THE ROLE OF INTERFERON GAMMA IN THE
STROMA OF GROWING AND REGRESSING
TUMOURS
vorgelegt von
BSc (Honours)
Felicia Pradera Elsley
Aus Perth
von der Fakultät III – Prozesswissenschaften
der Technischen Universität Berlin
zur Erlangung des akademischen Grades
Doktor der Naturwissenschaften
-Dr. rer. nat.-
genehmigte Dissertation
Promotionsausschuss
Vorsitzender..Prof. Dr. Roland. Tressl
Gutachter: Prof. Dr. Ulf Stahl
Gutachter: Prof. Dr. Roland Lauster
Tag der Aussprache: 21/09/06
Berlin 2006
D83
TABLE OF CONTENTS
TABLE OF CONTENTS
1 INTRODUCTION............................................................................................................ 1
1.1 Tumour Stroma............................................................................................................ 1
1.1.1 Extracellular Components of the Tumour Stroma ..................................................... 1
1.1.2 Molecules making up the Basement Membrane Associated Tumour Stroma ........... 2
A Proteoglycans (PG) .................................................................................................... 3
B Glycosaminoglycans (GAGs) ....................................................................................4
C Heparan Sulphate GAG Chains.................................................................................. 5
1.1.3 Cells within the Tumour Stroma................................................................................6
A Fibroblasts.................................................................................................................. 6
B Endothelial Cells ........................................................................................................ 7
C Haematopoietic Cells ................................................................................................. 7
1.1.4 Cytokines and Chemokines within the Tumour Stroma ............................................ 7
1.2 The Protein Interferon Gamma (IFN-γ) .................................................................... 8
1.2.1 Structure ..................................................................................................................... 8
1.2.2 IFN-γ Signal Transduction......................................................................................... 9
1.2.3 Interaction with the Matrix......................................................................................... 9
1.2.4 Role of IFN-γ in Tumours........................................................................................12
1.3 Tumour Therapy........................................................................................................ 13
1.3.1 Cyclophosphamide................................................................................................... 14
1.4 Aims............................................................................................................................. 15
2 MATERIALS AND METHODS................................................................................... 17
2.1 Materials ..................................................................................................................... 17
2.1.1 Cell Lines .................................................................................................................17
2.1.2 Mice.......................................................................................................................... 18
2.1.3 Peptide......................................................................................................................18
2.1.4 Oligonucleotides....................................................................................................... 18
A PCR Primers.............................................................................................................18
B Inserts ....................................................................................................................... 19
2.1.5 Reagents ................................................................................................................... 20
A Chemicals................................................................................................................. 20
B Kits ........................................................................................................................... 22
C Buffers......................................................................................................................23
D Antibodies ................................................................................................................24
E Equipment ................................................................................................................ 24
F Machinery................................................................................................................. 25
2.2 Methods....................................................................................................................... 27
2.2.1 Cell Culture .............................................................................................................. 27
A Incubation................................................................................................................. 27
B Adherent cell culture................................................................................................ 27
C Suspension cell culture............................................................................................. 27
I
TABLE OF CONTENTS
D Counting and viability..............................................................................................27
E Freezing and thawing ............................................................................................... 28
F Antibody production ................................................................................................28
2.2.2 In Vivo Experiments................................................................................................. 28
A Tumour cell injection and in vivo treatment............................................................. 28
B Neutralisation ........................................................................................................... 29
C Bone marrow chimeras (BMC)................................................................................ 29
D In vivo Cincinnati cytokine capture assay................................................................ 29
2.2.3 Immunohistochemistry.............................................................................................30
A Light Microscopy..................................................................................................... 30
B Fluorescent microscopy............................................................................................ 31
2.2.4 Laser Scanning Confocal Microscopy ..................................................................... 31
2.2.5 In Vivo Matrigel Binding Assay............................................................................... 31
2.2.6 Enzyme Linked Immunosorbent Assay (ELISA) ....................................................32
A IFN−γ ELISA ........................................................................................................... 32
2.2.7 Construction of Tumour cells expressing MC-2 peptide ......................................... 32
A Plasmid construction ................................................................................................ 32
B Fragment determination ...........................................................................................33
C Blunting....................................................................................................................33
D Dephosphorylation ................................................................................................... 34
E Ligation ....................................................................................................................34
F Transformation......................................................................................................... 34
2.2.8 Plasmid Preparations................................................................................................35
A Mini and Maxi preparations..................................................................................... 35
B Enzyme digestion..................................................................................................... 35
C Oligonucleotide annealing and insertion into pSecTag2-EF1α...............................35
D PCR of MC-2 insert.................................................................................................. 37
E Sequencing ............................................................................................................... 37
2.2.9 Tumour Cell Transfection........................................................................................ 37
2.2.10 Isolation of Genomic DNA .................................................................................. 38
2.2.11 Isolation of RNA.................................................................................................. 38
2.2.12 RT-PCR................................................................................................................ 39
2.2.13 Protein Determination .......................................................................................... 39
2.2.14 Iodine125 Labelled IFN-γ Experiments................................................................. 40
A Procedure for protein iodination and purification....................................................40
B Measuring the counts and quantity of radioactive protein....................................... 41
C Immobilisation of collagens and matrigel................................................................41
D Animals treated with I125-IFN-γ............................................................................... 42
E Measurement of the organ distribution of I125-IFN-γ...............................................42
3 RESULTS........................................................................................................................ 43
3.1 Genotyping Interferon Gamma Knockout Mice..................................................... 43
3.2 The Effect of Interferon Gamma on Tumour Growth ........................................... 44
3.3 Cyclophosphamide Induces IFN-γ Dependent Rejection of J558L Tumours....... 45
3.3.1 Wild type mice but not IFN-γ knockouts can reject J558L tumours........................ 45
3.3.2 Neutralisation of IFN-γ production in wild type mice effects tumour rejection......48
3.3.3 IFN-γ is released into the serum after cyclophosphamide treatment ....................... 50
3.3.4 Haematopoietic cells produce IFN-γ required for tumour rejection ........................ 51
II
TABLE OF CONTENTS
3.4 IFN-γ Co-localises to Heparan Sulphate Proteoglycan In Vivo and Binds Heparan
Sulphate Glycosaminoglycan In Vitro ...................................................................... 54
3.4.1 IFN-γ can co-localise with heparan sulphate in growing tumours........................... 54
3.4.2 IFN-γ influences the matrix production of tumours................................................. 59
3.4.3 Exogenous human IFN-γ can bind to matrigel plugs in vivo ................................... 63
3.5 The KRKRS Sequence of the Cytokine IFN-γ Influences In Vitro and In Vivo
Responses. ................................................................................................................... 64
3.5.1 Iodination of IFN-γ................................................................................................... 64
3.5.2 In Vitro inhibition.....................................................................................................65
A Radiolabelled IFN-γ can bind to matrix................................................................... 65
B The murine peptide can prevent binding of IFN-γ to matrigel................................. 66
C MC-2 peptide in the supernatants from transfected tumour cells can inhibit IFN-γ
binding......................................................................................................................67
3.5.3 In Vivo inhibition...................................................................................................... 68
A Radiolabelled IFN-γ can bind to matrix in vivo ....................................................... 68
B Human I125 IFN-γ binds less in MC-2 transfected tumours...................................... 69
C Transfected tumour response to cyclophospamide treatment .................................. 70
4 DISCUSSION ................................................................................................................. 72
4.1 The Origin of Interferon Gamma in Tumour Responses....................................... 72
4.2 The Binding and Influence of Interferon Gamma on the Tumour Extracellular
Matrix..........................................................................................................................75
4.3 The Blockade of Heparan Sulphate and Interferon Gamma Interactions within
the Tumour Stroma.................................................................................................... 79
5 SUMMARY..................................................................................................................... 83
6 ZUSAMMENFASUNG.................................................................................................. 85
7 ADDITIONAL ABBREVIATIONS ............................................................................. 87
8 REFERENCES............................................................................................................... 89
9 CURRICULUM VITAE................................................................................................ 95
10 PUBLICATIONS ........................................................................................................... 96
11 ACKNOWLEDGEMENTS........................................................................................... 97
12 APPENDIX ..................................................................................................................... 98
III
INTRODUCTION
1 INTRODUCTION
1.1 TUMOUR STROMA
The microenvironment, which encompasses the tumour body, is termed the tumour stroma
and its role in malignancy is poorly understood (Seljelid et al., 1999). All solid tumours are
made up of cancer cells and stroma. Tumour cells can modify the resident stroma by altering
the surrounding connective tissue and modulating the metabolism of the resident cells, thus
resulting in production of a stroma convenient for the tumour cells instead of maintaining the
physiological composition of the tissue. In addition to a variety of extracellular matrix
components, the stroma contains a rich cellular population, which includes fibroblasts that
provide the connective tissue framework for adipose, vasculature, resident immune cells, and
a milieu of cytokines and growth factors (Figure 1) (Pupa et al., 2002). Therefore, tumour
stroma is composed of a variety of normal cell types which appear to be actively recruited by
the tumour, for example, to provide blood supply (O'Reilly et al., 1994).
1.1.1 EXTRACELLULAR COMPONENTS OF THE TUMOUR
STROMA
A network of interacting extracellular macromolecules that constitute the extracellular matrix
(ECM) surrounds most cells in multicellular organisms. These proteins and polysaccharides
serve as ‘biological glue’, and are secreted locally into an organised meshwork within the
extracellular space of most tissues. They have the capability to form highly specialised
structures such as basal laminae, tendons and cartilage; and are an essential component of the
tumour stroma.
In recent years, it has become clear that the extracellular matrix plays an active and complex
role in regulating the behaviour of cells. The two main domains of extracellular matrix are the
basement membrane and the interstitial matrix, while the two primary classes of extracellular
macromolecules are collagens and glycosaminoglycans. Communication between the matrix
and cells are conducted through matrix receptors of which integrins and cadherins are the
most important classes. Additionally, almost all classes of matrix molecules are involved in
the control of proliferation, differentiation, and motion. Each tumour type has a different
composition of these molecules providing a stable microenvironment for tumour growth.
1
INTRODUCTION
collagen
HSPG
lymphocytes
macrophage
Basement Membrane
Tumour cells
Vessel-
extravasation
monocyte
fibroblast
Figure 1: Tumour Stroma
The tumour stroma consists of a variety of cellular and
extracellular molecules which mediate cell-cell and cell-
matrix interactions. The invasion of tumour cells requires
that the natural barriers between the various
compartments, such as basement membranes consisting of
heparan sulphate proteoglycan (HSPG) and collagen, are
overcome to facilitate the passage of tumour cells and
enable them to intravasate and generate distant metastasis.
Adapted from (Zigrino et al., 2005).
1.1.2 MOLECULES MAKING UP THE BASEMENT
MEMBRANE ASSOCIATED TUMOUR STROMA
Basement membranes constitute a specific compartment of the extracellular matrix, which
occur as thin sheet-like structures located in the adherent cell microenvironment, at the basal
side of a cell or as a pericellular envelope. These matrices are mainly composed of collagen
type IV, the glycoproteins laminin and entactin/nidogen and proteoglycans. Basement
membranes provide anchoring cell support, divide tissue compartments by forming selective
barriers, may control the access of regulatory molecules such as growth factors to the cell
surface, and can act as solid-phase regulators of growth and differentiation (Lortat-Jacob and
Grimaud, 1992).
2
INTRODUCTION
The basement membrane represents a barrier to invasive tumour cell growth, separating the
epithelium from connective tissue and the vascular endothelium. Remodelling or loss of the
basement membranes is required for tumour cells to reach vessels and thus allow for access
distant organs. This involves upregulation of the various matrix metalloproteinases (MMPs)
that act on the ECM (Kalluri, 2003).
A PROTEOGLYCANS (PG)
Proteoglycans are the most highly charged components of the basement membrane. They
contain 90-95% carbohydrate by weight, compared to the 1-60% for other glycoproteins.
Virtually all mammalian cells produce proteoglycans and secrete them into the ECM, insert
them into the membrane or store them in secretory granules.
In the panoply of signals from the extracellular matrix, proteoglycans appear to play a
predominant role. Proteoglycans can be grouped into several families based on their protein
core design (Iozzo, 1998). There are three main family members:
1. Lecticans:
Stimulate the proliferation of fibroblasts and chondrocytes.
2. Small Leucine Rich Repeats (SLRP’s):
Primary organisers of collagen networks, involved in signal transduction and
modulation/differentiation of epithelial and endothelial cells.
3. Heparan Sulphate Proteoglycans (HSPG):
Effects cell adhesion and migration, proliferation and differentiation. Binds cytokines
and growth factors. Involved in pathological processes.
Proteoglycans are continuously modified due to cleavage caused by heparanases,
exoglycosidases and sulphatases. They can also regulate enzyme activity by binding ligands
(such as various cytokines and chemokines) to their specific receptors, and protecting proteins
from degradation. Some of these proposed functions are likely to depend on a large
polyanionic domain supported by the glycosaminoglycan (GAG) component.
3
INTRODUCTION
B GLYCOSAMINOGLYCANS (GAGS)
Glycosaminoglycans are a group of negatively charged molecules present in many tissue
components of the extracellular matrix, basement and cellular membranes. Most GAGs exist
naturally as non-associated forms or as covalent complexes with the core proteins,
proteoglycans. Glycosaminoglycans assume extended structures in aqueous solution because
of their strong hydrophilic nature based on extensive sulphation, which is further exaggerated
when they are covalently linked to core proteins. They hold a large number of water
molecules in their domain and occupy hydrodynamic space in solution.
Figure 2: General schematic representation of
proteoglycan structure.
The core protein (thick line) undergoes post-translational
modification by covalently attached one or more (braces)
glycosaminoglycan chains (dotted lines, GAG) via the
linkage region (thick dots). Insert shows the general
structure of the disaccharide units in GAGs. The protein
core may also be modified by sulphation, phosphorylation,
myristilation or glycosylation of the amino acid chain
(γ). (Wegrowski and Maquart, 2004)
GAGs are linear polysaccharides whose building blocks consist of an amino sugar and an
uronic acid (Figure 2). Binding of proteins to the GAG chains of proteoglycans can result in
immobilisation of the respective proteins at their sites of production for matrix storage and
future mobilisation in response to immunological and environmental stimuli. As a result, the
interaction between GAGs and proteins can have physiological effects on cell growth,
migration and development, haemostasis, lipid transport and absorption.
4
INTRODUCTION
Within the tumour stroma and the tumour fibrotic tissue there are often higher proportions of
proteoglycans than in normal tissues. While the stromal proteoglycans of the SLRP family
possess antiproliferative properties, the GAGs liberated after PG degradation promote cancer
cell migration (Wegrowski and Maquart, 2004).
C HEPARAN SULPHATE GAG CHAINS
Heparan sulphate proteoglycan is a component of all basement membranes. Heparan sulphate
(HS), is a GAG composed of alternating sequences of glucosamine and either glucuronic or
iduronic acid (Gallagher and Walker, 1985). It is structurally related to heparin which is
produced mainly by mast cells. HS arises from the same biosynthetic pathway, but heparin is
more heavily sulphated and contains a greater amount of iduronic acid.
HS chains are characterised by complex sulphation patterns resulting in distinct protein
binding domains (Qiao et al., 2003). The HS side chains have been found to be involved in
the sequestration of heparin-binding growth factors (Noonan et al., 1991). In part this is due to
the molecular impermeability of these structures, imparting anionic charges across the
basement membrane.
HS molecules are ubiquitous in animal tissues where they function as ligands that are
involved in the regulation of the proteins that they bind (Lortat-Jacob et al., 2002). Cells have
the ability to rapidly adjust their HS in response to a changing microenvironment (Nurcombe
et al., 1993). In vivo the vast majority of HS exists in covalent linkage to core proteins, the
HSPGs. HSPGs can be divided into the cell surface forms (syndecans and glypicans) and
secreted extracellular matrix forms (e.g. perlecan). There are divergent reports suggesting that
the role of HSPGs (either stimulatory or inhibitory) depends on the core protein and its
association with a cell type (Qiao et al., 2003).
HSPGs at the tumour cell surface can actively modulate the tumourigenic process by
regulating autocrine signalling loops that lead to unregulated cell growth. They can also
influence how an organism responds to a growing tumour, including the recruitment of cells
of the immune system to the tumour site, the formation of a fibrin shell around the tumour
that acts as a protective barrier and the development of new blood vessels to the site of the
growing tumour (Sasisekharan et al., 2002).
5
INTRODUCTION
There are numerous biological effects of factors bound to HSPGs.
• Fibroblast growth factor (FGF) can result in receptor specific oligomerisation and in
cancer HSPGs bind FGF, thereby acting as a storage depot and activating an autocrine
signalling loop (Iozzo and San Antonio, 2001).
• Vascular endothelial growth factor (VEGF) bound to HSPGs results in low affinity
receptor interaction, which can block angiogenesis in cancer (Jiang and Couchman,
2003).
• Interleukin 8 (IL-8) can stimulate growth factor sequestration, which in turn can
modulate the host immune response to tumour cells (Capila and Lindhardt, 2002).
1.1.3 CELLS WITHIN THE TUMOUR STROMA
A FIBROBLASTS
The term fibroblast is ascribed to an heterogeneous multifunctional population of cells that
play a role in wound healing and developmental processes. They can be described as
connective tissue stem cells, matrix and protein synthesising cells (i.e. fibrocytes), contractile
cells (i.e. myofibroblasts) and, in some instances, tissue phagocytic cells (i.e. histiocytes)
(Silzle et al., 2004).
Fibroblasts are a relevant source of ECM and ECM-modulating molecules in tumours,
suggesting that tumour-associated fibroblast (TAF) alterations in ECM composition
profoundly impact the recruitment and function of immune cells (Silzle et al., 2004). They are
also capable of producing a number of paracrine immune modulators such as peptide growth
factors, cytokines, chemokines and inflammatory mediators. These modulators can contribute
to the development of transformed cells and the formation of tumour mass (Elenbaas and
Weinberg, 2001).
The production of MMPs by fibroblasts, enables tumour cells to cross structural barriers and
metastasise into distant organs, as well as regulate angiogenesis. These proteinases are also
responsible for the release of growth factors from the ECM such as FGF (Klein et al., 2004),
transforming growth factor-beta (TGF-β1) (Stamenkovic, 2003), and platelet derived growth
factor (PDGF) (Kalluri, 2003), which influence tumour development and rejection.
6
INTRODUCTION
B ENDOTHELIAL CELLS
The endothelium is a single layer of flattened, polygonal cells lining the vertebrate heart,
blood, and lymph vessels. Endothelial cells are the building blocks of angiogenesis, new
vessel formation. This is an important phenomenon during normal development and tissue
repair, as well as during various pathological processes such as tumour growth (Iivanainen et
al., 2003).
Angiogenesis depends on specific molecular interactions between vascular endothelial cells
and their surrounding microenvironment, composed of neighbouring cells and the
extracellular matrix (Iivanainen et al., 2003). This process involves a cascade of events
characterised by induction of vascular hyperpermeability, local degradation of the basement
membrane, migration and sprouting into the local stroma, cell proliferation and formation of
granulation tissue, reconstruction of the basement membrane and formation of new blood
vessels (Li and Thompson, 2003). Within tumours, vessels are organised in a chaotic fashion
and do not follow the hierarchical branching network of normal vascular networks (Jain and
Duda, 2003).
C HAEMATOPOIETIC CELLS
Tumours contain numerous different haematopoietic cell types, of the myeloid line as
granulocytes, macrophages and dendritic cells, and the lymphoid line, as B, T and natural
killer (NK) cells. Depending on the type of inflammation, haematopoietic cells can aid in
tumour growth (chronic inflammation) or tumour rejection (acute inflammation) (Philip et al.,
2004).
1.1.4 CYTOKINES AND CHEMOKINES WITHIN THE
TUMOUR STROMA
There is a functional relationship between inflammatory infiltrates and malignant growth, in
which the infiltrates can contribute to regression or development of cancer. In addition to
tumour cells, infiltrating cells such as macrophages and T cells are often responsible for
7
INTRODUCTION
cytokine and chemokine production. Normal cells produce cytokines only transiently,
however malignant cells can produce considerable and sustained amounts of cytokines and
pleiotropic chemokines such as IL-8, TGF-ß and macrophage chemotactic proteins (MCPs)
(De Wever and Mareel, 2003). These factors can enable the production of an array of
cytokines such as IL-2, IL-4, tumour necrosis factor (TNF) and interferon gamma (IFN-γ),
which influence tumour growth progression or inhibition.
Depending upon the cytokine milieu, cytokines and chemokines can be anti-tumour effector
molecules or alternatively they can contribute to tumour progression. Nevertheless, the origin
of these molecules (matrix bound or cellularly produced) and their influence in tumour
stromal interactions are not fully understood.
1.2 THE PROTEIN INTERFERON GAMMA (IFN-γ)
IFN-γ is a potent, pro-inflammatory cytokine produced primarily by activated T cells and NK
cells. It is a lymphoid factor that possesses powerful anti-viral and anti-parasitic functions,
which is also capable of inhibiting proliferation in a number of normal and transformed cells.
IFN-γ has several properties related to immuno-regulation:
i. Activator of macrophages to stimulate the release of reactive oxygen
species
ii. Increases major histocompatability class I (MHC-I) molecule
expression, and induces MHC-II molecules on a wide variety of cell
types
iii. Acts directly on T and B lymphocytes to promote differentiation and
maturation
iv. Strong activator of NK cells
1.2.1 STRUCTURE
IFN-γ is a dimeric protein. The active form of this cytokine is a homodimer consisting of two
intertwining 143-amino acid polypeptides (Sadir et al., 1998). The protein is glycosylated at
8
INTRODUCTION
two sites and the isoelectric point is 8.3-8.5. The (amino) N-terminus of IFN-γ is directly
involved in receptor binding and the integrity of the (carboxyl) C-terminus is critical for
biological activity. A basic amino acid cluster located in this C-terminal domain is involved in
the organisation of the three dimensional structure of the protein. This cluster is also
important as it increases the on rate of the IFN-γ - IFN-γ receptor (IFN-γ−R) binding reaction
(Lortat-Jacob and Grimaud, 1991).
Fernandez-Botran and co-workers, 1999, found that by removing stretches of more than 11
amino acids (aa) from the C-terminal portion of human IFN-γ, receptor binding and specific
activity were greatly reduced. This is due to the fact that the C-terminus of IFN-γ,
encompassing residues 95-133 in mouse and residues 95-134 in human IFN-γ, interact with
high affinity at a membrane proximal site on the cytoplasmic domain of the alpha chain of the
receptor (Subramaniam et al., 2000). It has been postulated that this domain could be a
regulatory element of the cytokine (Wetzel et al., 1990).
1.2.2 IFN-γ SIGNAL TRANSDUCTION
IFN-γ is a pleiotropic cytokine involved in aspects of immune regulation including
transcription. Upon engagement of IFN-γ with the receptor, janus tyrosine kinases (JAK) are
activated and subsequently phosphorylate the signal transducer and activator of transcription-
1α (STAT-1α) protein, which dimerises and translocates to the nucleus to induce target gene
transcription by binding to gamma activated sequences (GAS) in the promoter of IFN-γ-
responsive genes (Ma et al., 2005).
1.2.3 INTERACTION WITH THE MATRIX
IFN-γ molecules of different species display only modest homology at either cDNA or amino
acid levels (Fernandez-Botran et al., 1999). However, there is a conserved sequence within
the molecule present in many species and defined by the amino acids, KRKRS (Figure 3).
This homologous motif of the IFN-γ molecule is a high affinity binding site (Kd =10-9 M) for
heparan sulphate from basement membranes within the extracellular matrix (Lortat-Jacob and
Grimaud, 1992).
9
INTRODUCTION
ECM components have been hypothesised to be involved in storing IFN-γ and thereby
providing a local pool of cytokine (Lortat-Jacob et al., 1996b). Cell surface heparan sulphate
bound to IFN-γ was shown to delay the nuclear accumulation of IFN-γ, providing extra
evidence that HS molecules act as a storage depot around the cell for local delivery of the
cytokine (Sadir et al., 2000).
