Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Biology and receptor interactions of P97 and the transferrin receptors Walker, Brandie Laurel 2002

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

Item Metadata


831-ubc_2002-751613.pdf [ 12.93MB ]
JSON: 831-1.0090706.json
JSON-LD: 831-1.0090706-ld.json
RDF/XML (Pretty): 831-1.0090706-rdf.xml
RDF/JSON: 831-1.0090706-rdf.json
Turtle: 831-1.0090706-turtle.txt
N-Triples: 831-1.0090706-rdf-ntriples.txt
Original Record: 831-1.0090706-source.json
Full Text

Full Text

B I O L O G Y A N D RECEPTOR INTERACTIONS OF P97 A N D TFIE T R A N S F E R R I N RECEPTORS  by  BRANDIE L A U R E L W A L K E R  B . S c , University of Calgary, 1994  A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY In T H E F A C U L T Y OF G R A D U A T E STUDIES Department of Zoology and the Biotechnology Laboratory We accept this thesis as conforming to the required standard  T H E UNIVERSITY OF BRITISH C O L U M B I A June 2002 © Brandie Laurel Walker 2002  ln  presenting  degree freely  at  this  the  available  copying  of  department publication  of  in  partial  fulfilment  University  of  British  Columbia,  for  this or  thesis  reference  thesis by  this  for  his thesis  and  scholarly  or for  her  Department  D a t e  DE-6  (2/88)  3 l / o l  Columbia  I  further  purposes  gain  the  requirements  1 agree  shall  that  agree  may  representatives.  financial  permission.  T h e U n i v e r s i t y o f British Vancouver, Canada  study.  of  It not  be is  that  the  Library  permission  granted  by  understood be  for  allowed  an  advanced  shall for  the that  without  head  make  it  extensive of  my  copying  or  my  written  Abstract  Melanotransferrin, or p97, is an iron binding protein that is expressed as both a glycosylphosphatidylinositol-anchored form and as a soluble form. While the anchored form internalizes iron, the function of the soluble form is still unknown. Soluble p97 levels are increased in the serum and cerebral spinal fluid of Alzheimer disease patients, but the reasons for this increase are undetermined. In order to begin to address the question of function for this soluble protein, possible receptor interactions were studied. The interaction of p97 and transferrin receptor 1 was characterized with radioligand assays and immunofluorescent labeling assays. These experiments demonstrated that p97 interacts with the transferrin receptor 1 in cell binding experiments. However, p97 was not able to deliver iron into the cells via transferrin receptor 1. Furthermore, BIAcore studies were not able to measure any interaction between p97 and transferrin receptor 1. In the search for other likely candidate receptors of p97, a novel homologue of the transferrin receptor 1 was discovered through mining of the EST database. Expression of this protein was revealed by Northern blot to be largely restricted to the liver. In embryogenesis, the mouse transferrin receptor 2 is present by E15 and continues to increase in expression through E17, in contrast to transferrin receptor 1 that is discernable by E7, increases until E l 5 and decreases by E l 7 . Transferrin receptor 2 is present on the brain endothelial cells that form the blood-brain barrier implicating it as a candidate receptor for transport of p97 (and iron) into or out of the brain. Interestingly, in  transfected cells that over express both transferrin receptor 1 and 2, the receptor is present at the cell surface as a heterodimeric combination of the two receptors. p97 binds to the mouse transferrin receptor 2, and unlike transferrin receptor 1, also facilitates the uptake of Fe through this interaction. Thus, in addition to receptor 55  binding, the functional aspect of this interaction can be demonstrated.  Clearly,  identification of transferrin receptor 2 as a receptor of p97 is only one of the first important steps toward the ultimate goal of clarifying the function of soluble p97.  iii  Table of Contents Abstract  ii  Table of Contents  iv  List of Figures  viii  List,of Tables  x  List of Abbreviations  xi  Acknowledgement and Dedication  xiv  Chapter 1:  1  Introduction  I.  General introduction  1  II.  Iron absorption in the intestine  1  A.  Divalent metal transporter-1  2  B.  Heme absorption  6  C.  Parraferritin  7  D.  Ferroportin and hephaestin  8  E.  Regulation of iron uptake  9  III.  IV.  Serum iron transport proteins  11  A.  Transferrin  11  B.  Ceruloplasmin  12  C.  Melanotransferrin (p97)  14  Proteins involved in iron uptake  17  A.  Transferrin receptor 1  17  B.  HFE  24  iv  V.  C.  Transferrin receptor 2  28  D.  Stimulator of iron transport  30  Hypothesis and general approach  Chapter 2:  31  Materials and methods  35  I.  Cells and antibodies  35  II.  Binding of I-p97 and I-transferrin to transferrin receptors 1 and 2 125  125  36  A.  Iodination of proteins  36  B.  Binding to transferrin receptors  39  III.  Pandex experiments  40  IV.  Protein dialysis and iron loading  41  V.  Iron uptake studies  43  VI.  Immunoprecipitations  43  VII.  Surface plasmon resonance (BIAcore) experiments  46  VIII.  Northern blot analysis  47  IX.  Reverse transcriptase and polymerase chain reaction  48  X.  Immunofluorescence staining and confocal laser scanning microscopy  49  XI.  Statistical analysis  51  Chapter 3:  Interaction of soluble p97 with transferrin receptor 1 and integrin a (3 v  I.  52  3  Cellular based assays to examine p97 binding to transferrin receptor 1  52  A.  Rationale  52  B.  Results I:  C.  Results II: Particle concentration fluorescent immunoassay study  125  I p97 binding to the human transferrin receptor  v  55  of p97 binding to transferrin receptor 1 D.  E. II.  57  Results III: p97 can compete with transferrin for binding to the transferrin receptor 1  60  Discussion  62  Uptake of radioactive iron bound p97 by transferrin receptor 1  65  A.  Rationale  65  B.  Results 1: Fe -p97 uptake by cells expressing transferrin 55  receptor 1 C.  D. III.  66  Results II: Immunoprecipitation of p97, transferrin receptor 1, and ferritin  66  Discussion  69  Surface plasmon resonance studies to measure binding of soluble p97 to soluble transferrin receptor 1 and the integrin a (3 v  70  3  A.  Rationale  70  B.  Results I: p97 binding to soluble transferrin receptor 1 measured by BIAcore  76  C.  Results II: p97 binding to soluble integrin a (3  D.  Discussion  Chapter 4:  v  3  83  Characterization of the novel receptor transferrin receptor 2 including interaction with p97  I.  79  88  Tissue expression of transferrin receptor 2  89  A.  Rationale  89  B.  Results I: Northern blot of multiple human organs  91  vi  C.  Results II: Mouse embryo Northern  D.  Results III: Transferrin receptor 2 on human blood-brain  E. II.  93  barrier endothelial cells  93  Discussion  96  Subcellular localization  105  A.  Rationale  105  B.  Results I: Confocal immunofluorescence  106  C.  Results II: Western blot of heterodimers of transferrin receptors land 2  D. III.  Discussion  112  Cloning, constructs and transfections of mouse transferrin receptor 2  115  A.  Rationale  115  B.  Results I: Sequence of mouse transferrin receptor 2  C.  Results II: Myc-tagged mouse transferrin receptor 2 construct and  D. IV.  110  ,115  transfections  120  Discussion  122  Uptake of Fe -loaded p97 by mouse transferrin receptor 2  123  A.  Rationale  123  B.  Results 1: Fe -p97 uptake by mouse transferrin receptor 2  123  C.  Results II: Fe-p97 binding by mouse transferrin receptor 2  125  D.  Discussion  125  Conclusions and future directions  129  55  Chapter 5:  55  55  Bibliography  135  vii  List of Figures Figure 1.1: Flowchart of iron absorption pathway from the intestine  4  Figure 1.2: Model of the transferrin receptor 1 complex  19  Figure 3.1: Model of internalization of p97 via transferrin receptor 1 Figure 3.2:  125  53  Ip97 binding to transferrin receptor 1, with unlabeled p97  competition  56  Figure 3.3: Binding of soluble p97 to transferrin receptor 1, detected with the p97 antibody Hyb-C-FITC  59  Figure 3.4: Binding of transferrin to transferrin receptor 1, with and without competition by excess p97  61  Figure 3.5: Uptake of Fe-p97 by cells with and without transferrin receptor 1 55  67  Figure 3.6: Immunoprecipitation of ferritin, p97 and transferrin receptor 1 from cells after incubation with Fe-loaded p97 or transferrin  68  Figure 3.7: Diagram of surface plasmon resonance detection system  73  55  Figure 3.8: Sensorgram of transferrin and p97 binding to soluble transferrin receptor 1  78  Figure 3.9: Sensorgram of fibronectin and p97 binding to integrin a f3 v  3  Figure 4.1: Human multiple tissue Northern blot of transferrin receptor 2  81 92  Figure 4.2: Mouse embryo Northern blot of transferrin receptor 1, transferrin receptor 2, and (3-actin  94  Figure 4.3: Transferrin receptor 2 in brain endothelial cells  95  Figure 4.4: Uptake of I-p97 and I-Tf in vivo  101  125  125  Figure 4.5: Visualization of p97 and transferrin uptake by the brain  viii  103  Figure 4.6: Subcellular localization of transferrin receptor 2 in transfected human melanoma cells, SK-Mel-28  108  Figure 4.7: Western blot of heterodimers of transferrin receptors 1 and 2  111  Figure 4.8: Sequence alignment of the transferrin receptors  117  Figure 4.9: Chinese hamster ovary cell transfected with mouse transferrin receptor 2  121  Figure 4.10: Fe-loaded p97 uptake by Chinese hamster ovary cells or Chinese 55  hamster ovary cells transfected with mouse transferrin receptor 2  124  Figure 4.11: Binding of Fe-loaded p97 to cells with and without the 55  mouse transferrin receptor 2  126  ix  List of Tables Table 2.1: Primary antibodies  37  Table 2.2: Secondary antibodies  38  Table 2.3: Primers  50  x  List of Abbreviations: AP-1  Activator protein 1  ATCC  American Type Culture Collection  ATP  Adenosine triphosphate  BBB  Blood-brain barrier  BSA  Bovine serum albumin  cDNA  Complementary deoxyribonucleotide  CHO  Chinese hamster ovary  C-lobe  Carboxyl-terminal lobe  CNS  Central nervous system  Cp  Ceruloplasmin  CREB  Cyclic adenosine monophosphate response element binding protein  CSF  Cerebral spinal fluid  Dcytb  Duodenal cytochrome b  DIG  Digoxigenin  DMEM  Dulbecco modified eagle's media  DMT-1  Divalent metal transporter 1  DPM  Disintegrations per minute  E7  Embryological day 7  EDC  N-ethyl-N'-(dimethyl-aminopropyl) carbodiimide  EDTA  Ethylenediaminetetraacetic acid  EEA1  Early endosome associated antigen 1  EST  Expressed sequence tag  xi  FACS  Flow cytometry  Fe  2+  Ferrous iron .  Fe  3+  Ferric iron  FITC  Fluorescein isothiocyanate  GPI  Glycosylphosphatidylinositol  HEPES  2-hydroxyethyl-piperazine-2-ethansulfonic acid  HepG2  Human hepatocellular carcinoma derived cell line  HFE  Protein responsible for hereditary hemochromatosis  HH  Hereditary hemochromatosis  HO-1  Heme oxygenase 1  IRE  Iron responsive element  IRP  Iron regulatory protein  K562  Human erythroid leukemia cell line  mRNA  Messenger ribonucleotide  NHS  N-hydroxysuccinimide  N-lobe  Amino terminal lobe  PBS  Phosphate buffered saline  RNA  Ribonucleotide  RT-PCR  Reverse transcription and polymerase chain reaction  RU  Resonance units  SDS-PAGE  Sodium dodecyl sulfate-polyacrylamide gel electrophoresis  SFT  Stimulator of iron transport  SK-MEL-28  Human melanoma derived cell line  xii  TCA  Trichloroacetic acid  Tf  Transferrin  TfRl  Transferrin receptor 1  TfR2  Transferrin receptor 2  TRA  Transferrin receptor transcriptional control element  TRAC  Transferrin receptor transcriptional control element specific complex  TRVb  Chinese hamster ovary cell with defective endogenous transferrin receptor  xiii  Acknowledgements and Dedication Thanks are due to many individuals who have helped me with this work, directly or indirectly, over the years.  To begin, I must acknowledge the guidance and  encouragement of my supervisor, Dr. Wilf Jefferies. Wilf has always had a vision for this research, and an unshakeable faith in both the direction of study and in my ability to complete it. Many people in the lab were also essential to the progression of this thesis, including Dr. Maki Ujiie, who believed in and encouraged me constantly; and Dr. Jacqueline Tiong, who completed her thesis one year ahead of me (inspiring me and proving that it could be done).  Dr. Cheryl Pfeifer, Maya Kotturi, and Dara Dickstein,  helped with extensive proof-reading (and coffee break) duties. I would also like to acknowledge the contribution of all the other lab members (past and present) who have made my time at U B C enjoyable over the years. Thanks to Ian, Forrest, Gregor, Iku, Cyp, Joe, Judie, Daphne, Gaba, Malcolm, Greg, Ana Luiza, Alex, Kyung Bok (K.B.), Kendra, Siri, Jason X , Jason G , Laura, Quin Jin, Mei M e i , Kaan, Matt, Joanne, Mei, Linda, Gene, Ray, Andy, Francesca, and Sarah. I must also thank Dr. Alex Law of Oxford for allowing me to come to his facility to carry out BIAcore studies with the integrin cc (3 . His generous gift of reagents and v  friendly welcome were appreciated.  3  Thank-you also to Dr. Alister Dodds for his  technical assistance with the BIAcore assays in Oxford. Similarly, Dr. Pamela Bjorkman allowed me to work in her laboratory at Caltech for BIAcore studies with transferrin receptor 1 and 2. The BIAcore work could not have been accomplished without the help of Ph.D. candidate Anthony Giannetti in the Bjorkman lab, who worked very hard at trouble-shooting to get the assay working.  xiv  Finally, but most importantly, I must acknowledge my family for their indispensable contribution. M y Mom, who is my greatest cheerleader and unwavering support system. M y Dad, who believes I can accomplish anything. M y sisters, who understand the ups and downs of research and are always there with good advice, or just to listen when I need to vent. The rest of my family (old and new) who have refrained from asking "when will you be done?" Most of all, I need to thank my husband: without Greg, none of this could have been done or would have been worth doing. Greg, this thesis is dedicated to you.  xv  Chapter 1:  I.  Introduction  General introduction Protein-protein interactions are involved in some of the most important biological  processes discovered to date. From the development of the immune system to nerve impulse transmission, the ability of one protein to interact with another in a specific fashion to elicit a particular response is one fundamental way in which biological systems function.  This thesis examines one protein-protein interaction: namely the receptor  binding properties of melanotransferrin, a protein that binds iron and is a biomarker of Alzheimer disease. Melanotransferrin, also known as p97, shares significant sequence homology with the well-characterized transferrin receptor ligand, transferrin (Tf), and is implicated in iron transport across the blood-brain barrier. One of the main goals of p97 research has been to establish a function for the soluble form of the p97 protein. This investigation has begun to identify p97 receptors, including the binding to transferrin receptor 1 (TfRl). After the characterization of this interaction, the search for additional receptors for p97 led to the discovery of a novel transferrin receptor, now called transferrin receptor 2 (TfR2), which is characterized in detail.  II.  Iron absorption in the intestine p97 and the transferrin receptors make up only a few components in a very  complex group of players involved in iron uptake, transport, and storage. Iron is an essential element for life. However, it can also catalyze the formation of free radicals, and for this reason the uptake and dissemination of iron is closely regulated. Iron absorption takes place mainly in the duodenum and jejunum. It is a complex process  1  involving many proteins that play roles in the uptake and transfer of iron, or in the regulation of iron uptake.  A.  Divalent metal transporter 1 In the intestine, the uptake of inorganic iron occurs through the duodenal villus  cells via a protein called divalent metal transporter-1 (DMT-1).  Divalent metal  transporter-1 is a 561 amino acid protein with 12 membrane spanning regions. It is found on many different cell types in the body, with high levels of expression exhibited in the intestine, brain, thymus, proximal kidney and bone marrow (Griffiths et al, 2000; Gruenheid et al, 1995; Rabin et al, 1985). The highest concentration of DMT-1 protein is found in the upper villus region of the duodenum, and upon ingestion of high levels of iron, a redistribution of DMT-1 protein occurs from the apical cell surface to the cytosol (Sharp et al, 2002). Divalent metal transporter-1 's role in the iron uptake pathway is to transport the F e  2+  (ferrous) form of iron across the plasma membrane into the duodenal  cell (see Figure 1.1). Since dietary inorganic iron is largely in the F e  3+  (ferric) form, it  must be reduced by an NADH-ferric reductase before transport by DMT-1 (Conrad et al, 1999). The activity of this reductase has been demonstrated in both a cell culture model of the duodenal mucosa, and in the mucosa itself (Raja et al, 1992; Riedel et al, 1995). One candidate for the NADH-ferric reductase is duodenal cytochrome b (Dcytb). Recently it was shown that Dcytb is localized to the upper villus region of the duodenum, and that it has ferric reductase activity (McKie et al, 2001). Furthermore, both protein levels and mRNA levels of Dcytb are increased by iron deficiency as well as in a mouse model of chronic anemia (McKie et al, 2001). Based on these studies, Dcytb may be the  2  Figure 1.1: Flowchart of iron absorption pathway from the intestine. Ingested inorganic ferric iron is reduced to the ferrous form by a ferric reductase, possibly Dcytb, in the enterocyte, and transported across the cell membrane by D M T - 1 . Iron from heme is absorbed, possibly through a heme receptor, and reduced to the ferrous form by heme oxygenase 1 (HO-1). Ferric iron can also be associated with mucin in the gut, and absorbed by parraferritin which solubilizes the ferric iron, which is then reduced by mono-oxygenase. The ferrous iron is transported out of the cell to the blood stream. It is transported across the membrane by ferroportin, in association with hephaestin that oxidizes the ferrous iron to the ferric form that is bound by transferrin. Ceruloplasmin (Cp) oxidizes ferrous iron in serum for delivery to transferrin (Tf).  3  CD U_  ro  sz CL  I  CD JC  w CD  Pa rate (soli  N  ^ CD  15 LL 3  Q. <D O  fc [3  OJ  Q n < o>  2?  E x:  Z  *"  O  Q "D  o co .i=  0) C  CD CD  O  X —  4  CD  co o o E o tf) sz c/s co  iron reductase that works in concert with DMT-1 to bring dietary non-heme iron into the body. It should be noted that DMT-1 is not specific for F e , but also transports other 2+  metals such as Z n , M n , C o , C d , C u , N i , and P b 2+  2+  2+  2+  2+  2 +  2+  (Sharp et al, 2002).  Divalent metal transporter-1 expression, like the expression of a number of proteins involved in the iron uptake, transport and storage pathways, is regulated through the elegant iron responsive element (IRE)/ iron regulatory protein (IRP) system (Andrews, 1999). In this system, the IRE can be found at either the 3' or the 5' end of the mRNA. In general, IREs located at the 5' end of mRNA allow the IRP to bind in low iron conditions and thus result in decreased translation of the iron binding protein. The iron storage protein ferritin is an example of this regulation (Aziz and Munro, 1987). In this case, low levels of iron cause IRPs to bind to the IREs, thereby blocking translation of the ferritin mRNA and preventing the ferritin protein from being synthesized. When intracellular levels of iron are high, the IRPs change conformation and can no longer bind to the IREs. If this occurs, ferritin synthesis will proceed, resulting in a greater capacity to store pools of iron intracellularly. Conversely, i f IREs are found within the 3' end of mRNAs, IRPs will bind to the nascent m R N A molecule under low cellular iron conditions and stabilize it, thus allowing more protein to be translated.  The IREs within the T f R l and DMT-1 are  located in the 3' untranslated region of the mRNA. When iron concentrations are low and the IRPs are able to bind, the transcript is stabilized and increased amounts of the protein are produced (Mullner and Kuhn, 1988). If iron concentrations are high, the IRP cannot bind to the 5' IRE and the transcript becomes targeted for rapid degradation. Unlike the T f R l , which has five sets of IREs, and needs at least three to be bound by the  5  IRPs for stabilization of the mRNA (Casey et al, 1989), the common isoform of DMT-1 has only one IRE (while the other splice variant of DMT-1 has none) (Gunshin et al, 1997; Wardrop and Richardson, 1999). This single IRE within DMT-1 mRNA may lead to a less sensitive response to iron than that observed for the TfRl" (Sharp et al., 2002).  B.  Heme absorption  ,  •  Dietary iron from heme is absorbed through a different mechanism than inorganic iron. Hemoglobin and myoglobin-derived iron are absorbed more efficiently by the intestine than inorganic iron since the presence of amino acids from proteolytic digestion prevents polymerization and precipitation of heme (Majuri and Grasbeck, 1987; Raffin et al, 1974; Turnbull et al, 1989). There is evidence to support the presence of a heme receptor on the luminal side of villus cells that brings iron into the intestinal cell, largely in the metalloporphyrin form (Grasbeck et al,  1982), but this notion remains  controversial. Once heme is inside the cell, iron is released via heme oxygenase 1 (HO1), which cleaves the heme ring to generate bilirubin, carbon monoxide, and ferrous iron (see Figure 1.1) (Maines, 1997). In mice deficient in HO-1, tissue iron stores are high while serum iron is low, suggesting that the enzyme is involved in the discharge of iron from certain cells (Poss and Tonegawa, 1997a; Poss and Tonegawa, 1997b).  The  mechanism by which the iron liberated from heme is subsequently routed for systemic use is not well understood.  The liberation of iron from spent erythrocytes is of vital  importance, because almost all of the iron required for erythropoiesis is recycled in this manner (Andrews, 2000).  Reticuloendothelial macrophages  ingest senescent  erythrocytes and the iron is either stored within the ferritin of the macrophage, or released  6  to the serum where it can be oxidized by ceruloplasmin and bound by T f (Fleming and Sly, 2001). Recently, a mammalian 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 functionally coupled to this Fe-ATPase (Ferris et al., 1999; Poss and Tonegawa, 1997b). The novel Fe-ATPase is enriched in spleen, and the iron transport it mediates is dependent upon hydrolyzable nucleotide triphosphate, magnesium, time and temperature. Baranano and associates suggest a new model in which Fe-ATPase and HO-1 colocalize to the endoplasmic reticulum where heme is first degraded by HO-1 and the liberated iron is transported by the Fe-ATPase to the luminal side of the endoplasmic reticulum (Baranano et al., 2000). Moreover, they propose that the iron in the endoplasmic reticulum can then bind to T f and can be recycled back to the cell surface for exocytosis, rather than being released into the serum.  C.  Parraferritin The parraferritin complex is made up of (3-integrin, mobilferrin (a homologue of  the molecular chaperone calreticulin) and flavin mono-oxygenase.  Although the  mechanism is not well understood, parraferritin mediates the uptake of ferric iron associated with mucin in the gut lumen (Umbreit et al., 1998). The proposed mechanism is that ferric iron is solubilized by mucin in the intestinal lumen, and then transferred to parraferritin complexes containing (3-integrin and mobilferrin (Conrad et al., 1999). Following internalization of the iron, the complex becomes associated with the monooxygenase, and the ferric iron is reduced (see Figure 1.1).  7  Thus, the parraferritin  complex acts as an NADPH-dependent ferrireductase, and serves to make reduced iron available to the various iron transport proteins.  D.  Ferroportin and hephaestin Once inside the duodenal cell the absorbed iron can be stored in ferritin, or it can  be transported into the blood stream via the iron exporter protein ferroportin, (also known as iron regulated transporter-1) which is located on the basolateral surface of the duodenal enterocytes. Ferroportin also plays an important role in the release of iron from body stores in reticuloendothelial cells (Abboud and Haile, 2000). Furthermore, missense mutations in the ferroportin gene have been implicated in hereditary hemochromatosis (HH) (Montosi et al, 2001).  In this disease, the loss of function mutation causes  problems with iron recycling from the reticuloendothelial macrophages, leading to a decrease in iron available for the hematopoeitic system. This in turn leads to a feedback mechanism that increases iron absorption through the intestine, and thus to iron overload. Ferroportin was initially discovered through positional cloning of a zebrafish mutant with defects in circulating erythroid cells (Donovan et al., 2000).  The  mammalian version is expressed in the placenta, liver, spleen, macrophages and kidneys. Ferroportin has an IRE in the 5' untranslated region, much like ferritin, while its response to iron levels mimics that of the TfRl (McKie et al., 2000). When rodents are fed diets high in iron, ferritin expression increases while ferroportin expression decreases (Abboud and Haile, 2000).  Furthermore, in H H , the duodenal expression of ferroportin is  stimulated despite the increased level of iron in the body, perhaps due to a defect in the  8  iron-sensing mechanisms of the enterocyte (Zoller et al, 2001). The potential regulatory mechanisms of the promoter for this gene have yet to be elucidated. Ferroportin works in concert with a multicopper ferroxidase called hephaestin, which is closely related to ceruloplasmin (Vulpe et al.,-1999).  Hephaestin seems to  oxidize iron so that it can be transported by ferroportin into the bloodstream and ultimately be delivered throughout the body via Tf (see Figure 1.1). How these proteins interact to achieve this phenomenon has yet to be determined.  E.  Regulation of iron uptake Intestinal iron uptake is regulated in at least three distinct ways. The first  regulatory mechanism is the mucosal block, and refers to the fact that once a large oral dose of iron is received, enterocytes do not absorb iron again for 48 hours. While this has been shown experimentally, there is some dispute whether this would explain the ineffectiveness of daily oral iron supplementation (Benito et al., 1998). The molecular mechanism of this mucosal block is still under investigation. The second mechanism for iron uptake regulation is the stores regulator which seems to function at the level of the duodenal crypt cell and involves cell programming in response to existing body iron stores (Finch, 1994). A n absorptive capacity for the villus cell is thus established, and in this manner the amount of iron routed to the ferritin stores within the villus cells versus the amount transported across into the blood is regulated. If the body iron stores are high, more iron will be stored in the ferritin of the villus cells, and be expelled from the body as the villus cells are shed, reducing the overall amount of iron in the body. The current understanding is that iron levels are sensed via the iron  9  saturation level of the circulating soluble TfRl (Andrews, 2000). This in turn dictates the expression levels of DMT-1 and ferroportin on the villus cell. A recently discovered protein called hepcidin is also believed to be involved in communicating body iron storage levels to the intestine. Hepcidin was identified as a circulating antimicrobial peptide, and is produced in hepatocytes (Krause et al, 2000) (Park et al., 2001). In mice with a targeted disruption of the gene encoding upstream stimulatory factor 2 (USF2), an additional unintended knockout of the hepcidin gene (downstream of USF2) created a situation in which no hepcidin was expressed, and iron overload was observed (Nicolas et al, 2001). Deficiencies in four different proteins have now been shown to lead to the hereditary hemochromatosis-like iron overload phenotype in mice.  These include deficiencies in H F E (Feder et al., 1996), {32microglobulin  (Rothenberg and Voland, 1996), TfR2 (Camaschella et al, 2000) and hepcidin (Nicolas et al, 2001). This fact led to the hypothesis that hepcidin expression may be modulated by  iron levels in hepatocytes,  then function by interacting with the H F E ,  (32microglobulin, and T f R l system to regulate iron uptake by the crypt cells of the duodenum (Nicolas et al., 2001). This theory has yet to be tested experimentally, although it is now confirmed that an overexpression of hepcidin leads to a severe iron deficiency condition in transgenic mice (Nicolas et al., 2002). The final known regulation mechanism is the erythropoietic regulator (Finch, 1994). Using this mechanism, the body can adjust the level of intestinal iron absorption to the current demands of erythropoiesis, a major iron sink. It is speculated that an as yet unknown soluble signal from the hematopoeitic bone marrow to the duodenum must exist (Finch, 1994) that signals the need for increased dietary iron uptake. A great deal of  10  further work is necessary to fully understand how iron uptake is regulated in the dietary uptake system.  III.  A.  Serum iron transport proteins  Transferrin The major iron transport protein in the body is transferrin (Tf). Transferrin has  two lobes that possess about 40% sequence homology between them, connected by a short bridging peptide.  