-12. Baker, E. N., Baker, H. M., Smith, C. A., Stebbins, M. R., Kahn, M., Helstrom, K. E., and Helstrom, I. (1992). Human melanotransferin (p97) has only one functional iron-binding site, FEBS Let  298, 215-8. Baker, E. N., and Lindley, P. F. (1992). New perspectives on the structure and function of transferrins, J Inorg Biochem 47, 147-60. 178 Baranano, D. E., Wolosker, H., Bae, B. I., Barow, R. K., Snyder, S. H., and Ferris, C. D. (2000). A mammalin iron ATPase induced by iron, J Biol Chem 275, 15166-73. Barresi, G., and Tuccari, G. (1994). Immunocytochemical demonstration of melanotransferin (p97) in thyroid tumors of folicular cel origin, Pathology 26, 127-9. Barton, E. H., West, P. A., Rivers, C. A., Barton, J. C, and Acton, R. T. (2001). Transferin receptor-2 (tfr2) mutation y250x in alabama Caucasian and african american subjects with and without primary iron overload, Blood Cels Moi Dis 27, 279-84. Barton, J. C, Sawada-Hirai, R., Rothenberg, B. E., and Acton, R. T. (1999). Two novel missense mutations of the HFE gene (I105T and G93R) and identification of the S65C mutation in Alabama hemochromatosi  probands, Blood Cels Moi Dis 25, 147-55. Batey, R. G., Lai Chung Fong, P., Shamir, S., and Sherlock, S. (1980). A non-transferin-bound serum iron in idiopathic hemochromatosis, Dig Dis Sci 25, 340-6. Baynes, R. D., and Bothwel, T. H. (1990). Iron deficiency, Annu Rev Nutr 10, 133-48. Beard, J. L., Connor, J. R., and Jones, B. C. (1993). Iron in the brain, Nutr Rev 51, 157-70. Beard, J. L., Dawson, H., and Pinero, D. J. (1996). Iron metabolism: a comprehensive review, Nutr Rev 54, 295-317. Beinert, H. (1976). Iron-sulfur proteins, the most numerous and most diversified components of the mitochondrial electron transfer system, Adv Exp Med Biol 74, 137-49. Bernstein, S. E. (1987). Hereditary hypotransferinemia with hemosiderosis, a murine disorder resembling human atransferinemia, J Lab Clin Med 110, 690-705. 179 Beutler, E., Gelbart, T., Lee, P., Trevino, R., Fernandez, M. A., and Fairbanks, V. F. (2000). Molecular characterization of a case of atransferinemia, Blood 96, 4071-4. Brandts, J. F., and Lin, L. N. (1990). Study of strong to ultratight protein interactions using diferential scanning calorimetry, Biochemistry 29, 6927-40. Britigan, B. E., Serody, J. S., and Cohen, M. S. (1994). The role of lactoferrin as an anti-inflammatory molecule, Adv Exp Med Biol 357, 143-56. Brown, D. A., and London, E. (1998). Functions of lipid rafts in biological membranes, Annu Rev Cel Dev Biol 14,111-36. Brown, D. A., and Rose, J. K. (1992). Sorting of GPI-anchored proteins to glycolipid-enriched membrane subdomains during transport to the apical cel surface, Cel 68, 533-44. Brown, J. P., Woodbury, R. G., Hart, C. E., Helstrom, I., and Helstrom, K. E. (1981). Quantiative analysis of melanoma-associated antigen p97 in normal and neoplastic tissues, Proc Natl Acad Sci U S A 78, 539-43. Burdo, J. R., Martin, J., Menzies, S. L., Dolan, K. G., Romano, M. A., Fletcher, R. J., Garrick, M. D., Garrick, L. M., and Connor, J. R. (1999). Celular distribution of iron in the brain of the Belgrade rat, Neuroscience 93, 1189-96. Bush, A. I. (2000). Metals and neuroscience, Cur Opin Chem Biol 4, 184-91. Bush, A. I., Petingel, W. H, Multhaup, G., d Paradis, M., Vonsatel, J. P., Gusela, J. F., Beyreuther, K., Masters, C. L., and Tanzi, R. E. (1994). Rapid induction of Alzheimer A beta amyloid formation by zinc [see comments], Science 265, 1464-7. Buys, S. S., Martin, C. B., Eldridge, M., Kushner, J. P., and Kaplan, J. (1991). Iron absorption in hypotransferinemic mice, Blood 78, 3288-90. 180 Cairo, G., and Pietrangelo, A. (2000). Iron regulatory proteins in pathobiology, Biochem J352Pt2, 241-50. Camaschela, C, Fargion, S., Sampietro, M., Roeto, A., Bosio, S., Garozzo, G., Arosio, C, and Piperno, A. (1999). Inherited HFE-unrelated hemochromatosi  in Italian familes, Hepatology 29, 1563-4. Camaschela, C, Roeto, A., Cali, A., De Gobbi, M., Garozzo, G., Carela, M., Majorano, N., Totaro, A., and Gasparini, P. (2000). The gene TFR2 is mutated in a new type of haemochromatosi  mapping to 7q22, Nat Genet 25, 14-5. Camaschela, C, Roeto, A., Cicilano, M., Pasquero, P., Bosio, S., Gubeta, L., Di Vito, F., Gireli, D., Totaro, A., Carela, M., et al. (1997). Juvenile and adult hemochromatosis are distinct genetic disorders, Eur J Hum Genet 5, 371-5. Campbel, T., Skilton, R. A., Coombes, R. C, Shousha, S., Graham, M. D., and Luqmani, Y. A. (1992). Isolation of a lactoferrin cDNA clone and its expression in human breast cancer, Br J Cancer 65,19-26. Campuzano, V., Montermini, L., Lutz, Y., Cova, L., Hindelang, C, Jiralerspong, S., Trotier, Y., Kish, S. J., Faucheux, B., Trouilas, P., et al. (1997). Frataxin is reduced in Friedreich ataxia patients and is associated with mitochondrial membranes, Hum Moi Genet 6, 1771-80. Campuzano, V., Montermini, L., Molto, M. D., Pianese, L., Cossee, M., Cavalcanti, F., Monros, E., Rodius, F., Duclos, F., Monticeli, A., and et al. (1996). Friedreich's ataxia: autosomal recessive disease caused by an intronic GAA triplet repeat expansion [see comments], Science 271, 1423-7. 181 Canonne-Hergaux, F., Gruenheid, S., Ponka, P., and Gros, P. (1999). Celular and subcelular localization of the Nramp2 iron transporter in the intestinal brush border and regulation by dietary iron, Blood 93, 4406-17. Cazzola, M., Bergamaschi, G., Tonon, L., Arbustini, E., Grasso, M., Vercesi, E., Barosi, G., Bianchi, P. E., Cairo, G., and Arosio, P. (1997). Hereditary hyperferitnemia-cataract syndrome: relationship between phenotypes and specific mutations in the iron-responsive element of ferritin light-chain mRNA, Blood 90, 814-21. Cerneus, D. P., Uefing, E., Posthuma, G., Strous, G. J., and van der Ende, A. (1993). Detergent insolubility of alkaline phosphatase during biosynthetic transport and endocytosis. Role of cholesterol, J Biol Chem 268, 3150-5. Chen, Y., and Norkin, L. C. (1999). Extracelular simian virus 40 transmits a signal that promotes virus enclosure within caveolae, Exp Cel Res 246, 83-90. Colomb, E., Pianeta, C, Estevenon, J. P., Guy, O., Figarela, C, and Sarles, H. (1976). Lactoferrin in human pancreas. Immunohistological localization in normal and pathological pancreatic tissues, Digestion 14, 242-9. Congiu-Castelano, A., Boffi, F., Delia Longa, S., Giovanneli, A., Girasole, M., Natali, F., Pompa, M., Soldatov, A., and Bianconi, A. (1997). Aluminum site structure in serum transferrin and lactoferrin revealed by synchrotron radiation X-ray spectroscopy, Biometals 10, 363-7. Conrad, M. E., and Umbreit, J. N. (1993). A concise review: iron absorption—the mucin-mobilferrin-integrin pathway. A competive pathway for metal absorption, Am J Hematol 42, 67-73. 