UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A multifunctional protein : Phosphoglucose isomerase / autocrine motility factory / neuroleukin Y, Nathalie 2007

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

Item Metadata

Download

Media
831-ubc_2007-0290.pdf [ 9.69MB ]
Metadata
JSON: 831-1.0100830.json
JSON-LD: 831-1.0100830-ld.json
RDF/XML (Pretty): 831-1.0100830-rdf.xml
RDF/JSON: 831-1.0100830-rdf.json
Turtle: 831-1.0100830-turtle.txt
N-Triples: 831-1.0100830-rdf-ntriples.txt
Original Record: 831-1.0100830-source.json
Full Text
831-1.0100830-fulltext.txt
Citation
831-1.0100830.ris

Full Text

A multifunctional protein : Phosphoglucose isomerase / autocrine motility factor / neuroleukin by Nathalie Y B.Sc, Universite de Montreal, 2004 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Anatomy) THE UNIVERSITY OF BRITISH COLUMBIA April 2007 © Nathalie Y, 2007 II A B S T R A C T Phosphoglucose isomerase (PGI) is a glycolytic enzyme that moonlights as a cellular cytokine. The protein is also known as autocrine motility factor ( A M F ) , neuroleukin and maturation factor. P G I / A M F interaction with its receptor interaction is pH-dependent. Indeed, at neutral p H , P G I / A M F binds its receptor A M F R at the cell surface and can be endocytosed v ia two different pathways: caveolae/raft-dependent endocytosis to the smooth E R or clathrin-dependent endocytosis to multivesicular bodies (MVBs ) . Internalized P G I / A M F can recycle from M V B s to the plasma membrane where it can undergo further rounds of endocytosis and recycling. Recycl ing receptor-ligand complexes can also be sequestered via stable association with F N fibrils. Recent data show that, at acid p H , endocytosis is inhibited and P G I / A M F binds directly to F N fibrils or to HS . Heparan sulfate proteoglycans, when expressed on the surface of cells, modulate the actions of a large number of extracellular ligands while fibronectin is involved in many cellular processes such as tissue repair and cell migration/adhesion. However, the mechanisms that regulate P G I / A M F binding to its receptors still remain unclear. P G I / A M F cytokine activity, associated with several diseases, has been reported in rheumatoid synovial f luid and its deposition on synovial surfaces and ability to induce an autoimmune response in rheumatoid arthritis (RA) identified it as a possible autoantigen different from normal circulating P G I / A M F . However, more recent manuscripts have questioned the prevalence of an autoimmune response to PGI in R A . Ill In this study, recombinant PGI constructs were used to characterize PGI interactions and functions. We demonstrate that PGI behaves differently after N or C-terminal residue additions. Our data also suggest that monomerization but not enzymatic activity is necessary to induce cell motility at neutral p H . The putative function of PGI in R A was assessed and using the recombinant PGI constructs and PGI autoantibodies was found to be species and conformation-dependant. IV T A B L E O F C O N T E N T S pages Abstract II Table of contents IV List of tables VII List of figures VIII List of symbols and abbreviations X 1. Introduction 1 1.1 Historic 1 1.2 Molecular Biology of PGI 4 Gene structure of PGI 4 Gene 4 Minisatellites 4 Protein structure of PGI 5 Backbone structure 5 Active site 7 Interspecies homology 7 Mutations 7 1.3 Functions 8 Catalytic function o f PGI 8 Glycolysis 8 Active sites 9 Moonlighting functions of PGI 13 P G I / A M F / neuroleukin secretion 13 Neuroleukin 13 Autocrine motility factor and maturation factor 15 Involvement in mineralization during osteoblast differentiation. 19 Embryo implantation 20 1.4 Receptors 21 A M F R / gp78 22 Protein motifs implicated in A M F / P G I cytokine activity and receptor binding 24 Fibronectin 25 Heparan sulfate 26 IGFPB-3 27 Another receptor 27 1.5 P G I / A M F implication in diseases 28 Non-spherocytic haemolytic anaemia 28 Cancer 29 Rheumatoid Arthritis 30 V 2. Hypothesis 32 3. Mater ia l and methods 33 3.1 Protein purification 33 3.2 SDS-page and western blots 33 3.3 Enzymatic activity assay 34 3.4 Glutaraldehyde cross-linking assay 35 3.5 Circular dichroism 35 3.6 Fibronectin binding assay.... . 35 3.7 Cel l motility assay 36 3.8 Sera and synovial fluids 37 3.9 Human R A antisera E L I S A screening 37 3.10 Human R A antisera Western Blot screening 37 4. Results 39 4.1 Analysis of recombinant A M F / P G I expression and purification 39 4.2 Enzymatic activity of recombinant P G I / A M F 40 4.3 Recombinant P G I / A M F glutaraldehyde cross-linking 41 4.4 Circular dichroism of recombinant P G I / A M F 42 a. Far U V 42 b. N e a r U V 43 4.5 Binding to fibronectin 43 4.6 Recombinant cell-induced motility 44 4.7 Implication of A M F / P G I in Rheumatoid Arthritis 45 a. E L I S A essay 45 b. Western blot essay 46 5. Discussion 48 5.1 Conformational effects of residue additions to C-terminus and N-terminus. 48 5.2 Cell-induced motility and cell interaction of A M F / P G I 49 5.3 Implications for the Role of PGI in Rheumatoid Arthritis 51 6. Conclusion 53 VI 7. Figures, tables and legends 54 8. Bibliography 79 VII L I S T O F T A B L E S Table I Summary: Recombinant A M F / P G I properties 70 Table II Analysis of densitometry 75 VIII L I S T O F F I G U R E S 1. I N T R O D U C T I O N Figure 1 Structural organization of the human glucose phosphate isomerase gene 4 Figure 2 Ribbon representation of PGI from different species 6 Figure 3 Glycolysis pathway 8 Figure 4 Interconversion between glucose 6-phosphate and fructose 6-phosphate 9 Figure 5 Proposed catalytic mechanism for PGI 11 Figure 6 Molecular signaling in A M F / P G I motility stimulation 16 Figure 7 Interaction between A M F - A M F R and V E G F - V E G F R signal in tumor and host endothelial cells 17 Figure 8 Aptosis-related signal pathways induced by A M F / P G I overexpression. 18 Figure 9 The complex biology of P G I / A M F and its receptor 21 Figure 10 Structure of AMFR/gp78 22 Figure 11 Increased association of A M F / P G I to fibronectin at acid p H 26 4. R E S U L T S Figure 12 Vector and Constructs 55 Figure 13 Western blot and SDS-page analysis of recombinant A M F / P G I 57 Figure 14 Enzymatic activity of recombinant A M F / P G I 59 Figure 15 Glutaraldehyde cross-linking and semi-log graph analysis 1 61 Figure 15 Glutaraldehyde cross-linking and semi-log graph analysis II 62 Figure 16 Circular dichroism of recombinant A M F / P G I 64 Figure 17 Binding of P G I / A M F to dimeric F N at neutral and acid p H 66 IX Figure 18 Recombinant A M F / P G I cell-induced motility 68 Figure 19 Human R A antisera E L I S A screening 71 Figure 20 Western Blot control 73 Figure 21 Western Blot screening: species-specific recognition of human R A anti-sera to P G I / A M F 75 Figure 22 Western Blot screening: conformation-specific recognition of human R A anti-sera to recombinant P G I / A M F 77 L I S T O F S Y M B O L S A N D A B B R E V I A T I O N S AMF Autocrine motility factor AMFR Autocrine motility factor receptor ATP Adenosine triphosphate BSA Bovine Serum Albumine CD Circular Dichroism cDNA Complementary D N A ECM Extracellular matrix ER Endoplasmic reticulum ERAD Endoplasmic reticulum associated degradation F6P Fructose-6-phosphate FN Fibronectin G 6 P Glucose-6-phosphate GDI GDP-dissociation inhibitor GTP Guanosine triphosphate HS Heparan sulfate IGF Insulin growth factor IGFBP Insulin-like growth factor binding protein k D a Ki lo Daltons KDR kinase domain region mpCD methyl beta-cyclodextrin mRNA messenger R N A MVB Multivesicular bodies NAD Nicotamine adenine dinucleotide PGI Phosphoglucose isomerase RA Rheumatoid arthritis RBC Red blood cells SDS Sodium dodecyl sulfate Tfr Transferrin UV Ultraviolet VEGF Vascular endothelial growth factor 1 1 . I N T R O D U C T I O N 1.1 - HISTORY Phosphoglucose isomerase is a glycolytic enzyme present in all types of human cells. It had been studied for decades, but its molecular structure and the biological understanding of its numerous functions have emerged only during the past 20 years. Significant technical advances have led to the discovery of its structure, its identity as an extracellular cytokine and its receptor. The human enzyme is of medical interest because PGI is believed to be involved in several diseases. The interest of P G I / A M F and its receptor is growing not only because of its biological significance, but also because of its practical importance as a major target in the post-genomic era for developing therapeutics for diverse diseases. Glycolysis Glycolysis has been studied for over a century. This degradation pathway of sugar molecules leads to the formation of pyruvate releasing energy in the form of A T P and molecules with reduction potential (NAD) [1]. Initially described by Lohman in 1933, phosphoglucose isomerase (PGI) is known to be a ubiquitous cytosolic enzyme that catalyzes the interconversion between glucose-6-phosphate and fructose-6-phosphate during the second step of glycolysis [2]. Neuroleukin, lymphokin and phosphoglucose isomerase In response to partial denervation or paralysis, new neurotic processes, called terminal sprouts, can appear from the remaining motor axon terminal,[3]. In their attempt to identify factors produced by denervated or inactive muscle that might be necessary for 2 motor axon terminal sprouting, Gurney and colleagues purified in 1986 a ~56 k D protein, neuroleukin. Neuroleukin was subsequently shown to be capable of increasing survival of cultured sensory neurons [4] and, as they found later that year, to act as a lymphokine. Fol lowing secretion by lectin-stimulated T cells, neuroleukin was able to induce the maturation of B-cells into antibody secreting cells but was not necessary involved in the continued production of immunoglobulin by differentiated antibody-secreting cells [5]. Two years later, in 1988, the mouse phosphoglucose isomerase (PGI) c D N A was isolated and sequenced. The investigators, a group working on the molecular genetics of carbohydrate metabolism, surprisingly found a whole sequence identity between the 759 nucleotides at the 3 ' end of the mouse PGI clone and the sequence of mouse neuroleukin [6]. Moreover a second group showed that same year a 90% homology between PGI and neuroleukin [7]. Hence, the question as to how a ubiquitous, cytosolic enzyme could also function as an extracellular cytokine had been raised. Gurney et al. immediately reacted to this discovery and subsequently confirmed that both mouse and human neuroleukin expressed PGI enzymatic activity. However, they found that PGI activity was not blocked with monoclonal antibodies that were able to block neuroleukin activities. Thus, they hypothesized for the first time the existence of a PGI/neuroleukin receptor [8]. Autocrine motility factor, maturation factor and phosphoglucose isomerase Several steps are involved in the progression of a tumor, which include unrestrained growth and invasive behaviour or active locomotion of tumor cells is also one of its major properties. Since autocrine growth factors are essential to unrestrained growth of tumor cells, it was therefore suggested that tumor cell motility might also be regulated by autocrine mechanisms. This led to the discovery of the autocrine motility in 1986, by 3 Liotta et al. The group showed the presence of a cell motility-stimulating factor present in the serum-free conditioned medium of human melanoma cells. The ~55kDa motility factor, termed 'autocrine motility factor ( A M F ) ' , was believed to play a major role in the local invasive potential of tumor cells [9]. The hypothesis was later verified when A M F found in urine from patients with bladder cancer was shown to induce motility of transitional cell carcinoma from urinary bladder in a similar way to that of their own serum-free conditioned medium [10]. A M F sequence and structure remained unknown until the mid 90's. In 1996, Watanabe et al., working on A M F , microsequenced the tumor-secreted cytokine from a murine fibrosarcoma. They demonstrated that A M F corresponded in fact to the known enzyme and cytokine PGI/Neuroleukin and that A M F exhibited enzymatic/cytokine properties of PGI/neuroleukin which were inhibited by specific PGI inhibitors [11]. That same year, the identity of a maturation inducer capable of mediating the differentiation of human myeloid leukemic cells to terminal monocytic cells was investigated by X u et al. They also surprisingly found that the 54.3 kD inducer had PGI enzymatic activity and that its amino acid sequence was 100% homologue to neuroleukin and PGI [12]. Moreover, novel properties of PGI have been discovered more recently and the ubiquitous enzyme has been associated with functions such as involvement in mineralization during osteoblast differentiation [13] or embryo implantation [14], providing a possible hint on the evolutionary development of its original function as a cytokine. Several glycolytic enzymes serve multiple functions [15] and phosphoglucose isomerase and its extracellular homologues represent a brilliant example of the evolution of new functions from existing proteins. 4 1.2 - MOLECULAR BIOLOGY OF P G I GENE STRUCTURE Gene The gene encoding human phosphoglucose isomerase is located on chromosome 19. It is about 50 kb long and contains 18 exons and 17 introns [16]. Chr.;.19:' P t 3 i l 2 p I 3 . i i » •.*,«12»:...^.--w q l 9 J l ^ ^ l 3 i i 2 ^ q l J ' . S ^ » « 1 3 ; 3 2 8 « » . 3 : * l 3 .> 4 i s l 3 . « t « 1 3 ; M l Size (bp) 122 91 69 120 84 146 72 45 54 61 44 153 130 77 129 76 67 431 Exon 1* 2* 3 4 S 6 7 8 9 10 11 12 13 14 15 1617 18 ItltrOn Size Ikb" MBip 1.6kb ? 1 « b p 1.6«> 812bp1.HUi ?<!Bbp141bp I.Skb 15Qbp 2,Ekb 22Dbp 87bp 112bp X u e t a l . , 1995 Figure 1 S t ruc tu ra l o rgan iza t ion of the h u m a n g lucose phospha te i somerase gene . Minisatel l i tes Minisatellites are tandemly repeated D N A sequences found throughout the genomes of all eukaryotes. The repeat unit sequence is generally not conserved beyond closely related species [17]. Wil l iams and al. [18] have studied the minisatellite contained in the intron 9 of the human PGI and have found similar repeats in PGI of other species. Moreover, these repeat units did not appear in other locus of the genome. Minisatellite D N A has been reported to be involved in recombination activity, control of gene expression of nearby gene(s) (transcriptional and translational), whereas others form protein coding regions [19]. Thus, the high level of conservation exhibited by the GPI minisatellite, coupled with the unique location, strongly suggests a functional role for these repeated D N A sequences. 