EFFETS ANTI-TUMORAUX DE PEPTIDES ISSUS DE LA

Comments

Transcription

EFFETS ANTI-TUMORAUX DE PEPTIDES ISSUS DE LA
UNIVERSITE PARIS 12 –VAL DE MARNE
ÉCOLE DOCTORALE SCIENCES DE LA VIE ET DE LA SANTE
THESE
Présentée pour obtenir le grade de
DOCTEUR DE L’UNIVERSITE PARIS 12
Spécialités : Biochimie, Biologie Cellulaire et Moléculaire
par
Oya BERMEK
EFFETS ANTI-TUMORAUX DE PEPTIDES ISSUS DE LA
STRUCTURE DU FACTEUR DE CROISSANCE HARP
VIA LES RECEPTEURS ALK ET RPTPβ
Le directeur de thèse: Dr. José COURTY
Soutenue le 17 Décembre 2007 devant le jury composé de :
Dr. Gérard CHASSAING- Université Paris 6
Dr. Jean PLOUET- Hôpital Lariboisière
Pr. Hubert HONDERMARCK- Université de Lille
Dr. Josette BADET- Université Paris 5
Pr. David G. FERNIG- Université de Liverpool
Dr. José COURTY- Université Paris 12
Président
Rapporteur
Rapporteur
Examinateur
Examinateur
Examinateur
Laboratoire de Recherche sur la Croissance Cellulaire, la Réparation
et la Régénération Tissulaires CRRET-CNRS FRE 2412
61, avenue du Général De Gaulle, Créteil
UNIVERSITY PARIS 12 – VAL DE MARNE
SCHOOL OF LIFE AND HEALTH SCIENCES
THESIS
Submitted for the degree of
DOCTOR OF UNIVERSITY PARIS 12
Specialities : Biochemistry, Cellular and Molecular Biology
by
Oya BERMEK
ANTI-TUMORAL EFFECTS OF HARP PEPTIDES
THROUGH ALK AND RPTPβ PATHWAYS
Thesis Supervisor : José COURTY
Defended on December 17th 2007 on the jury composed of
phD Gérard CHASSAING- University Paris 6
phD Jean PLOUET- Hospital Lariboisière
Pr Hubert HONDERMARCK- University of Lille
phD Josette BADET- University Paris 5
Pr David G.FERNIG- University of Liverpool
phD José COURTY- University Paris 12
President
Referent
Referent
Examiner
Examiner
Examiner
Laboratoire de Recherche sur la Croissance Cellulaire, la Réparation
et la Régénération Tissulaires CRRET-CNRS FRE 2412
61, avenue du Général De Gaulle, Créteil
ACKNOWLEDGMENTS
This thesis has been realized over the last three years in the laboratory CRRET, at
the University Paris 12. First of all, I would like to express my sincere gratitude for my
supervisor José Courty for his continuous support, his guidance, his encouragement and
understanding during all those years. He has always shared my enthusiasm and his new
ideas sorted all the matters out. I also owe my sincere gratitude for his confidence in me
and for providing to me the opportunity to live a unique research experience in England.
I wish to express my special thanks to Jean Delbé who had a direct impact on the
implementation of this thesis as well as my scientific development. I feel a deep sense of
gratitude for his sympathetic help, detailed and constructive comments throughout this
work.
A part of this work has been realized in the School of Biological Chemistry, at the
University of Liverpool, under the supervision of David G. Fernig. It was a real pleasure to
work with him and I feel privileged for having a superb supervisor like him. The numerous
and extensive discussions that I had with him were very precious to me. They formed a
significant part of my vision and will help me also in the future. Thanks for his interest,
patience and careful reading of the first version of the thesis.
The synthesis of some peptides used in this work was done in “Laboratoire de
Synthèse, Structure et Fonction de Molécules Bioactives”, at the University Paris 6, under
the supervision of Gérard Chassaing. I would like to thank him for his support as well as
Fabienne Burlina for her help. I also wish to express my gratitude for the other members of
my PhD committee, Jean Plouet, Hubert Hondermarck, Josette Badet, who dedicated a part
of their invaluable time and who took an effort for reading the first version of the thesis.
I take this opportunity to express my most sincere thankfulness to Eric Huet for
helping me in all aspects during the whole thesis. His ideas and comments were of great
value for me. His criticism and his sceptical approach have always pushed me ahead. In his
endeavor to show me the essence of biology he suffered a lot from the early chemist whom
I was.
I have great regard for, Yamina Hamma-Kourbali, Patricia Albanese, Isabelle
Martelly, Angelica Keller and Sophie Besse for their friendly support and encouragement,
for their invaluable comments and for providing me advises and tips that helped me a lot.
I would like to thank to all the rest of the lab CRRET for providing a wonderful
atmosphere and good working environment and for accepting me in this great family.
Thanks to all my friends in the School of Biological Sciences. I would like to
express my special thanks for Laurence Duschesne for her precious help in all aspects
ACKNOWLEDGMENTS
during my stay in Liverpool. And Sophie, Tim, Paul, Carla, the French, Polish and Italian
friends for making me live an unforgettable time in Liverpool.
I owe a very particular thank to Daniele Caruelle. It has been longtime now, since
she left our lab and our team. Our discussions in the realm of science and life have been
always in my memories.
A thesis is a difficult task, and the force and the energy necessary to keep on it were
externally provided by my friends, without whom I would not have been able to
accomplish it. Zuleyha Geels was maybe my secret source of force. In a deep contrast to
my positivist approach, her occult and spiritual ideas provided me to evaluate things from a
completely different view. I thank her for this fruitful friendship and support.
Some people do not only influence people by their way of life but make things
change as well. This may be the definition of such a man and Jean-Claude Droyer is one of
them. He has influenced my personality, my ideas and had a direct impact on my life.
Thanks to Jean-Claude, for everything.
My research career in France, in my view, started already five years ago, in Ecole
Polytechnique where a wonderful friendship has started with Michiel de Greef. I could not
finish this work without his help. Our inspiring discussions, our eagerness to understand
the French people and the subtleties of the French language were our way to escape from
the frustrations of the initial stage in our research career. I am deeply grateful to him.
I would like to express my sincere thanks to Joel Berton, the secret hero, with
whom I shared my other passion, climbing, during all these years. Our climbing adventures
were my equilibrium and recreation during the challenging years of the thesis and each one
provided me to return to the lab with an additional energy and determination for the
continuation of my thesis work.
I feel a deep sense of gratitude for Denis Barritault who introduced me into the
world of biology in France and for everything he did for me. I hope that he does not regret
the support he has given me over these years.
At last, I would like to thank my father who was my source of inspiration. His
scientific approach has deeply influenced my vision in the science. I would like to thank
my mother, my brother and my sister who were a constant support for me during all these
years far from Turkey. I feel very lucky to have such a wonderful family and I am proud of
being a part of it.
Oya Bermek
Paris, October 21st, 2007
“Cet univers désormais sans maître ne lui paraît ni stérile ni futile.
Chacun des grains de cette pierre, chaque éclat minéral de cette
montagne pleine de nuit, à lui seul forme un monde. La lutte
elle même vers les sommets suffit à remplir un cœur d’homme.
Il faut imaginer Sisyphe heureux.”
Albert Camus, Le mythe de Sisyphe (1942).
ABBREVIATIONS
ALK
ALK ECD
BEL
Boc
BS3
BSA
CAM
CD
cDNA
CHO
CS-A
CS-D
CS-E
DIEA
DMEM
DMF
DSS
EC
ECM
EDTA
EGF
ELISA
eNOS
ERK
FAK
FBS
FCS
FGF
FGF-BP
FGFR
Fmoc
GAGs
GlcN
HARP
HBGF
HGF/SF
HIV
HS
HSPG
HUVEC
IdoA
IP3
IRS
ITC
anaplastic lymphoma kinase
anaplastic lymphoma kinase extracellular domain
bovine endothelial lens
tert-butyloxycarbonyl
bis(sulfosuccinimidyl)suberate
bovine serum albumin
chorioallantoic membrane
circular dichroism
complementary deoxyribonucleic acid
chinese hamster ovary
chondroitin sulfate A
chondroitin sulfate D
chondroitin sulfate E
N,N-diisopropylethylamine
Dulbecco’s modified eagle’s medium
dimetylformamide
disuccininimidyl suberate
endothelial cell
extracellular matrix
ethylene diamine tetracetic acid
epidermal growth factor
enzyme-linked immunosorbent assay
endothelial NO synthase
extracellular signal-regulated kinase
focal adhesion kinase
fetal bovine serum
fetal calf serum
fibroblast growth factor
fibroblast growth factor binding protein
fibroblast growth factor receptor
9-fluorenylmethyloxycarbonyl
glycosaminoglycans
glucosamine
heparin affin regulatory peptide
heparin binding growth factor
hepatocyte growth factor/scatter factor
human immunodeficiency virus
heparan sulfate
heparan sulfate proteoglycan
human umbilical vein endothelial cells
iduronic acid
inositol 1,4,5-triphosphate
insulin receptor substrate
isothermal titration calorimetry
ABBREVIATIONS
kb
kDa
Kd
LRP
LTP
MALDI-TOF
MAPK
MK
MMP
mRNA
NHS
nM
NMR
NPN
NPM
OPN
ORF
PBS
PBS-T
PDGF
PG
PI3-K
PKC
PLC-γ
PTPζ
PVDF
RPMI
RPTPβ
RTK
RT-PCR
SDS
SDS-PAGE
SH2
Shc
SPR
SPPS
SRCD
STAT
TGFβ
TSP
TSR1
VEGF
kilobase
kilodalton
dissociation constant
low-density-lipoprotein Receptor-related Protein
long-term potentiation
matrix-assisted laser desorption ionization-time of flight
mitogen activated protein kinase
midkine
matrix metalloprotease
messenger ribonucleic acid
N-hydroxysulfosuccinimide
nanomolar
nuclear magnetic resonance
neuropilin
nucleophosmin
osteopontin
open reading frame
phosphate buffer saline
phosphate buffer saline-tween 20
platelet-derived growth factor
proteoglycan
phosphatidylinositol3-kinase
protein kinase C
phospholipase C-gamma
protein tyrosine phosphatase zeta
polyvinylidiene fluride
Roswell Park Memorial Institute medium
receptor protein tyrosine phosphatase beta
receptor tyrosine kinase
reverse transcriptase-polymerase chain reaction
sodium dodecylsulfate
sodium dodecylsulfate-polyacrylamide gel electrophoresis
Src homology 2
Src homology and collagen domain
surface plasmon resonance
solid phase peptide synthesis
surface resonance circular dichroisme
signal transducers and activators of transcription
transforming growth factor beta
thrompospondin
thrompospondin type-1 repeat
vascular endothelial growth factor
TABLE OF CONTENTS
INTRODUCTION......................................................................................................... 1
1. HISTORICAL REMIND ........................................................................................ 3
1.1 GROWTH FACTORS........................................................................................... 3
1.2 REGULATION OF MITOGENIC RESPONSIVENESS TO GROWTH
FACTORS ...................................................................................................................... 3
1.3 THE GROWTH FACTOR CONNECTION TO CANCER.................................... 5
1.4 TUMOR ANGIOGENESIS................................................................................... 4
1.5 HEPARIN-BINDING GROWTH FACTORS ....................................................... 7
1.6 BIOLOGICAL ROLES OF HEPARIN.................................................................. 8
1.7 CELL SIGNALING BY RECEPTOR KINASES .................................................. 9
1.8 FGF MODELS...................................................................................................... 11
1.9 VEGF MODELS................................................................................................... 12
1.10 MIDKINE ........................................................................................................... 12
2. HARP....................................................................................................................... 17
2.1 GENE.................................................................................................................... 17
2.1.1 CHROMOSAL LOCATION......................................................................................... 17
2.1.2 GENE STRUCTURE..................................................................................................... 17
2.1.3 GENE EXPRESSION DURING DEVELOPMENT ................................................... 18
2.1.4 REGULATION OF EXPRESSION.............................................................................. 21
2.2 STRUCTURE ....................................................................................................... 25
2.3 RECEPTORS-INTERACTIONS-MECHANISMS OF ACTION .......................... 29
2.3.1 CELL SURFACE PROTEOLYCANS .......................................................................... 29
2.3.1.1 Heparan sulfate proteoglycans: Syndecans ........................................ 34
2.3.1.2 Chondroitin sulfate proteoglycans: Phosphacan................................. 36
2.3.2 ANAPLASTIC LYMPHOMA KINASE....................................................................... 38
2.3.3 NUCLEOLIN................................................................................................................... 39
2.4 BIOLOGICAL ACTIVITIES ................................................................................ 40
2.4.1 MITOGENIC ACTIVITY .............................................................................................. 40
2.4.2 DIFFERENTIATION PROMOTING ACTIVITY ....................................................... 42
2.4.3 ONCOGENIC ACTIVITY ............................................................................................. 43
2.4.4 ANGIOGENIC ACTIVITY............................................................................................ 43
2.4.5 MIGRATION................................................................................................................... 45
2.5 FUNCTIONAL DOMAINS OF HARP ................................................................. 45
3. AIM OF THE THESIS ........................................................................................... 48
MATERIALS AND METHODS .................................................................................. 50
1. MATERIALS .......................................................................................................... 51
2. METHODS .............................................................................................................. 53
2.1 SOLID-PHASE PEPTIDE SYNTHESIS............................................................... 53
2.1.1 SYNTHESIS ................................................................................................................... 54
2.1.2 PURIFICATION AND CHARACTERIZATION ....................................................... 54
2.2 PRODUCTION AND PURIFICATION OF RECOMBINANT HARP.................. 54
2.2.1 PRODUCTION OF HARP FROM BACTERIA ........................................................ 54
2.2.2 PRODUCTION OF HARP FROM CONDITIONED MEDIA................................... 55
2.2.3 PURIFICATION OF HARP .......................................................................................... 55
2.3 SAMPLE PREPARATION FOR SDS-PAGE ....................................................... 55
2.4 GEL ELECTROPHORESIS- SDS-PAGE ............................................................. 56
2.4.1 COOMASSIE BLUE AND SILVER STAINING ........................................................ 56
2.4.2 WESTERN BLOTTING (IMMUNOBLOTTING) ...................................................... 58
2.5 MEMBRANE STRIPPING ................................................................................... 59
2.6 LIGAND BLOT .................................................................................................... 59
2.7 HEPARIN-ELISA................................................................................................. 59
2.8 CELL BASED-ELISA (C-ELISA) ........................................................................ 60
2.9 IMMUNOFLUORESCENCE/CONFOCAL MICROSCOPY................................ 61
2.10 CELL SURFACE BINDING ASSAY ................................................................. 61
2.11 PROLIFERATION ASSAYS .............................................................................. 62
2.11.1 THYMIDINE INCORPORATION ASSAY.............................................................. 62
2.11.2 PROLIFERATION ASSAYS BY SPECTROPHOTOMETRY............................... 62
2.12 TUMOR GROWTH IN NUDE MICE................................................................. 64
2.13 SOFT AGAR ASSAY......................................................................................... 64
2.14 OVER-AGAR ASSAY........................................................................................ 64
2.15 EXTRACTION OF RNA .................................................................................... 65
2.16 RT-PCR .............................................................................................................. 65
2.17 ANALYSIS OF BIOMOLECULAR INTERACTIONS ...................................... 66
2.17.1 OPTICAL BIOSENSOR ANALYSIS (IAsys) .......................................................... 66
2.17.2 BINDING ASSAYS ................................................................................................... 67
2.17.2.1 Immobilization of oligosaccharides and proteins ............................. 67
2.17.2.2 Binding assays ................................................................................ 68
RESULTS ...................................................................................................................... 71
1.
IMPLICATION OF THE C-TERMINAL OF HARP IN THE
MITOGENIC, TRANSFORMING AND ANGIOGENIC ACTIVITIES................... 72
1.1 INTRODUCTION................................................................................................. 73
1.2 RESULTS ............................................................................................................. 74
1.2.1 INHIBITION OF TUMOR GROWTH BY SYNTHETIC PEPTIDES
(ARTICLE 1)......................................................................................................................................... 74
1.2.2 UNPUBLISHED RESULTS ......................................................................................... 98
1.2.2.1 Proliferation of cells transfected with cDNA of HARP Δ111-136
and HARP Δ124-136 ...................................................................................................... 98
1.2.2.2 Bio-distribution of radiolabeled-P111-124 in nude mice.................... 101
1.2.2.3 Isolation of cell-surface receptors for P111-136 ................................ 103
1.2.2.4 Structure-function relationships......................................................... 107
1.3 DISCUSSION ....................................................................................................... 111
2. BASIC DOMAIN OF C-TERMINAL OF HARP................................................. 116
2.1 INTRODUCTION................................................................................................. 117
2.2 RESULTS ............................................................................................................. 119
2.2.1 CHARACTERIZATION OF 10-AMINO ACID PEPTIDE (ARTICLE 2) ............ 119
2.2.2 ANTI-TUMORAL ACTIONS OF P122-131 THROUGH RPTPβ(ARTICLE 3)..... 130
2.3 DISCUSSION ....................................................................................................... 156
3.
MOLECULAR INTERACTION OF HARP AND HARP-DERIVED
PEPTIDES..................................................................................................................... 159
3A. MOLECULAR BASIS OF HARP - HEPARIN INTERACTION........................ 160
3A.1 INTRODUCTION.............................................................................................. 160
3A.2 RESULTS .......................................................................................................... 162
3A.2.1 HEPARIN-BINDING DOMAINS OF HARP .......................................................... 162
3A.2.2 INTERACTIONS OF HARP, HARPΔ124-136 AND CTSR WITH HEPARIN
AND CHEMICALLY MODIFIED HEPARINS................................................................................ 164
3A.2.3 INTERACTIONS OF HARP, HARPΔ124-136 AND CTSR WITH
OLIGOSACCHRIDES OF DIFFERENT LENGTHS ....................................................................... 170
3A.2.4 INTERACTIONS OF HARP, HARPΔ124-136 AND CTSR WITH HEPARIN
AND ITS CATION-BOUND DERIVATIVES ................................................................................. 173
3A.3 DISCUSSION .................................................................................................... 175
3B. HARP-BINDING PROTEINS - NEW IMPLICATION FOR HARP
FUNCTION................................................................................................................... 179
3B.1 INTRODUCTION .............................................................................................. 179
3B.2 RESULTS........................................................................................................... 180
3B.2.1 HARP-BINDING HEPARIN-BINDING PROTEINS ...................................... 180
3B.2.2 NEW IMPLICATIONS FOR HARP FUNCTION .......................................... 183
3B.2.2.1 S100 A4 ......................................................................................... 183
3B.2.2.2 Osteopontin .................................................................................... 185
3B.3 DISCUSSION..................................................................................................... 186
CONCLUSION AND SUGGESTIONS FOR FUTURE WORK ................................ 189
REFERENCES.............................................................................................................. 197
APPENDIX.................................................................................................................... 216
TABLE OF ILLUSTRATIONS
FIGURES
Figure 1: Models of the FGFR signaling complex ......................................................... 12
Figure 2: Three dimensional domain structures of MK.................................................. 14
Figure 3: FGF2-induced HARP gene regulation in LnCaP cells ..................................... 24
Figure 4: Amino acid sequence of HARP....................................................................... 25
Figure 5: Schematic representation of structural features of HARP ................................ 26
Figure 6: Structural comparison of human HARP and human MK ................................. 28
Figure 7: Disaccharide repeating units of totally sulfated and desulfated heparins .......... 31
Figure 8: Three isoforms of RPTPβ/ζ............................................................................. 36
Figure 9: Functional domains of HARP.......................................................................... 47
Figure 10: A binding assay............................................................................................. 69
Figure 11: Proliferation of transfected U87MG cells by HARP mutants ......................... 99
Figure 12: Growth properties of U87MG transfected with HARP constructs .................. 100
Figure 13: Test of bioactivity of tyrosinylated peptides .................................................. 101
Figure 14: Biodistribution of [I125]- P111-124 after 30 min and 6 h of the intravenous
injection .......................................................................................................................... 102
Figure 15: Specific binding of the biotinylated P111-136 to the 50-, 75- and 100-kDa
proteins ........................................................................................................................... 104
Figure 16: Inhibition of HARP mitogenic activity by P111-121 ..................................... 106
Figure 17: HARP mitogenic signaling through ALK in presence of HSPGs ................... 107
Figure 18: Interaction of proteins with immobilized biotinylated P111-136 and HARP .. 109
Figure 19: Binding of ALK ECD to immobilized heparin............................................... 110
Figure 20: Disordered profile plot of HARP................................................................... 112
Figure 21: Hypothesis for differential signaling of peptides derived from C-terminal ..... 118
Figure 22: α1-4 linked IdoA-GlcN dissacharide repeating structure of heparin/HS
chains.............................................................................................................................. 160
Figure 23: Binding of synthetic peptides and HARP to immobilized heparin.................. 163
Figure 24: Chemically modified polysaccharides ........................................................... 165
Figure 25: Competition of HARP, HARPΔ124-136 and CTSR binding to
immobilized heparin by modified forms of heparin ......................................................... 166
Figure 26: Competition of HARP, HARPΔ124-136 and CTSR binding to
immobilized heparin by heparin oligosaccharides of different lengths ............................. 170
Figure 27: Competition of HARP, HARPΔ124-136 and CTSR binding to
immobilized heparin by Mn2, K+, Cu2+ and Ca2-bound heparin derivatives...................... 173
Figure 28: Interaction of proteins with immobilized HARP determined in an optical
biosensor......................................................................................................................... 180
Figure 29: Competition by heparin for binding of growth factors to immobilized
HARP ............................................................................................................................. 181
Figure 30: Interaction of HARP-binding proteins with immobilized P111-136
determined in an optical biosensor .................................................................................. 182
Figure 31: Binding of S100A4 to HARP ........................................................................ 184
Figure 32: Binding of HARP and CTSR to OPN-derivatized surface and competition
by heparin ....................................................................................................................... 185
Figure 33: Molecular modelization of CTSR domain of HARP corresponding to the
amino acids 60-111. ........................................................................................................ 187
Figure 34: The two functional domains of C-terminal mediating biological functions
of HARP via two different HARP receptors .................................................................... 191
Figure 35: Model for HARP15 signaling via ALK receptor............................................ 192
Figure 36: Model for HARP18 signaling via ALK and RPTPβ receptors ....................... 193
Figure 37: Predicted secondary structure of P111-136 by PSIPRED studies .................. 194
TABLES
Table 1: Synthetic peptides............................................................................................. 52
Table 2: Synthesis of P122-131 and biotin-labeled forms ............................................... 54
Table 3: Primers used for amplification .......................................................................... 65
Table 4: IC50 values of chemically modified heparins (µg/ml) ........................................ 168
Table 5: IC50 values of oligosaccharides obtained from enzymatic cleavage of heparin
chains.............................................................................................................................. 172
Table 6: IC50s of the cation forms of heparin derivatives ................................................ 175
Table 7: Amino acid abbreviations, molecular formula and characteristics ..................... 216
INTRODUCTION
INTRODUCTION
In the first part of the introduction, growth factors will be described as
molecules, which play fundamental roles in normal growth and development and the
constitutive activation of their genes as a feature of many pathological diseases. For these
reasons, understanding the mechanisms that regulate the expression of growth factor genes
and the signaling pathways induced by their protein products is of fundamental importance
in both health and disease.
Indeed, the growth factors such as FGF or VEGF are the well known and the
most studied growth factors, among the members of HBGF family. However, our
laboratory has, in particular, interested in the heparin-binding growth factor HARP since it
was isolated from bovine brain by Courty et al., in 1991 (Courty et al., 1991). HARP is a
multifunctional polypeptide exerting a variety of biological activities such as neuritepromoting, neuronal cell survival and differentiation activities. It regulates inflammation
and tissue repair. In terms of clinical significance, the protein is specifically localized in
brain of patients with Alzheimer’s disease and increasing levels of HARP mRNA are
found in many human carcinomas. Therefore, HARP is also of considerable interest in
cancer biology.
My essential work during my thesis consisted of the elucidation of the molecular
mechanism of this protein in the domain of cancer. Therefore, in the second part of the
introduction, I will focus on the regulation of harp gene, the structure and functions of the
protein, the roles as a tumor promoter and angiogenic factor and the mechanisms by which
HARP signals through its receptors.
HARP displays a significant homology with MIDKINE (MK), not only in terms
of amino acid sequence and three-dimensional structure but also they exhibit the same
biological activities to similar extents including mitogenic, anti-apoptotic, angiogenic and
transforming ones. They also share some of their receptors. Because analysis of MK may
shed light on deeper understanding of the molecular mechanism of HARP, MK also will be
shortly described before introducing HARP.
2
INTRODUCTION
1. HISTORICAL REMIND
1.1 GROWTH FACTORS
Growth factors may be defined as polypeptides that stimulate cell proliferation
through binding to specific high-affinity cell membrane receptors. These molecules differ
from the well-known polypeptide hormones not only in the response elicited but also in
the mode of delivery from the secreting to the responding cell. Despite that few growth
factors have been identified in plasma and may act as hormones as well (Goustin et al.,
1986), they do not usually act in an endocrine manner; they also diffuse short-range
through intercellular spaces and act locally. Polypeptide growth factors also have an
extraordinary range of activities, including matrix protein deposition and resolution, the
maintenance of cell viability, cell differentiation and development, chemotaxis and
activation of inflammatory cells, tissue repair and disease (Deuel, 1987).
Growth factors have different cell type specificities; some factors stimulate only
one or few cell types while others stimulate a wide variety of cell types, both epithelial and
mesenchymal. The requirement of more than one growth factor is necessary for growth of
normal cells. Exposure of a cell to one growth factor can lower the threshold for
mitogenicity of a second growth factor. Moreover, growth factors operate at different
points of the cell cycle (Deuel, 1987).
1.2 REGULATION OF MITOGENIC RESPONSIVENESS TO GROWTH FACTORS
The cell cycle comprises four main phases. In the S-phase, the cell’s DNA is
replicated, while the physical division of the nucleus and then the cell itself occur during
the M-phase, or mitosis phase. Between these two phases are two ‘gap’ phases, G1 and G2,
during which the cell grows. In the G1 phase, the cell may pause its progress in the cycle
by entering a resting state, called the G0 phase, where it can remain indefinitely (Alberts,
1994). From here, the cell can either re-emerge into the S-phase to begin the process of
reproduction, or undergo apoptosis. Growth factors cause cells in the G0 phase to enter and
proceed through the cell cycle (Aaronson, 1991). The mitogenic response occurs in two
parts; the quiescent cell must first be advanced into the G1 phase of the cell cycle by
‘competence’ factors such as platelet-derived growth factor (PDGF), traverse the G1
phase, and then become committed to DNA synthesis under the influence of ‘progression’
factors. Transition through the G1 phase requires sustained growth factor stimulation over a
period of several hours. If the signal is disrupted for a short period of time, the cell reverts
3
INTRODUCTION
to the G0 state. There is also a critical period in G1 during which simultaneous stimulation
by both factors is needed to allow progression through the cell cycle. After this restriction
point, only the presence of a progression factor, such as insulin-like growth factor (IGF-1),
epidermal growth factor (EGF) is needed. Cytokines such as transforming growth factor β
(TGF-β), interferon or tumor necrosis factor (TNF) can antagonize the proliferative effects
of growth factors. In some cell types, the absence of growth factor stimulation causes the
rapid onset of programmed cell death or apoptosis (Aaronson, 1991). The receptors, which
stimulate apoptosis when expressed in the absence of their ligand are called dependence
receptors (Chao, 2003). The multiplicity of growth factors in various tissues, the varying
cell type specificity of growth factors and the requirement for multiple growth factors for
stimulation of specific cell types provide the fine tuning of relative proliferation rates
necessary for coordinated growth of cells to form tissues during development and to
maintain tissues in the adult state (Aaronson, 1991).
The interaction of growth factors, cytokines and hormones with specific
membrane receptors triggers a cascade of intracellular biochemical signals, resulting in the
activation and repression of various subsets of genes. Genetic aberrations in growth factor
signaling pathways are linked to developmental abnormalities and to a variety of chronic
diseases, including cancer.
1.3 THE GROWTH FACTOR CONNECTION TO CANCER
The involvement of growth factors in cancer dates to early work showing a
decreased serum requirement for growth of neoplastically transformed cells (Dulbecco and
Stoker, 1970). The altered serum requirement in transformed cells could be translated into
a diminished or absent requirement for exogenous growth factors for optimal growth and
multiplication than do their counterparts. Loss of requirement for specific growth factors is
a common finding in many types of cancer cells and could be explained by autocrine and
paracrine secretion of these factors (Goustin et al., 1986).
Autologous production of a growth factor by a cell bearing receptors for that
same factor could result in a growth advantage, allowing phenotypic expression of the
peptide by the same cell that produces it (Sporn and Roberts, 1985). The autocrine action
of an effector peptide may be amplified by an excessive production, expression and action
of autocrine growth factors. Enhanced cellular responsiveness to a growth factor may
result also from a change in the number or affinity of the receptors for the growth factor. A
further way to achieve an enhanced degree of signaling by the pathway normally activated
4
INTRODUCTION
by growth factors is to make the activation of the receptor independent of the effector.
Ligand-independent signaling can be achieved through structural alteration of receptors;
for example, truncated versions of the EGF receptor lacking much of its cytoplasmic
domain fire constitutively. The signaling pathways activated by an autocrine peptide need
not always evoke a positive growth response, as seem most strikingly in the response of
cells to TGF-β. This peptide was characterized initially by its ability to stimulate the
growth of non-neoplastic fibroblasts as colonies in soft agar (Sporn, 2005). The soft agar
assay was considered to be particularly important, because it was generally believed that
the growth in soft agar was a property of malignant cells that was not shared by normal
ones. However, TGF-β can be also a potent inhibitor of the growth of many cells. Thus,
the malignant transformation may be also result of the failure of the cells to synthesize,
express or respond to specific negative growth factors they normally release to control
their own growth.
The autocrine model may well be adequate to explain growth in soft agar and
relative growth factor independence of transformed cells, the serum factor independence
and even the pseudo malignant behavior of non-transformed cells (Goustin et al., 1986). A
second pathway involving a paracrine model might be at least as likely an explanation of
how growth factor production might operate in the development of cancer. Growth factors
released by one cell type and influencing proliferation of another cell may play important
roles in tumor progression. For instance, the ability of steroid hormones to stimulate
epithelial cell proliferation in sex-hormone-responsive tissues such as breast and prostate
appears to be mediated at least in part by hormonal effects on stromal cells. Stromal cells,
in turn, influence parenchymal cells by increasing production of growth factors, decreasing
production of inhibitory cytokines, or both. Angiogenesis, which is a vital component in
the development and progression of many human tumors, is believed to be mediated by
soluble factors released from tumor cells that then act on endothelial cells in a paracrine
manner.
1.4 TUMOR ANGIOGENESIS
Angiogenesis, the sprouting of new vessels from existing vasculature, is a
fundamental requirement for organ development and differentiation during embryogenesis,
wound healing and the female reproductive cycle. Pathologic angiogenesis is characterized
by either inadequate or excessive neovasculature.
5
INTRODUCTION
Tumor cells typically form a contigous growing cluster, which is reliant on
passive diffusion for the supply of oxygen and nutrients and the removal of waste products
(Sutherland et al., 1988). A tumor may persist in a diffusion-limited state, usually not more
than 2 mm diameter (Folkman, 1971), with cell proliferation balanced by cell death for
months or years unless the so-called angiogenic switch is turned on. It is a shifting of the
balance from the anti- to the pro-angiogenic factors that causes the transition from the
dormant to the angiogenic phase. These pro- and anti-angiogenic molecules can emanate
from cancer cells, endothelial cells (EC), stroma cells, blood and the extra-cellular matrix
(ECM) (Carmeliet and Jain, 2000). Their relative distribution is likely to change with
tumor type and tumor site.
The angiogenesis can be divided in three major steps:
•
On receiving a net angiogenic stimulus, EC in capillaries near the tumor become
activated and secrete proteolytic enzymes whose effect is to degrade extracellular
tissue. There are a large number of such enzymes, which may be broadly divided
into matrix metalloproteases (MMPs) and the plasminogen activator (PA)/plasmin
system. In the (PA)/plasmin system, PAs activate the widely expressed but inactive
substance, plasminogen, into the broad-spectrum protease, plasmin. Both of these
families of proteases have an associated class of inhibitors. MMPs are inhibited by
tissue inhibitors of metalloproteaes (TIMPs). PAs are inhibited by plasminogen
activator inhibitor (PAI). The first action of the angiogenic growth factors,
therefore, is to stimulate the production of proteases by EC (Pepper, 1997).
•
Following extravasation, the EC continue to secrete proteolytic enzymes, which
also degrade the extracellular matrix (ECM) and may release growth factors, such
as VEGF, that have been sequestered in the matrix, thus augmenting the angiogenic
signal. They continue to move away from the parent vessel and towards the tumor,
thus forming small sprouts. EC migration is governed mainly by a chemotactic
response to concentration gradients of diffusible growth factors produced by tumor.
Thus the second key function of angiogenic growth factors is to induce directed EC
migration towards the tumor.
•
For a short period following extravasation, the low mitosis levels continue: the
initial response is entirely migratory rather than proliferative. After this initial
period of migration, rapid EC proliferation begins, increasing the rate of sprout
elongation. EC proliferation is necessary for vascularization to take place, and its
6
INTRODUCTION
stimulation is the third and final key function of angiogenic growth factors (Plank
and Sleeman, 2003).
Quite a large number of growth factors and cytokines have been found to possess
angiogenic activity (Badet, 1994). Among them are acidic and basic growth factor (FGF1
and FGF2), transforming growth factors α and β (TGFα and TGFβ), epidermal growth
factor (EGF), tumor necrosis factor α (TNFα), platelet-derived growth factor (PDGF),
interleukin 1 (IL-1), interleukin 6 (IL-6) and vascular endothelial growth factor (VEGF).
VEGF, which was initially purified from the conditioned media of the pituitary cell line
(Plouët et al., 1989), is the most potent angiogenic factor so far detected (Yancopulos et
al., 2000) and it is the main driving force behind, not only tumor angiogenesis but, all
blood vessel formation. It was shown to be highly specific for EC and is capable of
inducing all three key activities of EC in angiogenesis. Several naturally occuring
angiogenic inhibitors have also been discovered, such as interferon-α/β, thrombospondin-1
(TSP-1), angiostatin and endostatin. Angiogenesis in general is controlled by the
coordinated action of positive and negative regulators and angiogenic factors have to
override the delicate balance between these regulators. The mechanism of action of each of
these factors is different. They interact with each other, with tumor cells, EC and immune
cells, and with the extra-cellular matrix to induce an apparently orderly formation of new
vessels.
Restoring the balance of angiogenesis stimulators and inhibitors in the tumor
microenvironment is one of the strategies for treatment of human malignancy.
1.5 HEPARIN-BINDING GROWTH FACTORS
The discovery of a great number of growth factors has enabled them to be
grouped into superfamilies based on amino acid sequence homology, function, as well as
their similar receptor-binding activity, such as that of fibroblast growth factor (FGF). The
nomenclature of these growth factors describes the source from which they are purified,
their target cells or their biological properties.
Growth factors, which play fundamental roles in regulation of growth,
differentiation and development, frequently bind to the glycosaminoglycans of
extracellular matrices. Since heparin is commonly used to probe these protein-sugar
interactions, this is often referred to as heparin-binding, which can be used to define
operationally, Heparin Binding Growth Factors (HBGFs) which encompass the majority of
7
INTRODUCTION
growth factors, cytokines and morphogens, including FGFs, VEGF, PIGF, heparin-binding
EGF-like growth factor, hepatocyte growth factor (HGF), TGF-β, PDGF, heparin affin
regulatory peptide (HARP) also called pleiotrophin (PTN) and midkine (MK).
Different family of Heparin-Binding: HARP/MK family
Among the heparin-binding growth factors, Heparin Affin Regulatory Peptide
(HARP) was originally discovered during routine preparation of FGF1 and was firstly
isolated from adult bovine brain as a novel heparin-binding protein (Bohlen et al., 1988;
Bohlen and Gautschi, 1989). Although the biochemical properties are characteristic of
heparin-binding growth factors, the chemical characterization has unambiguously shown
that the protein is not structurally related to any known member of FGFs and is a clearly
distinct molecule. Interestingly, it has been found that HARP shares 50% homology in
amino acid sequence with the protein product of a retinoic acid responsive gene which is
another heparin-binding protein; Midkine (MK). Thus, these two structurally and most
likely functionally related proteins constitute a different class of heparin-binding proteins
that are distinct from the heparin-binding growth factor family, which may be called the
HARP/MK family. The chicken counterpart of MK also called Retinoic acid-InducibleHeparin Binding protein (RI-HB) presents the third and last member of the HARP/MK
family.
1.6 BIOLOGICAL ROLES OF HEPARIN
Cell
surface
heparan
sulfate
proteoglycans
(HSPGs)
are
complex
macromolecules found in a wide variety of tissues of metazoans. They have been
implicated in many biological processes such as cell-cell and cell-matrix interactions,
assembly of extracellular matrix and basement membranes, receptor-mediated endocytosis
and tissue differentiation. Proteoglycans that are found predominantly on the cell surface
and in the extracellular matrix contain carbohydrates called glycosaminoglycans (GAGs).
The main GAGs in proteoglycans are chondroitin sulfate (CS), dermatan sulfate (DS),
keratan sulfate (KS), heparan sulfate (HS) and the specialised heparan sulfate, heparin
(Ruoslahti, 1989).
HBGFs have a strong affinity for the glycosaminoglycan heparin, the archetypal
member of the GAG family. This heparin-binding property has greatly facilitated the
purification and characterization of these growth factors. This is also a functional feature in
terms of storage of the factors in extracellular sites such as ECM and basement
8
INTRODUCTION
membranes, and in their controlled release. For example, biologically active FGF2 can be
displaced from the ECM by heparin or be released from by heparin-degrading enzymes
(Gospodarowicz et al., 1987; Klagsbrun, 1990). This reservoir serves to limit the
diffusibility of the growth factor and thus to regulate its bioavailability. Furthermore, the
importance of HBGFs-heparin interaction has been emphasized by the observation that
heparin protects them from denaturation and enzymatic proteolysis (Klagsbrun, 1990).
Heparin is also a potent modulator of their biological activities and binding of growth
factors to heparin has been found to regulate the assembly of ligand-receptor complexes
and can enhance or reduce their receptor activation, often depending on the concentrations
of ligand, receptor, and HSPG. The HS chains may catalyze encounters between ligand
and signaling receptor by bringing them together. Because binding to the HS chain reduces
the dimensionality of this interaction from three (when the ligand is soluble) to two (when
the ligand is bound to the HS chain), interaction could result from a localized increase in
ligand concentration at optimal HS concentrations. However, at HS levels lower or higher
than optimal, the effective ligand concentration for engaging the receptor will fall,
potentially accounting for the bell-shaped activity curve sometimes seen experimentally
when HSPG (or heparin) concentrations are varied (Lander, 1998).
1.7 CELL SIGNALING BY RECEPTOR KINASES
Many growth factors show their pleiotropic role by binding to and activating the
receptors endowed with intrinsic tyrosine kinase activity. All receptor tyrosine kinases
(RTKs) possess an extracellular ligand binding domain that is usually glycosylated, a
single hydrophobic transmembrane region, and a cytoplasmic domain that contains a
tyrosine kinase catalytic domain. The ligand-binding domain is connected to the
cytoplasmic domain by the single transmembrane helix. The cytoplasmic domain contains,
in addition to the catalytic protein tyrosine kinase (PTK) core, distinct regulatory
sequences with tyrosine, serine and threonine phosphorylation sites (Ullrich and
Schlessinger, 1990).
Tyrosine autophosphorylation of RTKs is crucial for activation of signaling
proteins. The phosphorylation of specific tyrosine residues within the activated receptor
creates binding sites for Src homology 2 (SH2) and phosphotyrosine binding (PTB)
domains containing proteins. Several SH2 containing proteins that have intrinsic enzymatic
activity include phospholipase C (PLC), the proto-oncogene c-Ras associated GTPase
activating protein (ras GAP), phosphotidylinositol-3-kinase (PI-3K), protein phosphatase9
INTRODUCTION
1C (PTP1C) as well as members of the Src family of protein tyrosine kinase (PTKs)
(Schlessinger, 1994). Other proteins that interact with the activated receptor act as adaptor
proteins and have no intrinsic enzymatic activity of their own. These adaptor proteins link
RTK activation to downstream signal transduction pathway, such as the Ras/MAP kinase
signaling cascade. MAP kinases were identified by virtue of their activation in response to
growth factor stimulation of cells in culture, hence the name mitogen-activated protein
kinases. They are also called ERKs for extracellular-signal regulated kinases. In addition to
activation of MAP kinase signaling cascade, Ras also activates PI-3 kinase that regulates
various metabolic processes and prevent apoptotic death (Schlessinger, 2000). In addition,
PI-3 kinase activation stimulates generation of hydrogen peroxide which in turn oxidizes
and blocks the action of the inhibitory protein tyrosine phosphatase (PTP) that
dephosphorylate activated PTK. The net tyrosine phosphorylation in higher eucaryotes is
controlled by the balance between activities of PTKs and PTPs. The activity of effector
proteins is dependent on PI-3 kinase activation which can be negatively regulated by
PTEN and SHIP, two phosphoinositide-specific phosphatases. PTEN is a tumor suppressor
protein that is mutated in a variety of human cancers leading to aberrant stimulation of cell
survival pathway.
Ligand-induced dimerization is a key event in transmembrane signaling by
receptors with tyrosine kinase activity. Receptor dimerization leads to an increase in kinase
activity, resulting in autophosphorylation and the induction of diverse biological responses.
Certain growth factors, such as PDGF, VEGF, colony stimulating factor-1 are themselves
dimeric proteins and can induce the dimerization of their respective receptors if each
molecule of the dimer binds independently to a separate receptor molecule, hence
providing the simplest mechanism for ligand-induced receptor dimerization. Other growth
factors are monomeric but contain two receptor-binding sites, enabling them to cross-link
two receptor molecules (Schlessinger, 2000).
Various growth factors can bind to two different classes of receptors. For
example, FGF and TGFβ both bind with high affinity to signaling receptors endowed with
tyrosine or serine/threonine kinase activities. However, the same growth factors also bind
with lower affinity to cell surface proteoglycans (HSPGs) that cannot transmit signals
alone, but somehow modulate the ability of the growth factor or the signaling receptor to
generate a biological response (Schlessinger, 1995). One of the best-studied examples for
this is that of the fibroblast growth factor receptor-1 (FGFR1) and its ligands FGF1 and
10
INTRODUCTION
FGF2. The mechanism that they represent for transmembrane signaling may account for
the action of many heparin-bound growth factors.
1.8 FGF MODELS
FGFRs are comprised of a common extracellular ligand binding portion
consisting of three immunoglobulin (Ig)-like domains (D1-D3) and thus belong to the
immunoglobulin family of RTKs. Interactions with FGFs occur via two Ig-like domains
D2 and D3 and FGF2 can generate mitogenic signals through its receptor tyrosine kinase
only when bound to either cell surface HSPGs or to free, soluble heparin or heparan sulfate
(Yayon et al., 1991; Ornitz et al., 1992). However, it is now accepted that FGF binds its
receptor without heparin and indeed can signal but there is no progression through the cell
cycle (Delehedde et al., 2002).
It has been proposed that binding of FGF to HSPG could induce a conformational
change in FGF and in consequence, its oligomerization that is required for a biologically
active interaction with its high affinity receptor (Spivak-Kroizman et al., 1994). Several
dimerization models have been proposed by different groups (Springer et al., 1994;
Spivak-Kroizman et al., 1994; Venkatamaran et al., 1996; Plotnikov et al., 1999;
Schlessinger et al., 2000; Pellegrini et al., 2000; Stauber et al., 2000). One of the most
popular model is that of Springer et al. (Springer et al., 1994). According to this model,
FGF2 and HS participate in a concerted bridge mechanism for the dimerization of FGFR1
and FGFR-binding sites on FGF are distinct from heparan sulfate binding domains (Figure
1). The crystal structure of co-complexes of FGF:heparin oligosaccharide:FGFR have
failed to conclusively distinguish between these models. For example, in one such cocomplex (Schlessinger et al., 2000), it is claimed that a dimeric 2:2:2 FGF:FGFR:heparin
ternary complex is formed, termed the “two-end” model, in which two 1:1 FGF:FGFR
complexes form a symmetric dimer. Within each 1:1 FGF:FGFR complex, heparin makes
numerous contacts with both FGF and FGFR, thereby augmenting FGF-FGFR binding.
Heparin also interacts with FGFR in the adjoining 1:1 FGF:FGFR complex to promote
FGFR dimerization (Schlessinger et al., 2000) (Figure 1). In this model the direction of the
saccharide chains is opposite, such that the non-reducing ends of the two saccharides
“point” at each other. An alternative structure, termed the “asymmetric model”, is
assembled around a central heparin decasaccharide linking two FGF1 ligands into a dimer
that bridges between two receptor chains (Pellegrini et al., 2000). The asymmetric heparin
binding involves contacts with both FGF1 molecule but only one receptor chain (Figure 1).
11
INTRODUCTION
Figure 1: Models of the FGFR signaling complex (from Duchesne (unpublished data).
1.9 VEGF MODEL
The multiple receptors of VEGFs include signaling receptor tyrosine kinases
(VEGFR1 and VEGFR2) and the nonsignaling co-receptor, Neuropilin-1. Multiple splice
forms of VEGF also exist: VEGF165 and VEGF121. Both VEGF isoforms bind to VEGF
receptor tyrosine kinases (VEGFRs) to induce signals. VEGF165 also interacts with
nonsignaling Neuropilin-1 co-receptors and with proteoglycans of the extracellular matrix
(Mac Gabhann and Popel AS, 2005). The binding sites on VEGF165 for VEGFR2 and
Neuropilin-1 (NPN-1) are distinct, so VEGF165 may bind simultaneously. It has been
shown that NPN-1 can likewise couple to VEGFR2 in the presence of VEGF165; which
potentiates the signal that VEGF165 transduces through VEGFR2. There are thus two
parallel pathways for VEGF165 to bind its signaling receptor: first by binding directly to
VEGFR2; and second by binding to NPN-1 and then diffusing laterally on the cell surface
to couple to VEGFR2. There is also evidence that NPN-1 and VEGFR2 do not directly
interact but are bridged by VEGF165. However, VEGF121, which does not bind NPN-1 can
only form VEGFR2 complexes directly (Soker et al., 2002).
1.10 MIDKINE
Retinoic acid, which is a derivative of retinol (vitamin A), is a key molecule in
the regulation of differentiation and development. Because intracellular receptors for
retinoic acid are members of the nuclear receptor superfamily, the receptor-retinoic acid
complex is expected to control the expression of certain genes, which, in turn, control
subsequent steps in development (Sun and Fukui, 1998). MIDKINE was first identified as
a retinoic acid-inducible gene in mouse embryos (Kadomatsu et al., 1988; Tomomura et
12
INTRODUCTION
al., 1990) during the midgestation period; among the adult organs, it was detected only in
the kidney. Because of its expression pattern, the protein encoded by the gene is called
midkine (midgestation and kidney). Chicken MK purified from chick basement membranes
(Raulais et al., 1991), is called Retinoic acid-Inducible Heparin-Binding protein (RI-HB).
GENE
The human mk gene is located on chromosome 11, band p11.2 (Kaname et al.,
1993), while the mouse gene is on chromosome 2 (Simon-Chazottes et al., 1992). Both
genes encompass 2 kb. The coding sequence is found to be arranged in four exons and the
exon-intron boundary is highly homologous between mouse and human (Uehera et al.,
1992). In the promoter region, in addition to retinoic acid- responsive element, there is a
binding site for WT1, the product of Wilms’ tumor suppressor gene.
The expression of MK is developmentally regulated. As an example, in the
cerebrum of the mouse or rat, its expression is strong in the midgestation period, but
scarcely expressed in the adult cerebrum. Strong expression is also found in the brain, in
the epithelial tissue in which epithelial-mesenchymal interactions are taking place, and in
the mesenchymal tissues undergoing remodeling (Kadomatsu et al., 1990; Mitsiadis et al.,
1995).
STRUCTURE
MK is a 13-kDa polypeptide rich in basic amino acids and contains 10 conserved
cysteine residues, all of which appeared to be disulphide linked (Tomomura et al., 1990).
The protein is structurally composed of two domains, each of which is held by intradomain
disulfide bridges (Fabri et al., 1993). The three dimensional structure of the N-terminal
(MK 1-59) and C-terminal (MK 62-104) half molecules was determined by NMR.
According to this solution structure, both domains consist of three anti-parallel β-sheets,
but in the difference of N-terminal domain, MK (62-104) has a long flexible hairpin loop
(Iwasaki et al., 1997). The tail portions located outside the domains do not form stable
structures, and the two domains are relatively independent.
Structural requirement for heparin-binding
MK is known to form cross-linkages by the action of transglutaminase and forms
non-covalent dimers in presence of heparin (Kojima et al., 1995a; Kojima et al., 1997).
The major heparin-binding site of MK is located on the C-terminal domain on the long
13
INTRODUCTION
hairpin loop. Basic residues on the β-sheet of the C-terminal domain form two basic
clusters which are another heparin-binding domain. It is noteworthy that the basic residues
involved in cluster 1 and cluster 2 except for Arg 89, are conserved in MK and HARP.
Upon binding to heparin, MK (62-104) forms a dimer in a head-to-head manner.
At the interface of the dimer, a fused heparin-binding surface is formed by basic residues
clusters on the long hairpin loop (Iwasaki et al., 1997). Both the N- and C-domains are
involved in dimerization since it has been demonstrated that free N-domain acts as an
inhibitor of dimerization (Kojima et al., 1997).
N-domain
C-domain
Figure 2: Three dimensional domain structures of MK. Dashed lines in C-domain indicate the
two-heparin binding sites (from Muramatsu, 2002).
BIOLOGICAL ACTIVITIES
MK has diverse activities. It promotes the growth of certain fibroblast cells,
neuro-ectoderm-derived cells, Wilms’ tumor cells and keratinocytes (Muramatasu, 2002).
It is known to promote the plasminogen activator activity in aortic endothelial cells,
leading to increased fibrinolysis (Kojima et al., 1995b). MK is involved in various
processes in development. It has been implicated in the regulation of epithelialmesenchymal interactions such as generation of epithelial tissues and remodeling of
mesoderm (Mitsiadis et al., 1995; Kadomatsu et al., 1990). In the nervous system, MK
enhances survival and neurite outgrowth of embryonic neurons and promotes neuronal
differentiation (Michiwaka et al., 1993a,b; Muramatsu et al., 1993).
MK is correlated with tumorigenesis and tumor progression (Tsutsui et al.,
1993; Choudhuri et al., 1997; Nakagwara et al., 1995). MK mRNA expression is increased
14
INTRODUCTION
in a number of malignant tumors compared to the non-cancerous tissue including
esophageal, hepatocellular, stomach, colon, breast, thyroid, lung and prostate carcinomas,
Hodgkin’s disease, neuroblastoma, glioblastoma and other brain tumors. Mk gene has been
identified as a target gene of the Wilms’ tumor suppression gene whose deletion leads to
Wilms’ tumor (Adachi et al., 1996). In experimental systems, transfection of MK cDNA
into NIH 3T3 cells results in oncogenic transformation (Kadomatsu et al., 1997) and antisense mk oligonucleotides suppresses the growth of some tumor cells (Takei et al., 2001).
MK is also correlated with other diseases. It is involved in the migration of
inflammatory leukocytes, an important process in tissue repair and defense against
infection (Horiba et al., 2000). In plus, MK stimulates collagen and glycosaminoglycan
synthesis in dermal fibroblasts. This finding reveals that MK may be also useful in the
therapy of delayed wound healing (Yamada et al., 1997). One of the important activities of
MK is its neuroprotective activity (Michiwaka et al., 1993b); MK can prevent the
degeneration of photoreceptor cells (Unoki et al., 1994). After ischemic injury, MK is
expressed by astrocytes in cerebral cortex and contributes to the survival of injured
neurons (Yoshida et al., 1995; Mitsiadis et al., 1995). MK has been found to be
accumulated in senile plaques of the brain of Alzheimer’s disease patients and closely
related to this pathogenesis (Yasuhara et al., 1993). Furthermore, MK inhibits HIV
infection by binding to the cell-surface nucleolin, which is implicated in the attachment of
HIV particles to cells (Callebout et al., 2001).
MECHANISM OF ACTION
N-syndecans, Protein Tyrosine Phosphatase (PTP), Low-density-lipoprotein
Receptor-related Protein (LRP) and Anaplastic Lymphoma Kinase (ALK) have been
described as the high affinity receptors for MK in the litterature. Although the precise
mechanism by which MK induces a signal through its receptors is not known, it is thought
that they might be differentially used by MK for specific biological activities or they might
form a molecular complex. Furthermore, α4β1- and α1β6- integrins have been proposed as
functional receptors of MK in recent years and the possibility to co-operate with other
receptors mentioned above (Muramatsu et al., 2004). Besides these receptors, nucleolin
has been identified as a low affinity receptor for MK (Take et al., 1994).
Syndecan family comprises 4 transmembrane heparan sulfate proteoglycans.
MK binds strongly to syndecan members namely syndecan-1, -3, -4 and the binding is
mediated by the heparan sulfate chains (Mitsiadis et al., 1995; Kojima et al., 1996;
15
INTRODUCTION
Nakanishi et al., 1997a). Syndecan-4 has been proposed to be the receptor of MK in the
MK-induced neurite outgrowth and in the enhancement of fibrinolytic activity. Digestion
of target cells with heparitinase reduced these MK activities, indicating the involvement of
the heparan sulfate proteoglycans in these systems (Akhter et al., 1998; Kaneda et al.,
1996).
RPTPζ is a receptor-like protein tyrosine phosphatase ζ (PTPζ) highly expressed
in nervous system as chondroitin sulfate proteoglycan. MK binds to the chondroitin sulfate
portion of PTPζ with high affinity and to the protein core with low affinity (Maeda et al.,
1999). PTPζ mediates the MK-induced migration of neurons (Maeda et al., 1999) and
osteoblasts (Qi et al., 2001), and the survival of neurons (Sakaguchi et al., 2003).
LRP has been identified as a MK-binding transmembrane protein other than
proteoglycans (Muramatsu et al., 2000). LRP is an endocytosis receptor and a member of
LDL receptor family. This receptor family has been described as a component of the
signaling receptor complex and it is currently regarded that LRP is part of MK receptor
complex including syndecans and PTPζ. Furthermore, a recent report indicates the nuclear
targetting by MK via LRP (Schibata et al., 2002). In this report, LRP delivers MK to the
nuclear transport system. Following to the MK endocytosis via LRP, MK binds to its low
affinity receptor nucleolin in the cytoplasm and is transported to the nucleus depending on
nucleolin. This protein was initially identified in nuclei as a protein involved in ribosome
biogenesis. It was also found that the 37-kDa-laminin-binding protein precursor binds to
MK and cotranslocated with MK in the nucleus, suggesting that MK uses both nucleolin
and laminin-binding protein as shuttle proteins (Salama et al., 2001). The nuclear
targetting process is necessary for the activity of MK in the promotion of survival cells and
is depending on the active entry by LRP and nuclear transport by nucleolin. Any physical
association between LRP and nucleolin has not been detected so far.
The forth receptor described as a high affinity receptor for MK is ALK. ALK is a
transmembrane tyrosine kinase with a molecular mass of 200 kDa. The roles of MK as a
growth, survival and angiogenic factor during tumorigenesis are linked to its signaling
through ALK receptor (Stoica et al., 2002). The downstream signaling system of MK
includes PI3 kinase and MAP kinase.
The common receptors of MK with HARP; RPTPζ, ALK and nucleolin will be
detailed in the section “HARP”.
16
INTRODUCTION
2. HARP
HARP has been described by several groups as a novel heparin-binding protein
variously termed heparin-binding growth factor-8 (HBGF-8) (Milner et al., 1989), heparinbinding neurotrophic factor (HBNF) (Kovesdi et al., 1990), heparin-binding brain mitogen
(HBBM) (Huber et al., 1990), heparin-binding growth-associated molecule (HB-GAM)
(Merenmies and Rauvala, 1990), osteoblast specific factor (OSF-1) (Tezuka et al., 1990),
pleiotrophin (PTN) (Wellstein et al., 1992; Li et al., 1990) and heparin affin regulatory
peptide (HARP) (Courty et al., 1991).
HARP was originally isolated from bovine brain by using the protein purification
described for the isolation of acidic and basic FGF (Bohlen et al., 1988; Bohlen and
Gautschi, 1989). Subsequently, other groups isolated the protein from different sources:
human, chicken (Huber et al., 1990), rat (Rauvala et al., 1989; Huber et al., 1990) and
bovine brain (Kuo et al., 1992), bovine uterus (Milner et al., 1989), adult bovine brain
(Courty et al., 1991), chicken heart (Hampton et al., 1992), conditioned media of human
breast cancer cells (Wellstein et al., 1992) and bovine follicular fluid (Ohyama et al.,
1994).
2.1 GENE
2.1.1 CHROMOSAL LOCATION
The human harp gene is localized to human chromosome 7, band q33 by fluorescence in
situ hybridization (Milner et al., 1992; Li et al., 1992a) and the mouse harp gene, to
chromosome 6 (Li et al., 1992a).
2.1.2 GENE STRUCTURE
The human harp gene spans at least 65 kb and was found to be arranged in five
exons and four introns. The mRNA transcript consists of 1650 nucleotides and the encoded
protein product, consists of 168 amino acids, of which the first 32 residues appear to be a
hydrophobic signal peptide sequence (Kretschmer et al., 1993; Lai et al., 1992). The
coding area for the mature protein was found in an open reading frame (ORF) located on 4
exons. Each of the four exons in the ORF coincided with one putative functional unit of the
HARP protein (Lai et al., 1992). Exon 1 is 271 nucleotides long and appears not to be
translated since the site of initiation of translation begins as the second nucleotide of exon
2. Exon 2, 3, 4 and 5 are 115, 175, 161 and 226 bp respectively. The 32 amino acid
hydrophobic signal sequence and the first six N-terminal amino acid of the mature protein
17
INTRODUCTION
are encoded by exon 2. Exon 3 encodes 57 amino acids (7-64) including six of the ten
cysteine residues. The remaining four cysteines are encoded among the 53 amino acids
(65-118) of exon 4. Exon 5 encodes the C-terminal 17 amino acids of the mature protein,
the TAA stop codon and 172 nucleotides of the 3’ untranslated sequence including the
polyadenylation signal AATAAA (Milner et al., 1992). The mouse harp gene contains two
additionnal 5’ untranslated regions (5’-UTR), exons (U2) and (U1) in its open-reading
frame and so utilizes two alternative promoters (Sato et al., 1997).
The transcription initiation site was analyzed for the presence of putative
transcription factor binding sites. A CCAAT box is located 17 bp upstream, but no TATA
box was identified. Interestingly, the sequence CCAAT is often present in promoter region
of genes, which are associated to the development processes. A functional serum response
element, which may account for the HARP up-regulation by PDGF, two sites AP1 and one
site GT1 were also identified (Li et al., 1992a). Several putative retinoic acid receptor
responsive elements can be observed in the HARP promoter. Additionally, the locations of
MyoD, NFkB, CArG, AP2 and CBP binding sites have been identified (Kretschmer et al.,
1993). The 3’ untranslated sequence shows 3 repeats of the ATTTA motif. This sequence
is implicated in mRNA stability and has been found in the 3’ region of several oncogenes.
2.1.3 GENE EXPRESSION DURING DEVELOPMENT
Several studies have demonstrated that HARP is expressed in a variety of
neuroectodermal and mesodermal tissues during the embryonic development of mouse
(Nakamato et al., 1992), rat (Rauvala et al., 1994; Vanderwinden et al., 1992) and chicken
(Vigny et al., 1989). HARP is differentially expressed during embryonic and postnatal
development, with a peak of expression in the immediate postnatal period in brain, but
generally down-regulated after birth and present only in minimal levels in few cell types.
Initially isolated from neonatal rat brain, the polypeptide was also shown to be present in
non-neuronal tissues, including heart (Hampton et al., 1992), uterus (Milner et al., 1989),
cartilage (Neame et al., 1993), meninges (Mailleux et al., 1992), iris, testis, bone extracts
(Zhou et al., 1992) and at sites where the epithelial-mesenchymal interactions take place,
typically in the teeth, lungs, and kidneys (Mitsiadis et al., 1995). Its expression distribution
suggests important functions in tissue growth and differentiation, in neurogenesis, cell
migration, second organogenesis induction and mesenchymal-epithelial interactions.
Moreover, the presence of HARP transcripts in some adult tissues such as testicular Leydig
cells or glia and neurons indicates a physiological role during adulthood as well.
18
INTRODUCTION
Nervous System
Although development is a continous process, during the growth and maturation
of an organ, various phases can be discerned. For example, in the cerebral development of
the rat, neuronal and glial precursors continue to multiply until birth. After this phase,
extensive sprouting of axons and dendrites occur until about the 10th postnatal day, during
which time neuronal connections begin to be formed. During this phase, the concentration
of neuronal constituents rapidly increases and myelination starts 1-2 weeks after the
perinatal growth phase (Jacobson, 1978). In perinatal rat brain, HARP is strongly
expressed at the time of birth until the postnatal age of 7-10 days. It is then clearly
decreased at the age of 14-19 days after birth and is just detectable in young adults
(Rauvala, 1989). The time period of rapid increase of HARP correlates to the rapid growth
spurt of brain, when axons and dendrites grow out and thus, its developmental profile
reflects cerebral development. The gene is expressed in a cell-type specific manner in
developping rat and it is mostly expressed in cerebral cortex (Vanderwinden et al., 1992,
Rauvala et al., 1994, Matsumoto et al., 1994). The mRNA is found in neurons as well as
glial cells suggesting its involvement in neural-glial interactions during development
(Wanaka et al., 1993). It is localized in radial glial process of the rat embryonic brain,
along which differentiating neuronal cells migrate, in accordance with the neurite
outgrowth activity of the protein (Matsumoto et al., 1994). The spatiotemporal regulation
of expression of HARP in developing cortex and thalamus is associated with the
development of thalamocortical pathway (Kinnunen et al., 1999). The expression pattern
of HARP during embryogenesis was also analyzed in developping mouse. The expression
of HARP was first observed at embryo day 8.5 (E8.5) in mouse and then became restricted
to the dorsal half of the ventricular zone (Silos-Santiago et al., 1996; Fan et al., 2000).
HARP is widely distributed in the peripheral nervous sytem, strongly expressed in the
neopallial cortex, midbrain and in meninges (Mitsiadis et al., 1995). The cellular
distribution of HARP mRNA was also reported in human meninges and meningioamas
(Mailleux et al., 1992). In adult brain, its expression persists in hippocampal and cortical
neurones (Vanderwinden et al., 1992; Wanaka et al., 1993). The high levels of HARP
mRNA have been detected in neural progenitor cells of mouse ventral mesencephalon
suggesting the action of HARP in promoting dopaminergic neurons from stem cells (Jung
et al., 2004).
19
INTRODUCTION
Cartilage and bone
HARP cDNA was cloned from murine osteosarcoma cell line MC3T3 (Tezuka et
al., 1990) and up to 3.5 mg/kg protein could be isolated from bone brain (Zhou et al.,
1992). Neame et al. isolated it from bovine nasal and fetal epipheseal cartilage (Neame et
al., 1993). In the developing rat bone, HARP is expressed in the cartilage template. HARP
can also be seen in the regions known as the sites where osteoblast precursors are recruited
for deposition of osteoid (Imai et al., 1998).
Organ and tissues undergoing epithelial-mesenchymal interactions
Vertebrate organs are typically composed of two dissimilar tissues: epithelial
and mesenchymal tissues. Inductive and reciprocal interaction between these two tissues,
that is often paracrine, is necessary for development and function of epitheliomesenchymal organs. HARP has been detected in the mesenchyme of several organs that
are formed through this interaction. An extensive study shows the presence of HARP in
facial processes and limb buds, at the sites of the hair and whisker follicules, in sense
organs, respiratory, digestive and skeletal systems (Mitsiadis et al., 1995). Isolated from
bovine uterus (Milner et al., 1989), bovine follicular fluid (Ohyama et al., 1994) and pig
uterine luminal fluid (Brigstock et al., 1996), HARP is therefore present in the urogenital
system.
HARP has been detected in various mammary cell lines isolated from breast,
prostate, muscle and ovary. In mammary breast tissue, the cellular localization has been
found in the normal human mammary gland (Fang et al., 1992; Ledoux et al., 1997;
Garver et al., 1994). In this tissue, HARP mRNA was localized in alveolar myoepithelial
cells. Interestingly, it has been found that the mRNA and the protein were not always colocalized. Although HARP mRNA was only expressed by myoepithelial cells, and alveolar
epithelial cells were negative for HARP mRNA, the protein has been localized in an area
including both alveolar epithelial and myoepithelial cells. This observation suggests that
HARP may be released from myoepithelial cells and act via a paracrine mechanism on the
adjacent epithelial cells. In the stroma, both HARP mRNA and protein were detected in
endothelial and smooth-muscle cells of blood vessels (Ledoux et al., 1997).
The
localization of HARP protein in the human mammary gland has prompted some authors to
investigate its biological function in this organ and to investigate its presence in human
breast milk (Bernard-Pierrot et al., 2004). HARP has been characterized in milk with a
concentration significantly greater in colostrum (milk collected after 1-4 days after
delivery) than in mature milk (milk collected at least 8 days after delivery). The difference
20
INTRODUCTION
may be explained by hormonal changes during lactation, as levels of progesterone and
estrogen decrease after birth. The responsiveness of harp gene expression to hormonal
changes will be discussed in the section “Regulation of Expression”.
In situ hybridization experiments show that HARP mRNA and the protein are
present in smooth muscle cells of both myometrium and blood vessels and in capillary
endothelial cells from central endometrium of the rat uterus. HARP protein was found in
both luminal and glandular epithelium, despite the absence of its transcript, again
suggesting a paracrine mechanism of action of the molecule (Milhiet et al., 1998).
Moreover, other groups also report no detection of HARP transcript along the entire
urogenital rat system in any epithelial structure (Vanderwinden et al., 1992), whereas the
protein has been detected on the surface of epithelial cells during mouse development
(Mitsiadis et al., 1995).
In the male reproductive system, HARP protein was found to be associated with
prostate and has been localized to the fibromuscular stroma. A strong expression was
observed in prostate cancer epithelial cells, but not in normal prostate or bening prostate
hyperplasia, although the corresponding mRNAs were localized to the stromal cells in each
tissue. This observation signals once more the paracrine mechanism of HARP, from
mesenchymal cells to epithelial cells (Vacherot et al., 1999b).
The opposition between mRNA and protein expression observed for HARP has
been earlier described for MK. Using tissue recombinant experiment, Mitsiadis et al. has
described that MK expression in the mesenchyme was regulated by the adjacent epithelium
and that epithelial cells express a molecule that controls the expression of MK in the
stroma (Mitsiadis et al., 1995). A similar paracrine mechanism may also account for
HARP that is produced by mesenchyme in various tissues such as breast, prostate gland or
mammary gland and thus mediate short-range signals, influencing epithelial growth.
2.1.4 REGULATION OF EXPRESSION
Although harp gene expression levels are strictly regulated during development,
its expression levels in adults are stable and limited to a few cell types, but significantly
increased in inflammatory, in wound repair following injury or in many pathological
processes.
Harp gene is a PDGF-inducible gene and the functional serum response element in
the promoter region may account for the PDGF-stimulated up-regulation of the expression
(Li et al., 1992b). As an example, the gene transcription is induced by PDGF and under
21
INTRODUCTION
hypoxic atmosphere in primary hepatic stellate cells of the liver (Antoine et al., 2005) or in
the cells when stimulated with PDGF (Li et al., 1992b). A very striking increased level of
the expression of the gene was found within macrophages and astrocytes in areas of
developing neovasculature and in the endothelial cells of the newly formed vessels after
ischemic brain injury in adult rat (Yeh et al., 1998). The up-regulation of HARP in cells at
the injury sites follows that of PDGF, supporting again the view that HARP functions as a
downstream effector of the PDGF signaling pathways. In addition, HARP expression
markedly increased in astrocytes surrounding the lesion following the hippocampal
neuronal injury (Takeda et al., 1995), in astrocytes of hippocampus after perforated path
lesion (Poulsen et al., 2000) and in corneal neovascularization as well (Usui et al., 2004).
The gene is up-regulated in myocardial infarction and ischemic myocardium (Christman et
al., 2005a,b). Another up-regulation of HARP was observed on the surface of damaged
bone in an injury model (Imai et al., 1998) and in fracture callus in a closed fracture model
(Petersen et al., 2004), while it was undetectable in adult bone under normal conditions,
pointing to its function in fracture healing. However, upon mechanical stimulus in primary
bone cells, a rapid decrease of HARP expression was observed (Liedert et al., 2006).
Endometrium from women with endometriosis has been found to have higher levels of
HARP mRNA when compared with endometrium from women who do not have
endometriosis (Chung et al., 2002).
In BALB/c 3T3 cells, the expression of HARP was enhanced in confluent and
quiescent cell cultures, and when treated with FGF2, the expression of HARP mRNA was
strongly reduced (Merenmies and Rauvala, 1992). In contrast, serum starved NIH 3T3
cells stimulated with FGF express a marked increase in levels of HARP mRNA (Li et al.,
1992b). Moreover, in NIH 3T3 cells, HARP expression was also associated with
quiescence such that its expression peaks after 2-4 days after confluence and its expression
was suppressed when the cells were transformed by ras or other oncogenes (Corbley,
1997). A positive regulation of the gene was observed in the human embryonal carcinoma
cell line (NF2/D1) induced by retinoic acid (Kretschmer et al., 1991), whereas it failed to
induce harp gene expression in NIH 3T3 cells (Li et al., 1992b) or osteoblasts (Tamura et
al., 1994). In contrast, mk gene expression was increased in retinoic-acid induced mouse
embryonic carcinoma cells in the mid-gestation period of mouse embryogenesis. The sites
of expression of HARP overlap extensively with those of MK, except that the mk gene is
expressed earlier in development than the harp gene. A very recent manuscript has
reported that HARP and MK are not only two proteins structurally and functionally related
22
INTRODUCTION
but, mk gene regulates harp gene transcription and HARP functions downstream of MK in
development (Herradon et al., 2005). The authors have demonstrated a remarkable
increase (230 fold) in HARP expression levels in heart of the genetically deficient MK -/mouse and lesser but significant increases in spinal cord, eye, aorta, bladder and urethra,
but not in brain, testis and lung and that this remarkable increase is to compensate for the
absence of mk gene in MK-/- mice, with a high degree of organ specifity.
