Receptor Mediated Catabolism of Plasminogen Activators

RECEPTOR MEDIATED

CATABOLISM OF PLASMINOGEN

ACTIVATORS

Philip George Grimsley

Thesis submitted for the degree of Doctor of Philosophy

from The University of New South Wales

School of Medical Sciences 2009 PLEASE TYPE THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: Grimsley

First name: Philip Other name/s: George

Abbreviation for degree as given in the University calendar: PhD

School: School of Medical Sciences Faculty: Medicine

Title: Receptor Mediated Catabolism of Plasminogen Activators

Abstract 350 words maximum: (PLEASE TYPE)

Humans have two plasminogen activators (PAs), tissue-type plasminogen activator (tPA) and urokinase-type plasminogen activator (uPA), which generate plasmin to breakdown fibrin and other barriers to cell migration. Both PAs are used as pharmaceuticals but their efficacies are limited by their rapid clearance from the circulation, predominantly by parenchymal cells of the liver. At the commencement of the work presented here, the hepatic receptors responsible for mediating the catabolism of the PAs were little understood. tPA degradation by hepatic cell lines was known to depend on the formation of binary complexes with the major PA inhibitor, plasminogen activator inhibitor type-1 (PAI-1). Initial studies presented here established that uPA was catabolised in a fashion similar to tPA by the hepatoma cell line, HepG2. Other laboratories around this time found that the major receptor mediating the binding and endocytosis of the PAs is Low Density Lipoprotein Receptor-related Protein (LRP1). LRP1 is a giant 600 kDa protein that binds a range of structurally and functionally diverse ligands including, activated α2-macroglobulin, apolipoproteins, β-amyloid precursor protein, and a number of serpin-enzymes complexes, including PA•PAI-1 complexes. Further studies for the work presented here centred on this receptor. By using radiolabelled binding assays, ligand blots, and Western blots on cultured cells, the major findings are that: (1) basal LRP1 expression on HepG2 is low compared to a clone termed, HepG2a16, but appears to increase in long term culture; (2) a soluble form of LRP1, which retains ligand-binding capacity, is present in human circulation; (3) soluble LRP1 is also present in cerebral spinal fluid where its role in neurological disorders such as Alzheimer‟s disease is a developing area of interest; and (4) the release of LRP1 is a mechanism conserved in evolution, possibly as distantly as molluscs. The discovery, identification, and characterisation of soluble LRP1 introduces this protein in the human circulation, and presents a possible further level of regulation for its associated receptor system.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

…………………………………………………………… ……………………………………..……………… ……….……………………... Signature Witness ……. Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS

To Leonie

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed ……………………………………………......

Date ……………………………………………......

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed ……………………………………………......

Date ……………………………………………......

ORIGINALITY STATEMENT

„I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.‟

Signed ……………………………………………......

Date ……………………………………………......

PUBLICATIONS

Philip G Grimsley, Kathryn A Quinn, Colin N Chesterman, Dwain A Owensby “Evolutionary conservation of circulating low density lipoprotein receptor-related protein-like („LRP1-like‟) molecules” Thrombosis Research 1999; 94(3): 153-164

Philip G Grimsley, Kathryn A Quinn, Dwain A Owensby “Soluble Low density lipoprotein receptor related protein” Trends in Cardiovascular Medicine 1998; 8(8): 363-368

Philip G Grimsley, Kathryn A Quinn, Colin N Chesterman, Dwain A Owensby “Low density lipoprotein receptor-related protein (LRP1) expression varies among HepG2 cell lines” Thrombosis Research 1997; 88(6): 485-98

Kathryn A Quinn, Philip G Grimsley, Yang-Ping Dai, Michael Tapner, Colin N Chesterman, Dwain A Owensby “Soluble low density lipoprotein receptor-related protein (LRP1) circulates in human plasma” Journal of Biological Chemistry 1997; 272(38): 23946-23951

Philip G Grimsley, John F Normyle, Ruth A Brandt, Georgina Joulianos, Colin N Chesterman, Philip J Hogg, Dwain A Owensby “Urokinase binding and catabolism by HepG2 cells is plasminogen activator inhibitor-1 dependent, analogous to interactions of tissue-type plasminogen activator with these cells” Thrombosis Research 1995; 79(4): 353-361

iii

CONFERENCE PRESENTATIONS

Philip G Grimsley, Kathryn A Quinn, Colin N Chesterman, Dwain A Owensby “Soluble low density lipoprotein receptor-related protein (LRP1) is present in the serum of mammals and birds” Australian Vascular Biology Society Fifth annual scientific meeting, The Fairmont Resort, Leura, Blue Mountains, NSW, 11-13 Sep 1997, page 65

Philip G Grimsley, Kathryn A Quinn, Colin N Chesterman, Dwain A Owensby “A soluble form of low density lipoprotein receptor-related protein (LRP1) circulates in plasma” Australian Vascular Biology Society Fourth annual scientific meeting, Marysville, Victoria, 17-20 Oct 1996, page 43

Philip G Grimsley, Kathryn A Quinn, Colin N Chesterman, Dwain A Owensby “Low density lipoprotein receptor-related protein exists as a soluble form in plasma” Australian Society for Experimental Pathology 28th annual meeting, University of NSW, Sydney, 2-4 Oct 1996, page 18

Philip G Grimsley, Kathryn A Quinn, Philip J Hogg, Colin N Chesterman, Dwain A Owensby “Variation in the level of expression of low density lipoprotein receptor related protein (LRP1) between sublines of HepG2 cells” Australian Vascular Biology Society Third annual conference, Terrigal, NSW, 5-7 Oct 1995

Philip G Grimsley, Fei Le, Philip J Hogg, Colin N Chesterman, Dwain A Owensby “Evidence for a receptor for plasminogen activatorplasminogen activator inhibitor type-1 complexes (PAPAI-1) distinct from the low density lipoprotein receptor-related protein (LRP1) on HepG2 cells” Australian Vascular Biology Society Second annual scientific meeting, Hahndorf, SA, 17-20 Nov 1994

P.G.Grimsley, J.F.Normyle, G.Joulianos, P.J.Hogg, C.N.Chesterman, D.A.Owensby “Specificity of the receptor for tissue type plasminogen activator-plasminogen activator inhibitor type-1 complex (tPAPAI-1) on human hepatoma cell line HepG2 extends to analogous complexes with urokinase (uPAPAI-1)” Australian Vascular Biology Society First scientific meeting, Oasis Resort, Caloundra, Qld, 19-21 Aug 1993, page 42

TABLE OF CONTENTS

Thesis / Dissertation Sheet Cover Page Dedication

Copyright Statement ...... i Authenticity Statement ...... i Originality Statement ...... ii Publications ...... iii Conference Presentations ...... iv Table of Contents ...... v List of Tables ...... ix List of Figures ...... x Abbreviations ...... xii Abstract ...... xiv Acknowledgements ...... xv

1 GENERAL INTRODUCTION 16

1.1 Overview, Aims, and Hypotheses 17 1.2 The Plasminogen Activator System 19 1.2.1 Overview of the Plasminogen Activator System 19 1.2.2 Components of Plasminogen Activator System 20 1.3 The Plasminogen Activators 20 1.3.1 Tissue-type plasminogen activator (tPA) 21 1.3.2 Urokinase-type plasminogen activator (uPA) 25 1.4 Serine Proteinases 27 1.4.1 Serpins, and serpin enzyme complexes (SECs) 27 1.4.2 Clinical uses of plasminogen activators 30 1.4.3 Clearance as a PA system regulator 31 1.4.4 Importance of PA clearance 31 1.4.5 Mechanism of Plasminogen Activators Clearance 32 1.5 Low density lipoprotein receptor-related protein (LRP1) 34 1.5.1 Historic Perspectives and Nomenclature 34 1.5.2 The LDL receptor family 36

2 GENERAL MATERIALS AND METHODS 43

2.1 Cell culture 44 2.1.1 Materials 44 2.1.2 Freezing of cells in liquid nitrogen 45 2.1.3 Culture of cells from frozen stocks 46 2.1.4 Passage of adherent cell lines 46 2.1.5 Characteristics and culture of HepG2 cells 47

v 2.1.6 Culture of other adherent cell lines 49 2.2 Cell homogenates 49 2.2.1 Preparation of cell lysates 49 2.2.2 Preparation of kidney membrane 50 2.3 Antibodies 50 2.3.1 Hybridomas. 50 2.3.2 LRP1-specific antibodies 52 2.3.3 A note concerning 8G1 species specificity 52 2.4 Flow Cytometry 53 2.4.1 Cell preparation, labelling, and analysis 53 2.4.2 Harvesting HepG2 for flow cytometric analysis 53 2.5 Protein manipulations 55 2.5.1 Protein assay 55 2.5.2 Iodination of proteins using iodogen coated tubes 55 2.5.3 Determination of specific activity 57 2.5.4 Purity of iodinated protein 58 2.5.5 Preparation of recombinant RAP 59 2.5.6 Production of GST-RAP in E.coli 60 2.5.7 Immobilisation of proteins on sepharose 2B 63 2.6 Electrophoresis and blotting 65 2.6.1 The Laemmli system of protein electrophoresis 65 2.6.2 Staining of gels for total proteins 66 2.6.3 Transfer of electrophoresed proteins to PVDF membrane 67 2.6.4 Blocking of membranes in milk 68 2.6.5 Western blotting 68 2.6.6 ¹2 5I-Ligand blotting 69

3 UROKINASE (UPA) BINDING AND ENDOCYTOSIS BY HEPG2 CELLS 70

3.1 Introduction 71 3.2 Methods 74 3.2.1 Separation of HMW-uPA and LMW-uPA 74 3.2.2 Iodination of HMW-uPA, LMW-uPA, and tPA 75 3.2.3 Autoradiography of 125I-uPA bound to HepG2 monolayers 75 3.2.4 Detection of PAI-1 associated with 125I-HMW-uPA following interaction with HepG2 76 3.2.5 Internalisation of surface-bound 125I-PAPAI-1 ligands 76 3.3 Results 77 3.3.1 Conversion of uPA to uPAPAI-1 ligands. 77 3.3.2 Identification of PAI-1 associated with 125I-HMW-uPA following exposure to HepG2 78 3.3.3 Internalisation of PAs 80 3.4 Discussion 82

4 LRP1 LEVELS VARY AMONG HEPG2 SUBLINES 85

4.1 Introduction 86 4.2 Materials and methods 88 4.2.1 HepG2 sublines 88 4.2.2 Materials 88 4.2.3 Flow cytometry 89

vi 4.2.4 Iodinations 90 4.2.5 Inhibition of 125I-tPA binding by RAP in serial assays 90 4.2.6 Protein assays (4.1-6) 91 4.2.7 Western blotting and 125I-RAP ligand blotting 91 4.2.8 Pseudomonas exotoxin A (PEA) sensitivity assay 91 4.2.9 Degradation assays 91 4.3 Results 92 4.3.1 Surface membrane phenotyping of HepG2 cells 92 4.3.2 ¹2 5I -tPA serial binding assays 94 4.3.3 ¹2 5I -tPA degradation by HepG2 cells 98 4.3.4 ¹2 5I-RAP ligand blots 98 4.3.5 Pseudomonas exotoxin A (PEA) sensitivity 99 4.3.6 Western blots for LRP1 in HepG2 lysates 101 4.3.7 LRP1 on HepG2 sublines determined by flow cytometry 102 4.3.8 Status of other endocytic receptors on HepG2 sublines. 104 4.3.9 Culture condition perturbations 105 4.4 Discussion 106

5 SOLUBLE LRP1 IN HUMAN CIRCULATION 109

5.1 Introduction 110 5.2 Materials and Methods 112 5.2.1 RAP, antibodies, and cell lines 112 5.2.2 ¹2 5I-RAP blots and Western blots of cell lysates and body fluids 112 5.2.3 Enrichment of soluble LRP1 on 8G1-sepharose 112 5.2.4 Ammonium sulfate precipitation of soluble LRP1 from serum 113 5.2.5 Investigation of LRP1 on resting and activated platelets 114 5.2.6 Release of soluble LRP1 from HepG2a16 (a16/HP) 114 5.2.7 LRP1 association with extracellular matrix 115 5.2.8 Endothelial cell culture 115 5.2.9 CSF, lacrimal fluid, and urine samples 116 5.3 Results 116 5.3.1 Initial detection of RAP-binding proteins in serum 116 5.3.2 High molecular weight RAP-binding proteins in human serum 118 5.3.3 Immunological detection of soluble LRP1 120 5.3.4 Stability of soluble LRP1 in serum 120 5.3.5 Identification of soluble LRP1 by serial 8G1-sepharose enrichment and 125I-RAP blotting 122 5.3.6 Enrichment of soluble LRP1 122 5.3.7 Further identification of soluble LRP1 125 5.3.8 Platelets as a potential source of soluble LRP1 125 5.3.9 Release of soluble LRP1 from HepG2a16 128 5.3.10 Association of serum-borne LRP1 with extracellular matrix 130 5.3.11 Detection of a RAP-binding protein in lacrimal fluid 132 5.3.12 Detection of soluble LRP1 in CSF 134 5.4 Discussion 135

6 EVOLUTIONARY CONSERVATION OF SOLUBLE LRP1 140

6.1 Introduction 141 6.2 Materials and Methods 142 6.2.1 Receptor Associated Protein and RAP-Sepharose 2B (RAP-2B). 142

vii 6.2.2 Sera and Haemolymph 142 6.2.3 Enrichment of soluble RAP-binding proteins on RAP-2B 142 6.2.4 Membrane LRP1/2MR and LRP2/megalin Standards 143 6.2.5 Primary Antibodies 143 6.2.6 Western and Ligand Blots 144 6.2.7 Iodinations 145 6.3 Results 145 6.3.1 RAP-binding protein in animal sera and haemolymph 145 6.3.2 Pseudomonas exotoxin A (PEA) ligand blots. 147 2 5 6.3.3 ¹ I-activated 2-macroglobulin ligand binding 149 6.3.4 Monoclonal antibody identification of soluble 500 kDa LRP1 -Chain in mammalian sera 150 6.3.5 -Chain fragments in soluble LRP1 152 6.4 Discussion 154

7 CONCLUSIONS 157

Extended Abstract ...... 161

8 REFERENCES 165

viii

LIST OF TABLES

Table 1-1 Examples of serpins and their common target serine proteases. 29

Table 2-1 Hybridomas producing LRP1 specific monoclonal antibodies. 51

Table 4-1 Characteristics and sources of HepG2 used in this study. 89

Table 4-2 Average level of inhibition (in percent) caused. to the specific binding of 3 nM 125I-tPA by HepG2 cell lines upon addition of 1 M RAP at 4C. 96

Table 4-3 Culture condition perturbations applied to HepG2 cells 105

LIST OF FIGURES

Figure 1-1 The PA system. 20

Figure 1-2 The primary sequence and secondary structure of tPA. 23

Figure 1-3 Cleavage sites on uPA. 26

Figure 1-4 PAI-1 dependent catabolism of tPA by HepG2. 33

Figure 1-5 Amino acid sequence of human LRP1. 35

Figure 1-6 Selected members of the LDL receptor family in humans. 37

Figure 2-1 HepG2 monolayers require contact factors. 47

Figure 2-2 Characteristics of HepG2 in culture. 48

Figure 2-3 Evaluation of methods for harvesting HepG2. 54

Figure 2-4 Production and characterisation of recombinant RAP. 62

Figure 3-1 Fractionation of Ukidan into high and low molecular weight forms of two-chain uPA (HMW-uPA, LMW-uPA). 75

Figure 3-2 HMW-uPA and LMW-uPA form SDS-stable complexes with PAI-1 during binding to HepG2 cells. 78

Figure 3-3 Inhibition of 125I-HMW-uPA binding to HepG2 cells by anti-PAI-1 antibody. 79

Figure 3-4 Internalisation of 125I-PAPAI-1 complexes by HepG2 cells. 81

Figure 4-1 Plasma membrane phenotyping of HepG2 sublines. 93

Figure 4-2 125I-tPA binding to HepG2 sublines. 95

Figure 4-3 Accumulation of cellular 125I-tPA and release of degradation products into the supernatant by HepG2 sublines at 37C. 97

Figure 4-4 125I-RAP ligand blot of HepG2 subline lysates. 99

Figure 4-5 Pseumonas exotoxin A (PEA) sensitivity. 100

Figure 4-6 Western blots of HepG2 subline lysates. 101

Figure 4-7 Expression of LRP1 on HepG2 sublines as a function of time in culture. 103

x Figure 4-8 Expression of endocytic receptors on HepG2 sublines. 104

Figure 5-1 Serendipitous initial detection of soluble LRP1 using 125I-RAP. 117

Figure 5-2 Initial detection of soluble LRP1 in humans by using 125I-RAP ligand blotting. 119

Figure 5-3 Immunological identification, investigation of stability, and enrichment of soluble LRP1 in human serum. 121

Figure 5-4 Enrichment of soluble LRP1 from serum by ammonium sulfate (AS) cutting. 123

Figure 5-5 Capture of soluble LRP1 on immobilised RAP or activated 2-macroglobulin. 124

Figure 5-6 Platelets as a potential source of soluble LRP1. 127

Figure 5-7 Soluble LRP1 is released by HepG2a16 (a16/HP). 129

Figure 5-8 Cellular and soluble LRP1 association with extracellular matrix (ECM). 131

Figure 5-9 RAP-binding proteins in lacrimal fluid. 133

Figure 5-10 Soluble LRP1 in cerebral spinal fluid (CSF). 134

Figure 6-1 125I-RAP ligand blotting of soluble LRP1-like molecules from various species. 146

Figure 6-2 125I-Pseudomonas exotoxin A (PEA) ligand blots of RAP-2B serum extracts. 148

125 125 Figure 6-3 Activated I-2macroglobulin ( I-2MM) blotting of RAP-2B extracted animal sera. 149

Figure 6-4 Western blotting of RAP-2B serum extracts for LRP1 -chain. 151

Figure 6-5 Western blotting of RAP-2B serum extracts for LRP1 -chain. 153

Figure 6-6 Composition of soluble LRP1. 156

ABBREVIATIONS

2M* 2-macroglobulin activated by methylamine or proteases 2MR 2-macroglobulin receptor ACD Acid Citrate Dextrose AS ammonium sulfate BCA bicinchoninic acid (4,4‟-dicarboxy-2,2‟-biquinoline) BSA bovine serum albumin CVR Centre for Vascular Biology, UNSW DMEM Dulbecco‟s minimum essential medium DMSO dimethylsulfoxide DTT dithiothreitol EACA -amino-n-caproic acid, (6-amino-n-hexanoic acid) EDTA ethylenediaminetetraacetic acid EMEM Eagles minimum essential medium with Earle‟s salts FBS foetal bovine serum FITC fluorescein isothiocyanate FPLC Fast Performance Liquid Chromatography GaM-Ig-HRP HRP-conjugated goat anti-mouse immunoglobulins GaR-Ig-FITC FITC-conjugated goat anti-rabbit immunoglobulins GaR-Ig-HRP HRP-conjugated goat anti-rabbit immunoglobulins GST glutathione S-transferase from Salmonella japonicum HBSS Hanks balanced salt solution HEPES (N-[2-hydroxyethyl]piperazine-N‟-[2-ethanesulfonic acid]) HMW-uPA high molecular weight uPA HRP horseradish peroxidase Ig Immunoglobulin LDL R low density lipoprotein receptor LMW-uPA low molecular weight uPA LRP1 low density lipoprotein receptor-related protein 1 M199 Medium 199 with Earle‟s salts MES 2-[N-morpholino]ethanesulfonic acid PA(s) plasminogen activator(s) PBS Dulbecco‟s phosphate buffer saline PBSabc PBS with 0.1% azide, 0.1% albumin, and 0.5mM calcium PBSc PBS with 0.5mM calcium PEA Pseudomonas exotoxin A Plg plasminogen Plm plasmin PMSF phenylmethylsulfonylfluoride PPACK D-Phe-Pro-Arg choromethyl ketone hydrochloride RAP 39-kDa receptor associated protein RAP-CL2B rat-RAP conjugated to Sepharose CL-2B SaM-Ig-FITC FITC-conjugated sheep anti-mouse immunoglobulins SEC R serpin enzyme complex receptor

xii Serpin serine proteinase inhibitor TPA tissue-type plasminogen activator Tris 2-amino-2-(hydroxymethyl)-1,3-propanediol UNSW University of New South Wales, Sydney, Australia UPA urokinase-type plasminogen activator uPA R uPA receptor

xiii

ABSTRACT

Humans have two plasminogen activators (PAs), tissue-type plasminogen activator (tPA) and urokinase-type plasminogen activator (uPA), which generate plasmin to breakdown fibrin and other barriers to cell migration. Both PAs are used as pharmaceuticals but their efficacies are limited by their rapid clearance from the circulation, predominantly by parenchymal cells of the liver. At the commencement of the work presented here from the early 1990s, the hepatic receptors responsible for mediating the catabolism of the PAs were little understood. tPA degradation by hepatic cell lines was known to depend on the formation of binary complexes with the major PA inhibitor, plasminogen activator inhibitor type 1 (PAI-1). Initial studies presented here established that uPA was catabolised in a fashion similar to tPA by the hepatoma cell line, HepG2. Other laboratories at the time found that the major receptor mediating binding and endocytosis of the PAs was Low Density Lipoprotein Receptor-relate Protein (LRP1). LRP1 is a giant 600 kDa protein that binds a range of structurally and functionally diverse ligands including, activated alpha2 macroglobulin, apolipoproteins, beta amyloid precursor protein, and a number of serpinenzymes complexes, including PA·PAI-1 complexes. Further studies for the work presented here centred on this receptor. By using radiolabelled binding assays, ligand blots, and Western blots, the major findings are that: (1) basal LRP1 expression on HepG2 cells is low compared to a clone termed, HepG2a16; but appears to increase with passage number; (2) a soluble form of LRP1, which retains ligand binding capacity, is present in human circulation; (4) soluble LRP1 is also present in cerebral spinal fluid where its role in neurological disorders such as Alzheimer's disease is a developing area of interest; and (5) the release of soluble LRP1 is a mechanism conserved in evolution, possibly as distantly as molluscs. The identification and characterisation of soluble LRP1 introduces a possible further level of regulation for this essential receptor system.

xiv

ACKNOWLEDGEMENTS

Foremost, I wish to thank my supervisor, Dr Yuri Veniaminovich Bobryshev for his wisdom, support, and encouragement in the difficult end phase of this work. Thanks also must go to Prof Ken Ashwell and Dr Pascal Carrive for providing the opportunity to complete the writing of this work. Prof Graham Johnston at the University of Sydney has my warm gratitude for a job in his team for a period.

My fellow students and workmates, Kylie Hotchkiss, Janette Burgess and John Normyle require mention for the community they created. Special recognition in this group goes to Anthony Ashton and Paul Stathakis for our academic discussions, arguably from which stemmed the discovery of soluble LRP1 in the human circulation. Other folk who provided tangible support in terms of arranging writing retreats are Hilary Lloyd and my brother John. Their efforts are much appreciated. My friend Donald Davidson is a credit for continually grounding me in the real world.

I am grateful to my initial supervisor and co-supervisor, A/Prof Dwain Owensby and Prof Colin Chesterman for their teachings, and guidance, and for their provision of the all-essential materials, laboratory space, and finances.

To my children, Randal and Kara, thank you for your patience. I am proud of your achievements. Finally, to my wife, Leonie, I express my deepest appreciation for surviving the hardships, and for your support at all levels.

Chapter 1

1 GENERAL INTRODUCTION

1.1 OVERVIEW, AIMS, AND HYPOTHESES

Among the first recombinant proteins to become available as pharmaceuticals were the two human plasminogen activators (PAs), tissue-type plasminogen activator (tPA) and urokinase-type plasminogen activator (uPA). PAs convert the zymogen plasminogen to the fibrinolytic protease, plasmin, and while useful as thrombolytic agents, the efficacies of the PAs are limited by their short plasma half lives. Consequently, the mechanisms involved in their removal have generated interest. The liver is the primary site of sequestration of injected tPA, with hepatic parenchymal cells responsible for the majority of uptake and catabolism mediated by the large endocytic receptor, low density lipoprotein-receptor-related protein (LRP1).

The commonly available hepatic cell line, HepG2, was used in early studies by several groups to investigate the catabolism of tPA by a model of liver parenchymal cells (Bu et al., 1993; Bu et al., 1992a; Hajjar and Reynolds, 1994; Nguyen et al., 1992; Otter et al., 1989; Owensby et al., 1988). Derived from an hepatocellular carcinoma, HepG2 is unusual for well-differentiated hepatic cells in long-term culture in that they retains the ability to secrete several proteins synthesised by the liver in vivo, including the principal inhibitor of the PAs, plasminogen activator inhibitor type-1 (PAI-1). In this system, the degradation of tPA follows in a multistep process whereby tPA initially interacts with PAI-1 bound to vitronectin in the cell line‟s extracellular matrix (ECM). The resulting tPA●PAI-1 binary complex is enzymatically inactive and so tightly bound that it remains stable in the presence of low concentration of the denaturing detergent, sodium dodecyl sulfate (SDS). Following release from the ECM, the tPA●PAI-1 complex binds to HepG2 via a receptor which is saturable, inhibited by uPA, and later identified as LRP1.

LRP1 (the α2-macroglobulin receptor, type V TGFβ receptor, IGFBP receptor, and CD91 antigen) is an endocytic receptor for a broad range of structurally and functionally distinct ligands including lipoproteins containing apolipoprotein E (apoE), activated α2-macroglobulin (α2M*), plasminogen activators (PAs), and PAs complexed with their primary inhibitor, PAI-1. LRP1 is abundant in liver, lungs, and brain, and at 600kDa, is among the largest proteins synthesised as a single polypeptide. In addition to its role as a scavenger receptor, LRP1 also performs signalling functions in concert

17 with adaptor proteins and other receptors. Together, these properties involve LRP1 in a diversity of diseases including atherosclerosis, metastasis, and Alzheimer‟s disease.

Initial studies for the work presented here aimed to investigate the catabolism of uPA using HepG2. At commencement, the tPA receptor on HepG2 had not been identified and uPA binding to HepG2 had been inferred from inhibition studies, but not demonstrated directly. Accordingly, uPA binding to HepG2 was examined and the published results are presented in Chapter 3. In these studies, uPA was shown to bind to HepG2 predominately in the form of uPA●PAI-1 complex, and co-authors in our team reinforced these findings with degradation and inhibition studies. All findings were anticipated from analogous patterns of binding and catabolism previously known from tPA interactions with HepG2.

Control experiments examining tPA binding to HepG2 revealed a surprising observation. A collection of HepG2 batches obtained from various sources yielded a variation in the extent of tPA binding among these sublines. Formal comparisons of four sublines showed that the degree of tPA binding at 4C, and degradation at 37C, correlated with relative levels of LRP1 determined immunologically, and by the binding of other LRP1 ligands. Of note are the low levels of LRP1, and consequently tPA degradation capacity, possessed by the parental HepG2 cell line, which is readily available from the American Type Culture Collection (ATCC). These findings allowed selection of the optimal batch of HepG2 for further studies.

In considering reasons for the variability in constitutive levels of LRP1 among HepG2 sublines, shedding of the receptor from the cell surface seemed a plausible mechanism. If so, soluble LRP1 might be expected in HepG2 culture supernatant. Indeed, a soluble form of LRP1 was found, but the source was tracked to the foetal bovine serum supplementing the medium. Logical extension of this serendipitous finding led to the discovery of soluble LRP1 in the circulation of humans (Chapter 5) and other vertebrates (Chapter 6).

18 Hypotheses:

Chapter 3 To determine whether uPA binding and internalisation by HepG2 cells is analogous to the PAI-1 dependent mechanism previously determined for tPA.

Chapter 4 To determine whether the level of binding and endocytic activity of LRP1, the major receptor responsible for mediating tPA clearance, is stable on HepG2 cells kept in continual culture, and in various laboratories..

Chapter 5 To determine whether a soluble form of LRP1 circulates in human plasma.

Chapter 6 To determine whether the release of soluble LRP1 is a feature conserved in evolution.

1.2 THE PLASMINOGEN ACTIVATOR SYSTEM

1.2.1 Overview of the Plasminogen Activator System

Plasminogen activators (PAs) convert the circulating zymogen, plasminogen, to the S1 serine protease plasmin (Castellino, 1988). The breakdown of fibrin clots is traditionally considered to be the primary function of plasmin (Novokhatny, 2008; Zorio et al., 2008), yet additional roles in wound-healing (Hayashi et al., 2009), cell migration (Wygrecka et al., 2009), angiogenesis (Koutsioumpa et al., 2009), and cancer (Ulisse et al., 2009), reflect broader activity. By initiating the production of plasmin, the two human PAs, tissue-type PA (tPA) and urokinase-type PA (uPA), facilitate the breakdown of extracellular matrix (ECM) and basal membranes, allowing cellular migration in many normal and pathological processes including ovulation (Cao et al., 2006), fertilisation (Sa et al., 2006), wound healing (Lund et al., 2006; Toriseva and Kahari, 2009), fibrinolysis (Cesarman-Maus and Hajjar, 2005; Rijken and Lijnen, 2009), and cancer (Danø et al., 2005).

19 1.2.2 Components of Plasminogen Activator System

Plasminogen and the PAs, together with α2-antiplasmin (α2AP, the predominate inhibitor of the plasmin) (Carpenter and Mathew, 2008), plasminogen activator inhibitor type-1 (PAI-1, the major inhibitor of the PAs) (Agirbasli, 2005; Dellas and Loskutoff, 2005), and the specific receptor for uPA (uPAR) (D'alessio and Blasi, 2009), constitute the PA system (Mcmahon and Kwaan, 2008) (see Figure 1-1). Plasminogen activator type-2 (PAI-2) is usually included in the PA system despite its predominantly intracellular location where it is largely inaccessible to plasma PAs (Medcalf and Stasinopoulos, 2005), except during pregnancy (Brenner, 2004; Kruithof et al., 1988). Protein C inhibitor (also known as PAI-3) is abundant in urine and plasma (Heeb et al., 1987; Stief et al., 1987), and is sometimes considered part of the PA System.

The initial function of the PA system is the removal of barriers to cell migration in both normal and pathological situation (Rijken and Lijnen, 2009), often with the assistance of metalloproteinases (MMPs) (Lukes et al., 1999). In addition to the proteolytic roles, the PA system also initiates a number of signalling pathways following interaction of the uPAR with a range of proteins such as integrins, EGF receptor, high molecular weight kininogen, caveolin, FPRL1, and intracellular matrix components (D'alessio and Blasi, 2009). Other functions are also assigned to the inhibitors, PAI-1 and PAI-2, most notably in the setting of apoptosis (Balsara and Ploplis, 2008; Jensen et al., 1994; Rossignol et al., 2006; Schneider et al., 2008).

1.3 THE PLASMINOGEN

ACTIVATORS

Humans possess two structurally similar PAs, tissue-type plasminogen activator (tPA, EC 3.4.21.68) and urokinase-type plasminogen activator (uPA,

EC EC 3.4.21.73). Both are trypsin-like serine proteases (family S1) whose substrate Figure 1-1 The PA system. specificity is highly restricted towards a

20 single Arg-Val bond in plasminogen (Sottrup-Jensen et al., 1975). Cleavage of this bond results in activation of plasminogen to produce another S1 serine protease, plasmin. In contrast to the PAs, the substrate specificity of plasmin is broad and includes fibrin as well as proteins in the extracellular matrix and basement membranes (Ulisse et al., 2009).

While the PAs share plasminogen as a common substrate and have similar structures, their roles are largely divided spatially, with tPA functioning predominantly in the circulation as an initiator of fibrinolysis, and uPA operating mainly within the extracellular space within tissues where it is localised to cell surfaces at focal contact points (Mcmahon and Kwaan, 2008; Pollanen et al., 1988). However, gene knock-out studies of the PAs have demonstrated that these roles overlap. Disruption of both PAs in mice causes extensive fibrin deposition with associated organ failure, whereas deficiency of tPA and uPA individually is less severe (Carmeliet et al., 1994).

The kinetics of plasmin generation is regulated by the immobilisation of the PAs to specific cellular and extracellular sites, offering three effects. Firstly, the rate of conversion of plasminogen to plasmin is modulated by co-localisation and concentration of the reactants. Secondly, immobilisation restricts activity to appropriate physiological sites. In this regard, both tPA and plasmin bind in close proximity to annexin II on endothelial cells and on fibrin (Stein et al., 2009), in a process moderated by matrilysin (Tsunezumi et al., 2008), and dependent on S100A10/p11 (He et al., 2008). In the case of uPA, cellular localisation is mediated by the uPA receptor (Roldan et al., 1990), which concentrates uPA-mediated proteolysis to the leading edge of migrating cells (Bastholm et al., 1994). Thirdly, immobilisation offers both tPA and uPA relative protection from inactivation by inhibitors, and via proteolytic degradation (Hajjar et al., 1987; Shih and Hajjar, 1993; Wiman et al., 1979).

1.3.1 Tissue-type plasminogen activator (tPA)

1.3.1-1 tPA gene, transcription, and translation

Following isolation of its RNA (Pennica et al., 1983), the single gene for tPA on chromosome 8 (Rajput et al., 1985; Tripputi et al., 1986; Verheijen et al., 1986b) was found to span a continuous stretch of 32,720 kb consisting of 14 exons with 13 introns

21 (Browne et al., 1985; Fisher et al., 1985; Friezner Degen et al., 1986) whose boundaries correlate with the protein‟s structural domains (Ny et al., 1984). RNA for tPA contains both a signal peptide of 21 or 23-residues, and an 11 to 13-residue propeptide (Figure 1-2). Both peptides are removed prior to secretion, but their exact length varies, leading to alternate starting peptides (gly33 or ser36) in the mature protein (Jörnvall et al., 1983). Two isoforms (long and short) of the protein exist as the result of alternative splicing with the short isoform missing the final 271 residues and differing by 23 amino acids at the carboxy terminus (Siebert and Fong, 1990).

1.3.1-2 tPA Biochemistry

A representation of the long form of the human protein based on the 562 amino acid sequence from the Swiss-Prot database (entry P00750) and the secondary structure of Ny et al (Ny et al., 1984) is shown in (Figure 1-2).

Further reference to tPA throughout this text is restricted to the long isoform. Consistent with the Swiss-Prot database, numbering of residues begins at the N-terminus of the full-length translated protein. Residue numbers from literature using the chymotrypsin numbering system (Schmidt et al., 2008) has been re-calculated for human tPA where necessary.

Unmodified, tPA has a calculated Mr of 62,917 and an actual Mr of ~70,000 when glycosylated (Pennica et al., 1983). From its N-terminus, tPA contains, a fibronectin type-1 finger domain (F), an epidermal growth factor-like domain (GF), two kringle units (K1 and K2), and a serine protease (SP). These domains largely correlate with the exon/intron boundaries, lending support to the concept that exon shuffling has been involved in the evolution of tPA (Patthy, 1985; Rogers, 1985).

The finger domain is homologous to fibrin-binding structures in fibronectin (Bányai et al., 1983) and is responsible, together with a greater contribution from K2, for the fibrin-binding properties of tPA (Ichinose et al., 1986b; Van Zonneveld et al., 1986a; Van Zonneveld et al., 1986c; Van Zonneveld et al., 1986b; Verheijen et al., 1986a). The GF domain is homologous to receptor-binding regions in epidermal growth factor (EGF) and transforming growth factor (TGF) (Derynck et al., 1984; Komoriya et al., 1984). This structure is found in many proteins where it usually

Figure 1-2 The primary sequence and secondary structure of tPA.

functions in direct protein-protein binding, or in the regulation of binding. For example, analogous to the binding of EGF and TGF to their respective receptors, the GF domain in uPA is responsible for the direct binding of uPA to the uPA receptor (Appella et al., 1987). In contrast, the GF domains in the low density lipoprotein (LDL) receptor regulate the release of internalised ligands (Hussain, 2001). Despite these well- characterised properties, the function of the GF domain in tPA remains unknown.

23 However, it is interesting to note that the receptor-binding region in the GF region of uPA differs markedly from the corresponding sequence in tPA (Belin et al., 1985), and this difference offers an explanation for the specificity of uPA over tPA for the uPA receptor.

The two kringle units in tPA are analogous to five similar domains in plasminogen, two in prothrombin, one in factor XII (Mcmullen and Fujikawa, 1985), four in hepatocyte growth factor (Nakamura et al., 1989), and up to thirty eight in apolipoprotein(a) (Mclean et al., 1987). These structural units are typically composed of six disulfide bonds arranged in the order C1-C6, C2-C4, C3-C5. While K2 in tPA is responsible for the majority of the protein‟s fibrin binding properties (Van Zonneveld et al., 1986b; Verheijen et al., 1986a), K1 has no known function. Lysine residues on fibrin (Christensen, 1985) mediate the kringle-dependent binding to both plasminogen (Fleury and Angles-Cano, 1991; Wiman and Wallen, 1977) and tPA (De Munk et al., 1989; Ichinose et al., 1986b; Radcliffe, 1983; Van Zonneveld et al., 1986b; Verheijen et al., 1986a) via lysine binding sites (LBSs). The basic components of LBSs are present in K1 of tPA and the single kringle of uPA, but their failure to bind fibrin is due to minimal sequence changes (Bürgin and Schaller, 1999).

The lysine binding properties of K2 are exploited in some tPA purification protocols which use lysine immobilised on chromatographic supports (Radcliffe and Heinze, 1978). Release of tPA from lysine-sepharose, and indeed from fibrin, can be achieved by the addition of lysine analogues such as -amino-n-caproic acid (EACA). This form of binding can be undesirable when investigating cellular receptor interactions. Accordingly, to circumvent the complication of additional binding via kringles, all binding experiments involving PAs presented in Chapters 3 and 4 were performed in the presence of 10 mM EACA.

The serine protease domain at the C-terminus is typical of similar enzymes (Strassburger et al., 1983) and contains a catalytic triad formed by his-357, asp-406, and ser-513. The protein is synthesised as a single-chained polypeptide (sc-tPA) which can be cleaved between arg-310 and ile-311 to form two-chained tPA (tc-tPA). An unusual feature of tPA among trypsin-like serine proteases is the enzymatic capacity of the uncleaved form (Lamba et al., 1996). Unlike pro-uPA, which has little enzymatic activity, both sc-tPA and tc-tPA readily activate plasminogen.

24 Three sites (152 in K1, 219 in K2, and 483 in the protease) carry variable N-glycosylation and some degree of sulfation (Pfeiffer et al., 1989). In addition, threonine 96 has an O-linked fucose (Harris et al., 1991). In early studies on the catabolism of tPAs by the liver, clearance mediated by a carbohydrate receptor presented a possibility. Although N-glycosylation patterns affect the kinetics of tPA clearance by cell lines, and from the circulation of animals (Bassel-Duby et al., 1992; Beebe and Aronson, 1988), both the galactose receptor on parenchymal cells and the mannose receptor on hepatic endothelial were initially excluded from involvement (Stang et al., 1992), as was fucosylation (O-glycosylation) of the finger region (Camani et al., 1998). These findings have some relevance to the results presented here, because glycosylated Human tPA produced from transfected Chinese hamster ovary cells was used for the studies in Chapter 4.

1.3.2 Urokinase-type plasminogen activator (uPA)

The results in Chapter 3 were generated from experiments examining the interactions of high (HMW) and low molecular mass (LMW) forms of uPA with HepG2 cells. The following brief introduction to the background of uPA refers to the positions of amino acid residues in uPA as they appear in the Swiss-Prot database (Accession number P00749). This numbering system begins with the signal peptide and differs from positions quoted in early literature by twenty.

The single gene for uPA is much smaller than the gene for tPA. Located on human chromosome 10 (Rajput et al., 1985; Tripputi et al., 1985) and spanning 6.4-kb, the 11 exons (Riccio et al., 1985; Verde et al., 1988) produce a 2.5 kb mRNA (Verde et al., 1984) and a protein with a predicted molecular mass of 48.5 kDa. Following secretion as a single-chained pro-enzyme (sc-uPA, or pro-uPA), the mature glycoprotein is reported to have a molecular mass in the range 47-56 kDa (Gunzler et al., 1982b).

The structure of uPA is similar to tPA except that the corresponding finger and kringle 2 domains of tPA are missing in uPA (Figure 1-3). In tPA, these domains endow fibrin-binding capacity, and their absence in uPA may be expected to exclude this property. pro-uPA requires cleavage to a two-chained form (tc-uPA) to attain full enzymatic activity. Both pro-uPA and tc-uPA bind to a specific cell surface receptor,

Figure 1-3 Cleavage sites on uPA.

the uPA receptor (uPAR), via a sequence in the N-terminal region of the growth factor domain. By localising on cell surfaces via the uPAR, uPA normally operates outside the vascular compartments. However, tPA knock-out mice display only mild thrombotic episodes, suggesting that uPA can also act as an effective fibrinolytic agent (Carmeliet et al., 1994).

Unlike sc-tPA, pro-uPA possesses little enzymatic activity (Blasi et al., 1987; Eaton et al., 1984; Ichinose et al., 1986a; Kasai et al., 1985; Pannell and Gurewich, 1987) prior to its cleavage between lys-178 and ile-179 (Figure 1-3). During the activation process the following bond (phe-177 to lys-178) is also cleaved resulting in the loss of lys-178 (Verde et al., 1984). Plasmin (Kasai et al., 1985), and to a lesser extent kallikrein (Frenette et al., 1997; Ichinose et al., 1986a), are capable of activating uPA in this way. The resulting two-chained uPA (tc-uPA) is a disulfide-bonded heterodimer consisting of an A-chain (formed from the GF and kringle domains), and a B-chain (composed mainly of the serine protease domain). Interestingly, thrombin hydrolyses the bond immediately preceding the activation site (arg-176 and phe-177), resulting in the complete inactivation of uPA (Ichinose et al., 1986a; Van Hinsberg et al., 1987).

Further processing by plasmin cleaves the A-chain between lys-155 and lys-156, resulting in the formation of a low molecular mass form of tc-uPA (LMW tc-uPA). This protein consists of the serine protease domain disulfide-bonded to a small residual

26 peptide from the A-chain (Gunzler et al., 1982a; Gunzler et al., 1982b). A similar molecule is secreted by a cell line (Stump et al., 1986). However, in this case the enzyme responsible for cleavage appeared not to be plasmin as the cleavage site was distinct (glu-163 – leu-164). Furthermore, the activation site was uncleaved so that the molecule remains single chained (LMW sc-uPA). The reports of these LMW forms of uPA are relevant here for two reasons. Firstly, they raise the possibility of the physiological importance of LWM forms of uPA. Secondly, a LMW form of uPA is present in the expired pharmaceutical preparations used as the starting materials for the work reported in Chapter 3.

1.4 SERINE PROTEINASES

PAs are classed as S1 proteinases (trypsin-like serine proteinases) on the basis of evolutionary and mechanistic features determined by Barrett and Rawlings (Rawlings and Barrett, 1993). Under this system, other serine proteases have been grouped into more than 25 other families. Remaining proteinases are categorised into the threonine-, cysteinyl-, aspartyl-, and metallo-, peptidase families, or peptidases of unknown catalytic mechanism. Serine proteinases operate by a charge relay system between the serine at the active site and a set of hydrogen-bonded histidine and aspartic acid residues (Bachovchin et al., 1981). These three amino acids form a catalytic triad, the locations of which are well conserved among the family. An example of the relative positions of these amino acids is shown in the representation of tPA in Figure 1-1. Other serine proteinases in family S1 include: (a) in the coagulation and fibrinolytic systems, kallikrein, thrombin, plasminogen, protein C, and Factors VII, IX, X, XI, and XII; (b) in the complement system, factors B, D, and I, and components C1r, C1s, and C2; (c) in neutrophils, elastase and cathepsin G; (c) in pancreatic fluid, trypsin and chymotrypsin; (d) among prohormone activators, hepatocyte growth factor activator; and (e) the unusual lipid associated protein, apolipoprotein(a) (Rawlings and Barrett, 1994).

1.4.1 Serpins, and serpin enzyme complexes (SECs)

PAI-1 and PAI-2 are members of the serine protease inhibitor (serpin) family of homologous proteins whose biological distribution extends to plants and viruses

27 (Gettins et al., 1992; Potempa et al., 1994; Schulze et al., 1994). Examples of serpins and their common targets are listed in Table 1-1. In plasma, serpins are important regulators of enzymes involved in the coagulation, fibrinolytic, and complement cascade, as well as enzymes released from neutrophils. As the family name suggests, many serpins are inhibitors of serine proteases. However, others, such as ovalbumin and angiotensinogen, serve other functions such as lipophilic molecule transporters and peptide hormone precursors rather than inhibitors (Gettins, 1989; Stein et al., 1990; Stein et al., 1989) but their sequence and structural homologies clearly place them in the same group.

Serpins characteristically contain a 30-40 amino acid reactive site loop (RSL, also called the reactive centre loop) in an exposed structure at their C-termini (Carrell and Owen, 1985). Sequences within this loop are recognised as normal substrates by target serine proteases, and upon interaction, serpins and their cognate enzymes form 1:1 covalent complexes (serpin enzyme complexes, SECs), which resist dissociation in boiling solutions of sodium dodecyl sulfate. Complex formation eventually results in the cleavage of the serpin at its P1-P1‟ site. The small C-terminal fragment produced inserts into one of the three -sheets (5-stranded -sheet A) of the serpin where it becomes strand 4 of an antiparallel 6-stranded -sheet. This process induces a large conformation changes in the serpin (Stratikos and Gettins, 1998) resulting a transition from a stressed (S) to a relaxed (R) state. This S  R transition is accompanied by the exposure of cryptic epitopes which are important in the binding of SECs to their clearance receptors (Knauer et al., 1999; Stefansson et al., 1998). In addition, structural changes can also occur in the enzyme.