182 Conrad, M. E., Umbreit, J. N., Moore, E. G., Uzel, C, and Berry, M. R. (1994). Alternate iron transport pathway. Mobilferrin and integrin in K562 cels, J Biol Chem 269, 7169-73. Conrad, M. E., Umbreit, J. N., Peterson, R. D., Moore, E. G., and Harper, K. P. (1993). Function of integrin in duodenal mucosal uptake of iron, Blood 81, 517-21. Corsi, B., Levi, S., Cozzi, A., Corti, A., Altimare, D., Albertini, A., and Arosio, P. (1999). Overexpression of the hereditary hemochromatosis protein, HFE, in HeLa cels induces and iron-deficient phenotype, FEBS Let  460,149-52. Craven, C. M., Alexander, J., Eldridge, M., Kushner, J. P., Bernstein, S., and Kaplan, J. (1987). Tissue distribution and clearance kinetics of non-transferin-bound iron in the hypotransferinemic mouse: a rodent model for hemochromatosis, Proc Natl Acad Sci U S A 84, 3457-61. Crouch, S. P., Slater, K. J., and Fletcher, J. (1992). Regulation of cytokine release from mononuclear cels by the iron-binding protein lactoferrin, Blood 80, 235-40. Csere, P., Lill, R., and Kispal, G. (1998). Identification of a human mitochondrial ABC transporter, the functional orthologue of yeast Atmlp, FEBS Let  441, 266-70. Cumberbatch, M., Dearman, R. J., Uribe-Luna, S., Headon, D. R., Ward, P. P., Conneely, O. M., and Kimber, I. (2000). Regulation of epidermal Langerhans cel migration by lactofeirin, Immunology 100, 21-8. Curtis, A. R., Fey, C, Morris, C. M., Bindof, L. A., Ince, P. G., Chinnery, P. F., Coulthard, A., Jackson, M. J., Jackson, A. P., McHale, D. P., et al. (2001). Mutation in the gene encoding ferritin light polypeptide causes dominant adult-onset basal ganglia disease, Nat Genet 28, 350-4. 183 Dalman, P. R. (1986). Biochemical basis for the manifestations of iron deficiency, Annu RevNutrt), 13-40. de Silva, D. M., Askwith, C. C, and Kaplan, J. (1996). Molecular mechanisms of iron uptake in eukaryotes, Physiol Rev 76, 31-47. Deckert, M., Ticchioni, M., and Bernard, A. (1996). Endocytosis of GPI-anchored proteins in human lymphocytes: role of glycolipid-based domains, actin cytoskeleton, and protein kinases, J Cel Biol 133, 791-9. Delatycki, M. B., Wilamson, R., and Forrest, S. M. (2000). Friedreich ataxia: an overview, J Med Genet 37, 1-8. Derijard, B., Hibi, M., Wu, I. H., Barret, T., Su, B., Deng, T., Karin, M., and Davis, R. J. (1994). JNK1: a protein kinase stimulated by UV light and Ha-Ras that binds and phosphorylates the c-Jun activation domain, Cel  76, 1025-37. Devary, Y., Gotlieb, R. A., Lau, L. F., and Karin, M. (1991). Rapid and preferential activation of the c-jun gene during the mammalin UV response, Mol Cel Biol 77, 2804-11. Donovan, A., Brownlie, A., Zhou, Y., Shepard, J., Prat, S. J., Moynihan, J., Paw, B. H., Drejer, A., Barut, B., Zapata, A., et al. (2000). Positional cloning of zebrafish ferroportinl identifies a conserved vertebrate iron exporter [see comments], Nature 403, 776-81. Duchange, N., Ochoa, A., Plowman, G. D., Roze, A., Amdjadi, M., and Zakin, M. M. (1992). Identification of an enhancer involved in the melanoma-specifc expression of the tumor antigen melanotransferin gene, Nucleic Acids Res 20, 2853-9. 184 Dupree, P., Parton, R. G., Raposo, G., Kurzchalia, T. V., and Simons, K. (1993). Caveolae and sorting in the trans-Golgi network of epithelial cels, Embo J 12, 1597-605. Dwork, A. J., Schon, E. A., and Herbert, J. (1988). Nonidentical distribution of transferin and ferric iron in human brain, Neuroscience 27, 333-45. Ebbel, B. (1937). The Papyrus Ebers, the greatest Egyptian medical document (Copenhagen, London, Levin & Munksgaard; H. Milford Oxford university press). Edwards, C. Q., Grifen, L. M., Ajioka, R. S., and Kushner, J. P. (1998). Screening for hemochromatosis: phenotype versus genotype, Semin Hematol 35, 72-6. Edwards, C. Q., Grifen, L. M., Goldgar, D., Drummond, C, Skolnick, M. H., and Kushner, J. P. (1988). Prevalence of hemochromatosi  among 11,065 presumably healthy blood donors, N Engl J Med 318, 1355-62. Egyed, A. (1988). Carier mediated iron transport through erythroid cel membrane, Br J Haematol 68,483-6. Eisenstein, R. S. (2000). Iron regulatory proteins and the molecular control of mammalin iron metabolism, Annu Rev Nutr 20, 627-62. Evans, R. W., and Wilams, J. (1978). Studies of the binding of diferent iron donors to human serum transferrin and isolation of iron-binding fragments from the N- and C-terminal regions of the protein, Biochem J 173, 543-52. Faber, H. R., Bland, T., Day, C. L., Norris, G. E., Tweedie, J. W., and Baker, E. N. (1996). Altered domain closure and iron binding in transferrins: the crystal structure of 185 the Asp60Ser mutant of the amino-terminal half-molecule of human lactoferrin, J Moi Biol 256, 352-63. Faucheux, B. A., Herero, M. T., Vilares, J., Levy, R., Javoy-Agid, F., Obeso, J. A., Hauw, J. J., Agid, Y., and Hirsch, E. C. (1995). Autoradiographic localization and density of [125I]ferrotransferrin binding sites in the basal ganglia of control subjects, patients with Parkinson's disease and MPTP-lesioned monkeys, Brain Res 691, 115-24. Feder, J. N., Gnirke, A., Thomas, W., Tsuchihashi, Z., Ruddy, D. A., Basava, A., Dormishian, F., Domingo, R., Jr., Elis, M. C, Fulan, A., et al. (1996). A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis, Nat Genet 13, 399-408. Feder, J. N., Tsuchihashi, Z., Irrinki, A., Lee, V. K., Mapa, F. A., Morikang, E., Prass, C. E., Starnes, S. M., Wolf, R. K., Parkkila, S., et al. (1997). The hemochromatosis founder mutation in HLA-H disrupts beta2- microglobulin interaction and cel surface expression, J Biol Chem 272, 14025-8. Ferris, C. D., Jafrey, S. R., Sawa, A., Takahashi, M., Brady, S. D., Barow, R. K., Tysoe, S. A., Wolosker, H., Baranano, D. E., Dore, S., et al. (1999). Haem oxygenase-1 prevents cel death by regulating celular iron, Nat Cel Biol 1, 152-7. Fielding, C. J., and Fielding, P. E. (2000). Cholesterol and caveolae: structural and functional relationships, Biochim Biophys Acta 1529, 210-22. Fielding, P. E., and Fielding, C. J. (1996). Intracelular transport of low density lipoprotein derived free cholesterol begins at clathrin-coated pits and terminates at cel surface caveolae, Biochemistry 35, 14932-8. 