Once IFN-γ is bound to heparan sulphate, the extent of its C-terminal cleavage is reduced to
less than 10 amino acids and this increases the cytokine activity. This may be due to the
protein folding in a new relaxed conformation with increased stability (Lortat-Jacob and
Grimaud, 1991). IFN-γ is generally thought of as a soluble factor and, in vivo, is eliminated
from the bloodstream with a half-life (t1/2) of 1.1 minutes. However, after heparin stabilisation
plasma clearance is reduced to a half-life (t1/2) of 99 minutes (Lortat-Jacob et al., 1996a).
It has been demonstrated (Fernandez-Botran et al., 1999) that IFN-γ molecules bound to
immobilised GAGs, such as heparan sulphate, are still capable of retaining their activity and
can induce MHC class II expression on target cells. Lortat-Jacob and colleagues, suggested a
novel model for the interaction of HS with a protein in which two sulphated terminal
sequences of the binding domain interact directly with the two IFN-γ C-termini and bridge the
two cytokine monomers through an internal N-acetyl-rich sequence (Figure 4).
This interaction could be biologically relevant as basement membranes could provide a local
concentration of this soluble cytokine, direct the range of its action and act as a physiological
storage depot around cells. IFN-γ is sequestered at the surface of endothelial cells by
electrostatic interactions between specific basic amino acid residues KRKRS of the conserved
motif, and the sulphated domains of HS, the most abundant endothelial GAG (50-90%)
(Douglas et al., 1997). At present little is known about the in vivo influence of IFN-γ on the
matrix of both normal and malignant tissues.
10
INTRODUCTION
SPECIES
TOTAL IFN-γ
HOMOLOGY
(%)
MC-2
SEQUENCE
% HOMOLOGY
OF MC-2
SEQUENCE
Mouse LRKRKRSR
Rat 87 ******** 100
Hamster 56 ******** 100
Duck 33 SKRKRSQS 60
Gallus 35 FKRKRSQS 60
Turkey 35 SKRKRSHP 60
Pheasant 36 SKRKRSQS 60
Quail 36 SKRKRSQC 70
Guinea fowl 36 LKRKRNQP 50
Fugu 18 LEKR*R*R 50
Zebra fish 23 *KNKE*K* 50
Rainbow trout 18 DNR***RQ 20
Guineapig 46 Q***R*TQ 60
Baboon 48 IG*****Q 70
Macaque 48 IG*****Q 70
Human 42 TG*****Q 60
Marmoset 49 IG**R**Q 60
Squirrel monkey 48 IG*****Q 70
Rabbit 47 *K*****Q 80
Woodchuck 51 *******Q 90
Donkey 53 *******Q 90
Cat 51 *******Q 90
Dog 52 *******Q 90
Camel 53 *******Q 90
Llama 52 ******RQ 80
Cow 51 *******Q 90
Water buffalo 51 *******Q 90
Goat 51 *******Q 90
Dolphin 50 ****R**Q 80
Pig 51 *******Q 90
Figure 3: Species consensus of the IFN-γ molecule and the C-terminal (mIFN-γ128-135)
fragment responsible for binding to heparan sulphate as determined by Vector-
AlignmentTM. Sequence homology was also described previously (Zou et al., 2005). (*)
Indicates consensus with the mouse sequence.
Amino acids; Green – non polar / hydrophobic, Black – No charge/ non-acidic amino acid/
polar/ hydrophilic, Blue – Positive charge/ basic amino acid/ polar /hydrophilic, Red –
Negative charge/ acidic amino acid/ polar / hydrophilic. For phylogenetic tree and complete
sequences see Appendix (Figure 27).
11
INTRODUCTION
A
B
Figure 4: Molecular organisation of the HS binding site
of Human IFN-γ
(A) IFN-γ is a C2-symmetric homodimer in solution and
binds (bold type-site of interaction) to (B) HS by virtue of
basic residues located at the C terminus of the two
subunits. A model of the interaction between IFN-γ and
HS was thus proposed in which the KRKRS domains,
located at the C terminus of each subunit of the IFN-γ
dimer, would interact with the highly charged NS domains
of the fragment (Lubineau et al., 2004).
N-acetylated regions (NA-domains) are mainly composed
of D-glucuronic acid and N-acetylated glucosamine, and
thus with low global charge, separate domains rich in L-
irduronic acid and N-sulphated glucosamine (NS-
domains), which are hypervariable and highly charged.
1.2.4 ROLE OF IFN-γ IN TUMOURS
Several mechanisms have been proposed to explain the role of IFN-γ in tumours using
defined experimental models. These include:
• Effects on tumour growth and survival. Studies have shown that IFN-γ and IFN−γ
receptor are essential for tumour rejection (Blankenstein and Qin, 2003). Tumours that
have been transfected to secrete IFN-γ are rejected in immunocompetent mice (Hock
et al., 1993) and blocking endogenous IFN-γ with neutralising antibodies inhibits
12
INTRODUCTION
tumour rejection (Dighe et al., 1994). The ability of CD8+ T cells to mediate tumour
rejection upon adoptive transfer correlates with interferon gamma production (Becker
et al., 2001). However, the mechanism by which IFN-γ exerts tumouricidal activity is
still not completely resolved.
• Angiogenesis. It is well known that a growing tumour requires new blood vessels.
Inhibition of angiogenesis by CD4+ (Qin and Blankenstein, 2000) and CD8+ (Qin et
al., 2003) T cell-derived IFN-γ, is an effective way to prevent rapid tumour burden
thereby allowing other, perhaps direct killing mechanisms to eliminate residual
tumour cells (Blankenstein and Qin, 2003). Evidence suggests that there is a critical
dependence on expression of IFN-γ−R on the vasculature in order to mediate
destruction of blood vessels, and ultimately of the tumour (Ibe et al., 2001).
• Effects on both innate and adaptive immune reponses against tumours. IFN-γ is
known to be a macrophage activating factor capable of inducing non-specific kill of a
variety of tumour targets as well as up-regulating expression of cytotoxic ligands such
as TNFα and FAS-ligand (Farrar and Schreiber, 1993). Additionally, IFN-γ also
stimulates B cell proliferation and differentiation and enhances MHC-I expression on
the tumour surface (Mocellin et al., 2001).
1.3 TUMOUR THERAPY
The reasons why the immune system does not effectively destroy tumour cells is not fully
understood and therefore is controversially discussed. The concept of immune surveillance
was first formulated by Thomas (Thomas, 1959). They assumed that the immune system
would recognise pre-cancerous and cancerous cells as non-self and reject them. This concept
is valid only for virally transformed cells. Tumours arising from non-virus related
mechanisms are regarded by the immune system as “self”, and therefore attempts to
manipulate the immune system to recognise the tumour as “non self” are difficult (Klein and
Klein, 2005).
Initially it was believed that tumours could escape immune recognition as there was no
detectable adhesion or costimulatory molecules, and no peptides that could be presented by
13
INTRODUCTION
MHC molecules, thereby creating low immunogenicity. Later it was thought tumours could
express antigens to begin with, to which the immune system could respond but that later these
were lost by antibody-induced internalisation or antigenic variation. Essentially, when
tumours were attacked by cells responding to a particular antigen, any tumour cell that did not
express that tumour antigen, would have a selective advantage.
Recently, it has been demonstrated that immune responses could be suppressed directly as
tumours can often produce substances such as TGF-ß and IL-10. Willimsky and Blankenstein,
(2005) demonstrated that spontaneously developed sporadic tumours of high antigenicity are
initially ignored by the immune system. The expansion of antigenic-spceific T cells occurs
later, but the T cells are dysfunctional at this later stage (Willimsky and Blankenstein, 2005).
The fact that cancer cells are genetically and phenotypically less stable than normal cells
enables them to rapidly change and escape immune destruction. Aided by the tumour stroma,
tumour cells can become resistant to destruction. In an effort to combat tumours various
therapies have been developed. These include radiation, surgery, chemotherapy, vaccination
and adoptive immunotherapy. Often combinations of these therapies have elicited the best
response in patients.
1.3.1 CYCLOPHOSPHAMIDE
Cyclophosphamide (Cy) is an alkylating agent widely used in chemotherapy. It has a bimodal
effect on the immune system, depending on the dose and the schedule of administration
(Matar et al., 2002). Large doses of Cy bring about impairment of host defence mechanisms,
therefore leading to severe immunodepression. However, the administration of low doses
leads to an enhancement of immune responses in both experimental animals (Awwad and
North, 1988) and humans (Berd et al., 1982).
Cy is an inert lipophilic prodrug that requires enzymatic conversion regulated by the hepatic
cytochrome P450 2B1 gene for its anticancer effect (Figure 5) (Pass et al., 2005). As Cy
toxicity is not dependent on a particular phase in the cell cycle, it is well suited for the
treatment of solid tumours.
14
INTRODUCTION
The anti-tumour response induced by Cy has been investigated previously, and the
interactions of stroma cell components were analysed during rejection of established tumours
(Ibe et al., 2001). It was shown that 6 h after Cy-treatment T cells in the tumour were
inactivated and tumour infiltrating macrophages (TIMs) switched to IFN-γ production. Both,
IL-10 production before and IFN-γ production after Cy-treatment by TIMs required T cells.
Under certain experimental conditions Cy can induce tumour rejection by host cell
modulation rather than direct tumouricidal activity (Awwad and North, 1988). For example, a
single injection of a defined amount of Cy induced tumour rejection in immunocompetent
mice but had no effect on tumour growth in immunodeficient mice (Hengst et al., 1980). It
was discovered that during low dose therapy using Cy, IFN-γ played a crucial role in
mediating tumour rejection, however whether the cytokine originates from T cells (Tsung et
al., 1998), or macrophages (Ibe et al., 2001) remains in question.
Figure 5: Pathway of the Bioactivation of
Cyclophosphamide (Chiocca, 1995)
1.4 AIMS
The goal of this study was to investigate the origin of IFN-γ during chemotherapy induced,
immune response mediated tumour rejection.
15
INTRODUCTION
The role of IFN-γ in mediating tumour rejection is known, yet all previous studies have
focused on the cellular origin of the cytokine. Matrix bound IFN-γ may play an important role
in immune responses. This hypothesis was based on experiments carried out by Fernandez-
Botran and coworkers, (2002) in which the survival of an allotransplanted skin graft was
prolonged by inhibiting this interaction. Additionally, it was shown that the tertiary structure
of IFN-γ changed when bound to the matrix and this provided the molecule with higher
affinity for its receptor. Therefore, major attention in this work was given to the presence and
role of matrix bound IFN-γ within the tumour microenvironment.
By analysing the tumour stroma interactions in an established model during
cyclophosphamide (Cy)-induced tumour rejection the aim was to determine:
• Whether IFN-γ played a role in tumour rejection in this model
• Which cells produced IFN-γ in response to tumour therapy?
• Could IFN-γ bind to the matrix in vivo?
• Could the interaction of IFN-γ with the matrix be disturbed and were there functional
consequences?
The role of IFN-γ in tumour immunity was investigated by inoculation of the plasmacytoma
J558L in IFN-γ-/- and IFN-γ+/- BALB/c mice and bone marrow chimeras followed by treatment
with a previously established amount of Cy (15mg/kg) (Ibe et al., 2001). Tumour growth
kinetics, histological analysis and radioactive cytokine accumulation were used to determine
whether there was a matrix/IFN-γ interaction within the model, and if interference of this
interaction could inhibit tumour rejection.
16
MATERIALS AND METHODS
2 MATERIALS AND METHODS
2.1 MATERIALS
2.1.1 CELL LINES
The plasmacytoma cell line J558L grows aggressively in BALB/c mice. It is a heavy chain
loss variant of the BALB/c derived line J558 that synthesises and secretes the λ light chain.
The cell line was induced using mineral oil and expresses H-2d and the antigen PC.1.
Additionally, J558L express cellular adhesion molecule ICAM-1 (Cavallo et al., 1995), but no
detectable MHC class II molecules. Cells are sensitive to dexamethasone and cortisol.
The J558L-interferon gamma transfectants were generated by Hock and colleagues, 1993.
Essentially the tumour cell line was transfected with the pLTR-IFN plasmid along with
pWLneo. Under G418 selection the cell line is capable of producing 120 ng/ml IFN-γ as
detected by ELISA.
TS/A is a spontaneous mammary adenocarcinoma, which developed from a 20 month old
retired BALB/c breeder. The cell line expressed MHC class I but not MHC class II molecules
and grows in H-2 matched, minor histocompatability antigen incompatible hosts such as
DBA/2 mice (Nanni et al., 1983).
Mc51.9 was induced by MCA in 129/Sv/Ev IFNγR-/- mice (H2b). These tumour cells express
MHC class I molecules (both H-2Db and H-2Kb) and are negative for MHC class II. The
interferon gamma transfectant was created using the pLTR-IFN and pWLneo plasmid and
under G418 selection pressure is capable of producing 70 ng/ml of IFN-γ.
The XMG-6 mAb neutralising IFNγ was obtained from ATCC (American Type Cell Culture).
It was originally derived from a Lewis rat after immunizing on eight sequential days with
soluble recombinant mouse IFN-γ (170 µg/injection). This rat was later boosted with IFN-γ
(100 µg) in CFA. After 21 days, the rat was boosted again with 100 µg IFN-γ without
adjuvant. 3 days later, the spleen cells were fused with P3X63Ag myeloma cells using 50%
polyethylene glycol. The cytokine production was identified by ELISA and neutralisation
capability determined. The isotype for XMG-6 was concluded to be IgG1.
17
MATERIALS AND METHODS
2.1.2 MICE
Wild type BALB/c mice and Severe Combined Immune Deficient (SCID) mice were obtained
from Jackson laboratories. IFN-γ deficient mice backcrossed to the BALB/c background, 11
generations, were bred and maintained from birth at the Max-Delbrueck-Centrum in Berlin.
All mice were used in experiments when they were 8-12 weeks old. Genotypes of IFN-γ
deficient mice (Dalton et al., 1993) were confirmed by polymerase chain reaction (PCR)
using a thermocycler. (PCR conditions: 12 cycles, annealing 64°C for 30 sec; 25 cycles,
annealing 58°C for 30 sec, elongation 72°C for 2 min).
2.1.3 PEPTIDE
The sequence specific for the basic amino acid cluster responsible for IFN-γ binding to
heparan sulphate was identified (Fernandez-Botran et al., 1999). The fragment of the murine
IFN-γ 128-135 named MC-2 (LRKRKRSR) was produced and purified by HPLC (Biosynthan,
Berlin). The peptide was diluted in phosphate buffered saline (PBS) to a 1 mM concentration
and sterile filtered. Aliquots were stored at –20°C.
2.1.4 OLIGONUCLEOTIDES
Oligonucleotides were obtained from TIBR Molbiol (Berlin, D).
A PCR PRIMERS
Intact IFN-γ:
fwd: 5’-AGA-AGT-AAG-TGG-AAG-GGC-CCA-GAA-G-3’
rev: 5’-AGG-GAA-ACT-GGG-AGA-GGA-GAA-ATA-T-3’
fragment: 260 bp
IFN-γ disrupted by neomycin:
fwd: 5’-TCA-GCG-CAG-GGG-CGC-CCG-GTT-CTT-T-3’
rev: 5’-ATC-GAC-AAG-ACC-GGC-TTC-CAT-CCG-3’
fragment: 320 bp
18
MATERIALS AND METHODS
pSecTag/EF1α:
fwd: 5’- ATC AGG GTT ATT GTC TCA T -3’
rev: 5’- GGA ACC CAG AGC AGC AGT -3’
fragment: 1466 bp
pMC2sec:
MC21: 5’- TAC CGA GCT CGG ATC CTC TCA GGA AG -3’
MC22: 5’- CAG CAT GCC TGC TAT TGT CTT CCC AA -3’
Fragment: 406 bp
β-Actin:
U12: 5’- TGG AAT CCT GTG GCA TCC ATG AAA CTA CAT-3’
U13: 5’- AAA CGC AGC TCA GTA ACA GTC GCG CTA GAA-3’
Fragment: 347 bp
B INSERTS
Murine IFN-γ 128−135 C-2) insert:
5’- GAT CCT CTC AGG AAG CGG AAA AGG AGT CGC TGC G -3’
5’- AA TTC GCA GCG ACT CCT TTT CCG CTT CCT GAG AG -3’
Fragment: 34 bp
Murine neutral insert:
5’- GAT CCT GCC GCA GCG GCT GCA GCC GCC GCT GCC G –3’
5’- AAT TCG GCA GCG GCG GCT GCA GCC GCT GCG GCA G –3’
Fragment: 34 bp
19
MATERIALS AND METHODS
2.1.5 REAGENTS
A CHEMICALS
CHEMICAL ABBREVIATION COMPANY
1-kb-DNA Ladder M Invitrogen, Karlsruhe, G
2-Mercaptoethanol ME Merck, Darmstadt, G
4-2-hydroxyethyl-1-piperazineethanesulfonic
acid
HEPES Serva, Heidelberg, G
5x First strand buffer Invitrogen, Karlsruhe, G
Acrylamide Bio-rad, München, G
Agarose Serva, Heidelberg, G
Ammonium chloride NH4Cl Merck, Darmstadt, G
Ampicillin (50ug/ul) Sigma, Taufkirchen, G
Ammonium persulphate (10%w/v) APS Bio-rad, Munich, G
Borgal solution (24%) Intervet,
Unterschleißheim, G
Bovine serum albumin BSA Serva, Heidelberg, G
Chloroform Roth, Karlsruhe, G
Concavalin A Con A Sigma, Taufkirchen, G
Cytofix/Cytoperm Becton Dickinson,
Heidelberg, G
Deoxy-nucleotide triphosphate dNTP’s Roche, Mannheim, G
Diethylether Otto Fischer, Berlin, G
Diethylpyrocarbonate DEPC Fluka, Taufkirchen, G
Dimethylsulfoxide DMSO Sigma, Taufkirchen, G
Dithiothritol DTT Merck, Darmstadt, G
Dnase I (Rnase-free) Roche, Mannheim, G
Dulbecco’s modified Eagles medium DMEM Gibco, Karlsruhe, G
Dulbecco’s PBS (1x) D-PBS Invitrogen, Karlsruhe, G
Ethanol Et-OH Merck, Darmstadt, G
Ethidium bromide Serva, Heidelberg, G
Ethylene diamine tetra-acetic acid EDTA Merck, Darmstadt, G
Extracellular matrix gel from Engelbreth-
Holm-Swarm murine sarcoma
Matrigel/ ECM gel Sigma,Taufkirchen,G
Fetal calf serum FCS Greiner, Solingen, G
20
MATERIALS AND METHODS
Gel/mount Vector Laboratories,
Peterborough, UK
Geneticin G418 Gibco, Karlsruhe, G
Glutamine Glu Gibco, Karlsruhe, G
Golgi plug Becton Dickinson,
Heidelberg, G
Haemotoxylin Merck, Darmstadt, G
Histoacryl B/Braun, Melsungen,G
Hydrochloric acid HCl Roth, Karlsruhe, G
Interferon gamma (murine) mIFN-γ R&D Systems,
Interferon gamma (human) hIFN-γ Bohringer Ingelheim,G
Isoamyl alcohol Merck, Darmstadt, G
Isopropanol Roth, Karlsruhe, G
Kaisers gelatine Merck, Darmstadt, G
Magnesium chloride MgCl2 Merck, Darmstadt, G
Magnesium sulfate MgSO4 Merck, Darmstadt, G
Nusieve GTG agarose Cambrex, Hess
Oldendorf,G
Tissue-Tek® OCT Sakura Finetek, USA
Penecillin/Streptomycin P/S Gibco, Karlsruhe, G
Perm/Wash Buffer Becton Dickinson,
Heidelberg, G
Phenol Gibco, Karlsruhe, G
Phenol/Chloroform (1:1) Roth, Karlsruhe, G
Polyoxyethylene sorbitan monolaureate Tween-20 Sigma, Taufkirchen, G
Phosphate buffered saline PBS Gibco, Karlsruhe, G
Potassium chloride KCl Merck, Darmstadt, G
Potassium bicarbonate KHCO3 Roth, Karlsruhe, G
Proteinase K Roche, Mannheim, G
Sodium acetate Fluka, Taufkirchen, G
Sodium chloride NaCl Fluka, Taufkirchen, G
Sodium citrate Sigma, Taufkirchen, G
Sodium-dodecyl-sulphate SDS Serva, Heidelberg, G
Sodium hydroxide NaOH Merck, Darmstadt, G
21
MATERIALS AND METHODS
Sodium pyruvate Gibco, Karlsruhe, G
RNAse-inhibitor Promega, Manheim, G
RPMI 1640-Medium RPMI Gibco, Karlsruhe, G
Reverse transcriptase (Superscript II) RT Invitrogen, Karlsruhe, G
T4 DNA ligase Promega, Manheim, G
T4 ligation buffer Promega, Manheim, G
Taq polymerase Roche, Mannheim, G
Taq polymerase buffer (10x) Roche, Mannheim, G
Tetramethylethylenediamine TEMED Bio-rad, Munich, G
TOTO-3 iodide TOTO-3 Invitrogen, Karlsruhe, G
Tricine Bio-rad, Munich, G
Tris (hydroxymethyl) aminomethane Tris Sigma, Taufkirchen, G
Tris-HCl Tris-HCl Sigma, Taufkirchen, G
Triton X-100 Serva, Heidelberg, G
Trizol Invitrogen, Karlsruhe, G
Trypan blue Sigma, Taufkirchen, G
Trypsin Gibco, Karlsruhe, G
Zeocin Zeo Invitrogen, Karlsruhe, G
B KITS
KIT COMPANY
DAB Substrate Kit Sigma, Taufkirchen, G
DAKO Fast-Red DAKO, Hamburg, G
DC Protein Assay Kit Bio-Rad, Munich, G
DNA Blunting/Ligation Kit TAKARA BIO. INC, Japan
Gel Extraction Kit Qiagen, Hilden, G
Human IFNγ ELISA Set Becton Dickinson, Heidelberg, G
In Vivo Capture Assay for Mouse IFN-γ Becton Dickinson, Heidelberg, G
Micro Protein Determination Sigma, Taufkirchen, G
Mouse IFNγ ELISA Set Becton Dickinson, Heidelberg, G
Plasmid Mini/Maxi Preparation Kit Qiagen, Hilden, G
22
MATERIALS AND METHODS
TitaniumTM One-Step RT-PCR Kit Clontech, Heidelberg, G
C BUFFERS
BUFFER INGREDIENTS
Annealing buffer 10 mM Tris, pH 7.5-8.0, 50 mM NaCl, 1 mM
EDTA
Blocking buffer (IHC) 10% FCS, 1 x PBS, 0.1% NaN3
Chloroform/Isoamyl alcohol Chloroform (29 ml), Isoamylalcohol (1 ml)
Diethylpyrocarbonate (DEPC) water 500 µl DEPC, 500 ml Aqua dest.
Ethidium bromide 10 mg ethidium bromide/ml A.dest
Gel buffer concentrate/500ml 181.5 g Tris, 1.5 g SDS, pH to 8.45 w/ HCl
LB medium 10 g NaCl, 5 g Yeast extract, 10 g trypton,
1L aqua dest
Lysis buffer for spleen 0.15 M NH4Cl, 1 mM KHCO3, 0.1 mM
Na2EDTA, pH 7.2
Lysis buffer for mouse tail biopsies 100 mM Tris-HCl, pH 8.5; 5 mM EDTA, pH
8; 0.2% SDS, 200 mM NaCl
PBS-T PBS containing 0.05% Tween-20
1x SDS PAGE buffer 24 g Tris base, 115.2 g glycine, 20 ml 20%
SDS, H2O to 4 litres
SOC-medium 2% Bactotrypton, 0.5% Bacto-yeast extract,
10 mM NaCl, 2.5 mM KCl, 10 mM MgCl2,
10 mM MgSO4, 20 mM Glucose
TAE (50x) 242 g Tris, 57.1 ml concentrated acetic acid,
100 ml EDTA pH8; 1 L H20
TE buffer (DNA dilution buffer) 100 mM Tris-HCL, 5 mM MgCl2, pH 7.6 at
16°C
23
MATERIALS AND METHODS
D ANTIBODIES
ANTIBODIES ISOTYPE COMPANY
Rat anti-mouse CD31, (390) Rat IgG2a, κ Becton Dickinson, Heidelberg, G
Rat anti-mouse Interferon gamma
(XMG1.2, F1, XMG6, R4-6A2-
Biotinylated).