5 Protein structure Backbone structure Whi le the extracellular cytokine activities have been associated with a 55 kd monomer, the dimeric structure of PGI was found to comprise two identical subunits, each of molecular mass 63 kDa [20]. PGI dimerization is required for its enzymatic activity because the active site of the enzyme is composed of polypeptide chains from both subunits [21,22]. Although the structural basis for the cytokine activity of the protein is not known yet [21], inhibitors of enzymatic activity are also capable of inhibiting the extracellular activites of P G I / A M F [11,12]. Human PGI was resolved by Read et al. in 2001. Crystallization of human PGI revealed that the enzyme is a tight dimer of identical subunits. Each monomer is composed of two globular domains, and the close association between them is reinforced by the presence of an 'arm' , a 45 residue extension structure at the C terminus which wraps around the other monomer, while, on the opposite site, another loop termed 'hook' also interacts with the adjacent monomer. The two domains are historically named large and small, although they are now known to be pretty similar in size. Both domains consist of an ahelix-Psheet-ahelix sandwich. The small domain contains a central five stranded parallel |3-sheet surrounded by helices whereas the large domain has a six-stranded mixed parallel/antiparallel p-sheet, also packed on both sides by a-helices. The polypeptide chain begins in the large domain, crosses to the small domain, then returns to the large domain and finishes at the end of the oc-helical arm [23]. 6 Rabbit PGI monomer 2.5A Davies & Muirhead, 2002 ~ 520" Largs ec domain I a21 a15l Small domain Rabbit PGI dimer Davies & Muirhead, 2002 Human PGI monomer L 6 A Read et al. , 2001 Large cfcjffiaih «tomai Bacillus stearothermophilus PGI monomer 2.3A Sun etal . , 1999 Figure 2 Ribbon representation of PGI from different species 7 Active sites For a better understanding of the relationship of PGI enzymatic function and structure, please see section C of the introduction. Interspecies homology To date, the pig [24], human [20], rabbit [25] and bacterial (Bacillus stearothermophilus) [21] PGI have been crystallized and their structure fully characterized. The studies showed that the overall structure of PGI in most species is very similar. A lso , the amino acids required in the active site have notably been shown to be conserved in all known PGI sequences in mammals, plants, flies, bacteria, and yeast [20,22,24]. Mutations in the human PGI PGI is an essential enzyme and PGI deficiency in humans is an autosomal recessive genetic disorder resulting in nonspherocytic haemolytic anaemia [26]. Mutations in PGI are homozygotes and can be classified into three groups: (a) those that impact the precise structure of the enzyme (b) those that disrupt or alter a dimer-dimer contact, and (c) those of residues at the active site, which may have a role in catalytic function, This disease w i l l be covered later on (Section D). 8 1.3 - FUNCTIONS CATALYTIC FUNCTIONS OF P G I Glycolysis Glycolysis is a degradation pathway of sugar molecules that leads to the formation of pyruvate. The breakdown of sugars also releases energy in the form of A T P and some reduction potential molecules [1]. Nine distinct reactions are required to convert glucose into pyruvate. Glucose ATP ADP hexokinase glucokinase Glucose-6-phosphate phosphohexose isomerase Fructose-6 -phosphate ATP- phospho-fructokinase-1 ADP' Fructose-1,6-bisphosphate aldolase Glyceraldehyde-3 -phosphate -Di hydro xyacetone triosephosphate phosphate isomerase M.W. King (1996) [112] Figure 3 G lyco lys i s pa thway The second step of glycolysis involves the conversion of glucose-6-phosphate (G6P) to fructose-6-phosphate (F6P) by phosphoglucose isomerase and as the name suggests, involves an isomerization reaction, leading to the interconversion between G6P and F6P. 9 The reaction is freely reversible at normal cellular concentrations of the two substrates and PGI is thus also used during gluconeogenesis [27]. The proposed multistep catalytic mechanism for PGI consists in ligand binding, ring opening, isomerization of the substrate, ring closing and ligand release. The isomerization involves an acid/base catalysis, where a rearrangement of the carbon-oxygen bond transforms the six-membered ring (G6P) into a five-membered ring (F6P) or vice-versa [28]. During glycolysis, the rearrangement takes place when the six-membered ring opens, v ia an enediolate intermediate, and then closes in such a way that the first carbon becomes now external to the ring. Phosphoglucose isomerase o II 8 -O— P — O — C H 2 f J Glucose 6-phosphat© Fructose 6-phosphate A G ' 0 = 1.7 kJ/mol Mathews A/an Holde/Ahern, 3rd Ed [27] Figure 4 In te rconvers ion be tween g lucose 6 -phospha te and f ruc tose 6 -phospha te Act ive sites The association between the monomers forms the putative active site for PGI's enzyme function in glycolysis and gluconeogenesis. The cleft between the large domain, small domain, and C-terminal arm forms a deep binding pocket in the dimer. This pocket is the 10 postulated binding site for glucose 6-phosphate in glycolysis and fructose 6-phosphate in gluconeogenesis. The catalytic mechanism of phosphoglucose isomerase should include the following steps: binding of the cyclic form of the substrate to the enzyme; ring-opening of the substrate; base-catalyzed isomerization v ia a cis-endiol intermediate; ring-closure of the product and release of product [29]. The following sketch represents the most recent isomerization mechanism as proposed by Jeffery and Lee in 2005 [28]. 5 JT V 1 It LB A f » 11 O U T J DOWN Figure 5 The proposed catalytic mechanism for PGI as proposed by Lee and Jeffery Lee and Jeffery, 2005 [28] 12 The proposed multistep catalytic mechanism is shown, including the ligand binding, ring opening, isomerization, ring closing, and ligand release steps. A helix containing amino acid residues 512-520 moves between an " i n " position, in which it interacts directly with the bound ligand, and an "out" position, in which there is a water molecule located between Lys518 and the ligand. A loop containing amino acid residues 210-214 moves "up" to interact with the phosphoryl group of the ligand upon ligand binding and "down" upon release of the ligand. The numbering of the steps refers to the F6P to G6P direction of catalysis. Amino acid residues involved in the ring-opening step are shown in green. Amino acid residues involved in the isomerization step are shown in blue. The cyclic and open chain forms of the substrates are shown in red. The dashed lines indicate hydrogen bond interactions. (1) Ligand binding: The F6P substrate binds in the active site, and the 210-214 loop moves "up." (2) Ring opening: His388 and a water molecule held by Lys518 and Thr214 catalyze the ring opening step. (3) Conformational changes: The substrate undergoes rotation about its C 3 - C 4 bond and extends so that C 1 - C 2 approaches Glu357. The 512-520 helix moves to the " i n " position. A n ordered water molecule located between the 512-520 helix and Lys518 is lost. (4) Isomerization: Glu357 abstracts a proton to yield the cis-enediol(ate) intermediate. (5) Glu357 transfers a proton to the intermediate to yield the open chain form of G6P. (6) Conformational changes: The 512-520 helix moves to the "out" position and an ordered water molecule inserts between the helix and the open chain form of G6P. The G6P undergoes rotation about its C 3 - C 4 bond to approach its cyclic conformation. (7) R ing closure: His388, Lys518, and Thr214 assist in ring closure. (8) Product release: The G6P product is released, and the 210-214 loop moves "down". 13 Moonlighting functions of PGI PGI/AMF/neuroleukin secretion One important feature of PGI/AMF/neuroleukin sequence is the lack of a hydrophobic signal sequence suggesting that the protein is not a classical endoplasmic reticulum/Golgi-dependent secretory protein [9,30]. P G I / A M F is predominantly secreted from some types of tumor cells [9,30,31] or T cells stimulated with lectins [5]. It is not clear yet how AMF/neuroleukin is secreted but it apparently follows the non-classical secretory pathway described for other cytosolic proteins such as galecting-3 [113] and F G F [114,115]. However, overexpression of P G I / A M F in normal or non-AMF-secreting tumor cells was able to induce its secretion [32,33] and that cells secreting P G I / A M F express higher PGI m R N A levels to that of normal cells [34]. A lso, some studies suggest that phosphorylation by casein kinase II is associated with the enzyme's secretion [35]. Neuroleukin, and phosphoglucose isomerase The formation of neuronal sprouts, fine neurotic processes, either from synaptic terminals or nearby nodes of Ranvier, is a widely known form of plasticity of motoneurons. A t least four cytokines or growth factors are believed to be involved in motoneuron sprouting, each of which using a distinctive signaling pathway. One of those cytokines is neuroleukin, shown previously to be the ubiquitous enzyme phosphoglucose isomerase [36]. The original biological role of sprouting was derived from the results of partial muscle denervation experiments. In an attempt to identify factors sufficient to induce motor axon terminal sprouting, Gurney et al. produced a polyclonal antisera [37] and monoclonal antibodies [8] against a 56kD protein. Mono and poly-clonal antibodies to 14 the protein derived from denervated or inactive muscle were capable of certain suppression of botulinum toxin-induced motor axon terminal sprouting. The 56kD protein was named neuroleukin. Gurney remarked that antiserum made against neuroleukin only suppressed about half of the sprouting but did not investigate further. Act ing as a neutrophic factor, neuroleukin promotes survival and sprouting at the neuromuscular junction and also plays a role in motor neuron regeneration in vivo and in the survival of peripheral and central neurons in vitro. In addition to Gurney's studies, more recent studies showed that neuroleukin is upregulated during Huntington's disease, a neurological disorder, and that inhibition of neuroleukin expression potentiates the induction of cell death in P C 12 cells, indicative of its protective role in neuronal cells [38]. T cells can be induced to secrete several factors. For example, stimulation with lectins stimulates T cell secretion of neuroleukin and increased in neuroleukin m R N A [5]. A lso , when added to cultured human peripheral blood mononuclear cells, recombinant neuroleukin induces B cell secretion of antibody. It has not been clearly shown whether neuroleukin acts directly on B lymphocytes, in an autocrine manner to stimulate the T-cell itself or even i f it was a T-cel l product capable of amplifying monocyte functions. Unfortunately, due to its identification as the enzyme phosphoglucose isomerase, interest in neuroleukin involvement in neuronal sprouting and B cell antibody secretion diminished, although some papers have recently cited neuroleukin in Huntington's disease [38]. Autocrine motility factor, maturation factor and phosphoglucose isomerase Tumorigenesis is a multi-step process during which cells acquire a characteristic set of properties which allow them to bypass the normal mechanisms of cellular growth control. Although the genetic basis of tumorigenesis can vary greatly, the steps required for metastasis are similar for all tumor cells. The various steps in the process of metastasis are angiogenesis, the attachment of tumor cells to other cells or matrix proteins, the invasion of the tumor cells and finally, the colonization of the secondary site by tumor [39]. Invasion of cancer cells into surrounding tissue and the vasculature is an initial step in tumor metastasis. This requires migration of cancer cells and can be stimulated by several factors. Through a variety of mechanisms, motility factors may cause one or more of the following which contribute to motility: changes in cell shape, cytoskeletal rearrangements, and changes in cell adhesion and/or membrane fluidity [40]. One of these motility factors is autocrine motility factor, also known as the enzyme phosphoglucose isomerase. Unl ike associated neuroleukin or T cells activities, the autocrine activations mechanisms of PGI in cancer cell motility have been assessed in detail. 16 . , 2 0 0 4 [49 ] Figure 6 Molecular signaling in AMF/PGI motility stimulation Members of the Rho family, such as Ras or Rho-like GTPases, induce changes in the actin cytoskeleton, an important step in cell motility. Studies showed that AMF/PGI stimulation of cell motility of some tumor cells via its receptor A M F R activates small Rho-like GTPase, Racl and RhoA but not Cdc42 in a time- and dose-dependent manner in human malignant melanoma cells [41]. In response to AMF/PGI , protein levels of JNK1 and JNK2, two M A P kinases working downstream of Rho-like GTPases, are upregulated, leading to actin fiber rearrangement and formation of heavy bundles of stress fiber-like structures transversing the cells. Expression of GDI-p, a Rho GTPase regulatory protein, is enhanced following stimulation of the tumor cell by AMF/PGI [33]. However, the role of GDI-P in invasion and metastasis is controversial. Upregulation of the regulatory protein have been associated with progression of ovarian carcinoma tumors [42] while GDI-p has been reported as an invasion and metastasis suppressor in bladder cancers [43-45]. 17 Upregulation of GDI-p might be induced as a negative signal in the putative feedback mechanisms against excess signals from AMF/PGI. Tumor cell Endothelial cell (autocrine) (paracrine) Angiogenesis Yaganawa et a l . , 2004 [49] Figure 7 In teract ion be tween A M F - A M F R and V E G F - V E G F R s igna l in the t u m o r and host endothe l ia l cel ls V E G F is an important signaling protein involved in angiogenesis and affects endothelial cells specifically [46]. V E G F is known to act on endothelial cell mitogenesis and migration via two tyrosine-phosphorylating receptors, fms-like tyrosine kinase (Flt-1), receptor for VEGF-1 [47] and KDR, receptor for VEGF-2 [48]. To better understand the A M F / A M F R stimulation pathway, crosstalk between A M F - A M F R and V E G F - V E G F R signals have been assessed. Secreted AMF/PGI was shown to stimulate the host tumor cell in an autocrine manner and to enhance the production of V E G F . Endothelial cells exposed to AMF/PGI secreting cells were shown to augment expression of Flt-1 but not of KDR, suggesting that proliferative signals of V E G F in endothelial cells depend on K D R while migrational activities depend on Flt-1. PKC and PI3K inhibitors were shown to inhibit Flt-1 expression induced by AMF/PGI , indicating that Flt-1 expression is dependent on activation of these two kinases [49]. 