During normal serine protease activity, the enzyme and substrate form a tetrahedral intermediate complex involving the hydrolytic water molecule. This structure is followed by an acyl intermediate formed after cleavage of the substrate. The nature of the stable serpinenzyme complexes has been the subject of debate, a full account of which is beyond the scope of this discussion. In brief, evidence suggests that SECs are stabilised in either a tetrahedral intermediate (Matheson et al., 1991), an acyl intermediate (Stratikos and Gettins, 1998; Stratikos and Gettins, 1999), or both (Nair and Cooperman, 1998). The acyl intermediate is associated with cleavage of the serpin and translocations up to 70Å or more (Stratikos and Gettins, 1999). However, existence

28 Table 1-1 Examples of serpins and their common target serine proteases.

Serpin Common Abbreviation of resulting Selected target Serpin Enzyme Complex Reference (SEC)

PAI-1 tPA. UPA tPAPAI-1, uPAPAI-1 PAI-2 tPA. UPA tPAPAI-2, uPAPAI-2 Protein C inhibitor Activated protCprotCI, IIaprotCI, (Suzuki et al., (also called PAI-3) protein C, FXaprotCI, FXIaprotCI, 1987) thrombin, kallikreinprotCI Factor Xa, Factor XIa, kallikrein Antithrombin III Thrombin IIaATIII Heparin cofactor II Thrombin IIaHCII (Tollefsen, 1995)

2-antiplasmin Plasmin, Plm2AP, CT2AP (Enghild et chymotrypsin al., 1994)

1-antitrypsin (also Elastase, El1-AT, Tryp1-AT, (Poller et al., called 1-proteinase Trypsin, ProtC1-AT 1999) inhibitor) activated protein C

1-antichymotrypsin Cathepsin G, CG1ACT, CT1ACT Chymotrypsin C1 inhibitor C1s, C1sCI, kallikrienCI, (Eldering et Kallikrien, FXIIaCI al., 1993; Factor XIIa Eldering et al., 1995) Protease nexin I Thrombin, IIaPN1, uPAPN1 (Low et al., uPA 1981) neuroserpin PAs, plasmin PANS, plmNS (Osterwalder et al., 1998) Angiotensinogen Non-inhibitory - (Stein et al., 1989) Ovalbumin Non-inhibitory - (Remold- O'donnell, 1993; Stein et al., 1990; Stein et al., 1989)

29 of the tetrahedral structure is supported by the observations that dissociation of SECs can lead to the release of serpins which retain inhibitory activity (Shieh et al., 1989), and by the ability of chemically inactivated serine proteases to form weak complexes with serpins (Enghild et al., 1994). Direct evidence for a tetrahedral structure was provided by Matheson et al (Matheson et al., 1991) using 13C NMR analysis of underivatised complexes in solution. Attempts to determine the structure of native SECs is difficult because of the fastidious conformational requirements of the serpin which may be sensitive to minor, artificial, alterations in the protein. Although the acyl intermediate readily explains many of the stability and conformational properties of SECs, their existence in vitro has been suggested to be artefacts resulting from in vitro processing (Potempa et al., 1994). This question of SEC structure has some relevance to the necessity for a clearance mechanism because dissociation of these two proposed intermediates, tetrahedral or acyl, leads to the predicted release of active or inactive serpins, respectively. Given that complex-free, inactive serpins are cleared more slowly than their active counterparts, the dissociation of cleaved serpins from SECs potentially could lead to their accumulation in the absence of a clearance mechanism.

The SECs as a group are an important consideration for the work presented here because, LRP1 (the receptor investigated in Chapters 4, 5, and 6) mediates the clearance of several SECs, additional to uPAPAI-1 and tPAPAI-1 which are investigated in Chapters 3 and 4.

1.4.2 Clinical uses of plasminogen activators

tPA has been used for the treatment of myocardial infarction (Bottiger et al., 2008) for many years and, despite concerns about neurotoxicity, has been approved for use in stroke in the USA and Canada (Chapman et al., 2000). In addition to cardiovascular conditions, tPA is used to complement eye surgery following the removal of cataracts (Mehta and Adams, 2000), and in the displacement of clots from submacular haemorrhage (Fang et al., 2009; Haupert et al., 2001). Urokinase (uPA) as the pharmaceutical, Abbokinase, was recalled in 2000 by the U.S. Food and Drug Administration (Hartnell and Gates, 2000), but has returned to the market (Traynor, 2002). It is applied mainly in pulmonary embolism (Gupta and Gupta, 2008) and the prevention of clots around catheters (Kethireddy and Safdar, 2008).

30 tPA is fibrin-specific, whereas uPA is non-fibrin specific. While this gives an advantage to tPA in terms of targeting fibrin clots, tPA also causes lysis of pre-formed hemostatic plugs, leading to haemorrhage (Perler, 2005).

tPA and uPA therapeutic agents are human proteins, giving them an advantage over the alternate (cheaper) recombinant bacterial (and antigenic) thrombolytic agents, streptokinase (Kunamneni et al., 2007) and staphylokinase (Ueshima and Matsuo, 2006). Streptokinase (SK) activates plasminogen to attack fibrin by altering the zymogen‟s conformation after forming a 1:1 complex (Boxrud et al., 2001). Staphylokinase (Sak) also forms 1:1 complexes with plasminogen, but in activating plasminogen‟s dormant serine protease, Sak imposes specificity for other plasminogen molecules, thereby cascading its activation (Jespers et al., 1998).

1.4.3 Clearance as a PA system regulator

The production of plasmin is tightly regulated by action of the PAs (Rånby and Brändström, 1988). In turn, the activity of the PAs is regulated by their rate of synthesis (Roychoudhury et al., 2006), posttranslational modification (Okajima et al., 2008), and the presence of plasminogen activator inhibitors (PAIs) (Kruithof, 1988). As a final regulator, PAs inactivated by inhibitors are rapidly cleared via endocytic receptors (Mattsson, 1988). In the case of PAsPAIs in circulation, sequestration is predominantly mediated by the liver (Otter et al., 1992b).

1.4.4 Importance of PA clearance

PAs and PAI-1 form tight 1:1 complexes (PAPAI-1), which are enzymatically inactive. However, these complexes are capable of re-releasing active PAs (Shieh et al., 1989) with potentially adverse systemic consequences. Their rapid removal ensures normal regulation of the PA system. The contribution of PA clearance has increased importance in pathological situations involving perturbations of the PA system, such as tumour metastasis (Danø et al., 1988) and atherosclerosis (Alessi and Juhan-Vague, 2008). Strategies for treating these diseases may target the dynamics of the PAs removal.

31 PA clearance has direct clinical relevance with the emergence of both tPA (Simpson et al., 2006) and uPA (Kunamneni et al., 2008; Ouriel, 1999) as pharmaceuticals for the dissolution of fibrin clots in myocardial infarction (Bottiger et al., 2008; Turcasso and Nappi, 2001), and ischaemic stroke (Broderick, 2009; Liang et al., 2008; Yepes et al., 2009). However, their rapid clearance results in short plasma half lives (Korninger et al., 1981; Sobel, 1988) which limits their efficacy. Methods devised to prolong the presence of these administered agents in vivo may be desirable for many reasons, one of which is cost.

The injection of exogenous PAs presents the in vivo clearance mechanisms with an unusually high burden. However, the system apparently can adapt and cope. There is a massive release of endogenous tPA after electroshock treatment. In this situation, immediately-available PAI-1 is overwhelmed, but protease inhibition and clearance is assisted by additional inhibitors, such as 2-macroglobulin (Bennett et al., 1990).

1.4.5 Mechanism of Plasminogen Activators Clearance

In the early 1980s, it was established that exogenously (injected) tPA spends little time in the circulation (T1/2 ~2 minutes for rabbits), and that the liver was its major site of sequestration in vivo (Korninger et al., 1981). By the end of that decade, the mechanism of tPA clearance had been elaborated to the processes depicted in Figure 1-4 (Morton et al., 1989; Owensby et al., 1989). In this system, (1) tPA initially interacts with PAI-1 bound to vitronectin in the extracellular matrix to form a tightly- bound complex of tPAPAI-1 (Owensby et al., 1991). This complex is then (2) released into the supernatant from where (3) it binds to a cell-surface receptor. At 37C, (4) the receptor, along with its ligand cargo, is internalised into endosomes. Upon release of the ligand, (5) the receptor recycles to the cell surface, while tPAPAI-1 is delivered to lysosomes for (6) degradation and subsequent secretion of the fragments into the culture supernatant.

Following elucidation of the clearance mechanism for tPA, the clearance of uPA was also shown to require analogous processing via a similar binding site (Grimsley et al., 1995) (results also reported in Chapter 3). These studies used the hepatocellular adherent cell line HepG2 which is unusual for an hepatic cell in that it retains many

Figure 1-4 PAI-1 dependent catabolism of tPA by HepG2.

features of well-differentiated hepatic parenchymal cells including the ability to secrete many proteins (Aden et al., 1979; Knowles and Aden, 1983; Knowles et al., 1980), fortuitously including PAI-1 (Wun et al., 1989).

The work presented here initially addresses uPA clearance by HepG2 cells (Chapter 3). It commences with the demonstration, from the early 1990‟s, that like tPA, uPA is also catabolised by HepG2 in a PAI-1 dependent fashion (Grimsley et al., 1995). Other laboratories at the time supported these findings by showing that both tPAPAI-1 (Bu et al., 1992b) and uPAPAI-1 complexes (Kounnas et al., 1993b; Nykjær et al., 1992) bind to the giant (600kDa) endocytic receptor, low density lipoprotein receptor- related protein (LRP1) (Herz et al., 1988). LRP1 had been simultaneously identified and isolated as the α2-macroglobulin receptor (α2-mMR) (Ashcom et al., 1990; Jensen et al., 1989) and later classified as the CD91 antigen (Schlossman et al., 1994). LRP1 is often termed LRP (especially in older literature), but for consistency within the LDL receptor family, LRP is now termed LRP1 (Lillis et al., 2008). HepG2 were shown to

33 express LRP1 (Kounnas et al., 1993b), a finding which is consistent with the results in Chapter 3.

The specific receptor for uPA is present on the surface of many migrating cells including monocytes (Plesner et al., 1994; Wang, 2001) and fibroblasts (Hildenbrand and Schaaf, 2009). This 3 domain receptor is GPI-linked to the cell surface (Casey et al., 1994) and as such does not posses a transmembrane region required for cellular internalisation. So, it was surprising that uPA internalisation could be mediated by uPAR (Olson et al., 1992). This result added a level of complexity given that uPAPAI-1 complexes were known to be internalised by binding to LRP1 (Nykjær et al., 1992). Subsequent studies solved the mystery by showing that uPAR bearing uPA was internalised via association with LRP1 (Conese et al., 1994).

LRP1 is expressed on HepG2 cells (Kounnas et al., 1993b). Preliminary studies in our laboratory revealed that the particular HepG2 batch which we maintained, expressed LRP1 at much lower levels than expected. This apparent discrepancy prompted a survey of comparative LRP1 protein levels present in HepG2 cell lines gathered from various laboratories. The findings (Grimsley et al., 1997), which are presented in Chapter 4, show that the relative levels of LRP1 expressed in HepG2 varies considerably among sublines held in various locations. This work allowed the subsequent confident selection of a subline expressing high levels of LRP1 for further studies.

1.5 LOW DENSITY LIPOPROTEIN RECEPTOR-RELATED

PROTEIN (LRP1)

1.5.1 Historic Perspectives and Nomenclature

The 1985 Nobel prize awarded to Brown and Goldstein recognised their contribution to the characterisation of the low density lipoprotein (LDL) receptor, mutations of which are a leading cause of familial hyperlipidemia (Brown and Goldstein, 1986). With the importance of this receptor well established, Herz et al. (Herz et al., 1988) searched for homologous proteins, and isolated cDNA coding for a large endocytic receptor, which they termed the low density lipoprotein receptor-related

34 protein (LRP) (see sequence and domain assignments in Figure 1-5). The name reflects both the method of discovery (screening for proteins homologous to the LDL receptor), and its actual sequence, which indeed displays striking homology to the LDL receptor. The major structural difference is the size of LRP, which is some four times larger than the LDL receptor. The sequence similarities alone were sufficient to predict that LRP would probably bind and endocytose lipoproteins, and this ability was soon demonstrated for apoE-containing lipoprotein (Beisiegel et al., 1989). Two other giant protein with similar structures to LRP have been recognised as belonging the same family of lipoprotein receptors, LRP1b (Liu et al., 2000) and LRP2/megalin (Kounnas et al., 1994; Moestrup, 1994). Consequently, LRP is now referred to as LRP1, and

1 MLTPPLLLLL PLLSALVAAA IDAPKTCSPK QFACRDQITC ISKGWRCDGE RDCPDGSDEA PEICPQSKAQ RCQPNEHNCL GTELCVPMSR LCNGVQDCMD 101 GSDEGPHCRE LQGNCSRLGC QHHCVPTLDG PTCYCNSSFQ LQADGKTCKD FDECSVYGTC SQLCTNTDGS FICGCVEGYL LQPDNRSCKA KNEPVDRPPV 201 LLIANSQNIL ATYLSGAQVS TITPTSTRQT TAMDFSYANE TVCWVHVGDS AAQTQLKCAR MPGLKGFVDE HTINISLSLH HVEQMAIDWL TGNFYFVDDI 301 DDRIFVCNRN GDTCVTLLDL ELYNPKGIAL DPAMGKVFFT DYGQIPKVER CDMDGQNRTK LVDSKIVFPH GITLDLVSRL VYWADAYLDY IEVVDYEGKG 401 RQTIIQGILI EHLYGLTVFE NYLYATNSDN ANAQQKTSVI RVNRFNSTEY QVVTRVDKGG ALHIYHQRRQ PRVRSHACEN DQYGKPGGCS DICLLANSHK 501 ARTCRCRSGF SLGSDGKSCK KPEHELFLVY GKGRPGIIRG MDMGAKVPDE HMIPIENLMN PRALDFHAET GFIYFADTTS YLIGRQKIDG TERETILKDG 601 IHNVEGVAVD WMGDNLYWTD DGPKKTISVA RLEKAAQTRK TLIEGKMTHP RAIVVDPLNG WMYWTDWEED PKDSRRGRLE RAWMDGSHRD IFVTSKTVLW 701 PNGLSLDIPA GRLYWVDAFY DRIETILLNG TDRKIVYEGP ELNHAFGLCH HGNYLFWTEY RSGSVYRLER GVGGAPPTVT LLRSERPPIF EIRMYDAQQQ 801 QVGTNKCRVN NGGCSSLCLA TPGSRQCACA EDQVLDADGV TCLANPSYVP PPQCQPGEFA CANSRCIQER WKCDGDNDCL DNSDEAPALC HQHTCPSDRF 901 KCENNRCIPN RWLCDGDNDC GNSEDESNAT CSARTCPPNQ FSCASGRCIP ISWTCDLDDD CGDRSDESAS CAYPTCFPLT QFTCNNGRCI NINWRCDNDN 1001 DCGDNSDEAG CSHSCSSTQF KCNSGRCIPE HWTCDGDNDC GDYSDETHAN CTNQATRPPG GCHTDEFQCR LDGLCIPLRW RCDGDTDCMD SSDEKSCEGV 1101 THVCDPSVKF GCKDSARCIS KAWVCDGDND CEDNSDEENC ESLACRPPSH PCANNTSVCL PPDKLCDGND DCGDGSDEGE LCDQCSLNNG GCSHNCSVAP 1201 GEGIVCSCPL GMELGPDNHT CQIQSYCAKH LKCSQKCDQN KFSVKCSCYE GWVLEPDGES CRSLDPFKPF IIFSNRHEIR RIDLHKGDYS VLVPGLRNTI 1301 ALDFHLSQSA LYWTDVVEDK IYRGKLLDNG ALTSFEVVIQ YGLATPEGLA VDWIAGNIYW VESNLDQIEV AKLDGTLRTT LLAGDIEHPR AIALDPRDGI 1401 LFWTDWDASL PRIEAASMSG AGRRTVHRET GSGGWPNGLT VDYLEKRILW IDARSDAIYS ARYDGSGHME VLRGHEFLSH PFAVTLYGGE VYWTDWRTNT 1501 LAKANKWTGH NVTVVQRTNT QPFDLQVYHP SRQPMAPNPC EANGGQGPCS HLCLINYNRT VSCACPHLMK LHKDNTTCYE FKKFLLYARQ MEIRGVDLDA 1601 PYYNYIISFT VPDIDNVTVL DYDAREQRVY WSDVRTQAIK RAFINGTGVE TVVSADLPNA HGLAVDWVSR NLFWTSYDTN KKQINVARLD GSFKNAVVQG 1701 LEQPHGLVVH PLRGKLYWTD GDNISMANMD GSNRTLLFSG QKGPVGLAID FPESKLYWIS SGNHTINRCN LDGSGLEVID AMRSQLGKAT ALAIMGDKLW 1801 WADQVSEKMG TCSKADGSGS VVLRNSTTLV MHMKVYDESI QLDHKGTNPC SVNNGDCSQL CLPTSETTRS CMCTAGYSLR SGQQACEGVG SFLLYSVHEG 1901 IRGIPLDPND KSDALVPVSG TSLAVGIDFH AENDTIYWVD MGLSTISRAK RDQTWREDVV TNGIGRVEGI AVDWIAGNIY WTDQGFDVIE VARLNGSFRY 2001 VVISQGLDKP RAITVHPEKG YLFWTEWGQY PRIERSRLDG TERVVLVNVS ISWPNGISVD YQDGKLYWCD ARTDKIERID LETGENREVV LSSNNMDMFS 2101 VSVFEDFIYW SDRTHANGSI KRGSKDNATD SVPLRTGIGV QLKDIKVFNR DRQKGTNVCA VANGGCQQLC LYRGRGQRAC ACAHGMLAED GASCREYAGY 2201 LLYSERTILK SIHLSDERNL NAPVQPFEDP EHMKNVIALA FDYRAGTSPG TPNRIFFSDI HFGNIQQIND DGSRRITIVE NVGSVEGLAY HRGWDTLYWT 2301 SYTTSTITRH TVDQTRPGAF ERETVITMSG DDHPRAFVLD ECQNLMFWTN WNEQHPSIMR AALSGANVLT LIEKDIRTPN GLAIDHRAEK LYFSDATLDK 2401 IERCEYDGSH RYVILKSEPV HPFGLAVYGE HIFWTDWVRR AVQRANKHVG SNMKLLRVDI PQQPMGIIAV ANDTNSCELS PCRINNGGCQ DLCLLTHQGH 2501 VNCSCRGGRI LQDDLTCRAV NSSCRAQDEF ECANGECINF SLTCDGVPHC KDKSDEKPSY CNSRRCKKTF RQCSNGRCVS NMLWCNGADD CGDGSDEIPC 2601 NKTACGVGEF RCRDGTCIGN SSRCNQFVDC EDASDEMNCS ATDCSSYFRL GVKGVLFQPC ERTSLCYAPS WVCDGANDCG DYSDERDCPG VKRPRCPLNY 2701 FACPSGRCIP MSWTCDKEDD CEHGEDETHC NKFCSEAQFE CQNHRCISKQ WLCDGSDDCG DGSDEAAHCE GKTCGPSSFS CPGTHVCVPE RWLCDGDKDC 2801 ADGADESIAA GCLYNSTCDD REFMCQNRQC IPKHFVCDHD RDCADGSDES PECEYPTCGP SEFRCANGRC LSSRQWECDG ENDCHDQSDE APKNPHCTSP 2901 EHKCNASSQF LCSSGRCVAE ALLCNGQDDC GDSSDERGCH INECLSRKLS GCSQDCEDLK IGFKCRCRPG FRLKDDGRTC ADVDECSTTF PCSQRCINTH 3001 GSYKCLCVEG YAPRGGDPHS CKAVTDEEPF LIFANRYYLR KLNLDGSNYT LLKQGLNNAV ALDFDYREQM IYWTDVTTQG SMIRRMHLNG SNVQVLHRTG 3101 LSNPDGLAVD WVGGNLYWCD KGRDTIEVSK LNGAYRTVLV SSGLREPRAL VVDVQNGYLY WTDWGDHSLI GRIGMDGSSR SVIVDTKITW PNGLTLDYVT 3201 ERIYWADARE DYIEFASLDG SNRHVVLSQD IPHIFALTLF EDYVYWTDWE TKSINRAHKT TGTNKTLLIS TLHRPMDLHV FHALRQPDVP NHPCKVNNGG 3301 CSNLCLLSPG GGHKCACPTN FYLGSDGRTC VSNCTASQFV CKNDKCIPFW WKCDTEDDCG DHSDEPPDCP EFKCRPGQFQ CSTGICTNPA FICDGDNDCQ 3401 DNSDEANCDI HVCLPSQFKC TNTNRCIPGI FRCNGQDNCG DGEDERDCPE VTCAPNQFQC SITKRCIPRV WVCDRDNDCV DGSDEPANCT QMTCGVDEFR 3501 CKDSGRCIPA RWKCDGEDDC GDGSDEPKEE CDERTCEPYQ FRCKNNRCVP GRWQCDYDND CGDNSDEESC TPRPCSESEF SCANGRCIAG RWKCDGDHDC 3601 ADGSDEKDCT PRCDMDQFQC KSGHCIPLRW RCDADADCMD GSDEEACGTG VRTCPLDEFQ CNNTLCKPLA WKCDGEDDCG DNSDENPEEC ARFVCPPNRP 3701 FRCKNDRVCL WIGRQCDGTD NCGDGTDEED CEPPTAHTTH CKDKKEFLCR NQRCLSSSLR CNMFDDCGDG SDEEDCSIDP KLTSCATNAS ICGDEARCVR 3801 TEKAAYCACR SGFHTVPGQP GCQDINECLR FGTCSQLCNN TKGGHLCSCA RNFMKTHNTC KAEGSEYQVL YIADDNEIRS LFPGHPHSAY EQAFQGDESV 3901 RIDAMDVHVK AGRVYWTNWH TGTISYRSLP PAAPPTTSNR HRRQIDRGVT HLNISGLKMP RGIAIDWVAG NVYWTDSGRD VIEVAQMKGE NRKTLISGMI 4001 DEPHAIVVDP LRGTMYWSDW GNHPKIETAA MDGTLRETLV QDNIQWPTGL AVDYHNERLY WADAKLSVIG SIRLNGTDPI VAADSKRGLS HPFSIDVFED 4101 YIYGVTYINN RVFKIHKFGH SPLVNLTGGL SHASDVVLYH QHKQPEVTNP CDRKKCEWLC LLSPSGPVCT CPNGKRLDNG TCVPVPSPTP PPDAPRPGTC 4201 NLQCFNGGSC FLNARRQPKC RCQPRYTGDK CELDQCWEHC RNGGTCAASP SGMPTCRCPT GFTGPKCTQQ VCAGYCANNS TCTVNQGNQP QCRCLPGFLG 4301 DRCQYRQCSG YCENFGTCQM AADGSRQCRC TAYFEGSRCE VNKCSRCLEG ACVVNKQSGD VTCNCTDGRV APSCLTCVGH CSNGGSCTMN SKMMPECQCP 4401 PHMTGPRCEE HVFSQQQPGH IASILIPLLL LLLLVLVAGV VFWYKRRVQG AKGFQHQRMT NGAMNVEIGN PTYKMYEGGE PDDVGGLLDA DFALDPDKPT 4501 NFTNPVYATL YMGGHGSRHS LASTDEKREL LGRGPEDEIG

Figure 1-5 Amino acid sequence of human LRP1 showing structural features highlighted. Regions in: green, indicate the signal peptide, furin-cleavage site and cytoplasmic NPxY motifs; purple, the 2-, 8-, 10-, and 11-membered clusters of LDL class A ligand-binding repeats; yellow, the clusters of EGF-like repeats, and; blue, the YWTD motifs (or related sequences) in spacer regions, and the transmembrane region.

35 antibodies recognising LRP1 are assigned to cluster of differentiation, CD91 (Schlossman et al., 1994). When LRP1 was cloned, the receptor responsible for endocytosis of activated 2-maroglobulin (2MR) was isolated using ligand affinity chromatography (Ashcom et al., 1990; Jensen et al., 1989; Moestrup and Gliemann,

1989). Sequence comparison revealed that LRP1 and 2MR are identical (Kristensen et al., 1990; Strickland et al., 1990), which meant that the same receptor was capable of recognising two large ligands with distinct structures of functions. That is, LPR1 bound lipoproteins and a protease inhibitor (2M). The repertoire of ligands recognised has broadened from these initial two, to over 30 today, including apoE-enriched lipoproteins, plasminogen activators and other protease, matrix proteins, and growth factors (Lillis et al., 2008). This range implicates LRP1 in a many biological systems including lipid transport, clearance of proteases, cell migration, and the target of microbial toxins (reviewed in Krieger and Herz, 1994; Moestrup, 1994; Strickland et al., 1995; Strickland et al., 1994).

Discussion of the LDL receptor family necessitates mention of receptor associated protein (RAP). RAP is a39kDa endoplasmic reticulum-resident chaperone which copurifies with LRP1 (Ashcom et al., 1990; Williams et al., 1994), and binds multiple site within the receptor (Williams et al., 1992). Along with LRP2, RAP forms part of the Heymann nephritis antigen, in experimental rats (Strickland et al., 1991). Its presence inhibits the binding of all known ligands to LRP1 (Herz et al., 1991; Nielsen et al., 1995) and other members of the LDL receptor family (Battey et al., 1994; Kounnas et al., 1992; Medh et al., 1995). This ability has allowed RAP to become a remarkable experimental tool. Use of recombinant RAP as an LRP1 ligand inhibitor, probe, and capture agent, was pivotal to the generation of results presented here in Chapters 3, 4, 5 and 6.

1.5.2 The LDL receptor family

Four representative human members of the LDL receptor family are depicted in Figure 1-6. In addition to the archetypal LDL receptor, the structures of the VLDL receptor, LRP1, and LRP2 are represented in the Figure. Other established human members (not shown) are apoER2/LRP8, MEGF7 and LRP1B (reviewed in Lillis et al., 2008). Some authors include the related proteins, SorLA/LR11, LRP5, and LRP6 in

36 this list (Jaeger and Pietrzik, 2008), and there are additional non-human members, such as the evolutionally conserved protein resembling LRP1 in the nematode, Caenorhabditis elegans (Yochem and Greenwald, 1993).

One receptor in particular is worth further comment here. LRP1B is closely related in structure to LRP1 and both exist in high levels in brain (Liu et al., 2001). LRP1 shares a number of features in common with LRP1B. Both bind a similar range of ligands including, Pseudomonas exotoxin (Pastrana et al., 2005), -amyloid precursor protein (Cam et al., 2004), and the plasminogen activators (Liu et al., 2001). Both LRP1 and LRP1B are cleaved by furin (Li et al., 2005). However, there are also stark differences. LRP1B was first identified as a candidate tumor suppressor gene (Langbein et al., 2002) that is inactivated in nearly 50% of non-small-cell lung cancer cell line (Liu et al., 2000). A possible contributing factor to this feature is its slow rate

Figure 1-6 Selected members of the LDL receptor family in humans.

37 of endocytosis, which impairs urokinase receptor regeneration (Li et al., 2002) and retains -amyloid precursor protein at the cell surface, causing reduced amyloid- peptide production (Cam et al., 2004). LRP1B has splice variants in different tissues, and most notably, unlike LRP1, LRP1B deletion is not lethal (Marschang et al., 2004). Like LRP1, LRP1B releases its ectodomain. This shedding is mediated by a metalloprotease, and the receptor is further processed to release its intracytoplasmic domain (ICD), which translocates to the nucleus via a nuclear localization signal present in this domain (Liu et al., 2007). This latter event mediates its tumor suppression function.

1.5.2-1 LRP1 biosynthesis, structure and characteristics

LRP1 is synthesised in the endoplasmic reticulum as a glycosylated single-chain polypeptide of 600kDa (Herz et al., 1990a). Upon passage through the trans-Gogli, the protein is cleaved by a furin-like enzyme on the C-terminal side of the recognition sequence, RHRR, between residues 3924 and 3925 (Willnow et al., 1996). Despite this posttranslational processing, the newly formed N-terminus 515 kD -chain remains tightly tethered to the C-terminus 85 kDa -chain, by non-covalent bonding. This heterodimer is inserted into the plasma membrane as a type 1 transmembrane protein (extracellular N-terminus), with the entire -chain extracellular, and the -chain harbouring the single transmembrane region and a cytoplasmic tail (Herz et al., 1990a). The 4,544 residue protein (Herz et al., 1988) is product of a 15 kb mRNA transcribed from a 92 kb gene (Van Leuven et al., 1994) on human chromosome 12 at q13-14 (Hilliker et al., 1992; Myklebost et al., 1989).

The -chain of LRP1 resembles four copies of the extracellular domain of the LDL receptor, which itself is comprised of seven LDL receptor class A repeats, two EGF-like repeats, and a spacer region containing the motif, YWTD (see Figure 1-5 and Herz et al., 1988). In LRP1, the LDL class A repeats (previously termed complement- like repeats) are arranged in clusters of 2, 8, 10, and 11, and these are flanked by up to two EGF-like repeats. The LDL class A repeats constitute the calcium-dependent ligand-binding regions in both the LDL receptor and LRP1 (Moestrup et al., 1993).

LDL class A repeats contain three characteristically arranged disulfide bonds and consists of approximately 40 amino acids (see insert in Figure 1-5). A sequence of

38 acidic residues within these repeats has been proposed to account for a preference by LRP1 to bind positively-charged ligands. However, the structure of the 5th repeat in the LDL receptor revealed that this negatively-charged sequence is involved in calcium binding and therefore more likely to provide structural support. Other LDL class A repeats contain homologous sequences and by similarity, these also are presumed to incorporate calcium.

The heterodimeric nature of LRP1 is a feature distinguishing it, and LRP1B, from most other members of the LDL receptor family (see Figure 1-6). While the role of this characteristic is not known, it is interesting to note that it is conserved in the homologous protein in Caenorhabditis elegans (Yochem and Greenwald, 1993).

The ectodomain of the -chain consists of a portion of the spacer region separating the last ligand-binding cluster on the -chain, followed by seven EGF- domains, and then a single transmembrane segment. On the 100 amino-acid cytoplasmic tail are two copies of the motif NPxY located within a seven residue sequence sharing homology to a region in the LDL receptor responsible for directing that receptor to clathrin coated pits for internalisation (Chen et al., 1990). By analogy, this region in LRP1 has been assumed to offer a similar function. However, NPxY is a motif also essential in docking the cytoplasmic adaptor protein SHC to facilitate signalling in receptor tyrosine kinases. Moreover, adaptor proteins Disbabled-1 and FE65, which are components of the neuronal Reelin/Disabled system, interact with NPxY in the cytoplasmic tail of LRP1 and other members of the LDL receptor family (Hiesberger et al., 1999; Trommsdorff et al., 1999; Trommsdorff et al., 1998). More recent evidence suggests that YxxL and di-leucine motifs in the LRP1 -chain are the predominant endocytic signals (Li et al., 2000).

1.5.2-2 LRP1 tissue distribution

LRP1 is most abundant in liver, brain, and lung (Herz et al., 1988), and is known to have endocytic competence in cortical neurons (Bu et al., 1994).

39 1.5.2-3 LRP1 as the 2-macroglobulin receptor

Human 2-macroglobulin (2M) is a 718 kDa tetrameric glycoprotein in human plasma (Bhattacharjee et al., 1999). Each subunit contains a „bait‟ region for a range of proteases, and a thiolester (Roche et al., 1989) which can react with the attacking protease (Sottrup-Jensen et al., 1980), small nucleophiles such as methylamine

(Christensen et al., 1989), and heavy metals (Hussain et al., 1995). In doing so, 2M undergoes a conformational change, and exposes a receptor-recognition site (Gonias et al., 1988). In this way, native 2M does not bind to the 2M receptor until it is activated (2M*). The endocytic activity of the 2M* receptor in the liver has been known since the early 1980s (Davidsen et al., 1985; Tycko and Maxfield, 1982).

Concurrent with the isolation of the 2M* receptor (Ashcom et al., 1990; Jensen et al., 1989; Moestrup and Gliemann, 1989), was the cloning of LRP1 (Herz et al., 1988), and the two receptors were soon after shown to be identical (Kristensen et al., 1990; Strickland et al., 1990).

2M is conserved in evolution, with homologous molecules existing in birds (pigeons) (Seo et al., 1997), and in crustaceans (Enghild et al., 1990; Hall et al., 1989; Iwaki et al., 1996). Of interest to results in Chapter 6, the latter possess a receptor- based clearance system for 2M, which is presumably related to the LDL receptor family (Melchior et al., 1995).

1.5.2-4 LRP1 is a chylomicron remnant receptor

Dietary lipids are broken down to fatty acids, transported into gut epithelium, and secreted into the lymph in the form of chylomicrons (CMs). CMs are conglomerates of triglycerides and cholesterol assembled into particles through association with the apolipoprotein B48. These particles remain suspended in the lymph through the amphoteric nature of apolipoprotein B in that its hydrophobic domain associates with the lipid core of CMs, and its hydrophilic region interfaces with the aqueous environment. Lipolysis of the CMs delivers fatty acids throughout the body. This process depletes lipid reserves in CMs and converts them to chylomicron remnants (CRs). CRs are cleared from the circulation by the liver. The LDL receptor had been identified as an endocytic receptor responsible for this process for some time. However,

40 a second receptor also appears to clear CRs. Foremost in this regard is the observation that mice deficient in the LDL receptor retain the capability to clear CRs.

LRP1 can clear CRs, (Hussain et al., 1991) but debate surrounded its requirement in this role in vivo (Van Dijk et al., 1991), because both the LDL-R and LRP1 can perform this function (Choi and Cooper, 1993). Attempts to determine the contribution of these two receptors initially used RAP injection (Martins et al., 2000), and RAP transfection (Willnow et al., 1994) into LDL -/- mice, but these approaches could not exclude the possible involvement of multiple LDL receptor members, all of which would be inhibited by RAP.

Deletion of LRP1 is lethal at the stage of embryo implantation (Herz et al., 1992; Herz et al., 1993). So, direct evidence for the involvement of LRP1 in CR clearance was impossible for some time. However, Rohlmann et al (Rohlmann et al., 1996) used an inducible knock-out system to deplete LRP1 in adult mice, and thereby provided convincing evidence that LRP1 is a clearance receptor for CRs.

1.5.2-5 The SEC Receptor and megalin/gp330

The serpin enzyme complex receptor (SEC-R) is defined by the binding of serpin mimetic peptides, termed 105Y (Perlmutter et al., 1990) and 105C (Joslin et al., 1991). These peptides were designed for homology to the putative pentapeptide receptor binding region in α1-antrypsin. As such, the SEC-R is a not a defined receptor, but rather a binding observation. It has never been cloned - and never will be. The binding profile of the SEC-R and LRP1 are generally indistinguishable. The term SEC-R persisted until 2004 (Verdier and Penke, 2004; Verdier et al., 2004; Ziady et al., 2004) because it is the subject of several international patents as a targeting system for the delivery of drugs into cells (US Patent 6242416, World patent WO/2001/008708, and European patent EP1200616). An unfortunate side-effect of the patenting system is the uncertainty surrounding a receptor‟s identity when it is ascribed a peculiar name. Under these circumstances, the realisation of important findings can be delayed. In this regard, the SEC-R was shown in early studies to bind amyloid- (the peptide associated with Alzheimer‟s disease). It took a further four year for the demonstration of amyloid precursor protein binding to LRP1 (Kounnas et al., 1995).

41 While the SEC-R terminology was persevered for commercial reasons, the terms megalin and gp330 for LRP2, endures for fondness. LRP2 was first identified as a component of the Heymann‟s nephritis antigen in experimental rats (Saito et al., 1994). Initial estimates of the size of this 600 kDa protein was 330 kDa, so the term gp330 was coined. In renal literature, use of gp330 continues, although usually in conjunction with megalin (i.e. megalin/gp330).

Chapter 2

2 GENERAL MATERIALS AND METHODS

43 2.1 CELL CULTURE

2.1.1 Materials

2.1.1-1 Water and glassware

Deionised water at more than 20 Mohm/cm (Milli Q water) was autoclaved (15 psi, 20 minutes) and used to dissolve or dilute all cell culture reagents and concentrates. Glassware for preparation and storage of medium and buffers was scrubbed with Pyroneg detergent, rinsed in tap water, rinsed with Milli Q water, and autoclaved.

2.1.1-2 Culture media, reagents and plasticware

Commercial culture reagents including prepared media (EMEM, DMEM, M199, HBSS), foetal bovine serum (FBS), glutamine, and 10x trypsin/EDTA were purchased from Life Technologies, Grand Island, NY. Plastic cultureware was Falcon brand as supplied by Becton Dickinson Labware, NJ.

2.1.1-3 Preparation of Media

Media reconstituted from powders into autoclaved Milli Q water were sterile filtered (0.2) into autoclaved one litre glass Schott bottles, and stored at 4C. When required, foetal bovine serum (FBS) was added under sterile conditions, generally to 10%. Where noted, antibiotics were similarly added, but cells were generally cultured in their absence.

2.1.1-4 Addition of NaHCO3

Sodium bicarbonate (Sigma, culture grade) was dissolved to 10% in Milli Q water in a robust glass bottle. The lid was screwed down and the bottle was autoclaved while enclosed inside double autoclave bags. This solution was added as the final component to medium (generally 20 ml / litre) through a 0.2 filter fitted to a syringe. This procedure avoided problems associated with negative pressure filtration of bicarbonate- containing media.

44 2.1.1-5 Specific Media

Eagles minimum essential medium with Earles salts (EMEM) used for the culture of HepG2 cells, was purchased either as a powder mix, or as a 10x concentrate. For the latter, glutamine was added to 2 mM at the time of dilution in autoclaved Milli Q water. Dulbecco‟s minimum essential medium with high glucose (DMEM, used for the culture of hybridomas and smooth muscle cells), M199 (used for endothelial cells), and Hanks balanced salt solution without calcium or magnesium (HBSS, used to dilute trypsin/EDTA concentrate), were prepared in one litre batches from powder. EMEM or

DMEM containing 10% FBS, 2 mM glutamine, and 2 g/L NaHCO3 is referred to as complete medium.

2.1.1-6 PBS

Filter (0.2) sterilised Dulbecco‟s phosphate buffered saline A (PBS) was prepared using analytical grade chemicals (see appendix A). PBS was not used in any procedure involving stock cell cultures, and only cells in the terminal stages of experimentation contacted PBS or other non-culture grade reagents. Azide to 0.1% (PBSa), BSA to 0.1% (PBSb), and calcium to 0.1 mM (PBSc) were added as required. Generally all three additives were included in PBS (PBSabc) for post culture work.

2.1.1-7 Trypsin/EDTA

A commercial 10x concentrate containing trypsin and EDTA was diluted to prepare a reagent containing 0.25% trypsin, 0.5 mM EDTA, and 2 g/L NaHCO3 in HBSS without calcium of magnesium.

2.1.2 Freezing of cells in liquid nitrogen

Cells were harvested as described in 2.1.4 and adjusted to the range 2 - 20 x 106 / ml with medium in which they are normally cultured. Freezing concentrate (20% DMSO / 80% FBS) was prepared, sterile filtered, and chilled. Immediately prior to freezing, equal volumes of freezing concentrate and cells were combined, and aliquots

45 of 1 ml were dispensed into freezing vials (Nunc, Roskilde, Denmark). The vials were inserted into an isopropanol-filled controlled rate freezing container (Nalgene) which regulates the rate of temperature decrease to approximately 1C per minute when placed in a -80C freezer. Following overnight equilibration in the -80C freezer, the vials were transferred to a liquid nitrogen tank for long-term storage.

2.1.3 Culture of cells from frozen stocks

Cells stored in liquid nitrogen were thawed rapidly in a 37C water bath, transferred to 10 ml of appropriate compete medium, and centrifuged at 300g for 5 minutes. The supernatant containing the majority of the freezing mix was discarded, and the cell pellet was resuspended in 1 to 5 ml of fresh complete medium. The cells were transferred to either a 25 cm2 flask, or serially diluted 1:2 in 24 well plates containing 1 ml of complete medium per well. The latter procedure prepares a range of seeding densities to ensure that optimal conditions for growth of fastidious cells are met in at least one well. Cells were cultured under 5% CO2 in a humidified incubator at 37C. Medium was exchanged after 24 hours, and at 48 hour intervals thereafter until sufficient numbers had accumulated for transfer to larger culture vessels. Generally this was 5x105 / ml for suspension cultures and 80% confluence for adherent cell lines.

2.1.4 Passage of adherent cell lines

The medium covering adherent cell lines was removed completely and the cells were rapidly rinsed once with trypsin/EDTA to reduce trypsin inhibitors and calcium. Rinse volumes of 1 ml for 25 cm2 flasks, and 3 ml for 75 cm2 flasks, were replaced with equal volumes of fresh trypsin/EDTA. Following incubation for 10 minutes at 37C, cells were dislodged by gentle aspiration using 3 ml (for 25 cm2 flasks) or 9 ml (75 cm2 flasks) of complete medium (containing 10% FBS), and the cell numbers in these single-cell suspensions were estimated. Samples delivering 10,000 cells / cm2 were transferred to flasks containing fresh complete medium (5 ml for 25 cm2 flasks, or 20 ml 2 for 75 cm flasks) then placed in an incubator maintained at 37C with 5% CO2 in humidified air.

46 2.1.5 Characteristics and culture of HepG2 cells

HepG2 cells were derived from hepatocellular carcinoma tissues in a 15-year-old male Caucasian, and are epithelial in morphology (Aden et al., 1979). They do not harbour Hepatitis B virus genome, and are not tumorigenic in nude mice (Knowles et al., 1980). HepG2 retain a well-differentiated phenotype in that they secrete several plasma proteins, and this ability is unusual for liver-derived cells maintained in long term culture (Knowles et al., 1980). For this reason, HepG2 are widely used as an in vitro model of hepatic parenchymal cells, albeit subject to patent restrictions (Knowles and Aden, 1983).

Vitronectin Control depleted

Figure 2-1 HepG2 monolayers require contact factors. Photomicrographs of HepG2 cultured in medium containing 5% foetal bovine serum which was untreated (left panel) or vitronectin-depleted (right panel). Depleted serum, produced by using immobilised anti-vitronectin antibody, was supplied courtesy of Dr M. Dziegielewski, Department of Pathology, University of New South Wales. Similar “golf ball” clusters (right panel) occur when HepG2 are cultured in non-culture grade plastic flasks.

47 HepG2 normally grow as an adherent monolayer (Figure 2-1, left panel). In the absence of contact factors such as vitronectin (Figure 2-1, right panel), or when cultured in non-culture grade flasks, the cells form suspended clusters which resemble golf balls. Cultures in which such clusters appeared were discarded.

Four sublines of HepG2 are used in this work. Details of their sources and differences are described in Chapter 4. Each subline displays similar growth kinetics, with an initial doubling time of approximately 24 hours as shown for three of the sublines in the left panel of Figure 2-2. This observation is in good agreement with similar data reported for the parental line (Liao and Florén, 1992). HepG2 are not contact-inhibited. After reaching 50-80% confluence, cells continue to divide and form multilayers (Figure 2-2, right panel) which are easily disrupted by washing procedures.

Stock cultures of HepG2 were maintained in 75 cm2 flasks using complete EMEM, and were passaged twice per week by re-seeding at ~25,000 per cm2 (~1:10 split). Cells for experimentation were seeded with the same medium, and at the same density, into 12-well plates (4 cm2 / well), 24-well plates (2 cm2 / well), or 25 cm2 flasks. Medium was exchanged every two days and 8-24 hours before passage or

Figure 2-2 Characteristics of HepG2 in culture. HepG2 sublines described in Chapter 4 have similar growth kinetics (initial doubling time ~ 24 hours) as indicated by the plot of cell numbers with time (left panel). In early phases of culture (less than 5 days) HepG2 form an adherent monolayer. However, the cells are not contact inhibited at confluence, and subsequently form three dimensional structures (right panel). The confocal photomicrograph in the right panel is a red/green stereo image showing multiple layers of cells at day six in culture.

48 experimentation. Cells for experimentation were used at 50-80% confluence following 3 to 5 days in culture.

2.1.6 Culture of other adherent cell lines

Bovine smooth muscle cells (SMC, supplied by Prof L. Khachigian, Centre for Vascular Research, University of New South Wales) were handled similarly to HepG2 and used as a cellular source of bovine LRP1 for the results in Chapter 5. The rabbit fibroblast cell line, Rab-9 (ATCC CRL-1414), was maintained by weekly passage using a 1:10 split. This line is a cellular source of rabbit LRP1 and was maintained to verify reactivity of the antibodies described in Table 2-1. The human fibrosarcoma cell line, HT-1080 (ATCC CCL-121, supplied by Dr L. Matthias, Centre Vascular Research, University of New South Wales), was passaged twice weekly (1:8 split) and used as controls in flow cytometric results in Chapter 4.

2.2 CELL HOMOGENATES

2.2.1 Preparation of cell lysates

For the preparation of cell lysates used as cellular sources of LRP1 in ligand and Western blotting, cells at approximately 80% confluence in 25 cm2 flasks were washed four times with PBS and treated with 2 ml of Laemmli sample buffer (without bromophenol blue). Lysis was aided by agitation on an orbital shaker for 10 minutes followed by 4 passages of the lysates progressively through 18, 21, and 23 gauge needles. Protein content was determined using a BCA assay kit (Pierce). The lysates were diluted to 5 mg/ml then frozen at -80C in 100 l aliquots. For blotting, 1-10 l (5 to 50 g) were diluted to 50 l with Laemmli buffer immediately prior to electrophoresis.

49 2.2.2 Preparation of kidney membrane

Kidney is a source of LRP2 (gp330), a 600 kDa membrane protein related to LRP1. A sample of human kidney from a cadaver was kindly supplied by Dr G. Higgins (Pathology Department, University of New South Wales). Membranes were prepared by a procedure based on a reported method for the isolation of brush border membranes (Malathi et al., 1979). A section of the sample (~2 grams) was diced with a scalpel and homogenised in cold 50 mM mannitol, 10 mM Tris -HCl pH 7.0, 0.5 mM PMSF (10 ml) by using a Dounce homogeniser. Calcium chloride was added to 10 mM and incubated with mixing on ice for 10 minutes. The homogenate was successively centrifuged at 500g and 3000g for 10 minutes at 4C, and the membranes in the supernatant were pelleted at 10,000g for 1 hour. The pellet was washed once in the same buffer, then resuspended in PBS and frozen in aliquots at -80C.