Rat IgG1 Abcam, Cambridge UK,
Becton Dickinson, Heidelberg, G
Rat anti-mouse Heparan sulphate
(A7L6)
Rat IgG2a, κ Abcam, Cambridge UK
Rat anti-mouse ERTR7 Rat IgG2a DPC Biermann, Bad Neuheim, G
Rat IgG1 (isotype standard) Rat IgG2a, κ Becton Dickinson, Heidelberg, G
Rat IgG2a (isotype standard) Rat IgG2a, κ Becton Dickinson, Heidelberg, G
Rat IgG2b (isotype standard) Rat IgG2a, κ Becton Dickinson, Heidelberg, G
Goat anti-rat FITC Goat IgG Abcam, Cambridge UK
Goat anti-rat Texas Red Goat IgG Abcam, Cambridge UK
Goat anti-rat Alkaline
Phosphatase
Goat IgG Dianova GmbH, Hamburg, G
Goat anti-rat Horse Radish
Peroxidase
Goat IgG Dianova GmbH, Hamburg, G
Goat anti-rabbit Alkaline
Phosphatase
Goat IgG Dianova GmbH, Hamburg, G
Rabbit anti-mouse Collagen I Rabbit IgG Abcam, Cambridge, UK
Rabbit anti-mouse Collagen II Rabbit IgG Abcam, Cambridge, UK
Rabbit anti-mouse Collagen III Rabbit IgG Abcam, Cambridge, UK
Rabbit anti-mouse Collagen IV Rabbit IgG Abcam, Cambridge, UK
E EQUIPMENT
PRODUCT COMPANY
96 well cell culture plates (flat or round
bottom)
Corning Costar, Bodenheim, G
96 well breakaway plates (flat bottom) Corning Costar, Bodenheim, G
Cell culture flasks (T-25, T-75, T-150) TPP, Trasafingen, Switzerland
24
MATERIALS AND METHODS
CELLineTM system Integra-Bioscience, Fernwald, G
Centrifuge tubes (15 ml, 50 ml) BD Falcon, Heidelberg G
Disposable cuvettes Roth, Karlsruhe, G
Disposable syringes (25 ml, 10 ml, 5 ml, 1ml) Braun, Mesungen, G
Eppendorf tubes (2 ml, 1.5 ml, 0.5 ml) Eppendorf, Hamburg, G
FACS tubes Becton Dickenson, Heidelberg G
Filter (0.2 µm, 0.45 µm) Schleicher and Schüll, Dassel, G
Needles (0.8 x40 bwz. 0.4 x20 mm) Sanimed, Berlin, G
PAPTM pen DAKO,Hamburg, G
Petri-dishes (10 cm Diameter) Greiner, Solingen, G
Pippette tips Roth, Karlsruhe, G
Quartz cuvettes Hellma, Müllheim,Baden, G
Scapel NeoLab, Heidelberg, G
Scissors Roth, Karlsruhe, G
Sieve (40 µm) Becton Dickinson, Heidelberg, G
Superfrost slides Roth, Karlsruhe, G
Sterile glass pippettes Brand, Wertheim, G
Sterile pipettes (5, 10, 25 ml) Costar, Bodenheim, G
Tweezers Roth, Karlsruhe, G
F MACHINERY
MACHINE COMPANY
Bio-Freezer (-80°C) Forma Scientific, Cotech, Berlin, G
Centrifuge 3K12 Sigma, Taufkirchen, G
Centrifuge 5415C Eppendorf, Hamburg, G
Centrifuge RT 6000D Sorvall, Langenselbold, G
Confocal LSM 510 Zeiss, G
Counter Mitutoyo, Neuss, G
Electrophoresis chamber for gels Bio-rad, München, G
Electroporation machine Amaxa, Köln, G
ELISA reader MR 5000 Dynatech, Berlington, USA
25
MATERIALS AND METHODS
FACSCalibur flow cytometer Becton Dickinson, Heidelberg, G
Fluorescent image analyser FLA-5000 FujiFilm, Dusseldorf, G
Frost free fridge Labotect, Göttingen, G
Gene ray UV-photometer Biometra, Göttingen, G
Haemocytometer Roth, Karlsruhe, G
Liquid nitrogen tank Messer Griesheim, Griesheim, G
Lysis machine Coulter, Krefeld, D; Bector Dickinson,
Heidelberg, G
Microscope Leitz DM IL Leica, Wetzlar, G
Olympus BX51 Olympus, Hamburg, G
PCR machine Biometra, Göttingen, G
pH meter Hanna Instruments, Kehl/Rhein, G
Pippetes (single or multichannel) Eppendorf, Hamburg, G
Print-Scale-Timer BF2306P Ortec, Meerbusch, G
Scales Sartorius, Göttingen, G
Spectrophotometer UV-160A Shimadzu, Berlin, G
Sterile bench BDK, Sonnbuhl, G
Thermomixer Eppendorf, Hamburg, G
Ultracentrifuge Beckmann, Krefeld, G
Vortex Janke & Kunkel IKA-labortechnik, Staufen,G
Wallac-Wizard-Gamma counter 1470 Perkin Elmer, Jügesheim, G
Waterbath GFL, Burgwedel, G
26
MATERIALS AND METHODS
2.2 METHODS
2.2.1 CELL CULTURE
A INCUBATION
All cells were incubated in a 5% CO2 oven at 37°C. All cells and cell lines were processed in
appropriate cell culture flasks and dishes.
B ADHERENT CELL CULTURE
When the cell monolayer (TS/A) had reached 80% confluency the cells were passaged.
Essentially, the culture medium was removed and cells washed with PBS. Cells were treated
with 0.3% Trypsin/2.7 mM EDTA diluted in PBS to cover the monolayer. After a 5 minute
incubation period at 37°C the cells had detached and were transferred to a tube for
centrifugation (5 min at 1200 rpm). The supernatant was discarded and cells were
resuspended in medium or frozen for storage.
C SUSPENSION CELL CULTURE
Semi-confluent cell cultures (J558L, J558L-IFN-γ or XMG6) were split and transferred to
new medium. This process involved centrifuging the cells for 5 minutes at 1200 rpm before
splitting or freezing for storage.
D COUNTING AND VIABILITY
10 µl of cell suspension was added to 90 µl of trypan blue and thoroughly mixed. After
incubation at room temperature for 2 minutes, trypan blue excluded cells were considered
viable and counted using a cell counting chamber (haemacytometer). The cell concentration is
then determined by the following equation:
Cell number x dilution factor x total volume x 104
27
MATERIALS AND METHODS
E FREEZING AND THAWING
Cells that had reached 80% confluency in a T-75 cell culture flask were centrifuged and
pelleted. They were washed in PBS and counted. After further centrifugation, the new pellet
was then resuspended in a solution of 10% DMSO and 90% FCS so that 1x107 cells per 1 ml
could be frozen slowly to –80°C. For long term storage cells were transferred to a liquid
nitrogen tank. When aliquots were required, cells were defrosted in a 37°C water bath and
then resuspended in 5 ml of medium.
F ANTIBODY PRODUCTION
The hybridoma XMG-6 produces a neutralising anti-IFN-γ monoclonal antibody. INTEGRA
CELLine CL1000 enables high cell culture densities and consequently augmented
concentrations of antibody via membrane technology. The lower chamber was inoculated
with 25 x 106 cells in 15 ml of complete DMEM supplemented with 10% FCS. The upper
nutrient compartment contained complete medium without serum. When harvesting the lower
chamber for antibody, 7.5 ml of mixed cell suspension was removed and replaced with 7.5 ml
of fresh complete medium. The harvest supernatant, which contained the antibody was
centrifuged for 15 min at 12000 rpm. The pellet was discarded and the supernatant stored
until purified by HPLC. Cultures were harvested every three days for 6-8 weeks.
2.2.2 IN VIVO EXPERIMENTS
A TUMOUR CELL INJECTION AND IN VIVO TREATMENT
J558L, J558L-IFN-γ, TS/A and plasmid transfected tumour cells were injected
subcutaneously in the left abdominal region of BALB/c IFN-γ competent or deficient mice at
a concentration of 1 x 106 – 5 x 106 cells per 0.2 ml of PBS. At day 11 mice were treated with
15 mg/kg Cyclophosphamide (Cy) in Dulbeccos PBS. Injections of Cy were administered
intraperitoneally (i.p). Tumour size was measured by a caliper and determined as the mean of
the largest diameter and the diameter at right angle. Tumours had an average size of 1 cm in
diameter or approximately 0.63 grams, 11 days after injection as previously determined by
Ibe and coworkers, 2001. Tumour rejection was defined as complete regression after
treatment and the absence of recurrent tumour for the entire follow up period (60 days).
28
MATERIALS AND METHODS
B NEUTRALISATION
To neutralize the IFN-γ activity in vivo, BALB/c mice were i.p. injected with 1 mg of purified
XMG6 (rat anti-mouse IFN-γ mAb) in 0.5 ml D-PBS 1 day before tumour cell injection. As
control, another group of mice were injected with affinity purified total Rat-IgG
(1mg/mouse).
C BONE MARROW CHIMERAS (BMC)
IFN-γ competent and deficient mice were separated one day before irradiation. On the day of
transfer five mice of each subset were euthanised and the bone marrow extracted from the
femurs. The bone marrow of one donor mouse was used to reconstitute four to five recipient
mice. To isolate the bone marrow, the ends of the femurs were cut off and the bone marrow
was flushed out with serum-free DMEM using a 25-gauge needle attached to a 1ml syringe.
The bone marrow was dispersed into a single cell suspension, and debris and cell clumps were
removed by filtration through a 40 µm filter. Typical yield from two femurs was between 1
and 1.5 x 107 cells. Recipient mice were irradiated on the day of transfer with 10 Gy and
subsequently injected intravenously with 0.2ml bone marrow cells from a 107 cells/ml bone
marrow cell suspension. Mice were monitored for 3 months and treated with Borgal
intermittently throughout that time period. After which the success of the chimerism was
determined by PCR of both blood and tail DNA. Mice were then injected with J558L tumour
cells and treated as above.
D IN VIVO CINCINNATI CYTOKINE CAPTURE ASSAY
Principle: Normal or experimentally treated mice are injected with a biotin-labelled cytokine-
binding monoclonal antibody. This allows the target cytokine, which normally has a very
short half-life, to accumulate in vivo for a defined period of time (2-72 hours) as a soluble
cytokine-anti-cytokine antibody complex. This complex increases the cytokines subsequent
cytokine measurement because it inhibits the cytokines utilisation, degradation or excretion.
The level of in vivo captured cytokine present in serum is measured using ELISA with a
monoclonal antibody directed against a different epitope.
29
MATERIALS AND METHODS
Tumour bearing IFN-γ wild type mice treated with or without Cy treatment were injected i.p.
with 10 µg of a no azide/low endotoxin biotin-conjugated anti-mouse IFN-γ antibody in 200
µl of sterile PBS. Blood was collected 24 hours later and allowed to clot for 30-60 minutes.
Samples were centrifuged for 10 minutes at 4°C at 4000 rpm. Serum was removed and
aliquoted for immediate ELISA and storage at –80°C.
2.2.3 IMMUNOHISTOCHEMISTRY
A LIGHT MICROSCOPY
Principle: Detection of protein in tissues can be determined by a specific antibody, which is
chemically coupled to an enzyme that converts a colourless substrate into a coloured product
in situ. The localised deposition of the coloured product where antibody has bound can be
directly observed under a light microscope. The antibody binds stably to its antigen, allowing
unbound antibody to be removed by thorough washing.
Mice were inoculated with tumours and 10 days later they were removed, embedded in OCT
and frozen in liquid nitrogen. Blocks were mounted and 4 µm sections were cut using a
Micron cryostat. Sections were allowed to dry on Superfrost slides for two hours and either
used directly or stored at –80°C long term. Sections were fixed in acetone and sections were
circled with a PAPTM pen. They were placed in a humid chamber and 50-75 µl of blocking
buffer was added to each section for 30 minutes to 1 hour at room temperature. After removal
of the blocking buffer, sections were incubated with 5 µg/ml dilution of primary antibody
(50-75µl per section) for 1 hour at room temperature or overnight at 4°C. Slides were washed
with blocking buffer and then the (7 µg/ml) secondary antibody (50-75 µl per section)
applied and incubated for 1 hour at room temperature. The enzyme substrate alkaline
phosphatase (AP) or diaminobenzidine (DAB) was used according to the manufacturers
instructions and left on the sections for 5-10 minutes. After thorough rinsing with water,
sections were washed in haematoxylin counterstain for 10 minutes. After a further wash,
slides were dried carefully and sealed using Kaisers Gelatine for analysis using an Olympus
BX51 microscope.
30
MATERIALS AND METHODS
B FLUORESCENT MICROSCOPY
Fluorescent antibodies can also be responsible for the visualisation of certain structures within
tissue sections. However, when two or more structures are visualised, fluorescence is the
most appropriate and safe method. Sections were not fixed. The slides were blocked with (50-
75µl per section) the appropriate buffer for 1 hour. They were washed and incubated for 1
hour with 50-75µl per section primary heparan sulphate unconguated antibody (5 µg/ml). A
secondary anti-Rat Texas red (615 nm emission) was applied at 5 µg/ml was applied for 1
hour. After washing with blocking buffer, 100 µl of an anti-IFN-γ-FITC conjugated antibody
(100 µg/ml) was incubated for 48-72 hours in the dark at 4°C. After washing slides were
sealed with Gel/mount (Vector Laboratories).
2.2.4 LASER SCANNING CONFOCAL MICROSCOPY
Principle: The confocal microscope uses computer-aided techniques to produce ultra-thin
optical sections of tissue. Confocal images have high resolution without the need for elaborate
sample preparation. The resolution of the confocal can be further increased using low-
intensity illumination so that two photons are required to excite the fluorochrome. A pulsed
laser beam is used, and only when it is focused into the focal plane of the microscope is the
intensity sufficient to excite fluorescence. By this the fluorescence emission itself can be
restricted to the optical section.
2.2.5 IN VIVO MATRIGEL BINDING ASSAY
Matrigel was thawed to 4°C and 0.2 ml was injected subcutaneously into the belly of IFN-γ-/-
or IFN-γ+/- mice. After a 5 day rest period, mice were treated with human IFN-γ. After 1-24
hours the gel was extracted from the mouse and homogenised with 500 µl cold PBS. A
diluted sample (100 µl sample: 1 ml PBS) was then analysed for IFN-γ by ELISA.
31
MATERIALS AND METHODS
2.2.6 ENZYME LINKED IMMUNOSORBENT ASSAY (ELISA)
Principle: A typical sandwich ELISA involves a specific capture antibody, samples, a
biotinylated detection antibody, a streptavidin-HRP conjugate (SAV), chromogen and stop
solution. Antigen will bind to the immobilised capture antibody and to the biotinylated
detection antibody. The SAV binds to the detection antibody to complete the sandwich. A
substrate solution is added, acted upon by the enzyme conjugated SAV, and effects a colour
change. The intensity of the colour change is proportional to the amount of antigen in the
original sample.
A IFN−γ ELISA
Murine interferon gamma produced by J558L-IFNγ cells or murine spleens after stimulation
with ConA were determined using the mouse IFN-γ kit. Essentially, supernatant samples from
murine cell culture were purified by filtration so that no debris was present.
Human interferon gamma was determined using a human IFN-γ kit. Human samples were
prepared by washing and mashing matrigel, isolated from mice, in cold D-PBS and filtering
the supernatant with 40µm filters.
Serum samples isolated from Cy treated and untreated mice for the in vivo capture assay were
analysed using the BD In Vivo Capture ELISA.
Standards and samples were added to 96 well plates that had been treated with the respective
capture antibody and were processed according to manufacturer’s instructions. The plates
were read at 450 nm with a λ correction at 570 nm.
2.2.7 CONSTRUCTION OF TUMOUR CELLS EXPRESSING
MC-2 PEPTIDE
A PLASMID CONSTRUCTION
PsecTag2 is a 5.2 kb expression vector, which was selected from Invitrogen Life
Technologies to produce high-level stable expression in mammalian cells. Proteins expressed
32
MATERIALS AND METHODS
from pSecTag2 are fused at the N-terminus to the murine Ig κ-chain leader sequence for
protein secretion and at the C-terminus to the c-myc epitope and six tandem histidine residues
for detection and purification.
Due to the presence of the human cytomegalovirus intermediate-early promoter/enhancer
(PCMV) the plasmid becomes methylated in vivo and subsequently shuts down. Hence, the
substitution of another promoter was required. Therefore, PCMV was substituted with the
human elongation factor 1α (EF-1α) promoter, which was removed from the PEF/Bsd vector
(Invitrogen Life Technologies) (4.3 kb). This promoter remains functional within the mouse
and thus is optimal to drive the constructed plasmid and the production of the MC-2 peptide.
B FRAGMENT DETERMINATION
The PCMV fragment was excised from pSecTag2 using the BglII restriction site and NheI. Two
fragments resulted, 883bp and 4276bp, the latter being required. PEF-1 was excised using the
restriction enzymes NheI and Acc65I to create 2 fragments of 1291bp and 3013bp. The
former was essential for ligation to create a functional plasmid (Figure 28-Appendix).
In order to identify the required fragments, digestions were subjected to gel electrophoresis
and run on 1.2% Nusieve GTG Agarose. DNA was eluted from the gel using QIAquick gel
extraction kit according to the manufacturers instructions using a vacuum manifold.
Essentially, the kit allows recovery of DNA and the removal of contaminants by incubating
the gel in a buffer containing guanidine thiocyanate and adsorbing the DNA to a silica-
membrane on the spin column. Guanidine thiocyanate solubilizes the DNA, denatures
proteins and acts as a pH indicator to ensure optimal adsorption.
C BLUNTING
DNA blunting took place using the DNA blunting kit from TAKARA according to the
manufacturer’s instructions. This converts the 3’ and 5’ protruding ends of DNA fragments to
blunt or flush ends. To deactivate the T4 polymerase the mixture was vigorously vortexed and
then diluted to a concentration of 1µg/ 50 µl in TE buffer.
33
MATERIALS AND METHODS
D DEPHOSPHORYLATION
In order to minimise self-circularisation/ligation the plasmid fragment was dephosphorylated.
Shrimp alkaline phosphatase was used as, unlike the calf enzyme, it is completely and
irreversibly inactivated by Tris-buffers at pH 8.0-8.5 by simply heating for 15 minutes at
65°C. Using 3 µl of DNA termini, 0.2 µl of phosphatase at 37°C for 1 hour was the minimum
effective amount.
E LIGATION
The DNA fragment for insertion should be present in 5 fold molar excess relative to the
vector DNA. Ligation was carried out according to manufacturer’s instructions, TAKARA
BIO INC. Essentially, buffer containing T4 DNA Ligase enabled sufficient ligation efficiency
to be achieved at 25°C for 3 minutes. The resultant solution was aliquoted for bacterial
transformation.
The remaining ligation DNA was extracted by phenol/chloroform 1:1. Then 1/10 volume of
sodium acetate (3 M) and 2.5 volumes of ethanol were added. The solution was left at –20°C
overnight and the following day it was centrifuged at 14000 rpm for 30 minutes at 4°C. The
supernatant was removed and the precipitated DNA kept at -80ºC for long term storage.
F TRANSFORMATION
UltraMAXTM DH5α-FTTM (Invitrogen) competent cells were used for generation of the
plasmid DNA. One of the easiest ways to get large amounts of DNA is to place the desired
DNA into bacteria, grow and harvest the bacteria, and subsequently isolate the DNA.
Transformation efficiency of the ligated circular DNA into competent cells could be improved
by the addition of one tenth the volume of Transformation Enhancer provided by TAKARA
Bio Inc. before transformation. 100 µl of cells and 10 ng of ligation DNA were then incubated
on ice for 30 minutes. The cells were heat shocked in a water bath at 42°C and placed on ice
for 2 minutes. SOC medium (Super optimal broth plus glucose) was added and the resultant
34
MATERIALS AND METHODS
1ml mixture was placed at 37°C for one hour. The solution was then divided and spread on
LB agar plates.
2.2.8 PLASMID PREPARATIONS
A MINI AND MAXI PREPARATIONS
Colonies from the LB transformation plates were used to inoculate 5 ml of LB medium (mini)
or 200 ml of LB medium (maxi). These cultures were incubated at 37°C overnight. 750 µl of
bacteria was mixed with 150 µl glycerol to form stocks which were stored at –80°C. The
cultures were then centrifuged and the resultant pellet was treated according to the
manufacturer instructions for QIAprep Spin Miniprep Protocol (using a vacuum manifold) or
Qiagen plasmid maxi protocol, respectively.
B ENZYME DIGESTION
In order to ascertain whether the transfer of the EF1-α promoter had correctly inserted into
the pSecTag2 vector, 3 µl of plasmid from the mini-preps were digested with 10 Units of each
enzyme for 2h at 37°C. Digestion was with either AgeI/BamHI for fragment sizes of
1236bp/4331bp or BglII/BamHI for fragment sizes of 748bp/4819bp. To check the resulting
fragments 5 µl of sample was run on a 1% agarose electrophoresis gel for 1 hour and
visualised under a 354 nm UV-light.
C OLIGONUCLEOTIDE ANNEALING AND INSERTION INTO
PSECTAG2-EF1α
The murine IFN-γ oligonucleotides (34bp) (LRKRKRSR – MC-2 insert) or the neutral
oligonucleotide (34bp) were resuspended in annealing buffer. Equimolar volumes of
oligonucleotides were mixed and placed in tubes within a thermal cycler. A program was
constructed to heat to 95ºC and remain at 95ºC for 2 minutes and then ramp cool to 25ºC over
a period of 45 minutes. Tubes were spun in a microfuge to draw all moisture from the lid.
Later unused samples were stored at 4ºC.
35
MATERIALS AND METHODS
PSecTag2-EF1α was cut using the EcoR1 and BamH1 restriction sites between the T7 and
TAA region (Stop codon). This created two fragments of 16bp and 5551bp. As the
oligonucleotide had been engineered to have the corresponding ends for insertion into the
vector, ligation and transformation was carried out as in 2.2.7E/F. Plasmids were prepared
according to protocol 2.2.8A/B. These plasmids were intentionally misaligned so that the
excreted peptide fragment did not contain the Myc/HIS-Tag. A realigned plasmid was created
by cutting PSecTag2-EF1α containing the MC-2 oligonucleotide with EcoR1, and filling with
Klenow before re-ligation (Figure 6).
INSERT HIS/MYC TAG COMMON NAME
Murine IFN-γ (128-135) Yes J558L-pMC2-Tag
Murine IFN-γ (128-135) No J558L-pMC2-NoTag
Alanine8 No J558L-pAlanine8-NoTag
None Yes J558L-pTag
Table 1: Components within the plasmid transfected cell lines.
All plasmids contained the EF1-α promoter and the Igκ-leader sequence responsible for
extracellular excretion of the peptide fragment.
Figure 6: Arrangement of the completed plasmid. In
this case the plasmid can secrete the murine IFN-γ
fragment responsible for ECM binding The Myc/HIS-Tag
was optional. For more detailed plasmid map and
sequences review Figures 28 and 29 (Appendix).