The crystal structure of rabbit T f produced by Bailey and  associates shows that there are 13 disulphide bridges within the protein, and that six of those are in homologous positions in each lobe (Bailey et al., 1988). The T f gene is located at 3q21-25, which is in close proximity to the TfRl gene (Rabin et al., 1985). Transferrin is mainly synthesized in the liver, and then is secreted into the plasma where it can transport iron to all areas of the body. Transferrin expression is controlled by both positive and negative regulatory elements found 5' of the transcription start site, although no IRE exists in the mRNA of this molecule. Transferrin binds one ferric ion in each lobe. Anderson et. al. have proposed a model where the open lobe structure of the protein changes configuration by closing tightly when the iron and its synergistic anion bind in the binding site (called the 'Venus fly-trap' mechanism) (Anderson et al, 1990). Since iron binds to T f with very high affinity, in order for T f to effectively supply iron to the cells, T f must also possess an efficient release mechanism. This mechanism appears to be quite complex. While past studies on iron release from Tf have been conducted on either single lobe or monoferric Tf, it has become clear that the physiological mechanism can only be understood by  11  taking into account the active role that the TfRl plays in iron release (Bali and Aisen, 1992). As will be discussed later, in the acidic environment of the endosome, iron is released from Tf, while the T f remains tightly bound to the T f R l . In general, iron release is thought to occur most readily from the amino-terminal-lobe (N-lobe) of the protein than the carboxyl-terminal-lobe (C-lobe), and has been proposed to involve a dilysine trigger mechanism (Dewan et al, 1993). The dilysine motif is, however, missing from lactoferrin (a Tf-like protein found in human milk and other epithelial secretions (Lonnerdal et al, 1987)) and the C-lobe of Tf. In addition, mutation of the iron ligand histidine at position 249 to glutamine in the N-lobe of human T f removes the dilysine "trigger" but does not significantly affect iron release (MacGillivray et al, 2000). Comparison of the two different conformations of the recombinant N-lobe of the human Tf crystal structure revealed that the protonation of the carbonate anion and the resulting partial removal of the anion from arginine 124 may be the initial step in pH-induced release of iron from T f (MacGillivray et al, 1998). Our understanding of the release of iron from T f will not be complete until iron release from the Tf/TfRl complex is directly examined.  B.  Ceruloplasmin Ceruloplasmin (Cp) is a copper binding protein secreted by hepatocytes into the  plasma, but interestingly, it plays no known role in copper transport (Gitlin, 1998). However, Cp does seem to be a vital component in the iron transport system, since it functions as a serum ferroxidase required to oxidate iron before the iron can be incorporated into transferrin (Osaki and Johnson, 1969; Roeser et al, 1970). It has also  12  been observed that under iron deficient conditions, the transcription and translation of Cp significantly increases (Mukhopadhyay et ai, 2000).  The role of this multicopper  oxidase has been more clearly defined in yeast, where a species orthologue multicopper oxidase protein (fet3P) is required for high affinity iron uptake (de Silva et al., 1997; Stearman et al., 1996). In a study on iron uptake using a human erythroleukemia (K562) cell line grown in iron deficient conditions, Cp was able to induce a 2- to 3-fold increase in non-Tf bound iron uptake, through a transcription-dependent manner (Attieh' et al., 1999).  In the presence of excess trivalent cations this effect could be completely  inhibited, suggesting the involvement of a trivalent cation specific mechanism (Attieh et al., 1999). Ceruloplasmin has three pairs of consensus hypoxia response elements in the 5' untranslated region of the gene (Mukhopadhyay et al., 2000). Hypoxic inducible factor 1 is believed to bind hypoxia response elements and increase transcription of the Cp, which is consistent with the fact that an iron deficiency leads to an increase in the level of Cp in the plasma Ceruloplasmin has been demonstrated to exist in both a soluble form and a glycosylphospatidylinositol (GPI)-linked cell surface form. Recently, the GPI-linked form was shown to be a result of alternative splicing (Patel et al., 2000), and to be expressed in glial cells associated with brain microvasculature. This may indicate a specialized role for Cp in iron metabolism and homeostasis within the central nervous system (Klomp and Gitlin, 1996; Sheth and Brittenham, 2000).  In the brain, all of the  Cp appears to be in the GPI-linked form, whereas the soluble form has been shown to be unable to cross the blood-brain barrier (BBB).  13  C.  Melanotransferrin (p97) Melanotransferrin, referred to as p97 in this thesis, is a 97 kDa protein which  shares 39% sequence identity with human transferrin (Rose et al, 1986). It was first identified as a cell surface marker for human skin cancer (Brown et al, 1981; Woodbury et al, 1981; Woodbury et al, 1980). However, the protein was subsequently found to be expressed at various levels in other tissues such as the liver, intestine, umbilical cord, placenta, sweat gland and more recently on human brain endothelium (Barresi and Tuccari, 1994; Rothenberger et al, 1996; Sciot et al, 1989). Unlike the other members of the transferrin family, p97 is present as two distinct forms. It can be found as a membrane protein attached to the cell surface via a GPI-anchor and as a soluble form in the serum and cerebrospinal fluid (Alemany et al, 1993; Brown et al, 1981; Food et al, 1994; Kennard et al, 1996). Like other members of the transferrin family, p97 is a bilobed metal binding protein. Sequence alignment with other members of the family (such as Tf, lactoferrin, and ovotransferrin) show that the amino acids involved in coordinating the iron atom of the N-lobe are precisely conserved (Rose et al, 1986). However, there is an aspartic acid at position 395 in the sequence of the C-lobe of T f that is a serine in the p97 sequence, which could affect iron binding to the C-lobe of the latter. Baker and associates have shown that p97 has only one functional iron binding site, but recent data using various techniques suggests that under certain conditions, both lobes are able to bind iron (Baker etal, 1992; Tiong, 2001; Tiong and Jefferies, 2002). Metal binding studies on a Chinese hamster ovary cell line transfected with human p97 show the iron uptake process by surface GPI-linked p97 is both temperature  14  dependent and saturable (Kennard et al, 1995). The GPI-linked cell surface form of p97 has been shown to deliver the internalized iron to ferritin (Tiong, 2001).  This  internalization, shown through immunofluorescence co-localization, is mediated through caveolae rather than clathrin coated pits (Tiong, 2001). The iron delivery to ferritin requires the presence of Rab 5 in the cells, which indicates that at some point the caveolae vesicle merges with the endosomal pathway (Tiong et al, 2002): • Similar to Tf, p97 is able to bind to other metals in addition to iron. Using a competitive metal binding assay, it has been demonstrated that metals such as A l , C u 3 +  and Z n  3 +  3 +  are able to compete for the iron binding site in p97 (Tiong, 2001). Also as with  Tf, there is no identifiable iron responsive element in p97 (Richardson, 2000). However, a regulatory element is located 2 kilobases (Kb) upstream from the promoter region of the p97 gene, and deletion of this element 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 the AP-1 transcription factor. AP-1 is formed by the dimerization of the proteins Jun and Fos through their leucine zipper motifs, and is upregulated after ultraviolet irradiation (a suspected risk factor in melanoma) (Devary et al., 1991). The transcription factor recognizes the phorbol 12-myristate 13-acetate responsive element (Angel and Karin, 1991), (Sassone-Corsi, 1994). By gel retardation assays it has been shown that the expression of p97 correlates with increased AP-1 binding activity (RozeHeusse et al, 1996). A secondary unidentified nuclear factor is involved with AP-1 to form a ternary complex at the two AP-1 sites in the p97 enhancer region (Roze-Heusse et al, 1996). This may explain how p97 is upregulated in melanoma cells.  15  It has long been believed that the Tf/TfRl system is exclusively responsible for iron delivery to the brain, but recent studies suggest this may not be the case. For example, hypotransferrinemic mice, deficient in Tf, have normal brain iron levels. Many pieces of evidence point to a role for p97 in iron transport to the brain. Expression of p97 has been examined in brains from Alzheimer disease patients as well as in brains from patients suffering from other neurological diseases. While immunohistochemical staining of p97 in normal brains shows distribution limited to the B B B , p97 in Alzheimer disease is also found to be highly expressed in a subset of reactive microglia cells associated with fj-amyloid plaques (Jefferies et al, 1996).  This was further confirmed by in situ  hybridization of p97 mRNA, which shows p97 is expressed in the reactive microglia cells in the Alzheimer disease brain but not in normal brain or non-reactive microglia in Alzheimer disease brains (Yamada et al, 1999). Soluble p97 was also found to be elevated 2 to 4 fold in the cerebral spinal fluid and serum of patients with Alzheimer disease compared with healthy age matched controls and patients with other forms of dementia (Feldman et al, 2001; Kennard et al, 1996; Kim, 2001; Moroo et al, 1999). Metals such as iron and zinc that bind to p97 have been shown to be able to nucleate the formation of insoluble |3-amyloid from soluble amyloid (Bush et al, 1994). The soluble form of p97 can cross the B B B and deliver iron across the B B B more efficiently than Tf, i f both are injected into the tail vein of the mouse (Moroo et al, 2002). These results may be an indication that p97 functions in shuttling iron across the B B B from the blood to the neuropil, transporting the iron needed for normal brain function.  How p97 is regulated in reactive microglia and how its function in metal  16  transcytosis interplays with its elevation in bodily fluids of Alzheimer disease patients is currently being investigated.  IV.  Proteins involved in iron uptake  A.  Transferrin receptor 1 The transferrin receptor 1 (TfRl) is a protein that functions to uptake iron into  cells through a tightly regulated process. The TfRl gene is found on chromosome 3q26.2 and consists of 19 exons (Evans and Kemp, 1997). It is highly expressed on rapidly dividing cells including immature erythroid cells and placental tissue.  Many  nonproliferating cell types also express T f R l , including hepatocytes, endothelial cells of the B B B , reticulocytes and Sertoli cells (Jefferies et al, 1984; Kuhn et al., 1990). Transferrin receptor 1 is a Type II membrane protein with a 61 residue N-terminal cytoplasmic domain, a 28 residue domain that crosses the membrane once, and a 671 amino acid extracellular C-terminal domain. The TfRl possesses three domains called the apical domain, protease-like domain, and helical domain (Lawrence et al., 1999) (See Figure 1.2). Electron cryomicroscopy studies have shown that the binding domain is separated from the membrane by a stalk of about 2.9 nm while the dimensions of the globular binding domains are approximately 6.4 nm high, 10.5 mn across the homodimer, and 7.5 nm deep (see Figure 1.2) (Fuchs et al., 1998). The receptor undergoes post-  17  Figure 1.2: Model of the transferrin receptor 1 complex. The T f R l complex is shown with transferrin (Tf) bound to one lobe of the T f R l homodimer. H F E and [32 microglobulin ((3) interact with one half of the TfRl homodimer. Each half of the TfRl has three domains, labeled H , A , and P for helical, apical, and protease-like, respectively. HFE has three domains termed a l , a2, and cc3. The TfRl has three sites of N-linked glycosylation (n) at N251, N317, and N727, and one O-linked site (o) at T104 (Lawrence etal, 1999; Lebronefa/., 1998).  18  19  translational modifications including N and O-linked glycosylation as well as phosphorylation. Recombinant receptors lacking all three asparagine-linked (N-linked) glycosylation sites show decreased transferrin binding, while mutations that eliminate phosphorylation of the cytoplasmic serine residue do not seem to have any effect on receptor internalization efficiency (Rothenberger et al, 1987b; Williams and Enns, 1991).  A tyrosine based internalization motif (tyrosine, threonine, arginine, and  phenylalanine at positions 20 - 23) is present in the cytoplasmic portion of the protein. Through mutational studies, this motif has been demonstrated necessary for endocytosis of the receptor (Collawn et al., 1993). The TfRl functions as a homodimer with two disulfide bonds formed between the cysteines at positions 89 and 98 (Jing and Trowbridge, 1987). Each of the two Cterminal lobes of one T f R l homodimer can bind one T f protein in a pH-dependent manner. The binding affinity of the TfRl for T f is dependent on the iron binding state of the Tf: the highest affinity at physiological pH is seen between diferric T f and T f R l , and the lowest for apo-Tf (Young et al., 1984). The residues involved in T f binding to the TfRl have not been fully characterized, although Lawrence and associates have produced a model based on crystal structure analysis that proposes that T f interacts with all three domains of the TfRl (apical, protease-like and helical) (see Figure 1.2) (Lawrence et al, 1999). The interaction of the T f R l with H F E also seems to have an impact on T f binding. Studies are underway in the Bjorkman laboratory to try to identify residues that are important for either binding to HFE or binding to Tf, but not both, in order to better understand the binding interactions (personal correspondence).  20  Mutational studies have shown that pH-dependent binding is a key component in the T f /TfRl recycling pathway, as the TfRl strongly binds T f saturated with iron at the pH of plasma (Kd ~ 10" to 10" mol/litre). The complex is then internalized via receptor 7  9  mediated endocytosis in a clathrin coated pit and routed to the endosome, where the lower p H allows T f to release iron. In fact, iron is released from T f at a low pH faster and more efficiently when the T f is complexed with the T f R l than i f T f is unbound, demonstrating that the interaction with the TfRl enhances iron release (Bali et al, 1991; Sipe and Murphy, 1991). Following iron release, the complex is recycled back to the cell surface, where the apo-Tf is released and replaced by holo-Tf. In order to study the physiological role of the T f R l , Levy and associates have created a knockout mouse lacking the TfRl (Levy et al, 1999). Transferrin receptor 1 -/mice die by embryonic day 12.5. Moreover, embryos from embryonic day 8.5 to 12.5 demonstrate growth retardation, pericardial effusions, and severe pallor, while histological analysis shows edema and diffuse necrosis throughout the tissues (Levy et al, 1999).  As late as embryonic day 10.5, however, some T f R l  -/- mice are  indistinguishable from wild-type mice, and do not show visible signs of anemia although they do not survive past day embryonic day 12.5. Both the anemic and non-anemic TfRl -/- mice have defective erythropoiesis and abnormal development of the nervous system, exhibiting kinking of neural tubes. Levy and associates speculate that the reason the observed phenotype for the T f R l knock-out mice is more severe than that of the hypotransferrinemic mice (mice with little or no transferrin) is that i f T f is missing from serum, unchelated iron is present and available for use in cells via an alternate iron uptake system. In the TfRl knockout mice, in contrast, the vast majority of iron is bound to  21  serum Tf, making it unavailable for use because the receptor is not present to internalize the Tf. Regulation of T f R l  expression is complicated and occurs at both the  transcriptional and post-transcriptional levels.  A region of 100 base pairs upstream of  the transcriptional start site is involved in driving basal and serum/mitogenic stimulation of the TfRl promoter activity via an "AP-1 like" site (Casey et al., 1988; Miskimins et al, 1986; Owen and Kuhn, 1987). "AP-1/CREB like" (cyclic adenosine monophosphate response element binding protein) factors have been shown to bind to the "AP-1 like" site of TfRl (Beard et al., 1991; Lok et al., 1996). The TfR transcriptional control element specific complex (TRAC), which is a nuclear protein that co-purifies with the K u autoantigen, binds specifically to the transcriptional control element T R A (TfR transcriptional control element) of the TfRl gene (located at nucleotides -77 to -70 and necessary for increased expression in proliferation) (Roberts et al., 1994). Differentiating erythroid cells exhibit specific up regulation of TfRl transcription linked to both the Etsbinding site (transforming-specific protein produced by the ets gene, first discovered in the E26 avian erythrobastosis virus) and the AP-1 (activator protein-1) binding site in the 5' flanking region of the transcription start site (Sieweke et al., 1996). Lok and Ponka have shown that a region (-118 to +14) of the TfRl promoter is necessary for erythroid differentiation induced promoter activity, and that mutation of EBS or the "AP-1/CREBlike" motif inhibit this inducible promoter activity (Lok and Ponka, 2000). Furthermore, they found that "CREB/ATF-like" factors and "Ets-like" factors bind to the identified element. Lok and Ponka also describe an element in the promoter region of the T f R l that is responsive to hypoxia. This region contains a binding site for hypoxia inducible factor  22  1 and the motif organization is similar to that of many hypoxia inducible genes such as erythropoietin and ceruloplasmin (Lok and Ponka, 1999). Post-transcriptional regulation of the T f R l is mediated by the binding of iron regulatory proteins to the five IREs in the 3' untranslated region of the T f R l mRNA. When cellular iron levels are low, the IRPs can bind to the IREs, thus stabilizing the transcript and allowing more TfRl to be translated (Mullner and Kuhn, 1988). This elegant system regulates several of the proteins involved in iron metabolism, allowing coordinated expression of the proteins involved in the pathway.  Another post-  transcriptional control of T f R l expression has recently been identified. Koeffler and associates have carried out experiments on cells with TfRl without the 3' untranslated region (Tong et ai, 2002). The level TfRl in these cells is increased when the cells are treated with desferroxamine (an iron chelator), even though the iron regulatory proteins can not bind to the transcript to stabilize it and increase protein expression level. The amount of transcript was not elevated, and the stability of the protein was unchanged. Transferrin receptor 2, which does not show iron-sensitive expression, was engineered with its 3' untranslated region replaced with that of T f R l . The addition of the transcript with the iron responsive elements made the expression increase about two-fold with the addition of desferroxamine. With T f R l , the observed increase is five-fold, indicating that the difference in responses is likely due to some RNA-independent method of regulation of TfRl expression. This new method of iron dependent regulation of T f R l expression needs to be examined further. Transferrin receptor 1 is also detectable as a soluble protein in serum, and is measured clinically to help differentiate between iron deficiency anemia and chronic  23  disease anemia. The amount of soluble TfRl in the blood correlates with the amount of erythropoiesis, thus indicating that the cleavage of TfRl may be the method for reducing the amount of T f R l when the red blood cell precursors differentiate into mature cells (Cook et al, 1993; Nair et al, 1990). The soluble form of the T f R l is the result of a proteolytic cleavage at the arginine at amino acid position 100 near the transmembrane domain in the extracellular portion of the protein, which occurs during transit through the endosomal pathway, after presentation at the cell surface (Rutledge et al., 1994; Shih et al, 1990). The receptor is found in the serum as a homodimer that is lacking the disulfide bonds of the cell surface form.  B.  HFE HFE is a protein that seems to have a substantial impact on the regulation of iron  absorption and the uptake of iron into cells. The HFE gene was identified by Feder and associates in 1996 and maps approximately 4 megabases telomeric to the HLA-A gene on chromosome 6 (Feder et al, 1996). HFE, a major histocompatibility complex class I-like protein, was originally found expressed in the gastrointestinal tract and placenta (Parkkila et al, 1997a; Parkkila et al, 1997b).  More recently, it has been found in tissue  macrophages as well as circulating monocytes and granulocytes (Parkkila et al, 2000). In the duodenum, H F E is expressed primarily in crypt cells, but there is very little H F E expressed in villus enterocytes (Parkkila et al, 1997b). The H F E protein forms a heterodimer with beta 2 microglobulin (P2m) and the two proteins together form a complex with the TfRl at the cell surface (See Figure 1.2) (Feder et al, 1997; Lebron et al, 1998; Parkkila et al, 1997a; Waheed et al, 1997). The  24  wild type HFE protein seems to have a significant effect on the uptake of T f and iron by the TfRl endocytosis pathway. Over-expression of HFE in transfected human hepatoma cells results in an approximately 50% decrease in receptor mediated Tf-iron uptake (Ikuta et al, 2000; Roy et al, 1999). A n increase in the dissociation constant from 1.9 to 4.3 n M is also observed, indicating a decrease in the affinity of TfRl for T f in the presence of H F E (Ikuta et al, 2000). It is not clear what the physiological significance of this decrease in affinity would be as the concentration of diferric T f in blood is high enough to saturate all T f R l s , despite the presence of H F E (Ikuta et al., 2000).  The  overexpression of H F E also seems to slow the rate at which the TfRl complex recycles to the cell surface (in hepatoma cells), and this may lead to a decrease in iron uptake by the cell (Ikuta et al, 2000). Expression of HFE in HeLa cells has been associated with a decrease in the intracellular iron pool, a decrease in the amount of ferritin translated due to a 5 fold increase in IRP binding, and a shift in the ratio of IRP1 to IRP2 in the cells (Roy et al, 2002). HFE, through the analysis of the knockout mouse phenotype, has been shown to be responsible for the most common type of hereditary hemochromatosis (HH) (Zhou et al, 1998). Hereditary hemochromatosis is a disease of iron overload, and is the most common autosomal recessive disorder in persons of European descent. The prevalence of H H has been estimated at 1 in 200 to 1 in 400, with a carrier rate of between 1 in 7 to 1 in 10 in Caucasian populations (Bulaj et al, 1996; Burke et al, 1998; Crawford et al, 1998; George et al, 1998; Goldwurm and Powell, 1997; Jazwinska and Powell, 1997; Ramm et al, 1997).  The disease is characterized by a 2- to 3-fold increase in dietary iron  absorption that leads to iron deposition in parenchymal cells of the liver, joints, pancreas,  25  heart, skin and pituitary gland. Prolonged iron deposition may lead to fibrosis and organ failure, frequently including hepatic cirrhosis, diabetes mellitus, cardiac dysfunction, arthritis and hypogonadism (Adams and Chakrabarti, 1998; Bacon et al, 1999; Burke et al, 1998; Cullen etal, 1997; Niederau et al, 1996; Niederau etal, 1985). The majority of patients with H H are homozygous for a cysteine to tyrosine mutation at amino acid residue 282 in the H F E protein (Feder et al, 1996). This mutation has been shown to prevent p^m from associating with the H F E protein, thus HFE is not expressed at the cell surface or associated with the T f R l . The mutant form of the protein is localized intracellularly, and colocalizes with calhexin, indicating that the protein may be retained in the endoplasmic reticulum and golgi due to improper folding (Ramalingam et al, 2000).  The mechanism .by which this HFE mutation leads to  deregulation of controls on iron absorption and thus a large increase in dietary iron uptake is under debate. Presumably, H F E must play a significant role in the iron absorption pathway, possibly in the "programming" of crypt cells. HFE -/- mice show increased expression of duodenal DMT-1, which supports the idea that HFE mutations in H H may lead to a decreased level of crypt cell iron. The crypt cell senses a decreased level of body iron and this prompts the cell, once it matures into a villus cell, to express increased levels of DMT-1 leading to increased dietary iron absorption (Fleming et al, 1999). If the crypt cell originally senses abundant iron, then the level of DMT-1 expressed by the mature villus cell will be low. In a recent publication, Townsend and Drakesmith hypothesized that the main role of H F E may be in programming of crypt cells through inhibiting the export of iron, rather than slowing the uptake of iron (Townsend and Drakesmith, 2002). In this theory,  26  they suggest that the binding of HFE to TfRl is competed for by holo-Tf, so that when there is an abundance of iron loaded serum Tf, the H F E is displaced from the T f R l complex and is available to interact with the iron export protein ferroportin. This interaction then inhibits the release of iron from the crypt cells (thus increasing overall iron stores within the cell). In effect this programs these cells to absorb less iron from the diet. This theory has yet to be proven experimentally. Levy and associates have created HFE knockout mice and bred them with other strains of mice that have mutations in their iron uptake or metabolism (Levy et al, 2000). One interesting mouse cross was between the HFE -/- and fcm -/- mice. The absence of p^m and HFE lead to a more severe iron overload phenotype than in the HFE knockout mice alone. A possible explanation is that the compromising of the immune system that results from the absence of fcm -/- itself leads to a more severe iron overload phenotype. Another set of crosses demonstrated that the iron overload observed in H H due to an HFE mutation is mediated through the D M T - 1 / hephaestin pathway. When HFE knockout mice were crossed with mk (microcytic anemia) mice, which are characterized by a loss of function in DMT-1, the mice resembled the mk phenotype rather than the iron overload phenotype characteristic of the HFE knockout, resulting in extremely low iron stores (Levy et al, 2000). This indicates that iron uptake in H H is probably mediated through a pathway involving DMT-1. Mice bred to possess a loss of function mutation in both HFE and hephaestin, a protein necessary for basolateral iron transport out of intestinal microvilli cells, also exhibit a decrease in iron loading in the liver. However, this did not lead to a complete block in iron transfer from enterocytes.  27  When HFE knockout mice are bred with TfRl heterozygotic mice, the animals have a more severe iron loading phenotype in hepatocytes than the HFE mutants alone. Levy and associates have hypothesized that this apparent contradiction occurs because the TfRl +/- mice have small erythrocytes with less hemoglobin than normal due to a decreased level of T f R l (Levy et al, 2000). This leads to an even greater increase in dietary iron absorption, possibly via an "erythroid signal", which leads to an increase in dietary iron absorption and compensates for the erythroid iron deficiency. Much work has yet to be done in order to discover what role HFE plays in iron absorption regulation, and to further illuminate the effect HFE has on the T f /TfRl pathway.  C.  Transferrin receptor 2 Transferrin receptor 2 (TfR2) is a recently identified member of the transferrin  receptor family (published first by Kawabata et. al., but simultaneously identified by others including this investigator.) The extracellular portion of this type II membrane protein shares 45% identity and 66% similarity with the TfRl (Kawabata et al, 1999). Two splice forms have been discovered: the a form which has a transcript size of approximately 2.9 kb, and a shorter (2.5 kb) P form which lacks the N-terminal portion of the protein. The shorter transcript probably encodes an intracellular protein as both the transmembrane and signal sequences are missing. Transferrin receptor 2 has a putative tyrosine based internalization motif (tyrosine, glutamine, arginine, and valine) similar, although not identical to that of the classical TfRl (tyrosine, threonine, arginine, and phenylalanine) (Kawabata et al, 1999). As well, two cysteine residues are present (109 and 112) in approximately the same location as those of the TfRl (89 and 98) that may  28  allow the formation of disulfide bridges, however, this remains to be demonstrated experimentally. West and associates have shown that a recombinant TfR2 binds iron loaded T f with a 25-fold lower affinity than T f R l , and that H F E can not be coimmunoprecipitated with TfR2 as it can with the TfRl (West et al, 2000). As with HFE, a form of H H has been linked to the TfR2 gene. Camaschella and associates found a cytosine to guanosine transversion in exon 6 of TfR2 (position 750 of the cDNA) causes a truncation of the extracellular portion of the protein (a stop codon replaces a tyrosine at residue 250) (Camaschella et al, 2000). It is unclear why this particular truncation, in which the T f binding portion of the TfR2 protein is removed, should lead to an iron overload phenotype. Further characterization of the role of TfR2 in iron metabolism is needed. Fleming and associates have demonstrated that the 3' untranslated region of the murine TfR2 is significantly shorter than that of T f R l , and does not contain any regions similar to the consensus sequences of iron responsive elements (Fleming et al., 2000). A n examination of the hepatic expression of mouse T f R l and TfR2 in iron overload supports this lack of iron regulation. A mouse model of H H shows down-regulation of mouse TfRl m R N A in the H H model mice, while no such down regulation occurs with mouse TfR2 (Fleming et al, 2000). It was recently demonstrated that Chinese Hamster Ovary (CHO) cells transfected with TfR2 are resistant to the iron deprivation effects of desferoxamine (an iron chelator), which suggests that TfR2 has the ability to increase the cellular iron pool and support cell growth, in the same manner as TfRl does (Kawabata et al, 2000). In addition, it was speculated that perhaps TfR2 is a lower affinity form of the  29  T f R l , and could in fact be a more primitive form, as TfR2 seems to be more closely related to the prostate specific membrane antigen than T f R l .  D.  Stimulator of iron transport The stimulator of iron transport (SFT) protein is a Tf-independent iron transporter  that was originally identified using Xenopus laevis oocytes (Gutierrez et al, 1997). The SFT protein is a 338 amino acid integral membrane protein with at least six membrane spanning regions ( Y u and Wessling-Resnick, 1998).  It is able to form a dimer,  suggesting similarity to the 12 membrane spanning domains of D M T - 1 , and is localized to the plasma membrane and endosomal vesicles. The transcript for SFT is expressed in a number of tissues including spleen, duodenum, colon, liver, kidney, heart, brain, and peripheral blood leukocytes (Gutierrez et al, 1997; Knutson et al, 2001). The SFT contains an R E X X E motif that is similar to the iron binding motif of ferritin, and it has been shown in HeLa cells to stimulate both Tf-dependent and independent iron uptake (Gutierrez etal, 1997). The mechanism of iron transport utilized by SFT remains unknown, however the expression of the protein does seem to be dependent upon intracellular iron levels. For instance, treatment of HeLa cells with desferroxamine, which depletes iron from the cell, leads to an increase in the SFT transcript (Yu et al, 1998). The mechanism for this iron dependent-response is unknown, since no IRE has been identified. Furthermore, liver samples from patients with H H have a 5-fold higher level of the SFT transcript, compared to liver samples from control subjects (Yu et al, 1998). In contrast, HFE knock-out mice do not show increased levels of the SFT mRNA (Knutson et al, 2001).  