2.3 ANTIBODIES

2.3.1 Hybridomas.

Monoclonal antibody-producing hybridomas were seeded at 105 / ml in 75 cm2 flasks containing 20 ml of complete DMEM. Medium was exchanged daily until cell numbers had expanded to approximately 5x105 / ml. Cells were pelleted, transferred to serum-free Hybridoma culture medium (Life Technologies) and cultured for a further two days. The supernatant was harvested, centrifuged at 1,000g for 30 minutes, sterile filtered, and either stored in the presence of 0.1% sodium azide at 4C, or frozen. The monoclonal antibodies which were raised from hybridomas for use in this work, are list in Figure 2-1.

Chain and species specificities of LRP1 antibodies used in this study are summarised in Figure 2-1 and Table 2-1. Antiserum 777 was raised in rabbits against purified human LRP1 -chain and affinity purified on immobilised immunogen. This antibody and 8G1 were kind gifts from Dr D. K. Strickland (American Red Cross, Rockville, Maryland). Monoclonal antibodies IgG-5D7 and 8G1 have -chain specificity, while IgG-1B3 and IgG-11 H4 are directed against the ectodomain and cytoplasmic tail of the -chain, respectively. Commercial monoclonal antibody, 3402

50 Table 2-1 Hybridomas producing LRP1 specific monoclonal antibodies.

Hybridoma Mouse Ig Specificity Source (References) Subclass

IgG-5D7 IgG1 LRP1 -chain. (Raised ATCC CRL-1938 against purified rabbit LRP1. (Herz et al., 1990b) Does not cross-react with other species tested in this work.)

IgG-1B3 IgG1 LRP1 -chain ectodomain. ATCC CRL-1937 (Herz (Raised against purified rabbit et al., 1990b) LRP1. Reacts also with human, bovine, and weakly, with rat antigens.)

IgG-11H4 IgG1 LRP1 -chain cytoplasmic ATCC CRL-1936 (Herz tail. (Raised against a et al., 1990b) synthetic peptide representing the C-terminal 13 amino acids of human LRP1. Known to cross react with various species.)

C7 IgG2b LDL receptor. (Raised ATCC CRL-1691 against purified bovine LDL (Beisiegel et al., 1981; receptor. Cross-reacts with Tolleshaug et al., 1983) human antigen.

(purchased from American Diagnostica), is directed against human -chain. All LRP1 monoclonal antibodies used are murine IgG1 immunoglobulins.

51 2.3.2 LRP1-specific antibodies

Two methods (Pharmacia Separation News, Vols 13.3 and 13.5) were applied interchangeably for the purification of monoclonal antibodies from cell-culture supernatants. In the first method, supernatant (500 ml) was diluted 1:1 with 3 M NaCl / 1.5 M glycine pH 8.9 (binding buffer) and applied overnight in a cold lab at 1 ml/minute to a column containing 2 ml of protein A-sepharose (Pharmacia). The column was washed with binding buffer for 1 hour and the purified antibody was eluted with 50 mM NaCl / 100 mM citrate pH 5.0. The eluate was immediately neutralised with 1/10th volume 1 M Tris. Protein A-sepharose was regenerated with 300 mM citric acid / NaOH pH 3.0 and stored in PBS / 0.1% azide.

In the second method, supernatant (500 ml) was diluted 1:5 with 20 mM citric acid / NaOH pH 5.5 (binding buffer) and applied overnight in a cold lab at 2 ml/minute to a column containing 2 ml of S-sepharose (Pharmacia). The column was washed with binding buffer for 1 hour and the purified antibody was eluted with 200 mM NaCl in binding buffer. S-sepharose was generally used once only, but could be regenerated by serially washing with 20ml volumes of 100 mM NaOH, 2 M NaCl, water, and methanol, then stored in PBS / 0.1% azide).

For both methods, columns were loaded by gravity, and the product (in ~5ml) was collected manually with the aid of a UV monitor. Purified antibodies were dialysed, with stirring, in a cold lab against 3 changes of PBS (500 ml) for at least 6 hours each.

Immunoglobulin content was estimated in mg/ml (as A280nm, 1 cm / 1.4), and the final product sterilised using 0.2 m filters for storage in aliquots either frozen or at 4C.

2.3.3 A note concerning 8G1 species specificity

The anti-LRP1 -chain monoclonal antibody, 8G1, was raised against purified human LRP1. In Chapter 5, this antibody is reported to cross-react with mammalian soluble LRP1 in rabbit and horse sera, but not in bovine serum. This recognition profile is likely to be the result of initial hybridoma selection procedures which normally include foetal bovine serum (FBS). Soluble LRP1 antigen in FBS (see Chapter 5) may have neutralised and excluded clones secreting bovine-recognising antibodies, and thereby influenced selection of antibodies selectively reactive with LRP1 from other

52 mammals. If so, the effect is not universal because IgG-1B3 recognises soluble LRP1 in bovine serum (see Chapter 5).

2.4 FLOW CYTOMETRY

2.4.1 Cell preparation, labelling, and analysis

HepG2 cells and other adherent cell lines were detached by incubation in HBSS containing 1 mM EDTA, 2 g/l NaHCO3 and 0.1% BSA for 10 minutes at 37C in

5% CO2. The cells were harvested and washed twice at 200g with ice-cold PBS containing 0.1% sodium azide, 0.1% BSA, and 0.2 mM CaCl2 (PBSabc). Suspensions were adjusted to 5 x 106 cells / ml in PBCabc containing heat aggregated (63C, 15 minutes) human immunoglobulins (CSL) at 1 mg/ml. Aliquots (100 l) containing 500,000 cells were labelled with primary antibodies (generally at 10 g/ml or according to the manufacturer‟s instruction) for 30 minutes on ice, then washed twice with cold PBSabc at 300g for 5 minutes. Immunofluorescence was developed for 30 minutes on ice with appropriate fluorescein-conjugated secondary anti sera (sheep anti-mouse immunoglobulins from Silenus, Melbourne, Australia, or goat anti-rabbit immunoglobulins from Dako). Following two further washes, cells were analysed in the presence of 5 g/ml propidium iodide on a Becton Dickinson FACStar plus cell sorter. Data was analysed using WinMDI, a PC-based freeware program available for download from Scripps University (Trotter, 2000).

2.4.2 Harvesting HepG2 for flow cytometric analysis

In preliminary studies, various methods for harvesting adherent cells to be analysed by flow cytometry were evaluated using the procedures outlined in the Legend to Figure 2-3. Results in the figure demonstrate that both scraping and EDTA- detachment are suitable for handling HepG2, in that reactivity with 8G1 (anti-LRP1 -chain) is retained. Harvesting with EDTA is shown to offer advantages with viability and was chosen for subsequent applications. Harvesting with trypsin is shown to be unsuitable for this work in that enzymatic detachment destroys the 8G1 epitope. Results

Figure 2-3 Evaluation of methods for harvesting HepG2 for analysis by flow cytometry. Shaded histograms of Log Green Fluorescence measurements in the first and second rows of panels (a-f) were generated using cells stained with anti-LRP1 antibodies. Open overlayed histograms in these panels are controls. Murine monoclonal antibodies, 8G1 (anti-LRP1), MOPC21 (negative control, faint overlays), and W6/32 (anti MHC-I, positive controls, heavy overlays) were used for the results in the first row (a-c). Cells represented in the second row (d-f), were stained with polyvalent anti-LRP1 antibody, “777” (antiserum immunopurified on immobilised LRP1 -chain), or purified normal rabbit immunoglobulins (negative control). Mouse and rabbit primary-layer antibodies were developed with FITC- conjugates of sheep anti-mouse immunoglobulins (a-c), and goat anti-rabbit immunoglobulins (d-f), respectively. Cells used for the results in the left column of panels (a, d, g) were harvested by scraping. The other panels show results using cells harvested by incubation for 10 mins at 37°C in Hanks buffer containing 0.25% trypsin / 0.53 mM EDTA (central column), or 0.1% BSA / 1 mM EDTA (right column). Cellular viability was measured simultaneously in each assay tube by including 5 g/ml propidium iodide (PI). Representative examples (the 8G1 labelled cells) of cells stained with PI are shown in third row (g-i) which are dot plots of Size versus Log PI (red) Fluorescence. The numbers inserted in these panels are the average viability (n=4) for each harvesting method determined by the percentage of cells recorded within the region displayed (R1). Results show that all methods retain high cellular viability and preserve recognition of surface membrane LRP1 by the rabbit polyvalent antiserum. However, trypsin treatment destroys the 8G1 epitope.

54 were similar for a LRP1 antibodies, 3402 and IgG-5D7 against HepG2 and Rab9, respectively (data not shown). However, recognition with rabbit antiserum 777 is retained on trypsin-harvested HepG2cells, indicating that a fragment of -chain survives tryptic cleavage and loss from the cell surface. Similarly, IgG-1B3 (anti-LRP1- chain ectodomain) retains recognition of trypsin-treated HepG2 and Rab-9 (data not shown), further supporting the retention of LRP1 antigens close to the surface membrane following exposure to trypsin.

2.5 PROTEIN MANIPULATIONS

2.5.1 Protein assay

Protein concentrations of SDS-containing cell lysates and general protein solutions were determined by the BCA assay method of Smith et al (Smith et al., 1985) using a commercial kit (Pierce). This technique was considered more convenient than the method of Lowry and more appropriate than the Bradford assay which is sensitive to the presence of detergents. Protein concentrations of purified immunoglobulins were estimated in mg/ml by dividing the absorbance at 280 nm (1 cm) by 1.4.

2.5.2 Iodination of proteins using iodogen coated tubes

2.5.2-1 Background and Materials

Na125I (half life 60 days) at 400 MBq (~10 mCi) in 100 mM NaOH (100 l) was purchased from Australian Radioisotopes, Lucas Heights and used within one month of the calibration date. Iodogen (Pierce), a water-insoluble oxidising agent, was used according to the method of Fraker et al (Fraker, 1978 #236). This compound functions similarly to chloramine T in producing 125I+ electrophiles which attach to the phenolic side chain of tyrosine residues. Insoluble iodogen coated inside a small glass tube offers the advantage that the iodination reaction is stopped upon removal of the protein from the tube. Protein probes iodinated by the following procedure have high specific radioactivity while retaining their specific binding activities.

55 2.5.2-2 Procedure

Iodogen (50 g) in 500l of chloroform or dichloroethane, was coated onto 12 x 75 mm glass tubes by slow (24 hour) evaporation of the solvent. These reaction vessels were hermetically sealed and stored for periods up to 12 months. The protein to be labelled (10 to 50 g) was diluted in 100 l of buffer (1:1 mixture of PBS and 100 mM sodium phosphate pH 7.5) then added to a reaction tube which had been washed once with buffer to remove any detached iodogen. Iodination was initiated by the addition of Na125I (20 MBq in 5 l) and the mixture was incubated at room temperature for 15 minutes. The reaction was terminated by separating the 125I-labelled protein from free 125I on a column containing 8 ml of P6DG biogel (BioRad) pre-equilibrated with PBS. Passage of 125I-labelled protein through the column was visually monitored by co- loading with 100l of bovine cytochrome c (Sigma) solution (20 mg/ml). 125I-labelled proteins were used within one week.

56 2.5.3 Determination of specific activity

The following parameters and values were used to calculate the concentration and specific activity of 125I-labelled proteins.

(1) Load. The amount of protein iodinated (in g). Generally this was 10-50 g. (2) Volume. The volume (in ml) of the P6DG column fraction which contained the labelled protein. This was calculated from the weight difference between the full and empty collection vial. (3) Recovery. Protein recovered (in %) from the P6DG columns. This was generally around 80%. (4) Counts. Radioactivity from a 10 l sample of the radiolabelled protein solution. This was expressed as counts per minute (CPM), (5) Efficiency. The counting efficiency of the -counter, measured as 40% using a standard with known radioactivity.

Conversions. 1 Becquerel = 1 disintegration / sec = 60 disintegration / minute 1 Mbq = 27 Ci (1 Ci = 3.7 x 1010 disintegrations / sec) For a counter with 40% efficiency, 1000 CPM = 42 Bq

Example:

35 g (500 pmol) of tPA in 1.5 ml, giving 1,300,000 CPM from a 10 l sample.

Load (in pmol) x Recovery 500 x 80 Concentration = ------= ------= 267 nM (in nM) Volume (in ml) x 100 1.5 x 100

Counts (in CPM) x Volume (in ml) x 100 1.3x106 x 1.5 x 100 Specific Activity = ------= ------(in CPM/fmol) Load (in pmol) x Sample (l) x Recovery 500 x 10 x 80

= 488 CPM/fmol

Specific Activity (in CPM/fmol) x 100 488 x 100 Specific Activity = ------= ------(in Bq/fmol) 60 x Efficiency 60 x 40

= 20 Bq/fmol

57 2.5.4 Purity of iodinated protein

The purity of a radiolabelled protein preparation is defined as the percentage of radioisotope attached to the protein compared to the total amount of radioisotope present. To determine their purity, 10 l samples of 125I-protein preparations were added to 490l aliquots of 0.1% BSA / PBS in duplicate. The proteins were precipitated by mixing with 500 l of 20% trichloroacetic acid / 4% phosphotungstic acid (TCA/PTA). The tubes were incubated on ice for 10 mins, centrifuged at 1,000g for 5 minutes, and 500 l of the supernatants (half supernatants) were transferred to a new tube. Purity was calculated using the following parameters and equation. Typically, purity was >98% on the day of iodination and >90% after one week of storage at 4ºC.

(1) ½sup. Counts (in CPM) from 500l of the supernatant above the TCA/PTA precipitated protein. Counts in this fraction represent half of the free iodine in the original 10l sample.

(2) pellet. Counts (in CPM) from TCA/PTA precipitated proteins together with the other half of the free iodine in the original 10l sample.

2 x ½sup x 100 Purity = 100 ------( pellet + ½sup )

58 2.5.5 Preparation of recombinant RAP

2.5.5-1 GST-RAP expressing plasmid

A pGEX-KG plasmid (a derivative of pGEX-2T from Pharmacia) harbouring sequence coding for a fusion protein between glutathione S-transferase from Salmonella japonicum, and 39-kDa receptor associated protein (RAP) from rat brain (Herz et al., 1991), was a kind gift provided by Dr J. Herz (University of Texas) through the courtesy of Dr K. K. Stanley (Heart Research Institute, Sydney).

2.5.5-2 Transformation

E. coli strain XL1blue were streaked on 0.3% agar (Oxoid) in 2*YT medium which consists of 16 g/L trypticase peptone (BBL), 10 g/L yeast extract (BBL), 5 g/L NaCl, and ~40 mg/L NaOH (pH7 final). Following over night incubation at 37C, a colony was transferred to 2*YT medium (5 ml) and cultured with swirling at 37C until the absorbance (1 cm) at 600 nm reached 0.6 (log phase). A sample (1 ml) was chilled on ice for 5 minutes, then centrifuged at 7,000g for 5 minutes in a cold room. The pellet was washed once with cold 0.1 M CaCl2 (1 ml), resuspended in the same solution, and incubated on ice for 30 minutes. A sample (100 l) of the resulting competent cells was mixed with plasmid DNA (1 pg), and held on ice for a further 45 minutes. The suspension was incubated at 42C for exactly 90 seconds, returned to ice for 2 minutes, mixed with 2*YT medium (900 l), and incubated at 37C for 45 minutes. The transformed cells were smeared onto sets of agar/2*YT plates which were either devoid of antibiotics (control), or contained ampicillin (50 g/ml). Cells subjected to the same procedure except for exposure to plasmid DNA were included as additional controls in separate tubes and sets of plates. Plasmid-transformed cells appearing as colonies on ampicillin-containing plates were transferred to 2*YT medium (5 ml), cultured to log phase, washed with 2*YT medium, resuspended in 2*YT medium containing 50% glycerol or 10% DMSO, and frozen on dry ice for storage.

59 2.5.6 Production of GST-RAP in E.coli

A loop of transformed XL1blue, was streaked onto agar containing 2*YT and 50 g/ml ampicillin (2*YT/amp). Following overnight incubation on these plates at 37C, a colony was transferred to 2*YT/amp liquid (10 ml) and cultured with swirling at 37C until log phase was reached (A600 nm, 1 cm = 0.6). The cells were transferred to 2*YT/amp (100 ml) in a 1 litre conical flasks and cultured to log phase at 37C on an orbital shaker at 110rpm. Expression was induced by the addition of isopropyl -D thiogalactopyranoside (IPTG) to 0.01% (w/v), and the cells were cultured for a further 4 to 6 hours at 30C. Cells were harvested by centrifugation at 1,000g for 15 minutes.

From this point, one of two protocols were followed. The initial procedure used to extract GST-RAP (Protocol 1) followed a published method (Herz et al., 1991). Later, a modified procedure (Protocol 2) was developed to facilitate extraction of larger batches. As determined by SDS-PAGE and inhibition of 125I-tPA binding to HepG2 cells, the purities and activities of RAP products derived from either protocol were indistinguishable (data not shown).

2.5.6-1 Isolation of GST-RAP, Protocol 1

Based on the method of Herz et al (Herz et al., 1991), cells from 100 ml of broth were resuspended in 10 ml of lysis buffer (15% sucrose, 50 mM Tris, 50 mM EDTA, pH 8) and incubated with lysozyme (1 mg/ml) on ice for 30 minutes. A 20 ml solution of 0.2% triton X100 and 0.5 mM PMSF was added and the cells were lysed by 4 passages each, sequentially through 18-, 21-, 23-, and 25-gauge needles. Dithiothreitol (DTT) was added to 1 mM and the lysate was mixed with 5 ml (bed volume) of glutathione-agarose (Sigma) which had previously been equilibrated in 20 mM HEPES pH 7.6, 100 mM KCl, 0.2 mM EDTA, 20% glycerol, 1 mM DTT, and 0.5 mM PMSF (buffer A). The mixture was rotated overnight in a cold lab, washed three times with cold buffer A by centrifugation (1,000g, 1 minute), then transferred to a column for further washing with buffer A (100 ml). GST-RAP was eluted using 25 mM glutathione in buffer A (readjusted to pH 7.5), and dialysed against 20 mM HEPES, 40 mM NaCl, pH 7.4.

60 2.5.6-2 Isolation of GST-RAP, Protocol 2

Cells from 1 litre of broth were resuspended with water (20 ml) containing enzyme inhibitors (5 mM EDTA, 1 mM DTT, 0.5 mM PMSF and 5 mM benzamidine, pH 6.8), then sonicated on ice for 30 minutes. The lysate was diluted with 300 ml of HEPES binding buffer (20 mM HEPES / 40 mM NaCl pH 6.5), centrifuged (10,000g, 30 minutes, 4C), filtered (0.45 m filter), and applied at 1 ml/minute to S-sepharose (10 ml) equilibrated with HEPES binding buffer. Following washing with the same buffer (1000 ml), GST-RAP was eluted with 300 mM NaCl in HEPES binding buffer, and dialysed against 20 mM Tris / 40 mM NaCl pH 8 (Tris binding buffer). The solution was applied to Q-sepharose (10 ml) equilibrated in Tris binding buffer and washed with the same buffer (1000 ml). GST-RAP was eluted with 300 mM NaCl in Tris binding buffer and dialysed 20 mM HEPES, 40 mM NaCl, pH 7.4 (same endpoint as Protocol 1).

2.5.6-3 Generation of RAP from GST-RAP by thrombin cleavage

Thrombin (30 U) was added to GST-RAP (50 mg) in 20 mM HEPES / 40 mM NaCl, 0.5 mM Ca++ pH 7.4 (50 ml, reaction buffer), sterile filtered (0.2 m), and incubated at 37C for 8 hours. Not exceeding these concentrations of calcium and protein was found to be important in avoiding precipitation. PPACK (a specific thrombin inhibitor) was added to 5 M, and the reaction mixture was applied to Heparin-sepharose (5 ml) pre-equilibrated in binding buffer (reaction buffer without calcium). Following washing with binding buffer (100 ml), RAP was eluted with 300 mM NaCl in binding buffer and gel filtrated on a P60 biogel column (70 ml) equilibrated with 1 M NaCl / 10 mM phosphate pH 7.3. Purified RAP in the eluate was dialysed against PBS (4 x 500 ml, 6 hrs each), filter sterilised (0.2 m), and stored at 4C or frozen. Figure 2-4 a shows an example of electrophoresed proteins sampled at stages throughout the cleavage reaction.

Figure 2-4 Production and characterisation of recombinant RAP. Panel a shows a Coomassie blue-stained 10 % polyacrylamide gel used to monitor the thrombin- cleavage of recombinant GST-RAP (lane 2) into RAP (lane 3) and GST (lane 4). RAP was separated from GST by binding to heparin-sepharose (at pH 7.4) and eluting GST in the fall-through. Panel b contains flow cytometric histograms which verify the binding activity of the freshly prepared RAP. RAP was conjugated to fluorescein isothiocyanate (FITC) and used at 10 nM to label EDTA-harvested HepG2 in the absence (solid histogram), and presence (open histogram) of a 100 fold excess of unlabelled RAP. The relative shift to the right of the solid histogram indicates specific binding and confirms the binding activity of RAP. Panel c shows Coomassie blue-stained 10% polyacrylamide gels used to determine that the pI of RAP falls within the range 7.3 to 7.7. The gels were loaded with supernatant (Free) and matrix extract (Bound) from pre-incubated mixtures of RAP and S-sepharose suspended in buffers at various pH, as indicated. RAP is observed to bind to S- sepharose at pH 7.3 and below, and is released at pH 7.7 and above.

62 2.5.6-4 pI determination, and monitoring of RAP for purity and biological activity

Gels displayed in Figure 2-4c revealed that the pI of RAP fell within the range 7.3 to 7.7. Determination of this range was used to select pH values for buffers employed in the ion exchange chromatography stages of the modified RAP purification procedure (Protocol 2). The activity of RAP as a ligand-binding inhibitor was evident in experiments reported in Chapters 4, 5,and 6. However, binding activity was also detectable directly by using Fluorescein isocthiocyanate (FITC) conjugated RAP and HepG2 cells in flow cytometric assays (Figure 2-4b).

2.5.6-5 Notes concerning RAP and RAP-GST

Proteins containing disulfide bonds can experience problems with expression of biological function when produced under the reducing conditions within bacteria. The absence of cysteine residues in RAP avoids these concerns. However, the relative ease with which biologically active RAP is obtained from prokaryotes does not extend to all proteins devoid of disulfide bonds. A notable example is PAI-1 which also lacks cysteine residues, yet its secondary structure is fundamentally important for protease inhibitory activity. In supporting studies for work presented in Chapters 3 and 4, PAI-1 was produced as a recombinant protein in E. coli, but the product required activation by a cycle of denaturation and renaturation (Alessi et al., 1988) (data not shown).

The plasmid, pGEX-KG, used to insert code for RAP, produces a polyglycine stretch between RAP and GST. Cleavage with coagulation factor Xa removes this stretch, whereas thrombin, which is more commonly used as a cleavage enzyme for this plasmid system, produces RAP with a polyglycine extension at the N-terminus. This extension imparts no known alteration to RAP function. Indeed, intact GST-RAP is known to display similar biological activity to RAP.

2.5.7 Immobilisation of proteins on sepharose 2B

In order to capture and enrich soluble LRP1 from serum, a method involving periodate oxidation and Schiff base conjugation (Grimsley et al., 1999; Hermanson et al., 1992) was used to immobilise RAP and anti-LRP1 -chain monoclonal antibody

63 8G1 on sepharose 2B (Pharmacia). This matrix was chosen for its large pore size, which is preferable in avoiding possible exclusion problems when dealing with a protein the size of LRP1 (~500-kDa). In preliminary experiments, RAP conjugated to NHS- sepharose CL-4B, with a smaller pore size, was investigated and found also to enrich soluble LRP1 from serum successfully (data not shown). However, an important consideration in dealing with this recently characterised protein is the possibility that some of the soluble receptor may be associated to very large ligands such as apoE- containing lipoproteins. Study of such complexes may require the largest pore size available.

Sepharose 2B (2 ml of a 50% slurry) was washed with water on a scintered glass filter, resuspended with water to 2 ml, and activated by addition of 20 mM sodium periodate (2 ml). In this reaction vicinal diols on agarose are converted to 20-40 mol of formyl groups per millilitre of gel. The mixture was incubated at room temperature for 90 minutes with occasional mixing before the reaction was terminated by addition of 50 l glycerol (330 mM) to consume any remaining periodate. The matrix was washed on a scintered glass filter with water and either all or part was added to a freshly prepared solution containing 6 mg/ml sodium cyanoborohydride (50 mM final concentration) and the protein of interest (2 mg RAP in 1 ml PBS plus 1 ml matrix, or 100 g 8G1 in 200 l PBS plus200 l matrix). The mixture was rotated overnight at room temperature to allow a Schiff base condensation between the activated sepharose and amines on the proteins. The presence of cyanoborohydride stabilises the product against hydrolysis by reducing C=N to C-N. Remaining active sites on the matrix were blocked by addition of ethanolamine (0.1 M in 0.1 M HCl, pH 7.5) to 20 mM and incubated for 60 minutes. The product was washed on a scintered glass filter with PBS and stored at 4C in the presence of 0.1% sodium azide.

64 2.6 ELECTROPHORESIS AND BLOTTING

2.6.1 The Laemmli system of protein electrophoresis

2.6.1-1 Background

Electrophoresis of proteins and cell lysates using the method of Laemmli (Laemmli, 1970) provided the foundation for much of the experimental work presented here. Subsequent molecular tracking by dye staining (see 2.6.2), Western blotting (see 2.6.5), and ligand blotting (see 2.6.6) allowed identification of purified proteins, radiolabelled probes, or the immunological targets in complex mixtures.

Given the elegance of the Laemmli system and it importance to this work, a brief outline of its development helps appreciate its components. Three innovations from earlier electrophoretic techniques are incorporated into the technique described by Laemmli. First, Raymond and Weintraub (Raymond and Weintraub, 1959) used polymerised mixtures of acrylamide and crosslinker to produce transparent gel supports with more consistent chemical definition and porosity than previous translucent systems based on starch. Next, Ornstein (Ornstein, 1964) and Davis (Davis, 1964) added a discontinuous pH buffer system which focuses broad zones of loaded proteins into narrow bands. Finally, sodium dodecyl sulfate (SDS) was introduced from work by Summers et al (Summers et al., 1965). This anionic detergent solubilises and unfolds proteins, and endows them with a uniform charge to mass ratio. Under electrophoretic conditions SDS-protein complexes separate according to size in a fashion similar to gel filtration. Laemmli integrated these features, and Neville described their use for the determination of molecular masses of proteins (Neville Jr, 1971).

2.6.1-2 Method

Samples of proteins or cell lysates were diluted in Laemmli sample buffer (62.5 mM Tris-HCl pH 6.8 2% SDS, 10% glycerol, and 0.001% bromophenol blue) and placed in hot water (>80C) for 3 minutes in preparation for electrophoresis.

The polyacrylamide gel system (usually 1.5 mm thick) was prepared from a stock solution (Bio-Rad) containing 40% acrylamide and 1.067% N,N‟-methylene-bis-

65 acrylamide crosslinker (ratio 37.5:1). Various volumes of this solution were diluted with pH 8.8 Tris-HCl buffer (final concentration 375 mM Tris) to form polyacrylamide resolving gels of various densities (3 to 15% acrylamide) in minigel formats. A stacking gel containing 2% or 3% acrylamide in 125 mM Tris-HCl at pH 6.8 was overlayed on the resolving gel to complete the discontinuous pH system. Both gels contained 0.1% SDS.

Polymerisation of the acrylamide mixtures was initiated by addition of 0.05% tetramethylethylenediamine (TEMED, Bio-Rad) and 0.1% ammonium persulfate (ICN, 1:100 dilution of a 10% solution). To aid the formation of a sharp interface, water- saturated iso-butyl alcohol was layered on the resolving gel during its polymerisation. The concentrations of acrylamide in the stacking and resolving gels varied depending on the application but were generally 4% and 8%, respectively, for experiments involving 125I-labelled plasminogen activators in Chapters 3 and 4, and 3% and 5% when examining the large receptor, LRP1, in Chapters 4, 5, and 6. In some experiments, which were designed for optimal separation of high molecular weight proteins, agarose was added to 1% as a highly porous support in stacking and resolving gels using only 2% and 3% acrylamide. (Chapter 5). For these systems, a frosted plate was required to maintain the acrylamide/agarose gels in a vertical configuration.

After casting and sample loading, gels were immersed in electrode buffer (pH 8.3) containing 25 mM Tris, 192 mM glycine, and 0.1% SDS (prepared using electrophoresis grade chemicals from ICN), and electrophoresed (100 V, ~90 minutes) until the bromophenol blue marker reached 2 mm above the bottom of the resolving gel. Apparatus (Mini-Protean II) and the power supply (PowerPac 300) were from Bio-Rad.

2.6.2 Staining of gels for total proteins

Total proteins within gels were stained by rocking the gel for 1 hour in staining solution (50 ml) consisting of 5% acetic acid, 40% methanol, and 0.05% Coomassie blue R-250. Excess stain was removed by washing in changes of destain (5% acetic acid in 40% methanol) until the background was clear. Stained gels were mounted on blotting paper and dried using a vacuum dryer (Bio-Rad). Coomassie blue staining of gels containing cell lysates and molecular weight markers, allowed visualisation of the protein controls and verified loading of the cell lysates in individual lanes.

66 2.6.3 Transfer of electrophoresed proteins to PVDF membrane

Proteins were transferred from gels to membranes as an initial step in Western and ligand blotting (see 2.6.5 and 2.6.6). In preliminary experiments, PVDF (polyvinylidene difluoride, from NEN) membranes were found to display enhanced signal detection over nitrocellulose for the proteins investigated in this work and therefore PVDF was chosen for all further studies (data not shown).

The gel was washed twice in transfer buffer (25 mM Tris, 192 mM glycine, 10% methanol, pH 8.3) for 10 minutes. PVDF membrane (10 x 8 cm) was prepared for transfer by, activating with methanol (1 minute), washing with water (2 minutes), and pre-equilibrating in transfer buffer (5 minutes or longer). A piece of thick blotting paper was soaked in transfer buffer and the membrane was placed on top. The gel was mounted on the membrane and features of the gel, including its boundaries, position of the tracking dye, locations of the wells, and position of stacking/resolving gel interface, were marked on the membrane with a soft pencil. A second piece of thick blotting paper soaked in transfer buffer was place on top and the sandwich was inserted between fibre pads in a cassette. With the membrane facing the anode, the cassette was placed in a wet transfer apparatus (Bio-Rad) together with a plastic cartridge containing ice, and the tank was filled with cold transfer buffer. Transfer proceeded at 100 V with stirring for 90 minutes.

Following transfer, the membrane was removed, and allowed to air dry for 2 hours. In some cases, the efficiency of transfer was verified by staining the gels for residual protein (see 2.6.2).

The area corresponding to gel lanes containing molecular mass markers was cut from the membrane and stained as described in 2.6.2). These were later re-united with the body of the membrane for molecular mass determinations. The body of the membranes were blocked with 5% milk in PBS (see 2.6.4) and further processed by either Western blotting (see 2.6.5), or by probing with 125I-labelled proteins in ligand blots (see 2.6.6).

67 2.6.4 Blocking of membranes in milk

In preparation for Western blotting (see 2.6.5) and ligand blotting (see 2.6.6), the highly adsorptive surfaces of PVDF membranes (containing electrophoresed proteins, (see 2.6.3 were blocked by activating the membrane with methanol (1 minutes), washing with water (2 minutes), and incubating with 5% skimmed bovine milk powder in PBS for 2 hours on a platform rocker. Prior to blotting, membranes were washed twice for 5 minutes with PBSc.

As described in 2.6.3, sections containing unstained molecular weight markers were cut from the membrane and stained with Coomassie blue as described in 2.6.2 Failure to remove these lanes prior to milk treatment resulted in heavy background staining and loss of marker position information.

In preliminary experiments, Tween 20 (2%), bovine serum albumin (5%), and Haemacel (10 g/L) in PBS or in 10mM Tris-HCl buffered saline were found to be less effective blocking agents than milk in minimising background signals. To verify its suitability as a blocking agent, samples of bovine skimmed milk powder (50 g) were electrophoresed, transferred to PVDF membrane, and verified for lack of reactivity with 125I-labelled ligands and antibodies used in this work. Tween 20 (2%) in PBS was used as the blocking agent for these studies (data not shown).

2.6.5 Western blotting

To detect the presence of specific antigens, PVDF membranes were blocked with milk (see 2.6.4), washed twice with PBS for 5 minutes, and incubated under gentle rocking conditions for one hour at room temperature with primary antibodies diluted to 2 g / ml (for monoclonal antibodies) in PBS (5 ml) containing 1% BSA. In some experiments, affinity purified rabbit antibody against LRP1 -chain (“777”) was used at 0.2 g / ml under the same conditions. The primary antibodies were removed and the membranes were washed four times for 10 minutes in PBS with rocking. Horseradish peroxidase (HRP) conjugated secondary antibody (sheep anti-mouse immunoglobulins, or goat anti-rabbit immunoglobulins, Dako) mixed in PBS / 1% BSA (5 ml) to a dilution recommended by the manufacturer (generally 1:1000), was incubated with the membranes for 1 hour. The secondary antibodies were removed and the membranes

68 were washed four times in PBS for 210 minutes. HRP was detected by treatment of the membranes for 3 minutes with freshly mixed chemiluminescence reagent (NEN) followed by exposure to autoradiography film (Hyperfilm-MP, Amersham) for up to 10 minutes in a stainless steel cassette (Sigma). Films were developed in an automatic processor.

2.6.6 ¹2 5I-Ligand blotting

To detect the presence of receptors, PVDF membranes were blocked with milk (see 2.6.4), washed twice (5 min) with PBS / 0.5 mM calcium (PBSc), and treated for 1 hour at room temperature with 125I-labelled ligands diluted in PBSc / 1% BSA. 125 Concentrations were generally 10 nM except for activated I-2macroglobulin which was 1 nM. Labelled membranes were washed four times with PBSc for 10 minutes under gentle rocking conditions, and dried in an oven at 60C for 2 hours. Bound radiolabelled ligands were detected by exposure of the membranes to autoradiography film (Hyperfilm-MP, Amersham) for 1 or 2 days at -80C in a stainless steel holder (Sigma) with an intensifying screen (NEN). Films were developed in an automatic processor.

Chapter 3

3 UROKINASE (UPA) BINDING AND ENDOCYTOSIS BY HEPG2 CELLS

70 3.1 INTRODUCTION

The generation of plasmin from plasminogen by the plasminogen activators (PAs), tissue-type PA (tPA) and urokinase-type-PA (uPA), is important in the breakdown of fibrin, extracellular matrix, and membranes (reviewed inRijken and Lijnen, 2009; Toriseva and Kahari, 2009). Removal of these barriers to cellular movement is fundamental to diverse physiological and pathophysiological processes including fibrinolysis, ovulation, embyrogenesis, wound healing, inflammation, and tumour metastasis. By the early 1980, it was know that injected tPA is rapidly and exponentially cleared from the circulation with a T1/2 of about 1 minute for rats (Emeis et al., 1985), 2 minutes for rabbits (Korninger et al., 1981), and 2 to 3 minutes for dogs (Fong et al., 1988). These studies showed that the main site of removal was the liver as demonstrated directly by liver perfusion (Emeis et al., 1985), and by prolongation of clearance time following hepatectomy (Korninger et al., 1981) to a T1/2 of 40 minutes (Nilsson et al., 1985). Following this rapid sequestration by the liver, fragments of tPA appeared in the blood (Beebe and Aronson, 1986; Korninger et al., 1981), stomach, and intestines (Nilsson et al., 1985). The disappearance of tPA injected into rabbits is biphasic with T1/2 and T1/2 values of < 1 minute and ~10 minutes (Beebe and Aronson, 1986; Beebe and Aronson, 1988).

tPA is endocytosed by liver parenchymal and endothelial cells (Davidsen et al., 1985; Einarsson et al., 1988), but little by Kupffer cells (Davidsen et al., 1985). In the intact liver, the majority of injected tPA accumulates in parenchymal cells (Bugelski et al., 1989), and degradation occurs in lysosomes (Bakhit et al., 1987; Einarsson et al., 1988).

The hunt for the hepatic receptor(s) responsible for tPA clearance initially focussed on the known glycoprotein receptors. Early studies discounted any involvement of these receptors (Bakhit et al., 1987; Emeis et al., 1985), although Type- II tPA, which lacks one of the three N-glycosylations structures on Type 1 tPA, displayed a longer β phase clearance time (Beebe and Aronson, 1988). The galactose receptor remained excluded in subsequent studies, along with the fucose receptor (Narita et al., 1995), but the mannose receptor was found to bind tPA (Otter et al., 1992c), and to contribute to its  phase clearance (Smedsrod and Einarsson, 1990). Receptor blocking studies suggest that the mannose receptor is one of two hepatic

71 receptor systems mediating tPA clearance (Biessen et al., 1997; Narita et al., 1995), but activity of this receptor is restricted to hepatic endothelial cells and is absent on parenchymal cells (Otter et al., 1992a).

The rate of clearance of tPA is enhanced by formation of complexes with PAI-1 (tPAPAI-1). In perfused rat liver, tPAPAI-1 cleared at twice the rate of tPA (Wing et al., 1991b), and binds specifically to high affinity sites on both isolated hepatocytes and HepG2 cells (Wing et al., 1991a). Sites in the growth factor domain of tPA (Bassel- Duby et al., 1992), and possibly the finger region (Camani and Kruithof, 1995) are important to its rate of endocytosis and degradation.

Early clearance studies in vitro used a variety of cell types including HepG2, which is a cell line derived from an hepatocellular carcinoma (Aden et al., 1979; Knowles et al., 1980), and is widely considered as a reliable model system for human hepatic parenchymal cells. With these cells, tPA binding, endocytosis, and degradation was shown to require initial formation of a complex with PAI-1 according to the model proposed in Figure 1-4 (Morton et al., 1989; Morton et al., 1990; Owensby et al., 1989; Owensby et al., 1991; Owensby et al., 1988). In this system exogenous tPA initially forms an SDS-stable complex with PAI-1 bound to vitronectin in the extracellular matrix. The formation of tPAPAI-1 complexes induces conformational changes in both the PAI-1 and vitronectin which result in the release of the complexes into the supernatant from where they bind to a receptor on the surface of the cells (Owensby et al., 1991). The entire ligand and receptor is subsequently internalised, and the PAPAI-1 complexes are eventually delivered to lysosomes for degradation while the receptor is recycled to the cell surface. Within the lysosomes, tPA and PAI-1 are degraded at different rates (Underhill et al., 1992).

While the catabolism of tPA by HepG2 had been extensively investigated, similar studies had not been performed on the other human PA, uPA. uPA is structurally homologous to tPA, except for the absence in uPA of the finger and second kringle domains of tPA. It is reasonable to suspect from this homology that the interactions of uPA with HepG2 might be similar to those of tPA, but supporting studies had not been performed to test this hypothesis in the early 1990s when the experiments presented here were commenced.

72 Urokinase is synthesised as a single-chain polypeptide, which possesses little proteolytic activity. Cleavage in the linker region between the kringle and serine protease domains generates the enzymatically active two-chain form of uPA (tc-uPA) (see Figure 1-3). tc-uPA, (referred to below as HMW-uPA), consists of an A-chain (the growth factor and kringle domains) and a B-chain (the serine proteinase domain) tethered by a single disulfide bond. Further cleavage in the linker region leads to almost complete loss of the A-chain and the formation of a low molecular mass form of uPA (LMW-uPA), which retains proteolytic activity. The interactions of both HMW and LMW forms of uPA with HepG2 cells were investigated to determine the role of the A-chain. An immediate advantage of including LMW-uPA is exclusion of the growth factor domain, which harbours the binding sequence for the uPA receptor. In the results to follow, similar patterns were established for the binding of both HMW-uPA and LWM-uPA. These findings allow the conclusion that the uPA receptor is not involved.

The comparison of uPA and tPA interactions with HepG2 was begun by previous members of our team who are the co-authors on our paper reporting these findings (Grimsley et al., 1995). Their initial studies utilised HMW-uPA and LMW-uPA and established that: (a) the dissociation constants (Kd) and number of binding sites per cell were similar for tPA, HMW-uPA, and LWM-uPA (in nM /and/ sites per cell - 1.8 // 86x103, 4.1 // 78x103, and 3.9 // 83x103 respectively); (b) the binding of the two individual forms of uPA was inhibited by tPA and cross-competed by each other; (c) the binding of both HMW-uPA and LMW-uPA could be inhibited by an antibody to PAI-1; and (d) the two forms of uPA were degraded by HepG2 at similar rates.

Together, this data provided convincing evidence that uPA and tPA were bound by a common receptor (findings a and b), and that binding was dependent on PAI-1 (finding c). Furthermore, the rate of uPA degradation (finding d) was shown to be similar to rates previously reported for tPA. Binding and degradation of HMW-uPA and LMW-uPA demonstrated that the growth factor and kringle domains of uPA are not required for these processes, indicating that the uPA receptor is not involved.

These results were derived from counting radioactivity in cell lysates and did not provide direct evidence for the formation of SDS-stable uPAPAI-1 complexes during the binding and internalisation of uPA. Accordingly, after verifying some of the previous findings, my contribution to this work centred on demonstrating the formation of PAI-1 complexes with HMW-uPA and LMW-uPA. Together, these results suggest

73 that HepG2 binds and internalises uPA in the form of SDS-stable uPAPAI-1 complexes, and that the scheme in Figure 1-4 is applicable to both tPA and uPA.

Hypothesis:

To determine whether uPA binding and internalisation by HepG2 cells is analogous to the PAI-1 dependent mechanism previously determined for tPA.

3.2 METHODS

3.2.1 Separation of HMW-uPA and LMW-uPA

Initial studies by co-workers used purified HMW-uPA and LMW-uPA provided by J. Henkin, Abbott Park, Illinios. These gifts had been consumed. To continue the work, a fresh source of HMW and LMW forms of uPA was needed. Similar forms of uPA were found in expired packages of pharmaceutical uPA (Ukidan, kindly provided by Serono Australia), although the precise cleavage site in this form of LMW-uPA was not determined.

To separate HMW-uPA and LMW-uPA, five vials of Ukidan containing 100,000 IU (~2 mg of protein) per vial were reconstituted in 2 ml of 20 mM MES-HCl at pH 6.0 (MES buffer) and dialysed twice against the same buffer. The solution was sterilised using a 0.2  filter (Minisart filter, Sartorius) fitted to a 5 ml syringe, then applied to a 2 ml S-sepharose column (Mono-S, Pharmacia) equilibrated with MES buffer on an FPLC apparatus (Pharmacia). The proteins were eluted with a linear gradient of NaCl in the range 0-500 mM. An example of the fractionation of Ukidan into HMW-uPA and LMW-uPA, as monitored by SDS-PAGE analysis, is shown in Figure 3-1 together with a typical elution profile from S-sepharose. Separated proteins represented 54% recovery of the starting material.

Figure 3-1 Fractionation of Ukidan into high and low molecular weight forms of two-chain uPA (HMW-uPA, LMW-uPA). The Coomassie-stained SDS-PAGE gels in Panel a show the presence of two proteins in the urokinase pharmaceutical, Ukidan, and the same proteins following their separation on S-sepharose. The gel on the right shows that HMW-uPA dissociates into two bands under reducing conditions, consistent with Ukidan consisting of urokinase in the form of catalytically active tc-uPA. Panel b contains a trace obtained during the separation of Ukidan by FPLC on S-sepharose. Representative data from multiple fractionations are shown.

3.2.2 Iodination of HMW-uPA, LMW-uPA, and tPA

PAs (50g) were iodinated using Iodogen as described in 2.5.2. For the results in Figure 3-2, the specific activities of 125I-HMW-uPA and 125I-LWM-uPA were 30 Bq/fmol and 6 Bq/fmol and this difference is reflected in the intensity of the bands.

3.2.3 Autoradiography of 125I-uPA bound to HepG2 monolayers

HepG2 cells, plated and cultured for three days in 12 well plates as described in Section 2.1, were washed three times in cold PBSc then incubated for 2 hours on ice with 2, 4, 8, or 16nM 125I-HMW-uPA or 125I-LMW-uPA in 0.5 ml of binding buffer consisting of Eagles minimum essential medium with Earle‟s salts without serum and bicarbonate (EMEM, which contains 2.5mM calcium), 20 mM HEPES (pH 7.3), and 10M EACA. The specificity of binding was determined in additional wells containing

75 unlabelled LMW-uPA, HMW-uPA, or tPA at 1M. Following removal of the reactants, the cells were washed three times with PBSc (Dulbecco‟s phosphate buffered saline with 0.5mM calcium) then lysed for 30 minutes at RT on a platform rocker using 100 l of Laemmli sample buffer containing 0.1% SDS. The use of this low concentration of SDS (0.1%) instead of the 2% specified in the original Laemmli sample buffer, allowed better preservation of uPAPAI-1 complexes. Lysates and radiolabelled starting ligands were electrophoresed (see Section 2.6) using 10% polyacrylamide gels containing 0.1% SDS (SDS-PAGE), then fixed and stained (see 2.6.2). Following drying on a BioRad gel drier, the gels were exposed to film (Hyperfilm MP, Amersham) for two weeks at room temperature. The molecular masses of bands were determined by overlaying the developed film with its dried gel and comparing migrations with those of known standards (BioRad low molecular weight standards).

3.2.4 Detection of PAI-1 associated with 125I-HMW-uPA following interaction with HepG2

Under conditions similar to those used to examine PAI-1 complex formation (see 3.2.3 above), HepG2 cultured for three days in 12-well plates were washed and incubated for 90 minutes on ice with 3 nM 125I-HMW-uPA in the presence of a range of concentration of anti-PAI-1 antibody. This antibody had been raised in rabbits by our team members, G. Joulianos and R. Brand, using PAI-1 purchased from American Diagnostica.

Following incubation, the cells were washed and lysed in 0.1% SDS Laemmli buffer. Lysates were processed to produce autoradiograms as detailed in 3.2.3 above.