36
MATERIALS AND METHODS
D PCR OF MC-2 INSERT
Principle: A powerful technique for amplifying small fragments of the genome is provided
by PCR. A preparation of DNA is denatured by heat and the single strands are annealed with
two short primer sequences that are complementary to sites on the opposite strands on either
the side of the target region. DNA polymerase is used to synthesise a single strand from the
3’-OH end of each primer. The entire cycle can then be repeated by denaturing the
preparation and starting again. The number of copies of the target sequence grows
exponentially. In practice, it doubles with each cycle until reaching a plateau at which more
primer-template accumulates than the enzyme can extend during the cycle; then the increase
in target DNA becomes linear. For all PCR’s the heat stable DNA-polymerase from Thermus
aquaticus (Taq-Polymerase) was used. It remains stable up to 95°C.
To ensure the correct insertion of the LRKRKRSR (MC-2) fragment into the plasmids PCR
was carried out. The temperature for annealing of the MC-2 primers had been determined by
the producers TIB® MolBiol to be around 66°C. However, when the PCR was run the
required band was not clear. Therefore, a temperature gradient PCR was run. It follows the
same principle as a normal PCR except that across the plate the annealing temperature in each
well increases in 0.8°C increments. It was determined that the strongest band for the MC-2
primers occurred at a temperature of 58°C.
Conditions using the MC-2 primers for the amplification of the MC-2 fragment from plasmid
and cellular DNA were: 94°C for 3 minutes; amplification by 40 cycles of 94°C for 40sec,
58°C for 40sec, 72°C for 40sec; elongation for 10 minutes at 72°C.
E SEQUENCING
To confirm the results of the PCR reactions, the sequences of pSecTag2-EF1α and the
plasmids containing the inserts, were determined by Invitek, Berlin.
2.2.9 TUMOUR CELL TRANSFECTION
J558L were transfected using the Gene Pulser® Electroporator and Electroprotocol from Bio-
Rad. The tumour cells were suspended in RPMI medium with 10% FCS at a concentration of
37
MATERIALS AND METHODS
1x107 cells/ml. 10 µl plasmid DNA (1 mg/ml) was suspended in TE buffer was added to 800
µl of cells. A cuvette gap of 0.4 cm was used and the voltage maintained between 0.2 to 0.4
kV. A field strength of 0.5 to 1 kV/cm and the capacitor at 960 µF was set. Cells were then
transferred to 5 ml RPMI-Medium in a T-25 flask and incubated overnight. The following day
1.5 mg/ml Zeocin was added to select for transfected cells.
2.2.10 ISOLATION OF GENOMIC DNA
Genomic DNA was isolated from 0.5 cm tail biopsies, blood and plasmid transfected tumour
cells for genotyping and analysis. Essentially samples were incubated in 700 µl lysis buffer
with 17.5 µl 20 mg/ml Proteinase K at 55°C overnight. A 0.5 ml solution of phenol-
chloroform-isoamyl alcohol (25:24:1) was added and the tube mixed well by repeated
inversion. After centrifugation at 14000 rpm for 5 minutes the aqueous phase was transferred
to a fresh tube and 0.5 ml chloroform-isoamyl alcohol (24:1) was added and the tube mixed.
The aqueous phase was again transferred to a fresh tube and 100% volume isopropanol added.
The tube was centrifuged and the supernatant removed. The pellet was then washed with 1ml
70% ethanol and later resuspended in 50 µl TE buffer.
2.2.11 ISOLATION OF RNA
Cells transfected with the 4 different plasmids were subject to total RNA isolation by
TRIzolTM reagent. RNA was prepared when 1x107 tumour cells were pelleted and
resuspended in 1ml of TRIzolTM. 0.2 ml chloroform per 1 ml TRIzolTM was added and the
tube vortexed for 15 seconds. Samples were then centrifuged for 10 minutes at 14000 rpm at
room temperature. The aqueous phase was transferred to a fresh tube and 0.5 ml of
isopropanol added and mixed by inversion. The sample was then centrifuged again at 12000 x
g but this time at 4°C for 15 minutes. The RNA was pelleted, the supernatant decanted, and 1
ml of ethanol added to wash the pellet. Once the ethanol was removed the pellet was air-dried
and resuspended in RNAse-free DEPC water.
The concentration of RNA was determined by diluting the recovered RNA 1:250 in a quartz
curvette for spectrophotometric analysis at 260 nm. An absorbance of 1 indicates around 40
38
MATERIALS AND METHODS
µl/ml RNA. To check the purity of the RNA a measurement at 280 nm was also carried out.
The quotient of the measurement between 260 nm and 280 nm was evaluated. An
OD260/OD280 greater than 1.6 indicated a clean preparation. The RNA was stored at –80°C.
2.2.12 RT-PCR
RT-PCR is a technique used for measuring gene expression in tissues and cultured cells.
Traditionally, RT-PCR is performed in two steps: a first strand cDNA synthesis step using
reverse transcriptase, followed by a PCR step using a thermostable DNA polymerase.
However, we used a one step procedure according to the manufacturer’s instructions. The
TITANIUM™ One-Step RT-PCR Kit allows cDNA synthesis and PCR to be performed in a
single optimised buffer with a single enzyme mix.
2.2.13 PROTEIN DETERMINATION
The concentration of protein in the samples from antibody production and tumour cells were
determined by the Bio-Rad DC Protein Assay, which is a colorimetric assay which functions
similarly to the Lowry assay.
Principle: The protein in the sample reacts with copper tartrate solution and Folin reagent.
Colour development is primarily due to the amino acids tyrosine and tryptophan and to a
lesser extent cystine, cysteine and histidine. Proteins effect a reduction of the Folin reagent by
the loss of 1, 2 or 3 oxygen atoms, thereby producing one or more of several reduced species
which have a characteristic blue colour with a maximum absorbance at 750 nm and a
minimum absorbance at 405 nm. Protein was determined by the microplate assay protocol,
essentially 5 µl of sample and standard was incubated with 25 µl of reagent A (alkaline
copper tartrate) and 200 µl of reagent B (dilute Folin reagent) and left for 15 minutes to
develop. The plate was then measured at 750 nm.
39
MATERIALS AND METHODS
2.2.14 IODINE125 LABELLED IFN-γ EXPERIMENTS
Iodine 125 is used to iodinate proteins on the cell surface and/or intracellular proteins. It is a
moderately volatile radioisotope, which emits relatively weak gamma radiation as it decays. It
is commonly used as a radiotracer in life sciences research as well as in clinical applications.
Principle: IODO-BEADS® Iodination Reagent is a N-chloro-benzenesulfonamide (sodium
salt) immobilsed on nonporous, polystyrene beads. Radioactive I125 can be incorporated into
protein by either enzymatic or chemical oxidation. IODO-BEADS is milder than the
traditional chloramine-T, generates sufficient radioactive iodine, and does not require a
reduction step, which makes it advantageous for maintaining biological activity of proteins.
IODO-BEADS allows easy separation of the reagent from the reaction mixture and allows for
a two-phase system, a more easily controlled reaction, which limits direct contact of the
oxidant with the protein.
A PROCEDURE FOR PROTEIN IODINATION AND PURIFICATION
Just before use, one or more beads were washed with 500 µl of reaction buffer per bead. This
wash step removes any loose particles and reagent. The beads were added to a solution of
carrier-free NaI125 (approximately 1 mCi per 100 µg of protein) diluted in PBS and allowed to
react for 5 minutes. The protein was dissolved or diluted in reaction buffer and added to the
Eppendorf tube, which was allowed to react for 2-15 minutes. The bead was removed from
the solution and the reaction subsequently stopped. The solution was then passed over D-
Salt™ Desalting Columns to remove excess NaI125 or unincorporated I125 from the iodinated
protein. The columns were equilibrated with 5 column volumes of PBS. The tip of the column
was placed in a test tube and the sample applied. The sample entered the gel and the column
stopped flowing. The tip of the column was placed in a new tube and a volume of buffer equal
to the fraction volume was added. After the buffer entered the gel, the column was transferred
to a new tube. This continued until the protein had emerged from the column.
40
MATERIALS AND METHODS
B MEASURING THE COUNTS AND QUANTITY OF RADIOACTIVE
PROTEIN
The counts of the extracted fractions were determined by the Print-Scale-Timer BF2306P.
Protein was visualised using a 12.5% SDS-PAGE gel to determine which fraction of the flow
through from the columns contained iodinated protein. Gels were run at 100V for 60 minutes
and then placed on a BIOMAX-MS film (Kodak) for 10 minutes. The film was then
developed according to the manufacturers instructions.
The quantity of protein could be monitored by measuring the absorbance of each fraction at
280nm. The absorbance depends on the presence of tyrosine and tryptophan in the protein and
this method is entirely advantageous as the sample can be recovered. The first peak in
absorbance will generally emerge when 1 void volume of buffer has been added after the
sample is applied. This peak is the protein. Molecules smaller than the exclusion limit of the
gel (i.e., buffer salts) will emerge from the column in subsequent fractions. These fractions
can be discarded after confirming that all fractions containing protein have been collected.
C IMMOBILISATION OF COLLAGENS AND MATRIGEL
The coating of microtitre plates and the calculation of coating efficiencies were performed as
described previously (Somasundaram et al., 2000). Native collagens were immobilised on
polystyrene microtitre plates at concentrations of 2µg/100µl/well for binding studies.
Immobilisation was done with 50 mM ammonium bicarbonate, pH 9.6, overnight at 4°C
followed by three washes with PBS, pH 7.4, with the exception of glycoproteins, which were
coated in PBS. Matrigel from an Engelbrecht Holm Swarm sarcoma was pre-coated 20
minutes at room temperature and then overnight at 4°C with 114 µg/well solution in D-PBS.
As 1-3% of Matrigel is made of heparan sulfate approximately 17 µg/ml coated the wells.
Non-specific binding sites were blocked with PBS-T for 2 hours at room temperature.
For binding studies 0.1-10 ng of I125-IFN-γ in PBS-T was added to the collagen or
matrigel coated wells and incubated for 2 hours at 4°C. Finally, after three washes in binding
buffer, radioactivity bound to the coated wells was measured using a gamma counter.
41
MATERIALS AND METHODS
For inhibition studies matrigel wells were incubated with either MC-2 peptide (1ng-1mg),
supernatants from transfected tumour cell lines or buffer overnight at 4°C. They were washed
with blocking buffer and then radioactive IFN-γ (7.5 ng/ml or 4x10-7 mol/L) was added to
wells to incubate from 2h to overnight at 4°C. The supernatants were discarded and the wells
washed three times with buffer (PBS-T). Bound IFN-γ was then was measured with a gamma
counter.
D ANIMALS TREATED WITH I125-IFN-γ
Adult female BALB/C interferon gamma competent and deficient mice were housed in the
Nuclear Medicine facility of the Benjamin Franklin Clinic on a 12 hour light/dark cycle, with
ad libitum access to food and water. After mice had grown J558L or transfected tumours over
a 8-10 day period they were injected intravenously (iv) through the tail vein with 5 µg of IFN-
γ containing counts between 100000 and 500000 cpm, depending on the success of protein
iodination, as a tracer. Mice were sacrificed at selected times (1h, 24h, 72h after injection),
and organs extracted (heart, lung, liver, spleen, kidney and tumour).
E MEASUREMENT OF THE ORGAN DISTRIBUTION OF I125-IFN-γ
Samples of tissues, excised at the various time points after I125-IFN-γ, were weighed and
counted. For each organ the radioactivity per gram was calculated as a percentage
accumulation of I125-IFN-γ.
42
RESULTS
3 RESULTS
3.1 GENOTYPING INTERFERON GAMMA
KNOCKOUT MICE
Mice with a non-functional IFN-γ gene were generated by Dalton and colleagues, 1993, and
obtained from Jackson Laboratories (USA). Mice were created by replacing one normal IFN-γ
allele with a defective allele in mouse embryonic stem cells. The targeting vector had a 2kb
neomycin resistance gene inserted into exon 2, which introduced a termination codon after the
first 30 amino acids of the mature IFN-γ protein. Mice appear to have no gross or histological
abnormalities, and no alterations of splenic and thymic populations. Knock-out mice remain
normal, healthy and fertile (Dalton et al., 1993). Thus, IFN-γ is not required for the
development of the immune system but is essential for the immune response.
Mice were crossed heterozygote to homozygote. In order to correctly type the IFN-γ+/-,
BALB/c and IFN-γ-/- BALB/c mice by PCR, genomic DNA from the tail was prepared and
the IFN-γ gene amplified. Mice that contained the neomycin resistance gene in exon 2
generated a PCR band of 320 bp, whilst mice with DNA for an intact IFN-γ gene produced a
PCR band of 260 bp.
B
260 bp
1 2 neo
r
3 4
1 2 3 4
A
Figure 7: PCR analysis of genomic DNA for typing IFN-γ mice.
M 1 2 3 4 5 6 7
320 bp
(A) IFN-γ gene structure on a genomic fragment. Primers were derived from exon 2. Insertion
of the 2-kb neor gene into exon 2, meant new primers were derived from the neomycin gene.
(B) IFN-γ gene 260 bp, neomycin disrupted gene 320 bp. Lane M Marker, Lanes 1-2 BALB/c
mouse, Lanes 3-4 IFN-γ+/- mouse (both bands), Lanes 5-6 IFN-γ-/- mouse (neo band only),
Lane 7 H2O.
43
RESULTS
3.2 THE EFFECT OF INTERFERON GAMMA ON
TUMOUR GROWTH
In several studies, IFN-γ has been associated with tumour rejection. To investigate this, the
influence of IFN-γ produced by tumour cells rather than host cells was investigated. The
J558L parental line had previously been transfected with a plasmid encoding IFN-γ.
Subcutaneous injection of J558L-IFN-γ producing cells demonstrated that IFN-γ inhibited the
outgrowth of the tumour in BALB/c mice whereas the parental cell line grew rapidly in
immuno-competent hosts.
Effective neutralisation of IFN-γ is important to determine the role it plays when derived from
the host. Therefore the ability of the neutralising antibody XMG6 was tested in a model where
J558L cells were known to produce a specific concentration of IFN-γ (120ng/ml). In this
experiment, 1mg/mouse of XMG6 (Isotype – Rat IgG1) was administered 1 day before
tumour cell injection. From the kinetics of the growth curve in Figure 8, the antibody was
capable of completely neutralising IFN-γ in vivo and allowed the transfected cells to grow
with similar kinetics compared to the parental line.
Figure 8: IFN-γ inhibits tumour growth.
BALB/c mice were inoculated with 1 x 106 J558L ( ) or 5 x 106 J558L-IFN-γ ( ) on day
0. Another group of BALB/c were neutralised for IFN-γ using 1mg/mouse XMG6 antibody
on day -1 and then inoculated with 5 x 106 J558L-IFN-γ ( ) on day 0. Tumour growth was
monitored over 30 days (n=5/group). Single experiment. NB. Indicates mouse removed from
experiment.
0
0,5
1
1,5
2
0 5 10 15 20 25 30
TUMOUR SIZE
-
(cm)
TIME - (Days)
44
RESULTS
3.3 CYCLOPHOSPHAMIDE INDUCES IFN-γ
DEPENDENT REJECTION OF J558L TUMOURS
3.3.1 WILD TYPE MICE BUT NOT IFN-γ KNOCKOUTS CAN
REJECT J558L TUMOURS.
The previous experiment demonstrated that IFN-γ inhibits tumour growth. Therefore, the role
of IFN-γ in Cy-mediated tumour rejection was analysed. Under certain conditions
cyclophosphamide can induce rejection through host cell modulation rather than direct
tumouricidal activity. It was shown that T cell deficient mice were incapable of rejecting a
solid tumour of J558L plasmacytoma cells, by this mode of treatment. However BALB/c
mice, which were treated similarly with a single injection of 15mg/kg of Cy 9-11 days after
tumour injection, rejected the tumour mass. Three days after Cy treatment, these tumours
became severely necrotic in immunocompetent mice and were rejected within 10-20 days
(Figure 9A).
Recently, it was demonstrated by Ibe and co-workers, 2001, that expression of IFN-γ−R on
host cells but not tumour cells was required for Cy-mediated tumour rejection. Therefore, Cy
was tested in a murine IFN-γ knockout model. The tumours grew similarly in this model, and
mice were treated when tumours ranged between 0.6 - 1cm. 11 days after inoculation with
J558L cells, the tumour was treated with Cy. IFN-γ-/- mice were unable to reject whereas IFN-
γ+/- mice rejected tumours. The tumours of the IFN-γ-/- mice did appear to reduce slightly in
size however, they eventually grew out aggressively (Figure 9B). All mice had succumbed to
tumour burden by day 21 and had to be euthanized.
From histological stainings of J558L tumours in Cy treated IFN-γ+/- and IFN-γ-/- mice there
appears to be rapid destruction of the endothelial vasculature (Figure 10). Approximately 24
hours after Cy treatment CD31 positive blood vessels in J558L tumours of IFN-γ+/- mice have
become shorter and appear to be less densely arranged when compared to IFN-γ-/- tumour
tissue. By 72 hours after Cy treatment tumour vasculature appears to be single dots with a few
small vessels intact.
45
RESULTS
A
0
0,5
1
1,5
2
0 5 10 15 20 25 30
TUMOUR SIZE - (cm)
TIME - (Days)
B
Figure 9: IFN-γ is required for Cy mediated tumour rejection.
Tumours were established by subcutaneous injection of 1x 106 J558L cells into mice. 11 days
later, when tumours reached a size of around 1 cm in diameter, mice were treated
intraperitoneally with 15 mg/kg Cy. (A) IFN-γ+/- mice ( , n=10) with treatment, IFN-γ+/-
mice ( , n=10) which had not received Cy treatment served as controls. All Cy-treated
BALB/c mice rejected the tumour. (B) IFN-γ-/- mice treated with Cy ( , n=10). IFN-γ-/-
mice treated with No Cy ( , n=10). Representative of 3 experiments. NB. Indicates
mouse removed from experiment.
0
0,5
1
1,5
2
0 5 10 15 20 25 30
TIME - (Days)
TUMOUR SIZE - (cm)
46
RESULTS
IFN-γ +/- IFN-γ -/-
D
A
Untreated
B E
24 hours
C F
72 hours
Figure 10: Cy induces IFN-γ dependent destruction of the vasculature. Tumours were
established in IFN-γ+/- and IFN-γ-/- BALB/c mice by subcutaneous injection of 1 x 106 J558L
cells. 11 d later, mice were injected intraperitoneally with 15 mg/kg Cy and tumours were
excised before, 24 h, and 72 h after Cy treatment. Immunohistochemical analysis of tissue
sections from IFN-γ+/- mice (A–C) and IFN-γ-/- mice (D–F) was performed with mAb CD31
(A-F). Scale bar equals 100µm. A representative staining of tumours from 3 mice per group.
47
RESULTS
3.3.2 NEUTRALISATION OF IFN-γ PRODUCTION IN WILD
TYPE MICE EFFECTS TUMOUR REJECTION.
Mice with targeted gene deletions, have a specific gene removed or inactivated. They are
unable to produce the gene product throughout the life of the animal. There are significant
advantages such as avoidance of inadequate dosing, specificity, affinity or penetration of
drugs and the lack of response to exogenous antibodies and soluble receptors. However, the
lack of clinical relevance arises from the absolute absence of the gene throughout the animals
life and the possible indirect effects such as the over compensatory expression of other genes
that can not be controlled.
The most effective way to validate the role of a particular cytokine in disease is by specific
blockage of its activity in a complex model. This works as a temporary blockade of a specific
signalling pathway. The results of the knockout model were confirmed by neutralisation of
IFN-γ using the anti-IFN-γ antibody XMG6. The mAb antibody was prepared from
hybridoma supernatant by the CELLineTM system and purified by HPLC to generated 76 mg
of XMG6 antibody.
The minimal neutralisation dose in which all J558L-IFN-γ producing cells grew in wild type
mice had been previously established to be 1 mg/mouse of anti-IFN-γ mAb (Figure 8). Wild
type mice were separated into groups of animals treated with (day -1 alone or day -1, day 5,
day 10) and without XMG6 mAb. On day 0 all animals received 1x106 J558L tumour cells.
Tumours in IFN-γ mAb treated animals grew with similar kinetics to the wild type untreated
controls. At day 11 all groups were treated with cyclophosphamide (15 mg/kg) and the
response monitored. Figure 11 demonstrates that neutralisation of IFN-γ prevents rejection of
tumours. All mice that were not treated with cyclophosphamide from each group (n=3),
including mice neutralised for IFNγ and injected with 1x106 J558L-IFNγ cells (n=9), grew out
by day 22.
48
RESULTS
Figure 11: The neutralisation of IFN-γ inhibits J558L tumour rejection induced by
cyclophosphamide.
IFN-γ competent mice were treated with the neutralising antibody XMG-6 (1 mg/mouse).
Mice were given a single injection of antibody on day -1 ( , n=6) or multiple injections on
day -1,5 and 10 ( , n=6 ). As control XMG-6 untreated IFN-γ+/- mice ( , n=6 ) and
IFN-γ-/- mice ( , n=6) were also included. Mice in these four groups were injected with
1x106 J558L tumour cells on day 0 and treated with cyclophosphamide (15 mg/kg) on day 11.
Additionally, IFN-γ+/- mice were treated with 1 mg of XMG on day –1 and injected with
1x106 J558L-IFNγ cells on day 0, there was no Cy treatment in this group ( ).
Representative of 2 experiments.
TIME - (Days)
0
0,5
1
1,5
2
0 5 10 15 20 25 30
TUMOUR SIZE - (cm)
49
RESULTS
3.3.3 IFN-γ IS RELEASED INTO THE SERUM AFTER
CYCLOPHOSPHAMIDE TREATMENT
The short in vivo lifespan of IFN-γ makes it difficult to measure its concentration. An assay
was developed to measure cytokine production in vivo (Finkelman and Morris, 1999). The
amount of cytokine measured is directly proportional to the amount produced and relatively
independent of the site of cytokine production. Mice were injected with 10 µg of a biotin-
labelled neutralising rat IgG anti-IFN-γ mAb (R4-6A2). The antibody is able to capture the
cytokine to produce a complex that has a relatively long in vivo half-life, and consequently
accumulates in the serum.
Six hours after treatment of the tumours with Cy there is a little difference in systemic IFN-γ
followed by increasing amounts of cytokine from 24-72 hours (Figure 12). The result shows
that three days after Cy treatment there is double the amount of IFN-γ in the serum compared
to naïve mice.
IFNγ concentration pg/ml
0
200
400
600
800
1000
1200
A B C 0-6h 6-24h 24-48h 72-96h
TIME - after Cy treatment
Figure 12: IFN-γ is released into the serum after Cy treatment.
Serum containing murine IFN-γ bound to a biotinylated detection antibody was isolated from
mice. C57BL/6 tumour free (A), BALB/C tumour free (B), and day 10 tumour bearing (C)
mice without Cy treatment served as baseline controls. Mice in groups (0-6h) were given the
antibody day 11 just before treatment with Cy on day 11. In the other groups antibody was
given on the initial time point and extracted on the final time point. Serum was analysed at
various time points after treatment by an IFN-γ ELISA developed by Finkelman and Morris
(n = 4/group). Representative of 1 experiment.
50
RESULTS
3.3.4 HAEMATOPOIETIC CELLS PRODUCE IFN-γ REQUIRED
FOR TUMOUR REJECTION
In experiments aimed at determining the origin of IFN-γ required for rejection of J558L
tumours, bone marrow chimeras were made using the IFN-γ BALB/c heterozygous and
homozygous negative mice. All four possible combinations were made (IFN-γ-/- BM to
IFN-γ-/- host, IFN-γ-/- BM to IFN-γ+/- host, IFN-γ+/- BM to IFN-γ-/- host, IFN-γ+/- BM to IFN-
γ+/- host). In bone marrow chimeras all blood borne cell lineages were derived from donor
cells. This was confirmed first by PCR from blood (Appendix - Figure 30), and then by ConA
stimulation of PBMC’s followed by a murine IFN-γ ELISA. The reconstituted, host derived
lymphoid compartments contained IFN-γ producing cell populations as determined after
ConA stimulation in Figure 13.
0
100
200
300
400
500
600
700
1
Amount of IFNγ (pg/ml)
Figure 13: Haematopoietic cells produce IFN-γ upon Con-A stimulation.