30  V.  Hypothesis and general approach This research aims at identifying and characterizing a receptor for the soluble  form of the p97 iron transport protein. The guiding hypothesis is that the soluble p97 binds specifically to a receptor and this receptor-ligand interaction leads to iron delivery to cells. A function has not yet been established for p97, although it is known that the GPI-linked, cell surface expressed form binds and internalizes iron, and can deliver that iron to the iron storage protein ferritin (Kennard et al., 1995; Tiong, 2001). It should be noted that the soluble form of p97 is not a cleavage product of the GPI-form, but likely a splice variant (Food et al, 1994). Since p97 resembles many of the other members of the Tf family, and as many of these members are in fact ligands for receptors, it seems reasonable that the soluble version of p97 also has a receptor and functions as a ligand. To determine the identity of this receptor, the most reasonable place to begin was with an examination of the receptor for Tf, the closest homologue of p97. Transferrin receptor 1 is located in many tissues throughout the body and is able to bind and internalize T f with high affinity, as well as enable the iron carried by the T f to remain in the cell while T f itself is recycled back to the cell surface and released. The first results chapter examines the hypothesis that soluble human p97 binds to the human T f R l , and that iron bound to the p97 can be delivered to the cell via this receptor-mediated mechanism. Initially, the general approach was to compare binding of the human p97 protein in two cell lines: one without T f R l , and one expressing human TfRl on the cell surface. The difference in binding and iron uptake between the two cell lines is thus assumed to be due to p97 binding to the receptor. Various assays were performed to examine cell surface binding, including iodination of the p97, and  31  immunofluorescent labeling and Pandex assay. In addition, binding between soluble human T f R l and p97 was measured in a surface plasmon resonance study (or BIAcore). For all of these assays, human T f binding to TfRl was measured as a positive control for the assays. To examine uptake of iron via p97, the total amount of radioactive iron taken up by the two cell lines was compared, as well as the amount of F e associated with 55  immunoprecipitated ferritin in the two cell lines and measured using a scintillation counter. A second hypothesis in the chapter 3 is that a different type of receptor exists for p97, namely integrin aV(33. This interaction was examined because p97 has a three amino acid motif, consisting of an arginine, glycine, and aspartate (referred to as an R G D motif). The R G D motif has been identified as an essential part of the recognition motif for a family of integrins. By comparing the p97 sequence to the available crystal structure of Tf, the R G D motif is likely exposed on the surface of p97. One prominent member of this heterodimeric family is integrin ccVP3, which also binds the widest variety of ligands (Xiong et al., 2001). For this reason, the interaction between the soluble p97 and a soluble engineered version of the heterodimeric integrin was measured by BIAcore. To identify other receptors that may bind p97, homologues of the T f R l were pursued. Database searches for proteins similar to the TfRl led to the discovery of TfR2. This novel receptor, which was simultaneously discovered by Kawabata and associates, is extremely similar to the classical T f R l , with the mouse version sharing 40% identity with the mouse T f R l .  The second results chapter begins with experiments aimed at  characterizing a number of features of the TfR2, including the sequence, the expression  32  during development, as well as subcellular and tissue localization. Some features of the two receptors are similar, though TfR2 is in many ways unique from T f R l . Specifically, the endocytosis and subcellular localization of the transferrin receptors are similar, but the tissue expression patterns are distinct. To examine some characteristics of the TfR2 that complement the published data, many approaches were utilized. First the entire mouse sequence of TfR2 was obtained through the technique of "primer walking". Northern blot analysis was employed to examine both embryological expression in the developing mouse, and adult tissue expression in the human.  To examine brain  microvascular expression, the reverse-transcriptase polymerase chain reaction was carried out on R N A isolated from primary brain micro vasculature. In addition, to study subcellular localization, immunofluorescence and confocal microscopy were used. The two receptors were also examined in the context of a cell line that expresses both TfRl and TfR2, through immunoprecipitation and Western blotting, to see i f they interact to form heterodimers. Finally, the second results chapter also examines the hypothesis that p97 binds to TfR2 and that F e bound to p97 is transported into cells via TfR2. To examine this 55  hypothesis, the approach was similar to that of results chapter 1. Cells transfected with mouse TfR2 were incubated with radioactive-iron loaded p97 and then the amount of iron within the cells compared. The difference in iron uptake between the two cell lines is deemed to be due to TfR2 mediated uptake. The model of p97 binding to the TfR2 and depositing iron within cells via this receptor was developed based on these results.  33  As a whole, the results substantially advance our understanding of the new iron transport receptor TfR2, and support a possible explanation for an additional iron uptake pathway via p97 in certain cells.  34  Chapter 2:  I.  Materials and methods  Cells and antibodies Many different cell lines were used during the experiments contained in this  thesis. The main group of cells is derived from Chinese Hamster Ovary (or CHO) cells. The parental C H O line was obtained from American Tissue Type Culture Collection (ATCC, Manassas, V A ) . TRVb cells are a CHO line that does not internalize T f and thus has no functional transferrin receptors expressed (McGraw et al, 1987). The phenotype arose as a spontaneous mutation and was selected with ricin A conjugated to transferrin. TRVb cells were then transfected with the constructs p C D T R l and pSV3Neo (to confer G418 resistance) to produce TRVb-1 that expresses functional human T f R l . These cells were the kind gift of Dr. F. Maxfield from New York University. For clarity, TRVb cells from this point on will be referred to as transferrin receptor minus, or TfR-, and TRVb-1 cells will be labeled transferrin receptor plus or TfRl+. The T R V b cells were also transfected with a pcDNA3.1 vector containing a myc tag and mouse TfR2 (alpha form) using F u G E N E 6 (Gibco).  This cell line will be referred to as mTfR2+.  Some  experiments also use a parental C H O cell line transfected with the same mTfR2 vector (CHO-mTR2). The media used to grow the mTfR2+ and the CHO-mTfR2 cells also contained 1000 u,g/ml G418 to maintain the expression of the TfR2. The T R V b transfected with human TfR2 (TfR2+), and TRVb-1 transfected with human TfR2 (TfRl,2+ respectively) were a gift from Dr. Caroline Enns, of the University of Oregon. A l l the C H O derived cell lines are maintained in Ham's F12 media (Invitrogen Life Technologies Inc., Burlington, ON) with 10% (v/v) fetal bovine serum (Invitrogen  35  Life Technologies Inc.), 2 m M glutamine (Invitrogen Life Technologies Inc.), 20 m M HEPES (Sigma-Aldrich Canada Ltd), pH 7.4, at 37 ° C in a 5% C0 -humidified 2  incubator. A human melanoma cell line called S K - M E L 28 (from A T C C ) was transfected with the mTfR2 wye-tagged construct pcDNA3.1 in order to examine TfR2 localization in a human cell line. The S K - M E L 28 cells were grown in D M E M medium (Invitrogen Life Technologies Inc.), supplemented with the same conditions as above for Hams F12. The human erythroid leukemia cell line K562 (ATCC), which naturally expresses high levels of human TfR2, was also maintained in this D M E M media. Primary human brain endothelial cells were obtained as a kind gift from Dr. Dorovini-Zis (Vancouver General Hospital). They were cultured in M l 9 9 medium (Invitrogen Life Technologies Inc.) supplemented with 10% heat inactivated horse serum (Sigma Chemical Co., St. Louis, M O ) , 10  Lig/ml  endothelial cell growth supplement  (Collaborative Biomedical Products, Bedford, M A ) , 5 |ig/ml insulin, 5 \ig/m\ T f and 5 ng/ml selenium (ITS Premix, Collaborative Biomedical Products), and 600 USP units/1 of heparin (Sigma). The primary and secondary antibodies used in this thesis are listed in Table 2.1 and 2.2 respectively.  II.  Binding of I-p97 and I-transferrin to transferrin receptors 1 and 2  A.  Iodination of proteins  125  125  The iodination of the proteins was carried out using the Chloramine T (Sigma Aldrich) method.  Equal volumes of p97 (provided by Synapse Technologies Inc., 36  Table 2.1:  Primary antibodies  Antibody  Antigen  Host  Isorvne  Concentration  Source  HybC  p97  mouse  IgGl  1 mg/ml  Dr. Shuen-Kuei Liau, McMaster  L235  p97  mouse  IgGl  15 mg/ml  ATCC  Transferrin-  Human transferrin  sheep  IgG  9.7 mg/ml  ICN Biomedicals  Human transferrin  mouse  IgGl  35 mg/ml  Cedarlane  FITC Transferrin  Laboratories ATCC  Human TfR  mouse  Anti-  Transferrin  rabbit  Transferrin  receptor(for  Trowbridge, Salk  receptor  Western blotting)  Institute  Anti-c-myc  Residues 410-419  clone 9E10  of human c-myc  EEA-1  Amino terminal  mouse  IgGl  12 mg/ml  OKT9  Gift from Dr. I.  Sigma  IgGl  200 ng/ml  Goat  Biotechnology  human EEA-1 Clathrin HC  Carboxy terminal  Santa Cruz  200 u,g/ml  rabbit  Santa Cruz Biotechnology  of human clathrin heavy chain Flag sequence  mouse  IgG  3.5 mg/ml  Sigma  Anti-FLAG  N-terminal Met-  mouse  IgG  3.5 mg/ml  Sigma  M5  F L A G fusion  1.3 mg/ml  DAKO,  Anti-FLAG M2  proteins Ferritin  Human ferritin L -  rabbit  Carpinteria C A .  chain  37  Table 2.2:  Secondary antibodies  Antibody  Concentration  Dilution Used  Source  Goat anti-mouse  1 mg/ml  1:1000  Jackson Immunoresearch Laboratories Inc., West  FITC  Grove, PA. Alexa 488 nm  2 mg/ml  1:1000  Molecular Probes Inc., Eugene, OR  conjugated Goat anti-mouse IgG Alexa 488 nm  2 mg/ml  1:500  Molecular Probes Inc.  2 mg/ml  1:500  Molecular Probes Inc.  2 mg/ml  1:500  Molecular Probes Inc.  0.8 mg/ml  1:10,000  Jackson Laboratories Inc.  0.8 mg/ml  1:10,000  Jackson Laboratories Inc.  conjugated rabbit anti-mouse IgG Alexa 568 nm conjugated rabbit anti-goat IgG Alexa 568 nm conjugated goat anti rabbit IgG Peroxidase conjugated goat anti-mouse IgG Peroxidase conjugated goat anti-rabbit IgG  38  Vancouver, BC) or human T f (Sigma) and 0.5 M phosphate buffer at pH 7.4 (20 p,l) were mixed together followed by 1 mCi N a l  1 2 5  (10u.l). To this mixture, 20 Lil of freshly  prepared Chloramine T (2.25 mg/ml in phosphate buffer) was added and kept at room temperature for one minute.  Next, 20 u.1 of sodium-metabisulfate (7mg/ml sodium  metabisulfate in 0.5M phosphate buffer) was added and again allowed to sit for one minute. Finally, 100 u,l of Nal (15 mM) was added, and the entire mixture was passed with 20 ml PBS + 1% bovine serum albumin (BSA) through a Sephadex G-25 column (Amersham). The eluate was collected in 1 ml fractions. A T C A (trichloroacetic acid) precipitation was carried out to determine the amount of labeling of the proteins as follows: Two microlitres of the labeled solution was added to 200 ul of 1% B S A in PBS. To this, 200 \il of 10% T C A was added and mixed, followed by centrifugation in a microfuge at maximum speed for 10 minutes. The supernatant was then transferred to a new eppendorf tube and both the supernatant and the pellet tubes were counted in the gamma counter.  The counts per minute of the  supernatants represent the non-TCA precipitable counts (or non-protein bound counts), while the counts per minute of the tubes with the pellets represent the T C A precipitable counts (or the counts bound to p97 or Tf).  B.  Binding to transferrin receptors  The binding experiment was carried out in the following manner: TfR positive (TfR+) and TfR negative (TfR-) cells were harvested from tissue culture plates with Versene (5.37 m M disodium E D T A , 1.37 M NaCl, 26.83 m M KC1, 81,01 m M N a H P 0 , 14.7 2  4  m M KH2PO4, 11.1 M glucose). The cells were counted with a hemocytometer and  39  aliquoted into BSA-treated tubes at one million cells per tube. The 5 ml polystyrene tubes were first coated with a PBS solution containing 5% B S A to reduce the amount of protein binding to the tubes, and thus reduce background counts. Following this step the cells were briefly acid washed with a mixture of 2.6 ml 0.1 M glycine buffer (0.1 M glycine, 0.05 N HC1, 10 m M NaCl), 5 ml D M E M serum free tissue culture media, and 182 ui 1 M NaCl. Each tube of cells was kept on ice and washed three times with 1 ml of the acid wash, followed by three washes with PBS. The radioactive protein was then added to the serum free media mixture and incubated with the cells for one hour at 4°C, followed by two washes with ice-cold PBS+ 1% BSA. The cell bound counts were then determined in the gamma counter. The pH of the incubation mix was monitored and adjusted with 1 N HC1 to be either pH 6.0 or pH 7.0. When competition experiments were carried out, the iodinated protein was first mixed with the unlabeled protein, then added to the cells, thus the cells were exposed to the radioactive and non-radioactive protein at the same time.  III.  Pandex experiments For the non-radioactive experiments, a Pandex Fluorescence Concentration  Analyzer (Idexx, Westbrook, Maine) was used to measure the fluorescence of the cells in a specialized 96 well plate. The plate has a 0.22 u.m cellulose acetate membrane rather than a solid bottom, to allow the wells to be drained under vacuum. This enables the washing of the cells to take place rapidly without removing the cells from the plate, and allows removal of the supernatant without centrifugation.  40  For these experiments, the cells were washed three times in serum free Hams F12 media (with 2 m M glutamine, 20 m M HEPES, pH 7.4) and placed at 37°C in a 5% C 0 humidified incubator for 20 minutes.  2  The cells were then harvested with Versene,  counted with a hemocytometer, and resuspended at 1.25 million cells per tube in 1 ml Pandex buffer (serum free D M E M media with 0.1% sodium azide and 1% BSA). The cells were then incubated for 45 minutes at 4°C with either 1 ml of Pandex buffer, 1 ml of 30 u.g/ml human T f in Pandex buffer, or 1 ml of 30 u,g/ml human p97 in Pandex buffer. The tubes were then centrifuged for 5 minutes at 4°C at 1200 revolutions per minute in a Sorval benchtop centrifuge. The cells were placed on ice and washed with the addition of 1 ml of cold PBS with 0.1%) sodium azide and 2 mg/ml B S A , and centrifuged as before. The final incubation involved the addition of fluorescein isothiocyanate (FITC) conjugated L235 or anti-Tf at a 1 in 75 dilution, for 45 minutes at 4°C followed by deposition of 40 u,l of cells (approximately 250, 000 cells) into the wells of the Pandex plate. The Pandex plate was then placed in the Pandex reader where the cells were washed 3 times, with fluorescent readings taken between each wash. For the FITC labeling of the antibodies, FITC was added at 1 mg/ml to phosphate buffer at pH 9.5 (0.15 M Na2HP04) and then 0.5 ml of the antibody at 4 mg/ml was added to 0.15 ml of the FITC solution at 1 mg/ml, and incubated overnight at room temperature in the dark.  IV.  Protein dialysis and iron loading Both p97 (Synapse Technologies, Inc., Vancouver) and T f (Sigma) were supplied  with iron partially bound to the protein. A dialysis protocol developed by Aisen and  41  Leibman was employed (Aisen and Leibman, 1968; Aisen and Leibman, 1972) in order to rid the protein of this tightly held ligand, and produce apo-p97 or apo-Tf. In this procedure, the solublized protein was injected into a Pierce 10,000 M W dialysis cassette and dialyzed in 1 L of 0.1 M sodium citrate buffer (pH 6.0) at 4°C for three hours, with buffer changes after each hour. Following this, the protein was transferred to a new cassette and then dialyzed for four hours (with three buffer changes) in 0.1 M sodium perchlorate, and 25 m M HEPES at pH 7.4 at 4°C. Finally, the protein (using the same cassette) was dialyzed for three hours (with two buffer changes) in 25 m M HEPES at pH 7.4 at 4°C. The protein was then removed from the dialysis cassette and concentrated in a Centricon-TMIO (Millipore, Bedford, M A ) , and the protein concentration was measured either by spectrophotometer readings at 280 nm (with an extinction coefficient of 1.214 for p97) or by B C A (bicinchoninic acid) protein assay (Pierce). To load the protein with iron, the iron chloride (either radioactive or not) was first mixed at a ratio of 1:1 with nitrilotriacetate (1 M each) for 10 minutes at room temperature.  Then the dialyzed apo-p97 or apo-Tf was mixed with the ferric  nitrilotriacetate at 2 moles iron per mole of protein. The pH for this step must be around pH 7.4 for proper iron loading, so the protein may need to be diluted in HEPES buffer to compensate for the low pH of the ferric nitrilotriacetate solution, especially for Fe, as it 55  is supplied in concentrated HC1. The mixture was allowed to sit at room temperature for one hour before the excess iron was dialyzed out of solution in three changes of 25 m M HEPES for 3 hours at ph 7.4, at 4°C.  The iron-loaded proteins undergo a color change,  from colorless in the apo-form to slightly pink in the iron-loaded form.  42  V.  Iron uptake studies The method for radioactive iron uptake in vitro involved measuring the  internalized radioactivity of the various TRVb cell lines transfected with the TfRl or TfR2.  The cells were grown in six well tissue culture plates (Nunclon) until  approximately 80 to 90% confluence was reached. The plates were then washed three times, each for 30 minutes at 37°C with 5% CO2 and humidity, with Hams F-12 serum free media plus 1% (w/v) B S A . After washing, the incubation mix containing either 55  Fe-loaded T f or p97 in Hams F-12 serum free media with 1% B S A was added to the  cells. After incubation at 37°C with 5% CO2 and humidity for the appropriate amount of time, the cells were washed three times (on ice) with 4°C wash buffer (PBS with 1% BSA) and then harvested with Versene. Finally, the amount of F e was counted in the 55  Beckman LS6000IS Liquid Scintillation counter.  VI.  Immunoprecipitations Immunoprecipitations were carried out on T R V b cells, with and without the  T f R l . The cells were grown to approximately 90% confluence in tissue culture plates. The adherent cells were incubated three times for 20 minutes each time at 37°C with serum free Hams F12 media (with 2 m M glutamine, 20 m M HEPES, and 1% B S A at pH 7.4) in a 5% CO2 humidified incubator. This step was designed to remove any residual bound bovine Tf, and to slightly iron starve the cells. The cells were then incubated with 1 ml Fe-loaded p97 or T f in serum free media for four hours at 37°C in a 5% CO2 55  humidified incubator. After the incubation, the incubation mix was removed, the plates were washed with PBS, and the cells were lysed in 1 ml 1%> (v/v) Nonidet-P40 (NP-40)  43  (ICN Pharmaceuticals Inc., Costa Mesa, CA) in lysis buffer (20 m M Tris-Cl pH 7.4, 150 m M NaCl, 2 m M E D T A and 1 tablet protease inhibitor cocktail (Roche)) for 30 minutes on ice. The lysate was then collected in an eppendorf tube and centrifuged in a Biofuge B microcentrifuge (VWR Canlab Mississauga, ON) at 11,000 rpm at 4°C for 10 minutes. The supernatant was then collected and 3 Ltl of normal rabbit serum was added. The tube was rotated at 4°C for an hour. Next, 30 | i l of protein G sepharose (50% (v/v) slurry) (Amersham-Pharmacia Biotech, Piscataway, NJ) was added, followed by additional rotation at 4°C for an hour. The samples were then centrifuged at 11,000 rpm for 5 minutes at 4°C, and the supernatant transferred to a new eppendorf tube. The lysate was then divided into the appropriate number of samples, and the primary antibody added. Antibodies used were L235 (against human p97), polyclonal anti-human T f R l (against classical tranferrin receptor), and anti-ferritin (against ferritin) (See Table 1). After an incubation at 4°C for at least one hour (up to overnight) the immuno-complexes were precipitated with protein G sepharose (30 Lil), for up to one hour at 4°C with rotation. The protein G immuno-complexes were washed twice in ice cold buffer B (0.2% (w/v) NP-40, 10 m M Tris-Cl pH 7.5, 150 m M NaCl, 2 m M E D T A ) , each time followed by centrifugation at 4°C and suction of the supernatant. Next, the samples were washed once as above with Buffer C (0.2% (w/v) NP-40, 10 m M Tris-Cl p H 7.5, 500 m M NaCl, 2 m M EDTA) and once with Buffer D (10 m M Tris-Cl pH 7.5) before resuspension in 1 ml of PBS and transfer to scintillation vials.  The vials had 5 ml of ReadySafe  scintillation counting fluid (Beckman Coulter Inc., Fullerton, C A ) added, and the disintigrations per minute were determined on a Beckman LS6000IS Liquid Scintillation counter.  44  The variation to this protocol involves the co-immunoprecipitation of the TfR 1 and 2 from the TRVb3 cells. In this case, no radioactivity was used. Instead, the cells were simply washed with PBS, then harvested as above. The protocol is the same as that above until the final step.  Here, instead of placing the lysate in PBS and adding  scintillation fluid, the beads were solubilized for an SDS-PAGE gel by adding 40 ul of Mix II (1 ml of Bromo M i x (36% (w/v) sucrose, 0.01%) (w/v) bromophenolblue, 0.1 M Tris-Cl p H 8.8), and 200 ul M i x I (18% (w/v) SDS, 0.05 M dithiothreitol (DTT) (Roche)). The beads in Mix II were heated at 95 °C for 2 minutes, then cooled to room temperature for 5 minutes, followed by a brief spin. Then, 5 u.1 of 0.5 M iodoacetamide (IAA) (Sigma-Aldrich Canada Ltd.) was added and the samples were allowed to sit for 15 minutes before being loaded on a standard 10% (w/v) S D S - P A G E gel and electrophoresed at 150 V for 1 hour. Next, the gel was transferred to a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, C A ) by electroblotting. Following the transfer, the blotted membrane was blocked with a 5%> (w/v) skim milk and 0.1 %> (v/v) Tween 20 (Biorad) in PBS for 1 hour (to overnight) and then the proteins were detected using antibodies against the TfRl (1:100) or against the flag-tag on TfR2 (1:100). The blotted membrane was incubated with the antibody for 1 hour at room temperature or overnight at 4°C (both with shaking) and then washed for 3 times 30 minutes with the 5%> (w/v) skim milk wash buffer. After the washes, the blotted membrane was incubated in peroxidase-conjugated goat anti-mouse or peroxidase-conjugated goat anti-rabbit (both at 1:10,000 dilution) in the skim milk buffer, followed by the same wash protocol used for the primary. The blots were then treated with the chemiluminescence E C L Western blotting detection system (Amersham Pharmacia Biotech) and exposed to x-ray film.  45  VII.  Surface plasmon resonance (BIAcore) experiments A BIAcore 2000 biosensor system (Pharmacia Biosensor, Uppsala, Sweden) was  used to assay the interaction of p97 with 0 ^ 3 integrin (kind gift of Alistair Henry of Celltech), vascular endothelial growth factor receptor 2 (VEGFR-2) (R&D Systems), and the T f R l (kind gift of Pamela Bjorkman, California Institute of Technology). The extracellular domains of the oc and P3 subunits were linked at the C-terminal end to the v  Fc of mouse IgGl to enforce heterodimerization .(Stephens et al., 2000) The soluble version of the human TfRl was generated through truncation at the C-terminal end, to remove the intracellular and transmembrane domains (Lebron et al., 1998). The soluble TfRl was purified in the Bjorkman lab from supernatants of baculovirus-infected High 5 cells using N i - N T A chromatography (Ni-NTA superflow; Qiagen) and followed by gel filtration chromatography with a Superdex-200 F P L C column (Pharmacia). On the BIAcore sensor chip carboxymethylated dextran matrix (Sensor Chip CM5, Research Grade, BIAcore) was preactivated with N-hydroxysuccinimide/ N-ethylN'-[3-(diethylamino)propyl] carbodiimide according to the  manufacturer's  recommendations. A l l steps were carried out at room temperature. For the experiments, random amine coupling of the p97 and 0 ^ 3 integrin to the chip was achieved by subsequent injection of 10 ng/ml p97 in 10 m M sodium acetate at pH 4.5, and 0 ^ 3 integrin at 50 |ig/ml at pH 4.5. One cell on the chip was left blank to serve as a control, and the sensorgram generated was subtracted as a baseline from the experimental sensorgrams. To perform binding assays, samples of p97, cc P3 integrin, or v  V E G F R 2 were injected in Hepes buffer (10 m M Hepes, 150 m M N a C l , 0.005% polysorbate 20, 2 m M M g C l , 0.6 m M CaCl , pH 7.4). To regenerate the chip for the 2  2  46  experiments (between injections) a wash step was performed with 0.01 M Hepes at pH 7.4 with 3 m M E D T A . For the TfRl experiments, the TfRl protein was coupled to the chip by flowing a 0.1 mg/ml solution in 3.46 m M maleate buffer at pH6.0 across the flow cell for various amounts of time. This yielded three flow cells with increasing amounts of protein coupling, and one blank flow cell with no bound protein. The iron-loaded p97 in this experiment was injected at 10 pi and 50 pi at p H 7.4, and the apo-p97 was injected at 50 ui at pH 6.0.  VIII.  Northern blot analysis The northern blots of multiple human tissues and mouse embryos, were both  commercially prepared by Clontech Laboratories. The probes used were both prepared through polymerase chain reaction, followed by labeling with P . The primers used to 3 2  generate the c D N A that was labeled to "make the probe are listed in Table 2.3. For probe labeling, 5 pi of gel-purified D N A probe (QIAEX II Gel Extraction Kit, Qiagen Inc., Mississauga, ON) was added to 4 u.1 sterile distilled water. This mixture was denatured by heating on a heat block to 100°C for five to ten minutes, followed by a quick chill in ice. Next, 1 ui each of dATP, dGTP, and dTTP was added, followed by 2 ui of reaction mixture (solution 6) (Random Primed D N A Labeling Kit, Roche). Finally, 5 u.1 of P 3 2  dCTP and 1 ui Klenow enzyme was added and mixed by pipetting. The mixture was incubated at 37°C for 30 minutes. After the incubation, the unincorporated label was removed by eluting the probe through a G50 sephadex spin column. The ExpressHyb solution was prewarmed to 68°C, and the membrane was prehybridized in at least 5 ml of  47  warmed ExpressHyb solution (Clontech) for 30 minutes at 68°C with rotation. The probe was then added to the pre-warmed solution and incubated with the blot for at least 1 hr (to overnight) at 68°C with continuous rotation. After incubation, the membrane was rinsed in wash solution 1 (0.3 M NaCl, 0.03 M sodium citrate pH 7.0, 0.05% (w/v) sodium dodecyl sulfate) several times at room temperature, then washed for 30-40 minutes with constant shaking, and several solution changes. Next, the blot was washed with wash solution 2 (15 m M NaCl, 1.5 m M sodium citrate pH 7.0, 0.1% (w/v) sodium dodecyl sulfate) for about 30 minutes. The blot was shaken to remove excess liquid, then wrapped in plastic wrap without allowing the blot to dry (so that it could be stripped and reprobed). The blot was then exposed to x-ray film or a phosphoimager and developed. To strip the blot, a solution of 0.5% (w/v) sodium dodecyl sulfate was heated to 90-100°C, then the membrane was added and incubated for 10 minutes with shaking. The solution was then allowed to cool for 10 minutes before removing the blot. The membrane was air-dried slightly before it was put in a plastic bag for storage.  IX.  Reverse transcriptase and polymerase chain reaction The R N A from various cell lines used for these experiments was isolated using  the RNeasy Mini Kit (QIAGEN, Inc.) according to the manufacturer's instructions. To make the complementary D N A (cDNA), 5 (ig of total R N A was used, along with 1 ul of oligo d T i , and RNAse-free distilled water to make the volume 12 ul. This solution was 8  heated to 70°C for 10 minutes, followed by a quick chill and centrifugation. Next, 4 ul of 5x first strand buffer, 2 ul 0.1 M dithiothreitol, and 1 ul 10 m M dNTP mix (10 m M of each dATP, dCTP, dGTP, dTTP) was added (Gibco). This was mixed gently at 42°C for  48  2 minutes. Finally, 1 ul Superscript II (Gibco) was added, the mixture was mixed up and down by pipetting, and then incubated at 42°C for 50 minutes. After incubating, the reaction was inactivated by heating to 70°C for 15 minutes, followed by the addition of 1 ui of RNase H (Gibco) and incubation at 37°C for 20 minutes. The polymerase chain reactions to create probes and to ascertain experession in brain endothelial cells were carried out in a UNOII or Tgradient thermocycler (Biometra), using primers made by Sigma (see Table 2.3).  X.  