Hormonal Regulation
The hormonal regulation of HARP mRNA expression was examined in several
osteoblast-like cell lines and the levels of HARP mRNA were down-regulated by treatment
with a calcitropic hormone, vitamin D3, suggesting a potential role of HARP in vitamin Ddependent regulation of bone metabolism (Tamura et al., 1994). In addition, HARP mRNA
synthesis is up-regulated by dihydrotestosterone, testosterone and estrogen androgens and
inhibited by anadron, a specific androgen inhibitor, in cultured prostatic epithelial cells
(Vacherot et al., 1995). In rat uterus, variations in HARP mRNA were observed
throughout the estrous cycle, with a maximum during diestrus, the progesterone-dominated
phase of the cycle, pointing to hormonal regulation and in vivo experiments with
ovariectomized rats treated with progesterone have demonstrated the progesteronedependent up-regulation of HARP mRNAs (Milhiet et al., 1998).
Pathologies
In contrast to greatly restricted expression profile in normal adult tissues, HARP
is significantly up-regulated in most tumors. HARP was purified from conditioned media
of the highly malignant human breast cancer cells (MDA-MB-231) (Wellstein et al.,
1992). HARP mRNA is highly expressed in various human tumor cell lines including
neuroblastoma (SK-N-SH, NBL-S, SH-ST5Y) (Nakagawara et al., 1995), glioblastoma
(U87MG, U373MG, LN229) (Lu et al., 2005), pancreas cancer (Colo357, Panc89) (Weber
et al., 2000), prostate cancer (PC-3) (Fang et al., 1992), breast cancer (MDA-MB-231,
MDA-MB-361, T47Dco or Hs-578T) (Fang et al., 1992), ovarian carcinoma (A1827, PA1) (Riegel and Wellstein, 1994), lung cancer (NCI-H187, NCI-H146) (Jager et al., 1997),
choriocarcinoma (Schulte et al., 1996), melanoma (1205LU) (Czubayko et al., 1996), but
not in non-tumor cell lines (Fang et al., 1992). Elevated serum levels were detected in
patients with pancreatic (Souttou et al., 1998; Klomp et al., 2002), colon (Souttou et al.,
1998), lung (Jager et al., 2002), testicular cancer (Aigner et al., 2003) and astrocytomas
(Ulbricht et al., 2003).
23
INTRODUCTION
In a recent work, it has been shown that the expression of harp gene was
regulated by PTEN/PI3K/AKT pathway and PTEN loss led to overexpression of HARP in
vitro in cell culture systems and in vivo in mammary tumors (Li et al., 2006). As we
discussed in the introduction, PTEN is a critical tumor suppressor and reduced expression
of pten gene is frequently found in a wide variety of human tumors including glioblastoma,
as well as endometrial, prostate, lung and breast cancers. Therefore, the overexpression of
harp gene may contribute to tumorigenesis caused by PTEN loss.
In recent years, intense studies were carried out to elucidate the interplay of
different growth factors and growth modulators within prostate cancer. In a model of
human prostate cancer LNCaP cells for which HARP is an autocrine growth factor, we
demonstrated an important relation between FGF2 and HARP in regulating prostate
cancer: FGF2 induces firstly NADP(H) oxidase activation which in turn generates H2O2,
and as a result, induces HARP expression and increases protein levels in LNCap cells
(Hatziapostolou et al., 2006) (Figure 3). Both AP-1 like binding sites in the HARP
promoter were required for the up-regulation upon activation of FGF receptors (FGFRs) by
FGF2. Hence, HARP seems an important mediator of FGF2 stimulatory effects in LNCaP
cells. A similar FGF-induced up-regulation was earlier observed in PC-3 cells (Connolly
and Rose, 1998) and another recent data which reports the hydrogen peroxide-induced,
AP-1-dependent transcriptional up-regulation of human harp gene in LNCaP cells supports
perfectly the latter finding (Polytarchou et al., 2005).
Figure 3: FGF2-induced harp gene regulation in LnCaP cells. FGF2-induced FGFR activation
increases the levels of H2 O2 and results in the transcriptional activation of harp gene through AP-1
binding sites and the consequent HARP protein release (modified from Hatziapostolou et al.,
2006).
24
INTRODUCTION
2.2 STRUCTURE
The primary structure of HARP deduced from cDNA is formed of 168 amino
acids, which contains a highly hydrophobic N-terminal sequence of 32 amino acid
corresponding to its signal peptide signal. The mature form of the protein corresponds to a
136 amino acid polypeptide. The analyses of amino acid composition indicated that the
protein is rich in cationic amino acids which are 24% of the residues (20.6% lysine and
3.7% arginine) and in cysteine (7.4%). The content of tryptophan of the protein (2.9%) is
also exceptionally high (Kuo et al., 1990; Bohlen et al., 1991).
40
20
M Q A Q Q Y Q Q Q RRKFA A AFLAFIFILA A V DTAEA G K K EKPEKK V K KSDCGE W Q W S VC VPT
*
60
*
100
80
SGDC GLGTREGTRTG AECK QT M K TQRCKIPCN W K K QFG AECK Y QFQ A W G EC DLNTALK
*
120
*
*
140
*
*
*
160
TRTGSLKR ALH N AECQ KTVTISKPCG KLTKPKPQ AESKK K K KEG K K Q EK M L D
*
*
Figure 4: Amino acid sequence of HARP. Amino acid numbers are given at the top. Ten
cysteines are denoted by asteriks. The arrow indicated the beginning of the mature form of human
HARP.
The mass of HARP as determined by mass spectrometry is 15 291 mass units,
which is in close agreement with the calculated mass of 15 289 based on amino acid
sequence, suggesting the absence of any post-translational modification of the protein,
other than cleavage of signal peptide. The calculated and determined masses differ from its
apparent molecular weight (18 kDa) which was estimated by SDS-polyacrylamide gel
(Hampton et al., 1992). It is likely that the slow migration on SDS-polyacrylamide gel is
due to its high content of basic amino acid residues (Kuo et al., 1990). The isoelectric point
(pI) of the molecule calculated from isoelectric focusing carried out in polyacrylamide gels
was 7.1 (Rauvala, 1989) and appears to be close to neutrality. However, this value differs
from theoretical pI when calculated by proteomics server such as EXPASY that was found
as 9.64. This point will be mentioned in the section “Results-Chaper 4-Discussion”. In
addition to the high content of basic residues, HARP contains 10 cysteine residues. The
studies of incorporation of (14C) iodoacetamide into the molecule in presence of 6 M urea
showed no free sulfhydryl group showing the implication of all the cysteines in the
formation of disulfide bond formation (Kuo et al., 1990; Fabri et al., 1992; Seddon et al.,
1994). The internal cysteine residues -15 and -44, -23 and -53, -30 and -57, -67 and -99,
25
INTRODUCTION
and -77 and -109 are linked as disulfide bonds (Fabri et al., 1992). The disulfide pairs
make the protein rigidly constrained and these conformational restraints are likely to
account for its significant stability (e.g. low pH or organic solvents). In contrast, Hampton
et al. has reported the presence of four free cystein residues and hence, whether or not all
cystein residues are engaged in the formation of disulfide bonds remains still controversial
(Hampton et al., 1992).
So far there is no data available about the three-dimensional structure of the
protein, but the arrangement of the five disulfides suggests N- and C-terminal two domain
structures (Hulmes et al., 1993). This two domain-structure is also supported by analysis of
the harp gene structure. The NMR studies by using the recombinant protein produced in
E.coli shows that the two domain structures are composed of two β-sheet domains
connected by a flexible linker. Both of these domains contain three antiparallel β-strands.
The β-sheet domains fold independently in solution and do not have domain-domain
interaction (Raulo et al., 2005). In contrast to these domain structures, the N- and Cterminal lysine rich sequences lack a detectable structure and appear to form random coils
(Kilpelainen et al., 2000).
Figure 5: Schematic representation of structural features of HARP. (from Raulo et al., 2006)
Studies using CD and NMR spectroscopy suggest that HARP undergoes a
conformational change upon binding to heparin, and that the binding occurs primarily to
the β-sheet domains of the protein. Reduction of the disulfide bonds dramatically reduced
26
INTRODUCTION
heparin binding, in agreement with the NMR results suggesting that the native protein
structure is essential for the binding (Kilpelainen et al., 2000). The heparin-binding sites of
HARP and the structural basis of the interaction will be discussed in detail in the section
“Receptors-Interactions-Mechanism of Action-Cell Surface Proteoglycans”.
Search of sequence databases shows that the β-sheet domains of HARP are
homologous to the thrombospondin type I repeats (TSR) (Kilpelainen et al., 2000). To
date, 41 proteins that contain TSRs have been found in human genome (Venter et al.,
2001) and all of them were in secreted proteins or in the extracellular portion of
transmembrane proteins. The finding that the β-sheet domains of HARP mediate heparin
binding agrees with the view that these domains correspond to the TSR motifs, because
several TSR repeats have been shown to bind heparin and heparan sulfate. In general,
TSR-containing proteins are implicated in cellular migration, communication, inhibition of
angiogenesis, GAG binding and inhibition of MMPs. The TSR domain consists of ∼60
amino acids, of which 12 are highly conserved. Most notably, they contain two or three
tryptophan residues separated by two to four amino acids and the majority of TSRs
possesses six cystein residues (Adams and Tucker, 2000). W∗∗W∗∗W (asteriks indicate
positions that are occupied by various amino acid) sequence has been proposed for GAG
binding site (Guo et al., 1992). The arrangement of intradomain cysteine bridges in HARP
is consistent with the predicted TSR motif and the amino acid sequences in the postulated
TSR domains show the conserved cysteine/tryptophan motif. Nevertheless, despite their
overall topology, the TSR motifs in HARP display apparently significant differences.
HARP has been produced as recombinant protein in mammary cells,
baculovirus, yeast, insect and Escherichia coli. It is now known that different forms of the
protein exist. The analysis of the NH2-terminal amino acid sequence of human
recombinant protein produced by NIH 3T3 transfected with human cDNA indicates the
protein with a N-terminal extended by 3 amino acids (AEA), as compared to tissue purified
HARP (Laaroubi et al., 1994). Lately, two naturally occuring forms of the protein with
migratory masses of 18 and 15 kDa have been characterized in the conditioned media of
glioblastoma cells (U87MG). The shorter form corresponds to a truncation of the 12 amino
acids (KKEGKKQEKMLD) from the C-terminal of HARP (Lu et al., 2005) and is most
probably generated by post-translational cleavage of the longer form.
27
INTRODUCTION
Homologies
HARP shares approximately 50% homology in amino acid sequence with MK
(Merenmies and Rauvala, 1990; Kovesdi et al., 1990). The C-domain in two proteins
displays the highest similarity and is evolutionary the most conserved domain (Figure 6).
1
5
15
10
20
25
30
35
40
HARP
MK
GKKEKPEKKVKK----SDCGEWQWSVCVPTSGDCGLGTREGTRTGAE
KKKD----KVKKGGPGSECAEWAWGPCTPSSKDCGVGFREGT-----
HARP
MK
CKQTMKTQRCKIPCNWKKQFGAECKYQFQAWGECDLNTALKTRTGSL
CGAQTQRIRCRVPCNWKKEFGADCKYKFENWGACDGGTGTKVRQGTL
HARP
MK
KRALHNAECQKTVTISKPCGKLTKPKPQAESKKKKKEGKKQEKMLD
KKARYNAQCQETIRVTKPCTPKTKAKAKAKKGKGKD
45
50
95
55
100
60
105
65
110
70
115
75
120
80
125
85
130
90
135
Figure 6: Structural comparison of human HARP and human MK. Conserved amino acids are
boxed.
HARP is found in many species from Drosophila to human and the structure of
HARP is highly conserved between species throughout evolution. The cDNA deduced
amino acid sequences of pre-HARP (168 residues) from human (Li et al., 1990;
Kretschmer et al., 1991), bovine (Li et al., 1990), mouse (Tezuka et al., 1990) and rat
(Merenmies and Rauvala , 1990; Li et al., 1990; Kovesdi et al., 1990) reveals the strikingly
high inter-species sequence conservation (98%), suggesting important biological functions
of the protein. For example, HARP from human and mice differ in only one amino acid
and all 10 cysteine residues are conserved in all vertebrates. In Drosophila, two members
of HARP/MK family have been characterized, named Miple1 and Miple2 from Midkine
and Pleiotrophin (Englund et al., 2005). The overall similarity of Miple1 and Miple2 with
human HARP is found as 60% and 58%, respectively. Miple1 and Miple2 consist of
domains that show a high degree of homologies to the C-domain of HARP, while they are
less within the N-domain. One of the most important differences is the lack of one pair of
cysteine and of the thrombospondin repeat (TSR) domain related to N-domain of the
protein. Importantly, amino acid residues implicated in binding to heparin are conserved in
Miple1 and Miple2.
28
INTRODUCTION
2.3 RECEPTORS-INTERACTIONS-MECHANISM OF ACTION
The term “receptor” refers to a molecule on the cell membrane or within the
cytoplasm or cell nucleus that binds a ligand and initiates the cellular response of the
ligand. Taking in account of this definition, HARP receptors can broadly be divided into
two types. The first class includes several cell surface components as potential receptors
and/or “co-receptors” such as heparan sulfate proteoglycans of syndecan, chondroitin
sulfate proteoglycan of receptor-type tyrosine phosphatase β/ζ (RPTPβ/ζ), and the second
class includes a “classic” high-affinity signaling tyrosine kinase receptor, Anaplastic
Lymphoma Kinase (ALK). In addition, the cell surface-expressed nucleolin which lacks a
cytoplasmic domain has been more recently identified as a receptor of HARP.
2.3.1 CELL SURFACE PROTEOLYCANS (PGs)
Heparin / Heparan Sulfate (HS)
HARP presents a strong affinity towards heparin. It was initially isolated from
brain and eluted from heparin-Sepharose by high salt concentration of ∼1 M NaCl.
Different methods were used to examine heparin interaction with HARP. One of them is
isothermal titration calorimetry (ITC), which is a thermodynamic, solution phase
measurement of the heat of interaction. ITC studies showed that, in solution, heparin binds
HARP with a dissociation constant of 460 nM and the stoichiometry of HARP / heparin
interaction is 3 mol of HARP / mol of heparin. Kinetic measurements made by surface
plasmon resonance (SPR) yielded a Kd of 4 nM, showing considerably higher binding
when HARP was immobilized on a surface (Fath et al., 1999). The HARP-heparin
interaction was further characterized by fast association kinetics. In these binding analyses
with biosensor, the Kd constant was found as 13 nM with an association rate constant kass =
1.6 x 106 M-1 s-1 and a dissociation rate constant kdiss = 0.68 x 106 s-1 (Vacherot et al.,
1999a). In order to define the GAG structure(s) responsible for HARP binding,
competitive binding assays with 3H-labeled heparin oligosaccharides were used. These
studies suggested a minimum heparin-binding site of 10 saccharides with an optimum
binding between 14 and 18 saccharide residues (Kinnunen et al., 1996), and hence
indicated the occurence of a distinct binding site for HARP within heparin.
The binding of HARP to heparin is thought to be mainly mediated by the two
TSR domains (Kilpelainen et al., 2000). Surface plasmon resonance analyses have
suggested that the individual TSR domains (N-TSR (13-58) and C-TSR (65-110)) bind
weakly to heparin and that the linker region (59-64) between the domains is not important
29
INTRODUCTION
for high affinity binding (Raulo et al., 2005). The dissociation constants from SPR
measurements were found as 13 µM for C-TSR and >70 µM for N-TSR, displaying a
dramatic change when compared to Kd of HARP (176 ± 37 nM). This result suggests that
co-operative action of the two TSR domains is essential for strong binding. Although the
lysine-rich N (1-12)- and C (111-136)-terminal tails are highly charged, they do not
contribute to the binding of HARP to heparin and the di-TSR (13-111) fragment showed
essentially the same affinity to heparin as the intact HARP (Kd = 153 ± 40 nM) (Raulo et
al., 2005).
A number of techniques have been used to characterize quantitatively
interactions between HARP and heparin. However, care is required for evaluation of
differences data regarding what is likely representative of actual binding events and what
might be subject to artifact. For example, ITC, measures interactions between molecules in
solution, where heparin is polydisperse. Similarly, filter trap methodology uses surfaces
only to capture the preexisting complexe. SPR and resonant mirror biosensor measure
interactions between molecules where the protein is in solution and heparin immobilized
on a surface. Such immobilization will reduce the degrees of freedom of heparin and will
affect the kinetics of the interaction. However, immobilization may be more representative
of the in vivo situation where HS chains are covalently attached to proteins as HSPGs
(Powell et al., 2004). Moreover, biosensor measurements, if done properly, will measure
the highest affinity class of binding sites. Therefore, one can reconcile the 100-fold or
more difference between these measurements. However, SPR measurements are subject to
considerable artifact. The optical biosensor measurements and the analysis of the data need
to incorporate specific quality controls to ensure measurements are not limited by diffusion
(mass transport), steric hindrance and other artefacts (Schuck et al., 1997; Fernig, 2001;
Rich and Myszka, 2006). From these constraints, one can suggest that the filter binding
assays (Kinnunen et al., 1996) may not produce fully quantitative data. The SPR biosensor
measurements of Fath et al. are reasonable in terms of establishing the Kd; the kinetic
measurements of Vacherot et al. are in accord with these, suggesting that these values
represent the intrinsic binding parameters of the interaction of HARP with its higher
affinity binding sites in heparin (Fath et al., 1999; Vacherot et al., 1999). However, the
measurements of Raulo et al. are not supported by any indication of experimental quality
control and, as suggested in a recent review in the field (Rich and Myszka, 2006), these
should be ignored. These considerations also impact on our understanding of the sites of
30
INTRODUCTION
interaction of the polysaccharide in the HARP polypeptide, which will be discussed at
length in the section “Results-Molecular Basis of HARP-heparin Interactions”.
The polysaccharide heparan sulfate (HS) is an abundant component of cell
surfaces and the pericellular matrix, including basement membranes (Gallagher et al.,
1986; Kjellen and Lindahl, 1991). HS is composed of α1-4 linked dissacharide repeating
units containing an uronic acid and amino sugar. Heparin is composed of the same
carbohydrate structures as HS, but is significantly higher in sulphate content and it lacks
the caracteristic non-sulphated segments of HS. Furthermore, heparin is not a component
of the cell surface (Gallagher et al., 1986). Nevertheless, heparin shares many of the
interactive properties of HS and because of its abundance, commercial availability and
perceived simplicity, the majority of experimental studies have utilized heparin (Delehedde
et al., 2001). The main dissacharide unit (75%) in heparin is iduronic acid (IdoA), 2SGlcNSO3 (N-sulphated glucosamine), 6S (Figure 7A). Experiments with selectively
desulfated heparins indicated that all 3 sulfate residues of the major dissaccharide unit of
heparin, in particular, 2-O-sulfated iduronic acid units, in a lesser extent glucosamine Nsulfate and 6-O-sulfate groups contribute to the interaction with HARP. Totally Odesulfated and N-desulfated N-reacetylated heparins didn’t bind to HARP (Figure 7B)
(Kinnunen et al., 1996). The importance of iduronic acid is that its sugar ring is flexible,
allowing substantial conformational change that appears to be essential for GAG-protein
interaction (Hricovini et al., 2001).
6-O-sulfate
A
AA
de-6-O-sulfated
B
2-O-sulfate
N-sulfate
de-2-O-sulfated
de-N-sulfate
Figure 7: Disaccharide repeating units of A) totally sulfated, B) totally desulfated heparins.
The arrows indicate N- and O-sulfate substitutions.
31
INTRODUCTION
Release of HARP from cells
Heparan sulfate is usually present on the surface of cells or in the ECM
(extracellular matrix), in the form of a proteoglycan. HARP was found to be associated
with heparan sulfate proteoglycans of the basement membrane/extracellular matrix of
MDA-MB-231 cells, MC 3T3 cells and NIH 3T3 cells overexpressing the protein. HARP
can be released from cells by an excess of soluble heparin and heparan sulfate, or by
enzymatic digestion of HSPGs by heparitinase and heparinase. The ineffectiveness of
chondroitinase treatment of the cells for the release of the extracellularly-stored HARP
indicates also that HARP is almost exclusively associated with HS on the cell surface.
Furthermore, other GAGs, dermatan sulfate and chondroitin sulfate A could displace
HARP from cells, showing the interaction of HARP with these GAGs (Vacherot et al.,
1999a).
Dermatan Sulfate (DS) / Chondroitin Sulfate (CS)
The affinity of HARP for the glycosaminoglycan DS also designated as CS-B
(Kd 51 ±14 nM) was found to be lower than that of heparin (Kd 13 ± 3 nM) (Vacherot et
al., 1999a).
The interactions of various CS preparations with HARP were analyzed using the
BIAcore system. HARP binds strongly to CS with E unit ((GlcAβ(2S)1-3GalNAc(6S))
with a Kd of 0.76 nM, to CS-D ((GlcAβ1-3GalNAc(4S,6S)) with a Kd of 2.7 nM,
moderately to CS-C, and ∼45 fold less to CS-B (Kd = 34 nM) in comparison with CS-E. In
contrast, CS-A showed no binding (Maeda et al., 2003). The CS 16∼18 mer has been
found to be the basic functional unit to interact with HARP. The authors have proposed
that one CS chain binds with two HARP units immobilized on the sensor chip and the
affinity was increased by multivalent binding of long CS to multiple HARP. The affinity
of CS for HARP is highly dependent on the structural variations of this glycosaminoglycan
and the oversulfated portion in CS contributes to the strong binding. Furthermore, it has
been suggested that the heparin binding sites of HARP also serve as CS binding sites
(Maeda et al., 2006).
Dimerization of HARP in presence of GAGs
The ability of HARP to dimerize (oligomerize) was investigated using chemical
cross-linking experiments (Bernard-Pierrot et al., 1999). In these experiments,
dissuccinimidyl suberate (DSS) was used as a cross-linker, known to link two molecules
through the primary amines. Incubation of HARP with various concentrations of heparin
32
INTRODUCTION
led to the formation of HARP dimers with a maximum dimerization at 10 µg/ml heparin,
corresponding to a molar ratio of 1:1 between HARP and heparin. The involvement of
other GAGs such as DS, CS-A, CS-C, HS and keratan sulfate (KS) were also tested. The
addition of these GAGs also resulted in the dimerization of HARP but to a weaker extent
than observed with heparin. No dimers were detected in presence of KS. Moreover, the
presence of secreted HARP dimers, in the conditioned medium of NIH 3T3 overexpressing
the growth factor is suggestive to the possible physiological relevance of this process. No
soluble dimers were detected without DSS, showing that HARP forms non-covalent
dimers. Chlorate treatment of the cells significantly decreased the amount of dimers,
therefore, the dimerization process is dependent on the sulfated GAGs (Bernard-Pierrot et
al., 1999). The ability of HARP to dimerize had already been suggested by Zhang et al.,
but in this work the mechanism of the dimerization was described through disulfide
interactions (Zhang et al., 1997). Similar to MK, HARP also is known to multimerize by
action of transglutaminase (Maeda, unpublished observation).
In a recent work, the CS-induced dimerization of HARP was further analyzed
using the CS oligosaccharides of various chain lengths. In the presence of the 10-mer
fraction, no dimers were observed. The 24-mer fraction induced dimer formation and in the
presence of 46-mer fraction, formation of trimers and tetramers was additionally observed
(Maeda et al., 2006).
As discussed in detail in “Cell signaling by receptor kinases”, ligand-induced
dimerization is necessary to activation of tyrosine kinase receptors and for monomeric
ligands, the dimerization of its receptors can be either occur by inducing a conformational
change in the receptor or by GAGs forming dimeric ligands. A similar ligand-induced
dimerization mechanism seems to underpin the mitogenic, angiogenic, transforming and
tumor formation activities of HARP and dimeric HARP to be biologically active form for
activating its receptors (Zhang et al., 1997; Bernard-Pierrot et al., 2002). It has been shown
that exogenous GAGs at concentrations inducing dimerization enhanced HARP
mitogenesis (Vacherot et al., 1999a). Interestingly, other groups have reported the
inhibitory effects of similar concentrations of GAGs; such that heparin inhibited HARPinduced neurite outgrowth of thalamic cells cultured on HARP matrix (Kinnunen et al.,
1999), and CS-C inhibited HARP-induced neuronal migration (Maeda and Noda, 1998). In
contrast, very low concentrations (10-30 ng/ml) of heparin were shown to inhibit the
neurite outgrowth activity induced by substrate-bound HARP (Kinnunen et al., 1996). The
inhibitory effect of heparin was lost upon selective 2-O-desulfation which resulted in
33
INTRODUCTION
shifting the dose-reponse curve 103-fold higher concentration. 6-O-desulfation and Ndesulfation followed by N-acetylation resulted in an IC50 102-fold higher that that of
heparin. In addition, a minimum of 10 monosaccharides was required for inhibition of
HARP-induced neurite outgrowth as well as for HARP binding (Kinnunen et al., 1996).
Taken together, all these data suggest that the mitogenic and neurite outgrowth activities of
HARP may occur via different mechanisms (Bernard-Pierrot et al., 1999). Other groups
have reported that the addition of heparin (0.1-1 µg/ml) has reversed the growth-arresting
role of HARP in the mesenchymal and epithelial cells of the developing rat limb,
suggesting that HARP may bind to a heparin-like carbohydrate epitope required for cell
proliferation in this system (Szabat and Rauvala, 1994).
2.3.1.1 Heparan Sulfate Proteoglycans: Syndecans
The initial studies on receptors for HARP found that the protein bound to a
specific receptor with a Kd of 8 nM and receptor numbers of 1.7x105 and 1x104 for 3T3
and PC12 cells, respectively (Kuo et al., 1990). Scatchard plot analysis of the binding of
125
I-labelled HARP to NIH 3T3 cells revealed a heparin-inhibitable binding of HARP to a
140-kDa cell surface component of NIH 3T3 cells (Kuo et al., 1992) and another report
indicated the phosphorylation of a ∼200 kDa protein in NIH 3T3 and NB41A3 cells, upon
stimulation with HARP (Li and Deuel, 1993). Lately, Souttou et al. have reported 190 and
215 kDa phosphoproteins in BEL cells in response to the HARP stimulu (Souttou et al.,
1997). However, so far there are no follow up data available on the identification of these
receptors.
The first receptor identified for HARP was the 200-kDa cell surface heparan
sulfate proteoglycan N-syndecan (syndecan-3) (Raulo et al., 1994). This protein was
isolated from cultured brain neurons and from brain as a detergent-solubilized component,
using the baculovirus-produced HARP as an affinity matrix. N-syndecan was found to
contain heparan sulfate chains, which are bound to a core protein with an apparent
molecular mass of 120 kDa. In a solid phase binding assay, N-syndecan was found to bind
to HARP with a Kd of 0.6 nM and the binding was inhibited by soluble heparin (Raulo et
al., 1994), hence indicating that binding was mediated by the heparan sulfate chains.
Structural analysis of N-syndecan from rat brain has indicated N-sulfated dissacharide
units with an unusual high proportion (82%) of 2-O-sulfate iduronic acid residues
(Kinnunen et al., 1996). Therefore, this carbohydrate sequence in N-syndecan has been
proposed to function as HARP-binding sites. Experiments with selectively desulfated
34
INTRODUCTION
heparins showed that elimination of IdoA 2-O-sulfate groups, in particular, yielded a
product which lost dramatically its ability to bind HARP (Kinnunen et al., 1996). In
addition, it is noteworthy that FGF2 in solution has competed for the binding, suggesting
that HARP and FGF2 bound to a similar carbohydrate structures in N-syndecan (Raulo et
al., 1994).
N-syndecan is strongly expressed in developing nervous tissues (Carey et al.,
1992) and a temporal co-expression for HARP and this receptor during brain development
has been demonstrated (Nolo et al., 1995). N-Syndecan has been proposed to be the
receptor of HARP in its neurite growth- promoting activity as N-syndecan inhibited
HARP-induced neurite outgrowth (Kinnunen et al., 1996). The inhibitory effect of Nsyndecan was attributed to the HS chains as it was abolished by treating N-syndecan with
heparitinase. N-syndecan mediates HARP-induced neurite growth signals via the Srccortactin pathway (Kinnunen et al., 1998).
Syndecans have structurally variable extracellular domains. Although syndecan
isoforms carry predominantly HS side chains, syndecan-1 and syndecan-4 (ryudocan) also
contain CS chains. A strong binding of HARP to these syndecans has been also reported
(Mitsiadis et al., 1995; Deepa et al., 2004). The kinetic parameters for the binding were
analyzed using a SPR biosensor. The observed Kd values were 27 nM for syndecan-1 and
16 nM for syndecan-4 (Deepa et al., 2004). Although a complete loss of FGF2 binding of
these syndecans was observed upon the removal of HS chains, it has been shown that
HARP and MK were still capable of interacting with CS/syndecan-1 and -4 devoid of HS.
In addition, the core protein of syndecan-1 was showed to participate to the binding of
HARP and MK, cooperating with the CS or HS chains or both. It has been proposed that
the CS chains accelerate the association and dissociation of HARP, MK and FGF2 from
the HS chains and suggests that they form a quaternary complex with the HS chains, the
growth factor and the core protein and transfer the growth factor to its cell surface receptor
(Deepa et al., 2004). Furthermore, in many epithelial-mesenchymal systems, HARP, MK,
FGFs and syndecan-1 are frequently co-localized, suggesting molecular interactions during
development (Mitsiadis et al., 1995).
In a similar way, another hybrid proteoglycan that carries mainly HS but also
CS chains, perlecan, was reported as a binding molecule for HARP (Peng et al., 1995).
35
INTRODUCTION
2.3.1.2 Chondroitin Sulfate Proteoglycans: Phosphacan
The chondroitin sulfate proteoglycan, 6B4 proteoglycan/phosphacan corresponds
to the extracellular region of a receptor protein tyrosine phosphatase beta (RPTPβ), also
known as protein tyrosine phosphatase zeta (PTPζ). RPTPβ/ζ is composed of a large
amino-terminal extracellular domain which carries CS chains, of a single transmembrane
domain and of an intracellular domain with tyrosine phosphatase activity (Levy et al.,
1993). Alternative splicing produces three isoforms of RPTPβ/ζ: a transmembrane fulllength form, a truncated transmembrane short form with a deletion in the extracellular
region and a secreted extracellular variant (Figure 8) (Nishiwaki et al., 1998).
Figure 8: Three isoforms of RPTPβ/ζ . Long RPTPβ/ζ, short RPTPβ/ζ and phosphacan (from
Lorente et al., 2005).
HARP seems to be the first soluble ligand to be identified for any of the
transmembrane class (receptor-type) tyrosine phosphatases. It has been shown that HARP
binds to the extracellular variant of RPTPβ/ζ, phosphacan (Maeda et al., 1996). Scatchard
analysis revealed the presence of two binding sites with different affinities (Kd = 3 nM and
0.25 nM). The binding of phosphacan to HARP depends in part on the CS portion of the
proteoglycan, since the removal of CS by enzymatic cleavage resulted in a substantial
decrease (60%) in the binding affinity of both sites without changing the number of
binding sites (Maeda et al., 1996). Using a BIAcore system, the importance of D unit
structure of CS in the binding affinity of phosphacan was demonstrated (Maeda et al.,
36
INTRODUCTION
2003). However, the affinity of phosphacan core protein for HARP was found to be
relatively high (1.5∼13 nM) (Maeda et al., 1996, 2003), so similar to that of HARP for the
CS chains, suggesting a cooperative interaction among the core protein, CS and HARP.
Recently, the heterogeneity of the chondroitin sulfate structure of RPTPβ/ζ in the
developing brain has been demonstrated. Analysis of the interaction between HARP and
various phosphacan preparations revealed that the differences in the chondroitin sulfate
structure on phosphacan markedly influence the binding affinity for HARP, hence
explaining the presence of different affinity binding sites (Maeda et al., 2003).
The HARP/RPTPβ/ζ signaling pathway presents a unique mechanism of receptor
signaling, which may be called “ligand-dependent receptor inactivation”. Upon binding
HARP, RPTPβ/ζ was thought to dimerize, leading to the inactivation of the intrinsic
tyrosine phosphatase activity (Meng et al., 2000). Recently, the dimeric (oligomeric)
inhibition model has been confirmed by Fukada et al. in vivo (Fukada et al., 2006). The
cytoskeletal proteins β-catenin (Meng et al., 2000) and β-adducin (Pariser et al., 2005b)
and Src family member Fyn (Pariser et al., 2005a) are the downstream targets of
HARP/RPTPβ/ζ signaling and HARP coordinately regulates the tyrosine phophorylation
levels of these RPTPβ/ζ substrates.
RPTPβ/ζ is predominantly expressed in the central nervous system (CNS) (Levy
et al., 1993) and is the neuronal receptor of HARP, involved in the migratory processes of
cortical neurons induced by HARP since polyclonal antibodies against the extracellular
domain of PTP, an extracellular secreted form of PTP, a protein tyrosine phosphatase
inhibitor and chondroitin sulfate C but not CS-A, added into culture medium inhibited
HARP-induced neuronal migration (Maeda and Noda, 1998). Moreover, RPTPβ/ζ and
HARP are overexpressed in human glioblastomas (Muller et al., 2003, Ulbricht et al.,
2003). HARP promotes haptotactic migration of gliablostomas through RPTPβ/ζ (Lu et
al., 2005). RNAi knockdown studies have established the functional role of RPTPβ/ζ in
HARP-induced glioblastoma cell motility (Ulbricht et al., 2006).
Although little is known about the functional role, another chondroitin sulfate
proteoglycan, the neurocan, synthesized by neurons, has been identified as a ligand of
HARP (Kd = 0.3-8 nM) (Milev et al., 1998).
37
INTRODUCTION
2.3.2 ANAPLASTIC LYMPHOMA KINASE
Anaplastic Lymphoma Kinase (ALK) is a receptor-type transmembrane tyrosine
kinase (RTK) of the insulin receptor superfamily (Morris et al., 1997). ALK was first
discovered as a chimeric nucleophosmin (NPM)-ALK fusion protein in anaplastic large
cell lymphomas (ALCL) (Morris et al., 1994). NPM-ALK contains the N-terminal domain
of NPM fused with the C-terminal cytoplasmic (catalytic) domain of ALK.
ALK cDNA encodes a protein of approximately 200 kDa with post-translational
modifications and consists of 1620 amino acids, of which 1030 amino acids are in the ECD
region (Pulford et al., 2004). Two major species of ALK were identified (220 and 140
kDa) and the 140 kDa species resulted from a cleavage of the full-length receptor, leading
to the release of the 80-kDa fragment into the culture medium (Moog-Lutz et al., 2005).
ALK is known to be activated through ligand-induced dimerization like other
RTKs. However, ALK has been regarded as an orphan receptor since the ligand(s) to
enforce dimerization and autoactivation was unknown. HARP was proposed to be the
ligand of ALK as a result of studies in which a phage display cDNA library was screened
against immobilized HARP (Stoica et al., 2001). In this study, the HARP binding site of
the receptor has been found at residues 391-401 and the apparent Kd as 32 ± 9 pM.
The same authors have also been reported that HARP-induced mitogenesis
results in an increase of the phosphorylation of ALK substrates and that IRS-1, Shc, PLC-γ
and PI3-kinase are the signal transducing molecules that are phosphorylated in response to
HARP (Stoica et al., 2001). These finding are in good agreement with those of Souttou et
al. who reported that the mitogenic activity of HARP is dependent on tyrosine
phosphorylation and utilizes MAP kinase and the PI3-kinase pathways (Souttou et al.,
1997). The same authors have later demonstrated that PI3-kinase also participates to the
proliferation response to HARP in endothelial cells, but the angiogenic response was not
dependent on eNOS pathway (Souttou et al., 2001). In contrast, the anti-apoptotic
signaling of HARP through ALK in fibroblastic NIH 3T3 cells was dependent on MAPK
pathway but not PI3-kinase (Bowden et al., 2002). In glioblastomas, ALK is overexpressed
together with HARP and the ribozyme targetting of ALK prevents HARP-stimulated
phosphorylation of the anti-apoptotic protein Akt (Powers et al., 2002). It appears that
different forms of HARP mediate distinct functional effects through different receptors in
glioblastomas. C-terminally processed HARP15 stimulation through ALK activates a
mitogenic signal transduction pathway via Akt or MAPK, whereas in the same cells
HARP18/RPTPβ/ζ induced haptotactic cell migration (Lu et al., 2005). The downstream
38
INTRODUCTION
pathways of these two HARP receptors seem to represent independent pathways leading to
dissimilar effects.
HARP has been proposed as a potential ligand of ALK and has been shown to
interact with the extracellular domain of ALK (Stoica et al., 2001; Bernard-Pierrot et al.,
2001), but this has not been confirmed by other groups (Dirks et al., 2002; Miyake et al.,
2002; Motegi et al., 2004) and still remains controversial. For example, in yet another
study, in which two anti-ALK monoclonal antibodies effectively activated ALK, a
mitogenic form of HARP, under the same conditions, failed to activate this receptor
(Moog-Lutz et al., 2005). In addition, in Drosophila, the protein jelly belly (Jeb) has been
identified as a ligand of ALK (Englund et al., 2003; Lee et al., 2003) and Jeb is clearly a
distinct protein from the Drosophila homolog of HARP, Miple1.
A very recent work presented a unique model of ALK activation (Perez-Pinera
et al., 2007). In this study, the authors demonstrated that HARP stimulates tyrosine
phosphorylation of ALK through HARP/RPTPβ/ζ signaling pathway and the
phosphorylation of ALK is independent of a direct interaction of HARP with ALK. ALK is
dephosphorylated by RPTPβ/ζ in cells not stimulated by HARP and is apparently a
substrate of RPTPβ/ζ. In contrast, the stimulation of cells by HARP leads to the
inactivation of RPTPβ/ζ. The inactive RPTPβ/ζ can no longer dephosphorylate the
autophosphorylated sites in ALK, thus the tyrosine phosporylation of ALK rapidly
increases. The mechanism proposed in this work suggests an alternative mechanism of
RTK activation and is the first demonstration for: a cytokine-dependent inactivation of a
tyrosine phosphatase is the mechanism to activate a transmembrane receptor kinase (PerezPinera et al., 2007).
2.3.3 NUCLEOLIN
In addition to its action through signaling receptors, HARP also can work
directly in cells. HARP has been reported to bind to cell-surface nucleolin (Said et al.,
2005). Nucleolin is a component of the cell surface implicated in the attachment of human
immunodeficiency virus (HIV) particles to cells and functions as a shuttle protein between
the cytoplasm and nucleus. Following its binding to surface-nucleolin, HARP is
internalized by an active process independent of HS and CS proteoglycans. HARP inhibits
HIV attachment to cells by its capacity to bind cell-surface nucleolin and the β-sheet
domains of HARP, especially located on C-terminal domain (60-110) have been found to
39
INTRODUCTION
be responsible for the inhibitory effect of HARP on HIV attachment (Said et al., 2005).
Furthermore, anti-nucleolin antibodies prevent the binding of HARP to its cell-surface
receptor and abolishes its inhibitory potential on HIV infection (unpublished observation).
2.4 BIOLOGICAL ACTIVITIES
2.4.1 MITOGENIC ACTIVITY
The cell growth effects of HARP have been described for a variety of cell lines.
HARP has been found to be a mitogen for endothelial cells (Li et al., 1990, Courty et al.,
1991, Fang et al., 1992, Laaroubi et al., 1994, Yeh et al., 1998), epithelial cells (Fang et
al., 1992, Delbé et al., 1995, Jager et al., 1997), fibroblastic cells (Milner et al., 1989, Li et
al., 1990, Fang et al., 1992), a subpopulation of hepatocytes (Sato et al., 1999),
oligodendrocytes and glial cells (Rumsby et al., 1999). Nevertheless, the mitogenic
potential of HARP has been challenged and has remained controversial for long time
because of the conflicting results of different groups (Kuo et al., 1990, Kretschmer et al.,
1991, Hampton et al., 1992, Raulo et al., 1992, Takamatsu et al., 1992). Several
laboratories have reported that the protein from insect cells using recombinant baculovirus
or from bacteria was mitogenically inactive but was still exhibiting neurite outgrowth
activity (Seddon et al., 1994, Raulo et al., 1992). Other investigators have attributed the
cellular growth activity to the co-purification of the protein with mitogenic proteins
including FGFs in heparin-Sepharose affinity chromatograhy (Raulo et al., 1992, Hampton
et al., 1992). Several experimental approaches were used to exclude the hypothesis of such
contamination by other growth factors. An enzyme immunoassay study showed no
detection of FGF1 and FGF2 in the mitogenically active form of HARP isolated from
bovine brain (Courty et al., 1991). To further rule out the possibility of contamination of
HARP fractions by FGF2, growth assays using bovine epithelial lens (BEL) cells for
which FGF2 is a potent mitogen, were carried out (Souttou et al., 1997). The authors have
reported that only the FGF2 activity was neutralized in presence of anti-FGF2 antibodies in
the cells stimulated with FGF2 and HARP. Furthermore, HARP, produced in a lung cell
line, which expresses neither FGF2 nor FGF1, was found mitogen and an anti-HARP
antibody specifically inhibited the HARP-induced growth of SW13 cells (Jager et al.,
1997). Taken together, it is now clear that the mitogenic activity of HARP is independent
of FGF2. The lack of the mitogenic activity or the limited activity of HARP obtained from
different sources may be due to the incorrect folding or improper (incomplete) processing
40
INTRODUCTION
of the recombinant protein from bacterial expression systems when highly expressed and it
seems that only recombinant forms of HARP produced in mammalian expression systems
display cellular growth activity.
As described in the section “Structure”, distinct molecular forms of HARP exist.
Firstly, Laaroubi et al. have reported that NIH 3T3 cells transfected with HARP cDNA
produce two forms of HARP, distinct in the structure and properties; the wild-type protein
and a protein, which was amino-terminally extended by 3 amino acids (AEA) (Laaroubi et
al., 1994). This extended form represented 90% of the preparation of the biologically
active recombinant protein and was found to be non-mitogenic. In plus, the same extended
form has been detected in the conditioned medium of non-transfected BEL and NIH 3T3
cells, pointing out the physiological relevance of the finding. However, in another work,
the same authors have reported that both molecular forms were found mitogenic for NIH
3T3 and BEL cells. Hence, the presence of these 3 amino acids cannot be considered as the
main determinant for the mitogenic activity (Bernard-Pierrot et al., 2001).
Recently, Lu et al. have reported the presence of two naturally occurring forms
of HARP with migratory masses of 15 and 18 kDa in the conditioned medium of
glioblastomas (Lu et al., 2005). The mitogenic activity of purified HARP18 and HARP15
was tested on serum-starved glioblastomas. The authors here utilized HARP produced in
human HEK 293T cells transfected with the wild type and truncated HARP constructs.
Their result suggests that the mitogenic activity of HARP in glioblastomas is mostly
dependent on the smaller HARP15 form and not the full length. This finding supports an
earlier work in which the mitogenic activity was associated with a C-terminally truncated
form of HARP (14 kDa) (Souttou et al., 1997). Taken together, the reports from different
groups suggest that the purified HARP is a mixture of mitogenic and non-mitogenic
proteins and that the mitogenic activity depends on the balance between the active and
non-active form(s) (Souttou et al., 1997). Interestingly, the mitogenic potential of the Cterminally intact, full-length HARP (18 kDa) has been previously reported (BernardPierrot et al., 2001, Zhang et al., 1999), but, it is worthy to note that these studies have
utilized mouse cells to express HARP in contrast to the others that used human cells. Thus,
one can speculate that the protein modifications which may vary from one clone to another
and even the type of mammalian expression system used for various HARP preparation
may influence the mitogenic activity of HARP (Bernard-Pierrot et al., 2001).
In contrast to its cell growth effect, recombinant protein has been found to show
the cell growth arrest in the mesenchymal and epithelial cells of the developing rat limb
41
INTRODUCTION
(Szabat and Rauvala, 1994). Subsequent studies showed that the chondrocyte proliferation
was also inhibited upon exposure to HARP and the synthesis of proteoglycan synthesis
was increased (Tapp et al., 1999).
2.4.2 DIFFERENTIATION PROMOTING ACTIVITY
Neurite outgrowth:
In contrast to the mitogenic activity of HARP, the
recombinant HARP expressed in insect cells or in bacteria exhibits neurite outgrowth
promoting activity for different cultured neuronal cell types, including embryonic
(Rauvala, 1989) and perinatal (Hampton et al., 1992) cortical neurons, neuroblastoma cells
(Kuo et al., 1990), PC12 cells (Li et al., 1992), rat embyonic neurons in primary culture
(Rauvala et al., 1994; Kinnunen et al., 1998) and thalamic neurons (Kinnunen et al., 1999).
The neurite promoting activity was observed only when the growth factor presented as a
matrix-bound form (Raulo et al., 1992).
HARP is required for the spontaneous differentiation of oligodendrocyte
progenitors in primary culture (Rumsby et al., 1999) and stimulates lineage-specific
differentiation of glial progenitor cells. HARP is implicated in the differentiation of the
endplates in muscle cells and in synapse formation at the developing neuromuscular
synapse, suggesting a role in postsynaptic differentiation at the developing neuromuscular
junction (Peng et al., 1995). In plus, this role is compatible with its expression in the
differentiating neuromuscular system (Szabat and Rauvala, 1994). Recently, HARP has
been described as a key molcule in the regulation of the stem cell differentiation (Hienola
et al., 2004, Jung et al., 2004). More recently, HARP has been shown to mediate the
effects of ciliary neurotrophic factor (CNTF) on retinal progenitor differentiation (Roger et
al., 2006). HARP is also involved in the regulation of hippocampal long-term potentiation
(Lauri et al., 1998; Pavlov et al., 2002).
Because HARP stimulates different progenitor cells to enter lineage-specific
differentiation pathway, the expression of HARP by activated monocytes/macrophages in
ischemic tissue has been questioned (Sharifi et al., 2006). Monocytes/macrophages are
known by their property to display a high degree plasticity and it has been demonstrated
that HARP induces the transdifferentiation of monocytes into functional endothelial cells.
Recently, it was discovered that HARP was required for the normal
differentiation of both cathecholamine (Ezquerra et al., 2004) and angiotensin (Herradon et
al., 2004) pathways during development.
42
INTRODUCTION
2.4.3 ONCOGENIC ACTIVITY
Dysregulated expression in tumors: As described in detail in the section “Gene”, the
constitutively high levels of HARP were found in a number of human tumor cells as well
as in tumor samples, suggesting an important role of HARP in dysregulated tumor growth
and that the constitutively expression may be essential to the malignant phenotype.
Transformation: HARP is a proto-oncogene. The oncogenic activity of HARP during
tumor growth was firstly established by the work of Chauhan et al. (Chauhan et al., 1993).
The introduction of the exogenous harp gene into non-tumorigenic NIH 3T3 (Chauhan et
al., 1993) or SW-13 adrenal carcinoma cells (Fang et al., 1992) resulted in the transformed
phenotype as anchorage-independent growth in soft agar and tumor formation in athymic
nude mice. It was also reported that HARP purified from the conditioned medium of the
human breast cancer cell line MDA MB-231 enhanced growth of SW-13 in soft agar as
well as normal rat kidney fibroblast (NRK) (Wellstein et al., 1992). The autocrine activity
of HARP is apparent from experiments with SW13 cells transfected with HARP. The
paracrine activity of HARP was demonstrated in studies in which HARP is harvested from
the supernatants of transfected cells and added back onto different cells to study soft agar
stimulating effects (Fang et al., 1992).
To directly assess whether the expression of HARP in tumor cells plays a
significant role for their malignant growth, the HARP mRNA and protein levels were
reduced in melanoma cells (Czubayko et al., 1994), in pancreatic cancer cell lines (Weber
et al., 2000), in glioblastomas (Grzelinski et al., 2005) by ribozyme-targeting. Depletion of
HARP by ribozymes resulted in a dramatic reduction of tumor growth and reverted the
transformed phenotype. These results suggest that HARP is a rate-limiting tumor growth
factor for these cells.
2.4.3 ANGIOGENIC ACTIVITY
Several studies have demonstrated that purified recombinant protein was
mitogenic for a variety of endothelial cells including human umbilical vein endothelial
cells (HUVEC), bovine brain capillary endothelial cells (Courty et al., 1991, Fang et al.,
1992, Laaroubi et al., 1994). To assess the significance of HARP for angiogenesis and
metastasis, the metastatic human melanoma cells (1205LU) were transfected with HARPtargeting ribozymes (Czubayko et al., 1996). In parallel with the reduced HARP mRNA, in
nude mice, tumor growth and angiogenesis were decreased as well as the rate of metastasis
and the latter, presumably by inhibiting tumor angiogenesis. In plus, the decreased
43
INTRODUCTION
angiogenesis due to decreased levels of HARP caused more cells into apoptosis. These
studies show a direct link between tumor angiogenesis and metastasis through HARP and
suggests that HARP is rate-limiting for human melanoma metastasis (Czubayko et al.,
1996). The in vivo angiogenic role for HARP was confirmed by Choudhuri et al. in a
rabbit corneal assay (Choudhuri et al., 1997). Experimental proof that HARP can act as an
angiogenic growth factor was further supplied from in vivo studies in which cells that
overexpress HARP were injected into athymic nude mice and in this model, HARP caused
the development of highly vascular tumors (Chauhan et al., 1993). In an in vitro model of
HUVEC, HARP induced EC migration and the formation of capillary-like structures by
HUVEC grown as 3D-culture in Matrigel (Souttou et al., 2001). In addition, the same
effects in vitro were observed using HARP secreted from prostatic LnCAP cells and the
angiogenic capability was further confirmed in the in vivo CAM assay, suggesting an
angiogenic role of tumor-derived HARP on prostate cancer (Hatziapostolou et al., 2005).
HARP localization also corroborates with its role in angiogenesis. As largely
discussed in the section “Gene”, HARP mRNA was detected in endothelial cells from
human mammary glands (Ledoux et al., 1997), rat endometrium (Milhiet et al., 1998),
human prostate (Vacherot et al., 1999b). Additionally, HARP was detected in human
plasma after intravenous heparin administration indicating the presence of HARP on the
endothelial cells (Novotny et al., 1993). Furthermore, HARP was shown to be upregulated
in macrophages, astrocytes and in newly forming blood vessels after cerebral ischemia and
an exuberant response of new blood formation was observed at the sites where it was upregulated (Yeh et al., 1998). Furthermore, when injected into ischemic myocardium,
HARP displayed more completely and more fully developed neovasculature, compared to
other angiogenic factors (Christman et al., 2005b). HARP also regulates the reninangiotensin pathway, a system known to have a critical role in angiogenesis (Herradon et
al., 2004).
Angiogenic switch is to switch the premalignant phenotype to a highly malignant
phenotype and is characterized by exuberant tumor growth and abundant new blood vessel
formation. Recently, it has been shown that the constitutive activation of harp gene into
premalignant SW-13 cells initiated an angiogenic switch in vivo (Zhang et al., 2006).
44
INTRODUCTION
2.4.5 MIGRATION
HARP is also functional in cell-cell interaction and migration (Kurtz et al., 1995)
and stimulates osteoblastc migration (Kinnunen et al., 1998). Using two cell migration
assay systems; in vitro with glass fibers and Boyden chambers, HARP induced cell
migration of cortical neurons (Maeda and Noda, 1998). Cell migration is classified into
several types, i.e. chemotaxis, haptotaxis and the Boyden chamber assay results suggest
that HARP induces neuronal migration by haptotaxis. Furthermore, HARP induces the loss
of cell-cell adhesion via degradation of cytoskeletal proteins and increases mobility and
invasiveness in HARP-stimulated cells (Perez-Pinera et al., 2007b).
2.5 FUNCTIONAL DOMAINS OF HARP
Separate independent domains in HARP have been identified to signal different
functions. To determine domain(s) of HARP functionally important for neoplastic
transformation, 12 mutants of HARP were tested for their transforming potential (Zhang et
al., 1999). It was established that HARP amino acids 41-64 were required for HARPdependent transformation and these residues were designated as the “transforming
domain”. In addition, either but not both the N- or C-terminal lysine rich domains were
required to enable the residues 41-64 to initiate transformation. Because the amino acids in
the two termini are very different, it was speculated that it was the net positive charge in
these two domains that establishes the requirement for transformation (Deuel et al., 2002).
Other structural basis that was uncovered in this study was the regulatory role of the amino
acid residues 41-45 and 64-68 which are the internal repeats (GAECK) since they were
able to suppress the transformation potential of HARP (Zhang et al., 1999). Previously, the
authors have reported the transforming activity of HARP was blocked by using a truncated
HARP (1-40) designed to heterodimerize with the endogenous product of the gene in a
model of breast cancer cells, MDA MB-231, suggesting that the dimeric form of the
protein was required for mediating the transforming activities.
In contrast to N-terminal “transforming” domain, the C-terminal of HARP 69136 has been termed as the “angiogenic domain” and it was sufficient to trigger an
angiogenic switch as effectively as wild type HARP, in a nude mouse model (Zhang et al.,
2006). Another important point that these studies indicate is that, the N- (transforming) and
C- (angiogenic) terminal domains, that both contain heparin-binding thrombospondin
domains, function independently, or in other words, the angiogenic activity functions
independently of the transforming activity (Deuel et al., 2002; Zhang et al., 2006). This
45
INTRODUCTION
conclusion is compatible with a recent study of Raulo et al. that reported that the two βsheet domains fold independently and that they do not interact each other in solution
conditions (Raulo et al., 2005).
Other authors have investigated the domain(s) responsible for neurite outgrowth
and synaptic plasticity and for this, TSR domains of HARP were tested, because previous
studies suggested that these activities of HARP were mediated through heparin-binding
domains of HARP. While the individual TSR domains; N-TSR (13-58) and C-TSR (65110) failed to produce these biological effects, both TSR domains (13-111) of HARP, but
not the lysine-rich tails were sufficient for the activity in neurite outgrowth and LTP
inhibition, suggesting the requirement of the co-operative action of the two TSR domains
(Raulo et al., 2005). In contrast, the regulation of neuronal migration seems to be different
in the mechanism since separate C-terminal TSR domain (65-110) was as active as di-TSR
(13-111) domain and wild type HARP (1-136) (Raulo et al., 2005).
Bernard-Pierrot et al. have reported the requirement of the C-terminal residues
for the mitogenic and tumor-forming activities of HARP, since the HARP mutant (1-111)
displayed none of these activities. In contrast, the HARP mutant lacking the last 26 amino
acid residues from C-terminal was still exhibiting the neurite outgrowth activity, hence
showing clearly that the structural determinants involved in HARP neurite outgrowth and
mitogenic activities were different (Bernard-Pierrot et al., 2001, Kuo et al., 1990).
Subsequently, these authors have shown the dominant negative effect of this HARP mutant
for its transforming and angiogenic activities, in addition to the mitogenic and tumorforming activities, by forming non-functional heterodimers with wild type HARP.
Moreover, a peptide corresponding to the deleted part of HARP, P111-136, displayed the
inhibition of wild-type HARP activities in vitro, but the proposed mechanism, in this case,
was the competition of the synthetic peptide with HARP for binding to ALK receptor
(Bernard-Pierrot et al., 2002). Another synthetic peptide corresponding to the C-terminal
TSR motif of HARP, P65-97, has been also shown to prevent the mitogenic, transforming
and angiogenic activities of HARP, but this time, by competing with HARP for binding to
its GAG co-receptors, by virtue of the ability of the peptide to bind heparin (HammaKourbali et al., 2007).
46
INTRODUCTION
N-lysine rich
1
12
TSR domain I
41
58
Linker
64 65
TSR domain II
69
C-lysine rich
111
Neurite Outgrowth (1)
N-TSR(13-58)+C-TSR(65-110)
Neuronal Migration (1)
C-TSR (65-110)
Transforming domain (41-64)
136
Mitogenic
Angiogenic
Tumor Growth (4)
C-lys(111-136)
Angiogenic domain (69-136) (3)
+ N- or C-lys rich (2)
Figure 9: Functional domains of HARP. (1) Raulo et al., 2005, (2) Zhang et al., 1999, (3) Zhang
et al., 2006, (4) Bernard-Pierrot et al., 2002
47
INTRODUCTION
3. AIMS OF THE THESIS
HARP is a growth factor with unusual and diverse functional activities. HARP
mediates different functions through different receptors. In addition, separate independent
domains and/or co-operative action of separate domains have been identified in HARP to
signal independently. HARP is also known for its mitogenic, tumor promoting and
angiogenic activities.
Determining how HARP signals the diverse functions is a challenging task and
structural studies are necessary to provide a better understanding of the HARP-mediated
functional responses. Such studies promise to provide insight also into HARP-involvement
in dysregulated growth, malignant phenotype of several tumor cells and in angiogenesis.
During the last decade, synthetic peptides have found expanded interest as tools
for research in biological and biomedical sciences and their
use represent a powerful
approach for structure/function analysis. In this approach, peptides corresponding to
individual domains or segments of proteins are synthesized and tested for their agonist and
antagonist activities, thereby helping to identify functional domains within a protein.
Synthetic peptides derived from the sequence of biologically active proteins are also useful
as tools to study protein-protein interactions and can be used as potential therapeutic agents
to combat cancer development.
This thesis concerns the elucidation of the structure-function relationship of
HARP by means of synthetic peptides, and characterization of the bioactive peptides
displaying antagonist action for the mitogenic, tumor promoting and angiogenic activities
of HARP. The work performed over a period of three years will be presented in three
chapters:
1. The first chapter will focus on the C-terminal end of HARP. The results from
the structure function studies carried out with a synthetic peptide P111-136 from the last
26 amino acid residues of the protein and with a smaller peptide P111-124 that has been
identified to mimic the function of P111-136, will be presented in the form of paper, which
will be submitted to Molecular Cancer Therapeutics, and of unpublished results, and a
discussion will be included at the end of the chapter.
2. In the second chapter of the thesis, in pursuit of the findings of the first
chapter, a synthetic peptide P122-131 corresponding to the basic cluster of residues at the
C-terminus of HARP will be studied and the results will be presented in the form of two
48
INTRODUCTION
papers, one, published in Cell Experimental Research and the other, submitted to Journal
of Carcinogenesis and will be completed with a discussion combining the results obtained
from the two articles.
3. The third and last chapter of the thesis is divided in two sub-chapters. The first
part attempts to elucidate the affinity of HARP and its functional domains for binding to
heparin and the structural requirements in heparin for these interactions. Finally, the ability
of HARP to interact with other heparin-binding proteins will be studied in the second part.
This thesis has yielded new informations over the mechanism of action of
HARP. At the end of the thesis, a “conclusion and suggestions for the future works” part
will be included and new models will be proposed for the possible signalization of HARP,
through its diverse receptors, as the basis of its multifunctional and regulatory properties.
49
MATERIALS AND METHODS
MATERIALS AND METHODS
1. MATERIALS
ANTIBODIES
Goat polyclonal anti-human HARP antibodies were purchased from R&D
(Oxon, United Kingdom). Polyclonal rabbit anti-PTPζ (H-300) antibodies were from Tebu
Bio (Le Perray, France). Rabbit polyclonal anti-ALK antibodies were from Zymed
Laboratories Inc. (South San Francisco, CA). Rabbit polyclonal antibodies against the
extracellular domain of human ALK were a gift from Pr. Marc Vigny (Université Pierre et
Marie Curie, France). All secondary antibodies were purchased from Jackson
ImmunoResearch (Suffolk, UK).
CELLS
DU145 cells were a generous gift of O. Cussenot (Hôpital St. Louis, Paris,
France). All other cell lines were purchased from the American Type Culture Collection
(Manassas, VA). They were maintained in a humidified atmosphere of 95% air and 5%
CO2, at 37°C. MDA-MB-231, PC3 and NIH 3T3 cells were grown in DMEM containing
10% fetal bovine serum
(FBS) and the antibiotics; penicilline (100 units/ml) and
streptomycin (100 µg/ml). CHO K1 cells were cultured in Ham’s F-12 medium
supplemented with 10% FBS, U87MG cells were cultured in MEMα supplemented with
10% FBS and DU145 cells were maintained in RPMI containing 10% heat-inactivated
serum (56°C 30 min) and 50 µg/ml gentamicin. All culture media and antibiotics were
supplied by Invitrogen (Cergy Pontoise, France). Fetal bovine serum was product of PAA
Laboratories (Les Mureaux, France).
CHEMICALS
Heparin-Sepharose gel and Mono-S column were from Amersham Biosciences.
All electrophoresis reagents were from Bio-Rad (Marne-La-Coquette, France). [methyl3
H]thymidine was provided by ICN (Orsay, France). (NHS) amino carporate (LC)-biotin
was from Pierce and Warriner (Chester, UK). Cell Counting Kit-8, ImmunoPure®TMB
substrate kits, BCA™ protein assay and Superblocker® solution were from Pierce
(Rockford, USA). BM chemiluminescence was obtained from Roche (Meylan, France).
Autoradioagraphic Kodak Biomax Light-1 films as well as BSA, agar, porcine intestinal
mucosal heparin and streptavidine were purchased from Sigma (Saint Quentin Fallavier,
France). Peptide synthesis grade solvents and other reagents were obtained from Applied
Biosystems (Courtaboeuf, France).
51
MATERIALS AND METHODS
PROTEINS AND PEPTIDES
Recombinant HARP was produced in the laboratory from bacteria (Seddon et
al., 1994) and conditioned media of eucaryotic cells (Laaroubi et al., 1994), according to
the procedure previously described. Recombinant extracellular domain of human ALK was
a gift from Pr. Marc Vigny (Université Pierre et Marie Curie, France). HGF/SF, NPN-1,
VEGF121, VEGF165, VEGFR1 were purchased from R&D systems (Abington, UK).
Recombinant FGF2 was produced as described (Ke et al., 1992). FGF-BP was a generous
gift of Dr Christian Heegaard and was purified according to the procedure as previously
described (Lamestsch et al, 2000). FGFR was produced as described (Duchesne et al.,
2006). Recombinant human S100A4 was a generous gift of Drs Thamir Ismail and Roger
Barraclough and was produced as previously described (Zhang et al., 2005).
The peptides indicated in table 1 were synthesized by Altergen (Schiltigheim,
France).
Name
Sequence
Molecular Weight
P111-136
KLTKPKPQAESKKKKKEGKKQEKMLD
3054
P111-124
KLTKPKPQAESKKK
1610
P1-21
AEAGKKEKPEKK
1600
P13-39
SDCGEWQWSVCVPTSGDCGLGTREGTR
2900
P65-97
AECKYQFQAWGECDLNTALKTRTGGSLKRALHNA
3630
Table 1: Synthetic peptides. Name, sequence and molecular weight (g/mol).
52
MATERIALS AND METHODS
2. METHODS
2.1 SOLID-PHASE PEPTIDE SYNTHESIS
Solid-phase peptide synthesis (SPPS), pionnered by Merrifield, is a process by
which chemical transformations are carried out on a solid support (Merrifield, 1963). The
peptide is covalently attached to the amino group of the insoluble resin through its Cterminal and hence the synthesis proceeds in a C-terminal to N-terminal fashion. Peptides
are synthesized by coupling the carboxyl group of one amino acid to the amino group of
another. There are two majorly used forms of SPPS: Boc (tert-butyloxycarbonyl) and
Fmoc (9-fluorenylmethyloxycarbonyl). The Boc and Fmoc are base-label protecting
groups for amines. The main difference between Fmoc- and Boc chemistry SPPS is the
reaction conditions for the deprotection of the α-amino group of the last coupled amino
acid. In Fmoc-chemistry, the removal of Fmoc group is carried out with bases like
piperidine to give a free amino group. In Boc-chemistry, however, the deprotection with
TFA leads to a protonated α-ammonium species which has to be neutralized before the
next amino acid is coupled, hence it uses in situ neutralization (i.e. neutralization
simultaneously with coupling) (Schnölzer et al., 1992).
The Boc-chemistry SSPS consists of the following steps in each cycle:
1. removal of Nα-Boc protecting group of the last coupled amino acid by treatment
with 100% TFA,
2. neutralization of the CF3COO-.NH3+ of the peptide-resin salt with DIEA,
3. coupling of the next pre-activated Boc-amino acids.
Between each of these steps, a flow wash with DMF is carried out. The simple
chemistry requires the activation of Boc-protected carboxyl groups of amino acids in a
polar solvent, HBTU, in combination with DIEA. Excess amount of DIEA is used to
activate the protected amino acid with concurrent neutralization of the peptide-resin
(Alewood, et al., 1997). Final cleavage from the resin as well as deprotection of side chain
protecting groups is achieved using strong acid, HF, while, in Fmoc chemistry this step is
achieved by incubating in TFA.
The manual in situ neutralization synthesis protocol described above has been
adapted for machine-assisted synthesis on the peptide synthesizer (Alewood et al., 1997).
53
MATERIALS AND METHODS
2.1.1 SYNTHESIS
The sequences of peptides synthesized are:
Name
Sequence
P122-131
(N-acetylated) CONH2- KKKKKEGKKQ-CONH2
Biot- P122-131
Biotin- GGGGKKKKKEGKKQ-CONH2
BiotD-P122-131
Biotin- GGGGKKKKKEGKKQ-CONH2 (D-amino acids)
Table 2: Synthesis of P122-131 and biotin-labeled forms. The synthesis was done in
collaboration with Laboratoire de Synthèse, Structure et Fonction de Molécules BioactivesUniversité Paris 6.
Peptides were synthesized by SPPS, using Boc or Fmoc strategies. MBHA
(methylbenzylhydrylamine) was used as a resin. The ten amino acids KKKKKEGKKQ
were coupled using machine-assisted protocol with Applied Biosystems 433A synthesizer
(Courtaboeuf, France), while the coupling of the linker (GGGG) and the labeling with the
biotin were performed using manual protocol.
2.1.2 PURIFICATION AND CHARACTERIZATION
The purification of the synthesized peptides was carried out by HPLC (High
Performance Liquid Chromatography) on RP-C8 column using a linear gradient of solvent
B/A (B = CH3CN/H20/TFA; 60/39.9/0.1, A = H2O/TFA; 99.9/0.1) over a period of 30 min.
After verifying the purity of fractions by analytical HPLC in isocratic conditions, fractions
with purity >95% were collected and lyophilized. Peptides were characterized with
MALDI-TOF MS (Voyager Elite, PerSeptive Biosystems) on matrix HCCA (α-cyano-4hydroxycinnamic acid). The m/z values of the protonated peptides were 1269 for P122-131
and 1683 for Biot-P122-131.
2.2 PRODUCTION AND PURIFICATION OF RECOMBINANT HARP
2.2.1 PRODUCTION OF HARP FROM BACTERIA
E. coli bearing the human HARP18-pETHH8 plasmid was cultured at 37°C in
LB media and expression of HARP was induced for 2 h at 37°C by addition of 2 mM
isopropylthiogalactoside. Cells from a 4 L-culture were centrifuged, resuspended in 30 ml
of 50 mM Tris-HCl, pH 7.4, 0.1 mM EDTA. The bacteries were lyzed by 4 cycles of
congelation and defrosting steps, the lysate was clarified by centrifugation and the
54
MATERIALS AND METHODS
inclusion bodies were solubilized in 30 ml of 50 mM Tris-HCl, pH 7.4, containing 10 mM
dithiothreitol, 0.1 mM EDTA, 1 M NaCl and 8 M urea. The insoluble material was
removed by centrifugation and the supernatant was dialyzed against 4 L of 25 mM HEPES,
pH 7.4, 1 M NaCl for 8 h, followed by a second 8-h dialysis against 4 L of 25 mM HEPES,
pH 7.4, at 4°C. Insoluble materiel formed during dialysis was removed by centrifugation
and the urea-solubilized protein was buffered to pH 7.4 with 20 mM Hepes, ionic strength
adjusted to 0.5 M NaCl.
Recombinant HARP15 was produced from E.coli bearing HARP15-pETHH8
plasmid and produced as described above.
2.2.2 PRODUCTION OF HARP FROM CONDITIONED MEDIA
HARP was purified from NIH 3T3 cells stably transfected by pCDNA3-HARP
plasmid, as previously described (Laaroubi et al., 1994). Cells were plated in 600-cm2
dishes and cultured for 72 h in complete medium. Conditioned medium (1 L) of 8 × 106
cells containing secreted HARP was buffered to pH 7.4 with 20 mM Hepes, ionic strength
adjusted to 0.5 M NaCl.
2.2.3 PURIFICATION OF HARP
The purification consists of two steps: a first purification by affinity
chromatography based on the heparin affinity of HARP and a cation-exchange
chromatography by FPLC in which highly positively charged ions of HARP bind to the
negatively charged resin.
Buffered HARP was loaded onto a 10-ml heparin-Sepharose column. Proteins
bound to heparin were eluted with 20 mM Hepes, 2 M NaCl, pH 7.4. The eluate was then
adjusted to 50 mM Tris-HCl, pH 7.4 and 0.4 M NaCl and subsequently purified using a
cation-exchange Mono-S column. The purification was carried out in 50 mM Tris-HCl, pH
7.4 and proteins were eluted using a 0.4 to 2 M NaCl gradient.
2.3 SAMPLE PREPARATION FOR SDS-PAGE
The cells were lyzed in TNE buffer (10 mM Tris-HCl pH 8.0, 150 mM NaCl, 1%
(v/v) NP-40, 2 mM EDTA, 1 mM sodium orthovanadate, 50 units/ml aprotinin, 0.45 mM
phenylmethylsulfonyl fluoride and 10 mM sodium fluoride) when native structure was
important in further analysis or when quantitation of proteins in samples by BCA assay
was required, or lyzed directly in sample buffer (5% (v/v) β-mercaptoethanol, 2% (v/v)
55
MATERIALS AND METHODS
SDS, 10% (v/v) glycerol, 10 mM Tris-HCl, pH 6.8, 1.25% (w/v) bromophenol blue) in ice.
The media were removed from all wells and each well was washed with PBS. The
appropriate lysis buffer was added and wells were scraped. After incubation for 15 min in
ice, protein lyzates were vortexed and insoluble materials were removed by centrifugation
for 15 min at 4°C.
In some cases, HARP-containing sample was concentrated by heparin-Sepharose
beads. The sample was buffered to pH 7.4 with 50 mM Tris, 1 mM EDTA and ionic
strength was adjusted to 0.5 M NaCl. The anti-proteases aprotinin, peptin, leupeptin
(×1000) and phenylmethylsulfonyl fluoride were added. The buffered sample was
incubated with 80 µl beads in 50 mM Tris pH 7.4 for 20 ml of solution overnight at 4°C.
The tubes were then centrifugated for 1 min at 4°C (5000 × g), the precipate was washed
twice in 50 mM Tris, pH 7.4, 0.5 M NaCl and once, in 50 mM Tris, pH 7.4. The precipitate
was resuspended in 40 µl of sample buffer for Western Blot or in 50 mM Tris for ELISA.
2.4 GEL ELECTROPHORESIS-SDS-PAGE
The proteins of the samples are separated by electrophoresis in a polyacrylamide
gel. The anionic detergent, SDS, denaturates secondary and non-disulfide-linked tertiary
structures. Sampled proteins become covered in the negatively charged SDS in proportion
to its mass and move to the positively charged electrode.