Whole blood was removed from individual mice of each group and red blood cells were
lysed. The remaining cells were washed and incubated with Con-A for 72 hours. Supernatants
were then subjected to an ELISA and the quantity (pg/ml) of IFN-γ determined. Groups
include WT to WT , KO to KO , WT to KO , KO to WT - Each group n = 3.
In general, J558L tumours grew with similar kinetics when compared with their unirradiated
counterparts. At day 11 all tumours were 0.8 cm in diameter with a standard deviation of 0.25
cm. The BMC tumour response to Cy is shown in Table 2. Four days after cyclophosphamide
injection those mice that received bone marrow from IFN-γ competent mice began to reduce
51
RESULTS
tumour burden. This was seen through a reduction of tumour size and central necrosis of the
tumour. Tumours in these mice were characterised by a bloody core and 3-7 days after
cyclophosphamide, holes became visible within the tumour mass (Figure 14).
A B
Figure 14: Mice with haematopoietic cells capable of producing IFN-γ reject a J558L
tumour.
IFN-γ competent mice, which bare 11 day old J558L tumours, were treated with Cy (15
mg/kg). Rejection was characterised by central necrosis and scab formation 3-7 days after Cy
(A) and complete destruction 10-15 days after Cy (B).
IFN-γ+/- to IFN-γ+/- IFN-γ+/- to IFN-γ-/- IFN-γ-/- to IFN-γ-/- IFN-γ-/- to IFN-γ+/-
CYCLOPHOSPHAMIDE
Experiment 1 6/6 6/7 0/5 0/6
Experiment 2 5/6 6/7 0/7 0/6
NO CYCLOPHOSPHAMIDE
Experiment 1 0/2 0/2 0/3 0/2
Experiment 2 0/2 0/2 0/2 0/2
Table 2: Haematopoietic cells produce IFN-γ that is required for Cy mediated tumour
rejection.
The table shows a summary of the number of mice, which rejected the tumour after Cy
treatment. IFN-γ bone marrow chimeras were made by irradiating competent and knock out
mice with 10 Gy and then transferring 2x106 bone marrow cells. After 12 weeks the chimeras
were tested for IFN-γ gene locus by PCR of tail and blood (Appendix - Figure 30). Mice were
injected with 1x106 J558L tumour cells on day 0 and treated with cyclophosphamide (15
mg/kg) on day 11.
Tumours of mice that received bone marrow unable to produce IFN-γ continued to grow after
Cy treatment. A few mice in the IFN-γ-/- haematopoietic to IFN-γ+/- non haematopoietic group
had a reduction in tumour burden of approximately 0.1 cm - 0.3 cm four days after Cy
treatment but shortly after continued to grow (Figure 15). The IFN-γ-/- haematopoietic to IFN-
52
RESULTS
γ+/- non haematopoietic group indicated that there was a small amount of IFN-γ that was
influencing the response of the mice. It was speculated that this could have been non-
haematopoeitic cell derived or from the matrix. Therefore, the relationship of matrix bound
cytokine was further investigated. All mice that were not treated with cyclophosphamide had
maximal tumour burden between days 20-25 and had to be euthanised.
A B
0
0,4
0,8
1,2
1,6
2
0 1020304050
0
0,4
0,8
1,2
1,6
2
0 1020304050
TUMOUR SIZE
-
(cm)
C D
0
0,4
0,8
1,2
1,6
2
0 10203040
0
0,4
0,8
1,2
1,6
2
0 1020304050
TIME - (Days)
Figure 15: Haematopoietic cells produce IFN-γ that is required for Cy mediated tumour
rejection.
Kinetics of tumour growth in IFN-γ+/- and IFN-γ-/- mice, which were irradiated with 10 Gy
and injected i.v. with 2x106 cells of bone marrow isolated from the femurs of IFN-γ deficient
or competent mice. Four groups of bone marrow chimeras were generated. (A) IFN-γ-/- BM to
IFN-γ-/- host, (B) IFN-γ-/- BM to IFN-γ+/- host, (C) IFN-γ+/- BM to IFN-γ-/- host, (D) IFN-γ+/-
BM to IFN-γ+/- host. After 12 weeks mice were given 1x106 J558L tumour cells and treated
11 days later with Cy (15mg/kg). Individual mice treated with Cy bold circles. Untreated mice
in dotted circles.
53
RESULTS
3.4 IFN-γ CO-LOCALISES TO HEPARAN SULPHATE
PROTEOGLYCAN IN VIVO AND BINDS HEPARAN
SULPHATE GLYCOSAMINOGLYCAN IN VITRO
It has been shown that in vitro IFN-γ can bind to HS found in the extracellular matrix
(Fernandez-Botran et al., 2004; Subramaniam et al., 1999). This association of IFN-γ has not
been validated in vivo. Therefore, the presence of this matrix bound cytokine within the
tumour microenvironment was examined due to the rapid destruction of the vasculature after
Cy treatment (Figure 10) and retarded outgrowth of tumours in the bone marrow chimeric
group where haematopoietic cells could not produce IFN-γ (Figure 5).
3.4.1 IFN-γ CAN CO-LOCALISE WITH HEPARAN SULPHATE
IN GROWING TUMOURS
To determine whether IFN-γ might associate with elements of the extracellular matrix in vivo,
we microscopically examined various tumours (J558L, J558L-IFN-γ) excised from mice for
evidence. BALB/c, IFN-γ deficient BALB/c and SCID mice were inoculated with 1x106
tumour cells. Tumours were then extracted at day 5 and 10 and prepared for histology.
Optimal conditions were determined for sectioning and staining the tumours. Similar to the
experiments carried out by Van der Loos and colleagues, 2001, four murine monoclonal
(XMG1.2, XMG6, R46A-2, F1) and two polyclonal IFN-γ antibodies were screened, as were
the fixatives used on the tissues. It was eventually determined that little difference could be
seen between acetone, paraformaldehyde and non fixed tissues. Two monoclonal antibodies
(XMG1.2 and R46A-2) showed poor but detectable staining in light microscopy and therefore
were considered for immunofluorescence co-localisation by confocal microscopy.
IFN-γ was detected using confocal microscopy (Figure 16) and specific staining was
confirmed by finding a lack of specific fluorescence in tissues from mutant mice deficient in
IFN-γ. Similarly, all sections with isotype control antibodies (IgG1-FITC) showed no
fluorescence under the confocal microscope.
54
RESULTS
The distribution of IFN-γ was limited in tumours from IFN-γ WT mice. Those tumours that
arose from haematopoietic cells (J558L) had detectable levels of IFN-γ in sections. The
pattern of staining was weak and widely dispersed in different areas of the tumour. However,
the most significant signal was generated from SCID mice that had J558L-IFN-γ producing
tumours and 129/Sv/Ev mice with MC51.9-IFN-γ producing tumours. As there was a strong
signal within these tumours, the parameters of the confocal were adjusted to identify only
those areas which had especially high deposits of IFN-γ. The staining was characterised by
single strong deposits with dull positive areas encircling.
Heparan sulphate was detected in all tumours in all mice (n=20). It was fairly evenly
distributed around the tumour. Nevertheless, within the IFN-γ knockout mice the deposits
were shorter and thicker than those seen in the wild type mice.
There was clearly binding of IFN-γ to heparan sulphate but it was restricted to some spots
around the tumour. Additionally, there were areas where IFN-γ was shown in close proximity
with what could be the nucleus (TOTO-3 - blue) as seen in Figure 16A. This possibly
indicates cellular accumulation of IFN-γ. The diffusion of interferon gamma into the matrix
can be seen through interaction with heparan sulphate. The strength of co-localisation is seen
by the yellow/orange colour in the overlaid images.
55
RESULTS
2
4
3
1
Figure 16 A: Matrix bound IFN-γ.
Interferon gamma (FITC) in association with heparan sulphate (TEXAS RED) in day 10
J558L-IFN-γ tumours in SCID mice. Nuclei are stained with TOTO-3 (blue). 1: Interferon
gamma, 2: Heparan Sulphate; 3: Nuclei; 4: Overlay. Representative of 3-4 mice per group.
Scale bar 50 µm.
56
RESULTS
3 4
1 2
Figure 16 B: Matrix bound IFN-γ.
Interferon gamma (FITC) in association with heparan sulphate (TEXAS RED) in day 10
MC51.9-IFN-γ tumours in wild type mice. Nuclei are stained with TOTO-3 (blue). 1:
Interferon gamma, 2: Heparan Sulphate; 3: Nuclei; 4: Overlay. Representative of 3-4 mice per
group. Scale bar 50 µm.
57
RESULTS
4
3
1 2
Figure 16 C: Matrix bound IFN-γ.
Interferon gamma (FITC) in association with heparan sulphate (TEXAS RED) in day 10
J558L tumours in wild type mice. Nuclei are stained with TOTO-3 (blue). 1: Interferon
gamma, 2: Heparan Sulphate; 3: Nuclei; 4: Overlay. Representative of 3-4 mice per group.
Scale bar 50 µm.
58
RESULTS
3.4.2 IFN-γ INFLUENCES THE MATRIX PRODUCTION OF
TUMOURS
Not only does IFN-γ bind to the matrix, but its presence can influence matrix structure.
Tumours have been described as wounds that do not heal (Dvorak, 1986), as there are many
parallels between tumour stroma generation and wound healing. In wound healing the
presence or absence of certain cytokines has been shown to influence the formation of scar
tissue (Azouz et al., 2004). While the majority of the components of the extracellular matrix
appear to remain the same in both the knock out and the wild type mice, the structural
formation of each matrix element varies.
To support the significance of the differences seen in the J558L model, the TS/A mammary
carcinoma was additionally analysed as it grows with similar kinetics in BALB/c mice. Mice
were injected with 1x106 cells of either tumour type and the tumours were extracted and
frozen for sectioning on day 10. Being of non-haematopoietic origin the TS/A tumours tended
to have larger quantities of matrix than the plasmacytoma J558L (grown as a solid tumour),
therefore the signal from histology was stronger due to the density of the matrix molecules.
The pan fibroblast marker, ERTR-7, was used to determine the distribution of fibroblasts
within growing tumours. In both the J558L and TS/A models there was dense staining in wild
type mice and tendrils were long and thin. These tended to be slightly thicker in the IFN-γ-/-
mice (Figure 17A-A & 17B-A).
The signal for heparan sulphate proteoglycan was markedly reduced in the wild type mice in
comparison with the IFN-γ-/- mice. Within the WT group the structures were very fine, long
and thin. The knockout mice had denser structure formations and these were a little more
restricted in distribution within the TS/A model. In general, there appeared to be more
heparan sulphate in the TS/A model.
Within the TS/A tumour collagen II had similar coverage in both WT and IFN-γ-/- groups. The
significant difference was that tendrils of collagen were long and wispy in the WT but short
and stubby in the IFN-γ-/- mice. There were larger amounts of deposited collagen II in WT
mice with J558L tumours compared with the tumours of IFN-γ-/- mice. However, the finding
of long tendrils in WT and shorter tendrils in knockout remained the same.
59
RESULTS
The most significant difference between the WT and IFN-γ-/- tumours was seen with collagen
IV. The staining was consistent in both the J558L and the TS/A model. Collagen IV is a
staple of basement membranes that make up the support for the blood vessels within the
tumours. The WT tumours were covered completely in thick, long, intertwined sheets of
collagen IV. Knockout tissues, while still having significant staining, consisted primarily of
shorter tendrils and dots of collagen IV.
60
RESULTS
IFN-
γ
+/- IFN-
γ
-/-
A
B
C
D
Figure 17 A: J558L matrix differences between wild type and IFN-γ knockout mice.
Mice were injected with 1x106 J558L tumour cells. Sections were made 10 days later
tumours were excised and cut. Tumours were stained for (A) ERTR-7; (B) Heparan sulphate
proteoglycan; (C) Collagen II; (D) Collagen IV, and then counterstained with Haematoxylin.
Scale Bar 100µm. Representative of 3-4 mice per group
61
RESULTS
IFN-
γ
+/- IFN-
γ
-/-
A
B
C
D
Figure 17 B: TS/A matrix differences between wild type and IFN-γ knockout mice.
Mice were injected with 1x106 TS/A tumour cells. Sections were made 10 days later tumours
were excised and cut. Tumours were stained for (A) ERTR-7; (B) Heparan sulphate
proteoglycan; (C) Collagen II; (D) Collagen IV, and then counterstained with Haematoxylin.
Scale Bar 100µm. Representative of 3-4 mice per group.
62
RESULTS
3.4.3 EXOGENOUS HUMAN IFN-γ CAN BIND TO MATRIGEL
PLUGS IN VIVO
Matrigel is a purified form of basement membrane from the Engelbreth-Holm-Swarm (EHS)
sarcoma (Lortat-Jacob et al., 1991). It contains collagen type IV, laminin and heparan
sulphate proteoglycans. It can provide a suitable biologically relevant environment where the
matrix is devoid cells. Previously, it had never been demonstrated that IFN-γ and heparan
sulphate could interact in vivo. In order to confirm the confocal histological analysis, matrigel
(0.2 ml) was injected subcutaneously into IFN-γ-/- mice. After 5 days, mice were treated with
200 µg of recombinant human IFN-γ. Upon removal of the plug, vasculature was
macroscopically visible throughout. The amount of IFN-γ detectable by ELISA had decreased
24 hours after the first matrigel extraction time point of 1 hour. The detection level was
limited to 1.6ng/ml (Figure 18).
0
10
20
30
40
50
60
70
80
90
1 hour 24 hours
IFN-γ Accunulation(ng/ml)
TIME
Figure 18: IFN-γ can bind to Matrigel plugs in vivo.
Five days after matrigel injection, mice received 200µg recombinant human IFN-γ, 1 hour or
24 hours later gel fragments ( ) were extracted and mulched with PBS. Serum IFN-γ levels
( ) were also analysed at the corresponding time points (n=3). Matrigel from mice that were
not treated with recombinant human IFN-γ had undetectable levels of cytokine.
63
RESULTS
3.5 THE KRKRS SEQUENCE OF THE CYTOKINE
IFN-γ INFLUENCES IN VITRO AND IN VIVO
RESPONSES.
3.5.1 IODINATION OF IFN-γ
It is known that IFN-γ has a dissociation constant for heparan sulphate similar to its affinity
for its receptor (Lortat-Jacob and Grimaud, 1992). As the competition for matrix free sites
within the tumour had never been analysed it seemed appropriate to investigate this
relationship. Human IFN-γ contains the conserved KRKRS sequence responsible for matrix
binding and, whilst it is capable of electrostatic interaction with the murine heparan sulphate,
it is incapable of binding to the murine IFN-γ receptor (Lortat-Jacob et al., 1996a).
Human IFN-γ was iodinated using the chloramine-T method primarily because the sites of
iodination (tyrosines) were not closely associated with the specific binding sequence. Purified
bound protein was obtained and identified in samples of the desalting column washes which
were run on a protein gel (Figure 19)
Unbound I-125
3 4 5
17kD
Figure 19: Radioactive labelling of IFN-γ.
Bands on an SDS gel of I125 IFN-γ from the different eluent fractions off a desalting column.
Lanes 3-5 show single bands of bound IFN-γ protein, Lanes 6-10 show unbound radioactivity.
Representative of 2 experiments.
64
RESULTS
3.5.2 IN VITRO INHIBITION
A RADIOLABELLED IFN-γ CAN BIND TO MATRIX
In vitro the best way to determine the binding of human I125 IFN-γ to heparan sulphate within
a tumour was to analyse its association with matrigel isolated from an EHS sarcoma. Whilst
matrigel contains basement membrane components such as laminin and collagen IV, it is also
composed of 1-3% heparan sulphate. Radioactively labelled cytokine was incubated in wells
with matrigel and individual matrix components such as Collagen I, III, IV, VI, XIVα and
fibronectin. Significant binding was seen only in wells containing matrigel and the optimal
concentration of IFN-γ binding per well was determined to be 5 ng – 7.5 ng using I125 IFN-γ
(Figure 20).
0
2000
4000
6000
8000
10000
12000
14000
16000
0 20 40 60 80 100 120
COUNTS (CPM)
IFN-
γ
(
n
g)
Figure 20: Radioactively (I125) labelled human recombinant IFN-γ can bind to EHS
matrigel.
Wells were incubated with 114 ug per well of matrigel overnight. After washing wells were
incubated for 24 hours with varying concentrations of I125 IFN-γ, washed again and counted.
Representative of 3 experiments.
65
RESULTS
B THE MURINE PEPTIDE CAN PREVENT BINDING OF IFN-γ TO
MATRIGEL
The basic sequence of amino acids at the C-terminus of the IFN-γ molecule are essential for
matrix binding (Fernandez-Botran et al., 2004) and for mediating receptor interaction (Sadir
et al., 1998). Utilising the murine peptide MC-2 (LRKRKRSR) in vitro inhibition of human
I125-IFN-γ binding to the matrigel complex was observed. Inhibition of 50% was seen using
400µg of the peptide and 1000µg of peptide caused 80% inhibition of radio-labelled cytokine
interaction with the matrigel (Figure 21). The molar ratio of IFN-γ, 4 x10-7 mol/L, to heparan
sulphate within the ECM gel 17 µg/ml was comparative to previously published reports.
0
2000
4000
6000
8000
10000
0 200 400 600 800 1000 1200
COUNTS
(
CPM
)
MC-2
p
e
p
tide
(
µ
g)
Figure 21: The murine peptide MC-2 (LRKRKRSR) can inhibit 7.5 ng/ml human I125-
IFN-γ binding to matrigel.
Matrigel coated wells were incubated with MC-2 peptide dissolved in PBS ( ) and the
MC-2 dissolved in 100% DMSO ( ). The negative charge of the DMSO interferes with
the effect of the peptide. Representative of 3 experiments.
66
RESULTS
C MC-2 PEPTIDE IN THE SUPERNATANTS FROM TRANSFECTED
TUMOUR CELLS CAN INHIBIT IFN-γ BINDING
The J558L tumour cell line was transfected with four different plasmids to create, 2 cell lines
that were capable of secreting the murine peptide and 2 cell lines that acted as negative
controls. At all times the transfected cell lines (bulk culture) were kept under selection with
ZeocinTM. The supernatants were conditioned by 4x106 cells/ml of RPMI medium overnight.
They were then collected and analysed for their ability to inhibit binding of human I125-IFN-γ
to matrigel coated wells. The results indicated that both tumour lines containing plasmids for
the secretion of MC-2 into the supernatant were indeed producing significant quantities of
MC-2 peptide, independent of the presence of a HIS-Tag. In both of these tumour lines IFN-γ
binding was inhibited by approximately 50%. The control supernatant with HIS-Tag alone
showed minimal difference to the parental untransfected J558L supernatant, while low level
inhibition was seen in the tumour line secreting the neutral alanine peptide fragment into the
supernatant. The results were titratable as seen in Figure 22.
0
2000
4000
6000
8000
10000
1 to 1 1 to 2 1 to 4
COUNTS (CPM)
J558L Tumour Supernatant Dilution
Figure 22: MC-2 from tumour cell supernatants inhibit human I125-IFN-γ binding to
plate bound matrigel.
Supernants were incubated for 24 hours on ECM gel coated plates at 4°C at various dilutions
in RPMI. Groups include; Parental – untransfected ( ), J558L – pMC-2NoTag ( ),
J558L – MC-2Tag ( ), J558L –pTag ( ), J558L – pAlanine8 NoTag ( ) tumour cells.
67
RESULTS
3.5.3 IN VIVO INHIBITION
A RADIOLABELLED IFN-γ CAN BIND TO MATRIX IN VIVO
It has been previously been shown that 20µg of radioactive IFN-γ injected into rats was
enough to determine the specific counts and location of the protein. As the IFN-γ-/- and wild
type mice are much smaller than the rat, the amount of protein thought to give similar results
was reduced to 5µg per mouse in order to get a similar result. Organs were extracted 1 hour,
24 hours and 72 hours after injection, however the most significant results were seen after the
1 hour time point. This corresponds to previously published reports where it was shown the
half life of IFN-γ was increased from 1.1 minutes to 99 minutes when bound to heparin
(Lortat-Jacob et al., 1996a).
0
2
4
6
8
10
12
14
Wild type Knockout
%IFN-γ ACCUMULATION
Figure 23: Distribution of I125 labelled human IFN-γ 1 hour after injection (%
Dose/gram).
Mice from both WT and IFN-γ-/- groups (n=3) were injected with 5 µg of I125 labelled IFN-
γ on day 5 or day 8 of tumour growth. One hour later blood and organs were extracted for
cpm analysis. Day 5 tumour burden ( ); Day 8 tumour burden ( ), representative organ -
spleen ( ) day 8.
The differences in the amount of radioactive IFN-γ that had accumulated per gram of tumour
can be seen in Figure 23. It was clear that between organs on both day 5 and day 8 there was
little difference in both the WT and IFN-γ-/- groups (spleen as representative organ other data
not shown). However, the tumours of the IFN-γ-/- groups, on both days, had larger amounts of
68
RESULTS
accumulated IFN-γ. This indicated that there was no competition by endogenous IFN-γ for
heparan sulphate binding sites in the knock out mice.
B HUMAN I125 IFN-γ BINDS LESS IN MC-2 TRANSFECTED
TUMOURS
To test the ability of the tumour transfectants in preventing exogenous human I125-IFN-γ
binding in vivo, the tumours without the HIS-Tag expressing MC-2 or alanine8 were injected
in IFN-γ deficient mice. At day 8 the mice were given a single shot of 5 µg human I125-IFN-γ
and one hour later the tumours were excised. The MC-2 secreting tumour line was capable of
preventing accumulation of IFN-γ by approximately 30% when compared to the alanine
secreting control tumour.
0
1
2
3
4
5
MC-2 Alanine
% IFN-γ Accumulation
J558L Tumour Transfectant (No HIS-Tag)
Figure 24: Tumours secreting the MC-2 peptide accumulate less human I125-IFN-γ in
vivo.
Percentage dose per gram of tissue. Transfected tumours grown for 8 days without the HIS-
Tag were treated with human I125-IFN-γ and 1 hour later the tumours (and organs) were
excised and counts determined. Little difference was seen in the other organs or blood
between the two groups, (n=3).
69
RESULTS
Similarly, single mice were given 3 tumour lines and matrigel in order to compare the
accumulation of human I125-IFN-γ within the organs and at the different sites. Tumours were
allowed to grow to 0.4-0.8cm in size. Both the pTag alone and parental tumours appeared to
grow with the same kinetics, however the pMC-2-Tag were always noticably larger. From
Figure 25 it is clear that there is less accumulation of I125-IFN-γ in the pMC-2-Tag tumours
compared with the parental tumour and the matrigel. The gel was the highest accumulator of
the human I125-IFN-γ even though less than 3% of the gel by weight is actually heparan
sulphate.
0
2
4
6
8
Parental pMC-2-Tag pTag ECM Gel
% IFN-γ Accumulation
J558L Tumour
Figure 25: MC-2 secreting tumours can prevent human I125-IFN-γ accumulation.
Percentage dose per gram of tissue. Eight days after s.c. injection of tumour cells and EHS-
ECM gel on single mice, 5µg human I125IFN-γ was injected i.v. and one hour later tumour and
gel excised and counts determined, (n=3).
C TRANSFECTED TUMOUR RESPONSE TO CYCLOPHOSPAMIDE
TREATMENT
The model in which the rejection of J558L tumours by cyclophosphamide is IFN-γ dependent
has been previously described in earlier experiments (Figure 9, 10 and 11). In order to show a
function of the LRKRKRS fragment in vivo, IFN-γ competent mice were given 1x106 tumour
cells and treated with Cy eleven days after tumour injection. Five groups were included:
70
RESULTS
J558L Parental, J558L-pMC2-NoTag, J558L-pMC2-Tag, J558L-pAlanine8-NoTag and
J558L-pTag. From the growth kinetics it was clear that mice that received tumours with the
MC2 fragment grew similarly to the control lines. Tumours that contained the neutral
fragment appeared to have a larger tumour volume, however after Cy treatment, tumours
grew out completely. Mice bearing the J558L-pMC2-NoTag and J558L-pMC2-Tag groups
were also unable of rejecting tumours upon treatment with Cy, whilst mice with the J558L-
pTag group rejected tumours 30% of the time. Parental J558L tumours and those mice
bearing parental tumours, treated daily with MC-2 peptide (i.p.) were rejected after Cy
treatment.