Immunofluorescence staining and confocal laser scanning microscopy The adherent cells were grown on sterile coverslips (No. 1, 18 mm, Fisher  Scientific Inc.) to approximately 60 to 80% confluence. The coverslips were washed gently twice with PBS, then blocked for 1 hour in 2%> (w/v) B S A in PBS. The cells were then fixed for 20 minutes in 2% (w/v) paraformaldehyde in PBS, and permeabolized with 0.1%> (w/v) saponin in PBS with 2% (w/v) BSA. The coverslips were stained with 200 pi primary antibody, diluted in the appropriate amount of PBS with 2%> (w/v) B S A (see Table 2.1) for 30 minutes at room temperature with gentle shaking. Control coverslips were incubated with no first antibody. When double labeling was carried out, both antibodies were mixed together in the PBS then added simultaneously to the coverslips. Following the 30 minute incubation, the coverslips were washed five times (for two minutes each time) with 0.1 %> (w/v) saponin in PBS with 2%o B S A . The cells were then incubated with the appropriate secondary antibody conjugated to either Alexa 488 or 568  49  Table 2.3: Primers  Template  Primer  Sequence  Application  Name CTT C T G CTC T A A A A G C T G C G  Sequencing  CAG A A G GAT GGA AGT CCC  Sequencing  FFS1  A C C C T G GTC C A A G A T A T C CTC G  Sequencing  mTfR2 a  FFS2  TGC G A G T T G G A A T T A C T A GCT T C G  Sequencing  mTfR2 a  FFS3  C A C CTC T C A G G C TCT C C T T A T C G  Sequencing  mTfR2 a  FFS4  A C C T C A A A G C T G T T G TGT A C G T G  Sequencing  mTfR2 a  FFS5  CCT G C C A G G CGT G T G G G G A C  Sequencing  mTfR2a  FFS6  C A G T G G CTC A G C T C G C G G  Sequencing  mTfR2 a  FFS7  GCC CTG GTA G A C C A C CTG C G  Sequencing  mTfR2 a  FFS8  G C C GTT GCT T A C C C A G A A A G C  Sequencing  HTfRl  Probe +  CTC A A A A A G A T G A A A A T C T T G C G  Northern/ RT PCR  HTfRl  Probe -  C C A A A G A A T G A A A G T TCT G C G  Northern/ RT PCR  Actin  Probe +  T G A A G T C T G A C G T G G A C A TC  Northern/ RT PCR  Actin  Probe -  A C T CGT C A T A C T CCT GCT T G  Northern/ RT PCR  MTfR.2  Probe +  C A A CGT T G G GGT C T A CTT C G G  Northern  MTfR2  Probe -  GAT C A G GGA C C A GAT A G G GGG  Northern  MTfRl  Probe+  C T A CCT G G G C T A T T G T A A G C G  Northern  MTfRl  Probe -  GGT C T G C C T C A A C A A C G G G  Northern  HTfR2  Probe A  CGT GGT C C A GCT TCT G G C G G G  Northern/ RT PCR  HTfR2  Probe E  G T A GCT G G G T C A CGT C C C  Northern/ RT PCR  HTfR2  Probe F  CCT G G A TTT C C A C C A G G G C  Northern/ RT PCR  HTfR2  Probe G  GGC C A T GTT CCT G C A GTT C  Northern/ RT PCR  mTfR2 a  FL3+1  mTfR2 a  Sep 13-2  mTfR2 a  50  (see Table 2.2) diluted in PBS with 2% (w/v) B S A for 30 minutes at room temperature, followed by washes as above. After washing, the coverslips were equilibrated with 6 drops of Slow Fade Equilibration buffer Component C (Molecular Probes, Eugene, OR) for 10 minutes at room temperature.  After the 10 minutes, all the liquid was suctioned  from the coverslip, and the coverslips were mounted on a slide with 5 u,l of an equal mixture of 50% glycerol in PBS and Molecular Probe Slow Fade Component A . The coverslips were then sealed onto the slide with nailpolish. The BioRad Radiance Plus confocal laser scanning microscope was used to capture the images of the cells, and an N I H imaging system was used to analyze the images produced.  XI.  Statistical analysis The difference between the means of experimental groups and control groups was  compared with the use of the Student's t-test, which compares the size of the difference between means with the standard error of that difference (Motulsky, 1995).  Null hypothesis = no average difference between the two means Alternate hypothesis = there is a significant difference between the two means  t = difference between means / Standard Error of difference  SE of difference (equal N) = square root ( S E M + S E M ) 2  a  Pooled SD = VrN -l)*SD, + f N h - l ) * S D h ~ N + N -2 2  a  _  a  b  SE of difference = pooled SD V 1/N + 1/N a  51  b  2  b  Chapter 3:  Interaction of soluble p97 with transferrin receptor 1 and integrin  I.  Cellular based assays to examine p97 binding to the transferrin receptor 1  A.  Rationale Melanotransferrin (p97) is a protein expressed as two distinct forms: the GPI-  linked cell surface form and a soluble form found in serum and cerebral spinal fluid (Alemany et al, 1993; Brown et al, 1981; Food et al, 1994; Kennard et al, 1996). The presence of a soluble form produced by a splice variant rather than as a cleavage product released from the cell surface raises the possibility that soluble p97 has a distinct function from that of the GPI-linked version (Hsu, 1997; McNagny et al, 1996). As it has been established that GPI-linked p97 binds iron and can transport the bound iron into cells, the search for a physiological function of soluble p97 in this thesis began by examining possible receptors for soluble p97. Considering that p97 has a 39% homology to T f (Rose et al, 1986), possible binding and internalization of p97 by T f R l was examined, and is detailed in the first section of this chapter (see model in Figure 3.1). Prior to the cloning and generation of recombinant forms of receptors through molecular biology techniques, the characterization of receptor interactions can be very difficult. Evaluating binding of a radioligand to tissue or cells is complicated i f more than one receptor on a cell binds to the ligand or when a cell line that does not bind the ligand of interest is unavailable. Even with the receptor cloned and expressed in cells, characterization of receptor interactions is not trivial. Often, one goal of receptor-ligand  52  Figure 3.1: Model of internalization of p97 via transferrin receptor 1. In this model, soluble p97 and Tf are present in the blood stream. The TfR1s on the cell surface can bind either p97 or Tf, and internalize either ligand. The complex is then present in the endosomes, and recycled to the cell surface after iron is dissociated from the p97 or Tf.  53  binding studies is to calculate a binding affinity for the interaction.  Scatchard  experiments, performed to generate these numbers, rely on a number of assumptions in order for the results to be accurate. These assumptions are that the receptors are equally accessible to ligands, that there is no partial binding, that neither the ligand nor receptor is altered by binding, and that the interaction is reversible. Often, not all of these assumptions apply, which can render the results very difficult to interpret and lead to incorrect binding affinities.  These experiments also require substantial amounts of  radioligand with high specificity. Alternatives to these types of studies exist. One of the more powerful techniques that can be used after a receptor is cloned is to create two cell lines that can be compared directly: a "negative" cell line that does not express the protein of interest, and a "positive" cell line which consists of the negative cell line transfected with the protein of interest. Under various experimental conditions, differences measured between the two cell lines can reasonably be assumed to be due to the presence of the transfected protein. In the case of the human T f R l , a cell line for these types of assays was created in 1987, when a Chinese hamster ovary (CHO) cell line was isolated through repeated founds of mutagenesis with ricin A chain conjugated T f (McGraw et al., 1987). The cells that internalized T f through the TfR were poisoned by the ricin through its ability to act at ribosomal target sites and inhibit cellular protein synthesis (Sundan et al., 1982). The isolated mutant cell line (called TRVb, referred to hereafter as TfR-) does not bind detectable amounts of T f and was characterized to lack functional TfRs. A human TfRl construct was then transfected into these cells to produce a cell line that can bind and internalize Tf, called TRVb-1 (for clarity in this thesis TRVb-1 will be termed TfRl+).  54  To test the hypothesis that soluble p97 specifically binds to human T f R l , two different cell-based assay systems were utilized. The first assay system uses iodinated p97, and examines cell surface binding of the protein at 4°C. This is carried out either in the presence or absence of excess unlabeled p97, to compete for binding sites. This assay also compares binding at the acidic pH of 6.0 to binding at the neutral pH of 7.0 to determine i f pH has an effect on binding. The second assay uses a particle concentration fluorescent immuno-assay (or Pandex assay) to measure the amount of fluorescence on the surface of a fixed number of cells in a specialized 96 well plate. This approach uses antibodies raised against p97 rather than radioactively labeled protein, followed by a FITC-labeled secondary antibody to detect p97 binding to T f R l .  B.  Results I:  125  I p97 binding to the human transferrin receptor in cells  Iodination is a common protein labeling technique, and was the method used successfully to examine uptake of  125  I labeled T f by Maxfield and associates in the  primary characterization of the T f R l + cell line (McGraw et al., 1987). Transferrin receptor 1 deficient C H O cells (TfR-) and those transfected with the human T f R l construct p C D T R l (TfRl+) were harvested with versene. Following the harvest, the cells were briefly acid washed with glycine buffer at pH 4.0 to remove any bound bovine Tf from the tissue culture medium, washed with PBS, and aliquoted into tubes. The cells were then incubated with iodinated protein (or iodinated protein plus 100 times unlabeled protein) for 1 hour at 4°C. Finally, the cells were washed and the bound radioactivity associated with both the TfR- and TfRl+ cells was measured in a gamma counter. The difference in bound counts between the cell lines (TfR- and TfRl+) was plotted (Figure  55  £.,WU  1,500  _co "CD  o  C  1,000  .2  I  500  Q_ Q  I CrtA  Ip97 p H 6.0  1 2 5  |p97 + 100X  1 2 5  |p97 p H 7.0  cold p97 p H 6.0  1 2 5  !p97 + 1 0 0 X cold p97 p H 7.0  Figure 3.2: lp97 binding to transferrin receptor 1, with unlabeled p97 competition. l p 9 7 and l p 9 7 with 100x unlabeled p97, incubated with TfR- and TfR1+ cells at either pH 6.0 or 7.0. Represents the mean +/- standard error for a representative experiment performed in triplicate, n=4. * Statistical significance was assessed with the use of a Student's t-test (p<0.005). 125  125  125  56  3.2). Therefore, the amount of bound radioactivity plotted per million cells represents the amount of iodinated protein bound specifically to the T f R l . Figure 3.2 shows that, at pH 6.0, there is no significant binding of  125  I p97 to the  T f R l , and the addition of 100 times unlabeled protein has no effect on the amount of radioactivity bound to the cells. The reason that the plotted results are less than zero is that, although the two cell lines did not have significantly different amounts of bound radioactivity, the "background binding" or the binding measured on the TfR- cells was slightly higher than the binding on the TfRl+ cells. At pH 7.0 the amount of binding on the cells with TfRl is significantly higher (assessed with Student's t-test, p<0.005) than the amount bound to the cells without the T f R l , with the difference being approximately 1,400 disintegrations per minute (DPM) per million cells. Furthermore, when unlabeled p97 is included to compete with the  125  I p97 (at a ratio of 100:1, unlabeled p97: I p97) l25  the bound radioactivity decreases to approximately 500 D P M , which is significantly lower than the amount bound in the absence of the cold competition (assessed by a Student's t-test, p<0.005).  C.  Results II: Particle concentration fluorescent immunoassay study of p97 binding  to transferrin receptor 1 The Pandex (particle concentration fluorescent immunoassay) is a technique that allows measurement of the bound fluorescence-labeled antibody on a pre-determined number of cells. In some instances, when examining lower affinity or transient proteinprotein interactions, the Pandex can be more sensitive than flow cytometry, as bound fluorescence in the flow cytometer is measured one cell at a time in a F A C S buffer  57  solution. This can lead to some interactions being lost with flow cytometry, as the ligand is highly diluted by FACS buffer, and thus can dissociate under the conditions present. In the Pandex, specialized 96-well plates are used that have wells with filtered bottoms to allow for incubation of the cells with antibodies within the wells. This is followed by rapid washing and draining steps, all performed within the context of the Pandex fluorescence concentration analyzer (Idexx, Westbrook, Maine) (Jolley et al., 1984). Figure 3.3 shows the results of an experiment in which TfR- and TfRl+ cells were harvested with versene, briefly acid washed to remove any bound bovine Tf, and aliquoted at 1.25 million cells per tube. The cells were then incubated with either Pandex buffer alone (serum free D M E M medium with 0.1% w/v sodium azide and 1% w/v BSA) or Pandex buffer with 0.3 ug/ml soluble iron-loaded p97 for 45 minutes at 4°C. The cells were washed with cold Pandex buffer, then incubated with 200 ul of anti-p97 antibody (hyb-C-FITC) for 45 minutes at 4°C. After the incubation, 800 ul of Pandex buffer was added to the tubes and the cells were loaded into the wells at 40 ul per well (thus approximately 50,000 cells per well). The wells were then drained and washed three times, and the bound fluorescence was read by the Pandex reader. In Figure 3.3, the difference between bound fluorescence from the cells incubated in the first step with 0.3 ug/ml p97 in Pandex buffer, and those incubated,in Pandex buffer without p97 (as a measure of background fluorescence) was plotted as a representation of the p97 bound to the cells. There was no bound fluorescence detected over background on the TfR- cells, but about 1,400 fluorescent units were measured bound to the TfRl+ cells. The fluorescence bound to the TfRl+ is assumed to represent the p97 bound to the T f R l . The mean fluorescence measured on the TfRl+ cells is  58  2,000 1,800 1,600 CO  "c D t-> c CD O CO  ©  O _3 LL  1,400 1,200 1,000 800 600 400 200 0  TfR-  TfR1 +  Cell Lines Figure 3.3: Binding of soluble p97 to transferrin receptor 1, detected with the p97 antibody Hyb-C-FITC. TfR- and TfR1+ cells incubated with p97 and the anti-p97 antibody Hyb-C-FITC, and bound fluorescence determined in a Pandex reader. Fluorescent units bound to the cells are plotted after subtraction of measured background fluorescence due to antibody alone. Represents the mean +/- standard error for a representative experiment, n=7. Statistical significance was assessed with the use of Student's t-test <p<0.005).  59  significantly higher than that measured for the TfR- cell line, as assessed by the Students's t-test (p<0.005).  D.  Results III: p97 can compete with transferrin for binding to the transferrin  receptor 1 The model that p97 binds to TfRl is based in part on the high homology between p97 and the main ligand of the T f R l , Tf. Using this model, it is assumed that p97 and T f use the same or a very similar binding pocket on T f R l . If this is true, p97 should be able to compete with T f for binding to that area, and excess p97 will interfere with T f binding to T f R l . In Figure 3 A, a Pandex experiment is depicted that demonstrates p97 competition with Tf. The experiment was performed as described in section C above, except that the primary protein incubation was Pandex buffer alone (as the negative control), 0.3 ug/ml Tf, or 0.3 Ug/ml T f with 30 Ug/ml p97 as a specific competitor. The first two columns in Figure 3.4 show that on TfR- cells (with no TfRl) there is very little binding of T f over background levels, and the presence of excess p97 does not have an effect on the bound fluorescence measured. In this graph, the bars represent fluorescence due to the antibody alone subtracted from the fluorescence measured in the wells where the cells were initially incubated with protein. In this way the graph depicts only the fluorescent units measured above background. The third column in Figure 3.4 shows the fluorescent units bound to the TfRl+ cells, which is approximately 24,500 fluorescent units. The T f clearly binds to the T f R l , as can be determined by comparing column 1 with column 3. The final column shows that, in the presence of a 100 fold excess of p97 (30 ug/ml of  60  Figure 3.4: Binding of transferrin to transferrin receptor 1, with and without competition by excess p97. TfR- and TfR1+ cells were incubated with Tf, or Tf and 100x concentration of p97, then Tf binding to the cells was measured with an anti-Tf antibody in a Pandex reader. Fluorescence is plotted after subtraction of background fluorescence due to antibody alone. Represents the mean +•/- standard error for a representative experiment n=8. Statistical significance was assessed with the use of a Student's t-test (p<0.005).  61  p97 with 0.3 jxg/ml Tf), the amount of bound T f is significantly reduced. The difference between binding measured in the presence of T f alone and binding of T f with excess p97 present is significant when analyzed with the Student's t-test (p<0.005).  E.  Discussion The results from the binding experiments support the hypothesis that p97 can bind  to T f R l , under the in vitro conditions used for these experiments. As outlined in Figure 3.2 it is clear that a significantly greater amount of radioactively labeled p97 is able to bind to the cells with TfRl compared to the cells without T f R l , and that this binding is pH dependent. The pH dependence of T f binding to TfRl and of iron binding to T f has been well documented in a study by Dautry-Varsat and associates (Dautry-Varsat et al, 1983). This study demonstrated that diferric T f binds strongly to T f R l at pH 7.0, and that as the pH lowers to that approximating the endosome (about pH 5.0) the T f loses its bound iron. At the lower pH, apo-Tf remains bound to the T f R l . Apo-transferrin has a high affinity for TfRl at the pH of approximately 5.5, but rapid dissociation of apo-Tf occurs at pH 7.3. This may be similar to the situation occurring with p97. The iron loaded p97 does not show any measurable binding affinity for T f R l at the lower pH of 6.0 used in this experiment, while significant binding was measured at the neutral pH. Consistent with this, Figure 3.2 also demonstrates that the iodinated p97 bind to the TfRl at pH 7.0 in a specific manner, in that it can be inhibited by the inclusion of excess unlabeled p97. The Pandex experiments (depicted in Figures 3.3 and 3.4) use a technique that allows the incubation of cells with protein and antibody followed by wash steps that are  62  carried out very rapidly to help minimize the effects of dilution on the dissociation of the ligand-receptor interaction. Prior to using the Pandex assay, multiple attempts to measure p97 binding to T f R l under various conditions with flow cytometry were not able to demonstrate measurable binding. Perhaps the period of time the cells are diluted in FACS buffer causes p97 to dissociate from the receptor, due to the low affinity nature of the interaction. Thus, the Pandex analysis is a more powerful technique than F A C S because the wash steps are minimized and the cells are not in solution for long periods after the antibody solution is washed away. Using the Pandex it was clearly demonstrated that the p97 associates with the TfRl+ cells and not the TfR- cells (Figure 3.3). These experiments were carried out at neutral pH, and confirm the cell surface binding observed with the iodinated p97 in Figure 3.2. Further, the presence of excess p97, but not excess B S A in the Pandex buffer, is able to compete for binding with the Tf, and thus decreases the binding of T f to TfRl+ cells (Figure 3.4). Together with the data shown in Figure 3.2, this demonstrates that p97 is binding specifically to the T f R l .  It is important to note that the difference in  fluorescence units for p97 and T f (shown in Figures 3.3 and 3.4 respectively) does not reflect relative binding affinities of T f and p97 for T f R l , but rather it is a measure of the amount of fluorescent labeling on the antibody. Another contributing factor is that the anti-Tf antibody is a polyclonal antibody while the anti-p97 antibody is monoclonal. This could also increase the fluorescent units observed in Figure 3.4, as more antibody may bind to the T f than to the p97 because the polyclonal T f antibody can bind multiple antigen sites on the Tf, possibly allowing greater amplification of the fluorescent signal. The fact that the presence of 100 times p97 only reduces T f binding to the TfRl by 30%  63  indicates that the affinity of p97 for the TfRl is much lower than the affinity of Tf, under the conditions of this experiment. The three experiments described above all measure p97 binding to the surface of cells, and were carried out at 4°C to ensure that internalization of the receptors from the cell surface did not take place (Dautry-Varsat et al., 1983). The Pandex assay is not designed to measure uptake and intracellular accumulation of ligand. The iodinated protein assay is often used to measure endocytosis in this way, but despite many attempts, uptake of the  125  I p97 at 37°C could not be conclusively shown to occur above the level  of background, while uptake of  125  I T f was clear (data not shown). This may indicate that  the binding of p97 to the T f R l has very low affinity or does not lead to protein internalization through endocytosis, but it may also be a problem with the iodination of the protein. Transferrin iodinates easily and the modification does not adversely affect the ability of T f to bind to the TfRl (Klausner et al., 1983). Lactoferrin, another iron binding protein that shares high homology with p97 and T f (Metz-Boutigue et al., 1984), does sustain damage to its monomeric form during iodination, and is prone to formation of tetramers after the modification (Rosenmund et al., 1986). Iodination does not seem to cause excessive polymer formation in p97, and iodinated p97 is able to bind iron after iodination, which is demonstrated through the color change of the solution from colorless to pink. It is not possible to test what proportion of the iodinated p97 is altered after iodination, since no functional assay exists. If the percentage of functionally labeled p97 that is present after iodination is very small, higher background may occur, and this would make the internalization assay very difficult to measure over non-specific counts. This may also explain why, despite many attempts and various protocols, kinetic data on  64  the interaction of p97 and T f R l was not obtainable.  Future experiments using  metabolically labeled p97 (such as S ) or iodination through other techniques that 35  produce less incorporation might be able to address these problems.  II.  Uptake of radioactive iron bound p97 by transferrin receptor 1  A.  Rationale It has been established that GPI-linked p97 expressed on the surface of cells is  capable of internalizing bound iron and delivering it to ferritin (Kennard et al., 1995; Tiong, 2001). If the soluble form of p97 binds to the T f R l in a manner similar to Tf, perhaps the function of this interaction is to allow additional iron delivery to the cells, or iron delivery under certain specialized conditions. The ability of p97 to bind to TfRl under controlled tissue culture conditions has limited application to identifying and understanding the function of the interaction.  To understand i f p97 is actually  functioning in a manner similar to that of Tf, an examination of net iron uptake is necessary. To test the hypothesis that iron bound to p97 is delivered to cells via the interaction between p97 and the T f R l , and that the internalized iron is delivered to the iron storage protein ferritin two cell lines (TfR- and TfRl+) were compared in an assay similar to those of section I.A.. In addition, the p97 was loaded with radioactive F e , 55  and the accumulation of the Fe within the cells over time was determined. To measure 55  if the iron from the p97 is delivered into the classical iron storage pathway of the cell (namely to the storage protein ferritin) an immunoprecipitation experiment was  65  performed. This allows the tracking of the radioactive iron through p97, T f R l , and ferritin.  B.  Results 1: Fe-p97 uptake by cells expressing transferrin receptor 1 55  In the iron uptake experiment, the TfR- and TfRl+ cells were grown in 6 well tissue culture plates. Upon reaching approximately 80% confluence, the plates were washed with PBS, then incubated with serum free Hams F12 media three times, for 30 minutes each to remove bound bovine T f and to slightly iron starve the cells. Following the washes the cells were incubated with the serum free Hams F12 medium with 1% B S A and the radioactive iron-loaded p97 for 2 hours at 37°C, washed at 4°C and then harvested. The D P M of the internalized Fe was determined in a scintillation counter. 55  Figure 3.5 shows that the internalized F e was not significantly different between the 55  two cell lines (as measured by Student's t-test analysis).  C.  Results II: Immunoprecipitation of p97, transferrin receptor 1, and ferritin If soluble p97 were to deliver its iron ligand to cells, then iron may enter the  regular iron storage pathway of the cell. If the iron follows the same path as that released from Tf, excess iron brought into the cells by p97 should be shuttled to the iron storage protein ferritin. In order to assess the contribution of iron loaded p97 to the standard iron storage pathway, the amount of radioactive iron bound to ferritin was measured. In this experiment, the TfR- and TfRl+ cells were incubated with Fe-loaded p97 55  and Fe-loaded T f for 2 hours at 37°C. 55  Following this incubation, the cells were  washed, harvested, and an immunoprecipitation was carried out using antibodies against  66  2,500  fl) C  2,000  "E  TfR-  TfR1 +  Figure 3.5: Uptake of Fe-p97 by ceils with and without transferrin receptor 1. 55  The TfR- and TfR1+ cells were allowed to internalize Fe-loaded p97 for 2 hours and the internalized counts measured in a scintillation counter. Represents the mean +/- standard deviation for a representative experiment, n=6. 55  67  •  nn  an6-p97 anli-T(R anti-tetntin  TfR»Fe-p97  TfR+ 55  TtRfe-T»  5S  jnnni TfR+ Fe-Tf  Figure 3.6: Immunoprecipitation of ferritin, p97 and transferrin receptor 1 from cells after incubation with ^Fe-loaded p97 or transferrin. The Tf R-  and TTR1+ ceils were incubated with Fe-loaded protein for 2 hours, then immunoprecipitated with antibody against ferritin, p97 or Tf R1. The immunoprecipitated counts were measured in a scintillation counter. A representative experiment is shown. 55  68  p97, T f and ferritin. The results shown in Figure 3.6 depict that significant amounts of radioactive iron are measured in conjunction with the iron storage protein ferritin only in the case of the TfRl positive cells incubated with Fe-loaded Tf, but not p97. Higher 55  levels of Fe remain associated with the p97 immunoprecipitated from both TfR- and 55  TfRl+ cell lines compared to Tf, which seems to retain almost no radioactive iron.  D.  Discussion Under the experimental conditions outlined above, it has been determined that  55  Fe-loaded p97 does not deliver a significant amount of iron to the cells examined via  the T f R l , therefore the null hypothesis is supported. Indeed, in Figure 3.6 it is clear that the amount of iron, while very similar between the two cell lines, may be slightly higher for the cell line without the TfRl (although the iron uptake is not statistically different between the two cell lines). Further, as shown in Figure 3.6, the iron that does get internalized into the cells via iron-loaded soluble p97 is not delivered to the iron storage protein ferritin.  In addition, the iron that enters the cells when the TfR- cells are  incubated with Fe-loaded T f also does not get delivered to ferritin. This may indicate 55  that in these cases the iron that is being measured may be background binding to the cell surface, or may be entering a different pathway within the cell. The amount of F e 55  associated with ferritin in the TfRl+ cells incubated with F e - T f is substantially higher 55  than the amount measured in any other case. This result is consistent with the published data on holo-Tf iron delivery via T f R l ! Bottomley and associates found that in K562 (erythroid leukemia) cells, 85% of the iron taken up via this pathway is delivered to  69  ferritin (Bottomley et al., 1985). It seems clear that holo-p97 is not functioning here in the same manner as holo-Tf. In Figure 3.6, the  amount  of iron  associated  with p97  after  the  immunoprecipitation of both cell lines is higher than that associated with Tf. The fact that the T f immunoprecipitated from the TfR- cell line does not have iron associated with it indicates either that very little T f associates with the cells i f the TfRl is not present, or that the T f that is non-specifically associated with the cells allows its iron to dissociate. From the known dissociation behavior of iron from T f the former explanation seems more plausible: that T f does not bind measurably to the TfR- cells and is thus not present to be immunoprecipitated after the wash steps. Unlike Tf, p97 shows a high degree of non-specific binding to the cells. This is apparent as only a small increase in bound counts are present in the immunoprecipitation from the TfRl+ cells compared to the TfRcontrol cells. This association with both cell types may indicate that p97 has a much higher propensity for non-specific binding than Tf, or that an as yet unidentified specific receptor for p97 exists on both cell types. This phenomenon should be investigated further.  III.  Surface plasmon resonance studies to measure binding of soluble p97 to  soluble transferrin receptor 1 and the integrin 0CvP3  A.  Rationale While cell based assays can be a powerful technique for examining interactions  between proteins, sometimes the results can be difficult to interpret due to the presence of many factors that can not be controlled in complex biological systems. For this reason,  70  biosensors to detect the interaction between isolated proteins can be a useful tool to supplement information gained in other ways. One biosensor, called BIAcore, measures biomolecular interactions in real time (Nice and Catimel, 1999). The system utilizes a sensor chip with a gold film to which a ligand (protein, peptide, oligonucleotide, or carbohydrate of interest) is immobilized, allowing measurement of analyte binding to a ligand. Unlike in the field of biological receptors, in BIAcore terminology, the ligand is the stationary substrate, fixed on the BIAcore chip. The analyte is the substance in solution, flowing over the BIAcore chip. The surface of the BIAcore chip has a dextran matrix (see Figure 3.7) that is covalently linked to the gold film (Hashimoto, 2000). Before the ligand can be bound to the chip, the surface of the chip is first activated by using a mixture of N-ethyl-N'(dimethyl-aminopropyl)carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS). Once activated, the NHS-ester groups react with amino groups in the ligand molecules and thus immobilize the ligand. Following this immobilization, the rest of the active groups in the matrix are deactivated by injecting 1 M ethanolamine-HCL through the flow cells. In the BIAcore 2000, four flow cells can be used on the chip, so three different ligands, or three concentrations of the same ligand can be loaded onto the cells, and the fourth is maintained as a blank cell to monitor background binding interactions. One chip can be used multiple times with regeneration steps between experiments, as long as the ligand-analyte interaction is reversible. Some interactions, such as many between an antibody and antigen, are of such high affinity that the treatments required to separate the two binding partners are very stringent, and damage the ligand or the chip in the attempt to regenerate the chip.  71  Figure 3.7: Diagram of Surface Plasmon Resonance detection system.  