186 Fleming, M. D., Romano, M. A., Su, M. A., Garrick, L. M., Garrick, M. D., and Andrews, N. C. (1998). Nramp2 is mutated in the anemic Belgrade (b) rat: evidence of a role for Nramp2 in endosomal iron transport, Proc Natl Acad Sci U S A 95,1148-53. Fleming, R. E., Migas, M. C., Holden, C. C, Waheed, A., Briton, R. S., Tomatsu, S., Bacon, B. R., and Sly, W. S. (2000). Transferrin receptor 2: continued expression in mouse liver in the face of iron overload and in hereditary hemochromatosis, Proc Natl Acad SciUS AP7, 2214-9. Fleming, R. E., Migas, M. C., Zhou, X., Jiang, J., Briton, R. S., Brunt, E. M., Tomatsu, S., Waheed, A., Bacon, B. R., and Sly, W. S. (1999). Mechanism of increased iron absorption in murine model of hereditary hemochromatosis: increased duodenal expression of the iron transporter DMT1, Proc Natl Acad Sci U S A 96, 3143-8. Food, M. R. (1992) The human melanoma associated antigen p97 is atached to the plasma membrane by a glycosylphosphatidylinositol anchor, M.Sc, University of British Columbia, Vancouver. Food, M. R., Rothenberger, S., Gabathuler, R., Haidl, I. D., Reid, G., and Jeferies, W. A. (1994). Transport and expression in human melanomas of a transferrin-like glycosylphosphatidylinositol-anchored protein, J Biol Chem 269, 3034-40. Freire, W. B. (1997). Strategies of the Pan American Health Organization/World Health Organization for the control of iron deficiency in Latin America, Nutr Rev 55, 183-8. Funk, W. D., MacGilivray, R. T., Mason, A. B., Brown, S. A., and Woodworth, R. C. (1990). Expression of the amino-terminal half-molecule of human serum transferrin in cultured cels and characterization of the recombinant protein, Biochemistry 29, 1654-60. 187 Gacy, A. M., Goelner, G. M., Spiro, C., Chen, X., Gupta, G., Bradbury, E. M., Dyer, R. B., Mikesel, M. J., Yao, J. Z., Johnson, A. J., et al. (1998). GAA instability in Friedreich's Ataxia shares a common, DNA-directed and intraalelic mechanism with other trinucleotide diseases, Mol Cel 1, 583-93. Gangaidzo, I. T., Moyo, V. M., Saungweme, T., Khumalo, H., Charakupa, R. M., Gomo, Z. A., Loyevsky, M., Stearman, R., La Vaute, T., Enquist, E. G., et al. (1999). Iron overload in urban Africans in the 1990s, Gut 45, 278-83. Garrat, R. C, and Jhoti, H. (1992). A molecular model for the tumour-associated antigen, p97, suggests a Zn-binding function, FEBS Let  305, 55-61. Gireli, D., Corocher, R., Bisceglia, L., Olivieri, O., Zelante, L., Panozzo, G., and Gasparini, P. (1997). Hereditary hyperferitnemia-cataract syndrome caused by a 29-base pair deletion in the iron responsive element of ferritin L-subunit gene, Blood 90, 2084-8. Gitlin, J. D. (1998). Aceruloplasminemia, Pediatr Res 44, 271-6. Glenney, J. R, Jr. (1989). Tyrosine phosphorylation of a 22-kDa protein is corelated with transformation by Rous sarcoma virus, J Biol Chem 264, 20163-6. Glenney, J. R., Jr., and Zokas, L. (1989). Novel tyrosine kinase substrates from Rous sarcoma virus-transformed cels are present in the membrane skeleton, J Cel Biol 108, 2401-8. Gorodinsky, A., and Harris, D. A. (1995). Glycolipid-anchored proteins in neuroblastoma cels form detergent- resistant complexes without caveolin, J Cel Biol 129, 619-27. Grasbeck, R., Majuri, R., Kouvonen, I., and Tenhunen, R. (1982). Spectral and other studies on the intestinal haem receptor of the pig, Biochim Biophys Acta 700, 137-42. 188 Griffiths, W. J., Kely, A. L., Smith, S. J., and Cox, T. M . (2000). Localization of iron transport and regulatory proteins in human cels, Ojm 93, 575-87. Grootveld, M., Bel, J. D., Haliwel, B., Aruoma, O. I., Bomford, A., and Sadler, P. J. (1989). Non-transferin-bound iron in plasma or serum from patients with idiopathic hemochromatosis. Characterization by high performance liquid chromatography and nuclear magnetic resonance spectroscopy, J Biol Chem 264,4417-22. Gross, C. N., Irrinki, A., Feder, J. N., and Enns, C. A. (1998). Co-traficking of HFE, a nonclassical major histocompatibilty complex class I protein, with the transferrin receptor implies a role in intracelular iron regulation, J Biol Chem 273, 22068-74. Grune, T., Reinheckel, T., and Davies, K. J. (1997). Degradation of oxidized proteins in mammalin cels, Faseb J 11, 526-34. Gunshin, H., Mackenzie, B., Berger, U. V., Gunshin, Y., Romero, M. F., Boron, W. F., Nussberger, S., Golan, J. L., and Hediger, M. A. (1997). Cloning and characterization of a mammalin proton-coupled metal-ion transporter, Nature 388, 482-8. Guteridge, J. M., Rowley, D. A., Griffiths, E., and Haliwel, B. (1985). Low-molecular-weight iron complexes and oxygen radical reactions in idiopathic haemochromatosis, Clin Sci (Colch) 68, A62>-1. Haber, F., and Weiss, J. (1934). The catalytic decompositon of hydrogen peroxide by iron salts, Proc R Soc London Ser A 147. Haile, D. J. (1999). Regulation of genes of iron metabolism by the iron-response proteins, Am J Med Sci 318, 230-40. Hamil, R. L., Woods, J. C, and Cook, B. A. (1991). Congenital atransferinemia. A case report and review of the literature, Am J Clin Pathol 96, 215-8. 189 Harney, J. H., Doyle, D., Brown, V., and Rogers, M. S. (1995). The celular isoform of the prion protein, PrPc, is associated with caveolae in mouse neuroblastoma (N2a) cels, Biochem Biophys Res Commun 210, 753-9. Harington, J. P., Stuart, J., and Jones, A. (1987). Unfolding of iron and copper complexes of human lactoferrin and transferrin, Int J Biochem 19, 1001-8. Harris, W. R. (1983). Thermodynamic binding constants of the zinc-human serum transferrin complex, Biochemistry 22, 3920-6. Harris, W. R., and Stenback, J. Z. (1988). The bicarbonate-dependence of zinc(I)-transferrin binding, J Inorg Biochem 33, 211-23. Harris, Z. L., Klomp, L. W., and Gitlin, J. D. (1998). Aceruloplasminemia: an inherited neurodegenerative disease with impairment of iron homeostasis, Am J Clin Nutr 67, 972S-977S. Harris, Z. L., Takahashi, Y., Miyajima, H., Serizawa, M., MacGilivray, R. T., and Gitlin, J. D. (1995). Aceruloplasminemia: molecular characterization of this disorder of iron metabolism, Proc Natl Acad Sci U S A 92, 2539-43. Haversen, L. A., Engberg, I., Baltzer, L., Dolphin, G., Hanson, L. A., and Matsby-Baltzer, I. (2000). Human lactoferrin and peptides derived from a surface-exposed helical region reduce experimental Escherichia coli urinary tract infection in mice, Infect Immun 68, 5816-23. Haynie (2001). Biological Thermodynamics (Cambridge, Cambridge University Press). He, Q. Y., Mason, A. B., Woodworth, R. C, Tarn, B. M., Wadsworth, T., and MacGilivray, R. T. (1997). Efects of mutations of aspartic acid 63 on the metal-binding 190 properties of the recombinant N-lobe of human serum transferrin, Biochemistry 36, 5522-8. Henley, J. R., Cao, H., and McNiven, M. A. (1999). Participation of dynamin in the biogenesis of cytoplasmic vesicles, Faseb J 13 Suppl 2, S243-7. Henley, J. R., Krueger, E. W., Oswald, B. J., and McNiven, M. A. (1998). Dynamin-mediated internalization of caveolae, J Cel Biol 141, 85-99. Hill, J. M., and Switzer, R. C, 3rd (1984). The regional distribution and celular localization of iron in the rat brain, Neuroscience 11, 595-603. Hooper, N. M. (1999). Detergent-insoluble glycosphingolipidVcholesterol-rich membrane domains, lipid rafts and caveolae (review), Moi Membr Biol 16, 145-56. Hsu, F. (1997) Studies on members of the transferrin family of proteins in humans and miceudies on members of transferrin family, M.Sc, University of British Columbia, Vancouver. Huggenvik, J. I., Craven, C. M., Idzerda, R. L., Bernstein, S., Kaplan, J., and McKnight, G. S. (1989). A splicing defect in the mouse transferrin gene leads to congenital atransferinemia, Blood 74,482-6. Hutchens, T. W., and Yip, T. T. (1991). Metal ligand-induced alterations in the surface structures of lactoferrin and transferrin probed by interaction with immobilzed copper(I) ions, J Chromatogr 536, 1-15. Ibrahim, H. R., Iwamori, E., Sugimoto, Y., and Aoki, T. (1998). Identification of a distinct antibacterial domain within the N-lobe of ovotransferin, Biochim Biophys Acta 1401, 289-303. 