Tumour Volume (mm3)
0
2000
4000
6000
8000
10000
12000
14000
0 5 10 15 20 25
Time - (Days)
Figure 26: Growth kinetics of plasmid transfected J558L cell lines treated with Cy.
Parental J558L , Parental J558L and MC-2 peptide , J558LpMC-2 NoTag , J558L
pMC-2-Tag ,J558L-Alanine8-NoTag , J558L pTag , Tumour volumes were
determined using a calliper, each group (n=7). Representative of 3 experiments. Tumours in
untreated mice grew out by day 20-25 and had to be euthanised (n=4).
71
DISCUSSION
4 DISCUSSION
4.1 THE ORIGIN OF INTERFERON GAMMA IN
TUMOUR RESPONSES
Interferon gamma is the principle effector cytokine of cell-mediated immunity. It plays a
critical role in promoting host resistance to microbial infection and is also involved in
pathological circumstances such as trauma, autoimmunity and cancer (Billiau et al., 1998).
Until recently, it was assumed that the main source of IFN-γ involved in anti-tumour
responses was lymphocyte derived, however it was demonstrated that macrophages could
secrete large amounts of IFN-γ and that this may operate by an autocrine positive feedback
loop in which IL-12/IL-18 stimulation was critical. During tumour rejection mediated by the
DNA-alkylating agent, cyclophosphamide, it was concluded that while production of IFN-γ
by macrophages was necessary, it was not entirely sufficient for mediating tumour rejection
(Ibe et al., 2001). In the experiments carried out by Ibe and colleagues, IFN-γR-/- mice were
unable to reject the tumour upon Cy treatment, however there was a delay in tumour growth,
which could not be explained. The purpose of this thesis was to further analyse the role of
IFN-γinvolved in Cy mediated rejection and determine whether matrix interactions with the
cytokine had any function within this model.
It has been shown that IFN-γ is an important participant in Cy mediated tumour rejection. The
model on which most of the animal experiments were based for the thesis demonstrated that
IFN-γ-/- mice are incapable of rejecting J558L tumours after Cy treatment and therefore
appear to be severely impaired when generating anti-tumour immune responses. There was a
short delay in the tumour growth but eventually tumours grew out. This behaviour is similar
to the response of the IFN-γR-/- mice with MC51.9 tumours, however the kinetics in this
model indicated a small decrease in size of the tumour before outgrowth. Additionally, the
rapid destruction of the vasculature demonstrated by the CD31 stainings confirmed what was
seen in the IFN-γR mice with MC51.9 tumours treated with Cy.
Confirmation of the results from the IFN-γ-/- mice was carried out by antibody neutralisation
of IFN-γ in wild type BALB/c mice, as it has been recently suggested that effects seen in mice
with targeted gene deletions may be misleading due to the possible indirect effects, such as
the over compensatory expression of other genes, that can not be controlled (Wu et al., 2004).
72
DISCUSSION
Neutralisation of IFN-γ works as a temporary blockade of a specific signalling pathway. The
neutralising antibody XMG6 was used to inactivate the effect of IFN-γ before tumour cell
injection. The potency of a single injection or multiple injections at different time points
throughout the growth period of the tumour had the same effect. Mice were incapable of
rejecting a tumour after Cy treatment. Neutralising antibodies are known to be effective for
two to four weeks, therefore it is reasonable to assume that XMG6 remained in the mice over
the observation period, presumably blocking the anti-angiogenic effect of IFN-γ and allowing
tumour outgrowth.
It was important to determine whether the source of IFN-γ required in tumour rejection arose
locally or from the whole body. It is clear from several studies that T cell and NK cell derived
IFN-γ is important in mediating tumour rejection. Nevertheless, it was never shown in vivo if
there was an augmentation of systemic IFN-γ production in response to Cy. By using the
Cincinnati Cytokine Capture (CCC) assay, over 96 hours in IFN-γ competent mice, there was
a steady increase of detectable IFN-γ in the serum after Cy treatment. Levels rose from the
baseline 250 pg/ml in tumour free mice to over 700 pg/ml, 96 hours after Cy treatment. This
result can be explained by two hypotheses, (i) that matrix may release IFN-γ into the serum in
response to Cy and/or (ii) there was homeostatic proliferation after lymphocyte depletion
caused by Cy. It is known that IFN-γ is released by haematopoietic cells following
homeostatic proliferation. This IFN-γ release into the serum correlates with the documented
recovery of lymphocytes, particularly T cells after Cy treatment. It is known that Cy treated
animals become lymphopenic within the first few hours after drug administration, and that it
takes approximately three days for the T and B lymphocytes to begin to recover and 10 days
for complete recovery of pretreatment total spleen numbers (Lutsiak et al., 2005)
In order to confirm the origin of IFN-γrequired for rejection, bone marrow chimeras were
made to elucidate the role of haematopoietic and non-haematopoietic cells in the
immunological response to cyclophosphamide. Four groups of bone marrow chimeras were
generated from IFN-γgene deficient and competent mice (IFN-γ-/- to IFN-γ-/-, IFN-γ-/- to IFN-
γ+/-, IFN-γ+/- to IFN-γ-/-, IFN-γ+/- to IFN-γ+/-). Those mice that had haematopoietic cells, which
were capable of producing IFN-γ were capable of rejecting tumours with similar kinetics to
the BALB/c wild type mice. It appeared in these two groups that the non-bone marrow
derived cells did not play a major role in the rejection mediated by Cy. It was only essential
that the donor bone marrow cells be capable of producing IFN-γ and these cells are most
73
DISCUSSION
likely responsible for the increase of IFN-γ after 48-72 hours in the serum detected by the
CCC assay.
Whilst, there was no rejection of J558L tumours in mice that had received IFN-γ-/-
haematopoietic cells, there was an interesting response in the IFN-γ-/- to IFN-γ+/- group. The
tumours within this group decreased in size for approximately four days, after which there
was expansion of the tumour mass. The total delay of outgrowth was approximately 6 days.
The unexpected stay in tumour growth within the IFN-γ-/- to IFN-γ+/- group led to the
hypothesis that matrix bound IFN-γ may influence immune responses.
The two variables in this experiment were the IFN-γ competent non-haematopoietic host cells
and an IFN-γ pre-loaded extracellular matrix. It was shown by Lortat-Jacob and colleagues,
that the half life (t1/2) of circulating IFN-γcould be increased from 1.5 to 99 minutes when it
was heparin bound, however the kinetics of matrix bound IFN-γ in vivo are presently
unknown. This delay in tumour rejection could be due to the release of matrix bound IFN-γ by
matrix metalloprotease 9 (MMP-9) produced by polymorphonuclear neutrophils in response
to Cy treatment (Hirsh et al., 2004). MMP-9 is important in the proteolytic modelling of the
matrix and is known to cleave and release cytokines and chemokines (Ma et al., 2005).
However, it is clear that for complete tumour rejection, sustained production of
haematopoietic cell derived IFN-γ was required.
While the IFN-γ derived from the haematopoietic system has a relatively short half-life in
vivo, it is expected that the t1/2 of matrix bound IFN-γ in vivo would be longer. A mechanism
has been proposed whereby the integrity of the C-terminal of IFN-γ that is crucial to
biological activity, remains intact from proteolytic cleavage by binding to heparan sulphate
(Lortat-Jacob et al., 1996a). Therefore, biological functions of the cytokine might be
prolonged in vivo through such an interaction (storage reservoir). It was shown that numerous
growth factors, including FGFs, insulin growth factor (IGF), TGF-β and VEGFs, which bind
to heparan sulphate, can be released from storage and/or be activated from latent forms
(Taipale and Keski-Oja, 1997). Stabilisation of IFN-γ with heparan sulfate results in
conformational changes of the secondary and tertiary structure enabling better recognition of
the molecule by the receptor (Balasubramanian and Ramanathan, 2000). Therefore, it was
appropriate to investigate whether IFN-γand heparan sulphate could interact within the
tumour matrix.
74
DISCUSSION
4.2 THE BINDING AND INFLUENCE OF INTERFERON
GAMMA ON THE TUMOUR EXTRACELLULAR
MATRIX
In general, the ability of growth factors and various cytokines to bind to the extracellular
matrix has been well documented, but it is less clear for IFN-γ. Therefore, to determine the
significance of the interaction of IFN-γ with heparan sulphate the evolutionary consensus of
the cytokine sequence was investigated. The homology of the basic amino acid cluster
responsible for IFN-γ/heparan sulphate interactions between different species was compared
(Figure 3, page 11) and was highly conserved. The most variability was seen amongst the two
fish species, however the amino acids were still basic. This indicated that it was likely to be
functionally important.
Until recently, all previous studies analysing the interactions between IFN-γ and heparan
sulphate were determined in vitro. Visualisation of the interaction between these molecules in
tumour sections was confirmed through confocal microscopy. In order to do this, a number of
murine IFN-γantibodies were screened. Caution in interpreting the results of
immunohistochemistry has been suggested (van der Loos et al., 2001). In total, van der Loos
and colleagues screened 13 anti-human IFN-γ antibodies. Twelve of the thirteen antibodies
recognised IFN-γ positive cells only upon stimulation and permeabilisation of the cell
membrane with saponin, the group also found staining for smooth muscle cells, endothelial
cells, extracellular matrix and CD138 plasma cells. At the time of this publication there was
little known regarding the possibility of cytokine interaction with the matrix. As a result they
considered the staining to be unspecific. Wrenshall and colleagues demonstrated that the
cytokine IL-2 can bind to the heparan sulphate. The tertiary structure of a cytokine is required
for ionic and electrostatic contact with the ECM. IL-2 and IFN-γ have similar tertiary
structures supporting the possibility of an IFN-γ/ heparan sulphate relationship (Wrenshall
and Platt, 1999).
Within the normal tumour microenvironment there is very little IFN-γ present. Therefore,
tumour models (J558l-IFN-γ and MC51.9-IFN-γ) in which there was overexpression of this
cytokine were used to create a basis for analysis of cytokine negative tumour cells. By doing
this, the threshold of staining with XMG1.2 was contained to those areas where there were
high IFN-γ deposits. Additionally, as matrix-cytokine interactions were of particular interest,
75
DISCUSSION
tumour sections were not treated with saponin. Specific staining of the IFN-γ transfected
tumours was confirmed by isotype controls and the fact that cellular IFN-γ was limited to
those cells that had been cross-sectioned by the microtome. Additionally, there was no
staining in IFN-γ-/- mice. The secreted IFN-γ being detected in these tissues could be seen in
scattered areas though out the central body and rim of the tumour section.
At least two forms of the basement membrane protein, HSPG, have been identified; a large
core protein (> 400 kD) and a small core protein (30 kD). The large HSPG is probably the
most abundant basement membrane proteoglycan. It is located predominantly in the lamina
lucida, where it forms clustered aggregates and interacts with other basement membrane
components to form the matrix. In addition, it also plays a critical role in attachment of cells
to the basal membrane via integrin receptors.
Heparan sulphate was bound by the monoclonal antibody A7L6 which specifically recognizes
domain IV of the core protein of the large heparan sulphate proteoglycan or perlecan. The
reactivity is independent of the galactosaminoglycan moieties. As such the epitope is not
sensitive to heparitinase treatment. However, HS chains can be degraded either enzymatically
by heparanase or non-enzymatically by nitric oxide (Mani et al., 2004), which is capable of
cleaving HS chains at glucosamines lacking N-substitution. Therefore, while the HS signal
from the different tumours within the different mice was strong there was noticeable variation
in their HS structures. This could be a result of the amount of heparan degrading enzymes
produced by the tumour and the mouse.
In co-localisation studies of the two molecules HS and IFN-γ in the J558L-IFN-γ producer,
approximately 10-15% of the tumour tissue was stained positive for IFN-γ (by visual
assessment of the defined tumour area of over 40 sections), and 60% of that IFN-γ positive
area was also heparan sulphate positive. In comparison, approximately 1-3% of the tumour
section was positive for IFN-γ and similarly 60% of that IFN-γ co-localised to heparan
sulphate in the cytokine negative parental J558L tumours. As expected, the staining for IFN-γ
was weaker than the over-expresser, however there was an association between this cytokine
and the basement membrane component heparan sulphate. By establishing the parameters of
analysis using the J558L-IFN-γ producer as a basis for the confocal protocols, this approach
allowed the specific interaction between heparan sulphate and IFN-γ to be seen in the normal
microenvironment of J558L tumours.
76
DISCUSSION
During the co-localisation studies using the confocal microscope slight differences in the
structure of HS were revealed, therefore more detailed analysis was carried out to determine if
there were any more structural alterations in the tumour matrices of IFN-γ-/- and WT mice.
IFN-γ is known to suppress matrix protein deposition (Niwa et al., 2004). Genes involved in
matrix metabolism that are negatively regulated by IFN-γ include MMP 1, -9, -13 and
stromelysin and type II collagen. However, the relevance of these differences for the tumour
microenvironment remains unclear.
Whilst there are no macroscopic differences in the organogenesis or development of the cells
of the immune system in IFN-γ gene deficient mice, there appears to be diminished capacity
to respond to infection and immunological stimuli. The tumour stroma is known to be an
important target in aiding tumour rejection. The extracellular microenvironment is constantly
being modified by matrix metalloproteinases and therefore is an active component of the
tumour. It is highly likely that there would be microscopic differences within the tumour
stroma of gene deficient and competent mice. Histological analysis of two different tumour
types (J558L and TS/A) revealed the presence of a panel of molecules of the ECM that are
important in the structural integrity of tumours (Collagen 1-4, heparan sulphate). However,
there were minor differences in the characteristics of the molecules between IFN-γ competent
and deficient mice. It is known from gene transcript studies that large amounts of IFN-γ are
required to modify protein expression of collagens (Diaz and Jimenez, 1997), and heparan
sulphate (Sharma and Iozzo, 1998), therefore, as there are limited amounts of circulating IFN-
γ within tumour bearing IFN-γ competent mice, only subtle differences can be seen. This
would account for the thin sheets of heparan sulphate and collagens in the tumours implanted
in immuno-competent mice compared with the thick, dense deposits in the tumours of gene
deficient mice.
Additional confirmation of the HS/IFN-γ interaction was provided by matrigel plug
accumulation of human IFN-γ, which could be detected ex vivo by ELISA up to 24 hours after
injection. The advantage of this system was that the human IFN-γ could bind to murine
heparan sulphate (Lortat-Jacob et al., 1991) but not the murine cell surface IFN-γ receptor
(Farrar and Schreiber, 1993). This experiment provided additional evidence that IFN-γ could
be stored in vivo as a non-receptor bound cytokine (Lortat-Jacob et al., 1991). The results
suggested that there were significant differences between the gene deficient and competent
77
DISCUSSION
mice, which could affect IFN-γ accumulation in vivo. Consequently, in vivo analysis was
carried out by iodinating human IFN-γ as it provided a more sensitive level of detection than
immunohistochemistry. The iodination was achieved by chloramine T and not the Bolton
Hunter method commonly used in cytokine labelling. Bolton Hunter usually labels the
arginines and lysines, which in this case are essential for IFN-γ binding to heparan sulphate,
therefore the tyrosines were labelled instead using chloramine T.
The amount of I125 labelled IFN-γ found in the blood was low (between 4-6%) indicating that
there was rapid clearance or blood to tissue transfer, which confirmed previous findings
(Lortat-Jacob et al., 1996b). Previously it had been described that the key time point in
determining organ accumulation was 1 hour. As cytokine retention capability was of
particular interest, analysis of I125-IFN-γ in organs was also made at 24 hours. At the early
time point of one hour after injection there was significantly more accumulation of IFN-γ in
the tumours of knockout mice compared with their wild type litter-mates. This was indicative
of competition between endogenous IFN-γ in the WT mice for the extracellular binding sites.
Day 5 tumours had higher accumulation of I125-IFN-γ compared to their day 8 counterparts, 1
hour after cytokine transfer in both gene deficient (three fold) and competent (two fold)
groups. Whilst the standard deviation remained high for the knockout mice there were still
clear differences. Early tumours are highly vascularised and matrix deposition has just begun.
Therefore, this could account for the high signal from day 5 gene deficient mice where there
was no competition for heparan sulphate binding sites and there was a high percentage dose
per gram ratio. The differences in the tumour matrix of the IFN-γ−/− mice compared to their
WT counterparts might also contribute to the higher accumulation of human I125-IFN-γ in the
IFN-γ−/− mice as a result of the structural differences as discussed earlier.
Accumulation of radioactive cytokine in the tumour is significantly different to other organs
due to large deposits of basement membrane matrix, which make up the major part of the
tumour body. The presence of I125-IFN-γ in organs (spleen, liver, kidney, heart, lung and
tumour) 24 hours and 72 hours after injection were analysed, however no significant
differences between the knockout and wild type mice was seen. The fact that by 24 hours
there is significant reduction in the amount of IFN-γ present in the organs might indicate that
there is rapid turnover of the cytokine.
78
DISCUSSION
For the first time binding and accumulation in vivo of IFN-γ to heparan sulphate of the tumour
matrix was shown. However, the exact function of this interaction remains unclear. Heparan
sulphate glycosaminoglycans, present at the cell surface and the extracellular matrix that
surrounds cells are important mediators of biological processes. Furthermore, it has become
apparent that cells dynamically regulate the structure of their heparan sulphate “coat” to
differentially regulate extracellular signals (Liu et al., 2002), or regulate matrix homeostasis
(Taipale and Keski-Oja, 1997). The histological confirmation of the electrostatic interaction
between IFN-γ and heparin sulphate in vivo, indicated a possible functional relationship as a
result of contact between the two molecules. Therefore, the purpose and relevance of this
interaction was investigated by interfering with the non-receptor mediated binding of IFN-γ to
the matrix.
4.3 THE BLOCKADE OF HEPARAN SULPHATE AND
INTERFERON GAMMA INTERACTIONS WITHIN
THE TUMOUR STROMA
IFN-γ is theoretically capable of interacting with heparan sulphate within the matrix and on
the endothelial vasculature (Yard et al., 1998), however reports vary regarding the function of
bound IFN-γ. Heparin bound IFN-γ has been shown to have an inhibitory outcome on MHC
class II expression as well as antiviral and anti-parasitic actions (Daubener et al., 1995).
Contra to this, immobilised plate bound IFN-γ molecules have been shown to induce HLA-
DR expression on COLO-205 tumour cells (Fernandez-Botran et al., 1999), suggesting cell
surface GAGs could present IFN-γ to its receptor and that there was an important regulatory
role for heparan sulphate on the activity of IFN-γ in vivo.
In an effort to determine the functional consequences of inhibiting IFN-γ binding to heparan
sulphate, a peptide was generated by Fernandez-Botran and co-workers, 1999, from the small
basic amino acid sequence (murine IFN-γ128-135) at the C-terminal end of IFN-γ. This
sequence is responsible for the electrostatic interaction with heparan sulphate and this cluster
was additionally identified as crucial in increasing the on rate of the IFN-γ/IFN-γ-R binding
reaction (Sadir et al., 1998). Therefore, confirmation of the functional inhibition in vitro of the
murine peptide (LRKRKRSR) was required in order to determine whether a response could
79
DISCUSSION
possibly be analysed in vivo. It was evident that the peptide was capable of preventing human
I125-IFN-γ binding to plate-bound heparan sulphate within the matrigel in vitro. This was the
first time that inhibition was shown in relation to matrigel, as previously the assay was carried
out with purified plate-bound heparan sulphate (Fernandez-Botran et al., 2002). In order to
study the effects of the peptide in vivo, plasmids were constructed that were capable of
secreting the peptide from transfected J558L tumour cells.
The murine IFN-γ128-135 sequence was inserted into the pSecTag-2 plasmid under the control
of the EF1-α promoter. Substitution of the original CMV promoter was necessary as it
becomes methylated and consequently shuts down in vivo (French Anderson, 1994).
Additionally, EF1-α is known to be a strong promoter of gene expression (Tokushige et al.,
1997). Four plasmids were constructed due to evidence suggesting that histidine repeats can
bind to heparin and heparan sulphate with high affinity (Jones et al., 2005). Hereby masking
the cleavage sites of the heparan sulphate chains recognised by the β-D-endoglycosidase
heparanase.
The supernatants from the cell lines expressing the MC-2 peptide were capable of inhibiting
the binding of human I125-IFN-γ to matrigel in vitro, when compared to the non-expressing
MC-2 controls and the untransfected parental tumour line. These results confirmed the
functional significance of the murine IFN-γ128-135 peptide sequence in preventing the binding
of I125-IFN-γ in vitro. Hence, production of the peptide from the tumours was confirmed, and
the influence of blocking the HS/IFN-γ interaction by the tumour lines was investigated in
vivo in the Cy model.
Non-receptor binding may be related to several aspects of cytokine activity and may have
physiological consequences. Thus, in concordance with the previous functional experiments
using cyclophosphamide, the transfected tumours were subjected to the same protocol. The
cell lines all appeared to grow with similar kinetics in vitro and in vivo, however the Alanine8
secreting tumours always grew slightly faster. At the same time, mice bearing the parental
J558L tumour were treated daily with 1mg of the murine IFN-γ128-135 peptide. Tumours were
slightly smaller compared with untreated parental tumours. This may have been a result of the
inflammatory response to the daily MC-2 peptide injections or that the peptide may have
prevented the non-receptor mediated binding of the IFN-γ molecule and therefore could
contribute to its influence on growth and the subsequent clearance of the cytokine.
80
DISCUSSION
Tumours expressing the MC-2 fragment were not rejected when treated with Cy at day 11.
There was a slight reduction in tumour mass but the tumour continued to grow out. Among
the mice of the J558L-pTAG group only 35% of the challenged mice were capable of
rejecting the tumour which may have been due to the influence of the 6-HIS-repeat sequence
that might have prevented remodelling and cleavage of the heparan sulphate side chains
(Jones et al., 2005) in response to Cy treatment. Mice challenged with the J558L-pAlanine-
NoTAG tumour did not reject the tumour. The neutral alanine fragment is unlikely to have
bound to the matrix through electrostatic or ionic interactions as the frame-shifted tail
sequence is relatively uncharged. Some other component of the plasmid backbone may have
influenced the way in which J558L tumour cells would normally be rejected by Cy. Recently,
it was shown that data obtained from plasmids containing the Zeocin resistance marker could
be misleading as Zeocin could not be completely detoxified by the Sh ble gene and was still
capable of cleaving DNA (Trastoy et al., 2005). Ultimately, recombinant cells could have
cumulative damages indicating that their genomic integrity and metabolism may have been
altered. Whilst there appears to be an effect of the murine IFN-γ128-135 peptide as seen by the
I125 IFN-γ binding assay, the role of IFN-γand heparan sulphate interactions in this model
remains to be further investigated. Ultimately, the development of a mouse in which the IFN-
γ molecule could bind to heparan sulphate and not to the receptor (or vice-versa) would
enable the role of this interaction to be fully understood in relation to Cy mediated tumour
rejection.
By using a chemotherapy induced, immune response mediated tumour rejection model the
critical importance of haematopoietic cell induced IFN-γ was elucidated. For the first time, it
has been demonstrated that IFN-γ levels were elevated in the serum after Cy treatment and the
presence of matrix bound IFN-γ in vivo was described. Additionally, the accumulation of
radioactive human IFN-γ in the tumour stroma, in vivo, confirmed and proved the presence of
non receptor mediated binding of the cytokine and that this accumulation could be inhibited in
the presence of the peptide fragment MC-2 (LRKRKRSR). Whilst the effect of inhibiting
IFN-γ/HS interactions in the plasmid transfected tumour model remains unclear at present, the
importance of heparan sulphate within the Cy tumour model could be investigated using
perlecan knock out mice. At present these mice are not on the BALB/c background and would
need to be backcrossed.