18 Yaganawa et a l . , 2004 [49] Figure 8 Aptosis-related signal pathways induced by AMF/PGI overexpression AMF-expressing cells were shown to be resistant to apoptosis induced by serum deprivation or by mitomycin-C. A M F / P G I signals can activate P I3K which activates A k t / P K B , which in turn inactivates the proapoptotic protein B A D and caspase-9, leading to suppression of apoptosis induced by serum deprivation [50]. A M F / P G I overexpression can also suppress the expression of Apaf-1 and caspase-9 which are important for apoptosis initiation, activating PI3K and M A P K and causing mitomycin-induced apoptosis resistance [51]. There is a possibility that increased PGI enzymatic activity may cause hyper-metabolism of glucose, an activity related to malignant tumors, thereby affecting the apoptosis pathways [49]. 19 Involvement in mineralization during osteoblast differentiation A M F / P G I has been identified as a key functional molecule in osteoblast differentiation, a multistep process that involves critical spatial and temporal regulation of cellular processes marked by the presence of a large number of differentially expressed molecules. Zh i et al. showed that A M F / P G I m R N A is temporally expressed during MC3T3-E1 osteoblast-like cell line cell differentiation and their studies revealed the presence of A M F / P G I in MC3T3-E1 cells as wel l as in the surrounding matrix, suggesting secretion of the protein. In addition, A M F R was detected primarily on the cell membrane. A M F / P G I was expressed at a high level in osteoblasts and superficial articular chondrocytes o f young mice, in fibroblasts and in proliferating chondrocytes. However, PGI expression was very low in fully differentiated bone cells such as hypertrophic chondrocytes or osteocytes. Treatment of MC3T3-E1 cells with 6-phosphogluconic acid, an A M F / P G I inhibitor [11,52], resulted in reduction in alkaline phosphatase activity and mineralization in MC3T3-E1 cells, especially during the matrix formation stage of differentiating cells [13]. The investigators showed specific expression of A M F / P G I in discrete populations of bone and cartilage cells, suggesting a possible role for this secreted protein in bone development and regeneration. 20 Embryo implantation and phosphoglucose isomerase Implantation is the first stage in development of the placenta. The role of implantation is to obtain very close apposition between embryonic and maternal tissues. There are substantial differences among species in the process of implantation, particularly with regard to "invasiveness," or how much the embryo erodes into the maternal tissue. However, current understanding of embryo implantation in carnivores is limited [14]. A group from Illinois discovered that a 60 kDa protein was necessary in embryo implantation in the domestic ferret. They later identified this protein as being phosphoglucose isomerase [14]. This discovery demonstrates an uncharacterized endocrine function o f the protein. This role may represent the natural motility-stimulating activity of PGI , a characteristic later acquired by tumor cells. 21 1.4 - Receptors AMF/PGI interacts with the cell through several receptors. A c i d P H Neutral pH <w " \ 7 Fibronectin / \ I&FBP-3 • PGI/AMF • AMF-R Mitochondria-Multivesicular associated endosome s m o o t h E R Lagana et al, 2005 Figure 9 The complex biology of PGI/AMF and its receptor. The figure above summarizes AMF/PGI extracellular interactions with the cell. There are 6 proposed pathways. Following AMF/PGI binding, A M F R is internalized via a clathrin-independent pathway to the smooth endoplasmic reticulum tubules (1) or via another route involving a clathrin-dependent pathway to the multivesicular bodies (MVB) or endosomes (2). Following clathrin-mediated uptake to M V B s , AMF/PGI can recycle to fibronectin fibrils (3-5). Under acidic pH conditions, endocytosis of AMF/PGI is inhibited and AMF/PGI binds directly to fibronectin fibrils (7) or to heparan sulfate (8). Finally, at neutral pH, AMF/PGI can also interact with the insulin-like growth factor binding protein 3 (IGFBP-3) (6). 22 AMFR/gp78 In 1990, Nabi et al. identified the receptor for A M F / P G I ( A M F R ) , demonstrating that B16-F1 melanoma cells expressed augmented glycosylation of a 78kDa glycoprotein (gp78) in response to cell shape modulation that correlated with an increased metastatic ability in vivo and motility in vitro [53]. The structure of A M F R was determined in 1999 to be a seven-transmembrane receptor, also called G-protein-coupled receptor, which contains a RING-type zinc finger [54]. M U ,, Extracellular daman NH; 0®©® • Domaina CUE © Domaina da liaison dala CaM O Domaina RING O Domains Leucine zipper O Sites de phosphorylation de CK2 © Sites de phosphorylation de PKC • Sitesde phosphorylation de TK • Recouvrement des domaines RING et CK2 • Recouvrement des domaines RINGetPKC O Recouvrement desdomaines CaM et CK2 Tran«nambr« i & ,=.©< domotnc .®& : 9fi»: •eta °<&® R I N G ®®©©®(D®®©®aa«®©©!B®©®»® ®€>© CUE ——————uS 99— # # a > « a e e @ ® B® COO H Figure 10 Structure of AMFR/gp78. AMFR/gp78 is a specific receptor for A M F / P G I and is an integral membrane protein that can be found in the endoplasmic reticulum (ER). This receptor can target both itself and other proteins including C D 3 D and A P O B for proteasomal degradation. Interestingly, the receptor possesses ubiquitin ligase activity. As a matter of fact, AMFR/gp78 is a R I N G finger-dependent ubiquitin protein ligase (E3) of the endoplasmic reticulum, 23 suggesting a potential l ink between ubiquitylation, ER-associated degradation ( E R A D ) , and metastasis [55]. AMFR/gp78 also contains a Cue domain, another ubiquitin-binding domain, which can interact directly to mono and poly-ubiquitin to promote proteasomal degradation and maybe vesicular trafficking [56]. The A A A ATPase p97 /VCP complex dislodges ubiquitin proteins from the E R and chaperones them to cytosol for proteosomal degradation. It has been reported recently that AMFR/gp78 can interact with the A A A ATPase p97 /VCP complex and that the interaction was enhancing p97 /VCP polyubiquitin-association. The investigators speculated that as a multiple membrane-spanning protein, AMFR/gp78 can form a channel for retrotranslocation [57] which might represent one way of coupling ubiquitination with retro translocation and degradation of E R A D substrates. In higher eukaryotes, p97 is bound to the E R membrane by a membrane protein complex containing Derlin-1 and VCP-interacting membrane protein (V IMP) . How the ubiquitination machinery is recruited to the p97/Derl in/VFMP complex is unclear. It was reported that p97 interacts directly with several ubiquitin ligases, such as AMFR/gp78 , and facilitates their recruitment to Derl in-1 [58]. Upon binding of its ligand A M F / P G I , A M F R was found to be internalized through two different pathways. Studies showed that A M F R can be localized to a smooth subdomain of the E R [59,60]. A M F - A M F R uptake to the smooth E R tubules in NIH-3T3 fibroblasts is sensitive to the cholesterol extracting reagent methyl-P-cyclodextrin (m(3CD), inhibited by the dynamin-1 K 4 4 A mutant and negatively regulated by caveolin-1 [61,62]. Thus, A M F - A M F R uptake to the E R suggests a caveolae or caveolae-like structure-mediated endocytic pathway. Studies describing A M F - A M F R uptake through a second pathway, 24 the clathrin-mediated pathway, have shown that the endocytosis was not totally inhibited by m(3CD, but that the complex receptor-ligand can be internalized to punctate structures in the cells showed by colocalization with L A M P - 1 but not T fR to be multivesicular bodies (MVBs) [62]. A M F - A M F R internalization to the M V B s has also been found to be recycled to the fibrils of fibronectin[62]. The nature and the reasons behind A M F - A M F R recycling and binding to fibronectin remain to be elucidated. It has been proposed that A M F activation of the A M F R recycling pathway could be actively involved in the remodeling of the fibronectin E C M of motile cells by regulating fibri l formation or turnover [63]. Protein motifs implicated in A M F / P G I cytokine activity and receptor binding In contrast to the well-characterized active site necessary for the catalytic activities of PGI , the sites involved in the cytokine functions of A M F / P G I have not yet been studied in details yet. A recent report provided some insights in A M F / P G I structure needed for cytokine functions [74]. The investigators demonstrated that mutation of residues in the wild-type human A M F / P G I that interact with the phosphate group of PGI substrates and mutations within the C-terminal region significantly reduced cell motility-stimulating activity and A M F R binding than that in the wild-type human A M F / P G I . The report further showed that mutant A M F R lacking the putative N-sugar chain attachment site was expressed on the cell membrane but did not respond to AMF/PGI-st imulat ion, and that N -glycosidase-treated A M F R did not compete with receptor binding of A M F / P G I . However, there results imply that the unique N-glycosylation site of A M F R is extracellular. Such topology would require an intracellular N-terminus and an extracellular C-terminus of A M F R . This is both inconsistent with the classical topology 25 of seven-transmembrane receptors [116], and would further argue that the long loops forming the C-terminus of A M F R and containing the Cue and Ring domains involved in the E3 ubiquitin ligase function of A M F R is extracellular (Figure 10). Fibronectin The extracellular matrix provides a framework for cell adhesion supports cell movement and serves to compartmentalize tissues into functional units. Fibronectin is an essential component of many extracellular matrices where it regulates a variety of cell activities through direct interactions with cell surface integrin receptors [64]. In addition to integrins, fibronectin also binds extracellular matrix components such as collagen, fibrin and heparin [65]. As discussed previously, internalized A M F / P G I can be recycled to the fibrils of fibronectin through the clathrin-mediated pathway [62]. In addition, A M F / P G I can interact directly with fibronectin under acidic conditions, corresponding to a change in PGI tertiary structure [66,67]. A M F / P G I recycling from M V B s did not increase cell motility but A M F / P G I sequestration by fibronectin at acid p H shows an increase in cell motility [67]. Therefore, sequestration of recycling A M F / A M F R complexes by association with fibronectin fibrils may regulate the extent of ligand-dependant A M F R signaling while A M F / P G I interacting directly with fibronectin at acid p H may be used to facilitate internalization after subsequent release upon pH. 26 Contro l pH 7.0 pH 6.5 I • pH 6.0 x jf>R 6.sJf J / -Amraei et al.. 2003 Figure 1 1 Stronger association of A M F / P G I to fibronectin at increasingly acid pH . Heparan sulfate Heparan sulfate (HS) is a sulfated polysaccharide that has a very large structural diversity. It is estimated that up to 48 different disaccharides can occur in heparan sulfate, although only 23 have been detected in vivo. Nevertheless, its structural heterogeneity allows HS to interact with a wide range of functionally diverse proteins, such as growth factors, cytokines, chemokines, proteases, lipases and cell-adhesion molecules [68]. Although fibronectin has two defined HS-binding sites [69], presence of HS does not enhance A M F / P G I binding to fibronectin fibrils but selectively increases A M F / P G I cellular binding under acidic conditions. HS therefore mediates fibronectin -independent interaction of A M F / P G I with the cell under acidic conditions [67]. Why A M F / P G I binding to fibronectin but not to HS upon acidic pH increases cell motility is still unclear. 27 IGFBP-3 Insulin-like growth factor binding proteins ( IGFBPs) proteins are regulators of insulin growth factors (IGFs). Significant data showed that IGFBPs play important roles in addition to their ability to modulate IGFs. IGFBP-3 is a 266 amino acid protein [70] which can variably be N-glycosylated [71] and is mainly secreted by the liver and to a lesser extent in fibroblasts, ovaries and placenta [72]. It has recently been shown that IGFBP-3 interacts with A M F / P G I , inhibiting A M F / P G I functions [73]. In their studies, Mishra and colleagues showed that IGFBP-3 was able to inhibit catalytic activities of A M F / P G I as well as its ability to induce migration of breast cancer cells. Moreover, A M F / P G I was capable of inhibiting IGFBP-3-induced cell apoptosis. Although A M F / P G I did not bind IGF/ IGFBP-3 , that binary complex seemed to be a more potent inhibitor of A M F / P G I than IGFBP-3 alone, suggesting a conformation change of IGFBP-3 after its binding to the IGF. Therefore, the results showed suggested a potential ability of IGFBP-3 to disrupt the interaction of A M F / P G I with its receptor A M F R . Another receptor The human acute monocytic leukemia line does not express gp78 and its motile activity is not enhanced by A M F / P G I though it is wel l differentiated by A M F / P G I exposure. Forced expression of AMFR/gp78 in leukemic cells recovered A M F / P G I motile stimulation with a reduced differentiation ability. Haga et al. detected two unknown proteins by crosslinking between A M F / P G I and leukemic cells, suggesting a new receptor molecule for A M F / P G I in leukemic differentiation [74]. 28 1.5 - A M F / P G I IMPLICATION IN DISEASES Non-spherocytic hemolytic anaemia Aberrations in expression or activity of PGI due to mutations or deletions are of significant clinical importance since in humans they are associated with hereditary non-spherocytic hemolytic anaemia disease. Mutations result in enzyme instability and the defect only affects mature erythrocytes because they are no longer are capable of enzyme synthesis. Since glucose is not digestible by PGI mutated R B C s , it accumulates within the cell to the point where it deforms the cell membrane disrupting the oxygen carrying function of the erythrocytes. The major cl inical features of haemolysis include variable degrees of jaundice, slight-to-moderate splenomegaly, an increased incidence of gallstones, and anaemia [75] and severe cases produce mental retardation and can even lead to death. Non-spherocytic hemolytic anaemia was first reported in 1968 [76] and has been found in many patients with PGI mutations [26]. Moreover, this autosomal recessive genetic disorder may be associated in some cases with neurological impairment [77]. Although it is a rare disease, spherocytic hemolytic anaemia is the third most common enzymatic defect resulting in hemolysis. However, it can also be caused by other enzymes deficiencies such as glucose-6-phosphate dehydrogenase, pyruvate kinase and possibly of pyrimidine 5'-nucleotidase [26,78,79]. In contrast to the other red cell enzyme deficiencies, most of the mutations observed in PGI seem to be of independent origin. Therefore, there is no indication of any selection for any particular PGI mutation [26]. 