72  73  Once the chip is prepared, the analyte can be injected and allowed to flow over the chip. The BIAcore measures the change in surface plasmon resonance that occurs due to the change in molecular mass near the gold film when the analyte binds to the ligand on the chip. Specifically, a polarized light is focused onto the gold surface of the chip, and while much of the light is reflected, some of the energy is transferred to electrons in the gold (plasmons), thus causing a drop in the amount of reflected light. The binding of the analyte to the ligand causes a change in mass, and thus a change in the refractive index of the gold. This is measured as a drop in the intensity of the reflected light at a specific angle of reflection. The changes in refractive index that are measured represent the change in mass that occurs as the ligand binds to or dissociates from the analyte. The information is formulated into a sensorgram that graphically depicts the association and dissociation phases of the binding interaction. The sensorgram shows resonance signal versus time, with the units of the resonance signal being the resonance unit. A shift in the resonance angle of 0.1° is represented by 1000 resonance units (RU) (Hashimoto, 2000).  With proper controls, including  measurements at various concentrations of bound ligand on the chip and analyte, as well as proper analysis, the BIAcore can generate a dissociation constant that is comparable to those generated through older methods, such as a Scatchard analysis. In a Scatchard experiment, binding experiments are carried out to equilibrium, and then bound ligand is plotted vs. free in a Scatchard plot (Rovati, 1993; Rovati, 1998). Two hypotheses were tested using this method: 1) p97 binding to TfRl and 2) p97 binding to the integrin owpV  Although the first major, effort towards revealing a  functional receptor for p97 focused on a role in iron transport and thus the T f R l ,  74  homology to T f is only one model that can be applied to predict possible interactions for p97. A second model is that p97 may bind to an integrin. The family of related heterodimeric receptors which integrate the intracellular cytoskeleton with the extracellular matrix in order to achieve migration and adhesion of the cells was first termed integrin in a review article (Hynes, 1987). The receptors are made up of two subunits, an a subunit (one of eighteen individual a subunits) and a |3 subunit (one of eight in mammals) (Plow et al., 2000). Individual receptors are often capable of binding multiple ligands. Integrins containing an ou, a$, a%, anb, or av component recognize the RGD motif (arginine, glycine, and aspartate amino acids) in their respective ligands (van der Flier and Sonnenberg, 2001). p97 has an R G D motif at residues 190-192. Through comparison of the p97 sequence to the sequence and crystal structure of Tf, it is likely that the R G D motif is exposed on the surface of p97, and may be available for integrin binding. The R G D motif is known to be an important component of the recognition motif of ligands for a family of integrins.  The significance of the R G D was first  identified through the determination that the fibronectin receptor (a5(3l) binds to fibronectin in a manner dependent upon the R G D motif of fibronectin (Pytela et al, 1985). It was subsequently determined that a related receptor binds vitronectin in the same R G D dependent manner (Pytela et al., 1985).  Integrin avP3 is a good choice to  study not only because it binds ligands through an R G D motif, but because p97 is an angiogenic protein (Sala et al., 2002), in that it leads to the formation of blood vessels in chorioallantoic assays, and the integrin av03 has been linked to angiogenesis in tumors (Brooks et ai, 1994a; Brooks et al., 1994b). For this reason, integrin avfo, which has been described as a "promiscuous" integrin and binds to many different ligands including  75  vitronectin, angiostatin and osteopontin (Xiong et al., 2001), was examined for its ability to bind p97.  B.  Results I: p97 binding to soluble transferrin receptor 1 measured by BIAcore The data from the cellular studies suggest that p97 may bind to the TfRl under  certain conditions, but kinetic data to determine binding constants were not obtainable through the traditional methods already employed. To determine i f soluble p97 binds to the soluble T f R l , and to measure this interaction, a series of BIAcore experiments were carried out. The soluble TfRlwas prepared and purified in Dr. Pamela Bjorkman's lab, as outlined in Chapter 2. The CM5 sensor ship was prepared with T f R l coupled to the chip by direct amine coupling. The coupling densities of the TfRl on the chip were varied, with the first cell blank (acting as the negative control), the second with 110 response units bound, the third with 309 response units, and the fourth with 670 response units. The T f injection did show binding to the T f R l , and the level of binding with the three flow cells corresponded to the amount of TfRl loaded onto each cell (see Figure 3.8a). The flow cell with the lowest amount of bound TfRl (shown as the bottom curve in figure 3.8a, just above the line indicating the blank flow cell), shows the least amount of association when the T f flows over. The upper-most curve, which is flow cell 4 with 670 response units of TfRl bound, shows the highest amount of association of Tf. There was, however, no binding observed with different concentrations of holo-p97 at pH 7.4 (Figure 3.8b shows 50 U.M p97, 10 u.M p97 not shown), or with apo-p97 at pH 6.0 (see Figure 3.8c). In both cases, with the blank flow cell used as a negative control and subtracted from the other flow  76  F i g u r e 3.8: S e n s o r g r a m o f t r a n s f e r r i n a n d p97 b i n d i n g t o s o l u b l e t r a n s f e r r i n r e c e p t o r 1. The sensor chip is loaded with 110, 309, and 670 response units of soluble TfRl on flow cells 2, 3, and 4. a) 1 p M holo-Tf binding to three concentrations of TfRl bound to the BIAcore chip, b) 50 u M holo-p97 at pH 7.4 binding to T f R l . c) 50 u M apop97 at pH 6.0 binding to T f R l . The bars above the sensorgrams indicate time of protein injection.  77  a)  1 uM h o l o - T f 140 | 120 100  g c o  Q. «  80 60 40 20 0  •Ml  -20 -40 200  250  300  350  450  400  Time (seconds)  b)  5 0 uM  -100  holo-p97  100  200  300  400  Time (seconds)  60  78  500  600  cells to zero the response units, the amount of binding is at zero or just above zero response units. No rise in signal indicating an association rate or down slope indicating a dissociation rate is observed for p97.  The length of injection is indicated on the  sensorgram as a bar.  C.  Results II: p97 binding to soluble integrin 0CvP3 A BIAcore 2000 biosensor system was used to assay the interaction of p97 with  a P3 integrin (the kind gift of Alistair Henry of Celltech pic) at 25°C. The extracellular v  domains of the oc and P3 subunits were linked at the C-terminal end to the Fc of mouse v  IgGl to enforce heterodimerization (Stephens et al., 2000). The carboxymethylated dextran matrix was preactivated with N H S according to the manufacturer's recommendations.  Random amine coupling of the p97 or a P 3 v  integrin to the chip was achieved by subsequent injection of 10 Ug/ml p97 in 10 m M sodium acetate at pH 4.5, and oc P3 integrin at 50 ug/ml at pH 4.5. One cell on the chip v  was left blank to serve as a control, and the sensorgram generated from the control cell was subtracted as a baseline from the experimental sensorgrams.  To perform binding  assays, samples of p97 or a P3 integrin were injected in Hepes buffer (10 m M Hepes, 150 v  m M NaCl, 0,005% polysorbate 20, 2 m M M g C l , 0.6 m M C a C l , pH 7.4). 2  2  In Figure 3.9a) the integrin 0 ^ 3 was bound to the chip, and fibronectin was injected over the flow cell at 50 ug/ml as a positive control. The sensorgram is shown with the blank subtracted. The binding of the fibronectin to the integrin is clear from the shape of the rising curve during the injection and the decreasing curve after the injection.  79  Figure 3.9: Sensorgram of fibronectin and p97 binding to the integrin oc P3. The sensor chip in a) through e) is loaded with the heterodimer oc p3, while the flow cell in f) is loaded with p97. a) 50 pg/ml fibronectin is injected over a flow cell with integrin 0 ^ 3 . b) 50 Ug/ml of apo-p97 injected over integrin oc p3. c) 50 Ug/ml of holo-p97 injected over integrin a p 3 . d) 1 mg/ml of holo-p97 injected over integrin a p 3 . e) 1 mg/ml of holo-p97 with 1 m M manganese injected over integrin 0 ^ 3 . f) 400 Ug/ml integrin 0 ^ 3 with 1 m M manganese injected over flow cell with p97 loaded. Bars above the sensorgrams indicate time of protein injection. v  v  v  v  v  80  a)  50 ug/ml fibronectin  "c Z>  CD  50  V)  C o Q. V) CD  CC  0  -100 .  1000  800  1200  Time (sec)  b)  , 00  0  C  =) w  50 ng/ml apo-p97  J ,  -50  C  CC -ioo -150  -200  300  400  500  600  Time (sec)  C) 50 ug/ml holo-p97 20  a ° CO OT.20 O CL  W-40 -60^  400  500  Time (sec)  81  600  1 mg/ml holo-p97  d) 1001 Xi  so  'c =) 0  CO  0  c o  CL CO  0?  5 0  -100 •  -150-  400  300  500  600  Time (sec)  e)  1 mg/ml holo-p97 + 1 m M Mn  W  200  9i c  o  o  CL CO  0  -200  GC -400  i  f  1000  500  0  Time (sec)  f)  400 Lig/ml  ap+ v  3  1mM Mn  GO  -f-J  c  ID  0  H  0  CO § "50 CL CO  0  DC  -ioo -150 1000  1500  Time (sec)  82  2000  Approximately 100 response units difference are measured. Figures 3.9b)- e) also show the integrin a p 3 bound to the chip, now with various concentrations of injected p97 as v  the analyte. Figure 3.9b) shows the injection of 50 ng/ml of iron-free (apo) p97, while Figure 3.9c) shows the same concentration of iron-loaded (holo) p97 injected over the integrin. In Figure 3.9 d) a much greater concentration (1 mg/ml holo-p97) is injected, and in Figure 3.9e) the same 1 mg/ml holo-p97 is injected, but with the addition of 1 m M manganese. The manganese is included as a known activator of integrins (Diamond and Springer, 1994). In all these cases, no difference from the baseline is observed in the sensorgrams. (All graphs are shown with the blank flow cell subtracted.) Figure 3.9f) shows the injection of the integrin a P 3 at 400 Lig/ml in buffer with 1 m M manganese v  over the chip bound with holo-p97. The binding of the analyte to the ligand was lower than the binding observed to the blank. No binding above background was observed in any of the conditions except the positive control. The length of the injections into the flow cells is indicated in Figure 3.9 by the gray bar over the sensorgram.  D.  Discussion In both BIAcore experiments, the null hypothesis is supported.  p97 does not  show measurable binding to the T f R l or integrin a p V The positive controls bound in v  both series of BIAcore experiments, which indicates that the loading of the chips was successful and that the ligands are functioning properly in terms of binding ability. The fact that no binding was observed in the experimental injections, where the binding of p97 to the TfRl or the integrin a P 3 was assessed, indicates that there is no measurable v  binding under the tested conditions. While this is strong evidence that no high affinity  83  interaction occurs between the two proteins, it does not rule out the possibility of biological interactions between the two proteins in very specific situations, or under particular conditions required for binding that have not been determined and tested. The level of sensitivity of the BIAcore 2000 approaches the low ng/ml concentration level, which enables the system to measure interactions at physiologically relevant concentrations (Nice and Catimel, 1999). The real sensitivity of the BIAcore, however, depends on a number of factors. These include the relative masses of the ligand and analyte, the level of immobilisation on the chip as well as the orientation of the ligand on the chip, the rates of both association and dissociation, and the volume and concentration of the injection (Nice and Catimel, 1999). For example, i f the analyte is of much smaller molecular mass than the ligand, the change in refractive index when the analyte molecule binds will be very small, creating an extremely low signal. This, however, should not be a major problem in these experiments, as both the ligands and analyte molecules are fairly large. In addition, while the immobilisation of the ligand to the chip seems to have been quite successful, there is no way to determine definitively how much of the ligand is oriented correctly. Direct amine coupling which was used in both experiments has the disadvantage of being orientationally non-specific, in that the protein of interest can be affixed to the matrix via any reactive amine group. Thus, this method does not guarantee that the majority of the bound ligand will be oriented in a way that promotes binding by the analyte. A n alternative to this method would be to attach a linker molecule, such as a ligand-specific antibody covalently to the matrix, and then attach the ligand to the antibody (although the binding of the ligand to the chip may be slightly less stable under  84  these conditions). In this way, all the bound ligand should be oriented in a similar fashion, and as long as the antibody of interest does not obscure the binding site of the analyte, this method may lead to better BIAcore results. This coupling method was attempted, as the TfRl was cloned with a 6 histidine tag on the C-terminus (Lebron et al, 1998), thus creating a protein that can be correctly and uniformly oriented on the chip with the use of an anti-his antibody. Unfortunately, the analyte (p97) bound very strongly to the anti-his antibody, thus making this fixation method untenable. problem of orientation is complicated even more with the integrin  a P3, v  The  as the  heterodimer has to be assembled correctly in order for the interaction with the ligand to occur.  In this case, however, the fact that the positive control proteins (Tf and  fibronectin) were able to bind to the immobilized ligands indicates that at least some of the ligand must be oriented in a position that favors binding. There was a rather smaller signal than expected generated for the positive controls, which might indicate that some of the ligand was not oriented correctly on the chip (West et al, 2000). If a greater signal (i.e. more response units) could have been measured for both T f binding to TfRl and fibronectin binding to the integrin cc p\ this would provide more confidence that the v  assay could measure a very small binding affinity interaction. Finally, i f the dissociation rate of the binding interaction is very fast, it may also be difficult to observe binding with the BIAcore technology. It seems clear that under the conditions examined in these BIAcore experiments, there is no strong interaction between p97 and T f R l . The results shown earlier in this chapter point to a binding interaction occurring between these two proteins.  This  apparent discrepancy may be explained by noting that many factors are present in the cell  85  binding studies that are absent in the surface plasmon resonance assay, such as additional proteins, media, and different solute and pH conditions. As well, the T f R l used in the BIAcore experiments is truncated so that the membrane-spanning and intracellular domains are not present. The construct used to produce the soluble T f R l was expressed in a lytic baculovirus/insect cell expression system. The human T f R l gene with the first 120 amino acid residues deleted, was fused 3' to a gene segment that encodes the hydrophobic leader peptide from the baculovirus protein gp67, plus a his tag and factor X a site, and then inserted into a modified pAcGP67a vector (West et al., 2000). In experiments used to examine the effect of point mutations on the binding of T f and HFE to the T f R l , Bjorkman and associates have found that some amino acid residues near the site of truncation of the TfRl are very important for high affinity T f binding (unpublished observations). If p97 and T f do not bind in exactly the same spot, or i f p97 binds with lower affinity than T f (which seems likely), then perhaps the truncation of the T f R l would lower binding of p97 to a level that is not detectable by the BIAcore. It does seem, from all the results in this chapter, that any interaction between p97 and the T f R l is of low affinity. For this interaction to be physiologically relevant it would likely have to take place in specialized circumstances to be useful (because in the bloodstream p97 is unlikely to compete with T f for binding to the TfRl). These specialized circumstances may include places like transport across the blood-brain barrier (discussed in the following chapter). Furthermore, no interaction could be measured between p97 and the integrin a (33 v  by BIAcore. It is not possible as yet to rule out p97 binding to integrin a p3, because v  ligand binding to integrins is a complex phenomenon. For example, the % integrins have  86  been shown to have two classes of ion binding sites within them, one that must have an ion present for the ligand to bind and a second that inhibits ligand binding when the ion is bound (Hu et ah, 1996). The binding of ligand is regulated by the coordination between these two ligand binding pockets. It is also quite possible that p97 may bind to a different integrin via the RGD peptide, as we have only tested the binding to one possible receptor. There have been eight to twelve identified integrins that recognize the R G D motif (Ruoslahti, 1996). The binding studies of p97 to various receptors lead to the conclusion that while p97 may bind to T f R l , another receptor may be important in the physiological role of p97, as the measured interaction between p97 and TfRl is of low affinity. Given the high level of T f in the blood, p97 would have to be acting in a very specialized area for this interaction to have a physiological role in the body.  The examination of p97 and  integrins will have to be continued before this possible mode of action can be ruled out.  87  Chapter 4:  Characterization of the novel receptor transferrin receptor 2,  including interaction with p97  The results of the binding of p97 to the T f R l did not produce clear and convincing evidence that T f R l is the main receptor for the soluble form of p97, so a search was initiated for other likely receptors. The first step in this process was to search databases to find proteins closely related to T f R l . Transferrin receptor homologues were initially targeted in this search for the same reason that TfRl was examined for binding to p97: p97 is very similar in sequence and structure to Tf, and the T f R l binds T f with a very high affinity.  This avenue was pursued further due to the evidence shown in  Chapter 3, that p97 does interact to some degree with T f R l .  Using the expressed  sequence tag (EST) database of the National Center for Biotechnology Information, a homologue of the TfRl was identified first in human and then in mouse tissues (Boguski et al, 1993). The EST database contains, short cDNA segments generated by researchers that are sequenced once and added to the public domain without further sequence conformation. They represent a survey of the cDNA of a tissue or organism, and have been useful in the past for applications including identification of new members of gene families and for discovering orthologues of genes (Boguski, 1995). After analyzing three EST clones of human and mouse each (all from American Type Culture Collection) and determining in each case that they were from the same gene, the longest cDNA of each species was used to pursue a full length clone. Several rounds of screening were carried out on a human liver library with a short probe made from the EST (ATCC # 363316) derived from human infant brain. A longer fragment of  88  the novel c D N A was successfully isolated from the liver library. Before a full-length cDNA could be isolated, Kawabata and associates (Kawabata et al., 1999) independently published the full length human TfR2 protein, so attention was turned instead to the mouse EST of TfR2.  I.  Tissue expression of transferrin receptor 2  A.  Rationale The identification of a new member of the TfR family, and potentially important  protein in iron uptake and transport, has raised many questions in the field of iron transport. Comparing this new receptor to the classical receptor in terms of its expression in adult and fetal tissue is one method of gaining some important information about the new receptor. For example, differential expression of the two related receptors may lead to some insight into the nature of TfR2's physiological role. Often the expression and function of a protein can be quite different when comparing adult tissue with embryonic tissue. For this reason, two Northern blots of adult and fetal organs were examined for TfR2 expression. First, a Northern blot analysis was carried out representing multiple human adult organs. These included heart, brain, placenta, lung, liver, skeletal muscle, kidney and pancreas. As well, a second Northern blot was completed to determine the level of expression in a mouse embryo at different stages of development, beginning with embryo day 7 (E7) and culminating at E l 7 . These various stages represent quite disparate occurrences in the development of the mouse. The hypothesis in these studies is that TfRl and TfR2 have distinct expression patterns, both in the adult and the developing organism.  89  The brain is an organ with a high requirement for iron, and the route by which iron is specifically delivered to the brain is not thoroughly understood.  It has been  proposed that the major route for iron transport to the brain is via Tf, but in a hypotransferrinemic mouse model that possesses less than 1% of the normal serum T f levels, brain iron uptake is quite adequate (Dickinson et ah, 1996; Malecki et ah, 1999b). This indicates that an additional route of iron uptake besides the Tf/TfRl system may be involved. With this in mind, an experiment was designed to determine i f TfR2 is present in cells that compose the blood-brain barrier (BBB). The B B B consists of endothelial cells of the cerebral microvessels with tight junctions. The blood is thus prevented from diffusional contact with the brain, unlike most other areas of the body, and transport across the B B B is very tightly regulated (Abbott et ah, 1999). The endothelia of the brain capillaries is different from most non-brain capillaries in that they have a greatly reduced number of pinocytotic vesicles, no fenestrae, and the mitochondial volume is higher (Brightman, 1992). A l l these factors combine to produce a protective barrier that allows the microenvironment of the brain to be carefully regulated. In addition, p97 is able to traverse the B B B to a much greater degree than Tf, and thus it is possible that p97 may function in shuttling iron into or out of the brain. It is very likely that in order to facilitate the amount and rapidity of transport across the B B B observed with p97, a specific receptor is involved in this movement. Kawabata and associates have previously reported that TfR2 is not expressed in brain tissue (Kawabata et ah, 1999). However, the brain capillary endothelial cells of the brain microvasculature make up such a small proportion of total brain material, a Northern blot or reverse transcription of the m R N A followed by polymerase chain  90  reaction (RT-PCR) (carried out to assess whether TfR2 is present in brain) may not necessarily be sensitive enough to detect any transcripts if the template material is drawn from whole brain homogenates.  To examine for the presence of TfR2 in the B B B ,  cultured primary human brain capillary endothelial cells were harvested for their R N A in these experiments, and then the RT-PCR was performed.  B.  Results I: Northern blot of multiple human organs Figure 4.1 shows a multiple organ Northern blot probed with a P-labeled c D N A  probe of TfR2. The Northern blot was purchased from Clontech Laboratories, and was loaded with approximately 2 ug of poly A+ R N A per lane from eight different tissues: heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas. The probe was created by P C R with EST 363316 (ATCC) used as the template. The primers used are listed in Table 2.3. The hybridization was carried out as suggested by Clontech, with a one hour incubation of the probe in ExpressHyb solution followed by several washes, and development on the phosphoimager. Figure 4.1 shows that the only band that can be visualized on the blot is in the liver at about 2.7 kb. The signal from the liver band is quite strong, and no other detectable signal was found in any other tissue lane, or at any other size on the blot. Following the hybridization with the probe of human TfR2, the blot was stripped and reprobed with a P-actin probe as a loading control. A l l lanes showed approximately equal P-actin signal (data not shown).  91  C/)  CD  1=L  CD  S>  CD 2 .  00 ^< CD  §  2.  S>. 0) CD JD 13 3 D  hTfR2  Figure 4.1: Human multiple tissue Northern blot of transferrin receptor 2. Northern blot with RNA from pancreas, kidney, skeletal muscle, liver, lung, placenta, brain and heart, hybridized with a P labeled human T f R 2 cDNA probe. 3 2  92  C.  Results II: Mouse embryo Northern The mouse embryo Northern blot represents the poly A+ R N A from four different  mouse embryos at days 7, 11, 15, and 17. It was commercially prepared with 2 \ig of R N A per lane on a denaturing formaldehyde gel, transferred to a nylon membrane and fixed by ultraviolet radiation. The blot was first probed with a mouse TfR2 c D N A short probe according to the Clontech protocol. The probe was approximately 250 base pairs long, generated by P C R from the cloned full-length mouse TfR2, gel purified and labeled with P . After the initial hybridization, the blot was stripped and re-hybridized with a 32  cDNA probe against the T f R l (approximately 640 base pairs, also generated by P C R and gel purification). Finally, it was stripped and re-hybridized with a c D N A probe from J3actin, to check that the loading of R N A on the blot was comparable between the various embryonic ages. The specific activity of all three probes was very similar. In Figure 4.2a, the mouse T f R l signal is clearly visible even in the youngest embryo at day E7. The signal increases through day E l 1 until it peaks at day E l 5 . A t day E l 7 , the signal has markedly decreased. In contrast, in Figure 4.2b the signal for mouse TfR2 is only visible beginning with E l 5 , and has increased by day E l 7 . Figure 4.2c shows that all four lanes on the Northern blot have comparable levels of P-actin.  D.  Results III: Transferrin receptor 2 on human blood-brain barrier endothelial cells To determine i f the transcript of the alpha, or transmembrane form, of TfR2 is  found in human brain microvascular cells, primary human brain microvascular cells from two individuals were cultured, and the R N A was harvested. The R N A was then used in RT-PCR to amplify a transcript specific to the alpha form of the protein. Figure 4.3a  93  a)  TT-  10  |S.  h"*  1—  1—  LU  111  HI  LU  4.6 k b - * -  b)  mTfR1  UT)  T—  I*-  T—  UJ  LU  s  T—  T—  LU  UJ  3.0 kb  mTfR2  c) LU  2 . 0 kb 1.6 kb  ""  UD  h->  LU  LU  LU  T  • • |»  •  •  •  p—sctin  Figure 4.2: Mouse embryo Northern blot of transferrin receptor 1,  transferrin receptor 2, and p-actin. The mouse embryo Northern blot contains R N A from four mouse embryo stages: day 7 , 1 1 , 1 5 and 17. a) shows the embryo blot hybridized with a P - l a b e l e d probe against TfR1 b) embryo blot hybridized with a P-labeled probe against TfR2. c) embryo blot hybridized with a P-labeled probe against p-actin. 32  32  32  94  a)  *Exon 1 TfR2  1  1 1  *Exon 2  *Exon 3  Exon 4  Exon 5  1 I Primer F  Primer A  b)  Figure 4.3: Transferrin receptor 2 in brain endothelial cells, a) Exon structure of TfR2 (Kawabata ef. al., 1999). Exons 1,2, and 3 are found only in the a-form of TfR2, while exon 4 is only found in the p-form. Exon 5 is common to both forms. Primer F is complementary to sequence within exon 3 and primer A is complementary to sequence in exon 5. b) Agarose gel of PCR product from K562 cells and human brain endothelial cells. PCR product generated with primers F and A.  95  shows the primers used for the amplification, and Figure 4.3b shows that both K562 cells (human erythroid leukemia) and the primary brain microvascular endothelial cells produce a band of approximately 200 bp, which is the predicted result. After the P C R reactions, the band was cut from the agarose gel, purified, and sequenced to confirm that the product was TfR2 a. This procedure was carried out on two separate samples of R N A , isolated from two sources of human brain microvascular cells, and confirmed by sequencing both times. As a negative control, the TfR2 and actin P C R was also carried out on c D N A from human heart tissue, which has previously been determined to be negative for TfR2a expression by RT-PCR (Kawabata et al., 1999).  E.  