University of Southern California Dissertations and Theses

Mechanistic studies of the disintegrin contortrostatin and characterization of the recombinant protein vicrostatin

(USC Thesis Other)

Mechanistic studies of the disintegrin contortrostatin and characterization of the recombinant protein vicrostatin

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content MECHANISTIC STUDIES OF THE DISINTEGRIN CONTORTROSTATIN
AND CHARACTERIZATION OF THE RECOMBINANT PROTEIN
VICROSTATIN

by

Corey Michael Helchowski

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOCHEMISTRY & MOLECULAR BIOLOGY)

August 2010

Copyright 2010 Corey Michael Helchowski
ii
EPIGRAPH

“It only ends once. Anything that happens before that… is just progress.”
— Jacob

iii
DEDICATION

For my father and grandfather, this was as much my dream as it was theirs.

iv
ACKNOWLEDGMENTS
I would like to thank members of my laboratory including Dr. Steven
Swenson, Dr. Radu Minea, Fritz Costa, Alexandria Harrold, Samuel Zidovetzki,
and Lesley Rakowski for their guidance and friendship throughout my time in the
laboratory. I would also like to thank my mentor, Dr. Francis Markland, for
believing that I could make a difference in his laboratory and for giving me that
chance.
Additionally, I offer thanks to the Department of Biochemistry and
Molecular Biology, the Clinical Reference Laboratory, and the Norris Cancer
Center Cell and Tissue Imaging Core facilities at the at the University of Southern
California, Keck School of Medicine for technical assistance. Also, I would like to
thank the members of my Ph.D. dissertation committee who guided me in the
completion of this work with their expertise.
Lastly, I would like to thank my family and friends for influencing me
throughout my life. Their presence in my life has made me the way I am today.

v
TABLE OF CONTENTS

Epigraph ii

Dedication iii
Acknowledgments iv
List of Figures viii
Abbreviations x
Abstract xiv
Chapter 1: Introduction 1
1.1 Integrins 1
1.1.1 Integrin Structure 2
1.1.2 Integrin Function 6
1.1.3 Integrin Activation 10
1.2 Disintegrins 11
1.2.1 Disintegrin Structure 12
1.2.2 Disintegrin Function 14
1.2.3 Contortrostatin and Vicrostatin 17
1.3 Specific Aims 23

Chapter 2: Contortrostatin Effects Gene Expression at the
mRNA Level 25
2.1 Summary 25
2.2 Introduction 26
2.3 Materials and Methods 27
2.3.1 Cell Culture 27
2.3.2 Oligo GEArrays 28
2.3.3 RT-PCR 30
2.4 Results 30
2.4.1 Gene Expression Changes Induced by
Contortrostatin (Oligo GEArrays) 30
Table 1: RT-PCR Primer Sequences 31
2.4.2 Gene Expression Changes by Induced
Contortrostatin (RT-PCR) 35
2.5 Discussion 35

Chapter 3: Contortrostatin Induced Integrin Internalization
And Development of a Novel Measurement
Technique 44
vi
3.1 Summary 44
3.2 Introduction 44
3.3 Materials and Methods 49
3.3.1 Cell Culture and Antibodies 49
3.3.2 Integrin/Disintegrin Internalization by
Western Blot 50
3.3.3 Integrin/Disintegrin Internalization by
FACS Analysis 51
3.3.4 Evaluating the Efficiency of Different
Stripping Buffers on Removing Cell
Surface Fluorescence 52
3.3.5 Internalization Rates of Beta1 Integrin
And Transferrin Receptors 53
3.3.6 Trypan Blue viability 54
3.3.7 Integrin/Disintegrin Internalization by
Confocal, Employing
Novel Pepsin Technique 54
3.4 Results 55
3.4.1 Contortrostatin Induces Integrin
Internalization (Western Blot) 55
3.4.2 Contortrostatin Induces Integrin
Internalization (FACS analysis) 55
3.4.3 Classical Stripping Buffers Fail to Remove
Residual Cell Surface Fluorescence 57
3.4.4 Internalization Rates of Beta1 Integrin and
Transferrin Receptors 57
3.4.5 Novel Pepsin Technique has no Effect
on Cell Viability 63
3.4.6 Contortrostatin Induces Integrin
Internalization (Confocal) 64
3.5 Discussion 64

Chapter 4: Beta1 Integrin Inactivation and Disruption of Focal
Adhesions by the Distintegrin Contortrostatin 73
4.1 Summary 73
4.2 Introduction 74
4.3 Materials and Methods 80
4.3.1 Cell Culture and Antibodies 80
4.3.2 Beta1 Integrin Inactivation 80
4.3.3 Immunofluorescence 81
4.3.4 Co-Immunoprecipitations and
Western Blotting 83
4.3.5 Active Rap1 Pull Down 84
4.3.6 Statistical Analysis 85
4.4 Results 85
vii
4.4.1 Beta1 Integrin Inactivation by
Contortrostatin 85
4.4.2 Focal Adhesion Protein Associations are
Disrupted by Contortrostatin 89
4.4.3 Disruption of Talin Based Focal
Adhesions by Contortrostatin 91
4.4.4 Inhibition of Talin Protein Expression
and Rap1 Activity by Contortrostatin 96
4.5 Discussion 99

Chapter 5: Characterization of Vicrostatin 107
5.1 Summary 107
5.2 Introduction 108
5.3 Materials and Methods 111
5.3.1 Cell Culture and Antibodies 111
5.3.2 Platelet Aggregation 112
5.3.3 Cell Invasion Through a Reconstituted
Basement Membrane 113
5.3.4 Endothelial Cell Actin
Cytoskeleton Disruption 114
5.3.5 Endothelial Cell Tube Formation 115
5.3.6 FAK Phosphorylation 116
5.4 Results 118
5.4.1 Inhibition of Platelet Aggregation
by Disintegrins 118
5.4.2 Inhibition of Cell Invasion Through a
Reconstituted Basement Membrane
by Disintegrins 118
5.4.3 Disruption of Endothelial Cell Actin
Cytoskeleton by Disintegrins 121
5.4.4 Inhibition of Endothelial Cell Tube
Formation by Disintegrins 121
5.4.5 FAK Hyperphosphorylation
by Disintegrins 125
5.5 Discussion 127

Chapter 6: Conclusions and Future Directions 133
Bibliography 138

viii
List of Figures
1: Integrin Architecture 3
2: Comparison Between Native CN and
Recombinant VCN Sequences 18

3: HEM & AM Microarray 33
4: HSTPF Microarray 34
5: CN Effect on MDA-MB-231 Gene Expression 36
6: CN Effect on MDA-MB-435 Gene Expression 37
7: Fc-specific Secondary Antibody
Requirement for Pepsin Removal
of the Residual Fluorescent Signal 48

8: CN Induced Integrin Internalization
(Western blot & FACS) 56

9: Effect of pepsin on labeled antibody
removal at maximum residual fluorescence 58

10: The Advantages of Using Pepsin/HCl
-stripping Buffer to Accurately Measure the
Internalization Rates of Receptors in Two System 60

11: Cell Viability Assessed by Trypan
Blue Staining65

12: CN Induced Integrin Internalization (confocal) 66

13: Inactivation of Beta1 Integrins by CN 87

14: Co-immunoprecipitation of Talin 90

15: Co-immunoprecipitation of Key Focal
Adhesion Proteins 92

16: Immunofluorescence of Actin Cytoskeleton
and Focal Adhesions 94

ix
17: Effect of Prolonged Exposure of CN on
Expression of Talin 97

18: Effect of Prolonged Exposure of CN on
Activity of Rap1 in HUVECs 98

19: VCN and Native CN Exhibit Identical
Effect on Platelet Aggregation 119

20: Inhibition of HUVECs invasion through
a reconstituted basement membrane 120

21: Actin Cytoskeleton Staining of HUVECs
Treated with FITC-VCN/CN 122

22: Inhibition of HUVEC Tube Formation 123

23: FAK Phosphorylation Levels in MDA-MB-435
Cells Treated with Soluble Disintegrins 126

x
ABBREVIATIONS
50% inhibitory concentrations (IC50)
adenosine diphosphate (ADP)
adjacent to MIDAS (ADMIDAS)
alpha ( α)
antibody (Ab)
arginine-glycine-aspartic acid (RGD)
beta ( β)
bicinchoninic acid (BCA)
bovine serum albumin (BSA)
CCAAT/enhancer binding protein beta (CEBPB)
complementary deoxyribonucleic acid (cDNA)
contortrostatin (CN)
Cortactin (CTTN)
Cyclin-dependent kinase 2 (CDK2)
Cyclo(Arg-Gly-Asp-DPhe-Val) (cRGDfV)
Dulbecco’s modified Eagle’s medium (DMEM)
E-cadherin (CDH1)
electrospray ionization-mass spectroscopy (ESI-MS)
extracellular matrix (ECM)
Fc-specific Alexa-488 conjugated
secondary antibody (Fc A488 secondary Ab)
fetal bovine serum (FBS)
xi
fibronectin (FN)
Fibronectin 1 (FN1)
fluorescein isothiocyanate (FITC)
fluorescence-activated cell sorting (FACS)
focal adhesion (FA)
focal adhesion kinase (FAK)
fragment, antigen-binding (F(ab ′)2)
fragment, crystallizable (Fc)
glutathione reductase (gor)
growth factor reduced (GFR)
high pressure liquid chromatography (HPLC)
human extracellular matrix and adhesion molecules (HEM & AM)
human signal transduction pathway finder (HSTPF)
human umbilical vein endothelial cell (HUVEC)
hypoxia-inducible transcription factors (HIFs)
immunoglobulin G (IgG)
intercellular cell-adhesion molecule (ICAM)
Jun B proto-oncogene (JUNB)
Jun oncogene (JUN)
leukocyte adhesion deficiencies (LAD)
liposomal contortrostatin (LCN)
liposomal vicrostatin (LVCN)

xii
messenger ribonucleic acid (mRNA)
metal ion-dependent adhesion site (MIDAS)
monoclonal antibodies (mAbs)
Myc (C-myc)
Nuclear factor of kappa light polypeptide
gene enhancer in B-cells 1 (NFKB1)

nuclear factor-kappa B (NF- κB)
ovarian carcinoma cells (OVCAR)
phosphate buffered saline (PBS)
platelet aggregation (PA)
platelet rich plasmas (PRPs)
plexin-sempahorin-integrin (PSI)
radioimmunoprecipitation (RIPA)
Rap1-interacting adaptor molecule (RIAM)
room temperature (RT)
Secreted phosphoprotein 1 (SPP1)
snake venom hemorrhagic metalloproteinases (SVMP)
sodium dodecyl sulfate polyacrylamide
gel electrophoresis (SDS-PAGE)
thioredoxin reductase (trxB)
TP53 (p53)transferrin receptor (TFR)
tween (0.1%) tris (1.5%, 7.5 pH) buffered saline (TTBS)
vascular endothelial growth factor (VEGF)
v-Fos (FOS)
xiii
vicrostatin (VCN)
vitronectin (VN)

xiv
ABSTRACT
Integrin mediated signaling is vital to several cellular pathways. These
pathways include the cell’s ability to migrate and invade through the extracellular
matrix, which is fundamental to more complex processes like angiogenesis and
cancer progression. Our ability to understand, manipulate, or even control these
events could have drastic implications for the field of cell biology. Disintegrins
act as integrin antagonists with the ability to disrupt adhesion based integrin
functions through direct binding. However, not much is know about how these
molecules initiate signaling events downstream after ligation to integrins.
Contortrostatin (CN), a homodimeric disintegrin isolated from southern
copperhead venom has been studied by the Markland laboratory for over fifteen
years. Although some initial studies were done on CN signaling, CN’s anti-
cancer ability has been the focus of the Markland laboratory since. In this
dissertation several mechanistic events, initiated by the ligation of CN to
integrins, were investigated. Through oligo arrays and RT-PCR, CN ligation
initiated expression changes at the mRNA level. It was also discovered through
investigations into the timing involved in CN’s internalization process, via a novel
technique developed for allowing more precise measurements of internalized
molecules, that CN allows for a faster integrin internalization rate. In addition,
CN treatment was shown to have a significant effect on talin, a key molecule
involved in integrin activation. Through a series of co-immunoprecipitations and
western blotting it was determined that CN binding leads to the displacement of
talin from the cytoplasmic domain of beta1 integrins, a decrease in calpain-II
xv
cleavage of talin, and an alteration of talin’s interaction with several other
proteins (Rap1, vinculin, and Rap1-interacting adaptor molecule). These events
combine to produce a global inactivation signal for beta1 integrins as indicated by
flow cytometry and a decrease in the ability of the cell to produce talin based
focal adhesions as shown by confocal microscopy.
Finally, the recombinant form of CN, called vicrostatin (VCN) was
characterized in multiple ways. It was established that VCN has the same anti-
angiogenic ability as CN in invasion and tube formation in vitro assays. In
addition, VCN was able to inhibit platelet aggregation with the same potency as
native CN. Lastly, it was shown that VCN initiates similar signaling events as CN
through focal adhesion kinase phosphorylation. These finding will help contribute
to the overall knowledge behind disintegrin function and further their
development for scientific applications.

1
Chapter 1: Introduction
1.1 Integrins
Integrins are a family of cell surface receptors that are essential to cell
biology. Many biological processes rely on integrins and they can easily be
distinguished from other receptor families by various means. First, evolutionarily
speaking, they are very old and are expressed in all multicellular animals
(Johnson, Lu et al. 2009). Second, the integrin receptor family is vast and
composed of many members. Third, integrin structure is multifaceted and the
receptor itself is relatively large. The overall purpose of integrins is to maintain
tissue architecture in a differentiated state and coordinate cell movement via firm,
reversible, and transient contacts with the extracellular matrix (ECM).
Integrins, and other cell adhesion molecules, bind with relatively low
affinity to their ligands and have weak signal strength, but because they are
expressed on the cell-surface at great numbers they can combine to produce a
strong signal. This differs greatly from cell-surface hormone receptors and
soluble extracellular signaling molecules in that they bind with a high affinity,
produce a strong signal, and are expressed at the cell-surface in low numbers
(Alberts 2002). Integrins are also capable of transducing signals in two
directions, a term called bi-directional signaling. What this means is that
integrins can bind an extracellular ligand and produce an intracellular change,
like modifying their attachment to the actin cytoskeleton, which is phrased
outside →in signaling, or they can alter their conformation or expression profile in
order to modify the way the cell interacts with the ECM, which is phrased
2
inside →out signaling (Hynes 2002). This unique and important characteristic of
integrins is vital to their behavior.
To indicate the importance of these receptors for maintaining the integrity
of the cytoskeletal-ECM linkage they were given the name “integrin” (Hynes
2004). The scientific community has been studying integrins for over 25 years
(Tamkun, DeSimone et al. 1986), and much is known about these molecules, yet
the overall complexity behind their mechanism of action is still not well
understood.

1.1.1 Integrin Structure
Integrins are transmembrane receptors that are composed of two non-
covalently associated distinct glycoprotein subunits, the alpha ( α) chain and the
beta ( β) chain (Hynes 2002) (Figure 1). There are a total of eighteen different α
subunits and eight different β subunits. These two chains combine to form
twenty-four distinct heterodimers that make up the integrin family (Takada, Ye et
al. 2007). The extracellular domain of the α subunit is >940 residues while the β
subunit is somewhat smaller, >640 residues (Takagi and Springer 2002). Two
long stalk regions from the α and β chains connect the extracellular domain to
the transmembrane domain while a globular ligand binding head piece makes up
the remaining section of the extracellular domain. The intracellular domains are
much smaller, with the exception of the specialized β4 subunit, which is adapted
to connect to the keratin actin cytoskeleton (de Pereda, Wiche et al. 1999).

Figure 1

Integrin Architecture (adapted from (Takagi and Springer 2002). (A)
Organization of domains within the primary structure of integrin. Depending on the α subunit, it
may contain an I-domain insertion as denoted by the dotted line. Asterisks show Mg
2+
(green)
and Ca
2+
(red) binding sites. Lines below the stick diagrams show disulfide bonds. (B) Integrin
domains arranged into a three-dimensional structure, with an I-domain added. Each domain is
color coded as in (A).

3
4
The N-terminal region of the α subunit is composed of a seven-bladed β-
propeller, which is connected to a thigh, a calf-1, and a calf-2 domain, together
forming the leg structure that supports the integrin head (Barczyk, Carracedo et
al. 2009). Half of integrin α subunits contain a domain that is inserted between β-
sheets 2 and 3 of the β-propeller domain (Springer 1997) known as the I domain.
An important site within the αI domain is the metal ion-dependent adhesion site
(MIDAS), which is responsible for binding negatively charged ions in residues via
a coordinating Mg
2+
ion (Lee, Bankston et al. 1995).
The β subunit contains a domain with weak but detectable sequence
homology to the I domain of the α subunit (Ponting, Schultz et al. 2000) called
the βI domain or the I-like domain. Like the αI domain, the βI domain contains an
Mg
2+
coordinating MIDAS, but the βI domain also contains a site adjacent to
MIDAS (ADMIDAS) binding an inhibitory Ca
2+
ion. Interestingly, the ADMIDAS
can also bind a Mn
2+
ion leading to a conformational change resulting in an active
integrin (Humphries, Symonds et al. 2003). The N-terminal region of the β
subunit contains a domain termed the plexin-sempahorin-integrin (PSI) domain
because it shares sequence homology with those particular membrane proteins
(Bork, Doerks et al. 1999). This region of the β subunit is cysteine rich,
containing seven cysteine residues that cooperate with another cysteine rich
region in the C-terminal region in order to restrain the integrin in an inactive
conformation (Zang and Springer 2001). The β subunit also contains a hybrid
domain that is a β-sandwich formed from folded amino acids on both sides of the
βI domain. This means that the hybrid domain, much like the β-propeller domain
5
in the α-subunit, has two covalent associations with the I domain. Movement of
one of these partial domains, relative to the other, may help transverse
conformational change throughout the entire integrin (Takagi and Springer 2002).
One of the difficulties in studying a family of receptors like the integrins is
that they bind a myriad of large ECM components, which in turn bind a large
number of other proteins. This made it notoriously difficult to isolate specific
binding sites for the integrins. The first and most utilized binding site to be
identified was the arginine-glycine-aspartic acid (RGD) sequence that is present
in ECM molecules such as fibronectin (FN), vitronectin (VN), and a variety of
other adhesive proteins (Hynes 1992). The RGD binding sequence was first
identified in 1984 by observing FN (Pierschbacher and Ruoslahti 1984) and was
met with skepticism that such a small recognition motif could be utilized in such
large molecules. But with the discovery that the RGD sequence was present in
other ECM proteins and the subsequent finding of their receptor, the integrins, its
importance was confirmed. Since then other integrin binding sequences have
been identified in additional ECM proteins that have slight variations on the RGD
motif. For instance, the αIIb β3 integrin is a potent binder to the specific lysine-
glycine-aspartic acid (KGD) sequence of certain ECM proteins (Plow,
Pierschbacher et al. 1985; Scarborough, Naughton et al. 1993). With the
discovery of these binding motifs, integrin science was truly beginning to emerge.

6
1.1.2 Integrin Function
Integrins are the cells primary receptor for attachment to the ECM and are
fundamental to many signal transduction pathways. Integrins provide the key link
between the ECM and the cytoskeleton by way of connections made to
numerous actin filaments. When an integrin binds one of its numerous ECM
ligands a series of conformational changes and phosphorylations lead to the
binding of several anchor proteins, namely talin, α-actinin, and filamin by the
cytoplasmic tail of the β subunit. These anchor proteins in turn bind other
proteins, such as vinculin, which make the final connections to the actin
cytoskeleton. The regulation of kinases like Focal Adhesion Kinase (FAK) and
Src kinase family members lead to the phosphorylation of many substrates,
initiating a signal transduction cascade. After these critical connections are
established, integrins cluster around each other at the ligation site commencing
with the formation of a focal adhesion (FA). Thus the cell makes an extensive
connections with the ECM, allowing for important mitogenic and motogenic
signals to proceed.
Integrins are responsible for many biological activities, but the basis
behind most of them is their ability to allow for cellular migration and invasion
through the ECM. As integrins engage their ECM ligands and establish focal
contacts at the leading edge of a cell (the lamellipodium), integrin connections at
the rear of the cell disassemble, allowing for forward movement. Integrins must
rapidly undergo conformational changes that enable the cell to rapidly attach to
and detach from the ECM in a coordinated fashion (Baker and Zaman 2009). In
7
this way a cell can translocate from one area to another by movement through
the ECM. Because integrins allow cells to migrate and invade through the ECM,
endothelial cells can therefore engage in more complex processes. For instance,
given the right environmental cues, endothelial cells can utilize their integrins to
develop into neovessels through the process of angiogenesis.
Angiogenesis can be described as an integrated set of cellular,
biochemical and molecular behaviors by which new blood vessels are formed
from pre-existing vessels (Folkman 2007). This process plays a key role in
various physiological and pathological conditions, including embryonic
development, wound repair, inflammation, and tumor growth (Garmy-Susini and
Varner 2008). In order for the process of angiogenesis to begin, the vascular
bed must be in hypoxic conditions (Dang, Chun et al. 2008). In homeostatic
conditions, tissues are oxygenated by simple diffusion, but as tissues and organs
begin to grow they require more oxygen. When hypoxic conditions are met,
vessel growth begins by the triggering of hypoxia-inducible transcription factors
(HIFs). HIFs up-regulate the gene for Vascular Endothelial Growth Factor
(VEGF-A) (Semenza 2003), as well as other important angiogenic genes, which
binds to two cell surface receptors named VEGFR-1 and VEGFR-2. These two
receptors will elicit signal transduction pathways leading to endothelial cell
migration, proliferation, and survival, which result in the establishment of new
blood vessels. Integrins may regulate the expression of pro-angiogenic
molecules (such as important proteases used to degrade the ECM) and facilitate
adhesion during crucial steps of angiogenesis (Contois, Akalu et al. 2009).
8
Given that specific integrins, including α5 β1, αvβ3, and αvβ5, were discovered to
be highly expressed in angiogenic blood vessels but weakly expressed in normal
cells and quiescent blood vessels (Brooks, Clark et al. 1994; Varner 1997; Kim,
Bell et al. 2000; Kumar, Armstrong et al. 2000) and that the integrins themselves
have extensive connections within and outside the cell, integrins are the
functional hubs that coordinate the process of angiogenesis (Hodivala-Dilke,
Reynolds et al. 2003).
Due to the important role integrins play in angiogenesis and their
involvement with the invasiveness of cells though the ECM; integrins are integral
to cancer progression. Without integrins a cancerous cell would not be able to
exhibit two of the “hallmarks of cancer”, sustained angiogenesis and tissue
invasion and metastasis (Hanahan and Weinberg 2000). Without these traits a
cell would most likely not progress to a disease state. In addition, new research
has begun to shed light on the ability of integrins to help control tumor cell
survival and proliferation (Vellon, Menendez et al. 2005; Han, Khuri et al. 2006).
This means that integrins are potentially involved in the remaining “hallmarks of
cancer”, limitless reproductive potential, evading apoptosis, insensitivity to anti-
growth signals, and self-sufficiency in growth signals as well. These new
discoveries pinpoint integrins as an excellent target for anti-cancer therapeutics.
Different combinations of integrins are expressed on the surface of
specific cell types, meaning that certain integrins are specialized for precise
reasons. For instance, the αIIb β3 integrin is only expressed on the surface of
platelets, making it the most specialized of all integrins. On the surface of a
9
single platelet cell there are only several hundred αvβ3 molecules, which
modulate platelet adhesion to osteopontin and VN (Bennett, Chan et al. 1997).
In contrast, there are over 80,000 αIIb β3 receptors that bind fibrinogen, von
Willebrand factor, and FN, crosslinking adjacent platelets and causing
aggregation (Bennett, Berger et al. 2009). Another example of specialization in
integrins is with the β2 subfamily whose expression is limited to white blood cells.
Integrins play a key role in immune function, so much so that a whole class of
diseases is named after them. Leukocyte adhesion deficiencies (LAD) are when
specific immune integrins have an altered expression profile or a defect in their
function results in a deficiency in the immune response. Common integrins
expressed on leukocytes include leukocyte function–associated antigen 1 (LFA-1
or αL β2), Mac-1 ( αM β2), and very late antigen 4 (VLA-4 or α4 β1) (Abram and
Lowell 2009). In addition to specialized integrins, it is possible for the same
integrin receptor to have different ligand-binding specificities in different cell
types, meaning that certain cell specific factors may be involved in altering the
level of integrin-binding (Alberts 2002).
Besides cell-ECM connections, it is also clear that some integrins can
make cell-cell or cell-pathogen associations as well. The leukocyte integrins are
mainly receptors for adhesion molecules on the surfaces of other cells, such as
intercellular cell-adhesion molecule (ICAM)-1 and ICAM-2 (Reyes-Reyes, Mora
et al. 2002), in the immune response. Also, it has been documented that
integrins can play a role in the cellular attachment to certain harmful bacterial
strains. When the Gram-negative bacterium Yersinia enterocolitica induces its
10
outer surface protein invasin to cluster, it activates eukaryotic β1 integrin
receptors that promote uptake of the bacteria into the host cell (Deuretzbacher,
Czymmeck et al. 2009). Though little is known about how integrins mediate
connections outside of the ECM, it is obvious that this will be at the forefront of
integrin science in the years to come.

1.1.3 Integrin Activation
Integrins can exist in one of two major conformational states, an open
extended state or a closed bent state. This was made evident by the
crystallization of the soluble heterodimer of αV β3 in 2002 (Xiong, Stehle et al.
2002). When in the bent conformation, the ligand-binding site is not in an optimal
orientation for binding macromolecular ligands in the ECM or on the surface of
other cells (Luo, Carman et al. 2007). Although it has recently been made clear
that it is possible for some integrins to bind ligands while in the bent conformation
(Askari, Buckley et al. 2009), in order for each integrin to take full advantage of
its binding capabilities it must be activated and in its open extended state.
The global conformation change observed when an integrin undergoes
activation is something shared among many integrins, including those with I
domains (Bazzoni, Shih et al. 1995; Yednock, Cannon et al. 1995). This can
easily be followed by the binding of specific monoclonal antibodies (mAbs) for the
activated or ligand-bound conformation. Commonly these mAbs map to the C-
terminal cysteine-rich domains in the β subunit, fairly distant from the ligand
binding site (Frelinger, Lam et al. 1988; Honda, Tomiyama et al. 1995).
11
Interestingly, the fact that some of these conformation specific mAbs could
induce an affinity change suggested that the activation of integrins is the result of
conformational changes within the extracellular domains of the integrin subunits
(Takagi and Springer 2002). Also, signals coming from within the cell
(inside →out signaling) can produce the same conformational changes involved in
activation. Many models have been developed to explain these structural
rearrangements, but none have been confirmed to be the true model.
Regardless of knowing the exact structural model for activation, it is clear that the
activity state of integrins is crucial to their function.

1.2 Disintegrins
In 1990, Robert Gould coined the word “disintegrin” to describe a newly
discovered class of cysteine-rich low molecular weight (40-100 amino acids)
proteins containing RGD motifs that were non-enzymatic and being isolated from
the venom of hemotoxic snakes (Gould, Polokoff et al. 1990). Since that time
over 100 different disintegrins have been discovered and they have proven to be
a chemically active and specific family of proteins that preferentially bind to and
inhibit integrin function.
Disintegrins, along with close to hundred other proteins, are synthesized in
the venom glands of snakes from the hemotoxic Viperidae family (Gould,
Polokoff et al. 1990; Niewiarowski, McLane et al. 1994). They can be
synthesized from messenger ribonucleic acid (mRNA) lacking the
metalloprotease-coding region (Okuda, Koike et al. 2002) or released through
12
proteolytic processing of Serpent-specific PII-snake venom hemorrhagic
metalloproteinase precursors in the venom gland (Kini and Evans 1992; Juarez,
Comas et al. 2008). Although disintegrins are present in toxic venom, the
molecule itself is not directly responsible for the more harmful effects of venom
such as immobilizing/paralyzing the prey or liquefying tissue. The purpose of
disintegrins, in regards to snake venom, is to bind to and inhibit integrin function
while the other toxins accomplish their specified tasks.