Besides SDS, proteins were optionally heated to 95°C for 5 min in sample buffer
to further denaturate the proteins by reducing disulfide linkages, according to the protocol
described by Laemmli (Laemmli, 1970) and were centrifuged for another 5 min. The
electrophoresis was performed at 150 volt (constant voltage) for 45 min using an apparatus
Mini-Protean II (Bio-Rad).
Electrophoretically separated polypeptides were visualized by stains like
Coomassie Blue, Colloidal or silver, or by immunochemical detection methods performed
after electrophoretic transfer (“blotting”).
2.4.1 COOMASSIE BLUE AND SILVER STAINING
Silver Staining
This protocol describes a method for silver staining proteins in a polyacrylamide
gel, modified from Wray et al. (Wray et al., 1981). The silver ions bind to the side chains
of the amino acids and are subsequently reduced to metallic silver by development
56
MATERIALS AND METHODS
solution. It is possible to detect as little as 0.1-1 ng of protein and this method is 100 to
1000 times more sensitive than Coomassie Blue staining.
Following to fixing step by placing it in a solution of 50% (v/v) methanol, 50%
(v/v) water, the gel was soaked in a freshly made ammoniacal silver nitrate solution for 15
min. To prepare 100 ml of this solution, 0.8 g of silver nitrate was dissolved in 4 ml of
distilled water, which was slowly mixed into a solution containing 73.75 ml of water, 21
ml of sodium hydroxide (NaOH) (0.36 N) and 1.25 ml of ammonium hydroxide (NH4OH)
(14.8 M). The gel was rinsed by successive washings for 30 min. The bands were
developped in a solution containing 5% (v/v) citric acid and 19% (v/v) formic acid, until
the bands were appeared. The development was stopped with 1% (v/v) acetic acid.
Coomassie Blue (R 250) Staining
The Coomassie dyes bind to proteins through interactions between dye sulfonic
acid groups and positive amine groups as well as through Van der Waals attractions. This
staining can detect as little as 0.1 µg of protein. The method is suitable for rapid detection
of proteins.
The gel was stained with 0.25% (w/v) Coomassie Blue in 10% (v/v) acetic acid,
50% (v/v) methanol and 40% (v/v) water (the solution was filtered through Whatman#1
paper) for thirty minutes and destained in 5% (v/v) acetic acid, 10% (v/v) methanol and
85% (v/v) water for 30 min or more, until background was clear.
Colloidal Coomassie Blue (G 250) Staining
The sensitivity of staining is about 10 ng/band. The method is especially suitable
in case of further analysis of bands for mass spectroscopy. If gels are supposed to be used
for this purpose, it is absolutely necessary that staining is performed in closed and clean
boxes so that keratin contamination is avoided as good as possible. The protocol is
modified from that of Neuhoff et al. (Neuhoff et al., 1988).
The gel was fixed in a solution containing 48.5 ml of water, 50 ml of ethanol and
1.5 ml of orthophosphoric acid (85%, v/v) for 2 h to overnight, in an orbital shaker. To
remove excess fixative, the gel was washed three times for 30 min per wash, each with
distilled water. The gel was equilibrated in the incubation solution for 1 h. To prepare 100
ml of this solution, 17 g of ammonium sulfate was dissolved in a maximum of distilled
water in which 34 ml of methanol was slowly mixed by small additions. 15 ml of 85%
(v/v) orthophosphoric acid was then added and the remaining water was added to the
solution. The gel was stained with Coomassie Blue G250 powder (0.67g/L) in the
57
MATERIALS AND METHODS
methanolic solution with orthophosphoric acid and ammonium sulfate. For optimal
staining, the gel was incubated for 3 to 4 days.
The gel was stored in 5% (v/v) acetic acid solution at 4°C, until in-gel digestion
and mass spectroscopy were performed.
2.4.2 WESTERN BLOTTING (IMMUNOBLOTTING)
Following electrophoretic separation by SDS-PAGE, proteins from the gel were
transferred onto a membrane by electrophoretic transfer and immunodetected.
Ponceau S Staining (optional): Transferred membranes may be stained with Ponceau S to
facilate location and identification of specific proteins, orientation of the blot. Ponceau S
can provide helpful landmarks such that one can mark the position of molecular weight
standards, trim away excess membrane more exactly and verify the efficacy of the transfer.
The membrane was stained in Ponceau S staining solution (0.5% in 1% (v/v) acetic acid,
w/v) for 5 min and was destained for an additional 5 min in water until the red staining
fades.
HARP: Proteins were transferred to a PVDF membrane using an apparatus Trans-Blot
(Bio-Rad) at 30 volt overnight at 4°C, or at 100 volt for 1 h, in 10 mM of 3(cyclohexylamino)-1-propane sulfonic acid (CAPS), pH 11, containing 10% (v/v)
methanol. After washing briefly in PBS, non-specific binding was prevented by incubating
the membrane in the Superblocker solution for 20 min at room temperature (RT). The
membrane was then incubated with goat anti-human HARP antibodies (250 ng/ml) diluted
in PBS containing 0.2% (v/v) Tween 20 (PBS-T) and 3% (v/v) Superblocker for 1 h at RT.
After three washes with PBS-T, the membrane was incubated with the peroxidaseconjugated anti-goat antibodies diluted to 1/100 000th in PBS-T, 3% (v/v) Superblocker for
30 min at RT. Bound antibodies were detected using the BM chemiluminescence reagent
and the bands were visualized using autoradioagraphic film.
ALK and RPTPβ: The transfer was done in the transfer buffer (25 mM Tris, pH 8.3, 200
mM glycine, 20% (v/v) methanol) overnight at 4°C. The membrane was blocked in PBS-T
containing 5% (w/v) powdered milk (Leader Price, France) for 1 h at RT. The blots were
then probed with the antibodies anti-ALK diluted to 1/7000th or anti-RPTPβ diluted to
1/200th in PBS-T containing 5% (w/v) milk overnight at 4°C. After three washings with
PBS-T, the blots were incubated with the peroxidase-conjugated anti-rabbit antibodies
diluted to 1/10 000th for ALK or to 1/50 000th for RPTPβ. After extensive washing in PBST, bound antibodies were visualized using BM chemiluminescence reagent.
58
MATERIALS AND METHODS
2.5 MEMBRANE STRIPPING
After washing five times in PBS-T, the membrane was incubated in stripping
buffer (62.5 mM Tris-HCl, pH 6.8, 2% (v/v) SDS and 100 mM β-mercaptoethanol) for 30
min at 56°C in a heating oven. The membrane was washed three times in PBS-T and the
normal protocol was followed after blocking the membrane in PBS-T, 5% (w/v) milk for 1
h.
2.6 LIGAND BLOT
The protocol is derived from the method described by Callebout et al.
(Callebout et al., 1997).
Following to separation by SDS-PAGE, samples were electrophoretically
transferred to a PVDF membrane. The electrophoretic blots were washed in TBS (0.1 M
Tris, pH 7.4, 0.15 M NaCl) for 5 min and saturated with casein-based blocking buffer
(Sigma) overnight at 4°C. The blots were incubated in blocking buffer containing 5 µM of
biotin-labeled peptides for 2 h at 4°C. The sheets were subsequently washed three times in
Tris-buffered saline containing 0.05% (v/v) Tween 20 (10 min each), followed by 2
washes in Tris-buffered saline (10 min each). Biotin was revealed by incubating the blot
with peroxidase-labeled anti-biotin diluted to 1/2000th in blocking buffer for 1 h at room
temperature and visualized using BM chemiluminescence reagent.
2.7 HEPARIN-ELISA
This assay is based on the high affinity of HARP for heparin. It consists of
adsorbing heparin-BSA covalent complexes to microtiter 96-well plate (Costar) and to
quantify the heparin-bound HARP by immunological methods. The method measures
concentrations ranging from 40-1200 pg/ml HARP (Soulié et al., 2002).
Heparin was covalently bound to BSA essentially as described by Gray and by
Najjam et al. (Gray, 1978, Najjam et al., 1997) and the solution was aliquoted and stored at
-80°C. Heparin-BSA complex equivalent to 0.25 µg BSA was dissolved in 50 mM TrisHCl, pH 7.4, 12.5 mM EDTA and 100 µl of this solution were incubated in each well of
96-well plate overnight at 4°C. The wells were washed three times with washing buffer
(PBS containing 0.05% (v/v) Tween 20) and blocked with PBS-T, 3% BSA (w/v) (300
µl/well) for 2 h at room temperature (RT). Samples (100 µl/well) were added to the wells,
and incubated overnight at 4°C. The wells were washed three times with washing buffer,
and the anti-HARP antibody (100 µl/well, 250 ng/ml) diluted in PBS-T, 1% (w/v) BSA
was added. The plates were incubated for 2 h at RT, washed three times with washing
59
MATERIALS AND METHODS
buffer and incubated with peroxidase-labeled rabbit anti-goat antibodies (100 µl/well)
diluted to 1/10 000th in PBS-T, 1% (w/v) BSA for another 2 h at RT. The plates were again
washed three times with washing buffer. Peroxidase activity was detected by adding 100 µl
of substrate solution, in each well. To prepare the colorimetric reagent, equal volumes of 3,
3′, 5, 5′-tetramethyl benzidine and dihydrochloride were mixed. The wells were incubated
with the substrate of peroxidase for 5 min to 1 h, in a shaker, protecting from light. The
reaction was stopped by addition of H2SO4 (100 µl/well, 2 N) and absorbances were read
at 450 nm. The concentration of HARP was determined with a titration curve established
with various concentration of HARP.
2.8 CELL-BASED ELISA (C-ELISA)
Cell-enzyme linked immunosorbent assay is described by a first cell-surface
binding assay and detection of the cell-bound proteins by immunological methods.
2 × 104 DU145 cells were seeded in a 96-well plate (culture plate) in 100 µl of
growth medium. The cells were allowed to adhere overnight at 37°C and 5% CO2 and
starved for 24 h. Before adding the samples, the wells were blocked in RPMI, 3% (w/v)
BSA, (300 µl/well) for 1 h. The incubation of cells with samples in RPMI, 1% (w/v) BSA
(100 µl/well) was performed by gentle shaking for 2 h at RT. For the competitive binding
assays, the cells were pre-incubated with competitors for 30 min at RT prior to adding the
samples. Unbound samples were removed by washing wells three times with 100 µl of
PBS, 1% (w/v) BSA and the cells were fixed with PBS, 4% (w/v) paraformaldehyde for 10
min at RT. The wells were then washed three times with PBS, BSA 1% (w/v) (100 µl/well)
(optional) and the non-specific sites were blocked with PBS, 1% BSA (w/v) (300 µl/well)
for 1 h at RT. For detection of cell surface-bound HARP, the protocol described above
(“Heparin-ELISA”) was used. For detection of biotinylated samples, 100 µl/well of
horseradish peroxidase-coupled anti-biotin antibody diluted to 1/2000th in blocking buffer
were added and the wells were incubated for 1 h at RT. The wells were washed three times
with PBS, BSA 1% (w/v) (100 µl/well) and the peroxidase activity was detected as
described above.
60
MATERIALS AND METHODS
2.9 IMMUNOFLUORESCENCE / CONFOCAL MICROSCOPY
The protocol is essentially based on C-ELISA binding assay described above,
just cell surface-bound biotinylated peptides are detected by immunofluorescence labelling
and analyzed by confocal microscopy.
5 × 104 DU145 cells (300 µl/well) were seeded in four-chamber LabTek slides
(Nunc) and were allowed to adhere on glass slides overnight. The cells were then serumstarved for 24 h, blocked with RPMI, 3% BSA (w/v) (500 µl/well) for 1 h at RT. The
biotinylated peptides diluted in RPMI/1% BSA (w/v) (200 µl/well) were added and the
slides were incubated for 2 h at RT, in absence or presence of non-biotinylated peptides.
After extensive washing in PBS, the Lab-Tek slides were fixed in PBS, 4% (w/v)
paraformaldehyde (300 µl/well) for 10 min at RT. If required, the cells were permeabilized
by washing three times the wells with PBS, 0.1% (v/v) Triton (300 µl/well), each for 5
min. The wells were further blocked with PBS, 1% BSA (w/v) (500 µl/well) for 1 h at RT
and incubated with fluorescein (FITC)-conjugated anti-biotin diluted to 1/100th in PBS, 1%
(w/v) BSA (200 µl/well) for 30 min, protecting from light. After several washes in PBS,
nuclei were stained with DAPI (4′, 6-diamidine-2′-phenylindole dihydrochloride) (1
µg/ml) (Roche) for 1 min and the slides were mounted by putting two-three drops of
mounting medium Immu-Mount (Thermo Electron Corporation, Pittsburgh PA) in each
chamber after wash. Slides were examined using a Zeiss LSM 510 META confocal laser
microscope (Zeiss, Iena, Germany) with a Plan Apochromate 63 × N.A. 1.4 objectif.
2.10 CELL SURFACE BINDING ASSAY
The protocol is derived from a method developped by the group of Hovanessian
(Callebaut et al., 1998) and is used to isolate and identify the cell-surface receptors of
P111-136 from PC3 cells.
PC3 cells were cultured in six 150-cm2 flasks, in RPMI 1640 medium containing
10% FBS and 1% PS, for 2 weeks. The cells in three 150-cm2 flasks (about 15 × 106
cells/flask) were incubated in 10 ml of culture medium (without serum) containing 5 µM
of biotin-labeled peptide for 2 h at room temperature. The other three flasks containing the
equivalent number of cells were used as a negative control. After washing extensively in
PBS containing 1 mM EDTA (PBS/EDTA), 1 ml of lysis buffer E (20 mM Tris-HCl, pH
7.6, 150 mM NaCl, 5 mM MgCl2, 0.2 mM phenylmethylsulfonyl fluoride, 0.2 mM βmercaptoethanol and 0.5% (v/v) Triton X-100) is added. After 5 min incubation in buffer E
(at room temperature), the flasks were declined and left at a vertical position to collect cell
61
MATERIALS AND METHODS
extracts. By this procedure, nucleus-free cells were recovered, since nuclei remain attached
to the plastic. The supernatants (V = 4 ml) were then centrifuged at 12,000 × g and stored
at -80°C.
The complex formed between cell-surface protein(s) and the biotin-labeled
peptide was isolated by purification using 100 µl of avidin-agarose (ImmunoPure
Immobilized Avidin, Pierce) in PBS/EDTA. After 2 h incubation at 4°C, the samples were
washed batchwise with PBS/EDTA (5 × 5 ml). The avidin-agarose pelet was resuspended
in 60 µl of 2-fold concentrated electrophoresis sample buffer (without β-mercaptoethanol),
heated at 95°C for 5 min. Two 30 µl aliquots of 60 µl of the purified preparation were
separately analyzed by SDS-PAGE in 10% polyacrylamide gel. One of the gels was
electrophoretically transferred to a PVDF membrane for ligand blotting, while the other
was used for further analysis of bands for mass spectroscopy, after staining of proteins
with colloidal coomassie blue (G 250).
2.11 PROLIFERATION ASSAYS
2.11.1 THYMIDINE INCORPORATION ASSAY
This assay is essentially carried out as described by Delbe et al. (Delbe et al.,
1995). 3 × 104 NIH 3T3 wild type cells per well were seeded in 48-well plates for 24 h in
DMEM supplemented by 10% FCS. Cells were then serum-starved for 24 h, and samples
were added. In the experiments involving inhibitors, competitors were added at the same
time as growth factors. Cells were then incubated for 18 h at 37°C and 5% CO2 and further
incubated for 6 h with 0.5 µCi of [methyl-3H]thymidine. Cells were then fixed with 10%
(v/v) cold trichloroacetic acid for 30 min at 4°C, washed three times with water and lyzed
with 0.2 N of NaOH for 30 min at 37°C or overnight at RT. Total incorporated
radioactivity was counted using a micro-beta scintillation counter (LKB, PerkinElmer Life
Sciences, Courtabœuf, France).
2.11.2 PROLIFERATION ASSAYS BY SPECTROPHOTOMETRY
Cell Counting Kit-8 (CCK-8) is a nonradioactive, sensitive colorimetric assay for
the determination of the number of viable cells in cell proliferation and cytotoxicity assays.
This test utilizes highly water-soluble tetrazolium salt, WST-8 [2-(2-methoxy-4nitrophenyl)-3-4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt].
WST-8 produces a water-soluble formazan dye upon reduction. In cells, WST-8 is reduced
62
MATERIALS AND METHODS
by dehydrogenases to give a yellow-colored product, formazan. The amount of formazan
dye generated is directly proportional to the number of living cells.
Cytotoxicity assay: 2 × 103 CHO K1 cells were seeded in triplicate in 96-well culture plate
(100 µl cell suspension/well) and were incubated in HAM’s F12 medium containing 10%
FBS for 24 h. Various concentrations of toxicants were added into the culture media and
the plate was incubated for 72 h at 37°C, 5% CO2. Doxorubicin hydrochloride (0.5 µM)
was used as a positive control. Ten µl of CCK-8 solution were added to each well of the
plate, the plate was then incubated for 3h, in the incubator. The absorbance at 450 nm was
measured using a microplate reader.
Cell Proliferation Assay: Tumor cells were seeded at 2 × 103 cells/well in triplicate in 96well plate, in the appropriate culture medium containing 10% FBS (100 µl/well). Medium
was removed 12 h later by fresh medium without serum or 1% serum, and the samples
were added. The plate was incubated for 72 h and was treated with samples each day. The
plate was incubated with 10 µl of CCK-8 solution, for 1-4 h, depending on the cell type
(usually 2 h) and absorbance at 450 nm was measured. To create a calibration curve, a
known number of viable cells at various cell density (103-105 cells/well) was seeded in a
separate plate and absorbance was read in the same manner.
Cell Adhesion Assay: The proliferation assay was slightly modified for the adhesion test.
The cell suspension was pre-incubated with various concentrations of samples for 30 min
at 37°C in a 15-ml conical tube. The tubes were centrifugated, the supernatants were
removed and the cells in the fresh media were transferred to the 96-well plate (100
µl/well). The plate was incubated in the incubator for 1 to 12 h. The non-adhering cells
were removed by gently aspirating and 100 µl/well media were added in each well. The
number of adherent cells was measured as described above.
63
MATERIALS AND METHODS
2.12 TUMOR GROWTH IN NUDE MICE
PC3 cells (2 × 106 cells/mouse) were injected into 12 female nude mice. Ten
days after the cell inoculation, all mice developed single tumors. Mice were arbitrarily
placed in control (n=6) and P111-124-treated groups (5 mg/kg/day). The volumes of
tumors were measured twice a week along two major axes with calipers. Tumor volumes
(cm3) were calculated as follows: V = (4/3) π R12 R2 where R1 and R2 are radius and
R1<R2. The tumors were treated 5 times/week and the mice were sacrificed after 25 days
of treatment.
2.13 SOFT AGAR ASSAY
This method assesses the ability of cells to grow in the absence of adhesion,
which is one indicator for assessing malignant cell transformation. Studies of anchorageindependent growth of cells in soft agar are carried out as described previously (Chauhan
et al., 1993).
2 × 103 cells were suspended in 0.35% (w/v) agar (500 µl/well) with or without
samples, and overlaid onto a 500 µl of solidified 0.6% (w/v) agar layer in 12-well plates, in
triplicate. Growth media with 10% FBS were included in both layer and 500 µl of growth
media were added on top layer. The cells were treated with samples three times within a
week. After one to three weeks, depending on the cell type, the colonies larger than 50 µm
were counted using a phase-contrast microscope equipped with a measuring grid.
2.14 OVER-AGAR ASSAY
The assay is used to harvest cells grown under anchorage-independent
conditions. The method is modified from a protocol developed by Dong and Cmarik (Dong
and Cmarik, 2002).
Thirteen ml of 0.5% (w/v) agar growth medium solution were pipetted into 100mm petri dish and allowed to harden for 40 min at room temperature, in the sterile laminar
flow hood. Seven ml of cell suspension containing a total of 1.5 × 105 cells were overlaid
onto the solidified 0.5% (w/v) agar layer. The cells were cultured on this agar growth
medium plate for three weeks. The liquid growth medium and any unattached cells were
then collected. To dislodge the remaining cells, the agar surface was washed gently with 34 ml PBS. All recovered cells were pelleted by centrifugation and subjected to biochemical
analyses.
64
MATERIALS AND METHODS
2.15 EXTRACTION OF RNA
Total RNA was extracted from cells at 60-70% confluence, using RNA Instapure
Kit (Eurogentec, Seraing, Belgium) according to the manufacturer’s instructions.
The concentration of RNA was determined spectrophotometrically at 260 nm
where 1 absorbance unit (A260) = 40 µg/ml of RNA/ml. The purity of RNA was estimated
from the ratio of absorbances at 260 nm and 280 nm (A260/A280), which should be between
1.9 and 2.1.
2.16 RT-PCR
cDNAs were synthesized from 1 µg of total RNA using random hexamers
primers and Superscript II™ reverse transcriptase (Invitrogen, Cergy Pontoise, France).
Then, 2 µl of RT product for HARP, RPTPβ and TFIID and 10 µl for ALK were subjected
to PCR amplification using the GenAmp9600 system (PE Applied Biosystems, Les Umis,
France). Primers used for amplification are indicated in the table 3.
Detection
HARP
RPTPβ
ALK
TFIID
Sequence
forward
5'-GAAAATTTGCAGCTGCCTT-3'
reverse
5'-CTTCTCCTGTTTCTTGCCT-3'
forward
5'-CTAAAGCGTTTCCTCGCTTG-3'
reverse
5'-TCTGAAACTCCTCCGCTGAC -3'
forward
5'-CAACGAGGCTGCAAGAGAGAT-3'
reverse
5'-GTCCCATTCCAACAAGTGAAGGA-3’
forward
5'-AGTGAAGAACAGTCCAGACTG-3'
reverse
5'-CCAGGAAATAACTCGGCTCAT-3'
Table 3: Primers used for amplification.
After 5 min at 94°C, 30 cycles (for HARP and TFIID), 35 cycles (for ALK and
RPTPβ) were used. Each cycle included denaturation at 94°C for 1 min, annealing at 60°C
(for HARP and TFIID) and 57°C (for ALK and RPTPβ) for 1 min, and primer extension at
72°C for 1 min. RT-PCR products were subjected to electrophoresis on 2% (w/v) agarose
gel.
65
MATERIALS AND METHODS
2.17 ANALYSIS OF BIOMOLECULAR INTERACTIONS
2.17.1 OPTICAL BIOSENSOR ANALYSIS (IAsys)
Optical biosensors provide a means of analyzing the biomolecular interactions
between immobilized ligands and soluble ligate molecules.
An optical biosensor consists of a sensor surface, on one side of which reside the
optics where the measurements are made and on the other side of which resides the liquid
phase in which experiments take place (Fernig D.G, 2001). Biosensor recognizes a
substance of interest on its sensor surface and converts this recognition event into an
electrical signal by means of the optical system. The optical system generates an
evanescent field, which penetrates into the liquid phase and extends away from the sensor
surface around 200 nm above the surface and decays exponentially away. Therefore, in this
system, only the interactions close to the sensor surface are monitored. The optics
essentially measure mass bound at the sensor surface. The binding on the sensor surface
causes changes in refractive index, which is linear with respect to the mass at the surface,
so as mass increases at the surface (binding), the signal will increase.
The most common commercially available optical biosensors on the market are
BIAcore and IAsys. BIAcore which may be called SPR biosensor uses surface plasmon
resonance (SPR) to produce the evanescent field in order to probe the liquid phase, while
IAsys (resonant mirror biosensor) uses an evanescent field generated by waveguiding
(Fernig D.G., 2001). Another powerful technique developed in recent years for analysis of
biomolecular interactions is dual polarization interferometry (DPI), which allows the
determination of thickness, density and mass of a biological layer on a waveguide surface
in real time (Analight). The IAsys resonant mirror technology was developed by Affinity
Sensors in the mid-1980s. The first commercial SPR biosensor appeared on the market in
1990 by BIAcore.
Regardless the type of instruments, the essence of experiments using optical
sensors is that one partner of molecular interaction is immobilized on the sensor surface
and the mobile partner at constant concentration is introduced into the liquid phase above
the surface (Schuck P., 1997). In optical biosensors, there are two main steps leading to the
increase in mass at the surface and so generating a response: sample diffusion and the
specific binding. The mobile partner has to be efficiently transported to and from the
reactive surface. To ensure rapid mixing and efficient sample delivery to the surface, hence
minimizing the limitations in the mass transport of the mobile reactant, IAsys technology
utilizes vibro-stirrer whereas BIAcore use laminar flow systems (Fernig D.G., 2001).
66
MATERIALS AND METHODS
Background binding on the sensor surface is a poor basis for any type of
measurement since the signal is a fraction of noise. However, there is almost no possible
true control to generate, since, by definition, only the experimental surface has
immobilized ligand. In a capture system, e.g., streptavidin for biotinylated ligands, is used,
the non-specific binding is tested on a surface with the capture system immobilized (Fernig
D.G., 2001).
2.17.2 BINDING ASSAYS
Binding assays using Iasys biosensor were performed to analyze biomolecular
interactions between heparin-HARP and the peptides derived from HARP as well as
HARP-protein. These assays consisted of an immobilization step of the heparin or protein
of interest on the aminosilane surface and of a binding step.
2.17.2.1 Immobilization of Oligosaccharides and Proteins
Instead of direct coupling of oligosaccharides and proteins to the surface, a
capture system, biotin-streptavidin was used. This system allows the oriented
immobilization of the ligand in a native conformation and accessible orientation, which
optimizes ligate binding.
A PREPARATION OF STREPTAVIDIN SURFACE
Streptavidin was coupled to the aminosilane cuvettes using bissulfosuccinimidyl
suberate (BS3 as the cross linker following the manufacturer’s instructions (Neosensors).
B BIOTINYLATION
Biotinylation of heparin: Heparin was biotinylated on free amino groups. Ten µl of a 50
mM solution of NHS-LC-biotin in dimethyl sulphoxide were added to 10 mg of heparin in
100 µl of distilled water. The solution was mixed end-over-end and the reaction was
allowed to proceed for 24 h at room temperature. Two further additions of 10 µl of NHSLC-biotin were made over the subsequent 48 h. Unreacted NHS groups were blocked by
addition of 10 µl of 2 M Tris-HCl, pH 7.2 and the sample was incubated for 10 min at RT.
Biotinylation of HARP: Prior to biotinylation, Tris buffer in which HARP fractions were
previously eluted in Mono-S column (see above “Purification of HARP”) was removed by
5 washes of 500 µl in PBS in Nanosep centrifugate tubes. Two µl of a 50 mM solution of
67
MATERIALS AND METHODS
NHS-LC-biotin (2.4 nmol) in dimethyl sulphoxide were added to 17 µg of Tris-free HARP
(1.2 nmol) in 10 µl of distilled water plus 10 µl of PBS and allowed to react 48 h at room
temperature.
Removal of free biotin: Free biotin was removed by fractionation on a Sephadex G-25
column (1 x 25 cm, flow rate 0.3 ml/min) equilibrated in PBS and calibrated with blue
dextran and potassium dichromate.
C CAPTURE
The cuvette was equilibrated in 50 µl of PBS, 0.02% (v/v) Tween 20 (PBS-T).
PBS-T was replaced with 30 µl of biotinylated heparin at 1 mg/ml, 50 µl of biotinylated
HARP or 30 µl of PBS-T containing biotinylated peptide P111-136 at 40 µg/ml. The
biotinylated ligand was generally incubated in the streptavidin-derivatized cuvette for 30
min. The cuvette was then washed 5 times with 50 µl of PBST-T. Cuvettes were stored dry
at 4°C, after washing 5 times with 50 µl of water and emptying it.
2.17.2.2 Binding assays
A BINDING ASSAY
The binding assay consists of the following events:
1- Addition of ligate: Binding is initiated by addition of 1 µl of the soluble binding partner
or ligate to a cuvette containing 29 or 49 µl of PBS-T. The association reaction is followed
over a set of time, usually 180 s, by which time binding is within 90% of the maximum
(Figure10-A).
2- PBS-T washes: At the end of the association reaction (equilibrium), the cuvette is
washed three times with 50 µl of PBS-T to initiate the dissociation reaction (Figure 10-B).
3- Regeneration: To regenerate the surface, the cuvette is washed three times with 50 µl of
suitable regeneration reagent, usually 2 M NaCl (Figure 10-C).
Regeneration of the surface serves to remove all bound ligate. It is essential that
regeneration protocols do not remove immobilized ligand or chemically alter the structure
of the immobilized ligand. Since glycosaminoglycans are robust ligands and the biotinstreptavidin bond is the known strongest non-covalent interaction (Ka = 1015 M-1), they are
resistant to 2 M NaCl, 20 mM HCl, 20 mM NaOH, 2 M guanidine. Most proteinglycosaminoglycan interactions depend on the ionic interaction between the sulphate
68
MATERIALS AND METHODS
groups of the GAG and the amino groups of the protein. Therefore regeneration of the
surface is usually performed with 2 M NaCl. When 2 M NaCl fails to remove all the bound
ligate, especially in case of protein-protein interactions, additional regeneration steps with
20 mM HCl, 20 mM NaOH and, in extreme cases, with 4 M urea, are applied.
4- Returning to starting conditions: The cuvette is washed three times with 50 µl of PBS-T
and then once with 29 or 49 µl PBS-T (Figure -10D).
B
A
C
D
Figure 10: A binding assay. Binding of HARP (final concentration 0.2 µg/ml) to biotinylated
heparin, immobilized on a streptavidin derivatized aminosilane cuvette. A response of 600 arc
seconds is equal to 1 ng/mm2 of protein in the aminosilane gel. A) Binding was initiated by the
addition of 1 µl of 100 µg/ml HARP to the 49 µl PBS-T cuvette. B) At the end of the association
reaction, the cuvette was washed three times with 50 µl of PBS-T to initiate the dissociation
reaction. C) To regenerate the surface, the surface was washed three times with 2 M NaCl. D) The
cuvette was washed three times with 50 µl PBS-T and then once with 49 µl PBS-T.
Calculation of extent of binding
The main measurement is the extent of ligate bound, which requires the binding
reaction to be at or near equilibrium. The extent of binding is calculated in two ways. First,
the extent of ligate binding is determined directly from the binding curve: it is equal to the
response at equilibrium minus the response before the addition of ligate, minus the bulk
shift. Second way is that the non-linear curve fitting software supplied by the manufacturer
is used to calculate the extent of binding.
69
MATERIALS AND METHODS
B COMPETITIVE BINDING ASSAY
These experiments test the hypothesis that the region of the ligate which
recognizes the ligand can also recognize the soluble competitor. The binding assays with a
soluble competitor of the ligand were performed in two ways:
a) In one set of competitive binding experiments, 5 µl of PBS-T containing heparin at 100
µg/ml were added to HARP- or P111-136-derivatized cuvettes containing 40 µl of PBST.
The soluble binding partner was then added and the association was followed for 210 s.
b) Another set of competitive binding assay used a cuvette with heparin-immobilized
surface and a customized program for the IAsys Auto+ instrument (available from the
manufacturer, NeoSensors). The cuvette was equilibrated at 20°C in 40 µl PBS-T and then
5 µl of the relevant dilution of an oligosaccharide in PBS-T were added. Once the base line
is stable, 5 µl of HARP or HARP domains were added to initiate the association phase.
The binding reaction was continued for 4 min. The surface was then washed three times
with 50 µl of PBS-T and 2 min later was regenerated with 20 mM HCl. All experiments
were performed with reagents stored in the instrument’s reagent trays at 4°C for the
duration of the experiment. Each set of curve comprised HARP-bound to immobilized
heparin in the absence of competing polysaccharide and the average of the amount of
HARP in absence of competitors at the start and the end of the each set of experiments was
used as a zero for the normalization of each set. The extent of binding was calculated by
fitting the association curve to a single site binding model using FastFit software (Affinity
Sensors).
70
RESULTS
1. IMPLICATION OF THE C-TERMINAL OF HARP
IN ITS MITOGENIC, TRANSFORMING
AND ANGIOGENIC ACTIVITIES
RESULTS
1. IMPLICATION OF THE C-TERMINAL OF HARP IN ITS MITOGENIC,
TRANSFORMING AND ANGIOGENIC ACTIVITIES
1.1 INTRODUCTION
The requirement for the C-terminal amino acid residues of HARP for its biological
activities have previously been shown in our laboratory (Bernard-Pierrot et al., 2001,
2002). The mutant protein lacking the whole C-terminal region has been shown to
heterodimerize with the wild type HARP, and hence inhibit the mitogenic, transforming,
tumor growth and angiogenic activities of HARP by forming non-functional dimers.
Several groups have been previously reported that the activities of HARP are mediated via
ALK receptor. Using cell-free binding assays, our team showed that HARP lacking its Cterminus unable to interact with its receptor ALK, showing the importance of the Cterminus residues of HARP for the binding capacity to ALK. Subsequently, a synthetic
peptide corresponding to the deleted 26 amino acids residues of the C-terminus, designated
P111-136, has been shown to bind to the extracellular domain of ALK and the in vitro
inhibition of biological activities of the wild type HARP has been explained by the
competition of the peptide with HARP for binding to ALK (Bernard-Pierrot et al., 2002).
Thus, we have firstly investigated the in vitro and in vivo effects of P111-136,
using prostatic cancer cell line, PC3, which expresses endogenous HARP and its receptor
ALK, in order to confirm the previous results in other models.
Recently, the laboratory of P.S. Mischel reported the isolation and
characterization of two forms of HARP secreted from human cells; HARP18 (full-length)
and HARP15 (C-terminally-processed) (Lu et al., 2005). According to this study,
HARP15, which lacks the last 12 C-terminal amino acid residues stimulated cell
proliferation in ALK-expressing cells, while the full-length HARP18 with an intact Cterminus was unable to do so. These results taken together with our previous results
reporting the whole C-terminal possessing the mitogenic activity, were pointing to the
importance of a 14 amino acid-C-terminal fragment. This prompted us to synthesize the
corresponding peptide P111-124 and to investigate whether this peptide mimics the effects
of P111-136.
73
RESULTS
1.2 RESULTS
1.2.1 INHIBITION OF TUMOR GROWTH BY SYNTHETIC PEPTIDES
Article 1: submitted to Molecular Cancer Therapeutics
Title: Synthetic Peptides Derived From the C-terminal Region of HARP Inhibit Tumor
Growth and Associated Angiogenesis
Authors: Oya Bermek, Yamina Hamma-Kourbali Isabelle Bernard-Pierrot, Vincent Rouet,
Sophie Dalle, Jean Delbé and José Courty
74
RESULTS
Synthetic Peptides from the HARP C-Terminus Inhibit Tumour Growth and
Associated Angiogenesis
O. Bermek,1,2 Y. Hamma-Kourbali,1,2 I Bernard-Pierrot,3 V. Rouet,4 S Dalle,1 J. Courty,1
and J. Delbé1.
1
Laboratoire de Recherche sur la Croissance Cellulaire, la Réparation et la Régénération
Tissulaires (CRRET), CNRS UMR 7149, Université Paris 12, Avenue du Général de
Gaulle, 94010 Créteil Cedex, France
Correspondence: J Courty, Laboratoire de Recherche sur la Croissance Cellulaire, la
Réparation et la Régénération Tissulaires (CRRET), CNRS UMR 7149, Université Paris
12, Avenue du Général de Gaulle, 94010 Créteil Cedex, France
E mail: [email protected]
Tel.: +33 145 171 797; Fax: +33 145 171 816
Running title: Tumour Growth Inhibition by Synthetic Peptides
Keywords: HARP, pleiotrophin, biological activities, synthetic peptides, tumour growth,
angiogenesis
Abbreviations: HARP, heparin affin regulatory peptide; DMEM, Dulbecco's modified
Eagle medium; PBS, phosphate buffer saline
Title footnotes:
2
These two authors contributed equally to this work.
3
The present address of I. Bernard-Pierrot is UMR 144, CNRS-Institut Curie, 26 rue
d’Ulm, 75248 Paris Cedex 05, France
4
The present address of V. Rouet is U845, Inserm Necker Enfant Malades, 156 rue de
Vaugirard 75015 Paris, France.
Heparin affin regulatory peptide
(HARP), also called pleiotrophin, is an
heparin-binding, secreted factor that
stimulates growth and angiogenesis.
HARP is overexpressed in several
tumours and tumour cell lines including
those of prostatic origin. In vivo, HARP
is a rate-limiting factor for tumour
growth and metastasis. Previously, we
showed that the biological activities of
HARP
involve
the
C-terminus,
composed of amino acids 111 to 136, and
we established that a synthetic peptide
composed of the same amino acids
(P111-136) was capable of binding to the
ALK receptor and of inhibiting the
biological activities of HARP. Here, we
investigated the biological effects of
P111-136 on the human prostatic
adenocarcinoma cell line PC-3, which
possesses an autocrine loop for HARP.
P111-136 inhibited the formation of PC3-cell colonies in the soft agar assay and
inhibited PC-3 tumour growth in the
xenograft model. Similar results were
obtained with peptide P111-124. Both
P111-136 and P111-124 inhibited
angiogenesis induced by HARP. Given
their ability to inhibit both tumour
growth and angiogenesis, these peptides
may hold promise for cancer therapy.
Prostate cancer is among the leading
malignancies in men throughout much of
the industrialized world and ranks second
among causes of death from cancer. The
lack of effective treatments indicates a
need for developing novel treatment
strategies.
Epithelial-stromal interactions play a
pivotal role in normal morphogenesis
(Cunha & Donjacour, 1987). Complex
interactions occur between peptide growth
factors and growth modulators, which may
be regulated either by androgens or by
other factors (Russell et al., 1998). Any
imbalance in these interactions, such as up75
RESULTS
or down-regulation of growth factors or
their receptors or a switch from paracrine
to autocrine mediation of growth-factor
pathways leads to prostate tumour
progression. Among the growth-factor
families involved in prostate-cancer
progression, transforming growth factorbeta (TGFβ), fibroblast growth factors
(FGFs), and epithelial growth factor (EGF)
seem to play a prominent role (Russell et
al., 1998).
Heparin affin regulatory peptide (HARP),
or pleiotrophin, is a growth-promoting and
angiogenic factor that has been shown to
play a key role in prostate cancer. Thus,
plasma HARP levels were elevated in
patients with prostate cancer (Soulie et al.,
2004; Souttou et al., 1998). Furthermore,
HARP protein was associated with
epithelial cells in prostate cancer but not in
normal prostate tissue, and the mRNAs
were located in the stromal compartment,
suggesting a paracrine mechanism of action
for HARP (Vacherot et al., 1999). In vitro,
HARP overexpression in normal prostate
epithelial PNT-1A cells induced both
anchorage-independent and anchoragedependent growth at low serum
concentrations. HARP was also mitogenic
for PC-3, LNCaP, and DU145 cell lines
(Vacherot et al., 1999). The growthpromoting effect of HARP on prostate
cancer cells was also confirmed using an
antisense strategy, which established
HARP as an important autocrine growth
factor for the LNCaP prostate-cancer cell
line and as a paracrine factor involved in
angiogenesis (Hatziapostolou et al., 2005).
HARP is a secreted polypeptide composed
of 136 amino acids. HARP and the related
protein midkine constitute a specific family
among the heparin-binding growth factors
(Muramatsu, 2002). During embryonic
development, HARP is expressed in tissues
originating in the mesoderm and
neuroectoderm, suggesting a role in
epithelium-mesenchyme interactions and in
neuronal migration. In adults, HARP
expression is limited except at sites such as
the mammary gland and uterus associated
with reproductive angiogenesis (Courty et
al.,
2000).
Furthermore,
HARP
overexpression has been documented in
several conditions associated with cell
proliferation and angiogenesis, such as
rheumatoid arthritis (Pufe et al., 2003) and
tumour growth (Heroult et al., 2004).
The transforming activity of HARP was
first established using NIH-3T3 cells,
whose transfection with HARP cDNA led
to
morphological
transformation,
anchorage-independent growth and tumour
formation in nude mice. Connecting with
its angiogenic activity, HARP was shown
to induce endothelial-cell migration in
collagen or Matrigel and to potentiate the
plasminogen activator (Laaroubi et al.,
1994; Muramatsu et al., 1994; Polykratis et
al., 2005). The mitogenic, angiogenic, and
transforming activities of HARP are linked
to the high-affinity tyrosine-kinase receptor
anaplastic lymphoma kinase (ALK)
(Bernard-Pierrot et al., 1999; Stoica et al.,
2001). Mature HARP consists of two betasheet domains containing thrombospondinI repeats (TSR-I), which are flanked by
flexible lysine-rich N- and C-terminal tails
having no apparent structure. The betasheet domains mediate binding of HARP to
glycosaminoglycans, which are involved in
its dimerization (Bernard-Pierrot et al.,
1999; Stoica et al., 2001), and to the
proteoglycan receptors syndecan-3 and
RPTPβ/ζ (Receptor
Protein
Tyrosine
Phosphatase β/ζ) involved in the neurite
outgrowth and cell migration activities of
HARP (Raulo et al., 2005). The Cterminus, composed of amino acids 111
through 136, is involved in the binding of
HARP to the ALK receptor (BernardPierrot et al., 2001; Bernard-Pierrot et al.,
2002). In these previous studies, a mutant
HARP protein lacking amino acids 111
through 136 (HARPΔ111-136) acted as a
dominant negative effector of HARP
mitogenic, angiogenic, and transforming
activities by heterodimerising with the wild
type protein. In addition, the synthetic
peptide composed of the deleted aminoacid segment (P111-136) competed with
76
RESULTS
HARP for binding to the ALK receptor
(Bernard-Pierrot et al., 2001). Thus, P111136 may hold promise for cancer treatment.
The objective of this study was to evaluate
in vitro and in vivo inhibition by P111-136
of the proliferation of PC-3, an androgenindependent prostate-cancer cell line. The
results showed that P111-136 and the
shorter peptide P111-124 inhibited the
growth of PC-3 cells as effectively as
Taxol in an in vivo model. Thus, P111-136
and P111-124 may be strong candidates for
cancer treatment.
MATERIALS AND METHODS
Materials
Culture medium, foetal calf serum (FCS),
and G418 were supplied by Invitrogen
(Cergy Pontoise, France). Methyl-[³H]
thymidine was purchased from ICN
(Orsay, France). Heparin-Sepharose gel
and Mono-S column were from GE
HealthCare (Orsay, France) and BM
ChemiLuminescence
from
Roche
Diagnostic (Meylan, France). Superblocker
solution was purchased from Perbio-Pierce
(Montluçon, France). Immobilon-P from
Millipore Corp (Saint-Quentin en Yvelines,
France). Goat antibody to human
pleiotrophin (HARP) was from R&D
(Oxon, United Kingdom). Goat anti-mouse
CD31 (PECAM) polyclonal antibody was
purchased
from
BD
Pharmingen
Biosciences (le Pont de Claix, France).
Mouse anti-phospho-p44/p42 MAP kinase
was from New England BioLabs (SaintQuentin en Yvelines, France). Horseradish
peroxidase-conjugated rabbit anti-goat IgG
and goat anti-rabbit IgG were purchased
from Jackson (Montluçon, France).
Matrigel was from BD PharMingen (Le
Pont de Calais, France). Taxol was
purshased from Sigma Aldrich (Saint
Quentin Fallavier, France). Altergen
(Schiltigheim, France) synthesized the
three following peptides: P111-136
(KLTKPKPQAESKKKKKEGKKQEKML
D), P111-124 (KLTKPKPQAESKKK),
and
P1-21
(AEAGKKEKPEKKVKKSDCGEW).
Purification of recombinant HARP from
conditioned media of NIH-3T3 cells
HARP was purified from conditioned
media of transfected NIH-3T3 cells as
previously described (Bernard-Pierrot et
al., 2001). Briefly, conditioned medium
was buffered in 20 mM Hepes pH 7.4 with
0.5 M NaCl then loaded on a heparinSepharose column. Bound proteins were
eluted using 20 mM Hepes pH 7.4 with 2
M NaCl and further purified using a cationexchange Mono-S column. Purification
was carried out in 50 mM Tris pH 7.4
using a 0.4-2 M NaCl gradient.
Thymidine incorporation assay
NIH-3T3 fibroblasts (2 × 104) were seeded
in 48-well plates for 24 h in Dulbecco
modified Eagle's medium supplemented
with 10% foetal calf serum. Cells were
serum-starved for 24 h before sample
addition then incubated at 37°C for 18 h.
After an additional 6-h incubation period
with 0.5 µCi of [³H]-thymidine, cells were
fixed with 10% trichloroacetic acid,
washed with water, and lysed with 0.2 N
NaOH. Total incorporated radioactivity
was counted using a micro-beta
scintillation counter (LKB, Perkin Elmer
Life Sciences, Courtaboeuf, France).
Reverse Transcriptase-Polymerase
Chain Reaction (RT-PCR) for ALK and
RPTPβ/ζ
Total RNA was extracted from cells using
the RNA Instapure Kit (Eurogentec, Liege
Belgium) according to the manufacturer’s
instructions. To synthesise cDNA, we used
1 µg of total RNA, random hexamer
primers, and Superscript II™ reverse
transcriptase (Invitrogen, Cergy Pontoise,
France). Reaction products diluted 1:10,
1:5, or 1:2 (v/v) diluted in PCR buffer were
subjected to PCR amplification using a
GenAmp9600 system (PE Applied
Biosystems, Les Ulis, France) for detection
of
HARP,
RPTPβ/ζ, and
ALK,
77
RESULTS
respectively. Primers were as follows:
HARP
detection,
5’GAAAATTTGCAGCTGCCTT-3’
(forward)
and
5’CTTCTCCTGTTTCTTGCCT-3’ (reverse);
RPTPβ/ζ
detection,
5’CTAAAGCGTTTCCTCGCTTG-3’
(forward)
and
5’TCTGAAACTCCTCCGCTGAC (reverse);
ALK
detection,
5’CAACGAGGCTGCAAGAGAGAT-3’
(forward)
and
5’GTCCCATTCCAACAAGTGAAGGA-3’
(reverse); and TFIID detection, 5’AGTGAAGAACAGTCCAGACTG-3’
(forward)
and
5’CCAGGAAATAACTCGGCTCAT-3’
(reverse). After 5 min at 94°C, PCR was
performed, with 30 cycles for HARP and
TFIID and 35 cycles for RPTPβ/ζ and
ALK. Each cycle included denaturation at
94°C for 1 min, annealing at 60°C (for
HARP and TFIID) or 57°C (for
RPTPβ/ζ or ALK) for 1 min, and primer
extension at 72°C for 1 min. RT-PCR
products were subjected to electrophoresis
on 2% agarose gels containing 0.5 mg/ml
ethidium bromide. The gels were then
photographed using a ChemiGenius system
(Syngene, Cambridge, UK).
incubated with monoclonal anti-phosphop42/p44 and anti-p42/p44 antibodies
diluted in PBS-Tween 20. After incubation
at 4°C overnight, membranes were washed
three times in PBS-Tween 20 and
incubated with horseradish peroxidaseconjugated secondary antibody diluted
1:1000 in PBS-Tween 20, for 1 h at room
temperature. Membranes were washed
three times with PBS, and immunoreactive
bands were detected using the BM
ChemiLuminescence detection kit (Roche
Diagnostics, Meylan, France) according to
the manufacturer's instructions.
MAP kinase phosphorylation assay
NIH-3T3 cells (2.5 × 105) were seeded in
35-mm culture dishes for 24 h, serumstarved for 24 h, and stimulated for 5 min
at 37 °C with FCS or HARP in presence or
absence of peptides. Cells were lysed with
electrophoresis sample buffer (50 mM TrisHCl pH 6.8, 10% glycerol, 0.02%
bromphenol blue, 2% SDS, and 5% βmercaptoethanol). Proteins were run on
SDS-10% polyacrylamide gel then
transferred to Immobilon-P membrane in
25 mM Tris, pH 8.3, containing 200 mM
glycine and 10% ethanol. Blocking was
achieved by incubating the membrane in
phosphate buffer saline (PBS)-Tween 20
(0.2% v/v), supplemented with 3% (w/v)
powdered milk for 1 h at room
temperature. Membranes were further
Tumour-cell inoculation to nude mice
All in vivo experiments were approved by
the appropriate ethics committee and
conducted in compliance with European
Community directives. PC-3 carcinoma
cells
(2
×
106)
were
injected
subcutaneously in the right flank of female
athymic nude mice (Janvier, Le Genest St
Isle, France). Two weeks after the
injection, each mouse had one palpable
tumour, about 60 mm3 in size. Groups of 5
mice were then given peritumoral
injections of 0.1 ml PBS alone, Taxol 10
mg/kg twice a week, P111-136 (5
mg/kg/day), or P111-124 (5 mg/kg/day).
Tumour size was determined twice a week
by using callipers to measure the lengths of
the two main axes, computing the
corresponding radii (labelled R1 and R2,
Colony formation in soft agar
PC-3 cells (ATCC, USA) were seeded at a
density of 2 x 104 in triplicate into 12-well
plates containing agar and DMEM
supplemented with 10% FCS and various
concentrations of P111-136 or P111-124
peptides, with or without anti-human
HARP antibody or control nonimmune
immunoglobulins. The compounds were
added to the culture medium twice a week.
After 12 days, colonies larger than 50 µm
in diameter were counted using a phasecontrast microscope equipped with a
reticle, in five fields in each of three wells.
The assay was repeated at least twice.
78
RESULTS
with R1<R2), and estimating tumour
volume as V = (4/3) π R12 R2.
Tissue
preparation,
immuno
histochemical staining, and image
analysis
The PC-3 tumours were removed surgically
then immediately frozen in liquid nitrogen
and fixed for 20 min in acetone at 4°C. The
6-µm sections were rehydrated in PBS then
saturated with PBS containing 1% bovine
serum albumin (BSA) and 2% normal goat
serum. Endogenous biotin was blocked
using the Vector blocking kit (Vector
Laboratories, Burlingame, CA). To
visualize endothelial cells within the
tumours, sections were incubated with goat
anti-mouse CD31 polyclonal antibody for 1
h at room temperature. After two washes in
PBS-Tween 20 (0.2% v/v), sections were
incubated for 1 h at room temperature with
biotinylated
goat
anti-rabbit
IgG
(Chemicon International Inc., Temecula,
CA) in saturation buffer, followed by three
washes and incubation with an avidinbiotinylated-alkaline phosphatase complex
(Vector
Laboratories).
Alkaline
phosphatase activity was revealed using the
Vector red substrate (Vector Laboratories,).
Finally, the sections were counterstained
with haematoxylin, washed with water, and
cover slipped with mounting medium
(Thermo Shandon, Pittsburgh, PA). For
each
CD31-labelled
section,
five
microscopic fields containing exclusively
viable tumour cells were randomly selected
for analysis. Endothelial-cell density was
expressed as the ratio of endothelial-cell
area/total area examined × 100. Mean
values were then computed for untreated
and treated tumours.
In vivo mouse angiogenesis assay using
the Matrigel plug model
Swiss mice (Janvier, Le Genest St Isle,
France) were injected subcutaneously with
0.3 ml of growth-factor reduced Matrigel
alone or containing P111-136 (1 µM),
HARP (5 nM), FGF-2 (10 nM), HARP and
P111-136, or FGF-2 and P111-136 (4
mice/group). The Matrigel rapidly formed
a single, solid, gel plug. After 8 days, the
skin was pulled back to expose the intact
plug, which was dissected out, frozen in
liquid nitrogen, and fixed with acetone.
Matrigel plug sections 8 µm in thickness
were cut using a cryostat CM3050 (Leica
Microsystems, Rueil, France) and stained
with Gomori-Trichrome for microscopic
observation. The area infiltrated by
endothelial cells was then measured using
an image analyser in six fields in each of
three Matrigel-plug sections from each
mouse.
Statistical analysis
Unpaired t-tests and Mann-Whitney
ANOVA were used to assess differences
between each group and the corresponding
control group. All results are reported as
mean ± SD determined from at least two
independent experiments.
RESULTS
Inhibition of HARP-induced mitogenesis
by peptide 111-136
P111-136 in various concentrations
inhibited the mitogenic effect of HARP on
serum-starved NIH-3T3 cells (Figure 1A):
[3H]-thymidine incorporation increased 10fold after the addition of 4 nM of HARP,
and this effect was dose-dependently
inhibited by adding P111-136. Inhibition
was complete with 1 µM of P111-136.
Given that the HARP mitogenic signal is
transduced through the MAP kinase
signalling pathway, we evaluated the
ability of HARP to induce MAP kinase
phosphorylation in the presence of P111136. As expected from the cell-stimulation
experiments, MAP kinase phosphorylation
was inhibited by 1 µM of P111-136 (Figure
1B). Furthermore, in the absence of
stimulation,
P111-136
completely
abolished the phosphorylation of P42 and
P44. To evaluate the specificity of the
inhibitory effect of P111-136, we evaluated
the synthetic peptide P1-21 composed of
the 21 first amino-acid residues of HARP.
79
RESULTS
P1-21 contains several lysine residues and
has a similar backbone to P111-136,
making it a suitable control for our
experiments. P1-21 in concentrations of up
to 1 µM failed to affect HARP-induced
proliferation, supporting the specificity of
the inhibitory effect of P111-136. Poly
arginine in a concentration of 1 µM also
failed
to
inhibit
HARP-induced
proliferation (Figure 1D). Taken together,
these results establish that P111-136
inhibits HARP-induced proliferation and
that this effect is not ascribable solely to
the presence of basic residues.
Inhibition of PC-3 colony formation by
peptide 111-136
The androgen-independent prostate-cancer
cell line PC-3 was shown in two
independent studies to both produce and
respond to HARP, suggesting an autocrine
loop for this growth factor (Soulie et al.,
2002; Vacherot et al., 1999). Using RTPCR, we found that PC-3 cells expressed
both HARP and its ALK receptor but not
its RPTPβ/ζ receptor, under exponential or
confluent conditions (Figure 2A,). U87MG cells, used as a control, expressed
HARP and both receptors. We then
evaluated the HARP autocrine loop, using
polyclonal anti-human HARP antibody in
the soft agar colony-formation assay,
which measures anchorage-independent
growth, a hallmark of malignant
transformation. Polyclonal anti-human
HARP antibody induced a dose-dependent
decrease in colony formation (Figure 2B),
whereas idiotypic immunoglobulins, used
as the control, had no effect. These data
established the existence of an autocrine
HARP-signalling loop for PC-3 cells,
prompting us to investigate the effect of
P111-136 on PC-3 proliferation. P111-136
dose-dependently
decreased
colony
formation (Figure 2C). With 1 µM of
P111-136, the colonies were smaller and
47% less numerous, compared to the
control.
Tumour growth inhibition by peptide
111-136 in a human xenograft model of
prostate cancer
To further investigate the effect of P111136 on HARP-induced PC-3 proliferation,
we injected mice with PC-3 cells, which
consistently led to tumour development
within 2 weeks. P111-136 treatment was
initiated at the end of the second week,
when the tumours were well established, in
order to simulate curative treatment. P111136 (5 mg/kg/day) significantly reduced
tumour growth as soon as the first
treatment week, compared to PBS used as
the control (Figure 3A). After 4 weeks,
tumour size was reduced by 77 % in the
P111-136 group. P111-136 treatment had
no effect on body weight (data not shown)
and induced no evidence of toxicity such as
diarrhoea, infection, weakness, or lethargy.
As expected, control treatment with Taxol
(10 mg/kg twice a week) strongly inhibited
tumour growth, by 93% compared to PBS,
after 4 weeks of treatment (Figure 3A). At
the end of the study, the animals were
sacrificed and tumour weight was
determined. Both P111-136 and Taxol
significantly decreased tumour weight, by
more than 75%, compared with PBS,
supporting the tumour-size data (Figure
3B). To determine whether angiogenesis
associated with tumour growth was also
affected by P111-136 treatment, we used
CD31 immunostaining to quantify blood
vessels. P111-136 injected near the tumour
significantly decreased endothelial-cell
density (Figure 4B and C) compared to the
untreated tumours (Figure 4A and C). The
mean percentage of endothelial cells in
viable fields of tumours treated with 5
mg/kg/day of P111-136 (1.7±0.58, based
on 25 fields in each of five tumours) was
inhibited by 64% (p<0.01) compared to the
control tumour value (4.8±2, based on 20
fields in each of four tumours). In addition,
vessels positive for the endothelial marker
GSL1 were scarcer in P111-136-treated
tumours than in control tumours (data not
shown), further confirming the antiangiogenic effect of P111-136. These
80
RESULTS
results suggested than P111-136 inhibited
the angiogenic effect of HARP.
Inhibition of in vivo HARP-induced
angiogenesis by peptide 111-136 in a
Matrigel plug model
The ability of P111-136 to inhibit the
angiogenic activity of HARP in vivo was
evaluated using a mouse Matrigel plug
assay. Incorporating 5 nM of HARP into
the Matrigel resulted in a 3.7-fold increase
in endothelial-cell infiltration (Figure 5, C
and E) compared to the control plug, which
contained only a few endothelial cells
(Figure 5, A and E). Adding 1 µM of P111136 to the HARP-containing Matrigel
inhibited the effect of HARP on endothelial
cell infiltration by 72% (Figure 5, C, D,
and E). P111-136 had no effect on
endothelial-cell infiltration seen with
Matrigel alone (Figure 5, A, B, and E).
These data indicate a role for the HARP Cterminus in the angiogenic effect of HARP.
Inhibition of the biological activities of
HARP by the C-terminus lacking its last
12 amino acids (peptide 111-124)
A recent study of HARP-induced
proliferation and migration of human
glioblastoma cells suggests that these two
activities may be mediated by two different
forms of HARP, acting via different
receptors (ALK or RPTPβ/ζ) (Lu et al.,
2005). The naturally occurring form of
HARP lacking the last 12 amino acids
induced glioblastoma-cell proliferation via
the ALK receptor pathway, whereas the
full-length form of HARP induced cell
migration via the RPTPβ/ζ receptor
pathway. These findings and our data
suggested that the synthetic peptide lacking
the last 12 amino acids of P111-136,
namely, P111-124, might inhibit the
biological effects of HARP. To explore this
possibility, we evaluated the mitogenic
effect of HARP on NIH-3T3 cells with or
without
P111-124
in
various
concentrations. P111-124 dose-dependently
inhibited the incorporation of [3H]thymidine induced by HARP (Figure 6A).
With 1 µM of P111-124, about 95%
inhibition was obtained. These results were
similar to those seen with P111-136.
Furthermore, evaluation of the HARP
signalling pathway mediated by MAP
kinase confirmed this effect of P111-124
(data not shown).
We evaluated the effect of P111-124
in the colony-formation assay using PC-3
cells. P111-124 dose-dependently inhibited
the growth of PC-3 cells in agar. With 1
µM of P111-124, the number of colonies
was decreased by 39%, compared to the
untreated control (Figure 6B). To
determine whether P111-124 exerted antiproliferative
effects
in
vivo,
we
administered soluble P111-124 to nude
mice with established primary tumours
induced by the injection of human PC-3
cells. P111-124 significantly inhibited
tumour growth when used in a dose of 5
mg/kg/day, which was also the active dose
of P111-136 (Figure 7A). Inhibition
compared to the vehicle-treated control
group was 70% with P111-124, which was
similar to the 77% inhibition seen with
P111-136 (Figure 3A). This inhibition was
confirmed by the determination of the
tumor weight at the end of the treatment
(Figure 7B). Thus, absence of the last 12
amino acids of the HARP C-terminus did
not alter the in vitro or in vivo inhibitory
effects of the peptide. As with P111-136,
no evidence of toxicity was observed in
mice treated with P111-124.
DISCUSSION
HARP is expressed in a wide range of
human tumours and tumour cell lines
including neuroblastoma; glioblastoma;
melanoma; and cancers of the pancreas,
breast, and prostate (Choudhuri et al.,
1997; Fang et al., 1992; Jager et al., 1997;
Souttou et al., 1998; Vacherot et al., 1999;
Weber et al., 2000). Numerous studies
using ribozyme, RNA interference, or
antisense strategies (Czubayko et al., 1994;
Czubayko et al., 1996; Grzelinski et al.,
2006; Hatziapostolou et al., 2005; Weber et
81
RESULTS
al., 2000) showed that HARP was a
potential in vivo rate-limiting angiogenic
factor in tumour growth and metastasis. As
a result, HARP and its two signalling
receptors ALK and RPTPβ/ζ are now
viewed as promising targets for cancer
therapy (Czubayko et al., 1994; Foehr et
al., 2006; Powers et al., 2002; Ulbricht et
al., 2006; Weber et al., 2000). Using
recombinant mutant HARP or a synthetic
peptide, we previously established that the
HARP C-terminus, composed of amino
acids 111 to 136, was closely involved in
the
mitogenic,
angiogenic,
and
transforming activities of HARP (BernardPierrot et al., 2001; Bernard-Pierrot et al.,
2002). Thus, P111-136 bound to the ALK
receptor, thereby acting as a dominant
inhibitor of the biological activities of
HARP. These findings, together with the
key role for HARP in prostate-tumour
growth, prompted us to further investigate
the potential anti-tumour effects of P111136, using the human androgenindependent adenocarcinoma PC-3 cell
line, which has an autocrine loop for
HARP (Vacherot et al., 1999).
In the in vitro and in vivo experiments
reported here, P111-136 inhibited the
growth of PC-3 cells. These results are
consistent with evidence that P111-136 is
an ALK-receptor antagonist, since PC-3
does not express RPTPβ/ζ. However, we
showed very recently that the shorter
synthetic peptide P122-131 – the Cterminal, basic, amino-acid motif derived
from P111-136 – inhibited the in vitro
growth
of
the
human
prostatic
adenocarcinoma cells DU145, which
expressed the RPTPβ/ζ receptor but not the
ALK receptor (Bermek et al., 2007). This
result suggested that P111-136 might bind
also to the RPTPβ/ζ receptor. In keeping
with this hypothesis, we found that P111136 inhibited the in vitro growth of DU145
cells, as well as of prostate cancer cells
(LNCaP), both devoid of ALK (data not
shown). The receptors normally involved
in the growth-promoting effect of HARP
are a matter of controversy (Mathivet et al.,
2007). Binding of HARP or its related
protein MK to the ALK receptor activated
the intracellular kinase domain and further
stimulated the downstream MAP and PI-3
kinase pathways (Bowden et al., 2002;
Stoica et al., 2001; Stoica et al., 2002).
Binding
of
HARP
to
RPTPβ/ζ
oligomerised the receptor and inactivated
its intracellular catalytic phosphatase
activity, leading to further activation of the
Src/Fyn kinase family and β-catenin
phosphorylation pathway (Meng et al.,
2000; Pariser et al., 2005). The RPTPβ/ζ
receptor was found to be involved in
HARP-induced cell migration and neurite
outgrowth (Maeda & Noda, 1998;
Polykratis et al., 2005) and was then shown
to play a role in glioblastoma cell
proliferation (Foehr et al., 2006; Ulbricht et
al., 2006). A very recent study (PerezPinera et al., 2007) identified an alternative
mechanism in which binding of HARP to
RPTPβ/ζ induced phosphorylation of ALK,
thereby activating the ALK receptor,
independently from a direct interaction of
HARP with ALK. These discrepancies may
be ascribable to differences in the cell
systems used and in the level of expression
of each receptor. However, a role for other
HARP cell-membrane receptors such as
cell-surface nucleolin (Said et al., 2005) or
unidentified co-receptors in the HARP
pathway remains a possibility that is being
investigated. The recent finding that HARP
occurs naturally as two variants, with the
shorter 124-amino acid variant lacking 12
amino acids in the C-terminus, prompted us
to evaluate the shorter C-terminus P111124 derived from P111-136. Our in vitro
and in vivo experiments showed that P111124 was as effective as P111-136 in
inhibiting the growth-promoting and
angiogenic effects of HARP. These data
suggest that the ALK-binding motif in the
HARP C-terminus may be located between
amino acids 111 to 124 and the RPTPβ/ζbinding motif between amino acids 125 to
131. Further experiments are needed to
investigate this possibility.
82
RESULTS
Evidence that C-terminus maturation
influences the biological effects of HARP
was obtained in an earlier study, in which a
14-kDa C-terminal truncated form of
HARP influenced the proliferation of BEL
cells (Souttou et al., 1997). Thus, natural
processing of the HARP molecule may be
pivotal in regulating the biodistribution and
biological effects of HARP in health and
disease. In keeping with this possibility, we
recently showed that plasmin and MMP-2
cleaved HARP in vitro, releasing various
peptides that may differentially affect the
angiogenic and mitogenic activities of
HARP (Dean et al., 2007; Polykratis et al.,
2004). GAGs in the microenvironment may
protect HARP from this enzymatic
cleaving. Furthermore, we found that
P111-136 bound to the C-TSR motif of
HARP involved in binding to heparin (data
not shown), which may constitute an
additional mechanism regulating the
bioavailability of HARP. Thus, the
biological effect of HARP is the net result
not only of HARP secretion and
degradation, but also of specific enzymatic
processing, which depends on the proteases
and GAGs present in the microenvironment
and may generate peptides that have
diverse (and perhaps opposite) biological
effects.
In vitro, P111-136 and P111-124 partially
inhibited the growth of PC-3 cells, and
similar partial inhibition was induced by
HARP antibody, suggesting that HARP
was not the only growth factor involved in
anchorage-independent growth of PC-3
cells. However, due to the rate-limiting
effect of HARP on in vivo PC-3 growth
and associated angiogenesis, each peptide
was more effective than in vitro in
inhibiting PC-3 xenograft tumour growth in
nude mice. This strong inhibitory effect,
similar to that of the clinical drug Taxol,
may be ascribable to the ability of each
peptide to inhibit both cell proliferation and
angiogenesis. Thus, P111-136 and P111124 may hold promise for the treatment of
prostate cancer. The effects of these
peptides in other animal models of human
cancer growth are in order.
83
RESULTS
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Cunha, G. R. and Donjacour, A. Stromal-epithelial interactions in normal and
abnormal prostatic development. Prog Clin Biol Res, 239: 251-272, 1987.
Russell, P. J., Bennett, S., and Stricker, P. Growth factor involvement in
progression of prostate cancer. Clin Chem, 44: 705-723, 1998.
Bernard-Pierrot, I., Delbe, J., Heroult, M., Rosty, C., Soulie, P., Barritault, D.,
Milhiet, P. E., and Courty, J. Heparin affin regulatory peptide in milk: its
involvement in mammary gland homeostasis. Biochem Biophys Res Commun,
314: 277-282, 2004.
Souttou, B., Juhl, H., Hackenbruck, J., Rockseisen, M., Klomp, H. J., Raulais, D.,
Vigny, M., and Wellstein, A. Relationship between serum concentrations of the
growth factor pleiotrophin and pleiotrophin-positive tumors. J Natl Cancer Inst, 90:
1468-1473, 1998.
Vacherot, F., Caruelle, D., Chopin, D., Gil-Diez, S., Barritault, D., Caruelle, J. P.,
and Courty, J. Involvement of heparin affin regulatory peptide in human prostate
cancer. Prostate, 38: 126-136, 1999.
Hatziapostolou, M., Delbe, J., Katsoris, P., Polytarchou, C., Courty, J., and
Papadimitriou, E. Heparin affin regulatory peptide is a key player in prostate cancer
cell growth and angiogenicity. Prostate, 65: 151-158, 2005.
Muramatsu, T. Midkine and pleiotrophin: two related proteins involved in
development, survival, inflammation and tumorigenesis. J Biochem (Tokyo), 132:
359-371., 2002.
Courty, J., Milhiet, P. E., Delbe, J., Caruelle, D., and Barritault, D. Heparin-Affin
Regulatory Peptide, HARP, p. 145-152: Springer, 2000.
Pufe, T., Bartscher, M., Petersen, W., Tillmann, B., and Mentlein, R. Expression of
pleiotrophin, an embryonic growth and differentiation factor, in rheumatoid
arthritis. Arthritis Rheum, 48: 660-667, 2003.
Heroult, M., Bernard-Pierrot, I., Delbe, J., Hamma-Kourbali, Y., Katsoris, P.,
Barritault, D., Papadimitriou, E., Plouet, J., and Courty, J. Heparin affin regulatory
peptide binds to vascular endothelial growth factor (VEGF) and inhibits VEGFinduced angiogenesis. Oncogene, 23: 1745-1753, 2004.
Bernard-Pierrot, I., Heroult, M., Lemaitre, G., Barritault, D., Courty, J., and
Milhiet, P. E. Glycosaminoglycans promote HARP/PTN dimerization. Biochem
Biophys Res Commun, 266: 437-442., 1999.
Stoica, G. E., Kuo, A., Aigner, A., Sunitha, I., Souttou, B., Malerczyk, C.,
Caughey, D. J., Wen, D., Karavanov, A., Riegel, A. T., and Wellstein, A.
Identification of anaplastic lymphoma kinase as a receptor for the growth factor
pleiotrophin. J Biol Chem, 276: 16772-16779, 2001.
Raulo, E., Tumova, S., Pavlov, I., Pekkanen, M., Hienola, A., Klankki, E.,
Kalkkinen, N., Taira, T., Kilpelainen, I., and Rauvala, H. The two thrombospondin
type I repeat domains of the heparin-binding growth-associated molecule bind to
heparin/heparan sulfate and regulate neurite extension and plasticity in
hippocampal neurons. J Biol Chem, 280: 41576-41583, 2005.
Bernard-Pierrot, I., Delbe, J., Caruelle, D., Barritault, D., Courty, J., and Milhiet, P.
E. The lysine-rich C-terminal tail of heparin affin regulatory peptide is required for
mitogenic and tumor formation activities. J Biol Chem, 276: 12228-12234, 2001.
Bernard-Pierrot, I., Delbe, J., Rouet, V., Vigny, M., Kerros, M. E., Caruelle, D.,
Raulais, D., Barritault, D., Courty, J., and Milhiet, P. E. Dominant negative
84
RESULTS
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
effectors of heparin affin regulatory peptide (HARP) angiogenic and transforming
activities. J Biol Chem, 277: 32071-32077, 2002.
Workman, P., Balmain, A., Hickman, J. A., McNally, N. J., Rohas, A. M.,
Mitchison, N. A., Pierrepoint, C. G., Raymond, R., Rowlatt, C., Stephens, T. C.,
and et al. UKCCCR guidelines for the welfare of animals in experimental
neoplasia. Lab Anim, 22: 195-201, 1988.
Bagheri-Yarmand, R., Kourbali, Y., Mabilat, C., Morere, J. F., Martin, A., Lu, H.,
Soria, C., Jozefonvicz, J., and Crepin, M. The suppression of fibroblast growth
factor 2/fibroblast growth factor 4-dependent tumour angiogenesis and growth by
the anti-growth factor activity of dextran derivative (CMDB7). Br J Cancer, 78:
111-118, 1998.
Mentlein, R. and Held-Feindt, J. Pleiotrophin, an angiogenic and mitogenic growth
factor, is expressed in human gliomas. J Neurochem, 83: 747-753, 2002.
Soulie, P., Heroult, M., Bernard, I., Kerros, M. E., Milhiet, P. E., Delbe, J.,
Barritault, D., Caruelle, D., and Courty, J. Immunoassay for measuring the heparinbinding growth factors HARP and MK in biological fluids. J Immunoassay
Immunochem, 23: 33-48, 2002.