29 Cancer Serum PGI activity has long been reported and is associated with tumor expression indicating that this protein is actively released from both normal and tumor cells [80]. This protein is elevated in the serum or urine of patients with malignant tumors such as gastro-intestinal, kidney, breast, colorectal and lung carcinomas, thereby being useful as a tumor marker [81]. When secreted, A M F / P G I promotes cellular locomotion or invasion [82] and regulates tumor malignancy proliferation [50], secretion [83] or apoptotic resistance [51]. Studies have shown that expression of glycolytic enzymes, including PGI , can modulate cellular life span [84]. Indeed, enhanced glycolysis seems to lead to an uncontrolled proliferation of mouse embryo fibroblasts. The paper correlates an increased glycolysis activity in tumors with their resistance to damaging free radical production. Since overexpression of PGI and phosphoglycerate mutase, an enzyme that catalyzes different steps in glycolysis, were both found to have similar effects on cell proliferation, immortalization seems likely to be caused by enhanced glycolysis rather then by specific properties of a glycolytic enzyme. Because both A M F / P G I and A M F R seem to be very specific to tumor and metastatic cells, interest in both the cytokine and its receptor has grown and the multifunctional enzyme and its receptor are of interest in the development of novel therapeutics against cancer. 30 Rheumatoid arthritis Autoimmunity is the failure of an organism to recognize its own constituent parts as " se l f , which results in an immune response against its own cells and tissues. Any disease that results from such an aberrant immune response is termed an autoimmune disease. Rheumatoid arthritis (RA) is a chronic, inflammatory, systemic disease. R A is strongly hypothesized to be an autoimmune disease. Although a wide range of autoantibodies antibodies can be found in R A , no common antigen has been specifically associated with the disease yet [85] and the true role of these antibodies remains controversial [86]. Studies suggesting implication of A M F / P G I in R A date from the mid 90's, when A M F / P G I was found in the synovial f luid of R A patients [87]. Watanabe et al. suggested that A M F was playing an essential role for communication among leukocytes in rheumatic disease [88]. However, a different role for A M F / P G I implication in R A was suggested later. Using the K / B x N T cell receptor transgenic mouse that spontaneously develops a joint disorder with many of the cl inical, histological, and immunological features of human R A , Matsumo and al. identified A M F / P G I this time as a possible autoantigen in R A [89]. The presence of antibodies to A M F / P G I in human R A was assessed and a first study by Schaller et al showed a strong binding of autoantibodies from the sera and synovial f luid of R A patients (64%) to the commercial rabbit A M F / P G I [90]. Two independent groups using recombinant human A M F / P G I contradicted these results, concluding that less than 5% of the sera/synovial f luid contained antibodies to human A M F / P G I [91,92]. Schaller and colleagues responded to these groups and stated that their new study using human 31 A M F / P G I showed that at least 49% of R A patients presented antibodies to both rabbit and human A M F / P G I . They attributed the differences between the studies to parameters such as antibody concentration, blocking solution type and time, incubation time, limitations of certain type of essays and conformation of the commercial and recombinant proteins. Since then, A M F / P G I implication in autoimmune diseases has remained very contradictory. More studies demonstrated no specificity of R A antibodies to A M F / P G I [93,94], while others revealed the importance of A M F / P G I in the disease [95-97]. Recent data have associated A M F / P G I antibodies to extra-articular complication in R A [98] to the pathogenesis of severe forms of arthritis [99] or correlated raised levels of anti-A M F / P G I to different subclasses of IgG [100]. It is unclear why A M F / P G I is secreted in the synovial joints of R A patients. However, A M F / P G I was identified as a hypoxic inducible gene [101] and hypoxia was found to be a characteristic feature of human rheumatoid arthritic joints and animal models of arthritis [102]. Therefore, it has been proposed that hypoxia, within the rheumatoid joints, might lead to upregulation of A M F / P G I which in turn would perpetuate R A [103]. As discussed earlier (section D), upon acidification, A M F / P G I is denatured and can bind directly to the fibrils of fibronectin. It has therefore been proposed that denaturation of PGI at acid p H could lead to increase binding to fibronectin-rich cartilage areas in the synovial joints permitting the generation of a localized autoimmune response against exposed non-native PGI epitopes [66]. 32 2. H Y P O T H E S I S Acid-induced conformational changes in phosphoglucose isomerase (PGI) result in its increased cell surface association and deposition on fibronectin fibrils [66]. We hypothesize that conformational changes of PGI , potentially due to disruption of the monomer-monomer interface under acidic conditions, such as those encountered in the synovial f luid of arthritic joints, could result in its deposition on the surface of joints and the induction of an autoimmune response. In this thesis, the acid-dependent binding of A M F / P G I to F N and conformation-dependent recognition of A M F / P G I by R A antisera are explored. 33 3. M A T E R I A L S A N D M E T H O D S 3.1 Protein purification PGI constructs were prepared by Zongjian Jia, a former post-doc. Transformed bacteria were grown in a 37°C shaker overnight in 3 m L Luria broth (LB) medium (Sigma) (Carbenicil l in (Sigma) 100 ug/mL; Kanamycin (Invitrogen) 30 ug/mL). 2 m L were used to inoculate 200 m L of L B medium (carbenicillin 100 ug/mL; kanamycine 30 ug/mL) and the culture was incubated in a 37°C shaker. After 2 hours (~0.6 OD600) protein expression was induced by addition of 1 m M of isopropyl-P-D-thiogalactopyranoside (IPTG) (Sigma) and the culture was agitated 4 more hours at room temperature. Afterwards, bacteria were harvested by centrifugation at 3,000g x 15min at 4°C and the pellet was used to purify his-tagged proteins by metal affinity resin following the B D Talon metal affinity resins protocol (Clontech). Proteins were dialyzed against P B S using dialyzing membranes ( V W R ) overnight at 4°C and concentrations were measured by B C A (Pierce). 3.2 SDS-PAGE and Western blots 5 ug of purified proteins were diluted in loading buffer with or without (3-mercaptoethanol and boiled at 95 °C for 5 minutes. Reduced samples, non-reduced samples and molecular weight markers (Pageruler prestained protein ladder, Fermentas) were loaded on an 8% polyacrylamide gel. Samples were run through the gel for 1 hour 30 minutes and protein bands revealed by Coomassie Blue staining. For western blot analysis, proteins were transferred to a nitrocellulose membrane (Amersham) using a semi-dry unit for 1 hour 30 minutes. Membranes were blocked in P B S containing 2% milk and 0.1% Tween-20 (S IGMA) overnight. Ant i -PGI (Raz, K C I , Detroit, U S A ) was 34 diluted 1:500 in blocking solution and incubated with the nitrocellulose membranes on a rotating support for 1 hour. Membranes were then washed 3 times 20 minutes and incubated with the secondary antibodies, anti-rabbit H R P (Molecular Probes), at a dilution of 1:5000 in blocking solution. Membranes were washed 20 minutes in blocking solution, 20 minutes in P B S - 2 % milk and twice 20 minutes in P B S . Membranes were revealed in E C L (1:1 solution A : 100 m M Tris p H 8.5, 2.5 m M Luminol , 0.4 m M p-Coumaric acid; solution B : 100 m M Tris p H 8.5, 0.02% H2O2) and exposed to X - R a y films. 3.3 Enzymatic activity assay A l l reagents were purchased from Sigma. Enzymatic activity of P G I / A M F was carried out using the Robert W . Gracy protocol [52]. Rabbit PGI and human constructs of PGI were prediluted in a 10 m M triethanolamine solution to a final concentration of l ug /mL. The reaction solution consisted of a 50mM triethanolamine buffer, 1 m M E D T A p H 8.0, 4 m M fructose-6-phosphate, 0.5 m M N A D P and 1 unit of glucose-6-phosphate dehydrogenase. Using a thermostatically regulated spectrophotometer ( U N I C A M U V visible spectrophotometer equipped with a circulating-water bath, Department of Chemistry, U B C , Vancouver, B C ) , the reaction solution was incubated for 5 minutes at 30°C in a 1 cm light-path cuvette. 25 u L of the prediluted enzyme was added to the reaction solution and the change in absorbance was read at 340 nm every minute for 5 minutes. The absorbance was then divided by 6.22 (the m M absorbance index for N A D P H ) to give the u M of glucose-6-phosphate formed per minute per m L of enzyme solution. 35 3.4 Glutaraldehyde cross-linking assay Glutaraldehyde cross-linking of PGI was carried out as previously described [66]. 50 pg PGI was incubated with 0.1% (v/v) glutaraldehyde (S IGMA) in 75 p i P B S for different times at room temperature. The reaction was stopped by addition of SDS sample buffer. Samples were boiled and reduced for 5 minutes, separated in 8% SDS-polyacrylamide gels, and protein bands were revealed by Coomassie Blue staining. 3.5 Circular dichroism Circular dichroism (CD) analysis of PGI and its constructs was performed using a Jasco J-810 spectropolarimeter ( U B C Laboratory of Molecular Biophysics). Spectra were recorded in a 2 m M quartz cuvette at room temperature in HEPES-based buffers at p H 7.5 and background signal obtained from parallel scans of the buffer alone were subtracted from the measurements. Far U V spectra of lug /ml PGI were recorded from 260 to 190 nm at a speed of 50nm/min in 1 nm steps with a signal averaging time of 4 seconds. Near U V spectra of 1 ug/ml PGI were recorded from 320 to 250 nm at a speed of 50 nm/min in 0.5 nm steps and with a signal averaging time of two seconds. 3.6 Fibronectin binding assay NIH-3T3 cells were plated at a density of 5000 cells/well on 96-well plates and cell containing wells and wells coated with soluble fibronectin at various concentration were fixed with 3% parafomaldehyde, rinsed extensively and labeled with anti-fibronectin and Alexa488 anti-rabbit secondary antibodies in order to determine the concentration (20 ug/ml) of soluble fibronectin that generated an equivalent signal to that of cell associated fibronectin fibrils. Subsequently, empty wells, NIH-3T3 cell containing wells and wells 36 coated with 20 ug/ml fibronectin or B S A were incubated in parallel with 25 ug/ml P G I / A M F - F I T C for 30 minutes at 37°C at p H 7.5 or p H 5.0, rinsed, fixed and then labeled with anti-FITC and Alexa488 anti-rabbit secondary antibodies. Fluorescence intensity of the labeled wells was measured using a Bio-Tek FL600 fluorescence plate reader. Fibronectin binding assay of constructs was carried as described above [67]. 25 ug/ml of constructs were added at various p H to 96-well plates coated with soluble 20 ug/mL fibronectin (S IGMA) and fixed with 3% paraformaldehyde, rinsed extensively and labeled with ant i -PGI /AMF as previously described [34] and A lexa 488 anti-rabbit (Jackson Laboratory) secondary antibodies, ant i -HA ( S I G M A ) and A lexa 488 anti-rabbit or anti-Flag (S IGMA) and A lexa 488 anti-mouse (Jackson Laboratory). 3 . 7 Cell motility assay 8 u M pore polycarbonate filters (Falcon) were placed in a 24-well plate and 500 p i A M F / P G I constructs, diluted in serum-free media (25 ug/mL), were added to the lower chamber of the filter. 200 p i of 5X10 5 MDA-231 cells were added to the upper chamber. After 24 hours of incubation at 37°C in a CO2 incubator, the top of the filters were gently cleaned with a Q-tip, rinsed with P B S and filters were transferred in a wel l containing methanol-acetone and fixed for 10 minutes at -20°C. Subsequently, filters were washed with P B S and transferred and incubated for 15 minutes in a new wel l containing 500 u L violet crystal and abundantly washed with water. Pictures of each filter were recorded using a Leica D M R A 2 microscope at x40 magnification. 37 3.8 Sera and synovial fluids Sera and synovial fluids samples were obtained from patients with rheumatoid arthritis or normal patients (Pascal Reboul, C H U M , Montreal, Canada; Hideomi Watanabe Gunma University, Japan) 3.9 Human RA antisera ELISA screening 96-well plates were coated with 100 u L of rabbit P G I / A M F at a 5pg/mL concentration diluted in 0.06 M Na-carbonate buffer p H 8.0 overnight. The plates were then blocked with 200 p L P B S - B S A 1% for 30 minutes. Blocking solution was removed and 50 p L of primary antibody (sera or synovial f luid 1:50) and incubated for 1 hour. Plates were washed 3 times with P B S - B S A 0.2% and detection antibody human IgG-HRP was added to the wells and incubated for 30 minutes. Plates were washed 2 times with P B S - B S A 0.2% and 1 time with P B S . lOOpL of A B T S solution (34.5 mg +25 p i of 30% H 2 0 2 in 100 m L H 2 0 ) was added to the wells and O D was read at 405 nm on a Bio-Tek FL600 plate reader. 3.10 RA antisera western blots screening 2 ug of PGI , PGI constructs or B S A were diluted in loading buffer with P-mercaptoethanol, boiled at 95°C for 5 minutes. Reduced samples and a molecular weight marker (Pageruler prestained protein ladder, Fermentas) were loaded on an 8% polyacrylamide gel. Samples were run through the gel for 1 hour 30 minutes. Proteins were transferred to a nitrocellulose membrane (Amersham) using a semi-dry unit for lh30min. Membranes were blocked in P B S containing 2% milk and 0.1% Tween-20 overnight. Primary antibodies, normal or R A sera/synovial f luid, were diluted 1:50 in blocking solution and incubated with the nitrocellulose membranes on a rotating support 38 for 1 hour. Membranes were then washed 3 times 20 minutes and incubated with the secondary antibodies, anti-human IgG-HRP (Molecular Probes), at a dilution of 1:5000 in blocking solution. Membranes were washed 1 time 20 minutes in blocking solution, 1 time 20 minutes in P B S - 2% mi lk and 2 times 20 minutes in P B S . Membranes were revealed in E C L (1:1 solution A : 100 m M Tris p H 8.5, 2.5 m M Luminol , 0.4 m M p-Coumaric acid; solution B : 100 m M Tris p H 8.5, 0.02% H 2 0 2 ) . 