1.2.1 Disintegrin Structure
Structural studies of disintegrins have adopted three techniques that are
useful in their own ways. They each draw upon one of the important steps in the
life of a protein. In the first technique, the transcriptomic approach, researchers
use the mRNA isolated from the cells of the venom glands of snakes to construct
complementary deoxyribonucleic acid (cDNA) libraries (Juarez, Wagstaff et al.
2006). In the second technique, the proteomic approach, the protein is purified
from the venom itself, meaning that it is fully translated. To isolate disintegrins, a
majority of scientists use variations of High Pressure Liquid Chromatography
(HPLC) (Trikha, Rote et al. 1994; Juarez, Wagstaff et al. 2006; Calvete,
Marcinkiewicz et al. 2007). The final technique, the genomic approach, takes the
actual tissue from snakes and examines the DNA present there constructing a
profile for that particular snake species (Bazaa, Juarez et al. 2007).
The “classical” disintegrin has a high number of disulfide bonds and little
secondary structure while maintaining the ability to bind integrins in a dose-
13
dependent manner, either as a monomeric or dimeric peptide, and possessing an
R/K/M/W/VGD, MLD, MVD or K/RTS sequence within what is known as the
“RGD adhesive loop”. This loop has a vital role in the way in which each
disintegrin can interact with specific integrin receptors (McLane, Marcinkiewicz et
al. 1998). The integrin-binding activity/avidity of disintegrins depends on the
appropriate pairing of cysteine residues, which determines the presentation of
the active part of the adhesive loop. Disintegrins are divided into four different
groups and are classified according to their polypeptide length, number of
subunits, and the number of disulfide bonds each has: monomeric short (41-51
residues and 4 disulfide bonds), monomeric medium (~70 amino acids and 6
disulfide bonds), monomeric long (~84 residues and 7 disulfide bonds), and
dimeric (~67 amino acids with 10 cysteines involved in 4 intrachain and 2
intrachain disulfide linkages) (McLane, Joerger et al. 2008). NMR solution
studies of several short (echistatin) and medium-sized (kistrin, flavoridin,
albolabrin) disintegrins revealed that the active tripeptide is located at the apex of
the adhesive loop protruding 14–17 Å from the peptide core (Adler, Lazarus et al.
1991; Saudek, Atkinson et al. 1991; Senn and Klaus 1993; Smith, Jaseja et al.
1996).
Like most venom proteins, disintegrins are extensively cross-linked by
disulfide bridges and have become functionally diverse while evolving into toxin
multigene families that exhibit interfamily, intergenus, interspecies, and
intraspecific variability (Juarez, Comas et al. 2008). The way that disintegrins
have mimicked native cell-surface receptors from plants and animals speaks
14
volumes for their structural evolution. A single amino acid change within the
RGD loop could be indicative to the ability of a certain species of snakes survival
or geographic locale (McLane, Joerger et al. 2008). The vast number of
structurally different disintegrins was most likely caused by gene duplication
events followed by accelerated evolution by positive selection and
neofunctionalization of duplicated gene copies (Juarez, Comas et al. 2008) in
specific species of snake based on the predator-prey relationship and integrin
profile of the snakes prey species.

1.2.2 Disintegrin Function
Once snakes of the Viperidae family bite their prey, disintegrins will be
released at the wound site with the purpose of selectively interfering or blocking
the function of integrins in order to facilitate hemorrhage in the envenomed prey
animal. Disintegrins accomplish this task by binding to the αIIb β3 integrin on the
surface of platelets and inhibiting platelet aggregation (PA), something they do
extremely well. The effect disintegrins have on PA can easily be measured in a
laboratory setting via a platelet aggregometer. In a comparative study of the 71
disintegrins described in the literature from 1998 – 2004, disintegrins were shown
to have a wide variety of potencies for inhibiting PA: 50% inhibitory
concentrations (IC50) less than 400 nM to over 1000 nM in collagen-induced PA;
less than 100 nM to over 1000 nM for ADP-induced PA; 28 – 1200 nM in
thrombin-induced PA; 7-165 nM with tumor-induced PA (Mclane 2006; McLane,
Joerger et al. 2008). This study indicates the importance of the RGD motif in the
15
effectiveness of the disintegrins ability to disrupt PA, because all of the tested
disintegrins that contained the RGD sequence had IC50s less than 200 nM while
disintegrins with a MGD, MLD, or VGD sequence in their active loop produced
much higher IC50s between 300 to 1800 nM.
Disintegrins also work extremely well as anti-adhesive, anti-migratory, and
anti-invasive molecules by binding to and interfering with integrins. For years
researchers have been showing that disintegrins can be tested for these
activities by means of in vitro assays. For instance, simple adhesion assays
consist of an ECM component fixed at the bottom of a well, which represents a
reconstituted basement membrane, and a cell line in question that attempts to
attach to the ECM element through the appropriate integrins. A disintegrin can
be placed within the well and the effectiveness of that disintegrin to disrupt the
adhesion of the particular cell line to that specific ECM molecule can be
measured. For example, the disintegrin echistatin exhibited an inhibitory effect
on the adhesion of T24 human urinary bladder carcinoma cells to FN with an
IC50 of 256 nM (Sanchez, Galan et al. 2006). Migratory assays begin with a
reconstituted basement membrane and a cell line seeded on top as well. After
the cells are allowed to incubate for a given amount of time their movement will
be measured across the ECM, by several different methods, in the presence or
absence of disintegrin. For instance, eristostatin has been shown to inhibit cell
migration on FN in a dose-dependent manner but had no effect on collagen IV or
laminin migration (Tian, Paquette-Straub et al. 2007). To demonstrate the anti-
invasive activity of disintegrins a slightly more complex assay must be performed.
16
A multiwell insert, or Boyden chamber, must be utilized to separate two
environments. One environment will contain a chemoattractant and the other will
lack the chemoattractant in a serum free environment. The cell line in question
will be seeded onto the non-chemoattractant well and allowed to invade through
an ECM matrix along the chemoattractant gradient in the presence or absence of
disintegrin. After a given time period the amount of cells that invaded to the other
side of the Boyden chamber will be counted. For example, viperistatin has been
shown to inhibit invasion of HS.939T cells through a primary adult dermal human
microvascular endothelial cell layer (Staniszewska, Walsh et al. 2009).
Interestingly, these researchers used a cell line instead of an ECM molecule in
order to mimic metastasis. Keep in mind that these assays are extremely integrin
and disintegrin specific. The integrin in question will only bind certain ECM
molecules and the disintegrin in question will only bind certain integrins, meaning
that all three molecules must coincide in order to see an effect.
Based on the realization of how important integrin function is to cancer
progression, many scientists have begun to investigate the potential of using
disintegrins as anti-cancer agents (Soszka, Knudsen et al. 1991; Swenson,
Costa et al. 2004; Yang, Tang et al. 2005; Staniszewska, Walsh et al. 2009).
These researchers, among others, believe that injecting disintegrins into cancer
patients can help slow or even stop the progression of cancer by halting the
tumors ability to recruit new blood vessels by means of anti-angiogenesis. Also,
since cancer cells utilize their invasive properties to enter the blood stream,
disintegrin binding directly to cancerous cells can potentially disrupt metastasis to
17
other areas of the body. This provides disintegrins with a distinct advantage over
other cancer therapeutics that only target one aspect of cancer’s deadly
phenotype.

1.2.3 Contortrostatin and Vicrostatin
Sixteen years ago the Markland laboratory reported the purification of the
snake venom disintegrin contortrostatin (CN) (Trikha, Rote et al. 1994) (Figure
2A). CN was purified from the venom of the Southern copperhead snake,
Agkistrodon contortrix contortrix, and since then has been the primary focus of
the Markland laboratory. CN was purified from Southern copperhead venom
using a three step chromatographic approach: hydrophobic interaction
chromatography-HPLC, and two steps on Cl8 reverse phase-HPLC. CN was
initially characterized by sodium dodecyl sulfate polyacrylamide gel
electrophoresis (SDS-PAGE) to migrate to approximately M
r
15,000 under non-
reducing conditions and M
r
6,000 under reducing conditions, which suggested
that CN was composed of two or three subunits (Trikha, Rote et al. 1994). The
mass of intact CN was further characterized by electrospray ionization-mass
spectroscopy (ESI-MS) and found to be 13,505 daltons while reduced
pyridylethyl-CN was found to have a mass of approximately 6750 daltons, which
provided good evidence that CN was in fact a dimer (Trikha, De Clerck et al.
1994), a characteristic that we now know from recent crystallography data is true
(Moiseeva, Swenson et al. 2002). Furthermore, by taking into account the
incorporation of approximately 1,250 mass units with 12 pyridylethyl groups, six
Figure 2

Comparison Between Native CN and Recombinant VCN Sequences. (A) CN’s
native sequence depicted with disulfide bonds present and the Arg-Gly-Asp tripeptide motif
depicted in bold. Crystallographic data (Moiseeva, Swenson et al. 2002) shows that CN is a
dimer with two identical chains configured in an antiparallel fashion. (B) VCN’s sequence with the
Arg-Gly-Asp tripeptide motif depicted in bold whereas VCN’s N-terminal extra residue and C-
terminal graft are italicized and underlined. Unlike CN, mass spectrometry data shows that VCN
is a monomer.

18

19
reduced disulfide bonds were recognized. At the same time, the finding that the
adhesion of M24 met cells to immobilized CN was completely blocked by RGD-
containing synthetic peptides and EDTA demonstrated that CN utilizes the
common RGD motif for binding (Trikha, De Clerck et al. 1994). Structurally, CN
is a cysteine-rich protein (10 cysteines per monomer) displaying almost no
secondary structure and a complex folding pattern that relies on multiple disulfide
bonds (four intrachain and two interchain disulfide bonds) to stabilize its tertiary
structure (Zhou, Hu et al. 2000).
CN was first established as a potent inhibitor of αIIb β3 mediated PA in
1994. PA was measured in human, canine, and rabbit platelet rich plasmas
(PRPs) and CN (0.73 pg/ml) inhibited 10 µM adenosine diphosphate (ADP)-
induced human PA by 50% (Trikha, Rote et al. 1994). In addition to αIIb β3, other
motogenic integrins were identified as binding sites for CN including α5 β1, αvβ3,
and αvβ5 (Trikha, De Clerck et al. 1994; Zhou, Nakada et al. 2000) and other in
vitro assays would soon follow confirming CN’s potential as an anti-integrin
antagonist. The binding of CN to α5 β1 and αvβ3 resulted in inhibition of
adhesion to FN and VN. Also, anti-adhesive and anti-invasive properties were
demonstrated by showing that the blockage of αvβ5 through CN binding could
inhibit αvβ5-mediated adhesion and invasion of ovarian carcinoma cells
(OVCAR-5) (Zhou, Nakada et al. 2000). Around the same time CN treatment
was demonstrated to cause severe disruption of actin cytoskeleton and
disassembly of focal adhesive structures without affecting cellular adhesion to a
reconstituted basement membrane (Ritter, Zhou et al. 2001). It was also shown
20
that CN not only blocks integrin function, but also induces an integrin-mediated
signal transduction pathway leading to tyrosine phosphorylation of different
intracellular proteins, including CAS, FAK and ERK2, in mammary and bladder
cancer cells (Ritter and Markland 2000; Ritter, Zhou et al. 2000). This discovery
indicates that not only is CN an integrin antagonist, but CN initiates a variety of
signaling pathways upon ligation to integrins. Lastly, it was established that CN
possesses potent anti-angiogenic activity when human umbilical vein endothelial
cell (HUVEC) adhesion to VN was inhibited through CN ligation via an αvβ3-
dependent mechanism and indirectly induced apoptosis by detaching HUVECs
from immobilized VN (Zhou, Nakada et al. 1999). HUVEC invasion, migration,
and ability to form neovessels were also shown to be significantly inhibited after
treatment with CN (Golubkov, Hawes et al. 2003). In addition, CN inhibited
angiogenesis in the chick embryo chorioallantoic membrane model (Zhou,
Nakada et al. 1999), solidifying CN’s anti-angiogenic capability.
In addition to the in vitro assays performed with CN, the Markland
laboratory has been investigating the potential of using CN as an anti-cancer
therapeutic since its discovery through in vivo models. The first model used
consisted of pretreating M24met cells with CN in order to prevent the cells from
forming pulmonary metastatic nodules after tail vein injection in a mouse model
(Trikha, De Clerck et al. 1994). A later model utilized MDA-MB-435 cells
implanted into the mammary fat pads of four-week-old female nude mice. In this
orthotopic xenograft model, intratumor injection of CN dramatically inhibited the
growth rate of the breast cancer and significantly reduced the pulmonary
21
metastases (Zhou, Nakada et al. 1999), but because of the inefficiency of this
delivery technique and the known problem of potential immunogenicity of long-
term delivery with purified venom proteins such as echistatin (Shebuski, Ramjit et
al. 1990) and kristrin (Gold, Yasuda et al. 1991; Barker, Bullens et al. 1992),
CN’s delivery strategy soon changed. The laboratory sought to deliver CN by
way of an in vivo liposomal formulation. CN was encapsulated within liposomes
composed of different ratios of high transition temperature lipids, cholesterol, and
lipids derivatized with long-chain polymers (polyethylene glycol) (Swenson, Costa
et al. 2004). Liposomal CN (LCN) drastically improved the half-life of CN within
the circulatory system, increased the amount of CN delivered at the tumor site,
did not react with platelets, and did not elicit an immune response. Liposomal
encapsulation of CN retains full biological activity while allowing for intravenous
administration, a method of drug delivery that is clinically relevant (Swenson,
Costa et al. 2004).
The yield of CN during the purification of disintegrin from pure southern
copperhead venom is very low. This is due to the naturally low amount of CN
present in the native venom as well as the inevitable loss of some of the peptide
during purification. In order for the advancement of CN to the clinic as a
therapeutic agent, this problem had to be overcome. The obvious solution was
to develop a recombinant molecule with the same biological activity as CN, and
that occurred in 2005 with the development of recombinant CN (Minea, Swenson
et al. 2005), now referred to as vicrostatin (VCN) (Figure 2B). Recombinant
production utilized a commercially available double-mutant strain [Origami B
22
(DE3) pLysS] in order to manufacture VCN within the cytoplasm of E. coli. This
Origami B strain contains dominant negative mutations for the enzymes
thioredoxin reductase and glutathione reductase that creates an oxidative
environment pivotal for driving disulfide bond formation and proper folding. The
formation of disulfide bonds in these mutants is dependent on the presence of
cytoplasmic thioredoxins that undergo a role reversal and actively assist the
formation of disulfide bridges, while maintaining the nascent target proteins in
solution. By expressing a recombinant disintegrin as a fusion with thioredoxin A,
directly in the cytoplasm of Origami B, it was possible to accelerate disulfide
bond formation and enhance the solubility of target protein. This scheme works
extremely well because the system essentially tricks the bacterial strain into
making more thioredoxin A, and thus recombinant disintegrin, in order to
overcome the oxidative stress induced by knocking out thioredoxin reductase
and glutathione reductase. A unique Tobacco etch virus protease cleavage site
was engineered just upstream of the recombinant disintegrin in order to facilitate
the subsequent cleavage of the target protein from its thioredoxin fusion partner.
Surprisingly, mass spectrometry (MALDI-TOF and ESI) demonstrated that, unlike
native CN, VCN is a monomer. Based on this analysis, it is hypothesized that
VCN folded correctly in the C-terminal half of the molecule, preserving the
structural element (the adhesive loop), which accounts for its potent biological
activity. Since LCN provided benefits that CN could not, VCN was encapsulated
by the same method producing LVCN, which could be administered in vivo.
LVCN was delivered to MDA-MB-435 human mammary cancer bearing mice with
23
no visible toxicity. Following administration of LVCN there is a significant
inhibitory effect on tumor growth (~80% inhibition) as compared to the PBS-
treated control (Minea, Swenson et al. 2005). The results of this experiment
indicate that VCN functions as effectively as native CN in vivo. However, limited
studies to date on the characterization of VCN in vitro have been performed, this
is important for anti-angiogenic research and progression to the clinic.

1.3 Specific Aims
The goals of this study were to provide insight to what occurs when
integrins engage disintegrin targets. Because integrins have the ability to
participate in inside →out and outside →in signaling (Hynes 2002), multiple
processes were examined intracellularly and extracellularly.
(1) Given that it has already been documented that CN participates in
several signaling pathways leading to multiple phosphorylation
events, it was hypothesized that prolonged CN ligation would have
an effect at the mRNA expression level for proteins involved in these
pathways as well as other molecules involved with integrin function
and signaling.
(2) Since integrins are cell surface transmembrane receptors that
become internalized and re-expressed at the surface in a constant
endocytic pathway, it was hypothesized that CN ligation would
interfere with the coordination of this pathway and influence the
rates at which this occurs. In order to do this it was necessary to
24
develop a novel way to measure receptor internalization that was
more precise than the classical techniques.
(3) Given that CN exhibits properties that cannot be mimicked by
synthesized molecules possessing the same binding sites of CN,
even though they possess similar binding kinetics, it was
hypothesized that CN induces integrin inactivation via an
inside →out signaling mechanism causing a more potent response to
ligation vs. a simple antagonistic molecule.
(4) Lastly, it was of utmost importance to the laboratory that the
recombinant molecule VCN be characterized in several in vitro anti-
invasive, signaling, and anti-angiogenic assays in order to determine
if VCN retains the same potency as the native CN molecule.

25
Chapter 2: Contortrostatin Effects Gene Expression at the
mRNA Level

2.1 Summary
The study of integrin/disintegrin interaction has primarily been at the
antagonist level and not much is known about what this interaction initiates within
cells. The disintegrin CN, isolated from the venom of Agkistrodon contortrix
contortrix (southern copperhead) and intensively studied in the Markland
laboratory, has been shown to possess potent anti-angiogenic and anti-migratory
activity, but CN has never been studied at the gene expression level. In this
study CN’s effect on gene expression is explored at the mRNA level after CN
engages integrin targets as a signaling molecule. It was discovered, through
Oligo GEArrays, that CN binding induces a number of genes to be up-regulated
while not causing down-regulation of any genes. Several important transcription
factors had their gene expression directly affected by CN treatment including
CCAAT/enhancer binding protein beta (CEBPB), Nuclear factor of kappa light
polypeptide gene enhancer in B-cells 1 (NFKB1), and Jun B proto-oncogene
(JUNB) in both MDA-MB-435 and MDA-MB231 cells. Interestingly, Jun
oncogene (JUN) was only up-regulated in MDA-MB-435 cells while Cyclin-
dependent kinase 2 (CDK2) and Fibronectin 1 (FN1) only demonstrated
significant up-regulation in MDA-MB-231 cells. To further test these results
several RT-PCR experiments verified that CN ligation induces transcription of
important genes in MDA-MB-435 and MDA-MB-231 cells in a cell specific
26
manner including FN1, JUN, v-Fos (FOS), TP53 (p53), Myc (C-myc), Secreted
phosphoprotein 1 (SPP1), E-cadherin (CDH1), and Cortactin (CTTN).

2.2 Introduction
Disintegrins are a complex family of snake venom proteins that were
discovered in the late 1980’s (Gould, Polokoff et al. 1990). Almost exclusively, all
of the work that has been on these molecules involves their potent antagonistic
association with integrin receptors. Because of this, disintegrins are well known
as anti-invasive, anti-adhesive, and anti-angiogenic molecules. CN, a disintegrin
isolated in the Markland laboratory (Trikha, Rote et al. 1994), has proven to
possess these defining characteristics (Zhou, Nakada et al. 2000; Golubkov,
Hawes et al. 2003), but not many experiments have been conducted regarding
the internal signaling of this molecule and no gene expression profiling has been
associated with CN.
Integrin signaling initiates many signal transduction pathways that are vital to
cell behavior. Also, because integrins possess the ability to signal in both
directions (outside →in and inside →out (Hynes 2002)), are the key receptor in the
formation of specialized structures known as FAs, and are the cells primary
means to communicate with the ECM, they can effect a broad range of signaling
proteins including cytoskeletal/scaffolding proteins (tensin, vinculin, paxillin, α-
actinin, parvin/actopaxin and talin), tyrosine kinases (Src, FAK, PYK2, Csk and
Abl), serine/threonine kinases (ILK, PKC and PAK), modulators of small
GTPases (ASAP1, Graf and PSGAP), tyrosine phosphatases (SHP-2 and LAR
27
PTP) and other enzymes (PI 3-kinase and the protease calpain II) (Zamir and
Geiger 2001). These proteins work together, in a complex network, in order to
facilitate integrin signaling. Interestingly, many of these components possess
multiple binding sites for additional proteins, theoretically allowing them to
complex in different ways leading to formation of multiple supramolecular
structures at the FA site (Zamir and Geiger 2001). This also makes regulation of
the expression of these molecules vital to the type of structure that can form
leading to specific regulation of certain pathways.
Because integrins are intimately involved with how a cell recognizes its
environment, it would stand to reason that any antagonistic signal utilizing
integrins, like CN, might cause the cell to change the expression level of proteins
associated with its ECM environment. It is also plausible that after a disintegrin
initiates integrin signaling the expression profile for scaffolding proteins
associated with the cytoplasmic tails of integrins would be altered. Lastly,
because it has already been shown that CN has a drastic effect on cytoskeletal
organization (Ritter, Zhou et al. 2001), it is possible that disintegrins up or down-
regulate critical proteins associated with the maintenance and regulation of the
cytoskeleton.

2.3 Materials and Methods
2.3.1 Cell Culture
MDA-MB-435 human cancer cells (Chambers 2009; Lacroix 2009), an
estrogen receptor-negative cell line, were obtained from Dr. Janet Price (MD
28
Anderson Cancer Center, University of Texas, Houston, TX) and MDA-MB-231
cells (ATTC, Manassas, VA) were maintained in Dulbecco’s modified Eagle’s
medium (DMEM) supplemented with 10% fetal bovine serum (FBS) in the
presence of 5% CO
2
. Cells were serum starved overnight and then harvested by
brief trypsinization with 10% trypsin stock (0.05% trypsin- 0.02% EDTA) in
phosphate buffered saline (PBS) for 5-min, followed by quenching in 0.2%
soybean trypsin inhibitor. Cells were resuspended in serum free media and left
in suspension at 37 °C for 1-hour before initiating experiments.

2.3.2 Oligo GEArrays
For the ‘human extracellular matrix and adhesion molecules’ (HEM & AM)
arrays (SABiosciences, Frederick, MD), twelve well plates were coated with
growth factor reduced (GFR) Matrigel (BD Biosciences, San Jose, CA) and left
overnight at 37°C. Residual media was aspirated out of the wells and MDA-MB-
435 cells (4x10
5
) cells were seeded in serum free media onto the polymerized
Matrigel layer. Cells were incubated for 1-hour at 37°C before being left
untreated or treated with 500 nM CN. After an incubation period of either 4 or
18-hours cells were washed twice with PBS. For the ‘human signal transduction
pathway finder’ (HSTPF) arrays (SABiosciences, Frederick, MD) either MDA-MB-
435 cells (4x10
5
) or MDA-MB-231 cells (3x10
5
) were kept in suspension while
being treated with or without 100 nM CN for 4-hours. After treatment, cells were
washed twice in PBS and retrieved via centrifugation at 4000 rpm.
29
Cells for both arrays were then lysed and RNA was retrieved using
RNAeasy kit (Qiagen, Valencia, CA). RNA purity was established though OD
and concentration was determined using the following method.

Absorbance in 10mM Tris HCl pH 7.5
Pure RNA = Abs 260/Abs 280 = 1.9-2.1

Concentration (µg/µl) = (OD 260) x 40 µg/ml x (Dilution Factor)
Yield (µg) = Concentration (µg/µl) x Xµl

A total of 1 µg of total RNA was isolated for each condition and converted to
cDNA, which was then used to synthesize cRNA while incorporating biotinylated-
UTP using Truelabeling-AMP 2.0 kit (SABiosciences, Frederick, MD). After
amplification, cRNA yield was determined and 2.0 µg cRNA was hybridized to
each array membrane following the manufactures protocol. Detection was
initiated through incubation with AP-Streptavidin (1:8000) for 10-min on the
bench top at room temperature (RT). After incubation, membranes were washed
three times in 4 ml RT buffer (supplied by the manufacturer) for 5-min each.
After a brief wash in a different buffer (supplied by the manufacturer) for 1-min
each membrane was incubated in 1.0 ml CDP-Star (developing agent) for 5-min.
Excess CDP-Star was removed by touching the corner of the array with a clean
tissue and the membranes were imaged by exposure to x-ray film.

30
2.3.3 RT-PCR
MDA-MB-435 cells (5x10
5
) or MDA-MB-231 cells (5x10
5
) were kept in
suspension while being treated with or without 100 nM CN for 4-hours. Cells
were washed twice in PBS and RNA was stabilized via treatment with RNAlater
solution (Qiagen, Valencia, CA) for 5-min. Following the procedure above, cDNA
was prepared from the isolated total RNA. Concentration and purity was
established by using the above formulas as well. Aliquots of cDNA were
prepared and frozen at -20
o
C for each condition.
Gene sequences were determined using the database from GenBank and
primers were constructed with the use of Primer 3.0 software (Table 1). For
each condition, 2 µl of cDNA was mixed with SYBR Green Supermix (Bio-Rad,
Hercules, CA). Appropriate primers were added to each well and PCR reaction
was performed for 3-hours.

2.4 Results
2.4.1 Gene Expression Changes Induced by Contortrostatin
(Oligo GEArrays)

In order to get a broad picture at gene expression change at the mRNA
level, induced by CN treatment, Oligo GEArrays were utilized. In this way it is
possible to examine over 100 genes at once in order to get an overall view of the
genome in specific areas of interest. The two gene arrays that were chosen
were ‘human extracellular matrix and adhesion molecules’ (HEM & AM) as well
as ‘human signal transduction pathway finder’ (HSTPF).
Table 1

RT-PCR Primer Sequences. Gene sequences were determined using the database from
GenBank and primers were constructed with the use of Primer 3.0 software.

31
32
For the HEM & AM array, MDA-MB-435 cells demonstrated very little
change at the gene expression level for both time points in conjunction with 500
nM CN treatment (Figure 3). What was noticeable was that all the genes that
were expressed for this array in MDA-MB-435 cells saw a slight increase in
expression and no genes seemed to demonstrate a decrease in expression.
This was a consistent trend that was verified by no change in the expression of
the control genes.
When the HSTPF array was utilized multiple genes were turned on during
4-hours of CN treatment at 100 nM. What was also evident was that there was a
stark difference between MDA-MB-435 cells and MDA-MB-231 cells (Figure 4).
MDA-MB-435 cells demonstrated highly up-regulated genes such as JUN and
FOS. In contrast, in MDA-MB-231 cells 100 nM CN induced a high expression
change in a large variety of genes including Baculoviral IAP repeat-2, baculoviral
IAP repeat-5, Bone morphogenetic protein 2, CDK2, Cathepsin D, FN1, Heat
shock 27kDa protein 1, Heat shock protein 90kDa alpha, Insulin-like growth
factor BP 3, Insulin-like growth factor BP 4, v-Myc myelocytomatosis, NGFI-A
binding protein 2, Nuclear factor kappa light enhancer in B-cells inhibitor,
Peroxisome proliferator-activated receptor gamma, Transferrin receptor
(p90,CD71), and p53. However, multiple genes were highly up-regulated in both
cell lines including CEBPB, Engrailed homeobox 1, JUNB, Kappa light enhancer
B-cells, v-Myc myelocytomatosis, NGFI-A binding protein 2, and NFKB1.
Interestingly it seems that no genes over both arrays were down-regulated.

Figure 3

HEM & AM Microarray. Twelve well plates were coated were coated with GFR Matrigel
and left overnight at 37°C. Cells were seeded in serum free media onto the polymerized Matrigel
layer. Cells were incubated for 1-hour at 37°C before being left untreated or treated with 500 nM
CN. After an incubation period of either (A) 4 or (B) 18-hours cells were washed twice with PBS.
Cells were then lysed and a total of 1 µg of total RNA was isolated for each condition and
converted to cDNA, which was then used to synthesize cRNA while incorporating biotinylated-
UTP. After amplification, cRNA yield was determined and 2.0 µg cRNA was hybridized to each
array membrane following the manufactures protocol. Detection was initiated through incubation
with AP-Streptavidin (1:8000) for 10-min on the bench top at RT. After incubation, membranes
were washed three times in 4 ml RT buffer F for 5-min each. After a brief wash in buffer G for 1-
min each membrane was incubated in 1.0 ml CDP-Star for 5-min. Excess CDP-Star was
removed by touching the corner of the array with a clean tissue and the membranes were imaged
by exposure to x-ray film.