191 Ibrahim, H. R., Sugimoto, Y., and Aoki, T. (2000). Ovotransferin antimicrobial peptide (OTAP-92) kils bacteria through a membrane damage mechanism, Biochim Biophys Acta 1523,196-205. Ide, C, and Saito, T. (1980). Electron microscopic histochemistry of ATPase and alkaline phosphatase activities in mouse digital corpuscles, J Neurocytol 9, 207-18. Ilangumaran, S., and Hoessli, D. C. (1998). Efects of cholesterol depletion by cyclodextrin on the sphingolipid microdomains of the plasma membrane, Biochem J 335, 433-40. Inoue, M., Yamada, J., Kitamura, N., Shimazaki, K., Andren, A., and Yamashita, T. (1993). Immunohistochemical localization of lactoferrin in bovine exocrine glands, Tissue Cel 25, 791-7. Isshiki, M., and Anderson, R. G. (1999). Calcium signal transduction from caveolae, Cel Calcium 26, 201-8. Jankovic, J., Kirkpatrick, J. B., Blomquist, K. A., Langlais, P. J., and Bird, E. D. (1985). Late-onset Halervorden-Spatz disease presenting as familial parkinsonism, Neurology 35, 227-34. Janssen, P. T., and van Bijsterveld, O. P. (1983). Origin and biosynthesis of human tear fluid proteins, Invest Ophthalmol Vis Sci 24, 623-30. Jeferies, W. A., Food, M. R., Gabathuler, R., Rothenberger, S., Yamada, T., Yasuhara, O., and McGeer, P. L. (1996). Reactive microglia specificaly associated with amyloid plaques in Alzheimer's disease brain tissue express melanotransferin, Brain Res 712, 122-6. 192 Jemmerson, R., Klier, F. G., and Fishman, W. H. (1985). Clustered distribution of human placental alkaline phosphatase on the surface of both placental and cancer cels. Electron microscopic observations using gold-labeled antibodies, J Histochem Cytochem 33, 1227-34. Jemmerson, R., and Low, M. G. (1987). Phosphatidylinositol anchor of HeLa cel alkaline phosphatase, Biochemistry 26, 5703-9. Johal, J. S., Thorp, J. W., and Oyer, C. E. (1998). Neonatal hemochromatosis, renal tubular dysgenesis, and hypocalvaria in a neonate, Pediatr Dev Pathol / , 433-7. Jonas, L., Kruger, B., and Tessenow, W. (1993). Immunohistochemical detection of lactoferrin in diferent human glandular tissues with special reference to the exocrine pancreas, Acta Histochem 95, 53-9. Kamen, B. A., Wang, M. T., Streckfuss, A. J., Peryea, X., and Anderson, R. G. (1988). Delivery of folates to the cytoplasm of MA 104 cels is mediated by a surface membrane receptor that recycles, J Biol Chem 263, 13602-9. Karuzina, II, and Archakov, A. I. (1994). Hydrogen peroxide-mediated inactivation of microsomal cytochrome P450 during monooxygenase reactions, Free Radic Biol Med 17, 557-67. Kasvosve, I., Gangaidzo, I. T., Gomo, Z. A., and Gordeuk, V. R. (2000). African iron overload, Acta Clin Belg 55, 88-93. Kawabata, H, Germain, R. S., Vuong, P. T., Nakamaki, T., Said, J. W., and Koefler, H. P. (2000). Transferrin receptor 2-alpha supports cel growth both in iron-chelated cultured cels and in vivo, J Biol Chem 275, 16618-25. 193 Kawabata, H., Yang, R, Hirama, T., Vuong, P. T., Kawano, S., Gombart, A. F., and Koefler, H. P. (1999). Molecular cloning of transferrin receptor 2. A new member of the transferrin receptor-like family, J Biol Chem 274, 20826-32. Kawamoto, T., Pan, H., Yan, W., Ishida, FL, Usui, E., Oda, R., Nakamasu, K., Noshiro, M., Kawashima-Ohya, Y., Fuji, M., et al. (1998). Expression of membrane-bound transferrin-like protein p97 on the cel surface of chondrocytes, Eur J Biochem 256, 503-9. Kennard, M. L., Feldman, H., Yamada, T., and Jeferies, W. A. (1996). Serum levels of the iron binding protein p97 are elevated in Alzheimer's disease, Nat Med 2, 1230-5. Kennard, M. L., Richardson, D. R., Gabathuler, R., Ponka, P., and Jeferies, W. A. (1995). A novel iron uptake mechanism mediated by GPI-anchored human p97, Embo J 74,4178-86. Kim, D. K., Seo, M. Y., Lim, S., Kim, S., Kim, J., Carrol, B. J., Do Yoon, K., Kwon, T., and Kang, S. S. (2001). Serum Melanotransferin, p97 as a Biochemical Marker of Alzheimer's Disease, Neuropsychopharmacology 25, 84-90. Kispal, G., Csere, P., Guiard, B., and Lill, R. (1997). The ABC transporter Atmlp is required for mitochondrial iron homeostasis, FEBS Let  418, 346-50. Klomp, L. W., Farhangrazi, Z. S., Dugan, L. L., and Gitlin, J. D. (1996). Ceruloplasmin gene expression in the murine central nervous system [see comments], J Clin Invest 98, 207-15. Koutnikova, H., Campuzano, V., Foury, F., Dole, P., Cazzalini, O., and Koenig, M. (1997). Studies of human, mouse and yeast homologues indicate a mitochondrial function for frataxin, Nat Genet 16, 345-51. 194 Kriegerbeckova, K., and Kovar, J. (2000). Role of melanotransferin (p97) in non-transferin iron uptake by HeLa and K562 cels, Folia Biol 46, 77-81. Kurokawa, H., Mikami, B., and Hirose, M. (1995). Crystal structure of diferric hen ovotransferin at 2.4 A resolution, J Moi Biol 254, 196-207. Kurose, I., Yamada, T., Wolf, R., and Granger, D. N. (1994). P-selectin-dependent leukocyte recruitment and intestinal mucosal injury induced by lactoferrin, J Leukoc Biol 55, 771-7. Kurzchalia, T. V., Dupree, P., and Monier, S. (1994). VIP21-Caveolin, a protein of the trans-Golgi network and caveolae, FEBS Let  346, 88-91. Lange, H., Kispal, G., and Lill, R. (1999). Mechanism of iron transport to the site of heme synthesis inside yeast mitochondria, J Biol Chem 274, 18989-96. Lange, S. J., and Que, L., Jr. (1998). Oxygen activating nonheme iron enzymes, Cur Opin Chem Biol 2, 159-72. Laufer, R. B. (1992). Iron and human disease (Boca Raton, CRC Press). Lebron, J. A., Bennet, M. J., Vaughn, D. E., Chirino, A. J., Snow, P. M., Mintier, G. A., Feder, J. N., and Bjorkman, P. J. (1998). Crystal structure of the hemochromatosis protein HFE and characterization of its interaction with transferrin receptor, Cel  93, 111-23. Leighton, J., and Schatz, G. (1995). An ABC transporter in the mitochondrial inner membrane is required for normal growth of yeast, Embo J 14,188-95. Lenarcic, B., Krishnan, G., Borukhovich, R., Ruck, B., Turk, V., and Moczydlowski, E. (2000). Saxiphiln, a saxitoxin-binding protein with two thyroglobulin type 1 domains, is an inhibitor of papain-like cysteine proteinases, J Biol Chem 275, 15572-7. 