39 4. RESULTS 4.1 Analysis of recombinant AMF/PGI expression and purification Denaturation of PGI is believed to lead to autoimmune disease [66,67]. To disrupt P G I / A M F structure, N - and C-terminal tags were added to generate chimeric forms of the protein. Six constructs using human P G I / A M F constructs made by Zongjian Jia, a former post-doc in the lab, were used to study conformational effects of tag additions to the C or N-terminus. The PGI c D N A was tagged with H A or Flag epitope tags at the N or the C-terminus ( H i s - H A - A M F , A M F - H A - H i s , H is -F lag -AMF and AMF-H is -F lag ) . In order to study whether dimerization was able to restore A M F / P G I properties, a cysteine residue was added to the C-terminus or within the N-terminal tag, H is -F lag-AMF-cys and His -cys-F lag-AMF, respectively. The six constructs, A M F - H i s - F l a g , H i s -F lag -AMF, A M F - H A - H i s , H i s - H A - A M F , His-cys-F lag-AMF, H is -F lag-AMF-cys (Fig. 12b), were transformed in E. col i and purified, yielding concentrations between 200 pg/mL and 800 pg/mL. Under similar conditions, A M F - H i s - F l a g and A M F - H A - H i s , were purified at lower concentrations than H i s -F lag -AMF and H i s - H A - A M F . H is -F lag-AMF-cys showed the highest protein expression/purification but H is -cys-F lag-AMF showed low protein concentration after purification. Some amino acids can resist degradation more than others and their presence at the beginning o f proteins can be used to protect proteins from degradation. For example, proteasome is known to have a preference for hydrophobic residues [110] and R G S (arginine, glycine, serine) residues found in the pQE-31 vector are highly polar (arginine, 40 serine) or neutral (glycine), therefore making the site more hydrophilic. Because the pQE-31 R G S residues were kept intact while cloning H is -F lag -AMF and H i s - H A - A M F but removed from the A M F - H i s - F l a g and A M F - H A - H i s constructs, this probably explains why A M F - H i s - F l a g and, to a lesser extent, A M F - H A - H i s showed expression levels lower than H i s -F lag -AMF and H i s - H A - A M F . The addition of a cysteine at the C-terminus of A M F / P G I (His-F lag-AMF-cys) increased levels of expression (800 ug/mL) but expression level remained low (200 pg/mL) when a cysteine was added to the N-terminus (His-cys-Flag-AMF). A M F / P G I samples were subjected to 8% SDS polyacrylamide gel. A plot o f log molecular weight versus the migration, measured from the top of the gel, was prepared to calculate the molecular weight of each band. The actual molecular weights are shown in place of their corresponding log values. Analyzes showed that under reducing conditions, rabbit PGI migrated as a band of about 60 kDa while recombinant human A M F / P G I consistently migrated a little bit slower, as a band of about 62 kDa (Fig. 13a). Under non-reducing conditions, H is -F lag-AMF-cys presented an additional higher band. That higher molecular band of 123 kDa corresponds to the molecular weight of a His-Flag-A M F - c y s dimer (Fig. 13b). 4.2 Enzymat ic activity Because A M F / P G I is an enzyme that mediates interconversion between g lucoses-phosphate and fructose-6-phosphate, the different recombinant human A M F / P G I were tested for their enzymatic abilities, using Gracy's PGI enzymatic assay [52]. His-Flag-A M F , H i s - H A - A M F , and H is -cys-F lag-AMF exhibited complete loss o f enzymatic 41 activity. However, A M F - H A - H i s remained 50% active relative to the positive control rabbit PGI , and H is -F lag-AMF-cys showed enzymatic activity recovery o f about 55% to that of rabbit PGI (Fig. 14). 4.3 Glutaraldehyde cross-l inking Cross-linking of recombinant A M F / P G I with glutaraldehyde for different times prior to S D S - P A G E was used to further characterize their ability to form larger complexes. It was previously demonstrated that in the absence of cross-linking, PGI was detected exclusively as a monomer of ~60 kDa but with increasing time of cross-linking distinct protein bands appeared progressively at ~120 kDa, corresponding to PGI dimers, and at 240 kDa, which was suggested to be a tetrameric form of PGI [66]. Recombinant A M F / P G I that did not show dimerization on non-reducing gels ( A M F - H i s -Flag, H i s -F lag -AMF, A M F - H A - H i s , H i s - H A - A M F , H is-cys-F lag-AMF) were assessed for they ability to form multimers with 0.1 % glutaraldehyde for different time and then analyzed by reducing SDS-polyacrylamide (8%) gel electrophoresis. Coomassie Blue staining revealed the progressive transition of monomeric PGI to dimeric and then to tetrameric forms of the protein with increasing times of cross-linking. A plot of log molecular weight versus the migration, measured from the top of the gel, was prepared to calculate the molecular weight of each band. The molecular weights were applied to their corresponding log values. Cross-linking of H i s -F lag -AMF, A M F - H i s - F l a g , A M F - H A - H i s and H i s - H A - A M F were similar to that of rabbit PGI (Fig. 15a,b,c,e,f). A t time 0 of incubation with glutaraldehyde, rabbit PGI and the different recombinant human P G I / A M F were detected 42 only as a monomer, and as time increased, bands at ~ 120 kDa and -240 kDa appeared. H is -Cys -F lag -AMF exhibited a monomeric band at time 0 of cross-linking. However, with increasing time of cross-linking, higher molecular weight bands did not appear, indicating that H is -cys-F lag-AMF was incapable of multimerization (Fig. 15d). Other bands visible on the gels are due to the staining of contaminating proteins. 4.4 C i r cu l a r D ichro ism 4.4a F a r U V Circular dichroism is a form of spectroscopy that measures the differential absorption o f left- and right-handed circularly polarized light. The six recombinant forms of human A M F / P G I were assessed by circular dichroism in both the far U V and near U V to identify loss or change of secondary and tertiary structure, respectively (Fig. 16). Far U V (190-260 nm) C D spectrum of recombinant human A M F / P G I show a slight difference from the rabbit enzyme and is sensitive to the presence of additional amino acids located at the N and C terminal of the enzyme. There is a decrease of signal in the negative bands at 208 and 222 nm and an increase in the amplitude of the positive band at 198 nm. The presence of a Flag tag (Fig. 16a A M F - H i s - F l a g and H is -F lag -AMF) seems to be more disruptive than that of a Ha tag (Fig. 16a A M F - H A - H i s and H i s - H A - A M F ) . However, loss o f secondary structure is increased when either tag is located at the N terminus (Fig. 16a H is -F lag -AMF and H i s - H A - A M F ) . Therefore, the addition of a Flag at the N-terminus (Fig. 16a H is -F lag-AMF) seems to show the most severe effects on the secondary structure. The presence of a cysteine at the C-terminus (Fig. 16a His-Flag-43 AMF-cys ) stabilized the secondary structure, but not at the N-terminus (Fig. 16a His-cys-F lag -AMF) . 4.4b Near UV Information regarding tertiary structure can be obtained from the near U V C D spectra (250 to 320 nm). A l l recombinant human P G I / A M F (AMF-H is -F lag , H i s -F lag -AMF, A M F - H A - H i s , H i s - H A - A M F , H is -cys-F lag-AMF, H is-F lag-AMF-cys) presented a change in ellipticity of the bands, reflecting an altered conformational state (Fig. 16b). A M F - H A - H i s seems to retain a near U V spectrum more comparable to rabbit A M F / P G I . It is interesting to point out the apparent contribution of the disulfide bond in the near U V C D spectrum His-F lag-AMF-cys . It can be observed that between 250 and 270 nm, the intensity of the C D spectrum is larger when the Cys residues are forming a disulfide bond. This is the region where the C D absorption bands of the disulfide bond are expected [111]. Although the amplitude of the bands showed some variation, the structure and shape of the spectrum was retained across the different forms of recombinant A M F / P G I . 4.5 Binding to fibronectin Recombinant human A M F / P G I were subsequently tested for its ability to interact with cell receptors. A M F / P G I is known to bind in an increasing manner to fibronectin (FN) at acid pH . A fibronectin binding assay was carried out to determine whether P G I / A M F was able to bind to the soluble form of fibronectin. P G I / A M F binding to soluble F N was assessed using a fluorescent plate reader assay. To compare P G I / A M F binding to soluble F N relative to cell associated F N , a soluble F N 44 concentration (20 ug/mL) that generated an equivalent fluorescent signal using anti-FN antibody relative to plated NIH-3T3 cells was determined. A t p H 7.5, P G I / A M F - F I T C did not exhibit detectably increased binding to soluble F N or to NIH-3T3 cells relative to control wells left empty or coated with an equivalent concentration of B S A . Binding to F N at p H 5 was detectable and was essentially equivalent to P G I / A M F binding to N I H -3T3 cells indicating that at acid p H , P G I / A M F binds to soluble F N (Fig. 17a). Recombinant human A M F / P G I binding to F N was assessed using the previously described assay (Fig. 17b). The control rabbit A M F / P G I strongly bound at p H 5.0 and binding to F N decreased as p H increased. A M F - H i s - F l a g , H i s -F lag -AMF, A M F - H A -His, H i s - H A - A M F and H is -F lag-AMF-cys showed an increased binding to F N at acid p H although to a lesser extent compared to that of rabbit A M F / P G I . H is -cys -F lag-AMF showed low binding to F N . 4.6 Recombinant cell-induced motility A M F / P G I is known to induce cell motility. Therefore, recombinant human A M F / P G I s were assessed for their ability to promote migration o f MDA-231 cells through 8 p M pore polycarbonate filters (Fig. 18). After 14 hours of treatment with rabbit A M F / P G I , M D A -231 cell migration through the filters was quantified, exhibiting a 2.6-fold increase above negative control (p<0.01) . Under similar conditions, A M F - H i s - F l a g and A M F - H A - H i s increased cell motility by a 1.6-fold while cell-induced migration by H i s - H A - A M F had a 2-fold increase compared to the control with no sera (p<0.01). However, H i s -F lag -AMF, H is -cys-F lag-AMF and His -F lag-AMF-cys did not significantly increased cell motility (p<0.01). 45 4.7 Implication of AMF/PGI in rheumatoid arthritis 4.7a ELISA PGI has been postulated to play a role as an autoantigen in R A , although this has been controversial. Some studies reported a high prevalence of PGI autoantibodies [90,95,96] while other reports showed no specificity of R A antibodies against PGI [91,92,93]. Previous findings in our lab demonstrated denaturation of P G I / A M F under an acidic environment and binding to fibronectin following conformational changes of P G I / A M F [66,67]. The synovial fluids of arthritic joints are often found to be acidic [108], therefore we attempted to demonstrate that conformational changes of PGI might be responsible for the autoimmune response in R A [66]. A n E L I S A essay against rabbit PGI was first used to determine the existence of autoantibodies in the sera of patients with R A (Fig. 19). 49 sera and 14 synovial fluids from patients with R A and 10 sera from 'normal ' patients were tested for their ability to bind rabbit PGI , using B S A as a negative control. The results showed reproducible variations between each sera or synovial fluids, with a small increase of binding at p H 5.5 compared to the binding at pH 7.5. R A sera binding to rabbit PGI did not show more specificity than 'normal' sera (Fig. 19), More importantly, binding of antibodies to control B S A showed a very similar binding pattern to the control B S A . As a matter of fact, differences between PGI binding compared to B S A binding were specific at 8% of the time at p H 7.5 and 5% of the time at p H 5.5, showing that the assay had no or very low specificity to rabbit PGI (Fig. 19) (p<0.05). 46 4.7b Western blot The presence of autoantibodies to A M F / P G I in the sera o f R A patients was then assessed by western blot. Control showed that rabbit PGI , recombinant proteins H is -F lag -AMF, H is -F lag-AMF-cys and A M F - H A - H i s , but not B S A could be recognized by polyclonal rabbit anti-PGI by western blot (Fig. 20). 20 R A sera, 7 R A synovial fluids and 7 sera from 'normal' patients were tested for their ability to bind rabbit PGI , H i s -F lag -AMF and B S A (Fig. 21). Western blots were quantified (Fig. 21b) and ratios are shown in Table II (Table II). From the results obtained, it can be seen that, among the samples tested, only one R A serum showed binding to rabbit PGI (Fig. 21 #002) (p<0.05), while other samples presented no significant binding to rabbit PGI . Comparatively, binding to the recombinant human PGI H is -F lag -AMF was detected. Hence, about 70% of R A sera significantly bound H is -F lag -AMF (p<0.05). Binding to H i s -F lag -AMF between the samples showed high variations, some exhibiting an increase of 4 or 5 times above background while others showed an increase of up to 15 times above background. Synovial fluids from R A patients demonstrated strong binding (Fig. 21 #554sf, 584sf) (p<0.05) to H i s -F lag -AMF, which correlated with the presence of P G I autoantibodies in R A sera from the same patients (Fig. 21 R A sera #554, #584). One 'normal' serum showed strong binding to H is -F lag -AMF (Fig. 21#573) (p<).05). Among the 34 samples used, 3 R A sera exhibited significant binding to B S A (Fig. 21 #502,512,554) (p<0.05) and none of the other samples significantly recognized the B S A control. 