33
Figure 4

HSTPF Microarray. Either (A) MDA-MB-435 cells (4x10
5
) or (B) MDA-MB-231 cells (3x10
5
)
were kept in suspension while being treated with or without 100 nM CN for 4-hours. Cells were
then lysed and RNA was retrieved. A total of 1 µg of total RNA was isolated for each condition
and converted to cDNA, which was then used to synthesize cRNA while incorporating
biotinylated-UTP. After amplification, cRNA yield was determined and 2.0 µg cRNA was
hybridized to each array membrane following the manufactures protocol. Detection was initiated
through incubation with AP-Streptavidin (1:8000) for 10-min on the bench top at RT. After
incubation, membranes were washed three times in 4 ml RT buffer F for 5-min each. After a brief
wash in buffer G for 1-min each membrane was incubated in 1.0 ml CDP-Star for 5-min. Excess
CDP-Star was removed by touching the corner of the array with a clean tissue and the
membranes were imaged by exposure to x-ray film.

34
35
2.4.2 Gene Expression Changes Induced by Contortrostatin (RT-
PCR)

RT-PCR allowed for a quantitative approach, compared to Oligo GEArrays,
in order to measure gene expression change at the mRNA level of specific
genes. In this experiment 100 nM CN treatment for 4-hours caused MDA-MB-
231 cells to increase mRNA expression of CTTN (2-fold increase), FN1 (4.2-fold
increase), C-myc (2.8-fold increase), SPP1 (3.6-fold increase), JUN (1.8-fold
increase), and FOS (1.8-fold increase) (Figure 5). Interestingly p53 and CDH1
mRNA levels remained unchanged with CN treatment. In the same experiment,
MDA-MB-435 cells demonstrated an increase in gene expression for JUN (1.6-
fold increase), FN1 (1.5-fold increase), C-myc (3.4-fold increase), and SPP1 (1.5-
fold increase) while CTTN, FOS, and p53 exhibited non-significant changes
(CDH1 primers failed to work for MDA-MB-435 cells) (Figure 6).

2.5 Discussion
During the initial studies for any molecule that may participate in differential
gene expression it is important to get a global view of the genome in order to
utilize as many genes as possible. In laboratories that are not able to utilize
whole genome analysis, the next best option is to use large oligo arrays that can
examine over 100 genes at once.
Two oligo arrays were initially chosen, oligo GEArray HEM & AM and
HSTPF microarrays. Logically these two were the appropriate ones to start with.
The HSTPF array profiles the expression of 113 genes representative of 18
Figure 5

CN Effect on MDA-MB-231 Gene Expression. MDA-MB-231 cells (5x10
5
) were kept
in suspension while being treated with or without 100 nM CN for 4-hours. cDNA was prepared
from the isolated total RNA. For each condition, 2 µl of cDNA was mixed with SYBR Green
Supermix. Appropriate primers were added to each well and PCR reaction was performed for 3-
hours.

36
Figure 6

CN Effect on MDA-MB-435 Gene Expression. MDA-MB-435 cells (5x10
5
) were kept
in suspension while being treated with or without 100 nM CN for 4-hours. cDNA was prepared
from the isolated total RNA. For each condition, 2 µl of cDNA was mixed with SYBR Green
Supermix. Appropriate primers were added to each well and PCR reaction was performed for 3-
hours.

37

38
signal transduction pathways that are vital to understanding what pathways are
initiated via CN treatment. The HEM & AM array profiles the expression of 113
genes key to the functions of cell adhesion. This array contains ECM proteins
including basement membrane constituents, collagens, and genes playing a role
in ECM structure. Also, proteases involved in degradation of the ECM were
included as well as their inhibitors. This array also represents molecules
important to cell adhesion including molecules involved in cell-cell and cell-matrix
adhesion, transmembrane molecules, and others. These molecules are integral
to the mechanics of integrin/disintegrin interaction and their mRNA expression
levels could be altered by CN treatment. Through a simple side-by-side
hybridization experiment it was possible to determine differential gene expression
between CN treated samples and untreated samples.
According to the HEM & AM microarray, CN failed to initiate the expression
of any genes outside of the basal expression in MDA-MB-435 cells (Figure 3).
Nonetheless, it was noticed that a great majority of the genes that were basally
expressed had their expression levels slightly increased as a result of CN
treatment. This could be a clear indication that disintegrin signaling through
integrins activates some transcription factors. This finding begs the question,
what specific transcription factors are being affected by CN treatment?
The HSTPF microarray provided a wealth of data as well shedding some
light on to which transcription factors CN treatment effects. With this array two
cell lines were compared, MDA-MB-435 cells and MDA-MB231 cells, and some
similarities as well as some stark differences were noticed. Similarly to the HEM
39
& AM microarray, there was a general trend of up-regulation and no genes were
observed to be down-regulated for either cell line (Figure 4). Some notable
genes whose transcription was significantly up-regulated in both cell lines
included CEBPB, NFKB1, and JUNB, all of which code for transcription factors.
The CEBPB gene encodes for a protein that is a basic leucine-zipper
transcription factor (Ramji and Foka 2002), which performs diverse functions
including the regulation of genes that contribute to the acute phase response,
glucose metabolism, and tissue differentiation, including adipogenesis and
hematopoiesis (Lekstrom-Himes and Xanthopoulos 1998; Matsuda, Kido et al.
2010). Another transcription factor that was up-regulated in both cell lines was
nuclear factor-kappa B (NF- κB) that is a product of the NFKB1 gene. Activated
NF- κB regulates transcription of over 400 genes involved in immunoregulation,
growth regulation, inflammation, carcinogenesis, and cell survival by binding to
specific DNA sequences in target genes, designated as κB elements (Li and
Sethi 2010). Lastly, JUNB encodes for the Jun-B protein, which is member of the
AP-1 (activator protein-1) family of dimeric transcription factors (Piechaczyk and
Farras 2008). The best-studied components of the AP-1 transcription complex
are the members of the FOS and JUN families. Jun-B is known as an inhibitor of
cell division (Bakiri, Lallemand et al. 2000; Passegue and Wagner 2000), an
inducer of senescence (Passegue and Wagner 2000) and a tumor suppressor
(Piechaczyk and Farras 2008).
There were also several key genes that were highly expressed, due to CN
treatment, but were specific to each cell line. In MDA-MB-435 cells the gene
40
encoding for the transcription factor c-jun, which is also an important member of
the AP-1 complex, was highly up-regulated while being unaffected in MDA-MB-
231 cells. This finding clearly indicates that CN induces differential transcription
between cell lines. Also, because multiple members of the AP-1 complex are
being expressed at different rates, transcription differentiation will be exacerbated
as this complex demonstrates different capabilities denoted by which
transcription factors are present (Chinenov and Kerppola 2001). In MDA-MB-231
cells the CDK2 gene and the FN1 gene, which produce cyclin-dependent kinase
2 and FN respectively, had significantly higher transcription rates than in MDA-
MB-435 cells after CN treatment. CDK2 is interesting because this provides
evidence that CN can potentially alter a cell’s fate within the cell-cycle, and that
this fate may be cell type specific. Also, the differential regulation of FN1
between cell lines may mean that CN treated cells try to alter their environmental
cues by changing the composition of certain ECM components, a way of
inside →out signaling (Hynes 2002).
Once it was clear that CN treatment was having an effect at the mRNA level
of many proteins it became necessary to perform focused profiling experiments
utilizing RT-PCR in order to validate the findings. Several genes were selected
based on the previous microarray data as well as some suspect genes that may
be influenced by CN treatment. As in the HSTPF microarray, two cell lines were
tested, MDA-MB-435 and MDA-MB-231 after being treated with 100 nM CN for
4-hours (Figure 5 and 6). The genes selected based on the microarray data
were FN1, JUN, FOS, p53, and C-myc. Several conditions correlated very nicely
41
with the data observed in the microarray experiments including a 4.2-fold
increase in expression of FN1 in MDA-MB-231 cells, a 1.6-fold increase in
expression of JUN in MDA-MB-435 cells, and no observable change in
expression of p53 in MDA-MB-435 cells. However some observation made by
RT-PCR experiments failed to correlate with microarray data such as p53 having
no observable change in MDA-MB-231 cells and FN1 demonstrating a 1.5-fold
increase in expression in MDA-MB-435 cells.
Interesting observations were also made on the three genes that were
selected for RT-PCR based on logical connections with CN treatment, they were
SPP1, ECDH, and CTTN. Osteopontin, coded for by SPP1, was chosen
because it is an extracellular structural protein that is organic component of bone
and the secreted protein interacts with integrins and CD44, mediates cell
adhesion, migration and tumor invasion, and has anti-apoptotic effects (Buback,
Renkl et al. 2009). Interestingly, MDA-MB-231 cells treated with CN
demonstrated a 3.6-fold increase in expression of SPP1 while MDA-MB-435 cells
only demonstrated a 1.5-fold increase. By means of secreted osteopontin in
combination with the effects observed on FN1, CN could further influence a
neighboring cells ability to migrate or invade without direct binding. The intent
behind choosing CDH1, as an RT-PCR candidate gene, was to see if CN
treatment could have an inside →out signaling effect on the expression of
receptors that were involved with adhesion, but were not in the integrin family.
Integrins are the cells primary means of attachment to the ECM while cadherins
are the cells primary means of attachment to other cells (Alberts 2002).
42
Cadherins are also capable of forming intercellular adhesive complexes
(adherens junctions) that are essential for the maintenance of tissue integrity
(May, Doody et al. 2005; Niessen 2007). Unfortunately, both of the tested cell
lines failed to show any expression change in CDH1. The last gene that had its
expression level measured by RT-PCR was CTTN, which encodes for cortactin.
Neither cadherins nor integrins act alone in their adhesive abilities. Instead, they
collaborate with the cytoskeleton via specific regulators to achieve biological
processes (Ren, Crampton et al. 2009). One of the more important molecules
that regulates actin is cortactin, a monomeric protein located in the cytoplasm of
cells that can be activated via external stimuli such as integrin ligation, to
promote polymerization and rearrangement of the actin cytoskeleton, especially
the actin cortex around the cellular periphery (Cosen-Binker and Kapus 2006;
Ammer and Weed 2008). Due to the already documented effect that CN has on
the disruption of the actin cytoskeleton (Ritter, Zhou et al. 2001), cortactin was a
logical molecule whose expression could be altered via CN signaling. What was
noticed was that CTTN expression increased 2-fold in MDA-MB-231 cells as a
result of CN treatment, but MDA-MB-435 cells remained unaffected.
The major differences between the two cell lines tested in these gene
expression experiments could be almost exclusively due to the integrin profile of
the cells. It is known that CN binds with high affinity to several different integrins
including αIIb β3, α5 β1, αvβ3, and αvβ5 (Trikha, De Clerck et al. 1994; Zhou,
Nakada et al. 2000), and it is also known that different cell types vary greatly in
not only the integrins they express but also the amount of each integrin on the
43
cell surface. MDA-MB-231 cells in particular seem to have an enormous amount
of beta1 integrins expressed on their surface while MDA-MB-435 cells express a
large quantity of the integrin αvβ3. Dissimilar integrin expression profiles could
lead to different outcomes when these cells are exposed to soluble disintegrin
binding in regards to many cellular characteristics including mRNA transcription.
For example, if one particular cell line has a majority of integrins on its surface
that bind FN and this cell line is treated with a disintegrin that binds these specific
integrins, then maybe the expression of the FN1 gene would be greatly altered in
response. On the contrary, if a different cell line that does not express the same
FN associated integrins is treated with the same disintegrin, then maybe that cell
line’s FN1 gene would not be affected.
In the future more experiments need to be conducted that utilize cell lines
which have their entire integrin expression profiles meticulously measured to
verify if the outcome of CN treatment is altered by specific integrin ligation. Also,
more gene constructs need to be made for RT-PCR experiments. Lastly, whole
genome analysis must be conducted to get a global view of expression change
initiated by CN ligation.

44
Chapter 3: Contortrostatin Induced Integrin Internalization and
Development of a Novel Measurement Technique

3.1 Summary
Integrins are important transmembrane receptors that participate in
endocytic pathways and are becoming implicated in viral entry into cells. In order
to investigate how disintegrin ligation effects integrin internalization an accurate
way to measure in verses out signal must be established. For internalization
experiments that use fluorescent antibody (Ab) staining to distinguish between
inside versus outside cellular localization of various receptor targeting ligands, it
is critical that there be efficient removal of all residual surface-bound fluorescent
Ab. To achieve this, a fluorescent Ab removal technique is commonly employed
in receptor internalization assays that utilizes low pH glycine-based buffers to
wash off the residual non-internalized fluorescent Ab retained on cell surfaces.
In this study, the shortcomings of this technique are highlighted and an
alternative in situ proteolytic approach that was found to be non-deleterious to
cells and significantly more effective in removing residual fluorescence resulting
from non-internalized surface-bound Ab is proposed. This novel technique was
utilized to shed some light on the effects that CN, a dimeric disintegrin, has on
beta1 integrin internalization rates.

3.2 Introduction
Integrins are heterodimeric transmembrane receptors that are vital to the
way cells transverse through the ECM. They provide the means for a cell to
45
migrate, invade, and grip to the ECM or other cells, but it is becoming clear that
integrins may have a more sinister role in nature. Integrins are prime examples
of physiologically important receptors that have been usurped by nonenveloped
and enveloped viruses for attachment and/or cell entry (Stewart and Nemerow
2007).
Disintegrins are small cysteine-rich proteins that are isolated from the
venom of hemotoxic snakes (Gould, Polokoff et al. 1990). They have been
shown to be potent integrin antagonists that firmly inhibit integrin function through
direct binding. Even though they have been studied for over 20 years now, little
is known about how they effect integrin receptor internalization or if they are even
internalized themselves.
Receptor internalization assays are important for understanding how a
particular ligand or receptor is involved in endocytosis. What is being measured
by these assays is the rate the ligand/receptor pair is internalized into the
cytoplasm of the cell. This information can be used for mechanistic studies in
multiple areas of research such as receptor-mediated endocytosis and drug
uptake.
One common way to explore the rate of cellular uptake for a particular
ligand is to label the ligand with a fluorescent tag and estimate the amount of
fluorescence that becomes internalized over time by either flow cytometry (FCM)
(Bernhagen, Krohn et al. 2007) or confocal microscopy (Casartelli, Cermenati et
al. 2008). During these ligand/receptor internalization experiments, the ability to
distinguish between the amount of ligand that remains on the cell surface and the
46
amount that becomes internalized during the course of the experiment is a
significant challenge. With either technique, in order to obtain an accurate
estimate of the amount of fluorescence inside the cell that is derived from
internalized ligands, it is critical to be able to efficiently wash off 100% of the
residual fluorescence (i.e., the fluorescent ligands that were not internalized
during the time course of the experiment). If the fluorescent signal remaining on
the surface is not close to zero after washing at the end of the internalization
experiment, what is measured as the intracellular signal is rendered unreliable
(Bernhagen, Krohn et al. 2007). When the ligand analyzed is a targeting Ab that
has been labeled with a fluorescent tag, removal of the non-internalized Ab from
the cell surface can be achieved, but depends on the strength of the washing
buffer. When the analyzed ligand is not an Ab but a protein that forms a stable
complex with its receptor, removal of residual cell surface fluorescence (i.e., the
non-internalized ligands) becomes quite difficult. Nonetheless, when analyzing
internalization of a non-antibody ligand that interacts tightly with its receptor, one
solution to the residual fluorescence problem would be to label the ligand with a
specific fluorescent Ab. The interaction between the fluorescent Ab and the
ligand is expected to be weaker than that between the ligand and its cell surface
receptor. Therefore, by using buffers that dissociate the Ab from the bound
ligand, one should be able to efficiently remove residual fluorescence associated
with the ligand. The solution to the residual fluorescence problem in
internalization experiments is to find a method that efficiently removes
fluorescent Ab from the cell surface.
47
The classical approach for stripping Ab from cell surfaces requires
washing cells at the end of the experiment in a simple acidic buffer containing
either 50mM glycine (Altankov and Grinnell 1995; Gao, Curtis et al. 2000;
Bernhagen, Krohn et al. 2007) or 100mM Na Acetate (Carroll, Beattie et al.
1999). However, after repeatedly testing this approach in the Markland
laboratory, it was concluded that these buffers are very inefficient at displacing
surface-bound Ab. To address this problem, it was decided to use a different
approach and explore the efficacy of removing surface-bound Ab by employing a
proteolytic enzyme known to cleave specific regions of the Ab.
The proteolytic enzyme pepsin, which is crucial for digestive processes in
the stomach, is synthesized from pepsinogen and secreted by the gastric chief
cells (Gritti, Banfi et al. 2000). Pepsin cleaves preferentially at the C-terminal
end of aromatic amino acids such as phenylalanine and tyrosine (Kageyama
2002). Pepsin worked most efficiently at removing surface-bound Ab without any
detrimental effects on the assay. When incubated with an immunoglobulin G
(IgG), pepsin is known to proteolytically separate the Ab into two fragments, the
bivalent fragment, antigen-binding (F(ab ′)2) region and the fragment,
crystallizable (Fc) region, by specifically cleaving the Ab between these two
regions (Andrew and Titus 2001) (Figure 7A). It was theorized that this
proteolytic action of pepsin could also be employed in situ to strip away
fluorescently labeled Ab from cell surfaces. In these experiments, either an anti-
integrin or anti-transferrin primary antibody and an Fc-specific Alexa-488
(Invitrogen, Carlsbad, CA) conjugated secondary antibody (Fc A488 secondary
Figure 7

Fc-specific Secondary Antibody Requirement for Pepsin Removal of the
Residual Fluorescent Signal. (A) Pepsin cleaves antibodies at the junction between the
bivalent F(ab ′)2 fragment and the Fc fragment. (B) Diagram illustrating the need for an Fc-
specific secondary Ab in order for the pepsin cleavage to eliminate the Alexa-488 signal from the
cell surface. After protelolytic cleavage, cells that were treated with a non-Fc-specific Ab still
have the Alexa-488 F(ab ′)2 attached to the primary F(ab ′)2, while cells that were treated with the
Fc-specific Ab no longer have any portion of the Fc A488 secondary Ab attached to the primary
F(ab ′)2.

48
49
Ab) were used in combination with pepsin digestion to demonstrate, by both FCM
and confocal microscopy, that in situ proteolysis could effectively remove labeled
secondary Ab.
Because pepsin specifically cleaves the Ab at the junction between the
F(ab ′)2 fragment and the Fc fragment (Andrew and Titus 2001), it was possible to
take advantage of this cleavage specificity by using an A488 secondary Ab that
is Fc-specific. Proteolysis by pepsin will cleave both the primary and the
secondary Ab bound on the cell surface. When this happens the F(ab ′)2
fragment of the primary antibody will be left in place, but the Fc fragment of the
primary Ab along with the bound F(ab ′)2 fragments from the secondary Ab will be
cleaved off the cellular surface (Figure 7B). This results in a very efficient
removal of any surface fluorescence. If a secondary Ab that was not Fc-specific
was used, the remaining F(ab ′)2 fragment from the primary Ab would still have
fluorescent F(ab ′)2 fragments from the Fc A488 secondary Ab bound to the cell
surface, resulting in persistent signal even after proteolysis.

3.3 Materials and Methods
3.3.1 Cell Culture and Antibodies
MDA-MB-435 human cancer cells (Chambers 2009; Lacroix 2009), an
estrogen receptor-negative cell line, were obtained from Dr. Janet Price (MD
Anderson Cancer Center, University of Texas, Houston, TX) and maintained in
DMEM supplemented with 10%FBS in the presence of 5% CO
2
. Cells were
serum starved overnight and then harvested by brief trypsinization with 10%
50
trypsin stock (0.05% trypsin- 0.02% EDTA) in PBS for 5-min, followed by
quenching in 0.2% soybean trypsin inhibitor. Cells were resuspended in serum
free media and left in suspension at 37 °C for 1-hour before initiating experiments.
Abs used were anti-contortrostatin mAb produced in the Markland laboratory by
hybridoma; anti-beta1 integrin (clone P5D2), anti-mouse HRP, and anti-
transferrin receptor (TFR) Ab (clone 9F81C11) from Santa Cruz Biotechnology
(Santa Cruz CA); anti-mouse Fc A488 secondary Ab from Jackson
ImmunoResearch (West Grove, PA) where the secondary Ab was conjugated to
Alexa 488 dye following Invitrogen’s protocol.

3.3.2 Integrin/Disintegrin Internalization by Western Blot
MDA-MB-435 cells (10
6
) were incubated with either 0 or 100 nM CN for
varying lengths (0-min, 5-min, 3-hours, or 24-hours) of time while in suspension
in serum free media. After each successive treatment, cells were immediately
put on ice for the remainder of the experiment. One group of cells was subjected
to trypsinization with 0.05 mg/ml cold trypsin-0.02% EDTA for 1-hour on ice while
the other group was left alone. Trypsinization was quenched by adding 0.2%
soybean trypsin inhibitor. Cells were then washed in PBS and lysed in a cold
modified radioimmunoprecipitation (RIPA) buffer (50mM tris, pH 8.0, 150mM
NaCl, 1% nonionic detergent Igepal CA-630, 0.5% deoxycholate, 0.1% SDS,
protease inhibitor cocktail, 1mM sodium pyrophosphate, 1mM sodium
orthovanadate, 50mM sodium fluoride) for 10 to 15-min at 4 °C. Lysates were
passed through a 23-gauge needle, to ensure membrane aggregates were
51
dissociated, and centrifuged at 10,000 RPM. Supernatants were separated from
insoluble material and protein levels were standardized by bicinchoninic acid
(BCA). Whole cell lysates (100 µg total protein) or were resolved on 4-20%
polyacrylamide LongLife gels (NuSep, Lawrenceville, GA). Proteins were
transferred to nitrocellulose membranes overnight at 4 °C and blocked with tween
(0.1%) tris (1.5%, 7.5 pH) buffered saline (TTBS) containing 2% bovine serum
albumin (BSA) for 1-hour at 20 °C. Membranes were probed with anti-beta1
integrin mAb (1:1000) for 2-hours at 20 °C. Membranes were washed five times
in TTBS before being probed with anti-mouse HRP (1:5000) for 2-hours at 20 °C.
Following five additional washes with TTBS the membranes were developed via
chemiluminescence with a SuperSignal kit (34080, Thermo Scientific, Rockford,
IL).

3.3.3 Integrin/Disintegrin Internalization by FACS Analysis
MDA-MB-435 cells (10
6
) were incubated with 100 nM fluorescein
isothiocyanate (FITC) conjugated CN for varying lengths of time (5-min, 5-hours,
or 24-hours) while in suspension in serum free media for at 4°C. Cells were
washed three times with PBS to remove unbound FITC-CN and the incubator
was adjusted to 37°C to allow internalization to proceed. Starting times for the
different experiments were adjusted in order for their completion times to
converge. Cells were washed three times and resuspended in PBS, pH 7.4, for
fluorescence-activated cell sorting (FACS) analysis.

52
3.3.4 Evaluating the Efficiency of Different Stripping Buffers on
Removing Cell Surface Fluorescence

MDA-MB-435 cells (10
6
) were incubated, in suspension, with anti-beta1
mAb for 30-min. Cells were washed three times with PBS and incubated with an
anti-mouse Fc A488 secondary Ab (1:1000, Jackson ImmunoResearch, West
Grove, PA) for 30-min (the secondary Ab was conjugated to Alexa 488 dye
following Invitrogen’s protocol). During the labeling process the cells were kept
at 4 °C to ensure that there was no membrane traffic and that Fc A488 secondary
Ab was not internalized by the MDA-MB-435 cells. These conditions allowed for
maximum surface fluorescence to be retained on the plasma membrane (i.e., no
membrane traffic) and for the evaluation of the efficiency of different buffers in
removing membrane-bound Fc A488 secondary Ab. Three different stripping
buffers were analyzed; two were traditional buffers containing either 50mM
glycine, 150mM NaCl, pH 2.5 (glycine/HCl buffer), or 100mM Na Acetate, 50mM
NaCl, pH 5.5 (acetate/HCl buffer). For the above buffers the cells were washed
for 30-min at 4 °C with gentle agitation. The third buffer was one of the traditional
buffers (glycine/HCl buffer) further supplemented with 0.01 mg/ml pepsin
(pepsin/HCl buffer). In buffer supplemented with pepsin, cells were washed for
15-min at 4 °C with gentle agitation. After the stripping step, the cells were either
fixed in 3.7% formaldehyde and mounted onto coverslips with fluorescent
mounting media (KPL, Gaithersburg Maryland) for confocal microscopy analysis,
or washed and resuspended in PBS, pH 7.4, for FACS analysis.

53
3.3.5 Internalization Rates of Beta1 Integrin and Transferrin
Receptors

MDA-MB-435 cells (10
6
) were labeled with either the anti-beta1 integrin Ab
(1:500) or the TFR Ab (1:500, clone 9F81C11, Santa Cruz Biotechnology, Santa
Cruz CA), washed and further stained with an anti-mouse Fc A488 secondary Ab
(1:1000). Cells were washed three times and incubated with Ab for 30-min at
4 °C. In evaluating beta1 integrin receptor internalization cells were transferred to
a 37 °C incubator and internalization of the integrin-bound Ab was allowed to
proceed for 5-min, 1-hour, or 3-hours. Cells with TFR-bound Ab were
resuspended in complete media, supplemented with 200 nM transferrin (Barroso-
Gonzalez, Machado et al. 2009), then left to internalize at 37 °C for 5-min, 20-min,
or 60-min. At the end of each time point, cells were either washed in glycine/HCl
stripping buffer for 30-min, pepsin/HCl stripping buffer for 15-min, or PBS for 15-
min. Washings were done at 4 °C with gentle agitation and the cells were then
fixed in 3.7% formaldehyde and mounted onto coverslips with fluorescent
mounting media for confocal microscopy.
The confocal images were quantified using the software program Simple
PCI (Hamamatsu, Sewickley, PA). Each representative image was scanned,
with this software, and every fluorescent pixel was counted that was between the
minimum and maximum signal strength which yielded the highest quality image.
These same boundaries were applied to all slides. Any pixel above or below the
boundary was excluded. The same images were then used to calculate the area
of the measured cells. From these two numbers, the amounts of pixels per unit
54
area for each image were obtained. This represents the rate that the beta1
integrins or TFRs are internalized per unit area of cell surface.

3.3.6 Trypan Blue Viability
MDA-MB-435 cells (10
6
) were either left untreated, exposed to UV light for
15-min, treated with actinomycin D (0.05 µg/ml) for 18 hours, or washed in
pepsin/HCl stripping buffer for 15-min at 4 °C with gentle agitation. When each
treatment concluded, cells were stained with 4% Trypan Blue (EMD Biosciences,
San Diego California) in PBS for 5-min and evaluated for cell viability by counting
stained cells on a hemocytometer at x10 magnification. Four separate counts
were taken and averaged, and the experiment was repeated three times.

3.3.7 Integrin/Disintegrin Internalization by Confocal,
Employing Novel Pepsin Technique

MDA-MB-435 cells (10
6
) were labeled with anti-beta1 integrin mAb
(1:500) for 30-min at 4°C, washed with PBS, and further stained with anti-mouse
Fc A488 secondary Ab (1:1000). After another incubation period of 30-min at
4°C, unbound secondary Ab was removed by three PBS washes and cells were
treated with 0 or 100 nM CN for various time intervals (10-min, 12-hours, or 24-
hours) at 37°C to allow internalization to proceed. At the end of each time point,
cells were washed in pepsin/HCl stripping buffer for 15-min. Washings were
done at 4 °C with gentle agitation and the cells were then fixed in 3.7%
55
formaldehyde and mounted onto coverslips with fluorescent mounting media for
confocal microscopy.

3.4 Results
3.4.1 Contortrostatin Induces Integrin Internalization (Western
Blot)

To measure the rate of beta1 integrin internalization induced by CN
ligation simple western blotting techniques were employed coupled with
trypsinization to remove surface bound protein. After an incubation period of 5-
min significant beta1 integrin signals were coming from inside of the cells
indicating that CN induces integrin internalization very rapidly. This signal
continues to decrease as the experiment progresses to 3 and 24-hours (Figure
8B), however it is clearly noticeable that the signal is significantly weaker when
the cells are initially subjected to trypsinization. This indicates that the
typsinization technique does remove some surface integrin.