81
DISCUSSION
The existence of this species conserved C-terminal sequence that facilitates IFN-γ binding to
the matrix indicates that there may be an important function for this interaction. Evidence
from the allotransplantation experiments carried out by Fernandez-Botran and coworkers,
2001, indicate that there may be an impaired immune response due to the presence of the MC-
2 peptide (Fernandez-Botran et al., 2002), however its function within this Cy dependent
tumour rejection model remains unclear. Lastly, it is clear that for tumour rejection IFN-γ is
required, however haematopoietic cell induced and matrix bound IFN-γ may not be mutually
exclusive and could possibly function together to generate an adequate anti-tumour immune
response.
82
SUMMARY
5 SUMMARY
Interferon gamma (IFN-γ) is a pro-inflammatory cytokine that is produced primarily by T and
NK cells. Previously, it has been shown that IFN-γ is necessary for cyclophosphamide
mediated tumour rejection, however, the source of this IFN-γ remained unclear. The aim of
this work was to identify the source and characterise the role of IFN-γ in this scenario. Using
the Cincinnati Cytokine Capture Assay it was shown for the first time that the IFN-γ serum
levels steadily increased over a 96 hour time period after cyclophosphamide administration.
Furthermore, with the help of bone marrow chimeras, I was able to show that IFN-γ required
for tumour rejection was produced by haematopoietic cells. Interestingly, there was a delay in
tumour outgrowth in bone marrow chimeras that had received IFN-γ deficient bone marrow
cells when compared to IFN-γ deficient mice. I concluded that there must have been an
additional source of IFN-γ and therefore extended the investigation.
The ability of IFN-γ to bind to matrix proteins such as heparan sulphate in vitro has been
reported. The binding of IFN-γ to the extracellular matrix in vivo has so far not been shown.
Therefore, I investigated whether matrix bound IFN-γ was responsible for the growth delay
observed in IFN-γ−/− to IFN-γ+/- bone marrow chimeras. Using confocal microscopy I was able
to show that IFN-γ could bind to heparan sulphate in various tumour models. In a xenotypic
model I used human IFN-γ in mice. The advantage of this system was that the human IFN-γ
could bind to murine heparan sulphate but not the murine cell surface receptor. I could
confirm the binding of human IFN-γ to heparan sulphate, as it had accumulated in matrigel
plugs. Additionally, using human I125 IFN-γ, binding and accumulation to heparan sulphate of
the tumour matrix was shown for the first time. The electrostatic co-operation between IFN-γ
and heparin sulphate in vivo supported the possibility there was functionality as a result of
contact between the two molecules.
In an effort to determine the functional consequences of inhibiting IFN-γ binding to heparan
sulphate, I made use of a previously identified peptide that was generated based on the highly
conserved, small basic amino acid sequence (murine IFN-γ128-135) at the C-terminal end of
IFN-γ. This peptide has been shown to block in vitro binding of IFN-γ to heparan sulphate. In
order to study the effects of the peptide in vivo, plasmid transfected J558L tumour cells were
constructed that secreted the peptide. Tumours that expressed the peptide could not be
83
SUMMARY
rejected after cyclophosphamide treatment. However, rejection of the empty vector control
tumours was also not complete.
This work demonstrates that haematopoietic cells are responsible for the production of IFN-γ
required for complete rejection of the tumour in our model, that there is a steady increase of
IFN-γ during the first 4 days after cyclophosphamide treatment, that IFN-γ binding to the
extracellular matrix occurs in vivo (this interaction can be partially inhibited by the MC-2
peptide), and that IFN-γ influences matrix deposition. Since the results of the functional
experiments using this model are inconclusive, further investigational modifications will be
required.
84
ZUSAMMENFASSUNG
6 ZUSAMMENFASSUNG
Interferon gamma (IFN-γ) ist ein pro-inflammatorisches Zytokin, das primär durch T- und
NK-Zellen gebildet wird. Es wurde kürzlich berichtet, dass IFN-γ für eine durch
Cyclophosphamid-vermittelte Tumorabstoßung notwendig ist. Die Quelle dieses IFN-γ blieb
jedoch unklar. Das Ziel dieser Arbeit war, diese Quelle zu identifizieren und die Rolle des
IFN-γ in diesem Szenario zu charakterisieren. Mit dem Cincinnati Zytokin Capture Assay
konnte ich zum ersten Mal zeigen, dass die IFN-γ Serumspiegel über einen Zeitraum von 96
Stunden nach der Verabreichung von Cyclophosphamid kontinuierlich anstiegen. Des
Weiteren konnte ich mit Hilfe von Knochenmarks-Chimären zeigen, dass das IFN-γ, welches
für die Tumorabstoßung erforderlich war, durch hämotopoetische Zellen gebildet wurde.
Interessanterweise gab es im Vergleich zu IFN-γ-defizienten Mäusen eine Verzögerung im
Tumorwachstum der Knochenmarks-Chimären, die IFN-γ-defiziente Knochenmarkzellen
erhalten hatten. Ich schloss daraus, dass es eine zusätzliche Quelle von IFN-γ geben musste,
und erweiterte daher die Untersuchung.
Über die Fähigkeit von IFN-γ, Matrix-Proteine wie Heparan-Sulfat in vitro zu binden, wurde
berichtet. Die Bindung von IFN-γ an die extrazelluläre Matrix in vivo war bisher nicht gezeigt
worden. Daher untersuchte ich, ob Matrix-gebundenes IFN-γ für das langsamere Wachstum,
das ich in IFN-γ-/- gegenüber IFN-γ+/- Knochenmarks-Chimären beobachtet hatte,
verantwortlich war. Durch konfokale Mikroskopie konnte ich zeigen, dass IFN-γ in
verschiedenen Tumormodellen Heparan-Sulfat binden konnte. In einem xenotypischen
Modell verwendete ich humanes IFN-γ in Mäusen. Der Vorteil dieses Systems war, dass das
humane IFN-γ zwar Maus-Heparan-Sulfat, nicht jedoch die Zelloberflächenrezeptoren der
Maus binden konnte. Ich konnte die Bindung von humanem IFN-γ an Heparan-Sulfat, wie es
in Matrigel Plugs akkumuliert war, bestätigen. Zusätzlich konnte ich durch humanes
I125 IFN-γ erstmalig die Bindung und Akkumulation an Heparan-Sulfat der Tumor-Matrix
zeigen. Die elektrostatische Wechselwirkung zwischen IFN-γ und Heparan-Sulfat in vivo
deutet darauf hin, dass Funktionalität als Ergebnis eines Kontaktes zwischen den zwei
Molekülen vorhanden war.
In einem Versuch, die funktionellen Konsequenzen der Inhibition der Wechselwirkung
zwischen IFN-γ Heparan-Sulfat aufzuklären, verwendete ich ein kürzlich identifiziertes
Peptid, das von der hoch konservierten Aminosäurensequenze am C-Terminal-Ende von IFN-
85
ZUSAMMENFASSUNG
γ (Maus-IFN-γ128-135) abgeleitet worden war und das reich an kleinen basischen Aminosäuren
ist. Es zeigte sich, dass dieses MC-2 Peptid die in vitro Bindung von IFN-γ an Heparan-Sulfat
blockierte. Um die Effekte dieses Peptids in vivo untersuchen zu können, wurden transfizierte
J558L Tumorzellen hergestellt, welche die Peptide absonderten. Tumore, die Peptide
exprimierten, konnten nach Cyclophosphamid-Behandlung nicht abgestoßen werden. Jedoch
war ebenso die Abstoßung der Kontrolltumore nicht vollständig.
Diese Arbeit demonstriert, dass hämatopoetische Zellen für die Produktion von IFN-γ,
welches für die komplette Tumorabstoßung in unserem Modell benötigt wird, verantwortlich
sind, dass es einen kontinuierlichen Anstieg von IFN-γ während der ersten vier Tage nach
Cyclophosphamid-Behandlung gibt, dass die IFN-γ-Bindung an die extrazelluläre Matrix
auch in vivo stattfindet (diese Interaktion kann durch das MC-2 Peptid partiell inhibiert
werden), und dass INF-γ die Zusammensetzung der Matrix beeinflusst. Da die Ergebnisse der
funktionalen Experimente, die dieses Model verwenden, nicht eindeutig sind, sind weitere,
modifizierte Untersuchungen erforderlich.
86
ABBREVIATIONS
7 ADDITIONAL ABBREVIATIONS
aa Amino acid
BM Basement membrane
BMC Bone marrow chimera
bp Base pair
C Carboxyl
Cpm Counts per minute
Cy Cyclophosphamide
DAB Diaminobenzidine
DMEM Dulbeccos modified Eagles medium
DNA Deoxyribonucleic acid
ECM Extracellular matrix
EGF Epidermal growth factor
EHS Engelbreth Holm Swarm
FCS Foetal calf serum
FGF Fibroblast growth factor
GAG Glycosaminoglycan
GAS Gamma activated sequences
Gy Gray
HPLC High performance liquid chromotography
HS Heparan sulphate
HSPG Heparan sulphate proteoglycan
IFN-γ Interferon gamma
IFN-γR Interferon gamma receptor
IGF Insulin growth factor
IL-2 Interleukin-2
IL-4 Interleukin-4
IL-8 Interleukin 8
JAK Janus tyrosine kinases
kb Kilobase
Kd Dissociation constant
kg Kilogram
LB Luria Bertani
M Marker
mAb Monoclonal antibody
MCA Methylcholanthrene
MCP Macrophages chemotactic proteins
mg Milligram
MHC Major histocompatability complex
ml Millilitre
MMP Matrix metalloproteases
N Amino
ng Nanogram
NK Natural killer
PBMC Peripheral blood monocyte cells
PCR Polymerase chain reaction
PDGF Platelet derived growth factor
PG Proteoglycan
87
ABBREVIATIONS
pg Picogram
SAV Streptavidin-HRP conjugate
SLRP Small leucine rich repeats
STAT-1α Signal transducer and activator of
transcription-1α
t1/2 half-life
TAF Tumour-associated fibroblast
Taq Thermus aquaticus
TGF-β1 Transforming growth factor beta
VEGF Vascular endothelial growth factor
µg Microgram
WT Wild type
88
REFERENCES
8 REFERENCES
Awwad, M., and North, R. J. (1988). Cyclophosphamide (Cy)-facilitated adoptive
immunotherapy of a Cy-resistant tumour. Evidence that Cy permits the expression of adoptive
T-cell mediated immunity by removing suppressor T cells rather than by reducing tumour
burden. Immunology 65, 87-92.
Azouz, A., Razzaque, M., El-Hallak, M., and Taguchi, T. (2004). Immunoinflammatory
response and fibrogenesis. Med Electron Microsc 37, 141-148.
Balasubramanian, V., and Ramanathan, M. (2000). Glycosaminoglycans alter the
conformation of interferon-gamma. Cytokine 12, 466-471.
Becker, C., Pohla, H., Frankenburger, B., Schuler, T., Assenmacher, M., Schendel, D. J., and
Blankenstein, T. (2001). Adoptive tumor therapy with T lymphocytes enriched through and
IFN-γ capture assay. Nat Med 7, 1159-1162.
Berd, D., Mastrangelo, M. J., Engstrom, P. F., Paul, A., and Maguire, H. (1982).
Augmentation of the human immune response by cyclophosphamide. Cancer Res 42, 4862-
4866.
Billiau, A., Heremans, H., Vermeire, K., and Matthys, P. (1998). Immunomodulatory
properties of interferon-gamma. An update. Ann N Y Acad Sci 856, 22-32.
Blankenstein, T., and Qin, Z. (2003). The role of IFN-γ in tumor transplantation immunity
and inhibition of chemical carcinogenesis. Curr Opin Immunol 15, 148-154.
Capila, I., and Lindhardt, R. J. (2002). Heparin - Protein Interactions. Angew Chem Int Ed
Engl 41, 390-412.
Cavallo, F., Martin-Fontecha, A., Bellone, M., Heltai, E., Tornaghi, P., Freschi, M., Forni, G.,
Dellabona, P., and Casorati, G. (1995). Co-expression of B7-1 and ICAM-1 on tumors is
required for rejection and the establishment of a memory response. Eur J Immunol 25, 1154-
1162.
Chiocca, E. A. (1995). Brain tumor therapy in mice with a novel "suicide" gene: The
cyclophosphamide-activating CYP2B1 gene).
Dalton, D. K., Pitts-Meek, S., Keshav, S., Figari, I. S., Bradley, A., and Stewart, T. A. (1993).
Multiple defects of immune cell function in mice with disrupted interferon-gamma genes.
Science 259, 1739-1742.
Daubener, W., Nockemann, S., Gutsche, M., and Hadding, U. (1995). Heparin inhibits the
antiparasitic and immune modulatory effects of human recombinant interferon-gamma. Eur J
Immunol 25, 688-692.
De Wever, O., and Mareel, M. (2003). Role of tissue stroma in cancer cell invasion. J Pathol
200, 429-447.
Diaz, A., and Jimenez, S. A. (1997). Interferon-gamma regulates collagen and fibronectin
gene expression by transcriptional and post-transcriptional mechanisms. Int J Biochem Cell
Biol 29, 251-260.
89
REFERENCES
Dighe, A. S., Richards, E., Old, L. J., and Schreiber, R. D. (1994). Enhanced in vivo growth
and resistance to rejection of tumor cells expressing dominant negative IFN gamma receptors.
Immunity 1, 447-456.
Douglas, M. S., Rix, D. A., Dark, J. H., Talbot, D., and Kirby, J. A. (1997). Examination of
the mechanism by which heparin antagonizes activation of a model endothelium by
interferon-gamma (IFN-gamma). Clin Exp Immunol 107, 578-584.
Dvorak, H. F. (1986). Tumors: wounds that do not heal. Similarities between tumor stroma
generation and wound healing. N Engl J Med 315, 1650-1659.
Elenbaas, B., and Weinberg, R. A. (2001). Heterotypic signaling between epithelial tumor
cells and fibroblasts in carcinoma formation. Exp Cell Res 264, 169-184.
Farrar, M. A., and Schreiber, R. D. (1993). The molecular cell biology of interferon-gamma
and its receptor. Annu Rev Immunol 11, 571-611.
Fernandez-Botran, R., Gorantla, V., Sun, X., Ren, X., Perez-Abadia, G., Crespo, F. A.,
Oliver, R., Orhun, H. I., Quan, E. E., Maldonado, C., et al. (2002). Targeting of
glycosaminoglycan-cytokine interactions as a novel therapeutic approach in
allotransplantation. Transplantation 74, 623-629.
Fernandez-Botran, R., Romanovskis, P., Sun, X., and Spatola, A. F. (2004). Linear basic
peptides for targeting interferon-gamma-glycosaminoglycan interactions: synthesis and
inhibitory properties. J Pept Res 63, 56-62.
Fernandez-Botran, R., Yan, J., and Justus, D. E. (1999). Binding of interferon gamma by
glycosaminoglycans: a strategy for localization and/or inhibition of its activity. Cytokine 11,
313-325.
Finkelman, F. D., and Morris, S. C. (1999). Development of an assay to measure in vivo
cytokine production in the mouse. Int Immunol 11, 1811-1818.
French Anderson, W. (1994). Making clinical grade gene therapy vectors. Hum Gene Ther 5,
925-926.
Gallagher, J. T., and Walker, A. (1985). Molecular distinctions between heparan sulphate and
heparin. Analysis of sulphation patterns indicates that heparan sulphate and heparin are
separate families of N-sulphated polysaccharides. Biochem J 230, 665-674.
Hengst, J. C., Mokyr, M. B., and Dray, S. (1980). Importance of timing in cyclophosphamide
therapy of MOPC-315 tumor-bearing mice. Cancer Res 40, 2135-2141.
Hirsh, M., Carmel, J., Kaplan, V., Levine, E., and Krausz, M. M. (2004). Activity of lung
neutrophils and matrix metalloproteinases in cyclophosphamide-treated mice with
experimental sepsis. Int J Exp Path 85, 147-157.
Hock, H., Dorsch, M., Kunzendorf, U., Qin, Z., Diamantstein, T., and Blankenstein, T.
(1993). Mechanisms of rejection induced by tumor cell-targeted gene transfer of interleukin 2,
interleukin 4, interleukin 7, tumor necrosis factor, or interferon gamma. Proc Natl Acad Sci U
S A 90, 2774-2778.
Ibe, S., Qin, Z., Schuler, T., Preiss, S., and Blankenstein, T. (2001). Tumor rejection by
disturbing tumor stroma cell interactions. J Exp Med 194, 1549-1559.
90
REFERENCES
Iivanainen, E., Kahari, V. M., Heino, J., and Elenius, K. (2003). Endothelial cell-matrix
interactions. Microsc Res Tech 60, 13-22.
Iozzo, R. V. (1998). Matrix proteoglycans: from molecular design to cellular function. Annu
Rev Biochem 67, 609-652.
Iozzo, R. V., and San Antonio, J. D. (2001). Heparan sulfate proteoglycans: heavy hitters in
the angiogenesis arena. J Clin Invest 108, 349-355.
Jain, R. K., and Duda, D. G. (2003). Role of bone marrow-derived cells in tumor angiogenesis
and treatment. Cancer Cell 3, 515-516.
Jiang, X., and Couchman, J. R. (2003). Perlecan and tumor angiogenesis. J Histochem
Cytochem 51, 1393-1410.
Jones, A. L., Hulett, M. D., and Parish, C. R. (2005). Histidine-rich glycoprotein: A novel
adaptor protein in plasma that modulates the immune, vascular and coagulation systems.
Immunol Cell Biol 83, 106-118.
Kalluri, R. (2003). Basement membranes: structure, assembly and role in tumour
angiogenesis. Nat Rev Cancer 3, 422-433.
Klein, G., and Klein, E. (2005). Surveillance against tumors - is it mainly immunological?
Immunol Lett 100, 29-33.
Klein, G., Vellenga, E., Fraaije, M. W., Kamps, W. A., and de Bont, E. S. (2004). The
possible role of matrix metalloproteinase (MMP)-2 and MMP-9 in cancer, e.g. acute
leukemia. Crit Rev Oncol Hematol 50, 87-100.
Li, A. C., and Thompson, R. P. (2003). Basement membrane components. J Clin Pathol 56,
885-887.
Liu, D., Shriver, Z., Venkataraman, G., El Shabrawi, Y., and Sasisekharan, R. (2002). Tumor
cell surface heparan sulfate as cryptic promoters or inhibitors of tumor growth and metastasis.
Proc Natl Acad Sci U S A 99, 568-573.
Lortat-Jacob, H., Baltzer, F., and Grimaud, J. A. (1996a). Heparin decreases the blood
clearance of interferon-gamma and increases its activity by limiting the processing of its
carboxyl-terminal sequence. J Biol Chem 271, 16139-16143.
Lortat-Jacob, H., Brisson, C., Guerret, S., and Morel, G. (1996b). Non-receptor-mediated
tissue localization of human interferon-gamma: role of heparan sulfate/heparin-like
molecules. Cytokine 8, 557-566.
Lortat-Jacob, H., and Grimaud, J. A. (1991). Interferon-gamma C-terminal function: new
working hypothesis. Heparan sulfate and heparin, new targets for IFN-gamma, protect, relax
the cytokine and regulate its activity. Cell Mol Biol 37, 253-260.
Lortat-Jacob, H., and Grimaud, J. A. (1992). Binding of interferon-gamma to heparan sulfate
is restricted to the heparin-like domains and involves carboxylic--but not N-sulfated--groups.
Biochim Biophys Acta 1117, 126-130.
Lortat-Jacob, H., Grosdidier, A., and Imberty, A. (2002). Structural diversity of heparan
sulfate binding domains in chemokines. Proc Natl Acad Sci U S A 99, 1229-1234.
91
REFERENCES
Lortat-Jacob, H., Kleinman, H. K., and Grimaud, J. A. (1991). High-affinity binding of
interferon-gamma to a basement membrane complex (matrigel). J Clin Invest 87, 878-883.
Lubineau, A., Lortat-Jacob, H., Gavard, O., Sarrazin, S., and Bonnaffe, D. (2004). Synthesis
of tailor-made glycoconjugate mimetics of heparan sulfate that bind IFN-gamma in the
nanomolar range. Chemistry 10, 4265-4282.
Lutsiak, M. E., Semnani, R. T., De Pascalis, R., Kashmiri, S. V., Schlom, J., and Sabzevari,
H. (2005). Inhibition of CD4(+)25+ T regulatory cell function implicated in enhanced
immune response by low-dose cyclophosphamide. Blood 105, 2862-2868.
Ma, Z., Chang, M. J., Shah, R. C., and Benveniste, E. N. (2005). Interferon-{gamma}-
activated STAT-1{alpha} suppresses MMP-9 gene transcription by sequestration of the
coactivators CBP/p300. J Leukoc Biol.
Mani, K., Cheng, F., Sandgren, S., Van der Born, J., Havsmark, B., Ding, K., and L.A, F.
(2004). The heparan sulfate-specific epitope 10E4 is NO-sensitive and partly inaccessible in
glypican-1. Glycobiology 14, 599-607.
Matar, P., Rozados, V. R., Gervasoni, S. I., and Scharovsky, G. O. (2002). Th2/Th1 switch
induced by a single low dose of cyclophosphamide in a rat metastatic lymphoma model.
Cancer Immunol Immunother 50, 588-596.
Mocellin, S., Wang, E., and Marincola, F. M. (2001). Cytokines and immune response in the
tumor microenvironment. J Immunother 24, 392-407.
Nanni, P., de Giovanni, C., Lollini, P. L., Nicoletti, G., and Prodi, G. (1983). TS/A: a new
metastasizing cell line from a BALB/c spontaneous mammary adenocarcinoma. Clin Exp
Metastasis 1, 373-380.
Niwa, T., Wada, H., Ohashi, H., Iwamoto, N., Ohta, H., Kirii, H., Fujii, H., Saito, K., and
Seishima, M. (2004). Interferon-gamma produced by bone marrow-derived cells attenuates
atherosclerotic lesion formation in LDLR-deficient mice. J Atheroscler Thromb 11, 79-87.
Noonan, D. M., Fulle, A., Valente, P., Cai, S., Horigan, E., Sasaki, M., Yamada, Y., and
Hassell, J. R. (1991). The complete sequence of perlecan, a basement membrane heparan
sulfate proteoglycan, reveals extensive similarity with laminin A chain, low density
lipoprotein-receptor, and the neural cell adhesion molecule. J Biol Chem 266, 22939-22947.
Nurcombe, V., Ford, M. D., Wildschut, J. A., and Bartlett, P. F. (1993). Developmental
regulation of neural response to FGF-1 and FGF-2 by heparan sulfate proteoglycan. Science
260, 103-106.
O'Reilly, M. S., Holmgren, L., Shing, Y., Chen, C., Rosenthal, R. A., Moses, M., Lane, W. S.,
Cao, Y., Sage, E. H., and Folkman, J. (1994). Angiostatin: a novel angiogenesis inhibitor that
mediates the suppression of metastases by a Lewis lung carcinoma. Cell 79, 315-328.
Pass, G. J., Carrie, D., Boylan, M., Lorimore, S., Wright, E., Houston, B., C.J., H., and Wolf,
C. R. (2005). Role of hepattic cytochrome P450s in the pharmacokinetics and toxicity of
cyclophosphamide: Studies with the hepatic cytochrome P450 reductase null mouse. Cancer
Res 65, 4211-4217.
Philip, M., Rowley, D. A., and Schreiber, H. (2004). Inflammation as a tumor promoter in
cancer induction. Semin Cancer Biol 14, 433-439.
92
REFERENCES
Pupa, S. M., Menard, S., Forti, S., and Tagliabue, E. (2002). New insights into the role of
extracellular matrix during tumor onset and progression. J Cell Physiol 192, 259-267.