195 Li, S., Song, K. S., and Lisanti, M. P. (1996). Expression and characterization of recombinant caveolin. Purification by polyhistidine tagging and cholesterol-dependent incorporation into defined lipid membranes, J Biol Chem 271, 568-73. Liang, Z., Sotrup-Jensen, L., Aspan, A., Hal, M., and Soderhal, K. (1997). Pacifastin, a novel 155-kDa heterodimeric proteinase inhibitor containing a unique transferrin chain, Proc Natl Acad Sci U S A 94, 6682-7. Lieu, P. T., Heiskala, M., Peterson, P. A., and Yang, Y. (2001). the roles of iron in health and disease, Mol Aspects Med 22, 1-87. Lill, R., Diekert, K., Kaut, A., Lange, H., Pelzer, W., Prohl, C, and Kispal, G. (1999). The essential role of mitochondria in the biogenesis of celular iron- sulfur proteins, Biol Chem 380, 1157-66. Lin, L. N., Mason, A. B., Woodworm, R. C, and Brandts, J. F. (1993). Calorimetric studies of the N-terminal half-molecule of transferrin and mutant forms modifed near the Fe(3+)-binding site, Biochem J 293, 517-22. Lin, L. N., Mason, A. B., Woodworm, R. C, and Brandts, J. F. (1994). Calorimetric studies of serum transferrin and ovotransferin. Estimates of domain interactions, and study of the kinetic complexites of ferric ion binding, Biochemistry 33, 1881-8. Lindahl, T., and Wood, R. D. (1999). Quality control by DNA repair, Science 286, 1897-905. Liochev, S. I. (1999). The mechanism of "Fenton-like" reactions and their importance for biological systems. A biologist's view, Met Ions Biol Syst 36, 1-39. Lisanti, M. P., Scherer, P. E., Vidugiriene, J., Tang, Z., Hermanowski-Vosatka, A., Tu, Y. H., Cook, R. F., and Sargiacomo, M. (1994). Characterization of caveolin-rich 196 membrane domains isolated from an endothelial-rich source: implications for human disease, J Cel Biol 126, 111-26. Lisanti, M. P., Tang, Z. L., and Sargiacomo, M. (1993). Caveolin forms a hetero-oligomeric protein complex that interacts with an apical GPI-linked protein: implications for the biogenesis of caveolae, J Cel Biol 123, 595-604. Liu, P., and Anderson, R. G. (1995). Compartmentalized production of ceramide at the cel surface, J Biol Chem 270, 27179-85. Liu, P., Ying, Y., and Anderson, R. G. (1997). Platelet-derived growth factor activates mitogen-activated protein kinase in isolated caveolae, Proc Natl Acad Sci USA 94, 13666-70. Liu, P., Ying, Y., Ko, Y. G., and Anderson, R. G. (1996). Localization of platelet-derived growth factor-stimulated phosphorylation cascade to caveolae, J Biol Chem 277, 10299-303. Lohn, M., Furstenau, M., Sagach, V., Elger, M., Schulze, W., Luft, F. C, Haler, H., and Golasch, M. (2000). Ignition of calcium sparks in arterial and cardiac muscle through caveolae, Circ Res 87, 1034-9. Loske, C, Gerdemann, A., Schepl, W., Wycislo, M., Schinzel, R, Palm, D., Riederer, P., and Munch, G. (2000). Transiton metal-mediated glycoxidation accelerates cross-linking of beta-amyloid peptide, Eur J Biochem 267,4171-8. Lovel, M. A., Robertson, J. D., Teesdale, W. J., Campbel, J. L., and Markesbery, W. R. (1998). Copper, iron and zinc in Alzheimer's disease senile plaques, J Neurol Sci 158, 47-52. 197 Lynch, T., Cherny, R. A., and Bush, A. I. (2000). Oxidative processes in Alzheimer's disease: the role of abeta-metal interactions, Exp Gerontol 35,445-51. Maines, M. D. (1997). The heme oxygenase system: a regulator of second messenger gases, Annu Rev Pharmacol Toxicol 37, 517-54. Majuri, R., and Grasbeck, R. (1987). A rosete receptor assay with haem-microbeads. Demonstration of a haem receptor on K562 cels, Eur J Haematol 38, 21-5. Makey, D. G., and Seal, U. S. (1976). The detection of four molecular forms of human transferrin during the iron binding process, Biochim Biophys Acta 453, 250-6. Makiya, R., Thornel, L. E., and Stigbrand, T. (1992). Placental alkaline phosphatase, a GPI-anchored protein, is clustered in clathrin-coated vesicles, Biochem Biophys Res Commun 183, 803-8. Marcus, D. L., Ibrahim, N. G., Gruenspecht, N., and Freedman, M. L. (1980). Iron requirement for isolated rat liver mitochondrial protein synthesis, Biochim Biophys Acta 607, 136-44. Mastick, C. C, and Saltiel, A. R. (1997). Insulin-stimulated tyrosine phosphorylation of caveolin is specific for the diferentiated adipocyte phenotype in 3T3-L1 cels, J Biol Chem 272, 20706-14. Mayor, S., Rothberg, K. G., and Maxfield, F. R. (1994). Sequestration of GPI-anchored proteins in caveolae triggered by cross- linking, Science 264, 1948-51. McCord, J. M. (1998). Iron, free radicals, and oxidative injury, Semin Hematol 35, 5-12. McGraw, T. E., Greenfield, L., and Maxfield, F. R. (1987). Functional expression of the human transferrin receptor cDNA in Chinese hamster ovary cels deficient in endogenous transferrin receptor, J Cel Biol 105, 207-14. 198 McKie, A., Barow, D., Latunde-Dada, G., Rolfs, A., Sager, G., Mudaly, E., Mudaly, M., Richardson, C., Barlow, D., Bomford, A., et al. (2001). An iron-regulated ferric reductase associated with the absorption of dietary iron (sciencexpress). McKie, A. T., Marciani, P., Rolfs, A., Brennan, K., Wehr, K., Barow, D., Miret, S., Bomford, A., Peters, T. J., Farzaneh, F., et al. (2000). A novel duodenal iron-regulated transporter, IREG1, implicated in the basolateral transfer of iron to the circulation, Mol Cel 5, 299-309. McNagny, K. M., Rossi, F., Smith, G., and Graf, T. (1996). The eosinophil-specifc cel surface antigen, EOS47, is a chicken homologue of the oncofetal antigen melanotransferin, Blood 87, 1343-52. McNamara, L., MacPhail, A. P., Mandishona, E., Bloom, P., Paterson, A. C, Rouault, T. A., and Gordeuk, V. R. (1999). Non-transferin-bound iron and hepatic dysfunction in African dietary iron overload, J Gastroenterol Hepatol 14, 126-32. Mikulits, W., Schranzhofer, M., Beug, H., and Mulner, E. W. (1999). Post-transcriptional control via iron-responsive elements: the impact of aberations in hereditary disease, Mutat Res 437,219-30. Miura, T., Suzuki, K., Kohata, N., and Takeuchi, H. (2000). Metal binding modes of Alzheimer's amyloid beta-peptide in insoluble aggregates and soluble complexes, Biochemistry 39, 7024-31. Moir, R. D., Atwood, C. S., Huang, X., Tanzi, R. E., and Bush, A. I. (1999). Mounting evidence for the involvement of zinc and copper in Alzheimer's disease, Eur J Clin Invest 29, 569-70. 