Discussion The expression of the TfR2 in adult tissue is markedly different from that of the  T f R l . In Figure 4.1, TfR2 can only be detected in the liver, while no detectable signal is observed for the pancreas, kidney, skeletal muscle, lung, placenta, brain or heart. This data agrees with that of Kawabata and associates, who subsequently published a Northern blot showing that the TfR2 a form can only be strongly detected in the liver, While a very weak signal was also detectable in the stomach, and that the band is of comparable size (Kawabata et al., 1999). The expression of human TfR2 is similar to that observed for mouse TfR2 (Fleming et al., 2000). A RT-PCR experiment also showed expression of the TfR2 a transcript in the liver derived cell line HepG2, and the erythroid leukemia cell line K562 (data not shown). This result was also confirmed by Kawabata (Kawabata et al., 1999). The high degree of liver expression of TfR2 suggests that the primary role of TfR2 may be within this organ. Hepatocytes have a high requirement for iron as they  96  have a high rate of metabolic activity, and iron is essential as a cofactor for cytochromes a, b, and c, cytochrome oxidase and the iron sulfur complexes of the oxidative chain (all involved in adenosine triphosphatase, or ATP, production) (Connor et al., 2001). This is likely one reason why TfR2 remains highly expressed in the liver throughout the life of the organism. Transferrin receptor 1 expression is very low in the adult liver, both under normal physiological conditions, and in the case of iron overload (Fleming et al., 2000). Thus, the iron loading of the liver that takes place early on in iron overload diseases such as hereditary hemochromatosis (HH) may be due to TfR2 on hepatocytes (Fleming et al., 2000). Unlike T f R l , TfR2 expression is not regulated by the amount of iron in the cell through the iron response element system mentioned earlier (Kawabata et al., 1999). This lack of regulation by cellular iron conditions may be advantageous under normal physiological conditions, as the liver is an organ with a constitutively high requirement for iron. The difference between expression of TfRl and TfR2 during embryogenesis is very distinct. Mouse TfRl is clearly visible by Northern blot quite early in development, by day E7 (Figure 4.2). At this stage, the head process is forming, along with the foregut pocket and the first somites, but most organs are not distinguishable (Rugh, 1990). Transferrin mRNA is detectable in whole embryos by E6 (Plowman, 1986). It is initially produced in the visceral yolk sac, by endodermal cells (Meehan et al., 1984), and the production of T f mRNA is several fold higher in these cells than in the adult liver. Transferrin receptor 2 transcript is not detectable at this stage (Figure 4.2b). The fetal liver develops from the gut ectoderm at E9. Transferrin mRNA can be detected in fetal  97  liver as early as E l 1 (Ekblom and Thesleff, 1985). By E l 1, the T f R l is expressed to a greater degree than at day E7 (Figure 4.2), but TfR2 is not yet discernable. At this point in development, organogenesis is occurring, but the organs are not completely formed. By E l 5 , which is the first day-TfR2 a transcript is detectable, organogenesis is complete, however rapid cell growth continues to take place. At this point, the liver is beginning to carry out some of its major functions and thus has a constitutive need for iron (discussed earlier). This situation continues through day E l 7 and beyond. The Northern blot shows that expression of TfRl peaks around day E l 5, and is expressed to a lesser degree after that. This may be due to the high demand of the body for iron at early stages to support extremely high rates of cell division and growth. As well, during embryogenesis, the liver accounts for approximately 5% of the body weight and is an extremely active hematopoeitic organ. The rise and fall of TfRl levels in the embryo roughly correspond to the proliferation of hematopoeitic stem cells in the liver (McNagny, 1990). Even after embryogenesis, the cells that continue to need high levels of iron, such as reticuloendothelial cells, continue to express high levels of T f R l . The examination of TfR2 expression on cells of the B B B was undertaken to try to understand the relationship between p97 and brain iron transport. The microvessels within the brain consist of specialized endothelial cells that have tight junctions. These cells make up the barrier between the brain and the blood and maintain control over the ionic microenvironment of the brain for important processes such as synaptic processing (Abbott et al., 1999). Therefore, specific transport systems exist to allow selective movement of specific compounds such as macromolecules and peptides that the brain requires for proper functioning.  98  The brain is an organ with high energy requirements, and thus a high demand for iron. While the brain weighs only 2% of total adult weight, it accounts for 25% of the body's oxygen consumption (Thompson and Thompson, 2001). As mentioned before, iron is essential for the production of ATP, and the brain requires A T P for such tasks as maintaining membrane ionic gradients, synthesizing neurotransmitters, and generating lipids and cholesterol for the synthesis of myelin (Beal, 1998; Connor and Menzies, 1996). The transport of iron into the brain is not clearly understood.  Studies have  demonstrated that serum T f does not cross the B B B (Crowe and Morgan, 1992). Instead the brain produces its own Tf, which is located in the interstitial fluid of the brain and CSF at levels about 100 times lower than the amount normally found in serum (Rouault, 2001). Using a mouse model in which oligodendrocytes fail to thrive, it was shown that oligodendrocytes in the brain produce the bulk of the T f transcripts (Bartlett et ah, 1991). The current theory of iron transport across the B B B is that iron en route to the brain is internalized via the regular Tf/TfRl pathway, released within the endosomes, and transported across the cell to the brain parenchyma, where it is picked up by T f in the CSF and transported to where it is needed within the CNS. One unexplained aspect of this system is that, while there is a molar excess of Tf in the blood stream to prevent free iron from causing cell damage through free radical formation, there is a molar excess of iron relative to the levels of T f in the CSF (Bradbury, 1997; Moos and Morgan, 1998). Whether this iron is bound to proteins such as p97 is unknown. As well, this explanation of iron transport into the brain does not explain how iron can be successfully delivered to the brain in hypotransferrinemic mice (mice with extremely low serum T f levels) (Dickinson et ah, 1996). Transferrin is required for correct distribution of iron within the  99  brain (Malecki et al., 1999a), as the hypotransferrinemic mice display abnormal iron accumulation in the choroid plexus. Clearly, Tf plays an important role in this system, but pieces of the puzzle to make the picture complete are still missing. In vivo studies involving injections of p97 and T f into mice have shown that a much greater proportion of p97 is transported into the brain compared to Tf, after intravenous injection of the proteins (Moroo et al, 2002). Figure 4.4 shows that the proportion of p97 transported into the brain is 7 times that of T f (Moroo et al., 2002). The proteins, labeled with radioactive  125  I were examined in the brain parenchyma  fraction of brain homogenate one hour after injection, followed by perfusion and capillary depletion of the brain to remove the microvessels (Triguero et al, 1990). While 125  I - T f is not detectable in the parenchyma after this treatment, p97 can be recovered and  appears to be intact, as shown by Western blot (Moroo et al., 2002). Transport across the B B B has also been demonstrated through injection of fluorescently labeled proteins (p97, Tf, and B S A labeled with Alexa fluor 488) into the tail vein of a mouse. While all three proteins can be found in the brain microvessels very clearly, only p97 can be visualized in the brain parenchyma (see Figure 4.5) (Moroo et al., 2002). The fact that p97 is transported across the B B B into the brain suggests that a specific receptor exists on the capillary endothelia to facilitate this transport. Previous publications have shown by RT-PCR and Northern blot that TfR2 is not expressed in the brain (Kawabata et al, 1999). The approach used for these experiments, however, was not sensitive enough to identify expression in certain lower frequency subsets of cells within the brain, as some of these specialized cells make up a very small proportion of the total cell mass of the brain. One of these specialized sub-types of cells  100  Figure  4.4:  Uptake of  l-p97 and  125  125  l - T f in vivo, a) The level of  125  l-p97  reaches a peak 3 hours after injection and is cleared from the brain within 5 hours. The maximum value of l - p 9 7 is approximately two fold higher than that of l-Tf. b)Ratio of brain dpm: serum dpm for both l - p 9 7 and l-Tf. (Moroo et al. 2002) 125  125  125  125  101  Figure 4.5: Visualization of p97 and T f uptake by the brain. Intraperitoneally injected (a) Alexa 488-p97 and (b) Alexa 488-Tf appear in the brain after one hr. Although p97 and T f can be seen in the microvessels of respective mice (solid arrows), p97 appears to transcytose the B B B more efficiently than T f and exhibits a punctate distribution in the cytoplasm of cerebral cortical cells (open arrows). Scale bar represents 10 pm. (c) After DIG-labeled p97 was injected into a mouse, the brain was harvested, sectioned, and the p97 localized with colloidal gold conjugated anti-DIG antibody and visualized by gold enhancement. Although parenchymal structures are weakly fixed, this electron micrograph shows that DIG conjugated p97 crosses the intact B B B and can be seen in the brain parenchyma. Scale bar represents 1 pm. (Moroo et. al, 2002).  102  103  is the endothelial cells of the brain microvascular system. When the R N A from these cells grown in culture was harvested and used in a RT-PCR reaction, the TfR2 a signal is clearly present (see Figure 4.3). Transferrin receptor 2 has 2 forms, due to alternate splicing events. One form is called the a form and includes exons 1, 2 and 3, and has both the full transmembrane domain and signal sequence, and is expressed on the surface of cells (Kawabata et al, 1999). The second, or (3 form of TfR2 lacks exons 1, 2, and 3, which code for the transmembrane and intracellular portions of the protein, including the signal sequence, and is likely to be located intracellularly. This form is detected in small amounts in nearly all tissues and cell lines examined through RT-PCR (Kawabata et al., 1999). To ensure that the form of the transcript present in the microvascular cells of the brain is the a  form, primers were designed such that the forward primer is  complementary to a sequence within exon 3, which is found in the transcript of the TfR2 a form, but not (3 form (Kawabata et al., 1999) (Figure 4.3a). The role of TfR2 in the cells of the B B B is not known.  It has not been  demonstrated i f this protein is located luminally or abluminally, or i f it is capable of undergoing transcytosis.  Future experiments must be undertaken to examine the  localization of TfR2 in various polarized cells. If the protein is expressed specifically on the apical surface of the polarized cells, this would suggest that it may be expressed on the luminal surface, or the surface facing the blood, in cerebral microvessels. This must also be confirmed with direct antibody staining once a good antibody is developed. It has been suggested that the high serum levels of p97 observed with Alzheimer disease may be due to an increased production of p97 in the brain, and that this p97 may be able to  104  cross the B B B from the brain to the serum via a receptor. More work will have to be done before one can conclude that TfR2 is involved in this process.  II.  Subcellular localization  A.  Rationale To further characterize TfR2, and as an initial step toward eventual transcytosis  experiments, it is important to examine where the receptor is located in the cell and what vesicles it travels through along the internalization pathway. The endocytosis pathway of TfRl is very well studied, and has long been cited as a classical model of receptor mediated endocytosis.  Endocytosis of the T f R l has been shown to require the  cytoplasmic domain of the receptor, specifically the Y T R F motif (which conforms to the classical Y X X O endocytosis signal that consists of an aromatic amino acid, often tyrosine, followed by two large hydrophobic residues) (Collawn et al, 1990; Iacopetta et al, 1988; Jing et al, 1990; Rothenberger et al, 1987a). The hypothesized endocytosis motif is also present in TfR2, with Y Q R V (tyrosine, glutamine, arginine, and valine) closely resembling the classical endocytosis motif (Kawabata et al,  1999).  To  investigate the endocytosis of this protein, confocal microscopy was used to examine TfR2 expressing cells that were immunofluorescently labeled with markers of the endocytic vesicles. The hypothesis to be tested is that TfR2 follows a similar endocytosis pathway to T f R l , and that it is localized to the early endosome during receptor mediated endocytosis. Another question that may increase the understanding of the function of TfR2 is whether TfR2 might function in tandem with T f R l . Transferrin receptor 1 exists at the  105  cell surface as a homodimer, with two disulphide bonds forming in the extracellular portion of the molecule to hold together the two monomers. TfR2 is also present as a homodimer (Kawabata et al, 1999), and has cysteines in a similar position to those found in T f R l .  The possibility that the two receptors can exist as heterodimers on cells that  express both receptors is addressed in this section. The hypothesis is that TfRl andTfR2 form heterodimers on the surface of cells.  B.  Results I: Confocal immunofluorescence Two sets of immunofluorescence experiments were carried out to assess which  subcellular compartments TfR2 is found in. In the first experiment, shown in Figure 4.6a) the SK-Mel-28 human melanoma cells were transiently transfected with a myctagged mouse TfR2 construct. These cells were then double stained with antibody against the myc tag (9E10) and antibody against the early endosomal antigen (EEA1), which recognizes a membrane bound protein that is both specific to the early endosome and essential for fusion between early endocytic vesicles. The cells were grown on coverslips, washed and blocked with 1% B S A in PBS, and permeablized with saponin. This permeablization step allows the antibody to enter the cell, as both antigens are intracellular. In this experiment, the EEA1 antigen is shown in green (visualized with secondary antibody rabbit anti-goat Alexa 488), and is depicted alone in the first frame of the figure. Antibody against c-myc (the marker on the mTfR2) is shown in red (with secondary antibody rabbit anti-mouse Alexa 568), alone in the second frame. The  106  Figure 4.6: Subcellular localization of transferrin receptor 2 in transfected human melanoma cells, SK-mel-28. a) EEA1 is shown in green, in the first panel, visualized an anti-EEAl antibody and with rabbit anti-goat Alexa 488 secondary antibody. Mouse TfR2 is stained with anti-myc antibody 9E10 in the center panel, and visualized with rabbit anti-mouse Alexa 568. The merged image is shown on the right, b) Anti-clathrin antibody staining the transfected human melanoma cell line is shown in green on the left, visualized with goat anti-rabbit Alexa 488. Mouse TfR2 is stained with 9E10, and visualized with the secondary goat anti-mouse Alexa 568. The right frame shows the merged image. In both a) and b) some yellow punctate staining indicating colocalization is indicated with large arrows, while the small arrows indicate no colocalization. Scale bar=5um  107  CO  108  109  merged image is seen in the third frame. The areas where the EEA1 and TfR2 colocalize are visualized as yellow, and indicated with the larger arrows. Another marker of endosomes is the clathrin heavy chain.-Clathrin heavy chain is found in vesicles that have originated from both the plasma membrane and trans-golgi network (Robinson and Pearse, 1986). In receptor mediated endocytosis, the receptors are associated with clathrin coated vesicles (Mellman, 1996). In this double labeling experiment, antibodies against the myc tag and those against the clathrin heavy chain were used. In figure 4.6b), the left frame shows the antibody against the clathrin heavy chain (in green, detected with goat anti-rabbit Alexa fluor 488). The middle frame shows anti-c-myc, which recognizes the myc tag on the TfR2, and is visualized in red (with goat anti-mouse Alexa 568). The right frame is the merged image, with yellow representing the areas where the clathrin heavy chain and the TfR2 are colocalized. In both a) and b), areas of colocalization are shown with large arrows, whereas areas that show single staining are indicated for comparison with small arrows.  C.  Results II: Western blot of heterodimers of transferrin receptors 1 and 2 To determine i f TfRl and 2 form heterodimers on the surface of the cells, cells  without transferrin receptors, cells with TfRl or TfR2, and cells with both T f R l and TfR2 were examined. A FACs profile showing the level of expression of both proteins on the four cell lines is shown in Figure 4.7a and b. The cells were harvested with versene, counted and aliquoted to 2 million cells per tube. The cells were then incubated with anti-TfRl antibody, normal rabbit serum, or no first antibody on ice, and washed to remove any unbound antibody. Next, the cells were solublized and immunoprecipitated  110  a)  b)  at* Legend: anti-TfR2 stain •  ...  Legend: anti-TfRl stain  TfR- cells  •  TfR- ceils  TfR 1+cells  —  T f R U cells  TfR2+ cells TfR1,2+cells  ...  TIR2+ cells TfR1,2+cells  Figure 4.7: Western blot of heterodimers of transferrin receptors 1 and 2. a) FACS profile showing expression of TfR2 on TfR-, TfR1+, TfR2+, and TfR1,2+ cells, b) FACS profile showing expression of TfR1 on TfR-, TfR1+, TfR2+, and TfR1,2+ cells, c) Western blot showing TfR-, TfR1+, TfR2+, and TfR1,2+ cells immunoprecipitated with anti-TfR1 antibody and blotted with anti-FLAG antibody to detect TfR2.  Ill  with protein G sepharose beads. Finally, the lysates were run on an 8% SDS-PAGE gel and transferred to nitrocellulose. The lysates were not reduced with D T T , so the disulphide bonds between receptors were not broken. The membrane was then blotted with an antibody against the FL AG-tag of hTfR2. Figure 4.7c shows that a band at 210 kDa is present only in the lane loaded with lysate from cells that express both TfRl and TfR2.  D.  Discussion Confocal immunofluorescence microscopy is a sensitive way to examine  subcellular localization of proteins. With this technique, optical slices allow one to distinguish vesicles within the cell, and double labeling can determine i f two proteins are co-localized within the same vesicle. The TfR2, shown in Figure 4.6a), co-localizes with a marker of early endosomes, E E A 1 , as well as with clathrin (Figure 4.6b), another marker of endosomes. This is shown by the punctate yellow staining pattern in the cells, indicated by a large arrow. This result supports the hypothesis that TfR2 follows the same endocytic pathway as TfRl (Tiong, 2001). This is logical, as the endocytic signal present in the cytoplasmic portion of TfR2 is similar to that of T f R l .  The endocytic  recycling process of TfRl is a very well studied phenomenon. The binding of ligand to receptor at the cell surface stimulates an accumulation of ligand-receptor complexes at clathrin-coated pits, which eventually bud off of the cell surface and become clathrincoated vesicles. These vesicles lose their clathrin coats and fuse with early endosomes. The pH of endosomes is approximately 6-6.8, and this slightly acidic nature promotes the dissociation of many ligands from their complexes (Forgac, 1992; Robinson and Pearse,  112  1986). The receptors accumulate in the tubular extension regions of the early endosomes after the removal of ligand, and these extensions bud off to become recycling vesicles that transport the receptor back to the cell surface (Mellman, 1996). The ligands that remain accumulate within the vesicles and these early endosomes then fuse with late endosomes or lysosomes. In the case of the T f R l , T f does not dissociate at the pH of the endosome, but iron which is bound to T f does dissociate. The apo-Tf at this pH remains bound to TfRl and is recycled to the cell surface. With TfR2, Kawabata and associates have demonstrated that F e can accumulate in the TfR2+ cells after incubation with Tf- Fe (Kawabata et 55  55  al., 1999). The confocal results support the hypothesis that TfR2 follows a similar pathway to that of T f R l , but further research must be completed to confirm this. One experiment that is underway involves the mutation of the endocytic signal of TfR2 from 23YRRV26 to A R R V .  In T f R l , this tyrosine to arginine mutation leads to an 80%  decrease in internalization of the TfRl (Collawn et al., 1990). It will be interesting to see if TfR2 is affected in a similar way. The Western blot experiment, shown in Figure 4.7 demonstrates that TfRl and TfR2 form heterodimers on the cell surface. The flow cytometry data in 4.7a shows that only the TfR2+ and T f R l , 2+ cell lines express TfR2. The high background fluorescence is likely due to the fact that the TfR2 F L A G marker is an intracellular antigen. In Figure 4.7b it is clear that only the TfRl+ and TfRl,2+ cells stain positively for presence of TfRl.  The immunoprecipitation experiment was performed by doing an identical  immunoprecipitation on the four cells lines: TfR-, TfRl+, TfR2+, and T f R l ,2+. Since the antibody for the immunoprecipitation was added to intact cells, the proteins being  113  immunoprecipitated are more likely to represent functional surface expressed receptors as opposed to those retained internally, such as in the endoplasmic reticulum. The Western blot was stained with an anti-FLAG antibody, which recognizes the F L A G tag of TfR2. The fact that no FLAG-tagged protein is seen in the third lane (that of TfR2+ cells) but is clearly present in the fourth lane (TfRl,2+ cells) indicates that the protein in lane 4 is present because it was immunoprecipitated in conjunction with T f R l , which was specifically immunoprecipitated by the anti-TfRl antibody. The size of the protein on the Western blot closely corresponds with the known size of the TfRl homodimer (-210 kDa), and once the gel was stripped and the blot detected with an anti-TfRl antibody, the bands correspond in size (data not shown). The presence of T f R l and TfR2 as a heterodimeric complex leads to some interesting new questions about TfR2 function. If the two receptors function as one unit in some cells in which they are both expressed, such as erythroid cells (Kawabata et al, 2000) and cells of the B B B (Jefferies etal, 1984), perhaps the presence of the TfRl confers some iron-responsiveness to the function of the heterodimer. Transferrin receptor 2 does not possess an iron response element (Kawabata et al, 1999), and does not exhibit any modulation of expression due to iron levels (Fleming et al, 2000). Transferrin receptor 1, on the other hand, is tightly regulated by the iron levels of the body, as discussed earlier. This iron-responsiveness is shared by many of the proteins involved in the iron uptake, transport and storage pathway, thus would seem very important for tight regulation of this system.  When T f R l expression is down-regulated due to high  intracellular levels of iron, this may down-regulate the heterodimer, and curtail any function that it may have had. Some level of iron-responsiveness conferred on the  114  function of TfR2 may be advantageous in certain situations. Further experiments to determine i f this heterodimerization occurs in vivo, rather than in transfected cell lines, will be important to carry out, as well as experiments designed to discover i f this phenomenon has a specific function, and i f it is actually iron-sensitive.  III.  Cloning, constructs and transfections of mouse transferrin receptor 2  A.  Rationale In order to study this new protein, the full sequence of the c D N A had to be  determined, followed by insertion into a cloning vector for manipulation. For ease of identification in the absence of a good antibody against the protein, the mouse TfR2 was expressed with a myc-tag on the C-terminal end of the protein and was transfected into cells. The myc tag represents a portion of the human c-myc oncogene product, and includes the epitope 410-419 of human c-myc recognized by the anti-c-myc monoclonal antibody, 9E10 (Campbell et al, 1992).  B.  Results I: Sequence of mouse transferrin receptor 2 The mouse EST, A T C C # 1978344 was originally derived from a 14 day mus  musculus embryo. Through sequence analysis and comparison to the published human TfR2 sequence, it was confirmed that the EST was from mouse TfR2, and that the clone represented the full-length TfR2 cDNA. Figure 4.8 shows the full length TfR2 c D N A sequence derived from "primer walking", where additional primers are designed from the sequence already obtained to generate additional sequence until the entire c D N A  115  Figure 4.8: Sequence alignment of the transferrin receptors. Aligned sequences of mouse TfR2, mouse TfRl (Stearne et al, 1985), human TfR2 (Kawabata et al, 1999), and human TfRl (Schneider et al, 1984). Shaded areas represent homologous sequence, while the boxed areas represent identical sequence.  116  PORWGLLRRVQQWS rfDOlA R S [ A ? ] S [ N ] L F | G ) G fERLWGLFQRAQQL S dlDQlA R S ( A ? ] S 1 ] L F ( G ] G  mTfR2 mTfR1 hT(R2 hTfR1 mTfR2 mTfRI hTfR2 hTfR1 mTfR2 mTfRI hTfR2 hTfR1  E  G  P E P E  RPjSQT P[TS - RjSQT p[f S  A O F [ C 1 P  A [ E ! L P  v  68 70 70 69  rnTfR2100 mTfR1102 hTfR2 102 hTfR1 105 mTfR2 rnTfRI hTfR2 hTfR1  132 126 138 137  mTfR2 mTfRI hTfR2 hTfR1  161 158 169 168  mTfR2lS 7 mTfRI 194 hTfR2 202 hT(R1 192 )  rnTfR2 233 mTfRI 227 hTfR2 238 hTfR1 225  D  E  LG  67  Y T R F  M I J L  E E E M A I D I H I M M K A E J E G A  Y0RV - - - EGPRKGH Y T R l F S L A R Q vlrJMDlH  33 33 33 33  pvlRfRlv - - - E G P O L E H  S  FC  i  G  L A R Q V P G D I N  E H  SCPGRS  V R K  E ( T T A H F [ C ] P M ( D L  E I E N A N A I D] I N I N  S  V  T  K  A  N  I  T  L  L[j]F  H  C  69  L | C | F A E  T  K d J c T G T T l c l - Y 68  K|PJ  P  69  R O  R Q P  -  -  -  -  -  -  A  G  T  99 101 101 104  V A F R G l s l C Q A C G D S V L V VlD[EDTt^p[E"r51SGRTT - - - - f L Y A ET EE TDK - - - S E T M E T E[D]V P T[S]S R|L Y V A F R G l S l C Q A C G D S V L V VlSlE D W Y E | p [ D L D l F H O O R L Y W E S P V RE - - E P G E D F P A A R R L YfwlD YP D Liic R K L [ S 1 E KFL|D  131 125 137 136  | W S D L Q A M F L l R l F L G E G R M E D T I R | L T S - - - ^ - - - L R E|R L S Q N T Y T P LlK T L L S E KlLlN S I E F A I D T I | K Q R S D L Q A M F L[Q[F L G E G R | L | E D T I R| - Q T|SJL R E R V A G STjpjF T S T I K L L N E N S Y V P R E A G S Q K D E N L A L  160 157 168  167  V A G S A R M A T V [ Q | D [ T 1 L D K L [ S ! R Q K LLJ .J D H V W 1 T D T H Y V G E A G S Q K D E S L A Y Y I E N Q l F H E F|K|F S K v w R D [ E ] H Y V K SAG - - -IMAIA LJT|Q]D R A A L [ S | R Q K L | D H V W T D T H Y V O YV E N Q I F R E F | K LIS K|VWIRIDIQIHIFIVIK  196 193 201 191  i A  P[I~P1WA[A1A G I  A  L  V  I  -  F  R K  LffJG  F  L[TJG  T P ] W A | A ] A G IJA  V  I  V  F  A  F  A  P[Y]L F  M  S  G  V  R R A A F  K  R  T | | ] A [ F ] L  V  E  O  K  E  E  - - C  V  K  L  L L H I F T [ G 1 A [ F ! L L I G Y I M T G  Y ( L G  Y\C  K  G  V  E  P  K  T  E  C  E  R  L  -  rw{A|D  F  P|5)P  V K S F  S  L | ] W J VDIADIGJS I  G I Q N  V T  v|o  P [ D ] P ( A H PJN"T)L|UW| V D  V K I D J S I A J I  Q N I S V  i  i  S | N G N  I  E A AJ |OG K  V D J K : I | NN GO J R  TlGNlAITG K L V Y A H Y G R P T E VTSJGK^L V|H|A|N F | G | T i^¥v T G [ E J L v Y A H YG R A  A T I V  T  G  K  L  VIHIAIN  s  V Q E I Q L P L | E P _ P L D  -  V G L  -  E D P I D  A  232  226  VV | Y C P | Y S A  L  V[EJN1PJG  E | D ] L Q | D T | K  A  K  G  V  E  L  s  -  -  -  Y  S  -JV]N[G]S  P  L Q[D  E  D  FIG T H K I D J F  EIFJL L  R A  E | D L J -  -  G Y  A  G  V  A  S  R  ^  S  237 224  K  L [ L T V  R G  V D[P  V G  -  Y T I P  VINIGIS  -  O  I|QJ  YVAFjSK  Y  F  V  E | G  „  L  E I V I Y C P I Y S  E | S J P  E I Q L P -  | ] K D  117  -  L Q I  L Q V  L | L L I  V  V I  268 257 273 256  rnTfR2269 mTIR1 258 h T f R 2 274 h T f R I 257 mTfR2303 mTfR1291 hTfR2 308 h T f R I 239 rnTfR2339 mTIRI 324 h T f R 2 343 h T f R I 322 mTfR2375 mTfR1 356 hTfR2 353 h T f R I 357 mTfR2409 mTfR1392 h T f R 2 387 h T f R I 390 mTfR2444 m  T f R 1 428  hTfR2 422 h T f R I 425 mTfR2478 rnTfRI 463 h T f R 2 456 h T f R I 460  [R V - - G IJTlS FA]Q|KvA[VJA!^D[F]G[A1Q|GVL I Y|P[D~P]SD[F S Q D Tjv R A[GJE I TJF A E K V A N A Q S F N A I G V L I Y M M N K F [RV|G V - - I[TF QJD|FJG A r p E F|A ] SQD V R A G K[T|T^1F A E K V A N AJEE S LN|A G V L I Y M D I Q T K  302 290 307 288  P H K | P G L(S]S H Q A j Y | G H|V|H L G T G D P Y T P G F P S F N Q T Q F VVEADLA H|A H L G T G D P Y T P G F P S F N H T Q F P P K P s L[S]SQQ VHLGTGDPYTPGFPSFNQTQ[T I V N A E L S F F[GH|AH L G T G D P Y T P G F P S F N!H|T Q F  338  S SGLPlsffPlAlQlPrfslAD I|A|D QfLlL R K L T G P f y l A F O E S S O L P N I P V Q T I S R A A A E K L F G - KME\O S C IK L(K]G - - P|V]A|P P P SlRlS S G L P N I P V Q T I S R A A A E K L F Q|N - RJHo D CypTsD  374  IWKGIH[L1S1G S P YIRIDOIP G PIDHIRL v V|N)N[H]R[V1S T P[JJS - N I P A RWN I DlsfSGK L E L S O N Q N V K L I V K N V L K E R R I L N I —i r~™i *.*>.«. • t i r""*i i * r~ i W|Q|O1S|L1LIO sTylHlLblP o PIRILIRL V V[NJN|H]RT ST P N W I WKlTDSTCRMVTSESK - - - N V K L T V S N 0 L K E K I LN I  408  1  1  1  m  ,—  r  [F|A c FG]V FGC FG V  n  E|G~F]A E P D H Y V V G A Q R D A W G P G JA A K S A V G T Aj K G [ Y E E P D d Y V V V G A Q R D A[E)G(A]G V A A K s H J V G T g EG R S E P D H Y V V G A Q R D A W G P G I A A K S A V G T A KIG TIVIE P D H Y v v V I G A Q R D A W G P G A A K SiGlVGTA  L L E VRTF L | L L | K | L | A Q VjF L L E L V R[f F LIL L l K I L l A Q  S SMVS S DM i l s S SMV• SIDWVL  N G F R P R|R S L L F I-IDGFRP RS I I FA S N - G F R P R|RS L L F - [DD GFIQIP R S I I F A  |S V G A T E W L E G Y L S V L H L K A A]V G A T E W L E G Y L S L H L K A S VGpDTEWLEGYLS V L H L K A S VGATEWLEGYL S LHLKA  V V Y V SL D N S V L G P G K F H F T Y I N L D K V V L G T[S]N F K V VYV S L D N A V L G D D K F F T Y I NILDIKIA V L G T[S]N  PN p|pN V K HP  mTfR2514 rnTfRI 499 hTfR2 492 h T f R I 496  118  S W DJG G D F G SWfjfAGDFG SWDJGGDFG SWfs A G D F G  H[S G Q T L Y EJQ V A L@ H vD]G[KS1LY[RDS|NW| I HIS G Q X L Y EJQ V VJF JJN VTIG - A  323 342 321  355 352 356  391 386 389 443 427 421 424 477 462 455 459 513 498 491 495 549 532 527 529  mTfR2550 rnTfRI 533 hTfR2 528 hTfRI 530  S|WDAfiE|V I Q P L P M D S S A Y S F T A F A J G V P A V | E | F | S | F M E D C E D LISFIDNAAYPFLAYfTG IP A V KV - EK F M E D L P M D S S A J j S F T A F " V J G V P A V SWDAEVIRP [LJTLIPJN A ^ l F YLtS i l P A V | S | F | C l CF |E D EJK SIK V  ?Mm  HKj D R G I T I P I A J / J V Q @ V [ A | Q L ( A R T A A EVA Y L G T R j j D T Y E 0 L L L Q K VlP  mT(R2586  . ,L H T E[ED T Y E N L  mTfRl564 hT(R2 564 hTfRI 561  LHTK[EDTYENLHJ1V[L YlLG T T M D T Y I K ^ L J I E R I  G I R L P O A Q A V A Q L A  EILJN KIVJA RIAJAIAJE V | A  mT(R2622 rnTfRI598 hTfR2 600 hTfRI 595  j G Q L L I R|DS|HD"1H L[L1P L D | F G [ p | G D 0 V [ L ] R H Q j ^ N J L l ^ E r F  mTfR2658  blLilR[Gl^TfLQWlv|Y S A RGD Y|I|RA|A E K ^ R K , E L Q W L Y S A R G D Y F R A T S R L T[TJD DIRDMOL  rnTfRI 634 hTfR2 636 hTfRI 631 IT.T1R2694 mTfRl670 hTfR2 672 hTfRI 667 mTfR2724 mTfRl700 hTfR2 708 hTfRI 697  |GQL[jT)l K  L D Y E M Y N S K L L S F M K D L N Q F | K T MY L T H D | E L D F OIR Y G D @ V L R H[T_G]N L N E | F R L S H D E L I L GQLL I LINILD Y E I R Y I N S Q L I L J S F V R D I L N I Q Y R A K L T H J p j E b Q[F~V] I 1  YY J I) I |R R A A| A A E E K K|L L|R R Q E ILYJS S D L I K J A R G L | T |LL Q QW W |VV Y | YSSAAR RGGD D L O W L I Y S A R O D I F F | R A | T S RILJTLTJD F G NA|EJ DIIIKETM G L AjD  Y H F L S P Y V S P[RE  N R F VMRIETIH IDIE R L T J R J M Y N | V | R PJRF V Q K K T  SAQ|VEF Y F L S Q Y V S P A J D  VEYHFLSPYVSPlKE  F R H I F ILJG1QJO]D|H T L G A L V D H L R M LJRJA D G S G A A S S R L [ T . . - R Q K N I F R H I F W GS G S H T L | H A L V E N L K L L - - F R H I F@o(l)ODHTLOALLDHLRLLRjSN S S G T P G A - F J R J # 1 F J J V ^ ^ SIHTLIFIA L I L E N 1 L 1 K | L ( R K R | R | Q L A L | L [ T W T L R N Q L A L A T W T I Q R R Q L A L L T W T L | Q RINIOLALIAITWTII O  mTfR2760 rnTfRI 728 hTfR2 742 hTfRI 722 mTfR2796  NN  rnTfRI 761 hTfR2 778  NEF NN F ME  hTfRI 758  F Y F L S Q Y V S P  [D]ERLIMRMYN  Q G A A N A L S G D VWN I D G 0 A N A L SGD|t)WN I D AA AN N AA L S G D V W N I D G A GAANALSGDVWIDIIDI  759 727 741 721 795 760 777 757 798 763 780 760  119  sequence is known. The sequence in Figure 4.8 has been aligned to the human TfR2, as well as both human and mouse T f R l . The mouse TfR2 shares 40% amino acid identity with mouse T f R l , and 82% identity with its human orthologue. Through comparison to the human form of the protein, it was determined that the mouse TfR2 clone is the ocform of the TfR2, or the transmembrane (surface expressed) form of the protein. Through sequence analysis it is clear that the newly discovered mouse TfR2 possesses an internalization signal at residues 23 to 26 that is very similar to that of the classical transferrin receptor. The mouse TfR2 internalization signal consists of a Y R R V (tyrosine, arginine, arginine, and valine) which is very similar to the internalization signal of the mouse T f R l  and human T f R l  of Y T R F (tyrosine, threonine, arginine,  phenylalanine). In addition, the RGD sequence (arginine, glycine, and aspartate) located at amino acid residue 673 is completely conserved throughout all four proteins as shown in Figure 4.8. The R G D sequence is known to be critical to T f binding to the T f R l (Dubljevice/a/., 1999).  