3.2.5 Internalisation of surface-bound 125I-PAPAI-1 ligands

To demonstrate that HepG2 cells are capable of internalising bound 125I-PAPAI-1, the sensitivity of these ligands to degradation by exogenous trypsin was observed following a brief cyclic exposure of the cells to 0C, 37C, and 0C. This procedure transiently restores endocytic competence. HepG2 cells at 3 days of culture

76 were washed twice in PBSc, then incubated in binding buffer on melting ice for 15 minutes to arrest metabolic activity and thereby inhibit endocytosis. Cells were retained on ice for a further hour while exposed to 3 nM 125I-tPA or 125I-HMW-uPA in the presence or absence of matching unlabelled PAs at 1M. Following removal of these reaction mixtures and washing twice in ice-cold PBSc, half of the cells were placed in a water bath at 37C for 10 minutes to re-activate endocytosis, while the other half remained on ice as matched controls. All cells were then collectively incubated on ice again for a further 15 minutes to inhibit endosomal delivery of the internalised complexes to lysosomes. At this point, 0.02% bovine trypsin (Sigma) was added to some cells to digest exposed ligands. Following incubation on ice for 30 minutes, a 10 molar excess of aprotinin (Sigma) was added and incubated for 5 minutes to inhibit further digestion by trypsin. Under these conditions, HepG2, which is an adherent cell line, is released from the culture plates by the action of the trypsin. Released cells were gently resuspended, transferred to centrifuge tubes, and centrifuged at 10,000g for 3 minutes. The supernatants in these tubes were carefully removed and the cell pellet was lysed in Laemmli sample buffer containing 0.1% SDS. Adherent cells (controls not exposed to trypsin) were lysed similarly after washing twice with PBSc. Lysates were separated by SDS-PAGE using 10% acrylamide, and the gels were fixed, stained, dried, and exposed to Hyperfilm for two weeks.

3.3 RESULTS

3.3.1 Conversion of uPA to uPAPAI-1 ligands.

To determine whether uPA forms complexes with PAI-1 during interactions with HepG2, 125I-HMW-uPA and 125I-LMW-uPA were incubated with these cells in the presence and absence of unlabelled tPA, HMW-uPA and LMW-uPA. Cell lysates prepared from these experiments were examined by autoradiography and the results are displayed in Figure 3-2. The molecular mass of each starting ligands is observed to increase by approximately 50-kDa, consistent with the formation of 125I-HMW-uPAPAI-1 and 125I-LMW-uPAPAI-1 complexes. Moreover, band intensities at these positions are observed to correspond with the level of specific binding in a dose-dependent manner. The specificity of these signals is demonstrated

Figure 3-2 HMW-uPA and LMW-uPA form SDS-stable complexes with PAI-1 during binding to HepG2 cells. The closed arrows show the positions of the starting reactants, 125I-HMW-uPA at 54-kDa (left panel) and 125I-LMW-uPA at 32-kDa (right panel). The open arrows show that the positions of the reactants increase by approximately 50-kDa when bound to HepG2 monolayers, suggesting that these signals correspond to the formation of 125I-HMW-uPAPAI-1 and 125I-LMW-uPAPAI-1 complexes. The positions of molecular weight markers are shown as arrow heads on the right of each panel. The ability of 1 M unlabelled LMW-uPA, HMW-uPA, and tPA to cross compete for binding of 125I-HMW-uPA and 125I-LMW-uPA is demonstrated for a single dose (4 nM) of these reactants in the last three lanes in each panel. The intensity of the signals corresponding to the complexes of each PA is observed to follow a dose dependency. Differences in the intensities between panels reflects the specific activity of radiolabelling. Representative blots from duplicate experiments are shown.

by their disappearance in the presence of matching unlabelled HMW-uPA or LMW-uPA. Moreover, binding of 125I-HMW-uPAPAI-1 and 125I-LMW-uPAPAI-1 is shown to be cross-inhibited by unlabelled LMW-uPA and HMW-uPA, and by tPA. This cross competition among the PAs implies that each is binding to common sites. These results suggest that uPA initially forms complexes with PAI-1 during interactions with HepG2 monolayers, and that identical sites are responsible for binding both tPA and uPA.

3.3.2 Identification of PAI-1 associated with 125I-HMW-uPA following exposure to HepG2

To verify that PAI-1 associates with 125I-HMW-uPA and is responsible for its an apparent 50-kDa increase in molecular mass during binding to HepG2, a purified rabbit

78 anti-PAI-1 antibody was added at various concentrations to 3 nM 125I-HMW-uPA and the mixtures were allowed to react with HepG2 monolayers. The resulting cell lysates were used to produce the autoradiogram in Figure 3-3 which shows that the presence of anti-PAI-1 at concentrations above 3 nM induced disruption of the bands at approximately 105-kDa. The pattern of signals appearing at the top of the resolving gel in the lanes corresponding to high concentrations of anti-PAI-1, suggests that this particular antibody causes precipitation of the ligand. These results provide evidence that the signals at approximately 105-kDa correspond to a complex between 125I-HMW-uPA and PAI-1.

Figure 3-3 Inhibition of 125I-HMW-uPA binding to HepG2 cells by anti-PAI-1 antibody. The autoradiogram shows the dose-dependent disruption of signals at ~105-kDa by increasing concentrations of anti-PAI-1 antibody, suggesting that the ligand bound to HepG2 is 125I-HMW-uPAPAI-1 complex. Results of a single set of determinations are shown.

79 3.3.3 Internalisation of PAs

To demonstrate that 125I-HMW-uPA binds to an endocytic receptor on HepG2, the ability of the cells to internalise the ligand was addressed. The autoradiograms in Figure 3-4 show the result of 125I-HMW-uPA (upper panel) and 125I-tPA (lower panel) interactions with HepG2 monolayers. In these experiments, cells were initially incubated on ice to inhibit their endocytic capacity, and were then exposed to radiolabelled ligands. This procedure prevents internalisation and thereby restricts ligand distribution to extracellular binding sites. Subsequently, by raising the temperature briefly, the cells were permitted to regain endocytic function, and some surface bound ligand was internalised. During the brief time allowed for metabolic re-activation, ligands would be expected to reach endosomal compartments, but not lysosomes, and would therefore remain intact inside the cells. To detect the presence of internalised ligands cells were treated with trypsin to digest exposed ligands and any signals remaining were deemed to be protected from tryptic digestion as a result of internalisation. In contrast, ligands bound to control cells kept entirely on ice throughout the experiment, are entirely degraded by the trypsin treatment.

The procedures followed to produce lysates for the autoradiograms in Figure 3-4 were identical for the upper and lower panels, except that different PAs were used in each. The final lane in each panel shows (somewhat overloaded) the 125I-labelled PAs used as starting reactants. The first two lanes were prepared from control cells which had been processed through the entire internalisation protocol, but had not been exposed to trypsin. The signals in these lanes reflect typical binding of PAs in the form of PAPAI-1 complexes as observed previously (Figure 3-2). The treatment of cells in the first two lanes differed only by their brief exposure to 37C conditions as indicated. The similarity of signal intensities between these lanes demonstrates that this brief temperature differential has little effect on the retention of bound ligands.

Cells in the third and fourth lanes were treated by the same procedures used on cells in the first two lanes, expect that these cells were treated with trypsin. The complete loss of signal in the third lane demonstrates that ligands on these cells were located entirely on extracellular binding sites. In contrast, the cells in the fourth lane were exposed to 37C conditions and show the retention of some signal (albeit weak) representing intact PAPAI-1 complexes. These conditions permit the cells to resume

80 endocytic functions with the result that ligands are internalised and protected from digestion by trypsin. In control experiments, non-specific binding was determined in the presence of matching unlabelled PA, and was found to be negligible (data not shown).

Together, these results show that pre-bound PAPAI-1 complexes can be internalised by HepG2, implying that at least some of these complexes were bound to an endocytic receptor. Moreover, the observation that both tPAPAI-1 and

Figure 3-4 Internalisation of 125I-PAPAI-1 complexes by HepG2 cells. The autoradiograms represent lysates of HepG2 loaded with 125I-HMW-uPA or 125I-tPA, on ice, washed, and either kept on ice (lane 1) or briefly (10 minutes) incubated at 37C (lane 2). Little change in ligand signal intensity is observed between these conditions for either PA. Additional cells similarly loaded with 125I-PAs, are observed to lose signal completely when treated with trypsin (lane 3) unless briefly incubated at 37C to reactivate endocytic function transiently and thereby protect some ligand from digestion (lane 4). Representative sets of blots from multiple experiments are shown.

81 HMW-uPAPAI-1 are internalised provides further support for a shared mechanism responsible for the processing of tPA and uPA by HepG2 cells.

3.4 DISCUSSION

This study demonstrates the conversion of two catalytically active forms of uPA (HMW-uPA and LMW-uPA) to complexes with the serpin inhibitor, PAI-1, during their interaction with HepG2 monolayers. The results are analogous to those reported previously for the homologous plasminogen activator, tPA, whose binding to HepG2 cells has been shown to be PAI-1 dependent. The predominant ligand bound to HepG2 following incubation with uPA is approximately 50kDa larger that the starting material. This is consistent with the formation of PAPAI-1 complexes and is observed for both HMW and LMW forms of uPA. These observations are made possible by the stability of PAPAI-1 complexes in low concentrations of the denaturing detergent, SDS, which is a necessary component of the electrophoretic system used. The association of uPA with PAI-1 was confirmed by using an anti-PAI-1 antibody, which caused inhibition of uPA binding to HepG2.

The levels of PAI complexes found associated with HepG2 was dose-dependent with respect to the concentration of 125I-uPA reactants applied, implying that conversion to the complex is a requirement for binding. Furthermore, binding of both HMW-uPA and LMW-uPA is cross-inhibited by each other, and by tPA, suggesting that the binding sites for tPA and uPA are identical. In addition to similarities in binding patterns, further processing by HepG2 is analogous for tPA and uPA in that both are internalised as PAI-1 complexes. Together, these results imply that both uPA and tPA bind, and are internalised, through a common PAI-1 dependent mechanism.

Previous studies in our lab had determined using quantitative analysis that binding parameters for tPA, HMW-uPA, and LMW-uPA with HepG2 were similar, and that binding of each could be inhibited by the addition of an antibody to PAI-1 (Grimsley et al., 1995). Our group also showed that the catalytically inactive form of uPA (pro-uPA, single-chain uPA), which does not form complexes with PAI-1, displayed no specific

82 binding to HepG2 monolayers. This finding provided further evidence that interactions with PAI-1 is a prerequisite for binding of uPA to HepG2.

These results from co-workers provided strong evidence to suggest that the binding of uPA to HepG2 is dependent on PAI-1, but the demonstration of uPAPAI-1 complexes in electrophoretic studies had not been performed. Evidence for the presence of these complexes during the binding and internalisation stages is presented in Figure 3-2 and Figure 3-4, and the identity of PAI-1 is evident in Figure 3-3. These results represent my contribution to these studies. In addition, a method was established for the purification of HMW-uPA and LMW-uPA from the pharmaceutical, Ukidan. This became necessary when original supplies of these reagents were exhausted.

Three lines of evidence strongly suggest that the signals for 125I-HMW-uPA at ~105-kDa correspond to complexes with PAI-1. Firstly, the magnitude of increase in molecular mass, for both HMW-uPA and LMW-uPA, is corresponds to the addition of 50kDa PAI-1. Secondly, the stability of this higher molecular mass species is consistent with the formation of an enzymeserpin-complex such as uPAPAI-1, because most non-covalent protein-protein interactions are disrupted in SDS. Thirdly, the inhibition of binding by anti-PAI-1 supports the identification of PAI-1 in the actual binding ligand (Figure 3-3). These considerations support assignment of the signals at ~105-kDa and ~80-kDa as HMW-uPAPAI-1 and LMW-uPAPAI-1, respectively.

A doublet of bands sometimes appears for uPAPAI-1 complexes. It is likely that these represent the slow loss of a 4 kDa fragment cleaved from PAI-1 during its normal process of inhibition. An example appears in Figure 3-3. The molecular species in both bands behave similarly in terms of binding.

The nature of the binding site was investigated by determining whether the ligands could be internalised by the cells. Results indicated that a portion of surface- bound ligand could be protected from digestion by trypsin following brief re-activation of endocytic competence. This finding is consistent with transport of ligands from extracellular locations to the interior of the cell where trypsin cannot penetrate. While some ligand is protected following internalisation, the intensity of their signals indicates a failure to protect the entire extracellular cohort. This observation may be explained by several considerations. Firstly, the time available for internalisation is insufficient for endocytosis of all surface-receptor bound ligand. Secondly, some ligand may be

83 associated with surface membrane receptors or accessory molecules which are not endocytic. Thirdly, some ligands may dissociation from their binding sites into the supernatant while awaiting internalisation. However, this process does not appear to be significant because controls not treated with trypsin show that most signal is retained during the brief incubation at 37C. Finally, 125I-HMW-uPA bound to HepG2 monolayers may not necessarily be associated with cellular receptors, but rather, with binding sites within the matrix. In this regard, the formation of complex with PAI-1 associated with vitronectin in the ECM is considered to result in the release of the complex from the ECM following conformational changes in both the inhibitor and vitronectin. However, it is possible that a portion of the PAPAI-1 complex could remain associated with ECM where it is not subject to internalisation. This possibility is supported by studies reporting that PAI-1 forms a significant binding site for tPA in endothelial cells monolayers.

The experiments presented here suggest that 125I-HMW-uPA, like tPA, is converted to a complex with PAI-1 during interaction with HepG2, and that the binding sites for uPA are shared with tPA. Initial characterisation indicates that at least a portion of these binding sites are endocytic receptors. Furthermore, the similarity of PAI-1 dependence and binding properties displayed between HMW-uPA and its homologue, LMW-uPA (which lacks the binding site for the uPA receptor), excludes a major contribution to PA binding to HepG2 by the uPA receptor.

These studies focussed on the nature of the actual ligand formed during binding of uPA to HepG2 and presents initial characterisation of the receptors involved. During the course of these studies, evidence was mounting from other groups, that the receptor mediating the majority of catabolism of tPA (Bu et al., 1993; Bu et al., 1992a; Camani et al., 1994) and uPA (Conese et al., 1995; Kounnas et al., 1993b; Nykjær et al., 1994; Nykjær et al., 1992) was low density lipoprotein receptor-related protein (LRP1). Furthermore, inhibition of LRP1 in vivo prolonged the clearance of tPA (Warshawsky et al., 1993). A little later, LRP1 was shown to mediate the endocytosis of uPA bound to the uPA receptor (Higazi et al., 1996), opening the involvement of LRP1 in cell migration. and cancer. While the results presented here are consistent with involvement of a receptor displaying characteristics of LRP1, total specific binding of tPA to HepG2 at 4ºC cannot be fully attributable to LRP1. These observations are explored further in the next Chapter.

Chapter 4

4 LRP1 LEVELS VARY AMONG HEPG2 SUBLINES

4.1 INTRODUCTION

The PAI-1 dependent binding of urokine-type plasminogen activator (uPA) to the hepatoma cell line, HepG2, was demonstrated in the previous Chapter. In those studies HepG2 was chosen as a model of hepatic parenchymal cells to extend previous investigations using the same cell line (Morton et al., 1989; Morton et al., 1990; Owensby et al., 1989; Owensby et al., 1991; Owensby et al., 1988; Underhill et al., 1992). The work of those authors provided initial characterisation of the receptors involved by showing that at least some plasminogen activators (PAs) bound to an endocytic receptor, and that PAI-1 participated in this process. These findings proved consistent with results from other laboratories which were providing strong evidence that the endocytic receptor mediating catabolism of PAs by HepG2 was low density lipoprotein receptor-related protein 1 (LRP1) (Bu et al., 1993; Bu et al., 1992b; Kounnas et al., 1993b; Narita et al., 1995; Warshawsky et al., 1993; Warshawsky et al., 1994).

LRP1 is a large multifunctional receptor capable of endocytosing a broad range of structurally and functionally unrelated ligands including lipoproteins and proteinaseinhibitor complexes (Blacklow, 2007; Lillis et al., 2008; May et al., 2007). To extend previous studies in our laboratory on the catabolism of PAs, both the presence of LRP1 and its involvement in the binding of tPA by HepG2 was verified in preliminary experiments. In agreement with the work of Bu et al (Bu et al., 1993), tPA binding to HepG2 was found to involve LRP1 and at least one additional binding site. However, in our studies the proportion of binding by tPA to sites unrelated to LRP1 was found to differ considerably from the reported values. Whereas Bu et al (Bu et al., 1993) had reported approximately 75% of tPA binding via LRP1, results using HepG2 maintained in our laboratory averaged only 26%.

To investigate the reason for this difference, fresh HepG2 cells were obtained from the American Type Culture Collection (ATCC) and examined for LRP1-dependent tPA binding in parallel with our existing stock cultures. This comparison revealed that the proportion of tPA binding attributable to LRP1 on standard HepG2 from the ATCC (ATCC HB8065) is even lower than our normal stock cultures (20%). These initial findings prompted the following investigation of comparative antigenic and ligand- binding activities of LRP1 on four HepG2 cell lines held in various laboratories.

86 A major difference between HepG2 available from the ATCC and those used by Bu et al, is that the latter were derived from a clone (termed a16), which was selected for high expression of the asialoglycoprotein receptor (Schwartz and Rup, 1983). Four HepG2 sublines representing parental HepG2 and the “a16” clone, both at high and low passage numbers, were obtained from various laboratories. These have been compared for relative LRP1 levels using anti-LRP1 antibodies, Pseudomonas exotoxin A (an LRP1-specific ligand), and a recombinant form of 39-kDa receptor associated protein (RAP). The most striking finding is the relatively low basal expression of LRP1 by the parental cells available from the ATCC.

Unexpected artefacts can confuse experimental results. For example, the catabolism of the human PAs is often studied in models derived from animals whose recognition of human proteins in clearance pathways may be a tenuous. This issue is not explored further here but deserves constant consideration. Another common assumption is that the characteristics of cell lines are invariant worldwide. The results presented here suggest otherwise for LRP1 on HepG2 and may offer clarification of apparent inconsistencies reported for binding and degradation of PAs. Some of these include: (a) the ability (Grimsley et al., 1995) and inability (Wing et al., 1991a) of tPA and uPA to crosscompete in binding studies; (b) the dependence (Herz et al., 1988; Moestrup et al., 1990; Orth et al., 1992) and independence (Kuiper et al., 1995) on calcium for binding of tPAPAI-1 and other ligands; and (c) the dependence (Morton et al., 1989; Morton et al., 1990; Underhill et al., 1992) and independence (Bu et al., 1992b; Iadonato et al., 1993) on PAI-1 for the binding of PAs to hepatic cells or isolted hepatocytes. These discrepancies may be attributable to the use of cell types with high and low levels of LRP1, especially if LRP1 is not the only binding site.

An example of the use of HepG2 cells apparently lacking LRP1 is the report by Storm et al (Storm et al., 1997) which shows that C1sC1-inhibitor complex is degraded via LRP1 but fails to be degraded by HepG2. Similarly, Patston et al (Patston et al., 1993) also described the inability of C1sC1-Inhibitor to bind to HepG2. These studies appear to conflict directly with the expression of LRP1 on HepG2.

The results presented here suggest that basal LRP1 expression on HepG2 cells varies considerably depending on their culture history. This variability is not normally acknowledged and presents a basis for caution when interpreting results derived from this commonly used cell line.

87 Hypothesis:

To Determine whether the level of binding and endocytic activity of LRP1, the major receptor responsible for mediating tPA clearance, is stable on HepG2 cells kept in continual culture, and in various laboratories..

4.2 MATERIALS AND METHODS

4.2.1 HepG2 sublines

The characteristics of the HepG2 sublines used in this study are described below and summarised in Table 4-1. Low passage number HepG2 cells (catalogue number HB8065) at passage 76 (HepG2 ATCC/LP) and HT1080 cells were purchased from the American Type Culture Collection (ATCC), Rockville, MD. These cells at a high passage number estimated at 300 (HepG2 ATCC/HP), were provided by M. Gallicchio, Monash University, Melbourne, Australia. The HepG2 clone, a16, was a gift from Dr A. L. Schwartz (Washington University, St Louis, MO). One batch of these cells had been stored as frozen stocks in our laboratory at around passage 300 (a16/LP) in 1990. Another, at higher passage number (estimated at 900) was obtained much later (a16/HP) from the same source. All four HepG2 sublines were cultured as described in section 2.1, and were assayed for LRP1 status within 20 passages after arrival in our laboratory.

4.2.2 Materials

Recombinant RAP was prepared as described in 2.5.5. Purified recombinant human tPA synthesised in Chinese hamster ovary cells was donated by Karl Thomae

GmbH, Biberach, Germany. Murine IgG1 monoclonal antibody 8G1 against the 515-kDa LRP1 -chain, and rabbit antibody “777” (affinity purified on immobilised 515-kDa LRP1) were gifts from Dr D K Strickland (American Red Cross, Rockville,

88 Table 4-1 Characteristics and sources of HepG2 used in this study.

Estimated Local Cells Passage Name Origin Number

HepG2 original parental 76 ATCC/LP Hepatocellular carcinoma, cell line archived at the American Type Culture Collection (ATCC)

HepG2 parental cell line 300 ATCC/HP Monash University at high passage number

HepG2a16 clone at low 300 a16/LP HepG2a16 clone stored in passage number nitrogen in our laboratory since 1990

HepG2a16 clone at high 900 a16/HP Washington University passage number

MD). Bovine serum albumin (BSA), cytochrome c, Pseudomonas exotoxin A (PEA), -amino caproic acid (EACA), and HEPES (N-2-hydroxyethylpiperazine-N‟-2- ethanesulfonic acid) were purchased from Sigma, St Louis, MO. Human plasma fibronectin (Mr = 500,000), used as a molecular weight marker in SDS-PAGE (polyacrylamide gel electrophoresis with SDS), was supplied by Life Technologies. All other chemicals were of analytical grade from BDH (Kilsyth, Australia).

4.2.3 Flow cytometry

To verify antigenic expression (results in Figure 4-1), HepG2 cells were harvested and labelled as described in 2.4.1. Primary antibodies, anti-CD7 antigen (Becton Dickinson, Mountain View, CA), Ta1 (anti-CD26 antigen, Coulter, Hialeah, Florida), anti-CD36 antigen (Dako, Glostrup, Denmark), anti-CD44 antigen (Becton Dickinson), 8G1 (anti-LRP1 -chain), or negative control antibody (MOPC21, murine

IgG1, Sigma) were applied at 10 g/ml, or according to the manufacturers‟ instructions.

89 For the experiment examining surface membrane expression of LRP1 as a function of the length of time in culture (results in Figure 4-7), HepG2 sublines were seeded into 6-well plates following initial harvest from stock cultures using trypsin/EDTA. After 1, 2, 3, 7, and 12 days in culture, cells were harvested using EDTA and analysed by flow cytometry using 8G1. Only three HepG2 sublines were included in this experiment as HepG2a16/HP had not arrived in our laboratory at the time.

4.2.4 Iodinations

Iodogen (Pierce, Rockford, IL) was used to label tPA (50g) with 125I (Australian Radioisotopes, Lucus Heights, Australia) to an average specific activity of 20 Bq/fmol by the method described in 2.5.2.

4.2.5 Inhibition of 125I-tPA binding by RAP in serial assays

HepG2 cell lines were plated at 50,000 cells/ml into 24 wells plates (2 cm2 per well) with 1 ml of complete medium (day 0) and subsequently fed by replacement with fresh medium at 48 hour intervals and 24 hours prior to assay. Binding was performed in PBS / 0.2 mM CaCl2 (PBSc) containing 10 mM EACA (binding buffer) in accordance with Bu et al (Bu et al., 1993), and all washing steps were in triplicate using PBSc on ice. On the day of assay, duplicate sets of plates containing the four individual cell lines were washed and preincubated in 1 ml of binding buffer on ice for 15 minutes to inhibit endocytosis. The binding buffer was replaced with 3 nM 125I-tPA in the presence of half log increasing concentrations of RAP or tPA as competitors. Cells incubated with 3 nM 125I-RAP in the absence and presence of unlabelled RAP were included as controls (data not shown). After 90 minutes on ice, the plates were washed and the cells lysed with 0.1 M NaOH / 0.2% SDS for determination of radioactivity in a Packard 12 well gamma counter. Results were expressed as fmol 125I-tPA bound per well. Cell counts were determined using a haemocytometer.

90 4.2.6 Protein assays (4.1-6)

As outlined in 2.5.1, cells lysates were assayed using a BCA protein assay kit (Pierce) by following the manufacturer‟s instructions.

4.2.7 Western blotting and 125I-RAP ligand blotting

HepG2 cells were seeded in 25 cm2 flasks at densities identical to those used in the serial binding assays (25,000 cells per cm2) and fed with the same schedule. On the day of assay, the cells were washed four times with cold PBSc, and cell lysates were prepared as described in 2.2.1. Lysates (50 g protein/lane) were separated by SDS-PAGE (5% acrylamide), then Western blotted using 5 g/ml 8G1 and ligand blotted using 10 nM 125I-RAP, according to procedures in 2.6.5 and 2.6.6 respectively.

4.2.8 Pseudomonas exotoxin A (PEA) sensitivity assay

HepG2 cells in 1 ml of complete medium in the absence and presence of 10 ng/ml PEA were cultured for 3 days, washed 3 times with PBSc, lysed in 0.1 M NaOH / 0.2% SDS, and assayed for total protein as an indication of cell number. Results were expressed as the percentage of protein from cells surviving PEA treatment compared to control cultures.

4.2.9 Degradation assays

Day 4 HepG2 plated and fed under identical conditions to those used in the serial binding assays (4.2.5), were washed three times in PBS / 0.5 mM calcium (PBSc) and incubated in a water bath in air at 37C with 1 ml of reaction buffer consisting of EMEM / 0.1% BSA / 20 mM HEPES (pH 7.4), and 5 nM 125I-tPA with and without added 1.5 M tPA or 1.5 M RAP. Periodically, 0.5 ml aliquots of supernatant from individual wells were transferred to 0.5 ml 20% trichloroacetic acid / 4% phosphotungstic acid and assayed for the presence of 125I-containing fragments by the procedure described in 2.5.4. Cell-associated radioactivity was also determined

91 following washing of the wells three times with PBSc and lysis of the cells with 0.1 M NaOH / 0.2% SDS.

4.3 RESULTS

4.3.1 Surface membrane phenotyping of HepG2 cells

As an initial step, the surface membrane phenotype of the four HepG2 sublines was verified. The panel of antibodies used were selected on the basis of results from the Fifth International Workshop on Leukocyte Differentiation Antigens (Shaw, 1994) which extensively characterised several adherent cell lines including HepG2 and the human fibrosarcoma cell line, HT-1080 (ATCC CCL-121). HepG2 were reported to express CD7, CD26, and CD36 antigens and to be negative for CD44 antigen. Of the adherent cells examined in the workshop, this pattern was highly specific for HepG2, and matching expression (positive or negative) of any one marker was rare for other adherent cell types. As shown in Figure 4-1, the four HepG2 sublines all displayed an identical phenotype of CD7pos, CD26pos, CD36pos, and CD44neg, which verified all cells as HepG2. Antibody specificity was controlled by inclusion of HT-1080 cells. Typical of many adherent cell lines, HT-1080 displays a phenotypic profile opposite that of HepG2 for the markers selected markers. That is, CD7neg, CD26neg, CD36neg, and CD44pos. The results provide strong evidence for the identification of the sublines as HepG2.

Figure 4-1 Plasma membrane phenotyping of HepG2 sublines. Each panel contains flow cytometric histograms obtained from cells treated with a detecting antibody (shaded histogram) as indicated on the abscissa, or with a negative control antibody (open histogram overlay). Primary antibodies were developed with FITC conjugated sheep anti-mouse immunoglobulins (FITC-SaMIg). Results in each row of panels were obtained using an individual HepG2 subline or HT1080 fibrosarcoma cells as indicated.

93 4.3.2 ¹2 5I -tPA serial binding assays

To compare the LRP1 ligand binding capacities of the HepG2 sublines, 125I-tPA binding at 4C in the absence and presence of RAP was performed following periods in culture between 1 and 9 days. The inhibition of 125I-tPA binding caused by the presence of increasing concentrations of unlabelled tPA and RAP using the four HepG2 sublines is depicted in Figure 4-2.

To allow direct comparisons, all cell lines were cultured and assayed simultaneously under identical conditions. Binding of 125I-tPA to all sublines was specific, as observed by the convergence of binding to a nonspecific binding (NSB) point of ~8 fmol/well or less in the presence of excess (1 M) unlabelled tPA (left column of panels in Figure 4-2). However, RAP inhibition of 125I-tPA binding (central column of panels) varied considerably with both the culture period and the cell line. The proportion of 125I-tPA specific binding inhibited by 1 M RAP (right column of panels) reveals the relative contribution to binding by LRP1. These studies indicate a consistently higher level of 125I-tPA binding to LRP1 on the high passage HepG2a16 clone (a16/HP), whereas RAP inhibition of 125I-tPA binding was least on the low passage ATCC line (ATCC/LP). The other two lines were intermediate between these results.

Total 125I-tPA binding by a16/HP was approximately 1.5 fold higher at day 7 compared to the other HepG2 lines even though cell numbers were similar for each of the lines (approximately 1.3x106 cells per well by day 7). However, the level of 125I-tPA binding in the presence of 1 M RAP appears relatively uniform among the four HepG2 lines, suggesting that the differences in total binding are caused by the differences in the expression of LRP1. These results are consistent with the existence of two binding sites for tPA on HepG2 at 4 C as previously observed by Bu et al (Bu et al., 1993).

Figure 4-2 125I-tPA binding to HepG2 sublines. HepG2 cells in culture for 1 to 9 days were washed and incubated at 4C with 3 nM 125I-tPA in the absence and presence of increasing concentrations of tPA (left column of panels) or RAP (central column of panels). The level of binding expressed as fmol / well is shown for cells in culture at day 1 ( ), 2( ), 3( ), 4( ), 5( ), 7( ), and 9( ). Each row of panels contains the results from an individual HepG2 subline, as indicated, and rows are arranged in decreasing order of observed LRP1 expression. The bar graphs in the right column of panels indicate the percentage by which 125I-tPA specific binding is inhibited in the presence of the 1 M RAP. Each data point is the mean of duplicate determinations. Error bars are omitted for clarity but the range was generally within the size of the symbol.

95 To facilitate direct comparisons of their relative tPA binding capacities, all four cell lines were analysed simultaneously in duplicate at each time point. Similar data to that displayed in Figure 4-2 were obtained in additional experiments and the average levels of 125I-tPA binding inhibition by RAP are summarised in Table 4-2. In these experiments, up to 80% (average 65%) of the 125I-tPA bound to the high passage HepG2a16 clone (a16/HP) was inhibited by the addition of RAP. In contrast, identical experiments using the low passage cells from the ATCC (ATCC/LP) showed only 0 - 45% (average 20%) inhibition. These results suggest that the variations observed among the HepG2 lines are reproducible and stable.

Table 4-2 Average level of inhibition (in percent) caused. to the specific binding of 3 nM 125I-tPA by HepG2 cell lines upon addition of 1 M RAP at 4C.

 Cell Line 

a16/HP a16/LP ATCC/HP ATCC/LP  Average 65 26 31 20 Std Dev + 8 + 10 + 16 + 10 (Number of determinations) (25) (23) (10) (16) 

Figure 4-3 Accumulation of cellular 125I-tPA and release of degradation products into the supernatant by HepG2 sublines at 37C. HepG2 sublines at day four of culture were incubated with 5 nM 125I-tPA in the absence ( ) and presence of 1.5 M RAP ( ) or 1.5 M tPA ( ). Each row of panels represents data for an individual HepG2 subline as indicated. The left column of panels shows 125I-tPA cellular uptake data and the right column shows the level of degradation products appearing in the supernatant. Numbers in ellipses are the average rates of 125I-tPA specific degradation expressed in molecules / cell per hour. Specific degradation is defined as the difference between total degradation and non specific degradation measured in the presence of 1.5 M tPA. Results from a single set of duplicate determinations are shown.

97 4.3.3 ¹2 5I -tPA degradation by HepG2 cells

The degradation of 125I-tPA subsequent to binding and endocytosis was measured using HepG2 cells incubated at 37C in the absence and presence of unlabelled excess tPA or RAP. At various time points cell lysates and media supernatants were collected and analysed for total radioactivity and labelled degradation fragments, respectively. Although 125I-tPA binds to more than one site at 4C, the results in Figure 4-3 (left column of panels) show that at 37C HepG2 binding is inhibited by RAP to nearly the same extent as unlabelled tPA, implying that LRP1 is the predominant binding receptor at 37ºC. Moreover, the appearance of ligand fragments in the supernatant (right column of panels in Figure 4-3) is attenuated more than 95% by the addition of RAP, indicating that LRP1 is the major receptor responsible for internalising tPA for subsequent degradation. The average rate of degradation of 125I-tPA over 6 hours for each cell line is indicated in ellipses on Figure 4-3. The comparative capacity of each HepG2 subline to degrade tPA is consistent with the relative levels of LRP1 expression determined in the other assays.

4.3.4 ¹2 5I-RAP ligand blots

Results in Figure 4-2 and Figure 4-3 indicate that RAP inhibits tPA binding to, and degradation by, HepG2 cell lines, suggesting that RAP competes with tPA for binding its receptor. To characterise the RAP receptor, cell lysates from HepG2 (at day 5 in culture) were probed with 125I-RAP and the resultant autoradiogram is shown in Figure 4-4. A prominent band at Mr ~500-kD in lysates from each of the HepG2 lines is consistent with the presence of LRP1. Bands at other positions have not been identified but collectively constitute only a minor proportion of the total 125I-RAP binding activity. The relative intensities of the bands at ~500-kD correlate with the degree to which RAP inhibits 125I-tPA binding (Figure 4-2) and degradation (Figure 4-3) by each of these cell lines. The results support the identification of LRP1 as the predominant member of the LDL receptor family present in HepG2. Moreover, its relative level on each sublines is consistent with it being the major receptor responsible for mediating tPA internalisation.

98 4.3.5 Pseudomonas exotoxin A (PEA) sensitivity

To examine the endocytic capabilities of LRP1 on HepG2 at 37C, cytotoxicity assays using the LRP1 ligand, Pseudomonas exotoxin A (PEA) were performed. PEA is inactive until LRP1 mediates its cellular internalisation (Fitzgerald et al., 1994). Within the cell, PEA is activated through cleavage by furin, and the activated toxin is transferred to the cytosol (Fitzgerald et al., 1995; Inocencio et al., 1994). Using protein content as a relative measure of cells remaining (Owensby et al., 1988), the results of exposure of the four HepG2 sublines to 10 ng/ml PEA for 3 days is shown in Figure

a16/HP a16/LP ATCC/HP ATCC/LP

500-kDa

205 -kDa

118 -kDa

85 -kDa

Figure 4-4 125I-RAP ligand blot of HepG2 subline lysates. Lysates (50 g protein per lane) of cells cultured for 5 days were separated by 5% PAGE-SDS, transferred to PVDF membranes, incubated with 10 nM 125I-RAP, washed, and exposed to autoradiography film. The migration distances for the major bands are similar to human plasma fibronectin (500-kDa) which was included in a separate lane together with other molecular weight markers whose positions are shown on the left. A representative blot from duplicate experiments is shown.

99 4-5. The degree of cell death among the sublines was found to correspond to the pattern of 125I-RAP binding evident in Figure 4-4. Cell death for the highest LRP1-expressing subline, a16/HP, was almost complete. In contrast, when compared to matched controls cultured in the absence of PEA, the other sublines survived 43%, 53%, and 63%. Occasional a16/HP cells survived PEA treatment and these were successfully expanded in number by subsequent culture in the absence of PEA. These cells displayed the surface membrane phenotype characteristic of HepG2 (CD7pos, CD26pos, CD36pos, CD44neg), but lacked detectable LRP1 (data not shown). These results imply that only cells expressing low levels of LRP1 survive PEA treatment and that sensitivity to PEA parallels 125I-RAP-binding and tPA-processing capabilities by the HepG2 sublines.

Figure 4-5 Pseumonas exotoxin A (PEA) sensitivity. HepG2 cells were plated in duplicate in the absence or presence of 10 ng/ml PEA and cultured for 3 days. The cells were washed, lysed, and assayed for total cellular proteins as a measure of surviving cell numbers. Results are expressed as a percentage of the protein content in the PEA treated wells compared to control wells. The results from a single determination are shown.

100 4.3.6 Western blots for LRP1 in HepG2 lysates

To determine whether variations in the degree of RAP inhibitable 125I-tPA binding among HepG2 cell lines correlate with levels of total cellular LRP1, HepG2 cell lysates were examined using western blotting and anti-LRP1 monoclonal antibody, 8G1. The results are shown in Figure 4-6. Single bands at Mr ~500,000 clearly demonstrate the presence of LRP1 in all HepG2 sublines, including the cells from the ATCC (ATCC/LP) which displayed reduced capacities for tPA binding and degradation (see Figure 4-2 and Figure 4-3). Considerable variation in signal intensities indicate marked differences in LRP1 antigenic levels among the sublines, and these findings are in accordance with 125I-tPA-degradation (Figure 4-3), and 125I-RAP-blotting (Figure 4-4) patterns. Moreover, the quantity of LRP1 per mg of cell lysate is observed to increase with time in culture, and this observation corresponds to time-dependent

Figure 4-6 Western blots of HepG2 subline lysates. Lysates (50 g protein per lane) prepared and stored frozen at -80C after 1, 2, 3, 4, 5, 7, and 9 days in culture were separated by 5% PAGE-SDS, transferred to PVDF membranes, probed with monoclonal antibody against the 515-kDa LRP1 ligand binding -chain (8G1, 5 g/ml), and developed using horseradish peroxide conjugated SaMIg (HRP-SaMIg). Each column represents lysates from a single time point, and each row from a single HepG2 subline as indicated. A complete set of blots from one of duplicate experiments is shown.

101 increases in tPA binding capacity shown in Figure 4-2, and in surface membrane expression presented in the next section (Figure 4-7).

4.3.7 LRP1 on HepG2 sublines determined by flow cytometry

The expression of LRP1 on the plasma membranes of three HepG2 sublines as a function of time in culture was determined by flow cytometry using anti-LRP1 antibody 8G1. HepG2a16/HP were not included in this study as this subline had not arrived in the laboratory at the time. For initial seeding, stock cell cultures were treated with trypsin/EDTA. This treatment removes the LRP1 epitope recognised by 8G1from the surface membrane. Subsequent harvests for assays on days 1, 2, 3, 7, and 12 used EDTA according to the procedure in 2.4.1. This method preserves the 8G1-recognised epitope. The results in Figure 4-7 show marked differences in LRP1 surface expression between the parental line from the ATCC (HepG2/LP) and the other two sublines. The levels of surface LRP1 correlate with the relative tPA (Figure 4-2), RAP (Figure 4-4), and PEA (Figure 4-5) binding capacities displayed by these sublines, and reflect the relative expression of total monolayer LRP1 determined by Western blotting Figure 4-6. Furthermore, in accordance with temporal tPA binding capacity (Figure 4-2) and total LRP1 levels (Figure 4-6), surface expression levels increase with time in culture to at least day 3.

LRP1 antigen was not convincingly detected on the surface of HepG2 from the ATCC (ATCC/LP), and the other assays described above support a finding of relatively low levels of expression by this cell line. However, flow cytometry is generally less sensitive than Western blotting, and the latter technique clearly showed the presence of LRP1 in all HepG2 sublines, including ATCC/LP (see Figure 4-6).

Together, ligand binding, ligand degradation, PEA sensitivity, and flow cytometric analysis display a consistent pattern demonstrating a stable variable expression of LRP1 levels among HepG2 sublines in the rank order HepG2a16/HP > HepG2a16/LP ~ ATCC/HP > ATCC/LP.

102

Hep G2a16/LP Hep G2/LP Hep G2/HP

Day 1

Day 2

Day 3 Relative Cell Number Cell Relative

Day 7

Day 12

Log Green Fluorescence

Figure 4-7 Expression of LRP1 on HepG2 sublines as a function of time in culture. Cells harvested using EDTA on days 1, 2, 3, 7, and 12 were labelled with 8G1 (shaded), or MOPC21 negative control (unshaded), and analysed by flow cytometry. LRP1 expression was consistently low on HepG2/LP and moderate on other sublines after 2 days in culture. The highest LRP1-expressing cell line, HepG2a16/HP, had not arrived in our laboratory at the time of this experiment. Interestingly a bimodal distribution is apparent on day one for HepG2a16/LP suggesting that a subpopulation more rapidly re-expressed LRP1 following trypsin treatment.

103 4.3.8 Status of other endocytic receptors on HepG2 sublines.

Confirmation of a difference in LRP1 expression on the plasma membranes between two of the HepG2 sublines, a16/HP and the parental line (ATCC/LP), was investigated using 8G1 and a second anti-LRP1 -chain monoclonal antibody, 3402. Flow cytometric analysis in Figure 4-8 shows near identical results for the two LRP1 antibodies, and further demonstrates a marked difference in LRP1 expression between the these sublines. In contrast, MHC-1 and CD-26 antigen (positive controls) displayed similar levels for both sublines.

In addition to LRP1, the expression of other internalising receptors on these sublines was investigated and the results are shown in Figure 4-8. While the LDL receptor has similar levels on both lines, expression of the thrombospondin/scavenger receptor (CD36 antigen) parallels LRP1 in being higher on the a16 clone than the parental line.

a16/HP a16/HP a b

ATCC/LP ATCC/LP Relative Cell Number Cell Relative c d

Log Green Fluorescence

Figure 4-8 Expression of endocytic receptors on HepG2 sublines. The sublines expressing the highest and lowest levels of LRP1, a16/HP (a, b) and ATCC/LP (c, d), were analysed by flow cytometry following labelling with antibodies against LRP1 (8G1, 3402), LDL receptor, (C7) and a member of the scavenger receptor family (Immunotech CD36). Controls were MPOC21 (negative), anti-MHC-1 (positive), and anti-CD26 antigen (positive). All cells were developed with FITC- SaMIg. Antibodies used to obtain each histogram are identified in the figure.

104 4.3.9 Culture condition perturbations

To investigate the basis for variation of LRP1 expression among the HepG2 sublines investigated, several culture condition perturbations over one to five passages were applied as listed in Table 4-3. After the culture period, alterations in LRP1 expression, compared to controls, were assayed using RAP inhibition of 125I-tPA binding or immunolabelling with 8G1 and flow cytometry. Except for treatment with Pseudomonas exotoxin A treatment (above) none of the conditions examined induced stable changes to LRP1 levels of the order observed between the lowest and highest LRP1-expressing HepG2 lines (ATCC/LP cf. a16/HP). However, minor increases were observed using increased cellular seeding densities and heat inactivated foetal bovine serum. These results imply that stable alterations to LRP1 expression to the degree observed among the HepG2 sublines is not induced by short-term variations in culture conditions.

Table 4-3 Culture condition perturbations applied to HepG2 cells from the time of seeding. Cells were harvested after the time interval indicated and assessed for alteration in LRP1 expression compared to controls.

Perturbation Mimics or affects: Duration Assay Method

Seeding density † growth kinetics 2 and 4 days i 2, 4, 6 nM glutamine supplement levels 4 days i 5-20% FBS supplement levels 4 days i 1 and 10 nM insulin gluconeogenesis 4 days i 100 nM dexamethasone gluconeogenesis 2 weeks / 3 passages fc intracellular 4 days i 3 mM NH4Cl acidification 2 and 4 g/L NaHCO3 medium alkalination 4 days i pH 6.8 adjusted with HCl medium acidification 4 days i 100 nM PMA protein kinase C 4 days i activation 100 U/ml penicillin intracellular Ca++ 5 passages fc +100 g/ml streptomycin alteration‡

† 5x104, 10x104, 15x104 cells per well seeded in 24 well plates (2 cm2/well) ‡ from (Bird et al., 1994) i = RAP inhibition of 125I-tPA binding, fc = flow cytometry using 8G1 anti-515-kD LRP1

105 4.4 DISCUSSION

Results in this study are consistent with LRP1 being the major receptor responsible for the internalisation and degradation of tPA by the human hepatoma cell line HepG2. However, wide variation in functional LRP1 expression was observed among HepG2 cells representing high and low passage numbers cultures of the parental line (from the ATCC), and the clone, a16.

LRP1 expression on these four lines was compared initially in binding studies at 4ºC using 125I-tPA as a representative LRP1 ligand, and RAP as a universal inhibitor of LRP1 ligand binding. The results revealed large differences in functional LRP1 levels among the HepG2 sublines. The highest and lowest activities are represented, respectively, by a high passage culture of clone a16 (a16/HP), and freshly acquired, low passage, cells from the ATCC (ATCC/LP).

Further comparisons of LRP1 status were determined using flow cytometric analysis for surface expression of EDTA-harvested cells in suspension, as well as 125I-RAP ligand and Western blotting of whole monolayer lysate. Relative functional levels of LRP1 were measured in Pseudomonas sensitivity (PEA) susceptibility assays, and 125I-tPA degradation assays. The latter provided quantitative comparisons by showing that a16/HP, a16/LP, and ATCC/HP degrade tPA at rates 8.9x, 3.2x, and 3.4x times faster than the parent line (ATCC/LP). Collectively, these immunological, binding, and functional results consistently demonstrate a wide variability of LRP1 levels in the order: a16/HP > a16/LP ~ ATCC/HP > ATCC/LP.

The lethality response towards PEA was consistent with a dose dependence on the level of LRP1 (see Figure 4-5). Interestingly, the rare cells that survived PEA exposure in the highest LRP1-expressing subline, a16/HP, could be expanded in subsequent culture. These cells were devoid of detectable LRP1 by flow cytometry (data not shown), which is consistent with the findings of FitzGerald et al (Fitzgerald et al., 1994). The survival of rare cells exposed to PEA implies that some individual cells express a level of LRP which departs from the bulk of the other cells in the culture. In the case of PEA exposure, cells expressing low levels of LRP1 are selected. In contrast, the correlation between high passage number and high LRP1 expression for both the clone and the parental line raises the possibility that long term culture of HepG2

106 naturally selects the outgrowth of cells exhibiting elevated constitutive expression of LRP1.

The genetic and/or posttranscriptional mechanism responsible for the variation in the level of LRP1 expression among the HepG2 sublines was not pursed rigorously here. LRP1 levels have been reported to be unregulated upon short term exposure to dexamethasone (Kancha and Hussain, 1996) and growth factors (Hussaini et al., 1990; Weaver et al., 1996). However, the magnitude of changes in LRP1 expression induced by these agents could not account for the level of stable differences observed among the HepG2 sublines investigated here. In an attempt to understand the basis of the large variations in HepG2 cell LRP1 expression, continual exposure to various stimuli was applied as listed in Table 4-3. None of these conditions could induce changes in LRP1 expression to the magnitude observed between the lowest and highest LRP1-expressing HepG2 cell lines. Rather, the most likely explanation for the differences is a change in the level of constitutive expression.