199 Monier, S., Parton, R. G., Vogel, F., Behlke, J., Henske, A., and Kurzchalia, T. V. (1995). VIP21 -caveolin, a membrane protein constiuent of the caveolar coat, oligomerizes in vivo and in vitro, Moi Biol Cel 6, 911-27. Morgan, E. H. (1988). Membrane transport of non-transferin-bound iron by reticulocytes, Biochim Biophys Acta 943, 428-39. Moroo, I., Ujie, M., Tiong, J. W. C, Walker, B. L., and Jeferies, W. A. (2001). The role of p97 in iron transport across the blood-brain barrier, manuscript in preperation. Mura, C, Raguenes, O., and Ferec, C. (1999). HFE mutations analysis in 711 hemochromatosis probands: evidence for S65C implication in mild form of hemochromatosis, Blood 93, 2502-5. Murata, M., Peranen, J., Schreiner, R., Wieland, F., Kurzchalia, T. V., and Simons, K. (1995). VIP21/caveolin is a cholesterol-binding protein, Proc Natl Acad Sci U S A 92, 10339-43. Nakamasu, K., Kawamoto, T., Shen, M., Gotoh, O., Teramoto, M., Noshiro, M., and Kato, Y. (1999). Membrane-bound transferrin-like protein (MTf): structure, evolution and selective expression during chondrogenic diferentiation of mouse embryonic cels, Biochim Biophys Acta 1447, 258-64. Neumann, M., Adler, S., Schluter, O., Kremmer, E., Benecke, R., and Kretzschmar, H. A. (2000). Alpha-synuclein accumulation in a case of neurodegeneration with brain iron accumulation type 1 (NBIA-1, formerly Halervorden-Spatz syndrome) with widespread cortical and brainstem-type Lewy bodies, Acta Neuropathol (Berl) 100, 568-74. 200 Nichols, B. J., Kenworthy, A. K., Polishchuk, R. S., Lodge, R., Roberts, T. H., Hirschberg, K., Phair, R. D., and Lippincot-Schwartz, J. (2001). Rapid cycling of lipid raft markers between the cel surface and Golgi complex, J Cel Biol 153, 529-41. Olynyk, J. K., Culen, D. J., Aquilia, S., Rossi, E., Summervile, L., and Powel, L. W. (1999). A population-based study of the clinical expression of the hemochromatosis gene, N Engl J Med 347,718-24. Orlandi, P. A., and Fishman, P. H. (1998). Filpin-dependent inhibition of cholera toxin: evidence for toxin internalization and activation through caveolae-like domains, J Cel Biol 141, 905-15. Otsuki, K., Yoda, A., Toma, Y., Shimizu, Y., Saito, H., and Yanaihara, T. (1998). Lactoferrin and interleukin-6 interaction in amniotic infection, Adv Exp Med Biol 443, 267-71. Palade, G. E. (1953). J Appl Phys 24, 1424. Palau, F. (2001). Friedreich's ataxia and frataxin: Molecular genetics, evolution and pathogenesis (Review), Int J Moi Med 7, 581-9. Parkkila, S., Waheed, A., Briton, R. S., Bacon, B. R., Zhou, X. Y., Tomatsu, S., Fleming, R. E., and Sly, W. S. (1997). Association of the transferrin receptor in human placenta with HFE, the protein defective in hereditary hemochromatosis, Proc Natl Acad Sci US A 94, 13198-202. Parton, R. G., Joggerst, B., and Simons, K. (1994). Regulated internalization of caveolae, JCel Biol 127, 1199-215. Parton, R. G., and Lindsay, M. (1999). Exploitation of major histocompatibilty complex class I molecules and caveolae by simian virus 40, Immunol Rev 168, 23-31. 201 Pelkmans, L., Kartenbeck, J., and Helenius, A. (2001). Caveolar endocytosis of simian virus 40 reveals a new two-step vesicular-transport pathway to the ER, Nat Cel Biol 3, 473-83. Penhalow, R. C, Mason, A. B., and Woodworth, R. C. (1991). Electrophoretic characterization of human, equine and bovine transferrins, Comp Biochem Physiol [B] 96,41-5. Perl, D. P., and Pendlebury, W. W. (1986). Aluminum neurotoxicity—potential role in the pathogenesis of neurofibrilary tangle formation, Can J Neurol Sci 13, 441-5. Pike, L. J., and Casey, L. (1996). Localization and turnover of phosphatidylinositol 4,5-bisphosphate in caveolin-enriched membrane domains, J Biol Chem 271, 26453-6. Ponka, P. (1999). Cel biology of heme, Am J Med Sci 318, 241-56. Poss, K. D., and Tonegawa, S. (1997a). Heme oxygenase 1 is required for mammalin iron reutilzation, Proc Natl Acad Sci U S A 94, 10919-24. Poss, K. D., and Tonegawa, S. (1997b). Reduced stress defense in heme oxygenase 1-deficient cels, Proc Natl Acad Sci U S A 94,10925-30. Poussin, C, Foti, M., Carpentier, J. L., and Pugin, J. (1998). CD14-dependent endotoxin internalization via a macropinocytic pathway, J Biol Chem 273, 20285-91. Puccio, H., Simon, D., Cossee, M., Criqui-Filipe, P., Tiziano, F., Melki, J., Hindelang, C, Matyas, R., Rustin, P., and Koenig, M. (2001). Mouse models for Friedreich ataxia exhibit cardiomyopathy, sensory nerve defect and Fe-S enzyme deficiency folowed by intramitochondrial iron deposits, Nat Genet 27, 181-6. 202 Radisky, D. C, Babcock, M. C, and Kaplan, J. (1999). The yeast frataxin homologue mediates mitochondrial iron eflux. Evidence for a mitochondrial iron cycle, J Biol Chem 274,4497-9. Rafin, S. B., Woo, C. FL, Roost, K. T., Price, D. C, and Schmid, R. (1974). Intestinal absorption of hemoglobin iron-heme cleavage by mucosal heme oxygenase, J Clin Invest 54, 1344-52. Rainard, P. (1993). Activation of the classical pathway of complement by binding of bovine lactoferrin to unencapsulated Streptococcus agalactiae, Immunology 79, 648-52. Recklinghausen, F. v. (1889). Ueber Haemachromatose, Vol 26). Reid, R. A., Moyle, J., and Mitchel, P. (1966). Synthesis of adenosine triphosphate by a protonmotive force in rat liver mitochondria, Nature 212, 257-8. Rich, P. R. (1984). Electron and proton transfers through quinones and cytochrome be complexes, Biochim Biophys Acta 768, 53-79. Richardson, D., and Baker, E. (1991a). The uptake of inorganic iron complexes by human melanoma cels, Biochim Biophys Acta 1093,20-8. Richardson, D. R. (2000). The role of the membrane-bound tumour antigen, melanotransferin (p97), in iron uptake by the human malignant melanoma cel, Eur J Biochem 267, 1290-8. Richardson, D. R., and Baker, E. (1991b). The release of iron and transferrin from the human melanoma cel, Biochim Biophys Acta 1091, 294-302. Riedel, H. D., Muckenthaler, M. U., Gehrke, S. G., Mohr, I., Brennan, K., Hermann, T., Fitscher, B. A., Hentze, M. W., and Stremmel, W. (1999). HFE downregulates iron 203 uptake from transferrin and induces iron- regulatory protein activity in stably transfected cels, Blood 94, 3915-21. Rodal, S. K., Skreting, G., Gared, O., Vilhardt, F., van Deurs, B., and Sandvig, K. (1999). Extraction of cholesterol with methyl-beta-cyclodextrin perturbs formation of clathrin-coated endocytic vesicles, Mol Biol Cel  10, 961-7'4. Roeto, A., Totaro, A., Cazzola, M., Cicilano, M., Bosio, S., D'Ascola, G., Carela, M., Zelante, L., Kely, A. L., Cox, T. M., et al. (1999). Juvenile hemochromatosis locus maps to chromosome lq, Am J Hum Genet 64, 1388-93. Rose, T. M., Plowman, G. D., Teplow, D. B., Dreyer, W. J., Helstrom, K. E., and Brown, J. P. (1986). Primary structure of the human melanoma-associated antigen p97 (melanotransferin) deduced from the mRNA sequence, Proc Natl Acad Sci U S A 83, 1261-5. Roskams, A. J., and Connor, J. R. (1990). Aluminum access to the brain: a role for transferrin and its receptor, Proc