Lu, K. V., Jong, K. A., Kim, G. Y., Singh, J., Dia, E. Q., Yoshimoto, K., Wang, M.
Y., Cloughesy, T. F., Nelson, S. F., and Mischel, P. S. Differential induction of
glioblastoma migration and growth by two forms of pleiotrophin. J Biol Chem,
280: 26953-26964, 2005.
Choudhuri, R., Zhang, H. T., Donnini, S., Ziche, M., and Bicknell, R. An
angiogenic role for the neurokines midkine and pleiotrophin in tumorigenesis.
Cancer Res, 57: 1814-1819, 1997.
Fang, W., Hartmann, N., Chow, D. T., Riegel, A. T., and Wellstein, A. Pleiotrophin
stimulates fibroblasts and endothelial and epithelial cells and is expressed in human
cancer. Journal of Biological Chemistry, 267: 25889-25897, 1992.
Jager, R., Noll, K., Havemann, K., Knabbe, C., Rauvala, H., and Zugmaier, G.
Differential expression and biological activity of the heparin-binding growthassociated molecule (HB-GAM) in lung cancer cell lines. Int J Cancer, 73: 537543, 1997.
Weber, D., Klomp, H. J., Czubayko, F., Wellstein, A., and Juhl, H. Pleiotrophin can
be rate-limiting for pancreatic cancer cell growth. Cancer Res, 60: 5284-5288.,
2000.
Czubayko, F., Riegel, A. T., and Wellstein, A. Ribozyme-targeting elucidates a
direct role of pleiotrophin in tumor growth. J Biol Chem, 269: 21358-21363., 1994.
Czubayko, F., Schulte, A. M., Berchem, G. J., and Wellstein, A. Melanoma
angiogenesis and metastasis modulated by ribozyme targeting of the secreted
growth factor pleiotrophin. Proceedings of the National Academy of Sciences of
the United States of America, 93: 14753-14758, 1996.
Grzelinski, M., Urban-Klein, B., Martens, T., Lamszus, K., Bakowsky, U., Hobel,
S., Czubayko, F., and Aigner, A. RNA interference-mediated gene silencing of
pleiotrophin through polyethylenimine-complexed small interfering RNAs in vivo
exerts antitumoral effects in glioblastoma xenografts. Hum Gene Ther, 17: 751766, 2006.
Foehr, E. D., Lorente, G., Kuo, J., Ram, R., Nikolich, K., and Urfer, R. Targeting
of the receptor protein tyrosine phosphatase beta with a monoclonal antibody
delays tumor growth in a glioblastoma model. Cancer Res, 66: 2271-2278, 2006.
85
RESULTS
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Powers, C., Aigner, A., Stoica, G. E., McDonnell, K., and Wellstein, A.
Pleiotrophin signaling through anaplastic lymphoma kinase (ALK) is rate- limiting
for glioblastoma growth. J Biol Chem, 23: 23, 2002.
Ulbricht, U., Eckerich, C., Fillbrandt, R., Westphal, M., and Lamszus, K. RNA
interference targeting protein tyrosine phosphatase zeta/receptor-type protein
tyrosine phosphatase beta suppresses glioblastoma growth in vitro and in vivo. J
Neurochem, 98: 1497-1506, 2006.
Bermek, O., Diamantopoulou, Z., Polykratis, A., Dos Santos, C., Hamma-Kourbali,
Y., Burlina, F., Delbe, J., Chassaing, G., Fernig, D. G., Katsoris, P., and Courty, J.
A basic peptide derived from the HARP C-terminus inhibits anchorageindependent growth of DU145 prostate cancer cells. Exp Cell Res, 2007.
Bowden, E. T., Stoica, G. E., and Wellstein, A. Anti-apoptotic signaling of
pleiotrophin through its receptor, anaplastic lymphoma kinase. J Biol Chem, 277:
35862-35868., 2002.
Stoica, G. E., Kuo, A., Powers, C., Bowden, E. T., Sale, E. B., Riegel, A. T., and
Wellstein, A. Midkine binds to anaplastic lymphoma kinase (ALK) and acts as a
growth factor for different cell types. J Biol Chem, 277: 35990-35998., 2002.
Meng, K., Rodriguez-Pena, A., Dimitrov, T., Chen, W., Yamin, M., Noda, M., and
Deuel, T. F. Pleiotrophin signals increased tyrosine phosphorylation of beta betacatenin through inactivation of the intrinsic catalytic activity of the receptor-type
protein tyrosine phosphatase beta/zeta. Proc Natl Acad Sci U S A, 97: 2603-2608,
2000.
Pariser, H., Ezquerra, L., Herradon, G., Perez-Pinera, P., and Deuel, T. F. Fyn is a
downstream target of the pleiotrophin/receptor protein tyrosine phosphatase
beta/zeta-signaling pathway: regulation of tyrosine phosphorylation of Fyn by
pleiotrophin. Biochem Biophys Res Commun, 332: 664-669, 2005.
Maeda, N. and Noda, M. Involvement of receptor-like protein tyrosine phosphatase
zeta/RPTP beta and its ligand pleiotrophin/heparin-binding growth-associated
molecule (HB-GAM) in neuronal migration. J. Cell. Biol., 142: 203-216, 1998.
Polykratis, A., Katsoris, P., Courty, J., and Papadimitriou, E. Characterization of
heparin affin regulatory peptide signaling in human endothelial cells. J Biol Chem,
280: 22454-22461, 2005.
Perez-Pinera, P., Zhang, W., Chang, Y., Vega, J. A., and Deuel, T. F. Anaplastic
Lymphoma Kinase Is Activated Through the Pleiotrophin/Receptor Proteintyrosine Phosphatase beta/{zeta} Signaling Pathway: AN ALTERNATIVE
MECHANISM OF RECEPTOR TYROSINE KINASE ACTIVATION. J Biol
Chem, 282: 28683-28690, 2007.
Souttou, B., Ahmad, S., Riegel, A. T., and Wellstein, A. Signal transduction
pathways involved in the mitogenic activity of pleiotrophin - Implication of
mitogen-activated protein kinase and phosphoinositide 3-kinase pathways. J Biol
Chem, 272: 19588-19593, 1997.
Polykratis, A., Delbe, J., Courty, J., Papadimitriou, E., and Katsoris, P.
Identification of heparin affin regulatory peptide domains with potential role on
angiogenesis. Int J Biochem Cell Biol, 36: 1954-1966, 2004.
Dean, R. A., Butler, G. S., Hamma-Kourbali, Y., Delbe, J., Brigstock, D. R.,
Courty, J., and Overall, C. M. Identification of Candidate Angiogenic Inhibitors
Processed by MMP-2 in Cell Based Proteomic Screens: Disruption of
VEGF/HARP (Pleiotrophin) and VEGF / CTGF Angiogenic Inhibitory Complexes
by MMP-2 Proteolysis. Mol Cell Biol, 2007.
86
RESULTS
ACKNOWLEDGEMENTS
This work was supported by grants from the CNRS, ANR-06-RIB-016-02, INCA, and
Association pour la Recherche sur le Cancer (#3242).
FIGURE LEGENDS
Figure 1: Inhibition of HARP mitogenic activity by peptide P111-136.
(A), stimulation of [³H]-thymidine incorporation in serum-starved NIH-3T3 cells treated
with 4 nM HARP in the presence or absence of different concentrations of synthetic
peptides P111-136. (B), stimulation of the phosphorylation of MAP kinases in serumstarved NIH-3T3 treated with FCS, or 4 nM HARP in presence or absence of 1 µM P111136. Phosphorylated ERKs and total ERKs present in cell lysates were detected by western
blotting using an anti-phospho p42/44 MAP kinase or anti-p42-44 antibodies. (C) and (D),
stimulation of [³H]-thymidine incorporation in serum-starved NIH-3T3 cells treated with 4
nM HARP in the presence or absence of different concentrations of P1-21 or poly-arginine.
Asterisks indicate a statistically significant difference from the corresponding untreated
cells. **, 0.001< p <0.01, ***, p < 0.001.
Figure 2: Inhibition of PC3 colonies formation by the peptide P111-136.
(A), RT-PCR analysis of the expression of HARP, ALK and RPTPβ/ζ in PC3 cells
cultured in various conditions. (1) negative control, (2) positive control from U87-MG
cells, (3) exponentially cultured PC-3 cells, (4) and (5) sub confluent cultured PC-3 cells.
TF2D was used as an internal control. PCR fragments corresponding to HARP, ALK,
RPTPβ/ζ and TF2D are 463 bp, 236 bp, 344 bp and 194 bp respectively. PC-3 cells were
seeded in triplicate into 12-well plates containing agar and DMEM supplemented by 10%
FCS and various concentration of (A) peptide P111-136, (B) polyclonal anti-human HARP
antibody or control IgG. Molecules were added in the culture medium twice a week. After
12 days, colonies with diameters greater than 50 µm were scored as positive using a phase
contrast microscope with a measuring grid. Asterisks denote a statistically significant
difference from the corresponding untreated cells. *0.01<p<0.1, **0.001<p<0.01, ***,
p<0.001.
Figure 3: Inhibition of tumour growth by the peptide P111-136.
6
(A) Tumour growth curves. PC3 carcinoma cells (2 × 10 ) were inoculated s.c. into the
3
right flank of female nude mice. When tumours reached about 60 mm (2 weeks), Peptide
P111-136 (5 mg/kg/day) or Taxol (10 mg/kg/once a week) were administrated s.c. for 4
weeks. Tumours were measured twice a week along two major axes (B) Tumour weights.
Experiments were performed as in A. The mice were sacrificed 6 weeks post injection, all
tumours were excised, and the tumour weights were determined. The results are presented
as the mean tumour volume or weight ± SE obtained from 5 mice in each group. *P<0.05,
**P<0.01, ***P<0.001; versus control (untreated tumours).
Figure 4: Inhibition of tumour angiogenesis by the peptide P111-136.
PC3 Tumour vascularization was analyzed immunohistologically with the endothelial cellspecific marker CD31 (Magnification × 200). Untreated tumours (A) and tumours treated
with peptide P111-136 (5 mg/kg/day) (B). (C), angiogenesis was quantified by image
analysis of CD31-labeled ECs. The results are presented as the mean areas ± SE (bars) of
ECs in tumour section obtained from 5-control and 5-P111-136 treated mice. **,
significantly different from control (P < 0.01).
87
RESULTS
Figure 5: Inhibition of HARP-induced in vivo angiogenesis by peptide P111-136.
Liquid Matrigel at 4°C was sub-cutaneously injected into Swiss mice alone or containing
molecules. Eight days later, the animals were sacrificed; the Matrigel plugs were removed,
sectioned, and stained using the Gomori-Trichrome method (magnification × 200).
Micrographs A-D represent Gomori-Trichrome staining of a sectioned Matrigel plug alone
(A) or containing 1 µM peptide P111-136 (B), 5 nM HARP (C), 5 nM HARP/1 µM P111136 (D). Quantification of endothelial cell invasion into the Matrigel was determined as a
mean of six fields per section from 3 sections of Matrigel plugs per mouse and the results
are expressed as the mean of four mice per group (E). Standard errors are indicated (**, p
< 0.01).
Figure 6: Inhibition of PC3 colonies formation in vitro and tumour growth in vivo by
peptide P111-124.
(A), PC3 cells suspended in soft agar were treated with peptide P111-124 for 12 days. At
that time, colonies were determined such as described in Figure 2. (B) PC3 tumour growth
treated with p111-124. (C) Tumour weights. Experiments were performed as in Figure 3.
Asterisks denote a statistically significant difference from the corresponding untreated
cells. *0.01<p<0.1, **0.001<p<0.01.
Figure 7: Inhibition of HARP mitogenic activity by peptide P111-124.
(A), Stimulation of [³H]-thymidine incorporation in serum-starved NIH-3T3 cells treated
with 4 nM HARP in the presence or absence of increased concentrations of P111-124 or
P111-136 peptides. (B), Lysate from serum-starved NIH-3T3 control or treated cells was
separated on 10% SDS-PAGE. Phosphorylated ERKs and total ERKs present in the lysate
were detected by western blotting as presented in figure 1. Asterisks indicate a statistically
significant difference from the corresponding untreated cells. **, 0.001< p <0.01, ***, p <
0.001.
88
RESULTS
Figure 1
89
RESULTS
Figure 1
90
RESULTS
91
RESULTS
Figure 2
B
5
4
*
3
***
2
1
0
0
0.1
0.5
1
P111-136 (µM)
C
5
4
3
*
2
**
1
0
0
0.01
0.05
0.1
Anti-HARP (µM)
0.1
IgG
92
RESULTS
Figure 3
93
RESULTS
Figure 4
94
RESULTS
Figure 5
95
RESULTS
Figure 6
96
RESULTS
Figure 7
97
RESULTS
1.2.2 UNPUBLISHED RESULTS
1.2.2.1 Proliferation of glioblastoma cells transfected with cDNA of HARP Δ111-136 and
HARP1 Δ124- 136
In article 1, we demonstrated that the peptide derived from the C-terminal of
HARP, designated P111-136 inhibited the cancer-related actions of HARP including
mitogenic activity, anchorage-independent growth, tumor growth in nude mice and in vitro
and in vivo angiogenesis and the peptide derived from a C-terminal lacking 12 amino
acids, designated P111-124 displayed the same antagonist effects of P111-136. Therefore,
the active sequence within the amino acid residues 111-136 was the sequence 111-124.
To evaluate these results by another approach, we transfected the human
glioblastoma cell line U87MG2 with eucaryotic expression vector pcDNA3.1 alone or
pcDNA3.1-HARPΔ124-136 and pcDNA3.1-HARPΔ111-136. HARP constructs were
engineered using a PCR primer to insert a TAG stop codon at the 124 or 111 translating
into a 3’ truncation of 12 or 26 amino acids. U87MG cells were then stably transfected
with HARP constructs using Fugene6 reagent according to the manufacturer’s protocol.
The expression of the mutant proteins was verified by Coomassie Blue coloration and
Western Blot analysis. Then, we observed the proliferation of these stably transfected
U87MG clones.
We expected two outcomes from these experiments, illustrated at the bottom
figure (Figure 11).
1
For clarity, it is important to note that HARP1-124, HARPΔ124-136 and HARP15 all designate the same C-
terminally processed-HARP and HARP18 designate the full-length intact HARP1-136.
2
It is important to remind that, according to the work of Lu et al., HARP15 induces ALK-mediated
proliferation and HARP18, RPTPβ-induced migration and both forms of the protein and both receptors of
HARP are endogenously produced in U87MG cells. HARP15 and HARP18 used in the migration and
proliferation assays were purified from the conditioned media of HEK293 cells transfected with pcDNA3.1HARP15 or pcDNA3.1-HARP18.
98
RESULTS
Overexpression of
mitogenically active
HARPΔ124-136
INDUCED-PROLIFERATION
Dominant negative
Formation of
inactive dimers
INHIBITION OF PROLIFERATION
Figure 11: Proliferation of transfected U87MG cells by HARP mutants.
Since HARP15 -that includes the putative sequence 111-124 was the
mitogenically active form in these cells, the over-expression of this form in these cells
induced U87MG proliferation (figure 12A). As a validation of this experiment, the overexpression of HARP mutant which lacks this sequence resulted in the inhibition of
proliferation (figure 12B).
Thus, to assess the implication of the sequence 111-124 in the biological activity
of HARP, in this and previous (article 1) studies, we used two different approaches. In
peptidic approach, we synthesized a peptide corresponding to this sequence and reported
the antagonist effect of the peptide for ALK-mediated biological activities of HARP
(article1). In a molecular biological approach, we transfected glioblastoma cells with a
HARP mutant including 111-124 and the over-expression of this mutant resulted in
induced-proliferation (figure 12A). We confirmed this result by transfecting the same cells
this time with a HARP mutant deleted of the putative sequence and this resulted in the
inhibition of proliferation as expected (figure 12B). Therefore, both approaches pointed
the implication of the amino acid residues 111-124 in the mitogenic activity of HARP.
99
RESULTS
500
**
A
400
HARPΔ124-136 clone(6)
300
HARPΔ124-136 clone(8)
200
pcDNA3 clone(4)
pcDNA3 clone(9)
100
0
HARPΔ124-136 clone(3)
0
1
2
days
3
4
300 B
pcDNA3 clone(9)
250
pcDNA3 clone(4)
200
150
HARPΔ111-136 clone(2)
100
**
HARPΔ111-136 clone(1)
50
0
HARPΔ111-136 clone(3)
0
1
2
3
4
days
Figure 12: Growth properties of U87MG transfected with HARP constructs. A) U87MG cells
transfected with eucaryotic expression vector pcDNA3.1-HARP∆124-136, B) cells transfected with
pcDNA3.1-HARP∆111-136. Control cells were transfected with the plasmid pcDNA3.1 alone. For
proliferation assays, 6 × 104 cells were seeded in the wells of 12-well plates in triplicates in the
MEMα medium supplemented with 10% FCS and 200 µg/ml of G418 and number of cells were
counted each day during 4 days.
100
RESULTS
1.2.2.2 Bio-distribution of radiolabeled [I125]-P111-124 in nude mice
In this study, we evaluated the potential use of P111-124 as a therapeutic tool. In
this aim, we3 have studied the pharmacokinetic parameters and the in vivo distribution of
the radiolabeled peptide. In order to label P111-124 with iodine-125, N- and C-terminaltyrosinylated two peptides were synthesized. The sequences of the peptides are as
following: (N-tyrosinylated peptide) NH2-Y-AAA- KLTKPKPQAESKKK-COOH and (Ctyrosinylated peptide) NH2- KLTKPKPQAESKKK-AAA-Y-COOH (Altergen). The
bioactivity of the peptides was tested by [methyl-3H]thymidine incorporation assays. The
results indicated that the activity of N-terminal tyrosine containing P111-124 was lost upon
tyrosinylation, while the C-terminal-tyrosinylated analog was active (Figure 13).
(-) HARP
30000
HARP (3.5 nM)
HARP + N-Tyr-P111-124 (0.1 µM)
*
20000
*
10000
**
HARP + N-Tyr-P111-124 (1 µM)
HARP + N-Tyr-P111-124 (10 µM)
HARP + C-Tyr-P111-124 (0.1 µM)
HARP + C-Tyr-P111-124 (1 µM)
HARP + C-Tyr-P111-124 (10 µM)
0
concentration (µM)
Figure 13: Test of bioactivity of tyrosinylated peptides. DNA synthesis was stimulated by
HARP on NIH 3T3 cells. Cells were labeled with [3H] thymidine after 18 h of addition of 3.5 nM
of HARP and peptides at the concentrations indicated. The standard errors indicated: *, 0.01 < p <
0.1; **, 0.001 < p < 0.01.
Hence, C-[tyrosin]-P111-124 peptide was labeled with iodine-125 [I125] using
iodobeads (Pierce) at 1 mCi/15 µg of peptide, in 100 µl of PBS and the products were
purified on a mini column C18 (sep-pak).
3
This work was done in collaboration with the Laboratory of Biochemical Radiopharmaceuticals (INSERM,
la Tronche, France).
101
RESULTS
First, the stability of peptides was evaluated by HPLC on C4 Waters column.
The radioactivity associated with the peptide was measured after 1, 3 or 6 h of incubation
in presence of physiological liquid or of blood (Cobra II - Packard). The results showed
that the peptide was stable for at least 6 hours. It is noteworthy that 15% of the peptide was
fixed to blood proteins, while 75% was free in blood.
Next, the biodistribution of the peptide in nude mice bearing tumors was studied.
The radiolabeled peptides (200 µCi) were injected into nude mice in which the human
breast cancer cells MDA MB-231 were previously implanted. After 30 min or 6 h of the
intravenous injection, the presence of the radiolabeled peptide was detected in following
organs: liver, kidney, thyroid, stomach and tumor (Figure 14).
100
50
0
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0100.0
heart
lung
liver spleen kidn. stom...intes.. musc.. bone
fat
tumor brain blood thyroid
Figure 14: Biodistribution of [I125]- P111-124 after 30 min and 6 h of the intravenous
injection. Approximately 200 mCi of peptide were injected to the anaesthetized nude mice and
after 30 min and 6 h of the injection, the following organs were taken: heart, lung, liver, spleen,
kidney, stomach, intestine, muscle, bone, fat, tumor, brain, blood, thyroid. Then, the radioactivity
was measured (Cobra II – Packard). The results were expressed as percentage of injected dose
versus weight of organ (% ID/g).
The results indicated that the peptide was rapidly and mainly eliminated by
kidney with 15% ID/g from 30 min after injection and a less important but still rapid
elimination was observed by liver. The activity observed in thyroid was mainly due to free
iodine and the activity in stomach may be explained by the presence of an enzyme, which
is known to cut iodine bonds. There is no radioactivity detected in the brain, probably due
to blood brain barrier which prevents the delivery of the peptide to the brain. The drop off
102
RESULTS
observed in the blood after 6 hours points the elimination of most of the circulating
peptide. The tumoral activity represented 1.6 ± 0.5% ID/g after 30 min of injection and
diminished to 0.18 ± 0.07% ID/g after 6 h. It should be noted that it is not known the
maximal tumoral activity and the profile of the activity of the peptide between 30 min and
6 h and it is necessary to do a kinetic of the activity. It is also noteworthy that the average
tumoral mass was 51.2 ± 13.2 mg and despite the low size of tumors, a high tumoral
activity was detected. The results of this very first in vivo bio-distribution test suggested
that P111-124 is successfully delivered to the tumors after 30 min of intravenous injection
and this, despite the low size of tumors, and eliminated from the tumors after 6 h of
injection by kidney and liver, hence promising a potential therapeutic tool in treatment of
these tumors.
1.2.2.3 Isolation of cell-surface receptors for P111-136
As demonstrated in Article 1, P111-136 inhibits most of the biological activities
of HARP including the mitogenic, transforming, tumor promoting and angiogenic ones.
Although the inhibitory effects of the peptide are explained by the hypothesis that the
peptide binds to ALK and competes with HARP for the binding to ALK, it is not known
whether these various activities of the protein are mediated by only one receptor, e.g.
ALK. Because HARP is a molecule having several signaling receptors, most notably ALK
and RPTPβ, we could not exclude the possibility that P111-136 might bind also to RPTPβ
and/or to the other cell surface-expressed proteins.
In view of this, we have investigated the potential molecular partners other than
ALK for P111-136. We used PC3 cells for these experiments, since P111-136 inhibited the
biological activities of the autologous HARP in these cells (Article 1). In order to isolate
P111-136-binding proteins from PC3 cells, biotin-labeled synthetic peptide was utilized
(Altergen). PC3 cells were incubated with the biotin-labeled P111-136, at a concentration
that displayed inhibitory effects (5 µM). Peptide-binding proteins were purified from crude
cell extracts (material corresponding to 4.5 × 107 cells) by affinity chromatography using
avidin-agarose (see “Materials and Methods”). Proteins were then analyzed by SDS-PAGE
using 10% polyacrylamide gel and the protein bands were revealed by staining with the
colloidal coomassie blue. Figure 15 shows the profile of the purified proteins. Three major
proteins of 50, 75 and 100 kDa were purified by this experimental procedure (lane 1). No
proteins were recovered by avidin-agarose in absence of peptide (control) (lane 2). The
specific binding of P111-136 to the purified proteins was further analyzed by ligand
103
RESULTS
blotting, incubating with biotin-labeled P111-136. By this experimental approach, each of
the proteins was shown to bind biotinylated P111-136, hence showing the specificity of the
binding (Figure 15B, lane 4).
A
B
250
150
100
75
50
37
1
2
3
4
5
6
Figure 15: Specific binding of the biotinylated P111-136 to the 50-, 75- and 100-kDa proteins.
PC3 cells were incubated with 5 µM of biotinylated P111-136 (2 h, room temperature). The
complexes formed between the biotin-labeled peptide and any cell surface protein was recovered
by purification using avidin-agarose. The proteins were analyzed by SDS-PAGE. A) The proteins
were revealed by staining by colloidal coomassie blue, B) The specificity was confirmed by ligand
blotting using biotin-labeled P111-136. The presence of biotinylated peptide was revealed by
peroxidase-coupled anti-biotin. Lanes 1 and 4 represent proteins bound to biotin-labeled P111-136,
lanes 2 and 5 represent proteins recovered from avidin-agarose in absence of peptide (non-specific
bands) and lanes 3 and 6 represent weight marker.
These results, therefore, demonstrate that P111-136 binds specifically to the cellsurface expressed 50-, 75- and 100- kDa proteins. The three proteins isolated from PC3
cell extracts were analyzed by mass spectroscopy4 after digestion with trypsin but so far
any attempts for the identification of the bands remained without result, and this, most
probably because of the ineffectiveness of the enzymatic digestion of the protein after
purification on avidin-agarose beads. The study is still in progress.
4
This work is realized by using the facilities of Institut Pasteur, Proteomics Platform.
104
RESULTS
Using the same experimental approach, we have also tested the smaller peptide
P111-124 derived from the C-terminal region lacking 12 amino acid from the carboxy end
since the peptide was reported to mimic the inhibitory effects of P111-136 to a similar
extent (Article 1). Interestingly, no protein bound to biotin-labeled P111-124 was detected
(data not shown). This may be explained by the loss5 of activity of the peptide upon
biotinylation. The biotinylation was performed at the primary amino group (ε-amino
group) which exists in the side chain of a lysine residue. Then, labeling of the peptide was
carried out by coupling the so-formed biotinylated lysine residue and a linker glycine
residue to the sequence of P111-124 to the carboxy side: NH2-KLTKPKPQAESKKKK(Biotin)G-COOH. Similarly, the coupling of a tyrosine residue to the amino terminal of
P111-124 also resulted in loss of activity (see previous “Unpublished Result”). In case of
P111-136, the biotinylation was carried out by coupling a glycine residue to the sequence
of P111-136 as a linker before introducing biotin to the free α-amino group of this glycine
residue (see “Appendix” for the molecular formula of the amino acids).
It is a known fact that, with large peptides, labeling does not usually harm
function or binding properties. With short peptides, however, biotinylation process is much
more likely to block binding sites and labeling may alter their functional properties. In
addition, considering the fact that P111-124 is the active sequence within the P111-136,
one can expect that P111-124 is more sensitive to any modification within its sequence
than P111-136 and any modification on its sequence, especially introducing such bulky
label to the sequence may create a sterical hindrance and alter the conformation of the
sequence in proximity to the label, so resulting in the loss of activity.
Inhibition of HARP mitogenic activity by P111-121:
The loss of activity upon labeling of P111-124 prompted us, somehow, to
question the requirement of the three lysine residues of P111-124 for the biological action
of the peptide. It is also important to note that P111-124 lacks already 5 lysine residues that
establish the strong net positive charge of P111-136, meanwhile retained the inhibitory
activity of the former peptide. Therefore, we synthesized a peptide corresponding to amino
acids from 111 to 121, lacking three lysines from the sequence of P111-124. So, the
sequence of the peptide is following: NH2-KLTKPKPQAES-COOH. We then tested the
activity of the peptide P111-121 by our routine biological test, i.e. thymidine incorporation
5
Nevertheless, the bioactivity of the biotin-labeled peptide hasn’t been tested because of technical problems.
105
RESULTS
assay and investigated the potential of the peptide to inhibit the mitogenic activity of
HARP on NIH 3T3 cells stimulated by HARP (Figure 16).
30000
(-) HARP
20000
10000
*
*
HARP (3.5 nM)
***
HARP + P111-121 (0.1 µM)
HARP + P111-121 (1 µM)
HARP + P111-121 (10 µM)
0
concentration (µM)
Figure 16: Inhibition of HARP mitogenic activity by P111-121. [3 H]thymidine incorporation of
serum-starved NIH-3T3 cells treated with 3.5 nM of HARP in presence of various concentrations
of P111-121. The results are the means of two separate experiments carried out in triplicate, and
the standard errors indicated: *, 0.01< p< 0.1; ***, p < 0.001.
The results showed that P111-121 inhibited thymidine incorporation into
fibroblast 3T3 stimulated by HARP. The profile of the inhibitory character of P111-121
was more similar to that of P111-136 than that of P111-124. The stimulation of cells by
HARP was completely abolished at 1 µM similar to P111-136, while the maximal
inhibition observed with P111-124 was at 10 µM. The results also showed that the peptide
became inactive at higher doses in the same manner with P111-136. In contrast, P111-124
presented a profile of inhibition in a dose-dependent manner. Further studies are needed to
confirm the antagonist effects of P111-121 but, this first study suggests that the three
lysines in the sequence of P111-124 seem not to be involved in its inhibitory action.
106
RESULTS
1.2.2.4 Structure-Function Relationships
One of the mechanisms proposed to account for HARP signaling through ALK
receptor, is a dual receptor system consisting of ALK and HSPGs and it is believed that
both receptors must be present to trigger a proliferative response. However, it is not known
how HARP and HS co-operate to induce the dimerization of ALK and thus activate its
intracellular activation. Furthermore, little is known about the mechanism of dimerization
of HARP through HSPGs.
Previous studies have demonstrated that HARP mitogenic activity is potentiated
in presence of heparin (Vacherot et al., 1999b), which, in other studies, was reported to be
related to the dimerization process of HARP (Bernard-Pierrot et al., 1999). Moreover,
HARP mitogenic activity is mediated through its receptor ALK and ALK-binding region
of HARP is reported as to be the C-terminal end of HARP (Bernard-Pierrot et al., 2002).
The HARP mitogenic signaling model through ALK in presence of HSPGs is illustrated at
the bottom figure.
Figure 17: HARP mitogenic signaling through ALK in presence of HSPGs. In this model,
inactive HARP monomers form active dimers in presence of HSPGs. Then, HARP dimers induce
dimerization of ALK, which in turn induce autophophorylation.
107
RESULTS
In the light of these data, we have examined the structure-function relationships
that have been proposed to account for the heparin-binding properties of HARP and its
receptor ALK, using two synthetic peptides and binding assays in an optical biosensor.
Two synthetic peptides used in the binding assays, P111-136 and P65-111
(CTSR) were shown to antagonize respectively; the interaction of HARP and its receptor
ALK (Article 1) and the interaction of HARP and its co-receptor heparin/HS (HammaKourbali et al., 2007) and both, to inhibit the mitogenic, transforming and angiogenic
activities of HARP.
A set of binding experiments was performed. In the first set of experiments, the
extent of binding of ALK ECD6, HARP, P111-136 and CTSR to immobilized P111-136
was measured (Figure 18A). Based on these binding assays, we confirmed once more the
C-terminal domain as ALK-interacting region of HARP. Interestingly, HARP, added as a
soluble ligate to the P111-136-derivatized surface, in absence of heparin, was also shown
to bind to its C-terminal tail. No binding of P111-136 was detected, excluding the possible
dimerization of the peptide and excluding also a possible dimerization of HARP through
their C-terminal domains. Furthermore, the binding assays revealed, for the first time, the
affinity of the heparin-binding region of HARP (CTSR) for the C-terminal part of the
protein. Taken together, these results propose a new insight for HARP dimerization such
that, the C-TSR domain of one molecule of HARP interacting by the C-terminal domain of
another molecule of HARP.
To further investigate the interactions of the internal domains of HARP, a second
type of experiments was performed. In these experiments, the extents of binding of soluble
ligates; ALK ECD, HARP and CTSR to immobilized HARP were measured (Figure 18B).
Under these experimental conditions, in which biotinylated HARP was immobilized to the
streptavidin-derivatized aminosilane surface, the binding capacity of ALK ECD to HARP
was significantly lower than its binding capacity to immobilized P111-136. The drop-off
observed in the extent of binding may be due to the biotinylation process of HARP and the
binding site in HARP recognized by ALK may not be accessible or lost upon
biotinylation/immobilization procedure. In this procedure, 2 mol of biotin molecules (NHS
(LC) biotin) was coupled to an amino group in 1 mol of HARP and an important lysine
residue from the very basic, lysine-rich C-terminal domain might be lost/blocked by this
6
Recombinant ALK corresponding to the extracellular domain (ECD) of the protein (from Vigny).
108
RESULTS
reaction. Nevertheless, under the same conditions, HARP, again in absence of heparin, and
its thrombospondin type domain related to C-terminal, CTSR, were still able to bind to
immobilized HARP. Based on the assumption that, the binding site of immobilized HARP
to ALK –which is in the C-terminal region- is lost upon biotinylation of HARP, these
results would suggest that the binding site of C-terminal fragment to ALK may be different
than the C-terminal fragment implicated in interacting with CTSR domain.
150
A
P111-136-derivatized surface
ALK ECD (0.2 µM)
100
HARP (0.18 µM)
50
P111-136 (500 µM)
CTSR (20 µM)
0
concentration (µM)
50 B
HARP-derivatized surface
40
ALK ECD (0.3 µM)
30
HARP (0.21 µM)
20
CTSR (10 µM)
10
0
concentration ( µM)
Figure 18: Interaction of proteins with immobilized biotinylated P111-136 and HARP.
Proteins were added as soluble ligates in 5 µl of PBS-T to A) biotinylated P111-136, B)
biotinylated HARP cuvette containing 25 µl PBS-T. The association data were analyzed with a
one-site binding model to determine the extent of binding of ALK ECD (0.3 µM), HARP (0.21
µM), P111-136, CTSR. Results are triplicates of one of two separate experiments carried out on
different biotinylated P111-136 and HARP derivatized cuvettes.
109
RESULTS
The generally believed model for HARP-mediated signaling is that binding of
heparin to HARP induces the formation of non-covalent dimers of HARP and this dimeric
form is required for the efficient interaction of HARP with ALK to induce, in turn, the
dimerization of ALK and consequently its activation and autophosphorylation (Figure 17).
HARP is a heparin-binding protein but so far nothing is known about the heparin-binding
potential of the receptor ALK. To test this possibility, we examined the ability of ALK to
bind to the heparin-derivatized surface. The result of this binding experiment indicated that
ALK binds to heparin with a high affinity and thus ALK is a heparin-binding protein
(Figure 19).
HARP-derivatized surface
100
75
ALK ECD (0.1 µM)
50
HARP (0.03 µM)
25
0
concentration (µM)
Figure 19: Binding of ALK ECD to immobilized heparin. ALK ECD in 2 µl of PBS-T was
added to the heparin-derivatized cuvette containing 28 µl of PBS-T (final concentration, 0.1 µM)
and the association was followed for 6 min. The experiment was carried out on two separate
heparin immobilized surface. The extent of binding was calculated by fitting the association curve
to a single site binding model using FastFit software (Affinity Sensors).
Most protein-heparin interactions depend on ionic bonding between the sulfate
groups of heparin and the amino groups of the protein, and washes of 2 M NaCl are
usually sufficient to regenerate the surface, following to the association. However, 2 M
NaCl failed to remove all the bound ALK ECD from heparin immobilized surface.
Additional regeneration steps with 20 mM HCl, 20 mM NaOH, 2 M guanidine, 2 M NaCl
supplemented with 0.1% of SDS or several combinations of the regenerating agents were
applied but we were unable to regenerate the surface completely.
The avidin-biotin interaction is the strongest known non-covalent, biological
interaction (Kass = 1015 M-1) between protein and ligand. The bond is unaffected by
extremes of pH, high temperature, organic solvents and other denaturating agents, and 8 M
110
RESULTS
guanidine-HCl is required for efficient dissociation of the complex. The use of detergents
like SDS decreases the binding strength. Concerning the heparin-ALK interaction, even
though we applied all extreme possible conditions, without disrupting the interaction of the
biotinylated heparin to the immobilized streptavidin in the surface, we could not overcome
the regeneration difficulty of ALK from heparin. These results point that heparin-ALK
bond is unexpectedly strong and approaches the strength of a covalent bond and to a Kdiss
close to that of biotin-avidin complex. It is noteworthy that the bond formation between
ALK and heparin was also very slow and once formed, was unaffected by extreme
conditions as we discussed.
1.3 DISCUSSION
HARP is principally composed of N- and C-domains held by a flexible linker
and disulfide bridges. There are short lysine-rich tails at both ends of the molecule. Both
N- and C-domains contain β-sheet structure and show weak homology to the
thrombospondin type I repeats (TSR) present in numerous extracellular proteins that
interact with the cell surface. The N- and C-terminal lysine rich domains have no apparent
structure and form random coils (Kilpelainen et al., 2000).
HARP displays several biological activities and different domains of HARP
appear to display different activities. For example, both β-sheet domains carry out the
neurite outgrowth activity of HARP (Raulo et al., 2005), while the transforming domain
was identified as N-domain of HARP in cooperation with N- or C-terminal highly basic
tails (Zhang et al., 1999). Furthermore, the β-sheet domains are the heparin-binding sites
of HARP as well (Kilpelainen et al., 2000) and the co-operative action of both domains is
required for efficient binding (Raulo et al., 2005). The understanding of the relationships
between function, amino acid sequence and structure of HARP has been one of the major
challenges of our team for many years and our group have shown that the cancer-related
activities were carried out by C-terminal basic terminal, which lacks a detectable structure
(Bernard-Pierrot et al., 2002).
More than 150 proteins, under native conditions contain disordered regions of 30
consecutive residues or longer. It has been experimentally shown that disordered regions
are involved in molecular recognition and molecular recognition is primarily used for
signaling (Iakoucheva et al., 2002). One of the advantages of disordered binding sites is
that their multiple conformations allow them to recognize several targets with high
111
RESULTS
specificity and low affinity-ideal properties for signal transduction (Dunker et al., 2002).
Some of the proteins undergo folding in presence of a ligand, leading to disorder-to-order
transition upon binding. Native disorder has been widely implicated in cancer-associated
proteins present in the human genome (Ward J.J. et al., 2004).
Native proteins can adopt one of three forms: fully folded, partially disordered
and fully disordered (Dunker A.K. and Obradovic Z., 2001). A study using a method for
predicting native disorder, DISOPRED2, showed that the C-terminal of HARP (amino
acids 111-136) lacking a detectable structure is such disordered region, hence HARP is a
partially disordered protein (Figure 20).
Figure 20: Disordered profile plot of HARP. The plot shows position in the sequence of HARP
against probability of being disordered. In total, 38 of the 136 residues were classed as disordered
at the default threshold. The horizontal line is the order/disorder threshold for the default false
positive rate of 5%. The ‘filter’ curve represents the outputs from DISOPRED2 and the ‘output’
curve the outputs from a linear SVM classifier (DISOPREDsvm). The outputs from
DISOPREDsvm are included to indicate shorter, low confidence predictions of disorder.
Considering the facts mentioned above, it is not surprising that deletion of the Cterminal end of HARP abolished the mitogenic and tumor forming activities of the protein.
Furthermore, our laboratory has also reported the dominant negative effects of the mutant,
HARP∆111-136 and the corresponding synthetic peptide P111-136 (Bernard-Pierrot et al.,
2002). Therefore, we focused on the C-terminal end of HARP and investigated the
structural determinants playing a role in biological activities of the protein, including the
112
RESULTS
mitogenic, angiogenic and tumor promoting ones. To establish a structural basis for these
activities, we synthesized a synthetic peptide corresponding to the whole C-terminal region
of HARP. In the meantime, Lu et al have isolated another form of HARP from
glioblastomas; HARP 15 (residues 1-124) lacking the 12 C-terminal amino acids
(KKEGKKQEKMLD) of HARP 18.
In their studies, they showed that HARP 15
promoted proliferation of glioblastoma cells in an ALK-dependent fashion while, HARP
18 was inactive. Combining this latter finding with our previous studies pointed the
importance of 14-amino acid-fragment in the C-terminal and it led us to synthesize the
peptide corresponding to this segment. Therefore, the sequences of the peptides
synthesized are as followed:
P111-136: KLTKPKPQAESKKKKKEGKKQEKMLD
P111-124: KLTKPKPQAESKKK
In the first part of our study, we demonstrated that the synthetic peptides
inhibited the in vitro and in vivo HARP biological activities. Not only we demonstrated the
importance of the C-terminal region of HARP in its activities by means of P111-136, we
also identified the functional domain within the C-terminal tail, since P111-124, lacking 12
amino acid residues from the sequence of P111-136 retained the inhibitory potential of
P111-136.
Once determined the sequence playing a role in the cancer-related activities of
HARP, we next examined the use of P111-124 as potential therapeutic agent to combat
cancers in which HARP is implicated. The studies carried out with radiolabeled P111-124
in nude mice showed the efficient targeting of the peptide to the tumor, upon intravenous
injection. Moreover, the stability of the peptide, which is at least 6 hours in physiological
conditions provides further experimental evidence, supporting the potential use of P111124 as an anti-tumoral agent.
HSPGs, RPTPβ and ALK have been identified as binding partners of HARP in
the literature. Among them, ALK has been proposed to be the physiological receptor of
HARP and the mitogenic, angiogenic and tumor promoting activities of HARP have been
linked to its binding to ALK. The previous studies in our laboratory have reported the Cterminal region of HARP as binding site for this receptor. Moreover, the inhibition of
biological activities of HARP by P111-136 was explained by the competition of the
peptide with HARP for binding to ALK (Bernard-Pierrot et al., 2002). Therefore, in our
current work, we focus, at first, on ALK as binding partner of P111-136 and we performed
a set of binding assay in an optical biosensor. In these assays, we immobilized the
113
RESULTS
synthetic peptide P111-136 to the surface and added ALK as a soluble ligate. The results
indicated that ALK specifically bound to immobilized P111-136 and the binding was
almost completely competed by soluble HARP, confirming by another experimental
approach that the binding site of HARP to ALK is the sequence 111-136.
Although ALK was proposed to be the physiological receptor of HARP (Stoica
et al., 2001); the conflicting views exist and the proposal of the group of Wellstein; HARP
to be the ligand of ALK, has not been confirmed by other groups (Dirks et al., 2002,
Miyake et al., 2002, Motegi et al., 2004, Moog-Lutz et al., 2005). This subject was
discussed in detail in the “Introduction” of the thesis (see “Receptors-InteractionsMechanism of Action –ALK”). In the light of this, we couldn’t exclude the possibility that
the peptide might display its antagonist effects via other receptors of HARP, e.g. RPTPβ.
Furthermore, the synthetic peptide, as a domain derived from the region implicated in
molecular recognition could be useful as tool for studying other unknown cell-surface
expressed partners of HARP. Using biotinylated P111-136, we isolated three proteins of
50, 75 and 100 kDa from the cell surface of PC3 cells. The identity of the proteins remains
to be elucidated but it is possible that these proteins are related to the variants of ALK or
RPTPβ. Although the two major species identified for ALK are 220- and 140-kDa protein
(Perez-Pinera et al., 2007), several authors also reported a phosphorylated 80-kDa protein
product of npm-alk gene detected by anti-ALK antibodies (Iwahara et al., 1997, Pulford et
al., 1997). The spliced products of RPTPβ correspond to 230-, 130- and 85-kDa proteins
(Meng et al., 2000). Furthermore, we can even speculate that the band that migrates at 100
kDa may correspond to the cell-surface nucleolin (95-kDa protein), which was recently
identified as another receptor of HARP. Therefore, the characterization of the bands seems
to be necessary for further studies.
Most heparin-binding growth factors have been found to require a dual-receptor
system for cell signaling. Concerning HARP, it is widely believed now that, to stimulate
cell growth and probably for some other biological activities, HARP engages its two
receptors, notably HSPGs and ALK, and the interaction with both receptors is required for
the growth-promoting action. However, the molecular mechanisms that underpin the
binding and signaling functions of HARP-heparin-ALK system remain unclear.
Originally, Kilpelainen et al. has reported that binding to heparin induces a
change in conformation of HARP (Kilpelainen et al., 2000). As well as inducing a
conformational change in HARP, there is also evidence that the polysaccharide induces the
growth factor to form dimers (Bernard-Pierrot et al., 1999). The dependence of the cellular
114
RESULTS
activities of HARP on heparin has been demonstrated elsewhere (Kinnunen et al., 1996,
Vacherot et al., 1999a). Thus, current evidence favors an activation mechanism whereby
HS induce the change in conformation and the dimerization of HARP molecules to expose
binding sites for ALK.
A different view arises from our present data, which indicate that heparin binds
to ALK. These results suggest that HS binds to both ALK and HARP, thus bringing two
interacting proteins into close proximity. Currently, there is no direct evidence that the
intermolecular association of HARP molecules is required to facilitate binding to ALK.
The biosensor studies indicated a direct binding of ALK to the immobilized
peptide corresponding to the C-terminal region of HARP. However, the binding of ALK to
the immobilized entire protein was much lower. Although this may be explained by the
loss of an important lysine upon biotinylation of HARP, we cannot disregard the
possibility of a requirement of heparin to expose the binding sites within HARP7 to ALK.
Another outcome of these assays was the association of HARP molecules
without requirement of heparin. However, this result is not in accordance with that of
Bernard-Pierrot et al. that has been previously reported that dimerization process was
dependent on the presence of GAGs. Based on detailed studies with FGF and their
receptors, a similar ligand-induced dimerization mechanism has been previously proposed
for HARP, but there is no direct evidence for HARP dimers to be the active forms.
Moreover, there is no structural basis for the dimerization process of HARP. The binding
assays performed using different internal domains of HARP indicated for the first time, the
interaction of ALK-binding domain of HARP (C-terminal tail) with its heparin-binding
domain (C-TSR domain), hence providing a new aspect for dimerization of HARP.
7
In our experiments, the exact structure of HARP immobilized on the surface isn’t known, but the low-
density coverage of the streptavidin surface by biotinylated HARP (< 100 arc s) suggest that the molecule is
likely to be in the monomeric form on the surface. Furthermore, the strong link of the biotin to the
streptavidin prevents subsequent self-association of HARP molecules.
115
2. BASIC DOMAIN OF C-TERMINAL OF HARP
RESULTS
2. BASIC DOMAIN OF C-TERMINAL REGION OF HARP
2.1 INTRODUCTION
HARP-mediated mitogenic, angiogenic and tumor promoting effects are thought
to occur through the stimulation of MAPK and PI-3K pathways via the intrinsic tyrosine
kinase activities of its receptor ALK. Ligand-induced dimerization is a key event in
transmembrane signaling by receptors with tyrosine kinase activity. Receptor dimerization
leads to receptor autophosphorylation in the tyrosine activation loop and an increase in
kinase activity, resulting in the induction of diverse biological responses. The net tyrosine
phosphorylation in cell signaling is controlled by the balance between activities of protein
tyrosine kinases (PTKs) and protein tyrosine phosphatases (PTPs).
HARP is also a soluble ligand of the receptor protein tyrosine phosphatase
(RPTPβ). This finding at first seems counterintuitive, because PTPs are potential tumor
suppressor proteins as they reverse the effects of PTKs. However, the HARP/RPTPβ
signaling pathway presents an unusual mechanism of receptor signaling such that HARP
inactivates the protein phosphatase activity of RPTPβ. The ability of HARP to inactivate
endogenous RPTPβ is thought to be responsible for the observed concomitant increases in
tyrosine phosphorylation of different substrate proteins of RPTPβ, notably β-catenin, βadducin and fyn.
The family of RPTPs is important in cell-cell and cell-matrix adhesion. The
HARP/RPTPβ signaling pathway is related to cytoskeletal disruption and decreases
homophilic cell-cell adhesion. In recent years, several groups have described the functional
role of RPTPβ in glioblastoma and RPTPβ is a potential target for treatment of this cancer.
In these studies, RPTPβ has been associated with proliferation (Ulbricht et al., 2006, Foehr
E. et al., 2006), adhesion and migration of glioblastoma cells (Muller et al., 2003, Lorente
et al., 2005).
Thus, RPTPβ seem to be important for cell growth in situations of contact
independence. The nature of RPTPβ biology is consistent with the studies of HARP;
HARP itself can transform cells with loss of contact inhibition, cell adhesion and with
disruption of cytoskeletal architecture. Based on this observation, the ability of HARP to
alter the reciprocal control of tyrosine phosphorylation of substrates by tyrosine kinases
and phosphatases, seems to be extremely important and may account for many of the
117
RESULTS
properties of HARP in the human cancer cells which constitutively express the protein
(Meng et al., 2000).
Recently, Lu et al. have reported that the 18-kDa form of HARP (amino acids 1
to 136) stimulates haptotactic cell migration by binding RPTPβ. Assuming that the whole
C-terminal of HARP is responsible of either ALK and RPTPβ-mediated biological
functions of the protein and as we demonstrated that the sequence 111-124 was sufficient
to mediate HARP biological activities in a ALK-dependent manner, the remaining portion
of C-terminal (the amino acids from 124 to 136) should be the functional domain of HARP
which signals through its receptor RPTPβ. This hypothesis is tested in the present chapter.
In order to establish a structural basis for RPTPβ-binding site within C-terminal
region, we were interested in a HARP fragment corresponding to the basic cluster of Cterminal amino acids (122 to 131) and so the following peptide was synthesized: P122131: NH2-KKKKKEGKKQ-COOH. The peptide corresponds to the middle of the putative
111-124 and 124-136 fragments, including most of the 12 HARP C-terminal last amino
acid residues. Furthermore, the peptide includes 64% of the lysine residues (7 lysines of
11) present in the C-terminus. Therefore, we used P122-131 as a model to determine
whether these amino acids have a role in RPTPβ-mediated HARP activity. The prostatic
cancer cells, DU145, were used, since they possess only RPTPβ receptor but not ALK.
Figure 21: Hypothesis for differential signaling of peptides derived from C-terminal.
118
RESULTS
2.2 RESULTS
2.2.1 CHARACTERIZATION OF 10-AMINO ACID BASIC PEPTIDE
Article 2: Accepted July 31st, 2007
Title: A Basic Peptide Derived from the HARP C-Terminus Inhibits AnchorageIndependent Growth of DU145 Prostate Cancer Cells
Authors: Oya Bermek, Zoi Diamantopoulou, Apostolis Polykratis, Celia Dos Santos,
Yamina Hamma-Kourbali, Fabienne Burlina, Jean Delbé, Gerard Chassaing, David G.
Fernig, Pagnagiotis Katsoris and José Courty
Journal: Experimental Cell Research
119
E XP E RI ME N TA L CE L L RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / y e x c r
Research Article
A basic peptide derived from the HARP C-terminus inhibits
anchorage-independent growth of DU145 prostate
cancer cells☆
Oya Bermek a , Zoi Diamantopoulou b , Apostolis Polykratis b , Celia Dos Santos a ,
Yamina Hamma-Kourbali a , Fabienne Burlina c , Jean Delbé a , Gerard Chassaing c ,
David G. Fernig d , Pagnagiotis Katsoris b , José Courty a,⁎
a
Laboratoire de recherche sur la Croissance Cellulaire, la Réparation et la Régénération Tissulaires (CRRET), CNRS UMR 7149,
Université Paris 12, 61 Avenue du Général de Gaulle, 94010 Créteil Cedex, France
b
Division of Genetics, Cell and Developmental Biology, Department of Biology, University of Patras, GR 26504, Greece
c
Laboratoire de Synthèse, Structure et Fonction des Molécules Bioactives, CNRS UMR 7613, Université Pierre et Maris Curie,
4 Place Jussieu, 75252 Paris Cedex 05, France
d
School of Biological Sciences, University of Liverpool, Liverpool L69 7ZB, UK
ARTICLE INFORMATION
ABS T R AC T
Article Chronology:
Heparin affin regulatory peptide (HARP) is an 18 kDa heparin-binding protein that plays a key
Received 2 May 2007
role in tumor growth. We showed previously that the synthetic peptide P(111–136) composed
Revised version received
of the last 26 HARP amino acids inhibited HARP-induced mitogenesis. Here, to identify the
31 July 2007
exact molecular domain involved in HARP inhibition, we investigated the effect of the shorter
Accepted 31 July 2007
basic peptide P(122–131) on DU145 cells, which express HARP and its receptor protein
Available online 7 August 2007
tyrosine phosphatase β/ζ (RPTPβ/ζ). P(122–131) was not cytotoxic; it dose-dependently
inhibited anchorage-independent growth of DU145 cells. Binding studies using biotinylated P
Keywords:
(122–131) indicated that this peptide interfered with HARP binding to DU145 cells.
Heparin affin regulatory peptide
Investigation of the mechanisms involved suggested interference, under anchorage-
Prostate cancer
independent conditions, of P(122–131) with a HARP autocrine loop in an RPTPβ/ζ-
Anchorage-independent growth
dependent fashion. Thus, P(122–131) may hold potential for the treatment of disorders
Basic peptide
involving RPTPβ/ζ.
Receptor protein tyrosine
© 2007 Elsevier Inc. All rights reserved.
phosphatase
Introduction
Prostate cancer is the most commonly diagnosed malignancy
in men in many industrialized countries [1]. Initially, the
growth of prostate cancer cells is mainly dependent on
androgen, and androgen deprivation is widely used as a
treatment [2]. However, androgen-independent growth occurs
eventually [3], leading to metastasis formation in 85% to 100%
of patients with advanced prostate cancer. No effective treatments are available to control the metastatic tumors, which
☆
HARP C-terminus inhibits anchorage-independent growth of DU145.
⁎ Corresponding author. Fax: +33 145 171 816.
E-mail address: [email protected] (J. Courty).
0014-4827/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.yexcr.2007.07.032
4042
E XP E RI ME N TA L CE LL RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
are responsible for patient death. The molecular mechanisms
responsible for the transition to androgen-independence are
unclear, and novel therapeutic strategies are needed for
patients at this stage of prostate cancer. In adults, the
functional integrity of the normal prostate depends on mesenchymal–epithelial interactions, which contribute to homeostasis
and repair of the glandular prostate epithelium. Disturbances in
prostate epithelium homeostasis lead to the development of
diseases such as cancer. Although the mechanisms that control
mesenchymal-epithelial interactions are poorly understood,
numerous studies suggest a crucial role for growth factors.
Heparin affin regulatory peptide (HARP), also called pleiotrophin (PTN), is a 136-amino acid secreted polypeptide. HARP
is one of the heparin-binding growth differentiation factors,
which induce transformation of several cell lines [4,5]. In
addition, HARP is mitogenic for endothelial cells [6] and
angiogenic both in vitro and in vivo [5,7,8]. The biological
activities of HARP are mediated by three distinct receptors:
syndecan 3 (N-syndecan), receptor protein tyrosine phosphatase β/ζ (RPTPβ/ζ), and anaplastic lymphoma kinase (ALK)
receptor [9]. The mitogenic, angiogenic, and phenotypetransforming effects of HARP are mediated by the ALK
receptor [10]. N-syndecan and RPTPβ/ζ are cell-surface proteoglycans involved in the neurite outgrowth-promoting
activity of HARP [11,12]. RPTPβ/ζ has also been reported to
mediate HARP-induced migration of tumor cells. HARP is
expressed in several human tumors and tumor cell lines.
Studies using an antisense and a dominant negative strategy
have established that HARP is the rate-limiting factor for
phenotype transformation, angiogenesis, and metastasis
[7,13,14]. More specifically, the functional significance of
HARP in the progression of prostate cancer has been convincingly demonstrated, and HARP expression has been documented in various prostate-derived cell lines including DU145,
PC3, and LNCaP [15], for which HARP acts as an autocrine
growth factor [16,17].
We showed that the C-terminus of HARP (amino acids 111
through 136, P(111–136)) plays a critical role in the interaction
of HARP with its high-affinity tyrosine kinase receptor ALK [13].
Another recent study using human glioblastoma cells suggested that processing of the HARP C-terminus might produce
two isoforms: the full-length 18 kDa polypeptide and a
truncated 15 kDa form lacking the last 12 amino acids (124
through 136). In a study of various glioblastoma cell lines [18],
the truncated form promoted cell proliferation in an ALKdependent fashion, whereas the full-length form promoted
cell migration via the RPTPβ/ζ-dependent pathway. Here, we
characterize a 10-amino acid peptide P(122–131) corresponding
to the basic cluster of residues at the C-terminus of HARP. P
(122–131) inhibited RPTPβ/ζ-dependent growth of DU145 cells
in soft agar.
purchased from Senn Chemicals (Cachan, France). Peptide
synthesis grade solvents and other reagents were obtained
from Applied Biosystems (Courtaboeuf, France). The control
peptide (5K) was purchased from Sigma (Saint-Quentin
Fallavier, France). Doxorubicin hydrochloride was from Teva
Classics (Paris, France). Recombinant human HARP, N, and C
TSR molecules were produced and purified from bacteria as
previously described [19,20]. We purchased [methyl-3H] thymidine from ICN (Orsay, France), ImmunoPure® TMB substrate
kits from Pierce (Rockford, USA), and Cell Counting Kit-8 from
Interchim (Montluçon, France). Antibodies to human HARP
were purchased from R&D systems (Oxon, UK) and antibodies
to human RPTPβ/ζ from Tebu Bio (Le Perray, France). FITCconjugated monoclonal mouse antibody to biotin and horseradish peroxidase-conjugated antibodies were obtained from
Jackson ImmunoResearch (Suffolk, UK). Heparitinase I, II, and
III were obtained from IBEX (San Leandro, USA) and Chondroitinase ABC from Sigma (Saint-Quentin Fallavier, France).
Cell culture media were from Invitrogen (Leek, The Netherlands). Rabbit polyclonal antibodies against the recombinant
extracellular domain of human ALK were a gift from Prof. M.
Vigny (INSERM U706, Paris, France).
Cell culture
The human prostate-cancer epithelial cell line DU145 [American Type Culture Collection (ATCC) # HTB-81] was grown in
RPMI-1640 medium supplemented with 10% (v/v) fetal calf
serum (FCS) and 50 μg/ml gentamicin. Cultures were maintained at 37 °C with 7% CO2 and 90% humidity. Chinese
hamster ovary (CHO-K1 line), MDA-MB231, and NIH-3T3 cells
were cultured as previously described [21].
Peptide synthesis and characterization
The peptides were produced using stepwise solid-phase
synthesis with an ABI 433A synthesizer (Applied Biosystems,
Courtaboeuf, France) and standard protocols for Boc chemistry (amino acid activation with dicyclohexylcarbodiimide/1hydroxybenzotriazole or HBTU). After synthesis of the KKKKKEGKKQ peptide, the peptidyl-resin was elongated by four
glycin residues and biotin to produce Biot-G4-P(122–131).
Peptides were cleaved from the resin by treatment with
anhydrous hydrofluoric acid (1 h, 0 °C) in the presence of
anisole (1.5 ml/g peptidyl-resin) and dimethylsulfide (0.25 ml/g
peptidyl-resin). Peptides with purity N95% were obtained by
high-performance liquid chromatography on an RP-C8 column
using a linear gradient of acetonitrile in water with 0.1%
trifluoroacetic acid. Peptides were characterized using MALDITOF MS (Voyager Elite, PerSeptive Biosystems) with matrix ácyano-4-hydroxycinnamic acid. The m/z values of the protonated peptides (monoisotopic peak) were 1269 for H-P(122–131)
and 1683 for Biot-P(122–131).
Experimental procedures
Proliferation assay
Materials
Standard Boc amino acids, p-methylbenzhydrylamine-polystyrene resin (0.81 mmol NH2/g), and O-(benzotriazol-1-yl)1,1,3,3-tetramethyluronium hexafluorophophate (HBTU) were
DU145 cells were seeded into a 12-well culture plate at 1 104
cells/well in culture medium with or without P(122–131) or
antibodies to HARP, RPTPβ/ζ, or ALK. After 48 h incubation, the
cells were trypsinized and counted.
E XP E RI ME N TA L CE L L RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
Thymidine incorporation assay
Incorporation of [methyl-3H] thymidine by serum-starved
NIH-3T3 cells was evaluated as previously described [21].
Briefly, NIH-3T3 cells were seeded in 48-well culture plates at
2·104 cells/well and incubated for 24 h in DMEM with 10% (v/v)
FCS. The cells were then serum-starved for 24 h, and samples
were added. The cells were incubated for 18 h in 5% CO2 at
37 °C then for 6 h with 0.5 μCi [methyl-3H] thymidine.
Macromolecules were precipitated with 10% (w/v) trichloroacetic acid, washed with water, and lysed with 0.2 N NaOH.
Radioactivity precipitated with trichloracetic acid was counted
using a micro-beta scintillation counter (LKB, PerkinElmer Life
Sciences, Courtaboeuf, France).
4043
body diluted 1:1000 in PBS containing 0.5% (w/v) BSA (w/v), for
1 h at room temperature; then exposed to peroxidase-conjugated anti-goat antibody diluted 1:100,000, for 1 h at room
temperature. Peroxidase activity was then as described above.
Heparitinase and chondroitinase treatment of cells
DU145 cells were plated in triplicate into 96-well plates at 2 104
cells/well, incubated overnight and starved for 24 h. Cells were
Cytotoxicity assay
Cell viability was measured using the CCK-8 Kit (Interchim,
Montluçon, France) according to the manufacturer's instructions. Briefly, CHO-K1 cells were seeded in 96-well plates at
2·103 cells/well and incubated for 24 h in HAM's F12 medium
with 10% (v/v) FCS. Cells were then incubated with or without
1, 10, or 50 μM of P(122–131) for 72 h. Doxorubicin hydrochloride (0.5 μM) served as the positive control. The number of
viable cells was assessed by using the cell counting reagent,
and absorbance at 450 nm was measured.
Soft agar assay
DU145 and MDA-MB231 cells were seeded into 12-well plates
containing 0.6% (w/v) agar at 2·103 cells/well in 0.35% (w/v) agar
and growth medium. After 2 weeks, colonies larger than 50 μm
in diameter were counted using a phase-contrast microscope
equipped with a measuring grid. The cells were treated with
the peptide three times within 1 week and with the antibodies
twice during the 2 weeks.
Cell ELISA binding assay
DU145 cells were seeded in triplicate on 96-well plates at 2·104
cells/well, incubated overnight, and subsequently starved for
24 h. Before the binding experiment, the wells were incubated
with 300 μl/well RPMI containing 3% bovine serum albumin
(BSA, w/v), for 1 h at room temperature. Binding of biotinylated
P(122–131) was achieved in RPMI containing 1% (w/v) BSA for
2 h at room temperature. Unbound biotinylated peptide was
removed by washing the cells three times with phosphatebuffer saline (PBS) containing 1% (w/v) BSA (washing buffer),
and the cells were fixed with PBS supplemented with 4% (w/v)
paraformaldehyde (PFA) for 10 min at room temperature
(100 μl/well). The cells were washed three times with washing
buffer, and non-specific binding sites were blocked for 1 h at
room temperature with PBS containing 3% (w/v) BSA (300 μl/
well). The bound peptide was characterized using a peroxidase-labeled antibody to anti-biotin antibody diluted 1:2000 in
PBS containing 0.5% BSA (v/v), for 1 h at room temperature.
Peroxidase activity was detected using 3,3,5,5-tetramethylbenzidine dihydrochloride substrate according to the supplier's instructions. Absorbance was measured at 450 nm. For
HARP binding, cells were incubated with goat anti-HARP anti-
Fig. 1 – P(122–131) inhibits HARP-induced mitogenesis.
Stimulation of [3H]thymidine incorporation in serum-starved
NIH-3T3 cells treated with various concentrations of
P(122–131) (A) or the control peptide P5K (B) with or without
4 nM HARP. The cytotoxicity of P(122–131) was investigated
using CCK-8 kit (C). Doxorubicin hydrochloride (doxo) served
as the positive control. Results are the means of three
separate experiments, each carried out in triplicate, and the
standard errors are indicated.
4044
E XP E RI ME N TA L CE LL RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
then incubated with either heparitinase I, II, and III (7.5 mU/ml)
or chondroitinase ABC (0.4 U/ml) for 2 h or 24 h at 37 °C in RPMI1640 medium. Binding experiments were performed as described according to the cell ELISA binding assay. The efficacy
of enzymes treatment was confirmed using the bioassay
developed by Barbosa et al. [22].
Confocal microscopy
DU145 cells (5·104 cells/well) were plated in 4-well glass slides
(Lab-Tek Brand, Nalge Nunc International, Naperville, IL) in
complete medium and then serum-starved for 24 h in RPMI
medium. The cells were incubated with 100 μM biot-P(122–131)
peptide for 2 h at room temperature with or without 1 mM P
(122–131). After extensive washing with PBS, the cells were
fixed with 4% PFA (w/v). Non-specific sites were saturated in
PBS containing 1% BSA (w/v), and bound biotinylated peptide
was revealed using anti-biotin antibody linked to FITC and
diluted 1:100 in PBS containing 1% (w/v) BSA. Nuclei were
stained with DAPI according to standard procedures. Slides
were examined using a Zeiss LSM 510 META confocal laser
microscope (Zeiss, Iena, Germany) with a Plan Apochromat 63×
N.A.1.4 objective.
Immunoprecipitation and Western blot analysis
The medium of DU145 cells grown in 60-mm plastic Petri
dishes was removed. The cells were washed twice with icecold PBS and lysed with 1 ml of cell lysis buffer (50 mM HEPES,
pH 7, 150 mM NaCl, 10 mM EDTA, 1% TritonX-100, 1% Nonidet
P-40 (both v/v), 1 mM phenylmethylsulfonyl fluoride, 1 mM
sodium orthovanadate, 5 μg/ml aprotinin, and 5 μg/ml
leupeptin). Cells were scraped from the plates, transferred to
microfuge tubes, sonicated for 4 min while kept on ice, and
then centrifuged at 20,000×g for 10 min at 4 °C. Approximately
400 μg of the supernatant was precleared by incubation with
30 μl of protein A-Sepharose or streptavidin–agarose beads for
60 min at room temperature, followed by centrifugation at
10,000×g for 5 min. The beads were collected by centrifugation
and the supernatants were transferred to new microfuge
tubes. After this precleaning step, supernatants were incubated overnight at 4 °C with RPTPβ/ζ primary antibodies diluted
1:200 or with biotinylated P(122–131). A suspension of protein
A-Sepharose or streptavidin–agarose beads in a volume of 80 μl
was added. After 3 h incubation at 4 °C, beads and bound
proteins were collected by centrifugation (10,000×g, 4 °C) and
washed by centrifugation three times with ice-cold cell lysis
buffer. The pellet was resuspended in 60 μl of 2× SDS loading
buffer (100 mM Tris–HCl, pH 6.8, 4% w/v) sodium dodecyl
sulfate (SDS), 0.2% (w/v) bromophenol blue, 20% glycerol, 0.1 M
dithiothreitol), and kept at 4 °C until use. Before electrophoresis, samples were heated at 95 °C for 5 min then centrifuged;
50 μl of the supernatant was subjected to SDS-polyacrylamide
gel electrophoresis, after which the separated polypeptides
were electrotransferred to an Immobilon P membrane (Millipore, St. Quentin en Yvelines, France). After 3 h incubation in
48 mM Tris, pH 8.3, 39 mM glycin, 0.037% (w/v) SDS, and 20%
(v/v) methanol, the membrane was blocked for 1 h at 37 °C in
5% (w/v) non-fat milk, 0.1 M Tris-buffered saline (TBS), Tween
20. The membrane was then probed overnight at 4 °C under
continuous agitation using anti-RPTPβ/ζ antibodies diluted
1:500. The blot was incubated with the secondary antibody
coupled to horseradish peroxidase, and bands were detected
using the ChemiLucent Detection System Kit (Chemicon
International Inc., Temecula, CA), according to the manufacturer's instructions.
Fig. 2 – P(122–131) inhibits proliferation of DU145 and MDA-MB231 cells. (A) Inhibition of DU145-cell colony formation on
soft agar by anti-HARP polyclonal antibodies toHARP and (B) by P(122–131). Similar experiments were performed with
MBA-MB231 cells. Inhibition of colony formation was tested with or without polyclonal antibodies to HARP (C) or
P(122–131) (D) Results are the means of three separate experiments, each carried out in triplicate, and the standard errors
are indicated.
E XP E RI ME N TA L CE L L RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
4045
(Perbio, Chester, UK), and captured on streptavidin-derivatized
planar aminosilane cuvettes (NeoSensors), as described [24–
26]. Neither HARP [24] nor peptides bound non-specifically to
streptavidin-derivatized surfaces (data not shown).
Statistical analysis
Fig. 3 – Expression by DU145 cells of mRNA encoding
receptors RPTPβ/ζ and ALK. Expression of mRNA encoding
RPTPβ/ζ, ALK, and TFIID in DU145 cells cultured on agar (lane
2) or plastic at medium (lane 3) or high confluence (lane 4).
Control RT-PCRs without RNA (lane 1) or with RNA from the
positive U87MG cell line (lane 5) are also shown. PCR
fragments for RPTPβ/ζ, ALK, and TFIID were 344 bp, 236 bp,
and 194 bp, respectively.
The significance of differences between results from the
various groups was evaluated using unpaired t-tests. Each
experiment included at least triplicate measurements for each
test condition. All results are expressed as mean ± S.E.M. from
at least three independent experiments.
Results
We previously reported that the biological effects of HARP were
inhibited by the truncated mutant HARPΔ111–136 and
Reverse transcriptase-polymerase chain reaction (RT-PCR) for
ALK and RPTPβ/ζ
Total RNA was extracted from cells using the RNA Instapure
Kit (Eurogentec, Seraing, Belgium) according to the manufacturer's instructions. To obtain DU145 cells growing under
anchorage-independent conditions, cells were plated and
cultured for 3 weeks over a layer of 0.5% agar in complete
medium, as described by Dong and Cmarik [23]. The growth
medium and unattached cells were then collected, and the
recovered cells were pelleted by centrifugation and subjected
to RNA extraction. cDNAs were synthesized from 1 μg of
total RNA using random hexamer primers and Superscript
II™ reverse transcriptase (Invitrogen, Cergy Pontoise,
France). Then, 1/5 or 1/2 (v/v) of the reaction products were
subjected to PCR amplification using the GenAmp9600
system (PE Applied Biosystems, Les Ulis, France) for detection of RPTPβ/ζ and ALK, respectively. Primers were as
follows: RPTPβ/ζ detection, 5′-CTAAAGCGTTTCCTCGCTTG-3′
(forward) and 5′-TCTGAAACTCCTCCGCTGAC (reverse); ALK
detection, 5′-CAACGAGGCTGCAAGAGAGAT-3′ (forward) and
5′-GTCCCATTCCAACAAGTGAAGGA-3′ (reverse); and TFIID
detection, 5′-AGTGAAGAACAGTCCAGACTG-3′ (forward) and
5′-CCAGGAAATAACTCGGCTCAT-3′ (reverse). After 5 min at
94 °C, the following number of cycles was used: 30 cycles for
TFIID, 35 cycles for RPTPβ/ζ and ALK. Each cycle included
denaturation at 94 °C for 1 min, annealing at 60 °C (for TFIID) or
57 °C (for RPTPβ/ζ or ALK) for 1 min, and primer extension at
72 °C for 1 min. RT-PCR products were subjected to electrophoresis on 2% (w/v) agarose gels containing 0.5 mg/ml
ethidium bromide. Gels were photographed using a ChimiGenius system (Syngene, Cambridge, UK).
Optical biosensor binding assays
Binding assays were performed in an IAsys optical biosensor
(NeoSensors, Sedgefield, UK) at 20 °C in which the response is
measured in arc s (1 arc s = 1/3600°; 600 arc s = 1 ng protein/mm2
sensor surface). Pig mucosal heparin (15–20 kDa, Sigma, Poole,
UK) was biotinylated on free amino groups with NHS biotin
Fig. 4 – Involvement of RPTPβ/ζ in the growth of DU145 cells.