47 To determine whether binding to human PGI was conformation specific (Fig. 22), His-F lag-AMF-cys and A M F - H A - H i s were chosen because the recombinant proteins retained enzymatic activity (Fig. 14), suggesting that their conformation is less altered, also confirmed by circular dichroism spectra (Fig. 16). Hence, in addition to rabbit PGI and H is -F lag -AMF, R A sera were tested for their ability to bind to H is -F lag-AMF-cys and A M F - H A - H i s by western blot. Sera that did not significantly bind H is -F lag -AMF did not show binding to any other recombinant A M F / P G I (Fig. 22 sera #009, 552, 556, 560, 572). When autoantibodies in R A sera showed binding to H i s -F lag -AMF, they exhibited binding to H is -F lag-AMF-cys (73%) and to A M F - H A - H i s (60%) as wel l . In many cases, autoantibodies bound H is -F lag -AMF more significantly than H is -F lag-AMF-cys or A M F - H A - H i s (66%). Several proteins can renature on a membrane following western transfer. Hence, PGI might renature following the transfer step. Therefore, stronger antibody binding to H is -F lag -AMF than to H is -F lag-AMF-cys and A M F - H A - H i s might be caused by the more aberrant conformation of H is -F lag -AMF. 48 5. Discussion 5.1 Conformational effects of residue additions to C-terminus and N-terminus PGI is a globular enzyme active as a dimer. The association between the monomers is reinforced by a C-terminus "a rm- l i ke " tail that w i l l embrace the other subunit when PGI forms a dimer. On the other hand, the N-terminus of PGI is found within the large domain [104]. When Flag and Ha were added to the C-terminus of PGI , only the Ha tag showed enzymatic activity (Table I A M F - H i s - F l a g , A M F - H A - H i s ) . However, both recombinant protein forms showed low structural alterations by C D (Fig. 16) and increased cell motility (Table I). When Flag and Ha tags were added to its N-terminus, PGI showed loss of enzymatic activity (Table I H i s -F lag -AMF, H i s - H A - A M F ) . However, H i s - H A - A M F showed less structural changes (Fig. 16) and increased cell motility while H is -F lag -AMF, which showed weak secondary structure C D spectra (Fig. 16), did not increase cell motility (Table I). There are some suggestions that not all residues make equal contributions to protein stability. In fact, it makes sense that amino acids located inside the protein, which become inaccessible to the solvent in the native state, would have a much greater effect than those on the surface [111]. These characteristics confer greater flexibil ity to the C-terminus tail but also cause the introduction of N-terminus residues to induce destabilization. Hence, because N -terminus is located between the dimer subunits, additions would be more disruptive than C-terminus additions, and interactions leading to dimerization would be less l ikely to happen, which explain loss of activity when tags are added to the N-terminus of PGI . 49 Disulfide bonds are believed to increase stability of the native state by decreasing the conformational entropy of the unfolded state due to the conformational constraints imposed by the cross-link [105]. C-terminal addition of a Cys residue stabilized PGI , showed by high levels o f protein expression (800ug/mL), by restoration o f enzymatic activity (Table I H is-F lag-AMF-cys) and by restoration of the Far U V spectra measured by C D (Fig. 16a). On the other hand, the addition of the cysteine to the N-terminus did not restore expression levels, enzymatic activity (Table I H is -cys-F lag-AMF) , the overall C D spectra showed altered conformation (Fig. 16a His-cys-F lag-AMF). The N-terminal Cys residue destabilized PGI even more, showed by loss of multimerization and of fibronectin binding (Summary Table I). Thus, cysteine increased PGI stability when added to the C-terminus but not to the N-terminus. One possible reason is that the cysteine residues introduced to the N-terminus would themselves lead to some destabilization. On the other hand, N-terminal cysteines may create a disulfide bond not strong enough to offset the destabilizing effects of the N-terminal Flag. A more simple explanation is that the added N-terminal cysteines can not interact with each other for steric reasons and therefore have no effect. 5.2 Cell-induced motility and cell interaction of AMF/PGI A M F was shown to be capable of glycolytic activities but A M F / P G I cell-induced motility does not correlate with enzymatic activity, although active sites from both enzymatic activity and motility overlap. A M F exhibited enzymatic/cytokine properties of PGI/neuroleukin which were inhibited by specific PGI inhibitors [11]. Site-directed mutagenesis of two residues involved in binding erythrose-4-phosphate results in 50 impaired autocrine motility factor activity, suggesting that the active site is involved in binding to the autocrine motility factor receptor [11]. W e showed (Fig. 14,18) that loss of enzymatic activity does not correlate with induction of cell motility, hence our results suggest that not all domains involved in the enzymatic active site are required for cytokine activity. It should also be noted that recombinant P G I / A M F were purified from bacterial culture and we cannot exclude the possibility that bacterial contaminants may contribute to the motile response observed. Indeed, A M F / P G I cell-induced motility requires a certain degree of structure: recombinant A M F / P G I showing loss of enzymatic activity (Table I H i s -F lag -AMF and His-cys-F lag-AMF), a strong change in structure (Fig. 16 H i s -F lag -AMF and His-cys-F lag -AMF) and low binding to fibronectin at acid p H (Table I H is -cys-F lag-AMF) did not increase cell motility (Table I H i s -F lag -AMF and His-cys-F lag-AMF) while proteins showing no enzymatic activity (Table I H i s - H A - A M F and A M F - H A - H i s ) but less changes in secondary structure (Fig. 5a H i s - H A - A M F and A M F - H A - H i s ) induced cell motility (Table I H i s - H A - A M F and A M F - H A - H i s ) . Therefore, alteration to a certain extent of A M F / P G I can disrupt cell-induced motility but enzymatic activity is not required, confirming that glycolytic and motility active sites are not totally identical. Thus interpretation is consistent with previous reports showing that enzymatic activity of phosphoglucose isomerase is not required for its cytokine function [61,62] PGI dimers are responsible for enzymatic activity but the monomer has been proposed to be responsible for neurotrophic activity of neuroleukin. In the presence of monomeric PGI , neuroblastoma cells have enhanced neurite extension and a reduced proliferation 51 rate [106]. Sites required to activate cell motility are not well-defined and it is still unclear whether dimerization is required for cell-induced motility [21]. The C-terminal covalently bound dimer (His-Flag-AMF-cys) , showed restoration of enzymatic activity (Table I H is-F lag-AMF-cys) and secondary structure (Fig. 5a His-F lag-AMF-cys) but was not able to induce cell motility (Table I H is -F lag-AMF-cys) . Thus, the inability of induced dimerization of H is -F lag-AMF-cys to restore motility stimulation argues that dimerization is not sufficient for the cytokine function of A M F / P G I and suggests that monomerization o f A M F / P G I might be required to activate cell motility pathway. Cross-linked A M F / P G I has been shown to endocytose at neutral p H [66]. Therefore, several reasons might explain the lack of cytokine activity of dimerized His -F lag-AMF-cys . For example, it might suggest that N-terminal modification wi l l prevent receptor activation or that structural changes or dimerization of H is -F lag-AMF-cys lead to A M F / P G I inability to react with the molecules involved in the cell motility pathway. On the other hand, dimerized H is -F lag-AMF-cys might bind different receptor or be endocytosed through a different pathway, explaining why cell motility did not occur. 5.3 Implications for the Role of PGI in Rheumatoid Arthritis Direct binding in large amounts of monomeric PGI to cell surface extracellular matrix fibrils at acid p H provides one possible explanation for the postulated role of PGI in R A . PGI cytokine activity has been found in rheumatoid synovial f luid [87] which has been reported to be acidic, as low as p H 6.0, particularly in rheumatoid patients [108]. A lso , PGI increased expression has been associated with hypoxia. In response to hypoxia, normal cells w i l l increase gene expression of glycolytic enzymes to adapt environmental 52 stress through activation of hypoxic-inducible transcription factor [102]. A t lower p H , A M F / P G I undergoes partial denaturation. Therefore, acid pH-association of PGI with the joint surface would result in exposure of epitopes different from the native form of the enzyme leading to an immune response and destruction of the cells through direct or indirect antibody mediated cytotoxicity [67]. Similarly, addition of tags to PGI results in a loss of structure and could mimic acid p H denaturation of PGI. Moreover, non-recognition of rabbit PGI by R A antibodies demonstrates a high specificity to human PGI of R A autoantibodies (Fig. 21) Differences between the ability of R A antisera to recognize different forms of recombinant human A M F / P G I at different levels (Fig. 22 H is -F lag -AMF, A M F - H A - H i s and His-F lag-AMF-cys) may be due to the degree of structural alteration of recombinant A M F / P G I . Recent studies have shown that R A autoantibodies to PGI are associated with the occurrence of extraarticular complications [98] and that elevated levels o f circulating anti-PGI antibodies in serum may be masked by non-specific binding [100]. However, patient information was unavailable for the current study and further research is required to determine whether high titers of PGI antibodies correlate with more severe forms of arthritis. It is interesting to note that the only normal sera that showed elevated anti-PGI antibodies was diagnosed with Sarcoidosis, an immune disorder which has also been correlated with high levels of secreted A M F [117] 53 6. Conclusion The prevalence of autoantibodies in R A sera was found in this study to be high (70%) (Fig. 21,Table II), consistent with original reports claiming elevated levels of antibodies to PGI in the sera/synovial fluids of R A patients (64%) [90]. Increased binding to His-F l a g - A M F compared to other forms of recombinant A M F / P G I may indicate more specific recognition by autoimmune Abs of conformationally modified A M F / P G I , confirming our original hypothesis that antisera in R A may be due to conformational changes in A M F / P G I . Although they have conserved a very similar structure throughout evolution, only mammalian but not bacterial or yeast forms of A M F / P G I acquired cytokine functions [109]. PGI extracellular functions, such as osteoblast differentiation [13] or embryo implantation [14], may have evolved in order to f i l l other functions. Bacteria and yeast do not have that complex multicellular system, which might explain that lack of PGI cytokine activity. Protein multifunctionalism is a clever mechanism for generation of complexity using existing proteins without requiring expansion of the genome [104]. However, loss of functional specificity may leave openings and opportunities for diseases. For instance, A M F / P G I structure may not have been primarily built to be secreted or to resist an acidic environment, resulting in the exposure of distinct motifs uncovered by the unfolding process, susceptible to autoantibody recognition. Therefore, A M F / P G I denaturation in the synovial fluid of R A patients might be responsible for R A autoantibody recognition and our data might provide new insights into the detection of epitopes involved.in autoimmune diseases. 7. FIGURES AND LEGENDS Vector and Constructs 5 5 Sma I EcoRI/RBS 6 x H " Bom HI Sph I Soc I Kpn I Xmo I Sol I Pstl Hind III fa I ATGTAGAGGATCI B M A C |GGATCCGCATGCGAGLTCGGTACCCCGGGTCGACCTGCAGCCAAGCTT]AATTAGCTGAG I 1 RGS His epitope B AMF-HA-His M - AMF - Y P Y D V P D Y A L H H H H H H K L Q stop... s i a r t l H H B AMHHHHHHHHBBHtHHIH— stop AMF-His-Flag M - AMF - H H H H H H D Y K D D D D K A stop start mmm: AMFflaHMi^i^HHHB^PM-st^ His-HA-AMF M R G S K H ^ D P H A S S V P Y HiHiBH^HBBB AMlvilHKHHr^st°P His-Flag-AMF M R G S j ^ ^ ^ ^ T D P H A S S V P R VD^y<r^^iDy^^ AMF_ ^ jp j jp^pjp A stop His-cys-Flag-AMF M R G S H hJH T D P H A S S V P R V D ^ ^ ^ ^ ) K - AMF - C stop s t a r t - — • • • • Flag AMF c y s s t o p His-Flag-AMF-cys M R G S H H C T D P H A S S V P R V D Y K D D D D ^ ^ ^ AMF - , ^ ^ ^ _ ^ _ s t o p start —-IHHHbys Flag AMF stop Figure 12 56 Figure 12 Vector and Constructs (A) Map of the expression vector pQE-31. (B) Maps of recombinant human P G I / A M F . P G I / A M F was tagged with H A or Flag epitope tags at the C and N-termini as well as 6XHis residues for purification ( A M F - H A - H i s , A M F - H i s - F l a g , H i s - H A - A M F , His-Flag-A M F ) . H i s -F lag -AMF was flanked with a cysteine residue at the C-terminus (His-Flag-AMF-cys ) and N-terminus (His-cys-Flag-AMF). 5 7 Western Blot and SDS-PAGE analysis of recombinant AMF/PGI Western blot SDS-PAGE Reducing Non-reducing Dimer Monomer c Semi-log graph 0 20 40 60 80 Relative migration Figure 13 58 Figure 13 Western blot and SDS-page analysis of recombinant A M F / P G I . Western blot (A) and S D S - P A G E (8%) (B) analysis of rabbit PGI and of the purified recombinant human A M F / P G I with (Reducing) or without (3-mercaptoefhanol (Non-reducing). Lane 1, molecular weight marker (Fermentas); lane 2, rabbit PGI (not shown on S D S - P A G E ) ; lane 3-8, purified A M F - H i s - F l a g , H i s - F l a g - A M F , A M F - H A - H i s , H i s - H A - A M F , H is -F lag-AMF-cys , and H is -cys-F lag-AMF N i - N T A affinity chromatography. (C) A plot of log molecular weight, measured from the top of the gel, was prepared to calculate the molecular weight of each band. The actual molecular weights are shown next to their corresponding log values. 