3.4.2 Contortrostatin Induces Integrin Internalization (FACS
Analysis)

The removal of CN bound to the surface of MDA-MB-435 cells, by
endocytosis of the integrin receptors, was measured through FACS analysis. At
time zero where the surface of the MDA-MB-435 cells are completely coated with
FITC-CN there is a FACS reading of 91.8% (Figure 8A). This signal drops to

Figure 8

CN Induced Integrin Internalization (Western blot & FACS). (A) MDA-MB-435
cells (10
6
) were incubated with 100 nM FITC-CN for varying times (5-min, 5-hours, or 24-hours)
while in suspension in serum free media at 4°C. Cells were washed three times with PBS to
remove unbound FITC-CN and cells were adjusted to 37°C to allow internalization to proceed.
Cells were washed three times and resuspended in PBS, pH 7.4, for FACS analysis. (B) MDA-
MB-435 cells (10
6
) were incubated with either 0 or 100 nM CN for varying lengths of time (0-min,
5-min, 3-hours, or 24-hours) while in suspension in serum free media. One group of cells was
subjected to trypsinization with 0.05 mg/ml cold trypsin-0.02% EDTA for 1-hour on ice while the
other group was left alone. Cells were then washed in PBS and lysed in RIPA buffer (50mM tris,
pH 8.0, 150mM NaCl, 1% nonionic detergent Igepal CA-630, 0.5% deoxycholate, 0.1% SDS,
protease inhibitor cocktail, 1mM sodium pyrophosphate, 1mM sodium orthovanadate, 50mM
sodium fluoride) for 10 to 15-min at 4°C. Whole cell lysates (100 µg total protein) or were
resolved on 4-20% polyacrylamide LongLife gels. Proteins were transferred to nitrocellulose
membranes overnight at 4°C and blocked for 1-hour at 20°C. Membranes were probed with anti-
beta1 integrin mAb (1:1000) for 2-hours at 20°C. Membranes were washed five times in TTBS
before being probed with anti-mouse HRP (1:5000) for 2-hours at 20°C. Membranes were
developed via chemiluminescence with a SuperSignal kit.

56
57
35.7% after an incubation period of 5-hours and continues to drop to 1.5% after
24-hours of incubation.

3.4.3 Classical Stripping Buffers Fail to Remove Residual Cell
Surface Fluorescence

When evaluated at maximum surface fluorescence, meaning that no
fluorescent Ab was internalized, the results clearly demonstrate the advantage of
using pepsin to remove surface bound Ab. Cells were visualized by confocal
microscopy, which showed that the acidic washes done with traditional buffers
were so ineffective that the retained fluorescent signal nearly matched the control
(i.e., cells washed in PBS), indicating very little secondary Ab removal. By
contrast, the pepsin-supplemented acid wash completely abolished the Alexa-
488 signal from the surface of the MDA-MB 435 cells (Figure 9A). These results
were similar to those observed using FACS analysis, where the glycine/HCl wash
showed only a minimal shift in the fluorescent signal as compared to that
observed in the control (i.e., cells washed in PBS); while the acetate wash
showed no shift at all. In contrast, the pepsin/HCl wash brought the FACS signal
down to 1.8% of control (Figure 9B).

3.4.4 Internalization Rates of Beta1 Integrin and Transferrin
Receptors

To demonstrate how the novel pepsin technique can be employed to
effectively and efficiently measure the internalization rates of different receptors,
Figure 9

Effect of pepsin on labeled antibody removal at maximum residual
fluorescence. MDA-MB-435 cells were incubated at 4°C with a mouse monoclonal Ab
against the cell surface beta 1 integrin (30-min), washed with PBS, and then further incubated
with an A488 secondary Ab that is specific for the Fc portion of mouse IgG (30-min). The cells
were then immediately washed with 3 different Ab stripping buffers or PBS at 4°C with gentle
agitation. Row 1: PBS; Row 2: 50mM glycine, 150mM NaCl, pH 2.5; Row 3: 100mM Na Acetate,
50mM NaCl, pH 5.5; Row 4: 0.1 mg/ml pepsin, 50mM glycine, 150mM NaCl, pH 2.5. The cells
were then either (A) fixed in 3.7% formaldehyde and mounted onto coverslips with fluorescent
mounting media (KPL, Gaithersburg Maryland) for confocal microscopy under either white light
(left panels) or light at a wavelength of 488 nM to visualize the Alexa-488 dye conjugated to the
Fc specific secondary Ab (right panels), or (B) washed and resuspended in PBS, pH 7.4, for
FACS analysis.

58
59
two experiments to measure internalization of the beta1 integrin receptor and the
TFR on MDA-MB-435 cells were conducted. These two systems were chosen
because of their different rates of internalization. Beta1 integrin is internalized
constitutively over time in a non-ligand induced manor, which leads to a slower
rate of internalization. Conversely, the TFR was internalized after binding its
natural ligand, transferrin, and is internalized via receptor-mediated endocytosis,
resulting in a faster rate of internalization. The time course chosen for each
experiment was determined by observations made in the Markland laboratory
using MDA-MB 435 cells, as well as internalization experiments conducted by
other laboratories (Gao, Curtis et al. 2000; Sever, Damke et al. 2000; Barroso-
Gonzalez, Machado et al. 2009). The time points of 5-min, 1-hour, and 3-hours
were chosen for the beta1 integrin and the time points of 5-min, 20-min, and 60-
min were chosen for the TFR, in order to allow the maximum amount of
internalized receptors before recycling became the predominant event in the
suspended MDA-MB 435 cells. Representative images from the beta1 integrin
internalization experiment are shown (Figure 10A). The fluorescent signal
coming from the MDA-MB 435 cells after 5-min of internalization at 37
o
C followed
by a PBS wash was considered the negative control with no removal of residual
fluorescence (Figure 10A, Row 1). By comparison, MDA-MB-435 cells washed
in pepsin/HCl buffer after 5-min of incubation at 37
o
C completely lose the cell
surface fluorescent signal (Figure 10A, Row 3). Cells washed in glycine/HCl
buffer after 5-min of incubation still maintained a high amount of surface
fluorescence (Figure 10A, Row 2). At the later time points the signal disparity
Figure 10

The Advantages of Using Pepsin/HCl-stripping Buffer to Accurately
Measure the Internalization Rates of Receptors in Two Systems. (A) MDA-MB-
435 human cancer cells were labeled at 4°C with a beta1-integrin Ab or a TFR Ab, washed, and
further stained with an anti-mouse Fc A488 secondary Ab. The cells treated with the beta1-
integrin Ab and secondary Ab were then transferred to 37°C and allowed to internalize the
integrin-bound antibody for three separate time intervals (5-min, 1-hour, and 3-hours), while cells
treated with the TFR Ab were resuspended in complete media supplemented with 200 nM
transferrin and transferred to 37°C and allowed to internalize the TFR-bound antibody for three
different time intervals (5-min, 20-min, and 60-min). Cells were either washed in glycine/HCl,
pepsin/HCl or PBS, after which they were fixed in 3.7% formaldehyde and mounted onto
coverslips with Fluorescent Mounting Media and analyzed by confocal microscopy.
Representative images only from the beta1 integrin internalization study are shown (A). Alexa-
488 column: images shown at a wavelength of 488 nM; White light column: images shown under
white light; Composite column: super imposed images showing views under both488 nM
wavelength and white light. Row 1: 5-min, PBS; Row 2: 5-min, glycine/HCl; Row 3: 5-min,
pepsin/HCl; Row 4: 1-hour, PBS; Row 5: 1-hour, glycine/HCl; Row 6: 1-hour, pepsin/HCl; Row 7:
3-hours, PBS; Row 8: 3-hours, glycine/HCl; Row 9: 3-hours, pepsin/HCl. Images from the beta1
integrin internalization experiment (B) and the TFR internalization experiment (C) were subjected
to quantification by the Simple PCI software. Fluorescent pixels were counted and divided by the
total area for the counted cells to obtain the amount of pixels per unit area for each image.
60
Figure 10: Continued

61
62
still persisted on the cell membrane with a strong signal coming from cells
washed in either PBS or glycine/HCl and considerably less signal from cells
washed in pepsin/HCl (Figure 10A, Rows 4-9). Thus, the residual surface signal
coming from non-internalized Ab in cells washed in PBS or glycine/HCl
represents a confounding variable in interpreting receptor internalization data. It
is believed that unlike other washing techniques previously employed in receptor
internalization studies, the combination of Fc-specific Ab and pepsin could be
employed to remove surface fluorescence and thereby generate a much more
interpretable signal coming from the internalized receptor-ligand complexes.
Quantitation of beta1 integrin internalization illustrates the efficiency of the
pepsin-based stripping buffer. After 5-min of incubation, the PBS control group
had a pixel per cell area count of 33.6, the glycine/HCl group had a count of 20.2,
while the pepsin/HCl treatment resulted in a count of 0.008 (Figure 10B).
Clearly, the most effective stripping buffer was the pepsin/HCl wash, since after
an incubation time of only 5-min, a very small amount of integrin should be
internalized by the cells. The signal disparity that was seen after a brief
incubation was the result of the inability of buffers, which do not contain pepsin,
to efficiently remove the non-internalized Fc A488 secondary Ab from the cell
surface. At the 1-hour time point, quantitation for the PBS group had a pixel per
cell area count of 39.2, glycine/HCl was 17.0, and pepsin/HCl 4.16. At the 3-hour
time point, the same trend was observed with the PBS pixel per cell area count of
57.5, glycine/HCl of 24.9, and pepsin/HCl 11.6 (Figure 10B). The only
63
internalization rate that demonstrated consistent progression throughout the
three time points was from the pepsin/HCl treatment group.
Quantitation of the TFR data yielded similar results to the beta1 integrin
experiment. After 5-min of incubation, the PBS control group had a pixel per cell
area count of 13.4, the glycine/HCl group had a count of 12.7, while the
pepsin/HCl treatment resulted in a count of 0.9. When compared to the 5-min
time point for the beta1 integrin, the only logical data set was from the pepsin/HCl
treatment group; TFR is expected to be internalized at a greater rate and 0.9
pixels per cell area was observed for TFR and only 0.008 pixels per cell area with
beta1 integrin was observed, while the other two treatment groups demonstrated
lower pixels per cell area counts for the TFR (Figure 10B and 10C) when
compared to the beta1 integrin pixels per cell area counts. At the 20-min time
point, quantitation for the PBS group had a pixel per cell area count of 26.3,
glycine/HCl was 16.1, and pepsin/HCl 4.00. At the 60-min time point, the same
trend was observed with the PBS pixel per cell area count of 27.3, glycine/HCl of
21.0, and pepsin/HCl 14.2 (Figure 10C). Again, the only treatment group that
established a consistent trend for all three time points was from the pepsin/HCl
wash.

3.4.5 Novel Pepsin Technique has no Effect on Cell Viability
Trypan Blue staining confirmed that the pepsin-supplemented acid wash
has no impact on the stability of the cellular membrane of MDA-MB-435 cells
exposed to the pepsin/HCl stripping buffer. Results of pepsin treatment were
64
compared to two agents known to initiate membrane instability and cell death,
UV light and Actinomycin D treatment, which left the majority of the MDA-MB-435
cells non-viable. The untreated cells showed that only 6.3% of the cells were
non-viable, while the positive controls of UV light and Actinomycin D
demonstrated 69.1% and 52.7% to be non-viable. Exposure of these cells to
pepsin treatment resulted in very low cell death of 6.5%, similar to results seen in
the untreated control group (Figure 11).

3.4.6 Contortrostatin Induces Integrin Internalization
(Confocal)
The novel pepsin/HCl wash technique was employed in a new
internalization experiment measuring beta1 integrin internalization influenced by
CN ligation, utilizing confocal microscopy. After an incubation period of 10-min
hardly any integrin is within the MDA-MB-435 cells that were treated with 0 nM
CN (Figure 12). However, at time points 12 and 24-hours the amount of integrin
present within steadily increases. In contrast, the MDA-MB-435 cells that were
treated with 100 nM CN show a large presence of integrin within each cell
already at the 10-min time point which drops off to some extent as the cells
progress to the 12 and 24-hour time points.

3.5 Discussion
In order to evaluate if CN caused any effect on the rate that integrin is
internalized into the cell two crude internalization experiments were conducted.
Figure 11

Cell Viability Assessed by Trypan Blue Staining. (A) Composite of MDA-MB-435
cells after no treatment, exposed to UV light for 15-min, treated with actinomycin D for 18 hours,
or washed in pepsin/HCl stripping buffer for 15-min. When each treatment concluded, cells were
stained with 4% Trypan Blue for 5-min. This figure contains representative images from multiple
experiments. (B) Bar graph showing the percent of non-viable cells for each treatment group.

65
Figure 12

CN Induced Integrin Internalization (confocal). MDA-MB-435 cells (10
6
) were
labeled with anti-beta1 integrin mAb (1:500) for 30-min at 4°C and further stained with anti-
mouse Fc A488 secondary Ab (1:1000). After another incubation period of 30-min at 4°C, cells
were treated with 0 (Row A) or 100 nM CN (Row B) for various time intervals (10-min, 12-hours,
or 24-hours) at 37°C to allow internalization to proceed. At the end of each time point, cells were
washed in pepsin/HCl stripping buffer for 15-min. Washings were done at 4°C with gentle
agitation and the cells were then fixed in 3.7% formaldehyde and mounted onto coverslips with
fluorescent mounting media for confocal microscopy.

66
67
One utilized trypsinization and centrifugation to eliminate CN that was not
internalized into the cells. This experiment indicated that a large portion of beta1
integrin entered the cells within 5-mins after integrin ligation followed by a
degradation of the signal over time (Figure 8B). This would mean that CN was
being rapidly cleared from the cellular surface. To investigate this phenomenon
further a FACS experiment was conducted with FITC-CN bound to the surface of
MDA-MB-435 cells (Figure 8A). By manipulating the exterior temperature of the
cells, the internalization time points could easily be manipulated. With this
experiment it was shown conclusively that CN is completely removed from the
surface of MDA-MB-435 cells after 24-hours of ligation. Even though this
experiment doesn’t prove CN is being internalized, it provides further evidence
that integrin internalization rates could be greatly influenced by disintegrin
treatment.
The same trypsinization technique that was utilized in the beta1 integrin
internalization western blots proved useless when it was employed to precisely
track integrin internalization via Abs during CN treatment. This was because the
trypsinization process failed to remove all of the residual surface fluorescence
that was present in the integrins still at the surface of the cells. It was necessary
to completely remove this signal if the internalization rates were ever to be
measured accurately.
The method described in this dissertation is a simple but efficient
technique for removing residual surface bound fluorescence in receptor
internalization assays. The utility of this procedure becomes evident when it is
68
compared to the classical acid buffer wash recommended by most standard
protocols (Figure 9). Without including pepsin in the wash, the fluorescent signal
generated with classical acidic buffer washes almost exactly matches the PBS
treated controls, indicating that stripping Ab off the cellular surface with traditional
buffers is very ineffective. Furthermore, the traditional Ab stripping buffers have
demonstrated poor results in other types of assays. For example, one group
found that in a surface plasmon resonance experiment the glycine/HCl acid wash
was totally ineffective at disrupting an Ab-receptor complex (Brigham-Burke
1992). The only limiting factor of the method presented herein is that, compared
to the traditional methods, the novel method calls for the usage of an Fc specific
secondary Ab, but these are common and fairly inexpensive reagents.
While evaluating the pepsin technique by means of two internalization
systems, one slow (beta1 integrin) and one fast (TFR), a true advantage for the
novel system emerged. Although all three-treatment groups demonstrated
increasing internalization rates when comparing beta1 integrin to TFR, the PBS
and glycine/HCl groups were inconsistent and unreliable. The reason for this
was due to strong cell-membrane signals (non-internalized complexes) that were
not present in the pepsin/HCl treatment groups. This produced a key difference
between the data sets. Given the conditions for each experiment, from the 1-
hour to the 3-hour time point in the beta1 integrin experiment, and from the 20-
min to the 60-min time point in the TFR experiment, the amount of internalized
receptor should be expected to triple in an almost linear process. The
pepsin/HCl group demonstrated a consistent rate that confirmed expectation,
69
while the PBS and glycine/HCl groups did not (Figure 10B and 10C). The false
data points with these washing conditions also led to other inaccuracies. From
the 5-min to 1-hour time points in the beta1 integrin experiment, the glycine/HCl
group actually showed a negative internalization rate going from 20.2 to 17.0
pixels per cell area. This result would not be possible in a properly conducted
internalization experiment, but because of the false signal coming from the non-
internalized receptors, this type of result was obtained.
Other interesting observations can be made when examining the confocal
images generated from the beta1 integrin internalization experiment (Figure
10A). The PBS and glycine/HCl images clearly showed a pronounced
fluorescent signal (a fluorescent corona) coming from the plasma membrane of
the MDA-MB-435 cells at all time points. This residual signal, which was
generated by the Fc A488 secondary Ab that was not washed off the cell surface,
was very pronounced as compared to the pepsin/HCl treatment group. The
residual surface fluorescence problem in the previously conducted internalization
experiments was overcame by successfully cleaving off the Fc fragment of the Fc
A488 secondary Ab with pepsin. Another solution to this problem would be to
use computer software to only count the internal area of the cell, excluding the
fluorescent corona, rendering the stripping of the non-internalized signal
unnecessary. However, this would not be appropriate for several reasons. First,
artificially eliminating the residual corona would not be possible when conducting
studies such as flow cytometry. Second, one would not be able to distinguish an
internal from an external membrane, meaning that newly released vesicles would
70
not be counted. Third, a closer look at the 3-hour pepsin/HCl treatment images
(Figure 10A, Row 9) shows that in distinct pockets at the cell surface a faint
corona starts forming again, even after very efficient stripping of all fluorescence
on the cellular surface. This phenomenon could represent the beginning of
recycling vesicles carrying the beta1 integrin receptor still bound to the Ab ligand
back to the plasma membrane, but before the vesicle fused to the plasma
membrane. This new level of complexity should be investigated as a separate
event. The use of traditional buffers would make the recycling phenomenon
impossible to quantitate because these methods never fully eliminate the
fluorescence coming from the cell surface.
The recycling aspect of receptors demonstrates another possible
application for this technique – analysis of the rate of recycling of receptors after
internalization. A method to investigate recycling of receptors requires a reliable
procedure for removing the initial surface signal, similar to internalization studies
(Pagano, Crottet et al. 2004; Barroso-Gonzalez, Machado et al. 2009). One way
to utilize the pepsin/HCl buffer would be to bring the cells, after initial Ab
stripping, back to 37
o
C, a temperature that would allow the internalization
process to resume. After a predetermined time interval, in which recycling of the
receptor being studied was shown to occur, the surface of the cells would be
stripped again with the pepsin/HCl buffer. Then comparisons could be made,
using flow cytometry, between the signals generated from cells that were
stripped once to cells stripped a second time. The difference in the two
measurements should provide an indication of the amount of receptor recycled to
71
the cell surface. However, this would only be possible if you were able to
completely remove all residual fluorescence from the cell surface.
The novel pepsin Ab fragmentation technique was utilized to further
investigate the internalization rates of integrins after being ligated to CN (Figure
12). This experiment finally showed that the integrins themselves, beta 1
integrins in this case, are having their internalization rates drastically increased
after CN ligation. This could indicate one possible reason that disintegrins work
so well as integrin antagonists in anti-adhesive, anti-invasive, and anti-
angiogenic assays. Disintegrins not only bind to and inhibit the ability of cells to
use their integrins but they also remove a large portion of the integrins from the
cell surface. This experiment also provides further evidence that the western blot
and FACS experiments (Figure 8) were correct and the integrins are being
rapidly internalized, possibly bringing CN inside the cells with them. This finding
suggests that CN may cause intracellular changes as well as extracellular.
In summary, the method described herein can be effectively used to
completely remove residual cell surface fluorescent signal in receptor
internalization assays. By doing an even more detailed internalization time
course, the pepsin/HCl wash would be expected to generate a clearer, but more
complex image of receptor internalization. This would allow for better
understanding of the internalization rates for specific receptors. Now that
integrins are being linked to more and more cases of viral internalization events
(Wickham, Mathias et al. 1993; Chiu, Mathias et al. 1999; Maginnis, Forrest et al.
2006), its important to demonstrate how disintegrins influence their internalization
72
rate and discover new techniques that are more precise at measuring in verses
out cellular signals.

73
Chapter 4: Beta1 Integrin Inactivation and Disruption of Focal
Adhesions by the Disintegrin Contortrostatin
4.1 Summary
Integrins mediate signaling in a variety of cellular pathways and their
activation status plays an important role in their ability to interact with the
extracellular matrix. Disintegrins act as integrin antagonists with the ability to
disrupt adhesion based integrin functions through direct binding. Based on
evidence accumulated while investigating the action of CN, a homodimeric
disintegrin isolated from southern copperhead venom, on cellular invasion and
actin cytoskeleton disruption, it was hypothesized that CN acts on signaling
pathways that affect the activation state of integrins. In this study it is established
that talin, a key molecule in integrin activation, can be directly affected by CN
ligation of beta1 integrins. Through a series of co-immunoprecipitations and
western blotting it was determined that CN binding leads to the displacement of
talin from the cytoplasmic domain of beta1 integrins, a decrease in the cleavage
of talin into its subdomains, and an alteration of talin’s interaction with several
other proteins (Rap1, vinculin, and Rap1-interacting adaptor molecule). These
events combine to produce a global inactivation signal for beta1 integrins as
indicated by flow cytometry and a decrease in the ability of the cell to produce
talin based FAs as shown by confocal microscopy.

74
4.2 Introduction
Integrins are a family of αβ heterodimeric cell surface receptors that are key
to the ability of cells to ECM proteins. Integrins can switch between two different
affinity conformations, one with high ligand affinity (on state) and one with low
ligand affinity (off state). A unique feature of integrins compared to other
adhesion molecules is that their extracellular domains can be activated on a
timescale of <1s by signals within the cell (inside →out signaling) (Takagi and
Springer 2002). Both integrin clustering and activation are important functions for
a myriad of cellular processes including the immune response, migration,
proliferation, and survival. The binding of a cytoskeletal protein, called talin, to
the β subunit cytoplasmic tail, is a common final step in the integrin activation
process (Tadokoro, Shattil et al. 2003; Wegener, Partridge et al. 2007).
Talin (~270 kDa) is a high molecular weight protein containing a globular
N-terminal head (talin-H, ~50 kDa, residues 1–433) and a flexible C-terminal rod
domain (talin-R, ~220 kDa, residues 434–2541) (Rees, Ades et al. 1990;
Critchley 2009). Talin-H is composed of a FERM (four-point-one, ezrin, radixin,
moesin) domain, which in turn contains 3 subdomains: F1, F2, and F3 (Rees,
Ades et al. 1990; Garcia-Alvarez, de Pereda et al. 2003; Papagrigoriou, Gingras
et al. 2004). Talin-R contains 62 α helices arranged into multiple helical bundles,
and a dimerization site in the carboxy-terminal H62 which leads to the formation
of an antiparallel homodimer (Gingras, Ziegler et al. 2005; Gingras, Bate et al.
2008). Autoregulatory interactions between talin-H and talin-R domains establish
an inhibited intact protein with cryptic integrin β tail cytoplasmic binding domains.
75
Intact talin can be cleaved into two separate domains by calpain-II cleavage at
the linker region (residues 401–481) between the two domains (Beckerle,
Burridge et al. 1987; Rees, Ades et al. 1990; Azam, Andrabi et al. 2001).
Interestingly, when the talin-H and the talin-R domains are dissociated, their
binding affinity for β integrin tails is increased due to both the generation of free
talin-H and the cooperative binding of talin-R to the talin-H- β tail complex (Yan,
Calderwood et al. 2001). Specifically, talin-H has been shown to have a 6-fold
higher affinity for beta3 integrin tail when compared with intact talin. Talin-R also
contains up to 11 vinculin binding sites (Gingras, Ziegler et al. 2005) and, via
talin, vinculin connects the actomyosin cytoskeleton with β integrin subunits
(Mierke 2009). Although vinculin’s role as a key component of the
mechanosensor apparatus downstream of integrins is not clearly understood, it
has recently been shown, through the use of magnetic tweezers, that the
stretching of the talin-R domain results in the exposure of additional binding sites
for vinculin that are cryptic in the absence of force (del Rio, Perez-Jimenez et al.
2009).
Although an exact mechanism for how talin is activated or regulated has
yet to be established, it is well known that the small monomeric Ras GTPases
play a vital role in this process (Kinbara, Goldfinger et al. 2003). Like all
GTPases, the Ras proteins cycle between an active GTP-bound state and an
inactive GDP-bound state. The GTPase function is regulated by two other
protein families: the guanine exchange factors which, as their name implies,
exchange a molecule of GDP for GTP and, thus, inactivate the GTPases, and the
76
GTPase activating proteins which promote hydrolysis of GTP to GDP. Of the
members of the Ras family, Rap1 was shown to interact with integrins and to
have an instrumental role in integrin activation (Caron, Self et al. 2000; Bertoni,
Tadokoro et al. 2002; Bos 2005; Kooistra, Dube et al. 2007). In addition to Rap1,
one of its effector molecules, the Rap1-interacting adaptor molecule (RIAM), was
also linked to integrin activation based on a number of observations: RIAM was
shown to promote αL β2-integrin and α4 β1-integrin dependent adhesion of Jurkat
cells, and to activate αIIb β3 integrin when ectopically expressed in Chinese
hamster ovary cells (Lafuente, van Puijenbroek et al. 2004; Han, Lim et al. 2006).
RIAM interacts preferentially with active Rap1 and recently Rap1-GTP/RIAM/talin
complex has been shown to be important in the relocalization and recruitment of
talin to the plasma membrane, which could represent a key step in the regulation
of integrin activation (Lee, Lim et al. 2009).
Talin is a key protein of FA complexes, which are dynamic structures that
behave like network hubs for relaying the signals generated by integrin-ECM
interaction through the interior of the cell. FAs are composed of a number of
anchorage (adaptor) and enzymatic signaling proteins that form large biological
macromolecular assemblies. Although FAs are the most complex
supramolecular structures generated at the integrin-ECM interface, several less
exotic structures have also been designated based upon morphological and
molecular criteria including focal contacts, fibrillar adhesions, and podosomes
(Wu 2007). Without these dynamic structures the cell could not efficiently grip the
77
ECM and perform relatively simple tasks such as cell migration or more elaborate
tasks such as muscle contraction.
Disintegrins are a group of polypeptides isolated from the venom of
hematotoxic snakes. They are disulfide-rich polypeptides, of which many contain
an Arg-Gly-Asp (RGD) integrin recognition motif which enables them to bind with
high affinity to integrins (McLane, Joerger et al. 2008). CN, purified from the
venom of Agkistrodon contortrix contortrix (southern copperhead snake) is a
homodimeric disintegrin with a molecular weight of 13.5 kDa, and contains two
identical 6.75 kDa subunits (Trikha, De Clerck et al. 1994). Each subunit of CN
contains an RGD motif and 10 cysteines at conserved positions. CN was first
established as a potent inhibitor of αIIb β3 mediated platelet aggregation in 1994
(Trikha, Rote et al. 1994) in the Markland laboratory. In addition to αIIb β3, a
number of motogenic integrins relevant to tumor angiogenesis and metastasis
were also identified as binding sites for CN including integrins α5 β1, αvβ3, and
αvβ5 (Trikha, De Clerck et al. 1994; Zhou, Hu et al. 2000). CN binds to α5 β1,
αvβ3, and αvβ5 with low nanomolar Kds demonstrating a tight association of CN
with these integrin receptors (Minea, Helchowski et al. 2010). One of the
hallmarks of disintegrins is their ability to interact with and disrupt integrin
function. For instance, CN has been shown to significantly inhibit invasion of
HUVEC trough a Matrigel matrix (Golubkov, Hawes et al. 2003), disrupt the actin
cytoskeleton, inhibit HUVEC tube formation, and act as an anti-angiogenic agent
in an in vivo model of human breast cancer (Swenson, Costa et al. 2004). Along
with the antagonistic behavior of CN, other signaling aspects have been
78
identified. For instance, FAK, a major non-receptor tyrosine kinase component of
FAs, as well as one of its adaptor molecules, the Crk-associated substrate, are
both highly phosphorylated compared to controls when cancer cells are exposed
to CN while in suspension (Ritter et al., 2000). However, if the same cells were
attached to an ECM before being exposed to this disintegrin, the phosphorylation
state of FAK in these adherent cells changes in the opposite direction (i.e., it
goes down) as a result of integrin engagement by CN. This indicates integrin
mediation of an inward signal transduction process (outside →in signaling), at the
FA complex, that is taking place as a result of receptor ligation by CN.
In this chapter it is shown that CN has the ability to inactivate the
extracellular domain of β1 integrins through the displacement of talin bound to
the beta1 integrin cytoplasmic domain via a mechanism that involves an initial
integrin-mediated transduction of an extracellular signal (outside →in signaling)
followed by a propagation of this signal to the neighboring integrins from inside
the cell (inside →out signaling). It was demonstrated, by co-IP, that CN treatment
causes an increase in the talin-vinculin interaction (at low concentrations) and a
decrease in the talin-RIAM interaction. Through confocal microscopy, it was
shown that CN interferes with and disrupts the actin cytoskeleton and FA
complexes in HUVEC seeded on GFR Matrigel. In addition, an exposure of
these cells to CN for 18-hours causes a decrease in protein expression for both
talin-R and talin-H domains as well as a decrease in Rap1 activity in HUVEC.
These observations are very interesting because they couldn’t be reproduced
when a small cyclic RGD peptide was used in this setting as an integrin
79
antagonist instead of an intact disintegrin such as CN. The control cyclic RGD
peptide that was included in the experiments was the cyclo(Arg-Gly-Asp-DPhe-
Val) or cRGDfV peptide (Alghisi et al., 2009; Dechantsreiter et al., 1999). The
laboratory was very interested to see how this cyclic RGD peptide compared to
CN because cRGDfV is a precursor to Cilengitide, a small integrin antagonist
developed by Merck KGaA and currently in clinical trials against a variety of solid
tumors (Reardon et al., 2008). It was particularly important to determine whether
the effects induced by CN, which is a complex polypeptide that utilizes beside its
RGD motif a number of additional and spatially distinct residues for integrin
engagement, could also be observed with a much simpler ligand that only
displays an RGD motif. The cRGDfV peptide has been demonstrated, by
competitive binding assays, to bind to several integrin receptors including αIIb β3,
α5 β1, and αvβ3 (Dechantsreiter et al., 1999; Pfaff et al., 1994), all of which are
shared targets with CN. Unlike Cilengitide, however, which was designed to bind
with high affinity to αv integrins (Alghisi et al., 2009), cRGDfV is more
promiscuous demonstrating some binding affinity for most RGD dependent
integrins. Nonetheless, as a simple ligand built around the RGD motif, cRGDfV
binds to these integrins with a lower affinity compared to CN, mainly because it
lacks the spatial complexity that is characteristic of disintegrin loops. In this
chapter, the highlights and the differences between disintegrins and cyclic RGD
peptides in regards to their ability to disrupt beta1 integrin signaling and their
subsequent effects on talin are examined.