Representative phase-contrast microphotography of colony
formation of DU145 cells treated with anti-RPTPβ/ζ (A),
anti-ALK (B), control IgG (C); in (D), no treatment was used.
Magnification, ×50; scale bar, 250 μm. Colonies larger than
50 μm in diameter were counted and presented in (E). Results
are the means of three separate experiments, each carried
out in triplicate, and the standard errors are indicated.
4046
E XP E RI ME N TA L CE LL RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
corresponding synthetic peptide P(111–136) (KLTKPKPQAESKKKKKEGKKQEKMLD) [13]. Here, we sought to identify the
minimum sequence responsible for the inhibition of HARP
activity. Since an obvious feature of P(111–136) is the stretch of
basic residues, which may be involved in ionically driven
molecular interactions, we investigated whether the basic
sequence P(122–131) (KKKKKEGKKQ) mediated the proliferation-inhibiting effect of P(111–136).
Effect of P(122–131) on inhibition of HARP-induced
mitogenesis
The mitogenic activity of HARP in the presence of P(122–131)
was investigated using serum-starved NIH-3T3 cells with or
without HARP stimulation, as described in Experimental
procedures. As shown in Fig. 1A, stimulation with 4 nM HARP
induced a 5.3-fold increase in [3H]-thymidine incorporation
and 1, and 10 μM of peptide P(122–131) inhibited this effect by
30% (Fig. 1A). Since P(122–131) contains a cluster of five lysine
residues, we performed a control experiment with a five-lysine
peptide (P5K) to evaluate the specificity of the inhibitory effect of
P(122–131). P5K at concentrations of up to 1 μM failed to inhibit
HARP-induced mitogenesis (Fig. 1B) whereas 10 μM P5K
increased slightly HARP-induced mitogenesis. On NIH-3T3
cells, no significant effect was observed when P(122–131) or P5K
was tested alone (Figs. 1A and B). To confirm the inhibitory effect
of P(122–131) on thymidine incorporation, we performed a
cytotoxic assay using CHO-K1 cells. In contrast to doxorubicin
used as the positive control, increasing concentrations of P(122–
131) ranging from 1 to 50 μM did not inhibit CHO-K1 cell growth
(Fig. 1C). Results were similar with NIT 3T3 cells (data not shown).
Taken together, these results demonstrated that P(122–131) was
not cytotoxic and inhibited HARP-induced cell proliferation.
P(122–136) inhibits DU145 cell proliferation
The effect of P(122–131) on HARP-induced tumorigenesis was
evaluated by measuring DU145 growth on agar. This prostate
cancer cell line produces HARP [15], which in turn stimulates
the growth of the cells [17]. We studied DU145 growth under
anchoring-independent conditions in the presence of an antiHARP polyclonal antibody. The number of cell colonies
decreased by 30% in the presence of 5 μg/ml of HARP antibodies
(Fig. 2A). Thus, endogenous HARP may act as an autocrine
growth factor for DU145 cells. We then investigated inhibition
of DUI45 growth by P(122–131) by seeding cells on agar and
adding various amounts of P(122–131) to the culture medium.
The number of cell colonies decreased by 50% in the presence
of 10 μM P(122–131) (Fig. 2B). To confirm these results, we
looked for inhibitory effects of P(122–131) on the growth of
MDA-MB231 tumor cells, previously shown to be dependent on
a HARP autocrine loop [21]. Anti-HARP antibodies caused a
dose-dependent decrease in the number of colonies to a
maximum of −55% with 0.8 μM of anti-HARP (Fig. 2C). When
increasing concentrations of P(122–131) were added to the
culture medium, significant inhibition was also observed,
confirming that P(122–131) inhibited the tumorigenic effect of
HARP (Fig. 2D). We used RT-PCR to investigate the expression
of RPTPβ/ζ and ALK receptors, two of the receptors involved in
mediating the biological activities of HARP. The human
glioblastoma cell line U87MG served as the positive control
for the expression of the two receptors, as previously described
[18]. mRNA corresponding to RPTPβ/ζ was the only receptor
expressed in DU145 cells (Fig. 3). Therefore, we investigated
whether RPTPβ/ζ was involved in the HARP-dependent autocrine stimulation of DU145 seeded on soft agar by using a
polyclonal antibody against the extracellular domain of RPTPβ/
Fig. 5 – P(122–131) binding to DU145 cells. Binding of HARP and biot-P(122–131) was investigated as described in Experimental
procedures. (A) Inhibition of HARP binding to DU145 cells by increasing concentrations of P(122–131). (B) Kinetics of
biot-P(122–131) binding to DU145 cells. Saturation binding curve (C). Binding of biot-P(122–131) to DU145 cells with (black
triangle) or without (black square) excess P(122–131) (1 mM). The specific resulting binding is shown (inverted black triangle).
E XP E RI ME N TA L CE L L RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
4047
dose-dependent manner up to 30% at 5 μM (Fig. 5A). We then
evaluated P(122–131) binding to DU145 using biotinylated-P
(122–131) (biot-P(122–131), which displays similar biological
effects to the non-biotinylated peptide (not shown). When
DU145 cells were incubated for 30 to 180 min with 20 μM biot-P
(122–131) and binding was evaluated as described in Experimental procedures, binding reached a maximum after 120 min
(Fig. 5B). The specific binding of biot-P(122–131) to DU145 was
specific and saturable (Fig. 5C), with an estimated Kd value of
25 μM. This value is in the same range of concentration in
which P(122–131) acts on cell proliferation. To strengthen these
data, competition experiments were performed using biot-P
(122–131) as ligand and either HARP or P(122–131) as competitor. As shown in Fig. 5D, HARP and P(122–131) inhibited the
Fig. 6 – Confocal immunohistological analysis of
biot-P(122–131) binding to DU145. DU145 cells were
incubated with biot-P(122–131) with (B) or without (A) a
10-fold P(122–131) excess. Bound peptide was revealed as
described in Experimental procedures. Scale bar, 80 μm.
(C) Whole DU145 cell lysates were incubated (1) with
anti-RPTPβ/ζ immobilized on protein A sepharose beads,
(2) with biot-P(122–131) immobilized on streptavidin agarose
beads, or (3) with streptavidin immobilized on agarose beads.
The precipitates were analyzed by 5% SDS-PAGE and
electroblotted; and the membrane was probed against
RPTPβ/ζ using specific antibodies.
ζ idiotypic immunoglobulins and antibody to ALK served as
controls. Anti-RPTPβ/ζ decreased the anchorage-independent
growth of DUI45, compared to untreated cells (Fig. 4). No
growth inhibition occurred with control immunoglobulin oranti ALK.
P(122–131) binds to a DU145 cell-surface component
Since P(122–131) inhibited the biological activity of HARP, we
investigated whether it prevented HARP from binding to
DU145. P(122–131) inhibited HARP binding to DU145 in a
Fig. 7 – Chondroitinase treatment of DU145 cells and binding
of P(122–131) to heparin immobilized on a
streptavidin-derived IAsys sensor surface. (A) Binding of
biot-P(122–131) (20 μM) was performed on DU145 cells
treated with heparitinase I, II, or III (hatched bar) or with
chondroitinase (grey bar). The binding of biot-P(122–131)
performed without enzymatic treatment and control
performed without biotinylated peptide were shown by
white and black bar, respectively. (B) The binding of HARP
(30 nM), C-TSR (3 μM), N-TSR (3 μM), and P(122–131) (80 μM)
to immobilized heparin was measured in an IAsys optical
biosensor, as described under Experimental procedures.
Results are the mean of three experiments carried out in
triplicate.
4048
E XP E RI ME N TA L CE LL RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
binding of biot-P(122–131) to DU145 in a dose-dependent
manner with EC50 of 0.87 ± 0.08 μM and 20 ± 0.4 μM, respectively.
These data indicated that the binding affinity of HARP is 22fold more potent than P(122–131). As previously suggested by
Maeda et al. [27], this result suggests that other domain(s) of
HARP are involved in the binding of HARP to strengthen the
interaction.
Confocal microscopy showed that biot-P(122–131) bound to
a cell-surface component of cultured DU145 cells (Fig. 6A).
With a 10-fold excess of P(122–131), fluorescence associated
with cell surfaces was effectively competed, confirming that
binding was specific (Fig. 6B).
Characterization of the P(122–131) binding site
Since RPTPβ/ζ was expressed by DU145, these results
suggested an interaction of P(122–131) with RPTPβ/ζ?. These
observations prompted us to conduct immunoprecipitation
experiments using whole DU145 cell lysates with either biotP(122–131) or anti-RPTPβ/ζ. RPTPβ/ζ was detected in biotP(122–131) immunoprecipitates (lane 2), indicating that this
HARP receptor also interacted with P(122–131) (Fig. 6C). No
band was detected when biot-P(122–131) was omitted from
the assay (lane 3). Since the RPTPβ/ζ-derived chondroitin
sulfate was involved in the binding to HARP, we investigated
the effect of chondroitinase ABC digestion to the P(122–131)
binding to the cell surface. As shown in Fig. 7A, treatment of
cells with chondroitinase or heparitinase I, II or III as control
had no effect on the binding of P(122–131) to DU145. These
results suggest that P(122–131) did not bind to the RPTPβ/ζderived glycoaminoglycans, in spite of its basicity. To directly
confirm the hypothesis that P(122–131) does not bind to
glycoaminoglycans, the interactions analyzed in a optical
biosensor. P(122–131) failed to bind to a heparin surface even
at concentration as high as 80 μM. Similar results were
obtained as control using the N-TSR domain of HARP. In
contrast, HARP and its heparin domain C-TSR bound to the
heparin surface (Fig. 7B).
Discussion
A role for HARP in prostate cancer progression was first suggested by Vacherot et al. [17], who showed that HARP was
associated with epithelial cells in human prostate cancer but
not in normal prostate or benign prostate hyperplasia. Thus,
HARP may contribute to unregulated prostate cancer growth.
Using an antisense strategy directed against HARP mRNA,
Hatziapostolou et al. demonstrated that HARP was essential
for the migration and anchorage-dependant and -independent
growth of the human prostate tumor cell line LnCap [16]. The
same group established HARP as a key mediator of FGF-2induced stimulation of LnCap growth and migration in vitro
[28]. The importance of HARP in prostate tumor growth
prompted us to evaluate the effect on DU145 cell growth of
the synthetic basic peptide P(122–131) derived from the HARP
C-terminus. HARP was found to act as an autocrine growth
factor on DU145 cell growth [17].
Our data demonstrate clearly that P(122–131) specifically
and dose-dependently inhibits anchorage-independent
growth of DU145 cells. This effect is likely to involve binding
of P(122–131) to cell-surface RPTPβ/ζ, which is the only HARP
receptor expressed in DU145 cells?. Early studies showed that
RPTPβ/ζ was linked to neural cell migration. Subsequently,
RPTPβ/ζ was found to be overexpressed in tumor cells such as
astrocytomas and to correlate with malignancy [29]. RPTPβ/ζ
overexpression was documented in melanoma [30] and in
carcinomas of the lung, colon, breast, and prostate, compared
with normal tissue [9], in which expression was very low.
These observations suggested that RPTPβ/ζ might be an
excellent tumor marker and therapeutic target.
The growth factor HARP exerts potent biological effects. It
increases cell proliferation, migration, and differentiation;
tumor growth; and angiogenesis. These effects are mediated
by interactions with two receptors, ALK and RPTPβ/ζ? HARP,
expressed in several tumors and can be a rate-limiting factor
for tumor growth and metastasis in vivo. Thus, HARP holds
promise as a treatment target in patients with cancer. We
recently reported that the HARP C-terminus (amino acids 111
through 136) and a corresponding synthetic peptide P(111–136)
inhibited the biological effects of HARP [13]. In this previous
study, P(111–136) was found to bind to the extracellular domain
of ALK and to inhibit HARP-induced cell proliferation. In
addition, P(111–136) was active in vivo and inhibited tumor
growth when injected daily around the tumor (manuscript in
preparation). To better characterize this peptide and to identify
the likely minimum amino acid sequence involved in its
biological activity, we evaluated the effects of the peptide P
(122–131) comprising the most basic tract of P(111–136). We
studied the effects of P(122–131) on HARP-induced cell
proliferation using the prostate tumor-derived cell line
DU145, known to undergo HARP stimulation via an autocrine
loop. We found that P(122–131) dose-dependently inhibited
HARP-induced cell proliferation and anchorage-independent
DU145 growth. Since P(111–136) was previously reported to bind
to the ALK receptor, and P(122–131) was derived from P(111–136),
we hypothesized that P(122–131) might inhibit DU145 growth by
interacting with ALK and further blocking the HARP-activating
pathways. However, DU145 cells expressed the RPTPβ/ζ receptor
but not the ALK receptor, suggesting mediation of the HARP
autocrine loop in DU145 cells by the RPTPβ/ζ receptor.
Previous studies showed that the extracellular RPTPβ/ζ
domain 6B4 proteoglycan/phosphacan bound to HARP at both
high- and low-affinity binding sites [12]. Chondroitinase ABC
treatment decreased the affinity of 6B4 proteoglycan/phosphacan for HARP without significantly affecting the number of
binding sites. These results indicated that both the chondroitin
sulfate chain and the core 6B4 proteoglycan protein were
involved in binding to HARP. It is noteworthy that we have
shown that chondroitin sulfate was not involved in the binding
of P(122–131) suggesting that it does not bind to the glycoaminoglycan chains of RPTPβ/ζ-derived, but instead to the RPTPβ/ζ
core protein. This possibility has received support from
another study demonstrating that the HARP domain corresponding to the 122–131 region is not involved in HARP
interactions with glycosaminoglycans [31]. A recent study
using various human glioblastoma cell lines established that
extracellular processing of the HARP C-terminus produced two
isoforms: a full-length molecule whose binding to RPTPβ/ζinduced cell migration and a truncated form lacking the last 12
E XP E RI ME N TA L CE L L RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
amino acids (124–136) whose binding to the ALK receptor led to
a proliferative response [18]. Our earlier data on P(111–136) and
those reported here suggest that P(122–131), which includes
most of the last 12 HARP amino acids present in the full length,
but not the truncated isoform, may bind to RPTPβ/ζ and that
the binding site of the HARP C-terminus to ALK may include
the amino acids 111–121. Further experiments are needed to
investigate this possibility. In tumors such as gliomas, HARP
binding to RPRPβ/ζ inactivated the intracellular catalytic
domain of the receptor, leading to tyrosine phosphorylation
of the downstream effectors Fyn, β-catenin, and β-adducin; as
well as to cytoskeleton disruption, increased cell plasticity, and
loss of cell-cell adhesion [32–34]. However, RPTPβ/ζ may also be
associated with tumor cell proliferation in vitro and in vivo,
since interfering molecules such as siRNA or monoclonal antiRPRPβ/ζ antibodies suppressed tumor cell growth [35,36]. In
addition, we have shown that RPRPβ/ζ has a direct and positive
signaling role acting as phosphatase and activating Src kinase
[37]. Similar results have been obtained using P(122–131)
(manuscript in preparation). Therefore, a peptide that antagonizes the interaction of HARP with RPRPβ/ζ may hold promise
as a therapeutic tool for cancer.
Acknowledgments
This work was supported by grants from the CNRS, ANR-06-567
RIB-016-02, INCA, and Association pour la Recherche sur le
Cancer (#3242), the Cancer and Polio Research Fund, the North
West Cancer Research Fund, the Human Frontiers Science
Programme and a short-term Marie Curie early stage training
fellowship (MolFun) to OB. We thank Nicolas Setterblad at the
Imagery Department of the Institut Universitaire d'Hématologie IFR105 for the confocal microscopy studies. The imaging
department is supported by grants from the Conseil Regional
d'Ile-de-France and the Ministère de la Recherche.
REFERENCES
[1] H. Gronberg, Prostate cancer epidemiology, Lancet 361 (2003)
859–864.
[2] D.G. Bostwick, J. Qian, Effect of androgen deprivation therapy
on prostatic intraepithelial neoplasia, Urology 58 (2001)
91–93.
[3] B.J. Feldman, D. Feldman, The development of
androgen-independent prostate cancer, Nat. Rev., Cancer 1
(2001) 34–45.
[4] A.K. Chauhan, Y.S. Li, T.F. Deuel, Pleiotrophin transforms NIH
3T3 cells and induces tumors in nude mice, Proc. Natl. Acad.
Sci. U. S. A. 90 (1993) 679–682.
[5] R. Choudhuri, H.T. Zhang, S. Donnini, M. Ziche, R. Bicknell, An
angiogenic role for the neurokines midkine and pleiotrophin
in tumorigenesis, Cancer Res. 57 (1997) 1814–1819.
[6] J. Courty, M.C. Dauchel, D. Caruelle, M. Perderiset, D.
Barritault, Mitogenic properties of a new endothelial cell
growth factor related to pleiotrophin, Biochem. Biophys. Res.
Commun. 180 (1991) 145–151.
[7] F. Czubayko, A.M. Schulte, G.J. Berchem, A. Wellstein,
Melanoma angiogenesis and metastasis modulated by ribozyme
targeting of the secreted growth factor pleiotrophin, Proc. Natl.
Acad. Sci. U. S. A. 93 (1996) 14753–14758.
4049
[8] E. Papadimitriou, A. Polykratis, J. Courty, P. Koolwijk, M.
Heroult, P. Katsoris, HARP induces angiogenesis in vivo and in
vitro: implication of N or C terminal peptides, Biochem.
Biophys. Res. Commun. 282 (2001) 306–313.
[9] M. Grzelinski, N. Bader, F. Czubayko, A. Aigner,
Ribozyme-targeting reveals the rate-limiting role of pleiotrophin
in glioblastoma, Int. J. Cancer 117 (2005) 942–951.
[10] G.E. Stoica, A. Kuo, A. Aigner, I. Sunitha, B. Souttou,
C. Malerczyk, D.J. Caughey, D. Wen, A. Karavanov, A.T. Riegel,
A. Wellstein, Identification of anaplastic lymphoma kinase as
a receptor for the growth factor pleiotrophin, J. Biol. Chem.
276 (2001) 16772–16779.
[11] E. Raulo, M.A. Chernousov, D.J. Carey, R. Nolo, H. Rauvala,
Isolation of a neuronal cell surface receptor of heparin
binding growth-associated molecule (HB-GAM). Identification
as N-syndecan (syndecan-3), J. Biol. Chem. 269 (1994)
12999–13004.
[12] N. Maeda, T. Nishiwaki, T. Shintani, H. Hamanaka, M. Noda,
6B4 proteoglycan/phosphacan, an extracellular variant of
receptor-like protein-tyrosine phosphatase zeta/RPTPbeta,
binds pleiotrophin/heparin-binding growth-associated
molecule (HB-GAM), J. Biol. Chem. 271 (1996) 21446–21452.
[13] I. Bernard-Pierrot, J. Delbe, V. Rouet, M. Vigny, M.E. Kerros,
D. Caruelle, D. Raulais, D. Barritault, J. Courty, P.E. Milhiet,
Dominant negative effectors of heparin affin regulatory
peptide (HARP) angiogenic and transforming activities, J. Biol.
Chem. 277 (2002) 32071–32077.
[14] F. Czubayko, A.T. Riegel, A. Wellstein, Ribozyme-targeting
elucidates a direct role of pleiotrophin in tumor growth,
J. Biol. Chem. 269 (1994) 21358–21363.
[15] P. Soulie, M. Heroult, I. Bernard, M.E. Kerros, P.E. Milhiet,
J. Delbe, D. Barritault, D. Caruelle, J. Courty, Immunoassay for
measuring the heparin-binding growth factors HARP and MK
in biological fluids, J. Immunoass. Immunochem. 23 (2002)
33–48.
[16] M. Hatziapostolou, J. Delbe, P. Katsoris, C. Polytarchou,
J. Courty, E. Papadimitriou, Heparin affin regulatory peptide is
a key player in prostate cancer cell growth and angiogenicity,
Prostate 65 (2005) 151–158.
[17] F. Vacherot, D. Caruelle, D. Chopin, S. Gil-Diez, D. Barritault,
J.P. Caruelle, J. Courty, Involvement of heparin affin regulatory
peptide in human prostate cancer, Prostate 38 (1999) 126–136.
[18] K.V. Lu, K.A. Jong, G.Y. Kim, J. Singh, E.Q. Dia, K. Yoshimoto,
M.Y. Wang, T.F. Cloughesy, S.F. Nelson, P.S. Mischel,
Differential induction of glioblastoma migration and growth
by two forms of pleiotrophin, J. Biol. Chem. 280 (2005)
26953–26964.
[19] Y. Hamma-Kourbali, I. Bernard-Pierrot, M. Heroult, S. Dalle, D.
Caruelle, P.E. Milhiet, D.G. Fernig, J. Delbe, J. Courty, Inhibition
of the mitogenic, angiogenic and tumorigenic activities of
pleiotrophin by a synthetic peptide corresponding to its
C-thrombospondin repeat-I domain, J. Cell Physiol. (in press).
[20] M. Heroult, I. Bernard-Pierrot, J. Delbe, Y. Hamma-Kourbali,
P. Katsoris, D. Barritault, E. Papadimitriou, J. Plouet, J. Courty,
Heparin affin regulatory peptide binds to vascular endothelial
growth factor (VEGF) and inhibits VEGF-induced angiogenesis,
Oncogene 23 (2004) 1745–1753.
[21] I. Bernard-Pierrot, J. Delbe, D. Caruelle, D. Barritault, J. Courty,
P.E. Milhiet, The lysine-rich C-terminal tail of heparin affin
regulatory peptide is required for mitogenic and tumor
formation activities, J. Biol. Chem. 276 (2001) 12228–12234.
[22] I. Barbosa, S. Garcia, V. Barbier-Chassefiere, J.P. Caruelle, I.
Martelly, D. Papy-Garcia, Improved and simple micro assay
for sulfated glycosaminoglycans quantification in biological
extracts and its use in skin and muscle tissue studies,
Glycobiology 13 (2003) 647–653.
[23] Z. Dong, J.L. Cmarik, Harvesting cells under
anchorage-independent cell transformation conditions for
biochemical analyses, Sci. STKE (2002) PL7.
4050
E XP E RI ME N TA L CE LL RE S E A RCH 3 1 3 ( 2 00 7 ) 4 0 4 1 –40 5 0
[24] F. Vacherot, J. Delbe, M. Heroult, D. Barritault, D.G. Fernig,
J. Courty, Glycosaminoglycans differentially bind HARP and
modulate its biological activity, J. Biol. Chem 274 (1999)
7741–7747.
[25] H. Rahmoune, P.S. Rudland, J.T. Gallagher, D.G. Fernig,
Hepatocyte growth factor/scatter factor has distinct classes of
binding site in heparan sulfate from mammary cells,
Biochemistry 37 (1998) 6003–6008.
[26] H. Rahmoune, H.L. Chen, J.T. Gallagher, P.S. Rudland, D.G.
Fernig, Interaction of heparan sulfate from mammary cells
with acidic fibroblast growth factor (FGF) and basic FGF.
Regulation of the activity of basic FGF by high and low affinity
binding sites in heparan sulfate, J. Biol. Chem. 273 (1998)
7303–7310.
[27] N. Maeda, M. Noda, Involvement of receptor-like protein
tyrosine phosphatase zeta/RPTP beta and its ligand
pleiotrophin/heparin-binding growth-associated molecule
(HB-GAM) in neuronal migration, J. Cell. Biol. 142 (1998) 203–216.
[28] M. Hatziapostolou, C. Polytarchou, P. Katsoris, J. Courty, E.
Papadimitriou, Heparin affin regulatory peptide/pleiotrophin
mediates fibroblast growth factor 2 stimulatory effects on
human prostate cancer cells, J. Biol. Chem. 281 (2006)
32217–32226.
[29] S. Muller, P. Kunkel, K. Lamszus, U. Ulbricht, G.A. Lorente,
A.M. Nelson, D. von Schack, D.J. Chin, S.C. Lohr, M. Westphal,
T. Melcher, A role for receptor tyrosine phosphatase zeta in
glioma cell migration, Oncogene 22 (2003) 6661–6668.
[30] T. Goldmann, F. Otto, E. Vollmer, A receptor-type protein
tyrosine phosphatase PTP zeta is expressed in human
cutaneous melanomas, Folia Histochem. Cytobiol. 38 (2000)
19–20.
[31] I. Kilpelainen, M. Kaksonen, H. Avikainen, M. Fath, R.J.
Linhardt, E. Raulo, H. Rauvala, Heparin-binding
growth-associated molecule contains two heparin-binding
[32]
[33]
[34]
[35]
[36]
[37]
beta -sheet domains that are homologous to the
thrombospondin type I repeat, J. Biol. Chem. 275 (2000)
13564–13570.
K. Meng, A. Rodriguez-Pena, T. Dimitrov, W. Chen, M. Yamin,
M. Noda, T.F. Deuel, Pleiotrophin signals increased tyrosine
phosphorylation of beta beta-catenin through inactivation of
the intrinsic catalytic activity of the receptor-type protein
tyrosine phosphatase beta/zeta, Proc. Natl. Acad. Sci. U. S. A.
97 (2000) 2603–2608.
H. Pariser, P. Perez-Pinera, L. Ezquerra, G. Herradon, T.F.
Deuel, Pleiotrophin stimulates tyrosine phosphorylation of
beta-adducin through inactivation of the transmembrane
receptor protein tyrosine phosphatase beta/zeta, Biochem.
Biophys. Res. Commun. 335 (2005) 232–239.
H. Pariser, L. Ezquerra, G. Herradon, P. Perez-Pinera, T.F.
Deuel, Fyn is a downstream target of the
pleiotrophin/receptor protein tyrosine phosphatase
beta/zeta-signaling pathway: regulation of tyrosine
phosphorylation of Fyn by pleiotrophin, Biochem. Biophys.
Res. Commun. 332 (2005) 664–669.
E.D. Foehr, G. Lorente, J. Kuo, R. Ram, K. Nikolich,
R. Urfer, Targeting of the receptor protein tyrosine
phosphatase beta with a monoclonal antibody delays tumor
growth in a glioblastoma model, Cancer Res. 66 (2006)
2271–2278.
U. Ulbricht, C. Eckerich, R. Fillbrandt, M. Westphal, K.
Lamszus, RNA interference targeting protein tyrosine
phosphatase zeta/receptor-type protein tyrosine
phosphatase beta suppresses glioblastoma growth in vitro
and in vivo, J. Neurochem. 98 (2006) 1497–1506.
A. Polykratis, P. Katsoris, J. Courty, E. Papadimitriou,
Characterization of heparin affin regulatory peptide signaling
in human endothelial cells, J. Biol. Chem. 280 (2005)
22454–22461.
RESULTS
2.2.2 ANTI-TUMORAL ACTIONS OF P122-131 THROUGH RPTPβ
In the first part of the chapter, we characterized for the first time a 10-amino acid
peptide derived from to the carboxy end of HARP and corresponding to the basic cluster of
the C-terminus. We reported that the peptide binds to RPTPβ and exerts an antagonist
effect for transforming signaling of HARP.
To pursue our investigation with P122-131, we have particularly interested in
further effects of the peptide in other biological activities mediated by RPTPβ. The
implication of RPTPβ in the biological actions of the peptide as well as the nature of the
interaction of P122-131 with RPTPβ was explored.
In order to elucidate the mechanism of action of the peptide, finally, we analyzed
the intracellular signaling cascades activated by P122-131 in DU145 cells.
Article 3: submitted to Journal of Carcinogenesis
Title: A 10 Amino Acid Peptide Corresponding to the C-Terminal Part of HARP, Exhibits
Anti-Cancer Related Actions Through RPTPβ
Authors: Zoi Diamantopoulou, Oya Bermek, Apostolis Polykratis, Yamina HammaKourbali, Jean Delbé, José Courty and Pagnagiotis Katsoris
130
RESULTS
A NOVEL MOLECULE RELATED TO THE HARP C-TERMINUS INDUCES ANTITUMOR EFFECTS THROUGH RPTPβ/ζ
Zoi Diamantopoulou1, Oya Bermek2, Apostolis Polykratis1, Yamina HammaKourbali2, Jean Delbé2, José Courty2, and Panagiotis Katsoris1
1
Division of Genetics, Cell and Developmental Biology, Department of Biology,
University of Patras, Greece
2
Laboratoire de recherche sur la Croissance Cellulaire, la Réparation et la Régénération
Tissulaires (CRRET), CNRS UMR 7149, Université Paris XII, Avenue du Général de
Gaulle, 94010 Créteil Cedex, France
Running head: P122-131 inhibits tumor phenotypes via RPTPβ/ζ
Address correspondence to: Panagiotis Katsoris, University of Patras, Department of
Biology, GR 26500, Patras, Greece. Fax: +302610994797; E-mail: [email protected]
Heparin Affin Regulatory Peptide
(HARP) is a 15 kDa growth factor
expressed in various tissues and cell
lines. HARP participates in multiple
biological actions including the induction
of cellular proliferation, migration and
angiogenesis, and it is thought to be
involved in carcinogenesis. Recently, we
identified and characterized several
HARP proteolytic fragments with
biological activities similar or opposite to
that of HARP. Here, we investigated the
biological actions of P122-131, a
synthetic peptide corresponding to the
carboxy terminal region of HARP. Our
results show that P122-131 inhibits in
vitro adhesion, anchorage-dependent and
-independent
proliferation,
and
migration in DU145 cells, a human
cancer prostate cell line. In addition,
P122-131 inhibits angiogenesis in vivo, as
determined by the chicken embryo CAM
assay. Investigation of the transduction
mechanisms revealed that P122-131
binds to RPTPβ/ζ and induces the
phosphorylation of SRC-kinase, FAK,
PTEN, and ERK1/2. RPTPβ/ζ mediated
the anti-tumor effects of P122-131 as
well as activation of these other signaling
molecules as demonstrated by selective
knockdown of RPTPβ/ζ expression with
siRNA. Cumulatively, these results
suggest that P122-131 may be a useful
tool for tumor therapy.
Heparin Affin Regulatory Peptide (HARP,
also known as Pleiotrophin) is a 136-amino
acid, secreted growth factor that, along
with Midkine, constitutes a two-member
sub family of heparin binding growth
factors (HBGFs). Although HARP has
been shown to promote neurite outgrowth
in the developing brain (1), elevated
concentrations of this growth factor are
found in many types tumors as well as in
the plasma of patients with different types
of cancer (2-4).
HARP induces a
transformed phenotype in several cell lines
(5, 6) and exhibits mitogenic, anti
apoptotic, chemotactic, and angiogenic
actions in vitro as well as in vivo (7-10).
The biological activities of HARP are
mediated by three distinct receptors:
Syndecan-3 (N-Syndecan) (11), Receptor
Protein Tyrosine Phosphatase (RPTPβ/ζ)
(12), and Anaplastic Lymphoma Kinase
(ALK) (13). N-Syndecan and RPTPβ/ζ
have been implicated in neurite outgrowth
(10, 11), while RPTPβ/ζ and ALK have
been shown to mediate cellular migration
induced by HARP as well as the mitogenic,
angiogenic, and transforming activities of
this protein (14-18).
Growth factors can be hydrolyzed by
proteases, leading to the production of
biological active peptides. Previous studies
indicate that HARP is cleaved by enzymes
in the extracellular environment, such as
plasmin, trypsin, chymotrypsin, and
131
RESULTS
MMPs. Moreover, the resulting peptides
exert altered biological functions compared
to the whole molecule. The proteolytic
cleavage of HARP is also affected by the
presence of glycosaminoglycans (GAGs),
suggesting that a complex system serves to
regulate the overall effect of this growth
factor (19 and unpublished data).
Furthermore, HARP and HARP peptides
modulate the biological actions of other
growth factors such as VEGF, contributing
to the complex mode of growth factor
actions (20).
Prostate cancer (PCa) is the most common
cancer among men in Western countries,
although the development of PCa as well as
the signals contributing to the transformed
phenotype of PCa cells remains
incompletely understood (21).
During
adulthood, maintenance of normal prostate
function depends on mesenchymalepithelial interactions, which contribute to
the homeostatic equilibrium of the
glandular
prostate
epithelial
cells.
Disturbances in this equilibrium lead to the
development of diseases like PCa.
Although the mechanisms that control the
mesenchymal-epithelial interactions are
poorly understood, numerous studies
suggest that growth factors have a key role
in prostate homeostasis. HARP has been
implicated in PCa progression and acts as
an autocrine growth factor in various
prostate-derived cell lines including
DU145, PC3, and LNCaP (22, 23).
Truncated forms of HARP or synthetic
peptides corresponding to defined domains
of this growth factor have been studied in
an
attempt
to
understand
the
structure/function relationship of HARP
(24-26).
Studies of truncated HARP
lacking the amino acids 111-136 and
synthetic peptides corresponding to
residues 121-136 have revealed that the
carboxy terminal region is important for
several biological actions (27). Here, we
investigated the effect of a 10-amino acid
peptide (P122-131), corresponding to the
basic cluster of the C-terminal region of
HARP, on the adhesion, proliferation, and
growth of a prostate epithelial cell line as
well as on in vivo angiogenesis.
EXPERIMENTAL PROCEDURES
Materials
Standard
Boc
amino
acids,
pmethylbenzhydrylamine-polystyrene resin
(0.81 mmol NH2/g), and O-(benzotriazol-1yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate
(HBTU)
were
purchased from Senn Chemicals. Solvents
(peptide synthesis grade) and other
reagents were obtained from Applied
Biosystems. Cell culture reagents were
from BiochromKG (Seromed, Germany).
All other reagents were purchased from
Sigma-Aldrich.
Polyclonal antibodies against npSRCkinase (Tyr527), pFAK (Tyr925), pPTEN
(Ser380), pAKT (Ser473), pERK1/2
(Thr202/Tyr204), AKT, and PTEN, as well
as monoclonal antibodies against SRCkinase (36D10) were purchased from Cell
Signaling Technology.
Polyclonal
antibodies against total ERK1/2 and
monoclonal antibodies against SRC-kinase
were purchased from Upstate. Monoclonal
anti-RPTPβ/ζ antibodies were from BD
Transduction Laboratories (San Diego,
CA), and polyclonal ALK antibodies were
from Zymed (San Francisco, CA). Finally,
actin polyclonal antibody was purchased
from Sigma-Aldrich.
Cell culture
The human prostate cancer epithelial cell
line DU145 (ATCC) was grown in RPMI1640 medium supplemented with 10% fetal
bovine serum (FBS), 100 U/ml penicillin,
and 100 µg/ml streptomycin. Cultures
were maintained in 5% CO2 and 100%
humidity at 370C.
Peptide synthesis and characterization
Peptides were produced using stepwise
solid-phase synthesis with an ABI 433A
synthesizer
(Applied
Biosystems,
Courtaboeuf, France) and standard
protocols for Boc chemistry (amino acid
132
RESULTS
activation
with
dicyclohexylcarbodiimide/1hydroxybenzotriazole or HBTU). After
synthesis of the KKKKKEGKKQ domain,
the peptidyl-resin was elongated using four
glycine residues and biotin to produce
Biot-G4-P122-131 (B122-131). Peptides
were cleaved from the resin by treatment
with anhydrous hydrofluoric acid (1 h, 00C)
in the presence of anisole (1.5 ml/g
peptidyl-resin) and dimethylsulphide (0.25
ml/g peptidyl-resin). Peptides with purity
> 95% were obtained by high-performance
liquid chromatography on an RP-C8
column using a linear gradient of
acetonitrile
in
water
with
0.1%
trifluoroacetic acid.
Peptides were
characterized using MALDI-TOF MS
(Voyager Elite, PerSeptive Biosystems)
with matrix α-cyano-4-hydroxycinnamic
acid. The m/z values of the protonated
peptides (monoisotopic peak) were 1269
for H-P122-131 and 1683 for Biot-P122131.
The control peptide (5K) was
purchased from Sigma-Aldrich (SaintQuentin Fallavier, France).
Crystal violet assay
Adherent cells were fixed with methanol
and stained with 0.5% crystal violet in 20%
methanol for 20 min. After gentle rinsing
with water, the retained dye was extracted
with 30% acetic acid, and the absorbance
was measured at 590 nm.
MTT assay
Cell proliferation was measured by
reduction of 3-(4,5-dimethylthiazol-2-yl)diphenyltetrazolium bromide (MTT) as
previously described (28, 29).
Soft agar growth assay
Anchorage-independent
growth
was
assessed by measuring the formation of
colonies in soft agar. Twelve-well plates
were layered with bottom agar, which
consisted of growth medium containing
10% FBS and 0.8% agar. After the bottom
agar had solidified, 2000 cells were
resuspended in growth medium containing
10% FBS, 0.3% agar, and peptide, then
seeded onto the bottom agar. The top agar
was then allowed to solidify, and standard
growth media supplemented with peptide
was added to each well. The cells were
incubated at 370C, in 5% CO2 for 12 days.
Cell colonies larger than 50 µm were
quantified by counting the entire area of
each well.
Transwell assay
Migration assays were performed in 24well microchemotaxis chambers (Costar,
Avon,
France)
using
uncoated
polycarbonate membranes with 8 µm pores,
as described elsewhere (10).
Chicken
embryo
chorioallantoic
membrane (CAM) assay
The in vivo CAM angiogenesis model was
used as previously detailed (10).
Immunofluorescence
microscopy
confocal
DU145 cells grown in 8-well tissue culture
slides (Nunc) were incubated with
biotinylated P122-131 (B122-131) at 40C
for the indicated time. The cells were then
fixed in 4% paraformaldehyde for 10 min
at room temperature, rinsed three times
with PBS, quenched with 50 mM Tris
buffer pH 8.0 and 100 mM NaCl,
permeabilized for 15 min in PBS
containing 0.3% Triton X-100 and 0.5%
bovine serum albumin (BSA), and blocked
in PBS containing 3% BSA for 1 h at room
temperature. Cells were incubated for 1 h
with streptavidin-FITC (1:100), antiRPTPβ/ζ antibody (1:100), and rhodamineconjugated goat anti-mouse IgG (1:600) in
permeabilization buffer. After three rinses
in PBS, cells were mounted using Sigma
mounting fluid. Labeling was observed
using a Nikon confocal microscope and
photographed.
Reverse transcriptase-polymerase chain
reaction (RT-PCR) for ALK, RPTPβ/ζ,
and GAPDH
Total RNA was extracted using the
Nucleospin RNA II kit (Macherey-Nagel,
Germany), according to the manufacturer’s
133
RESULTS
instructions. The integrity of he isolated
RNA was examined by electrophoresis on
a 1% agarose gel containing 0.5 mg/ml
ethidium bromide. Specific primers were
as
follows:
hRPTPβ/ζ,
5΄AGATGGTTTACCCTTCTGAAAG C-3΄
and 5΄-TTTCACTAGAAGCAGGGTCAG
AG-3΄;
hALK,
5΄AAGCCTTCATAGGCGGCG ACATGC3΄ and 5΄-TGTGACTTCCACCAGGA
CTGTGCCC-3΄;
hGAPDH,
5΄CCACCCATGGC AAATTCCATGGCA3΄
and
5΄-TCTAGACGGC
AGGTCAGGTCCACC-3΄. The RT-PCR
reactions were performed in a single step
with 250 ng of total RNA, using the Qiagen
RT-PCR system. The RT-PCR products
were subjected to electrophoresis on 1%
agarose gel containing 0.5 mg/ml ethidium
bromide, digitally photographed, and
quantified using image analysis software
(Scion Image PC, Scion Corporation,
Frederick, MD).
siRNA transfection
RNA oligonucleotide primers and the
siPORT NeoFX Transfection Agent were
obtained from Ambion Inc. The following
sequences were used: RPTPβ/ζ sense, 5΄AAAUGCGAAUCCUAAAGCGUU-3΄;
RPTPβ/ζ
antisense,
5΄AACGCUUUAGGAUUCGCAU UU-3΄.
The annealing of the primers and the
transfection was performed according to
Ambion’s instructions. Briefly, siPORT
NeoFX and siRNA were mixed at a final
ratio of 1:10 in OPTI-MEM media. The
transfection complexes were then overlaid
onto 6-well plate cultures grown in RPMI1640 supplemented with 10% FBS.
Transfection efficiency was evaluated
using Silencer FAM-Labeled GAPDH
siRNA (Ambion).
Double-stranded
negative control siRNA from Ambion was
also used.
Immunoprecipitation
Media from DU145 cultures grown in 60
mm plastic dishes were aspirated, cells
were washed twice with ice-cold PBS, and
cells were lysed in 1 ml buffer containing
50 mM HEPES pH 7.0, 150 mM NaCl, 10
mM EDTA, 1% Triton X-100, 1% Nonidet
P-40, 1 mM phenylmethylsulfonyl fluoride,
1 mM sodium orthovanadate, 5 µg/ml
aprotinin, and 5 µg/ml leupeptin. Cells
were harvested, sonicated for 4 min on ice,
and centrifuged at 20000 g for 10 min at
40C.
Approximately 400 µg of the
supernatant was then incubated with 30 µl
of protein A-Sepharose bead suspension for
60 min at room temperature. Beads were
collected by centrifugation, and the
supernatants were incubated overnight at
40C with anti-RPTPβ/ζ (1:200) or antiSRC-kinase (1:1.000) primary antibodies.
The mixtures were then incubated with 80
µl protein A-Sepharose beads for 3 h at
40C. The beads and bound proteins were
collected by centrifugation (10000 g, 40C),
washed three times with ice-cold lysis
buffer, and resuspended in 60 µl 2X SDS
loading buffer (100 mM Tris-HCl pH 6.8,
4% SDS, 0.2% bromphenol blue, 20%
glycerol, 0.1 M dithiothreitol). Samples
were then heated to 95–1000C for 5 min
and centrifuged. Fifty microliters of the
supernatant were analyzed by Western
blotting.
Western blot analysis
Cells were starved for 4 h, then incubated
with P122-131 for varying times. Cells
were subsequently washed twice with PBS
and lysed in 250 µl 2X SDS loading buffer
under reducing conditions. Proteins were
separated by SDS-PAGE and transferred to
an Immobilon-P membrane for 3 h in 48
mM Tris pH 8.3, 39 mM glycine, 0.037%
SDS, and 20% methanol. The membrane
was blocked in TBS containing 5% non-fat
milk and 0.1% Tween 20 for 1 h at 370C.
Membranes were then probed with primary
antibody overnight at 40C under continuous
agitation. Antibodies against total ERK1/2
and SRC-kinase were used at a 1:5000
dilution. Anti-RPTPβ/ζ antibody was used
at a 1:500 dilution and anti-actin antibody
at a 1:3000 dilution. All other antibodies
were used at a 1:1000 dilution. The blot
was then incubated with the appropriate
134
RESULTS
secondary antibody coupled to horseradish
peroxidase, and bands were detected with
the ChemiLucent Detection System Kit
(Chemicon International Inc., CA),
according
to
the
manufacturer’s
instructions. Where indicated, blots were
stripped in buffer containing 62.5 mM Tris
HCl pH 6.8, 2% SDS, 100 mM 2mercaptoethanol for 30 min at 500C and
reprobed. Quantitative estimation of band
size and intensity was performed through
analysis of digital images using the
ImagePC image analysis software (Scion
Corporation, Frederick, MD).
ELISA assay
The ELISA plate wells were positively
charged by the sequential reaction of
glutaraldehyde (GA) and spermine, as
previously described (30). Briefly, wells
were incubated with 100 µl of 0.2% GA in
0.1 M sodium phosphate (pH 5.0) for 20 h
at 250C. The wells were subsequently
washed twice with the same buffer for 5
sec, then incubated with 100 µl of spermine
(10 µg/ml) in 0.1 M sodium phosphate (pH
9.0) for 4 h at 370C. After the incubation
was complete, the wells were washed twice
for 5 sec with 100 µl 0.15 M NaCl. Any
remaining free sites were blocked by
incubating wells with 1% (w/v) BSA in 0.1
M sodium phosphate (pH 8.0) for 1 h at
370C. The wells were then washed four
times for 5 sec with 100 µl of 0.15 M
NaCl–0.01 M sodium phosphate buffer (pH
4.5) containing 0.1% Tween 20 (PBS-T,
pH 4.5). Next, 100 µl of 1 µg/ml heparin
or chondroitin sulfate c (Cs-C) was added
to the modified wells, allowing these
compounds to become immobilized onto
amino groups via their negative charges.
Following four, 5-sec washings in 100 µl
PBS-T (pH 7.2), 100 µl of the sample to be
tested (suspended in PBS-T) was incubated
for 16 h at 40C. The wells were washed
four times with 100 µl PBS-T (pH 7.2),
then incubated with 100 µl streptavidinHRP diluted in PBS-T (pH 7.2) for 1 h at
370C. The wells were washed four times
for 5 sec with 100 µl of PBS-T (pH 7.2)
and incubated with 100 µl TMB liquid
substrate system (Sigma-Aldrich) for 30
min at 250C. The reaction was terminated
by the addition of 100 µl of 2 N sulfuric
acid, and the reaction product was
measured at 450 nm.
Statistical analysis
Comparison of variability among groups
was performed using an unpaired t-test.
Each experiment included at least triplicate
measurements for each condition tested.
All results are expressed as mean ± S.E.M.
from
at
least
three
independent
experiments. Values of p less than 0.05
were accepted as significant (*p < 0.05,
**p < 0.01, ***p < 0.001).
RESULTS
To investigate whether P122-131 may have
biological activities that are related to the
induction of a transformed phenotype in
PCa cells, we tested the effect of P122-131
on tumor phenotypes in the wellestablished prostate carcinoma cell line,
DU145. We also investigated the effect of
this peptide on angiogenesis in vivo, using
the CAM assay. Since P122-131 contains
seven lysines and is highly charged, we
also examined the effects of two "mock
peptides in parallel. One consisted of Damino acids (designated AAD), while the
other consisted of five lysines (designated
5K).
P122-131 inhibits adhesion of DU145
cells
The effect of P122-131 on the adhesion of
DU145 cells was tested using three
approaches. First, an equal number of cells
was
incubated
with
increasing
concentrations of peptide and immediately
seeded on plastic culture plates. In the
second approach, cells were incubated with
different concentrations of the peptides for
30 min before seeding.
In the final
approach, cells were pre-incubated for 30
min with increasing concentrations of
peptide, then washed and seeded on plastic
plates. After a 1-h incubation period,
135
RESULTS
adherent cells were measured by the crystal
violet assay. Under all conditions, a
maximal inhibition of 30% was observed at
the concentration of 5 µΜ P122-131 (Fig.
1A).
P122-131 inhibits proliferation of DU145
cells
The effect of P122-131 on the proliferation
of pre-plated and serum-starved cells was
investigated. We found that P122-131
inhibited proliferation in a concentrationdependent manner (data not shown),
having a maximal effect (25% inhibition
relative to control) at a concentration of 5
µΜ (Fig. 1B). To determine whether the
inhibitory effect of P122-131 was also
anchorage independent, soft agar growth
assays were performed. As shown in Fig.
1C, a similar inhibitory effect was found,
with a maximal inhibition of 30%
occurring in the presence of 5 µM P122131. Taken together, these results indicate
that P122-131 can inhibit both anchoragedependent and -independent growth of
DU145 cells.
P122-131 inhibits migration of DU145
cells
We next investigated the effect of P122131 on DU145 chemotaxis, as measured
using Transwell assays. Similar to its
effects on adhesion and proliferation,
P122-131 inhibited chemotactic migration
in a concentration-dependent manner (data
not shown). The maximum effect (25%
inhibition relative to control) was observed
at the concentration of 5 µM (Fig. 1D).
P122-131
inhibits
angiogenesis
the
in
vivo
The inhibitory effects of P122-131 on
DU145 adhesion, proliferation, and
migration are consistent with a possible
anti-tumor action for this peptide.
Therefore, we tested the effect of this
peptide on in vivo angiogenesis. Tumor
angiogenesis plays a key role in cell
proliferation by providing nutrients and
oxygen.
It also facilitates metastasis
through the formation of new, leaky
vessels.
We observed that P122-131
reduced the total length of blood vessels in
the CAM assay in a concentrationdependent manner (data not shown).
Angiogenesis was inhibited up to 40%,
with maximal inhibition occurring in the
presence of 100 µM P122-131 (Fig. 1E).
In contrast to P122-131, neither AAD nor
5K had any measurable effect on
angiogenesis (Fig. 1E). Similarly, neither
affected adhesion, proliferation, or
migration of DU145 cells (Fig. 1A-D).
DU145 cells express RPTPβ/ζ, but not
ALK
P122-131 harbors a cluster of basic
residues known to bind to cell receptors
(24).
To begin to understand the
mechanism through which P122-131 exerts
its biological actions, we investigated the
effect of this peptide on signaling mediated
by the HARP receptors, ALK and
RPTPβ/ζ.
First, we investigated the
expression profile of ALK and RPTPβ/ζ in
DU145 cells, both at the mRNA and
protein level. In contrast to HUVECs,
which expressed both receptors, DU145
cells expressed only RPTPβ/ζ (Fig. 2).
P122-131 binds to RPTPβ/ζ receptor and
is endocytosed
Next, DU145 cells were co-labeled for
B122-131 and RPTPβ/ζ to determine
whether P122-131 may exert its effects
through RPTPβ/ζ.
After a 30-sec
incubation of cells with B122-131,
confocal microscopy revealed that B122131 bound to the cell surface and colocalized with RPTPβ/ζ. After a 20-min
incubation, endocytotic vesicles containing
both B122-131 and RPTPβ/ζ were detected
in the cytoplasm, while non-interacting
B122-131 and RPTPβ/ζ were located at the
cell surface (Fig. 3A).
As expected,
incubation of cells with only streptavidinFITC or rhodamine conjugated secondary
antibodies produced no signal (data not
shown).
To provide additional support for an
interaction between P122-131 and this
receptor, B122-131 was precipitated from
136
RESULTS
DU145 lysates using streptavidin-bound
agarose beads, and precipitates were
probed for RPTPβ/ζ by Western blotting.
As control, DU145 lysates were incubated
with anti-RPTPβ/ζ monoclonal antibodies
immobilized on protein-G Sepharose
beads.
Western blot analysis of
streptavidin precipitates revealed that
RPTPβ/ζ specifically interacted with P122131 (Fig. 3B).
RPTPβ/ζ mediates the biological actions
of P122-131
To determine whether RPTPβ/ζ may
mediate the effects of P122-131 in DU145
cells, we transiently transfected cells with
siRNA targeting the 5΄UTR of RPTPβ/ζ
mRNA. Time course experiments revealed
that maximal down-regulation of RPTPβ/ζ
mRNA and protein was observed at 2 and 3
days after transfection, respectively. No
reduction of the RPTPβ/ζ expression was
observed in cells transfected with a
negative control siRNA (Fig. 4). As shown
in Fig. 5, siRNA-mediated knockdown of
RPTPβ/ζ reduced the inhibitory effects of
P122-131 on adhesion, anchoragedependent and -independent proliferation,
and migration. These results indicate that
RPTPβ/ζ helps mediate the biological
effects of P122-131.
P122-131-RPTPβ/ζ binding is associated
with SRC-kinase activation
SRC-kinase is known to mediate HARP
signaling (18, 31). We have previously
shown that HARP binding to RPTPβ/ζ
leads to SRC-kinase activation (18). To
determine whether SRC-kinase may be
activated by P122-131, DU145 cells were
serum starved for 4 h, then incubated with
P122-131 for 5 to 60 min. SRC-kinase
activation was indirectly assessed by
Western blot analysis of phosphorylated
and total SRC-kinase.
Preliminary
experiments revealed that maximal
activation occurred 15 min after P122-131
treatment (data not shown). Thus, this
incubation time was selected for doseresponse experiments. As shown in Fig.
6A, P122-131 elicited a 65% decrease in
SRC-kinase phosphorylation at the
concentration of 0.1 µΜ.
To determine whether RPTPβ/ζ mediates
SRC-kinase activation in response to P122131,
we
probed
RPTPβ/ζ
immunoprecipitates for SRC-kinase by
Western blot. As can be seen in Fig. 6B,
SRC-kinase was detected in RPTPβ/ζ
immunoprecipitates.
Since RPTPβ/ζ is a chondroitin sulfate
proteoglycan (32), we also tested the effect
of Cs-C on SRC-kinase activation. As
shown in Fig. 8A, inhibiting P122-131RPTPβ/ζ interactions with Cs-C (10 µg/ml,
45 min pretreatment) partially blocked
P122-131-induced SRC-kinase activation.
These results support the hypothesis that
P122-131 binding to RPTPβ/ζ leads to s
SRC-kinase activation.
P122-131 activates FAK, ERK1/2, and
PTEN
To investigate the extent of P122-131induced signaling, we also explored the
effects of P122-131 on activation of other
molecules known to interact with SRCkinase.
We found that
FAK
phosphorylation was increased 15 min after
incubation of DU145 cells with P122-131.
Similar
to
SRC-kinase,
FAK
phosphorylation increased up to 50%, with
maximal
responses
occurring
at
concentrations of 0.1 µΜ (Fig. 7).
Preincubation of cells for 45 min with 10
nM of the SRC-kinase inhibitor, PP1,
inhibited FAK phosphorylation in response
to P122-131 (Fig. 8B). As shown in Fig. 7,
ERK1/2 and PTEN were also activated after
a 15-min incubation P122-131, with
phosphorylation increasing up to 40 and
45%, respectively. (Fig. 7). No interaction
between PTEN and either SRC-kinase or
FAK
could
be
detected
by
immunoprecipitation experiments (data not
shown).
Finally, in the same set of
experiments, no change in AKT activation
(phosphorylation of Ser473) was observed
(Fig. 7, blot not shown).
137
RESULTS
The biological actions of P122-131 are
inhibited by Cs-C or PP1
DU145 cells were treated with Cs-C or PP1
to determine whether RPTPβ/ζ and SRCkinase may participate in the inhibitory
effects of P122-131 on adhesion,
proliferation, and migration. As shown in
Fig. 9, both Cs-C and PP1 partially
reversed the effects of P122-131,
implicating P122-131-RPTPβ/ζ interactions
and subsequent SRC-kinase activation in
all of these anti-tumor phenotypes.
P122-131-induced kinase activation is
mediated by RPTPβ/ζ
To determine whether P122-131 signaling
is RPTPβ/ζ-dependent, we tested the effect
of RPTPβ/ζ-specific siRNA on activation of
SRC-kinase, FAK, PTEN, and ERK1/2. As
shown in Fig. 10, RPTPβ/ζ knockdown
inhibited activation of all four molecules
by P122-131.
P122-131 competes with growth factors
for binding to Heparin and Cs-C
Since P122-131 binding to RPTRβ/ζ is
inhibited by Cs-C, we tested the hypothesis
that this peptide could compete with other
growth factors for binding sites on the cell
surface or the extracellular matrix. Cs-C or
heparin was immobilized on ELISA plates
(at approximately 80% saturation) and
incubated with increasing concentrations of
B122-131. We found that B122-131 binds
to Cs-C and heparin in a concentrationdependent manner (data not shown).
Incubation of B122-131 with either P122131, HARP, FGF-2 or VEGF revealed that
HARP and VEGF, but not FGF-2,
competed with P122-131 for binding to CsC or heparin (Fig. 11). This suggests that,
in vivo, this peptide could regulate growth
factor activity by inhibiting binding to their
receptors and/or inducing the release of
bound growth factors from the ECM. The
fact that FGF-2 did not dissociate P122131 from Cs-C indicates that it does not
compete for regions specific to Cs-C.
Indeed, FGF-2 is known to bind
phosphacan (a spliced form of RPTRβ/ζ) at
the protein core of the molecule and not at
the Cs-C portion of the receptor (33).
DISCUSSION
During the last decade, HARP has come to
be recognized as a pleiotropic growth
factor that participates not only in neurite
outgrowth in the developing brain (1), but
also in cell proliferation, migration,
angiogenesis, and malignant transformation
of many cell types. HARP is elevated in
sera or tumors from patients with colon,
stomach, pancreatic, and breast cancer (210). Moreover, the differential expression
of HARP mRNA and protein among
normal and malignant prostate epithelial
cells, implicates this protein in the
induction of a transformed phenotype (22).
HARP has a high affinity for heparin and
other GAGs, with GAG length and sulfate
content strongly influencing binding.
Binding is also affected by HARP
multimerization, which is catalyzed by
tissue transglutaminase.
The HARP
monomer contains two heparin binding
sites that also serve as CS binding sites
(34-37), and multimerization is greatly
enhanced by GAGs such as heparin.
HARP also binds to specific cell surface
receptors such as ALK (13), RPTPβ/ζ (12),
and N-Syndecan (11). Since N-Syndecan
is a low-affinity receptor mediating HARP
signal transduction mainly in neurons, we
focused our attention on RPTPβ/ζ and
ALK. Our results revealed that DU145
cells exclusively express RPTPβ/ζ.
RPTPβ/ζ is synthesized as a membranebound
CS
proteoglycan
and
its
extracellular variant, which is generated by
alternative splicing, is phosphacan, a major
soluble CS proteoglycan (38). HARP
binding to RPTPβ/ζ depends on the CS
portion of this receptor, and the removal of
CS results in a remarkable decrease in
binding affinity (39).
Contradictory results have been published
concerning HARP binding and activation
of RPTPβ/ζ.
Studies have provided
evidence that HARP induces the
138
RESULTS
dimerization of RPTPβ/ζ, which leads to an
increase in the tyrosine phosphorylation of
specific substrates (31, 40 and 41). Other
results indicate that HARP binding to
RPTPβ/ζ
induces
SRC-kinase
dephosphorylation (18, 42). Since HARP
binds to GAGs and is multimerized, it
seems logical that dimerization of RPTPβ/ζ
could be induced. On the other hand,
HARP could also activate RPTPβ/ζ
phosphatase upon direct binding to this
receptor.
To date, HARP activities have been
attributed either to the entire molecule or to
specific domains.
HARP peptide
fragments have been detected in cell
supernatants as well as in tissues (29, 43),
and such peptides can also be generated in
vitro by proteolytic cleavage of HARP
(19).
Our laboratory has already
characterized the biological actions of
several HARP peptides and proposed that
the biological activity of HARP is mainly
attributed to the two central domains of the
molecule as well as its C-terminal region
(19, 24, 27 and 29).
Therefore, the
biological actions of HARP should be
always considered to be the overall
outcome of its secretion, degradation, and
specific cleavage, with latter event possibly
generating HARP peptides with diverse, or
even opposite, biological actions.
To
illustrate this point, a recent study on
glioblastoma cell
proliferation and
migration has revealed that cleavage of the
12 C-terminal-most amino acids from
HARP (124-136) leads to distinct
biological activities through differential
activation of RPTPβ/ζ or ALK signaling
pathways (16).
In this study, we further characterized this
C-terminal region of HARP using the
synthetic peptide P122-131. We have
shown that P122-131 inhibits DU145 cell
adhesion,
proliferation
(anchoragedependent
and
-independent),
and
migration, while HARP exerts opposite
effects (unpublished data). The CAM
assay revealed that P122-131 suppressed
the formation of new blood vessels, a
process important for tumor growth and
metastasis. Conversely, HARP has been
shown to induce angiogenesis (10).
Our results demonstrate that P122-131
actions are mediated by RPTPβ/ζ. P122131 was colocalized with RPTPβ/ζ at the
cell surface and eventually become
cytoplasmic, likely as a result of
endocytosis.
Immunoprecipitation/
Western blotting confirmed the interaction
between P122-131 and RPTPβ/ζ. The
finding that P122-131 actions could be
abolished by RPTPβ/ζ siRNA or Cs-C
demonstrates that the RPTPβ/ζ receptor
meditates P122-131 actions and suggests
that binding occurs at the GAG portion of
the receptor.
The biological activity of P122-131 could
be attributed solely to its high positive
charge. Nevertheless, this does not seem to
be the case, since, in the same set of
experiments, neither AAD nor 5K exerted
any detectable biological activity. Thus,
the action of P122-131 is more likely due
to its specific amino acid sequence and
charge.
Previous studies have revealed that HARP
activates SRC-kinase after binding to
RPTPβ/ζ (18). We have shown that this is
also the case for P122-131, since RPTPβ/ζ
co-immunoprecipitates with SRC-kinase
and
P122-131-induced
SRC-kinase
activation is blocked by Cs-C or RPTPβ/ζ
siRNA. P122-131 also activated PTEN,
ERK1/2, and FAK, with activation of the
latter being reversible by SRC-kinase
inhibition. Importantly, all four signaling
molecules lay downstream of P122-131induced RPTPβ/ζ activation, as shown by
siRNA-mediated RPTPβ/ζ knockdown.
The activation of the above-mentioned
kinases was concentration-dependent.
Maximal activation occurred at a P122-131
concentration of 0.1 µM, while higher
concentrations (up to 10 µM) returned
activity to control levels. Since activation
of this pathway induces cell adhesion,
proliferation,
and
migration,
we
investigated the possibility that P122-131
may interplay with HARP and/or other
139
RESULTS
growth factors. Using a modified ELISA
mimicking components of the extracellular
matrix, we found that P122-131 displaced
heparin- or Cs-C-bound HARP or VEGF in
a concentration-dependent manner. We
speculate that the maximum induction of
signaling at a concentration of 0.1 µM is
due to competition of P122-131 with other
growth factors for ECM binding sites,
leading to the release of these growth
factors from the ECM and increased
availability for binding with their cognate
receptors. At 5 µM, the concentration at
which P122-131 had maximal biological
activity, P122-131 competes with growth
factors not only for binding to the ECM,
but also for binding to their receptors. In
this way, P122-131 binding to the ECM
may regulate the biological activity of
growth factors. Indeed, several studies
have shown that VEGF peptides not only
interact
with
cell
surface
glycosaminoglycans (44), but also prevent
the binding and mitogenic activity of
exogenously added growth factors like
FGF-2 (45).
Although the overall
biological actions can be measured,
understanding the contribution interplay
between growth factors, the ECM,
coreceptors, and growth factor cleavage
products remains very complicated.
Finally, since P122-131 does enter DU145
cells, direct binding to intracellular targets
may also modulate biological actions.
In conclusion, our results demonstrate
that P122-131, a HARP derived peptide,
not
only
inhibits
cell
adhesion,
proliferation, and migration, but also
exhibits a strong anti-angiogenic actions.
These biological actions are mediated, at
least in part, by RPTPβ/ζ and may involve
downstream signaling molecules such as
SRC-kinase, FAK, PTEN and ERK1/2.
P122-131 may act directly, after binding to
RPTPβ/ζ, although binding to undefined
cytoplasmic molecules should not be
excluded. At the same time, P122-131
might also antagonize the actions of growth
factors, such as HARP and VEGF, by
regulating their release from the
extracellular matrix and their receptors.
Cumulatively, these results indicate that
P122-131 is a potential anti-cancer agent,
and they warrant further study of this
peptide.
140
RESULTS
REFERENCES
1. Rauvala, H. (1989) EMBO J. 8(10), 2933-41
2. Fang, W., Hartman, N., Chow, D.T., Riegel, A.T., and Wellstein, A. (1992) J. Biol.
Chem. 267(36), 25889-97
3. Souttou, B., Juhl, H., Hackenbruck, J., Rockseisen, M., Klomp, H.J., Raulais, D.,
Vigny, M., and Wellstein, A. (1998) J. Natl. Cancer Inst. 90(19), 1468-73
4. Soulié, P., Héroult, M., Bernard-Pierrot, I., Caruelle, D., Oglobine, J., Barritault, D.,
and Courty, J. (2004) Cancer Detect Prev. 28(5), 319-24
5. Chauhan, A.K., Li, Y.S., and Deuel, T.F. (1993) PNAS 90(2), 679-82
6. Choudhuri, R., Zhang, H.T., Donnini, S., Ziche, M., and Bicknell, R. (1997) Cancer
Res. 57(9), 1814-9
7. Courty, J., Dauchel, M.C., Caruelle, D., Perderiset, M., and Barritault, D. (1991)
Biochem. Biophys. Res. Com. 180(1), 145-51
8. Czubayko, F., Riegel, A.T., and Wellstein, A. (1994) J. Biol. Chem. 269(33), 21358-63
9. Bowden, E.T., Stoica, G.E., and Wellstein, A. (2002) J. Biol. Chem. 277(39), 35862-8
10. Papadimitriou, E., Polykratis, A., Courty, J., Koolwijk, P., Heroult, M., and Katsoris, P.
(2001) Biochem. Biophys. Res. Commun. 282(1), 306-13
11. Raulo, E., Chernousov, M.A., Carey, D.J., Nolo, R., and Rauvala, H. (1994) J. Biol.
Chem. 269(17), 12999-3004
12. Maeda, N., Nishiwaki, T., Shintani, T., Hamanaka, H., and Noda, M. (1996) J. Biol.
Chem. 271(35), 21446-52
13. Stoica, G.E., Kuo, A., Aigner, A., Sunitha, I., Souttou, B., Malerczyk, C., Caughey,
D.J., Wen, D., Karavanovi, A., Riegel, A.T., and Wellstein, A. (2001) J. Biol. Chem.
276(20), 16772-9
14. Ulbricht, U., Eckerich, C., Fillbrandt, R., Westphal, M., and Lamszus, K. (2006) J.
Neurochemistry 98(5), 1497-506
15. Grzelinski, M., Bader, N., Czubayko, F., and Aigner, A. (2005) Int. J. Cancer 117(6),
942-51
16. Lu, K.V., Jong, K.A., Kim, G.Y., Singh, J., Dia, E.Q., Yoshimoto, K., Wang, M.Y.,
Cloughesy, T.F., Nelson, S.F., and Mischel, P.S. (2005) J. Biol. Chem. 280(29), 2695364
17. Powers, C., Aigner, A., Stoica, G.E., McDonnell, K., and Wellstein, A. (2002) J. Biol.
Chem. 277(16), 14153-8
18. Polykratis, A., Katsoris, P., Courty, J., and Papadimitriou, E. (2005) J. Biol. Chem.
280(23), 22454-61
19. Polykratis, A., Delbé, J., Courty, J., Papadimitriou, E., and Katsoris, P. (2004) Int. J.
Biochem. Cell. Biol. 36(10), 1954-66
20. Heroult, M., Bernard-Pierrot, I., Delbe, J., Hamma-Kourballi, Y., Katsoris, P.,
Barritault, D., Papadimitriou, E., Plouet, J., and Courty, J. (2004) Oncogene 23(9),
1745-53.
21. Gronberg, H. (2003) Lancet 361(9360), 859-864
22. Vacherot, F., Caruelle, D., Chopin, D., Gil-Diez, S., Barritault, D., Caruelle, J.P., and
Courty, J. (1999) Prostate 38(2), 126-36
23. Hatziapostolou, M., Delbe, J., Katsoris, P., Polytarchou, C., Courty, J., and
Papadimitriou, E. (2005) Prostate 65(2), 151-8
24. Bernard-Pierrot, I., Delbe, J., Rouet, V., Vigny, M., Kerros, M.E., Caruelle, D.,
Raulais, D., Barritault, D., Courty, J., and Milhiet, P.E. (2002) J. Biol. Chem. 277(35),
32071-7
25. Zhang, N., Zhong, R., and Deuel, T.F. (1999) J. Biol. Chem. 274(19), 12959-62
141
RESULTS
26. Zhang, N., Zhong, R., Wang, Z.Y., and Deuel, T.F. (1997) J. Biol. Chem. 272(27),
16733-6
27. Bernard-Pierrot, I., Delbe, J., Caruelle, D., Barritault, D., Courty, J., and Milhiet, P.E.
(2001) J. Biol. Chem. 276(15), 12228-34
28. Mosmann, T. (1983) J. Immunol. Methods 65, 55–58
29. Papadimitriou, E., Heroult, M., Courty, J., Polykratis, A., Stergiou, C., and Katsoris, P.
(2000) Biochem. Biophys. Res. Com. 274(1), 242-8
30. Triantaphyllidou, I.E., Sklaviadis, T., and Vynios, D.H. (2006) Anal. Biochem. 359(2),
176-82
31. Pariser, H., Ezquerra, L., Herradon, G., Perez-Pinera, P., and Deuel, T.F. (2005)
Biochem. Biophys. Res. Com. 332(3), 664-9
32. Maeda, N., Hamanaka, H., Shintani, T., Nishiwaki, T., and Noda, M. (1994) FEBS
Lett. 354, 67-70
33. Milev, P., Monnerie, H., popp, S., Margolis, R.N., and Margolis, R.U. (1998) J. Biol.
Chem. 274(34), 21439-42
34. Vacherot, F., Delbe, J., Heroult, M., Barritault, D., Fernig, D.G., and Courty, J. (1999)
J. Biol. Chem. 274(12), 7741-7
35. Maeda, N., Fukazawa, N., and Hata, T. (2005) J. Biol. Chem. 281(8), 4894-902
36. Bao, X., Muramatsu, T., and Sugahara, K. (2005) J. Biol, Chem. 280(42), 35318-28
37. Bernard-Pierrot, I., Heroult, M., Lemaitre, G., Barritault, D., Courty, J., and Milhiet,
P.E. (1999) Biochem. Biophys. Res. Commun. 266(2), 437-42
38. Maurel, P., Rauch, U., Flad, M., Margolis, R.K., and Margolis, R.U. (1994) PNAS
91(7), 2512-6
39. Maeda, N., He, J., Yajima, Y., Mikami, T., Sugahara, K., and Yabe, T. (2003) J. Biol.
Chem. 278, 35805-11
40. Meng, K., Rodríguez-Pena, A., Dimitrov, T., Chen, W., Yamin, M., Noda, M., and
Deuel, T.F. (2000) PNAS 97(6), 2603-8
41. Kawachi, H., Fujikawa, A., Maeda, N., and Noda, M. (2001) PNAS 98(12), 6593-8
42. Nam, H.J., Poy, F., Saito, H., and Frederick, C.A. (2005) J. Exp. Med. 201(3), 441-52
43. Bohlen, P., Muller, T., Gautschi-Sova, P., Albrecht, U., Rasool, C.G., Decker, M.,
Seddon, A., Fafeur, V., Kovesdi, I., and Kretschmer, P. (1991) Growth Factors 4(2),
97-107
44. Jonca, F., Ortega, N., Gleizes, P.E., Bertrand, N., and Plouet, J. (1997) J. Biol. Chem.
272(39), 24203–9
45. Betsholtz, C., Rorsman, F., Westermark, B., Ostman, A., and Heldin, C.H. (1990)
Nature 344(6264), 299
142
RESULTS
FOOTNOTES
This work was supported by a grant from the Research Committee of the University of
Patras (Karatheodoris) and by grants from the CNRS (ANR-06-567 RIB-016-012), INCA,
and Association pour la Recherche sur le Cancer (#3242).
The abbreviations used are: HARP, heparin affin regulatory peptide; RPTPβ/ζ, receptor
type protein tyrosine phosphatase β/ζ; ALK, anaplastic lymphoma kinase; MAPK,
mitogen-activated protein kinase; FAK, focal adhesion kinase; ERΚ, extracellular signalrelated kinase; PTEN, phosphatase with tensin homology which is detected on
chromosome 10; Cs-C, chondroitin sulfate-C; BSA, bovine serum albumin; TBS, Trisbuffered saline; PBS, Phosphate-buffered saline.