The a16 clone was selected for high expression of the asialoglycoprotein receptor (Schwartz and Rup, 1983). Results in Figure 4-8 show that, compared with the parental line (ATCC/LP), the a16 clone (a16/HP) displays an increase in surface levels of another internalising receptor, the caveolin-independent, scavenger receptor class B type I (SR-BI; CD36 antigen) (Zeng et al., 2003). In contrast, the LDL receptor displays similar expression between these sublines (Figure 4-8). This is not surprising because the LDL receptor and LRP1 are subject to separate transcriptional regulators. Despite the similarities in their protein structures (Herz et al., 1988). LRP1 and the LDL receptor share little homology in their promoter regions, where LRP1 has features of a house keeping gene and lacks the sterol regulatory element (SRE) present in the LDL receptor promoter (Kütt et al., 1989). LRP1 does contain an SRE in its 5‟untranslated region (Gaeta et al., 1994), but this region induces upregulation in the presence of aggregated LDL, which is the opposite response of the LDL receptor (Llorente-Cortes et al., 2002). It is possible that selection for high expression of the asialoglycoprotein receptor may have coincidentally also isolated a cell type enriched in high expression for certain other endocytic receptors, including LRP1.

Binding of tPA to HepG2 cells at 4C was reported by Bu et al (Bu et al., 1993) to be mediated largely by LRP1 in that they observed 70-80% inhibition of 125I-tPA binding upon the addition of RAP. The result here of 65% for a16/HP is similar, but the

107 value for the other HepG2 sublines tested deviated considerably. Clearly, LRP1 does not account for all 125I-tPA binding to HepG2 and indeed is only a minor contributor to binding by the low LRP1-expressing line, ATCC/LP. The other molecule(s) involved in tPA interactions with HepG2 have not been elucidated. VLDL receptor mRNA is present in HepG2 (Webb et al., 1994) and this RAP-inhibitable receptor binds PAs (Heegaard et al., 1995). However, 125I-RAP blots in Figure 4-4 show that LRP1 is the major RAP-binding receptor in all the HepG2 sublines investigated. Another tPA receptor described, the serpin enzyme complex receptor (SEC-R), has similar ligand binding profiles to LRP1 (Kounnas et al., 1996; Stefansson et al., 1996) and in the absence of its cloning, may be assumed to be LRP1. Similarly, a binding activity dependent on the presence of -fucose at threonine 61 of tPA has not been fully characterised (Hajjar and Reynolds, 1994). Other studies concluded that PAI-1 localised in the extracellular matrix is a major tPA binding site for in cultures of endothelial cells (Ramakrishnan et al., 1990; Russell et al., 1990) and this possibility cannot be excluded for HepG2. Whatever constitutes the non-LRP1 tPA binding site at 4C, its contribution to endocytosis appears to be minimal. In assays at 37C, specific tPA cellular accumulation and degradation are almost completely eliminated by the addition of RAP (see Figure 4-3).

These results demonstrate that HepG2 sublines vary in their basal expression of LRP1 under standard culture conditions and that the magnitude of this variation correlates with passage number in the limited number of sublines examined. These differences in LRP1 levels may provide a plausible explanation for apparent discordant results in the published literature regarding the capacity of HepG2 to interact with known LRP1 ligands such as tPA.

108

Chapter 5

5 SOLUBLE LRP1 IN HUMAN CIRCULATION

5.1 INTRODUCTION

The widely used hepatic cell line, HepG2 appears to increase its expression of LRP1 over long periods in continuous culture, as presented in Chapter 4. This artefact is likely the result of the selective outgrowth of cells with higher constitutive expression of LRP1. However, alternative mechanisms are possible. The structure of LRP1 is unusual for a surface membrane receptor in that the two chains of the heterodimer remain associated by non-covalent interactions, rather than via disulfide bridges. It follows that dissociation of the  and  chains offers a hypothetical means by which surface levels may be reduced. If such a mechanism was operating, then a consequence might be the shedding of a soluble form of the receptor in HepG2 culture supernatant. This Chapter presents the initial experiments in the search for a released form of LRP1 in culture supernatant, with a serendipitous finding. A soluble form of LRP1 indeed was detected in HepG2 supernatant, but its was found overwhelmingly to be a bovine protein originating from serum supplementing the culture medium, rather than human LRP1 released from by the cells. This outcome soon lead to the detection of a previously unrecognised component of human blood, soluble LRP1.

Dissociation of LRP1 had been previously proposed, but earlier studies were unable to detect the presence of a soluble form of the receptor (Herz et al., 1990a; Moestrup, 1994). The initial approach for the results presented here used 125I-RAP as a probe for LRP1 in ligand blots. This technique offers high sensitive and selectivity and these factors may have contributed to successful detection in our laboratory.

The protein identified as soluble LRP1 in human serum demonstrates characteristics of the LRP1 -chain in that it displays: (a) a high molecular weight of ~500-kDa; (b) recognition by three antibodies against the LRP1 -chain, but not by an antibody against the cytoplasmic tail of the -chain, and (c) the ability to bind the LRP1 ligands, activated 2-macroglobulin, and RAP.

Soluble LRP1 was not initially detected in the supernatant of HepG2 or fibroblasts. However, my colleague, Dr Kathryn A Quinn, found that primary rat hepatocytes synthesise and release soluble LRP1 into culture supernatants, suggesting that the liver may be a source in plasma. Dr Quinn also established the surprisingly narrow normal range in serum of 6.2 + 1.2 g/ml, and provided data suggesting that

110 patients with liver abnormalities have slightly elevated levels. Potential value as a diagnostic marker in liver disease is evident. Later, her work on the trophoblast cell line, BeWo, showed marked shedding of the receptor from this cell line, and fittingly, raised levels were found in cord blood serum (Quinn et al., 1999). These findings suggest that soluble LPR may play a role in gestation. The partial amino acid sequence of the protein purified from serum was included in the same report. That data confirmed the initial immunological findings presented here and in Chapter 6, that, in addition to the large -chain, soluble LRP1 retains a fragment from the ectodomain of the -chain.

Several membrane receptor systems are known to have soluble counterparts and many of these have distinct biological activities (reviewed in Ehlers and Riordan, 1991; Levine, 2008; Muller-Newen et al., 1996; Rose-John and Heinrich, 1994). LRP1 is a large endocytic receptor capable of binding, and mediating the degradation of several ligands. The retention of ligand binding function by the soluble form, may have consequences in pathological conditions known to be associated with disturbance in levels of LRP1 ligands, such as atherosclerosis and tumour metastasis. The normal range for soluble LRP1 levels in serum is surprisingly narrow suggesting that its concentration may be regulated. If so, deviations from its homeostatic levels might occur in various conditions in which cellular LRP1 is involved.

The initial detection of soluble LRP1 involved the recognition a bovine protein by RAP derived from rat. The same RAP was used to extend these findings to a third species, human. This cross-recognition among species was investigated further by examining the sera from other animals, and those studies are presented in the next Chapter. Mention here will be made of observations from those studies which clarify the findings relating to the human protein discussed in this Chapter.

The work here presents the rationale and initial investigations by the author (me) which identified soluble LRP1 in the human circulation. This work was disclosed in two conference presentations (Grimsley et al., 1996a; Grimsley et al., 1996b). Some of the data following are preliminary in nature and have not been published except for selected mentions in our later review (Grimsley et al., 1998). The formal publication of initial identification was a collaboration (Quinn et al., 1997).

111 Hypothesis:

To determine whether a soluble form of LRP1 circulates in human plasma.

5.2 MATERIALS AND METHODS

5.2.1 RAP, antibodies, and cell lines

Recombinant RAP was prepared as described in 2.5.5. Details of antibodies against LRP1 -chain (8G1) and -chain (IgG-1B3 and IgG-11H4) are outlined in Table 2-1. Sources and characteristics of HepG2 cell lines are presented in 4.2.1.

5.2.2 ¹2 5I-RAP blots and Western blots of cell lysates and body fluids

Serum (1 l), plasma (1 l), CSF (10 l), lacrimal fluid (10 l), culture supernatant (10 l), and cell lysate (50 g, preparation described in 2.2.1) were electrophoresed on 5% polyacrylamide (see 2.6.1), transferred to PVDF membrane (2.6.3), blocked with 5% bovine skimmed milk in PBS for 1 hour (2.6.4), and either Western blotted (2.6.5), or ligand blotted using 10 nM 125I-RAP (2.6.6). Bovine smooth muscle cells (SMC) used as a control in Figure 5-9b were cultured as described in 2.1.6. To facilitate distinction of Western blots from ligand blots, and to aid their interpretation, a small icon representing cellular LRP1 and the specificity of the antibody, is displayed on Western blot Figures. An example of the Antiserum  777 8G1 icon is presented here. IgG-1B3 IgG-5D7  IgG-11H4

5.2.3 Enrichment of soluble LRP1 on 8G1-sepharose

Anti LRP1 -chain monoclonal antibody, 8G1, was immobilised on sepharose 2B (8G1-2B) as described in 2.5.7, equilibrated with PBS, and loaded (200 l bed volume) into a small column (Miniprep, BioRad). Serum (100 l) was applied and permitted to flow into the matrix without restriction. Under these conditions, the matrix was able to remain unbroken when run dry as a result of the small volumes involved. Binding was

112 promoted by incubation at room temperature for 10 minutes before the matrix was washed with PBS (4 times with 1 ml and twice with 5 ml). Bound antigen was eluted by adding two aliquots (100 l each) of elution buffer (50 mM citric acid / NaOH / 10 mM NaCl pH 4) with a 2 minute incubation period between each aliquot. Eluate was collected following addition of the second aliquot of elution buffer and the solution was neutralised with 1 M Tris (10 l). A sample (10 l) was mixed with 40 l of Laemmli sample buffer and processed for blotting with 10 nM 125I-RAP (see 2.6.6).

The pH of the elution buffer was selected from preliminary studies which demonstrated destruction of ~500-kDa 125I-RAP-binding molecules in serum subjected to buffer at pH 3 (data not shown).

5.2.4 Ammonium sulfate precipitation of soluble LRP1 from serum

Ammonium sulfate (10 g) was added to water (10 ml) and stirred at room temperature overnight. The suspension was allowed to settle and a sample (4 ml) of saturated ammonium sulfate solution (SASS) in the supernatant was centrifuged (10,000g, 5 minutes, room temperature) to remove any residual solids. Serum (700 l, 600 l, and 500 l) was added to SASS (300 l, 400 l, and 500 l respectively) to form mixtures containing AS at 30, 40, and 50% saturation. The mixtures were incubated at room temperature with mixing for 1 hour, then centrifuged as before. Samples of the supernatants (100 l) from the three conditions were diluted 1:7, 1:6, and 1:5 respectively with PBS. Using this system, materials in each of the diluted supernatants represented extracts of the original serum uniformly diluted 1:10. The rest of the supernatant was discarded and the pellets were washed once in SASS diluted with water to match the treatment concentration (i.e. 30%, 40%, 50% SASS respectively). Pellets were resuspended in PBS to the original volume of serum used (700, 600, 500 l). Samples of the diluted supernatant (10 l) and resuspended pellets (1 l) were diluted to 50 l with Laemmli sample buffer and examined for LRP1 -chain by Western blotting (see 2.6.5) using monoclonal antibody 8G1.

113 5.2.5 Investigation of LRP1 on resting and activated platelets

Platelet rich plasma (PRP) from a volunteer was prepared by centrifugation (150 g for 15 minutes at room temperature) of blood (10 ml) collected into ACD anticoagulant (1.4 ml). The resulting PRP was centrifuged at 940 g for 15 minutes and the platelets were re-suspended in calcium and albumin free Tyrode buffer (pH 6.2) containing 2 U/ml apyrase (Sigma) to deplete ADP. Following incubation at 37C for 15 minutes, the platelets were washed twice (940 g for 15 minutes) in Tyrode buffer (pH 6.2) and resuspended at 200 x 109/L in Tyrode buffer at pH 7.4 containing 5 mM EDTA. Platelets were activated with 30 nM thrombin for 2 minutes at 37C and the reaction was terminated by addition of 5 M PPACK (to inhibit thrombin) and 10 M leupeptin (as a general proteinase inhibitor). Resting and activated platelets were labelled with anti-LRP1 antibodies (8G1 and 777), negative controls (MOPC21 and normal rabbit immunoglobulins), or positive controls (AK6 and W6/32). All cells were treated with FITC conjugates of antisera to mouse or rabbit immunoglobulins to develop the appropriate primary antibodies, and assayed by flow cytometry. Activated platelet lysate (~10 g), and their supernatant (10 l), were examined for the presence of LRP1 by Western blotting using 8G1.

5.2.6 Release of soluble LRP1 from HepG2a16 (a16/HP)

Cells were seeded and cultured overnight in 12 well plates at 105 / cm2 (four times usual plating density) in HepatoZYME-SFM serum free medium (Life Technologies) supplemented with 2 mM glutamine and foetal bovine serum at concentrations of 8, 6, 4, 2, 1, 0.5, and nil percent. Supernatants were centrifuged at 10,000g for 30 minutes and 10l was examined by Western blotting using anti-LRP1 monoclonal antibody 8G1 which recognises human, but not bovine, LRP1. Purified placental LRP1 (40 ng, kindly provided by Dr K. Quinn, Centre for Thrombosis and Vascular Research, University of New South Wales) was used as the cellular LRP1 control for this experiment. Results appear in Figure 5-7.

114 5.2.7 LRP1 association with extracellular matrix

Well-washed HepG2a16 cells harvested with trypsin/EDTA were plated in 24 well plates at 25 x 103 / cm2 in EMEM containing 10% rabbit serum. Medium was replaced every 2 days and 24 hour before assay. Following various times in culture, cells were washed 4 times with cold PBS and lysed with Laemmli sample buffer (200 l), either directly as a whole monolayer, or as cells in suspension following harvesting (37C, 10 minutes) and further washing (twice) with 1 mM EDTA / 0.1% BSA in PBS (release buffer). The matrix remaining after removal of the cells with release buffer was washed four times with release buffer, then extracted with Laemmli buffer (100 l). All preparations were stored frozen at -80C until analysed. Samples were investigated by 125I-RAP blotting to detect total LRP1, and by Western blotting using monoclonal antibody IgG-5D7, which recognises rabbit, but not human, LRP1.

5.2.8 Endothelial cell culture

Human umbilical vein endothelial cells (HUVEC) frozen at passage 5, and bovine brain extract were kindly provided by Dr A Ashton (Centre for Thrombosis and Vascular Research, University of New South Wales). HUVEC were originally isolated within 6 hours of delivery from umbilical cords collected into PBS containing antibiotics. The cord vein was flushed with Hartman‟s buffer (1.7% sodium lactate, 0.1 g/l glucose, 1 mM KH2PO4, and 1.25 mM Na2HPO4 pH 7.6), incubated for 10 minutes at 37C in the same buffer containing 50 U/ml collagenase, and the cells were harvested in medium M199 containing 20% human serum. Bovine brain extract containing ~80% endothelial cell growth factor (ECGF) was originally isolated according to Maciag et al (Maciag et al., 1979). T25 culture flasks were coated for 1 hour at 37C with 0.1 g/L Hemaccel (polygeline, a form of collagen) in PBS, prior to seeding with HUVEC at 4 x 104/cm2 using medium M199 supplemented with 150 g/ml ECGF, 20% human serum, 10 U/ml heparin, and antibiotics according to Minter et al (Minter et al., 1992). Medium was replaced daily and cell lysates were prepared on day 4 (see 2.1.1).

115 5.2.9 CSF, lacrimal fluid, and urine samples

Six cerebral spinal fluid (CSF) samples were randomly selected from specimens submitted to the Pathology laboratories at the Prince of Wales Hospital, NSW. The patients from whom these originated had a range of disorders, predominantly related to infections but not knowingly to dementia. These samples (10 l) were blotted with 125I-RAP (2.6.6) to produce the autoradiogram in Figure 5-10a. A further four CSF samples were generously provided by Dr W. Brooks, Centre for Education and Research on Ageing, Concord Hospital, NSW. These samples were collected for other purposes from patients with dementia, which, although not confirmed, was most likely related to Alzheimers disease. Western blotting (2.6.5) of these samples (10 l) using 8G1 produced the results in Figure 5-10b.

Lacrimal fluid was collected from an individual with influenza, and a sample (10 l) was blotted with 125I-RAP to produce the autoradiogram shown in Figure 5-9a. The presence in urine of soluble gp330/megalin has previously been reported. A sample of urine (10 l) drawn from a specimen from a normal individual was blotted with 125I-RAP and found to produce a band at 600-kDa corresponding to soluble gp330/megalin. This finding has been previously reported and repeated here as a control only.

All CSF, tear, and urine samples were centrifuged at 10,000g for 10 minutes prior to analysis to remove any cellular debris or precipitates.

5.3 RESULTS

5.3.1 Initial detection of RAP-binding proteins in serum

The possible presence of a soluble form of LRP1 in the supernatants from HepG2 was investigated by ligand blotting using 125I-RAP. HepG2 sublines described in Chapter 4 were cultured in medium containing 10% foetal bovine serum (FBS) and equal amounts of cell lysates (50 g) prepared from these cells (see 2.2.1) were blotted with 125I-RAP (see 2.6.6). The resulting autoradiogram shown in Figure 5-1 (left panel)

116 display bands at ~500-kDa with intensities paralleling the level of LRP1 expressed by each subline. This experiment represents a verification of results in Figure 4-4. Supernatant from these cultures also produced signals at ~500-kDa (Figure 5-1, right panel), but in contrast to the cell lysates (left panel), their intensities were uniform. Subsequently, similar signals were revealed when analysing fresh FBS-containing medium, but not serum-free medium (data not shown). These initial findings implied that serum was the source of RAP-binding activity.

Figure 5-1 Serendipitous initial detection of soluble LRP1 using 125I-RAP. In the panel on the left, cell lysates (50 g) prepared from the HepG2 sublines described in Section 4.2.1, display signals at ~500-kDa with intensities corresponding to cellular LRP1 expression. Culture supernatants (right panel) from these cells also show bands at ~500-kDa. In contrast to the cell lysates, 125I-RAP binding to the supernatants appears uniform and was later shown to be detecting bovine soluble LRP1 in fresh serum-supplemented culture medium. The initial single determination is shown.

117 5.3.2 High molecular weight RAP-binding proteins in human serum

Investigation of the RAP-binding activity in bovine serum was extended to determine its possible presence in humans. In pre-empting a positive finding, experimental conditions were designed to establish optimal collection and storage requirements for this previously uncharacterised blood protein. Serum and heparinised plasma samples were collected from two volunteers on two consecutive days. Specimens from the first venipuncture were stored at 4C overnight and analysed with fresh samples collected the following day. Samples from each collection (2 l) were analysed by 125I-RAP blotting (see 2.6.6) and the resulting autoradiogram is shown in Figure 5-2 a. Similar RAP-binding activity to bovine serum, was displayed by human sera with bands appearing in a high molecular weight region. The molecular mass of these signals corresponded to ~500-kDa by comparison to the similar migration of fibronectin, which is known to be 500-kDa. Little variation in the intensity of these bands is observed between the individuals, or among samples of differing age and type (serum or plasma). This information was helpful in determining that the RAP-binding activity is robust and its detection imposes no unusual sample handling requirements. Activity was also shown for plasma prepared from blood samples collected into EDTA and ACD anticoagulants (data not shown).

To expand the number of individuals investigated and commence a screen of pathological conditions, seven randomly selected sera from patients with various disorders were blotted using 125I-RAP and the results are shown in Figure 5-2b. In contrast to the results in Figure 5-2a (normal individuals), levels of RAP-binding activity in this small cohort of patients is observed to vary considerably. These preliminary results bear parallels to more formal assays performed by colleagues who found a restricted range of soluble LRP1 levels (3.7-10.8 g/ml) among 50 normal individuals, and a broader range (2.0-22.4 g/ml) among 45 patients with abnormal liver function (Quinn et al., 1997).

118

Figure 5-2 Initial detection of soluble LRP1 in humans by using 125I-RAP ligand blotting. 125I-RAP signals appear characteristically at ~500-kDa in: (a) two volunteers whose serum and plasma were examined immediately or after one day in storage at 4C; and (b) a random selection of sera from patients with various conditions. Results of a single determination are shown.

119 5.3.3 Immunological detection of soluble LRP1

At this point, the protein of interest in serum had been demonstrated to display RAP-binding ability and possess a molecular mass of approximately 500-kDa. These criteria are indicative, but not definitive of LRP1. To characterise the protein further, serum was Western blotted using the anti-LRP1 monoclonal antibody, 8G1, and the results are displayed in Figure 5-3a. A band is observed for fresh serum (lane 2) at a similar migration distance to the single band developed from HepG2 lysate (lane 1), suggesting the presence of LRP1 antigen in both lanes at this position. Results in other lanes of this blot are discussed in the next section (5.3.4). The heavy background in regions less than 500-kDa in Figure 5-3a is due to non-specific binding which occurs when using whole serum. Reduction of this background in addressed in section 5.3.6.

5.3.4 Stability of soluble LRP1 in serum

The stability of soluble LRP1 in serum stored under sterile conditions at 4C was investigated for a single individual using samples collected at various times up to one year. The Western blot of these samples (2 l) shown in Figure 5-3. Figure 5-3a reveals similar signal intensities at ~500-kDa for all time points, indicating that the majority of soluble LRP1 in serum appears stable for several months without additives. Furthermore, the apparent consistency of activity over this time period suggests that physiological levels (at least for this individual) have long-term uniformity.

Attempts to exploit this stability by using serum diluted in Laemmli sample buffer as a simple standard were thwarted by loss of signals at ~500-kDa within a week of storage at 4C (Figure 5-3a, lane 6). Similar loss of activity was observed for HepG2 lysates stored at 4C (data not shown). However, this degradation is prevented by freezing lysates and at -80C.

120

Figure 5-3 Immunological identification, investigation of stability, and enrichment of soluble LRP1 in human serum. Panel a shows a Western blot (using anti-LRP1 antibody, 8G1) of an individual‟s whole serum which was stored at 4C for various times up to one year (lanes 2-5). Bands at ~500-kDa indicate the presence of soluble LRP1. Reduction of the heavy non-specific binding to proteins less than 500-kDa is addressed in Chapter 6 by pre-enriching samples on immobilised RAP. The experiment also included serum mixed with Laemmli buffer and shows that long term stability of soluble LRP1 is not retained in the presence of SDS (lane 7). Panel b is an 125I-RAP blot of soluble LRP1 enriched by applying serum to 8G1-sepharose, washing, and eluting with pH 4 buffer. The signal at ~500- kDa in lane 3 provides strong evidence for the presence of a soluble form of LRP1 in serum based on its initial antigenic recognition by 8G1 (an LRP1 antibody), and followed by its functional recognition by RAP (an LRP1 ligand). Results of a single determination are shown.

121 5.3.5 Identification of soluble LRP1 by serial 8G1-sepharose enrichment and 125I-RAP blotting

To demonstrate that molecules recognised immunologically by 8G1 in Western blots are the same as those binding RAP in 125I-ligand blots, serum was enriched on 8G1 immobilised to sepharose 2B (8G1-2B), and the eluate was investigated in the 125I-RAP blot shown in Figure 5-3b. Under the conditions used, not all RAP-binding activity was captured on the 8G1-2B minicolumn as evident from the appearance of RAP-binding molecules in the fall through (data not shown). However, some material was successfully captured and produced signals at ~500-kDa (Figure 5-3b, lane 3). This finding suggests that a single class of molecule in human serum displays three characteristics of LRP1 -chain, namely: (1) molecular mass of ~500-kDa; (2) immunological recognition by an anti-LRP1 antibody; and (3) functional binding of an LRP1 ligand (RAP).

5.3.6 Enrichment of soluble LRP1

As mentioned in section 5.3.3, the Western blot in Figure 5-3a contains heavy non-specific binding bands associated with serum proteins at molecular masses less than 500-kDa. To reduce this problem, techniques for pre-enriching soluble LRP1 from serum were investigated. Results in 5.3.5 which used immobilised 8G1 (anti- LRP1 -chain antibody), represent the first of these approaches. Enrichment of soluble LRP1 by this method was successful, but limited by the availability of the antibody. A further concern with this technique is the exposure of the enriched proteins to acidic conditions (pH 4). However, destruction of soluble LRP1 had been found to require stronger acidic conditions (pH 3, data not shown).

A simpler protocol was sought using ammonium sulfate (AS) cutting and results in Figure 5-4 show that soluble LRP1 is precipitated in the saturation range 30-40%. Although results in both Figure 5-3 and Figure 5-4 are 8G1 Western blots, background problems are less obvious in the latter because the lanes containing serum extracts received half the loading (equivalent to 1 l), and one third the time exposed to X-Ray film. Nevertheless, background is observed to be improved by pre-cutting serum with AS at 30% saturation. This technique is useful in achieving the original aim of

122

Figure 5-4 Enrichment of soluble LRP1 from serum by ammonium sulfate (AS) cutting. Samples (1 l) of the supernatant and the re-dissolved pellet from serum brought to various concentrations of AS were examined by Western blotting using 8G1. Bands at ~500-kDa corresponding to soluble LRP1 are observed to disappear from the soluble fraction and appear intact in the precipitate at 40% AS or greater. Results of a single determination are shown.

background reduction in Western blotting, but has limited use as a step in a purification protocol because only a small proportion of serum proteins are precipitated at 30% saturation, and inadequate removal of unwanted proteins occurs in the cut range 30- 40%. Furthermore, resulting fractions contain AS which requires removal by dialysis prior to additional purification steps.

A technique offering enrichment properties similar to immobilised 8G1 was clearly preferable. Such a technique is the use of immobilised ligands, RAP or activated

2macroglobulin (see Figure 5-5). In later experiments, immobilised activated

2macroglobulin was successfully used by colleagues (Quinn et al., 1997) and offers the advantage of higher specificity for LRP1 over RAP which is known to bind to other members of the LDL receptor family. However, RAP was chosen for initial work for its availability in sufficient quantity and its stability. As a capture ligand, RAP potentially offers other advantages, when compared to activated 2-macroglobulin. The smaller molecular size of RAP is less constricting on pore sizes of matrix and offers higher numbers of potential binding sites per milligram. Furthermore, RAP has multiple

123 binding sites on LRP1 and binds to the receptor in the presence of other LRP1 ligands. This property may prove beneficial for capture of soluble LRP1 from serum which contains multiple LRP1 ligands.

Background problems encountered in Western blotting (illustrated in Figure 5-3a) were found to be conveniently and effectively reduced by pre-enriching serum for soluble LRP1 using RAP immobilised to sepharose 2B (see 2.5.7 for coupling method). Although the need for pre-treatment of serum is obviated by less background in ligand blots, preliminary enrichment on immobilised RAP was applied to all serum samples subsequently examined by both Western and ligand blotting. An example of serum enriched on RAP-2B is not shown in this Chapter because the technique was applied to all serum samples for the results in Chapter 6. A comparison of Figure 5-3a and Figure 6-4 demonstrates the improvement.

Figure 5-5 Capture of soluble LRP1 on immobilised RAP or activated 2-macroglobulin.

124 5.3.7 Further identification of soluble LRP1

In support of the identification of soluble LRP1 in serum, seven (including 8G1) anti-LRP1 antibodies were applied in Western blots of serum enriched on immobilised RAP. Data for five of these appears in the next Chapter and show that three anti-LRP1 -chain antibodies (8G1, IgG-5D7, and rabbit antibody 777) recognise ~500-kDa RAP- binding proteins in sera from a range of mammals. Two additional -chain antibody, 3402 (used for results in Figure 4-8), and an incompletely characterised preparation of rabbit immunoglobulins raised “in-house” against purified human placental LRP1, also recognised soluble LRP1 by Western blotting (data not shown). The other two anti- LRP1 antibodies (IgG-1B3 and IgG-11H4) are -chain specific. As reported and discussed in the next Chapter, their reactivity pattern suggests that part of the -chain is retained in soluble LRP1.

Data additional to the results in this Chapter are available to support the identification of a soluble form of LRP1 in serum. These include, recognition of a 500-kD protein in serum by Western blotting using a further LRP1 antibody (3501) against the -chain ectodomain (Quinn et al., 1999), and in ligand blots by tPAPAI-1 complexes and the archetype LRP1 ligand, activated 2macroglobulin (2M*) (see

Figure 5-5 and Quinn et al., 1997). Recognition of was 2M* also demonstrated by enrichment of soluble LRP1 on immobilised 2M* (Quinn et al., 1997).

Together, these findings provide strong immunological and functional evidence for the identification in human serum of a soluble form of LRP1.

5.3.8 Platelets as a potential source of soluble LRP1

The possibility that cellular components of blood may be a source of soluble LRP1 in serum was investigated by examining resting and activated platelets, and products released upon activation. Platelets were chosen for study based on 1 of 3 experiments from the 5th International Workshop on Leukocyte Differentiation Antigens which suggested that platelets express LRP1 (Shaw, 1994). In the same study, monocytes were also identified as an LRP1-expressing cell type in peripheral blood, but their total mass is usually much less than platelets in normal individuals. Flow

125 cytometric analysis of platelets using two antibodies against LRP1 -chain, 8G1 (a monoclonal antibody) and 777 (a ligand-affinity purified rabbit antiserum), is shown in Figure 5-6a. While neither antibody detected any change in surface expression of LRP1 on platelets following activation, a marked difference in reactivity between the antibodies was observed in that LRP1 was detected and not detected using 777 and 8G1, respectively. This finding suggests that a remnant fragment of the -chain (lacking the 8G1 epitope) is present on platelets. This pattern is reminiscent of trypsin-harvested HepG2, which lose their 8G1 epitope but retain activity with 777 (see Figure 4-8).

A lysate prepared from activated platelets revealed a faint band at ~500-kDa in a Western blot using 8G1 (Figure 5-6b). Although platelets were negative by flow cytometry using the same antibody, the findings are consistent with differences in sensitivity of the two techniques, or with a possible intracellular location of the antigen. The presence of strong signals at positions less than 500-kDa (Figure 5-6b) implies the existence of LRP1 -chain fragments, which in these cases, retain the 8G1 epitope.

Examination of the supernatant from activated platelets was unable to detect the presence of released LRP1. The sample analysed (10 l) was derived from platelets in suspension at a similar concentration to blood and thereby represented released platelet products in approximately 10 l of serum. This is 10 times the volume required to detect strong signals for soluble LRP1 in serum. These findings allow the conclusion that platelets are unlikely to be a major source of soluble LRP1 in clotted blood, and that its presence is unlikely to be an artefact of collection. The detection of similar quantities of RAP-binding molecules at ~500-kDa in serum and in plasma prepared using various anticoagulants provides further support for this conclusion.

126

Figure 5-6 Platelets as a potential source of soluble LRP1. Panel a shows flow cytometric histograms obtained on resting and thrombin-activated washed platelets labelled with LRP1 antibodies 777 and 8G1. The activation status of these cells was controlled by labelling with monoclonal antibodies AK6 and W6/32 which recognise the activation sensitive and insensitive markers, P-selectin (CD62P) and MHC-1, respectively. MOPC21 and purified normal rabbit Ig were included as the negative control in the appropriate assays (open overlays). Both platelet preparations are observed to react with polyclonal anti-LRP1 antibody, 777, whereas labelling by monoclonal antibody, 8G1 is undetectable. Panel b shows a Western blot (using 8G1) of activated platelet lysate (10 g) and products released during their activation (supernatant). A faint band at ~500-kDa is present in the activated-platelet lysate together with several apparent LRP1 degradation products. Soluble LRP1 in the supernatant was undetectable.

127 5.3.9 Release of soluble LRP1 from HepG2a16

The original hypothesis for the possible release of soluble LRP1 proposed simple dissociation of - and - chains. For this mechanism to operate, soluble LRP1 of cell line-origin might be expected in culture supernatants. To investigate this possibility the supernatant of HepG2 cells was re-investigated specifically for the presence of soluble human LRP1. Two precautions were taken. Firstly, to assess possible interference from bovine LRP1, culture of the HepG2 was performed in a medium supplemented with varying concentrations of foetal bovine serum (FBS). Secondly, the Western blotting detection system utilised 8G1, which recognises human, but not bovine, LRP1 (see Figure 6-4). The design of this experiment contrasts, the original 125I-RAP blot of HepG2 supernatant (see Figure 5-1), which detected LRP1 from both human and bovine sources. Results in Figure 5-7 show that detectable quantities of soluble human LRP1 indeed are released into the supernatant from the cell line HepG2a16/HP which displays high expression of cellular LRP1 (see results in Chapter 4). In the complete absence of serum, HepG2 failed to attach to the culture flask, as observed previously for vitronectin-depleted serum (Figure 2-1). In the presence of 1% serum or more, the cells released soluble LRP1 at a rate largely independent of the serum supplementation concentration. These findings show that HepG2a16 are capable of soluble LRP1 release and imply that cell attachment is a prerequisite for its release by this cell line.

128

Figure 5-7 Soluble LRP1 is released by HepG2a16 (a16/HP). Cells were seeded at 105 / cm2 (four times usual plating density) and cultured overnight in Hepatozyme supplemented with foetal bovine serum at various concentrations as indicated. Supernatant was centrifuged (10,000g for 30 minutes) and examined by Western blotting using 8G1 which does not recognise bovine LRP1. Faint bands are observed at ~500-kDa in all serum-containing supernatants, and their intensities are similar in the range 2-8%. This consistency contrasts with the intensities of non-specific bands near the bottom of the gel which correlate with the serum concentration. Results of a single determination are shown.

129 5.3.10 Association of serum-borne LRP1 with extracellular matrix

LRP1 binds numerous ligands and some, such as thrombospondin, can associate with extracellular matrix (ECM), often through affinity for glycosaminoglycans (GAGs). The presence of these ligands in ECM raises the possibility that soluble LRP1 in serum may also become associated with ECM by binding its immobilised ligands. HepG2 cultured in the presence of rabbit serum, were divided into cellular and ECM components with the use of EDTA. This technique also promotes dissociation of LRP1 from its ligands because calcium is required for their binding. Whole monolayers, EDTA-isolated cells, and cell-fee ECM were treated with Laemmli sample buffer and analysed for the presence of LRP1 by ligand blotting using 125I-RAP, which detects both human and rabbit LRP1, and by Western blotting using monoclonal antibody, IgG- 5D7 which recognises rabbit, but not human, LRP1. The results are shown in Figure 5-8. Bands appearing at ~500-kDa in all lanes of the 125I-RAP blot (panel a) indicate the presence of LRP1 (human or rabbit) in all culture fractions examined, including the ECM. In panel b, LRP1 derived from rabbit serum (rabbit soluble LRP1) is observed by IgG-5D7 Western blotting, to be associated with the intact monolayers after day 4 in culture. This finding is consistent with the ECM, and LRP1 ligands within it, requiring time to reach maturity. Although rabbit soluble LRP1 is observed to be associate with the intact monolayer, its presence in the ECM and cellular fractions was not detected. This finding is consistent with dissociation and loss by washing of rabbit soluble LRP1 as a consequence of calcium chelation by EDTA. Together the results suggest that soluble LRP1 from serum can associate with cells or ECM in culture. Furthermore, the association of non-cellular LRP1 with ECM is tight for HepG2-derived LRP1 (panel a), and less avid for soluble rabbit LRP1 derived from serum (panel b). These observations are consistent with the possibilities that cell-derived LRP1 is incorporated into the matrix during its synthesis, whereas serum-derived LRP1 adheres onto components within the ECM.

130

Figure 5-8 Cellular and soluble LRP1 association with extracellular matrix (ECM). HepG2a16 cultured in EMEM / 10% rabbit serum were washed and lysed in Laemmli buffer as either whole monolayers or as suspended cells separated from extracellular matrix using release buffer (1 mM EDTA in PBS / 0.1% BSA). The remaining matrix was also treated with Laemmli buffer following further washing with release buffer. Panel a shows an 125I-RAP blot detecting total (human and rabbit) LRP1 in samples prepared on day 4 of culture. The band at ~500-kDa in lane 3 indicates that LRP1 is incorporated into the matrix and is resistant to dissociation by EDTA. Panel b shows a Western blot of whole monolayers at various times in culture, along with EDTA-harvested cells and matrix at day 8. The blot used antibody IgG-5D7 to detect rabbit soluble LRP1 specifically in the presence of human cellular LRP1. Bands at ~500-kDa in lanes 6 and 7 indicate that exogenous soluble rabbit LRP1 in the medium associates with the human monolayers at times after 2 days of culture. In contrast to cell-derived LRP1, bands corresponding to serum-derived LRP1 are not detected in EDTA-treated cells (lane 8) or ECM (lane 9), suggesting that soluble LRP1 has a looser association with matrix. The initial single determination is shown.

131 5.3.11 Detection of a RAP-binding protein in lacrimal fluid

The existence of LRP1 in serum raises the possibility of its presence in other body fluids. To address this question, a sample of a tear was investigated using 125I-RAP blotting with an unusual result as shown in Figure 5-9a. A weak signal is observed at ~500-kDa which is presumed to be soluble LRP1 or soluble gp330/megalin. However, the strongest signal appears at a much faster position of migration estimated to correspond to ~195-kDa. The identity of this species is currently unknown.

A prominent band at a similar running position is independently observed for lysate of human umbilical vein endothelial cells (HUVEC) in a Western blotting using the LRP1 -chain specific monoclonal antibody, IgG-1B3 (Figure 5-9b). The specificity of labelling at this position is controlled by the band‟s absence in a similar blot of HUVEC using 8G1, indicating that detection is not an artefact of the secondary antiserum. As positive controls, IgG-1B3 is observed to react in bands at 600-kDa and 85-kDa corresponding to pre-cleaved and mature LRP1 -chain in lysates of HepG2 and bovine smooth muscle cells (SMC). Interestingly, these lysates also display minor bands at ~195-kDa.

The results suggest that tears and endothelial cells posses an unusually sized molecule related to the -chain of LRP1. The estimated molecular weight does not appear to correspond to normal -chain (85-kDa), dimeric -chain (170-kDa), or related receptors (LDL receptor at 160-kDa, VLDL receptor at 160-kDa, or apoER2/LRP8 at 160-kDa). Normal processing of 600-kDa LRP1 produces a 515/85-kDa -/-chain heterodimer, the chains of which are not associated by disulfide bonds. Assuming, the specificity of IgG-1B3 is restricted to the LRP1 -chain, and given that electrophoresis in the presence of SDS dissociates the - and -chain, a molecule at ~195-kDa would be expected to be derived from the 600-kDa receptor in a pre-Golgi compartment. The results are curious and further work is required to characterise further this dominant RAP-binding molecule in tears.

132

Figure 5-9 RAP-binding proteins in lacrimal fluid. Panel a shows a 125I-RAP blot of lacrimal fluid. In contrast to serum and CSF, the strongest band appears at 195-kDa and only a weak band is observed at ~500-kDa. Panel b contains Western blots of lysates prepared from human umbilical vein endothelial cells (HUVEC, at passage 6) with HepG2 and bovine smooth muscle cells (SMC) as positive controls. HUVEC display a weak signal at ~500-kDa using LRP1 -chain antibody, 8G1. As expected, the -chain specific monoclonal antibody IgG-1B3 recognises an epitope in pre-processed (600-kDa) and processed (85-kDa) LRP1 -chain in HepG2 and SMC lysates. IgG-1B3 also recognises a protein in HUVEC with a similar molecular weight (195-kDa) to the RAP-binding protein in lacrimal fluid. The positions of molecular weight markers are shown on the left and calculated molecular masses on the right. Results of a single determination are shown.

133 5.3.12 Detection of soluble LRP1 in CSF

To investigate the possible existence of soluble LRP1, in cerebral spinal fluid (CSF), six samples were randomly selected from specimens submitted for examination by the Pathology laboratories at the Prince of Wales Hospital, NSW. The patients from whom the specimens originated had a range of disorders, predominantly related to infections, and not to dementia. Blotting with 125I-RAP (Figure 5-10, left panel) identified reactivity with a ~500-kDa protein in four of the six CSF samples, suggesting that detectable soluble LRP1 is present in the CSF of some, but not all, individuals, and that its level may vary considerably within the human population, or with disease states.

Figure 5-10 Soluble LRP1 in cerebral spinal fluid (CSF). The left panel is a 125I-RAP blot showing the appearance of bands at ~500-kDa in CSF from four of five samples (10 l) selected at random from specimens submitted to a Pathology laboratory. These CSF samples originated from patients with various conditions unrelated to dementia. The Western blot using 8G1 (anti-LRP1) in the right panel confirms the presence of soluble LRP1 immunologically in CSF samples from four patients with dementia. Results of a single set of determinations are shown.

134 The identity of the high molecular weight RAP-binding proteins in CSF was further investigated by Western blotting using anti-LRP1 antibody 8G1. To extend the study of CSF, samples were obtained from 4 patients with dementia-likely related to Alzheimers disease. As shown in (Figure 5-10, left panel) all four samples produced bands at ~500-kDa, which support the identity of molecules in these bands, as well as the RAP-binding molecules in CSF from the other patients investigated, as soluble LRP1. The results may have important implications in Alzheimers disease.

5.4 DISCUSSION

The initial studies leading to the detection and characterisation of a soluble form of LRP1 in serum are described. These preliminary experiments revealed features of the molecule yet to be reported, as well as the presence of related molecules in other body fluids. Further studies on the characterisation of soluble LRP1 have been published (Grimsley et al., 1999; Grimsley et al., 1998; Quinn et al., 1997), and the release mechanism has subsequently been addressed by Dr K A Quinn and colleagues (Quinn et al., 1999).

Serum is shown to contain a class of high molecular mass molecules which can be extracted on immobilised anti-LRP1 -chain antibody, and subsequently bind RAP. These experiments demonstrate that the molecules are related to the -chain of LRP1 by three characteristic features: (1) ~500-kDa molecular mass; (2) antigenic recognition; and (3) functional ability to bind an LRP1 ligand (RAP). Additional ligand-binding and immunological evidence to support the identification of LRP1 -chain and a fragment of the -chain, is presented in Chapter 6. In particular, these high molecular mass, molecules extracted from sera by immobilised RAP, are shown to bind the archetypal

LRP1 ligand, activated 2-macroglobulin (Figure 6-3 in Chapter 6).

An early concern was that the presence of soluble LRP1 in serum may have been an artefact of clotting. Platelets express the related receptor, apoE2R/LRP8 (Riddell et al., 1999), and megakaryocytes express LRP1 (Bouchard et al., 2008). Despite a report to the contrary (Riddell et al., 1998), one study suggested that platelets also contain LRP1 (Shaw, 1994). Given that platelets shed a number of surface molecules during

135 activation (Fox, 1994), a survey of LRP1 distribution in platelets was appropriate. Indeed, Western blotting detected the presence of some full-length LRP1 in a lysate of activated platelets, and flow cytometric analysis using a rabbit antiserum against LRP1, detected antigens on resting platelets. However, soluble LRP1 was not detected in the released products from activated platelets suggesting that its presence in serum was not an artefact and reflected its presence in circulation.

The availability of a plasmid to generate recombinant RAP (see 2.5.5 for acknowledgments and methods) aided these studies as both a detection probe and as a basis for isolation. RAP could be conjugated with fluorescein for flow cytometric studies (2.5.5), radiolabelled with 125I (2.5.2) for ligand blotting (2.6.6), and immobilised to sepharose (2.5.7) to enrich soluble LRP1 from serum (5.3.6). At 6 mg/L, soluble LRP1 is a rare protein in serum which contains ~70 g/L, and the extraneous protein load causes problems with background in Western blotting. Techniques investigated to reduce this background included enrichment on immobilised 8G1 and ammonium sulfate cutting. While these were effective, enrichment on immobilised RAP provided the most convenient procedure and was adopted for further studies (see 5.3.6 for further discussion).

Previous studies had reported an inability to detect soluble LRP1. It is possible that the high sensitivity offered by initial screening with 125I-RAP, and the subsequent reduced background following enrichment of serum on RAP-sepharose, may have contributed to the success in the current approach.

Soluble LRP1 in serum appears stable at 4C for several months. However, the receptor contains numerous disulfide bonds in the LDL receptor class A repeats and EGF repeats, which will hold the protein together even following limited proteolysis. Under the non-reducing conditions usually employed in this work, cleavage in these repeats may have little effect on the antigenic properties of the molecule, or on its function as a soluble receptor for LRP1 ligands. However, under reducing conditions, Western blotting of stored serum failed to detect bands at 500-kDa (data not shown) suggesting that soluble LRP1 in aged serum contains little, if any, 515-kDa -chain as a single chained polypeptide. Whether the protein is uncleaved in circulation is unknown. However, it is clear from blotting studies that soluble LRP1 in stored serum (albeit cleaved) retains ligand binding function.

136 LRP1 appears important in the clearance of -amyloid (A) (Narita et al., 1997b) and soluble LRP1 shed from Schwann cells modulates signalling in glia (Gaultier et al., 2008). Soluble LRP1 acts as a „sink‟ for A transcytosed across the blood-brain barrier to the plasma (Deane et al., 2008; Sagare et al., 2007). The presence of soluble LRP1 within the brain is likely to influence this efflux, so cerebral spinal fluid (CSF) from five randomly selected samples from patient without dementia were examined in 125I-RAP ligand blots and 4 samples from patients with dementia in Western blots using anti-LRP1 -chain antibody (Figure 5-10). The results show the presence of detectable high molecular mass soluble LRP in 4 of the 5 non-dementia patients, and all of those with dementia. Unlike serum, where the level of soluble LRP is fairly constant, the intensity range in the bands in Figure 5-10, suggests that the concentration is less regulated. This is a preliminary examination, so the exact role of soluble LRP in CSF in these conditions will require a much larger study.

While the major signals from serum and CSF samples usually appear at ~500-kDa in both Western blots using anti-LRP1 ligands, and ligand blots using 125I-RAP, several bands at lower molecular mass also appear on occasions. These are assumed to be artefacts caused by cleavage in regions outside disulfide bonded loops, and are generally ignored. Signals in fresh samples are largely confined to a single band at ~500-kDa, although minor signals at lower molecular masses appear in stored samples (compare results in panels a and b of Figure 5-2). In some samples, the signals at lower molecular mass can be prominent, as observed in CSF samples from patients with dementia (Figure 5-10). These samples were taken from frozen material stored in small aliquots which may have been thawed and frozen a number of times. As such the significance of these additional bands is unknown.