80
4.3 Materials and Methods
4.3.1 Cell Culture and Antibodies
MDA-MB-231 (ATTC, Manassas, VA) human cancer cells were grown in
DMEM containing 10% FBS, and HUVEC were grown in endothelial cell growth
medium (Promo Cell, Heidelberg, Germany) in tissue culture flasks. Cells were
serum starved overnight and then harvested by brief trypsinization with 10%
trypsin stock (0.05% trypsin-0.02% EDTA) in PBS for 5-min, followed by
quenching in 0.2% soybean trypsin inhibitor. Cells were resuspended in serum
free media and left in suspension at 37 °C for 1-hour before initiating experiments.
Abs used were anti-active beta1 integrin (clone HUTS-4), anti-alphaV beta5
integrin (clone P1F6), anti-alpha4 integrin (clone P1H4), and anti-alphaV integrin
(clone P3G8) from Millipore (Billerica, MA); anti-mouse FITC-conjugated
secondary from Jackson ImmunoResearch (West Grove, PA); anti-talin (clone
TA205 and clone H-300), anti-beta1 integrin (clone P5D2), anti-vinculin (clone G-
11), anti-actin (clone I-19), anti-mouse HRP, anti-rabbit HRP, anti-goat HRP and
A/G PLUS-Agarose from Santa Cruz Biotechnology (Santa Cruz, CA); anti-RIAM
(clone EP2806) from Epitomics (Burlingame, CA).

4.3.2 Beta1 Integrin Inactivation
MDA-MB-231 cells were washed in 150 mM NaCl, 25 mM Tris-Cl, pH 7.4,
containing 1 mg/ml BSA and then left in this solution for 10 min at 37 °C. Cells
(10
6
) were treated with divalent cations (2 mM Ca
2+
, 2 mM Mn
2+
), CN (1 nM, 10
81
nM, 100 nM), or cRGDfV (0.1 µM, 1 µM, 10 µM) for 30-min at 37 °C. Cells were
then labeled with anti-active beta1 integrin Ab (1:1000) for 1-hour at 4 °C. After 3
washes in PBS, cells were further stained with anti-mouse FITC-conjugated
secondary Ab (1:1000) for 1-hour at 4 °C. Following 3 more washes in PBS, cells
were subjected to FACS analysis.
To determine the relative amount of active or inactive beta1 integrins on
the surface of the MDA-MB-231 cells, an arbitrary signal intensity was chosen as
a reference point based on the FACS readouts generated from the 2 mM Ca
2+

(minimum intensity) and the 2 mM Mn
2+
(maximum intensity) treated cells. This
rule was applied to all conditions and the relative fluorescence shifts were taken
into account to determine the levels of beta1 integrin activity for each treatment
group.

4.3.3 Immunofluorescence
Eight well chamber slides were coated with GFR matrigel (BD
Biosciences, San Jose, CA) and left overnight at 37 °C. Residual media was
aspirated off the chamber slides and HUVECs (2.5 x 10
4
) were seeded in serum
free media onto the polymerized matrigel layer. Cells were incubated for 1-hour
at 37 °C before being treated with either 100 nM CN or 10 µM cRGDfV. After
overnight incubation, cells were washed with PBS, fixed in ice-cold acetone for
10-min at 4 °C, and allowed to air dry for 2-min at 20 °C. Content of each
chamber was then treated with PBS containing 5% BSA for 1-hour at 20 °C. Talin
82
was stained with a mouse anti-talin monoclonal Ab (1:50) in the presence of 5%
BSA for 2-hours at 20 °C. After three washes with PBS, each for 5-min, chamber
contents were counter stained with anti-mouse FITC-conjugated secondary Ab
(1:200). The nuclei were stained with Hoechst stain (1:500, Lab Vision
Products, Freemont, CA) and the actin cytoskeleton was stained with rhodamine
phalloidin (1:40 Invitrogen, Carlsbad, CA) for 20-min at 20 °C. Following three
more washes in PBS the chambers were separated and mounted with
fluorescent mounting media (KPL, Gaithersburg, MD) for confocal microscopy.
The confocal images were quantified using the software program Simple
PCI (Hamamatsu, Sewickley, PA). The background nuclei staining was digitally
subtracted, each representative image scanned, and every fluorescent pixel that
fell between a arbitrarily set minimum and maximum signal strength range was
counted. The same rule was applied to all images with all pixels falling above or
below the arbitrary signal strength range excluded. The same images were then
used to calculate the area of the measured cells. Using the above approach,
numbers for the amounts of pixels per unit area for each image were obtained.
These numbers represent the relative size of the FAs present in the treatment
groups. The same approach was used to dilate and fill in the gaps in between
pixels approximately 1 µm apart, which allowed us to digitally count the objects
and determine the relative number of FA sites for each treatment condition. This
number was divided by the total cellular area to obtain the number of FAs per
unit area.

83
4.3.4 Co-Immunoprecipitations and Western Blotting
For co-immunoprecipitations (co-IP), anti-beta1 integrin (clone P5D2, 10
µl) or anti-talin (clone TA205 or H-300, 10 µl) monoclonal antibodies (mAb) were
incubated with protein A/G PLUS-Agarose (20 µl) beads for 2-hours at 4 °C.
Following incubation, the Ab-bead conjugates were fixed in 3.7% formaldehyde
for 10-min at 20°C and washed three times in PBS. MDA-MB-231 cells (1 x 10
6
)
were either left untreated, treated with CN (1 nM, 10 nM, 100 nM), or treated with
cRGDfV (0.1 µM, 1 µM, 10 µM) for 30-min at 37 °C. Cells were then washed in
PBS and lysed in cold lysis buffer (50 mM tris, pH 8.0, 150 mM NaCl, 1%
nonionic detergent Igepal CA-630, 0.5% deoxycholate, 0.1% SDS, protease
inhibitor cocktail, 1 mM sodium pyrophosphate, 1 mM sodium orthovanadate, 50
mM sodium fluoride) for 10-15-min at 4 °C. Lysates were passed through a 23-
gauge needle, to ensure membrane aggregates were dissociated, and
centrifuged at 10,000 RPM. Supernatants were separated from insoluble
material and incubated with the Ab-bead conjugates overnight at 4 °C. After
incubation, immunoprecipitates were washed three times in lysis buffer, without
inhibitors, and dissociated by adding SDS-PAGE sample buffer and boiling for 5-
min.
For protein expression and Rap1 activity, HUVEC (1.5 x 10
5
) were plated
in 4 well chamber slides that were previously coated with GFR Matrigel that was
allowed to polymerize into a matrix overnight at 37 °C. Cells were incubated for
1-hour at 37 °C and either left untreated, treated with CN (1 nM, 10 nM, 100 nM),
84
or treated with cRGDfV (0.1 µM, 1 µM, 10 µM). After an incubation period of 18-
hours, cells were lysed in 100 µl of lysis buffer as previously described. Whole
cell lysates (30 µg total protein) or immunoprecipitates (30 µl) were resolved on
either 12% or 4-20% polyacrylamide LongLife gels (NuSep, Lawrenceville, GA).
Proteins were transferred to nitrocellulose membranes overnight at 4 °C and
blocked in TTBS containing 2% BSA for 1-hour at 20 °C. Membranes were
probed with anti-talin (clone TA205, 1:1000), anti-vinculin (clone G-11, 1:500), or
anti-RIAM (clone EP2806, 1:1000) mAb for 2-hours at 20 °C or overnight at 4 °C.
Membranes were washed five times in TTBS before being probed with anti-
mouse HRP, anti-rabbit HRP, or anti-goat HRP (1:5000) for 2-hours at 20 °C.
Following five additional washes with TTBS the membranes were developed via
chemiluminescence with a SuperSignal kit (34080, Thermo Scientific, Rockford,
IL). Each membrane was exposed for 30-sec, 2-min, and 10-min. The exposure
with the best signal was subjected to densitometry. All protein levels were
standardized by BCA assay for equal protein loading. Due to the need for non-
reducing conditions with clone P5D2, loading controls were established using
polymerized actin levels for each lysate.

4.3.5 Active Rap1 Pull Down
Assays were performed with active Rap1 pull-down and detection kit
(89872, Thermo Scientific) according to the manufactures instructions.

85
4.3.6 Statistical Analysis
For each of the experiments above a student T-test was performed for
statistical analysis, generating a p-value. Each experiment was performed three
separate times and standard deviation error bars were included on the graphs to
illustrate the data ranges between the multiple experiments.

4.4 Results
4.4.1 Beta1 Integrin Inactivation by Contortrostatin
After making the previous observation that the treatment of endothelial
cells in the presence of low nanomolar concentrations of CN significantly
inhibited their invasion through a reconstituted basement membrane (Zhou et al.,
2000a; Zhou et al., 2000b), while small cyclic-RGD peptides required micromolar
concentrations to induce a similar inhibitory effect (Nisato et al., 2003), it was
hypothesized here that the ligation of integrins by CN may lead to a global
inhibitory effect due to the propagation of an inhibitory signal from the receptors
engaged by CN to the adjacent unligated ones. One important characteristic of
integrins is that they can be switched to an inactive state from inside. Therefore,
one way to explain the potent inhibitory effects of disintegrins on cell invasion at
low nanomolar concentrations and the discrepancy in effects between
disintegrins and cyclic RGD peptides was to understand whether the ligation of
integrins by CN leads to intracellular changes that in turn affect the state of
activation of unligated integrins via a mechanism involving inside →out signaling.
86
The activation status of integrins can be easily manipulated by divalent
cations (Mould et al., 2002) and this technique was utilized to generate a FACS
readout for a cellular state where the majority of beta1 integrins on the surface of
a cancer cell line kept in suspension were either active (Mn
2+
treatment) or
inactive (Ca
2+
treatment). Beta1 integrins were chosen as an appropriate target
for our integrin activation studies because these receptors were the only class of
integrin to meet the following experimental constrains: (i) the availability of a
monoclonal antibody that only recognizes the active conformation of beta1
integrins, (ii) a shared target for both CN and cRGDfV, and (iii) the availability of
a cell line (i.e., the human breast carcinoma MDA-MB-231) that expresses large
amounts of beta1 integrin. In order to probe the beta1 integrin activation status
on the surface of human breast carcinoma MDA-MB-231 line, an Ab (HUTS-4),
was used that maps to a cryptic beta1 integrin epitope that is only exposed when
the receptor is switched to the active conformation state. Furthermore, it was
confirmed by FACS that the MDA-MB-231 line used in our experiments did
indeed express a large amount of beta1 integrins (see supplementary data). As
mentioned above, treatment with divalent cations was used to generate the
FACS readout for active and inactive beta1 integrin state in these cells.
Accordingly, after treatment with 2 mM Mn
2+
,

the active control showed that
98.2% of the beta1 integrins were active (Figure 13C), while the inactive control,
following treatment with 2 mM Ca
2+
, showed that only 1.2% of these receptors
were in the active conformation (Figure 13B). Since the tested cells were cancer
cells and, as a general rule, cancer cells express high amounts of dysregulated
Figure 13

Inactivation of Beta1 Integrins by CN. MDA-MB-231 cells were incubated in 150mM
NaCl, 25mM Tris-Cl, pH 7.4, containing 1 mg/ml BSA for 10-min at 37°C. Cells (10
6
) were
treated with divalent cations, CN, or cRGDfV for 30-min at 37°C. Cells were then labeled with
anti-active beta1 integrin Ab (clone HUTS-4) for 1-hour at 4°C. Cells were counter stained with
an anti-mouse FITC-conjugated secondary Ab for 1-hour at 4°C and subjected to FACS analysis.
Panel A: Untreated; panel B: Ca
2+
; panel C: Mn
2+
; panel D: 0.1 µM cRGDfV; panel E: 1 µM
cRGDfV; panel F: 10 µM cRGDfV; panel G: 1 nM CN; panel H: 10 nM CN; panel I: 100 nM CN;
panel J: bar graph of beta1 integrin activation state for all treatment groups.
87
88
integrins in an active state, it was not unexpected that the untreated control
showed an active beta1 integrin level of 85.7% (Figure 13A). For the cRGDfV
treatment, the percentage of active beta1 integrins detected by FACS dropped
with concentration titration from 60.7% at 0.1 µM (Figure 13D), to 57.5% at 1.0
µM (Figure 13E), and to 47.6% for 10 µM (Figure 13F). In contrast, for the CN
treatment it was noticed that the highest percentage of receptor inactivation
happened at the lowest disintegrin concentration (1 nM) and a steady increase in
receptor activation was recorded as the concentrations of CN was increased,
from 54.2% at 1 nM (Figure 13G), to 62.1% at 10 nM (Figure 13H), and to
80.8% at 100 nM (Figure 13I). Thus, the receptor activation trends observed
with CN and cRGDfV treatments evolved in opposite directions as their
concentrations increased. Because the HUTS-4 mAb only binds to the active
conformation of beta1 integrins, the logical explanation for our results was that
the binding of either cRGDfV or CN interfered with the ability of this antibody to
recognize its epitope. This appears to be true with the cRGDfV peptide, in which
case the subsequent Ab binding (indicative of activation level) was inversely
proportional to the concentration used for this peptide. However, the above
rationale cannot explain the results obtained with the CN treatment, because in
this case the Ab binding increases as the CN concentrations rises (Figure 13J).
Most interestingly, the biggest drop in activation occurred when cells were
exposed to the lowest concentration of CN tested (1 nM) and, by comparison, to
obtain a similar effect with a cyclic RGD peptide the highest of the tested
concentrations was required (10 µM).
89

4.4.2 Focal Adhesion Protein Associations are Disrupted by
Contortrostatin

To further investigate this phenomenon, co-IP was utilized to study how
the interactions of several key proteins involved in beta1 integrin activation were
affected in MDA-MB-231 cells exposed to either CN or cRGDfV. Since binding of
talin to the cytoplasmic domain of integrin beta subunits is the final step in
integrin activation, this particular association was the logical choice to examine
first. Therefore, the beta1 integrin receptor was immunoprecipitated and further
probed it for association with talin. Interestingly, after the cRGDfV treatment, no
significant change in the binding of talin-H to beta1 integrin tails in these cells
was observed, while the binding of talin-R decreased slightly (63% of untreated
control at 10 µM). By comparison, CN treatment produced a considerable drop
in the binding of both talin-H (8.5% of untreated control at 100 nM) and talin-R
(10.4% of untreated at 100 nM) (Figure 14A & B). It is important to mention,
however, that the intact talin protein failed to be immunoprecipitated by this
technique. These findings suggest that CN might inactivate the beta1 integrin
receptors through the displacement of bound talin from their cytoplasmic tails.
Moreover, the association between talin and two of its binding partners, vinculin
and RIAM, in cells exposed to either CN or cRGDfV were probed for. The level
of vinculin, which links talin to the actin cytoskeleton, increases with low
concentrations of both cRGDfV and CN (182% of untreated control at 0.1 µM
cRGDfV and 171% of control at 1 nM CN). At higher treatment concentrations,

Figure 14

Co-immunoprecipitation of Talin. Anti-beta1 integrin (clone P5D2) was incubated with
protein A/G PLUS-Agarose beads for 2-hours at 4°C. Following incubation, the Ab-bead
conjugates were fixed in 3.7% formaldehyde for 10-min at 20°C and washed three times in PBS.
MDA-MB-231 cells (1 x 10
6
) were either left untreated, or treated with CN (1 nM, 10 nM, 100 nM),
or cRGDfV (0.1 µM, 1 µM, 10 µM) for 30-min at 37°C. Cells were then washed in PBS and lysed
in cold lysis buffer for 10-15-min at 4°C. Supernatants were separated from insoluble material
and incubated with the Ab-bead conjugates overnight at 4°C. Immunoprecipitates were washed
three times in lysis buffer and dissociated by adding SDS-PAGE sample buffer and boiling for 5-
min. Whole cell lysates (30 µg total protein) were resolved on 12% polyacrylamide gels. Proteins
were transferred to nitrocellulose membranes overnight at 4°C and developed via
chemiluminescence. All protein levels were standardized by BCA assay and actin was used as a
loading control. Panel A: Western blot image showing IP of beta1 integrin with immunoblot for
talin (clone TA205). Panel B: Densitometry plot for panel A.

90
91
however, there were non-significant changes for cRGDfV, while a decrease in
binding was observed (35.6% of untreated control) at 100 nM CN (Figure 15A &
C). It was also noticed that the RIAM-talin interaction, which is involved in the
recruitment of talin to the plasma membrane, was significantly decreased at 1 nM
and 100 nM CN (14-21% of untreated control), while cRGDfV showed only a
modest effect at 0.1 µM (Figure 15B & D).

4.4.3 Disruption of Talin Based Focal Adhesions by
Contortrostatin

To further investigate the effects of these treatments on talin and FA
formation, it was decided to directly visualize talin using confocal microscopy.
Even though the staining for some of these FA proteins (e.g. paxillin, talin) tend
to generate a high background around the nucleus with the available monoclonal
antibodies (Himmel et al., 2009; Shan et al., 2009), it was possible to obtain an
acceptable pattern of discrete FA staining with the talin antibody used in our
experiments. The cells that were examined were migratory cells that were
actively involved in establishing FAs before being exposed to various treatments
with either cRGDfV or CN. For these studies, rather than MDA-MB-231, HUVEC
were utilized because these endothelial cells possess a higher ability than the
MDA-MB-231 cells to assemble FA complexes while they migrate, invade, and
assemble into vessels when plated on Matrigel (Enis et al., 2005; Samarzija et
al., 2009) and, thus, these complexes formed by HUVEC were easier to visualize
by confocal microscopy. Accordingly, the Fas of untreated HUVEC can be
Figure 15

Co-immunoprecipitation of Key Focal Adhesion Proteins. Anti-talin (clone
TA205 or clone H-300) was incubated with protein A/G PLUS-Agarose beads for 2-hours at 4°C.
Following incubation, the Ab-bead conjugates were fixed in 3.7% formaldehyde for 10-min at
20°C and washed three times in PBS. MDA-MB-231 cells (1 x 10
6
) were either left untreated, or
treated with CN (1 nM, 10 nM, 100 nM), or cRGDfV (0.1 µM, 1 µM, 10 µM) for 30-min at 37°C.
Cells were then washed in PBS and lysed in cold lysis buffer for 10-15-min at 4°C. Supernatants
were separated from insoluble material and incubated with the Ab-bead conjugates overnight at
4°C. Immunoprecipitates were washed three times in lysis buffer and dissociated by adding
SDS-PAGE sample buffer and boiling for 5-min. Whole cell lysates (30 µg total protein) were
resolved on 12% polyacrylamide gels. Proteins were transferred to nitrocellulose membranes
overnight at 4°C and developed via chemiluminescence. All protein levels were standardized by
BCA assay and actin was used as a loading control. Panel A: Western blot image showing IP of
talin (clone TA205) with immunoblot for vinculin; Panel B: Western blot image showing IP of
talin (clone H-300) with immunoblot for RIAM; Panel C: Densitometry plot for panel A; Panel D:
Densitometry plot for panel B.

92
93
visualized as well-defined, cylindrical structures (Figure 16, Row A), while the
smaller, but numerous focal contacts can be seen at the leading edge of their
lamellipodia. Furthermore, the actin cytoskeleton of these untreated cells is
highly organized in a complex network and under little stress. In every aspect,
from the number of focal contacts made by talin to the organization of the actin
cytoskeleton, the staining pattern in cells treated with 10 µM cRGDfV looks
almost identical to the untreated group (Figure 16, Row B). On the other hand
and in stark contrast, the CN treatment reveals an ill-defined pattern for talin
staining with broader structures that cannot be easily classified as FAs (Figure
16, Row C). It was also noticed that in cells exposed to CN the number of focal
contacts at the leading edges of lamellipodia was greatly reduced. Moreover, the
actin cytoskeleton for the CN treatment group looks highly disorganized as if the
cells were under extreme stress conditions. The quantitation of the confocal
images revealed that the average pixels/cell area (which represents the average
size of an individual FA) was 10.7 for untreated control, 9.7 for 10 µM cRGDfV
treatment, and 4.1 for 100 nM CN treatment (Figure 16D). In addition, the
average FA/cell area was 0.0033 for untreated control, 0.0034 for 10 µM cRGDfV
treatment, and 0.002 for 100 nM CN treatment (Figure 16E). This data indicates
that cRGDfV treatment has virtually no effect on cytoskeletal organization and
FAs based on their talin associations, while CN treatment dramatically disrupts
both the cytoskeleton and talin-based FAs, a finding that provides further
evidence for the loss of beta1-talin association induced by this type of treatment.
Figure 16

Immunofluorescence of Actin Cytoskeleton and Focal Adhesions. Chamber
slides coated with GFR Matrigel were seeded with HUVEC cells (2.5x10
4
) in serum free media.
Cells were incubated for 1-hour at 37°C either with no treatment (Row A), treatment with 10 µM
cRGDfV (Row B), or 100 nM CN (Row C). After overnight incubation, cells were washed with
PBS, fixed in acetone for 10-min at 4°C, and allowed to air dry for 2-min at 20°C. Each chamber
was then blocked with PBS containing 5% BSA for 1-hour at 20°C. Talin was stained with a
mouse monoclonal Ab (clone TA205) for 2-hours at 20°C. Followed by counter staining with anti-
mouse FITC-conjugated secondary Ab (488 nM), hoechst stain (460 nM), and rhodamine
phalloidin (580 nM). After three washes in PBS the chambers were separated and mounted with
fluorescent mounting media for confocal microscopy (63x). Composite column: superimposed
images shown after counter-staining by all three agents; actin column: rhodamine phalloidin
staining; talin column: talin staining; zoom talin column: close up images of talin staining;
yellow arrows: talin-based focal adhesions; white arrows: talin-based focal contacts; yellow
line: 10 µm. Panel D: Pixel per cell area quantitation. Panel E: Focal adhesions per cell area
quantitation. Panel D: Pixel per cell area quantitation. Panel E: Focal adhesions per cell area
quantitation.

94
Figure 16: Continued

95
96
4.4.4 Inhibition of Talin Protein Expression and Rap1 Activity by
Contortrostatin

These observations on talin and the FAs it associates with led us to
hypothesize that the overall activity and expression of talin was being inhibited by
prolonged treatment with CN in HUVEC. To examine the relative levels of talin
protein expression, Western blotting analysis on whole cell lysates from HUVEC
exposed to either CN or cRGDfV treatments for 18-hours was examined(Figure
17A). Unlike what was seen in the previous pull down experiments, this time the
total cell lysate blots produced a distinct band for the intact talin protein. For all
cRGDfV concentrations tested, the levels of protein expression for talin, talin-R,
and talin-H were constant showing virtually non-significant changes (Figure
17B). However, the CN treatment resulted in a significant decrease in the
expression of both talin-R (15% of untreated control at 100 nM) and talin-H (1.4%
of untreated control at 100 nM) subdomains, but only a modest decrease in intact
talin expression (44.4% of untreated control at 100 nM). This differential in the
relative levels of talin subdomains compared to the whole protein in CN-treated
cells suggests a possible influence of this disintegrin, downstream of integrin
ligation, on the mechanism through which talin is cleaved physiologically. In this
context, relative activity level of talin by focusing on Rap1, a key small GTPase
that links integrin activation and talin function was probed for. Therefore, a GST-
fusion protein of the Rap1-binding domain from human RalGDS (a downstream
effector of Rap1) was used to pull-down GTP-bound Rap1 (Figure 18A). Once
again, the cRGDfV treatment resulted in non-significant changes relative to the
Figure 17

Effect of Prolonged Exposure of CN on Expression of Talin. HUVEC cells
(1.5x10
5
) were plated on 4 well chamber slides that had been previously coated with GFR
Matrigel and allowed to polymerize overnight at 37°C. Cells were then incubated for 1-hour at
37°C before being treated with CN (1 nM, 10 nM, 100 nM) or cRGDfV (0.1 µM, 1 µM, 10 µM).
After an incubation period of 18 hours, cells were lysed in 100 µl of lysis buffer. Whole cell
lysates (30 µg total protein) were resolved on 4-20% polyacrylamide gels. Proteins were
transferred to nitrocellulose membranes overnight at 4°C, blocked, and incubated with anti-talin
mAb (clone TA205). Membranes were then developed via chemiluminescence. All protein levels
were standardized by BCA assay and actin was used as loading control. Panel A: Western blot
images for talin, talin-R domain, and talin-H domain protein levels. Panel B: Densitometry plot for
talin levels.

97
Figure 18

Effect of Prolonged Exposure of CN on Activity of Rap1 in HUVECs. For
Rap1 activity, Thermo Scientific’s protocol for active Rap1 pull-down and detection was followed.
Untreated lysates were spiked with GTP γS (0.1mM) or GDP (0.1mM) to establish controls where
Rap1 was completely active or in an inactive state. All protein levels were standardized by BCA
assay and actin was used as loading control. Panel A: Western blot image of active Rap1 pull-
down; Panel B: Densitometry plot for active Rap1 pull-down.

98
99
amount of GTP-bound Rap1 blotted at different cRGDfV concentrations (Figure
18B). This indicates that the overall Rap1 activity level in HUVEC plated on a
GFR Matrigel matrix remains fairly constant regardless of the amount of cRGDfV
peptide these cells are exposed to. However, a significant drop in GTP-bound
Rap1 was observed in the presence of 100 nM CN (10.3% of the untreated
control). These findings suggest that CN treatment leads to a decrease in the
average activity level of Rap1 GTPase, a signaling event that may contribute to
the observed displacement of talin from the beta1 integrin cytoplasmic tail.
These results were obtained following a prolonged exposure of HUVEC to CN
(18-hours). A shorter incubation time (less than 2-hours) produced no significant
changes in the levels of these proteins (data not shown).