59 Enzymatic activity of recombinant AMF/PGI Control AMF/PGI Rabbit i AMF-His-Flag His-Flag-AMF AMF-HA-His His-HA-AMF His-cys-Flag-AMF His-Flag-AMF-cys en 0.02 0.04 0.06 0.08 pM glucose 6-phosphate / min / mL 0.1 Figure 14 60 Figure 14 Enzymatic activity of recombinant AMF/PGI. A n enzymatic activity assay was used to determine the abilities of the purified constructs to convert fructose-6-phosphate to glucose-6-phosphate. The reaction was initiated by the addition of rabbit A M F / P G I or recombinant A M F / P G I (0.1 unit/mL) to l m L of reaction mixture [50 m M triethanolamine buffer (pH 8.3), 1 m M E D T A , 4 m M fructose 6-phosphate as a substrate, 0.5 m M N A D P , and 1 unit of glucose 6-phosphate dehydrogenase]. The graph shows activity for the construct A M F - H A - H i s and restoration of activity to the inactive H is -F lag -AMF by insertion of a cysteine residue at the C-terminus of the protein. Glutaraldehyde cross-linking of recombinant AMF/PGI 61 (T)~240--> (D)~120-> (M)~60--> A Rabbit AMF/PGI Min 0 1 5 10 15 30 60 B AMF-His-Flag Min 0 1 5 10 15 30 60 (T)~240~> f (D)~120-> (M)~60--> C His-Flag-AMF Min 0 1 5 10 15 30 60 (T)~240~> (D)~120--> (M)~60~> Semi-Log Graphs •Jretamer (~240 kDa) 20 40 60 Relative migration Jretamer (~-240_kQa) Relative migration Monomer (-62 kDa) 10 20 30 43 Relative migration 50 Figure 15 Glutaraldehyde cross-linking of recombinant AMF/PGI (continuation) 62 D His-cys-Flag-AMF Min 0 1 5 15 30 60 (M)~60-AMF-HA-His (T)~240~> (D)~120--> (M)~60»> His-HA-AMF Min 0 1 5 15 Semi-Log Graphs Monomer (-62 klpa) 10 20 30 40 Relative migration Tretamer (~ 240 kDa) 50 Monomer (-62 kba) 10 2 0 30 40 50 Relative migration Tretamer (~ 240 kDa) kDa) 10 20 30 40 Relative migration 50 Figure 15 63 Figure 15 Glutaraldehyde cross-linking and semi-log graph analysis. (A) Rabbit PGI and (B-F) recombinant human PGI were cross-linked with 0.1% glurataldehyde for 0, 1, 5, 10, 20, 30 and 60 min, as indicated, and then analyzed by reducing SDS-polyacrylamide (8%) gel electrophoresis. (A -C , E-F) Coomassie Blue staining reveals the progressive transition of monomeric PGI to dimeric and then to tretrameric forms of the protein with increasing times of cross-linking. (D) Coomassie Blue staining reveals no multimerization of the purified recombinant protein His-cys-F lag -AMF. (A-F) Plots of log molecular weight versus distance (in mm), measured from the top of the gel, were used to calculate the molecular weight of each band. The calculated molecular weights are shown next to their corresponding log values. Circular Dichroism of recombinant AMF/PGI B Far UV Near UV AMF-His-Flag His-Flag-AMF 0 His-cys-Flag-AMF a His-Flag-AMF-cys AMF-HA-His • His-HA-AMF • rabbit PGI CD 65 Figure 16 C i r cu la r d ichroism spectra of recombinant P G I / A M F . Circular dichroism (CD) analysis performed by recording spectra in a 2 m M quartz cuvette at room temperature in HEPES-based buffers at pH 7.5. Far U V spectra of recombinant P G I / A M F were recorded from 260 to 190 nm and near U V spectra of recombinant P G I / A M F were recorded from 320 to 250 nm. (A) Effects of tag additions upon the far U V C D spectra of P G I / A M F . The far U V C D spectrum of A M F - H A - H i s ()IOIOK) shows less structural alteration than that of the other recombinant P G I / A M F . H is -F lag-AMF-cys (eee) , relatively to H is -F lag -AMF (AAA) , shows restoration of its far U V C D spectra, to an extent comparable to that of A M F - H A - H i s (HOIQH). (B) Effects of tag additions upon the near U V C D spectra of P G I / A M F . The near U V C D spectrum of A M F - H A - H i s ()ipipi() show less structural alteration than that of other recombinant P G I / A M F s . The addition of a cysteine at the C-terminal of H i s - F l a g - A M F (AAA) leads to the apparition of a strong band between 250 and 270 nm. Binding of PGI/AMF to dimeric FN at neutral and acid pH 66 Labeling Figure 17 67 Figure 17 Binding of PGI/AMF to dimeric FN at neutral and acid pH. (A) The fluorescent signal due to binding of P G I / A M F - F I T C to uncoated wells of a 96-well plate, to wells coated with 20ug/mL B S A or F N , or to wells plated with NIH-3T3 cells was amplified by anti-FITC and Alexa488 anti-rabbit secondary antibodies and measured with a fluorescence plate reader. Absolute relative fluorescence values were normalized to maximal values and binding at pH 5.0 and p H 7.5 in the presence or absence of P G I / A M F - F I T C was determined. To assess relative F N levels in the wells containing soluble F N or NIH-3T3 cells, parallel wells were labeled with anti-FN and Alexa 488 anti-rabbit secondary antibodies. (B) Recombinant P G I / A M F binding to fibronectin was assessed using the same assay and showed increased binding at acid pH except for H is -cys-F lag-AMF. 68 Recombinant AMF/PGI cell-induced motility Figure 18 69 Figure 18 Recombinant AMF/PGI cell-induced motility. MDA-231 cells plated on 8 u M pore polycarbonate filters were treated with 25 ug/mL of each form of recombinant human A M F / P G I or rabbit PGI each construct or rabbit PGI . After 24 hours, the cells were fixed with methanol/acetone, dyed with crystal violet and pictures o f each filter taken. The number o f migrating cells was quantified. Rabbit PGI , A M F - H A - H i s , A M F - H i s - F l a g and H i s - H A - A M F significantly (*) stimulated cell motility, and H is -F lag -AMF, H is -cys-F lag-AMF and H is -F lag-AMF-cys did not. Table I Summary of recombinant AMF/PGI properties Construct Tag Caractenstics Non-reducing gel Enzymatic activity Glutaraldehyde cross-link FN binding Cell motility AMF-His-Flag Flag-Hi s at C-terminal 0 Monomer, Dimer, Tetramer ** His-Flag-AMF His-Flag at N-terminal 0 Monomer, Dimer, Tetramer * His-cys-Flag-AMF His-Flag at N-terminal Cysteine at N terminal 0 Monomer, no multimerization IC His-Flag-AMF-cys His-Flag at N-terminal Cysteine at C terminal Presence of dimer Monomer Dimer Tetramer * AMF-Ha-His Ha-His at ^terminal *** Monomer Dimer Tetramer ** His-Ha-AMF His-Ha at N-terminal 0 Monomer Dimer Tetramer *** 71 Aijsueiuj JO|OO aAi}B|9u Ajisuejui JO|OO 8Ai}B|ey Figure 19 .72 Figure 19 H u m a n R A antisera E L I S A screening. The binding of anti-sera to wells coated with 5ug/mL of rabbit PGI or B S A at pH 7.5 (A) or pH 5.5 (B) was detected using anti-human-HRP, visualized with a chromogenic substrate and quantified using a plate reader. (A) A t p H 7.5, 8% of the sera/synovial fluid showed significant (*) differences of binding to rabbit PGI and B S A and (B) at pH 5.5, 5% of the sera/synovial fluid showed significant (*) differences of binding to rabbit PGI and B S A . 73 RA antisera Western Blot screening Western Blot control Figure 20 74 Figure 20 Western Blot control Rabit PGI , H i s - F l a g - A M F , H is -F lag-AMF-cys and A M F - H A - H i s were run through a polyacrylamide gel, transferred to a nitrocellulose membrane and botted for anto-PGI. Polyclonal rabbit anti-PGI is able to specifically recognize the - 6 0 kDa rabbit PGI , His-F l a g - A M F , H is -F lag-AMF-cys and A M F - H A - H i s monomeric bands but not 66.4 kDa B S A band. Western Blot screening : species-specific recognition of RA anti-sera to AMF/PGI 75 RA sera 554 RA synovial fluid 552sf 554sf 555sf 560sf 570sf 572sf 584sf Densitometry Normal sera Pa Jo Al 558 n-ra 559 n-ra 573 n-ra 583 n-ra l l Audi iniAi T - N O J t - N N O f N I O ^ U l N O T - M ' t l O O D ^ O O O O O in in ll) if. in m Hi t'.i I I ' I-- 1^ - | - Qj o o o i n i n i n i n m m m m i n m m m m i n i n m m M V ) CA 1(1 H OT OT <N • * i n O O <N i f i n ifi i n U3 N s oo in in m i n i n in in a ;= o 0. < -> Ra sera I rahhit PGI • his-flag-Ah co co TO — *— i— i— c c c c oo C D ro <*> m m h co in m to m Dvial fluid 'Normal' sera Figure 21 Binding to (p<0.05) RA sera Ra synovial fluids 'normal' sera Rabbit PGI 5% (1/20) 0% 0% His-Flag-AMF 70% (14/20) 30% (2/7) 15% (1/7) BSA 15% (3/20) 0% 0% 76 Figure 21 Western Blot screening : species-specific recognition of human R A anti-sera to P G I / A M F Table II Analysis of densitometry (A) R A sera and synovial fluids and normal sera were assessed for the presence of autoantibodies against rabbit PGI , H is -F lag -AMF and BSA by western blot. (B) The bands were quantified. (See Table II) Quantification showed high prevalence of antibodies to H i s -F lag -AMF in R A sera (70%). R A synovial fluids that presented antibodies to H i s - F l a g - A M F correlated to patient sera that showed elevated levels of antibodies to His-Flag-AMF (#554, #584). 77 Western Blot screening : conformation-specific recognition of R A anti-sera to recombinant A M F / P G I Figure 22 78 Figure 22 Western Blot screening : conformation-specific recognition of human R A anti-sera to recombinant PGI/AMF (A) R A sera and synovial fluids and normal sera were assessed for the presence of autoantibodies against rabbit PGI, His-Flag-AMF and B S A by western blot. (B) The bands were quantified (see table II). Quantification showed a high prevalence of antibodies to His-Flag-AMF in R A sera (70%). R A synovial fluids that presented antibodies to His-Flag-AMF correlated to patient sera that showed high levels of antibodies to His-Flag-AMF (#554, #584). 79 8. Bibliography 1. Rapoport S: The regulation of glycolysis in mammalian erythrocytes. Essays Biochem 1968, 4:69-103. 2. Lohman: Biochem. Z . ; 1933. 3. Brown M C , Holland R L , Hopkins W G : Motor nerve sprouting. Annu Rev Neurosci 1981,4:17-42. 4. Gurney M E , Heinrich S P , Lee M R , Y i n H S : Molecular cloning and expression of neuroleukin, a neurotrophic factor for spinal and sensory neurons. Science 1986,234:566-574. 5. Gurney M E , Apatoff B R , Spear G T , Baumel M J , Antel JP, Bania M B , Reder A T : Neuroleukin: a lymphokine product of lectin-stimulated T cells. Science 1986, 234:574-581. 6. Faik P, Walker JI, Redmi l l A A , Morgan M J : Mouse glucose-6-phosphate isomerase and neuroleukin have identical 3' sequences. Nature 1988, 332:455-457. 7. Chaput M , Claes V , Portetelle D, Cludts I, Cravador A , Burny A , Gras H, Tartar A : The neurotrophic factor neuroleukin is 90% homologous with phosphohexose isomerase. Nature 1988, 332:454-455. 8. Gurney M : Reply : The neurotrophic factor neuroleukin is 90% homologous with phosphohexose isomerase. Nature 1988, 332:455-456. 9. Liotta L A , Mandler R, Murano G , Katz D A , Gordon R K , Chiang P K , Schiffmann E: Tumor cell autocrine motility factor. Proc Natl Acad Sci U S A 1986, 83:3302-3306. 10. Guirguis R, Schiffmann E, L iu B, Birkbeck D, Engel J , Liotta L: Detection of autocrine motility factor in urine as a marker of bladder cancer. J Natl Cancer Inst 1988, 80:1203-1211. 11. Watanabe H, Takehana K, Date M , Shinozaki T, Raz A : Tumor cell autocrine motility factor is the neuroleukin/phosphohexose isomerase polypeptide. Cancer Res 1996, 56:2960-2963. 12. X u W , Seiter K , Feldman E, Ahmed T, Chiao JW: The differentiation and maturation mediator for human myeloid leukemia cells shares homology with neuroleukin or phosphoglucose isomerase. B lood 1996, 87:4502-4506. 13. Zh i J , Sommerfeldt D W , Rubin C T , Hadjiargyrou M : Differential expression of neuroleukin in osseous tissues and its involvement in mineralization during osteoblast differentiation. J Bone Miner Res 2001,16:1994-2004. 14. Schulz L C , Bahr J M : Glucose-6-phosphate isomerase is necessary for embryo implantation in the domestic ferret. Proc Natl Acad Sci U S A 2003,100:8561-8566. 15. Roberts R M , Ealy A D , Alexenko A P , Han C S , Ezashi T: Trophoblast interferons. Placenta 1999, 20:259-264. 16. X u W , Lee P, Beutler E: Human glucose phosphate isomerase: exon mapping and gene structure. Genomics 1995, 29:732-739. 17. Jeffreys A J , Wi lson V , Thein SL : Hypervariable 'minisatellite' regions in human DNA. Nature 1985, 314:67-73. 80 18. Wil l iams R R , Hassan-Walker A F , Lavender F L , Morgan M , Faik P, Ragoussis J : The minisatellite of the GPI/AMF/NLK/MF gene: interspecies conservation and transcriptional activity. Gene 2001, 269:81-92. 19. Nakamura Y , Koyama K, Matsushima M : VNTR (variable number of tandem repeat) sequences as transcriptional, translational, or functional regulators. J Hum Genet 1998, 43:149-152. 20. Read J , Pearce J , L i X , Muirhead H, Chirgwin J , Davies C : The crystal structure of human phosphoglucose isomerase at 1.6 A resolution: implications for catalytic mechanism, cytokine activity and haemolytic anaemia. J M o l B io l 2001,309:447-463. 21. Sun Y J , Chou C C , Chen W S , W u R T , Meng M , Hsiao C D : The crystal structure of a multifunctional protein: phosphoglucose isomerase/autocrine motility factor/neuroleukin. Proc Nat l Acad Sc i U S A 1999, 96:5412-5417. 22. Jeffery C J , Bahnson B J , Chien W , Ringe D, Petsko G A : Crystal structure of rabbit phosphoglucose isomerase, a glycolytic enzyme that moonlights as neuroleukin, autocrine motility factor, and differentiation mediator. Biochemistry 2000, 39:955-964. 23. Shaw P J , Muirhead H: Crystallographic structure analysis of glucose 6-phosphate isomerase at 3-5 A resolution. J M o l B io l 1977,109:475-485. 24. Davies C , Muirhead H: Crystal structure of phosphoglucose isomerase from pig muscle and its complex with 5-phosphoarabinonate. Proteins 2002, 49:577-579. 25. Davies C , Muirhead H: Structure of native phosphoglucose isomerase from rabbit: conformational changes associated with catalytic function. Acta Crystallogr D B io l Crystallogr 2003, 59:453-465. 26. Beutler E, West C , Britton H A , Harris J , Forman L: Glucosephosphate isomerase (GPI) deficiency mutations associated with hereditary nonspherocytic hemolytic anemia (HNSHA). B lood Cells M o l Dis 1997, 23:402-409. 27. Matthews C K , Van Holde K E , Ahern K G : Biochemistry edn 3rd. San Francisco: Addison Wesley Longman Publishing Co ; 2000. 28. Lee J H , Jeffery C J : The crystal structure of rabbit phosphoglucose isomerase complexed with D-sorbitol-6-phosphate, an analog of the open chain form of D-glucose-6-phosphate. Protein Sci 2005,14:727-734. 29. Chou C C , Sun Y J , Meng M , Hsiao C D : The crystal structure of phosphoglucose isomerase/autocrine motility factor/neuroleukin complexed with its carbohydrate phosphate inhibitors suggests its substrate/receptor recognition. J B io l Chem 2000, 275:23154-23160. 30. Silletti S, Watanabe H, Hogan V , Nabi IR, Raz A : Purification of B16-F1 melanoma autocrine motility factor and its receptor. Cancer Res 1991, 51:3507-3511. 31. Watanabe H, Carmi P, Hogan V , Raz T, Silletti S, Nabi IR, Raz A : Purification of human tumor cell autocrine motility factor and molecular cloning of its receptor. J B io l Chem 1991, 266:13442-13448. 32. Tsutsumi S, Hogan V , Nabi IR, Raz A : Overexpression of the autocrine motility factor/phosphoglucose isomerase induces transformation and survival of NIH-3T3 fibroblasts. Cancer Res 2003, 63:242-249. 81 33. Yanagawa T, Watanabe H, Takeuchi T, Fujimoto S, Kurihara H, Takagishi K: Overexpression of autocrine motility factor in metastatic tumor cells: possible association with augmented expression of KIF3A and GDI-beta. Lab Invest 2004, 84:513-522. 34. Niinaka Y , Paku S, Haga A , Watanabe H, Raz A: Expression and secretion of neuroleukin/phosphohexose isomerase/maturation factor as autocrine motility factor by tumor cells. Cancer Res 1998, 58:2667-2674. 35. Haga A , Niinaka Y , Raz A: Phosphohexose isomerase/autocrine motility factor/neuroleukin/maturation factor is a multifunctional phosphoprotein. Biochim Biophys Acta 2000, 1480:235-244. 36. English AW: Cytokines, growth factors and sprouting at the neuromuscular junction. J Neurocytol 2003,32:943-960. 37. Gurney M E , Apatoff BR, Heinrich SP: Suppression of terminal axonal sprouting at the neuromuscular junction by monoclonal antibodies against a muscle-derived antigen of 56,000 daltons. J Cell Biol 1986,102:2264-2272. 38. Romagnoli A , Oliverio S, Evangelisti C, Iannicola C, Ippolito G, Piacentini M : Neuroleukin inhibition sensitises neuronal cells to caspase-dependent apoptosis. Biochem Biophys Res Commun 2003, 302:448-453. 39. Yamaguchi H , Wyckoff J , Condeelis J : Cell migration in tumors. Curr Opin Cell Biol 2005,17:559-564. 40. Woodhouse EC, Chuaqui RF, Liotta L A : General mechanisms of metastasis. Cancer 1997, 80:1529-1537. 41. Tsutsumi S, Gupta SK, Hogan V , Collard JG, Raz A: Activation of small GTPase Rho is required for autocrine motility factor signaling. Cancer Res 2002, 62:4484-4490. 42. Tapper J , Kettunen E, El-Rifai W, Seppala M , Andersson LC, Knuutila S: Changes in gene expression during progression of ovarian carcinoma. Cancer Genet Cytogenet 2001,128:1-6. 