The predominant band appearing at ~195-kDa in the 125I-RAP blot of lacrimal fluid is of more interest (Figure 5-9). This sample was analysed immediately following collection and is therefore unlikely to be a degradation artefact. While tears were not investigated further, their pattern of reactivity with 125I-RAP, bears striking resemblance to patterns in IgG-1B3 Western blotting of endothelial cell lysate. A further similarity between endothelial cell lysate and tears is the weakness of signals observed at 500-kDa in Western blots using antibody against LRP1 -chain, and in ligand blots using 125I-RAP, respectively. These results indicate that neither contains substantial quantities of full-length LRP1. Further investigation is required to establish the relationship, if

137 any, between the dominant RAP-binding molecule in tears and the molecule of similar molecular mass harbouring a LRP1 -chain epitope in endothelial cells. At ~195-kDa, clearly neither is full length LRP1, and the mechanism leading to inclusion of -chain epitope in a molecule greater than 85-kDa and less that 600-kDa (mature -chain, and pre-processed LRP1 respectively) has not been described.

The presence of soluble LRP1 in serum raises the question of its clinical significance.. Given the role of cellular LRP1 in lipid metabolism, examination of serum from patients with altered lipid levels may be of interest. In a preliminary study, purified placental LRP1 in solution at concentrations mimicking physiological levels of authentic soluble LRP1 (10 mg/L) was shown to retard tPA clearance (Quinn et al., 1997). This finding provides evidence that soluble LRP1 may also disturb the catabolism of fibrinolytic system components and raises the possibility that altered levels of the soluble receptor may influence pathological conditions involving any of the numerous LRP1 ligands.

Some ligands of LRP1 are immobilised by glycosaminoglycans raising the question of whether soluble LRP1 can, in turn, be immobilised by binding to these ligands. Antigenically distinct soluble LRP1 (rabbit receptor) was investigated for its ability to associate with HepG2 monolayers. These hepatoma cells are an appropriate selection for this study given that they are known to secrete many serum proteins. Rabbit soluble LRP1 was shown to associate with HepG2 monolayers and be dissociated by the removal of calcium. These findings suggest that blood-borne LRP1 can be immobilised by cells or extracellular matrix. Further work is required to determine the influence this process may have on the masking matrix associated molecules.

Soluble LRP1 has been shown to be released from rat hepatocytes (Quinn et al., 1997) and from the choriocarcinoma cell line BeWo (Quinn et al., 1999). While the latter finding may have implications for pregnancy, the release from hepatocytes is suggestive that the liver is a source of soluble LRP1 in the blood of both sexes. A previous study was unable to detect the release of synthesised LRP1 into the supernatant of cultured HepG2 or fibroblasts (Quinn et al., 1997). However, using a different subline of HepG2 (HepG2a16), which expresses higher levels of LRP1, and a more direct examination for released receptor, this hepatoma cell line was shown here to shed

138 trace amounts of LRP1. This finding further supports the suggestion that liver may be a major source of soluble LRP1 in serum.

The suggestion that liver is the source of soluble LRP1 in serum is unlikely to extend to its presence in CSF. Neurons and astrocytes are known to express the receptor, lending credence to the concept that a released form is likely for any body fluid in contact with LRP1-expressing tissue. The presence of soluble LRP1 in CSF may have important implications for neurological disorders. In particular, Alzheimers disease has several known associations with LRP1 (Donahue et al., 2006; Van Uden et al., 2000a; Van Uden et al., 2000b) and certain genetic polymorphism within the receptor increase predisposition to the disease (Beffert et al., 1999; Bullido et al., 2000; Kolsch et al., 2003; Sanchez et al., 2001; Verpillat et al., 2001). Moreover, LRP1 itself and many of its ligands are present in the characteristic senile plaques (Rebeck et al., 1995). The source of this LRP1 and its role in promoting the accumulation of these molecules is unknown. The finding in the results reported here that serum-derived LRP1 can associate with HepG2 monolayers (Figure 5-8), raises the possibility that soluble LRP1 in CSF may associate with plaque by a similar mechanism.

The most poignant relationship between Alzheimers disease and LRP1 is the fact that the major component of plaque in Alzheimers disease, the 39-42 amino acid peptide amyloid  (A), is derived from the processing of amyloid precursor protein, which, as a soluble form of the variant containing a Kunitz domain, is a ligand of LRP1 (Rebeck et al., 2001; Ulery et al., 2000). Given the preliminary findings reported here that a soluble form of LRP1 is presence in the CSF of all 4 patients with dementia examined, but only 4 of 5 patients with unrelated disorders, it is tempting to speculate that soluble LRP1 in CSF may play a direct role in its aetiology, and be a useful marker of the disease.

139

Chapter 6

6 EVOLUTIONARY CONSERVATION OF SOLUBLE LRP1

6.1 INTRODUCTION

A soluble form of LRP1 in serum was described in the previous Chapter. While results presented there were focussed on human serum, use was made of serum from other species to allow antigenic tracking of soluble LRP1 in various compartment. Intrinsic to these experiments was an assumption that conservation of cellular LRP1 ligand recognition extended to the generation of a similar soluble forms in other mammals. A strong indication for this conservation became obvious early in these studies when RAP of rat origin was found to detect soluble LRP1 in bovine and human serum.

Cellular LRP1 is known to be widely conserved in that homologous molecules have been detected in birds, (Seo et al., 1997) and species as primitive as nematodes (Yochem and Greenwald, 1993) and crustaceans (Iwaki et al., 1996; Melchior et al., 1995). In the studies presented here, the extent to which generation of soluble forms of these receptors is evolutionarily conserved was investigated using the specific high- affinity binding properties of RAP to probe for their presence in the circulation of a range of species. Described here is the presence of molecules which bind to rat RAP in the circulation of species ranging from humans to a mollusc. The predominant RAP- binding molecules are of high molecular weight and appear, by ligand binding profiles to be homologous to LRP1, at least in mammals, birds, and reptiles. These findings extend the range of species found to synthesise LRP1-like receptors to reptiles and molluscs, and suggest that generation of soluble forms is a conserved companion feature in the biology of the receptor.

Hypothesis:

To determine whether the release of soluble LRP1 is a feature conserved in evolution.

141 6.2 MATERIALS AND METHODS

6.2.1 Receptor Associated Protein and RAP-Sepharose 2B (RAP-2B).

Recombinant rat RAP immobilised on sepharose 2B (RAP-2B) was prepared by the method described in 2.5.7 and used to enrich all serum samples for soluble LRP1. A similar procedure was attempted on molluscan haemolymph without success. Contributing factors to this finding may be the apparent lower concentration of detectable RAP-binding proteins in haemolymph compared to serum and the possibility that molluscan receptor has low affinity for rat RAP. Alteration of conditions, such as pH and salinity, during the application of haemolymph to RAP-2B matrix, and the use of larger volumes, may be more informative, but were not attempted here for sample availability reasons.

6.2.2 Sera and Haemolymph

Human sera were obtained from volunteers. Mammalian sera were standard culture reagents purchased from Gibco/Life Technologies, Grand Island, CA (foetal bovine, rabbit, and horse) or Sigma (St Louis, MO, mouse and rat). Chicken serum was collected by W. Bryden, University of Sydney, Camden, Australia. Reptilian sera from an aldaboran tortoise (Geochelone gigantea), a diamond python (Morelia spilota spilota), and an eastern diamond backed rattlesnake (Crotalus adamanteus) were supplied by F. Hulst, Taronga Park Zoo, Sydney, Australia. Cell-free haemolymph from a giant clam (Tridacna gigas) was provided by D. Yellowlees, James Cook University of North Queensland, Australia. All specimens were collected under approval from the ethics committees of the various institutions, and were centrifuged at 10,000g for 10 minutes prior to use of the particle-free supernatants.

6.2.3 Enrichment of soluble RAP-binding proteins on RAP-2B

Sera (1 ml) were added to 400 l aliquots of 50% RAP-2B slurry, mixed for 1 hour, transferred to a small disposable column (BioRad) and washed with 5 aliquots of 10 ml

142 PBS / 0.5 mM Ca++. The column exit was plugged and the RAP-2B resuspended with 200 l PBS / 10 mM EDTA (elution buffer). After 5 minutes the column was unplugged and released proteins in the eluate were collected and combined with the void volume by flushing with a further 100 l of elution buffer. The fractions were pooled and frozen in 50 l aliquots at -80C until used.

6.2.4 Membrane LRP1/2MR and LRP2/megalin Standards

The hepatoma cell line, HepG2 clone a16, was a gift from A. L. Schwartz, Washington University, St Louis, MO. This subline of HepG2 has high expression of

LRP1/2MR (Grimsley et al., 1997) and was maintained in continuous culture as described (Owensby et al., 1988). It was referred to as a16/HP in Chapter 4. Four days after plating and 12 hours after medium exchange, HepG2 monolayers in 25 cm2 flasks were washed four times in cold phosphate buffered saline (PBS), lysed in 1 ml of 25 mM Tris pH 6.8 / 2% sodium dodecyl sulfate (SDS) / 10% glycerol / 5 mM leupeptin / 1 mM phenylmethylsulfonyl fluoride (Laemmli sample buffer with inhibitors), and homogenised by six passages through a 25 gauge needle. Lysates were diluted to 5 mg of protein per ml and frozen until used. A kidney biopsy taken from a cadaver by G. Higgins (University of New South Wales, Sydney, Australia) was homogenised in cold PBS with inhibitors, centrifuged at 300g for 10 minutes, and the crude membrane preparation in the supernatant washed once at 40,000g for 30 minutes before freezing. HepG2 and kidney extracts were used in ligand and western blots as membrane sources of LRP1/2MR and LRP2/megalin, respectively.

6.2.5 Primary Antibodies

Murine IgG1 monoclonal antibody 8G1, and an affinity purified rabbit antiserum, both raised against human 515 kDa LRP1 -chain, were generous gifts from D. K. Strickland (American Red Cross, Rockville, MD). Hybridomas, IgG-5D7, IgG-11H4, and IgG-1B3, producing murine IgG1 against LRP1/2MR components were purchased from the American Type Culture Collection, Rockville, MD (ATCC). Their specificities have been previously described (Herz et al., 1990b). Antibodies were

143 purified on protein A-agarose (Sigma) as described (Jiskoot et al., 1989), dialysed against PBS, 0.2 sterile filtered, and stored at 4C.

6.2.6 Western and Ligand Blots

RAP-2B extracts (10 l) were added to 40 l of Laemmli sample buffer (Laemmli, 1970) and electrophoresed on 5% polyacrylamide running gels with 4% stacking gels containing 0.1% sodium docedyl sulfate (SDS) except as follows. Western and 125I-RAP ligand blots shown in Figure 6-5c and Figure 6-5d were derived from non-denaturing gels by omitting SDS in the gels and the Laemmli buffer, and the results in Figure 6-1a were produced on a high resolution system using 1% agarose throughout 3% running and 2% stacking gels. Separated proteins in all systems were transferred to PVDF (polyvinylidene difluoride) membranes (NEN Research Products, Boston, MA) which were subsequently blocked for 1 hour with PBS / 0.5 mM Ca++ /

0.05% Tween 20 (PBSCT) containing 5% skimmed milk. After four 10 minutes washes with PBSCT, iodinated ligands (10 nM), monoclonal antibodies (5 g/ml), or rabbit anti-LRP1/2MR antibody (0.6 g/ml) in PBSCT / 1% BSA were added for 30 minutes and further washed four times. Radiolabelled ligand blots were dried and exposed to autoradiography film (Hyperfilm, Amersham, Sweden) with an intensifying screen for 24 to 48 hours at -80C. Western blots were incubated with horseradish peroxidase- conjugated secondary antiserum (Dako, Glostrup, Denmark) of appropriate species specificity (anti-mouse, or -rabbit immunoglobulins) for 30 minutes in PBSCT / 1% BSA at dilutions recommended by the manufacturer. Following washing, membranes were developed with chemiluminescence reagents (Renaissance, NEN Research Products) and exposed to Hyperfilm for 10 to 20 minutes. One lane of each gel contained prestained molecular weight standards (10 l, high range, Gibco) mixed with 5 g fibronectin (500 kDa, ICN or Gibco), and another lane containing 10 g HepG2 protein extract.

144 6.2.7 Iodinations

Ligands (20 g) were radiolabelled with 125I (Australian Radioisotopes, Lucas Heights, Australia) using iodogen (Pierce, Rockford, IL) as described (Owensby et al., 1991) (see Section 2.5.2).

6.3 RESULTS

6.3.1 RAP-binding protein in animal sera and haemolymph

To determine whether RAP-binding proteins exist in other species, RAP-2B extracts from human, bovine, rabbit, chicken, rat, mouse, and horse sera were separated on 3% polyacrylamide / 1% agarose, transferred to PVDF membrane, and blotted with 125I-RAP. The resultant autoradiogram in Figure 6-1a shows that 125I-RAP bound to serum components from all species tested and that the most prominent band for each animal has an identical electrophoretic mobility to 515 kDa membrane-derived human LRP1 -chain in HepG2 lysate. Furthermore, the bands migrate faster than LRP2/megalin (the major band in kidney membranes corresponding to 600 kDa) and coincide with the running position of human fibronectin (500 kDa). These results show that high molecular weight RAP-binding molecules are present in the sera of a range of mammals and also in the circulation of a separate phylogenetic class, aves.

In separate experiments, similar results were obtained in 125I-RAP ligand blots examining 2 l of whole serum substituted for the RAP-2B extracts (data not shown). However, Western blots of unextracted sera were found to produce complex multiple bands in regions less than 500 kDa. In order to maintain consistency between 125I-RAP blots (Figure 6-1) and western blots (Figure 6-4 and Figure 6-5), RAP-2B extracts were used throughout except for the haemolymph experiment in Figure 6-1b.

The evolutionary conservation of high molecular weigh RAP-binding molecules was further explored by examining the sera of another phylogenetic class, reptilia, and the circulating haemolymph of a member of a different phylum, mollusca. Figure 6-1a

145

Figure 6-1 125I-RAP ligand blotting of soluble LRP1-like molecules from various species. Results using serum extracted on RAP-sepharose 2B (RAP-2B) are shown for: (Panel a) six mammals, and a bird (chicken); and (Panel c) three species of reptile. These experiments were controlled internally by inclusion of kidney membranes (containing gp330/megalin at 600-kDa), HepG2 cell lysate (containing cellular LRP1 at ~500-kDa), and molecular weight standards whose positions are indicated on the left. Panel b shows external controls consisting of HepG2 and HUVEC lysates representing high and low cellular sources of LRP1. Panel d contains results using 3, 10, and 30 l samples of centrifuged whole haemolymph from the giant clam, Tridacna gigas, and lyophilised material representing 50 l of the same haemolymph.. Tissue (mantle) homogenate from T gigas, HepG2 lysate, and purified placental LRP1, are included as positive controls. Samples were separated on; (a) 3%, and (b, c, d) 5% polyacrylamide. 125I-RAP binding signals at ~500-kDa are evident in the circulation of all species tested. Due to the scarcity of the biological materials, the results of a single experiment was performed.

146 shows the 125I-RAP blots of serum extracts from three species of reptiles, and unextracted circulating haemolymph from a mollusc (giant clam). Bands at approximately 500 kDa are observed from each species. Haemolymph extracted on RAP-2B by a protocol identical to that used for serum failed to produce bands in 125I-RAP blots presumably due to lower concentrations of RAP-binding molecules in the circulation of this mollusc compared to vertebrates. Their reactivity with RAP, their high molecular mass, and their presence in each of the species examined, suggests that the soluble molecules identified belong to the LDL R family and that their synthesis and release is widely conserved.

6.3.2 Pseudomonas exotoxin A (PEA) ligand blots.

To characterise the identity of the molecules in the 500 kDa RAP-binding bands, ligand blots using 125I-PEA at pH 6.8 were performed and the resultant autoradiograms are shown in Figure 6-2. PEA is a known ligand of human LRP1/2MR, but like activated 2-macroglobulin, does not cross-bind to human gp330 (Kounnas et al., 1993a; Kounnas et al., 1994). No signals in 125I-PEA blots of haemolymph from the giant clam, T. gigas were observed although the presence of PEA-binding molecules below the level of detection could not be excluded. In contrast, RAP-2B serum extracts from all species examined demonstrated PEA-binding bands at 500 kDa suggesting that soluble RAP-binding proteins present in the circulation of mammals, birds, and reptiles share characteristics attributable to human LRP1.

147

Figure 6-2 125I-Pseudomonas exotoxin A (PEA) ligand blots of RAP-2B serum extracts. Bands are observed at ~500-kDa in all species blotted with 125I-PEA indicating that the RAP-binding molecules detected in Figure 6-1 a have the additional “LRP1-like” characteristic of PEA recognition. Insufficient sample was available to test python serum extract adequately, and haemolymph failed to be enriched on RAP-2B. HepG2 lysate is included as a positive control. The specificity of 125I-PEA for LRP1 is controlled by its lack of reactivity with kidney membranes which contains the 600-kDa homologous receptor, LRP2/megalin. Due to the scarcity of the biological materials, the results of a single set of determinations are shown.

148 2 5 6.3.3 ¹ I-activated 2-macroglobulin ligand binding

The high molecular weight molecules extracted from mammalian and avian sera on immobilised RAP were examined for binding with the classic LRP1 ligand, activated 125 2-macroglobulin ( I-2M*) in ligand blots using radiolabelled human 2M activated with methylamanine (1 nM). The results in Figure 6-3 show weak bands at ~500kDa for human, bovine, rabbit, and chicken proteins. The lack of reactivity with rat, mouse and horse proteins may reflect species specificity, or the condition of the particular samples at the time.

125 125 Figure 6-3 Activated I-2macroglobulin ( I-2MM) blotting of RAP-2B extracted animal sera. Radiolabelled methylamine-activated 2-macroglobulin 125 ( I-2MM, 1 nM) was used to blot RAP-2B extracted animal sera. In Panel a, very faint bands are observed at ~500-kDa in lanes containing human, bovine, rabbit, and chicken proteins. The specificity of these signals is demonstrated by their absence in the presence of (b) EDTA, and (c) RAP. Panel d contains a 125I-RAP control blot to verify loading. Results of a single set of determinations are shown.

149 6.3.4 Monoclonal antibody identification of soluble 500 kDa LRP1 -Chain in mammalian sera

To characterise the RAP-binding molecules antigenically, antibodies against LRP1 -chain were used in Western blots of RAP-2B serum extracts. IgG-5D7 and 8G1, two monoclonal antibodies raised against rabbit and human 515 kDa LRP1 -chain respectively, reacted as expected with serum extracts from the species corresponding to their immunogens (Figure 6-4a and Figure 6-4b). That is, IgG-5D7 reacted with rabbit serum extract (Figure 6-4a) while 8G1 reacted with human serum extract (Figure 6-4b). As expected, 8G1 also detected human membrane LRP1/2MR in the HepG2 lysate (Figure 6-4b). IgG-5D7 displayed mono-species specificity for rabbit LRP1 with no band appearing at 500 kDa in lanes containing RAP-binding molecules from any species other than rabbit (compare Figure 6-1a with Figure 6-4a). In contrast, 8G1 reacted with RAP-2B serum extracts from mammals additional to human, including horse and rabbit (Figure 6-4b). Bands at approximately 150 kDa in the mouse and rat extracts (Figure 6-4a and Figure 6-4b) are contaminating mouse and rat immunoglobulins detected by the secondary anti-mouse antiserum.

The presence of soluble LRP1 antigen in mammalian sera was further supported in blots using an affinity purified rabbit anti-human LRP1 -chain antibody (Figure 6-4c). As expected human material of both membrane and serum origins contained detectable LRP1 antigen (Figure 6-4c, first two lanes). This antibody also recognised 500 kDa RAP-binding proteins from foetal bovine, mouse, rat, and horse sera. The strong band at 160 kDa in the rabbit lane of Figure 6-4c represents detection of contaminating rabbit immunoglobulins by the secondary anti-rabbit antiserum.

Together, the results in Figure 6-4a, b and c utilise three different antibodies to identify a soluble form of antigenic LRP1 -chain in the sera of all mammals investigated. The species-specificities of these antibodies are summarised in Table 1 together with the reactivity patterns using the LRP1 ligands, RAP and PEA.

150

Figure 6-4 Western blotting of RAP-2B serum extracts for LRP1 -chain. RAP- 2B extracts were separated on 5% polyacrylamide to show: (a) species specific identification of rabbit soluble LRP1 -chain at ~500-kDa using monoclonal antibody IgG-5D7; (b) limited mammalian cross-reactivity with monoclonal antibody, 8G1, raised against human LRP1 -chain; and (c) detection of soluble LRP1 in multiple mammalian species using an affinity purified rabbit antiserum (“777”) raised against human LRP1 -chain. Together, these antibodies identify soluble LRP1 -chain antigen in the sera of all mammals tested. Major bands at molecular weights lower than 500-kDa represent contaminating immunoglobulins detected by the secondary antisera. Each blot includes at least one negative control in the form of chicken serum extract which is known to contain RAP- and PEA-binding molecules of ~500-kDa (see Figure 6-1 and Figure 6-2). A representative set of blots from duplicate determinations is shown.

151 6.3.5 -Chain fragments in soluble LRP1

Having confirmed the presence of the LRP1 extracellular -chain among the RAP-binding proteins of mammalian sera, the possible presence of components of the LRP1 -chain was investigated using two monoclonal antibodies IgG-11H4 and IgG-1B3.

IgG-11H4 was raised against a synthetic peptide representing the 13 amino acids at the intracellular carboxy-terminus of the human LRP1 -chain and has previously been demonstrated to react with rabbit and human membrane LRP1 (Herz et al., 1990b), but not with human soluble LRP1 isolated from plasma on activated 2-macroglobulin- coated matrix (Quinn et al., 1997). In the Western blots of HepG2 lysate shown in Figure 6-4a, IgG-11H4 detected 85 kDa and 600 kDa bands corresponding, respectively, to mature LRP1/2MR -chain, and single chain polypeptide present in pre-golgi compartments. The band at ~300 kDa in the HepG2 lystate lane is currently unidentified. IgG-11H4 did not detect any relevant bands using serum extracts from multiple species (Figure 6-4a) even though loading controls confirmed the presence of 125I-RAP binding proteins in each lane (Figure 6-4b).

The second -chain monoclonal antibody used, IgG-1B3, reacts with the ectodomain of rabbit LRP1 -chain and is known to cross-react with rat -chain (Herz et al., 1990b). In a Western blot of serum extracts separated under non-denaturing conditions (absence of SDS), this antibody produced bands (Figure 6-5c) which have identical electrophoretic mobilities to 125I-RAP binding activities separated under the same conditions (Figure 6-5d). In the absence of SDS the - and -chain would be expected to remain associated unlike the disruption caused under the denaturing conditions used in the other experiments. The observation that the 125I-RAP binding activity (corresponding to LRP1 -chain) co-migrates with molecules recognised by

IgG-1B3 under non-denaturing conditions suggests that soluble LRP12MR in mammalian sera retains an attached portion of the -chain ectodomain. However, the absence of reactivity with IgG-11H4 implies that the -chain is incomplete and lacks intracellular components at the carboxy terminus. IgG-1B3 reactivity was evident with three mammalian sera tested suggesting that the retention of a portion of the -chain is a conserved feature in the release of soluble LRP1-like molecules, at least in mammals.

152

Figure 6-5 Western blotting of RAP-2B serum extracts for LRP1 -chain. RAP-2B serum extracts were examined: (a) under denaturing (SDS present) conditions using the cytoplasmic tail specific antibody, IgG-11H4, and (c) under non-denaturing (SDS absent) conditions using the -chain ectodomain specific antibody, IgG-1B3. Loading for each lane was verified using 125I-RAP blotting of duplicate gels for each condition (b, d). In Panel a, IgG-11H4 reactivity with HepG2 lysate is observed to produce bands at 600-kDa (representing single-chain LRP1 running slightly slower than the 500-kDa fibronectin marker) and at 85-kDa (representing mature -chain). Neither of these bands was detectable in the sera of the mammals tested. In contrast, IgG-1B3 detected the presence of an epitope in the -chain ectodomain in human, bovine, rabbit, and horse sera. Under non-denaturing conditions this -chain fragment co-migrates with 125I-RAP binding proteins (LRP1) suggesting that the two proteins remain associated. IgG-1B3 reactivity with mouse and rat extracts is uninterpretable in this non-denaturing system due to interference from immunoglobulins detected by the secondary antiserum. A representative set of blots from duplicate determinations is shown.

153 6.4 DISCUSSION

A soluble form of LRP1 exists in human plasma that retains the capacity to bind the LRP1 ligands, RAP, and methylamine-activated 2-macroglobulin (see Chapter 5). In the work presented here, RAP derived from rat cDNA is shown to bind to soluble molecules of approximately 500 kDa in the sera from five additional mammals, a bird (chicken), and three species of reptiles. Moreover, RAP demonstrated a remarkable evolutionary conservation of receptor-recognition by binding to a high molecular weight soluble molecule in the haemolymph of a giant clam. Because this is the circulatory fluid of a member of phylum mollusca, this finding represents a cross-species ligand recognition between vertebrates and invertebrates, and suggests that the generation of soluble RAP-binding molecules may be required in fundamental receptor-mediated functions common to widely divergent evolutionary species.

The nature of the 500 kDa RAP-binding protein in the sera of mammals, reptiles, and chickens was investigated using the LRP1-specific ligand, Pseudomonas exotoxin A (PEA). The only known RAP-binding proteins with the high molecular mass in the detected bands are LRP1, LRP1b, and LRP2/megalin. A feature distinguishing LRP1 from LRP2/megalin is the ability of LRP1, but not LPR2/megalin, to bind PEA (Kounnas et al., 1993a). Each of the mammalian, avian, and reptilian sera extracted on RAP-2B produced a band at 500 kDa in 125I-PEA ligand blots suggesting that the RAP-binding molecules in these sera have ligand binding qualities more closely related to LRP1 than other members of the LDL R family.

Further evidence for the presence of soluble LRP1 was obtained from Western blots of RAP-2B serum extracts using monoclonal and polyclonal antibodies against LRP1 -chain (Figure 6-4). Bands at 500 kDa were evident using one or two of these antibodies against extracts from each of the six mammalian sera tested. Insufficient material was available to perform these experiments on the reptilian samples. None of the three antibodies raised against mammalian LRP1/2MR reacted with chicken serum extract suggesting that the cross-species specificities of the antibodies are restricted.

The composition of soluble LRP1 compared to the membrane form of the receptor was investigated in western blots using anti-LRP1 -chain antibodies, IgG-11H4 and IgG-1B3 which are directed against cytoplasmic and extracellular epitopes respectively.

154 IgG-11H4 did not react with RAP-2B serum extracts from any of the mammals tested, suggesting that a component (or all) of the intracellular carboxy-terminus of the -chain is absence in soluble LRP1. Conversely, for each of the mammalian serum extracts tested, IgG-1B3 recognised molecules co-migrating with the RAP-binding activity in non-denaturing gels. These results suggest that a fragment of the -chain, incorporating an epitope on the ectodomain, remains associated with soluble LRP1 -chain in mammalian sera. A model of the findings is shown in Figure 6-6.

Soluble forms of membrane receptors are quite common and the mechanisms leading to their generation are known in some cases (reviewed in Ehlers and Riordan, 1991; Levine, 2008; Montuori and Ragno, 2009; Rose-John and Heinrich, 1994). Metalloproteases acting independently can mediate the shedding of multiple cell surface proteins, including receptors (Arribas et al., 1996). These enzymes can also act in concert with protein disulfide isomerase which reduces disulfide bonds anchoring cleaved fragments to their membrane-bound receptor remnants (Couët et al., 1996). Alternative splicing can generate soluble homologues of membrane-bound receptors by causing the removal of the transmembrane regions (Astier et al., 1994; Müller-Newen et al., 1996) or by introducing an extra sequence into the receptor‟s ectodomain which becomes the target of proteases (Tolchinsky et al., 1996). Once generated these soluble receptors can have profound biological effects. For example, transgenic mice expressing soluble human IL-6 receptor are hypersensitised to the effects of IL-6 (Peters et al., 1996), or conversely, soluble TNF receptor can antagonise responses to TNF (Williams et al., 1996).

LRP1 is known to be released by BACE, a transmembrane protease with -secretase activity, which also cleaves the amyloid precursor protein (APP) (Von Arnim et al., 2005). Dissociation of the non-covalent interactions holding the LRP1 - and -chains together has been proposed as a possible release mechanism (Moestrup, 1994), but this proposal, on its own, does not explain the presence of the -chain fragment in soluble LRP Figure 6-5. The existence of one established release mechanism (BACE) does not necessarily preclude other mechanisms, especially is tissue specific environments. For example, further experiments would be required to show that all soluble LRP molecules contain this fragment. It remains a possibility that a portion could be released by the original dissociation concept.

155 The demonstration of high molecular weight RAP-binding proteins in the circulation of several species representing a wide evolutionary spectrum provides further evidence for the conserved nature of LRP1-like receptors. The presence of soluble forms of these receptors in multiple species indicates that their generation is also highly conserved and suggests that they may play an essential role in the overall function of this receptor system.

Figure 6-6 Composition of soluble LRP1. 8G1, IgG-5D7, and 3402 are murine monoclonal antibodies which recognise the LRP -chain. IgG-1B3 and IgG-11H4 are murine monoclonal antibodies which recognise, respectively, the -chain‟s ectodomain proximal to the membrane, and the -chain‟s intracellular C-terminus. Soluble LRP1 is composed of the entire -chain, and the extracellular portion of the -chain.

156

Chapter 7

7 CONCLUSIONS

CONCLUSIONS AND FUTURE STUDIEs

This work initially determined that urokinase-type plasminogen activator (uPA) binds to HepG2 as a complex with PAI-1 in a mechanism analogous to that previously characterised for tPA.

The major receptor involved in this process has since been identified by others as LRP1. Studies presented here confirmed that LRP1 mediates the degradation of tPA, but LRP1 is not solely responsible for the binding of tPA at 4ºC by the cell line, HepG2. Indeed, LRP1 was found to be only a minor contributor to total specific binding of tPA by the original HepG2 cell line available from the ATCC. In contrast, the clone, HepG2a16, expressed much higher levels of LRP1 and proportionally bound 65% via LRP1. The non-LRP1 binding site for tPA on HepG2 was not identified, and but was shown not to mediate endocytosis of tPA. LRP1 accounted for the entire degradative capacity for tPA on all HepG2 sublines examined.

To investigate the binding of tPA to HepG2 further, four subclones of the line were studied and found to express LRP1 at a wide range of basal levels. Expression could not be stably altered by a variety of perturbations to culture conditions suggesting that constitutive levels of expression varied among the individual lines and that long term culture may favour outgrowth of cells expressing higher levels of LRP1. This would explain the rise in expression with passage number observed for both the parental HepG2 cells, and the clone, HepG2a16.

Variable expression of LRP1 among HepG2 sublines raised a possibility that shedding via /-chain dissociation may alter surface membrane concentrations and thereby account for the range in basal levels. If so, dissociated receptor might be expected in HepG2 culture supernatant. However, soluble LRP1 found in abundance in the supernatant of these human cells was a bovine protein originating from the serum supplementing the culture medium. This diversionary finding led to the identification of soluble LRP in human plasma and serum.

Soluble receptors have been an emerging area of interest since the late 1970s, with many displaying profound effects on the activity of their ligands, and on their membrane-bound congeners. The large repertoire of structurally and functionally diverse ligands endocytosed by LRP1, lodges the receptor in many physiological and pathological settings, and additional signalling functions extend its range further. Of

158 particular interest are the associations with Alzheimer disease (AD). Certain polymorphisms in LRP1 itself present a weak predisposition to AD, but possession of the apoE4 isoform of apoE is a major risk factor (Higgins et al., 1997; Kuusisto et al., 1994). As a receptor for apoE, LRP1 is thought to play a role in neuronal apoptosis (Hashimoto et al., 2000), and neuronal plasticity (Narita et al., 1997a), but any modulatory effect imposed by the presence of soluble LRP has not been investigated. Similarly, LRP1 regulates signalling by the NMDA glutamate receptor, (Qiu et al., 2002) but the effect of soluble LRP1 on this system is untested. LRP1 mediates the clearance of -amyloid (A) (Sagare et al., 2007), which can form fibrils and be deposited as the major component of senile plaque. Soluble LRP1 is thought to be a „sink‟ for A in plasma, and to mediate its efflux from the brain (Deane et al., 2008). While A is the major component of the senile plaques in Alzheimers disease, many LRP ligands are also present in these deposits, as is LRP1 itself (Rebeck et al., 1995). It is tempting to speculate on the circumstances leading to the extracellular deposition of these once-soluble proteins. Soluble LRP is likely to have been accessible these protein prior to their incorporation into plaque. The presence of soluble LRP1 in cerebral spinal fluid (CSF) is an area likely to prove increasingly important in the development, diagnosis, or management, of AD.

The generation of soluble LRP1 is an ancient process, with its presence detected in the sera of all vertebrates tested, and possibly in the haemolymph of an invertebrate. In humans, soluble LRP is present at levels in a narrow normal range, suggesting that its concentration may be regulated. If so, this implies a functional role, possibly as a reservoir, such as the „sink‟ role proposed for A (Sagare et al., 2007). Completely unexplored is any potential fragmentation. While the protein in serum appears largely as a 500kDa species, it is possible that small peptides could be cleaved off for signalling or hormonal purposes without apparent change in its size.

Finally, the lifespan and ultimate fate of soluble LRP1 as a circulating protein is unknown. The immense capacity of LRP1 to recognise ligands implies that soluble LRP1 will interact with other components of blood, and may acquire binding partners such as lipoproteins, protease-activated α2-macroglobulin, or plasminogen activators. If so, these complexes composed of soluble receptors and ligands might be expected to be cleared. An interesting question is the identity of such a clearance mechanism. Perhaps LRP1 clears its own soluble form.

159

Extended Abstract

160 EXTENDED ABSTRACT

This work describes an investigation of some of the processes involved in the clearance and degradation of the plasminogen activators (PAs), tissue-type plasminogen activator (tPA) and urokinase-type plasminogen activator (uPA). In humans, these two serine proteinases are responsible for the generation of a third serine proteinase, plasmin, from its abundant zymogen, plasminogen. Plasmin degrades a broad spectrum of substrates and its activity is central to several physiological and pathophysiological processes including fibrinolysis, cell migration, atherosclerosis, and tumour metastasis. Plasmin activity is tightly regulated in these processes and is predominantly controlled by a balance between rapid inhibition (mainly by 2-antiplasmin), and generation by the PAs. In turn, PAs are regulated by the presence of specific inhibitors, notably plasminogen activator inhibitor type-1 (PAI-1), and by the rates of their clearance. PA clearance is therefore important in the many processes in which plasmin is involved. The introduction of both tPA and uPA as pharmaceuticals in the treatment myocardial infarction and stroke provides further impetus for understanding their clearance because the efficacy of the PAs is limited by their rapid removal from circulation, predominantly by the liver. Accordingly, the work described here addresses the catabolism of the PAs - the final phase in their regulation..

Mechanisms in the hepatic catabolism of uPA. tPA was previously known to be bound, endocytosed, and degraded by the hepatocellular cell line, HepG2. This process required initial reaction of tPA with PAI-1 located on vitronectin in the cell line‟s extracellular matrix. The resulting tPAPAI-1 binary complexes are tightly bound as evident from their stability the presence of the denaturing detergent, sodium dodecyl sulfate (SDS). uPA is structurally homologous to tPA and forms similar stable complexes with PAI-1 suggesting that the binding and endocytosis of both PAs may be governed by common mechanisms. The investigation of this possibility described here from the early 1990s revealed that, like tPA, uPA formed analogous uPAPAI-1 complexes prior to binding to HepG2. Furthermore, the receptor involved appeared to be distinct from the uPA receptor (uPAR), which binds to a region near the amino terminus of uPA. Both low and high molecular weight forms of catalytically active uPA (LMW uPA and HMW-uPA) bind in a similar fashion to HepG2. Given that

161 LMW-uPA lacks the amino terminal region and that this region mediates binding to the uPAR, these findings suggest involvement of a receptor distinct from the uPAR.

Variation of LRP1 expression among HepG2 cell lines. At the time these experiments were performed, other laboratories showed that the predominant receptor responsible for mediating PA degradation by the liver, is the giant (600 kDa) multifunctional member of the low density lipoprotein receptor family, low density lipoprotein receptor-related protein (LRP1).

The HepG2 cell line maintained in our laboratory was examined for the contribution made by LRP1 in binding tPA. Surprisingly, less than 30% of specific tPA binding by this particular cell line could be attributed to LRP1. This finding appeared inconsistent with values reported for HepG2 maintained in other laboratories. Accordingly, a systematic survey was undertaken of LRP1 levels in four HepG2 sublines obtained from various centres. Two of these sublines were derived from the parental line and two from a clone of HepG2 termed “a16”. Collectively the group represented examples of parental line and the a16 clone, each at low and high passage numbers.

In addition to the PAs, LRP1 mediates the uptake of numerous structurally and functionally distinct ligands. The binding of each of these ligands can be inhibited by the presence of recombinant forms of a 39 kDa endoplasmic reticulum-resident chaperone molecule termed receptor associated protein (RAP). The contribution of LRP1 to the specific binding of 125I labelled tPA for each of the four HepG2 sublines was determined by performing binding assays in the presence and absence of RAP, and comparing these results to binding in the presence and absence of unlabelled tPA. The a16 clones were found to bind higher levels of tPA than the parental lines, and this increase was attributable to higher levels of LRP1. Higher levels of LRP1 also corresponded to higher passage number within both the a16 pair and the parental line pair. The contribution of LRP1 to specific tPA binding was found to vary considerably among the sublines, ranging from 10% to 65%.

Determination of variable LRP1 levels among the four HepG2 sublines was confirmed by performing: (1) tPA degradation assays; (2) western blots using an LRP1- specific antibody; and (3) cytotoxic susceptibility assays using the LRP1-specific ligand, Pseudomonas exotoxin A (PEA). The observation that tPA binding could be

162 only partially inhibited by RAP indicated that LRP1 is not the only binding site for tPA in HepG2 cultures. However, the presence of RAP consistently ablated the ability of each of the sublines to degrade tPA despite a considerable range in their degradative capacities. These findings suggest that LRP1 is the sole binding site responsible for mediating tPA degradation, and that its level varies considerably among HepG2 cell sublines.

Soluble LRP1 in humans. Original experiments examining the possibility of LRP1 shedding used media containing foetal bovine serum (FBS). If shedding was operating, then a soluble form of LRP1 might be expected in HepG2 culture supernatants. Indeed, the expected signal at ~500 kDa was detected in ligand blots of culture supernatants probed with 125I-RAP. However, fresh media in controls gave identical results, and FBS was recognised as the source of this activity. This serendipitous finding soon extended to the identification of a previously undescribed protein in human circulation, soluble LRP1.

Human serum enriched on RAP immobilised to sepharose CL-2B (RAP-2B) and released by EDTA, was examined in Western blots using LRP1-specific antibodies, and in radiolabelled ligand blots probed with the LRP1 ligand, Psuedomonas exotoxin A (PEA). Both experiments produced a specific ~500 kDa signal, confirming that these RAP-binding molecules in normal human serum and plasma are a soluble form of LRP1 which retains ligand binding capacity.

LRP1 is synthesised as a 600 kDa single-chained polypeptide which is cleaved in the trans-Golgi to form a heterodimer. At the cell surface the mature receptor consists of the LRP1 515 kDa ligand-binding -chain, anchored via tight non-covalent interactions to its 85 kDa -chain. The latter is comprised of an ectodomain, a transmembrane region, and a cytoplasmic tail. Because the - and -chains are associated non-covalently, simple reasoning leads to the suspicion that soluble LRP1 may result for dissociation of the - and - chains.

Further characterisation using region-specific antibodies revealed that soluble LRP1 in human plasma contains the -chain and a fragment of the -chain‟s ectodomain, but lacks at least part of the cytoplasmic region. The presence of a -chain fragment in soluble LRP1 is not consistent with spontaneous dissociation of the - and

163 - chains as the mechanism releasing soluble LRP1 into the circulation, although it cannot be excluded as a minor contributor.

Other body fluids were examined for the presence of similar high molecular weight RAP-binding proteins, and signals were detected in urine. However, urine is known to contain a soluble form of another giant member of the LDL receptor family, LRP2/megalin. A fluid considered likely to contain soluble LRP1 was cerebral spinal fluid (CSF) as LRP1 is expressed in brain. Using 125I-RAP ligand blotting, signals corresponding to ~500 kDa were detected in CSF from four patients with dementia, and four of five CSF samples collected from patients for conditions unrelated to dementia. These preliminary results raised the possibility the association between LRP1 and Alzheimers disease may extend to an involvement of a soluble form of the receptor.

Soluble LRP1 in other species. Soluble LRP1 was initially detected in bovine serum by probing with RAP of rat origin. Investigation of this cross species recognition was extended to other mammals (mouse, rat, rabbit, and horse), as well as sera from non-mammalian species including three species of reptile, and a bird (chicken). In all cases, serum components recognising rat-RAP on CL-2B were enriched and shown to contain ~500 kDa signals in Western blots using polyclonal and monoclonal antibodies, and in ligand blots using 125I-PEA. Furthermore, 125I-RAP produced a signal at ~500 kDa in the haemolymph of a mollusc. These findings suggest that the generation of soluble LRP1 is a conserved process and that soluble forms of LRP1 may play an important role in the operation of this receptor system.

Conclusions. The results presented here are consistent with tPA and uPA being cleared by the liver via similar mechanisms and that the major receptor responsible is LRP1. A soluble form of this receptor retaining ligand-binding functions has been detected and characterised in the circulation. The role of soluble LRP1, and the extent of its influence on the biology of its membrane-bound counterpart, has yet to be determined. Evolutionary conservation of its generation suggests that it may play an essential role in the receptor‟s function.