4.5 Discussion
Both the activation status of integrins and the ability of the cell to form FAs
play important roles in many biological processes. Disintegrins are known to be
integrin antagonists and can influence many integrin related cellular processes.
However, little is known about how disintegrin binding to integrins influences
integrin mediated cellular processes. In this chapter, it was shown that CN, a
homodimeric disintegrin isolated from southern copperhead venom, has the
ability to inactivate beta1 integrins and disrupt talin-containing FAs.
By using a mAb tool that only binds to active beta1 integrin subunits and
FACS analysis it was shown here that both CN and a cyclic RGD peptide (the
cRGDfV) lead to beta1 integrin inactivation, with a dramatic change induced by
100
CN in the low nanomolar range. Interestingly, as CN concentration increases the
percentage of beta1 integrins in the active conformation follows a counterintuitive
trend (i.e., it goes up), whereas as the cRGDfV concentration is titrated up the
percentage of active beta1 integrins follows a downward trend. A couple of
explanations come to mind when attempting to explain the above findings. In the
case of cRGDfV, this small peptide receptor antagonist might either directly
inhibit the HUTS-4 mAb binding through epitope masking (steric hindrance) or,
once bound, the ligand might force the receptor into its inactive conformation.
However, the former possibility (i.e., a steric hindrance induced by the binding of
a small cyclic RGD peptide) is unlikely to be true because this effect was
definitely not observed with CN, a much larger molecule. In an attempt to explain
this rather intriguing integrin inactivation/activation observation with CN it was
hypothesized that: (i) once bound to integrins, CN forces the receptor to remain
in the active conformation, and (ii) a signal is propagated downstream of the CN-
ligated integrin (outside →in signaling) to the neighboring integrins which
inactivates the adjacent unligated receptors (inside →out signaling).
To further investigate this hypothesis a co-IP experiment was conducted
that examined the binding of talin to the cytoplasmic domain of beta1 integrins in
cells treated with either cRGDfV or CN, which is an event that represents the
common final step in integrin activation (Tadokoro et al., 2003; Wegener et al.,
2007). The data reveals that the exposure of MDA-MB-231 cells to the cRGDfV
peptide caused essentially no change in talin-H binding and only a slight
decrease in talin-R binding. By comparison, CN induced a significant decrease
101
in binding of both talin-H and talin-R to the beta1 cytoplasmic tail as the CN
concentration was increased. Interestingly, it was possible to immunoprecipitate
both the talin-H and talin-R domains, but not the intact talin protein. This could
be due to the fact that the beta integrin tails have a higher affinity for each of the
talin subdomains than for the intact protein and, therefore the intact protein was
lost in this pull down experiment. When combined with the data generated by
FACS analysis, the talin data seems to suggest that CN may have the ability to
inactivate beta1 integrins through a mechanism involving the displacement of
talin from the cytoplasmic domain which may be the result of an inside →out
signaling event.
After these initial observations CN’s effects on talin’s binding partners
through co-IP was further investigated. Talin was immunoprecipitated and
probed for two of its critical binding partners, vinculin and RIAM. Vinculin binding
increased following treatment by micromolar concentrations of cRGDfV or low
nanomolar concentrations of CN. Since the physical stretching of the talin-R
domain has revealed additional cryptic binding sites for vinculin (del Rio et al.,
2009), our data could suggest that while surface integrins are engaged by a
disintegrin, a mechanical torque may be exerted on the talin-R domain, revealing
additional binding sites to vinculin. It is possible that this is the result of integrin
receptors attempting to cluster around the ligated integrin, possibly stretching the
talin-R domain as FAs close the gap. This event would be less dramatic as more
and more integrins become bound to disintegrin because there would be less
need for integrin relocation in the attempt to cluster around a ligated integrin.
102
Talin’s recruitment to the plasma membrane was then investigated, a process in
which RIAM binding has been shown to be an integral component (Lee et al.,
2009). Although there was not a dose dependent trend, a consistent decrease in
the RIAM-talin association for all concentrations of CN that were significantly
greater than the observed results following cRGDfV treatment was observed. It
is speculated that as talin is displaced from the cytoplasmic tail, and the integrin
is inactivated, there is less need for recruitment of talin to the plasma membrane,
leading to a decrease of talin interaction with RIAM.
We further attempted to directly visualize the mature FAs (of which talin is
an important component) in HUVEC plated on a polymerized GFR matrigel layer
and exposed to CN or cRGDfV treatment. The results indicate that cRGDfV has
a negligible effect on the number and intensity of talin-based FAs as well as the
cytoskeletal structure in HUVEC. In contrast, CN induced a dramatic decrease in
both the number of FAs and their intensity. The FAs seen in both the untreated
and cRGDfV groups were very similar, as small, cylindrical, and relatively uniform
in size and shape. In comparison, the structures seen in the CN treatment group
were large, broad, and dissimilar. As previously reported for CN (Schmitmeier et
al., 2005), its ability to disrupt and disorganize the actin cytoskeleton was also
visualized . It is possible that CN ligation induces the cell to produce less
extensive focal contacts resulting in the formation of disorganized FAs that are
unable to make all protein associations characteristic of mature FAs, which may
explain the observed staining pattern with fluorescently-labeled phalloidin. These
effects induced by CN in HUVEC plated on Matrigel seem to occur despite the
103
fact that this artificial matrix has a basement membrane composition (laminins
and collagen IV) and is rich in peptide motifs that bind integrins in a non-RGD-
dependent manner, which, are not competing with CN for the same integrin
binding sites. This apparent conundrum, however, further supports our
observations that CN, through active signaling via integrins, may deactivate
adjacent integrins that are unoccupied by disintegrin ligands.
Once bound, due to its strong affinity for integrins, CN probably
dissociates from these receptors at a very slow rate while sterically trapping them
in the active conformation. Taken together, these findings may explain how at
low CN concentrations an initial global inactivation signal is generated and
propagated inside the cells, and why an increase in HUTS-4 Ab binding is
observed by FACS at increasing concentrations of CN. The theoretical ability of
the dimeric disintegrin CN to crosslink two integrins together, via its two RGD
adhesive loops, which are positioned 69 Å apart (Moiseeva et al., 2008), could
also play a distinct role in the observed phenomenon. In the future the laboratory
plans to investigate this phenomenon further using other disintegrins, including a
recombinant version of CN, called VCN. Because VCN possesses a higher
affinity for integrin α5 β1 (Minea et al., 2010) and because VCN is a monomer,
the laboratory should be able to determine if the quality of signal is any different
with VCN and if the dimeric structure of CN is at all responsible for influencing
the activation status of beta1 integrins.
In the future, the laboratory also intends to examine the effect of
disintegrins on other focal adhesion proteins (e.g., FAK, Src, paxillin) in a similar
104
manner and in greater detail. It is also important to note that due to the dramatic
effect shown here on cell morphology and the cytoskeleton, the possibility of
disintegrins inducing cell death/apoptosis in HUVEC grown on Matrigel was
investigated. Interestingly, the results indicate that disintegrins have no effect on
HUVEC viability grown on Matrigel despite the observed morphological changes,
but do have modest pro-apoptotic effects when the HUVEC are sandwiched
between two layers of Matrigel (Minea et al., 2010).
Finally, the effect CN had on global talin levels after prolonged exposure
(18-hours of incubation) of cells to the disintegrin was also examined. It was
shown that cRGDfV has little effect on the amounts of talin protein after
prolonged exposure, while CN causes a significant decrease in protein levels of
both talin-R and talin-H domains. Interestingly, while CN caused a significant
decrease in the levels of talin’s subdomains, there was only a 55% decrease in
the amount of the intact talin protein. It is proposed that CN may be inhibiting the
calpain-II cleavage of talin into its head and rod domains resulting in talin being
left in its autoinhibited conformation. Given that liberated talin-H and talin-R
domains have a higher binding affinity than intact talin to the cytoplasmic
domains of integrins (Yan et al., 2001), the greater percentage of intact talin in
HUVEC treated with CN results in less talin-integrin complexes and hence less
formation of talin-based FAs. This model is further supported by the confocal
microscopy images of HUVEC stained for talin. Prolonged exposure to CN (100
nM CN for 18-hours) also caused a significant decrease in GTP-bound Rap1
while treatment by cRGDfV had virtually no effect. Fluctuations in Rap1 activity
105
has been directly linked to cells being able to proceed through mitosis (Dao et al.,
2009). When the visual data of the effect of CN treatment on HUVEC actin
cytoskeleton and the Rap1 activity data are combinded, it is believed that CN
treatment may result in cells being sequestered in a segment of the cell cycle,
inhibiting their progression through mitosis. Therefore, the laboratory plans to
investigate this possibility in the future by determining the cell-cycle status of CN
treated cells by FACS analysis as well as by examining levels of cyclins, through
RNA expression, using real time-PCR.
To conclude, a mechanism by which CN can impact cellular processes in
a manner that requires more than a simple antagonistic ligation of integrins is
proposed. Most of these actions cannot be mimicked by a small cyclic-RGD
peptide indicating that for these activities the full structure of CN, in addition to
the RGD integrin-binding motif, is required. Taking into account the observations
presented here and corroborating them with what was previously shown
(Golubkov et al., 2003; Ritter et al., 2000; Swenson et al., 2004; Trikha et al.,
1994b), it seems that the exposure of motile cells to CN forces these cells into a
state of inactivity (or paralysis). The connection appears to involve the
dissociation of talin from the beta1 integrin cytoplasmic domain and is linked to
the subsequent disruption of the actin cytoskeleton. As a result, these motile
cells are unable to complete simple processes such as the formation of FAs and
migration, as well as more complex activities such as invasion through a
basement membrane and endothelial tube formation. Due to the global cellular
effects they seem to elicit, CN and other disintegrins, have the potential to
106
become a novel class of anti-invasive cancer therapeutics. Furthermore, this
global integrin inactivation phenomenon they induce at very low concentrations
opens up the possibility for disintegrins to function as efficient blockers of viral
uptake, which is another integrin-mediated process (Nemerow, 2009; Ylipaasto
et al., 2009).

107
Chapter 5: Characterization of Vicrostatin
5.1 Summary
Native disintegrins are known to bind with high affinity to several mammalian
integrin receptors, among them αvβ3, αvβ5, and α5 β1. From the therapeutic
standpoint it would be very advantageous to battle pathological processes like
angiogenesis and metastasis through multiple integrin pathways stemming from
disintegrin ligation to these integrins. The native CN, for instance, a polypeptide
originally isolated from the venom of Agkistrodon contortrix contortrix (southern
copperhead snake) and thoroughly studied in the Markland laboratory, showed
potent anti-angiogenic and anti-metastatic activities in several animal models.
Despite their proven therapeutic potential, disintegrins have one major drawback.
Disintegrin isolation and purification from crude venom is laborious and
prohibitively expensive for translation into the clinic, meaning it is imperative that
a recombinant version be produced. VCN is a chimeric recombinant disintegrin
generated in an E. coli system as a genetic fusion between the C-terminal tail of
a viperid snake venom disintegrin, echistatin, and the crotalid disintegrin CN. In
this study, VCN is characterized as an anti-angiogenic, anti-invasive, and
signaling molecule through in vitro assays, which help to confirm that VCN
retains all of the activities that the native molecule possesses. VCN exerts a
potent inhibitory effect on HUVEC migration and tube formation in a dose-
dependent manner, by forcing these cells to undergo significant actin
cytoskeleton reorganization when exposed to this agent in vitro. Moreover, as a
direct effect on cancer cells, VCN blocks the motility of MDA-MB-435 cells when
108
they are allowed to invade through a reconstituted basement membrane barrier.
As previously shown with native CN, VCN appears to engage its integrin targets
in an agonistic manner since it differentially alters the global tyrosine
phosphorylation status of FAK when suspended MDA-MB-435 cells encounter
this signaling molecule.

5.2 Introduction
Many human cancers possess the ability to evade various therapeutic
agents through the exploitation of several survival pathways. This redundancy
paves the way for development of drugs that attack cancer through multiple
pathways. Some of the most important characteristics of cancer include
angiogenesis, growth, differentiation, invasion and metastasis. In fact the ability
of transformed cells to evade the microenvironment control and disturb the
normal tissue architecture by proliferating in an anchorage-independent fashion
is one of cancer’s hallmarks (Hanahan and Weinberg 2000; De Wever and
Mareel 2003). How a particular transformed epithelial cell acquires an
anchorage-independent phenotype is an interesting process that revolves around
key cell surface receptors, the integrins (Hood and Cheresh 2002).
In non-cancerous tissue, integrins are a group of heterodimeric cellular
receptors that mediate cell-ECM interactions; they have a critical role in
maintaining the normal tissue architecture by keeping epithelial cells in a
differentiated state. Conversely, transformed cells evade the microenvironment
pressure by acquiring the ability to 1) alter from inside (inside →out) their integrin
109
receptor affinity and avidity for separate ECM proteins, and 2) become tolerant to
specialized cues coming from the microenvironment, a process that will
ultimately favor the overexpression of those integrins that are preferentially
utilized in the process of tumor progression. Because of the roles they play in
sustaining tumor progression, being involved in processes ranging from
angiogenesis to metastasis, integrins emerged as important therapeutic targets
in cancer leading to the development of new agents designed to disrupt integrin
binding/signaling. Among these agents, snake venom disintegrins hold a
significant potential based on their anti-angiogenic and anti-metastatic effects
demonstrated in various experimental settings (Kang, Lee et al. 1999; Zhou,
Sherwin et al. 2000; Huang, Yeh et al. 2001; Markland, Shieh et al. 2001;
Marcinkiewicz, Weinreb et al. 2003; Swenson, Costa et al. 2004).
Disintegrins are a class of cysteine-rich polypeptides isolated from the
venom of Viperinae and Crotalinae families of snakes that are powerful,
antagonistic, and soluble ligands for integrins (Gould, Polokoff et al. 1990;
Niewiarowski, McLane et al. 1994). Integrins have many important structural
characteristics, but none is more important to integrin function than the tri-peptide
RGD motif, which is conserved in many disintegrins (Gould, Polokoff et al. 1990;
Niewiarowski, McLane et al. 1994). The natural target of snake venom
disintegrins is the platelet fibrinogen receptor αIIb β3, and antagonistic binding
results in the inhibition of fibrinogen-dependent PA (Savage, Marzec et al. 1990).
Except for barbourin, a KGD-containing disintegrin which is a relatively specific
ligand for αIIb β3 (Scarborough, Rose et al. 1991), most other snake venom
110
disintegrins are rather promiscuous in that they can bind to more than one β1, β3
or β5 integrins yet showing different levels of binding affinity and selectivity
(McLane, Marcinkiewicz et al. 1998).
CN is the disintegrin isolated from the venom of Agkistrodon contortrix
contortrix (southern copperhead) (Trikha, Rote et al. 1994). As an anti-cancer
agent, CN proved to be a potent anti-angiogenic and anti-metastatic agent in
several in vivo animal models (Trikha, De Clerck et al. 1994; Trikha, Rote et al.
1994; Schmitmeier, Markland et al. 2000; Zhou, Hu et al. 2000; Markland, Shieh
et al. 2001). CN was also shown to directly engage tumor cells and suppress
their grow in a cytostatic manner (Trikha, De Clerck et al. 1994; Trikha, Rote et
al. 1994; Schmitmeier, Markland et al. 2000). The antitumor activity of CN is
based on its high affinity interaction with integrins α5 β1, αvβ3 and αvβ5 on both
cancer cells and newly growing vascular endothelial cells resulting in an anti-
angiogenic effect in adhesive and apoptotic assays (Trikha, De Clerck et al.
1994; Zhou, Nakada et al. 1999; Zhou, Nakada et al. 2000; Zhou, Sherwin et al.
2000). In addition, CN’s anti-angiogenic and anti-invasive properties were shown
to inhibit HUVEC invasion, migration, and tube formation (Golubkov, Hawes et al.
2003). One of the ways integrins initiate signaling is by the disruption of an
autoinhibitory mechanism within FAK causing autophosphorylation, which allows
FAK to complex with Src and other cellular proteins. FAK’s internal kinase activity
or scaffolding function will then trigger downstream signaling. Both integrins and
FAK are found to play crucial roles in the maintenance of stem cells in studies
using mouse models, suggesting that integrin signaling through FAK may serve
111
as a functional marker for many types of stem cells (Guan 2010). CN has been
shown to differentially hyper-phosphorylate FAK in mammary and bladder cancer
cells (Ritter and Markland 2000; Ritter, Zhou et al. 2000), which provided the first
evidence that CN has potential as a soluble signaling molecule and not just as an
integrin antagonist. This diverse mechanism of action provides CN with a distinct
advantage over many antiangiogenic agents that only block a single angiogenic
pathway and/or do not directly target tumor cells.
For eventual clinical use, however, the direct isolation of native CN from
snake venom could not be contemplated as an economically viable solution since
this polypeptide represents a very minor fraction relative to other venom
components. For this reason, a recombinant expression system was sought that
was amenable to large-scale production of a biologically active recombinant
product that could serve as a clinically translatable version of CN. In this study
the in vitro functional characterization of a biologically active chimeric variant of
CN, called VCN, that possesses the same anti-angiogenic, anti-invasive, and
signaling capabilities as CN is reported.

5.3 Materials and Methods
5.3.1 Cell Culture and Antibodies
MDA-MB-435 human cancer cells (Chambers 2009; Lacroix 2009), an
estrogen receptor-negative cell line, were obtained from Dr. Janet Price (MD
Anderson Cancer Center, University of Texas, Houston, TX) and MDA-MB-231
112
cells (ATTC, Manassas, VA) were maintained in DMEM supplemented with 10%
FBS in the presence of 5% CO
2
. HUVECs were obtained from PromoCell
(Heidelberg, Germany) and maintained in complete media with supplements at
37° C and 5% CO
2
according to the manufacturer’s protocol. Cells were serum
starved overnight and then harvested by brief trypsinization with 10% trypsin
stock (0.05% trypsin- 0.02% EDTA) in PBS for 5-min, followed by quenching in
0.2% soybean trypsin inhibitor. Cells were resuspended in serum free media and
left in suspension at 37 °C for 1-hour before initiating experiments. Abs used
were anti-FAK (clone A-17), anti-mouse HRP, anti-goat HRP, and A/G PLUS-
Agarose from Santa Cruz Biotechnology (Santa Cruz, CA); anti-p-Tyrosine (clone
P-Tyr-102) from Cell Signaling Technology (Danvers, MA); anti-FAK (clone 77)
from BD Biosciences (Bedford, MA).

5.3.2 Platelet Aggregation
The inhibition of ADP-induced PA by recombinant disintegrins was
determined by measuring the light absorption of human platelet-rich plasma
(PRP) in a specialized spectrophotometer (Platelet Aggregometer IV Plus;
Helena Laboratories, Beaumont, TX). Whole human blood (40 ml) obtained from
volunteers was drawn into 4 ml of 0.1M sodium citrate. The donors had not
received any NSAID-based medication for at least 2 weeks prior to the blood
draw. The citrated blood was centrifuged (150xg for 30-min) at RT and the
supernatant (PRP) was carefully transferred to a new tube. A volume of 1 ml of
PRP was further centrifuged for an additional 1-min in a benchtop centrifuge at
113
full speed. This supernatant was retained as platelet-poor plasma (PPP) and
utilized as a baseline reference in the assay. Inhibition of ADP-induced PA was
monitored at RT by adding samples (native CN or recombinant disintegrins) over
a range of concentrations (0–250 nM) to PRP 1-min prior to the addition of ADP.
The level of aggregation inhibition was plotted against the final concentration of
disintegrins, either native or recombinant, added to each sample.

5.3.3 Cell Invasion Through a Reconstituted Basement
Membrane

To assess the ability VCN to block the invasion of either HUVEC, MDA-
MB-231, or MDA-MB-435 cells through a reconstituted basement membrane a
fluorometric cell invasion assay kit (QCM
TM
24-Well Cell Invasion) from Chemicon
(Temecula, CA) was utilized. This kit contains a number of invasion chambers
consisting of plastic inserts (pre-coated with ECM proteins) that fit into a 24-well
tissue culture plate. The cells to be analyzed are seeded in these inserts and
allowed to adhere to and invade through a thin ECM layer (ECMatrix
TM
) and a
porous polycarbonate membrane (8 µM-pore size) toward a chemical gradient.
The ECMatrix
TM
serves as a reconstituted basement membrane and is an ECM
extract prepared from the Engelbreth Holm-Swarm (EHS) mouse sarcoma. The
ECM layer occludes the membranes pores by blocking the non-invasive cells
from migrating through. The cells that made it all the way through the porous
membrane into the lower chamber are then harvested and quantitated. In the
studies, either HUVEC or MDA-MB-435 cells were serum starved overnight,
114
harvested and resuspended in serum-free media (10
6
) and further incubated in
the presence of various concentrations (0-1000 nM) of either native CN or VCN
for 30-min at 37°C. At the end of the incubation period, cell aliquots (300 µl) from
each condition were added to the invasion inserts, whereas the bottom wells of
each chamber received 500 µl of chemoattractant (conditioned media from
human HT1080 fibrosarcoma cells). The invasion plates were then incubated for
either 24-hours (highly invasive MDA-MB-435 cells) or 48-hours (HUVECs) at
37°C in the presence of 5% CO2 and the cells were allowed to invade through the
ECM toward the chemoattractant gradient. Once invaded, the cells that moved
through the pores to the lower chamber (i.e., to the other side of the porous
membrane) were detached and lysed according to the manufacturer protocol.
The total DNA content from each cell lysate was labeled with a DNA-binding
fluorescent dye (CyQUANT) and the invaded cells quantitated by measuring the
relative fluorescence of each condition in a SPECTRAmax GeminiEM fluorescent
plate reader (Molecular Devices, Sunnyvale, CA). The results were calculated in
% invasion, where the untreated control was considered as 100% invasion.

5.3.4 Endothelial Cell Actin Cytoskeleton Disruption
Chamber slides (8-well format) were coated with complete Matrigel (BD
Biosciences, Bedford, MA) that was allowed to polymerize overnight at 37°C, 5%
CO
2
. The next day, the excess matrix was aspirated out of the chamber slides
and HUVECs (4x10
4
) were seeded in triplicate in the presence of either FITC-CN
(10 nM) or FITC-VCN (10 nM) or without any treatments. Following 1-hour of
115
incubation at 37°C in the presence of 5% CO
2
, cells were washed twice with PBS
and then fixed in 3.7% formaldehyde solution in PBS for 10-min at RT. The cells
were washed again twice with PBS before being permeabilized in 0.1% Triton X-
100 in PBS for 5-min. After washing the cells two more times in PBS, the
chambers were stained with 200 µl rhodamine-phalloidin in PBS for 20-min at
RT. The labeled phallotoxin stock was resuspended in methanol according to
the manufacturer protocol and further diluted 1:40 in PBS before being added to
the chambers. At the end of the staining procedure, the HUVECs were washed
again with PBS and then imaged by multiphoton laser scanning confocal
microscopy (LSM 510 Confocal/Titanium Sapphire Laser, Confocal Core at
University of Southern California).

5.3.5 Endothelial Cell Tube Formation
‘Endothelial Tube Formation’ plates were obtained that were precoated
with Matrigel by the manufacturer (BD Biosciences, Bedford, MA). ‘Endothelial
Tube Formation’ plates were brought out of storage at -20°C and allowed to thaw
at 4°C for 24-hours. Prior to treatment the plates were allowed to polymerize at
37°C for 30-min. HUVECs (3x10
4
) were seeded in triplicate in the presence of
various concentrations of either native CN or VCN (0-1000 nM) or a known tube
formation inhibitor Suramin (Calbiochem, San Diego, CA) and the plates were
incubated for 18-hours at 37°C in the presence of 5% CO
2
. At the end of the
incubation period, cells were washed twice with PBS, stained with Calcein AM
(Invitrogen, Carlsbad, CA) at 8 µg/ml for 30-min and re-washed twice with PBS.
116
The plates were imaged by multiphoton laser scanning confocal microscopy
(LSM 510 14 Confocal/Titanium Sapphire Laser, Confocal Core at University of
Southern California) and micrographs acquired. The total length of tubes formed
by HUVECs in the presence or absence of disintegrins was quantitated with
Zeiss LSM Image software, which was fed images collected from multiple
independent experiments to generate each data point. The captured confocal
images were quantitated for the amount of formed tubes based on total tube
length measurements using the Zeiss LSM image Browser. Individual tubes from
each image were measured blindly by three separate individuals and the total
tube length from each data set was averaged (p<0.005). Individuals doing the
measurements were instructed to discount any structures that were less than 5
µM in length and use their own judgment when deciding which structures were in
fact endothelial tubes.

5.3.6 FAK Phosphorylation
MDA-MB-435 (10
6
) cells were treated with various concentrations (0-500
nM) of either CN or VCN for 10 to 30-min, washed with PBS, and further lysed in
a modified RIPA buffer containing: 50mMol/L of Tris (pH 7.4), 150mMol/L of
NaCl, 1% Nonidet P-40, 0.5% desoxycholate, 0.1% sodium dodecyl sulfate,
1mMol/L sodium orthovanadate, 1mMol/L sodium pyrophosphate, 50mMol/L
sodium fluoride, and a protease inhibitor cocktail. The cell lysates were passed
through a 26-gauge syringe needle to remove most of the insoluble debris and
incubated on ice for 15-min. The insoluble material was further separated by
117
centrifugation in a benchtop centrifuge at top speed for 15-min. Supernatants
were collected and total protein contents standardized by BCA protein assay
(Pierce, Rockford, IL). The immunoprecipitation was carried out for each
condition by incubating whole cell lysates (300 µg of total protein per condition)
with 4 µg of rabbit polyclonal anti-total FAK at 4°C for 2-4 hours followed by 40 µl
protein A/G PLUS-Agarose overnight at 4°C. Immune complexes were washed
three times in lysis buffer without protease inhibitors and dissociated by adding
SDS-PAGE sample buffer and boiling them for 5-min. Immunoprecipitates were
resolved by 4-20% SDS-PAGE and transferred overnight at 4°C and 125 mAmp
to nitrocellulose membranes (BioRad Laboratories, Hercules, CA) in a transfer
buffer containing 0.025% SDS. The blots were blocked with 2% BSA for 1-hour
at RT then washed twice with TTBS. The incubation with a mouse anti-p-
Tyrosine antibody was performed in blocking buffer for 2-hours at RT. After
washing three times with TTBS, membranes were incubated with an anti-goat
HRP in blocking buffer 1-hour at RT and then washed extensively. The
immunoblots were developed using the Super Signal® West Pico
Chemiluminescent Substrate from Pierce (Pierce Chemical Co., Rockford, IL).
The same blots were stripped using the ReBlot Plus kit and reprobed with a
mouse monoclonal anti-FAK to demonstrate equal loading. All experiments were
repeated at least three times to ensure the significance of the results.

118
5.4 Results
5.4.1 Inhibition of Platelet Aggregation by Disintegrins
The disintegrins were tested as inhibitors of ADP-induced PA.
Interestingly, the data from this assay showed that VCN had a potent inhibitory
effect on PA with a calculated IC50 of ~60 nM (Figure 19). The VCN slope
generated by graphing the PA data was very similar to native CN, indicating that
they behave almost identically for inhibiting PA.

5.4.2 Inhibition of Cell Invasion Through a Reconstituted
Basement Membrane by Disintegrins

In order to assess the effect of VCN on HUVEC, MDA-MB-231, and MDA-
MB-435 motility, cells were seeded in invasion chambers (modified Boyden
chambers) and allowed to invade in the presence of disintegrins through a
reconstituted laminin-rich basement membrane against a chemoattractant
gradient. Cytochalasin D control wells demonstrated below 20% invaded cells
for each cell type. Invasion data showed that VCN significantly (p<0.001)
affected HUVEC, MDA-MB-231, and MDA-MB-435 motility in this setting with a
potency similar to native CN, bringing the percentage of invaded cells below 40%
at a low concentration of 10 nM (Figure 20).

Figure 19

VCN and Native CN Exhibit Identical Effect on Platelet Aggregation. Different
concentrations of either VCN or native CN were incubated with platelet-rich plasma. The addition
of ADP in the absence of an inhibitor induces platelet aggregation. When added immediately
prior to ADP, both VCN and native CN inhibit platelet aggregation with an IC50 of ~60 nM.

119
Figure 20

Inhibition of HUVEC invasion through a reconstituted basement membrane.
HUVECs, MDA-MB-231, or MDA-MB-435 cells were preincubated with various concentrations of
either CN or VCN (0-1000 nM) for 10-min before being seeded on porous inserts coated with
ECMatrix
TM
and allowed to migrate against a chemoattractant gradient for 42-hours. The invaded
cells were detached, lysed, stained with CyQuant, a DNA-binding fluorescent dye, and
quantitated in a fluorescent plate reader.