C.  Results II: Myc-tagged mouse transferrin receptor 2 construct and transfections Figure 4.9 depicts Chinese hamster ovary cells transfected with the vector  containing the mouse TfR2 c D N A tagged with a human c-myc tag. As the antigen (cmyc) is located on the intracellular tail of the receptor, the cells were initially fixed with 2% paraformaldehyde then treated with 0.1 % saponin to permeablize the membrane and stained with anti-myc antibody. In Figure 4.9, the TfR2 can be seen on the cell surface of the four cells, as well as in vesicles within the cell, but is excluded from the nuclei. This  120  Figure 4.9: Chinese hamster ovary ceils transfected with mouse transferrin receptor 2. Cells are permeablized and stained with the anti-c-myc antibody 9E10 followed by goat anti-mouse Alexa 488 to show subcellular localization of mouse TfR2. Scale bar = 5 um  121  staining was performed to confirm the expression of the TfR2 and the presence of the mouse TfR2 on the cell surface prior to experiments to examine iron uptake.  D.  Discussion The full-length sequence of the mouse TfR2 reveals a 798 amino acid protein that  shares 40% sequence identity with the classical mouse T f R l . It also shows 82% amino acid identity with its human orthologue. The proteins are very similar in many important aspects, including the internalization signals. The fact that the mouse TfR2 possesses an internalization signal that is very similar to that of the classical T f R l , and conforms to the typical format of an endocytosis signal, indicates that the receptor is probably able to internalize from the cell surface and targeted to the endosome in a manner similar to that of the classical receptor. This targeting was confirmed experimentally (Figure 4.6). The tyrosine-X-X-(hydrophobic residue) motif has been recognized as the motif involved in internalization of clathrin-coated pits by endocytosis, and is recognized by the u2 subunit of AP2 adaptors (Owen and Evans, 1998). The AP2 subunit links the internalization signal to the clathrin coat through the p.2 subunit. The putative transmembrane domain, determined through alignment of the mouse and human TfR2 proteins is 92% identical at the amino acid level to that of the human orthologue, with 22 of 24 amino acids shared between the two sequences. Overall, the level of cross species conservation is extremely high.  122  IV.  Uptake of Fe-loaded p97 by mouse transferrin receptor 2  A.  Rationale  55  In order to determine i f p97 is a ligand of TfR2, the ability of p97 to deliver its bound iron to cells via TfR2 was determined. This was accomplished by comparing the amount of iron taken up by TfR- and mouse TfR2+ cells. In chapter 3 it was shown that p97 does bind to the TfRl in both radioligand studies and Pandex fluorescence studies. However, the Fe uptake studies with p97 and TfRl showed that p97 does not internalize 55  iron via this method. For this reason, the more functional aspects of p97 uptake by mouse TfR2 were addressed initially. In addition, binding of p97 loaded with Fe to the 55  surface of cells at 4°C (therefore with no receptor internalization) was also examined. The hypothesis being tested is that Fe-loaded p97 is able to both bind to cells via mouse 55  TfR2, and internalize within the cells via the receptor.  B.  Results 1: Fe-p97 uptake by mouse transferrin receptor 2 55  Figure 4.10 shows that significantly more radioactive iron (p<0.005 by Student's t-test) is internalized by the CHO-mouse TfR2+ cells than the cells transfected with vector alone. The cells were initially washed to remove any bound Tf, and blocked with B S A to try to minimize non-specific background binding of the protein to the cells. The cells were then incubated with Fe-loaded p97 for 2 hours at 37°C, washed on ice and 55  harvested for counting in the Beckman scintillation counter. The results are from a representative experiment, performed in triplicate.  123  **************  ************** ******** ************** -  ••••• ••••< ••••<  mi  ****<  >,***»*»»,  nun?  tmt  . . • • • • • • • » •  w*•••••• ••••••  • • • • • •  $*************< „ •••••••••••^ »••••••••••••<  ************** ************** ************** , **************{ **************! ************** >••••••••••••• ************** ************** ************** ************** ************** ************** ,  *********** ************** ************** **************  • • • • • • • • • • • • • « • • • • • • • • • • • 4 * »  ************** ************** CHO-vector alone  CHO-mTR2  Figure 4.10: Fe4oaded p97 uptake by Chinese hamster ovary cells and Chinese hamster ovary cells transfected with mouse transferrin 55  receptor 2. The cells with and without the mouse Tf R2 were allowed to internalize the F e - l o a d e d p97 for 2 hours and the internalized counts were measured in a scintillation counter. Represents the mean +/- standard deviation for a representative experiment, n=6. 'Statistical significance w a s assessed with the use of Student's t-test (p<0.005). 55  124  C.  Results II: Fe-p97 binding by mouse transferrin receptor 2 55  The cells in this experiment (TfR- and mouse TfR2+) were incubated with Fe55  loaded p97 at 4°C. The cells were then washed, harvested, and the D P M of the bound 55  Fe was determined in a scintillation counter. The TfR- cells in this experiment (Figure  4.11) have significantly less Fe bound to the surface of the cells than the mouse TfR2+ 55  cells, after being incubated with the iron loaded p97 (Student's t-test, p<0.005). The results are from a representative experiment, performed in triplicate.  D.  Discussion Two experiments were designed to show i f p97 binds to TfR2, and if iron bound  to p97 can be internalized by TfR2. These experiments both use human p97 and mouse TfR2. While human and mouse TfR2 are highly conserved (82% identity at the amino acid level), cross-species binding experiments are not ideal.  They were, however,  necessary in this case for a number of reasons. To begin, mouse p97 has been cloned, but is not yet available in the recombinant, soluble form at the quantities necessary for binding experiments. Secondly, while human TfR2 is available, and cloned into TfRcells, (such as those used earlier in this chapter for the heterodimer experiments) these cells do not function as expected in the binding and uptake experiments. A series of 125  55  experiments looking at  Fe-loaded p97 and T f as well as  125  I-p97 and  I-Tf uptake and  cell surface binding were undertaken, and the results were rather surprising.  While the  TfRl+ cells were able to both bind and internalize Tf, the TfR2+ cells were not. This is in contrast to the published results for TfR2+ cells taking up Fe via T f (Kawabata et al., 55  1999). In fact, the TfR2+ cells had significantly less iron uptake than the TfR- cells (data  125  8001 700 • c OJ CL  to c o CD  600 500 400 300  E  "55 b  200 100 0  Hi  TfR-  mouse  TfR2+  Figure 4.11: Binding of F e - l o a d e d p97 to cells with and without the m o u s e transferrin receptor 2. TfR- and TfR2+ 55  cells were incubated with F e - p 9 7 on ice, and the bound radioactivity was measured in a scintillation counter. Represents the mean +/- standard deviation for a representative experiment, n=6. 'Statistical significance was assessed with Student's t-test (p<0.005). 55  126  not shown).  The TfR2 F L AG-tagged construct was generated by Kawabata and  associates to transfect TRVb cells. However, the individual clone of TfR2+ cells used for these experiments is not the same cell clone as that published by the Koeffler group at the Cedars-Sinai Medical Centre, but was instead cloned in Caroline Enns' laboratory at Oregon Health Sciences University. To resolve the discrepancy between the activities of TfR2 in the two cell lines and to be able to draw conclusions about the function of TfR2 with regard to T f uptake, additional studies on new clones should be undertaken. It is very likely that the unexpected decrease in internalized iron observed in the TfR2+ clones is a result of a problem with the expression of TfR2, or with the integration of the vector into the chromosomes of the cell. Even though these experiments use mouse TfR2 and human p97, both the uptake of F e via p97 (Figure 4.10) and the cell surface binding of Fe-loaded p97 (Figure 55  55  4.11) are clear. The differences between the D P M measured in both cases is statistically significant by Student's t-test (p<0.005).  It seems logical that once the appropriate  reagents are generated, the intra-species experiments (mouse p97 binding to mouse TfR2 and human p97 binding to human TfR2) may give even more convincing binding results. Since p97 is very homologous to Tf, it is likely that the interaction between p97 and TfR2 is similar to that of Tf and TfR2, in terms of the binding pocket. While the binding pocket has not been studied in TfR2, some binding information is known. For example, it has been established that TfR2, like T f R l , binds T f (although likely with lower affinity) and that TfR2, in contrast to T f R l , does not bind HFE (West et al, 2000). One part of the binding pocket of human T f R l , the R G D motif (arginine, glycine, aspartate at 646-648), is known to be crucial to T f binding (Dubljevic et al, 1999). A  127  future experiment that is currently underway involves the mutation of this R G D site (which is fully conserved in both the mouse and human TfR2) by replacing the arginine with a lysine residue. This substitution leads to a reduction of T f binding to T f R l by 95% of the wild-type value (Dubljevic et al, 1999). The effect that this mutation has on both p97 and T f binding will answer some important questions regarding the action of TfR2.  128  Chapter 5:  Conclusions and future directions  The field of iron uptake and transport has generated research interest for many years.  Iron is essential to life because it participates in single electron oxidation-  reduction reactions involved in a range of important biological processes including the production of A T P (Trinder et al, 2000). Unregulated uptake and transport of iron is detrimental because it has the potential to lead to the formation of reactive oxygen species that cause molecular damage (Sayre et al, 1999). New components of the iron uptake system are being discovered all the time, such as hepcidin, which is a small, soluble protein believed to be involved in communicating the iron storage status of the body to the absorptive cells of the intestine (Fleming and Sly, 2001). On the other hand, the function of some proteins involved in the iron uptake and transport system is still a mystery, despite the fact they were identified long ago. p97 is one of these proteins, with a function that remains controversial. The ability of GPI-p97 to take up iron and deliver the iron to the storage protein ferritin has been demonstrated (Tiong, 2001; Tiong and Jefferies, 2002), and this suggests that some niche does exist for p97 in the iron transport scheme.  Some studies, however, have questioned the role of GPI-linked p97 in  physiological iron uptake, both because p97 expression is not regulated by iron (Richardson, 2000), and because the Tf/TfRl iron uptake system seems so efficient as an iron uptake system that additional systems seem redundant. The existence of a soluble form of p97 created through alternative splicing suggests that there maybe a unique function for this form of the protein. Elucidating this function through examining possible receptor interactions has been the major focus of this work.  129  The interaction of p97 with the most logical receptor choice, T f R l , was characterized with the use of radioligand assays and immunofluorescent labeling Pandex assays. These experiments have shown that p97 does interact with the TfRl in binding experiments, that the interaction is pH specific (in that it occurs at pH 7.0 but not 6.0), and can be specifically competed with excess unlabeled p97. The binding also seems specific, as excess p97 can compete with the binding of labeled T f to T f R l , and reduce the binding of T f by approximately 30%.  The failure of p97 to facilitate the  internalization of iron via TfRl indicates that this binding interaction is not likely to function in terms of classical iron uptake into the cell.  This does not rule out the  possibility that this interaction could function in some other capacity, such as to transport p97 across the B B B from the brain to the blood. The BIAcore studies were not able to measure any interaction between p97 and T f R l . This indicates that if an interaction does exist between the two proteins, it may require other cellular components in order for the interaction to occur, or that the binding affinity was too low to measure with the sensitivity of the BIAcore. In the future it may be possible to increase the sensitivity of the system by increasing the percentage of the TfRl that is immobilized on the chip in the proper orientation. In other studies, the TfRl was coupled to the chip through an anti-His-tag antibody which recognized the His tag on the C-terminus of T f R l , thus a large percentage of the receptor was oriented correctly. In these studies the resonance signal generated for T f binding was about 3 times that observed in the experiments included here (West et al., 2000). BIAcore studies were also undertaken to test another theory of p97 function: that due to the presence of an RGD motif in p97 similar to the R G D motifs present in the  130  ligands of many integrins, p97 may interact with a receptor such as the  0CvP3  integrin.  p97 has been implicated in angiogenesis, since it appears to stimulate vasculogenesis measured in a chick chorioallantoic assay (Sala et al, 2002), thus an association with the integrin av|33, known to interact with the vascular endothelial growth factor receptor 2, seems possible (Borges et al., 2000; Brooks et al, 1994b). This experiment did not demonstrate binding between p97 and the integrin. The work should be expanded, however, to examine both the possible interaction of p97 with other integrins, and to include the study of binding to integrins in cell based systems, which would allow the interaction with other factors that may be crucial to the interaction. In the search for other likely candidate receptors for p97, a novel homologue of the T f R l was discovered through mining of the EST database.  This receptor,  independently published by Kawabata (Kawabata et al., 1999), shows 45% identity and 66%) similarity at the amino acid level, with T f R l .  Through various experiments  designed to characterize this new receptor, the expression was revealed to be largely restricted to the liver. In embryogenesis, the mouse TfR2 m R N A is present by E l 5 and continues to increase in expression through E l 7 , in contrast to T f R l m R N A that is discernable by E7, increases until E l 5 and decreases by E17. This research has also demonstrated that TfR2 is present on the brain endothelial cells that form the B B B , in contrast with results produced by Kawabata et al. This result is important as soluble p97 has been demonstrated to cross the B B B , and thus candidate receptors for this protein are likely to be expressed in the brain microvasculature (Moroo et al., 2002). Subcellular localization experiments have indicated that the TfR2 protein is internalized into vesicles that also stain for both EEA1 (a marker of early endosomes) and  131  clathrin heavy chain. Transferrin receptor 1 is known to localize to these same vesicles during endocytosis, thus it is likely that the two receptors follow the same internalization pathway. One future experiment to confirm this expression involves modification of the internalization signal of the TfR2, to determine i f this modification will inhibit internalization of the receptor, as it does with T f R l .  Another direction that will be  interesting to explore in the future will be whether TfR2 is able to undergo transcytosis in polarized cells in a similar manner to T f R l .  Transferrin receptor 1 has a basolateral  sorting signal contained within residues 29 to 35 (Dargemont et al., 1993; Odorizzi and Trowbridge, 1997), which do not seem to be highly conserved between the two receptors. Since TfR2 is present in the polarized cells of the B B B , the question of whether it undergoes sorting to the apical or basolateral side of a polarized cell, and i f it can undergo transcytosis, are very pertinent. Interestingly, in transfected cells that over express both T f R l and TfR2, the receptor is present as a heterodimeric combination of the two receptors. This result must be examined further, as it poses some new and interesting questions. For instance, what unique role, i f any, does this heterodimeric receptor play in cells that express both TfRl and TfR2, such as cells in the erythroid lineage (Kawabata et al., 2001). Since TfRl is closely regulated by intracellular iron levels, would the heterodimerization of the two proteins confer some level of iron-regulation on whatever function this heterodimeric receptor carries out?  The results of these functional experiments could be very  interesting and lead to a new level of understanding in cellular iron uptake. Mouse TfR2, which was isolated from the EST database and sequenced through the technique of "primer walking" to confirm that the full-length c D N A was generated,  132  displays 40% amino acid identity with mouse T f R l , and 82% amino acid identity with human TfR2.  The mouse TfR2, cloned with a myc-tag on the C-terminus for  identification, was transfected into both CHO cells and TfR- cells and tested for both p97 binding and iron uptake via p97. Human p97 was able to facilitate the uptake of Fe 55  through mouse TfR2, and the p97 was demonstrated to bind to the surface of the cells. These results are intriguing, because while p97 was able to bind to T f R l , the iron uptake or functional aspect of the interaction was not present. With TfR2, the functional aspect seems to occur along with the binding. The next steps in this research are already underway. First, a soluble form of mouse p97 is being generated for use in these studies, so that an inter-species binding system will no longer be necessary. As well, new clones of the human TfR2 in TfR- cells generated in the Enns' laboratory will soon be isolated for further work on the human p97 interaction with TfR2. In collaboration with Pamela Bjorkman's lab at the California Institute of Technology, new soluble recombinant forms of human TfR2 for BIAcore studies are being produced. While the Bjorkman group has produced soluble TfR2 for BIAcore studies in the past (West et al, 2000), a problem with protein stability generated BIAcore results in this study that did not show T f binding to TfR2 (data not shown), thus were also unsuccessful at demonstrating any interaction with p97.  Hopefully this protein stability problem will be rectified in the near future and the  binding interactions between p97 and TfR2 can be measured by BIAcore. Clearly, identification of TfR2 as a receptor of p97 is only the first step toward the ultimate goal of clarifying the function of soluble p97. To reach this final conclusion, functional assays must be developed. The difficulty now will be to design these assays  133  and try to answer the pressing biological questions such as what p97 does as a soluble, iron binding protein, and why it is increased in Alzheimer disease.  134  Bibliography  Abbott, N . J., Chugani, D. C , Zaharchuk, G., Rosen, B. R. and Lo, E. H . (1999). Delivery of imaging agents into brain. Adv Drug Deliv Rev 3 7 , 253-277. Abboud, S. and Haile, D. J. (2000). A novel mammalian iron-regulated protein involved in intracellular iron metabolism. J Biol Chem 2 7 5 , 19906-12. Adams, P. C. and Chakrabarti, S. (1998). Genotypic/phenotypic correlations in genetic hemochromatosis: evolution of diagnostic criteria. Gastroenterology 114, 319-23. Aisen, P. and Leibman, A. (1968). The stability constants of the Fe3+ conalbumin complexes. Biochem Biophys Res Commun 3 0 , 407-13. Aisen, P. and Leibman, A . (1972). Lactoferrin and transferrin: a comparative study. Biochim Biophys Acta 2 5 7 , 314-23. Alemany, R., Vila, M . R., Franci, C , Egea, G., Real, F. X . and Thomson, T. M . (1993). Glycosyl phosphatidylinositol membrane anchoring of melanotransferrin (p97): apical compartmentalization in intestinal epithelial cells. J Cell Sci 104, 1155-62. Anderson, B . F., Baker, H . M . , Norris, G . E., Rumball, S. V . and Baker, E. N . (1990). Apolactoferrin structure demonstrates ligand-induced conformational change in transferrins. Nature 3 4 4 , 784-7. Andrews, N . C. (1999). The iron transporter D M T 1 . Int J Biochem Cell Biol 3 1 , 991-4. Andrews, N . C. (2000). Iron homeostasis: insights from genetics and animal models. Nat Rev Genet 1, 208-17.  135  Angel, P. and Karin, M . (1991). The role of Jun, Fos and the AP-1 complex in cell-proliferation and transformation. Biochim Biophys Acta 1072, 129-57. Attieh, Z. K., Mukhopadhyay, C. K., Seshadri, V., Tripoulas, N . A . and Fox, P. L . (1999). Ceruloplasmin ferroxidase activity stimulates cellular iron uptake by a trivalent cation-specific transport mechanism. J Biol Chem 274, 1116-23. Aziz, N . and Munro, H . N . (1987). Iron regulates ferritin m R N A translation through a segment of its 5' untranslated region. Proc Natl Acad Sci USAS4,  8478-82.  Bacon, B. R., Powell, L . W., Adams, P. C , Kresina, T. F. and Hoofnagle, J. H . (1999). Molecular medicine and hemochromatosis: at the crossroads. Gastroenterology 116, 193-207. Bailey, S., Evans, R. W., Garratt, 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 transferrin at 3.3-A resolution. Biochemistry 27, 5804-12. Baker, E. N . , Baker, H . M . , Smith, C. A., Stebbins, M . R., Kahn, M . , Hellstrom, K . E. and Hellstrom, I. (1992). Human melanotransferrin (p97) has only one functional iron-binding site. FEBS Lett 298, 215-8. Bali, P. K . and Aisen, P. (1992). Receptor-induced switch in site-site cooperativity during iron release by transferrin. Biochemistry 31, 3963-7. Bali, P. K . , Zak, O. and Aisen, P. (1991). A new role for the transferrin receptor in the release of iron from transferrin. Biochemistry 30, 324-8. Baranano, D. E., Wolosker, H . , Bae, B . I., Barrow, R. K . , Snyder, S. H . and Ferris, C. D . (2000). A mammalian iron ATPase induced by iron. J Biol Chem 275, 15166-73.  136  Barresi, G. and Tuccari, G. (1994). Immunocytochemical demonstration of melanotransferrin (p97) in thyroid tumors of follicular cell origin. Pathology 26, 127-9. Bartlett, W. P., L i , X . S. and Connor, J. R. (1991). Expression of transferrin mRNA in the CNS of normal and jimpy mice. J Neurochem 57, 318-22. Beal, M . F. (1998). Mitochondrial dysfunction in neurodegenerative diseases. Biochim Biophys Acta 1366, 211-23. Beard, P., Offord, E., Paduwat, N . and Bruggmann, H . (1991). SV40 activates transcription from the transferrin receptor promoter by inducing a factor which binds to the CRE/AP-1 recognition sequence. Nucleic Acids Res 19, 7117-23. Benito, P., House, W. and Miller, D. (1998). Comparison of oral and intraperitoneal iron supplementation in anaemic rats: a re-evaluation of the mucosal block theory of iron absorption. Br JNutr 79, 533-40. Boguski, M . S. (1995). The turning point in genome research. Trends Biochem Sci 20, 295-6. Boguski, M . S., Lowe, T. M . and Tolstoshev, C. M . (1993). dbEST-database for expressed sequence tags. Nat Genet 4, 332-3. Borges, E., Jan, Y . and Ruoslahti, E. (2000). Platelet-derived growth factor receptor beta and vascular endothelial growth factor receptor 2 bind to the beta 3 integrin through its extracellular domain. J Biol Chem 275, 39867-73. Bottomley, S. S., Wolfe, L . C. and Bridges, K . R. (1985). Iron metabolism in K562 erythroleukemic cells. J Biol Chem 260, 6811-5. Bradbury, M . W. (1997). Transport of iron in the blood-brain-cerebrospinal fluid system. J Neurochem 69, 443-54.  137  Brightman, M . W. (1992). Ultrastructure of brain endothelium. In Physiology and Pharmacology of the Blood-brain Barrier (ed. M . W. B. Bradbury), pp. 549. Berlin; New York: Springer Verlag. Brooks, P. C , Clark, R. A . and Cheresh, D. A . (1994a). Requirement of vascular integrin alpha v beta 3 for angiogenesis. Science 264, 569-71. Brooks, P. C , Montgomery, A . M . , Rosenfeld, M . , Reisfeld, R. A., Hu, T., Klier, G. and Cheresh, D. A . (1994b). Integrin alpha v beta 3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell 79, 1157-64. Brown, J. P., Woodbury, R. G., Hart, C. E., Hellstrom, I. and Hellstrom, K . E. (1981). Quantitative analysis of melanoma-associated antigen p97 in normal and neoplastic tissues. Proc Natl Acad Sci USA 78, 539-43. Bulaj, Z. J., Griffen, L . M . , Jorde, L . B., Edwards, C. Q. and Kushner, J. P. (1996). Clinical and biochemical abnormalities in people heterozygous  for  hemochromatosis. N Engl J Med 335, 1799-805. Burke, W., Thomson, E., Khoury, M . J., McDonnell, S. M . , Press, N . , Adams, P. C , Barton, J. C , Beutler, E., Brittenham, G., Buchanan, A., Clayton, E. W., Cogswell, M . E., Meslin, E. M . , Motulsky, A . G., Powell, L . W., Sigal, E., Wilfond, B . S. and Collins, F. S. (1998). Hereditary hemochromatosis: gene discovery and its implications for population-based screening. Jama 280, 172-8. Bush, A . I., Pettingell, W. H . , Multhaup, G., d Paradis, M . , Vonsattel, J. P., Gusella, J. F., Beyreuther, K., Masters, C. L . and Tanzi, R. E. (1994). Rapid induction of Alzheimer A beta amyloid formation by zinc. Science 265, 1464-7.  138  Camaschella, C , Roetto, A., Cali, A., De Gobbi, M . , Garozzo, G., Carella, M . , Majorano, N . , Totaro, A . and Gasparini, P. (2000). The gene TFR2 is mutated in a new type of haemochromatosis mapping to 7q22. Nat Genet 25, 14-5. Campbell, A . M , Kessler, P. D. and Fambrough, D. M . (1992). The alternative carboxyl termini of avian cardiac and brain sarcoplasmic reticulum/endoplasmic reticulum Ca(2+)-ATPases are on opposite sides of the membrane. J Biol Chem 267, 9321-5. Casey, J. L., D i Jeso, B., Rao, K . K., Rouault, T. A., Klausner, R. D. and Harford, J. B. (1988). Deletional analysis of the promoter region of the human transferrin receptor gene. Nucleic Acids Res 16, 629-46. Casey, J. L . , Koeller, D. M . , Ramin, V . C , Klausner, R. D. and Harford, J. B . (1989). Iron regulation of transferrin receptor m R N A levels requires iron-responsive elements and a rapid turnover determinant in the 3' untranslated region of the mRNA. EmboJ8, 3693-9. Collawn, J. F., Lai, A., Domingo, D., Fitch, M . , Hatton, S. and Trowbridge, I. S. (1993). Y T R F is the conserved internalization signal of the transferrin receptor, and a second Y T R F signal at position 31-34 enhances endocytosis. J Biol Chem 268, 21686-92. Collawn, J. F., Stangel, M . , Kuhn, L . A., Esekogwu, V., Jing, S. Q., Trowbridge, I. S. and Tainer, J. A . (1990). Transferrin receptor internalization sequence Y X R F implicates a tight turn as the structural recognition motif for endocytosis. Cell 63, 106172. Connor, J. R. and Menzies, S. L . (1996). Relationship of iron to oligodendrocytes and myelination. Glia 17, 83-93.  139  Connor, J. R., Menzies, S. L., Burdo, J. R. and Boyer, P. J. (2001). Iron and iron management proteins in neurobiology. Pediatr Neurol 25, 118-29. Conrad, M . E., Umbreit, J. N . and Moore, E. G. (1999). Iron absorption and transport. Am J Med Sci 318, 213-29. Cook, J. D., Skikne, B. S. and Baynes, R. D. (1993). Serum transferrin receptor. Annu Rev Med 44, 63-74. Crawford, D. H . , Jazwinska, E. C , Cullen, L . M . and Powell, L . W. (1998). Expression of HLA-linked hemochromatosis in subjects homozygous or heterozygous for the C282Y mutation. Gastroenterology 114, 1003-8. Crowe, A . and Morgan, E. H . (1992). Iron and transferrin uptake by brain and cerebrospinal fluid in the rat. Brain Res 592, 8-16. Cullen, L . M . , Summerville, L., Glassick, T. V., Crawford, D. H., Powell, L . W. and Jazwinska, E. C. (1997). Neonatal screening for the hemochromatosis defect [letter]. Blood 90, 4236-7. Dargemont, C , Le Bivic, A . , Rothenberger, S., Iacopetta, B. and Kuhn, L . C. (1993). The internalization signal and the phosphorylation site of transferrin receptor are distinct from the main basolateral sorting information. Embo J12, 1713-21. Dautry-Varsat, A . , Ciechanover, A . and Lodish, H . F. (1983). p H and the recycling of transferrin during receptor-mediated endocytosis. Proc Natl Acad Sci  USA  80, 2258-62. de Silva, D., Davis-Kaplan, S., Fergestad, J. and Kaplan, J. (1997). Purification and characterization of Fet3 protein, a yeast homologue of ceruloplasmin. J Biol Chem 272, 14208-13.  140  Devary, Y . , Gottlieb, R. A . , Lau, L . F. and Karin, M . (1991). Rapid and preferential activation of the c-jun gene during the mammalian U V response. Mol Cell 5/0/11,2804-11. Dewan, J. C., Mikami, B . , Hirose, M . and Sacchettini, J. C. (1993). Structural evidence for a pH-sensitive dilysine trigger in the hen ovotransferrin N-lobe: implications for transferrin iron release. Biochemistry 32, 11963-8. Diamond, M . S. and Springer, T. A . (1994). The. dynamic regulation of integrin adhesiveness. Curr Biol 4, 506-17. Dickinson, T. K . , Deyenyi, A . G..'and .Connor, J. R. (1996). Distribution of injected iron 59 and manganese 54 in hypotransferrinemic mice. J Lab Clin Med 128, 270-8. Donovan, A., Brownlie, A., Zhou, Y., Shepard, J., Pratt, S. J., Moynihan, J., Paw, B. H., Drejer, A., Barut, B., Zapata, A., Law, T. C , Brugnara, C , Lux, S. E., Pinkus, G. S., Pinkus, J. L., Kingsley, P. D., Palis, J., Fleming, M . D., Andrews, N . C. and Zon, L . I. (2000). Positional cloning of zebrafish ferroportin 1 identifies a conserved vertebrate iron exporter. Nature 403, 776-81. Dubljevic, V . , Sali, A . and Goding, J. W. (1999). A conserved R G D (Arg-GlyAsp) motif in the transferrin receptor is required for binding to transferrin. Biochem J 341, 11-4. Duchange, N . , Ochoa, A., Plowman, G. D., Roze, A., Amdjadi, M . and Zakin, M . M . (1992). Identification of an enhancer involved in the melanoma-specific expression of the tumor antigen melanotransferrin gene. Nucleic Acids Res 20, 2853-9.  