164

8 REFERENCES

165

BIBLIOGRAPHY

1 Aden, D.P., Fogel, A., Plotkin, S., Damjanov, I. and Knowles, B.B. 1979 "Controlled synthesis of HBsAg in a differentiated human liver carcinoma-derived cell line." Nature 282(5739), 615-616

2 Agirbasli, M. 2005 "Pivotal role of plasminogen-activator inhibitor 1 in vascular disease" Int J Clin Pract 59(1), 102-106

3 Alessi, M.C., Declerck, P.J., De Mol, M., Nelles, L. and Collen, D. 1988 "Purification and characterization of natural and recombinant human plasminogen activator inhibitor-1 (PAI-1)" European Journal of Biochemistry 175(3), 531-540

4 Alessi, M.C. and Juhan-Vague, I. 2008 "Metabolic syndrome, haemostasis and thrombosis" Thrombosis & Haemostasis 99(6), 995-1000

5 Appella, E., Robinson, E.A., Ullrich, S.J., Stoppelli, M.P., Corti, A., Cassani, G. and Blasi, F. 1987 "The receptor-binding sequence of urokinase: A biological function for the growth-factor module of proteases" Journal of Biological Chemistry 262(10), 4437-4440

6 Arribas, J., Coodly, L., Vollmer, P., Kishimoto, T.K., Rose-John, S. and Massagué, J. 1996 "Diverse cell surface protein ectodomains are shed by a system sensitive to metalloprotease inhibitors" Journal of Biological Chemistry 271(19), 11376-11382

7 Ashcom, J.D., Tiller, S.E., Dickerson, K., Cravens, J.L., Argraves, W.S. and Strickland, D.K. 1990

"The human 2-macroglobulin receptor: Identification of a 420-kD cell surface glycoprotein specific for the activated conformation of 2-macroglobulin" Journal of Cell Biology 110(4), 1041-1048

8 Astier, A., De La Salle, H., De La Salle, C., Bieber, T., Esposito-Farese, M.-E., Freund, M., Cazenave, J.-P., Fridman, W.-H., Teillaud, J.-L. and Hanau, D. 1994 "Human epidermal langerhans cells secrete a soluble receptor for IgG (FcII/CD32) that inhibits the binding of immune complexes to FcR+ cells" Journal of Immunology 152(1), 201-212

9 Bachovchin, W.W., Kaiser, R., Richards, J.H. and Roberts, J.D. 1981 "Catalytic mechanism of serine proteases: reexamination of the pH dependence of the histidyl 1J13C2-H coupling constant in the catalytic triad of alpha-lytic protease" Proceedings of the National Academy of Sciences of the United States of America 78(12), 7323-7326

10 Bakhit, C., Lewis, D., Billings, R. and Malfroy, B. 1987 "Cellular catabolism of recombinant tissue- type plasminogen activator. Identification and characterization of a novel high affinity uptake system on rat hepatocytes" Journal of Biological Chemistry 262(18), 8716-8720

11 Balsara, R.D. and Ploplis, V.A. 2008 "Plasminogen activator inhibitor-1: the double-edged sword in apoptosis" Thrombosis & Haemostasis 100(6), 1029-1036

12 Bányai, L., Váradi, A. and Patthy, L. 1983 "Common evolutionary origin of the fibrin-binding structures of fibronectin and tissue-type plasminogen activator" FEBS Letters 163(1), 37-41

13 Bassel-Duby, R., Jiang, N.Y., Bittick, T., Madison, E., Mcgookey, D., Orth, K., Shohet, R., Sambrook, J. and Gething, M.-J. 1992 "Tyrosine 67 in the epidermal growth factor-like domain of tissue-type plasminogen activator is important for clearance by a specific hepatic receptor" Journal of Biological Chemistry 267(14), 9668-9677

166 14 Bastholm, L., Nielsen, M.H., De Mey, J., Dano, K., Brunner, N., Hoyer-Hansen, G., Ronne, E. and Elling, F. 1994 "Confocal fluorescence microscopy of urokinase plasminogen activator receptor and cathepsin D in human MDA-MB-231 breast cancer cells migrating in reconstituted basement membrane" Biotechnic & Histochemistry 69(2), 61-67

15 Battey, F.D., Gåfvels, M.E., Fitzgerald, D.J., Argraves, W.S., Chappell, D.A., Strauss Iii, J.F. and Strickland, D.K. 1994 "The 39-kDa receptor-associated protein regulates ligand binding by the very low density lipoprotein receptor" Journal of Biological Chemistry 269(37), 23268-23273

16 Beebe, D.P. and Aronson, D.L. 1986 "Turnover of human tissue plasminogen activator (tPA) in rabbits" Thrombosis Research 43(6), 663-674

17 Beebe, D.P. and Aronson, D.L. 1988 "Turnover of tPA in rabbits: influence of carbohydrate moieties" Thrombosis Research 51(1), 11-22

18 Beffert, U., Arguin, C. and Poirier, J. 1999 "The polymorphism in exon 3 of the low density lipoprotein receptor-related protein gene is weakly associated with Alzheimer's disease" Neuroscience Letters 259(1), 29-32

19 Beisiegel, U., Schneider, W.J., Goldstein, J.L., Anderson, R.G. and Brown, M.S. 1981 "Monoclonal antibodies to the low density lipoprotein receptor as probes for study of receptor-mediated endocytosis and the genetics of familial hypercholesterolemia" Journal of Biological Chemistry 256(22), 11923-11931

20 Beisiegel, U., Weber, W., Ihrke, G., Herz, J. and Stanley, K.K. 1989 "The LDL-receptor-related protein, LRP, is an apolipoprotein E-binding protein" Nature 341(6238), 162-164

21 Belin, D., Vassalli, J.D., Combepine, C., Godeau, F., Nagamine, Y., Reich, E., Kocher, H.P. and Duvoisin, R.M. 1985 "Cloning, nucleotide sequencing and expression of cDNAs encoding mouse urokinase-type plasminogen activator" European Journal of Biochemistry 148(2), 225-232

22 Bennett, B., Croll, A., Ferguson, K. and Booth, N.A. 1990 "Complexing of tissue plasminogen

activator with PAI-1, 2 -macroglobulin, and C1-inhibitor: Studies in patients with defibrination and a fibrinolytic state after electroshock of complicated labour" Blood 75(3), 671-676

23 Bhattacharjee, G., Gron, H. and Pizzo, S.V. 1999 "Incorporation of non-proteolytic proteins by murine alpha(2)-macroglobulin" Biochimica et Biophysica Acta - Protein Structure & Molecular Enzymology 1432(1), 49-56

24 Biessen, E.A., Van Teijlingen, M., Vietsch, H., Barrett-Bergshoeff, M.M., Bijsterbosch, M.K., Rijken, D.C., Van Berkel, T.J. and Kuiper, J. 1997 "Antagonists of the mannose receptor and the LDL receptor-related protein dramatically delay the clearance of tissue plasminogen activator" Circulation 95(1), 46-52

25 Bird, S.D., Walker, R.J. and Hubbard, M.J. 1994 "Altered free calcium transients in pig kidney cells (LLC-PK1) cultured with penicillin/streptomycin" In Vitro Cellular & Developmental Biology 30A(7), 420-424

26 Blacklow, S.C. 2007 "Versatility in ligand recognition by LDL receptor family proteins: advances and frontiers" Current Opinion in Structural Biology 17(4), 419-426

27 Blasi, F., Vassalli, J.D. and Dano, K. 1987 "Urokinase-type plasminogen activator: proenzyme, receptor, and inhibitors" Journal of Cell Biology 104(4), 801-804

28 Bottiger, B.W., Arntz, H.R., Chamberlain, D.A., Bluhmki, E., Belmans, A., Danays, T., Carli, P.A., Adgey, J.A., Bode, C. and Wenzel, V. 2008 "Thrombolysis during resuscitation for out-of-hospital cardiac arrest" New England Journal of Medicine 359(25), 2651-2662

167 29 Bouchard, B.A., Meisler, N.T., Nesheim, M.E., Liu, C.X., Strickland, D.K. and Tracy, P.B. 2008 "A unique function for LRP-1: a component of a two-receptor system mediating specific endocytosis of plasma-derived factor V by megakaryocytes" J Thromb Haemost 6(4), 638-644

30 Boxrud, P.D., Verhamme, I.M.A., Fay, W.P. and Bock, P.E. 2001 "Streptokinase triggers conformational activation of plasminogen through specific interactions of the amino-terminal sequence and stabilizes the active zymogen conformation" Journal of Biological Chemistry 276(28), 26084-26089

31 Brenner, B. 2004 "Haemostatic changes in pregnancy" Thrombosis Research 114(5-6), 409-414

32 Broderick, J.P. 2009 "Endovascular therapy for acute ischemic stroke" Stroke 40(3 Suppl), S103-106

33 Brown, M.S. and Goldstein, J.L. 1986 "A receptor-mediated pathway for cholesterol homeostasis" Science 232(4746), 34-47

34 Browne, M.J., Tyrrell, A.W., Chapman, C.G., Carey, J.E., Glover, D.M., Grosveld, F.G., Dodd, I. and Robinson, J.H. 1985 "Isolation of a human tissue-type plasminogen-activator genomic DNA clone and its expression in mouse L cells" Gene 33(3), 279-284

35 Bu, G., Morton, P.A. and Schwartz, A.L. 1992a "Identification and partial characterization by chemical cross-linking of a binding protein for tissue-type plasminogen activator (t-PA) on rat hepatoma cells: A plasminogen activator inhibitor type-1 independent t-PA receptor" Journal of Biological Chemistry 267(22), 15595-15602

36 Bu, G., Williams, S., Strickland, D.K. and Schwartz, A.L. 1992b "Low density lipoprotein receptor-

related protein/2-macroglobulin receptor is an hepatic receptor for tissue-type plasminogen activator" Proceedings of the National Academy of Sciences of the United States of America 89(16), 7427-7431

37 Bu, G., Maksymovitch, E.A. and Schwartz, A.L. 1993 "Receptor-mediated endocytosis of tissue-type plasminogen activator by low density lipoprotein receptor-related protein on human hepatoma HepG2 cells" Journal of Biological Chemistry 268(17), 13002-13009

38 Bu, G., Maksymovitch, E.A., Nerbonne, J.M. and Schwartz, A.L. 1994 "Expression and function of the low density lipoprotein receptor-related protein (LRP) in mammalian central neurons" Journal of Biological Chemistry 269(28), 18521-18528

39 Bugelski, P.J., Fong, K.L., Klinkner, A., Sowinski, J., Rush, G. and Morgan, D.G. 1989 "Uptake of human recombinant tissue-type plasminogen activator by rat hepatocytes in vivo: an electron microscope autoradiographic study" Thrombosis Research 53(3), 287-303

40 Bullido, M.J., Guallar-Castillon, P., Artiga, M.J., Ramos, M.C., Sastre, I., Aldudo, J., Frank, A., Coria, F., Rodriguez-Artalejo, F. and Valdivieso, F. 2000 "Alzheimer's risk associated with human apolipoprotein E, alpha-2 macroglobulin and lipoprotein receptor related protein polymorphisms: absence of genetic interactions, and modulation by gender" Neuroscience Letters 289(3), 213-216

41 Bürgin, J. and Schaller, J. 1999 "Expression, isolation and characterization of a mutated human plasminogen kringle 3 with a functional lysine binding site" Cellular & Molecular Life Sciences 55(1), 135-141

42 Cam, J.A., Zerbinatti, C.V., Knisely, J.M., Hecimovic, S., Li, Y. and Bu, G. 2004 "The low density lipoprotein receptor-related protein 1B retains beta-amyloid precursor protein at the cell surface and reduces amyloid-beta peptide production" Journal of Biological Chemistry 279(28), 29639-29646

43 Camani, C., Bachmann, F. and Kruithof, E.K.O. 1994 "The role of plasminogen activator inhibitor type 1 in the clearance of tissue-type plasminogen activator by rat hepatoma cells" Journal of Biological Chemistry 269(8), 5770-5775

168 44 Camani, C. and Kruithof, E.K. 1995 "The role of the finger and growth factor domains in the clearance of tissue-type plasminogen activator by hepatocytes" Journal of Biological Chemistry 270(44), 26053-26056

45 Camani, C., Gavin, O., Bertossa, C., Samatani, E. and Kruithof, E.K. 1998 "Studies on the effect of fucosylated and non-fucosylated finger/growth-factor constructs on the clearance of tissue-type plasminogen activator mediated by the low-density-lipoprotein-receptor-related protein" European Journal of Biochemistry 251(3), 804-811

46 Cao, M., Buratini, J., Jr., Lussier, J.G., Carriere, P.D. and Price, C.A. 2006 "Expression of protease nexin-1 and plasminogen activators during follicular growth and the periovulatory period in cattle" Reproduction 131(1), 125-137

47 Carmeliet, P., Schoonjans, L., Kieckens, L., Ream, B., Degen, J., Bronson, R., De Vos, R., Van Den Oord, J.J., Collen, D. and Mulligan, R.C. 1994 "Physiological consequences of loss of plasminogen activator gene function in mice" Nature 368(6470), 419-424

48 Carpenter, S.L. and Mathew, P. 2008 "Alpha2-antiplasmin and its deficiency: fibrinolysis out of balance" Haemophilia 14(6), 1250-1254

49 Carrell, R.W. and Owen, M.C. 1985 "Plakalbumin, 1-antitrypsin, antithrombin and the mechanism of inflammatory thrombosis" Nature 317(6039), 730-732

50 Casey, J.R., Petranka, J.G., Kottra, J., Fleenor, D.E. and Rosse, W.F. 1994 "The structure of the Urokinase-type plasminogen activator receptor gene" Blood 84(4), 1151-1156

51 Castellino, F.J. Custom 5 1988 "Structure/function relationships of human plasminogen and plasmin" in Tissue-type plasminogen (t-PA): Physiological and Clinical Aspects Vol. I of 2 pp. 145-169 Ed. Plasmin(Ogen)_Chemistry_Review Pub CRC Press Inc., Boca Raton, Florida.

52 Cesarman-Maus, G. and Hajjar, K.A. 2005 "Molecular mechanisms of fibrinolysis" Br J Haematol 129(3), 307-321

53 Chapman, K.M., Woolfenden, A.R., Graeb, D., Johnston, D.C.C., Beckman, J., Schulzer, M. and Teal, P.A. 2000 "Intravenous tissue plasminogen activator for acute ischemic stroke - A Canadian hospital's experience" Stroke 31(12), 2920-2924

54 Chen, W.-J., Goldstein, J.L. and Brown, M.S. 1990 "NPXY, a sequence often found in cytoplasmic tails, is required for coated pit-mediated internalization of low density lipoprotein receptor`" Journal of Biological Chemistry 265(6), 3116-3123

55 Choi, S.Y. and Cooper, A.D. 1993 "A comparison of the roles of the low density lipoprotein (LDL)

receptor and the LDL receptor-related protein/2-macroglobulin receptor in chylomicron remnant removal in the mouse in vivo" Journal of Biological Chemistry 268(21), 15804-15811

56 Christensen, U. 1985 "C-terminal lysine residues of fibrinogen fragments essential for binding to plasminogen" FEBS Letters 182(1), 43-46

57 Christensen, U., Simonsen, M., Harrit, N. and Sottrup-Jensen, L. 1989 "Pregnancy zone protein, a proteinase-binding macroglobulin. Interactions with proteinases and methylamine" Biochemistry 28(24), 9324-9331

58 Conese, M., Olson, D. and Blasi, F. 1994 "Protease nexin-1-urokinase complexes are internalized and degraded through a mechanism that requires both urokinase receptor and alpha 2-macroglobulin receptor" Journal of Biological Chemistry 269(27), 17886-17892

169 59 Conese, M., Nykjær, A., Petersen, C.M., Cremona, O., Pardi, R., Andreasen, P.A., Gliemann, J.,

Christensen, E.I. and Blasi, F. 1995 "-Macroglobulin receptor/LDL receptor-related protein (LRP)- dependent internalization ot the Urokinase receptor" Journal of Cell Biology 131(6 Pt 1), 1609-1622

60 Couët, J., De Bernard, S., Loosfelt, H., Saunier, B., Milgrom, E. and Misrahi, M. 1996 "Cell surface protein disulfide-isomerase is involved in the shedding of human thyrotropin receptor ectodomain" Biochemistry 35(47), 14800-14805

61 D'alessio, S. and Blasi, F. 2009 "The urokinase receptor as an entertainer of signal transduction" Front Biosci 14, 4575-4587

62 Danø, K., Nielsen, L.S., Pyke, C. and Kellerman, G.M. Custom 5 1988 "Plasminogen activators and neoplasia" in Tissue-type plasminogen (t-PA): Physiological and Clinical Aspects Vol. I of 2 pp. 19- 46 Ed. Pas_Cancer_Review Pub CRC Press Inc., Boca Raton, Florida.

63 Danø, K., Behrendt, N., Hoyer-Hansen, G., Johnsen, M., Lund, L.R., Ploug, M. and Romer, J. 2005 "Plasminogen activation and cancer" Thrombosis & Haemostasis 93(4), 676-681

64 Davidsen, O., Christensen, E.I. and Gliemann, J. 1985 "The plasma clearance of human 2- macroglobulin-trypsin complex in the rat is mainly accounted for by uptake into hepatocytes" Biochimica et Biophysica Acta 846(1), 85-92

65 Davis, B.J. 1964 "Disc electrophoresis, II Method and application to human serum proteins" Annals of the New York Academy of Sciences 121, 404-427

66 De Munk, G.A., Caspers, M.P., Chang, G.T., Pouwels, P.H., Enger-Valk, B.E. and Verheijen, J.H. 1989 "Binding of tissue-type plasminogen activator to lysine, lysine analogues, and fibrin fragments" Biochemistry 28(18), 7318-7325

67 Deane, R., Sagare, A. and Zlokovic, B.V. 2008 "The role of the cell surface LRP and soluble LRP in blood-brain barrier Abeta clearance in Alzheimer's disease" Curr Pharm Des 14(16), 1601-1605

68 Dellas, C. and Loskutoff, D.J. 2005 "Historical analysis of PAI-1 from its discovery to its potential role in cell motility and disease" Thrombosis & Haemostasis 93(4), 631-640

69 Derynck, R., Roberts, A.B., Winkler, M., Chen, E.Y. and Goeddel, D.V. 1984 "Human transforming growth factor alpha:precursor structure and expression in E. coli" Cell 38(1), 287-297

70 Eaton, D.L., Scott, R.W. and Baker, J.B. 1984 "Purification of human fibroblast urokinase proenzyme and analysis of its regulation by proteases and protease nexin" Journal of Biological Chemistry 259(10), 6241-6247

71 Ehlers, M.R. and Riordan, J.F. 1991 "Membranes proteins with soluble counterparts: Role of proteolysis in the release of transmembrane proteins" Biochemistry 30(42), 10065-10074

72 Einarsson, M., Smedsrod, B. and Pertoft, H. 1988 "Uptake and degradation of tissue plasminogen activator in rat liver" Thrombosis & Haemostasis 59(3), 474-479

73 Eldering, E., Huijbregts, C.C., Nuijens, J.H., Verhoeven, A.J. and Hack, C.E. 1993 "Recombinant C1 inhibitor P5/P3 variants display resistance to catalytic inactivation by stimulated neutrophils" Journal of Clinical Investigation 91(3), 1035-1043

74 Eldering, E., Verpy, E., Roem, D., Meo, T. and Tosi, M. 1995 "COOH-terminal substitutions in the serpin C1 inhibitor that cause loop overinsertion and subsequent multimerization" Journal of Biological Chemistry 270(6), 2579-2587

75 Emeis, J.J., Van Den Hoogen, C.M. and Jense, D. 1985 "Hepatic clearance of tissue-type plasminogen activator in rats" Thrombosis & Haemostasis 54(3), 661-664

170 76 Enghild, J.J., Thogersen, I.B., Salvesen, G., Fey, G.H., Figler, N.L., Gonias, S.L. and Pizzo, S.V. 1990 "-macroglobulin from Limulus polyphemus exhibits proteinase inhibitory activity and participates in a hemolytic system" Biochemistry 29(43), 10070-10080

77 Enghild, J.J., Valnickova, Z., Thogersen, I.B. and Pizzo, S.V. 1994 "Complexes between serpins and inactive proteinases are not thermodynamically stable but are recognized by serpin receptors" Journal of Biological Chemistry 269(31), 20159-20166

78 Fang, I.M., Lin, Y.C., Yang, C.H., Yang, C.M. and Chen, M.S. 2009 "Effects of intravitreal gas with or without tissue plasminogen activator on submacular haemorrhage in age-related macular degeneration" Eye 23(2), 397-406

79 Fisher, R., Waller, E.K., Grossi, G., Thompson, D., Tizard, R. and Schleuning, W.D. 1985 "Isolation and characterization of the human tissue-type plasminogen activator structural gene including its 5' flanking region" Journal of Biological Chemistry 260(20), 11223-11230

80 Fitzgerald, D.J., Fryling, C.M., Zdanovsky, A., Saelinger, C., Kounnas, M., Strickland, D.K. and Leppla, S. 1994 "Selection of Pseudomonas Exotoxin-resistant cells with altered expression of

2MR/LRP" Annals of the New York Academy of Sciences 737, 138-144

81 Fitzgerald, D.J., Fryling, C.M., Zdanovsky, A., Saelinger, C.B., Kounnas, M., Winkles, J.A., Strickland, D. and Leppla, S. 1995 "Pseudomonas exotoxin-mediated selection yields cells with altered expression of low-density lipoprotein receptor-related protein. [erratum appears in J Cell Biol 1995 Aug;130(4):1015]" Journal of Cell Biology 129(6), 1533-1541

82 Fleury, V. and Angles-Cano, E. 1991 "Characterization of the binding of plasminogen to fibrin surfaces: the role of carboxy-terminal lysines" Biochemistry 30(30), 7630-7638

83 Fong, K.L., Crysler, C.S., Mico, B.A., Boyle, K.E., Kopia, G.A., Kopaciewicz, L. and Lynn, R.K. 1988 "Dose-dependent pharmacokinetics of recombinant tissue-type plasminogen activator in anesthetized dogs following intravenous infusion" Drug Metab Dispos 16(2), 201-206

84 Fox, J.E.B. 1994 "Shedding of adhesion receptors from the surface of activated platelets [Review]" Blood Coagulation & Fibrinolysis 5(2), 291-304

85 Frenette, G., Tremblay, R.R., Lazure, C. and Dube, J.Y. 1997 "Prostatic kallikrein hK2, but not prostate-specific antigen (hK3), activates single-chain urokinase-type plasminogen activator" International Journal of Cancer 71(5), 897-899

86 Friezner Degen, S.J., Rajput, B. and Reich, E. 1986 "The human tissue plasminogen activator gene" Journal of Biological Chemistry 261(15), 6972-6985

87 Gaeta, B.A., Borthwick, I. and Stanley, K.K. 1994 "The 5'-flanking region of the alpha 2MR/LRP gene contains an enhancer-like cluster of Sp1 binding sites" Biochimica et Biophysica Acta 1219(2), 307-313

88 Gaultier, A., Arandjelovic, S., Li, X., Janes, J., Dragojlovic, N., Zhou, G.P., Dolkas, J., Myers, R.R., Gonias, S.L. and Campana, W.M. 2008 "A shed form of LDL receptor-related protein-1 regulates peripheral nerve injury and neuropathic pain in rodents" Journal of Clinical Investigation 118(1), 161-172

89 Gettins, P. 1989 "Absence of large-scale conformational change upon limited proteolysis of ovalbumin, the prototypic serpin" Journal of Biological Chemistry 264(7), 3781-3785

90 Gettins, P., Patston, P.A. and Schapira, M. 1992 "Structure and mechanism of action of serpins" Hematology/Oncology Clinics of North America 6(6), 1393-1408

171 91 Gonias, S.L., Allietta, M.M., Pizzo, S.V., Castellino, F.J. and Tillack, T.W. 1988 "Electron microscopic identification of exposed plasmin epitopes in alpha 2-macroglobulin-plasmin complex using monoclonal antibody-colloidal gold adducts" Journal of Biological Chemistry 263(22), 10903- 10906

92 Grimsley, P.G., Normyle, J.F., Brandt, R.A., Joulianos, G., Chesterman, C.N., Hogg, P.J. and Owensby, D.A. 1995 "Urokinase binding and catabolism by Hep G2 cells is plasminogen activator inhibitor-1 dependent, analogous to interactions of tissue-type plasminogen activator with these cells" Thrombosis Research 79(4), 353-361

93 Grimsley, P.G., Quinn, K.A., Chesterman, C.N. and Owensby, D.A. 1996a "Low Density Lipoprotein Receptor-Related Protein Exists a a Soluble Form in Plasma" Australian Society for Experimental Biology - 28th annual meeting, University of New South aWales, Sydney, Australia 2-4 Oct, 18

94 Grimsley, P.G., Quinn, K.A., Chesterman, C.N. and Owensby, D.A. 1996b "A soluble form of low density lipoprotein receptor-related protein (LRP) circulates in plasma" Australian Vascular Biology Society - 4th Annual Scientific Meeting, Marysville, Victoria, Australia October 17-20(Abstract 9), 43

95 Grimsley, P.G., Quinn, K., Chesterman, C.N. and Owensby, D.A. 1997 "Low density lipoprotein receptor-related protein (LRP) expression varies among Hep G2 sublines" Thrombosis Research 88(6), 485-498

96 Grimsley, P.G., Quinn, K.A. and Owensby, D.A. 1998 "Soluble low-density lipoprotein receptor- related protein" Trends in Cardiovascular Medicine 8(8), 363-368

97 Grimsley, P.G., Quinn, K.A., Chesterman, C.N. and Owensby, D.A. 1999 "Evolutionary conservation of circulating soluble low density lipoprotein receptor-related protein-like ("LRP-like") molecules" Thrombosis Research 94(3), 153-164

98 Gunzler, W.A., Steffens, G.J., Otting, F., Buse, G. and Flohe, L. 1982a "Structural relationship between human high and low molecular mass urokinase" Hoppe-Seylers Zeitschrift fur Physiologische Chemie 363(2), 133-141

99 Gunzler, W.A., Steffens, G.J., Otting, F., Kim, S.M., Frankus, E. and Flohe, L. 1982b "The primary structure of high molecular mass urokinase from human urine. The complete amino acid sequence of the A chain" Hoppe-Seylers Zeitschrift fur Physiologische Chemie 363(10), 1155-1165

100 Gupta, S. and Gupta, B.M. 2008 "Acute pulmonary embolism advances in treatment" J Assoc Physicians India 56, 185-191

101 Hajjar, K.A., Hamel, N.M., Harpel, P.C. and Nachman, R.L. 1987 "Binding of tissue plasminogen activator to cultured human endothelial cells" Journal of Clinical Investigation 80(6), 1712-1719

102 Hajjar, K.A. and Reynolds, C.M. 1994 "-Fucose-mediated binding and degradation of tissue-type plasminogen activator by HepG2 cells" Journal of Clinical Investigation 93(2), 703-710

103 Hall, M., Soderhall, K. and Sottrup-Jensen, L. 1989 "Amino acid sequence around the thiolester of alpha 2-macroglobulin from plasma of the crayfish, Pacifastacus leniusculus" FEBS Letters 254(1-2), 111-114

104 Harris, R.J., Leonard, C.K., Guzzetta, A.W. and Spellman, M.W. 1991 "Tissue plasminogen activator has an O-linked fucose attached to threonine-61 in the epidermal growth factor domain" Biochemistry 30(9), 2311-2314

105 Hartnell, G.G. and Gates, J. 2000 "The case of Abbokinase and the FDA: the events leading to the suspension of Abbokinase supplies in the United States" J Vasc Interv Radiol 11(7), 841-847

172 106 Hashimoto, Y., Jiang, H., Niikura, T., Ito, Y., Hagiwara, A., Umezawa, K., Abe, Y., Murayama, Y. and Nishimoto, I. 2000 "Neuronal apoptosis by apolipoprotein E4 through low-density lipoprotein receptor-related protein and heterotrimeric GTPases" Journal of Neuroscience (Online) 20(22), 8401-8409

107 Haupert, C.L., Mccuen, B.W., Jaffe, G.J., Steuer, E.R., Cox, T.A., Toth, C.A., Fekrat, S. and Postel, E.A. 2001 "Pars plana vitrectomy, subretinal injection of tissue plasminogen activator, and fluid-gas exchange for displacement of thick submacular hemorrhage in age-related macular degeneration" American Journal of Ophthalmology 131(2), 208-215

108 Hayashi, M., Matsuzaki, Y. and Shimonaka, M. 2009 "Impact of plasminogen on an in vitro wound healing model based on a perfusion cell culture system" Mol Cell Biochem 322(1-2), 1-13

109 He, K.L., Deora, A.B., Xiong, H., Ling, Q., Weksler, B.B., Niesvizky, R. and Hajjar, K.A. 2008 "Endothelial cell annexin A2 regulates polyubiquitination and degradation of its binding partner S100A10/p11" Journal of Biological Chemistry 283(28), 19192-19200

110 Heeb, M.J., Espana, F., Geiger, M., Collen, D., Stump, D.C. and Griffin, J.H. 1987 "Immunological identity of heparin-dependent plasma and urinary protein C inhibitor and plasminogen activator inhibitor-3" Journal of Biological Chemistry 262(33), 15813-15816

111 Heegaard, C.W., Simonsen, A.C.W., Oka, K., Kjøller, L., Christensen, A., Madsen, B., Ellgaard, L., Chan, L. and Andreasen, P.A. 1995 "Very low density lipoprotein receptor binds and mediates endocytosis of urokinase-type plasminogen activator-type-1 plasminogen activator inhibitor complex" Journal of Biological Chemistry 270(35), 20855-20861

112 Hermanson, G.T., Mallia, A.K. and Smith, P.K. Custom 5 1992 "Immobilized affinity ligand techniques" Ed. Methods_Immobilisation Pub Academic Press Inc., Harcourt Brace Jovanovich, Sydney.

113 Herz, J., Hamann, U., Rogne, S., Myklebost, O., Gausepohl, H. and Stanley, K.K. 1988 "Surface location and high affinity for calcium of a 500-kd liver membrane protein closely related to the LDL- receptor suggest a physiological role as a lipoprotein receptor" EMBO Journal 7(13), 4119-4127

114 Herz, J., Kowal, R.C., Goldstein, J.L. and Brown, M.S. 1990a "Proteolytic processing of the 600kd low density lipoprotein receptor-related protein (LRP) occurs in a trans-golgi compartment" EMBO Journal 9(6), 1769-1776

115 Herz, J., Kowal, R.C., Ho, Y.K., Brown, M.S. and Goldstein, J.L. 1990b "Low density lipoprotein receptor-related protein mediates endocytosis of monoclonal antibodies in cultured cells and rabbit liver" Journal of Biological Chemistry 265(34), 21355-21362

116 Herz, J., Goldstein, J.L., Strickland, D.K., Ho, Y.K. and Brown, M.S. 1991 "39-kDa protein

modulates binding of ligands to low density lipoprotein receptor-related protein/2-macroglobulin receptor" Journal of Biological Chemistry 266(31), 21232-21238

117 Herz, J., Clouthier, D.E. and Hammer, R.E. 1992 "LDL receptor-related protein internalizes and degrades uPA-PAI-1 complexes and is essential for embryo implantation" Cell 71(3), 411-421

118 Herz, J., Clouthier, D.E. and Hammer, R.E. 1993 "Correction: LDL receptor-related protein internalizes and degrades uPA-PAI-1 complexes and is essential for embryo implantation" Cell 73(3), 428

119 Hiesberger, T., Trommsdorff, M., Howell, B.W., Goffinet, A., Mumby, M.C., Cooper, J.A. and Herz, J. 1999 "Direct binding of Reelin to VLDL receptor and ApoE receptor 2 induces tyrosine phosphorylation of disabled-1 and modulates tau phosphorylation" Neuron 24(2), 481-489

173 120 Higazi, A.A., Upson, R.H., Cohen, R.L., Manuppello, J., Bognacki, J., Henkin, J., Mccrae, K.R., Kounnas, M.Z., Strickland, D.K., Preissner, K.T., Lawler, J. and Cines, D.B. 1996 "Interaction of single-chain urokinase with its receptor induces the appearance and disappearance of binding epitopes within the resultant complex for other cell surface proteins" Blood 88(2), 542-551

121 Higgins, G.A., Large, C.H., Rupniak, H.T. and Barnes, J.C. 1997 "Apolipoprotein e and alzheimers disease - a review of recent studies" Pharmacology, Biochemistry & Behavior 56(4), 675-685

122 Hildenbrand, R. and Schaaf, A. 2009 "The urokinase-system in tumor tissue stroma of the breast and breast cancer cell invasion" Int J Oncol 34(1), 15-23

123 Hilliker, C., Van Leuven, F. and Van Den Berghe, H. 1992 "Assignment of the gene coding for the

2-macroglobulin receptor to mouse chromosome 15 and to human chromosome 12q13-q14 by isotopic and nonisotopic in situ hybridization" Genomics 13(2), 472-474

124 Hussain, M.M., Maxfield, F.R., Mas-Oliva, J., Tabas, I., Ji, Z.-S., Innerarity, T.L. and Mahley, R.W.

1991 "Clearance of chylomicron remnants by the low density lipoprotein receptor-related protein/2- macroglobulin receptor" Journal of Biological Chemistry 266(21), 13936-13940

125 Hussain, M.M., Kancha, R.K. and Tulenko, T.N. 1995 "Nickel is a specific inhibitor for the binding of activated alpha 2-macroglobulin to the low density lipoprotein receptor-related protein/alpha 2- macroglobulin receptor" Biochemistry 34(49), 16074-16081

126 Hussain, M.M. 2001 "Structural, biochemical and signaling properties of the low-density lipoprotein receptor gene family [Review]" Frontiers in Bioscience 6, D417-D428

127 Hussaini, I.M., Srikumar, K., Quesenberry, P.J. and Gonias, S.L. 1990 "Colony-stimulating factor-1

modulates 2-macroglobulin receptor expression in murine bone marrow macrophages" Journal of Biological Chemistry 265(32), 19441-19446

128 Iadonato, S.P., Bu, G., Maksymovitch, E.A. and Schwartz, A.L. 1993 "Interaction of a 39kDa protein with the low-density-lipoprotein-receptor-related protein (LRP) on rat hepatoma cells" Biochemical Journal 296(Pt 3), 867-875

129 Ichinose, A., Fujikawa, K. and Suyama, T. 1986a "The activation of pro-urokinase by plasma kallikrein and its inactivation by thrombin" Journal of Biological Chemistry 261(8), 3486-3489

130 Ichinose, A., Takio, K. and Fujikawa, K. 1986b "Localization of the binding site of tissue-type plasminogen activator to fibrin" Journal of Clinical Investigation 78(1), 163-169

131 Inocencio, N.M., Moehring, J.M. and Moehring, T.J. 1994 "Furin activates Pseudomonas exotoxin A by specific cleavage in vivo and in vitro" Journal of Biological Chemistry 269(50), 31831-31835

132 Iwaki, D., Kawabata, S., Miura, Y., Kato, A., Armstrong, P.B., Quigley, J.P., Nielsen, K.L., Dolmer, K., Sottrup-Jensen, L. and Iwanaga, S. 1996 "Molecular cloning of Limulus alpha 2-macroglobulin" European Journal of Biochemistry 242(3), 822-831

133 Jaeger, S. and Pietrzik, C.U. 2008 "Functional role of lipoprotein receptors in Alzheimer's disease" Curr Alzheimer Res 5(1), 15-25

134 Jensen, P.H., Moestrup, S.K. and Gliemann, J. 1989 "Purification of the human placental 2- macroglobulin receptor" FEBS Letters 255(2), 275-280

135 Jensen, P.H., Cressey, L.I., Gjertsen, B.T., Madsen, P., Mellgren, G., Hokland, P., Gliemann, J., Doskeland, S.O., Lanotte, M. and Vintermyr, O.K. 1994 "Cleaved intracellular plasminogen activator inhibitor 2 in human myeloleukaemia cells is a marker of apoptosis" British Journal of Cancer 70(5), 834-840

174 136 Jespers, L., Van Herzeele, N., Lijnen, H.R., Van Hoef, B., De Maeyer, M., Collen, D. and Lasters, I. 1998 "Arginine 719 in human plasminogen mediates formation of the staphylokinase:plasmin activator complex" Biochemistry 37(18), 6380-6386

137 Jiskoot, W., Van Hertrooij, J.J.C.C., Klien Gebbinck, J.W.T.M., Van Der Velden-De Groot, T., Crommelin, D.J.A. and Beuvery, E.C. 1989 "Two-step purification of a murine monoclonal antibody intended for therapeutic application in man. Optimisation of purification conditions and scaling up" Journal of Immunological Methods 124(1), 143-156

138 Jörnvall, H., Pohl, G., Bergsdorf, N. and Wallén, P. 1983 "Differential proteolysis and evidence for a residue exchange in tissue plasminogen activator suggest possible association between two types of protein microheterogeneity" FEBS Letters 156(1), 47-50

139 Joslin, G., Fallon, R.J., Bullock, J., Adams, S.P. and Perlmutter, D.H. 1991 "The SEC receptor

recognizes a pentapeptide neodomain of -antitrypsin-protease complexes" Journal of Biological Chemistry 266(17), 11282-11288

140 Kancha, R.K. and Hussain, M.M. 1996 "Up-regulation of low density lipoprotein receptor-related protein by dexamethasone in HepG2 cells" Biochimica et Biophysica Acta 1301(3), 213-220

141 Kasai, S., Arimura, H., Nishida, M. and Suyama, T. 1985 "Proteolytic cleavage of single-chain pro- urokinase induces conformational change which follows activation of the zymogen and reduction of its high affinity for fibrin" Journal of Biological Chemistry 260(22), 12377-12381

142 Kethireddy, S. and Safdar, N. 2008 "Urokinase lock or flush solution for prevention of bloodstream infections associated with central venous catheters for chemotherapy: a meta-analysis of prospective randomized trials" J Vasc Access 9(1), 51-57

143 Knauer, M.F., Crisp, R.J., Kridel, S.J. and Knauer, D.J. 1999 "Analysis of a structural determinant in thrombin-protease nexin 1 complexes that mediates clearance by the low density lipoprotein receptor-related protein" Journal of Biological Chemistry 274(1), 275-281

144 Knowles, B.B., Howe, C.C. and Aden, D.P. 1980 "Human hepatocellular carcinoma cell lines secrete the major plasma proteins and hepatitis B surface antigen" Science 209(4455), 497-499

145 Knowles, B.B. and Aden, D.P. 1983 "Human, hepatoma derived cell line, process for preparation thereof, and uses therefore" United-States-Patent 4,393,133

Methods_HepG2,

146 Kolsch, H., Ptok, U., Mohamed, I., Schmitz, S., Rao, M.L., Maier, W. and Heun, R. 2003 "Association of the C766T polymorphism of the low-density lipoprotein receptor-related protein gene with Alzheimer's disease" Am J Med Genet B Neuropsychiatr Genet 121B(1), 128-130

147 Komoriya, A., Hortsch, M., Meyers, C., Smith, M., Kanety, H. and Schlessinger, J. 1984 "Biologically active synthetic fragments of epidermal growth factor: Localization of a major receptor-binding region" Proceedings of the National Academy of Sciences of the United States of America 81(5), 1351-1355

148 Korninger, C., Stassen, J.M. and Collen, D. 1981 "Turnover of human extrinsic (tissue-type) plasminogen activator in rabbits" Thrombosis & Haemostasis 46(3), 658-661

149 Kounnas, M.Z., Argraves, W.S. and Strickland, D.K. 1992 "The 39-kDa receptor-associated protein

interacts with two members of the low density lipoprotein receptor family, 2-macroglobulin receptor and glycoprotein 330" Journal of Biological Chemistry 267(29), 21162-21166

175 150 Kounnas, M.Z., Chappell, D.A., Strickland, D.K. and Argraves, W.S. 1993a "Glycoprotein 330, a member of the low density lipoprotein receptor family, binds lipoprotein lipase in vitro" Journal of Biological Chemistry 268(19), 14176-14181

151 Kounnas, M.Z., Henkin, J., Argraves, W.S. and Strickland, D.K. 1993b "Low density lipoprotein

receptor-related protein/2-macroglobulin receptor mediates cellular uptake of pro-urokinase" Journal of Biological Chemistry 268(29), 21862-21867

152 Kounnas, M.Z., Stefansson, S., Loukinova, E., Argraves, K.M., Strickland, D.K. and Argraves, W.S.

1994 "An overview of the structure and function of glycoprotein 330, a receptor related to the 2- macroglobulin receptor" Annals of the New York Academy of Sciences 737, 114-123

153 Kounnas, M.Z., Moir, R.D., Rebeck, G.W., Bush, A.I., Argraves, W.S., Tanzi, R.E., Hyman, B.T. and Strickland, D.K. 1995 "LDL receptor-related protein, a multifunctional ApoE receptor, binds secreted -amyloid precursor protein and mediates its degradation" Cell 82(2), 331-340

154 Kounnas, M.Z., Church, F.C., Argraves, W.S. and Strickland, D.K. 1996 "Cellular internalization

and degradation of antithrombin III-thrombin, heparin cofactor II-thrombin, and 1-antitrypsin- trypsin complexes is mediated by the low density lipoprotein receptor-related protein" Journal of Biological Chemistry 271(11), 6523-6529

155 Koutsioumpa, M., Hatziapostolou, M., Mikelis, C., Koolwijk, P. and Papadimitriou, E. 2009 "Aprotinin stimulates angiogenesis and human endothelial cell migration through the growth factor pleiotrophin and its receptor protein tyrosine phosphatase beta/zeta" Eur J Pharmacol 602(2-3), 245- 249

156 Krieger, M. and Herz, J. 1994 "Structures and functions of multiligand lipoprotein receptors: Macrophage scavenger receptor and LDL receptor-related protein (LRP)" Annual Reviews of Biochemistry 63, 601-637

157 Kristensen, T., Moestrup, S.K., Gliemann, J., Bendtsen, L., Sand, O. and Sottrup-Jensen, L. 1990

"Evidence that the newly cloned low-density-lipoprotein receptor related protein (LRP) is the 2- macroglobulin receptor" FEBS Letters 276(1,2), 151-155

158 Kruithof, E.K., Gudinchet, A. and Bachmann, F. 1988 "Plasminogen activator inhibitor 1 and plasminogen activator inhibitor 2 in various disease states" Thrombosis & Haemostasis 59(1), 7-12

159 Kruithof, E.K.O. Custom 5 1988 "Inhibitors of plasminogen activators" in Tissue-type plasminogen (t-PA): Physiological and Clinical Aspects Vol. I of 2 pp. 189-210 Ed. Pai-1_Review Pub CRC Press Inc., Boca Raton, Florida.

160 Kuiper, J., Otter, M., Voorschuur, A.H., Van Zonneveld, A.J., Rijken, D.C. and Van Berkel, T.J.C. 1995 "Characterization of the interaction of a complex of tissue-type plasminogen activator and plasminogen activator inhibitor type 1 with rat liver cells" Thrombosis & Haemostasis 74(5), 1298- 1304

161 Kunamneni, A., Abdelghani, T.T. and Ellaiah, P. 2007 "Streptokinase--the drug of choice for thrombolytic therapy" J Thromb Thrombolysis 23(1), 9-23

162 Kunamneni, A., Ravuri, B.D., Saisha, V., Ellaiah, P. and Prabhakhar, T. 2008 "Urokinase-a very popular cardiovascular agent" Recent Pat Cardiovasc Drug Discov 3(1), 45-58

163 Kütt, H., Herz, J. and Stanley, K.K. 1989 "Structure of the low-density lipoprotein receptor-related protein (LRP) promoter" Biochimica et Biophysica Acta 1009(3), 229-236

176 164 Kuusisto, J., Koivisto, K., Kervinen, K., Mykkanen, L., Helkala, E.L., Vanhanen, M., Hanninen, T., Pyorala, K., Kesaniemi, Y.A. and Riekkinen, P. 1994 "Association of apolipoprotein E phenotypes with late onset Alzheimer's disease: population based study" British Medical Journal 309(6955), 636-638

165 Laemmli, U.K. 1970 "Cleavage of structural proteins during the assembly of the head of bacteriophage T4" Nature 227(259), 680-685

166 Lamba, D., Bauer, M., Huber, R., Fischer, S., Rudolph, R., Kohnert, U. and Bode, W. 1996 "The 2.3 A crystal structure of the catalytic domain of recombinant two-chain human tissue-type plasminogen activator" Journal of Molecular Biology 258(1), 117-135

167 Langbein, S., Szakacs, O., Wilhelm, M., Sukosd, F., Weber, S., Jauch, A., Lopez Beltran, A., Alken, P., Kalble, T. and Kovacs, G. 2002 "Alteration of the LRP1B gene region is associated with high grade of urothelial cancer" Lab Invest 82(5), 639-643

168 Levine, S.J. 2008 "Molecular mechanisms of soluble cytokine receptor generation" Journal of Biological Chemistry 283(21), 14177-14181

169 Li, Y., Knisely, J.M., Lu, W., Mccormick, L.M., Wang, J., Henkin, J., Schwartz, A.L. and Bu, G. 2002 "Low density lipoprotein (LDL) receptor-related protein 1B impairs urokinase receptor regeneration on the cell surface and inhibits cell migration" Journal of Biological Chemistry 277(44), 42366-42371

170 Li, Y., Lu, W. and Bu, G. 2005 "Striking differences of LDL receptor-related protein 1B expression in mouse and human" Biochemical & Biophysical Research Communications 333(3), 868-873

171 Li, Y.H., Marzolo, M.P., Van Kerkhof, P., Strous, G.J. and Bu, G.J. 2000 "The YXXL motif, but not the two NPXY motifs, serves as the dominant endocytosis signal for low density lipoprotein receptor-related protein" Journal of Biological Chemistry 275(22), 17187-17194

172 Liang, B.A., Lew, R. and Zivin, J.A. 2008 "Review of tissue plasminogen activator, ischemic stroke, and potential legal issues" Arch Neurol 65(11), 1429-1433

173 Liao, W. and Florén, C.-H. 1992 "Endotoxins inhibit endocytotic catabolism of low-density lipoproteins in Hep G2 cells" Hepatology 16(1), 224-231

174 Lillis, A.P., Van Duyn, L.B., Murphy-Ullrich, J.E. and Strickland, D.K. 2008 "LDL receptor-related protein 1: unique tissue-specific functions revealed by selective gene knockout studies" Physiol Rev 88(3), 887-918

175 Liu, C.X., Musco, S., Lisitsina, N.M., Yaklichkin, S.Y. and Lisitsyn, N.A. 2000 "Genomic organization of a new candidate tumor suppressor gene, LRP1B" Genomics 69(2), 271-274

176 Liu, C.X., Li, Y., Obermoeller-Mccormick, L.M., Schwartz, A.L. and Bu, G. 2001 "The putative tumor suppressor LRP1B, a novel member of the low density lipoprotein (LDL) receptor family, exhibits both overlapping and distinct properties with the LDL receptor-related protein" Journal of Biological Chemistry 276(31), 28889-28896

177 Liu, C.X., Ranganathan, S., Robinson, S. and Strickland, D.K. 2007 "gamma-Secretase-mediated release of the low density lipoprotein receptor-related protein 1B intracellular domain suppresses anchorage-independent growth of neuroglioma cells" Journal of Biological Chemistry 282(10), 7504-7511

178 Llorente-Cortes, V., Otero-Vinas, M., Sanchez, S., Rodriguez, C. and Badimon, L. 2002 "Low- density lipoprotein upregulates low-density lipoprotein receptor-related protein expression in vascular smooth muscle cells: possible involvement of sterol regulatory element binding protein-2- dependent mechanism" Circulation 106(24), 3104-3110

177 179 Low, D.A., Baker, J.B., Koonce, W.C. and Cunningham, D.D. 1981 "Released protease-nexin regulates cellular binding, internalization, and degradation of serine proteases" Proceedings of the National Academy of Sciences of the United States of America 78(4), 2340-2344

180 Lukes, A., Mun-Bryce, S., Lukes, M. and Rosenberg, G.A. 1999 "Extracellular matrix degradation by metalloproteinases and central nervous system diseases" Mol Neurobiol 19(3), 267-284

181 Lund, L.R., Green, K.A., Stoop, A.A., Ploug, M., Almholt, K., Lilla, J., Nielsen, B.S., Christensen, I.J., Craik, C.S., Werb, Z., Dano, K. and Romer, J. 2006 "Plasminogen activation independent of uPA and tPA maintains wound healing in gene-deficient mice" EMBO Journal 25(12), 2686-2697

182 Maciag, T., Cerundolo, J., Ilsley, S., Kelley, P.R. and Forand, R. 1979 "An endothelial cell growth factor from bovine hypothalamus: identification and partial characterisation" Proceedings of the National Academy of Sciences of the United States of America 76(11), 5674-5678

183 Malathi, P., Preiser, H., Fairclough, P., Mallett, P. and Crane, R.K. 1979 "A rapid method for the isolation of kidney brush border membranes" Biochimica et Biophysica Acta 554(1), 259-263

184 Marschang, P., Brich, J., Weeber, E.J., Sweatt, J.D., Shelton, J.M., Richardson, J.A., Hammer, R.E. and Herz, J. 2004 "Normal development and fertility of knockout mice lacking the tumor suppressor gene LRP1b suggest functional compensation by LRP1" Molecular & Cellular Biology 24(9), 3782- 3793

185 Martins, I.J., Hone, E., Chi, C., Seydel, U., Martins, R.N. and Redgrave, T.G. 2000 "Relative roles of LDLr and LRP in the metabolism of chylomicron remnants in genetically manipulated mice" Journal of Lipid Research 41(2), 205-213

186 Matheson, N.R., Van Halbeek, H. and Travis, J. 1991 "Evidence for a tetrahedral intermediate complex during serpin-protease interactions" Journal of Biological Chemistry 266(21), 13489-13491

187 Mattsson, C. Custom 5 1988 "Turnover and clearance of t-PA" in Tissue-type plasminogen (t-PA): Physiological and Clinical Aspects Vol. II of 2 pp. 37-48 Ed. Tpa_Catabolism, Tpa-Receptors Pub CRC Press Inc., Boca Raton, Florida.