FIGURE LEGENDS
Figure 1: P122-131 promotes anti-tumor phenotypes in DU145 cells and inhibits
angiogenesis in vivo
(A), Number of adherent DU145 cells in the presence of 5 µM P122-131, AAD, or 5K, as
determined by the crystal violet assay. (B), Anchorage-dependent proliferation, as
measured by crystal violet assay and the MTT assay. (C), Soft agar growth assays
showing anchorage-independent proliferation. (D), Migration of cells through Transwell
filters. (E), Effect of 100 µM P122-131 on angiogenesis as measured by the chicken
embryo CAM assay.
Figure 2: DU145 cells express RPTPβ/ζ, but not ALK
Protein expression was analyzed by immunoprecipitation/Western blotting and mRNA
expression by RT-PCR. HUVECs served as a positive control.
Figure 3: P122-131 mediates its actions through binding to RPTPβ/ζ
(A), DU145 cells co-labeled for B122-131 (green) and RPTPβ/ζ (red) at 30 sec or 20 min
after incubation with P122-131. Overlapping labeling appears yellow in merged images.
(B), Co-immunoprecipitation/Western blot analysis of P122-131 and RPTPβ/ζ.
Precipitates obtained with (1) anti-RPTPβ/ζ antibody and protein-G sepharose, (2) B122131 and agarose-bound streptavidin, (3) or only agarose-bound streptavidin were probed
with RPTPβ/ζ specific antibodies.
Figure 4: siRNA-mediated RPTPβ/ζ knockdown in DU145 cells
RT-PCR analysis (A), and Western blot analysis (B) of RPTPβ/ζ. (1) non-transfected, (2)
transfected with negative control siRNA, (3) transfected with RPTPβ/ζ siRNA.
Figure 5: RPTPβ/ζ knockdown inhibits P122-131 biological activity in DU145 cells
Effect of P122-131 on adhesion (A), anchorage-dependent proliferation (B), anchorageindependent proliferation (C), and migration (D). siRNA-transfected (+), non-transfected
(-).
Figure 6: RPTPβ/ζ interacts and dephosphorylates SRC-kinase after P122-131
treatment
(A), Western blot analysis of phosphorylated SRC-kinase in cells stimulated with the
indicated concentration of P122-131 for 15 minutes. The blot was stripped and reprobed
for total SRC-kinase. (B), Co-immunoprecipitation/Western blot analysis of cells
143
RESULTS
stimulated with 0.1 µΜ P122-131 for 15 min. RPTPβ/ζ co-immunoprecipitates were
probed for SRC-kinase or RPTPβ/ζ.
Figure 7: PTEN FAK, and ERK1/2 are activated by P122-131
Western blot analysis of phosphorylated PTEN FAK, and ERK1/2 in cells stimulated with
the indicated P122-131 concentration for 15 min. The blots were stripped and reprobed for
PTEN, AKT, ERK1/2, or actin.
Figure 8: P122-131-induced SRC-kinase and FAK activation is inhibited by Cs-C or
PP1, respectively
Western blot analysis of phosphorylated SRC-kinase (A), and FAK (B) in DU145 cells
pretreated with Cs-C (10 µg/ml) or PP1 (10 nM), then stimulated with 0.1 µΜ P122-131
for 15 min.
Figure 9: RPTPβ/ζ, SRC-kinase, and FAK participate in the biological actions of
P122-131
Adhesion (A), anchorage-dependent proliferation (B), anchorage-independent proliferation
(C), and migration (D) of DU145 cells pretreated with Cs-C (10 µg/ml) or PP1 (10 nM),
then stimulated with 0.1 µΜ P122-131 for 15 min.
Figure 10: RPTPβ/ζ knockdown inhibits P122-131-induced kinase signaling
Western blot analysis of phosphorylated SRC-kinase, PTEN, FAK, and ERK1/2 in siRNAtransfected (+) or non-transfected (-) DU145 cells stimulated with 0.1 µM P122-131 for 15
min.
Figure 11: P122-131 competes with growth factors for binding to Heparin and Cs-C
Cs-C or heparin was immobilized on ELISA plates and incubated with B122-131 alone or
in combination with P122-131, HARP, FGF-2, or VEGF. B122-131 that remained bound
to Cs-C or heparin was quantified using streptavidin-conjugated HRP.
144
RESULTS
Figure 1
145
RESULTS
Figure 2
146
RESULTS
Figure 3
147
RESULTS
Figure 4
148
RESULTS
Figure 5
149
RESULTS
Figure 6
150
RESULTS
Figure 7
151
RESULTS
Figure 8
152
RESULTS
Figure 9
153
RESULTS
Figure 10
154
RESULTS
Figure 11
155
RESULTS
2.3 DISCUSSION
To pursue the investigation of the structure-function relationships of HARP, the
properties of a 10-amino acid peptide, P122-131, corresponding to the basic cluster of the
C-terminal region was characterized (article 2). We reported that P122-131 inhibited the
anchorage-independent (article 2) and anchorage-dependent cell proliferation, cell
migration, cell adhesion of prostatic cancer cell line, DU145 and in vivo angiogenesis
(article 3).
Previously, several studies have reported the functional significance of HARP in
development and progression of human prostate cancer (Vacherot et al., 1999,
Hatziapostolou et al., 2005, 2006). Although in normal prostate tissue, epithelial-stromal
interactions, often paracrine play an important role, in case of prostate cancer, some
growth factor pathways can become autocrine, enabling the epithelial cells to grow
independently of stromal cells (Russel et al., 1998). To assess the role of HARP in the
prostate-derived epithelial cell-line, DU145, the endogenous HARP produced by these
cells was targeted and monoclonal HARP antibodies inhibited the anchorage-independent
growth of DU145 cells. Hence, HARP acts as an autocrine growth factor and so regulates
prostatic tumor cell growth. In addition, HARP induced cell growth of DU145 cells
through binding to RPTPβ, since this was the only HARP receptor expressed in these cells.
The treatment of DU145 cells with polyclonal anti-RPTPβ directed against the
extracellular region of the protein, resulted in strong suppression of colony formation in
soft agar, indicating the direct and positive signaling role of RPTPβ. Based on these
studies, we concluded that HARP/RPTPβ- ligand/receptor pair plays an important role for
prostate cell growth and the inhibitory effects of P122-131 in DU145 cells may be due to
its antagonizing HARP/RPTPβ. To study the relevance of this hypothesis, we used several
experimental approaches. By cell-ELISA binding assays, we showed that the peptide
displayed a saturable and specific binding on DU145 cells that can be competed by HARP.
Immunoprecipitation experiments using DU145 cell lysates indicated that RPTPβ
specifically interacted with the peptide (article 2). Finally, confocal microscopy indicated
the co-localization of P122-131 with RPTPβ on the cell surface, indicating a likely
interaction of P122-131 with RPTPβ (article 3). It should be noted that further confocal
fluorescence studies carried out with permeabilized cells reported the detection of the
peptide in the nucleus as well. During the last decade, cell-penetrating basic peptides have
been the subject of the extensive studies and it is possible that the internalization and
156
RESULTS
translocation to the nucleo-cytoplasmic space of the peptide is due to its highly basic
character. From this, one might speculate that the biological activities might simply be due
to its highly positive charge and not recognition of the primary amino acid sequence.
However, this is clearly not the case for P122-131, since the control peptides; KKKKK and
the peptide P122-131 synthesized from D-amino acids didn’t exert any effect under the
same experimental conditions.
RPTPβ has two binding sites with different affinities for HARP: The binding of
RPTPβ to HARP depends in part on the CS portion of RPTPβ, since CSase digestion of
cells resulted in decrease of binding affinity. In addition, HARP binds to the RPTPβ core
protein with high affinity. It seems that the CS chains play a regulatory role for HARP
binding. Since P122-131 has been shown to bind to RPTPβ, we investigated whether the
GAG moiety of RPTPβ was involved in the interaction with the peptide. The results
presented in two articles that comprise this chapter represent an ambiguous response to this
question. In the first report, we showed that the removal of CS from cells didn’t alter the
binding affinity to RPTPβ, so we deduced that the peptide binds to the core protein. In
contrast, the soluble CS, when added to cell culture medium, abolished the effects of P122131 (article 3), which implicates at least an indirect interaction with CS, which does not
need to be physically associated with RPTPβ.
Since the mechanism of signaling of HARP through RPTPβ is ligand-induced
inactivation of the tyrosine phosphatase activity (Meng et al., 2001), P122-131 could exert
its antagonistic effects, by potentially enhancing total cellular phosphatase activity, and
hence could inhibit cell growth. Thus, we studied the signaling pathways of P122-131 in
DU145 cells. The results indicated the activation of SRC-kinase, FAK, PTEN and ERK
signaling molecules. No change in AKT activation was observed. The SRC-kinase family
member, FYN, has been previously shown to be phosphorylated upon HARP binding to
RPTPβ, increasing its tyrosine kinase activity (Pariser et al., 2005a). In a similar vein,
other authors reported that HARP binding to RPTPβ increased the dephosphorylation of
SRC at tyrosine 527, which leads to its activation (Polykratis et al., 2005). Our second
work on P122-131 reported the phosphorylation of substrate molecules of RPTPβ with a
maximum activation at 0.1 µM, but this effect was no longer apparent at higher
concentrations of peptide: 1 and 10 µM (article 3). Here, it is important to note that the
biological effect of the peptide was observed at 5 µM. Therefore, the current experimental
evidence may not be sufficient to provide a convincing molecular mechanism for the
157
RESULTS
observed biological actions of P122-131. The regulation of RPTPβ signalization is
complex and it is likely that it is the result of different tyrosine phosphatases and multiple
protein tyrosine kinases. Therefore, the exact mechanism of P122-131 in DU145 cells after
binding to RPTPβ is unclear and remains to be elucidated.
158
3. MOLECULAR INTERACTIONS OF
HARP AND HARP-DERIVED PEPTIDES
RESULTS
3. MOLECULAR INTERACTIONS OF HARP AND HARP-DERIVED PEPTIDES
This chapter of the thesis will tackle two aspects of HARP function. In the first
part of the chapter, the molecular basis underlying HARP-heparin interactions and the
heparin-binding domains of HARP will be studied. In the second part, the interactions of
HARP with other proteins will be investigated.
3A. MOLECULAR BASIS OF HARP-HEPARIN INTERACTION
3A.1 INTRODUCTION
The glycosaminoglycans (GAGs) are a major class of extracellular complex
polysaccharides. The GAGs heparan sulfate and its experimental proxy heparin, a
specialized form of heparan sulfate (HS) produced by mast cells, have been shown to bind
many proteins and regulate their activities. Both polymers are linear, acidic
polysaccharides composed of disaccharide repeating units of a uronic acid linked to a
hexosamine. The uronic acid can be either β-D-glucuronic acid (GlcA) or its epimer α-Liduronic acid (IdoA). The amino sugar is a D-glucosamine (Powell et al., 2004). Each
GAG backbone can be modified by sulfation at the uronic acid and hexosamine. The
uronic acid can be sulfated at position 2. Glucosamine derivatives can be either Ndeoxyacetamido (GlcNAc) or N-deoxysulfonamido (GlcNS) substituted and O-sulfated at
positions 3 (relatively rare) and 6 (Figure 22), though a small proportion of the amino
groups are not substituted, which can have important functional implications (Vanpouille
et al., 2007).
GlcN
IdoA
Figure 22: α1-4 linked IdoA-GlcN dissacharide repeating structure of heparin/HS chains. For
clarity, the figure shows the uronic acid moiety as α-L-IdoA but this can also be the C-5 epimer βD-GlcA.
Heparin is far more homogenous than heparan sulfate (HS) with around 75% of
its disaccharides being trisulfated and so it displays higher N and O-sulfation than HS
(Gallagher and Walker, 1985). HS further differs from heparin, displaying a far greater
160
RESULTS
chemical heterogeneity in terms of sulfation pattern and backbone chemical structure. In
HS, only a fraction of glucosamines are N-sulfated and these are clustered. The clustering
of the N-sulfation and post polymerization modifications result in the formation of a
distinct domain structure. Domains of low sulfation separate the highly modified sulfated
S-domains and proteins bind these S-domain structures (Lyon and Gallagher, 2000). Most
experimental approaches use heparin as a proxy for heparan sulfate because heparin is
readily available, due to its clinical use as an anticoagulant, though this assumes that
heparin is roughly equivalent to the protein-binding S-domains of heparan sulfate. Protein
recognition of heparin depends upon substitution pattern of both IdoA- and GlcNcontaining sequences. A first attempt at defining the GAG structures responsible for HARP
binding measured the ability of selectively desulfated polysaccharides to displace 3Hlabeled heparin from HARP (Kinnunen et al., 1996). The results of competitive binding of
unlabeled polysaccharides indicated that particularly 2-O-sulfated iduronic acid units and
to a lesser extent glucosamine N-sulfate and 6-O sulfate groups within heparin were
important for binding to HARP. The results of neurite outgrowth assays using heparin and
its modified derivatives reflected essentially those obtained in the competitive binding
assays. Thus, HARP-induced neurite outgrowth was observed to be inhibited by heparin,
explained by the heparin-dependence of this biological activity and the inhibitory activity
of heparin was essentially lost upon selective 2-O-desulfation and to a lesser but
appreciable manner upon selective 6-O-desulfation and N-desulfation followed by Nacetylation (Kinnunen et al., 1996). The minimum size of oligosaccharide required for
HARP-binding, as well as for its neurite outgrowth activity, was found to be a
deccasaccharide (Kinnunen et al., 1996). Similar to neurite outgrowth activity, heparin has
been shown to be required for the mitogenic activity of the protein. However, these studies
reported that heparin potentate the mitogenic activity of HARP and antagonized only at
higher concentrations of heparin, hence displaying a dual effect (Vacherot et al., 1999a).
However, the GAG structure implicated in the mitogenic property of HARP has not been
investigated. Beside regulating the activities of the protein, there is also evidence that
binding to the chains of proteoglycans also serves to regulate the storage and diffusion of
the protein at the cell surface and in the extracellular matrix, since HARP can be released
either by addition of heparin or by enzymatic cleavage of GAGs (Vacherot et al., 1999a).
Heparin has been suggested to induce a conformational change in HARP
(Kilpelainen et al., 2000), which is thought to be important for efficient binding of the
growth factor to its receptor. Heparin-binding sites of HARP have been suggested to be in
161
RESULTS
both N- and C-terminal domains, homologous to TSR motifs, whereas the poly lysine-rich
tails of HARP have, intriguingly, not been found to be important (Kilpelainen et al., 2000,
Raulo et al., 2005). Interactions between heparin and HARP have been quantitatively
characterized using a number of techniques, including optical biosensors (SPR, resonant
mirror), ITC, filter trap. However, values for the affinity of heparin-HARP interaction vary
considerably depending on the techniques and the laboratory. Thus, Kd values range from 4
nM to 460 nM, a 100-fold difference between these measurements. The subject was
discussed at length in the “Review” section, and because of the constraints described, care
is required for evaluation of the current published experimental data (see the section
“Heparin/HS”) and GAG-HARP interactions need to be further defined. Moreover, the
consensus motifs that underlie the interaction of HARP with heparin require a far more
precise definition.
In order to assess these issues, as a first step, we explored the ability of various
peptides corresponding to different domains of HARP, using binding assays in a resonant
mirror optical biosensor. Once the heparin-binding sequences within the protein were
determined, we then examined whether the structural determinants in heparin that have
been proposed to account for HARP binding were also required for interactions with
different domains of the protein. For this, heparin-derivatized aminosilane surfaces were
used. In three sets of experiments, the ability of 1) chemically modified polysaccharides, 2)
oligosaccharides of different lengths and 3) cation-bound heparin derivatives, to compete
with immobilized heparin for the binding of HARP and its putative heparin-binding
domains were examined.
3A.2 RESULTS
3A.2.1 HEPARIN-BINDING DOMAINS OF HARP
To identify the heparin-binding sites of HARP, we examined the heparin-binding
properties of several putative sequences of HARP using synthetic peptides. The peptides
corresponding to residues 1-12 (N1-12), 13-39 (NTSR), 65-97 (CTSR), 111-136 (P111136), 111-121 (P111-121), 111-124 P(111-124) and 122-131 (P122-131) were synthesized
(Altergen) and their apparent affinities for immobilized heparin were examined. Human
recombinant HARP produced from bacteria (see “Materials and Methods”) was used as
control (Figure 23).
CTSR bound substantially to immobilized heparin, while the thrombospondin
domain corresponding to the N-terminus (NTSR) at a 10-fold higher concentration failed
162
RESULTS
to bind. P111-136, corresponding to the lysine-rich C-terminal and at a concentration 10fold higher than that of CTSR exhibited a moderate binding to heparin whereas N1-12
corresponding to N-terminal lysine-rich tail, at a 2-fold higher concentration than P111136 bound less. Since peptides related to the HARP C-terminus, 122-131, 111-121 and
111-124 didn’t exhibit a significant binding comparing to P111-136, the results suggest
that the residues from 124 to 136 are mainly responsible for the heparin-binding properties
of the C-terminal tail. Interestingly, the scrambled peptide from 111-136 also bound to
immobilized heparin, suggesting that the net positive charge of the C-terminal tail might
account for its heparin affinity, rather than its primary structure. However, a series of such
scrambled peptides would need to be examined to establish this, since the orientation of the
basic groups in the scrambled peptide might, by chance be appropriate for binding to
heparin.
50
40
30
20
10
0
Figure 23: Binding of synthetic peptides and HARP to immobilized heparin. Various peptides
derived from HARP were added as soluble ligates in 5 µl PBS-T to heparin-derivatized cuvette
containing 45 µl of PBS-T. The extent of binding was measured using a one-site binding model.
Results are triplicates of one of two separate experiments carried out two different heparinderivatized surface.
163
RESULTS
These data demonstrate that the residues 65-97 are likely to form a primary
recognition site for heparin within HARP. Since peptide P111-136 bound to heparin, but
peptide P111-124 failed to do so significantly, the additional basic residues in the Cterminal tail, probably the residues from 124 to 136, are also likely to contribute to the
heparin affinity of the protein. Thus, we used the CTSR peptide and the C-terminal
truncated HARP designated HARPΔ124-136 for further analysis and the full-length HARP
was used as a control.
3A.2.2 INTERACTIONS OF HARP, HARPΔ124-136 AND CTSR WITH HEPARIN AND
CHEMICALLY MODIFIED HEPARINS
It has been suggested that the interaction of HARP with heparin requires
particular sulfation patterns of the iduronic acid and hexosamine residues. HARP interacts
mainly with 2-O-sulfate groups of IdoA unit, while N-sulfates and 6-O-sulfates of
GlcNSO3 residues are less important for HARP interaction. In this study, we explored the
specificity of heparin-CTSR and heparin-HARPΔ124-136 interactions, using binding
studies involving chemically modified heparins. In this approach, one or more sulfate
groups from disaccharide units of heparin is/are selectively removed by chemical
treatment. This is a widely used approach to investigate the involvement of particular
groups within heparin chains for binding, because the modification in the structure of
heparin is often correlated with a change in binding to the protein, though most studies
only use three modified heparins. Uniquely, by collaboration with Dr E. Yates, University
of Liverpool, a complete set of chemically modified heparins was available for this study.
There are three positions at which substitutions can be made: these are sulfate or
hydroxyl at position 2 of iduronate or at position 6 of glucosamine and N-sulfate or Nacetyl at position 2 of glucosamine. Following orthogonal chemical desulfation at these
positions, there are 23 = 8 possible combinations of substitutions pattern (Yates et al.,
2004). The 8 so-formed disaccharides are shown in figure 24.
164
RESULTS
1
5
2
6
3
7
4
8
Figure 24: Chemically modified polysaccharides. The numbers indicate; 1) fully sulfated heparin
(2S, 6S, NS) 2) N-desulfated/acetylated (2S, 6S, De-NS Re-NAc) 3) 2-O-desulfated (2OH, 6S, NS)
4) 6-O-desulfated (2S, 6OH, NS) 5) 2-O-desulfated, N-desulfated/acetylated (2OH, 6S, De-NS ReNAc) 6) 6-O-desulfated, N-desulfated/acetylated (2S, 6OH, De-NS Re-NAc) 7) 2-O- and 6-Odesulfated (2OH, 6OH, NS) 8) 2-O- and 6-O-desulfated, N-desulfated/acetylated (2OH, 6OH, DeNS Re-NAc) saccharides.
The competition assays, in which soluble modified heparins competed with
heparin immobilized on the sensor surface, for the binding of HARP (0.15 µg/ml) (figure
25A), HARPΔ124-136 (0.7 µg/ml) (figure 25B) and CTSR (4 µg/ml) (figure 25C), were
used to explore the roles of distinct sulfate groups of heparin for distinct domains of
HARP.
165
RESULTS
100
90
HARP
A
totally desulfated
80
70
2-O-desulfated,6-O-desulfated
60
6-O-desulfated,N-desulfated
50
2-O-desulfated,N-desulfated
40
N-desulfated
30
2-O-desulfated
20
6-O-desulfated
10
heparin (totally sulfated)
0
0.001
0.01
0.1
1
10
concentration of competing saccharide (µg/ml)
100
HARPΔ124-136
90
80
B
70
totally desulfated
6-O-desulfated,N-desulfated
60
2-O-desulfated,N-desulfated
50
2-O-desulfated,6-O-desulfated
40
N-desulfated
30
20
10
heparin (fully sulfated)
2-O-desulfated
6-O-desulfated
0
0.001
0.01
0.1
1
10
concentration of competing saccharide (µg/ml)
166
RESULTS
100
90
CTSR
C
80
70
6-O-desulfated, N-desulfated
totally desulfated
60
2-O-desulfated,6-O-desulfated
50
2-O-desulfated,N-desulfated
40
30
2-O-desulfated
N-desulfated
20
10
0
0.001
0.01
0.1
1
10
concentration of competing saccharide (µg/ml)
6-O-desulfated
heparin (fully sulfated)
Figure 25: Competition of HARP (A), HARPΔ124-136 (B) and CTSR (C) binding to
immobilized heparin by modified forms of heparin. The extent of binding of soluble ligates to
heparin immobilized on an aminosilane surface was measured in the presence of increasing
concentrations of heparin and its modified forms. Errors for individual datum points generated by
the curve fitting were less than 1% of the mean and were omitted for clarity. Results are expressed
as a percentage of a maximal binding of HARP (25 arc s), of HARPΔ124-136 (20 arc s), of CTSR
(50 arc s) to heparin observed in the absence of competing polysaccharides. Following to Ndesulfation, N-acetyl groups were added to the desulfated backbone (N-desulfated/acetylated). As a
matter of clarity, only N-desulfated is indicated in the figures.
167
RESULTS
The results obtained in four separate experiments were used to calculate the IC50 values,
reported in table 4.
HARP
HARPΔ124-136
CTSR
totally desulfated
-
>10
-
2-O-desulfated, 6-O-desulfated
-
>10
-
6-O-desulfated, N-desulfated
-
0,9
-
2-O-desulfated, N-desulfated
8,1
3
>30
N-desulfated
4,8
0,3
>30
2-O-desulfated
0,875
0,3
>30
6-O-desulfated
0,5
0,5
5,2
0,0875
0,2
0,85
Chemically modified heparins
Heparin (totally sulfated)
Table 4: IC50 values of chemically modified heparins (µg/ml).
Intact heparin induced a 50% inhibition of HARP binding to immobilized heparin
at 0.09 µg/ml. Five-fold higher concentrations of selectively 6-O-desulfated, 10-fold of 2O-desulfated and 50-fold of N-desulfated/re-acetylated were required for a 50% inhibition
as compared with intact heparin. However, the results with the multiple desulfations were,
at first glance counter intuitive in some cases. Thus, an 80-fold higher concentration of 2O-desulfated and N-desulfated/re-acetylated heparin compared to native heparin was
required for 50% inhibition of binding, whereas the additive effect of the component
individual modifications (Table 4) predicts an IC50 around 250-fold higher than that
observed with native heparin, which would not have been detected at the range of
concentrations used. Similarly, totally O-desulfated heparin was observed to not compete
for HARP binding, yet the constituent individual desulfations predict an IC50 only 48-fold
higher than that of heparin, which is detectable with the range of competing sugar used in
these experiments. The absence of competition observed with 6-O-desulfated and Ndesulfated/re-acetylated heparin and totally O-desulfated and N-desulfated/re-acetylated
heparin is consistent with the effects of the individual saccharide modifications. Therefore,
while the single desulfations indicate that glucosamine-N-sulfates in particular and to a
lesser extent glucosamine 6-O-sulfate and iduronic acid 2-O-sulfate groups play a role in
HARP binding, it is clear that the effect of sulfation is more subtle than that suggested by
the simple recognition of charge.
168
RESULTS
As may be seen in table 4 HARPΔ124-136 (Fig. 25B) and CTSR (Fig. 25C)
differed from each other and from HARP (Fig. 25A) in their interaction with selectively
desulfated heparins. The concentration of heparin required to induce a 50% inhibition of
HARPΔ124-136 binding to immobilized heparin was 2-fold lower than for HARP (0.2
µg/ml). The desulfation of 6-O, 2-O and N-sulfate groups slightly shifted the IC50 values
comparing to intact heparin (approximately 2-fold), which were less than observed for
HARP. However, a 15-fold higher concentration of 2-O-desulfated and N-desulfated/reacetylated heparin was required for a 50% inhibition of HARPΔ124-1364 binding.
Interestingly, totally O-desulfated heparin and totally O-desulfated and N-desulfated/reacetylated heparin also could inhibit the binding of HARPΔ124-136, at 250-fold higher
concentrations than IC50 of intact heparin, whereas these polysaccharides failed to
detectably inhibit HARP binding to immobilized heparin. These results suggest that the
structural requirement for the interaction of heparin with HARPΔ124-136 is distinct from
that of heparin-HARP and the binding of HARPΔ124-136 binding to heparin didn’t just
depend on the interaction with sulfate groups of the heparin chains, since it could still bind
to totally desulfated heparins. This, in turn, suggests that the carboxyl groups of heparin
also contribute to the interaction of truncated HARP, whereas the interaction of HARP is
more strongly dependent on the sulfate groups of heparin. The interaction of HARPΔ124136 with heparin does, however, share one important feature with that of HARP: the
inhibition observed with individual desulfations does not predict the effect of multiple
desulfations.
In contrast to HARPΔ124-136, CTSR was generally far more sensitive to the
removal of individual sulfate groups than HARP (Fig. 25C). Thus, whereas the IC50 for 6O-desulfated heparin was 6-fold higher than for heparin, similar to HARP, a greater IC50
(more than 35-fold shift) was observed for 2-O-desulfated; and N-desulfated/re-acetylated
heparins with respect to intact heparin. However, 2-O-desulfated and N-desulfated/reacetylated heparin had a similar ability to compete for CTSR binding as either of its
constituent individual desulfations. The data also indicated that totally O-desulfated
heparin and totally O-desulfated and N-desulfated/re-acetylated heparin didn’t compete for
the binding of CTSR to heparin, hence reflecting the results obtained with HARP.
169
RESULTS
3A.2.3
INTERACTIONS
OF
HARP,
HARPΔ124-136
AND
CTSR
WITH
OLIGOSACCHARIDES OF DIFFERENT LENGTHS
As a second step, we determined the relationship between the length of heparinderived oligosaccharides and their binding capacities. To identify the minimum size of
heparin required for binding, we performed a set of competition assays in which soluble
oligosaccharides of different lengths competed with immobilized heparin for the binding of
HARP, HARPΔ124-136 and CTSR binding (Fig. 26). Heparin-derived oligosaccharides
(DP 2 to DP 16) used for these experiments were prepared from partial heparinase I digests
of pig mucosal heparin were obtained from Iduron (Manchester, UK); the main
disaccharide unit in these saccharides (> 75%) is IdoA, 2S-GlcNSO3.
The results obtained from four separate experiments were indicated in figure 26
and IC50 values were reported in table 5.
100
HARP
90
80
A
70
60
DP4
DP2
DP6
DP8
50
40
30
20
10
0
0.001
0.01
0.1
1
10
concentration of competing oligosaccharide (µg/ml)
DP10
DP12
DP14
DP16
Heparin
170
RESULTS
100
90
HARPΔ124-136
B
DP2
80
DP4
70
DP6
60
DP8
50
DP10
40
DP12
30
DP14
20
DP16
10
Heparin
0
0.001
0.01
0.1
1
10
concentration of competing oligosaccharide (µg/ml)
100
90
80
CTSR
C
DP2,DP4,DP6,DP8
70
60
50
40
DP10
DP12
DP14
30
20
10
DP16
heparin
0
0.1
1
10
concentration of competing saccharide(µg/ml)
Figure 26: Competition of HARP (A), HARPΔ124-136
(B) and CTSR (C) binding to
immobilized heparin by heparin oligosaccharides of different lengths. The extent of binding of
HARP, HARPΔ124-136 and CTSR to heparin immobilized on an aminosilane surface was
measured in the presence of increasing concentrations of oligosaccharides of different lengths. The
amount of soluble ligate bound to the immobilized heparin at each concentration of oligosaccharide
was calculated from the binding curve by non-linear curve fitting using a one-site binding model.
Results are expressed as a percentage of the maximal binding of HARP (0.15 µg/ml) (25 arc s), of
HARPΔ124-136 (0.7 µg/ml) (20 arc s) and of CTSR (4 µg/ml) (50 arc s) to heparin observed in the
absence of competing oligosaccharides.
171
RESULTS
Oligosaccharides of
different lengths
HARP
HARPΔ124-136
CTSR
DP2
-
>50
-
DP4
-
>50
-
DP6
>100
>50
-
DP8
>100
8,5
-
DP10
0,7
0,09
>60
DP12
0,4
0,08
>60
DP14
0,1
0,09
>60
DP16
0,07
0,07
60
heparin
0,3
0,5
1
Table 5: IC50 values of oligosaccharides obtained from enzymatic cleavage of heparin chains.
The concentrations of oligosaccharides (µg/ml) required inhibiting 50% of the binding to heparin
(mean of at least 4 separate experiments).
The results summarized in table 5 indicate a very clear difference of dependence
on polymer size between HARP, HARPΔ124-136 and CTSR. In accordance with the
literature, disaccharides (DP2) and tetrasaccharides (DP4) in solution failed to inhibit the
binding of HARP to the immobilized heparin on the biosensor surface and DP6 and DP8
could inhibit 50% of the binding, though only at concentrations higher than 100 µg/ml.
IC50s decreased from DP8 to DP14, as the number of disaccharide units increased: DP10,
0.7 µg/ml; DP12, 0.4 µg/ml; DP14, 0.1 µg/ml. IC50 values of the longest oligosaccharide,
DP16 and of heparin were similar to that of DP14. These results indicated that DP ≥ 6 all
bind HARP, but that DP6 and DP8 were the least potent. Therefore, a deccasaccharide is
probably required for efficient binding and there was no difference in IC50s for DP14,
DP16 and full-length heparin.
These results were in marked contrast to those obtained with HARPΔ124-136,
where even disaccharides could inhibit the binding of HARPΔ124-136, though at IC50 > 50
µg/ml. IC50s decreased progressively as the oligosaccharides increased in length up to
DP10. Surprisingly, DP10-16 had similar IC50 values (∼0.08 µg/ml) and they were even
more potent than heparin (0.5 µg/ml). However, it should be noted that the inhibition curve
172
RESULTS
for heparin was much steeper and as a consequence, 1 µg/ml heparin was sufficient to
completely abolish HARPΔ124-136 binding, which is similar to what is observed with
DP12-16. CTSR displayed a totally opposite competition profile, such that DP10 could
inhibit 50% inhibition of CTSR binding at 600-fold higher concentration than IC50 of
DP10 for HARP binding to heparin. Furthermore, the IC50 values were not affected by the
increase of length, at least up to DP16 and all the oligosaccharides were much less potent
than heparin (1 µg/ml). These results indicated that CTSR required a sequence longer than
16 disaccharide units for high affinity binding to heparin, whereas the polymer size DP6
was sufficient to compete for HARP lacking its 12 C-terminus amino acids.
3A.2.4 INTERACTIONS OF HARP, HARPΔ124-136 AND CTSR WITH HEPARIN AND ITS
CATION-BOUND DERIVATIVES
To complete the structural definition of heparin in HARP interaction, in a third
and last set binding experiments, we studied the effects of heparin derivatives converted to
the K+, Cu2+, Ca2 and Mn2+ forms.
SRCD spectra and NMR studies reported that binding to K+, Cu2+, Ca2 and Mn2+
ions resulted in unique spectra, hence showing that each was structurally distinct and
binding to K+ and Ca2 ions resulted in subtle changes in conformation and flexibility of
heparin. Furthermore, these structural modifications modified FGF2-FGFR1 signaling
(Rudd et al., 2007).
100
90
A
80
70
60
HARP
(-)cation
Mn
50
40
30
K
Ca
20
10
Cu
0
0.0001
0.001
0.01
0.1
1
concentration of competing saccharide (µg/ml)
173
RESULTS
100
B
90
80
HARPΔ124-136
(-)cation
70
Mn
60
50
K
40
30
Ca
20
Cu
10
0
0.0001
0.001
0.01
0.1
1
concentration of competing saccharide (µg/ml)
100
C
90
CTSR
80
(-)cation
70
60
Mn
50
K
40
30
Cu
20
Ca
10
0
0.0001
0.001
0.01
0.1
1
concentration of competing saccharide (µg/ml)
Figure 27:
Competition of HARP (A), HARPΔ124-136 (B) and CTSR (C) binding to
immobilized heparin by Mn2, K+, Cu2+ and Ca2+-bound heparin derivatives and native
heparin (designated (-) cation).
174
RESULTS
Cation-bound heparin
derivatives
Mn2+
HARP
HARPΔ124-136
CTSR
0,09
0,095
0,8
K+
0,09
0,1
0,7
Cu2+
0,1
0,075
0,8
0,09
0,075
0,6
0,09
0,25
0,8
Ca
2+
Native Heparin
Table 6: IC50s of the cation forms of heparin derivatives
As seen in table 6, IC50 values were not affected by conversion of heparin to
cation forms, for HARP and CTSR binding. However, heparins converted to Cu2+ and Ca+
ion forms could inhibit 50% of HARP1-124 binding at 3-fold lower concentration than
native heparin (-cation) and Mn2+ and K+ ions at 2.5-fold lower concentration.
3A.3 DISCUSSION
HARP is a heparin binding growth factor and many of its important properties
and the regulation of some of the biological activities of the protein are associated with its
interaction with glycosaminoglycans. Therefore, an understanding of this interaction at the
molecular level is of fundamental importance. In this study, we focused on the molecular
basis underlying HARP-heparin interaction.
According to the NMR studies of Kilpaleinen et al. (Kilpelainen et al., 2000),
which examined shifts in the
15
N spectrum of HARP upon heparin binding without
15
rigorous assignment of the N signals, two domains of HARP, homologous to TSR motifs
were suggested to interact with heparin: NTSR-(13-58) and CTSR-(65-110). These NMR
1
H-15N spectra also indicated that the lysine rich tails of the protein didn’t contribute to the
binding. Subsequently, in order to identify the heparin-binding domains of HARP using
direct binding assays, our group synthesized two peptides derived from these heparinbinding consensus sequences. The peptides correspond to the residues NTSR-(13-39) and
CTSR-(65-97). The CTSR peptide bound heparin, but with 10-fold lower affinity (Kd =
120 nM) (Hamma-Kourbali et al., 2007) than HARP (Kd = 13 nM) (Vacherot et al.,
1999a), while the NTSR peptide didn’t bind at all. However, this may be explained by the
observation that the synthetic peptide lacks 2 cysteines of the whole NTSR domain and the
175
RESULTS
importance of the disulfide bonds in heparin binding property of HARP has been already
reported (Kilpelainen et al., 2000). In parallel, we studied several functional sequences of
C-terminal tail that have been identified on the basis of their ability to interact with HARP
receptors and that can act as antagonist in biological assays of HARP activity (chapter one
and two) as well as N-terminus. The results indicated that P111-136 displayed a modest
binding to heparin whereas the synthetic peptides P111-121, P111-124 and P122-131
bound with a very low capacity, suggesting that the sequence 124 to136 near the Cterminus was responsible for the intrinsic heparin-binding activity of the C terminus tail.
There is a body of evidence that indicates that distinct structural requirements are
necessary for the interaction of heparin with proteins. These interactions depend on the
length of polysaccharide chain and on distinct sulfate groups. The minimum HARP
binding sequence in heparin has been previously suggested to be a deccasaccharide, with
the 2-O-sulfate iduronic acid units making a major contribution to the interaction
(Kinnunen et al., 1996). In contrast, the present study demonstrates, using a quantitative
competitive binding assays that of the single desulfations of heparin, the loss of the Nsulfate from glucosamine has by far the greatest effect on the IC50. The IC50 is far less
sensitive to the loss of the 6-O-sulfate from glucosamine and the iduronic acid 2-O-sulfate
group, which has been claimed to be the major structural feature responsible for HARP
binding was found to be the least potent competitor. It is important to note that all the
chemically modified heparins used here have been structurally validated by NMR (Yates et
al., 1996). The competition for HARP binding with multiple desulfated heparins provides
further insights into the specificity of the HARP-sugar interaction. The results indicate that
the effect of multiple desulfations doesn’t reflect the inhibition observed with individual
desulfations and so couldn’t be predicted by simple additive effect of individual
modifications. The simplest explanation for this observation is that the conformation of the
polysaccharide chain plays a major role in the interaction of HARP with heparin, since the
substitution of sulfate at the 2 position of iduronate alters the equilibrium between its
different conformers, in turn altering the position of the carboxylic acid group, whereas
loss of N-sulfate from glucosamine results in the loss of an inter saccharide H-bond
between the N-sulfate and the neighboring uronic acid. The effects of chain substitution on
the conformation of the polysaccharide are complex and have only just begun to be
investigated (Rudd et al., 2007). Full structural definition of the saccharide structures
recognized by HARP will require a substantial effort with respect to probing the solution
conformation of the protein and the sugar.
176
RESULTS
Then, we investigated the dependence on the length of heparin chains for these
interactions. Although HARP could bind the polysaccharides as small as DP6 and DP8, a
decasaccharide (DP10) was required for efficient binding, confirming the previous studies.
In similar ways, we attempted of the structural elucidation of CTSR peptideheparin complex. Our studies indicated that all sulfate groups; especially glucosamine Nand iduronic acid 2-O-sulfate groups were necessary for interaction of CTSR with heparin.
Although these results essentially reflected those obtained with HARP, it is important to
note that the IC50 values observed with CTSR for each modified heparin and also for intact
heparin, were 10-fold or more higher than those with HARP. Furthermore, CTSR showed
a marked contrast in the minimum size requirement with HARP, since CTSR required
much longer oligosaccharide (at least 16 disaccharide units).
Based on assumption that HARP could have more than one heparin interacting
site and such binding sites might be specific for certain sequences on the polysaccharide
chain, in parallel to CTSR, we tested the structural requirements in heparin for the putative
residues of HARP 124-136. However, in this approach, instead of using synthetic peptides,
we used the protein lacking the sequence 124-136, designated HARPΔ124-136. It is
noteworthy to remember that this short form of HARP is a naturally occurring form of the
protein, which has been isolated from glioblastoma (Lu et al., 2005). The interaction of Cterminally truncated HARP with heparin differed from CTSR and also from full-length
HARP in dependence on the size of the polysaccharide chain and on the particular sulfate
groups in heparin. HARPΔ124-136 displayed a relaxed specificity but still exhibits
differences, in particular the double desulfations all show greater IC50s than the sum of
individual desulfations. Furthermore, HARPΔ124-136 could bind even disaccharide units,
but the short oligosaccharides were less potent competitors than the longer ones.
Taken together, these data demonstrate that the heparin binding properties of
HARP are dictated by structural features more complex than the currently believed
thrombospondin homologue heparin-binding sites. The above observations support the
hypothesis that HARP has multiple heparin-binding sites, each contributing to the overall
affinity of the protein for heparin and providing synergistic structural specificities on
heparin chains. An appropriate 3D-structure of HARP is necessary for optimal interaction,
because the separate CTSR domain binds with a lower affinity to heparin and the spatial
orientation of the sequence on protein seems to be the determining factor. Thus, we
propose a model whereby residues in the region of 65-97 form a primary recognition site
for heparin and heparin interacts additionally with C-terminus basic residues 124-136 -or
177
RESULTS
the basic residues in the C-terminal directly interacts with CTSR-, so inducing the optimal
shape of the protein for an efficient binding.
178
RESULTS
3B. HARP-BINDING PROTEINS-NEW IMPLICATIONS FOR HARP FUNCTION
3B.1 INTRODUCTION
Angiogenesis is the process of sprouting of new vessels and pathological
angiogenesis, which can be described as an excessive angiogenesis (neovasculature), is a
hallmark of the cancer. Angiogenesis is governed by the coordinated action of positive and
negative regulators and angiogenic growth factors play an important role in this delicate
balance. The complex interactions exist among these molecules and often a combination of
the angiogenic molecules is required to stimulate functional angiogenesis. Quite a large
number of growth factors have been found to possess angiogenic activity and among them,
VEGF is considered as the most potent angiogenic factor.
HARP is a multifunctional protein that interacts with heparin and with several
signaling molecules. HARP also is an angiogenic factor and contributes to the regulation
of angiogenesis. In a recent work of our group, HARP has been reported to negatively
regulate angiogenesis (Heroult et al., 2004). In this study, HARP inhibited VEGF-induced
angiogenic activity by inhibiting the binding of VEGF to its receptors (KDR and NPN-1,
but not Flt-1). The mechanism proposed in this work was not by binding of HARP to
VEGF receptors, but by direct binding of the molecule to VEGF. The VEGF-interacting
site in HARP has been suggested to be encompassed by the TSR motifs, which also
contain the primary heparin-binding domain of the protein. These experiments employed a
solid-phase ligand binding assay with HARP-coated on ELISA plates and radiolabeled
VEGF. It should be pointed out that the HARP-interacting isoforms of VEGF used in these
assays was VEGF165, one of the heparin-binding isoform of VEGFs, and its binding to
HARP was inhibited by heparin (maximal inhibition at 10 µg/ml), suggesting that the
interaction may be in part due to a heparin interacting site on one of the two partners.
Neuropilin (NPN-1) is a co-receptor for semaphorin-3 family. In the
endothelium, NPN-1 interacts with the heparin-binding isoforms of VEGF (Soker et al.,
1998). However, it doesn’t interact with VEGF121, the isoform lacking the basic heparinbinding module of VEGF165 (Lee et al., 2003). A recent study on NPN-1 has reported that
the molecule could also interact with several proteins other than VEGF, including FGF,
FGFR, HGF/SF and FGF-binding protein. All these proteins have a common property, in
that they are heparin-binding proteins, so suggesting that NPN-1 was a heparin-binding
protein-binding protein. This binding activity would be due to the presence of a heparin
mimetic site on NPN-1 (West et al., 2005).
179
RESULTS
In the first chapter of the thesis, we studied in detail the interaction of HARP with
ALK, which we reported to be another heparin-binding protein. In the first part of this
chapter, we studied the interactions of HARP with heparin. Thus, in this work, we tested
the hypothesis that HARP might interact with other heparin-binding proteins, using an
optical biosensor as a rapid screen for molecular interactions.
3B.2 RESULTS
3B.2.1 HARP-BINDING HEPARIN-BINDING PROTEINS
A biotinylated HARP-derivatized surface was used in the binding assays. Seven
structurally unrelated proteins -VEGF165, VEGFR, NPN-1, FGF2, FGFR1, FGF-BP and
HGF/SF- all known to bind to heparin were tested for their ability to interact with HARP
(Fig. 28). The non-heparin binding isoform of VEGF; VEGF121 was used as a control.
60
50
40
30
HARP-derivatized surface
20
10
0
Figure 28: Interaction of proteins with immobilized HARP determined in an optical
biosensor. Proteins were added as soluble ligates in 2 µl of PBS-T to a HARP-derivatized cuvette
containing 28 µl of PBS-T (see “Materials and Methods”). The association data were analyzed with
a one-site binding model to determine the extent of binding: VEGF165, 33 µg/ml; VEGF121, 16
µg/ml; VEGFR, 33 µg/ml; NPN-1, 3.3 µg/ml; FGF2, 17 µg/ml; FGFR1, 13 µg/ml; FGF-BP, 17
µg/ml and HGF/SF, 10 µg/ml. Results are triplicates of one of at least four separate experiments
carried out on the same HARP-derivatized surface.
Strong binding to immobilized HARP was observed with FGF-BP at 17 µg/ml
and HGF/SF at 10 µg/ml. Lower levels of binding with FGF2 at 17 µg/ml and with NPN-1
at 3.3 µg/ml were observed. It is important to note that the interaction was relatively strong
for NPN-1, considering the low concentration used. VEGF165 bound to HARP, though to a
180
RESULTS
lesser extent than the above proteins, whereas no detectable binding was observed with
VEGF121. The binding of VEGFR1 (33 µg/ml) was also detected. No interaction was
observed with FGFR1 (13 µg/ml). Therefore, six proteins tested; NPN-1, FGF-BP,
HGF/SF, VEGF165, VEGFR1 and FGF2 were shown to bind to immobilized HARP.
The interaction between these six proteins and HARP was dissociated, at least in
part, with 2 M NaCl. The % regeneration of the HARP surface following three 2 M NaCl
washes were as following: 78% for VEGF, 100% for VEGFR1, 33% for NPN-1, 95% for
FGF2, 79% for FGF-BP, 100% for HGF/SF. These observations suggested that the driving
force for the binding was the ionic interactions between HARP and its binding proteins.
We then tested the possibility that the heparin binding sites of these proteins
might be involved in binding to HARP. Consequently, we examined the ability of heparin
to act as a competitor between HARP and these factors (Fig. 29).
100
HARP-derivatized surface
50
0
Figure 29: Competition by heparin for binding of growth factors to immobilized HARP. The
extent of binding of the growth factors to immobilized HARP was calculated by non-linear curve
fitting using a one-site model for binding of VEGF165, VEGFR1, NPN-1, FGF2, FGF-BP and
HGF/SF (protein concentrations are as for Fig. 24) to immobilized HARP (open bars) competed by
heparin (dark shading, 100 µg/ml). Data are representative of at least two experiments.
Heparin (100 µg/ml) inhibited completely the binding of NPN-1, reduced the
binding of HGF/SF to HARP by 96%, of FGF2 by 91%, of VEGF by 86% and that of
FGF-BP by 73% (Figure 29). The binding of VEGFR1 to HARP was competed only 37%
by heparin.
The results of the competition assays by heparin suggested that the heparinbinding sites of HARP or of HARP-binding proteins were involved in their interaction
181
RESULTS
with HARP. However, since the different HARP binding proteins are structurally unrelated
and their only common property is a heparin-binding site, it is most likely that the heparinbinding sites of these proteins are involved in their interaction with HARP. The
observation that the binding of VEGF121, an isoform of VEGF lacking the heparin-binding
domain, to HARP is not detectable, is in support of this argument. We then sought
interacting site(s) in HARP responsible for the interaction with these proteins and
examined whether the heparin-binding sites of the proteins might interact with the largely,
but not exclusively basic C-terminal region of HARP. The rationale for this choice was
that the C-terminus of HARP binds to heparin and this may, therefore present a binding
site for proteins such as NPN-1 that bind heparin-binding proteins via a large acidic patch
on their surface (West et al., 2005). In this way, one might distinguish between a protein
that employs its heparin-binding site to bind to an acidic region of HARP and one that has
an acidic region that can interact with a heparin-binding site in HARP. For these
experiments, we employed biotinylated P111-136-derivatized streptavidin surfaces (Fig.
30).
35
30
25
20
P111-136-derivatized surface
15
10
5
0
Figure 30: Interaction of HARP-binding proteins with immobilized P111-136 determined in
an optical biosensor. Proteins were added as soluble ligates in 2 µl of PBS-T to a HARP Cterminal-(111-136)-derivatized cuvette containing 28 µl of PBS-T (see “Materials and Methods”).
The association data were analyzed with a one-site binding model to determine the extent of
binding: VEGF165, 27 µg/ml; VEGFR1, 27 µg/ml; NPN-1, 7 µg/ml; FGF2, 7 µg/ml; FGF-BP, 33
µg/ml; HGF/SF, 7 µg/ml. Results are triplicates of one of at least four separate experiments carried
out on the two different P111-136-derivatized surfaces.
182
RESULTS
The results suggested two classes of HARP-binding proteins: a group including
VEGF165, VEGFR1 and HGF/SF which didn’t bind at all -or bound very weakly- to the Cterminal tail of HARP; and another group, NPN-1, FGF2 and FGF-BP which showed a
high affinity to the basic region of the protein. The majority of bound NPN-1, FGF2 and
FGF-BP (82%, 56% and 100%) could be regenerated from the P111-136-immobilized
surface by 2 M NaCl washes, showing that these interactions depended on the ionic
strength. While the interactions of FGF-BP and NPN-1 may be ascribed to their possessing
an acidic region that binds the heparin-binding site of proteins, the interaction of FGF2
suggests that the C-terminal tail of HARP may actually possess an acidic patch, even
though it is overwhelmingly basic. Thus, the present data allowed us to conclude that all
the heparin-binding proteins that were observed to interact with HARP may not recognize
a single particular binding site in HARP. While the binding of NPN-1, FGF2 and FGF-BP
was mediated, at least in part by the basic C-terminal of HARP, these proteins may interact
differently with this structure and moreover, the structural requirements of VEGF165,
VEGFR1 and HGF/SF are clearly different and remained to be identified.
3B.2.2 NEW IMPLICATIONS FOR HARP FUNCTION
As HARP interacted with a subset of HBGFs of considerable interest in terms of
tumor angiogenesis, we pursued our investigation of potential molecular partners for
HARP by studying two molecules: S100A4 and osteopontin (OPN).
S100A4 belongs to a family of small, acidic, Ca2+ binding proteins, S100 family.
OPN is an acidic glycoprotein containing arginine-glycine-aspartic acid (RGD) sequence
and a heparin-binding domain. Both proteins are strongly associated with the tumor
progression and promotion of metastases and they occur at higher levels in the primary
human carcinomas (de Silva Rudland et al., 2006).
3B.2.2.1 S100A4
In a first set of experiment, we demonstrated that S100A4 bound to immobilized
HARP and that heparin acted as a competitor for the interaction between S100A4 and
HARP, in a dose-dependent manner (Fig. 31A). As indicated in the figure, the binding of
67 µg/ml of S100 A4 was almost completely inhibited by 1 mg/ml of heparin.
To confirm these results, in a second set of experiments, we added HARP as a
soluble ligate (0.25 µg/ml) to a S100A4-derivatized surface and we obtained a similar
response. Moreover, the 70% of the binding of HARP to S100 A4 could be competed by
183
RESULTS
heparin. To assess the binding site of HARP responsible of the interaction, we tested the
heparin-binding domain of HARP-(65-97) designated CTSR, to the S100A4 immobilized
surface. The results indicated that CTSR mediated the interaction of HARP with S100A4
and the binding of CTSR to S100A4 could be inhibited 80% by 100 µg/ml of heparin (Fig.
31B).
15
A
HARP-derivatized surface
10
**
S100 A4 (67µg/ml)
S100 A4 (67µg/ml)+heparin (0.1mg/ml)
5
**
S100 A4 (67µg/ml)+heparin (1mg/ml)
0
ligate (µg/ml)
100
B
S100 A4-derivatized surface
75
HARP (0.25 µg/ml)
50
25
HARP (0.25 µg/ml) + heparin (100 µg/ml)
***
**
CTSR (4 µg/ml)
CTSR 4 µg/ml) + heparin (100 µg/ml)
0
concentration (µg/ml)
Figure 31: Binding of S100A4 to HARP. A) Binding of S100A4 at 67 µg/ml to immobilized
HARP competed by increasing concentrations of heparin, B) binding of 0.25 µg/ml of HARP and 4
µg/ml of CTSR to immobilized S100A4 competed by 100 µg/ml of heparin.
184
RESULTS
3B.2.2.2 Osteopontin
The ability of HARP to bind OPN was tested on an OPN-derivatized surface.
Similar to the results obtained with S100A4, HARP and its thrombospondin heparin-affin
domain CTSR bound to OPN and the interaction was competed by 100 µg/ml of heparin
(Fig. 32).
100
OPN-derivatized surface
75
HARP 0.5 µg/ml
HARP 0.5 µg/ml+heparin 100 µg/ml
50
CTSR 4 µg/ml
25
***
**
CTSR 4 µg/ml+heparin 100 µg/ml
0
ligate (µg/ml)
Figure 32: Binding of HARP and CTSR to OPN-derivatized surface and competition by
heparin.
185
RESULTS
3B.3 DISCUSSION
In the previous chapters, we studied the structural features in HARP required for
its interactions with its heparin-binding receptor ALK, its co-receptor chondroitin sulfate
proteoglycan RPTPβ and the archetypal member of heparan sulfate proteoglycan, heparin.
In the latter half of this last chapter of the thesis, we were interested in exploring the
interactions of HARP with other proteins, notably other heparin-binding proteins. Thus, we
reported that HARP8 bound a series of structurally unrelated heparin-binding proteins:
FGF2, FGF-BP, HGF/SF, VEGF165, VEGFR1 and NPN-1. These interactions were
characterized by ionic bonding, which are also characteristics for interactions between
proteins and HS and evidenced by the removal of much of the bound protein with 2 M
NaCl. Furthermore, binding of the proteins to HARP was inhibited by heparin. Therefore,
the heparin-binding domains of the proteins would be involved for interacting with HARP.
Further evidence for this was provided by the observation that VEGF121 lacking a heparinbinding domain failed to bind HARP.
In these studies, we also demonstrated that the interactions of FGF2, FGF-BP and
NPN-1 involved the basic region of HARP. Among them, FGF-BP and NPN-1 possess an
acidic region and they can bind the highly basic C-terminus of HARP bind via this region.
Since FGF2 doesn’t have such region, the logical deduction would be that the heparinbinding sites in FGF2 interact with an acidic patch of HARP within the basic C-terminal
tail. Therefore, FGF2, FGF-BP and NPN-1 may interact differently with the C-terminal
structure, and these motifs differ clearly than those recognized by HGF/SF, VEGF165 and
VEGFR1.
8
Because HARP binds heparin, one can speculate that HARP used in these experiments carried heparin upon
the purification and that the interactions observed in this study are due to heparin being associated with
HARP which is immobilized on the sensor surface. We exclude this possibility by providing evidence that:
HARP-bound to the sensor surface is always subjected to 2 M NaCl washes before using it for any binding
assays. If any, there is a contamination with heparin on HARP, usually 2 M NaCl is sufficient to remove any
non-covalent bound polysaccharide from the protein.
186
RESULTS
The primary structure of HARP is characterized by an exceptionally high content of
cationic residues and the theoretical isoelectric point (pI) is 9.6. However, the isoelectric
focusing experiments on polyacrylamide gels reported an overall pI close to the neutrality
(pI = 7.1). This observation suggested the distribution of positively and negatively charged
clusters in 3D-structure of the protein and prompted us to investigate the electrostatic
potential of the individual domains, which seemed to be crucial in the understanding of
these interactions. According to the model structure of CTSR domain of HARP (60-111)
(Fig. 33), acidic and basic clusters exist on the CTSR domain and the acidic residues are
localized at the two extremities.
Figure 33: Molecular modeling of CTSR domain of HARP corresponding to the amino acids
60-111. Electrostatic potential of basic residues were indicated in blue and acidic residues in red.
The modeling employed GRASP (Graphical Representation and Analysis of Structural Properties)
program.
Therefore, the charged residues in HARP might be segregated on the surface, such
that presenting an electropositive patch and an adjacent electronegative face, which may
contribute to binding of basic heparin binding sites of FGF2, HGF/SF, VEGF and
VEGFR1. The binding of NPN-1 may be due to either its recognition of the heparinbinding site of HARP, as found for the interactions of NPN-1 with other proteins (West et
al., 2005) or to the interaction of the heparin-binding site of NPN-1 with an acidic patch on
HARP or both. These considerations also raise the possibility that HARP may be able to
dimerize through electrostatic interactions, though evidence for this is currently lacking.
187
RESULTS
In this study, we discovered two further molecular partners for HARP: S100A4
and osteopontin. The interactions of S100A4 and osteopontin with HARP also depended
on ionic strength and were inhibited by heparin. Moreover, we reported that the
thrombospondin domain related to the C-terminus of HARP, HARP-(65-97) was involved
in both interactions. Both proteins are strongly associated with the tumor progression and
promotion of metastases. OPN is known to signal through binding to integrins at the cell
surface. Although the mechanism of action of S100A4 is not known, it is thought that it
acts in association with other angiogenic factors or even by forming a physical complex
with some of them, to induce a full angiogenic response (Ambartsumian et al., 2001).
Thus, our preliminary studies indicated new potential partners for HARP, but
binding assays reflect only an interaction with a single protein and the most fascinating
aspect of the function of these interactions in the cellular basis remains to elucidate
188
CONCLUSION
SUGGESTIONS FOR FUTURE WORK
CONCLUSION
HARP is a mitogenic, transforming and angiogenic growth factor in addition to
many other biological actions and is implicated in the development and progression of
numerous human cancers. Targeting the expression of the molecule by several strategies in
several tumors of diverse origin resulted in the inhibition of the angiogenesis and
associated tumor growth. In our studies, we focused on prostate cancer for which HARP is
an autocrine growth factor and has been shown to be a rate-limiting factor in the growth
and progression of his cancer. We developed a peptide-based strategy, whereby specific
inhibition of the activation of receptors for HARP was achieved, by designing synthetic
peptides, which correspond to the receptor-interacting domains of the growth factor.
The C-terminal region of HARP may be designated as the ‘molecular recognition
domain’ of the protein. The two HARP receptors -ALK and RPTPβ- bind HARP through
the amino acids 111-136. On this sequence, two functionally and structurally distinct
domains exist: 111-124 and 122-131. Our studies suggest that HARP amino acid residues
111-124 functions as a mitogenesis, angiogenesis and tumor growth-signaling domain of
the protein via the ALK receptor. Experimental evidence from our previous studies also
supports the present data (Bernard-Pierrot et al., 2001). These studies using HARP mutant
proteins indicated that HARP wild type and HARPΔ129-136 mutant proteins exhibit
similar mitogenic activities, whereas HARPΔ111-136 fails to induce mitogenic activity in
NIH 3T3 cells, emphasizing the importance of the HARP residues 111-129. Our present
work goes further, indicating that these activities are encompassed in the sequence 111124. In vitro and in vivo experiments using the peptide P111-124 inhibited all the cancerrelated activities of the protein using human prostatic PC3 cells and the pharmacokinetic
experiments in a nude mice model indicated the peptide as a promising therapeutical tool,
in virtue of its high stability, non-toxicity and its efficient targeting to the tumor.
Furthermore, the preliminary studies showed that the deletion of three lysine residues from
the sequence 111-124 didn’t alter the inhibitory potential of the peptide on mitogenic
activity of the protein.
Another outcome of our studies was that HARP amino acid residues 122-131 are
required to promote cell transformation, adhesion and migration of the another human
prostatic cell line, DU145 cells, in RPTPβ-dependent manner. Firstly, our studies showed
very strong dependence of the HARP response in these cells on RPTPβ and blocking the
receptor resulted in the growth arrest. The very basic peptide P122-131 inhibited the
190
CONCLUSION
anchorage-dependent and -independent growth of these cells, the chemotactic migration
and cell adhesion and in vivo angiogenesis, by binding specifically to this receptor.
Thus, two synthetic peptides derived from two adjacent sequences of the Cterminal of HARP constitute promising tools for treatment of the prostate cancer: one
specifically targeting the ALK receptor and the other RPTPβ.
Collating the data obtained in the present work allows the definition of the
functional separate domains of HARP within the C-terminal tail of HARP may be
summarized as following:
P111-124
ALK-dependent mitogenic, angiogenic and tumor
promoting activities
111
124
136
KLTKPKPQAESKKKKKEGKKQEKMLD
P111-121
mimics P111-124?
122
131
136?
P122-131
RPTPβ-dependent transforming,
migratory and angiogenic activities
Figure 34: The two functional domains of C-terminal mediating biological functions of HARP
via two different HARP receptors. Two distinct domains 111-124 (red) and 122-131 (blue) are
indicated on the figure. The putative 111-121 sequence is shown with purple arrow. The
discontinuous blue arrow indicates the possible implication of the sequence 131-136 in RPTPβdependent activities. The basic residues are shown in bold on the primary structure.
It is a well-known fact that different forms of HARP exist. They can be generated
from enzymatic cleavage of the protein in the cells or by post-translational modifications
following secretion and they result in the C-terminally truncated forms of HARP. Their
biological activities differ in function of the type of cells: while some authors report that
the short forms are not able to stimulate mitogenesis (Souttou et al., 2001) and that the
mitogenesis requires full-length HARP (Bernard-Pierrot et al., 2001), others described Cterminally processed HARP15 to be mitogenic in glioblastomas (Lu et al., 2005).
Combining the data obtained from our studies with optical biosensor with the experimental
191
CONCLUSION
evidence summarized above (figure 34), new insights into the molecular mechanism of
HARP that underlie the activities of HARP are gained. As represented in figure 35,
hypothesis 1 could explain the molecular mechanism used by HARP15 to stimulate the
ALK pathway.
Hypothesis 1
Figure 35: Model for HARP15 signaling via ALK receptor. CTSR motif-65-97 was shown in
black, the sequence 111-124 in purple. HARP15 designates C-terminally processed HARP-(1-124),
which lacks the sequence 124-136. Heparan sulfate chains of HSPG were shown in orange and the
proteoglycan in orange-gray.
In this model, C-terminally processed HARP15 induces ALK-dimerization,
which in turn autophosphorylates. The dimerization of HARP and the role of heparin are
not clear. HARP15, which lacks 12 amino acids from carboxy end, appears to differ in its
structural requirements for binding to heparin. This form of HARP either may not be
dependent on heparin to activate its receptor and may bind ALK without heparin or it binds
heparin differently, but still allows HARP dimerization. Two 1:1 HARP:ALK complexes
may form a symmetrical dimer and the role of heparin may be to reinforce the HARP-ALK
binding through making numerous contacts with both molecules, hence bridging two
complexes. Or, another possibility could be that the transglutaminase which is thought to
dimerize HARP may play a role to induce the dimerization of its receptor.
192
CONCLUSION
Hypothesis 2
Figure 36: Model for HARP18 signaling via ALK and RPTPβreceptors. CTSR motif-65-97
was shown in black, the sequence 111-124 in purple, and the sequence 124-136 in pink. HARP18
designates the full-length HARP-1-136. Autophosphorylation of ALK receptors upon their
dimerization was indicated by the letter P. Heparan sulfate chains of HSPG were shown in orange
and the proteoglycan in orange-gray.
As represented in figure 36, this model describes a dual-receptor system for cell
signaling and HARP18 has an obligate requirement for its co-receptor heparin. Since the
basic amino acid residues in C-terminal are not directly implicated in heparin binding,
heparin induces a conformational change in HARP via CTSR and so monomeric or
dimeric HARP could expose its binding site-111-124 for ALK. There are two possibilities
for action of heparin. From the classical model, heparin, besides inducing a conformational
change in HARP, will cause HARP to form dimers. Hence, the combination of heparininduced conformational change in HARP and the formation of dimer may explain the
activation process of ALK. The other scenario is that the HS chain contains two
recognition sites, one for HARP and one for ALK with the two sites being in close
approximity to facilitate HARP-ALK interaction. This hypothesis is supported by our
findings which indicate that CTSR requires saccharides of a minimum size of 16 sugar
residues and ALK 10 sugar residues (preliminary studies, data not shown). The model in
which heparin links one molecule of HARP and one molecule of ALK is illustrated in
figure 36, on the second ALK molecule in the right side.
193
CONCLUSION
The role of RPTPβ-interacting domain of HARP is unclear. The sequence 124136 may interact with RPTPβ and the protein may signal through receptor inactivation as
proposed by Meng et al. or may regulate ALK-signaling through ALK-RPTPβ pathway as
proposed by Perez-Pinera et al. (Meng et al., 2000, Perez-Pinera et al., 2007). It is
noteworthy that this 10-amino acid basic fragment seems to adopt an alpha-helical
structure in solution (data not shown and the molecular modelization is in progress9), in
contrast to the flexible structure of the rest of C-terminal. The results of DISOPRED
studies also are consistent with the presence of such structure in the sequence of 111-136
(Figure 37). These observations may be adequate to explain the differential signaling of
separate domains within the same C-terminal region.
Figure 37: Predicted secondary structure of P111-136 by PSIPRED studies.
9
in collaboration with Dr Daniel Rigden - Bioinformatic Platform, School of Biological Sciences, University
of Liverpool.
194
CONCLUSION
Considering the complexity of the molecular mechanisms of HARP signaling and
the fine and subtle regulation of its biological activities through ALK/RPTPβ/HS, it is now
clear that the structural basis for the mitogenic activity of this growth factor is still far from
resolved. Moreover, the discovery that HARP interacts with a number of other signaling
molecules, including VEGFR1, suggests that HARP lies within an extremely complex
molecular signaling network. In practical terms, there are issues with the HARP protein in
that HARP produced from bacterial, insect expression systems and yeast, suffers from an
inappropriate folding and even recombinant protein (HARP18) produced in mammalian
expression systems doesn’t satisfy the spatial requirements for an efficient interaction with
ALK, in some cell systems. Thus whilst there has been substantial progress in
understanding the molecular basis for the activities of HARP, it is clear that this relatively
small protein possesses an extremely complex interactome and a substantial effort is
required to define its complexes at the structural level and then to establish in vivo how the
partitioning of HARP between these complexes results in the observed activity in diverse
biological systems.
195
SUGGESTIONS FOR FUTURE WORK
The thesis realized over a period of three years has yielded new informations over the
structure-function of the protein and implies further studies for future researchers. We
propose here some suggestions for the improvement of the present work.
1. The investigation of the sequence of 124-136 seem to be crucial for further
understanding the function of HARP, especially in the pathologies where RPTPβ is
implicated. Particularly, KEG motif in this fragment deserves consideration since It
has been identified as integrin-recognizing motif and is thought to be implicated in
cell adhesion. Furthermore, the fact that GER motifs can be substituted by GEK
(Knight et al., 2000, Zhang et al., 2003) reinforces the hypothesis.
2. Our studies reported the inter-domain interaction of HARP CTSR domain with Cterminal region of HARP. Therefore, it would be interesting to investigate the
possible contribution of the sequence 124-136 in this interaction.
3. We studied extensively HARP and heparin interactions during the thesis and we
defined the thrombospondin repeat type domain of CTSR (65-97) as a primary
recognition site for heparin. However, the studies with the complete N-TSR (13-58)
and C-TSR (60-110) domains of HARP are necessary for the full definition of these
interactions.
4. The tyrosine kinase receptor of HARP, ALK has been identified for the first time as
a heparin-binding protein. Therefore, its heparin-binding properties also need to be
completely studied.
5. In recent years, a body of evidence indicated the regulatory interplay among growth
factor pathways. The present work indicated the interactions of FGF, FGF-BP,
HGF/SF, VEGF165, VEGFR1, NPN-1, S100 A4 and OPN with HARP. However,
the significance of these interactions in cellular systems remained to be elucidated.
196
REFERENCES
REFERENCES
Aaronson S.A. Growth factors and cancer. Science, 1991, 254, 1146-1153.
Adachi Y., Matsubara S., Pedraza C., Ozawa M., Tsutsui J., Takamatsu H., Noguchi H.,
Akiyama T. & Muramatsu T. Midkine as a novel target gene for the Wilm's tumor
suppressor gene (WT1). Oncogene, 1996, 13, 2197-2203.
Adams J.C.a.T., R.P. The thrombospondin type I repeat (TSR) superfamily; diverse protein
with related roles in neuronal development. Dev.Dyn., 2000, 218, 280-299.
Aigner A., Brachmann P., Beyer J., Jager R., Raulais D., Vigny M., Neubauer A.,
Heidenreich A., Weinknecht S., Czubayko F. & Zugmaier G. Marked increase of
the growth factors pleiotrophin and fibroblast growth factor-2 in serum of testicular
cancer patients. Ann Oncol, 2003, 14, 1525-1529.
Akhter S., Ichihara-Tanaka K., Kojima S., Muramatsu H., Inui T., Kimura T., Kaneda N.,
Talukder A.H., Kadomatsu K., Inagaki F. & Muramatsu T. Clusters of basic amino
acids in midkine: roles in neurite-promoting activity and plasminogen activatorenhancing activity. J Biochem (Tokyo), 1998, 123, 1127-1136.
Alberts B., Bray D., Lewis J., Raff M., Roberts K. & Watson J.D. The Molecular Biology
of the Cell. 3rd Ed. New York: Garland, 1994.
Alewood P., Alewood D., Miranda L., Love S. & Meutermans W. Rapid in situ
neutralization protocols for Boc and Fmoc solid-phase chemistries. Methods
Enzymol, 1997, 289, 14-29.
Ambartsumian N., Klingelhofer J., Grigorian M., Christensen C., Kriajevska M.,
Tulchinsky E., Georgiev G., Berezin V., Bock E., Rygaard J., Cao R., Cao Y. &
Lukanidin E. The metastasis-associated Mts1(S100A4) protein could act as an
angiogenic factor. Oncogene, 2001, 20, 4685-4695.
Antoine M., Tag C.G., Wirz W., Borkham-Kamphorst E., Sawitza I., Gressner A.M. &
Kiefer P. Upregulation of pleiotrophin expression in rat hepatic stellate cells by
PDGF and hypoxia: implications for its role in experimental biliary liver
fibrogenesis. Biochem Biophys Res Commun, 2005, 337, 1153-1164.
Badet J. Angiogenic role of growth factors. Contracept Fertil Sex, 1994, 213, 375-382.
Bernard-Pierrot I., Heroult M., Lemaitre G., Barritault D., Courty J. & Milhiet P.E.
Glycosaminoglycans promote HARP/PTN dimerization. Biochem Biophys Res
Commun, 1999, 266, 437-442.
Bernard-Pierrot I., Delbe J., Caruelle D., Barritault D., Courty J. & Milhiet P.E. The
lysine-rich C-terminal tail of heparin affin regulatory peptide is required for
mitogenic and tumor formation activities. J Biol Chem, 2001, 276, 12228-12234.
Bernard-Pierrot I., Delbe J., Rouet V., Vigny M., Kerros M.E., Caruelle D., Raulais D.,
Barritault D., Courty J. & Milhiet P.E. Dominant negative effectors of heparin affin
regulatory peptide (HARP) angiogenic and transforming activities. J Biol Chem,
2002, 277, 32071-32077.
Bernard-Pierrot I., Delbe J., Heroult M., Rosty C., Soulie P., Barritault D., Milhiet P.E. &
Courty J. Heparin affin regulatory peptide in milk: its involvement in mammary
gland homeostasis. Biochem Biophys Res Commun, 2004, 314, 277-282.