141  Ekblom, P. and Thesleff, I. (1985). Control of kidney differentiation by soluble factors secreted by the embryonic liver and the yolk sac. Dev Biol 110, 29-38. Evans, P. and Kemp, J. (1997). Exon/intron structure of the human transferrin receptor gene. Gene 199, 123-31. Feder, J. N . , Gnirke, A., Thomas, W., Tsuchihashi, Z., Ruddy, D. A., Basava, A., Dormishian, F., Domingo, R., Jr., Ellis, M . C , Fullan, A., Hinton, L . M . , Jones, N . L . , Kimmel, B . E . , Kronmal, G. S., Lauer, P., Lee, V . K . , Loeb, D . B . , Mapa, F. A . , McClelland, E., Meyer, N . C , Mintier, G. A . , Moeller, N . , Moore, T., Morikang, E., Wolff, R. K . and et al. (1996). A novel M H C 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 . , Wolff, R. K., Parkkila, S., Sly, W. S. and Schatzman, R. C. (1997). The hemochromatosis founder mutation in H L A - H disrupts beta2-microglobulin interaction and cell surface expression. J Biol Chem 272, 14025-8. Feldman, H., Gabathuler, R., Kennard, M . , Nurminen, J., Levy, D., Foti, S., Foti, D., Beattie, B . and Jefferies, W. (2001). Serum p97 levels as an aid to identifying Alzheimer's Disease. Journal of Alzheimer's Disease 3, 507-516. Ferris, C. D., Jaffrey, S. R., Sawa, A., Takahashi, M . , Brady, S. D., Barrow, R. K., Tysoe, S. A . , Wolosker, H . , Baranano, D. E., Dore, S., Poss, K . D. and Snyder, S. H . (1999). Haem oxygenase-1 prevents cell death by regulating cellular iron. Nat Cell Biol 1, 152-7. Finch, C. (1994). Regulators of iron balance in humans. Blood 84, 1697-702.  142  Fleming, R. E., Migas, M . C , Holden, C. C , Waheed, A., Britton, 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 Sci USA91,  2214-9.  Fleming, R. E., Migas, M . C , Zhou, X . , Jiang, J., Britton, 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 USA 96, 3143-8. Fleming, R. E. and Sly, W. S. (2001). Hepcidin: a putative iron-regulatory hormone relevant to hereditary hemochromatosis and the anemia of chronic disease. Proc Natl Acad Sci USA9S,  8160-2.  Food, M . R., Rothenberger, S., Gabathuler, R., Haidl, I. D., Reid, G. and Jefferies, W. A . (1994). Transport and expression in human melanomas of a transferrin-like glycosylphosphatidylinositol-anchored protein. JBiol Chem 269, 3034-40. Forgac, M . (1992). Structure and properties of the coated vesicle proton pump. Ann N YAcad Sci 671, 273-83. Fuchs, H., Lucken, U., Tauber, R., Engel, A . and Gessner, R. (1998). Structural model of phospholipid-reconstituted human transferrin receptor derived by electron microscopy. Structure 6, 1235-43. George, D. K . , Ramm, G. A . , Powell, L . W., Fletcher, L . M . , Walker, N . I., Cowley, L . L . and Crawford, D. H . (1998). Evidence for altered hepatic matrix degradation in genetic haemochromatosis. Gut 42, 715-20. Gitlin, J. D. (1998). Aceruloplasminemia. Pediatr Res 44, 271-6.  143  Goldwurm, S. and Powell, L . W. (1997). Haemochromatosis after the discovery of HFE (HLA-H). Gut 41, 855-6. 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, 13742. Griffiths, W. J., Kelly, A . L., Smith, S. J. and Cox, T. M . (2000). Localization of iron transport and regulatory proteins in human cells. Qjm 93, 575-87. Gruenheid, S., Cellier, M . , Vidal, S. and Gros, P. (1995). Identification and characterization of a second mouse Nramp gene. Genomics 25, 514-25. Gunshin, H., Mackenzie, B., Berger, U . V . , Gunshin, Y . , Romero, M . F., Boron, W. F., Nussberger, S., Gollan, J. L . and Hediger, M . A . (1997). Cloning and characterization of a mammalian proton-coupled metal-ion transporter. Nature 388, 4828. Gutierrez, J. A . , Y u , J., Rivera, S. and Wessling-Resnick, M . (1997). Functional expression cloning and characterization of SFT, a stimulator of Fe transport [published erratum appears in J Cell Biol 1999 Oct 4;147(l):following 204]. J Cell Biol 139, 895905. Hashimoto, S. (2000). Principles of B I A C O R E . In Real-Time Analysis  of  Biomolecular Interactions: Applications of Biacore (ed. K . Nagata and H . Handa), pp. 256. Toyko: Springer-Verlag. Hsu, F. (1997). Melanotransferrin. In Biotechnology. Vancouver: University of British Columbia.  144  Hu, D. D., Barbas, C. F. and Smith, J. W. (1996). A n allosteric Ca2+ binding site on the beta3-integrins that regulates the dissociation rate for R G D ligands. J Biol Chem 271,21745-51. Hynes, R. O. (1987). Integrins: a family of cell surface receptors. Cell 48, 549-54. Iacopetta, B . J., Rothenberger, S. and Kuhn, L . C. (1988). A role for the cytoplasmic domain in transferrin receptor sorting and coated pit formation during endocytosis. Cell 54, 485-9. Ikuta, K . , Fujimoto, Y., Suzuki, Y., Tanaka, K . , Saito, H., Ohhira, M . , Sasaki, K . and Kohgo, Y . (2000). Overexpression of hemochromatosis protein, H F E , alters transferrin recycling process in human hepatoma cells. Biochim Biophys Acta 1496, 22131. Jazwinska, E. C. and Powell, L . W. (1997). Hemochromatosis and H L A - H : definite! Hepatology 25, 495-6. Jefferies, W. A . , Brandon, M . R., Hunt, S. V . , Williams, A . F., Gatter, K . C. and Mason, D. Y . (1984). Transferrin receptor on endothelium of brain capillaries. Nature 312, 162-3. Jefferies, W. A . , Food, M . R., Gabathuler, R., Rothenberger, S., Yamada, T., Yasuhara, O. and McGeer, P. L . (1996). Reactive microglia specifically associated with amyloid plaques in Alzheimer's disease brain tissue express melanotransferrin. Brain Res 712, 122-6. Jing, S. Q., Spencer, T., Miller, K . , Hopkins, C. and Trowbridge, I. S. (1990). Role of the human transferrin receptor cytoplasmic domain in endocytosis: localization of a specific signal sequence for internalization. J Cell Biol 110, 283-94.  145  Jing, S. Q. and Trowbridge, I. S. (1987). Identification of the intermolecular disulfide bonds of the human transferrin receptor and its lipid-attachment site. Embo J6, 327-31. Jolley, M . E., Wang, C. H . , Ekenberg, S. J., Zuelke, M . S. and Kelso, D. M . (1984). Particle concentration fluorescence immunoassay (PCFIA): a new, rapid immunoassay technique with high sensitivity. J Immunol Methods 67, 21-35. Kawabata, H . , Germain, R. S., Vuong, P. T., Nakamaki, T., Said, J. W. and Koeffler, H . P. (2000). Transferrin receptor 2-alpha supports cell growth both in ironchelated cultured cells and in vivo. JBiol Chem 275, 16618-25.. Kawabata, H . , Nakamaki, T.,Tkonomi, P., Smith, R. D.; Germain, R. S. and Koeffler, H . P. (2001). Expression of transferrin receptor 2 in normal and neoplastic hematopoietic cells. Blood 98, 2714-9. Kawabata, H., Yang, R., Hirama, T., Vuong, P. T., Kawano, S., Gombart, A . F. and Koeffler, H . P. (1999). Molecular cloning of transferrin receptor 2. A new member of the transferrin receptor-like family. JBiol Chem 274, 20826-32. Kennard, M . L . , Feldman, H . , Yamada, T. and Jefferies, 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 Jefferies, W. A . (1995). A novel iron uptake mechanism mediated by GPI-anchored human p97. Embo J 14,4178-86.  146  Kim, D. K . , Seo, M . Y . , Lim, S.W., Kim, S., Kim, J.W., Carroll, B.J., Kwon, D. Y., Kwon, T., Kang, S.S. (2001). Serum melanotransferrin, p97 as biochemical marker of Alzheimer's disease. Neuropsychopharmacology 24, In press. Klausner, R. D., Ashwell, G., van Renswoude, J., Harford, J. B . and Bridges, K . R. (1983). Binding of apotransferrin to K562 cells: explanation of the transferrin cycle. Proc Natl Acad Sci U S A 80, 2263-6. Klomp, L . W. and Gitlin, J. D. (1996). Expression of the ceruloplasmin gene in the human retina and brain: implications for a pathogenic model in aceruloplasminemia. Hum Mol Genet 5, 1989-96. Knutson, M . D., Levy, J. E., Andrews, N . C. and Wessling-Resnick, M . (2001). Expression of stimulator of Fe transport is not enhanced in Hfe knockout mice. J Nutr 131, 1459-64. Krause, A . , Neitz, S., Magert, H . J., Schulz, A . , Forssmann, W. G., SchulzKnappe, P. and Adermann, K . (2000). L E A P - 1 , a novel highly disulfide-bonded human peptide, exhibits antimicrobial activity. FEBS Lett 480, 147-50. Kuhn, L. C , Schulman, H. M . and Ponka, P. (1990). Iron-transferrin requirements and transferrin receptor expression in proliferating cells. In Iron Transport and Storage (ed. P. Ponka, R. C. Schulman and R. C. Woodworm), pp. 149-191. Boca Ratan, Fl.: CRC. Lawrence, C. M . , Ray, S., Babyonyshev, M . , Galluser, R., Borhani, D. W. and Harrison, S. C. (1999). Crystal structure of the ectodomain of human transferrin receptor. Science 286, 779-82.  147  Lebron, J. A., Bennett, 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. Cell 93, 11123. Levy, J. E., Jin, O., Fujiwara, Y., Kuo, F. and Andrews, N . C. (1999). Transferrin receptor is necessary for development of erythrocytes and the nervous system. Nat Genet 21, 396-9. Levy, J. E., Montross, L . K . and Andrews, N . C. (2000). Genes that modify the hemochromatosis phenotype in mice. J Clin Invest 105, 1209-16. Lok, C. N . , Chan, K . F. and Loh, T. T. (1996). Transcriptional regulation of transferrin receptor expression during phorbol-ester-induced HL-60 cell differentiation. Evidence for a negative regulatory role of the phorbol-ester-responsive element-like sequence. Eur J Biochem 236, 614-9. Lok, C. N . and Ponka, P. (1999). Identification of a hypoxia response element in the transferrin receptor gene. J Biol Chem 274, 24147-52. Lok, C. N . and Ponka, P. (2000). Identification of an erythroid active element in the transferrin receptor gene. J Biol Chem . Lonnerdal, B . , Woodhouse, L . R. and Glazier, C. (1987). Compartmentalization and quantitation of protein in human milk. JNutr 111, 1385-95. MacGillivray, R. T., Bewley, M . C , Smith, C. A . , He, Q. Y . , Mason, A . B . , Woodworm, R. C. and Baker, E. N . (2000). Mutation of the iron ligand His 249 to Glu in the N-lobe of human transferrin abolishes the dilysine trigger; but does not significantly affect iron release. Biochemistry 39, 1211-6.  148  MacGillivray, R. T., Moore, S. A., Chen, J., Anderson, B . F., Baker, FL, Luo, Y., Bewley, M . , Smith, C. A., Murphy, M . E., Wang, Y., Mason, A . B . , Woodworth, R. C , Brayer, G. D. and Baker, E. N . (1998). Two high-resolution crystal structures of the recombinant N-lobe of human transferrin reveal a structural change implicated in iron release. Biochemistry 37, 7919-28. 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 rosette receptor assay with haemmicrobeads. Demonstration of a haem receptor on K562 cells. Eur J Haematol 38, 21-5. Malecki, E. A . , Cook, B. M . , Devenyi, A . G., Beard, J. L . and Connor, J. R. (1999a). Transferrin is required for normal distribution of 59Fe and 54Mn in mouse brain. J Neurol Sci 170, 112-8. Malecki, E. A., Devenyi, A . G., Beard, J. L . and Connor, J. R. (1999b). Existing and emerging mechanisms for transport of iron and manganese to the brain. J Neurosci Res 56, 113-22. McGraw, T. E., Greenfield, L . and Maxfield, F. R. (1987). Functional expression of the human transferrin receptor c D N A in Chinese hamster ovary cells deficient in endogenous transferrin receptor. J Cell Biol 105, 207-14. McKie, A. T., Barrow, D., Latunde-Dada, G. O., Rolfs, A., Sager, G., Mudaly, E., Mudaly, M . , Richardson, C , Barlow, D., Bomford, A., Peters, T. J., Raja, K . B . , Shirali, S., Hediger, M . A . , Farzaneh, F. and Simpson, R. J. (2001). A n iron-regulated ferric reductase associated with the absorption of dietary iron. Science 291, 1755-9.  149  McKie, A . T., Marciani, P., Rolfs, A., Brennan, K., Wehr, K . , Barrow, D., Miret, S., Bomford, A . , Peters, T. J., Farzaneh, F., Hediger, M . A., Hentze, M . W. and Simpson., R. J. (2000). A Novel Duodenal Iron-Regulated Transporter, IREG1, Implicated in the Basolateral Transfer of Iron to the Circulation. Molecular Cell 5, 299-309. McNagny, K . M . (1990). Identification and Characterization of the Murine BP-3 Antigen. In Microbiology, pp. 119. Birmingham: University of Alabama at Birmingham. McNagny, K . M . , Rossi,, F., Smith, G. and Graf, T. (1996). The eosinophilspecific cell surface antigen, EOS47, is a chicken homologue of the oncofetal antigen melanotransferrin. Blood 87, 1343-52. Meehan, R. R., Barlow, D. P., Hill, R. E., Hogan, B. L . and Hastie, N . D. (1984). Pattern of serum protein gene expression in mouse visceral yolk sac and foetal liver. EmboJ3,  1881-5.  Mellman, I. (1996). Endocytosis and molecular sorting. Annu Rev Cell Dev Biol 12, 575-625. Metz-Boutigue, M . H . , Jolles, J., Mazurier, J., Schoentgen, F., Legrand, D., Spik, G., Montreuil, J. and Jolles, P. (1984). Human lactotransferrin: amino acid sequence and structural comparisons with other transferrins. Eur J Biochem 145, 659-76. Miskimins, W. K., McClelland, A . , Roberts, M . P. and Ruddle, F. H . (1986). Cell proliferation and expression of the transferrin receptor gene: promoter sequence homologies and protein interactions. J Cell Biol 103, 1781-8. Montosi, G., Donovan, A . , Totaro, A . , Garuti, C , Pignatti, E., Cassanelli, S., Trenor, C. C , Gasparini, P., Andrews, N . C. and Pietrangelo, A . (2001). Autosomal-  150  dominant hemochromatosis is associated with a mutation in the ferroportin (SLC11A3) gene. J Clin Invest 108, 619-23. Moos, T. and Morgan, E. H . (1998). Evidence for low molecular weight, nontransferrin-bound iron in rat brain and cerebrospinal fluid. J Neurosci Res 54, 486-94. Moroo, I., Ujiie, M . , Tiong, J. W. C , Walker, B . L . and Jefferies, W. A . (2002). The role of p97 in iron transport across the blood-brain barrier. Manuscript  in  preperation. Moroo, I., Yamada, T., Gabathuler, R., Kennard, M . L . , Nurminen, J. and Jefferies, W. A . (1999). Use of p97 in the diagnosis of Alzheimer's disease. Alzheimer's Reports 2, 353-8. Motulsky, H . (1995). Intuitive Biostatistics, pp. 386. Oxford: Oxford University Press. Mukhopadhyay, C. K . , Mazumder, B . and Fox, P. L . (2000). Role of hypoxiainducible factor-1 in transcriptional activation of ceruloplasmin by iron deficiency. J Biol Chem 275,21048-54. Mullner, E. W. and Kuhn, L . C. (1988). A stem-loop in the 3' untranslated region mediates iron-dependent regulation of transferrin receptor m R N A stability in the cytoplasm. Cell 53, 815-25. Nair, M . K., Ebner, K. E. and Cook, J. D. (1990). Isolation and characterization of a transferrin binding protein from rat plasma. Biochim Biophys Acta 1035, 306-12. Nice, E . C. and Catimel, B . (1999). Instrumental biosensors: new perspectives for the analysis of biomolecular interactions. Bioessays 21, 339-52.  151  Nicolas, G., Bennoun, M . , Devaux, I., Beaumont, C., Grandchamp, B., Kann, A . and Vaulont, S. (2001). Lack of hepcidin gene expression and severe tissue iron overload in upstream stimulatory factor 2 (USF2) knockout mice. Proc Natl Acad Sci USA  98,  8780-5. Nicolas, G., Bennoun, M . , Porteu, A., Mativet, S., Beaumont, C., Grandchamp, B., Sirito, M . , Sawadogo, M . , Kahn, A . and Vaulont, S. (2002). Severe iron deficiency anemia in transgenic mice expressing liver hepcidin. Proc Natl Acad Sci USA 99, 4596601. Niederau, C., Fischer, R., Purschel, A . , Stremmel, W., Haussinger, D. and Strohmeyer, G. (1996). Long-term survival in patients with hereditary hemochromatosis. Gastroenterology 1 1 0 , 1107-19. Niederau, C., Fischer, R., Sonnenberg, A., Stremmel, W., Trampisch, H . J. and Strohmeyer, G. (1985). Survival and causes of death in cirrhotic and in noncirrhotic patients with primary hemochromatosis. N Engl J Med 3 1 3 , 1256-62. Odorizzi, G. and Trowbridge, I. S. (1997). Structural requirements for basolateral sorting of the human transferrin receptor in the biosynthetic and endocytic pathways of Madin-Darby canine kidney cells. J Cell Biol 1 3 7 , 1255-64. Osaki, S. and Johnson, D. A . (1969). Mobilization of liver iron by ferroxidase (ceruloplasmin). J Biol Chem 244, 5757-8. Owen, D. and Kuhn, L . C. (1987). Noncoding 3' sequences of the transferrin receptor gene are required for mRNA regulation by iron. Embo J 6, 1287-93. Owen, D. J. and Evans, P. R. (1998). A structural explanation for the recognition of tyrosine-based endocytotic signals. Science 282, 1327-32.  152  Park, C. H., Valore, E. V., Waring, A . J. and Ganz, T. (2001). Hepcidin, a urinary antimicrobial peptide synthesized in the liver. J Biol Chem 276, 7806-10. Parkkila, S., Parkkila, A . K., Waheed, A., Britton, R. S., Zhou, X . Y . , Fleming, R. E., Tomatsu, S., Bacon, B. R. and Sly, W. S. (2000). Cell surface expression of H F E protein in epithelial cells, macrophages, and monocytes. Haematologica 85, 340-5. Parkkila, S., Waheed, A., Britton, R. S., Bacon, B. R., Zhou, X . Y . , Tomatsu, S., Fleming, R. E. and Sly, W. S. (1997a). Association of the transferrin receptor in human placenta with HFE, the protein defective in hereditary hemochromatosis. Proc Natl Acad Sci USA94,  13198-202.  Parkkila, S., Waheed, A., Britton, R. S., Feder, J. N . , Tsuchihashi, Z., Schatzman, R. C , Bacon, B . R. and Sly, W. S. (1997b). Immunohistochemistry of H L A - H , the protein defective in patients with hereditary hemochromatosis, reveals unique pattern of expression in gastrointestinal tract. Proc Natl Acad Sci USA 94, 2534-9. Patel, B . N . , Dunn, R. J. and David, S. (2000). Alternative R N A splicing generates a glycosylphosphatidylinositol-anchored form of ceruloplasmin in mammalian brain. JBiol Chem 275, 4305-10. Plow, E. F., Haas, T. A., Zhang, L . , Loftus, J. and Smith, J. W. (2000). Ligand binding to integrins. JBiol Chem 275, 21785-8. Plowman, G. D. (1986). Characterization and expression of the melanotransferrin (p97) gene. In Pathology, pp. 229. Seattle: University of Washington, Seattle. Poss, K . D. and Tonegawa, S. (1997a). Heme oxygenase 1 is required for mammalian iron reutilization. Proc Natl Acad Sci USA 94, 10919-24.  153  Poss, K . D. and Tonegawa, S. (1997b). Reduced stress defense in heme oxygenase 1-deficient cells. Proc Natl Acad Sci U S A 94, 10925-30. Pytela, R., Pierschbacher, M . D. and Ruoslahti, E. (1985). Identification and isolation of a 140 kd cell surface glycoprotein with properties expected of a fibronectin receptor. Cell 40, 191-8. Rabin, M . , McClelland, A . , Kuhn, L . and Ruddle, F. H . (1985). Regional localization of the human transferrin receptor gene to 3q26.2q. Am J Hum Genet 37, 1112-6. Raffin, S. B . , Woo, C. H . , 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. Raja, K . B . , Simpson, R. J. and Peters, T. J. (1992). Investigation of a role for reduction in ferric iron uptake by mouse duodenum. Biochim Biophys Acta 1135, 141-6. Ramalingam, T. S., West, A . P., Lebron, J. A . , Nangiana, J. S., Hogan, T. H . , Enns, C. A . and Bjorkman, P. J. (2000). Binding to the transferrin receptor is required for endocytosis of H F E and regulation of iron homeostasis. Nat Cell Biol 2, 953-7. Ramm, G . A., Crawford, D. H., Powell, L . W., Walker, N . I., Fletcher, L . M . and Halliday, J. W. (1997). Hepatic stellate cell activation in genetic haemochromatosis. Lobular distribution, effect of increasing hepatic iron and response to phlebotomy. J Hepatol 26, 584-92. Richardson, D. R. (2000). The role of the membrane-bound tumour antigen, melanotransferrin (p97), in iron uptake by the human malignant melanoma cell. Eur J Biochem 267, 1290-8.  154  Riedel, H . D., Remus, A . J., Fitscher, B . A . and Stremmel, W. (1995). Characterization and partial purification of a ferrireductase from human duodenal microvillus membranes. Biochem J 309, 745-8. Roberts, M . R., Han, Y., Fienberg, A., Hunihan, L . and Ruddle, F. H . (1994). A DNA-binding activity, T R A C , specific for the T R A element of the transferrin receptor gene copurifies with the K u autoantigen. Proc Natl Acad Sci U S A 91, 6354-8. Robinson, M . S. and Pearse, B . M . (1986). Immunofluorescent localization of 100K coated vesicle proteins. J Cell Biol 102, 48-54. Roeser, H . P., Lee, G. R., Nacht, S. and Cartwright, G. E. (1970). The role of ceruloplasmin in iron metabolism. J Clin Invest 49, 2408-17. Rose, T. M . , Plowman, G. D., Teplow, D. B., Dreyer, W. J., Hellstrom, K . E. and Brown, J. P. (1986). Primary structure of the human melanoma-associated antigen p97 (melanotransferrin) deduced from the mRNA sequence. Proc Natl Acad Sci USA 83, 1261-5. Rosenmund, A . , Kuyas, C. and Haeberli, A . (1986). Oxidative radioiodination damage to human lactoferrin. Biochem J240, 239-45. Rothenberg, B . E. and Voland, J. R. (1996). beta2 knockout mice develop parenchymal iron overload: A putative role for class I genes of the major histocompatibility complex in iron metabolism. Proc Natl Acad Sci U SA 93, 1529-34. Rothenberger, S., Food, M . R., Gabathuler, R., Kennard, M . L . , Yamada, T., Yasuhara, O., McGeer, P. L . and Jefferies, W. A . (1996). Coincident expression and distribution of melanotransferrin and transferrin receptor in human brain capillary endothelium. Brain Res 712, 117-21.  155  Rothenberger, S., Iacopetta, B. J. and Kuhn, L . C. (1987a). Endocytosis of the transferrin receptor requires the cytoplasmic domain but not its phosphorylation site. Cell 49, 423-31. Rothenberger, S., Iacopetta, B. J. and Kuhn, L . C. (1987b). Endocytosis of the transferrin receptor requires the cytoplasmic domain but not its phosphorylation site. Cell 49, 423-31. Rouault, T. A . (2001). Systemic iron metabolism: a review and implications for brain iron metabolism. Pediatr Neurol 25, 130-7. Rovati, G. E. (1993). Rational experimental design and data analysis for ligand binding studies: tricks, tips and pitfalls. Pharmacol Res 28, 277-99. Rovati, G . E. (1998). Ligand-binding studies: old beliefs and new strategies. Trends Pharmacol Sci 19, 365-9. Roy, C. N . , Blemings, K . P., Deck, K . M . , Davies, P. S., Anderson, E. L . , Eisenstein, R. S. and Enns, C. A . (2002). Increased IRP1 and IRP2 R N A binding activity accompanies a reduction of the labile iron pool in HFE-expressing cells. J Cell Physiol 190,218-26. Roy, C. N . , Penny, D. M . , Feder, J. N . and Enns, C. A . (1999). The hereditary hemochromatosis protein, HFE, specifically regulates transferrin-mediated iron uptake in HeLa cells. JBiol Chem 274, 9022-8. Roze-Heusse, A., Houbiguian, M . L., Debacker, C., Zakin, M . M . and Duchange, N . (1996). Melanotransferrin gene expression in melanoma cells is correlated with high levels of Jun/Fos family transcripts and with the presence of a specific API-dependent ternary complex. Biochem J318, 883-8.  156  Rugh, R. (1990). The Mouse: It's reproduction and development: Oxford University Press. Ruoslahti, E. (1996). R G D and other recognition sequences for integrins. Annu Rev Cell Dev Biol 12, 697-715. Rutledge, E. A . , Green, F. A . and Enns, C. A . (1994). Generation of the soluble transferrin receptor requires cycling through an endosomal compartment. J Biol Chem 269,31864-8. Sala, R., Jefferies, W. A., Walker, B., Yang, J., Tiong, J., Law, S. A., Carlevaro, M . F., Vacca, A . , Cancedda, R., Cancedda, F. D. and Ribatti, D. (2002). The human melanoma associated protein melanotransferrin promotes endothelial cell migration and angiogenesis in vivo. Submitted. Sassone-Corsi, P. (1994). Goals for signal transduction pathways: linking up with transcriptional regulation. Embo J13, 4717-28. Sayre, L . M . , Perry, G. and Smith, M . A . (1999). Redox metals and neurodegenerative disease. Curr Opin Chem Biol 3, 220-5. Schneider, C., Owen, M . J., Banville, D. and Williams, J. G. (1984). Primary structure of human transferrin receptor deduced from the mRNA sequence. Nature 311, 675-8. Sciot, R., de Vos, R., van Eyken, P., van der Steen, K., Moerman, P. and Desmet, V . J. (1989). In situ localization of melanotransferrin (melanoma-associated antigen P97) in human liver. A light- and electronmicroscopic immunohistochemical study. Liver 9, 110-9.  157  Sharp, P., Tandy, S., Yamaji, S., Tennant, J., Williams, M . and Singh Srai, S. K . (2002). Rapid regulation of divalent metal transporter (DMT1) protein but not mRNA expression by non-haem iron in human intestinal Caco-2 cells. FEBS Lett 510, 71-6. Sheth, S. and Brittenham, G. M . (2000). Genetic disorders affecting proteins of iron metabolism: clinical implications. Annu Rev Med 51, 443-64. Shih, Y . J., Baynes, R. D., Hudson, B. G., Flowers, C. H . , Skikne, B . S. and Cook, J. D. (1990). Serum transferrin receptor is a truncated form of tissue receptor. J Biol Chem 265, 19077-81. Sieweke, M . H . , Tekotte, H . , Frampton, J. and Graf, T. (1996). MafB is an interaction partner and repressor of Ets-1 that inhibits erythroid differentiation. Cell 85, 49-60. Sipe, D. M . and Murphy, R. F. (1991). Binding to cellular receptors results in increased iron release from transferrin at mildly acidic pH. J Biol Chem 266, 8002-7. Stearman, R., Yuan, D. S., Yamaguchi-Iwai, Y . , Klausner, R. D. and Dancis, A . (1996). A permease-oxidase complex involved in high-affinity iron uptake in yeast. Science 271, 1552-7. Stearne, P. A . , Pietersz, G. A . and Goding, J. W. (1985). c D N A cloning of the murine transferrin receptor: sequence of trans-membrane and adjacent regions. J Immunol 134, 3474-9. Stephens, P. E., Ortlepp, S., Perkins, V . C , Robinson, M . K . and Kirby, H . (2000). Expression of a soluble functional form of the integrin alpha4betal in mammalian cells. Cell Adhes Commun 7, 377-90.  158  Sundan, A . , Olsnes, S., Sandvig, K . and Pihl, A . (1982). Preparation and properties of chimeric toxins prepared from the constituent polypeptides of diphtheria toxin and ricin. Evidence for entry of ricin A-chain via the diphtheria toxin pathway. J Biol Chem 257, 9733-9. Thompson, A . M . and Thompson, G. C. (2001). Serotonin projection patterns to the cochlear nucleus. Brain Res 907, 195-207. Tiong, J. (2001). Metal Binding Properties of p97 and the Trafficking of a Novel Iron Internalizatiion Pathway by GPI-anchored p97. In Microbiology and Immunology and the Biotechnology Laboratory, pp. 205. Vancouver: University of British Columbia. Tiong, J., Tian, M . and Jefferies, W. (2002). A novel iron uptake pathway through caveolae by GPI-anchored p97. manuscript in preparation . Tiong, J. W. C. and Jefferies, W. A . (2002). Metal binding properties of melanotransferrin., manuscript in preperation. Tong, X . , Kawabata, H . and Koeffler, H . P. (2002). Iron deficiency can upregulate expression of transferrin receptor at both the m R N A and protein level. Br J Haematol 116, 458-64. Townsend, A . and Drakesmith, H . (2002). Role of H F E in iron metabolism, hereditary haemochromatosis, anaemia of chronic disease, and secondary iron overload. Lancet 359, 786-90. Triguero, D., Buciak, J. and Pardridge, W. M . (1990). Capillary depletion method for quantification of blood-brain barrier transport of circulating peptides and plasma proteins. J Neurochem 54, 1882-8.  159  Trinder, D., Macey, D. J. and Olynyk, J. K . (2000). The new iron age. Int J Mol Med 6, 607-12. Turnbull, A . , Cleton, F. and Finch, C. A . (1989). The Journal of clinical Investigation, Volume 41, 1962: iron absorption. IV. The absorption of hemoglobin iron. NutrRev 47, 51-3. Umbreit, J. N., Conrad, M . E., Moore, E. G. and Latour, L . F. (1998). Iron absorption and cellular transport: the mobilferrin/paraferritin paradigm. Semin Hematol 35, 13-26. van der Flier, A . and Sonnenberg, A . (2001). Function and interactions of integrins. Cell Tissue Res 305, 285-98. 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 intestinal iron transport, is defective in the sla mouse. Nat Genet 21, 195-9. Waheed, A., Parkkila, S., Zhou, X . Y., Tomatsu, S., Tsuchihashi, Z., Feder, J. N . , Schatzman, R. C , Britton, R. S., Bacon, B . R. and Sly, W. S. (1997). Hereditary hemochromatosis: effects of C282Y and H63D mutations on association with beta2microglobulin, intracellular processing, and cell surface expression of the H F E protein in COS-7 cells. Proc Natl Acad Sci USA 94,123 84-9. Wardrop, S. L . and Richardson, D. R. (1999). The effect of intracellular iron concentration and nitrogen monoxide on Nramp2 expression and non-transferrin-bound iron uptake. Eur J Biochem 263, 41-9. West, A . P., Jr., Bennett, M . J., Sellers, V . M . , Andrews, N . C , Enns, C. A . and Bjorkman, P. J. (2000). Comparison of the interactions of transferrin receptor and  160  transferrin receptor 2 with transferrin and the hereditary hemochromatosis protein HFE. J Biol Chem 275, 38135-8. Williams, A . M . and Enns, C. A . (1991). A mutated transferrin receptor lacking asparagine-linked glycosylation sites shows reduced functionality and an association with binding immunoglobulin protein. JBiol Chem 266, 17648-54. Woodbury, R. G., Brown, J. P., Loop, S. M . , Hellstrom, K . E. and Hellstrom, I. (1981). Analysis of normal neoplastic human tissues for the tumor-associated protein p97. Int J Cancer 27, 145-9. Woodbury, R. G., Brown, J. P., Yeh, M . Y., Hellstrom, I. and Hellstrom, K . E. (1980). Identification of a cell surface protein, p97, in human melanomas and certain other neoplasms. Proc Natl Acad Sci USA 77, 2183-7. Xiong, J. P., Stehle, T., Diefenbach, B . , Zhang, R., Dunker, R., Scott, D. L . , Joachimiak, A . , Goodman, S. L . and Arnaout, M . A . (2001). Crystal structure of the extracellular segment of integrin alpha Vbeta3. Science 294, 339-45. Yamada, T., Tsujioka, Y . , Taguchi, J., Takahashi, M . , Tsuboi, Y . , Moroo, I., Yang, J. and Jefferies, W. A . (1999). Melanotransferrin is produced by senile plaqueassociated reactive microglia in Alzheimer's disease. Brain Res 845, 1-5. Young, S. P., Bomford, A . and Williams, R. (1984). The effect of the iron saturation of transferrin on its binding and uptake by rabbit reticulocytes. Biochem J 219, 505-10. Y u , J. and Wessling-Resnick, M . (1998). Structural and functional analysis of SFT, a stimulator of Fe Transport. JBiol Chem 273, 21380-5.  161  Y u , J., Y u , Z. K . and Wessling-Resnick, M . (1998). Expression of SFT (stimulator of Fe transport) is enhanced by iron chelation in HeLa cells and by hemochromatosis in liver. J Biol Chem 273, 34675-8. Zhou, X . Y . , Tomatsu, S., Fleming, R. E., Parkkila, S., Waheed, A., Jiang, J., Fei, Y., Brunt, E. M . , Ruddy, D. A., Prass, C. E., Schatzman, R. C , O'Neill, R., Britton, R. S., Bacon, B. R. and Sly, W. S. (1998). H F E gene knockout produces mouse model of hereditary hemochromatosis. Proc Natl Acad Sci U S A 95, 2492-7. Zoller, H., Koch, R. O., Theurl, I., Obrist, P., Pietrangelo, A., Montosi, G., Haile, D. J., Vogel, W. and Weiss, G. (2001). Expression of the duodenal iron transporters divalent-metal transporter 1 and ferroportin 1 in iron deficiency and iron overload. Gastroenterology 120, 1412-9.  162  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items