188 May, P., Woldt, E., Matz, R.L. and Boucher, P. 2007 "The LDL receptor-related protein (LRP) family: an old family of proteins with new physiological functions" Ann Med 39(3), 219-228

189 Mclean, J.W., Tomlinson, J.E., Kuang, W.J., Eaton, D.L., Chen, E.Y., Fless, G.M., Scanu, A.M. and Lawn, R.M. 1987 "cDNA sequence of human apolipoprotein(a) is homologous to plasminogen" Nature 330(6144), 132-137

190 Mcmahon, B. and Kwaan, H.C. 2008 "The plasminogen activator system and cancer" Pathophysiol Haemost Thromb 36(3-4), 184-194

191 Mcmullen, B.A. and Fujikawa, K. 1985 "Amino acid sequence of the heavy chain of human alpha- factor XIIa (activated Hageman factor)" Journal of Biological Chemistry 260(9), 5328-5341

192 Medcalf, R.L. and Stasinopoulos, S.J. 2005 "The undecided serpin. The ins and outs of plasminogen activator inhibitor type 2" Febs J 272(19), 4858-4867

193 Medh, J.D., Fry, G.L., Bowen, S.L., Pladet, M.W., Strickland, D.K. and Chappell, D.A. 1995 "The 39-kDa receptor-associated protein modulates lipoprotein catabolism by binding to LDL receptors" Journal of Biological Chemistry 270(2), 536-540

194 Mehta, J.S. and Adams, G.G.W. 2000 "Recombinant tissue plasminogen activator following paediatric cataract surgery" British Journal of Ophthalmology 84(9), 983-986

178 195 Melchior, R., Quigley, J.P. and Armstrong, P.B. 1995 "2-macroglobulin-mediated clearance of proteases from the plasma of the American horseshoe crab, Limulus polyphemus" Journal of Biological Chemistry 270(22), 13496-13502

196 Minter, A.J., Dawes, J. and Chesterman, C.N. 1992 "Effects of heparin and endothelial cell growth supplement on haemostatic functions of vascular endothelium" Thrombosis & Haemostasis 67(6), 718-723

197 Moestrup, S.K. and Gliemann, J. 1989 "Purification of the rat hepatic -macroglobulin receptor as an approximately 440-kDa single chain protein" Journal of Biological Chemistry 264(26), 15574- 15577

198 Moestrup, S.K., Kaltoft, K., Sottrup-Jensen, L. and Gliemann, J. 1990 "The human 2-macroglobulin receptor contains high affinity calcium binding sites important for receptor conformation and ligand recognition" Journal of Biological Chemistry 265(21), 12623-12628

199 Moestrup, S.K., Holtet, T.L., Etzerodt, M., Thogersen, H.C., Nykjaer, A., Andreasen, P.A.,

Rasmussen, H.H., Sottrup-Jensen, L. and Gliemann, J. 1993 "2-macroglobulin-proteinase complexes, plasminogen activator inhibitor type-1-plasminogen activator complexes, and receptor-

associated protein bind to a region of the 2-macroglobulin receptor containing a cluster of eight complement-type repeats" Journal of Biological Chemistry 268(18), 13691-13696

200 Moestrup, S.K. 1994 "The 2-macroglobulin receptor and epithelial glycoprotein-330: two giant receptors mediating endocytosis of multiple ligands" Biochimica et Biophysica Acta 1197(2), 197- 213

201 Montuori, N. and Ragno, P. 2009 "Multiple activities of a multifaceted receptor: roles of cleaved and soluble uPAR" Front Biosci 14, 2494-2503

202 Morton, P.A., Owensby, D.A., Sobel, B.E. and Schwartz, A.L. 1989 "Catabolism of tissue-type plasminogen activator by the human hepatoma cell line HepG2: Modulation by plasminogen activator inhibitor type 1" Journal of Biological Chemistry 264(13), 7228-7235

203 Morton, P.A., Owensby, D.A., Wun, T.-C., Billadello, J.J. and Schwartz, A.L. 1990 "Identification of determinants involved in binding of tissue-type plasminogen activator-plasminogen activator inhibitor type 1 complexes to HepG2 cells" Journal of Biological Chemistry 265(24), 14093-14099

204 Muller-Newen, G., Kohne, C. and Heinrich, P.C. 1996 "Soluble receptors for cytokines and growth factors" International Archives of Allergy & Immunology 111(2), 99-106

205 Müller-Newen, G., Köhne, C., Keul, R., Hemmann, U., Müller-Esterl, W., Wijdenes, J., Brakenhoff, J.P.J., Hart, M.H.L. and Heinrich, P.C. 1996 "Purification and characterization of the soluble interleukin-6 receptor from human plasma and identification of an isoform generated through alternative splicing" European Journal of Biochemistry 236(3), 837-842

206 Myklebost, O., Arheden, K., Rogne, S., Geurts Van Kessel, A., Mandahl, N., Herz, J., Stanley, K., Heim, S. and Mitelman, F. 1989 "The gene for the human putative apoE receptor is on chromosome 12 in the segment q13-14" Genomics 5(1), 65-69

207 Nair, S.A. and Cooperman, B.S. 1998 "Antichymotrypsin interaction with chymotrypsin. Reactions following encounter complex formation" Journal of Biological Chemistry 273(28), 17459-17462

208 Nakamura, T., Nishizawa, T., Hagiya, M., Seki, T., Shimonishi, M., Sugimura, A., Tashiro, K. and Shimizu, S. 1989 "Molecular cloning and expression of human hepatocyte growth factor" Nature 342(6248), 440-443

179 209 Narita, M., Bu, G., Herz, J. and Schwartz, A.L. 1995 "Two receptor systems are involved in the plasma clearance of tissue-type plasminogen activator (t-PA) in Vivo" Journal of Clinical Investigation 96(2), 1164-1168

210 Narita, M., Bu, G., Holtzman, D.M. and Schwartz, A.L. 1997a "The low-density lipoprotein receptor-related protein, a multifunctional apolipoprotein E receptor, modulates hippocampal neurite development" Journal of Neurochemistry 68(2), 587-595

211 Narita, M., Holtzman, D.M., Schwartz, A.L. and Bu, G. 1997b "2-macroglobulin complexes with and mediates the endocytosis of -amyloid peptide via cell surface low-density lipoprotein receptor- related protein" Journal of Neurochemistry 69(5), 1904-1911

212 Neville Jr, D.M. 1971 "Molecular weight determination of protein-dodecyl sulfate complexes by gel electrophoresis in a discontinuous buffer system" Journal of Biological Chemistry 246(20), 6328- 6334

213 Nguyen, G., Self, S.J., Camani, C. and Kruithof, E.K.O. 1992 "Characterization of the binding of plasminogen activators to plasma membranes from human liver" Biochemical Journal 287(Pt 3), 911-915

214 Nielsen, M.S., Nykjaer, A., Warshawsky, I., Schwartz, A.L. and Gliemann, J. 1995 "Analysis of

ligand binding to the 2-macroglobulin receptor/low density lipoprotein receptor-related protein. Evidence that lipoprotein lipase and the carboxyl-terminal domain of the receptor-associated protein bind to the same site" Journal of Biological Chemistry 270(40), 23713-23719

215 Nilsson, S., Einarsson, M., Ekvarn, S., Haggroth, L. and Mattsson, C. 1985 "Turnover of tissue plasminogen activator in normal and hepatectomized rabbits" Thrombosis Research 39(4), 511-521

216 Novokhatny, V. 2008 "Structure and activity of plasmin and other direct thrombolytic agents" Thrombosis Research 122 Suppl 3, S3-8

217 Ny, T., Elgh, F. and Lund, B. 1984 "The structure of the human tissue-type plasminogen activator gene: correlation of intron and exon structures to functional and structural domains" Proceedings of the National Academy of Sciences of the United States of America 81(17), 5355-5359

218 Nykjær, A., Petersen, C.M., Moller, B., Jensen, P.H., Moestrup, S.K., Holtet, T.L., Etzerodt, M.,

Thogersen, H.C., Munch, M., Andreasen, P.A. and Gliemann, J. 1992 "Purified 2-macroglobulin receptor/LDL receptor-related protein binds urokinaseplasminogen activator inhibitor type-1

complex: Evidence that the 2-macroglobulin receptor mediates cellular degradation of urokinase receptor-bound complexes" Journal of Biological Chemistry 267(21), 14543-14546

219 Nykjær, A., Kjøller, L., Cohen, R.L., Lawrence, D.A., Garni-Wagner, B.A., Todd Iii, R.F., Van Zonneveld, A.-J., Gliemann, J. and Andreasen, P.A. 1994 "Regions involved in binding of urokinase-

type-1 inhibitor complex and pro-urokinase to the endocytic 2-macroglobulin receptor/low density lipoprotein receptor-related protein. Evidence that the urokinase receptor protects prourokinase against binding to the endocytic receptor" Journal of Biological Chemistry 269(41), 25668-25676

220 Okajima, T., Matsuura, A. and Matsuda, T. 2008 "Biological functions of glycosyltransferase genes involved in O-fucose glycan synthesis" J Biochem 144(1), 1-6

221 Olson, D., Pollanen, J., Hoyer-Hansen, G., Ronne, E., Sakaguchi, K., Wun, T.-C., Appella, E., Dano, K. and Blasi, F. 1992 "Internalization of the urokinase-plasminogen activator inhibitor type-1 complex is mediated by the urokinase receptor" Journal of Biological Chemistry 267(13), 9129-9133

222 Ornstein, L. 1964 "Disc electrophoresis, I Background and theory" Annals of the New York Academy of Sciences 121, 321-349

180 223 Orth, K., Madison, E.L., Gething, M.-J., Sambrook, J.F. and Herz, J. 1992 "Complexes of tissue-type plasminogen activator and its serpin inhibitor plasminogen-activator inhibitor type 1 are internalized

by means of the low density lipoprotein receptor-related protein/2-macroglobulin receptor" Proceedings of the National Academy of Sciences of the United States of America 89(16), 7422-7426

224 Osterwalder, T., Cinelli, P., Baici, A., Pennella, A., Krueger, S.R., Schrimpf, S.P., Meins, M. and Sonderegger, P. 1998 "The axonally secreted serine proteinase inhibitor, neuroserpin, inhibits plasminogen activators and plasmin but not thrombin" Journal of Biological Chemistry 273(4), 2312-2321

225 Otter, M., Van Berkel, T.J.C. and Rijken, D.C. 1989 "Binding and degradation of tissue-type plasminogen activator by the human hepatoma cell line Hep G2" Thrombosis & Haemostasis 62(2), 667-672

226 Otter, M., Kuiper, J., Bos, R., Rijken, D.C. and Van Berkel, T.J.C. 1992a "Characterization of the interaction both in vitro and in vivo of tissue-type plasminogen activator (t-PA) with rat liver cells. Effects of monoclonal antibodies to t-PA" Biochemical Journal 284(Pt 2), 545-550

227 Otter, M., Kuiper, J., Van Berkel, T.J.C. and Rijkin, D.C. 1992b "Mechanisms of tissue-type plasminogen activator (tPA) clearance by the liver" Annals of the New York Academy of Sciences 667, 431-442

228 Otter, M., Zockova, P., Kuiper, J., Van Berkel, T.J.C., Barrett-Bergshoeff, M.M. and Rijken, D.C. 1992c "Isolation and characterisation of the mannose receptor from human liver potentially involved in the plasma clearance of tissue-type plasminogen activator" Hepatology 16(1), 54-59

229 Ouriel, K. 1999 "Urokinase and the US Food and Drug Administration" J Vasc Surg 30(5), 957-958

230 Owensby, D.A., Sobel, B.E. and Schwartz, A.L. 1988 "Receptor-mediated endocytosis of tissue-type plasminogen activator by the human hepatoma cell line Hep G2" Journal of Biological Chemistry 263(22), 10587-10594

231 Owensby, D.A., Morton, P.A. and Schwartz, A.L. 1989 "Interactions between tissue-type plasminogen activator and extracellular matrix-associated plasminogen inhibitor type 1 in the human hepatoma cell line Hep G2" Journal of Biological Chemistry 264(30), 18180-18187

232 Owensby, D.A., Morton, P.A., Wun, T.-C. and Schwartz, A.L. 1991 "Binding of plasminogen activator inhibitor type-1 to extracellular matrix of Hep G2 cells: Evidence that the binding protein is vitronectin" Journal of Biological Chemistry 266(7), 4334-4340

233 Pannell, R. and Gurewich, V. 1987 "Activation of plasminogen by single-chain urokinase or by two- chain urokinase--a demonstration that single-chain urokinase has a low catalytic activity (pro- urokinase)" Blood 69(1), 22-26

234 Pastrana, D.V., Hanson, A.J., Knisely, J., Bu, G. and Fitzgerald, D.J. 2005 "LRP 1 B functions as a receptor for Pseudomonas exotoxin" Biochimica et Biophysica Acta 1741(3), 234-239

235 Patston, P.A., Medcalf, R.L., Kourteva, Y. and Schapira, M. 1993 "C1-inhibitor-serine proteinase complexes and the biosynthesis of C1-inhibitor by Hep G2 and U 937 cells" Blood 82(11), 3371- 3379

236 Patthy, L. 1985 "Evolution of the proteases of blood coagulation and fibrinolysis by assembly from modules" Cell 41(3), 657-663

237 Pennica, D., Holmes, W.E., Kohr, W.J., Harkins, R.N., Vehar, G.A., Ward, C.A., Bennett, W.F., Yelverton, E., Seeburg, P.H., Heyneker, H.L., Goeddel, D.V. and Collen, D. 1983 "Cloning and expression of human tissue-type plasminogen activator cDNA in E. coli" Nature 301(5897), 214-221

181 238 Perler, B. 2005 "Thrombolytic therapies: the current state of affairs" J Endovasc Ther 12(2), 224-232

239 Perlmutter, D.H., Glover, G.I., Rivetna, M., Schasteen, C.S. and Fallon, R.J. 1990 "Identification of a serpin-enzyme complex receptor on human hepatoma cells and human monocytes" Proceedings of the National Academy of Sciences of the United States of America 87(10), 3753-3757

240 Peters, M., Jacobs, S., Ehlers, M., Vollmer, P., Müllberg, J., Wolf, E., Brem, G., Meyer Zum Büschenfelde, K.-H. and Rose-John, S. 1996 "The function of the soluble interleukin 6 (IL-6) receptor in vivo: Sensitization of human soluble IL-6 receptor transgenic mice towards IL-6 and prolongation of the plasma half-life of IL-6" Journal of Experimental Medicine 183(4), 1399-1406

241 Pfeiffer, G., Schmidt, M., Strube, K.H. and Geyer, R. 1989 "Carbohydrate structure of recombinant human uterine tissue plasminogen activator expressed in mouse epithelial cells" European Journal of Biochemistry 186(1-2), 273-286

242 Plesner, T., Ralfkiaer, E., Wittrup, M., Johnsen, H., Pyke, C., Pedersen, T.L., Hansen, N.E. and Dano, K. 1994 "Expression of the receptor for urokinase-type plasminogen activator in normal and neoplastic blood cells and hematopoietic tissue" American Journal of Clinical Pathology 102(6), 835-841

243 Pollanen, J., Hedman, K., Nielsen, L.S., Dano, K. and Vaheri, A. 1988 "Ultrastructural localization of plasma membrane-associated urokinase-type plasminogen activator at focal contacts" Journal of Cell Biology 106(1), 87-95

244 Poller, W., Merklein, F., Schneider-Rasp, S., Haack, A., Fechner, H., Wang, H., Anagnostopoulos, I. and Weidinger, S. 1999 "Molecular characterisation of the defective alpha 1-antitrypsin alleles PI Mwurzburg (Pro369Ser), Mheerlen (Pro369Leu), and Q0lisbon (Thr68Ile)" European Journal of Human Genetics 7(3), 321-331

245 Potempa, J., Korzus, E. and Travis, J. 1994 "The serpin superfamily of proteinase inhibitors: Structure, function, and regulation" Journal of Biological Chemistry 269(23), 15957-15960

246 Qiu, Z., Strickland, D.K., Hyman, B.T. and Rebeck, G.W. 2002 "alpha 2-Macroglobulin exposure reduces calcium responses to N-methyl-D-aspartate via low density lipoprotein receptor-related protein in cultured hippocampal neurons" Journal of Biological Chemistry 277(17), 14458-14466

247 Quinn, K.A., Grimsley, P.G., Dai, Y.-P., Tapner, M., Chesterman, C.N. and Owensby, D.A. 1997 "Soluble low density lipoprotein receptor-related protein (LRP) circulates in Human plasma" Journal of Biological Chemistry 272(38), 23946-23951

248 Quinn, K.A., Pye, V.J., Dai, Y.P., Chesterman, C.N. and Owensby, D.A. 1999 "Characterization of the soluble form of the low density lipoprotein receptor-related protein (LRP)" Experimental Cell Research 251(2), 433-441

249 Radcliffe, R. and Heinze, T. 1978 "Isolation of plasminogen activator from human plasma by chromatography on lysine-sepharose" Archives of Biochemistry & Biophysics 189(1), 185-194

250 Radcliffe, R. 1983 "A critical role of lysine residues in the stimulation of tissue plasminogen activator by denatured proteins and fibrin clots" Biochimica et Biophysica Acta 743(3), 422-430

251 Rajput, B., Degen, S.F., Reich, E., Waller, E.K., Axelrod, J., Eddy, R.L. and Shows, T.B. 1985 "Chromosomal locations of human tissue plasminogen activator and urokinase genes" Science 230(4726), 672-674

252 Ramakrishnan, V., Sinicropi, D.V., Dere, R., Darbonne, W.C., Bechtol, K.B. and Baker, J.B. 1990 "Interaction of wild-type and catalytically inactive mutant forms of tissue-type plasminogen activator with human umbilical vein endothelial cell monolayers" Journal of Biological Chemistry 265(5), 2755-2762

182 253 Rånby, M. and Brändström, A. Custom 5 1988 "Control of t-PA-mediated fibrinolysis" in Tissue- type plasminogen (t-PA): Physiological and Clinical Aspects Vol. I of 2 pp. 211-224 Ed. Tpa_Activity_Review, Tpa_Regulation, Fibrinolysis Pub CRC Press Inc., Boca Raton, Florida.

254 Rawlings, N.D. and Barrett, A.J. 1993 "Evolutionary families of peptidases" Biochemical Journal 290(Pt 1), 205-218

255 Rawlings, N.D. and Barrett, A.J. 1994 "Families of serine peptidases" Methods in Enzymology 244, 19-61

256 Raymond, S. and Weintraub, L. 1959 "Acrylamide gel as a suporting medium for zone electrophoresis" Science 30(3378), 711-711

257 Rebeck, G.W., Harr, S.D., Strickland, D.K. and Hyman, B.T. 1995 "Multiple, diverse senile plaque-

associated proteins are ligands of an apolipoprotein E receptor, the 2-macroglobulin receptor/low- density-lipoprotein receptor-related protein" Annals of Neurology 37(2), 211-217

258 Rebeck, G.W., Moir, R.D., Mui, S., Strickland, D.K., Tanzi, R.E. and Hyman, B.T. 2001 "Association of membrane-bound amyloid precursor protein APP with the apolipoprotein E receptor LRP" Molecular Brain Research 87(2), 238-245

259 Remold-O'donnell, E. 1993 "The ovalbumin family of serpin proteins" FEBS Letters 315(2), 105-108

260 Riccio, A., Grimaldi, G., Verde, P., Sebastio, G., Boast, S. and Blasi, F. 1985 "The human urokinase- plasminogen activator gene and its promoter" Nucleic Acids Research 13(8), 2759-2771

261 Riddell, D.R., Siripurapu, V., Vinogradov, D.V., Gliemann, J. and Owen, J.S. 1998 "Blood platelets do not contain the low-density receptor-related protein (LRP)" Biochemical Society Transactions 26(3), S244

262 Riddell, D.R., Vinogradiv, D.V., Stannard, A.K., Chadwick, N. and Owen, J.S. 1999 "Identification and characterization of LRP8 (apoER2) in human blood platelets" Journal of Lipid Research 40(10), 1925-1930

263 Rijken, D.C. and Lijnen, H.R. 2009 "New insights into the molecular mechanisms of the fibrinolytic system" J Thromb Haemost 7(1), 4-13

264 Roche, P.A., Moncino, M.D. and Pizzo, S.V. 1989 "Independent analysis of bait region cleavage dependent and thiolester bond cleavage dependent conformational changes by cross-linking of a2- macroglobulin with cis-dichlorodiammineplatinum(II) and dithiobis(succinimidyl propiate)" Biochemistry 28(19), 7629-7636

265 Rogers, J. 1985 "Exon shuffling and intron insertion in serine protease genes [news]" Nature 315(6019), 458-459

266 Rohlmann, A., Gotthardt, M., Willnow, T.E., Hammer, R.H. and Herz, J. 1996 "Sustained somatic gene inactivation by viral tranfer of Cre recombinase" Nature Biotechnology 14(11), 1562-1565

267 Roldan, A.L., Cubellis, M.V., Masucci, M.T., Behrendt, N., Lund, L.R., Dano, K., Appella, E. and Blasi, F. 1990 "Cloning and expression of the receptor for human urokinase plasminogen activator, a central molecule in cell surface, plasmin dependent proteolysis. [erratum appears in EMBO J 1990 May;9(5):1674]" EMBO Journal 9(2), 467-474

268 Rose-John, S. and Heinrich, P.C. 1994 "Soluble receptors for cytokines and growth factors: generation and biological function" Biochemical Journal 300(Pt 2), 281-290

183 269 Rossignol, P., Angles-Cano, E. and Lijnen, H.R. 2006 "Plasminogen activator inhibitor-1 impairs plasminogen activation-mediated vascular smooth muscle cell apoptosis" Thrombosis & Haemostasis 96(5), 665-670

270 Roychoudhury, P.K., Khaparde, S.S., Mattiasson, B. and Kumar, A. 2006 "Synthesis, regulation and production of urokinase using mammalian cell culture: a comprehensive review" Biotechnol Adv 24(5), 514-528

271 Russell, M.E., Quertermous, T., Declerck, P.J., Collen, D., Haber, E. and Homey, C.J. 1990 "Binding of tissue-type plasminogen activator with human endothelial cell monolayers: Characterization of the high affinity interaction with plasminogen activator inhibitor-1" Journal of Biological Chemistry 265(5), 2569-2575

272 Sa, S.J., Rhee, H.H., Cheong, H.T., Yang, B.K. and Park, C.K. 2006 "Effects of plasmin on sperm- oocyte interactions during in vitro fertilization in the pig" Anim Reprod Sci 95(3-4), 273-282

273 Sagare, A., Deane, R., Bell, R.D., Johnson, B., Hamm, K., Pendu, R., Marky, A., Lenting, P.J., Wu, Z., Zarcone, T., Goate, A., Mayo, K., Perlmutter, D., Coma, M., Zhong, Z. and Zlokovic, B.V. 2007 "Clearance of amyloid-beta by circulating lipoprotein receptors" Nature Medicine 13(9), 1029-1031

274 Saito, A., Pietromonaco, S., Loo, A.K. and Farquhar, M.G. 1994 "Complete cloning and sequencing of rat gp330/"megalin," a distinctive member of the low density lipoprotein receptor gene family" Proceedings of the National Academy of Sciences of the United States of America 91(21), 9725-9729

275 Sanchez, L., Alvarez, V., Gonzalez, P., Gonzalez, I., Alvarez, R. and Coto, E. 2001 "Variation in the LRP-associated protein gene (LRPAP1) is associated with late-onset Alzheimer disease" American Journal of Medical Genetics 105(1), 76-78

276 Schlossman, S.F., Boumsell, L., Gilks, W., Harlan, J.M., Kishimoto, T., Morimoto, C., Ritz, J., Shaw, S., Silverstein, R.L., Springer, T.A., Tedder, T.F. and Todd, R.F. 1994 "CD antigens 1993" Blood 83(4), 879-880

277 Schmidt, A.E., Sun, M.F., Ogawa, T., Bajaj, S.P. and Gailani, D. 2008 "Functional role of residue 193 (chymotrypsin numbering) in serine proteases: influence of side chain length and beta-branching on the catalytic activity of blood coagulation factor XIa" Biochemistry 47(5), 1326-1335

278 Schneider, D.J., Chen, Y. and Sobel, B.E. 2008 "The effect of plasminogen activator inhibitor type 1 on apoptosis" Thrombosis & Haemostasis 100(6), 1037-1040

279 Schulze, A.J., Huber, R., Bode, W. and Engh, R.A. 1994 "Structural aspects of serpin inhibition" FEBS Letters 344(2-3), 117-124

280 Schwartz, A.L. and Rup, D. 1983 "Biosynthesis of the Human asialoglycoprotein receptor" Journal of Biological Chemistry 258(18), 11249-11255

281 Seo, T., Wang, H.-C., Feldman, S.R. and St Clair, R.W. 1997 "Characterization of 2-macroglobulin receptor low density lipoprotein receptor-related protein (2-MR/LRP) in White Carneau pigeon peritoneal macrophages: its role in lipoprotein metabolism" Biochimica et Biophysica Acta 1344(2), 171-188

282 Shaw, S. 1994 "Leukocyte Differentiation Antigen Database [computer database] Version 1.10 (Fifth International workshop on Leukocyte Differentiation Antigens. A database of the results is available from Stephen Shaw, National Institutes of Health, or can be dowloaded from ftp://ncbi.nlm.nih.gov/pub/shaw/LDAD/ldad1.zipNIH , or obtained from Flow Cytometry Standards, P.O.Box 194344, San Juan Puerto Rico 00919-4344, Tel: 809-753-9341, FAX: 809-758-3267)."

184 283 Shieh, B.-H., Potempa, J. and Travis, J. 1989 "The use of 2-antiplasmin as a model for the demonstration of complex reversibility in serpins" Journal of Biological Chemistry 264(23), 13420- 13423

284 Shih, G.C. and Hajjar, K.A. 1993 "Plasminogen and plasminogen activator assembly on the human endothelial cell" Experimental Biology & Medicine 202(3), 258-264

285 Siebert, P.D. and Fong, K. 1990 "Variant tissue-type plasminogen activator (PLAT) cDNA obtained from human endothelial cells" Nucleic Acids Research 18(4), 1086

286 Simpson, D., Siddiqui, M.A., Scott, L.J. and Hilleman, D.E. 2006 "Reteplase: a review of its use in the management of thrombotic occlusive disorders" Am J Cardiovasc Drugs 6(4), 265-285

287 Smedsrod, B. and Einarsson, M. 1990 "Clearance of tissue plasminogen activator by mannose and galactose receptors in the liver" Thrombosis & Haemostasis 63(1), 60-66

288 Smith, P.K., Krohn, R.I., Hermanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olson, B.J. and Klenk, D.C. 1985 "Measurement of protein using bicinchoninic acid [published erratum appears in Anal Biochem 1987 May 15;163(1):279]" Analytical Biochemistry 150(1), 76-85

289 Sobel, B.E. Custom 5 1988 "Thrombolytic therapy with tPA" in Tissue-type plasminogen (t-PA): Physiological and Clinical Aspects Vol. II of 2 pp. 109-128 Ed. Tpa_Clinical Pub CRC Press Inc., Boca Raton, Florida.

290 Sottrup-Jensen, L., Zajdel, M., Claeys, H., Petersen, T.E. and Magnusson, S. 1975 "Amino-acid sequence of activation cleavage site of plasminogen: Homology with "pro" part of prothrombin" Proc Nat Acad Sci 72(7), 2577-2581

291 Sottrup-Jensen, L., Petersen, T.E. and Magnusson, S. 1980 "A thiol-ester in 2-macroglobulin cleaved during proteinase complex formation" FEBS Letters 121(2), 275-279

292 Stang, E., Roos, N., Schluter, M., Berg, T. and Krause, J. 1992 "Evidence for carbohydrate- independent endocytosis of tissue-type plasminogen activator by liver cells" Biochemical Journal 285, 799-804

293 Stefansson, S., Lawrence, D.A. and Argraves, W.S. 1996 "Plasminogen activator inhibitor-1 and vitronectin promote the cellular clearance of thrombin by low density lipoprotein receptor-related proteins 1 and 2" Journal of Biological Chemistry 271(14), 8215-8220

294 Stefansson, S., Muhammad, S., Cheng, X.F., Battey, F.D., Strickland, D.K. and Lawrence, D.A. 1998 "Plasminogen activator inhibitor-1 contains a cryptic high affinity binding site for the low density lipoprotein receptor-related protein" Journal of Biological Chemistry 273(11), 6358-6366

295 Stein, E., Mcmahon, B., Kwaan, H., Altman, J.K., Frankfurt, O. and Tallman, M.S. 2009 "The coagulopathy of acute promyelocytic leukaemia revisited" Best Pract Res Clin Haematol 22(1), 153- 163

296 Stein, P.E., Tewkesbury, D.A. and Carrell, R.W. 1989 "Ovalbumin and angiotensinogen lack serpin S-R conformational change" Biochemical Journal 262(1), 103-107

297 Stein, P.E., Leslie, A.G., Finch, J.T., Turnell, W.G., Mclaughlin, P.J. and Carrell, R.W. 1990 "Crystal structure of ovalbumin as a model for the reactive centre of serpins" Nature 347(6288), 99- 102

298 Stief, T.W., Radtke, K.P. and Heimburger, N. 1987 "Inhibition of urokinase by protein C-inhibitor (PCI). Evidence for identity of PCI and plasminogen activator inhibitor 3" Biological Chemistry Hoppe-Seyler 368(10), 1427-1433

185 299 Storm, D., Herz, J., Trinder, P. and Loos, M. 1997 "C1 Inhibitor-C1s complexes are internalized and degraded by the low density lipoprotein receptor-related protein" Journal of Biological Chemistry 272(49), 31043-31050

300 Strassburger, W., Wollmer, A., Pitts, J.E., Glover, I.D., Tickle, I.J., Blundell, T.L., Steffens, G.J., Gunzler, W.A., Otting, F. and Flohe, L. 1983 "Adaptation of plasminogen activator sequences to known protease structures" FEBS Letters 157(2), 219-223

301 Stratikos, E. and Gettins, P.G. 1998 "Mapping the serpin-proteinase complex using single cysteine

variants of 1-proteinase inhibitor Pittsburgh" Journal of Biological Chemistry 273(25), 15582- 15589

302 Stratikos, E. and Gettins, P.G. 1999 "Formation of the covalent serpin-proteinase complex involves translocation of the proteinase by more than 70 A and full insertion of the reactive center loop into beta-sheet A" Proceedings of the National Academy of Sciences of the United States of America 96(9), 4808-4813

303 Strickland, D.K., Ashcom, J.D., Williams, S., Burgess, W.H., Migliorini, M. and Argraves, W.S.

1990 "Sequence identity between the 2-macroglobulin receptor and low density lipoprotein receptor-related protein suggests that this molecule is a multifunctional receptor" Journal of Biological Chemistry 265(29), 17401-17404

304 Strickland, D.K., Ashcom, J.D., Williams, S., Battey, F., Behre, E., Mctigue, K., Battey, J.F. and

Argraves, W.S. 1991 "Primary structure of 2-macroglobulin receptor-associated protein. Human homologue of a Heymann nephritis antigen" Journal of Biological Chemistry 266(20), 13364-13369

305 Strickland, D.K., Kounnas, M.Z., Williams, S.E. and Argraves, W.S. 1994 "LDL receptor-related protein (LRP): a multiligand receptor" Fibrinolysis 8(Suppl 1), 204-215

306 Strickland, D.K., Kounnas, M.Z. and Argraves, W.S. 1995 "LDL receptor-related protein: a multiligand receptor for lipoprotein and proteinase catabolism" FASEB Journal 9(10), 890-898

307 Stump, D.C., Lijnen, H.R. and Collen, D. 1986 "Purification and characterization of a novel low molecular weight form of single-chain urokinase-type plasminogen activator" Journal of Biological Chemistry 261(36), 17120-17126

308 Summers, D.F., Maizel, J.V. and Darnell Jr, J.E. 1965 "Evidence for virus-specific noncapsid proteins in polyvirus-infected HeLa cells" Proceedings of the National Academy of Sciences of the United States of America 54(2), 505-513

309 Suzuki, K., Deyashiki, Y., Nishioka, J., Kurachi, K., Akira, M., Yamamoto, S. and Hashimoto, S. 1987 "Characterization of a cDNA for human protein C inibitor: a new member of the plasma serine protease inhibitor superfamily" Journal of Biological Chemistry 262(2), 611-616

310 Tolchinsky, S., Yuk, M.H., Ayalon, M., Lodish, H.F. and Lederkremer, G.Z. 1996 "Membrane- bound versus secreted forms of human asialoglycoprotein receptor subunits. Role of a juxtamembrane pentapeptide" Journal of Biological Chemistry 271(24), 14496-14503

311 Tollefsen, D.M. 1995 "Insights into the mechanisms of action of Heparin Cofactor II" Thrombosis & Haemostasis 74(5), 1209-1214

312 Tolleshaug, H., Hobgood, K.K., Brown, M.S. and Goldstein, J.L. 1983 "The LDL receptor locus in familial hypercholesterolemia: multiple mutations disrupt transport and processing of a membrane receptor" Cell 32(3), 941-951

313 Toriseva, M. and Kahari, V.M. 2009 "Proteinases in cutaneous wound healing" Cell Mol Life Sci 66(2), 203-224

186 314 Traynor, K. 2002 "Urokinase to return to market" Am J Health Syst Pharm 59(23), 2278

315 Tripputi, P., Blasi, F., Verde, P., Cannizzaro, L.A., Emanuel, B.S. and Croce, C.M. 1985 "Human urokinase gene is located on the long arm of chromosome 10" Proceedings of the National Academy of Sciences of the United States of America 82(13), 4448-4452

316 Tripputi, P., Blasi, F., Ny, T., Emanuel, B.S., Letosfsky, J. and Croce, C.M. 1986 "Tissue-type plasminogen activator gene is on chromosome 8" Cytogenetics & Cell Genetics 42(1-2), 24-28

317 Trommsdorff, M., Gotthardt, M., Hiesberger, T., Shelton, J., Stockinger, W., Nimpf, J., Hammer, R.E., Richardson, J.A. and Herz, J. 1999 "Reeler/Disabled-like disruption of neuronal migration in knockout mice lacking the VLDL receptor and ApoE receptor 2" Cell 97(6), 689-701

318 Trommsdorff, R., Borg, J.P., Margolis, B. and Herz, J. 1998 "Interaction of cytosolic adaptor proteins with neuronal apolipoprotein E receptors and the amyloid precursor protein" Journal of Biological Chemistry 273(50), 33556-33560

319 Trotter, J. 2000 "WinMDI, Windows Multiple Document Interface Flow Cytometry application",

320 Tsunezumi, J., Yamamoto, K., Higashi, S. and Miyazaki, K. 2008 "Matrilysin (matrix metalloprotease-7) cleaves membrane-bound annexin II and enhances binding of tissue-type plasminogen activator to cancer cell surfaces" Febs J 275(19), 4810-4823

321 Turcasso, N.M. and Nappi, J.M. 2001 "Tenecteplase for treatment of acute myocardial infarction" Ann Pharmacother 35(10), 1233-1240

322 Tycko, B. and Maxfield, F. 1982 "Rapid acidification of endocytic vesicles containing 2- macroglobulin" Cell 28(3), 643-651

323 Ueshima, S. and Matsuo, O. 2006 "Development of new fibrinolytic agents" Curr Pharm Des 12(7), 849-857

324 Ulery, P.G., Beers, J., Mikhailenko, I., Tanzi, R.E., Rebeck, G.W., Hyman, B.T. and Strickland, D.K. 2000 "Modulation of beta-amyloid precursor protein processing by the low density lipoprotein receptor-related protein (LRP). Evidence that LRP contributes to the pathogenesis of Alzheimer's disease" Journal of Biological Chemistry 275(10), 7410-7415

325 Ulisse, S., Baldini, E., Sorrenti, S. and D'armiento, M. 2009 "The urokinase plasminogen activator system: a target for anti-cancer therapy" Curr Cancer Drug Targets 9(1), 32-71

326 Underhill, D.M., Owensby, D.A., Morton, P.A. and Schwartz, A.L. 1992 "Endocytosis and lysosomal delivery of tissue plasminogen activator-inhibitor 1 complexes in Hep G2 cells" Blood 80(11), 2746-2754

327 Van Dijk, M.C.M., Ziere, G.J., Boers, W., Linthorst, C., Bijsterbosch, M.K. and Van Berkel, T.J.C. 1991 "Recognition of chylomicron remnants and -migrating very-low-density lipoproteins by the

remnant receptor of parenchymal liver cells is distinct from the liver 2-macroglobulin-recognition site" Biochemical Journal 279(Pt 3), 863-870

328 Van Hinsberg, V.W., Sprengers, E.D. and Kooistra, T. 1987 "Effect of thrombin on the production of plasminogen activators and PA inhibitor-1 by human foreskin microvascular endothelial cells" Thrombosis & Haemostasis 57(2), 148-153

329 Van Leuven, F., Stas, L., Hilliker, C., Lorent, K., Umans, L., Serneels, L., Overbergh, L., Torrekens,

S., Moechars, D. and De Strooper, B. 1994 "Structure of the gene (LRP1) coding for the human 2- macroglobulin receptor lipoprotein receptor-related protein" Genomics 24(1), 78-89

187 330a Van Uden, E., Kang, D.E., Hoo, E.H. and Masliah, E. 2000a "LDL receptor-related protein (LRP) in Alzheimer's disease: Towards a unified theory of pathogenesis [Review]" Microscopy Research & Technique 50(4), 268-272

330b Van Uden, E., Sagara, Y., Van Uden, J., Orlando, R., Mallory, M., Rockenstein, E. and Masliah, E. 2000b "A protective role of the low density lipoprotein receptor-related protein against amyloid beta-protein toxicity" Journal of Biological Chemistry 275(39), 30525-30530

331 Van Zonneveld, A.J., Veerman, H., Macdonald, M.E., Van Mourik, J.A. and Pannekoek, H. 1986a "Structure and function of human tissue-type plasminogen activator (t-PA)" Journal of Cellular Biochemistry 32(3), 169-178

332 Van Zonneveld, A.J., Veerman, H. and Pannekoek, H. 1986b "On the interaction of the finger and the kringle-2 domain of tissue-type plasminogen activator with fibrin. Inhibition of kringle-2 binding to fibrin by epsilon-amino caproic acid" Journal of Biological Chemistry 261(30), 14214-14218

333 Van Zonneveld, A.J., Veerman, H. and Pannekoek, H. 1986c "Autonomous functions of structural domains on human tissue-type plasminogen activator" Proceedings of the National Academy of Sciences of the United States of America 83(13), 4670-4674

334 Verde, P., Stoppelli, M.P., Galeffi, P., Di Nocera, P. and Blasi, F. 1984 "Identification and primary sequence of an unspliced human urokinase poly(A)+ RNA" Proceedings of the National Academy of Sciences of the United States of America 81(15), 4727-4731

335 Verde, P., Boast, S., Franze, A., Robbiati, F. and Blasi, F. 1988 "An upstream enhancer and a negative element in the 5' flanking region of the human urokinase plasminogen activator gene" Nucleic Acids Research 16(22), 10699-10716

336 Verdier, Y. and Penke, B. 2004 "Binding sites of amyloid beta-peptide in cell plasma membrane and implications for Alzheimer's disease" Curr Protein Pept Sci 5(1), 19-31

337 Verdier, Y., Zarandi, M. and Penke, B. 2004 "Amyloid beta-peptide interactions with neuronal and glial cell plasma membrane: binding sites and implications for Alzheimer's disease" J Pept Sci 10(5), 229-248

338 Verheijen, J.H., Caspers, M.P., Chang, G.T., De Munk, G.A., Pouwels, P.H. and Enger-Valk, B.E. 1986a "Involvement of finger domain and kringle 2 domain of tissue-type plasminogen activator in fibrin binding and stimulation of activity by fibrin" EMBO Journal 5(13), 3525-3530

339 Verheijen, J.H., Visse, R., Wijnen, J.T., Chang, G.T., Kluft, C. and Meera Khan, P. 1986b "Assignment of the human tissue-type plasminogen activator gene (PLAT) to chromosome 8" Human Genetics 72(2), 153-156

340 Verpillat, P., Bouley, S., Campion, D., Hannequin, D., Dubois, B., Belliard, S., Puel, M., Thomas- Anterion, C., Agid, Y., Brice, A. and Clerget-Darpoux, F. 2001 "Use of haplotype information to test involvement of the LRP gene in Alzheimer's disease in the French population" European Journal of Human Genetics 9(6), 464-468

341 Von Arnim, C.A., Kinoshita, A., Peltan, I.D., Tangredi, M.M., Herl, L., Lee, B.M., Spoelgen, R., Hshieh, T.T., Ranganathan, S., Battey, F.D., Liu, C.X., Bacskai, B.J., Sever, S., Irizarry, M.C., Strickland, D.K. and Hyman, B.T. 2005 "The low density lipoprotein receptor-related protein (LRP) is a novel beta-secretase (BACE1) substrate" Journal of Biological Chemistry 280(18), 17777-17785

342 Wang, Y. 2001 "The role and regulation of urokinase-type plasminogen activator receptor gene expression in cancer invasion and metastasis" Med Res Rev 21(2), 146-170

343 Warshawsky, I., Bu, G. and Schwartz, A.L. 1993 "39-kD protein inhibits tissue-type plasminogen activator clearance in vivo" Journal of Clinical Investigation 92(2), 937-944

188 344 Warshawsky, I., Bu, G. and Schwartz, A.L. 1994 "LRP and the receptor-mediated endocytosis of plasminogen activators" Annals of the New York Academy of Sciences 737, 70-87

345 Weaver, A.M., Mccabe, M., Kim, I., Allietta, M.M. and Gonias, S.L. 1996 "Epidermal growth factor and platelet-derived growth factor-BB induce a stable increase in the activity of low density lipoprotein receptor-related protein in vascular smooth muscle cells by altering receptor distribution and recycling" Journal of Biological Chemistry 271(40), 24894-24900

346 Webb, J.C., Patel, D.D., Jones, M.D., Knight, B.L. and Soutar, A.K. 1994 "Characterization and tissue-specific expression of the human 'very low density lipoprotein (VLDL) receptor' mRNA" Human Molecular Genetics 3(4), 531-537

347 Williams, L.M., Gibbons, D.L., Gearing, A., Maini, R.N., Feldmann, M. and Brennan, F.M. 1996 "Paradoxical effects of a synthetic metalloproteinase inhibitor that blocks both p55 and p75 receptor shedding and TNF processing in RA synovial membrane cell cultures" Journal of Clinical Investigation 97(12), 2833-2841

348 Williams, S.E., Ashcom, J.D., Argraves, W.S. and Strickland, D.K. 1992 "A novel mechanism for

controlling the activity of 2-macroglobulin receptor/low density lipoprotein receptor-related protein: Multiple regulatory sites for 39-kDa receptor-associated protein" Journal of Biological Chemistry 267(13), 9035-9040

349 Williams, S.E., Kounnas, M.Z., Argraves, K.M., Argraves, W.S. and Strickland, D.K. 1994 "The 2- macroglobulin receptor/low density lipoprotein receptor-related protein and the receptor associated protein: An Overview" Annals of the New York Academy of Sciences 737, 1-13

350 Willnow, T.E., Sheng, Z., Ishibashi, S. and Herz, J. 1994 "Inhibition of hepatic chylomicron remnant uptake by gene transfer of a receptor antagonist" Science 264(5164), 1471-1474

351 Willnow, T.E., Moehring, J.M., Inocencio, N.M., Moehring, T.J. and Herz, J. 1996 "The low- density-lipoprotein receptor-related protein (LRP) is processed by furin in vivo and in vitro" Biochemical Journal 313(Part 1), 71-76

352 Wiman, B. and Wallen, P. 1977 "The specific interaction between plasminogen and fibrin. A physiological role of the lysine binding site in plasminogen" Thrombosis Research 10(2), 213-222

353 Wiman, B., Lijnen, H.R. and Collen, D. 1979 "On the specific interaction between the lysine-binding sites in plasmin and complementary sites in alpha2-antiplasmin and in fibrinogen" Biochimica et Biophysica Acta 579(1), 142-154

354 Wing, L.R., Bennett, B. and Booth, N.A. 1991a "The receptor for tissue plasminogen activator (t- PA) in complex with its inhibitor, PAI-1, on human hepatocytes" FEBS Letters 278(1), 95-97

355 Wing, L.R., Hawksworth, G.M., Bennett, B. and Booth, N.A. 1991b "Clearance of t-PA, PAI-1, and t-PA-PAI-1 complex in an isolated perfused rat liver system" Journal of Laboratory and Clinical Medicine 117(2), 109-114

356 Wun, T.-C., Palmier, M.O., Siegel, N.R. and Smith, C.E. 1989 "Affinity purification of active plasminogen activator inhibitor-1 (PAI-1 using immobilized anhydrourokinase. Demonstration of the binding, stabilization, and activation of PAI-1 by vitronectin" Journal of Biological Chemistry 264(14), 7862-7868

357 Wygrecka, M., Marsh, L.M., Morty, R.E., Henneke, I., Guenther, A., Lohmeyer, J., Markart, P. and Preissner, K.T. 2009 "Enolase-1 promotes plasminogen-mediated recruitment of monocytes to the acutely inflamed lung" Blood doi:10.1182/blood-2008-08-170837, 0

358 Yepes, M., Roussel, B.D., Ali, C. and Vivien, D. 2009 "Tissue-type plasminogen activator in the ischemic brain: more than a thrombolytic" Trends in Neurosciences 32(1), 48-55

189 359 Yochem, J. and Greenwald, I. 1993 "A gene for a low density lipoprotein receptor-related protein in the nematode Caenorhabditis elegans" Proceedings of the National Academy of Sciences of the United States of America 90(10), 4572-4576

360 Zeng, Y., Tao, N., Chung, K.N., Heuser, J.E. and Lublin, D.M. 2003 "Endocytosis of oxidized low density lipoprotein through scavenger receptor CD36 utilizes a lipid raft pathway that does not require caveolin-1" Journal of Biological Chemistry 278(46), 45931-45936

361 Ziady, A.G., Kim, J., Colla, J. and Davis, P.B. 2004 "Defining strategies to extend duration of gene expression from targeted compacted DNA vectors" Gene Ther 11(18), 1378-1390

362 Zorio, E., Gilabert-Estelles, J., Espana, F., Ramon, L.A., Cosin, R. and Estelles, A. 2008 "Fibrinolysis: the key to new pathogenetic mechanisms" Curr Med Chem 15(9), 923-929

190