Natl Acad Sci U S A 87, 9024-7. Rothberg, K. G., Heuser, J. E., Donzel, W. C, Ying, Y. S., Glenney, J. R., and Anderson, R. G. (1992). Caveolin, a protein component of caveolae membrane coats, Cel  68, 673-82. Rothberg, K. G., Ying, Y. S., Kolhouse, J. F., Kamen, B. A., and Anderson, R. G. (1990). The glycophospholipid-linked folate receptor internalizes folate without entering the clathrin-coated pit endocytic pathway, J Cel Biol 110, 637-49. Rothenberger, S., Food, M. R., Gabathuler, R., Kennard, M. L., Yamada, T., Yasuhara, O., McGeer, P. L., and Jeferies, W. A. (1996). Coincident expression and distribution of 204 melanotransferin and transferrin receptor in human brain capilary endothelium, Brain Res 712, 117-21. Rotig, A., de Lonlay, P., Chretien, D., Foury, F., Koenig, M., Sidi, D., Munnich, A., and Rustin, P. (1997). Aconitase and mitochondrial iron-sulphur protein deficiency in Friedreich ataxia, Nat Genet 77, 215-7. Roy, C. N., Penny, D. M., Feder, J. N., and Enns, C. A. (1999). The hereditary hemochromatosi  protein, HFE, specificaly regulates transferin-mediated iron uptake in HeLa cels, J Biol Chem 274, 9022-8. Ruoslahti, E. (1996). RGD and other recogniton sequences for integrins, Annu Rev Cel Dev Biol 12, 697-715. Sahlin, M., and Sjoberg, B. M. (2000). Ribonucleotide reductase. A virtual playground for electron transfer reactions, Subcel Biochem 35,405-43. Saito, K., and Nakanuma, Y. (1992). Lactoferrin and lysozyme in the intrahepatic bile duct of normal livers and hepatolithiasis. An immunohistochemical study, J Hepatol 15, 147-53. Salter-Cid, L., Brunmark, A., Li, Y., Leturcq, D., Peterson, P. A., Jackson, M. R., and Yang, Y. (1999). Transferin receptor is negatively modulated by the hemochromatosis protein HFE: implications for celular iron homeostasis, Proc Natl Acad Sci U S A 96, 5434-9. Salter-Cid, L., Brunmark, A., Peterson, P. A., and Yang, Y. (2000a). The major histocompatibilty complex-encoded class I-like HFE abrogates endocytosis of transferrin receptor by inducing receptor phosphorylation, Genes Immun 7, 409-17. 205 Salter-Cid, L., Peterson, P. A., and Yang, Y. (2000b). The major histocompatibilty complex-encoded HFE in iron homeostasis and immune function, Immunol Res 22, 43-59. Sasaki, N., Toki, S., Chowei, H, Saito, T., Nakano, N., Hayashi, Y., Takeuchi, M., and Makita, Z. (2001). Immunohistochemical distribution of the receptor for advanced glycation end products in neurons and astrocytes in Alzheimer's disease, Brain Res 888, 256-262. Sayre, L. M., Perry, G., Atwood, C. S., and Smith, M. A. (2000). The role of metals in neurodegenerative diseases, Cel Mol Biol (Noisy-le-grand) 46, 731-41. Schade, A. L., and Caroline, L. (1946). An iron-binding component in human blood plasma, Science 104, 340. Scheifele, P., Verkade, P., Fra, A. M., Virta, H, Simons, K., and Ikonen, E. (1998). Caveolin-1 and -2 in the exocytic pathway of MDCK cels, J Cel Biol 140, 795-806. Schelman, J. A. (1975). Macromolecular Binding, Biopolymers 14, 1019-1026. Scherer, P. E., Lewis, R. Y., Volonte, D., Engelman, J. A., Galbiati, F., Couet, J., Kohtz, D. S., van Donselaar, E., Peters, P., and Lisanti, M. P. (1997). Cel-type and tissue-specific expression of caveolin-2. Caveolins 1 and 2 co-localize and form a stable hetero-oligomeric complex in vivo, J Biol Chem 272, 29337-46. Scherer, P. E., Okamoto, T., Chun, M., Nishimoto, I., Lodish, H. F., and Lisanti, M. P. (1996). Identification, sequence, and expression of caveolin-2 defines a caveolin gene family, Proc Natl Acad Sci U S A 93,131-5. Schipper, H. M. (1999). Glial HO-1 expression, iron depositon and oxidative stress in neurodegenerative diseases, Neurotoxicity Res 1, 57-70. 206 Schipper, H. M. (2000). Heme oxygenase-1: role in brain aging and neurodegeneration, Exp Gerontol 35, 821-30. Schipper, H. M., Chertkow, H., Mehindate, K., Frankel, D., Melmed, C., and Bergman, H. (2000). Evaluation of heme oxygenase-1 as a systemic biological marker of sporadic AD, Neurology 54, 1297-304. Schnitzer, J. E., Oh, P., and Mcintosh, D. P. (1996). Role of GTP hydrolysis in fission of caveolae directly from plasma membranes, Science 274, 239-42. Sciot, R., de Vos, R., van Eyken, P., van der Steen, K., Moerman, P., and Desmet, V. J. (1989). In situ localization of melanotransferin (melanoma-associated antigen P97) in human liver. A light- and electronmicroscopic immunohistochemical study, Liver 9, 110-9. Sekyere, E., and Richardson, D. R. (2000). The membrane-bound transferrin homologue melanotransferin: roles other than iron transport?, FEBS Let  483, 11-6. Senbongi, H., Ling, F., and Shibata, T. (1999). A mutation in a mitochondrial ABC transporter results in mitochondrial dysfunction through oxidative damage of mitochondrial DNA, Moi Gen Genet 262, 426-36. Sheldon, J. H. (1935). Haemochromatosis (London, Oxford University Press H. Milford). Shin, J. S., and Abraham, S. N. (2001a). Co-option of endocytic functions of celular caveolae by pathogens, Immunology 102, 2-7. Shin, J. S., and Abraham, S. N. (2001b). Glycosylphosphatidylinositol-anchored receptor-mediated bacterial endocytosis, FEMS Microbiol Let  197, 131-8. 207 Shin, J. S„ Gao, Z., and Abraham, S. N. (2000). Involvement of celular caveolae in bacterial entry into mast cels, Science 289, 785-8. Shyng, S. L., Heuser, J. E., and Harris, D. A. (1994). A glycolipid-anchored prion protein is endocytosed via clathrin-coated pits, J Cel Biol 125, 1239-50. Simionescu, M., and Simionescu, N. (1991). Endothelial transport of macromolecules: transcytosis and endocytosis. A look from cel biology, Cel Biol Rev 25, 5-78. Simons, K., and Toomre, D. (2000). Lipid Rafts and Signal Transduction, Nat Rev Mol Cel Biol 7, 31-39. Skreting, G., Torgersen, M. L., van Deurs, B., and Sandvig, K. (1999). Endocytic mechanisms responsible for uptake of GPI-linked diphtheria toxin receptor, J Cel Sci 112, 3899-909. Smart, E. J., Estes, K., and Anderson, R. G. (1995). Inhibitors that block both the internalization of caveolae and the return of plasmalemmal vesicles, Cold Spring Harb Symp Quant Biol 60, 243-8. Smart, E. J., Mineo, C, and Anderson, R. G. (1996). Clustered folate receptors deliver 5-methyltetrahydrofolate to cytoplasm of MA104 cels, J Cel Biol 134, 1169-77. Smith, C. A., Anderson, B. F., Baker, H. M., and Baker, E. N. (1992). Metal substiution in transferrins: the crystal structure of human copper-lactoferrin at 2.1-A resolution, Biochemistry 31, 4527-33. Sofic, E., Riederer, P., Heinsen, H, Beckmann, H., Reynolds, G. P., Hebenstreit, G., and Youdim, M. B. (1988). Increased iron (III) and total iron content in post mortem substantia nigra of parkinsonian brain, J Neural Transm 74, 199-205. 