43. Seraj M J , Harding M A , Gildea JJ, Welch DR, Theodorescu D: The relationship of BRMS1 and RhoGDI2 gene expression to metastatic potential in lineage related human bladder cancer cell lines. Clin Exp Metastasis 2000,18:519-525. 44. Gildea JJ, Seraj M J , Oxford G, Harding M A , Hampton G M , Moskaluk CA, Frierson HF, Conaway MR, Theodorescu D: RhoGDI2 is an invasion and metastasis suppressor gene in human cancer. Cancer Res 2002, 62:6418-6423. 45. Harding M A , Arden K C , Gildea JW, Gildea JJ, Perlman EJ, Viars C, Theodorescu D: Functional genomic comparison of lineage-related human bladder cancer cell lines with differing tumorigenic and metastatic potentials by spectral karyotyping, comparative genomic hybridization, and a novel method of positional expression profiling. Cancer Res 2002, 62:6981-6989. 46. Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N : Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989, 246:1306-1309. 47. Shibuya M , Yamaguchi S, Yamane A , Ikeda T, Tojo A , Matsushime H, Sato M : Nucleotide sequence and expression of a novel human receptor-type tyrosine kinase gene (fit) closely related to the fms family. Oncogene 1990, 5:519-524. 82 48. Terman B l , Carrion M E , Kovacs E, Rasmussen B A , Eddy R L , Shows T B : Identification of a new endothelial cell growth factor receptor tyrosine kinase. Oncogene 1991, 6:1677-1683. 49. Yanagawa T, Funasaka T, Tsutsumi S, Watanabe H, Raz A : Novel roles of the autocrine motility factor/phosphoglucose isomerase in tumor malignancy. Endocr Relat Cancer 2004,11:749-759. 50. Tsutsumi S, Yanagawa T, Shimura T, Fukumori T, Hogan V , Kuwano H, Raz A : Regulation of cell proliferation by autocrine motility factor/phosphoglucose isomerase signaling. J B io l Chem 2003, 278:32165-32172. 51. Haga A , Funasaka T, Ni inaka Y , Raz A , Nagase H: Autocrine motility factor signaling induces tumor apoptotic resistance by regulations Apaf-1 and Caspase-9 apoptosome expression. Int J Cancer 2003,107:707-714. 52. Gracy R W , Ti l ley B E : Phosphoglucose isomerase of human erythrocytes and cardiac tissue. Methods Enzymol 1975, 41:392-400. 53. Nabi IR, Watanabe H, Raz A : Identification of B16-F1 melanoma autocrine motility-like factor receptor. Cancer Res 1990, 50:409-414. 54. Shimizu K, Tani M , Watanabe H, Nagamachi Y , Ni inaka Y , Shiroishi T, Ohwada S, Raz A , Yokota J : The autocrine motility factor receptor gene encodes a novel type of seven transmembrane protein. F E B S Lett 1999,456:295-300. 55. Fang S, Ferrone M , Yang C , Jensen JP, Tiwari S, Weissman A M : The tumor autocrine motility factor receptor, gp78, is a ubiquitin protein ligase implicated in degradation from the endoplasmic reticulum. Proc Nat l Acad Sci U S A 2001, 98:14422-14427. 56. Donaldson K M , Y i n H, Gekakis N , Supek F, Joazeiro C A : Ubiquitin signals protein trafficking via interaction with a novel ubiquitin binding domain in the membrane fusion regulator, Vps9p. Curr B io l 2003,13:258-262. 57. Zhong X , Shen Y , Ballar P, Apostolou A , Agami R, Fang S: AAA ATPase p97/valosin-containing protein interacts with gp78, a ubiquitin ligase for endoplasmic reticulum-associated degradation. J B io l Chem 2004, 279:45676-45684. 58. Y e Y , Shibata Y , Kikkert M , van Voorden S, Wiertz E, Rapoport T A : Inaugural Article: Recruitment of the p97 ATPase and ubiquitin ligases to the site of retrotranslocation at the endoplasmic reticulum membrane. Proc Nat l Acad Sci U S A 2005,102:14132-14138. 59. Benlimame N , Simard D, Nabi IR: Autocrine motility factor receptor is a marker for a distinct membranous tubular organelle. J Ce l l B io l 1995,129:459-471. 60. Wang H J , Benlimame N , Nabi I: The AMF-R tubule is a smooth ilimaquinone-sensitive subdomain of the endoplasmic reticulum. J Cel l Sci 1997, 110 ( Pt 24):3043-3053. 61. Benlimame N , Le P U , Nabi IR: Localization of autocrine motility factor receptor to caveolae and clathrin-independent internalization of its ligand to smooth endoplasmic reticulum. M o l B io l Ce l l 1998, 9:1773-1786. 62. Le P U , Benlimame N , Lagana A , Raz A , Nabi IR: Clathrin-mediated endocytosis and recycling of autocrine motility factor receptor to fibronectin fibrils is a limiting factor for NIH-3T3 cell motility. J Cel l Sci 2000, 113 ( Pt 18):3227-3240. 83 63. Silletti S, Paku S, Raz A : Autocrine motility factor and the extracellular matrix. II. Degradation or remodeling of substratum components directs the motile response of tumor cells. Int J Cancer 1998, 76:129-135. 64. Mao Y , Schwarzbauer J E : Fibronectin fibrillogenesis, a cell-mediated matrix assembly process. Matr ix B io l 2005, 24:389-399. 65. Wierzbicka-Patynowski I, Schwarzbauer JE : The ins and outs of fibronectin matrix assembly. J Ce l l Sci 2003,116:3269-3276. 66. Amraei M , Jia Z , Reboul P, Nabi IR: Acid-induced conformational changes in phosphoglucose isomerase result in its increased cell surface association and deposition on fibronectin fibrils. J B io l Chem 2003, 278:38935-38941. 67. Lagana A , Goetz J G , Y N , Altschuler Y , Nabi IR: pH-specific sequestration of phosphoglucose isomerase/autocrine motility factor by fibronectin and heparan sulphate. J Cel l Sci2005,118:4175-4185. 68. Parish C R : The role of heparan sulphate in inflammation. Nat Rev Immunol 2006, 6:633-643. 69. Homandberg G A , Wil l iams J E , Grant D, Schumacher B, Eisenstein R: Heparin-binding fragments of fibronectin are potent inhibitors of endothelial cell growth. A m J Pathol 1985,120:327-332. 70. Shimasaki S, L ing N : Identification and molecular characterization of insulin-like growth factor binding proteins (IGFBP-1, -2, -3, -4, -5 and -6). Prog Growth Factor Res 1991,3:243-266. 71. Corps A N , Brown K D : Mitogens regulate the production of insulin-like growth factor-binding protein by Swiss 3T3 cells. Endocrinology 1991,128:1057-1064. 72. Lemmey A B , Glassford J , Fl ick-Smith H C , Hol ly J M , Pel l J M : Differential regulation of tissue insulin-like growth factor-binding protein (IGFBP)-3, IGF-I and IGF type 1 receptor mRNA levels, and serum IGF-I and IGFBP concentrations by growth hormone and IGF-I. J Endocrinol 1997, 154:319-328. 73. Mishra S, Raz A , Murphy L J : Insulin-like growth factor binding protein-3 interacts with autocrine motility factor/phosphoglucose isomerase (AMF/PGI) and inhibits the AMF/PGI function. Cancer Res 2004, 64:2516-2522. 74. Haga A , Komazaki S, Funasaka T, Hashimoto K, Yokoyama Y , Watanabe H , Raz A , Nagase H: AMF/G6PI induces differentiation of leukemic cells via an unknown receptor that differs from gp78. Leuk Lymphoma 2006, 47:2234-2243. 75. Kugler W , Lakomek M : Glucose-6-phosphate isomerase deficiency. Baillieres Best Pract Res C l in Haematol 2000,13:89-101. 76. Baughman F A , Jr., Vander Ko lk K J , Mann JD, Valdmanis A : Two cases of primary amenorrhea with deletion of the long arm of theX chromosome (46, XXq-). A m J Obstet Gynecol 1968,102:1065-1069. 77. Eber S W , Gahr M , Lakomek M , Prindull G , Schroter W : Clinical symptoms and biochemical properties of three new glucosephosphate isomerase variants. Blut 1986,53:21-28. 84 78. Hirono Y , Fushida S, Yonemura Y , Yamamoto H, Watanabe H, Raz A : Expression of autocrine motility factor receptor correlates with disease progression in human gastric cancer. B r J Cancer 1996, 74:2003-2007. 79. Kanno H, We i D C , Chan L C , Mizoguchi H, Ando M , Nakahata T, Narisawa K, Fuj i i H , M i w a S: Hereditary hemolytic anemia caused by diverse point mutations of pyruvate kinase gene found in Japan and Hong Kong. B lood 1994, 84:3505-3509. 80. Bodansky O: Serum phosphohexose isomerase in cancer. II. As an index of tumor growth in metastatic carcinoma of the breast. Cancer 1954, 7:1200-1226. 81. Haga A : [Possibility that AMF will serve as a target molecule for the diagnosis and treatment of a metastatic neoplasm]. Yakugaku Zasshi 2005,125:169-175. 82. Liotta L A , Wewer U , Rao N C , Schiffmann E, Stracke M , Guirguis R, Thorgeirsson U , Muschel R, Sobel M : Biochemical mechanisms of tumor invasion and metastases. Adv Exp M e d B io l 1988, 233:161-169. 83. Funasaka T, Haga A , Raz A , Nagase H: Autocrine motility factor secreted by tumor cells upregulates vascular endothelial growth factor receptor (Flt-1) expression in endothelial cells. Int J Cancer 2002,101:217-223. 84. Kondoh H , Lleonart M E , G i l J , Wang J , Degan P, Peters G , Martinez D, Carnero A , Beach D: Glycolytic enzymes can modulate cellular life span. Cancer Res 2005,65:177-185. 85. Weyand C M , Goronzy JJ , Takemura S, Kurt in PJ : Cell-cell interactions in synovitis. Interactions between T cells and B cells in rheumatoid arthritis. Arthritis Res 2000, 2:457-463. 86. Maccioni M , Zeder-Lutz G , Huang H, Ebel C , Gerber P, Hergueux J , Marchal P, Duchatelle V , Degott C , van Regenmortel M , et al.: Arthritogenic monoclonal antibodies from K/BxN mice. J Exp Med 2002,195:1071-1077. 87. Watanabe H, Takeuchi K, Chigira M : Expression of autocrine motility-like factor in rheumatoid synovial fluid. J Rheumatol 1994, 21:37-40. 88. Takeuchi K, Watanabe H, Takagishi K: Biochemical investigation of cell motile activity in rheumatoid synovial fluid. J Rheumatol 1998, 25:9-15. 89. Matsumoto I, Staub A , Benoist C , Mathis D: Arthritis provoked by linked T and B cell recognition of a glycolytic enzyme. Science 1999, 286:1732-1735. 90. Schaller M , Burton D R , Ditzel H J : Autoantibodies to GPI in rheumatoid arthritis: linkage between an animal model and human disease. Nat Immunol 2001, 2:746-753. 91. Schubert D, Schmidt M , Zaiss D, Jungblut P R , Kamradt T: Autoantibodies to GPI and creatine kinase in RA. Nat Immunol 2002, 3:411; author reply 412-413. 92. Kassahn D, Ko lb C , Solomon S, Bochtler P, Illges H: Few human autoimmune sera detect GPI. Nat Immunol 2002, 3:411-412; author reply 412-413. 93. Matsumoto I, Lee D M , Goldbach-Mansky R, Sumida T, Hitchon C A , Schur P H , Anderson R J , Coblyn JS, Weinblatt M E , Brenner M , et al.: Low prevalence of antibodies to glucose-6-phosphate isomerase in patients with rheumatoid arthritis and a spectrum of other chronic autoimmune disorders. Arthritis Rheum 2003, 48:944-954. 94. Herve C A , Wait R, Venables P J : Glucose-6-phosphate isomerase is not a specific autoantigen in rheumatoid arthritis. Rheumatology (Oxford) 2003, 42:986-988. 85 95. K i m J Y , Lee M H , Jung K I , N a H Y , Cha H S , K o E M , K i m TJ : Detection of antibodies against glucose 6-phosphate isomerase in synovial fluid of rheumatoid arthritis using surface plasmon resonance (BIAcore). Exp M o l M e d 2003,35:310-316. 96. Cha H S , K i m TJ , K i m J Y , Lee M H , Jeon C H , K i m J , Bae E K , A h n K S , K o h E M : Autoantibodies to glucose-6-phosphate isomerase are elevated in the synovial fluid of rheumatoid arthritis patients. Scand J Rheumatol 2004, 33:179-184. 97. Muraki Y , Matsumoto I, Chino Y , Hayashi T, Suzuki E, Goto D, Ito S, Murata H , Tsutsumi A , Sumida T: Glucose-6-phosphate isomerase variants play a key role in the generation of anti-GPI antibodies: possible mechanism of autoantibody production. Biochem Biophys Res Commun 2004, 323:518-522. 98. van Gaalen F A , Toes R E , Ditzel H J , Schaller M , Breedveld F C , Verweij C L , Huizinga T W : Association of autoantibodies to glucose-6-phosphate isomerase with extraarticular complications in rheumatoid arthritis. Arthritis Rheum 2004, 50:395-399. 99. Hayashi T, Matsumoto I, Muraki Y , Takahashi R, Chino Y , Goto D, Ito S, Tsutsumi A , Sumida T: Clinical characteristics of anti-glucose-6-phosphate isomerase antibody-positive Japanese patients with rheumatoid arthritis. M o d Rheumatol 2005,15:258-263. 100. Schaller M , Stohl W , Tan S M , Benoit V M , Hilbert D M , Ditzel H J : Raised levels of anti-glucose-6-phosphate isomerase IgG in serum and synovial fluid from patients with inflammatory arthritis. Ann Rheum Dis 2005, 64:743-749. 101. Yoon D Y , Buchler P, Saarikoski ST, Hines OJ , Reber H A , Hankinson O: Identification of genes differentially induced by hypoxia in pancreatic cancer cells. Biochem Biophys Res Commun 2001, 288:882-886. 102. Taylor P C , Sivakumar B: Hypoxia and angiogenesis in rheumatoid arthritis. Curr Opin Rheumatol 2005,17:293-298. 103. Naughton D P : Hypoxia-induced upregulation of the glycolytic enzyme glucose-6-phosphate isomerase perpetuates rheumatoid arthritis. M e d Hypotheses 2003, 60:332-334. 104. Copley SD: Enzymes with extra talents: moonlighting functions and catalytic promiscuity. Curr Opin Chem B io l 2003, 7:265-272. 105. Welker E, Wedemeyer W J , Narayan M , Scheraga H A : Coupling of conformational folding and disulfide-bond reactions in oxidative folding of proteins. Biochemistry 2001, 40:9059-9064. 106. Mizrachi Y : Neurotrophic activity of monomeric glucophosphoisomerase was blocked by human immunodeficiency virus (HIV-1) and peptides from HIV-1 envelope glycoprotein. J Neurosci Res 1989, 23:217-224. 107. Benell i R, Lorusso G , A lb in i A , Noonan D M : Cytokines and chemokines as regulators of angiogenesis in health and disease. Curr Pharm Des 2006, 12:3101-3115. 108. Hadjigogos K: The role of free radicals in the pathogenesis of rheumatoid arthritis. Panminerva Med 2003, 45:7-13. 109. Amraei M , Nabi IR: Species specificity of the cytokine function of phosphoglucose isomerase. F E B S Lett 2002, 525:151-155. 86 110. Haga A , Tanaka N , Funasaka T, Hashimoto K, Nakamura K T , Watanabe H, Raz A , Nagase H: The autocrine motility factor (AMF) and AMF-receptor combination needs sugar chain recognition ability and interaction using the C-terminal region of AMF. J M o l B io l 2006, 358:741-753. 111. Woody R W , Sugeta H, Kodama TS: [Circular dichroism of proteins: recent developments in analysis and prediction]. Tanpakushitsu Kakusan Koso 1996, 41:56-69. 112. K i n g M W : http://www.dentistry.leeds.ac.uk/biochem/thcme/glycolysis.pdf 1996 113. Hugues R C : Secretion of the galectin family of mammalian carbohydrate-binding proteins. B iochim Biophys Acta 1999 1473:172-85. 114. Mignatti P, Morimoto T, Ri fk in D B : Basic fibroblast growth factor, a protein devoid of secretory signal sequence, is released by cells via a pathway independent of the endoplasmic reticulum-Golgi complex. J . Ce l l Physiol 1992,15:81-93 115. Florkiewicz R Z , Majack R A , Buechler R D , Florkiewicz E: Quantitative export of FGF-2 occurs through an alternative, energy dependent, non-ER/Golgi pathway. J . Ce l l Physiol 1995,162:388-399 116. Gether U : Uncovering molecular mechanisms involved in activation of G protein-coupled receptors. Endocr Rev 2000, 21:90-113 117. Dobashi Y , Watanabe H, Matsubara M , Yanagawa T, Raz A , Shimamiya T, Oo i A : Autocrine motility factor/glucose-6-phosphate isomerase is a possible predictor of metastasis in bone and soft tissue tumours. J Pathol 2006, 208:44-53. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0100830/manifest

Comment

Related Items