Qiao, D., Meyer, K., Mundhenke, C., Drew, S. A., and Friedl, A. (2003). Heparan sulfate
proteoglycans as regulators of fibroblast growth factor-2 signaling in brain endothelial cells.
Specific role for glypican-1 in glioma angiogenesis. J Biol Chem 278, 16045-16053.
Qin, Z., and Blankenstein, T. (2000). CD4+ T cell--mediated tumor rejection involves
inhibition of angiogenesis that is dependent on IFN gamma receptor expression by
nonhematopoietic cells. Immunity 12, 677-686.
Qin, Z., Schwartzkopff, J., Pradera, F., Kammertoens, T., Seliger, B., Pircher, H., and
Blankenstein, T. (2003). A critical requirement of interferon gamma-mediated angiostasis for
tumor rejection by CD8+ T cells. Cancer Res 63, 4095-4100.
Sadir, R., Forest, E., and Lortat-Jacob, H. (1998). The heparan sulfate binding sequence of
interferon-gamma increased the on rate of the interferon-gamma-interferon-gamma receptor
complex formation. J Biol Chem 273, 10919-10925.
Sadir, R., Lortat-Jacob, H., and Morel, G. (2000). Internalization and nuclear translocation of
IFN-gamma and IFN-gammaR: an ultrastructural approach. Cytokine 12, 711-714.
Sasisekharan, R., Shriver, Z., Venkataraman, G., and Narayanasami, U. (2002). Roles of
heparan-sulphate glycosaminoglycans in cancer. Nat Rev Cancer 2, 521-528.
Seljelid, R., Jozefowski, S., and Sveinbjornsson, B. (1999). Tumor stroma. Anticancer Res
19, 4809-4822.
Sharma, B., and Iozzo, R. V. (1998). Transcriptional silencing of perlecan gene expression by
interferon-gamma. J Biol Chem 273, 4642-4646.
Silzle, T., Randolph, G. J., Kreutz, M., and Kunz-Schughart, L. A. (2004). The fibroblast:
sentinel cell and local immune modulator in tumor tissue. Int J Cancer 108, 173-180.
Somasundaram, R., Ruehl, M., Tiling, N., Ackermann, R., Schmid, M., Riecken, E. O., and
Schuppan, D. (2000). Collagens serve as an extracellular store of bioactive interleukin 2. J
Biol Chem 275, 38170-38175.
Stamenkovic, I. (2003). Extracellular matrix remodelling: the role of matrix
metalloproteinases. J Pathol 200, 448-464.
Subramaniam, P. S., Larkin, J., 3rd, Mujtaba, M. G., Walter, M. R., and Johnson, H. M.
(2000). The COOH-terminal nuclear localization sequence of interferon gamma regulates
STAT1 alpha nuclear translocation at an intracellular site. J Cell Sci 113 (Pt 15), 2771-2781.
Subramaniam, P. S., Mujtaba, M. G., Paddy, M. R., and Johnson, H. M. (1999). The carboxyl
terminus of interferon-gamma contains a functional polybasic nuclear localization sequence. J
Biol Chem 274, 403-407.
Taipale, J., and Keski-Oja, J. (1997). Growth factors in the extracellular matrix. Faseb J 11,
51-59.
Tokushige, K., Moradpour, D., Wakita, T., Geissler, M., Hayashi, N., and Wands, J. R.
(1997). Comparison between cytomegalovirus promoter and elongation factor-1 alpha
93
REFERENCES
promoter-driven constructs in the establishment of cell lines expressing hepatitis C virus core
protein. J Virol Methods 64, 73-80.
Trastoy, M. O., Defais, M., and Larminat, F. (2005). Resistance to the antibiotic Zeocin by
stable expression of the Sh ble gene does not fully suppress Zeocin-induced DNA cleavage in
human cells. Mutagenesis 20, 111-114.
Tsung, K., Meko, J. B., Tsung, Y. L., Peplinski, G. R., and Norton, J. A. (1998). Immune
response against large tumors eradicated by treatment with cyclophosphamide and IL-12. J
Immunol 160, 1369-1377.
van der Loos, C. M., Houtkamp, M. A., de Boer, O. J., Teeling, P., van der Wal, A. C., and
Becker, A. E. (2001). Immunohistochemical detection of interferon-gamma: fake or fact? J
Histochem Cytochem 49, 699-710.
Wegrowski, Y., and Maquart, F. X. (2004). Involvement of stromal proteoglycans in tumour
progression. Crit Rev Oncol Hematol 49, 259-268.
Wetzel, R., Perry, L. J., Veilleux, C., and Chang, G. (1990). Mutational analysis of the C-
terminus of human gamma-interferon. Protein Eng 4, 105.
Willimsky, G., and Blankenstein, T. (2005). Sporadic immunogenic tumors avoid destruction
by inducing T-cell tolerance. Nature 437, 141-146.
Wrenshall, L. E., and Platt, J. L. (1999). Regulation of T cell homeostasis by heparan sulfate-
bound IL-2. J Immunol 163, 3793-3800.
Wu, T. H., Pabin, C. N., Qin, Z., Blankenstein, T., Philip, M., Dignam, J., Schreiber, K., and
Schreiber, H. (2004). Long-term suppression of tumor growth by TNF requires a Stat1- and
IFN regulatory factor 1-dependent IFN-gamma pathway but not IL-12 or IL-18. J Immunol
172, 3243-3251.
Yard, B. A., Lorentz, C. P., Herr, D., and van der Woude, F. J. (1998). Sulfation-dependent
down-regulation of interferon-gamma-induced major histocompatibility complex class I and
II and intercellular adhesion molecule-1 expression on tubular and endothelial cells by
glycosaminoglycans. Transplantation 66, 1244-1250.
Zigrino, P., Löffek, S., and Mauch, C. (2005). Tumor-stroma interactions: their role in the
control of tumor cell invasion. Biochemie 87, 321-328.
Zou, J., Carrington, A., Collet, B., Dijkstra, J. M., Yoshiura, Y., Bols, N., and Secombes, C.
(2005). Identification and bioactivities of IFN-γ in Rainbow Trout Oncorhynchus mykiss: The
first Th-1-type cytokine characterised functionally in fish. J Immunol 175, 2484-2494.
94
CURRICULUM VITAE
9 CURRICULUM VITAE
DATE OF BIRTH: 31.05.1977
PLACE OF BIRTH: Swan District, Perth, Western Australia
NATIONALITY: Australian, Spanish
MARITAL STATUS: Single
EDUCATION:
PRIMARY SCHOOL:
1983 Normanhurst Primary, Normanhurst, New South
Wales
1984-1985 Nightcliff Primary, Nightcliff, Northern Territory
1986-1988 Saint Patrick’s Primary, Katanning, Western
Australia
1988 Star of the Sea Primary, Rockingham, Western
Australia
SECONDARY SCHOOL:
1989-1994 Iona Presentation College, Mosman Park,
Western Australia
UNIVERSITY:
1996-1998 Bachelor of Science in Biotechnology at Murdoch
University, Perth, Western Australia
1999 Honours in Pathology at the University of
Melbourne, Melbourne, Victoria
PROMOTION: since 2001
Group: Molekulare Immunologie und Gentherapie
Prof. Dr. Thomas Blankenstein
Institute: Max-Delbrück-Centrum für Molekulare Medizin
und
Institut für Immunologie, Charite-UKBF
95
PUBLICATIONS
10 PUBLICATIONS
Qin Z, Schwartzkopff J, Pradera F, Kammertoens T, Seliger B, Pircher H, Blankenstein T. A
critical requirement of interferon gamma-mediated angiostasis for tumor rejection by CD8+ T
cells. Cancer Res. 2003 Jul 15;63(14):4095-100
96
ACKNOWLEDGEMENTS
11 ACKNOWLEDGEMENTS
I would like to acknowledge the help of many people during my PhD.
First, I wish to thank Thomas Blankenstein for the opportunity to work within the laboratory,
and for offering direction.
Thomas Kammertoens my arm of support and trust when things were going well or were
falling apart. Your dedication to science is infectious, your inspirational ideas either verged
on insanity or genius. Thankyou, I wouldn’t be here without you.
To my colleagues: Zhihai Qin for teaching me the basics of science and the yin and yan of
life, Boris Engels for his patience, whenever I needed to give someone an earbashing, Hye-
Jung Kim for her wisdom, friendship, strength and intelligence, Martin Textor for the laughs
and coffee in the stairwell, Jehad Charo the Molecular Biology Master for teaching me the
ways of the force, and the technical assistance I received from Tanja Specowius, Nahid Hakiy
and Christel Westen
Most importantly I wish to thank my family. My parents Juan and Christine who showed me
the infinite possibilities the world had to offer. My brothers Michael and Robert, who were
here in Europe and supported me during this time. I hope one day I can return all the kindness
and love you have given me.
.
97
APPENDIX
12 APPENDIX
A
98
APPENDIX
B
99
APPENDIX
Figure 27: Phylogenetic tree (A) and IFN-γ sequences (B) showing the homology of
known IFN-γ molecules from different species as determined by Vector AlignmentTM.
In (A) the phylogenetic tree is built using the Neighbour Joining (NJ) method of Saitou and
Nei. The NJ method works on a matrix of distances between all pairs of sequences to be
analysed and these distances are related to the degree of divergence between the sequences.
The values in parenthesis are the calculated distance values. The colours in (B) indicate
identical (yellow), similar (blue) and weakly similar (green) amino acids. The red box
highlights the C-terminal LRKRKRSR sequence capable of binding to heparan sulphate.
100
APPENDIX
A
B
C
Figure 28: Plasmid Map of the constructs used to transfect J558L tumour cells. The
Human EF-1 alpha promoter from the pEF/Bsd vector (A) replaced the pCMV promoter from
the pSecTag2 vector (B) to create the backbone plasmid pTag (C).
101
APPENDIX
A
gacggatcgggagatcctcgaggagacctgcaaagatggataaagttttaaacagagaggaatctttgcagctaatggaccttctaggt
cttgaaaggagtgggaattggctccggtgcccgtcagtgggcagagcgcacatcgcccacagtccccgagaagttgtggggagggg
tcggcaattgaaccggtgcctagagaaggtggcgcggggtaaactgggaaagtgatgtcgtgtactggctccgcctttttcccgaggg
tgggggagaaccgtatataagtgcagtagtcgccgtgaacgttctttttcgcaacgggtttgccgccagaacacaggtaagtgccgtgt
gtggttcccgcgggcctggcctctttacgggttatggcccttgcgtgccttgaattacttccacctggctgcagtacgtgattcttgatccc
gagcttcgggttggaagtgggtgggagagttcgaggccttgcgcttaaggagccccttcgcctcgtgcttgagttgaggcctggcctg
ggcgctggggccgccgcgtgcgaatctggtggcaccttcgcgcctgtctcgctgctttcgataagtctctagccatttaaaatttttgatg
acctgctgcgacgctttttttctggcaagatagtcttgtaaatgcgggccaagatctgcacactggtatttcggtttttggggccgcgggcg
gcgacggggcccgtgcgtcccagcgcacatgttcggcgaggcggggcctgcgagcgcggccaccgagaatcggacgggggtag
tctcaagctggccggcctgctctggtgcctggcctcgcgccgccgtgtatcgccccgccctgggcggcaaggctggcccggtcggc
accagttgcgtgagcggaaagatggccgcttcccggccctgctgcagggagctcaaaatggaggacgcggcgctcgggagagcg
ggcgggtgagtcacccacacaaaggaaaagggcctttccgtcctcagccgtcgcttcatgtgactccacggagtaccgggcgccgtc
caggcacctcgattagttctcgagcttttggagtacgtcgtctttaggttggggggaggggttttatgcgatggagtttccccacactgagt
gggtggagactgaagttaggccagcttggcacttgatgtaattctccttggaatttgccctttttgagtttggatcttggttcattctcaagcct
cagacagtggttcaaagtttttttcttccatttcaggtgtcgtgaggaattagcttggtacctagccaccatggagacagacacactcctgc
tatgggtactgctgctctgggttccaggttccactggtgacgcggcccagccggccaggcgcgccgtacgaagcttggtaccgagctc
ggatccactccagtgtggtggaattctgcagatatccagcacagtggcggccgctcgaggagggcccgaacaaaaactcatctcaga
agaggatctgaatagcgccgtcgaccatcatcatcatcatcattgaatagcgccgtcgaccatcatcatcatcatcattgagtttaaaccc
gctgatcagcctcgactgtgccttctagttgccagccatctgttgtttgcccctcccccgtgccttccttgaccctggaaggtgccactccc
actgtcctttcctaataaaatgaggaaattgcatcgcattgtctgagtaggtgtcattctattctggggggtggggtggggcaggacagca
agggggaggattgggaagacaatagcaggcatgctggggatgcggtgggctctatggcttctgaggcggaaagaaccagctgggg
ctctagggggtatccccacgcgccctgtagcggcgcattaagcgcggcgggtgtggtggttacgcgcagcgtgaccgctacacttgc
cagcgccctagcgcccgctcctttcgctttcttcccttcctttctcgccacgttcgccggctttccccgtcaagctctaaatcggggcatcc
ctttagggttccgatttagtgctttacggcacctcgaccccaaaaaacttgattagggtgatggttcacgtagtgggccatcgccctgata
gacggtttttcgccctttgacgttggagtccacgttctttaatagtggactcttgttccaaactggaacaacactcaaccctatctcggtctat
tcttttgatttataagggattttggggatttcggcctattggttaaaaaatgagctgatttaacaaaaatttaacgcgaattaattctgtggaat
gtgtgtcagttagggtgtggaaagtccccaggctccccagcaggcagaagtatgcaaagcatgcatctcaattagtcagcaaccaggt
gtggaaagtccccaggctccccagcaggcagaagtatgcaaagcatgcatctcaattagtcagcaaccatagtcccgcccctaactcc
gcccatcccgcccctaactccgcccagttccgcccattctccgccccatggctgactaattttttttatttatgcagaggccgaggccgcc
tctgcctctgagctattccagaagtagtgaggaggcttttttggaggcctaggcttttgcaaaaagctcccgggagcttgtatatccattttc
ggatctgatcagcacgtgttgacaattaatcatcggcatagtatatcggcatagtataatacgacaaggtgaggaactaaaccatggcca
agttgaccagtgccgttccggtgctcaccgcgcgcgacgtcgccggagcggtcgagttctggaccgaccggctcgggttctcccggg
acttcgtggaggacgacttcgccggtgtggtccgggacgacgtgaccctgttcatcagcgcggtccaggaccaggtggtgccggaca
acaccctggcctgggtgtgggtgcgcggcctggacgagctgtacgccgagtggtcggaggtcgtgtccacgaacttccgggacgcc
tccgggccggccatgaccgagatcggcgagcagccgtgggggcgggagttcgccctgcgcgacccggccggcaactgcgtgcac
ttcgtggccgaggagcaggactgacacgtgctacgagatttcgattccaccgccgccttctatgaaaggttgggcttcggaatcgttttc
cgggacgccggctggatgatcctccagcgcggggatctcatgctggagttcttcgcccaccccaacttgtttattgcagcttataatggtt
acaaataaagcaatagcatcacaaatttcacaaataaagcatttttttcactgcattctagttgtggtttgtccaaactcatcaatgtatcttatc
atgtctgtataccgtcgacctctagctagagcttggcgtaatcatggtcatagctgtttcctgtgtgaaattgttatccgctcacaattccaca
caacatacgagccggaagcataaagtgtaaagcctggggtgcctaatgagtgagctaactcacattaattgcgttgcgctcactgcccg
ctttccagtcgggaaacctgtcgtgccagctgcattaatgaatcggccaacgcgcggggagaggcggtttgcgtattgggcgctcttcc
gcttcctcgctcactgactcgctgcgctcggtcgttcggctgcggcgagcggtatcagctcactcaaaggcggtaatacggttatccac
agaatcaggggataacgcaggaaagaacatgtgagcaaaaggccagcaaaaggccaggaaccgtaaaaaggccgcgttgctggc
gtttttccataggctccgcccccctgacgagcatcacaaaaatcgacgctcaagtcagaggtggcgaaacccgacaggactataaaga
taccaggcgtttccccctggaagctccctcgtgcgctctcctgttccgaccctgccgcttaccggatacctgtccgcctttctcccttcgg
gaagcgtggcgctttctcaatgctcacgctgtaggtatctcagttcggtgtaggtcgttcgctccaagctgggctgtgtgcacgaacccc
ccgttcagcccgaccgctgcgccttatccggtaactatcgtcttgagtccaacccggtaagacacgacttatcgccactggcagcagcc
actggtaacaggattagcagagcgaggtatgtaggcggtgctacagagttcttgaagtggtggcctaactacggctacactagaagga
cagtatttggtatctgcgctctgctgaagccagttaccttcggaaaaagagttggtagctcttgatccggcaaacaaaccaccgctggta
gcggtggtttttttgtttgcaagcagcagattacgcgcagaaaaaaaggatctcaagaagatcctttgatcttttctacggggtctgacgct
cagtggaacgaaaactcacgttaagggattttggtcatgagattatcaaaaaggatcttcacctagatccttttaaattaaaaatgaagtttt
102
APPENDIX
aaatcaatctaaagtatatatgagtaaacttggtctgacagttaccaatgcttaatcagtgaggcacctatctcagcgatctgtctatttcgtt
catccatagttgcctgactccccgtcgtgtagataactacgatacgggagggcttaccatctggccccagtgctgcaatgataccgcga
gacccacgctcaccggctccagatttatcagcaataaaccagccagccggaagggccgagcgcagaagtggtcctgcaactttatcc
gcctccatccagtctattaattgttgccgggaagctagagtaagtagttcgccagttaatagtttgcgcaacgttgttgccattgctacagg
catcgtggtgtcacgctcgtcgtttggtatggcttcattcagctccggttcccaacgatcaaggcgagttacatgatcccccatgttgtgca
aaaaagcggttagctccttcggtcctccgatcgttgtcagaagtaagttggccgcagtgttatcactcatggttatggcagcactgcataa
ttctcttactgtcatgccatccgtaagatgcttttctgtgactggtgagtactcaaccaagtcattctgagaatagtgtatgcggcgaccgag
ttgctcttgcccggcgtcaatacgggataataccgcgccacatagcagaactttaaaagtgctcatcattggaaaacgttcttcggggcg
aaaactctcaaggatcttaccgctgttgagatccagttcgatgtaacccactcgtgcacccaactgatcttcagcatcttttactttcaccag
cgtttctgggtgagcaaaaacaggaaggcaaaatgccgcaaaaaagggaataagggcgacacggaaatgttgaatactcatactcttc
ctttttcaatattattgaagcatttatcagggttattgtctcatgagcggatacatatttgaatgtatttagaaaaataaacaaataggggttcc
gcgcacatttccccgaaaagtgccacctgacgtc
B
1 50
pAlaNoTag (1)
pMC2NoTag (1)
pMC2Tag (1)
pTag (1)
51 100
pAlaNoTag (51)
pMC2NoTag (51)
pMC2Tag (51)
pTag (51)
101 150
pAlaNoTag (101) TGC GCACGGC GC-----AGCCGCCCTGC
pMC2NoTag (101) TCT G A CGG ------AGGGTC CTGC
pMC2Tag (101) TCT -GCGGAAAAGGAGTCGCTCGA
pTag (101) CT -CA----------------TGT
151 200
pAlaNoTag (144) CG
pMC2NoTag (145) -G
pMC2Tag (148) TT
pTag (130) GG
201 250
pAlaNoTag (194) -----------------------
pMC2NoTag (194) -----------------------
pMC2Tag (198) TGAATAGCGCCGTCGACCATCATC
pTag (180) TGAATAGCGCCGTCGACCATCATC
251 261
pAlaNoTag (220) -----------
pMC2NoTag (220) -----------
pMC2Tag (248) ATCATCATCAT
pTag (230) ATCATCATCAT
ATGGAGACAGACACACTCCTGCTATGGGTACTGCTGCTCTGGGTTCCAGG
ATGGAGACAGACACACTCCTGCTATGGGTACTGCTGCTCTGGGTTCCAGG
ATGGAGACAGACACACTCCTGCTATGGGTACTGCTGCTCTGGGTTCCAGG
ATGGAGACAGACACACTCCTGCTATGGGTACTGCTGCTCTGGGTTCCAGG
TTCCACTGGTGACGCGGCCCAGCCGGCCAGGCGCGCCGTACGAAGCTTGG
TTCCACTGGTGACGCGGCCCAGCCGGCCAGGCGCGCCGTACGAAGCTTGG
TTCCACTGGTGACGCGGCCCAGCCGGCCAGGCGCGCCGTACGAAGCTTGG
TTCCACTGGTGACGCGGCCCAGCCGGCCAGGCGCGCCGTACGAAGCTTGG
TACCGAGCTCGGATCC C-- G G
TACCGAGCTCGGATCC CAG G G
TACCGAGCTCGGATCC C-- G G
TACCGAGCTCGGATCC C-- G-- GTG
AATTCTGCAGATATCCAGCACAGTGGCGGCCGCTCGAGGAGGGCCCGA
AATTCTGCAGATATCCAGCACAGTGGCGGCCGCTCGAGGAGGGCCCGA
AATTCTGCAGATATCCAGCACAGTGGCGGCCGCTCGAGGAGGGCCCGA
AATTCTGCAGATATCCAGCACAGTGGCGGCCGCTCGAGGAGGGCCCGA
ACAAAAACTCATCTCAGAAGAGGATC-
ACAAAAACTCATCTCAGAAGAGGATC-
ACAAAAACTCATCTCAGAAGAGGATC
ACAAAAACTCATCTCAGAAGAGGATC
T
A AAA A
AGAAA
A
103
APPENDIX
C
1 50
pAlaNoTag (1) ---------
pMC2NoTag (1) LR---------
pMC2Tag (1) LRKRKRSRCEL
pTag-puro (1) L-----CGG
51 86
pAlaNoTag (42) --AAAA AEFCRY PL SQKRI--
pMC2NoTag (42) -- SR CRY PL SQKRI--
pMC2Tag (51) ILQISSTVAAARGG IS V HH
pTag-puro (45) ILQISSTVAAARGG IS V HH
METDTLLLWVLLLWVPGSTGDAAQPARRAVRSLVPSSDPAA
METDTLLLWVLLLWVPGSTGDAAQPARRAVRSLVPSSDP
METDTLLLWVLLLWVPGSTGDAAQPARRAVRSLVPSSDP
METDTLLLWVLLLWVPGSTGDAAQPARRAVRSLVPSSDP
PAQWR EEGPN
PAQWR EEGPN
PEQKL EE NSA
PEQKL EE NSA
Q-
AA KNS
KRKRCEF KNS
DLDHHHH
DLDHHHH
Figure 29: Nucleotide and Amino acid sequences of the plasmids used to transfect J558L
tumour cells. The sequence of the plasmid backbone pTag (A) was used to create plasmids
pAla8NoTag, pMC2NoTag and pMC2Tag. Oligonucleotides were cloned between BamH1
and Eco R1. The nucleotide sequence (B) and amino acid sequences (C) of the secreted
fragments are included.
104
APPENDIX
Tail Blood
Figure 30: IFN-γ gene PCR of the blood and tail from selected mice from the bone
marrow chimera experiment.
Each number represents an individual mouse within the experimental group. 12 weeks after
irradiation with 10 Gy and reconstitution with bone marrow mouse tail and blood were lysed
and DNA was isolated for PCR using primers specific for IFN-γ (See materials and methods).
105
Berlin 2006
Ich erkläre an Eides Statt, dass die vorliegende Dissertation in allen Teilen von mir
selbständig angefertigt wurde und die benutzten Hilfsmittel vollständig angegeben worden
sind.
Veröffentlichungen von irgendwelchen Teilen der vorliegenden Dissertation sind von mir
nicht / wie folgt vorgenommen worden.
Weiter erkläre ich, dass ich nicht schon anderweitig einmal die Promotionsabsicht angemeldet
oder ein Promotionseröffnungsverfahren beantragt habe.
Felicia Pradera, Berlin, 2006