Bohlen P., Gautschi-Sova P., Albrecht U., Lehmann S. & Huber D. J Cell Biochem Suppl,
1988, 12A, 221.
Bohlen P. & Gautschi P. European Patent Application, 1989, EP 326,075,
Böhlen P., Muller T., Gautschi-Sova P., Albrecht U., Rasool C.G., Decker M., Seddon A.,
Fafeur V., Kovesdi I. & Kretschmer P. Isolation from bovine brain and structural
characterization of HBNF, a heparin-binding neurotrophic factor. Growth Factors,
1991, 4, 97-107.
Bowden E.T., Stoica G.E. & Wellstein A. Anti-apoptotic signaling of pleiotrophin through
its receptor, anaplastic lymphoma kinase. J Biol Chem, 2002, 277, 35862-35868.
198
REFERENCES
Brigstock D.R., Kim G.Y. & Steffen C.L. Pig uterine luminal fluid contains the
developmentally regulated neurotrophic factor, pleiotrophin. J Endocrinol, 1996,
148, 103-111.
Callebaut C., Jacotot E., Krust B., Guichard G., Blanco J., Valenzuala A., Svab J., Muller
S., Briand J.P. & Hovanessian A.G. Pseudopeptide TASP inhibitors of HIV entry
bind specifically to a 95-kDa cell surface protein. J Biol Chem, 1997, 272, 71597166.
Callebaut C., Blanco J., Benkirane N., Krust B., Jacotot E., Guichard G., Seddiki N., Svab
J., Dam E., Muller S., Briand J.P. & Hovanessian A.G. Identification of V3 Loopbinding proteins as potential receptors implicated in the bining of HIV particles to
CD4+ cells. J Biol Chem, 1998, 279 21988-21997.
Callebaut C., Nisole S., Briand J.P., Krust B. & Hovanessian A.G. Inhibition of HIV
infection by the cytokine midkine. Virology, 2001, 281, 248-264.
Carey D.J., Evans D.M., Stahl R.C., Asundi V.K., Conner K.J., Garbes P. & GizmeciSmith G. Molecular cloning and characterization of N-syndecan, a novel
transmembrane heparan sulphate proteoglycan. J. Cell. Biol., 1992, 117, 191-201.
Carmeliet P. & Collen D. Molecular basis of angiogenesis. Role of VEGF and VEcadherin. Ann N Y Acad Sci, 2000, 902, 249-262; discussion 262-244.
Chao M.V. Dependence Receptors:What is the mechanism? Sci STKE, 2003, 200, 38.
Chauhan A.K., Li Y.S. & Deuel T.F. Pleiotrophin transforms NIH 3T3 cells and induces
tumors in nude mice. Proceedings of the National Academy of Sciences of the
United States of America, 1993, 90, 679-682.
Choudhuri R., Zhang H.T., Donnini S., Ziche M. & Bicknell R. An angiogenic role for the
neurokines midkine and pleiotrophin in tumorigenesis. Cancer Res, 1997, 57,
1814-1819.
Christman K.L., Fang Q., Yee M.S., Johnson K.R., Sievers R.E. & Lee R.J. Enhanced
neovasculature formation in ischemic myocardium following delivery of
pleiotrophin plasmid in a biopolymer. Biomaterials, 2005a, 26, 1139-1144.
Christman K.L., Fang Q., Kim A.J., Sievers R.E., Fok H.H., Candia A.F., Colley K.J.,
Herradon G., Ezquerra L., Deuel T.F. & Lee R.J. Pleiotrophin induces formation of
functional neovasculature in vivo. Biochem Biophys Res Commun, 2005b, 332,
1146-1152.
Chung H.W., Wen Y., Choi E.A., Hao L., Moon H.S., Yu H.K. & Polan M.L. Pleiotrophin
(PTN) and midkine (MK) mRNA expression in eutopic and ectopic endometrium
in advanced stage endometriosis. Mol Hum Reprod, 2002, 8, 350-355.
Connolly J.M. & D.P. R. Angiogenesis in two human prostate cancer cell lines with
differing metastatic potential when growing as solid tumors in nude mice. I Urol,
1998, 160, 932-936.
Corbley M.J. Transformation by ras suppresses expression of the neurotrophic growth
factor pleiotrophin. J Biol Chem, 1997, 272, 24696-24702.
Courty J., Dauchel M.C., Caruelle D., Perderiset M. & Barritault D. Mitogenic properties
of a new endothelial cell growth factor related to pleiotrophin. Biochemical &
Biophysical Research Communications, 1991, 180, 145-151.
Czubayko F., Riegel A.T. & Wellstein A. Ribozyme-targeting elucidates a direct role of
pleiotrophin in tumor growth. J Biol Chem, 1994, 269, 21358-21363.
Czubayko F., Schulte A.M., Berchem G.J. & Wellstein A. Melanoma angiogenesis and
metastasis modulated by ribozyme targeting of the secreted growth factor
pleiotrophin. Proceedings of the National Academy of Sciences of the United States
of America, 1996, 93, 14753-14758.
199
REFERENCES
De Silva Rudland S., Martin L., Roshanlall C., Winstanley J., Leinster S., Platt-Higgins A.,
Carroll J., West C., Barraclough R., Rudland P. Association of S100A4 and
osteopontin with specific prognostic factors and survival of patients with minimally
invasive breast cancer. Clin Cancer Res, 2006, 15, 1192-1200.
Deepa S.S., Yamada S., Zako M., Goldberger O. & Sugahara K. Chondroitin sulfate chains
on syndecan-1 and syndecan-4 from normal murine mammary gland epithelial cells
are structurally and functionally distinct and cooperate with heparan sulfate chains
to bind growth factors. A novel function to control binding of midkine,
pleiotrophin, and basic fibroblast growth factor. J Biol Chem, 2004, 279, 3736837376.
Delbe J., Vacherot F., Laaroubi K., Barritault D. & Courty J. Effect of heparin on bovine
epithelial lens cell proliferation induced by heparin affin regulatory peptide. J Cell
Physiol, 1995, 164, 47-54.
Delehedde M., Lyon M., Sergeant N., Rahmoune H. & Fernig D.G. Proteoglycans:
pericellular and cell surface multireceptors that integrate external stimuli in the
mammary gland. Journal of Mammary Gland Biology and Neoplasia, 2001, 6, 253273.
Delehedde M., Sergeant N., Lyon M., Rudland P.S. & Fernig D.G. Hepatocyte growth
factor/scatter factor stimulates migration of rat mammary fibroblasts through both
mitogen-activated protein kinase and phosphatidylinositol 3-kinase/Akt pathways.
Eur J Biochem, 2002, 268, 4423-4429.
Delehedde M., Lyon M., Gallagher J.T., Rudland P.S. & Fernig D.G. Fibroblast growth
factor-2 binds to small heparin-derived oligosaccharides and stimulates a sustained
phosphorylation of p42/44 mitogen-activated protein kinase and proliferation of rat
mammary fibroblasts. J Biochem, 2002, 366, 235-244.
Deuel T.F. Polypeptide Growth Factors: Roles in normal and abnormal cell growth. Ann
Rev Cell Biol, 1987, 3, 443-492.
Deuel T.F., Zhang N., Yeh H.-J., Silos-Santiago I. & Wang Z.-Y. Pleiotrophin: a cytokine
with diverse functions and a novel signaling pathway. Arch Biochem Biophys,
2002, 397, 162-171.
Dirks W.G., Fahnrich S., Lis Y., Becker E., MacLeod R.A. & Drexler H.G. Expression and
functional analysis of the anaplastic lymphoma kinase (ALK) gene in tumor cell
lines. Int J Cancer, 2002, 100, 49-56.
Dong Z. & Cmarik J.L. Harvesting cells under anchorage-independent cell transformation
conditions for biochemical analyses. Sci STKE, 2002, 130, pl7.
Duchesne L., Tissot B., Rudd T.R., Dell A. & Fernig D.G. N-glycosylation of fibroblast
growth factor receptor 1 regulates ligand and heparan sulfate co-receptor binding. J
Biol Chem, 2006, 281, 27178-27189.
Dulbecco R. & Stoker M.G. Conditions determining initiation of DNA synthesis in 3T3
cells. Proc Natl Acad Sci U S A, 1970, 66, 204-210.
Dunker A.K. & Obradovic Z. The protein trinity-linking function and disorder. Nature
Biotechnol., 2001, 19, 805-806.
Dunker A.K., Brown C.J., Lawson J.D., Iakoucheva L.M. & Obradovic Z. Intrinsic
disorder and protein function. Biochemistry, 2002, 41, 6573-6582.
Englund C., Loren C.E., Grabbe C., Varshney G.K., Deleuil F., Hallberg B. & Palmer R.H.
Jeb signals through the Alk receptor tyrosine kinase to drive visceral muscle fusion.
Nature, 2003, 425, 512-516.
Englund C., Birve A., Falileeva L., Grabbe C. & Palmer R.H. Miple1 and miple2 encode a
family of MK/PTN homologues in Drosophila melanogaster. Dev Genes Evol,
2005, 216, 10-18.
200
REFERENCES
Ezquerra L., Herradon G., Nguyen T., Vogt T.F., Bronson R., Silos-Santiago I. & Deuel
T.F. Pleiotrophin is a major regulator of the catecholamine biosynthesis pathway in
mouse aorta. Biochem Biophys Res Commun, 2004, 323, 512-517.
Fabri L., Nice E.C., Ward L.D., Maruta H., Burgess A.W. & Simpson R.J.
Characterization of bovine heparin-binding neurotrophic factor (HBNF):
assignment of disulfide bonds. Biochemistry International, 1992, 28, 1-9.
Fabri L., Maruta H., Muramatsu H., Muramatsu T., Simpson R.J., Burgess A.W. & Nice
E.C. Structural characterisation of native and recombinant forms of the
neurotrophic cytokine MK. J Chromatogr, 1993, 646, 213-225.
Fan Q.W., Muramatsu T. & Kadomatsu K. Distinct expression of midkine and pleiotrophin
in the spinal cord and placental tissues during early mouse development. Dev
Growth Differ, 2000, 42, 113-119.
Fang W., Hartmann N., Chow D.T., Riegel A.T. & Wellstein A. Pleiotrophin stimulates
fibroblasts and endothelial and epithelial cells and is expressed in human cancer.
Journal of Biological Chemistry, 1992, 267, 25889-25897.
Fath M., VanderNoot V., Kilpelainen I., Kinnunen T., Rauvala H. & Linhardt R.J.
Interaction of soluble and surface-bound heparin binding growth-associated
molecule with heparin. FEBS Lett, 1999, 454, 105-108.
Fernig D.G. Optical biosensor techniques to analyze protein-polysaccharide interactions.
Methods Mol Biol, 2001, 171, 505-518.
Foehr E.D., Lorente G., Kuo J., Ram R., Nikolich K. & Urfer R. Targeting of the receptor
protein tyrosine phosphatase beta with a monoclonal antibody delays tumor growth
in a glioblastoma model. Cancer Res, 2006, 66, 2271-2278.
Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med, 1971, 285, 11821186.
Fukada M., Fujikawa A., Chow J.P., Ikematsu S., Sakuma S. & Noda M. Protein tyrosine
phosphatase receptor type Z is inactivated by ligand-induced oligomerization.
FEBS Lett, 2006, 580, 4051-4056.
Gallagher J.T. & Walker A. Molecular distinctions between heparan sulphate and heparin.
Analysis of sulphation patterns indicates that heparan sulphate and heparin are
separate families of N-sulphated polysaccharides. Biochem J, 1985, 230, 665-674.
Gallagher J.T., Lyon M. & Steward W.P. Structure and function of heparan sulphate
proteoglycans. Biochem J, 1986, 236, 313-325.
Garver R.I., Jr., Radford D.M., Donis-Keller H., Wick M.R. & Milner P.G. Midkine and
pleiotrophin expression in normal and malignant breast tissue. Cancer, 1994, 74,
1584-1590.
Gospodarowicz D., Neufeld G. & Schweigerer L. Fibroblast growth factor: structural and
biological properties. J Cell Physiol Suppl, 1987, Suppl 5, 15-26.
Goustin A.S., Leof E.B., Shipley G.D. & Moses H.L. Growth factors and cancer. Cancer
Res, 1986, 46, 1015-1029.
Gray G.R. Preparation of antigens by coupling carbohydrates to proteins by reductive
amination with cyanoborohydride. Methods Enzymol, 1978, 50, 155-160.
Grzelinski M., Bader N., Czubayko F. & Aigner A. Ribozyme-targeting reveals the ratelimiting role of pleiotrophin in glioblastoma. Int J Cancer, 2005, 117, 942-951.
Guo N.H., Krutzsch H.C., Negre E., Zabrenetzky V.S. & Roberts D.D. Heparin-binding
peptides from the type I repeats of thrombospondin. Structural requirements for
heparin binding and promotion of melanoma cell adhesion and chemotaxis. Journal
of Biological Chemistry, 1992, 267, 19349-19355.
Hamma-Kourbali Y., Bernard-Pierrot I., Heroult M., Dalle S., Caruelle D., Milhiet P.E.,
Fernig D.G., Delbe J. & Courty J. Inhibition of the mitogenic, angiogenic and
201
REFERENCES
tumorigenic activities of pleiotrophin by a synthetic peptide corresponding to its Cthrombospondin repeat-I domain. J Cell Physiol, 2007,
Hampton B.S., Marshak D.R. & Burgess W.H. Structural and functional characterization
of full-length heparin-binding growth associated molecule. Molecular Biology of
the Cell, 1992, 3, 85-93.
Hatziapostolou M., Delbe J., Katsoris P., Polytarchou C., Courty J. & Papadimitriou E.
Heparin affin regulatory peptide is a key player in prostate cancer cell growth and
angiogenicity. Prostate, 2005, 65, 151-158.
Hatziapostolou M., Polytarchou C., Katsoris P., Courty J. & Papadimitriou E. Heparin
affin regulatory peptide/pleiotrophin mediates fibroblast growth factor 2
stimulatory effects on human prostate cancer cells. J Biol Chem, 2006, 281, 3221732226.
Heroult M., Bernard-Pierrot I., Delbe J., Hamma-Kourbali Y., Katsoris P., Barritault D.,
Papadimitriou E., Plouet J. & Courty J. Heparin affin regulatory peptide binds to
vascular endothelial growth factor (VEGF) and inhibits VEGF-induced
angiogenesis. Oncogene, 2004, 23, 1745-1753.
Herradon G., Ezquerra L., Nguyen T., Vogt T.F., Bronson R., Silos-Santiago I. & Deuel
T.F. Pleiotrophin is an important regulator of the renin-angiotensin system in
mouse aorta. Biochem Biophys Res Commun, 2004, 324, 1041-1047.
Herradon G., Ezquerra L., Nguyen T., Silos-Santiago I. & Deuel T.F. Midkine regulates
pleiotrophin organ-specific gene expression: evidence for transcriptional regulation
and functional redundancy within the pleiotrophin/midkine developmental gene
family. Biochem Biophys Res Commun, 2005, 333, 714-721.
Hienola A., Pekkanen M., Raulo E., Vanttola P. & Rauvala H. HB-GAM inhibits
proliferation and enhances differentiation of neural stem cells. Mol Cell Neurosci,
2004, 26, 75-88.
Horiba M., Kadomatsu K., Nakamura E., Muramatsu H., Ikematsu S., Sakuma S., Hayashi
K., Yuzawa Y., Matsuo S., Kuzuya M., Kaname T., Hirai M., Saito H. &
Muramatsu T. Neointima formation in a restenosis model is suppressed in midkinedeficient mice. J Clin Invest., 2000, 105, 489-495.
Huber D., Gautschi-Sova P. & Bohlen P. Amino-terminal sequences of a novel heparinbinding protein from human, bovine, rat, and chick brain: high interspecies
homology. Neurochem Res, 1990, 15, 435-439.
Hulmes J.D., Seddon A.P., Decker M.M. & Bohlen P. Comparison of the disulfide bond
arrangements of human recombinant and bovine brain heparin binding neuritepromoting factors. Biochem Biophys Res Commun, 1993, 192, 738-746.
Iakoucheva L.M., Brown C.J., Lawson J.D., Obradovic Z. & Dunker A.K. Intrinsic
disorder in cell-signaling and cancer-associated proteins. J Mol Biol, 2002, 323,
573-584.
Imai S., Kaksonen M., Raulo E., Kinnunen T., Fages C., Meng X.J., Lakso M. & Rauvala
H. Osteoblast recruitment and bone formation enhanced by cell matrix-associated
heparin-binding growth-associated molecule (HB-GAM). J Cell Biol, 1998, 143,
1113-1128.
Iwasaki W., Nagata K., Hatanaka H., Inui T., Kimura T., Muramatsu T., Yoshida K.,
Tasumi M. & Inagaki F. Solution structure of midkine, a new heparin-binding
growth factor. Embo J, 1997, 16, 6936-6946.
Jacobson M. Developmental Neurobiology 1978, Plenum, New York,
Jager R., Noll K., Havemann K., Knabbe C., Rauvala H. & Zugmaier G. Differential
expression and biological activity of the heparin-binding growth-associated
molecule (HB-GAM) in lung cancer cell lines. Int J Cancer, 1997, 73, 537-543.
202
REFERENCES
Jager R., List B., Knabbe C., Souttou B., Raulais D., Zeiler T., Wellstein A., Aigner A.,
Neubauer A. & Zugmaier G. Serum levels of the angiogenic factor pleiotrophin in
relation to disease stage in lung cancer patients. Br J Cancer, 2002, 86, 858-863.
Jung C.G., Hida H., Nakahira K., Ikenaka K., Kim H.J. & Nishino H. Pleiotrophin mRNA
is highly expressed in neural stem (progenitor) cells of mouse ventral
mesencephalon and the product promotes production of dopaminergic neurons
from ebryonic stem cell-derived nestin-positive cells. The FASEB Journal, 2004,
18, 1237-1239.
Kadomatsu K., Tomomura M. & Muramatsu T. cDNA cloning and sequencing of a new
gene intensely expressed in early differentiation stages of embryonal carcinoma
cells and in mid- gestation period of mouse embryogenesis. Biochem Biophys Res
Commun, 1988, 151, 1312-1318.
Kadomatsu K., Hagihara M., Akhter S., Fan Q.W., Muramatsu H. & Muramatsu T.
Midkine induces the transformation of NIH3T3 cells. Br J of Cancer, 1997, 75,
354-359.
Kaname T., Kuwano A., Murano I., Uehara K., Muramatsu T. & Kajii T. Midkine gene
(MDK), a gene for prenatal differentiation and neuroregulation, maps to band
11p11.2 by fluorescence in situ hybridization. Genomics, 1993, 17, 514-515.
Kaneda N., Talukder A.H., Nishiyama H., Koizumi S. & Muramatsu T. Midkine, a
heparin-binding growth/differentiation factor, exhibits nerve cell adhesion and
guidance activity for neurite outgrowth in vitro. J Biochem (Tokyo), 1996, 119,
1150-1156.
Ke Y., Wilkinson M.C., Fernig D.G., Smith J.A. & Rudland P.S. A rapid procedure for
production of human basic fibroblast growth factor in E. coli cells. Biochim
Biophys Acta, 1992, 1131, 307-310.
Kilpelainen I., Kaksonen M., Avikainen H., Fath M., Linhardt R.J., Raulo E. & Rauvala H.
Heparin-binding growth-associated molecule contains two heparin-binding beta sheet domains that are homologous to the thrombospondin type I repeat. J Biol
Chem, 2000, 275, 13564-13570.
Kinnunen T., Raulo E., Nolo R., Maccarana M., Lindahl U. & Rauvala H. Neurite
outgrowth in brain neurons induced by heparin-binding growth-associated
molecule (HB-GAM) depends on the specific interaction of HB-GAM with
heparan sulfate at the cell surface. J Biol Chem, 1996, 271, 2243-2248.
Kinnunen T., Kaksonen M., Saarinen J., Kalkkinen N., Peng H.B. & Rauvala H. CortactinSrc kinase signaling pathway is involved in N-syndecan- dependent neurite
outgrowth. J Biol Chem, 1998, 273, 10702-10708.
Kinnunen A., Niemi M., Kinnunen T., Kaksonen M., Nolo R. & Rauvala H. Heparan
sulphate and HB-GAM (heparin-binding growth-associated molecule) in the
development of the thalamocortical pathway of rat brain [In Process Citation]. Eur
J Neurosci, 1999, 11, 491-502.
Kjellen L. & Lindahl U. Proteoglycans: Structures and interactions. Ann Rev Biochem,
1991, 60, 443-475.
Klagsbrun M. The affinity of fibroblast growth factors (FGFs) for heparin; FGF-heparan
sulfate interactions in cells and extracellular matrix. Curr Opin Cell Biol, 1990, 2,
857-863.
Klomp H.J., Zernial O., Flachmann S., Wellstein A. & Juhl H. Significance of the
expression of the growth factor pleiotrophin in pancreatic cancer patients. Clin
Cancer Res, 2002, 8, 823-827.
Kojima S., Muramatsu H., Amanuma H. & Muramatsu T. Midkine enhances fibrinolytic
activity of bovine endothelial cells. J Biol Chem, 1995a, 270, 9590-9596.
203
REFERENCES
Kojima S., Inui T., Kimura T., Sakakibara S., Muramatsu H., Amanuma H., Maruta H. &
Muramatsu T. Synthetic peptides derived from midkine enhance plasminogen
activator activity in bovine aortic endothelial cells. Biochemical & Biophysical
Research Communications, 1995b, 206, 468-473.
Kojima T., Katsumi A., Yamazaki T., Muramatsu T., Nagasaka T., Ohsumi K. & Saito H.
Human ryudocan from endothelium-like cells binds basic fibroblast growth factor,
midkine, and tissue factor pathway inhibitor. J Biol Chem, 1996, 271, 5914-5920.
Kojima S., Inui T., Muramatsu H., Suzuki Y., Kadomatsu K., Yoshizawa M., Hirose S.,
Kimura T., Sakakibara S. & Muramatsu T. Dimerization of midkine by tissue
transglutaminase and its functional implication. J Biol Chem, 1997, 272, 94109416.
Kovesdi I., Fairhurst J.L., Kretschmer P.J. & Bohlen P. Heparin-binding neurotrophic
factor (HBNF) and MK, members of a new family of homologous,
developmentally regulated proteins. Biochemical & Biophysical Research
Communications, 1990, 172, 850-854.
Kretschmer P.J., Fairhurst J.L., Decker M.M., Chan C.P., Gluzman Y., Bohlen P. &
Kovesdi I. Cloning, characterization and developmental regulation of two members
of a novel human gene family of neurite outgrowth-promoting proteins. Growth
Factors, 1991, 5, 99-114.
Kretschmer P.J., Fairhurst J.L., Hulmes J.D., Popjes M.L., Bohlen P. & Kovesdi I.
Genomic organization of the human HBNF gene and characterization of an HBNF
variant protein as a splice mutant. Biochemical & Biophysical Research
Communications, 1993, 192, 420-429.
Kuo M.D., Oda Y., Huang J.S. & Huang S.S. Amino acid sequence and characterization of
a heparin-binding neurite- promoting factor (p18) from bovine brain. J Biol Chem,
1990, 265, 18749-18752.
Kuo M.D., Huang S.S. & Huang J.S. Characterization of heparin-binding growthassociated factor receptor on NIH 3T3 cells. Biochemical & Biophysical Research
Communications, 1992, 182, 188-194.
Kurtz A., Schulte A.M. & Wellstein A. Pleiotrophin and midkine in normal development
and tumor biology. Crit Rev Oncog, 1995, 6, 151-177.
Laaroubi K., Delbe J., Vacherot F., Desgranges P., Tardieu M., Jaye M., Barritault D. &
Courty J. Mitogenic and in vitro angiogenic activity of human recombinant heparin
affin regulatory peptide. Growth Factors, 1994, 10, 89-98.
Laemmli U.K. Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature, 1970, 227, 680-685.
Lai S., Czubayko F., Riegel A.T. & Wellstein A. Structure of the human heparin-binding
growth factor gene pleiotrophin. Biochem Biophys Res Commun, 1992, 187, 11131122.
Lametsch R., Rasmussen J.T., Johnsen L.B., Purup S., Sejrsen K., Petersen T.E. &
Heegaard C.W. Structural characterization of the fibroblast growth factor-binding
protein purified from bovine prepartum mammary gland secretion. J Biol Chem,
2000, 275, 19469-19474.
Lander A.D. Proteoglycans: master regulators of molecular encounter? Matrix Biol, 1998,
17, 465-472.
Lauri S.E., Rauvala H., Kaila K. & Taira T. Effect of heparin-binding growth-associated
molecule (HB-GAM) on synaptic transmission and early LTP in rat hippocampal
slices. Eur J Neurosci, 1998, 10, 188-194.
Ledoux D., Caruelle D., Sabourin C., Liu J., Crepin M., Barritault D. & Courty J. Cellular
distribution of the angiogenic factor heparin affin regulatory peptide (HARP)
204
REFERENCES
mRNA and protein in the human mammary gland. J. Histochem. Cytochem., 1997,
45, 1-7.
Lee C.C., Kreusch A., McMullan D., Ng K. & Spraggon G. Crystal structure of the human
neuropilin-1 b1 domain. Structure, 2003, 11, 99-108.
Lee H.H., Norris A., Weiss J.B. & Frasch M. Jelly belly protein activates the receptor
tyrosine kinase Alk to specify visceral muscle pioneers. Nature, 2003, 425, 507512.
Levy J.B., Canoll P.D., Silvennoinen O., Barnea G., Morse B., Honegger A.M., Huang
J.T., Cannizzaro L.A., Park S.H., Druck T. & et al. The cloning of a receptor-type
protein tyrosine phosphatase expressed in the central nervous system. J Biol Chem,
1993, 268, 10573-10581.
Li Y.S., Milner P.G., Chauhan A.K., Watson M.A., Hoffman R.M., Kodner C.M.,
Milbrandt J. & Deuel T.F. Cloning and expression of a developmentally regulated
protein that induces mitogenic and neurite outgrowth activity. Science, 1990, 250,
1690-1694.
Li Y.S., Hoffman R.M., Le Beau M.M., Espinosa R.d., Jenkins N.A., Gilbert D.J.,
Copeland N.G. & Deuel T.F. Characterization of the human pleiotrophin gene.
Promoter region and chromosomal localization. Journal of Biological Chemistry,
1992a, 267, 26011-26016.
Li Y.S., Gurrieri M. & Deuel T.F. Pleiotrophin gene expression is highly restricted and is
regulated by platelet-derived growth factor. Biochemical & Biophysical Research
Communications, 1992b, 184, 427-432.
Li Y.S. & Deuel T.F. Pleiotrophin stimulates tyrosine phosphorylation in NIH 3T3 and
NB41A3 cells. Biochemical & Biophysical Research Communications, 1993, 195,
1089-1095.
Li G., Hu Y., Huo Y., Liu M., Freeman D., Gao J., Liu X. & Wu D. PTEN deletion leads
to up-reguation of a secreted growth factor pleiotrophin. Journal of Biological
Chemistry, 2006, 16, 10663-10668.
Liedert A., Kaspar D., Claes L. & Ignatius A. Signal transduction pathways involved in
mechanical regulation of HB-GAM expression in osteoblastic cells. Biochem
Biophys Res Commun, 2006, 342, 1070-1076.
Lorente G., Nelson A., Mueller S., Kuo J., Urfer R., Nikolich K. & Foehr E.D. Functional
comparison of long and short splice forms of RPTPbeta: implications for
glioblastoma treatment. Neuro Oncol, 2005, 7, 154-163.
Lu K.V., Jong K.A., Kim G.Y., Singh J., Dia E.Q., Yoshimoto K., Wang M.Y., Cloughesy
T.F., Nelson S.F. & Mischel P.S. Differential induction of glioblastoma migration
and growth by two forms of pleiotrophin. J Biol Chem, 2005, 280, 26953-26964.
Lyon M. and Gallagher J.T. Heparan Sulphate: Molecular structure and interaction with
growth factors and morphogens. In R. V. Iozzo (ed.) Proteoglycans: Structure,
Biology and Molecular Interactions, 2000, Marcel Dekker, New York, 27-60.
Mac Gabhann F. & Popel A.S. Differential binding of VEGF isoforms to VEGF receptor 2
in the presence of neuropilin-1: A computational model. Am J Physiol Heart Circ
Physiol, 2005, 288, 2851–2860.
Maeda N., Nishiwaki T., Shintani T., Hamanaka H. & Noda M.
6B4
proteoglycan/phosphacan, an extracellular variant of receptor-like protein-tyrosine
phosphatase zeta/RPTPbeta, binds pleiotrophin/heparin-binding growth-associated
molecule (HB-GAM). Journal of Biological Chemistry, 1996, 271, 21446-21452.
Maeda N.and.Noda M. Involvement of receptor-like protein tyrosine phosphatase
zeta/RPTP beta and its ligand pleiotrophin/heparin-binding growth-associated
molecule (HB-GAM) in neuronal migration. J. Cell. Biol., 1998, 142, 203-216.
205
REFERENCES
Maeda N., Ichihara-Tanaka K., Kimura T., Kadomatsu K., Muramatsu T. & Noda M. A
receptor-like protein-tyrosine phosphatase PTPzeta/RPTPbeta binds a heparinbinding growth factor midkine. Involvement of arginine 78 of midkine in the high
affinity binding to PTPzeta. J Biol Chem, 1999, 274, 12474-12479.
Maeda N., He J., Yajima Y., Mikami T., Sugahara K. & Yabe T. Heterogeneity of the
chondroitin sulfate portion of phosphacan/6B4 proteoglycan regulates its binding
affinity for pleiotrophin/heparin binding growth-associated molecule. J Biol Chem,
2003, 278, 35805-35811.
Maeda N., Fukazawa N. & Hata T. The binding of chondroitin sulfate to
pleiotrophin/heparin-binding growth-associated molecule is regulated by chain
length and oversulfated structures. J Biol Chem, 2006, 281, 4894-4902.
Mailleux P., Vanderwinden J.M. & Vanderhaeghen J.J. The new growth factor
pleiotrophin (HB-GAM) mRNA is selectively present in the meningothelial cells of
human meningiomas. Neurosci Lett, 1992, 142, 31-35.
Matsumoto K., Wanaka A., Takatsuji K., Muramatsu H., Muramatsu T. & Tohyama M. A
novel family of heparin-binding growth factors, pleiotrophin and midkine, is
expressed in the developing rat cerebral cortex. Brain Res Dev Brain Res, 1994, 79,
229-241.
Meng K., Rodriguez-Pena A., Dimitrov T., Chen W., Yamin M., Noda M. & Deuel T.F.
Pleiotrophin signals increased tyrosine phosphorylation of beta beta-catenin
through inactivation of the intrinsic catalytic activity of the receptor-type protein
tyrosine phosphatase beta/zeta. Proc Natl Acad Sci U S A, 2000, 97, 2603-2608.
Merenmies J. and Rauvala H. Molecular cloning of the 18-kDa growth-associated protein
of developing brain. Journal of Biological Chemistry, 1990, 265, 16721-16724.
Merrifield R.B. Solid Phase Peptide Synthesis. I. The synthesis of a tetrapeptide. J Am
Chem Soc, 1963, 85, 2149-2154.
Michikawa M., Kikuchi S., Muramatsu H., Muramatsu T. & Kim S.U. Retinoic acid
responsive gene product, midkine, has neurotrophic functions for mouse spinal
cord and dorsal root ganglion neurons in culture. J Neurosci Res, 1993b, 35, 530539.
Michikawa M., Xu R.Y., Muramatsu H., Muramatsu T. & Kim S.U. Midkine is a mediator
of retinoic acid induced neuronal differentiation of embryonal carconoma cells.
Biochem Biophys Res Commun, 1993a, 192, 1312-1318.
Milev P., Chiba A., Haring M., Rauvala H., Schachner M., Ranscht B., Margolis R.K. &
Margolis R.U. High affinity binding and overlapping localization of neurocan and
phosphacan/protein-tyrosine phosphatase-zeta/beta with tenascin-R, amphoterin,
and the heparin-binding growth-associated molecule. J Biol Chem, 1998, 273,
6998-7005.
Milhiet P.E., Vacherot F., Caruelle J.P., Barritault D., Caruelle D. & Courty J.
Upregulation of the angiogenic factor heparin affin regulatory peptide by
progesterone in rat uterus. J Endocrinol, 1998, 158, 389-399.
Milner P.G., Li Y.S., Hoffman R.M., Kodner C.M., Siegel N.R. & Deuel T.F. A novel 17
kD heparin-binding growth factor (HBGF-8) in bovine uterus: purification and Nterminal amino acid sequence. Biochemical & Biophysical Research
Communications, 1989, 165, 1096-1103.
Milner P.G., Shah D., Veile R., Donis-Keller H. & Kumar B.V. Cloning, nucleotide
sequence, and chromosome localization of the human pleiotrophin gene.
Biochemistry, 1992, 31, 12023-12028.
Mitsiadis T.A., Salmivirta M., Muramatsu T., Muramatsu H., Rauvala H., Lehtonen E.,
Jalkanen M. & Thesleff I. Expression of the heparin-binding cytokines, midkine
206
REFERENCES
(MK) and HB-GAM (pleiotrophin) is associated with epithelial-mesenchymal
interactions during fetal development and organogenesis. Development, 1995, 121,
37-51.
Miyake I., Hakomori Y., Shinohara A., Gamou T., Saito M., Iwamatsu A. & Sakai R.
Activation of anaplastic lymphoma kinase is responsible for hyperphosphorylation
of ShcC in neuroblastoma cell lines. Oncogene, 2002, 21, 5823-5834.
Moog-Lutz C., Degoutin J., Gouzi J.Y., Frobert Y., Brunet-de Carvalho N., Bureau J.,
Creminon C. & Vigny M. Activation and inhibition of anaplastic lymphoma kinase
receptor tyrosine kinase by monoclonal antibodies and absence of agonist activity
of pleiotrophin. J Biol Chem, 2005, 280, 26039-26048.
Morris S.W., Naeve C., Mathew P., James P.L., Kirstein M.N., Cui X. & Witte D.P. ALK,
the chromosome 2 gene locus altered by the t(2;5) in non-Hodgkin's lymphoma,
encodes a novel neural receptor tyrosine kinase that is highly related to leukocyte
tyrosine kinase (LTK). Oncogene, 1997, 14, 2175-2188.
Motegi A., Fujimoto J., Kotani M., Sakuraba H. & Yamamoto T. ALK receptor tyrosine
kinase promotes cell growth and neurite outgrowth. J Cell Sci, 2004, 117, 33193329.
Muller S., Kunkel P., Lamszus K., Ulbricht U., Lorente G., Nelson A., von Schack D.,
Chin D.J., Lohr S.C., Westphal M. & Melcher T. A role for receptor tyrosine
phosphatase zeta in glioma cell migration. Oncogene, 2003, 22, 6661-6668.
Muramatsu H., Shirahama H., Yonezawa S., Maruta H. & Muramatsu T. Midkine, a
retinoic acid-inducible growth/differentiation factor: immunochemical evidence for
the function and distribution. Dev Biol, 1993, 159, 392-402.
Muramatsu H., Zou K., Sakaguchi N., Ikematsu S., Sakuma S. & Muramatsu T. LDL
receptor-related protein as a component of the midkine receptor. Biochem Biophys
Res Commun, 2000, 270, 936-941.
Muramatsu H., Zou P., Suzuki H., Oda Y., Chen G.Y., Sakaguchi N., Sakuma S., Maeda
N., Noda M., Takada Y. & Muramatsu T. alpha4beta1- and alpha6beta1-integrins
are functional receptors for midkine, a heparin-binding growth factor. J Cell Sci,
2004, 117, 5405-5415.
Muramatsu T. Midkine and pleiotrophin: two related proteins involved in development,
survival, inflammation and tumorigenesis. J Biochem (Tokyo), 2002, 132, 359-371.
Najjam S., Gibbs R.V., Gordon M.Y. & Rider C.C. Characterization of human
recombinant interleukin 2 binding to heparin and heparan sulfate using an ELISA
approach. Cytokine, 1997, 9, 1013-1022.
Nakagawara A., Milbrandt J., Muramatsu T., Deuel T.F., Zhao H., Cnaan A. & Brodeur
G.M. Differential expression of pleiotrophin and midkine in advanced
neuroblastomas. Cancer Res, 1995, 55, 1792-1797.
Nakamoto M., Matsubara S., Miyauchi T., Obama H., Ozawa M. & Muramatsu T. A new
family of heparin binding growth/differentiation factors: differential expression of
the midkine (MK) and HB-GAM genes during mouse development. J Biochem
(Tokyo), 1992, 112, 346-349.
Nakanishi T., Kadomatsu K., Okamoto T., Ichihara-Tanaka K., Kojima T., Saito H.,
Tomoda Y. & Muramatsu T. Expression of syndecan-1 and -3 during
embryogenesis of the central nervous system in relation to binding with midkine. J
Biochem (Tokyo), 1997a, 121, 197-205.
Neame P.J., Young C.N., Brock C.W., Treep J.T., Ganey T.M., Sasse J. & Rosenberg L.C.
Pleiotrophin is an abundant protein in dissociative extracts of bovine fetal
epiphyseal cartilage and nasal cartilage from newborns. Journal of Orthopaedic
Research, 1993, 11, 479-491.
207
REFERENCES
Neuhoff V., Stamm R. & Eibl H. Improved staining of proteins in polyacrylamide gels
including isoelectric focusing gels with clear background at nanogram sensitivity
using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis, 1988, 6, 255262.
Nishiwaki T., Maeda N. & Noda M. Characterization and developmental regulation of
proteoglycan-type protein tyrosine phosphatase zeta/RPTPbeta isoforms. J
Biochem (Tokyo), 1998, 123, 458-467.
Nolo R., Kaksonen M., Raulo E. & Rauvala H. Co-expression of heparin-binding growth
associated molecule (HB-GAM) and N-syndecan (syndecan-3) in developing rat
brain. Neurosci Lett, 1995, 191, 39-42.
Novotny W.F., Maffi T., Mehta R.L. & Milner P.G. Identification of novel heparinreleasable proteins, as well as the cytokines midkine and pleiotrophin, in human
postheparin plasma. Arteriosclerosis & Thrombosis, 1993, 13, 1798-1805.
Ohyama Y., Miyamoto K., Minamino N. & Matsuo H. Isolation and identification of
midkine and pleiotrophin in bovine follicular fluid. Mol Cell Endocrinol, 1994,
105, 203-208.
Ornitz D.M., Yayon A., Flanagan J.G., Svahn C.M., Levi E. & Leder P. Heparin is
required for cell-free binding of basic fibroblast growth factor to a soluble receptor
and for mitogenesis in whole cells. Mol Cell Biol, 1992, 12, 240-247.
Pariser H., Ezquerra L., Herradon G., Perez-Pinera P. & Deuel T.F. Fyn is a downstream
target of the pleiotrophin/receptor protein tyrosine phosphatase beta/zeta-signaling
pathway: regulation of tyrosine phosphorylation of Fyn by pleiotrophin. Biochem
Biophys Res Commun, 2005a, 332, 664-669.
Pariser H., Perez-Pinera P., Ezquerra L., Herradon G. & Deuel T.F. Pleiotrophin stimulates
tyrosine phosphorylation of beta-adducin through inactivation of the
transmembrane receptor protein tyrosine phosphatase beta/zeta. Biochem Biophys
Res Commun, 2005b, 335, 232-239.
Pavlov I., Voikar V., Kaksonen M., Lauri S.E., Hienola A., Taira T. & Rauvala H. Role of
heparin-binding growth-associated molecule (HB-GAM) in hippocampal LTP and
spatial learning revealed by studies on overexpressing and knockout mice. Mol Cell
Neurosci, 2002, 20, 330-342.
Pellegrini L., Burke D.F., von Delft F., Mulloy B. & Blundell T.L. Crystal structure of
fibroblast growth factor receptor ectodomain bound to ligand and heparin. Nature,
2000, 407, 1029-1034.
Peng H.B., Ali A.A., Dai Z., Daggett D.F., Raulo E. & Rauvala H. The role of heparinbinding growth-associated molecule (HB-GAM) in the postsynaptic induction in
cultured muscle cells. J Neurosci, 1995, 15, 3027-3038.
Pepper M.S. Manipulating angiogenesis. Arterio Thromb Vasc Biol, 1997, 17, 605-619.
Perez-Pinera P., Garcia-Suarez O., Menendez-Rodriguez P., Mortimer J., Chang Y.,
Astudillo A. & Deuel T.F. The receptor protein tyrosine phosphatase
(RPTP)beta/zeta is expressed in different subtypes of human breast cancer.
Biochem Biophys Res Commun, 2007a, 362, 5-10.
Perez-Pinera P., Zhang W., Chang Y., Vega J.A. & Deuel T.F. Anaplastic lymphoma
kinase (ALK) is activated through the pleiotrophin (PTN)/receptor protein tyrosine
phosphatase (RPTP)beta /zeta signaling pathway: An "alternative mechanism of
receptor tyrosine kinase (RTK) activation". J Biol Chem, 2007b,
Petersen W., Wildemann B., Pufe T., Raschke M. & Schmidmaier G. The angiogenic
peptide pleiotrophin (PTN/HB-GAM) is expressed in fracture healing: an
immunohistochemical study in rats. Arch Orthop Trauma Surg, 2004, 124, 603607.
208
REFERENCES
Plank M.J. & Sleeman B.D. Tumour-induced angiogenesis. J Theor Med, 2004, 5, 137153.
Plotnikov A.N., Schlessinger J., Hubbard S.R. & Mohammadi M. Structural basis for FGF
receptor dimerization and activation. Cell, 1999, 98, 641-650.
Plouët J., Schilling J. & Gospodarowicz D. Isolation and characterization of a newly
identified endothelial cell mitogen produced by AtT-20 cells. EMBO J, 1989, 8,
3801-3806.
Polykratis A., Katsoris P., Courty J. & Papadimitriou E. Characterization of heparin affin
regulatory peptide signaling in human endothelial cells. J Biol Chem, 2005, 280,
22454-22461.
Polytarchou C., Hatziapostolou M. & Papadimitriou E. Hydrogen peroxide stimulates
proliferation and migration of human prostate cancer cells through activation of
activator protein-1 and up-regulation of the heparin affin regulatory peptide gene. J
Biol Chem, 2005, 280, 40428-40435.
Poulsen F.R., Lagord C., Courty J., Pedersen E.B., Barritault D. & Finsen B. Increased
synthesis of heparin affin regulatory peptide in the perforant path lesioned mouse
hippocampal formation. Exp Brain Res, 2000, 135, 319-330.
Powell A.K., Yates E.A., Fernig D.G. & Turnbull J.E. Interactions of heparin/heparan
sulfate with proteins: appraisal of structural factors and experimental approaches.
Glycobiology, 2004, 14, 17R-30R.
Powers C., Aigner A., Stoica G.E., McDonnell K. & Wellstein A. Pleiotrophin signaling
through anaplastic lymphoma kinase (ALK) is rate- limiting for glioblastoma
growth. J Biol Chem, 2002, 23, 23.
Pulford K., Lamant L., Morris S.W., Butler L.H., Wood K.M., Stroud D., Delsol G. &
Mason D.Y. Detection of anaplastic lymphoma kinase (ALK) and nucleolar protein
nucleophosmin (NPM)-ALK proteins in normal and neoplastic cells with the
monoclonal antibody ALK1. Blood, 1997, 89, 1394-1404.
Pulford K., Morris S.W. & Turturro F. Anaplastic lymphoma kinase proteins in growth
control and cancer. J Cell Physiol, 2004, 199, 330-358.
Qi M., Ikematsu S., Maeda N., Ichihara-Tanaka K., Sakuma S., Noda M., Muramatsu T. &
Kadomatsu K. Haptotactic migration induced by midkine. Involvement of proteintyrosine
phosphatase
zeta.
Mitogen-activated
protein
kinase,
and
phosphatidylinositol 3-kinase. J Biol Chem, 2001, 276, 15868-15875.
Raulais D., Lagente-Chevallier O., Guettet C., Duprez D., Courtois Y. & Vigny M. A new
heparin binding protein regulated by retinoic acid from chick embryo. Biochemical
& Biophysical Research Communications, 1991, 174, 708-715.
Raulo E., Chernousov M.A., Carey D.J., Nolo R. & Rauvala H. Isolation of a neuronal cell
surface receptor of heparin binding growth-associated molecule (HB-GAM).
Identification as N-syndecan (syndecan-3). Journal of Biological Chemistry, 1994,
269, 12999-13004.
Raulo E., Julkunen I., Merenmies J., Pihlaskari R. & Rauvala H. Secretion and biological
activities of heparin-binding growth-associated molecule. Neurite outgrowthpromoting and mitogenic actions of the recombinant and tissue-derived protein.
Journal of Biological Chemistry, 1992, 267, 11408-11416.
Raulo E., Tumova S., Pavlov I., Hienola A., Klankki E., Kalkkinen N., Taira T.,
Kilpelainen I. & Rauvala H. The two thrombospondin type I repeat domains of the
heparin-binding growth-associated molecule bind to heparin/heparan sulfate and
regulate neurite extension and plasticity in hippocampal neurons. J Biol Chem,
2005, 280, 41576-41583.
209
REFERENCES
Raulo E., Tumova S., Pavlov I., Hienola A., Lauri S., Taira T. & Rauvala H. The Two
Thrombospondin Type I Repeat Domains of HB-GAM display a cooperative
function in N-syndecan binding and regulation of synaptic plasticity.
TheScientificWorldJournal, 2006, 6, 406-409.
Rauvala H. An 18-kd heparin-binding protein of developing brain that is distinct from
fibroblast growth factors. EMBO Journal, 1989, 8, 2933-2941.
Rauvala H., Vanhala A., Castren E., Nolo R., Raulo E., Merenmies J. & Panula P.
Expression of HB-GAM (heparin-binding growth-associated molecules) in the
pathways of developing axonal processes in vivo and neurite outgrowth in vitro
induced by HB-GAM. Brain Research. Developmental Brain Research, 1994, 79,
157-176.
Rich R.L. & Myszka D.G. Survey of the year 2005 commercial optical biosensor literature.
J Mol Recognit, 2006, 19, 478-534.
Riegel A.T. & Wellstein A. The potential role of the heparin-binding growth factor
pleiotrophin in breast cancer. Breast Cancer Res Treat, 1994, 31, 309-314.
Roger J., Brajeul V., Thomasseau S., Hienola A., Sahel J.A., Guillonneau X. & Goureau
O. Involvement of Pleiotrophin in CNTF-mediated differentiation of the late retinal
progenitor cells. Dev Biol, 2006, 298, 527-539.
Rudd T.R., Guimond S.E., Skidmore M.A., Duchesne L., Guerrini M., Torri G., Cosentino
C., Brown A., Clarke D.T., Turnbull J.E., Fernig D.G. & Yates E.A. Influence of
substitution pattern and cation binding on conformation and activity in heparin
derivatives. Glycobiology, 2007, 17, 983-993.
Rumsby M., Suggitt F., Haynes L., Hughson E., Kidd D. & McNulty S. Substratum of
pleiotrophin (HB-GAM) stimulates rat CG-4 line oligodendrocytes to adopt a
bipolar morphology and disperse: primary O- 2A progenitor glial cells disperse
similarly on pleiotrophin. Glia, 1999, 26, 361-367.
Ruoslahti E. Proteoglycans in cell regulation. J Biol Chem, 1989, 264, 13369-13372.
Russell P.J., Bennett S. & Stricker P. Growth factor involvement in progression of prostate
cancer. Clin Chem, 1998, 44, 705-723.
Said E.A., Courty J., Svab J., Delbe J., Krust B. & Hovanessian A.G. Pleiotrophin inhibits
HIV infection by binding the cell surface-expressed nucleolin. Febs J, 2005, 272,
4646-4659.
Sakaguchi N., Muramatsu H., Ichihara-Tanaka K., Maeda N., Noda M., Yamamoto T.,
Michikawa M., Ikematsu S., Sakuma S. & Muramatsu T. Receptor-type protein
tyrosine phosphatase? as a component of the signaling receptor complex for
midkine-dependent survival of embryonic neurons. Neurosci Res, 2003, 45, 219224.
Salama R.H., Muramatsu H., Zou K., Inui T., Kimura T. & Muramatsu T. Midkine binds
to 37-kDa laminin binding protein precursor, leading to nuclear transport of the
complex. Exp Cell Res, 2001, 270, 13-20.
Sato H., Funahashi M., Kristensen D.B., Tateno C. & Yoshizato K. Pleiotrophin as a Swiss
3T3 cell-derived potent mitogen for adult rat hepatocytes. Exp Cell Res, 1999, 246,
152-164.
Sato M., Miyado K., Okazaki M. & Kimura M. Regulation of osteoblast-specific factor-1
(OSF-1) mRNA expression by dual promoters as revealed by RT-PCR. Biochem
Biophys Res Commun, 1997, 238, 831-837.
Schlessinger J., Lax I. & Lemmon M. Regulation of growth factor activation by
proteoglycans: what is the role of the low affinity receptors? Cell, 1995, 83, 357360.
210
REFERENCES
Schlessinger J., Plotnikov A.N., Ibrahimi O.A., Eliseenkova A.V., Yeh B.K., Yayon A.,
Linhardt R.J. & Mohammadi M. Crystal structure of a ternary FGF-FGFR-heparin
complex reveals a dual role for heparin in FGFR binding and dimerization. Mol
Cell, 2000, 6, 743-750.
Schlessinger J. Cell signaling by receptor tyrosine kinases. Cell, 2000, 103, 211-225.
Schnolzer M., Alewood P., Jones A., Alewood D. & Kent S.B.H. In situ neutralization in
Boc-chemistry solid phase peptide synthesis. Int J Peptide Protein Res, 1992, 40,
180-193.
Schuck P. Use of surface plasmon resonance to probe the equilibrium and dynamic aspects
of interactions between biological macromolecules Annu Rev Biophys Biomol
Struct, 1997, 26, 541-566.
Schulte A.M., Lai S., Kurtz A., Czubayko F., Riegel A.T. & Wellstein A. Human
trophoblast and choriocarcinoma expression of the growth factor pleiotrophin
attributable to germ-line insertion of an endogenous retrovirus. Proc Natl Acad Sci
U S A, 1996, 93, 14759-14764.
Seddon A.P., Hulmes J.D., Decker M.M., Kovesdi I., Fairhurst J.L., Backer J., DougherVermazen M. & Bohlen P. Refolding and characterization of human recombinant
heparin-binding neurite-promoting factor. Protein Expr Purif, 1994, 5, 14-21.
Sharifi B.G., Zeng Z., Wang L., Song L., Chen H., Qin M., Sierra-Honigmann M.R.,
Wachsmann-Hogiu S. & Shah P.K. Pleiotrophin induces transdifferentiation of
monocytes into functional endothelial cells. Arterioscler Thromb Vasc Biol, 2006,
26, 1273-1280.
Shibata Y., Muramatsu T., Hirai M., Inui T., Kimura T., Saito H., McCormick L.M., Bu G.
& Kadomatsu K. Nuclear targeting by the growth factor midkine. Mol Cell Biol,
2002, 22, 6788-6796.
Silos-Santiago I., Yeh H.J., Gurrieri M.A., Guillerman R.P., Li Y.S., Wolf J., Snider W. &
Deuel T.F. Localization of pleiotrophin and its mRNA in subpopulations of
neurons and their corresponding axonal tracts suggests important roles in neuralglial interactions during development and in maturity. J Neurobiol, 1996, 31, 283296.
Simon-Chazottes D., Matsubara S., Miyauchi T., Muramatsu T. & Guenet J.L.
Chromosomal localization of two cell surface-associated molecules of potential
importance in development: midkine (Mdk) and basigin (Bsg). Mamm Genome,
1992, 2, 269-271.
Soker S., Takashima S., Miao H.Q., Neufeld G. & Klagsbrun M. Neuropilin-1 is expressed
by endothelial and tumor cells as an isoform-specific receptor for vascular
endothelial growth factor. Cell, 1998, 92, 735-745.
Soker S, Miao HQ, Nomi M, Takashima S & Klagsbrun M. VEGF165 mediates formation
of complexes containing VEGFR-2 and neuropilin-1 that enhance VEGF165receptor binding. J Cell Biochem, 2002, 85, 357-368.
Soulie P., Heroult M., Bernard I., Kerros M.E., Milhiet P.E., Delbe J., Barritault D.,
Caruelle D. & Courty J. Immunoassay for measuring the heparin-binding growth
factors HARP and MK in biological fluids. J Immunoassay Immunochem, 2002, 23,
33-48.
Souttou B., Ahmad S., Riegel A.T. & Wellstein A. Signal transduction pathways involved
in the mitogenic activity of pleiotrophin - Implication of mitogen-activated protein
kinase and phosphoinositide 3-kinase pathways. J Biol Chem, 1997, 272, 1958819593.
Souttou B., Juhl H., Hackenbruck J., Rockseisen M., Klomp H.J., Raulais D., Vigny M. &
Wellstein A. Relationship between serum concentrations of the growth factor
211
REFERENCES
pleiotrophin and pleiotrophin-positive tumors. J Natl Cancer Inst, 1998, 90, 14681473.
Souttou B., Carvalho N.B., Raulais D. & Vigny M. Activation of anaplastic lymphoma
kinase receptor tyrosine kinase induces neuronal differentiation through the
mitogen-activated protein kinase pathway. J Biol Chem, 2001, 276, 9526-9531.
Spivak-Kroizman T., Lemmon M.A., Dikic I., Ladbury J.E., Pinchasi D., Huang J., Jaye
M., Crumley G., Schlessinger J. & Lax I. Heparin-induced oligomerization of FGF
molecules is responsible for FGF receptor dimerization, activation, and cell
proliferation. Cell, 1994, 79, 1015-1024.
Sporn M.B. & Roberts A.B. Autocrine growth factors and cancer. Nature, 1985, 313, 745747.
Sporn M.B. The early history of TGF-β, and a brief glimpse of its future. Cytokine Growth
Factor Rev, 2006, 17, 3-7.
Springer B.A., Pantoliano M.W., Barbera F.A., Gunyuzlu P.L., Thompson L.D., Herblin
W.F., Rosenfeld S.A. & Book G.W. Identification and concerted function of two
receptor binding surfaces on basic fibroblast growth factor required for
mitogenesis. J Biol Chem, 1994, 269, 26879-26884.
Stauber D.J., DiGabriele A.D. & Hendrickson W.A. Structural interactions of fibroblast
growth factor receptor with its ligands. Proc Natl Acad Sci U S A, 2000, 97, 49-54.
Stoica G.E., Kuo A., Aigner A., Sunitha I., Souttou B., Malerczyk C., Caughey D.J., Wen
D., Karavanov A., Riegel A.T. & Wellstein A. Identification of anaplastic
lymphoma kinase as a receptor for the growth factor pleiotrophin. J Biol Chem,
2001, 276, 16772-16779.
Stoica G.E., Kuo A., Powers C., Bowden E.T., Sale E.B., Riegel A.T. & Wellstein A.
Midkine binds to anaplastic lymphoma kinase (ALK) and acts as a growth factor
for different cell types. J Biol Chem, 2002, 277, 35990-35998.
Sun X.-Z. & Yoshifiro F. Midkine, A nex heparin binding growth/differeniation factor:
Expression and Distribution during embryogenesis and pathological status. Cong
Anom, 1998, 38, 25-38.
Sutherland R.M., Rasey J.S. & Hill R.P. Tumor biology. Am J Clin Oncol, 1988, 11, 253274.
Szabat E. & Rauvala H. Role of HB-GAM (heparin-binding growth-associated molecule)
in proliferation arrest in cells of the developing rat limb and its expression in the
differentiating neuromuscular system. Developmental Biology, 1996, 178, 77-89.
Takamatsu H., Motofumi I., Kimura M., Gospodarowicz D. & Amann E. Expression and
purification of biologically active human OSF-1 in Esherichia coli. Biochem
Biophys Res Commun, 1992, 185, 224-230.
Take M., Tsutsui J., Obama H., Ozawa M., Nakayama T., Maruyama I., Arima T. &
Muramatsu T. Identification of nucleolin as a binding protein for midkine (MK)
and heparin-binding growth associated molecule (HB-GAM). Journal of
Biochemistry, 1994, 116, 1063-1068.
Takeda A., Onodera H., Sugimoto A., Itoyama Y., Kogure K., Rauvala H. & Shibahara S.
Induction of heparin-binding growth-associated molecule expression in reactive
astrocytes following hippocampal neuronal injury. Neurosci 1995, 68, 57-64.
Takei Y., Kadomatsu K., Matsuo S., Itoh H., Nakazawa K., Kubota S. & Muramatsu T.
Antisense oligodeoxynucleotide targeted to Midkine, a heparin-binding growth
factor, suppresses tumorigenicity of mouse rectal carcinoma cells. Cancer Res,
2001, 61, 8486-8491.
212
REFERENCES
Tamura M., Ichikawa F., Guillerman R.P., Deuel T.F. & Noda M. 1-Alpha,25Dihydroxyvitamin D-3 downregulates pleiotrophin messenger RNA expression in
osteoblast-like cells. Endocrinology, 1994, 3, 21-24.
Tapp H., Hernandez D.J., Neame P.J. & Koob T.J. Pleiotrophin inhibits chondrocyte
proliferation and stimulates proteoglycan synthesis in mature bovine cartilage.
Matrix Biology, 1999, 18, 543-556.
Tezuka K.I., Takeshita S., Hakeda Y., Kumegawa M., Kikuno R. & Hashimoto-Gotoh T.
Isolation of mouse and human cDNA clones encoding a protein expressed specially
in osteoblasts and brain tissues. Biochem. Biophys. Res. Commun., 1990, 173, 246251.
Tomomura M., Kadomatsu K., Nakamoto M., Muramatsu H., Kondoh H., Imagawa K. &
Muramatsu T. A retinoic acid responsive gene, MK, produces a secreted protein
with heparin binding activity. Biochem Biophys Res Commun, 1990, 171, 603-609.
Tsutsui J., Kadomatsu K., Matsubara S., Nakagawara A., Hamanoue M., Takao S.,
Shimazu H., Ohi Y. & Muramatsu T. A new family of heparin-binding
growth/differentiation factors: increased midkine expression in Wilms' tumor and
other human carcinomas. Cancer Res, 1993, 53, 1281-1285.
Uehara K., Matsubara S., Kadomatsu K., Tsutsui J. & Muramatsu T. Genomic structure of
human midkine (MK), a retinoic acid-responsive growth/differentiation factor. J
Biochem (Tokyo), 1992, 111, 563-567.
Ulbricht U., Brockmann M.A., Aigner A., Eckerich C., Muller S., Fillbrandt R., Westphal
M. & Lamszus K. Expression and function of the receptor protein tyrosine
phosphatase zeta and its ligand pleiotrophin in human astrocytomas. J Neuropathol
Exp Neurol, 2003, 62, 1265-1275.
Ulbricht U., Eckerich C., Fillbrandt R., Westphal M. & Lamszus K. RNA interference
targeting protein tyrosine phosphatase zeta/receptor-type protein tyrosine
phosphatase beta suppresses glioblastoma growth in vitro and in vivo. J
Neurochem, 2006, 98, 1497-1506.
Ullrich A. & Schlessinger J. Signal transduction by receptors with tyrosine kinase activity.
Cell, 1990, 61, 203-212.
Unoki K., Ohba N., Arimura H., Muramatsu H. & Muramatsu T. Rescue of photoreceptors
from the damaging effects of constant light by midkine, a retinoic acid-responsive
gene product. Invest Ophthalmol Vis Sci, 1994, 35, 4063-4068.
Usui T., Yamagami S., Yokoo S., Mimura T., Ono K. & Amano S. Gene expression profile
in corneal neovascularization identified by immunology related macroarray. Mol
Vis, 2004, 10, 832-836.
Vacherot F., Laaroubi K., Caruelle D., Delbe J., Barritault D., Caruelle J.P. & Courty J.
Upregulation of heparin-affin regulatory peptide by androgen [letter]. In Vitro Cell
Dev Biol Anim, 1995, 31, 647-648.
Vacherot F., Delbe J., Heroult M., Barritault D., Fernig D.G. & Courty J.
Glycosaminoglycans differentially bind HARP and modulate its biological activity.
J Biol Chem, 1999a, 274, 7741-7747.
Vacherot F., Caruelle D., Chopin D., Gil-Diez S., Barritault D., Caruelle J.P. & Courty J.
Involvement of heparin affin regulatory peptide in human prostate cancer. Prostate,
1999b, 38, 126-136.
Vanderwinden J.M., Mailleux P., Schiffmann S.N. & Vanderhaeghen J.J. Cellular
distribution of the new growth factor pleiotrophin (HB-GAM) mRNA in
developing and adult rat tissues. Anatomy & Embryology, 1992, 186, 387-406.
Vanpouille C., Deligny A., Delehedde M., Denys A., Melchior A., Lienard X., Lyon M.,
Mazurier J., Fernig D.G. & Allain F. The heparin/heparan sulfate sequence that
213
REFERENCES
interacts with cyclophilin B contains a 3-O-sulfated N-unsubstituted glucosamine
residue. J Biol Chem, 2007, 282, 24416-24429.
Venkataraman G., Sasisekharan V., Herr A.B., Ornitz D.M., Waksman G., Cooney C.L.,
Langer R. & Sasisekharan R. Preferential self-association of basic fibroblast
growth factor is stabilized by heparin during receptor dimerization and activation.
Proc Natl Acad Sci U S A, 1996, 93, 845-850.
Venter J.C., Adams M.D., Myers E.W., Li P.W., Mural R.J., Sutton G.G., Smith H.O.,
Yandell M., Evans C.A. & Holt R.A. The sequence of human genome. Science,
2001, 291, 1304-1351.
Vigny M., Raulais D., Puzenat N., Duprez D., Hartmann M.P., Jeanny J.C. & Courtois Y.
Identification of a new heparin-binding protein localized within chick basement
membranes. Eur J Biochem, 1989, 186, 733-740.
Wanaka A., Carroll S.L. & Milbrandt J. Developmentally regulated expression of
pleiotrophin, a novel heparin binding growth factor, in the nervous system of the
rat. Brain Research. Developmental Brain Research, 1993, 72, 133-144.
Ward L.D., Sodhi J.S., McGuffin L.J., Buxton B.F. & Jones D.T. Prediction and functional
analysis of native disorder in proteins from the three kingdoms of life J Mol Biol,
2004, 337, 635-645.
Weber D., Klomp H.J., Czubayko F., Wellstein A. & Juhl H. Pleiotrophin can be ratelimiting for pancreatic cancer cell growth. Cancer Res, 2000, 60, 5284-5288.
Wellstein A., Fang W.J., Khatri A., Lu Y., Swain S.S., Dickson R.B., Sasse J., Riegel A.T.
& Lippman M.E. A heparin-binding growth factor secreted from breast cancer cells
homologous to a developmentally regulated cytokine. Journal of Biological
Chemistry, 1992, 267, 2582-2587.
West D.C., Rees C.G., Duchesne L., Patey S.J., Terry C.J., Turnbull J.E., Delehedde M.,
Heegaard C.W., Allain F., Vanpouille C., Ron D. & Fernig D.G. Interactions of
multiple heparin binding growth factors with neuropilin-1 and potentiation of the
activity of fibroblast growth factor-2. J Biol Chem, 2005, 280, 13457-13464.
Wray W., Boulikas T., Wray W.P. & Hancock R. Silver staining of proteins in
polyacrylamide gels. Anal Biochem, 1981, 18, 197-203.
Yamada H., Inazumi T., Tajima S., Muramatsu H. & Muramatsu T. Stimulation of
collagen expression and glycosaminoglycan synthesis by midkine in human skin
fibroblasts. Arch Dermatol Res, 1997, 289, 429-433.
Yancopoulos G.D., Davis S., Gale N.W., Rudge J.S., Wiegand S.J. & Holash J. Vascularspecific growth factors and blood vessel formation. Nature, 2000, 407, 242-248.
Yasuhara O., Muramatsu H., Kim S.U., Muramatsu T., Maruta H. & McGeer P.L.
Midkine, a novel neurotrophic factor, is present in senile plaques of Alzheimer
disease. Biochem Biophys Res Commun, 1993, 192, 246-251.
Yates E.A., Santini F., Guerrini M., Naggi G. & Casu B. 1H and 13C NMR spectral
assignments of the major sequences of twelve systematically modified heparin
derivatives. Carbohydr Res, 1996, 20, 15-27.
Yates E.A., Guimond S.E. & Turnbull J.E. Highly diverse heparan sulfate analogue
libraries: providing access to expanded areas of sequence space for bioactivity
screening. J Med Chem, 2004, 47, 277-280.
Yayon A., Klagsbrun M., Esko J.D., Leder P. & Ornitz D.M. Cell surface, heparin-like
molecules are required for binding of basic fibroblast growth factor to its high
affinity receptor. Cell, 1991, 64, 841-848.
Yeh H.J., He Y.Y., Xu J., Hsu C.Y. & Deuel T.F. Upregulation of pleiotrophin gene
expression in developing microvasculature, macrophages, and astrocytes after acute
ischemic brain injury. J Neurosci, 1998, 18, 3699-3707.
214
REFERENCES
Yoshida Y., Goto M., Tsutsui J., Ozawa M., Sato E., Osame M. & Muramatsu T. Midkine
is present in the early stage of cerebral infarct. Biochem Biophys Res Commun,
1995, 85, 25-30.
Zhang N., Zhong R., Wang Z.Y. & Deuel T.F. Human breast cancer growth inhibited in
vivo by a dominant negative pleiotrophin mutant. J Biol Chem, 1997, 272, 1673316736.
Zhang N., Zhong R. & Deuel T.F. Domain structure of pleiotrophin required for
transformation. Journal of Biological Chemistry, 1999, 274, 12959-12962.
Zhang N., Zhong R., Perez-Pinera P., Herradon G., Ezquerra L., Wang Z.Y. & Deuel T.F.
Identification of the angiogenesis signaling domain in pleiotrophin defines a
mechanism of the angiogenic switch. Biochem Biophys Res Commun, 2006, 343,
653-658.
Zhang S., Wang G., Liu D., Bao Z., Fernig D.G., Rudland P.S. & Barraclough R. The Cterminal region of S100A4 is important for its metastasis-inducing properties.
Oncogene, 2005, 24, 4401-4411.
Zhou H.Y., Ohnuma Y., Takita H., Fujisawa R., Mizuno M. & Kuboki Y. Effects of a bone
lysine-rich 18 kDa protein on osteoblast-like MC3T3-E1 cells. Biochem. Biophys.
Res. Commun., 1992, 186, 1288-1293.
215
REFERENCES
Aaronson S.A. Growth factors and cancer. Science, 1991, 254, 1146-1153.
Adachi Y., Matsubara S., Pedraza C., Ozawa M., Tsutsui J., Takamatsu H., Noguchi H.,
Akiyama T. & Muramatsu T. Midkine as a novel target gene for the Wilm's tumor
suppressor gene (WT1). Oncogene, 1996, 13, 2197-2203.
Adams J.C.a.T., R.P. The thrombospondin type I repeat (TSR) superfamily; diverse protein
with related roles in neuronal development. Dev.Dyn., 2000, 218, 280-299.
Aigner A., Brachmann P., Beyer J., Jager R., Raulais D., Vigny M., Neubauer A.,
Heidenreich A., Weinknecht S., Czubayko F. & Zugmaier G. Marked increase of
the growth factors pleiotrophin and fibroblast growth factor-2 in serum of testicular
cancer patients. Ann Oncol, 2003, 14, 1525-1529.
Akhter S., Ichihara-Tanaka K., Kojima S., Muramatsu H., Inui T., Kimura T., Kaneda N.,
Talukder A.H., Kadomatsu K., Inagaki F. & Muramatsu T. Clusters of basic amino
acids in midkine: roles in neurite-promoting activity and plasminogen activatorenhancing activity. J Biochem (Tokyo), 1998, 123, 1127-1136.
Alberts B., Bray D., Lewis J., Raff M., Roberts K. & Watson J.D. The Molecular Biology
of the Cell. 3rd Ed. New York: Garland, 1994.
Alewood P., Alewood D., Miranda L., Love S. & Meutermans W. Rapid in situ
neutralization protocols for Boc and Fmoc solid-phase chemistries. Methods
Enzymol, 1997, 289, 14-29.
Ambartsumian N., Klingelhofer J., Grigorian M., Christensen C., Kriajevska M.,
Tulchinsky E., Georgiev G., Berezin V., Bock E., Rygaard J., Cao R., Cao Y. &
Lukanidin E. The metastasis-associated Mts1(S100A4) protein could act as an
angiogenic factor. Oncogene, 2001, 20, 4685-4695.
Antoine M., Tag C.G., Wirz W., Borkham-Kamphorst E., Sawitza I., Gressner A.M. &
Kiefer P. Upregulation of pleiotrophin expression in rat hepatic stellate cells by
PDGF and hypoxia: implications for its role in experimental biliary liver
fibrogenesis. Biochem Biophys Res Commun, 2005, 337, 1153-1164.
Badet J. Angiogenic role of growth factors. Contracept Fertil Sex, 1994, 213, 375-382.
Bernard-Pierrot I., Heroult M., Lemaitre G., Barritault D., Courty J. & Milhiet P.E.
Glycosaminoglycans promote HARP/PTN dimerization. Biochem Biophys Res
Commun, 1999, 266, 437-442.
Bernard-Pierrot I., Delbe J., Caruelle D., Barritault D., Courty J. & Milhiet P.E. The
lysine-rich C-terminal tail of heparin affin regulatory peptide is required for
mitogenic and tumor formation activities. J Biol Chem, 2001, 276, 12228-12234.
Bernard-Pierrot I., Delbe J., Rouet V., Vigny M., Kerros M.E., Caruelle D., Raulais D.,
Barritault D., Courty J. & Milhiet P.E. Dominant negative effectors of heparin affin
regulatory peptide (HARP) angiogenic and transforming activities. J Biol Chem,
2002, 277, 32071-32077.
Bernard-Pierrot I., Delbe J., Heroult M., Rosty C., Soulie P., Barritault D., Milhiet P.E. &
Courty J. Heparin affin regulatory peptide in milk: its involvement in mammary
gland homeostasis. Biochem Biophys Res Commun, 2004, 314, 277-282.
Bohlen P., Gautschi-Sova P., Albrecht U., Lehmann S. & Huber D. J Cell Biochem Suppl,
1988, 12A, 221.
Bohlen P. & Gautschi P. European Patent Application, 1989, EP 326,075,
Böhlen P., Muller T., Gautschi-Sova P., Albrecht U., Rasool C.G., Decker M., Seddon A.,
Fafeur V., Kovesdi I. & Kretschmer P. Isolation from bovine brain and structural
characterization of HBNF, a heparin-binding neurotrophic factor. Growth Factors,
1991, 4, 97-107.
Bowden E.T., Stoica G.E. & Wellstein A. Anti-apoptotic signaling of pleiotrophin through
its receptor, anaplastic lymphoma kinase. J Biol Chem, 2002, 277, 35862-35868.