120
121
5.4.3 Disruption of Endothelial Cell Actin Cytoskeleton by
Disintegrins

HUVECs were assessed for how CN or VCN treatment would affect cell
morphology and actin cytoskeleton organization. Cells were treated with FITC-
labeled polypeptides, fixed, permeabilized and counter-stained with rhodamine-
phalloidin. Untreated cells demonstrate nicely formed actin filaments distributed
throughout the cells that show no signs of stress (Figure 21A). In stark contrast,
both CN (Figure 21B-21D) and VCN (Figure 21E-21G) treatments were found to
have a significant impact on HUVEC actin cytoskeleton organization as shown by
representative confocal images. These HUVECs have almost no clear actin
filaments that are visible and appear to be under extreme stress conditions.
Moreover, VCN demonstrated the same ability to disrupt the actin cytoskeleton
as native CN.

5.4.4 Inhibition of Endothelial Cell Tube Formation by
Disintegrins

To evaluate VCN’s capability as an anti-angiogenic agent in vitro, HUVEC
tube formation assays were conducted. Given the presence of certain growth
factors, a precisely polymerized Matrigel matrix layer, and perfect timing,
HUVECs will form endothelial vessels in vitro. The untreated well demonstrated
exactly how extensive these tubes could be with long and multi-branching
structures (Figure 22A). In contrast, the control wells that were treated with
suramin were unable to form widespread tubes (Figure 22B and 22C).
Figure 21

Actin Cytoskeleton Staining of HUVECs Treated with FITC-VCN/CN. Chamber
slides were coated with complete Matrigel that was allowed to polymerize overnight at 37°C, 5%
CO
2
. HUVECs (4x10
4
) were seeded in triplicate in the presence of either FITC-CN (10 nM) or
FITC-VCN (10 nM) or without any treatments. Following 1-hour of incubation at 37°C in the
presence of 5% CO
2
, cells were washed twice with PBS and then fixed in 3.7% formaldehyde
solution in PBS for 10-min at RT. The cells were permeabilized with 0.1% Triton X-100 in PBS
for 5-min. Chambers were stained with rhodamine-phalloidin in PBS for 20-min at RT. HUVECs
were imaged by multiphoton laser scanning confocal microscopy. Representative confocal
images of double-stained cells are shown at 63x. Panel A: untreated; Panels B-D: 10 nM CN;
Panels E-G: 10 nM VCN.

122
Figure 22

Inhibition of HUVEC Tube Formation. HUVECs were plated on ‘Endothelial Cell Tube
Formation’ plates in the presence of various concentrations of either CN or VCN (0-1000 nM). A
known tube formation inhibitor (Suramin) was used as a control. Representative figures from
three independent experiments were shown above: panel A: untreated control; panel B: 50 µM
Suramin; panel C: 100 µM Suramin; panel D: 1 nM CN; panel E: 10 nM CN; panel F: 1000 nM
CN; panel G: 1 nM VCN; panel H: 10 nM VCN; panel I: 1000 nM VCN. Cells were stained with
Calcein AM and imaged using confocal microscopy (2.5x). Panel J: The captured confocal
images were quantitated for the amount of formed tubes based on total tube length
measurements using the Zeiss LSM image Browser. Individual tubes from each image were
measured blindly by three separate individuals and the total tube length from each data set was
averaged (p<0.005).

123
Figure 22: Continued

124
125
Furthermore, the tube formation data showed that VCN significantly inhibited the
ability HUVECs to form tubes (Figure 22G-22I) with a similar potency as native
CN (Figure 22D-22F).

5.4.5 FAK Hyperphosphorylation by Disintegrins
CN was previously shown to engage its integrin targets in an agonistic
manner (Ritter, Zhou et al. 2000; Schmitmeier, Markland et al. 2000; Ritter, Zhou
et al. 2001). To understand whether this effect is a function of the native
molecule’s ability to cross-link its target receptors on the cell surface (i.e., a
consequence of being a homodimer) or rather an intrinsic property of CN’s
sequence (i.e., the residues flanking the tripeptide motif at the tip of the
disintegrin loop), the ability of VCN to function as a signaling molecule and
hyperphosphorylate FAK was assessed. In this assay, serum starved and
suspended MDA-MB-435 cells were exposed to various concentrations of either
VCN or CN and the levels of global FAK phosphorylation was measured by
Western blotting. The untreated control established the natural level of FAK
phosphorylation for MDA-MB-435 cells in suspension (Figure 23). When these
same cells are exposed to CN treatment at 1 nM, FAK becomes
hyperphosphorylated, a trend that continues in a dose dependent manner. It
appears that VCN acts as signaling molecule in a similar fashion to native CN,
because the global levels of FAK phosphorylation in MDA-MB-435 cells after
treatment with various concentrations of VCN follows the same dose dependent
trend that native CN demonstrated.
Figure 23

FAK Phosphorylation Levels in MDA-MB-435 Cells Treated with Soluble
Disintegrins. MDA-MB-435 (10
6
) cells were treated with various concentrations (0-500 nM)
of either CN or VCN for 10 to 30-min, washed with PBS, and further lysed in a modified RIPA
buffer. Immunoprecipitation was carried out for each condition by incubating whole cell lysates
(300 µg of total protein per condition) with 4 µg of rabbit polyclonal anti-total FAK at 4°C for 2-4
hours followed by 40 µl protein A/G PLUS-Agarose overnight at 4°C. Immune complexes were
dissociated by adding SDS-PAGE sample buffer and boiling for 5-min. Immunoprecipitates were
resolved by 4-20% SDS-PAGE and transferred overnight at 4°C and 125 mAmp to nitrocellulose
membranes. The immunoblots were probed with anti-p-Tyrosine and developed via x-ray film.
The same blots were stripped reprobed with a mouse monoclonal anti-FAK to demonstrate equal
loading.

126
127
5.5 Discussion
The process of angiogenesis is dependent on the integrin cell surface
receptor family and has been successfully repressed through antagonistic
binding to these proteins by therapeutic agents (Mahabeleshwar, Feng et al.
2006). Cancer angiogenesis is a multifaceted process that involves numerous
paracrine and autocrine loops and a still poorly understood cross-talk between
transformed epithelial cells and activated endothelial cells originating from
preexisting vessels. In this process, the activated endothelial cells are recruited
into the tumor microenvironment in a multistep manner characterized by: (1) the
activation and loss of polarity of the endothelial cells found in the tumor vicinity;
(2) the rapid proliferation of these endothelial cells which manage to escape their
normal microenvironment pressure; (3) the activation of a migratory program in
these cells; (4) the migration of the angiogenic endothelial cells that will populate
the tumor microenvironment against chemokine gradients; and (5) the self-
assembly of the migrated cells into neovessels along a provisional basement
membrane (Folkman 1997). Integrins are involved in several of the steps
delineated above (Hood and Cheresh 2002) and a number of motogenic integrins
(e.g., αvβ3, αvβ5, α4 β1, and α5 β1), which were shown to be instrumental to
angiogenesis, appear to be activated and/or up-regulated by signals (i.e., VEGF
and basic fibroblast growth factor) channeled through the growth factor/receptor
tyrosine kinase system (Byzova, Goldman et al. 2000). Furthermore, the
existence of cross-talk between a number of motogenic integrins has been
suggested to regulate the survival and migration of angiogenic endothelial cells in
128
the avascular tissue before they assemble into neovessels (Kim, Bell et al. 2000;
Kim, Harris et al. 2000).
Because disintegrins have evolved to bind a wide variety of integrins with
high affinity, they have the potential to become broad-spectrum anti-cancer
agents. Generally, these agents are thought to interfere with the earliest steps of
angiogenesis, which are mainly characterized by adhesive and migratory events
(Zhou, Nakada et al. 1999; Schmitmeier, Markland et al. 2000; Zhou, Sherwin et
al. 2000; Markland, Shieh et al. 2001; Swenson, Costa et al. 2004) thus
preventing the subsequent neovessel assembly. Nonetheless, it has been
reported that a few angiogenesis inhibitors can actually cause a regression of the
already assembled neovessels and/or a process of normalization (i.e., reversal of
structural and functional abnormalities in tumor vessels) in those vessels that are
mature enough to resist regression (Baluk, Hashizume et al. 2005). For
instance, Huang et al. reported in 2003 that a high-affinity blockade of VEGF by a
VEGF-Trap molecule abolished preexisting vasculature in established tumor
xenografts and that was followed by marked tumor regression, including
regression of lung micrometastases (Huang, Frischer et al. 2003). The downside
of this powerful VEGF blockade is that the basement membrane of tumor vessels
tended to persist after endothelial cells/pericytes underwent regression. This
provides a scaffold for rapid repopulation of these ‘tracks’ with endothelial cells
once the anti-VEGF treatment is discontinued (Baluk, Hashizume et al. 2005).
For this reason, the use of disintegrins, agents that may mainly interfere with the
129
migratory steps of angiogenesis, might represent an attractive adjuvant treatment
to VEGF neutralizing agents.
Although the native disintegrin CN has already been fully characterized in
the Markland laboratory and shown to possess anti-angiogenic and anti-invasive
properties (Zhou, Nakada et al. 1999; Zhou, Sherwin et al. 2000; Swenson,
Costa et al. 2005) in in vitro and in vivo experiments, the yield of the purified
disintegrin from venom is so low it makes the transition of such a molecule to the
clinic a difficult endeavor. Hence, the production of a recombinant version of CN
commenced with a number of attempts directed at recombinant disintegrin
expression in eukaryotic cells, which yielded disappointing levels of expression.
However, VCN was eventually synthesized in a in a soluble active form in an
engineered E. coli system and purified with significant yields (10-20 mg of active
protein per L of bacterial culture). By folding in a compact structure stabilized by
multiple disulfide bonds, VCN as a therapeutic has a design advantage which
makes it more resistant to proteolysis and better able to survive the harsh tumor
microenvironment that involves wide pH variations. Moreover, its high stability in
very acidic environments makes it ideal for rapid reverse phase HPLC
purification in organic solvents (acetonitrile) directly from bacterial lysates. In
order for VCN to have any value to the Markland laboratory it was necessary to
compare VCN to CN in several in vitro assays to verify that the recombinant
protein maintained full biological activity.
VCN showed potent biological activity in four classical in vitro assays
aimed at evaluating the molecule’s anti-invasive and anti-angiogenic capabilities.
130
First, VCN at low nanomolar concentrations was shown to disrupt the ability of
platelets to aggregate after being stimulated by ADP (Figure 19). Because this
is one of the hallmark features of native disintegrins, this provides evidence that
VCN adopted a correct disintegrin fold. Secondly, VCN demonstrated anti-
angiogenic potential by inhibiting HUVEC tube formation (Figure 22A and 22B).
Untreated HUVECs formed distinct microvessels in a specific environment
facilitating angiogenesis. These microvessels are representatives of an excellent
in vitro mimic for in vivo angiogenesis, and the fact that VCN abolishes almost all
tube formation present in the untreated control clearly indicates VCN’s potential
as an anti-angiogenic molecule. HUVEC, MDA-MB-231, and MDA-MB-435
motility through a reconstituted basement membrane was also severely hindered
due to VCN treatment (Figure 20). This in vitro experiment was intended to
mimic the invasive properties required by endothelial cells, in order to form
neovessels, and cancer cells, in order to become metastatic. VCN proved to be
as good an anti-invasive molecule as native CN by inhibiting invasion at similar
concentrations. Lastly, HUVEC morphology, as indicated by actin cytoskeleton
staining, was significantly altered by soluble VCN treatment (Figure 21). This
finding suggests that persistent stimulation with a soluble agonistic integrin ligand
in the absence of integrin tethering, a process that normally happens in the
presence of a solid ECM support (stimulation sans tethering), may have a
significant impact on the process of neovessel assembly. In support of this
theory, it has been previously shown that persistent interference with beta3
integrins leads to a trans-dominant inhibition of other integrins (Diaz-Gonzalez,
131
Forsyth et al. 1996), a process that may make a target cell unresponsive to
further stimulation by soluble motogenic cues.
In an effort to provide a mechanistic explanation for its anti-migratory
effects, it was shown that VCN globally alters the phosphorylation status of
tyrosine residues in FAK in suspended MDA-MB-435 cells (Figure 23) receiving
no other input except VCN, much like CN has been demonstrated to do in the
past (Ritter and Markland 2000). By doing so, VCN may interfere with the
efficient assembly of the locomotor apparatus in these cells presumably with an
impact on tumor progression (Mitra, Mikolon et al. 2006; Mitra and Schlaepfer
2006). The hyperphosphorylation of FAK could indicate the assembly or
misassembly of FAs in suspended cells. When cells are exposed to these
conditions it may put them into a state of confusion, since they are not bound to
an ECM substrate for which they are accustomed. Once the cells are confused
they may adopt a defiant phenotype, not willing to do anything, much like what is
observed in disintegrin treated cells. Globally altering the phosphorylation status
of FAK may also lead to further phosphoylation events in different intracellular
proteins, downstream of FAK, that could help to decipher the exact signaling
pathways engaged by disintegrins.
In summary, it has now been demonstrated by several in vitro assays, that
recombinant VCN behaves very similarly to native CN. These experiments
indicate that the chimeric VCN retains the ability of its parent disintegrins in an in
vitro setting, helping to confirm that the recombinant molecule folded properly in
the E. coli production method and resembles a classical disintegrin. The results
132
from these assays provide evidence for the acceptance of VCN as a fully
functional anti-angiogenic and anti-invasive molecule in the treatment of cancer.
Furthermore, these observations will help to accelerate VCN’s transition to a
clinical setting.

133
Chapter 6: Conclusions and Future Directions
Although disintegrins have been studied for over twenty years now, their
true capability as signaling molecules is still poorly understood. The intent of this
dissertation was to provide mechanistic studies on the native disintegrin CN and
characterization of the recombinant protein VCN. Because integrins are a
complex family of receptors that have the ability to conduct in inside →out and
outside →in signaling (Hynes 2002), both intracellular and extracellular events
were examined. These studies also called for the development of a novel
measurement procedure as well as the redefining of some classical techniques.
In an attempt to see if CN treatment has any influence on gene expression
at the mRNA level, Olgio GEArrays were employed to study over 100 genes at
once. Two specific arrays were examined, HEM & AM and HSTPF. The reason
the HEM & AM was selected was because of the close relationship integrin
signaling plays with the ECM and other adhesion molecules. The important
observations that were made from this microarray were that CN signaling seems
to be somewhat cell specific and, generally speaking, CN acts as an up-regulator
of gene expression. The HSTPF oligo array was chosen because it incorporates
genes from multiple signal transduction pathways, which gave the best chance at
discovering CN influenced pathways. Many genes were determined to be up-
regulated with CN treatment, including several transcription factors.
Furthermore, noticeable differences between CN treated cell lines were
observed, providing further evidence that CN signaling seems to be influenced in
a cell specific manner. To verify these results RT-PCR experiments were
134
conducted, which were more quantitative and specific. These experiments made
it clear that CN treatment up-regulates transcription factors as well as several
other important genes explicitly from cell type to cell type. It is theorized that this
specificity is due to the integrin expression profile of the tested cell lines and
each integrin may react to CN ligation in a slightly different way. Based on this
work, new RT-PCR experiments need to be conducted with different cell lines
that over-express or under-express particular integrins. Each cell line’s integrin
expression profile needs to be carefully measured or engineered in order to fully
characterize which integrins are present. These cell lines can then be compared
and contrasted with each other for the determination of integrin specificity in
regards to CN signaling and gene expression. Also, because genes are not
individual entities, meaning they work together in the context of a complete
organism, the entire genome needs to be evaluated as a whole in order to
examine exactly what occurs during CN treatment. This will allow for total
pathway analysis without any breaks in the pattern and determination of
beginning and end points of mRNA transcriptional changes.
It was quickly established that CN influenced the rate at which integrins
become internalized through Western blotting and FACS analysis, but because
the techniques employed here were both non-quantitative and unable to be
visualized, a new procedure was deemed necessary. Drawing on a classical
technique that employs acidic Ab stripping buffers to remove cell surface signal
for an in vs. out approach, a new procedure was designed that overcame the
flaws in the classical technique. In this new approach, the proteolytic enzyme
135
pepsin was utilized to remove Ab off the cell surface and cleave it into two
fragments, the F(ab ′)2 region and the Fc region. In conjunction with Fc specific
secondary Ab, this phenomenon was utilized to effectively and efficiently remove
all residual cell surface fluorescence in assays that were designed to distinguish
between in vs. out cellular signals. This technique was employed to confirm that
CN treatment causes the internalization rates for beta1 integrins to escalate. In
the future, this same technique could be employed to track CN directly through
internalization studies utilizing specific Abs designed against CN. This research
could explore the possibilities that disintegrins are utilized within the interior of
cells and not just as antagonistic ligands on the exterior. Also, with this novel
technique it is now possible for CN induced integrin internalization to be precisely
measured in specific cell lines exhibiting particular integrins. It is possible that
CN ligation only affects bound integrins, but it is also possible that neighboring
integrins are affected as well, integrins that may not even be capable of binding
CN.
The mechanistic investigation to how CN works led to many comparisons
with cyclic RGD molecules, in particular the cilengitide precursor cRGDfV. The
most staggering difference between these molecules is that even though they
exhibit the same RGD binding motif and have similar binding kinetics, CN
maintains around 1000-fold better potency in in vitro assays designed to
measure anti-invasive properties. In order to explain this phenomenon it was
shown that CN causes β1 integrin inactivation in MDA-MB-231 cells via
displacement of talin from the cytoplasmic β tail. It was further demonstrated that
136
other FA molecules, vinculin and RIAM, were impacted by CN treatment as well.
Confocal images of HUVECs actively forming endothelial neovessels revealed
that talin based FAs were all but non-existent after CN treatment, when
compared to untreated control. Furthermore, it was established that the activity
level of talin was being inhibited by CN treatment. Rap1, the key GTPase
involved with talin recruitment to the plasma membrane, was found to bind less
GTP after CN treatment. This leaves open the possibility that CN causes cells to
be arrested in some part of the cell-cycle, which could easily be measured in the
future by cell-cycle analysis through FACS or measuring cyclin levels via RT-
PCR experiments. Also, after examining total cell lysate Western blots of
HUVECs after a prolonged exposure to CN, it was revealed that the cells
significantly express less talin-H and talin-R domains when compared to the
whole protein. This suggests that the cleavage of talin into the more active
protein subdomains is being inhibited. These experiments pinpoint talin
interactions at the forefront of CN’s mechanism of action. With this discovery,
other talin related proteins and pathways need to be examined during CN
treatment. For instance, the cleavage of talin by calpain-II should be measured
kinetically at different time points after CN treatment. This could be done through
serial Western blotting after protein synthesis was inhibited by cycloheximide
treatment. Also, the precise phosphorylation status of talin should be examined
to determine if this plays a role in calpain-II recognition during CN treatment.
In order to facilitate the transition of the recombinant protein VCN from the
laboratory to the clinic, several in vitro assays were conducted to establish its
137
effectiveness as an anti-angiogenic and anti-invasive molecule. Assays included
HUVEC tube formation, invasion through a reconstituted basement membrane,
actin cytoskeleton staining, and PA for which VCN performed as well as CN in all
aspects. In addition, VCN was able to elicit the same hyperphosphorylation
events in FAK that were previously demonstrated by CN (Ritter and Markland
2000). These assays help to confirm that the chimeric molecule VCN, acts as
native disintegrins in vitro and the utilized production method allows for proper
folding and display of the RGD adhesive loop. Additional mechanistic studies
need to be conducted in order to elucidate whether VCN is able to differentially
alter the phosphorylation status of key FAK tyrosine residues. The precise
understanding of what key tyrosine residues in FAK are being impacted by
disintegrins will shed a light on the intimate mechanism through which these
small soluble ECM-mimetics exert their potent effect on cell motility. Also, these
events should be examined in multiple contexts, meaning that suspended cells
should be compared to cells that are actively invading through an ECM matrix.
This would allow for theoretical differences to be measured between cells in an in
vitro context vs. cell in a more in vivo context.
In this dissertation the disintegrin CN has been studied mechanistically.
The results from the experiments discussed here will help to advance in
disintegrin research in the future. Furthermore, the characterization of the
recombinant disintegrin VCN may help to provide the world with the first clinically
relevant disintegrin molecule.

138
Bibliography
Abram, C. L. and C. A. Lowell (2009). "The ins and outs of leukocyte integrin
signaling." Annu Rev Immunol 27: 339-62.

Adler, M., R. A. Lazarus, et al. (1991). "Solution structure of kistrin, a potent
platelet aggregation inhibitor and GP IIb-IIIa antagonist." Science
253(5018): 445-8.

Alberts, B., Johnson, A., Lewis, J., Raff, M., Roberts, K., Walter, P. (2002). The
Molecular Biology of the Cell, Garland Science.

Alghisi, G. C., L. Ponsonnet, et al. (2009). "The integrin antagonist cilengitide
activates alphaVbeta3, disrupts VE-cadherin localization at cell junctions
and enhances permeability in endothelial cells." PLoS One 4(2): e4449.

Altankov, G. and F. Grinnell (1995). "Fibronectin receptor internalization and AP-
2 complex reorganization in potassium-depleted fibroblasts." Exp Cell Res
216(2): 299-309.

Ammer, A. G. and S. A. Weed (2008). "Cortactin branches out: roles in regulating
protrusive actin dynamics." Cell Motil Cytoskeleton 65(9): 687-707.

Andrew, S. M. and J. A. Titus (2001). "Fragmentation of immunoglobulin G." Curr
Protoc Immunol Chapter 2: Unit 2 8.

Askari, J. A., P. A. Buckley, et al. (2009). "Linking integrin conformation to
function." J Cell Sci 122(Pt 2): 165-70.

Azam, M., S. S. Andrabi, et al. (2001). "Disruption of the mouse mu-calpain gene
reveals an essential role in platelet function." Mol Cell Biol 21(6): 2213-20.

Baker, E. L. and M. H. Zaman (2009). "The biomechanical integrin." J Biomech.

Bakiri, L., D. Lallemand, et al. (2000). "Cell cycle-dependent variations in c-Jun
and JunB phosphorylation: a role in the control of cyclin D1 expression."
Embo J 19(9): 2056-68.

Baluk, P., H. Hashizume, et al. (2005). "Cellular abnormalities of blood vessels
as targets in cancer." Curr Opin Genet Dev 15(1): 102-11.

Barczyk, M., S. Carracedo, et al. (2009). "Integrins." Cell Tissue Res 339(1): 269-
80.

139
Barker, P. L., S. Bullens, et al. (1992). "Cyclic RGD peptide analogues as
antiplatelet antithrombotics." J Med Chem 35(11): 2040-8.

Barroso-Gonzalez, J., J. D. Machado, et al. (2009). "Moesin regulates the
trafficking of nascent clathrin-coated vesicles." J Biol Chem 284(4): 2419-
34.

Bazaa, A., P. Juarez, et al. (2007). "Loss of introns along the evolutionary
diversification pathway of snake venom disintegrins evidenced by
sequence analysis of genomic DNA from Macrovipera lebetina
transmediterranea and Echis ocellatus." J Mol Evol 64(2): 261-71.

Bazzoni, G., D. T. Shih, et al. (1995). "Monoclonal antibody 9EG7 defines a novel
beta 1 integrin epitope induced by soluble ligand and manganese, but
inhibited by calcium." J Biol Chem 270(43): 25570-7.

Beckerle, M. C., K. Burridge, et al. (1987). "Colocalization of calcium-dependent
protease II and one of its substrates at sites of cell adhesion." Cell 51(4):
569-77.

Bennett, J. S., B. W. Berger, et al. (2009). "The structure and function of platelet
integrins." J Thromb Haemost 7 Suppl 1: 200-5.

Bennett, J. S., C. Chan, et al. (1997). "Agonist-activated alphavbeta3 on platelets
and lymphocytes binds to the matrix protein osteopontin." J Biol Chem
272(13): 8137-40.

Bernhagen, J., R. Krohn, et al. (2007). "MIF is a noncognate ligand of CXC
chemokine receptors in inflammatory and atherogenic cell recruitment."
Nat Med 13(5): 587-96.

Bertoni, A., S. Tadokoro, et al. (2002). "Relationships between Rap1b, affinity
modulation of integrin alpha IIbbeta 3, and the actin cytoskeleton." J Biol
Chem 277(28): 25715-21.

Bork, P., T. Doerks, et al. (1999). "Domains in plexins: links to integrins and
transcription factors." Trends Biochem Sci 24(7): 261-3.

Bos, J. L. (2005). "Linking Rap to cell adhesion." Curr Opin Cell Biol 17(2): 123-8.

Brigham-Burke, M., O'Shannessy, D.J. (1992). "A micro-scale method employing
surface plasmon resonance detection for the determination of conditions
for immunoaffinity chromatograpghy of proteins." Chromatographia 35: 45-
49.

140
Brooks, P. C., R. A. Clark, et al. (1994). "Requirement of vascular integrin alpha v
beta 3 for angiogenesis." Science 264(5158): 569-71.

Buback, F., A. C. Renkl, et al. (2009). "Osteopontin and the skin: multiple
emerging roles in cutaneous biology and pathology." Exp Dermatol 18(9):
750-9.

Byzova, T. V., C. K. Goldman, et al. (2000). "A mechanism for modulation of
cellular responses to VEGF: activation of the integrins." Mol Cell 6(4): 851-
60.

Calvete, J. J., C. Marcinkiewicz, et al. (2007). "Snake venomics of Bitis gabonica
gabonica. Protein family composition, subunit organization of venom
toxins, and characterization of dimeric disintegrins bitisgabonin-1 and
bitisgabonin-2." J Proteome Res 6(1): 326-36.

Caron, E., A. J. Self, et al. (2000). "The GTPase Rap1 controls functional
activation of macrophage integrin alphaMbeta2 by LPS and other
inflammatory mediators." Curr Biol 10(16): 974-8.

Carroll, R. C., E. C. Beattie, et al. (1999). "Dynamin-dependent endocytosis of
ionotropic glutamate receptors." Proc Natl Acad Sci U S A 96(24): 14112-
7.

Casartelli, M., G. Cermenati, et al. (2008). "A megalin-like receptor is involved in
protein endocytosis in the midgut of an insect (Bombyx mori,
Lepidoptera)." Am J Physiol Regul Integr Comp Physiol 295(4): R1290-
300.

Chambers, A. F. (2009). "MDA-MB-435 and M14 cell lines: identical but not M14
melanoma?" Cancer Res 69(13): 5292-3.

Chinenov, Y. and T. K. Kerppola (2001). "Close encounters of many kinds: Fos-
Jun interactions that mediate transcription regulatory specificity."
Oncogene 20(19): 2438-52.

Chiu, C. Y., P. Mathias, et al. (1999). "Structure of adenovirus complexed with its
internalization receptor, alphavbeta5 integrin." J Virol 73(8): 6759-68.

Contois, L., A. Akalu, et al. (2009). "Integrins as "functional hubs" in the
regulation of pathological angiogenesis." Semin Cancer Biol 19(5): 318-
28.

Cosen-Binker, L. I. and A. Kapus (2006). "Cortactin: the gray eminence of the
cytoskeleton." Physiology (Bethesda) 21: 352-61.
141

Critchley, D. R. (2009). "Biochemical and structural properties of the integrin-
associated cytoskeletal protein talin." Annu Rev Biophys 38: 235-54.

Dang, D. T., S. Y. Chun, et al. (2008). "Hypoxia-inducible factor-1 target genes
as indicators of tumor vessel response to vascular endothelial growth
factor inhibition." Cancer Res 68(6): 1872-80.

Dao, V. T., A. G. Dupuy, et al. (2009). "Dynamic changes in Rap1 activity are
required for cell retraction and spreading during mitosis." J Cell Sci 122(Pt
16): 2996-3004.

de Pereda, J. M., G. Wiche, et al. (1999). "Crystal structure of a tandem pair of
fibronectin type III domains from the cytoplasmic tail of integrin
alpha6beta4." Embo J 18(15): 4087-95.

De Wever, O. and M. Mareel (2003). "Role of tissue stroma in cancer cell
invasion." J Pathol 200(4): 429-47.

Dechantsreiter, M. A., E. Planker, et al. (1999). "N-Methylated cyclic RGD
peptides as highly active and selective alpha(V)beta(3) integrin
antagonists." J Med Chem 42(16): 3033-40.

del Rio, A., R. Perez-Jimenez, et al. (2009). "Stretching single talin rod molecules
activates vinculin binding." Science 323(5914): 638-41.

Deuretzbacher, A., N. Czymmeck, et al. (2009). "Beta1 integrin-dependent
engulfment of Yersinia enterocolitica by macrophages is coupled to the
activation of autophagy and suppressed by type III protein secretion." J
Immunol 183(9): 5847-60.