208 Song, K. S., Scherer, P. E., Tang, Z., Okamoto, T., Li, S., Chafel, M., Chu, C, Kohtz, D. S., and Lisanti, M. P. (1996). Expression of caveolin-3 in skeletal, cardiac, and smooth muscle cels. Caveolin-3 is a component of the sarcolemma and co-fractionates with dystrophin and dystrophin-associated glycoproteins, J Biol Chem 277, 15160-5. Spik, G., and Montreuil, J. (1983). The role of lactotransferrin in the molecular mechanisms of antibacterial defense, Bul Eur Physiopathol Respir 19, 123-30. Stohs, S. J. (1995). The role of free radicals in toxicity and disease, J Basic Clin Physiol Pharmacol 6, 205-28. Stohs, S. J., and Bagchi, D. (1995). Oxidative mechanisms in the toxicity of metal ions, Free Radic Biol Med 18, 321-36. Stubbe, J. (2000). Ribonucleotide reductases: the link between an RNA and a DNA world?, Cur Opin Struct Biol 10, 731-6. Subtil, A., Gaidarov, I., Kobylarz, K., Lampson, M. A., Keen, J. H., and McGraw, T. E. (1999). Acute cholesterol depletion inhibits clathrin-coated pit budding, Proc Natl Acad Sci U S A 96, 6775-80. Tabuchi, M., Yoshimori, T., Yamaguchi, K., Yoshida, T., and Kishi, F. (2000). Human NRAMP2/DMT1, which mediates iron transport across endosomal membranes, is localized to late endosomes and lysosomes in HEp-2 cels, J Biol Chem 275, 22220-8. Tang, S., MacCol, R., and Parsons, P. J. (1995). Spectroscopic study ofthe interaction of aluminum ions with human transferrin, J Inorg Biochem 60, 175-85. Tang, Z., Scherer, P. E., Okamoto, T., Song, K., Chu, C, Kohtz, D. S., Nishimoto, I., Lodish, H. F., and Lisanti, M. P. (1996). Molecular cloning of caveolin-3, a novel 209 member of the caveolin gene family expressed predominantly in muscle, J Biol Chem 277,2255-61. Tong, W. H., and Rouault, T. (2000). Distinct iron-sulfur cluster assembly complexes exist in the cytosol and mitochondria of human cels, Embo J 19, 5692-700. Trenor, C. C, 3rd, Campagna, D. R., Selers, V. M., Andrews, N. C, and Fleming, M. D. (2000). The molecular defect in hypotransferinemic mice, Blood 96, 1113-8. Trousseau, A. (1865). Glycosurie, diab te sucr . In Clinique M dicale de l'H tel-Dieu de Paris, pp. 663-98. Turnbul, A., Cleton, F., and Finch, C. A. (1989). The Journal of clinical Investigation, Volume 41, 1962: iron absorption. IV. The absorption of hemoglobin iron, Nutr Rev 47, 51-3. Ueda, F., Raja, K. B., Simpson, R. J., Trowbridge, I. S., and Bradbury, M. W. (1993). Rate of 59Fe uptake into brain and cerebrospinal fluid and the influence thereon of antibodies against the transferrin receptor, J Neurochem 60, 106-13. van Rensburg, S. J., Potocnik, F. C, De Viliers, J. N., Kotze, M. J., and Taljaard, J. J. (2000). Earlier age of onset of Alzheimer's disease in patients with both the transferrin C2 and apolipoprotein E-epsilon 4 aleles, Ann N Y Acad Sci 903, 200-3. Vohra, P., Haler, C, Emre, S., Magid, M., Holzman, I., Ye, M. Q., Iofel, E., and Shneider, B. L. (2000). Neonatal hemochromatosis: the importance of early recogniton of liver failure, J Pediatr 136, 537-41. Vulpe, C. D., Kuo, Y. M., Murphy, T. L., Cowley, L., Askwith, C, Libina, N., Gitschier, J., and Anderson, G. J. (1999). Hephaestin, a ceruloplasmin homologue implicated in So intestinal iron transport, is defective in the sla mouse, Nat Genet 21, 195-9. 210 Waheed, A., Parkkila, S., Zhou, X. Y., Tomatsu, S., Tsuchihashi, Z., Feder, J. N., Schatzman, R. C, Briton, R. S., Bacon, B. R., and Sly, W. S. (1997). Hereditary hemochromatosis: efects of C282Y and H63D mutations on association with beta2-microglobulin, intracelular processing, and cel surface expression of the HFE protein in COS-7 cels, Proc Natl Acad Sci U S A 94, 12384-9. Walker, B. L., Kennard, M. L., Gabathuler, R., and Jeferies, W. A. (1999). Paper presented at: Proceedings of the World Congress on Iron Metabolism (Sorento, Naples, Italy). West, A. P., Jr., Bennet, M. J., Selers, V. M., Andrews, N. C, Enns, C. A., and Bjorkman, P. J. (2000). Comparison of the interactions of transferrin receptor and transferrin receptor 2 with transferrin and the hereditary hemochromatosi  protein HFE, J Biol Chem 275, 38135-8. Wilson, R. B., and Roof, D. M. (1997). Respiratory deficiency due to loss of mitochondrial DNA in yeast lacking the frataxin homologue, Nat Genet 16, 352-7. Wong, A., Yang, J., Cavadini, P., Gelera, C, Lonnerdal, B., Taroni, F., and Cortopassi, G. (1999). The Friedreich's ataxia mutation confers celular sensitivity to oxidant stress which is rescued by chelators of iron and calcium and inhibitors of apoptosis, Hum Moi Genet 8,425-30. Woodbury, R. G., Brown, J. P., Loop, S. M., Helstrom, K. E., and Helstrom, I. (1981). Analysis of normal neoplastic human tissues for the tumor-associated protein p97, Int J Cancer 27, 145-9. 211 Woodbury, R. G., Brown, J. P., Yeh, M. Y., Helstrom, I., and Helstrom, K. E. (1980). Identification of a cel surface protein, p97, in human melanomas and certain other neoplasms, Proc Natl Acad Sci U S A 77, 2183-7. Wuebbens, M. W., Roush, E. D., Decastro, C. M., and Fierke, C. A. (1997). Cloning, sequencing, and recombinant expression of the porcine inhibitor of carbonic anhydrase: a novel member of the transferrin family, Biochemistry 36,4327-36. Yamada, E. (1955). J Biophys Biochem Cytol 1, 445-58. Yamada, T., Tsujioka, Y., Taguchi, J., Takahashi, M., Tsuboi, Y., Moroo, I., Yang, J., and Jeferies, W. A. (1999). Melanotransferin is produced by senile plaque-associated reactive microglia in Alzheimer's disease, Brain Res 845, 1-5. Yan, S. D., Chen, X., Fu, J., Chen, M., Zhu, H., Roher, A., Slatery, T., Zhao, L., Nagashima, M., Morser, J., et al. (1996). RAGE and amyloid-beta peptide neurotoxicity in Alzheimer's disease, Nature36\2, 685-91. Yang, J. (2000) Deletion of the GPI pre-anchor sequence in Mtf as a mechanism for generating the soluble form of the protein, M.Sc, University of British Columbia, Vancouver. Yoon, S. J., Koh, Y. H., Floyd, R. A., and Park, J. W. (2000). Copper, zinc superoxide dismutase enhances DNA damage and mutagenicity induced by cysteine/iron, Mutat Res 448, 97-104. Yu, J., Yu, Z. K., and Wessling-Resnick, M. (1998). Expression of SFT (stimulator of Fe transport) is enhanced by iron chelation in HeLa cels and by hemochromatosis in liver, J Biol Chem 273, 34675-8. 212 Zhou, B., Westaway, S. K., Levinson, B., Johnson, M. A., Gitschier, J., and Hayflick, S. J. (2001). A novel pantothenate kinase gene (PANK2) is defective in Halervorden- Spatz syndrome, Nat Genet 28, 345-9. Zurzolo, C., van't Hof, W., van Meer, G., and Rodriguez-Boulan, E. (1994). VIP21/caveolin, glycosphingolipid clusters and the sorting of glycosylphosphatidylinositol-anchored proteins in epithelial cels, Embo J 13,42-53. 213 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090775/manifest

Comment

Related Items