198
REFERENCES
Brigstock D.R., Kim G.Y. & Steffen C.L. Pig uterine luminal fluid contains the
developmentally regulated neurotrophic factor, pleiotrophin. J Endocrinol, 1996,
148, 103-111.
Callebaut C., Jacotot E., Krust B., Guichard G., Blanco J., Valenzuala A., Svab J., Muller
S., Briand J.P. & Hovanessian A.G. Pseudopeptide TASP inhibitors of HIV entry
bind specifically to a 95-kDa cell surface protein. J Biol Chem, 1997, 272, 71597166.
Callebaut C., Blanco J., Benkirane N., Krust B., Jacotot E., Guichard G., Seddiki N., Svab
J., Dam E., Muller S., Briand J.P. & Hovanessian A.G. Identification of V3 Loopbinding proteins as potential receptors implicated in the bining of HIV particles to
CD4+ cells. J Biol Chem, 1998, 279 21988-21997.
Callebaut C., Nisole S., Briand J.P., Krust B. & Hovanessian A.G. Inhibition of HIV
infection by the cytokine midkine. Virology, 2001, 281, 248-264.
Carey D.J., Evans D.M., Stahl R.C., Asundi V.K., Conner K.J., Garbes P. & GizmeciSmith G. Molecular cloning and characterization of N-syndecan, a novel
transmembrane heparan sulphate proteoglycan. J. Cell. Biol., 1992, 117, 191-201.
Carmeliet P. & Collen D. Molecular basis of angiogenesis. Role of VEGF and VEcadherin. Ann N Y Acad Sci, 2000, 902, 249-262; discussion 262-244.
Chao M.V. Dependence Receptors:What is the mechanism? Sci STKE, 2003, 200, 38.
Chauhan A.K., Li Y.S. & Deuel T.F. Pleiotrophin transforms NIH 3T3 cells and induces
tumors in nude mice. Proceedings of the National Academy of Sciences of the
United States of America, 1993, 90, 679-682.
Choudhuri R., Zhang H.T., Donnini S., Ziche M. & Bicknell R. An angiogenic role for the
neurokines midkine and pleiotrophin in tumorigenesis. Cancer Res, 1997, 57,
1814-1819.
Christman K.L., Fang Q., Yee M.S., Johnson K.R., Sievers R.E. & Lee R.J. Enhanced
neovasculature formation in ischemic myocardium following delivery of
pleiotrophin plasmid in a biopolymer. Biomaterials, 2005a, 26, 1139-1144.
Christman K.L., Fang Q., Kim A.J., Sievers R.E., Fok H.H., Candia A.F., Colley K.J.,
Herradon G., Ezquerra L., Deuel T.F. & Lee R.J. Pleiotrophin induces formation of
functional neovasculature in vivo. Biochem Biophys Res Commun, 2005b, 332,
1146-1152.
Chung H.W., Wen Y., Choi E.A., Hao L., Moon H.S., Yu H.K. & Polan M.L. Pleiotrophin
(PTN) and midkine (MK) mRNA expression in eutopic and ectopic endometrium
in advanced stage endometriosis. Mol Hum Reprod, 2002, 8, 350-355.
Connolly J.M. & D.P. R. Angiogenesis in two human prostate cancer cell lines with
differing metastatic potential when growing as solid tumors in nude mice. I Urol,
1998, 160, 932-936.
Corbley M.J. Transformation by ras suppresses expression of the neurotrophic growth
factor pleiotrophin. J Biol Chem, 1997, 272, 24696-24702.
Courty J., Dauchel M.C., Caruelle D., Perderiset M. & Barritault D. Mitogenic properties
of a new endothelial cell growth factor related to pleiotrophin. Biochemical &
Biophysical Research Communications, 1991, 180, 145-151.
Czubayko F., Riegel A.T. & Wellstein A. Ribozyme-targeting elucidates a direct role of
pleiotrophin in tumor growth. J Biol Chem, 1994, 269, 21358-21363.
Czubayko F., Schulte A.M., Berchem G.J. & Wellstein A. Melanoma angiogenesis and
metastasis modulated by ribozyme targeting of the secreted growth factor
pleiotrophin. Proceedings of the National Academy of Sciences of the United States
of America, 1996, 93, 14753-14758.
199
REFERENCES
De Silva Rudland S., Martin L., Roshanlall C., Winstanley J., Leinster S., Platt-Higgins A.,
Carroll J., West C., Barraclough R., Rudland P. Association of S100A4 and
osteopontin with specific prognostic factors and survival of patients with minimally
invasive breast cancer. Clin Cancer Res, 2006, 15, 1192-1200.
Deepa S.S., Yamada S., Zako M., Goldberger O. & Sugahara K. Chondroitin sulfate chains
on syndecan-1 and syndecan-4 from normal murine mammary gland epithelial cells
are structurally and functionally distinct and cooperate with heparan sulfate chains
to bind growth factors. A novel function to control binding of midkine,
pleiotrophin, and basic fibroblast growth factor. J Biol Chem, 2004, 279, 3736837376.
Delbe J., Vacherot F., Laaroubi K., Barritault D. & Courty J. Effect of heparin on bovine
epithelial lens cell proliferation induced by heparin affin regulatory peptide. J Cell
Physiol, 1995, 164, 47-54.
Delehedde M., Lyon M., Sergeant N., Rahmoune H. & Fernig D.G. Proteoglycans:
pericellular and cell surface multireceptors that integrate external stimuli in the
mammary gland. Journal of Mammary Gland Biology and Neoplasia, 2001, 6, 253273.
Delehedde M., Sergeant N., Lyon M., Rudland P.S. & Fernig D.G. Hepatocyte growth
factor/scatter factor stimulates migration of rat mammary fibroblasts through both
mitogen-activated protein kinase and phosphatidylinositol 3-kinase/Akt pathways.
Eur J Biochem, 2002, 268, 4423-4429.
Delehedde M., Lyon M., Gallagher J.T., Rudland P.S. & Fernig D.G. Fibroblast growth
factor-2 binds to small heparin-derived oligosaccharides and stimulates a sustained
phosphorylation of p42/44 mitogen-activated protein kinase and proliferation of rat
mammary fibroblasts. J Biochem, 2002, 366, 235-244.
Deuel T.F. Polypeptide Growth Factors: Roles in normal and abnormal cell growth. Ann
Rev Cell Biol, 1987, 3, 443-492.
Deuel T.F., Zhang N., Yeh H.-J., Silos-Santiago I. & Wang Z.-Y. Pleiotrophin: a cytokine
with diverse functions and a novel signaling pathway. Arch Biochem Biophys,
2002, 397, 162-171.
Dirks W.G., Fahnrich S., Lis Y., Becker E., MacLeod R.A. & Drexler H.G. Expression and
functional analysis of the anaplastic lymphoma kinase (ALK) gene in tumor cell
lines. Int J Cancer, 2002, 100, 49-56.
Dong Z. & Cmarik J.L. Harvesting cells under anchorage-independent cell transformation
conditions for biochemical analyses. Sci STKE, 2002, 130, pl7.
Duchesne L., Tissot B., Rudd T.R., Dell A. & Fernig D.G. N-glycosylation of fibroblast
growth factor receptor 1 regulates ligand and heparan sulfate co-receptor binding. J
Biol Chem, 2006, 281, 27178-27189.
Dulbecco R. & Stoker M.G. Conditions determining initiation of DNA synthesis in 3T3
cells. Proc Natl Acad Sci U S A, 1970, 66, 204-210.
Dunker A.K. & Obradovic Z. The protein trinity-linking function and disorder. Nature
Biotechnol., 2001, 19, 805-806.
Dunker A.K., Brown C.J., Lawson J.D., Iakoucheva L.M. & Obradovic Z. Intrinsic
disorder and protein function. Biochemistry, 2002, 41, 6573-6582.
Englund C., Loren C.E., Grabbe C., Varshney G.K., Deleuil F., Hallberg B. & Palmer R.H.
Jeb signals through the Alk receptor tyrosine kinase to drive visceral muscle fusion.
Nature, 2003, 425, 512-516.
Englund C., Birve A., Falileeva L., Grabbe C. & Palmer R.H. Miple1 and miple2 encode a
family of MK/PTN homologues in Drosophila melanogaster. Dev Genes Evol,
2005, 216, 10-18.
200
REFERENCES
Ezquerra L., Herradon G., Nguyen T., Vogt T.F., Bronson R., Silos-Santiago I. & Deuel
T.F. Pleiotrophin is a major regulator of the catecholamine biosynthesis pathway in
mouse aorta. Biochem Biophys Res Commun, 2004, 323, 512-517.
Fabri L., Nice E.C., Ward L.D., Maruta H., Burgess A.W. & Simpson R.J.
Characterization of bovine heparin-binding neurotrophic factor (HBNF):
assignment of disulfide bonds. Biochemistry International, 1992, 28, 1-9.
Fabri L., Maruta H., Muramatsu H., Muramatsu T., Simpson R.J., Burgess A.W. & Nice
E.C. Structural characterisation of native and recombinant forms of the
neurotrophic cytokine MK. J Chromatogr, 1993, 646, 213-225.
Fan Q.W., Muramatsu T. & Kadomatsu K. Distinct expression of midkine and pleiotrophin
in the spinal cord and placental tissues during early mouse development. Dev
Growth Differ, 2000, 42, 113-119.
Fang W., Hartmann N., Chow D.T., Riegel A.T. & Wellstein A. Pleiotrophin stimulates
fibroblasts and endothelial and epithelial cells and is expressed in human cancer.
Journal of Biological Chemistry, 1992, 267, 25889-25897.
Fath M., VanderNoot V., Kilpelainen I., Kinnunen T., Rauvala H. & Linhardt R.J.
Interaction of soluble and surface-bound heparin binding growth-associated
molecule with heparin. FEBS Lett, 1999, 454, 105-108.
Fernig D.G. Optical biosensor techniques to analyze protein-polysaccharide interactions.
Methods Mol Biol, 2001, 171, 505-518.
Foehr E.D., Lorente G., Kuo J., Ram R., Nikolich K. & Urfer R. Targeting of the receptor
protein tyrosine phosphatase beta with a monoclonal antibody delays tumor growth
in a glioblastoma model. Cancer Res, 2006, 66, 2271-2278.
Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med, 1971, 285, 11821186.
Fukada M., Fujikawa A., Chow J.P., Ikematsu S., Sakuma S. & Noda M. Protein tyrosine
phosphatase receptor type Z is inactivated by ligand-induced oligomerization.
FEBS Lett, 2006, 580, 4051-4056.
Gallagher J.T. & Walker A. Molecular distinctions between heparan sulphate and heparin.
Analysis of sulphation patterns indicates that heparan sulphate and heparin are
separate families of N-sulphated polysaccharides. Biochem J, 1985, 230, 665-674.
Gallagher J.T., Lyon M. & Steward W.P. Structure and function of heparan sulphate
proteoglycans. Biochem J, 1986, 236, 313-325.
Garver R.I., Jr., Radford D.M., Donis-Keller H., Wick M.R. & Milner P.G. Midkine and
pleiotrophin expression in normal and malignant breast tissue. Cancer, 1994, 74,
1584-1590.
Gospodarowicz D., Neufeld G. & Schweigerer L. Fibroblast growth factor: structural and
biological properties. J Cell Physiol Suppl, 1987, Suppl 5, 15-26.
Goustin A.S., Leof E.B., Shipley G.D. & Moses H.L. Growth factors and cancer. Cancer
Res, 1986, 46, 1015-1029.
Gray G.R. Preparation of antigens by coupling carbohydrates to proteins by reductive
amination with cyanoborohydride. Methods Enzymol, 1978, 50, 155-160.
Grzelinski M., Bader N., Czubayko F. & Aigner A. Ribozyme-targeting reveals the ratelimiting role of pleiotrophin in glioblastoma. Int J Cancer, 2005, 117, 942-951.
Guo N.H., Krutzsch H.C., Negre E., Zabrenetzky V.S. & Roberts D.D. Heparin-binding
peptides from the type I repeats of thrombospondin. Structural requirements for
heparin binding and promotion of melanoma cell adhesion and chemotaxis. Journal
of Biological Chemistry, 1992, 267, 19349-19355.
Hamma-Kourbali Y., Bernard-Pierrot I., Heroult M., Dalle S., Caruelle D., Milhiet P.E.,
Fernig D.G., Delbe J. & Courty J. Inhibition of the mitogenic, angiogenic and
201
REFERENCES
tumorigenic activities of pleiotrophin by a synthetic peptide corresponding to its Cthrombospondin repeat-I domain. J Cell Physiol, 2007,
Hampton B.S., Marshak D.R. & Burgess W.H. Structural and functional characterization
of full-length heparin-binding growth associated molecule. Molecular Biology of
the Cell, 1992, 3, 85-93.
Hatziapostolou M., Delbe J., Katsoris P., Polytarchou C., Courty J. & Papadimitriou E.
Heparin affin regulatory peptide is a key player in prostate cancer cell growth and
angiogenicity. Prostate, 2005, 65, 151-158.
Hatziapostolou M., Polytarchou C., Katsoris P., Courty J. & Papadimitriou E. Heparin
affin regulatory peptide/pleiotrophin mediates fibroblast growth factor 2
stimulatory effects on human prostate cancer cells. J Biol Chem, 2006, 281, 3221732226.
Heroult M., Bernard-Pierrot I., Delbe J., Hamma-Kourbali Y., Katsoris P., Barritault D.,
Papadimitriou E., Plouet J. & Courty J. Heparin affin regulatory peptide binds to
vascular endothelial growth factor (VEGF) and inhibits VEGF-induced
angiogenesis. Oncogene, 2004, 23, 1745-1753.
Herradon G., Ezquerra L., Nguyen T., Vogt T.F., Bronson R., Silos-Santiago I. & Deuel
T.F. Pleiotrophin is an important regulator of the renin-angiotensin system in
mouse aorta. Biochem Biophys Res Commun, 2004, 324, 1041-1047.
Herradon G., Ezquerra L., Nguyen T., Silos-Santiago I. & Deuel T.F. Midkine regulates
pleiotrophin organ-specific gene expression: evidence for transcriptional regulation
and functional redundancy within the pleiotrophin/midkine developmental gene
family. Biochem Biophys Res Commun, 2005, 333, 714-721.
Hienola A., Pekkanen M., Raulo E., Vanttola P. & Rauvala H. HB-GAM inhibits
proliferation and enhances differentiation of neural stem cells. Mol Cell Neurosci,
2004, 26, 75-88.
Horiba M., Kadomatsu K., Nakamura E., Muramatsu H., Ikematsu S., Sakuma S., Hayashi
K., Yuzawa Y., Matsuo S., Kuzuya M., Kaname T., Hirai M., Saito H. &
Muramatsu T. Neointima formation in a restenosis model is suppressed in midkinedeficient mice. J Clin Invest., 2000, 105, 489-495.
Huber D., Gautschi-Sova P. & Bohlen P. Amino-terminal sequences of a novel heparinbinding protein from human, bovine, rat, and chick brain: high interspecies
homology. Neurochem Res, 1990, 15, 435-439.
Hulmes J.D., Seddon A.P., Decker M.M. & Bohlen P. Comparison of the disulfide bond
arrangements of human recombinant and bovine brain heparin binding neuritepromoting factors. Biochem Biophys Res Commun, 1993, 192, 738-746.
Iakoucheva L.M., Brown C.J., Lawson J.D., Obradovic Z. & Dunker A.K. Intrinsic
disorder in cell-signaling and cancer-associated proteins. J Mol Biol, 2002, 323,
573-584.
Imai S., Kaksonen M., Raulo E., Kinnunen T., Fages C., Meng X.J., Lakso M. & Rauvala
H. Osteoblast recruitment and bone formation enhanced by cell matrix-associated
heparin-binding growth-associated molecule (HB-GAM). J Cell Biol, 1998, 143,
1113-1128.
Iwasaki W., Nagata K., Hatanaka H., Inui T., Kimura T., Muramatsu T., Yoshida K.,
Tasumi M. & Inagaki F. Solution structure of midkine, a new heparin-binding
growth factor. Embo J, 1997, 16, 6936-6946.
Jacobson M. Developmental Neurobiology 1978, Plenum, New York,
Jager R., Noll K., Havemann K., Knabbe C., Rauvala H. & Zugmaier G. Differential
expression and biological activity of the heparin-binding growth-associated
molecule (HB-GAM) in lung cancer cell lines. Int J Cancer, 1997, 73, 537-543.
202
REFERENCES
Jager R., List B., Knabbe C., Souttou B., Raulais D., Zeiler T., Wellstein A., Aigner A.,
Neubauer A. & Zugmaier G. Serum levels of the angiogenic factor pleiotrophin in
relation to disease stage in lung cancer patients. Br J Cancer, 2002, 86, 858-863.
Jung C.G., Hida H., Nakahira K., Ikenaka K., Kim H.J. & Nishino H. Pleiotrophin mRNA
is highly expressed in neural stem (progenitor) cells of mouse ventral
mesencephalon and the product promotes production of dopaminergic neurons
from ebryonic stem cell-derived nestin-positive cells. The FASEB Journal, 2004,
18, 1237-1239.
Kadomatsu K., Tomomura M. & Muramatsu T. cDNA cloning and sequencing of a new
gene intensely expressed in early differentiation stages of embryonal carcinoma
cells and in mid- gestation period of mouse embryogenesis. Biochem Biophys Res
Commun, 1988, 151, 1312-1318.
Kadomatsu K., Hagihara M., Akhter S., Fan Q.W., Muramatsu H. & Muramatsu T.
Midkine induces the transformation of NIH3T3 cells. Br J of Cancer, 1997, 75,
354-359.
Kaname T., Kuwano A., Murano I., Uehara K., Muramatsu T. & Kajii T. Midkine gene
(MDK), a gene for prenatal differentiation and neuroregulation, maps to band
11p11.2 by fluorescence in situ hybridization. Genomics, 1993, 17, 514-515.
Kaneda N., Talukder A.H., Nishiyama H., Koizumi S. & Muramatsu T. Midkine, a
heparin-binding growth/differentiation factor, exhibits nerve cell adhesion and
guidance activity for neurite outgrowth in vitro. J Biochem (Tokyo), 1996, 119,
1150-1156.
Ke Y., Wilkinson M.C., Fernig D.G., Smith J.A. & Rudland P.S. A rapid procedure for
production of human basic fibroblast growth factor in E. coli cells. Biochim
Biophys Acta, 1992, 1131, 307-310.
Kilpelainen I., Kaksonen M., Avikainen H., Fath M., Linhardt R.J., Raulo E. & Rauvala H.
Heparin-binding growth-associated molecule contains two heparin-binding beta sheet domains that are homologous to the thrombospondin type I repeat. J Biol
Chem, 2000, 275, 13564-13570.
Kinnunen T., Raulo E., Nolo R., Maccarana M., Lindahl U. & Rauvala H. Neurite
outgrowth in brain neurons induced by heparin-binding growth-associated
molecule (HB-GAM) depends on the specific interaction of HB-GAM with
heparan sulfate at the cell surface. J Biol Chem, 1996, 271, 2243-2248.
Kinnunen T., Kaksonen M., Saarinen J., Kalkkinen N., Peng H.B. & Rauvala H. CortactinSrc kinase signaling pathway is involved in N-syndecan- dependent neurite
outgrowth. J Biol Chem, 1998, 273, 10702-10708.
Kinnunen A., Niemi M., Kinnunen T., Kaksonen M., Nolo R. & Rauvala H. Heparan
sulphate and HB-GAM (heparin-binding growth-associated molecule) in the
development of the thalamocortical pathway of rat brain [In Process Citation]. Eur
J Neurosci, 1999, 11, 491-502.
Kjellen L. & Lindahl U. Proteoglycans: Structures and interactions. Ann Rev Biochem,
1991, 60, 443-475.
Klagsbrun M. The affinity of fibroblast growth factors (FGFs) for heparin; FGF-heparan
sulfate interactions in cells and extracellular matrix. Curr Opin Cell Biol, 1990, 2,
857-863.
Klomp H.J., Zernial O., Flachmann S., Wellstein A. & Juhl H. Significance of the
expression of the growth factor pleiotrophin in pancreatic cancer patients. Clin
Cancer Res, 2002, 8, 823-827.
Kojima S., Muramatsu H., Amanuma H. & Muramatsu T. Midkine enhances fibrinolytic
activity of bovine endothelial cells. J Biol Chem, 1995a, 270, 9590-9596.
203
REFERENCES
Kojima S., Inui T., Kimura T., Sakakibara S., Muramatsu H., Amanuma H., Maruta H. &
Muramatsu T. Synthetic peptides derived from midkine enhance plasminogen
activator activity in bovine aortic endothelial cells. Biochemical & Biophysical
Research Communications, 1995b, 206, 468-473.
Kojima T., Katsumi A., Yamazaki T., Muramatsu T., Nagasaka T., Ohsumi K. & Saito H.
Human ryudocan from endothelium-like cells binds basic fibroblast growth factor,
midkine, and tissue factor pathway inhibitor. J Biol Chem, 1996, 271, 5914-5920.
Kojima S., Inui T., Muramatsu H., Suzuki Y., Kadomatsu K., Yoshizawa M., Hirose S.,
Kimura T., Sakakibara S. & Muramatsu T. Dimerization of midkine by tissue
transglutaminase and its functional implication. J Biol Chem, 1997, 272, 94109416.
Kovesdi I., Fairhurst J.L., Kretschmer P.J. & Bohlen P. Heparin-binding neurotrophic
factor (HBNF) and MK, members of a new family of homologous,
developmentally regulated proteins. Biochemical & Biophysical Research
Communications, 1990, 172, 850-854.
Kretschmer P.J., Fairhurst J.L., Decker M.M., Chan C.P., Gluzman Y., Bohlen P. &
Kovesdi I. Cloning, characterization and developmental regulation of two members
of a novel human gene family of neurite outgrowth-promoting proteins. Growth
Factors, 1991, 5, 99-114.
Kretschmer P.J., Fairhurst J.L., Hulmes J.D., Popjes M.L., Bohlen P. & Kovesdi I.
Genomic organization of the human HBNF gene and characterization of an HBNF
variant protein as a splice mutant. Biochemical & Biophysical Research
Communications, 1993, 192, 420-429.
Kuo M.D., Oda Y., Huang J.S. & Huang S.S. Amino acid sequence and characterization of
a heparin-binding neurite- promoting factor (p18) from bovine brain. J Biol Chem,
1990, 265, 18749-18752.
Kuo M.D., Huang S.S. & Huang J.S. Characterization of heparin-binding growthassociated factor receptor on NIH 3T3 cells. Biochemical & Biophysical Research
Communications, 1992, 182, 188-194.
Kurtz A., Schulte A.M. & Wellstein A. Pleiotrophin and midkine in normal development
and tumor biology. Crit Rev Oncog, 1995, 6, 151-177.
Laaroubi K., Delbe J., Vacherot F., Desgranges P., Tardieu M., Jaye M., Barritault D. &
Courty J. Mitogenic and in vitro angiogenic activity of human recombinant heparin
affin regulatory peptide. Growth Factors, 1994, 10, 89-98.
Laemmli U.K. Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature, 1970, 227, 680-685.
Lai S., Czubayko F., Riegel A.T. & Wellstein A. Structure of the human heparin-binding
growth factor gene pleiotrophin. Biochem Biophys Res Commun, 1992, 187, 11131122.
Lametsch R., Rasmussen J.T., Johnsen L.B., Purup S., Sejrsen K., Petersen T.E. &
Heegaard C.W. Structural characterization of the fibroblast growth factor-binding
protein purified from bovine prepartum mammary gland secretion. J Biol Chem,
2000, 275, 19469-19474.
Lander A.D. Proteoglycans: master regulators of molecular encounter? Matrix Biol, 1998,
17, 465-472.
Lauri S.E., Rauvala H., Kaila K. & Taira T. Effect of heparin-binding growth-associated
molecule (HB-GAM) on synaptic transmission and early LTP in rat hippocampal
slices. Eur J Neurosci, 1998, 10, 188-194.
Ledoux D., Caruelle D., Sabourin C., Liu J., Crepin M., Barritault D. & Courty J. Cellular
distribution of the angiogenic factor heparin affin regulatory peptide (HARP)
204
REFERENCES
mRNA and protein in the human mammary gland. J. Histochem. Cytochem., 1997,
45, 1-7.
Lee C.C., Kreusch A., McMullan D., Ng K. & Spraggon G. Crystal structure of the human
neuropilin-1 b1 domain. Structure, 2003, 11, 99-108.
Lee H.H., Norris A., Weiss J.B. & Frasch M. Jelly belly protein activates the receptor
tyrosine kinase Alk to specify visceral muscle pioneers. Nature, 2003, 425, 507512.
Levy J.B., Canoll P.D., Silvennoinen O., Barnea G., Morse B., Honegger A.M., Huang
J.T., Cannizzaro L.A., Park S.H., Druck T. & et al. The cloning of a receptor-type
protein tyrosine phosphatase expressed in the central nervous system. J Biol Chem,
1993, 268, 10573-10581.
Li Y.S., Milner P.G., Chauhan A.K., Watson M.A., Hoffman R.M., Kodner C.M.,
Milbrandt J. & Deuel T.F. Cloning and expression of a developmentally regulated
protein that induces mitogenic and neurite outgrowth activity. Science, 1990, 250,
1690-1694.
Li Y.S., Hoffman R.M., Le Beau M.M., Espinosa R.d., Jenkins N.A., Gilbert D.J.,
Copeland N.G. & Deuel T.F. Characterization of the human pleiotrophin gene.
Promoter region and chromosomal localization. Journal of Biological Chemistry,
1992a, 267, 26011-26016.
Li Y.S., Gurrieri M. & Deuel T.F. Pleiotrophin gene expression is highly restricted and is
regulated by platelet-derived growth factor. Biochemical & Biophysical Research
Communications, 1992b, 184, 427-432.
Li Y.S. & Deuel T.F. Pleiotrophin stimulates tyrosine phosphorylation in NIH 3T3 and
NB41A3 cells. Biochemical & Biophysical Research Communications, 1993, 195,
1089-1095.
Li G., Hu Y., Huo Y., Liu M., Freeman D., Gao J., Liu X. & Wu D. PTEN deletion leads
to up-reguation of a secreted growth factor pleiotrophin. Journal of Biological
Chemistry, 2006, 16, 10663-10668.
Liedert A., Kaspar D., Claes L. & Ignatius A. Signal transduction pathways involved in
mechanical regulation of HB-GAM expression in osteoblastic cells. Biochem
Biophys Res Commun, 2006, 342, 1070-1076.
Lorente G., Nelson A., Mueller S., Kuo J., Urfer R., Nikolich K. & Foehr E.D. Functional
comparison of long and short splice forms of RPTPbeta: implications for
glioblastoma treatment. Neuro Oncol, 2005, 7, 154-163.
Lu K.V., Jong K.A., Kim G.Y., Singh J., Dia E.Q., Yoshimoto K., Wang M.Y., Cloughesy
T.F., Nelson S.F. & Mischel P.S. Differential induction of glioblastoma migration
and growth by two forms of pleiotrophin. J Biol Chem, 2005, 280, 26953-26964.
Lyon M. and Gallagher J.T. Heparan Sulphate: Molecular structure and interaction with
growth factors and morphogens. In R. V. Iozzo (ed.) Proteoglycans: Structure,
Biology and Molecular Interactions, 2000, Marcel Dekker, New York, 27-60.
Mac Gabhann F. & Popel A.S. Differential binding of VEGF isoforms to VEGF receptor 2
in the presence of neuropilin-1: A computational model. Am J Physiol Heart Circ
Physiol, 2005, 288, 2851–2860.
Maeda N., Nishiwaki T., Shintani T., Hamanaka H. & Noda M.
6B4
proteoglycan/phosphacan, an extracellular variant of receptor-like protein-tyrosine
phosphatase zeta/RPTPbeta, binds pleiotrophin/heparin-binding growth-associated
molecule (HB-GAM). Journal of Biological Chemistry, 1996, 271, 21446-21452.
Maeda N.and.Noda M. Involvement of receptor-like protein tyrosine phosphatase
zeta/RPTP beta and its ligand pleiotrophin/heparin-binding growth-associated
molecule (HB-GAM) in neuronal migration. J. Cell. Biol., 1998, 142, 203-216.
205
REFERENCES
Maeda N., Ichihara-Tanaka K., Kimura T., Kadomatsu K., Muramatsu T. & Noda M. A
receptor-like protein-tyrosine phosphatase PTPzeta/RPTPbeta binds a heparinbinding growth factor midkine. Involvement of arginine 78 of midkine in the high
affinity binding to PTPzeta. J Biol Chem, 1999, 274, 12474-12479.
Maeda N., He J., Yajima Y., Mikami T., Sugahara K. & Yabe T. Heterogeneity of the
chondroitin sulfate portion of phosphacan/6B4 proteoglycan regulates its binding
affinity for pleiotrophin/heparin binding growth-associated molecule. J Biol Chem,
2003, 278, 35805-35811.
Maeda N., Fukazawa N. & Hata T. The binding of chondroitin sulfate to
pleiotrophin/heparin-binding growth-associated molecule is regulated by chain
length and oversulfated structures. J Biol Chem, 2006, 281, 4894-4902.
Mailleux P., Vanderwinden J.M. & Vanderhaeghen J.J. The new growth factor
pleiotrophin (HB-GAM) mRNA is selectively present in the meningothelial cells of
human meningiomas. Neurosci Lett, 1992, 142, 31-35.
Matsumoto K., Wanaka A., Takatsuji K., Muramatsu H., Muramatsu T. & Tohyama M. A
novel family of heparin-binding growth factors, pleiotrophin and midkine, is
expressed in the developing rat cerebral cortex. Brain Res Dev Brain Res, 1994, 79,
229-241.
Meng K., Rodriguez-Pena A., Dimitrov T., Chen W., Yamin M., Noda M. & Deuel T.F.
Pleiotrophin signals increased tyrosine phosphorylation of beta beta-catenin
through inactivation of the intrinsic catalytic activity of the receptor-type protein
tyrosine phosphatase beta/zeta. Proc Natl Acad Sci U S A, 2000, 97, 2603-2608.
Merenmies J. and Rauvala H. Molecular cloning of the 18-kDa growth-associated protein
of developing brain. Journal of Biological Chemistry, 1990, 265, 16721-16724.
Merrifield R.B. Solid Phase Peptide Synthesis. I. The synthesis of a tetrapeptide. J Am
Chem Soc, 1963, 85, 2149-2154.
Michikawa M., Kikuchi S., Muramatsu H., Muramatsu T. & Kim S.U. Retinoic acid
responsive gene product, midkine, has neurotrophic functions for mouse spinal
cord and dorsal root ganglion neurons in culture. J Neurosci Res, 1993b, 35, 530539.
Michikawa M., Xu R.Y., Muramatsu H., Muramatsu T. & Kim S.U. Midkine is a mediator
of retinoic acid induced neuronal differentiation of embryonal carconoma cells.
Biochem Biophys Res Commun, 1993a, 192, 1312-1318.
Milev P., Chiba A., Haring M., Rauvala H., Schachner M., Ranscht B., Margolis R.K. &
Margolis R.U. High affinity binding and overlapping localization of neurocan and
phosphacan/protein-tyrosine phosphatase-zeta/beta with tenascin-R, amphoterin,
and the heparin-binding growth-associated molecule. J Biol Chem, 1998, 273,
6998-7005.
Milhiet P.E., Vacherot F., Caruelle J.P., Barritault D., Caruelle D. & Courty J.
Upregulation of the angiogenic factor heparin affin regulatory peptide by
progesterone in rat uterus. J Endocrinol, 1998, 158, 389-399.
Milner P.G., Li Y.S., Hoffman R.M., Kodner C.M., Siegel N.R. & Deuel T.F. A novel 17
kD heparin-binding growth factor (HBGF-8) in bovine uterus: purification and Nterminal amino acid sequence. Biochemical & Biophysical Research
Communications, 1989, 165, 1096-1103.
Milner P.G., Shah D., Veile R., Donis-Keller H. & Kumar B.V. Cloning, nucleotide
sequence, and chromosome localization of the human pleiotrophin gene.
Biochemistry, 1992, 31, 12023-12028.
Mitsiadis T.A., Salmivirta M., Muramatsu T., Muramatsu H., Rauvala H., Lehtonen E.,
Jalkanen M. & Thesleff I. Expression of the heparin-binding cytokines, midkine
206
REFERENCES
(MK) and HB-GAM (pleiotrophin) is associated with epithelial-mesenchymal
interactions during fetal development and organogenesis. Development, 1995, 121,
37-51.
Miyake I., Hakomori Y., Shinohara A., Gamou T., Saito M., Iwamatsu A. & Sakai R.
Activation of anaplastic lymphoma kinase is responsible for hyperphosphorylation
of ShcC in neuroblastoma cell lines. Oncogene, 2002, 21, 5823-5834.
Moog-Lutz C., Degoutin J., Gouzi J.Y., Frobert Y., Brunet-de Carvalho N., Bureau J.,
Creminon C. & Vigny M. Activation and inhibition of anaplastic lymphoma kinase
receptor tyrosine kinase by monoclonal antibodies and absence of agonist activity
of pleiotrophin. J Biol Chem, 2005, 280, 26039-26048.
Morris S.W., Naeve C., Mathew P., James P.L., Kirstein M.N., Cui X. & Witte D.P. ALK,
the chromosome 2 gene locus altered by the t(2;5) in non-Hodgkin's lymphoma,
encodes a novel neural receptor tyrosine kinase that is highly related to leukocyte
tyrosine kinase (LTK). Oncogene, 1997, 14, 2175-2188.
Motegi A., Fujimoto J., Kotani M., Sakuraba H. & Yamamoto T. ALK receptor tyrosine
kinase promotes cell growth and neurite outgrowth. J Cell Sci, 2004, 117, 33193329.
Muller S., Kunkel P., Lamszus K., Ulbricht U., Lorente G., Nelson A., von Schack D.,
Chin D.J., Lohr S.C., Westphal M. & Melcher T. A role for receptor tyrosine
phosphatase zeta in glioma cell migration. Oncogene, 2003, 22, 6661-6668.
Muramatsu H., Shirahama H., Yonezawa S., Maruta H. & Muramatsu T. Midkine, a
retinoic acid-inducible growth/differentiation factor: immunochemical evidence for
the function and distribution. Dev Biol, 1993, 159, 392-402.
Muramatsu H., Zou K., Sakaguchi N., Ikematsu S., Sakuma S. & Muramatsu T. LDL
receptor-related protein as a component of the midkine receptor. Biochem Biophys
Res Commun, 2000, 270, 936-941.
Muramatsu H., Zou P., Suzuki H., Oda Y., Chen G.Y., Sakaguchi N., Sakuma S., Maeda
N., Noda M., Takada Y. & Muramatsu T. alpha4beta1- and alpha6beta1-integrins
are functional receptors for midkine, a heparin-binding growth factor. J Cell Sci,
2004, 117, 5405-5415.
Muramatsu T. Midkine and pleiotrophin: two related proteins involved in development,
survival, inflammation and tumorigenesis. J Biochem (Tokyo), 2002, 132, 359-371.
Najjam S., Gibbs R.V., Gordon M.Y. & Rider C.C. Characterization of human
recombinant interleukin 2 binding to heparin and heparan sulfate using an ELISA
approach. Cytokine, 1997, 9, 1013-1022.
Nakagawara A., Milbrandt J., Muramatsu T., Deuel T.F., Zhao H., Cnaan A. & Brodeur
G.M. Differential expression of pleiotrophin and midkine in advanced
neuroblastomas. Cancer Res, 1995, 55, 1792-1797.
Nakamoto M., Matsubara S., Miyauchi T., Obama H., Ozawa M. & Muramatsu T. A new
family of heparin binding growth/differentiation factors: differential expression of
the midkine (MK) and HB-GAM genes during mouse development. J Biochem
(Tokyo), 1992, 112, 346-349.
Nakanishi T., Kadomatsu K., Okamoto T., Ichihara-Tanaka K., Kojima T., Saito H.,
Tomoda Y. & Muramatsu T. Expression of syndecan-1 and -3 during
embryogenesis of the central nervous system in relation to binding with midkine. J
Biochem (Tokyo), 1997a, 121, 197-205.
Neame P.J., Young C.N., Brock C.W., Treep J.T., Ganey T.M., Sasse J. & Rosenberg L.C.
Pleiotrophin is an abundant protein in dissociative extracts of bovine fetal
epiphyseal cartilage and nasal cartilage from newborns. Journal of Orthopaedic
Research, 1993, 11, 479-491.
207
REFERENCES
Neuhoff V., Stamm R. & Eibl H. Improved staining of proteins in polyacrylamide gels
including isoelectric focusing gels with clear background at nanogram sensitivity
using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis, 1988, 6, 255262.
Nishiwaki T., Maeda N. & Noda M. Characterization and developmental regulation of
proteoglycan-type protein tyrosine phosphatase zeta/RPTPbeta isoforms. J
Biochem (Tokyo), 1998, 123, 458-467.
Nolo R., Kaksonen M., Raulo E. & Rauvala H. Co-expression of heparin-binding growth
associated molecule (HB-GAM) and N-syndecan (syndecan-3) in developing rat
brain. Neurosci Lett, 1995, 191, 39-42.
Novotny W.F., Maffi T., Mehta R.L. & Milner P.G. Identification of novel heparinreleasable proteins, as well as the cytokines midkine and pleiotrophin, in human
postheparin plasma. Arteriosclerosis & Thrombosis, 1993, 13, 1798-1805.
Ohyama Y., Miyamoto K., Minamino N. & Matsuo H. Isolation and identification of
midkine and pleiotrophin in bovine follicular fluid. Mol Cell Endocrinol, 1994,
105, 203-208.
Ornitz D.M., Yayon A., Flanagan J.G., Svahn C.M., Levi E. & Leder P. Heparin is
required for cell-free binding of basic fibroblast growth factor to a soluble receptor
and for mitogenesis in whole cells. Mol Cell Biol, 1992, 12, 240-247.
Pariser H., Ezquerra L., Herradon G., Perez-Pinera P. & Deuel T.F. Fyn is a downstream
target of the pleiotrophin/receptor protein tyrosine phosphatase beta/zeta-signaling
pathway: regulation of tyrosine phosphorylation of Fyn by pleiotrophin. Biochem
Biophys Res Commun, 2005a, 332, 664-669.
Pariser H., Perez-Pinera P., Ezquerra L., Herradon G. & Deuel T.F. Pleiotrophin stimulates
tyrosine phosphorylation of beta-adducin through inactivation of the
transmembrane receptor protein tyrosine phosphatase beta/zeta. Biochem Biophys
Res Commun, 2005b, 335, 232-239.
Pavlov I., Voikar V., Kaksonen M., Lauri S.E., Hienola A., Taira T. & Rauvala H. Role of
heparin-binding growth-associated molecule (HB-GAM) in hippocampal LTP and
spatial learning revealed by studies on overexpressing and knockout mice. Mol Cell
Neurosci, 2002, 20, 330-342.
Pellegrini L., Burke D.F., von Delft F., Mulloy B. & Blundell T.L. Crystal structure of
fibroblast growth factor receptor ectodomain bound to ligand and heparin. Nature,
2000, 407, 1029-1034.
Peng H.B., Ali A.A., Dai Z., Daggett D.F., Raulo E. & Rauvala H. The role of heparinbinding growth-associated molecule (HB-GAM) in the postsynaptic induction in
cultured muscle cells. J Neurosci, 1995, 15, 3027-3038.
Pepper M.S. Manipulating angiogenesis. Arterio Thromb Vasc Biol, 1997, 17, 605-619.
Perez-Pinera P., Garcia-Suarez O., Menendez-Rodriguez P., Mortimer J., Chang Y.,
Astudillo A. & Deuel T.F. The receptor protein tyrosine phosphatase
(RPTP)beta/zeta is expressed in different subtypes of human breast cancer.
Biochem Biophys Res Commun, 2007a, 362, 5-10.
Perez-Pinera P., Zhang W., Chang Y., Vega J.A. & Deuel T.F. Anaplastic lymphoma
kinase (ALK) is activated through the pleiotrophin (PTN)/receptor protein tyrosine
phosphatase (RPTP)beta /zeta signaling pathway: An "alternative mechanism of
receptor tyrosine kinase (RTK) activation". J Biol Chem, 2007b,
Petersen W., Wildemann B., Pufe T., Raschke M. & Schmidmaier G. The angiogenic
peptide pleiotrophin (PTN/HB-GAM) is expressed in fracture healing: an
immunohistochemical study in rats. Arch Orthop Trauma Surg, 2004, 124, 603607.
208
REFERENCES
Plank M.J. & Sleeman B.D. Tumour-induced angiogenesis. J Theor Med, 2004, 5, 137153.
Plotnikov A.N., Schlessinger J., Hubbard S.R. & Mohammadi M. Structural basis for FGF
receptor dimerization and activation. Cell, 1999, 98, 641-650.
Plouët J., Schilling J. & Gospodarowicz D. Isolation and characterization of a newly
identified endothelial cell mitogen produced by AtT-20 cells. EMBO J, 1989, 8,
3801-3806.
Polykratis A., Katsoris P., Courty J. & Papadimitriou E. Characterization of heparin affin
regulatory peptide signaling in human endothelial cells. J Biol Chem, 2005, 280,
22454-22461.
Polytarchou C., Hatziapostolou M. & Papadimitriou E. Hydrogen peroxide stimulates
proliferation and migration of human prostate cancer cells through activation of
activator protein-1 and up-regulation of the heparin affin regulatory peptide gene. J
Biol Chem, 2005, 280, 40428-40435.
Poulsen F.R., Lagord C., Courty J., Pedersen E.B., Barritault D. & Finsen B. Increased
synthesis of heparin affin regulatory peptide in the perforant path lesioned mouse
hippocampal formation. Exp Brain Res, 2000, 135, 319-330.
Powell A.K., Yates E.A., Fernig D.G. & Turnbull J.E. Interactions of heparin/heparan
sulfate with proteins: appraisal of structural factors and experimental approaches.
Glycobiology, 2004, 14, 17R-30R.
Powers C., Aigner A., Stoica G.E., McDonnell K. & Wellstein A. Pleiotrophin signaling
through anaplastic lymphoma kinase (ALK) is rate- limiting for glioblastoma
growth. J Biol Chem, 2002, 23, 23.
Pulford K., Lamant L., Morris S.W., Butler L.H., Wood K.M., Stroud D., Delsol G. &
Mason D.Y. Detection of anaplastic lymphoma kinase (ALK) and nucleolar protein
nucleophosmin (NPM)-ALK proteins in normal and neoplastic cells with the
monoclonal antibody ALK1. Blood, 1997, 89, 1394-1404.
Pulford K., Morris S.W. & Turturro F. Anaplastic lymphoma kinase proteins in growth
control and cancer. J Cell Physiol, 2004, 199, 330-358.
Qi M., Ikematsu S., Maeda N., Ichihara-Tanaka K., Sakuma S., Noda M., Muramatsu T. &
Kadomatsu K. Haptotactic migration induced by midkine. Involvement of proteintyrosine
phosphatase
zeta.
Mitogen-activated
protein
kinase,
and
phosphatidylinositol 3-kinase. J Biol Chem, 2001, 276, 15868-15875.
Raulais D., Lagente-Chevallier O., Guettet C., Duprez D., Courtois Y. & Vigny M. A new
heparin binding protein regulated by retinoic acid from chick embryo. Biochemical
& Biophysical Research Communications, 1991, 174, 708-715.
Raulo E., Chernousov M.A., Carey D.J., Nolo R. & Rauvala H. Isolation of a neuronal cell
surface receptor of heparin binding growth-associated molecule (HB-GAM).
Identification as N-syndecan (syndecan-3). Journal of Biological Chemistry, 1994,
269, 12999-13004.
Raulo E., Julkunen I., Merenmies J., Pihlaskari R. & Rauvala H. Secretion and biological
activities of heparin-binding growth-associated molecule. Neurite outgrowthpromoting and mitogenic actions of the recombinant and tissue-derived protein.
Journal of Biological Chemistry, 1992, 267, 11408-11416.
Raulo E., Tumova S., Pavlov I., Hienola A., Klankki E., Kalkkinen N., Taira T.,
Kilpelainen I. & Rauvala H. The two thrombospondin type I repeat domains of the
heparin-binding growth-associated molecule bind to heparin/heparan sulfate and
regulate neurite extension and plasticity in hippocampal neurons. J Biol Chem,
2005, 280, 41576-41583.
209
REFERENCES
Raulo E., Tumova S., Pavlov I., Hienola A., Lauri S., Taira T. & Rauvala H. The Two
Thrombospondin Type I Repeat Domains of HB-GAM display a cooperative
function in N-syndecan binding and regulation of synaptic plasticity.
TheScientificWorldJournal, 2006, 6, 406-409.
Rauvala H. An 18-kd heparin-binding protein of developing brain that is distinct from
fibroblast growth factors. EMBO Journal, 1989, 8, 2933-2941.
Rauvala H., Vanhala A., Castren E., Nolo R., Raulo E., Merenmies J. & Panula P.
Expression of HB-GAM (heparin-binding growth-associated molecules) in the
pathways of developing axonal processes in vivo and neurite outgrowth in vitro
induced by HB-GAM. Brain Research. Developmental Brain Research, 1994, 79,
157-176.
Rich R.L. & Myszka D.G. Survey of the year 2005 commercial optical biosensor literature.
J Mol Recognit, 2006, 19, 478-534.
Riegel A.T. & Wellstein A. The potential role of the heparin-binding growth factor
pleiotrophin in breast cancer. Breast Cancer Res Treat, 1994, 31, 309-314.
Roger J., Brajeul V., Thomasseau S., Hienola A., Sahel J.A., Guillonneau X. & Goureau
O. Involvement of Pleiotrophin in CNTF-mediated differentiation of the late retinal
progenitor cells. Dev Biol, 2006, 298, 527-539.
Rudd T.R., Guimond S.E., Skidmore M.A., Duchesne L., Guerrini M., Torri G., Cosentino
C., Brown A., Clarke D.T., Turnbull J.E., Fernig D.G. & Yates E.A. Influence of
substitution pattern and cation binding on conformation and activity in heparin
derivatives. Glycobiology, 2007, 17, 983-993.
Rumsby M., Suggitt F., Haynes L., Hughson E., Kidd D. & McNulty S. Substratum of
pleiotrophin (HB-GAM) stimulates rat CG-4 line oligodendrocytes to adopt a
bipolar morphology and disperse: primary O- 2A progenitor glial cells disperse
similarly on pleiotrophin. Glia, 1999, 26, 361-367.
Ruoslahti E. Proteoglycans in cell regulation. J Biol Chem, 1989, 264, 13369-13372.
Russell P.J., Bennett S. & Stricker P. Growth factor involvement in progression of prostate
cancer. Clin Chem, 1998, 44, 705-723.
Said E.A., Courty J., Svab J., Delbe J., Krust B. & Hovanessian A.G. Pleiotrophin inhibits
HIV infection by binding the cell surface-expressed nucleolin. Febs J, 2005, 272,
4646-4659.
Sakaguchi N., Muramatsu H., Ichihara-Tanaka K., Maeda N., Noda M., Yamamoto T.,
Michikawa M., Ikematsu S., Sakuma S. & Muramatsu T. Receptor-type protein
tyrosine phosphatase? as a component of the signaling receptor complex for
midkine-dependent survival of embryonic neurons. Neurosci Res, 2003, 45, 219224.
Salama R.H., Muramatsu H., Zou K., Inui T., Kimura T. & Muramatsu T. Midkine binds
to 37-kDa laminin binding protein precursor, leading to nuclear transport of the
complex. Exp Cell Res, 2001, 270, 13-20.
Sato H., Funahashi M., Kristensen D.B., Tateno C. & Yoshizato K. Pleiotrophin as a Swiss
3T3 cell-derived potent mitogen for adult rat hepatocytes. Exp Cell Res, 1999, 246,
152-164.
Sato M., Miyado K., Okazaki M. & Kimura M. Regulation of osteoblast-specific factor-1
(OSF-1) mRNA expression by dual promoters as revealed by RT-PCR. Biochem
Biophys Res Commun, 1997, 238, 831-837.
Schlessinger J., Lax I. & Lemmon M. Regulation of growth factor activation by
proteoglycans: what is the role of the low affinity receptors? Cell, 1995, 83, 357360.
210
REFERENCES
Schlessinger J., Plotnikov A.N., Ibrahimi O.A., Eliseenkova A.V., Yeh B.K., Yayon A.,
Linhardt R.J. & Mohammadi M. Crystal structure of a ternary FGF-FGFR-heparin
complex reveals a dual role for heparin in FGFR binding and dimerization. Mol
Cell, 2000, 6, 743-750.
Schlessinger J. Cell signaling by receptor tyrosine kinases. Cell, 2000, 103, 211-225.
Schnolzer M., Alewood P., Jones A., Alewood D. & Kent S.B.H. In situ neutralization in
Boc-chemistry solid phase peptide synthesis. Int J Peptide Protein Res, 1992, 40,
180-193.
Schuck P. Use of surface plasmon resonance to probe the equilibrium and dynamic aspects
of interactions between biological macromolecules Annu Rev Biophys Biomol
Struct, 1997, 26, 541-566.
Schulte A.M., Lai S., Kurtz A., Czubayko F., Riegel A.T. & Wellstein A. Human
trophoblast and choriocarcinoma expression of the growth factor pleiotrophin
attributable to germ-line insertion of an endogenous retrovirus. Proc Natl Acad Sci
U S A, 1996, 93, 14759-14764.
Seddon A.P., Hulmes J.D., Decker M.M., Kovesdi I., Fairhurst J.L., Backer J., DougherVermazen M. & Bohlen P. Refolding and characterization of human recombinant
heparin-binding neurite-promoting factor. Protein Expr Purif, 1994, 5, 14-21.
Sharifi B.G., Zeng Z., Wang L., Song L., Chen H., Qin M., Sierra-Honigmann M.R.,
Wachsmann-Hogiu S. & Shah P.K. Pleiotrophin induces transdifferentiation of
monocytes into functional endothelial cells. Arterioscler Thromb Vasc Biol, 2006,
26, 1273-1280.
Shibata Y., Muramatsu T., Hirai M., Inui T., Kimura T., Saito H., McCormick L.M., Bu G.
& Kadomatsu K. Nuclear targeting by the growth factor midkine. Mol Cell Biol,
2002, 22, 6788-6796.
Silos-Santiago I., Yeh H.J., Gurrieri M.A., Guillerman R.P., Li Y.S., Wolf J., Snider W. &
Deuel T.F. Localization of pleiotrophin and its mRNA in subpopulations of
neurons and their corresponding axonal tracts suggests important roles in neuralglial interactions during development and in maturity. J Neurobiol, 1996, 31, 283296.
Simon-Chazottes D., Matsubara S., Miyauchi T., Muramatsu T. & Guenet J.L.
Chromosomal localization of two cell surface-associated molecules of potential
importance in development: midkine (Mdk) and basigin (Bsg). Mamm Genome,
1992, 2, 269-271.
Soker S., Takashima S., Miao H.Q., Neufeld G. & Klagsbrun M. Neuropilin-1 is expressed
by endothelial and tumor cells as an isoform-specific receptor for vascular
endothelial growth factor. Cell, 1998, 92, 735-745.
Soker S, Miao HQ, Nomi M, Takashima S & Klagsbrun M. VEGF165 mediates formation
of complexes containing VEGFR-2 and neuropilin-1 that enhance VEGF165receptor binding. J Cell Biochem, 2002, 85, 357-368.
Soulie P., Heroult M., Bernard I., Kerros M.E., Milhiet P.E., Delbe J., Barritault D.,
Caruelle D. & Courty J. Immunoassay for measuring the heparin-binding growth
factors HARP and MK in biological fluids. J Immunoassay Immunochem, 2002, 23,
33-48.
Souttou B., Ahmad S., Riegel A.T. & Wellstein A. Signal transduction pathways involved
in the mitogenic activity of pleiotrophin - Implication of mitogen-activated protein
kinase and phosphoinositide 3-kinase pathways. J Biol Chem, 1997, 272, 1958819593.
Souttou B., Juhl H., Hackenbruck J., Rockseisen M., Klomp H.J., Raulais D., Vigny M. &
Wellstein A. Relationship between serum concentrations of the growth factor
211
REFERENCES
pleiotrophin and pleiotrophin-positive tumors. J Natl Cancer Inst, 1998, 90, 14681473.
Souttou B., Carvalho N.B., Raulais D. & Vigny M. Activation of anaplastic lymphoma
kinase receptor tyrosine kinase induces neuronal differentiation through the
mitogen-activated protein kinase pathway. J Biol Chem, 2001, 276, 9526-9531.
Spivak-Kroizman T., Lemmon M.A., Dikic I., Ladbury J.E., Pinchasi D., Huang J., Jaye
M., Crumley G., Schlessinger J. & Lax I. Heparin-induced oligomerization of FGF
molecules is responsible for FGF receptor dimerization, activation, and cell
proliferation. Cell, 1994, 79, 1015-1024.
Sporn M.B. & Roberts A.B. Autocrine growth factors and cancer. Nature, 1985, 313, 745747.
Sporn M.B. The early history of TGF-β, and a brief glimpse of its future. Cytokine Growth
Factor Rev, 2006, 17, 3-7.
Springer B.A., Pantoliano M.W., Barbera F.A., Gunyuzlu P.L., Thompson L.D., Herblin
W.F., Rosenfeld S.A. & Book G.W. Identification and concerted function of two
receptor binding surfaces on basic fibroblast growth factor required for
mitogenesis. J Biol Chem, 1994, 269, 26879-26884.
Stauber D.J., DiGabriele A.D. & Hendrickson W.A. Structural interactions of fibroblast
growth factor receptor with its ligands. Proc Natl Acad Sci U S A, 2000, 97, 49-54.
Stoica G.E., Kuo A., Aigner A., Sunitha I., Souttou B., Malerczyk C., Caughey D.J., Wen
D., Karavanov A., Riegel A.T. & Wellstein A. Identification of anaplastic
lymphoma kinase as a receptor for the growth factor pleiotrophin. J Biol Chem,
2001, 276, 16772-16779.
Stoica G.E., Kuo A., Powers C., Bowden E.T., Sale E.B., Riegel A.T. & Wellstein A.
Midkine binds to anaplastic lymphoma kinase (ALK) and acts as a growth factor
for different cell types. J Biol Chem, 2002, 277, 35990-35998.
Sun X.-Z. & Yoshifiro F. Midkine, A nex heparin binding growth/differeniation factor:
Expression and Distribution during embryogenesis and pathological status. Cong
Anom, 1998, 38, 25-38.
Sutherland R.M., Rasey J.S. & Hill R.P. Tumor biology. Am J Clin Oncol, 1988, 11, 253274.
Szabat E. & Rauvala H. Role of HB-GAM (heparin-binding growth-associated molecule)
in proliferation arrest in cells of the developing rat limb and its expression in the
differentiating neuromuscular system. Developmental Biology, 1996, 178, 77-89.
Takamatsu H., Motofumi I., Kimura M., Gospodarowicz D. & Amann E. Expression and
purification of biologically active human OSF-1 in Esherichia coli. Biochem
Biophys Res Commun, 1992, 185, 224-230.
Take M., Tsutsui J., Obama H., Ozawa M., Nakayama T., Maruyama I., Arima T. &
Muramatsu T. Identification of nucleolin as a binding protein for midkine (MK)
and heparin-binding growth associated molecule (HB-GAM). Journal of
Biochemistry, 1994, 116, 1063-1068.
Takeda A., Onodera H., Sugimoto A., Itoyama Y., Kogure K., Rauvala H. & Shibahara S.
Induction of heparin-binding growth-associated molecule expression in reactive
astrocytes following hippocampal neuronal injury. Neurosci 1995, 68, 57-64.
Takei Y., Kadomatsu K., Matsuo S., Itoh H., Nakazawa K., Kubota S. & Muramatsu T.
Antisense oligodeoxynucleotide targeted to Midkine, a heparin-binding growth
factor, suppresses tumorigenicity of mouse rectal carcinoma cells. Cancer Res,
2001, 61, 8486-8491.
212
REFERENCES
Tamura M., Ichikawa F., Guillerman R.P., Deuel T.F. & Noda M. 1-Alpha,25Dihydroxyvitamin D-3 downregulates pleiotrophin messenger RNA expression in
osteoblast-like cells. Endocrinology, 1994, 3, 21-24.
Tapp H., Hernandez D.J., Neame P.J. & Koob T.J. Pleiotrophin inhibits chondrocyte
proliferation and stimulates proteoglycan synthesis in mature bovine cartilage.
Matrix Biology, 1999, 18, 543-556.
Tezuka K.I., Takeshita S., Hakeda Y., Kumegawa M., Kikuno R. & Hashimoto-Gotoh T.
Isolation of mouse and human cDNA clones encoding a protein expressed specially
in osteoblasts and brain tissues. Biochem. Biophys. Res. Commun., 1990, 173, 246251.
Tomomura M., Kadomatsu K., Nakamoto M., Muramatsu H., Kondoh H., Imagawa K. &
Muramatsu T. A retinoic acid responsive gene, MK, produces a secreted protein
with heparin binding activity. Biochem Biophys Res Commun, 1990, 171, 603-609.
Tsutsui J., Kadomatsu K., Matsubara S., Nakagawara A., Hamanoue M., Takao S.,
Shimazu H., Ohi Y. & Muramatsu T. A new family of heparin-binding
growth/differentiation factors: increased midkine expression in Wilms' tumor and
other human carcinomas. Cancer Res, 1993, 53, 1281-1285.
Uehara K., Matsubara S., Kadomatsu K., Tsutsui J. & Muramatsu T. Genomic structure of
human midkine (MK), a retinoic acid-responsive growth/differentiation factor. J
Biochem (Tokyo), 1992, 111, 563-567.
Ulbricht U., Brockmann M.A., Aigner A., Eckerich C., Muller S., Fillbrandt R., Westphal
M. & Lamszus K. Expression and function of the receptor protein tyrosine
phosphatase zeta and its ligand pleiotrophin in human astrocytomas. J Neuropathol
Exp Neurol, 2003, 62, 1265-1275.
Ulbricht U., Eckerich C., Fillbrandt R., Westphal M. & Lamszus K. RNA interference
targeting protein tyrosine phosphatase zeta/receptor-type protein tyrosine
phosphatase beta suppresses glioblastoma growth in vitro and in vivo. J
Neurochem, 2006, 98, 1497-1506.
Ullrich A. & Schlessinger J. Signal transduction by receptors with tyrosine kinase activity.
Cell, 1990, 61, 203-212.
Unoki K., Ohba N., Arimura H., Muramatsu H. & Muramatsu T. Rescue of photoreceptors
from the damaging effects of constant light by midkine, a retinoic acid-responsive
gene product. Invest Ophthalmol Vis Sci, 1994, 35, 4063-4068.
Usui T., Yamagami S., Yokoo S., Mimura T., Ono K. & Amano S. Gene expression profile
in corneal neovascularization identified by immunology related macroarray. Mol
Vis, 2004, 10, 832-836.
Vacherot F., Laaroubi K., Caruelle D., Delbe J., Barritault D., Caruelle J.P. & Courty J.
Upregulation of heparin-affin regulatory peptide by androgen [letter]. In Vitro Cell
Dev Biol Anim, 1995, 31, 647-648.
Vacherot F., Delbe J., Heroult M., Barritault D., Fernig D.G. & Courty J.
Glycosaminoglycans differentially bind HARP and modulate its biological activity.
J Biol Chem, 1999a, 274, 7741-7747.
Vacherot F., Caruelle D., Chopin D., Gil-Diez S., Barritault D., Caruelle J.P. & Courty J.
Involvement of heparin affin regulatory peptide in human prostate cancer. Prostate,
1999b, 38, 126-136.
Vanderwinden J.M., Mailleux P., Schiffmann S.N. & Vanderhaeghen J.J. Cellular
distribution of the new growth factor pleiotrophin (HB-GAM) mRNA in
developing and adult rat tissues. Anatomy & Embryology, 1992, 186, 387-406.
Vanpouille C., Deligny A., Delehedde M., Denys A., Melchior A., Lienard X., Lyon M.,
Mazurier J., Fernig D.G. & Allain F. The heparin/heparan sulfate sequence that
213
REFERENCES
interacts with cyclophilin B contains a 3-O-sulfated N-unsubstituted glucosamine
residue. J Biol Chem, 2007, 282, 24416-24429.
Venkataraman G., Sasisekharan V., Herr A.B., Ornitz D.M., Waksman G., Cooney C.L.,
Langer R. & Sasisekharan R. Preferential self-association of basic fibroblast
growth factor is stabilized by heparin during receptor dimerization and activation.
Proc Natl Acad Sci U S A, 1996, 93, 845-850.
Venter J.C., Adams M.D., Myers E.W., Li P.W., Mural R.J., Sutton G.G., Smith H.O.,
Yandell M., Evans C.A. & Holt R.A. The sequence of human genome. Science,
2001, 291, 1304-1351.
Vigny M., Raulais D., Puzenat N., Duprez D., Hartmann M.P., Jeanny J.C. & Courtois Y.
Identification of a new heparin-binding protein localized within chick basement
membranes. Eur J Biochem, 1989, 186, 733-740.
Wanaka A., Carroll S.L. & Milbrandt J. Developmentally regulated expression of
pleiotrophin, a novel heparin binding growth factor, in the nervous system of the
rat. Brain Research. Developmental Brain Research, 1993, 72, 133-144.
Ward L.D., Sodhi J.S., McGuffin L.J., Buxton B.F. & Jones D.T. Prediction and functional
analysis of native disorder in proteins from the three kingdoms of life J Mol Biol,
2004, 337, 635-645.
Weber D., Klomp H.J., Czubayko F., Wellstein A. & Juhl H. Pleiotrophin can be ratelimiting for pancreatic cancer cell growth. Cancer Res, 2000, 60, 5284-5288.
Wellstein A., Fang W.J., Khatri A., Lu Y., Swain S.S., Dickson R.B., Sasse J., Riegel A.T.
& Lippman M.E. A heparin-binding growth factor secreted from breast cancer cells
homologous to a developmentally regulated cytokine. Journal of Biological
Chemistry, 1992, 267, 2582-2587.
West D.C., Rees C.G., Duchesne L., Patey S.J., Terry C.J., Turnbull J.E., Delehedde M.,
Heegaard C.W., Allain F., Vanpouille C., Ron D. & Fernig D.G. Interactions of
multiple heparin binding growth factors with neuropilin-1 and potentiation of the
activity of fibroblast growth factor-2. J Biol Chem, 2005, 280, 13457-13464.
Wray W., Boulikas T., Wray W.P. & Hancock R. Silver staining of proteins in
polyacrylamide gels. Anal Biochem, 1981, 18, 197-203.
Yamada H., Inazumi T., Tajima S., Muramatsu H. & Muramatsu T. Stimulation of
collagen expression and glycosaminoglycan synthesis by midkine in human skin
fibroblasts. Arch Dermatol Res, 1997, 289, 429-433.
Yancopoulos G.D., Davis S., Gale N.W., Rudge J.S., Wiegand S.J. & Holash J. Vascularspecific growth factors and blood vessel formation. Nature, 2000, 407, 242-248.
Yasuhara O., Muramatsu H., Kim S.U., Muramatsu T., Maruta H. & McGeer P.L.
Midkine, a novel neurotrophic factor, is present in senile plaques of Alzheimer
disease. Biochem Biophys Res Commun, 1993, 192, 246-251.
Yates E.A., Santini F., Guerrini M., Naggi G. & Casu B. 1H and 13C NMR spectral
assignments of the major sequences of twelve systematically modified heparin
derivatives. Carbohydr Res, 1996, 20, 15-27.
Yates E.A., Guimond S.E. & Turnbull J.E. Highly diverse heparan sulfate analogue
libraries: providing access to expanded areas of sequence space for bioactivity
screening. J Med Chem, 2004, 47, 277-280.
Yayon A., Klagsbrun M., Esko J.D., Leder P. & Ornitz D.M. Cell surface, heparin-like
molecules are required for binding of basic fibroblast growth factor to its high
affinity receptor. Cell, 1991, 64, 841-848.
Yeh H.J., He Y.Y., Xu J., Hsu C.Y. & Deuel T.F. Upregulation of pleiotrophin gene
expression in developing microvasculature, macrophages, and astrocytes after acute
ischemic brain injury. J Neurosci, 1998, 18, 3699-3707.
214
REFERENCES
Yoshida Y., Goto M., Tsutsui J., Ozawa M., Sato E., Osame M. & Muramatsu T. Midkine
is present in the early stage of cerebral infarct. Biochem Biophys Res Commun,
1995, 85, 25-30.
Zhang N., Zhong R., Wang Z.Y. & Deuel T.F. Human breast cancer growth inhibited in
vivo by a dominant negative pleiotrophin mutant. J Biol Chem, 1997, 272, 1673316736.
Zhang N., Zhong R. & Deuel T.F. Domain structure of pleiotrophin required for
transformation. Journal of Biological Chemistry, 1999, 274, 12959-12962.
Zhang N., Zhong R., Perez-Pinera P., Herradon G., Ezquerra L., Wang Z.Y. & Deuel T.F.
Identification of the angiogenesis signaling domain in pleiotrophin defines a
mechanism of the angiogenic switch. Biochem Biophys Res Commun, 2006, 343,
653-658.
Zhang S., Wang G., Liu D., Bao Z., Fernig D.G., Rudland P.S. & Barraclough R. The Cterminal region of S100A4 is important for its metastasis-inducing properties.
Oncogene, 2005, 24, 4401-4411.
Zhou H.Y., Ohnuma Y., Takita H., Fujisawa R., Mizuno M. & Kuboki Y. Effects of a bone
lysine-rich 18 kDa protein on osteoblast-like MC3T3-E1 cells. Biochem. Biophys.
Res. Commun., 1992, 186, 1288-1293.
215
APPENDIX
Amino Acid Abbrev
Alanine
A
Molecular Formula
Side Chain
Acidicity or
Basicity
Molecular
Mass (Da)
neutral
89
Arginine
R
basic (strongly)
174
Asparagine
N
neutral
132
Aspartic acid
D
acidic
133
Cysteine
C
neutral
121
Glutamic acid
E
acidic
146
Glutamine
Q
neutral
147
Glycine
G
neutral
75
Histidine
H
basic (weakly)
155
Isoleucine
I
neutral
131
Leucine
L
neutral
131
Lysine
K
basic
146
Methionine
M
neutral
149
Phenylalanine
F
neutral
165
216
Proline
P
neutral
115
Methionine
M
neutral
149
APPENDIX
Phenylalanine
F
neutral
165
Proline
P
neutral
115
Serine
S
neutral
105
Threonine
T
neutral
119
Tryptophan
W
neutral
204
Tyrosine
Y
neutral
181
Valine
V
neutral
117
Table 7: Amino acid abbreviations, molecular formula and characteristics.
1 Da = 1.66 × 10-24 g (1 Dalton is the unit of mass defined as one twelfth of the mass of the carbon
atom 12C, that is 12/12 gram or 1/N gram). N (Avogadro number) = 6.023 x× 1023.
217
RESUMÉ
HARP est un facteur de croissance jouant un rôle-clé dans la progression et dans l’invasion
tumorale. C’est une molécule multifonctionnelle possédant des activités mitogène, transformante
et angiogène. L’activité de HARP est mediée par deux types de récepteurs transmembranaires;
RPTPβ et ALK. La structure de HARP est constituée de deux domaines homologues aux TSR
motifs et de deux régions basiques aux extrémités N- et C-terminaux. Les travaux précédents de
notre équipe ont montré l’implication de la région C-terminal de HARP dans ses activités
dépendantes de ALK. Le but de mon travail de thèse a été d’approfondir ces recherches et plus
particulièrement sur les relations structure-fonction de HARP. Dans ce dessein, nous avons
synthétisé deux peptides dérivant de l’extrémité C-terminal de HARP: les peptides P111-124 et
P122-131. Le peptide P111-124 a inhibé les activités mitogène et transformante de HARP dans
un modèle de cancer de prostate humain, entrant en compétition avec la fixation de HARP sur
son récepteur ALK. Ce peptide inhibe également la croissance tumorale de ces cellules dans les
souris athymiques et des études pharmacocinétiques ont confirmé la présence du peptide dans la
tumeur et sa bio distribution efficace après l’injection intraveineuse. Le peptide P122-131 a
inhibé la migration cellulaire, l’adhésion et l’angiogénèse induite par HARP via RPTPβ. Des
études de biosensor, utilisant les peptides correspondant aux domaines C-terminal et TSR de
HARP nous ont permis de mieux comprendre le mécanisme d’action de HARP via les voies de
signalisation HARP/ALK/héparine/RPTPβ. Enfin, des études portant sur la recherche d’autres
partenaires ont mis en évidence que HARP entre en interaction avec de nombreuses molécules
telles que FGF2, FGF-BP, VEGF, NPN-1, HGF/SF, S100A4 et OPN. En conclusion, ce travail
ouvre de nouvelles perspectives portant sur l’étude du mécanisme moléculaire de HARP et sur
l’utilisation potentielle de deux peptides ciblant ALK et RPTPβ dans le traitement du cancer.
MOTS-CLÉS
Facteur de croissance; HARP; Thérapie peptidique; Angiogénèse; Croissance tumorale;
Domaine moléculaire; Structure-fonction.
ABSTRACT
HARP is a heparin-binding growth factor, which plays a key role in tumor growth and invasion.
It is a multifunctional polypeptide with mitogenic, transforming and angiogenic activities. HARP
mediates its diverse functions through its main transmembrane receptors: RPTPβ and ALK.
Structurally, HARP contains two TSR homologous domains and two basic clusters in its N and
C-termini. Previous studies from our laboratory have reported that the ALK-dependent
biological activities of HARP were related to its C-terminal region including the amino acid
residues 111-136. The aim of this thesis was to further elucidate the structure-function
relationship of HARP. In this aim, we designed two synthetic peptides derived from the Cterminal region. P111-124 inhibited the mitogenic and transforming activities of HARP in a
prostatic cancer cell model, by specifically competing with HARP for binding to ALK. The
peptide inhibited the tumor growth of these cells in nude mice and the in vivo pharmacokinetic
studies indicated the efficient targeting of the peptide to the tumor. P122-131 inhibited HARPinduced cell migration, adhesion and in vivo angiogenesis, by binding to RPTPβ. The biosensor
studies using the C-terminal and TSR peptides allowed us to better understand the mechanisms
that underpin the signaling features of HARP/ALK/heparin/RPTPβ. Finally, the binding assays
with biosensor revealed the interaction of HARP with a subset of proteins including FGF2, FGFBP, VEGF, NPN-1, HGF/SF, S100A4 and OPN. In conclusion, this thesis yielded new insights
in mechanism of action of HARP. The discovery of new molecular partners for HARP implies
new avenues for regulatory interplay of the protein. The thesis also highlights the potential of the
two anti-tumoral peptides targeting ALK and RPTPβ, as therapeutical agents against cancer.
KEYWORDS
Growth factor; Peptide therapy; Angiogenesis; Tumor growth; Molecular domain; Structurefunction.

Similar documents