Diaz-Gonzalez, F., J. Forsyth, et al. (1996). "Trans-dominant inhibition of integrin
function." Mol Biol Cell 7(12): 1939-51.

Enis, D. R., B. R. Shepherd, et al. (2005). "Induction, differentiation, and
remodeling of blood vessels after transplantation of Bcl-2-transduced
endothelial cells." Proc Natl Acad Sci U S A 102(2): 425-30.

Folkman, J. (1997). Tumor Angiogenesis, in Cancer Medicine. Baltimore,
Williams & Wilkins.

Folkman, J. (2007). "Angiogenesis: an organizing principle for drug discovery?"
Nat Rev Drug Discov 6(4): 273-86.

142
Frelinger, A. L., 3rd, S. C. Lam, et al. (1988). "Occupancy of an adhesive
glycoprotein receptor modulates expression of an antigenic site involved in
cell adhesion." J Biol Chem 263(25): 12397-402.

Gao, B., T. M. Curtis, et al. (2000). "Increased recycling of (alpha)5(beta)1
integrins by lung endothelial cells in response to tumor necrosis factor." J
Cell Sci 113 Pt 2: 247-57.

Garcia-Alvarez, B., J. M. de Pereda, et al. (2003). "Structural determinants of
integrin recognition by talin." Mol Cell 11(1): 49-58.

Garmy-Susini, B. and J. A. Varner (2008). "Roles of integrins in tumor
angiogenesis and lymphangiogenesis." Lymphat Res Biol 6(3-4): 155-63.

Gingras, A. R., N. Bate, et al. (2008). "The structure of the C-terminal actin-
binding domain of talin." Embo J 27(2): 458-69.

Gingras, A. R., W. H. Ziegler, et al. (2005). "Mapping and consensus sequence
identification for multiple vinculin binding sites within the talin rod." J Biol
Chem 280(44): 37217-24.

Gold, H. K., T. Yasuda, et al. (1991). "Animal models for arterial thrombolysis and
prevention of reocclusion. Erythrocyte-rich versus platelet-rich thrombus."
Circulation 83(6 Suppl): IV26-40.

Golubkov, V., D. Hawes, et al. (2003). "Anti-angiogenic activity of contortrostatin,
a disintegrin from Agkistrodon contortrix contortrix snake venom."
Angiogenesis 6(3): 213-24.

Gould, R. J., M. A. Polokoff, et al. (1990). "Disintegrins: a family of integrin
inhibitory proteins from viper venoms." Proc Soc Exp Biol Med 195(2):
168-71.

Gritti, I., G. Banfi, et al. (2000). "Pepsinogens: physiology, pharmacology
pathophysiology and exercise." Pharmacol Res 41(3): 265-81.

Guan, J. L. (2010). "Integrin signaling through FAK in the regulation of mammary
stem cells and breast cancer." IUBMB Life.

Han, J., C. J. Lim, et al. (2006). "Reconstructing and deconstructing agonist-
induced activation of integrin alphaIIbbeta3." Curr Biol 16(18): 1796-806.

143

Han, S., F. R. Khuri, et al. (2006). "Fibronectin stimulates non-small cell lung
carcinoma cell growth through activation of Akt/mammalian target of
rapamycin/S6 kinase and inactivation of LKB1/AMP-activated protein
kinase signal pathways." Cancer Res 66(1): 315-23.

Hanahan, D. and R. A. Weinberg (2000). "The hallmarks of cancer." Cell 100(1):
57-70.

Himmel, M., A. Ritter, et al. (2009). "Control of high affinity interactions in the talin
C terminus: how talin domains coordinate protein dynamics in cell
adhesions." J Biol Chem 284(20): 13832-42.

Hodivala-Dilke, K. M., A. R. Reynolds, et al. (2003). "Integrins in angiogenesis:
multitalented molecules in a balancing act." Cell Tissue Res 314(1): 131-
44.

Honda, S., Y. Tomiyama, et al. (1995). "Topography of ligand-induced binding
sites, including a novel cation-sensitive epitope (AP5) at the amino
terminus, of the human integrin beta 3 subunit." J Biol Chem 270(20):
11947-54.

Hood, J. D. and D. A. Cheresh (2002). "Role of integrins in cell invasion and
migration." Nat Rev Cancer 2(2): 91-100.

Huang, J., J. S. Frischer, et al. (2003). "Regression of established tumors and
metastases by potent vascular endothelial growth factor blockade." Proc
Natl Acad Sci U S A 100(13): 7785-90.

Huang, T. F., C. H. Yeh, et al. (2001). "Viper venom components affecting
angiogenesis." Haemostasis 31(3-6): 192-206.

Humphries, M. J., E. J. Symonds, et al. (2003). "Mapping functional residues
onto integrin crystal structures." Curr Opin Struct Biol 13(2): 236-43.

Hynes, R. O. (1992). "Integrins: versatility, modulation, and signaling in cell
adhesion." Cell 69(1): 11-25.

Hynes, R. O. (2002). "Integrins: bidirectional, allosteric signaling machines." Cell
110(6): 673-87.

Hynes, R. O. (2004). "The emergence of integrins: a personal and historical
perspective." Matrix Biol 23(6): 333-40.

144
Johnson, M. S., N. Lu, et al. (2009). "Integrins during evolution: evolutionary
trees and model organisms." Biochim Biophys Acta 1788(4): 779-89.

Juarez, P., I. Comas, et al. (2008). "Evolution of snake venom disintegrins by
positive Darwinian selection." Mol Biol Evol 25(11): 2391-407.

Juarez, P., S. C. Wagstaff, et al. (2006). "Molecular cloning of Echis ocellatus
disintegrins reveals non-venom-secreted proteins and a pathway for the
evolution of ocellatusin." J Mol Evol 63(2): 183-93.

Kageyama, T. (2002). "Pepsinogens, progastricsins, and prochymosins:
structure, function, evolution, and development." Cell Mol Life Sci 59(2):
288-306.

Kang, I. C., Y. D. Lee, et al. (1999). "A novel disintegrin salmosin inhibits tumor
angiogenesis." Cancer Res 59(15): 3754-60.

Kim, S., K. Bell, et al. (2000). "Regulation of angiogenesis in vivo by ligation of
integrin alpha5beta1 with the central cell-binding domain of fibronectin."
Am J Pathol 156(4): 1345-62.

Kim, S., M. Harris, et al. (2000). "Regulation of integrin alpha vbeta 3-mediated
endothelial cell migration and angiogenesis by integrin alpha5beta1 and
protein kinase A." J Biol Chem 275(43): 33920-8.

Kinbara, K., L. E. Goldfinger, et al. (2003). "Ras GTPases: integrins' friends or
foes?" Nat Rev Mol Cell Biol 4(10): 767-76.

Kini, R. M. and H. J. Evans (1992). "Structural domains in venom proteins:
evidence that metalloproteinases and nonenzymatic platelet aggregation
inhibitors (disintegrins) from snake venoms are derived by proteolysis from
a common precursor." Toxicon 30(3): 265-93.

Kooistra, M. R., N. Dube, et al. (2007). "Rap1: a key regulator in cell-cell junction
formation." J Cell Sci 120(Pt 1): 17-22.

Kumar, C. C., L. Armstrong, et al. (2000). "Targeting integrins alpha v beta 3 and
alpha v beta 5 for blocking tumor-induced angiogenesis." Adv Exp Med
Biol 476: 169-80.

Lacroix, M. (2009). "MDA-MB-435 cells are from melanoma, not from breast
cancer." Cancer Chemother Pharmacol 63(3): 567.

145
Lafuente, E. M., A. A. van Puijenbroek, et al. (2004). "RIAM, an Ena/VASP and
Profilin ligand, interacts with Rap1-GTP and mediates Rap1-induced
adhesion." Dev Cell 7(4): 585-95.

Lee, H. S., C. J. Lim, et al. (2009). "RIAM activates integrins by linking talin to ras
GTPase membrane-targeting sequences." J Biol Chem 284(8): 5119-27.

Lee, J. O., L. A. Bankston, et al. (1995). "Two conformations of the integrin A-
domain (I-domain): a pathway for activation?" Structure 3(12): 1333-40.

Lekstrom-Himes, J. and K. G. Xanthopoulos (1998). "Biological role of the
CCAAT/enhancer-binding protein family of transcription factors." J Biol
Chem 273(44): 28545-8.

Li, F. and G. Sethi (2010). "Targeting transcription factor NF-kappaB to overcome
chemoresistance and radioresistance in cancer therapy." Biochim Biophys
Acta.

Luo, B. H., C. V. Carman, et al. (2007). "Structural basis of integrin regulation
and signaling." Annu Rev Immunol 25: 619-47.

Maginnis, M. S., J. C. Forrest, et al. (2006). "Beta1 integrin mediates
internalization of mammalian reovirus." J Virol 80(6): 2760-70.

Mahabeleshwar, G. H., W. Feng, et al. (2006). "Integrin signaling is critical for
pathological angiogenesis." J Exp Med 203(11): 2495-507.

Marcinkiewicz, C., P. H. Weinreb, et al. (2003). "Obtustatin: a potent selective
inhibitor of alpha1beta1 integrin in vitro and angiogenesis in vivo." Cancer
Res 63(9): 2020-3.

Markland, F. S., K. Shieh, et al. (2001). "A novel snake venom disintegrin that
inhibits human ovarian cancer dissemination and angiogenesis in an
orthotopic nude mouse model." Haemostasis 31(3-6): 183-91.

Matsuda, T., Y. Kido, et al. (2010). "Ablation of C/EBPbeta alleviates ER stress
and pancreatic beta cell failure through the GRP78 chaperone in mice." J
Clin Invest 120(1): 115-26.

May, C., J. F. Doody, et al. (2005). "Identification of a transiently exposed VE-
cadherin epitope that allows for specific targeting of an antibody to the
tumor neovasculature." Blood 105(11): 4337-44.

McLane, M. A., T. Joerger, et al. (2008). "Disintegrins in health and disease."
Front Biosci 13: 6617-37.
146

McLane, M. A., C. Marcinkiewicz, et al. (1998). "Viper venom disintegrins and
related molecules." Proc Soc Exp Biol Med 219(2): 109-19.

Mclane, M. A., Zhang, X., Tian, J., and Paquette-Straub, C. (2006). "Monomeric
and dimeric disintegrins: Platelet active agents from viper venom." Toxin
Rev 25: 435-464.

Mierke, C. T. (2009). "The role of vinculin in the regulation of the mechanical
properties of cells." Cell Biochem Biophys 53(3): 115-26.

Minea, R., S. Swenson, et al. (2005). "Development of a novel recombinant
disintegrin, contortrostatin, as an effective anti-tumor and anti-angiogenic
agent." Pathophysiol Haemost Thromb 34(4-5): 177-83.

Minea, R. O., C. M. Helchowski, et al. (2010). "Vicrostatin – An Anti-Invasive
Multi-Integrin Targeting Chimeric Disintegrin with Tumor Anti-Angiogenic
and Pro-Apoptotic Activities." PLoS ONE 6(5): e10929.

Mitra, S. K., D. Mikolon, et al. (2006). "Intrinsic FAK activity and Y925
phosphorylation facilitate an angiogenic switch in tumors." Oncogene
25(44): 5969-84.

Mitra, S. K. and D. D. Schlaepfer (2006). "Integrin-regulated FAK-Src signaling in
normal and cancer cells." Curr Opin Cell Biol 18(5): 516-23.

Moiseeva, N., R. Bau, et al. (2008). "Structure of acostatin, a dimeric disintegrin
from Southern copperhead (Agkistrodon contortrix contortrix), at 1.7 A
resolution." Acta Crystallogr D Biol Crystallogr 64(Pt 4): 466-70.

Moiseeva, N., S. D. Swenson, et al. (2002). "Purification, crystallization and
preliminary X-ray analysis of the disintegrin contortrostatin from
Agkistrodon contortrix contortrix snake venom." Acta Crystallogr D Biol
Crystallogr 58(Pt 12): 2122-4.

Mould, A. P., J. A. Askari, et al. (2002). "Integrin activation involves a
conformational change in the alpha 1 helix of the beta subunit A-domain."
J Biol Chem 277(22): 19800-5.

Nemerow, G. R. (2009). "A new link between virus cell entry and inflammation:
adenovirus interaction with integrins induces specific proinflammatory
responses." Mol Ther 17(9): 1490-1.

Niessen, C. M. (2007). "Tight junctions/adherens junctions: basic structure and
function." J Invest Dermatol 127(11): 2525-32.
147

Niewiarowski, S., M. A. McLane, et al. (1994). "Disintegrins and other naturally
occurring antagonists of platelet fibrinogen receptors." Semin Hematol
31(4): 289-300.

Nisato, R. E., J. C. Tille, et al. (2003). "alphav beta 3 and alphav beta 5 integrin
antagonists inhibit angiogenesis in vitro." Angiogenesis 6(2): 105-19.

Okuda, D., H. Koike, et al. (2002). "A new gene structure of the disintegrin family:
a subunit of dimeric disintegrin has a short coding region." Biochemistry
41(48): 14248-54.

Pagano, A., P. Crottet, et al. (2004). "In vitro formation of recycling vesicles from
endosomes requires adaptor protein-1/clathrin and is regulated by rab4
and the connector rabaptin-5." Mol Biol Cell 15(11): 4990-5000.

Papagrigoriou, E., A. R. Gingras, et al. (2004). "Activation of a vinculin-binding
site in the talin rod involves rearrangement of a five-helix bundle." Embo J
23(15): 2942-51.

Passegue, E. and E. F. Wagner (2000). "JunB suppresses cell proliferation by
transcriptional activation of p16(INK4a) expression." Embo J 19(12): 2969-
79.

Pfaff, M., K. Tangemann, et al. (1994). "Selective recognition of cyclic RGD
peptides of NMR defined conformation by alpha IIb beta 3, alpha V beta 3,
and alpha 5 beta 1 integrins." J Biol Chem 269(32): 20233-8.

Piechaczyk, M. and R. Farras (2008). "Regulation and function of JunB in cell
proliferation." Biochem Soc Trans 36(Pt 5): 864-7.

Pierschbacher, M. D. and E. Ruoslahti (1984). "Cell attachment activity of
fibronectin can be duplicated by small synthetic fragments of the
molecule." Nature 309(5963): 30-3.

Plow, E. F., M. D. Pierschbacher, et al. (1985). "The effect of Arg-Gly-Asp-
containing peptides on fibrinogen and von Willebrand factor binding to
platelets." Proc Natl Acad Sci U S A 82(23): 8057-61.

Ponting, C. P., J. Schultz, et al. (2000). "Evolution of domain families." Adv
Protein Chem 54: 185-244.

Ramji, D. P. and P. Foka (2002). "CCAAT/enhancer-binding proteins: structure,
function and regulation." Biochem J 365(Pt 3): 561-75.

148
Reardon, D. A., K. L. Fink, et al. (2008). "Randomized phase II study of
cilengitide, an integrin-targeting arginine-glycine-aspartic acid peptide, in
recurrent glioblastoma multiforme." J Clin Oncol 26(34): 5610-7.

Rees, D. J., S. E. Ades, et al. (1990). "Sequence and domain structure of talin."
Nature 347(6294): 685-9.

Ren, G., M. S. Crampton, et al. (2009). "Cortactin: Coordinating adhesion and the
actin cytoskeleton at cellular protrusions." Cell Motil Cytoskeleton 66(10):
865-73.

Reyes-Reyes, M., N. Mora, et al. (2002). "beta1 and beta2 integrins activate
different signalling pathways in monocytes." Biochem J 363(Pt 2): 273-80.

Ritter, M. R. and F. S. Markland, Jr. (2000). "Contortrostatin activates ERK2 and
tyrosine phosphorylation events via distinct pathways." Biochem Biophys
Res Commun 274(1): 142-8.

Ritter, M. R., Q. Zhou, et al. (2000). "Contortrostatin, a snake venom disintegrin,
induces alphavbeta3-mediated tyrosine phosphorylation of CAS and FAK
in tumor cells." J Cell Biochem 79(1): 28-37.

Ritter, M. R., Q. Zhou, et al. (2001). "Contortrostatin, a homodimeric disintegrin,
actively disrupts focal adhesion and cytoskeletal structure and inhibits cell
motility through a novel mechanism." Cell Commun Adhes 8(2): 71-86.

Samarzija, I., P. Sini, et al. (2009). "Wnt3a regulates proliferation and migration
of HUVEC via canonical and non-canonical Wnt signaling pathways."
Biochem Biophys Res Commun 386(3): 449-54.

Sanchez, E. E., J. A. Galan, et al. (2006). "Isolation and characterization of two
disintegrins inhibiting ADP-induced human platelet aggregation from the
venom of Crotalus scutulatus scutulatus (Mohave Rattlesnake)." Toxicol
Appl Pharmacol 212(1): 59-68.

Saudek, V., R. A. Atkinson, et al. (1991). "Three-dimensional structure of
echistatin, the smallest active RGD protein." Biochemistry 30(30): 7369-
72.

Savage, B., U. M. Marzec, et al. (1990). "Binding of the snake venom-derived
proteins applaggin and echistatin to the arginine-glycine-aspartic acid
recognition site(s) on platelet glycoprotein IIb.IIIa complex inhibits receptor
function." J Biol Chem 265(20): 11766-72.

149
Scarborough, R. M., M. A. Naughton, et al. (1993). "Design of potent and specific
integrin antagonists. Peptide antagonists with high specificity for
glycoprotein IIb-IIIa." J Biol Chem 268(2): 1066-73.

Scarborough, R. M., J. W. Rose, et al. (1991). "Barbourin. A GPIIb-IIIa-specific
integrin antagonist from the venom of Sistrurus m. barbouri." J Biol Chem
266(15): 9359-62.

Schmitmeier, S., F. S. Markland, et al. (2000). "Anti-invasive effect of
contortrostatin, a snake venom disintegrin, and TNF-alpha on malignant
glioma cells." Anticancer Res 20(6B): 4227-33.

Schmitmeier, S., F. S. Markland, et al. (2005). "Potent mimicry of fibronectin-
induced intracellular signaling in glioma cells by the homodimeric snake
venom disintegrin contortrostatin." Neurosurgery 57(1): 141-53; discussion
141-53.

Semenza, G. L. (2003). "Targeting HIF-1 for cancer therapy." Nat Rev Cancer
3(10): 721-32.

Senn, H. and W. Klaus (1993). "The nuclear magnetic resonance solution
structure of flavoridin, an antagonist of the platelet GP IIb-IIIa receptor." J
Mol Biol 232(3): 907-25.

Sever, S., H. Damke, et al. (2000). "Dynamin:GTP controls the formation of
constricted coated pits, the rate limiting step in clathrin-mediated
endocytosis." J Cell Biol 150(5): 1137-48.

Shan, Y., L. Yu, et al. (2009). "Nudel and FAK as antagonizing strength
modulators of nascent adhesions through paxillin." PLoS Biol 7(5):
e1000116.

Shebuski, R. J., D. R. Ramjit, et al. (1990). "Prevention of canine coronary artery
thrombosis with echistatin, a potent inhibitor of platelet aggregation from
the venom of the viper, Echis carinatus." Thromb Haemost 64(4): 576-81.

Smith, K. J., M. Jaseja, et al. (1996). "Three-dimensional structure of the RGD-
containing snake toxin albolabrin in solution, based on 1H NMR
spectroscopy and simulated annealing calculations." Int J Pept Protein
Res 48(3): 220-8.

Soszka, T., K. A. Knudsen, et al. (1991). "Inhibition of murine melanoma cell-
matrix adhesion and experimental metastasis by albolabrin, an RGD-
containing peptide isolated from the venom of Trimeresurus albolabris."
Exp Cell Res 196(1): 6-12.
150

Springer, T. A. (1997). "Folding of the N-terminal, ligand-binding region of integrin
alpha-subunits into a beta-propeller domain." Proc Natl Acad Sci U S A
94(1): 65-72.

Staniszewska, I., E. M. Walsh, et al. (2009). "Effect of VP12 and viperistatin on
inhibition of collagen receptors-dependent melanoma metastasis." Cancer
Biol Ther 8(15): 1507-16.

Stewart, P. L. and G. R. Nemerow (2007). "Cell integrins: commonly used
receptors for diverse viral pathogens." Trends Microbiol 15(11): 500-7.

Swenson, S., F. Costa, et al. (2005). "Contortrostatin, a snake venom disintegrin
with anti-angiogenic and anti-tumor activity." Pathophysiol Haemost
Thromb 34(4-5): 169-76.

Swenson, S., F. Costa, et al. (2004). "Intravenous liposomal delivery of the snake
venom disintegrin contortrostatin limits breast cancer progression." Mol
Cancer Ther 3(4): 499-511.

Tadokoro, S., S. J. Shattil, et al. (2003). "Talin binding to integrin beta tails: a final
common step in integrin activation." Science 302(5642): 103-6.

Takada, Y., X. Ye, et al. (2007). "The integrins." Genome Biol 8(5): 215.

Takagi, J. and T. A. Springer (2002). "Integrin activation and structural
rearrangement." Immunol Rev 186: 141-63.

Tamkun, J. W., D. W. DeSimone, et al. (1986). "Structure of integrin, a
glycoprotein involved in the transmembrane linkage between fibronectin
and actin." Cell 46(2): 271-82.

Tian, J., C. Paquette-Straub, et al. (2007). "Inhibition of melanoma cell motility by
the snake venom disintegrin eristostatin." Toxicon 49(7): 899-908.

Trikha, M., Y. A. De Clerck, et al. (1994). "Contortrostatin, a snake venom
disintegrin, inhibits beta 1 integrin-mediated human metastatic melanoma
cell adhesion and blocks experimental metastasis." Cancer Res 54(18):
4993-8.

Trikha, M., W. E. Rote, et al. (1994). "Purification and characterization of platelet
aggregation inhibitors from snake venoms." Thromb Res 73(1): 39-52.
Varner, J. A. (1997). "The role of vascular cell integrins alpha v beta 3 and alpha
v beta 5 in angiogenesis." Exs 79: 361-90.

151
Vellon, L., J. A. Menendez, et al. (2005). "AlphaVbeta3 integrin regulates
heregulin (HRG)-induced cell proliferation and survival in breast cancer."
Oncogene 24(23): 3759-73.

Wegener, K. L., A. W. Partridge, et al. (2007). "Structural basis of integrin
activation by talin." Cell 128(1): 171-82.

Wickham, T. J., P. Mathias, et al. (1993). "Integrins alpha v beta 3 and alpha v
beta 5 promote adenovirus internalization but not virus attachment." Cell
73(2): 309-19.

Wu, C. (2007). "Focal adhesion: a focal point in current cell biology and
molecular medicine." Cell Adh Migr 1(1): 13-8.

Xiong, J. P., T. Stehle, et al. (2002). "Crystal structure of the extracellular
segment of integrin alpha Vbeta3 in complex with an Arg-Gly-Asp ligand."
Science 296(5565): 151-5.

Yan, B., D. A. Calderwood, et al. (2001). "Calpain cleavage promotes talin
binding to the beta 3 integrin cytoplasmic domain." J Biol Chem 276(30):
28164-70.

Yang, R. S., C. H. Tang, et al. (2005). "Inhibition of tumor formation by snake
venom disintegrin." Toxicon 45(5): 661-9.

Yednock, T. A., C. Cannon, et al. (1995). "Alpha 4 beta 1 integrin-dependent cell
adhesion is regulated by a low affinity receptor pool that is
conformationally responsive to ligand." J Biol Chem 270(48): 28740-50.

Ylipaasto, P., M. Eskelinen, et al. (2009). "Vitronectin receptors, {alpha}V-
integrins, are recognized by several non-RGD containing echoviruses in a
continuous laboratory cell line and also in primary human Langerhans'
islets and endothelial cells." J Gen Virol.

Zamir, E. and B. Geiger (2001). "Molecular complexity and dynamics of cell-
matrix adhesions." J Cell Sci 114(Pt 20): 3583-90.

Zang, Q. and T. A. Springer (2001). "Amino acid residues in the PSI domain and
cysteine-rich repeats of the integrin beta2 subunit that restrain activation
of the integrin alpha(X)beta(2)." J Biol Chem 276(10): 6922-9.

Zhou, Q., P. Hu, et al. (2000). "Molecular cloning and functional expression of
contortrostatin, a homodimeric disintegrin from southern copperhead
snake venom." Arch Biochem Biophys 375(2): 278-88.

152
Zhou, Q., M. T. Nakada, et al. (1999). "Contortrostatin, a dimeric disintegrin from
Agkistrodon contortrix contortrix, inhibits angiogenesis." Angiogenesis
3(3): 259-69.

Zhou, Q., M. T. Nakada, et al. (2000). "Contortrostatin, a homodimeric
disintegrin, binds to integrin alphavbeta5." Biochem Biophys Res Commun
267(1): 350-5.

Zhou, Q., R. P. Sherwin, et al. (2000). "Contortrostatin, a dimeric disintegrin from
Agkistrodon contortrix contortrix, inhibits breast cancer progression."
Breast Cancer Res Treat 61(3): 249-60.

Abstract (if available)

Abstract Integrin mediated signaling is vital to several cellular pathways. These pathways include the cell’s ability to migrate and invade through the extracellular matrix, which is fundamental to more complex processes like angiogenesis and cancer progression. Our ability to understand, manipulate, or even control these events could have drastic implications for the field of cell biology. Disintegrins act as integrin antagonists with the ability to disrupt adhesion based integrin functions through direct binding. However, not much is know about how these molecules initiate signaling events downstream after ligation to integrins. Contortrostatin (CN), a homodimeric disintegrin isolated from southern copperhead venom has been studied by the Markland laboratory for over fifteen years. Although some initial studies were done on CN signaling, CN’s anti-cancer ability has been the focus of the Markland laboratory since.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Effect of vicrostatin on integrin based signaling molecules in cancer

PDF

In vivo study with a novel liposomal formulation of a recombinant ADAM disintegrin-like domain showing both anti-angiogenic and tumor growth inhibition activities

PDF

Quantitative evaluation of the effectiveness of an integrin antagonist, vicrostatin, in cell migration

PDF

Effect of vicrostatin (an integrin based therapy) on canine osteosarcoma

PDF

Structural characterization of pseudorepeat shuffled alpha-synuclein

PDF

The folding, misfolding and refolding of membrane proteins and the design of a phosphatidylserine-specific membrane sensor

PDF

The cancer stem-like phenotype: therapeutics, phenotypic plasticity and mechanistic studies

PDF

The mechanism of R-loop formation in mammalian immunoglobulin class switch recombination

PDF

Calorie restriction reduces liver steatosis in liver specific Pten deleted mouse model

PDF

Dual effects of transmembrane proline residues on integrin function

PDF

Molecular basis of CD33 receptor function, and effects of a destabilizing transmembrane motif on αXβ2 integrin

PDF

Studies on the role of a novel protein, TMEM 56 in tumorigenic growth for MCF-7 cells

PDF

The role of CD1d transmembrane and cytoplasmic tail domain in CD1d trafficking pathway

PDF

Characterization and functional study of a novel human protein SFMBT, and PR-Set7 histone methyltransferase

PDF

Study of integrin αIIbβ3 transmembrane and cytosolic domain interaction and the expression of extracellular domain by Pichia Pastoris

PDF

Harnessing the power of stem cell self-renewal pathways in cancer: dissecting the role of BMI-1 in Ewing’s sarcoma initiation and maintenance

PDF

Membrane curvature sensors and inducers studied by site-directed spin labeling

PDF

The structure and function of membrane curving proteins on different membrane shapes and their regulation by post-translational modifications

PDF

Role of GRP78 and its novel cytosolic isoform GRP78va in regulating the unfolded protein response in cancer

PDF

Studies on the expression and function of the human TMEM56 protein

Asset Metadata

Creator Helchowski, Corey Michael (author)

Core Title Mechanistic studies of the disintegrin contortrostatin and characterization of the recombinant protein vicrostatin

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Biochemistry and Molecular Biology

Degree Conferral Date 2010-08

Publication Date 07/14/2010

Defense Date 03/26/2010

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag disintegrin,integrin,internalization,OAI-PMH Harvest,signaling,talin

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Markland, Francis S., Jr (committee chair), Hamm-Alvarez, Sarah F. (committee member), Ulmer, Tobias (committee member)

Creator Email cmhelchowski@gmail.com,corey8316@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3191

Unique identifier UC1210264

Identifier etd-Helchowski-3650 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-354551 (legacy record id),usctheses-m3191 (legacy record id)

Legacy Identifier etd-Helchowski-3650.pdf

Dmrecord 354551

Document Type Dissertation

Rights Helchowski, Corey Michael

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu