Isoform specific Loss of CD44 Interferes with Different Aspe

Isoform specific Loss of CD44 Interferes with Different Aspe
submitted to the
Combined Faculties for the Natural Sciences and for Mathematics
of the
Ruperto-Carola University of Heidelberg, Germany
for the degree of
Doctor of Natural Sciences
presented by
Dipl. Biol. Pamela Klingbeil
born in Berlin
Isoform-specific Loss of CD44 Interferes with
Different Aspects of the Metastatic Process
PD Dr. Jochen Wittbrodt
Prof. Dr. Margot Zöller
Whenever you fall,
pick something up.
Oswald Theodore Avery
Table of Contents
Table of Contents
Summary.................................................................................................................................. 1
Zusammenfassung.................................................................................................................... 2
Introduction.................................................................................................................. 4
Cancer evolves as a multistep process..................................................................................
The BSp73 cell system.........................................................................................................
The cell-cell and cell-matrix adhesion molecule CD44.......................................................
Structural properties of CD44.........................................................................................
Different modes of interactions for CD44......................................................................
Physiological and pathological functions ascribed to CD44..........................................
CD44 in tumor progression.............................................................................................
RNAinterference as a tool to study isoform specific gene functions...................................
Aims of the thesis.................................................................................................................
Results........................................................................................................................... 25
Establishment of stable CD44vk.d. cell lines and rescue clones.........................................
RNAi construct evaluation by FACS and fluorescence microscopy..............................
Establishment of stable CD44vk.d. clones by selection and recloning..........................
Restoring CD44 expression by introduction of mutated cDNAs...................................
Characterization of the knock-down cell lines in vivo.........................................................
CD44vk.d. cells exhibit a reduced metastatic capacity in vivo.......................................
Characterization of knock-down cell lines in vitro..............................................................
CD44vk.d. cells show no phenotypic changes................................................................
CD44vk.d. cells show no altered growth behaviour.......................................................
CD44vk.d. cells do not differ in MMP2 and MMP9 expression....................................
ASMLwt but not CD44vk.d. cells aggregate in stromal cell culture supernatant..........
ASMLwt cells produce an adhesive matrix, which is impaired in the CD44vk.d.........
Adhesion promoting components are secreted..........................................................
The secreted matrix contains HA, collagen and laminin...........................................
Adhesion to the secreted matrix is mediated by β1 integrin......................................
CD44vk.d. cells lack a secreted 180 kDa protein...........................................................
CD44vk.d. cells exhibit a reduced resistance to apoptotic triggers................................
Apoptosis resistance is increased by elongated pre-cultivation prior to irradiation..
PI3K-Akt, rather than MAPK signalling is involved in apoptosis resistance of
ASML cells................................................................................................................
Discussion..................................................................................................................... 52
Materials and Methods............................................................................................... 66
Chemical inhibitors.........................................................................................................
Nucleotide and protein standards....................................................................................
Primers and oligos...........................................................................................................
cDNAs and constructs.....................................................................................................
Primary antibodies.....................................................................................................
Secondary antibodies/ reagents..................................................................................
Table of Contents
4.1.10 Cell lines .........................................................................................................................
4.1.11 Animals...........................................................................................................................
4.2 Methods...................................................................................................................................
Molecular biology...........................................................................................................
Plasmid preparation...................................................................................................
RNAinterference construct design and cloning.........................................................
PCR-based mutagenesis for rescue constructs ..........................................................
RNA-isolation and reverse transcription-PCR (RT-PCR).........................................
Cell biology.....................................................................................................................
Cell culture.................................................................................................................
Cryo-conservation of eukaryotic cells.......................................................................
Transfection of eukaryotic cells................................................................................ .
Recloning of transfected cells by limiting dilution....................................................
Collection of conditioned cell culture supernatant....................................................
Coating of plastic surfaces.........................................................................................
Adhesion assay...........................................................................................................
Agglomeration assay..................................................................................................
Proliferation assay...................................................................................................... Soft agar assay........................................................................................................... Drug treatment............................................................................................................ γ-irradiation of adherent cells.................................................................................... MTT staining of respiratory active cells.................................................................... Crystal violet staining of adherent cells ..................................................................... FACS analysis............................................................................................................ Immunofluorescence staining of cells grown on coverslips...................................... Cryo-sectioning of tumor tissue................................................................................. Immunohistological staining of cryo sections...........................................................
Animal experiments........................................................................................................
In vivo metastasis assay.............................................................................................
Protein biochemistry.......................................................................................................
Surface bioninylation of molecules...........................................................................
Immunoprecipitation (IP)..........................................................................................
Lysis of intact cells for SDS-PAGE..........................................................................
SDS-polyacrylamide gel electrophorese (SDS-PAGE)............................................
Western blotting.......................................................................................................
Colloidal Coomassie staining of protein gels...........................................................
Silver staining of protein gels...................................................................................
Gelatine zymography for detection of MMP activity...............................................
Gel-filtration............................................................................................................. Ultracentrifugation of cell culture supernatant......................................................... TCA-precipitations of proteins ................................................................................. Analysis of proteins by mass spectrometry..............................................................
References................................................................................................................. 84
Acknowledgements................................................................................................................. 98
Abbreviations......................................................................................................................... 99
The cell-cell and cell-matrix adhesion molecule CD44 and its numerous splice
variants are involved in a multitude of physiological and pathological processes, including
tumour progression. Especially variant CD44 has been implicated in metastasis formation.
For long term in vivo experiments on metastasis formation, a plasmid based RNAi
technique was applied to generate stable splice variant ‘v7’-specific CD44 knock-down
clones of a highly metastatic rat pancreatic adenocarcinoma cell line (BSp73ASML). The
resulting phenotype was characterized with an emphasis on interactions of CD44v with the
tumour surrounding during the course of metastasis formation. Loss of CD44v is
accompanied in vivo by a marked reduction in metastatic growth in the lymph nodes and
particularly in the lung, which could be reverted by restoring CD44v expression in the knockdown cells. The impaired metastatic growth was not due to a lower proliferative activity or a
reduced anchorage-independence of these cells in vitro. Instead, they display several defects,
which can be attributed to perturbed interactions of CD44v with the microenvironment.
Compared to ASMLwt cells CD44vk.d. cells do not form cell aggregates in stromal
surroundings, such as lymph nodes and the lung, due to lost cell-cell adhesion, mediated by
interactions of CD44v and hyaluronic acid (HA). Furthermore, CD44vk.d. cells exhibit an
impaired matrix production, as CD44v is most likely involved in the assembly of matrix
components, containing HA, collagen and laminin. The matrix supports rapid adhesion of
ASML cells through β1 integrin and in addition contributes to survival. Finally, the loss of
CD44v is accompanied by a marked decrease in apoptosis resistance. Impaired PI3K-Akt
survival signalling, activated by CD44v was identified as the cause of this defect.
In conclusion, CD44v contributes to the metastatic phenotype of ASML cells as a
multifunctional player interacting with the surrounding in several ways. First, as cell-cell
adhesion molecule by mediating cell aggregation, second, as cell-matrix adhesion molecule
by organizing matrix generation and last, as signalling molecule supporting survival. This
highlights the role of variant CD44 in the metastatic spread of tumour cells through complex
interactions with the tumour microenvironment and underlines the important role of a highly
regulated interplay between tumour cells and their surrounding for metastasis formation.
Das Zell-Zell und Zell-Matrix Adhäsionsmolekül CD44, sowie seine zahlreichen
Spleißvarianten sind an einer Vielzahl physiologischer und pathologischer Prozesse beteiligt,
zu denen auch die Tumorprogression zählt. Besonders variante CD44 Formen wurden mit
Metastasierung in Verbindung gebracht.
Um langwierige in vivo Experimente zur Untersuchung von Metastasenausbildung zu
ermöglichen, wurden stabile CD44-‘knock-down’-Klone einer stark metastasierenden
Pankreas-Adenokarzinomlinie der Ratte (BSp73ASML) generiert. Über plasmidbasierte
‘RNA interference’ (RNAi) wurde die CD44-Expression spleißvarianten-‘v7’-spezifisch
reguliert. Der resultierende Phänotyp wurde besonders im Hinblick auf Interaktionen
zwischen CD44v und der Mikroumgebung im Verlauf der Metastasierung charakterisiert. In
vivo führt der Verlust von CD44v zu deutlich reduzierter Metastasenbildung in den
Lymphknoten und besonders in der Lunge, und dieser Effekt war durch wiederhergestellte
CD44v-Expression in den ‘knock-down’-Zellen wieder umkehrbar. In vitro zeigen die
verankerungsunabhängige Wachstumsvermögen wie der Wiltyp. Demgegenüber weisen sie
mehrere Defekte auf, die auf Interaktionsverlust von CD44v mit der Mikroumgebung
beruhen. In stromaler Umgebung, wie in den Lymphknoten und der Lunge, bilden CD44vk.d.-Zellen im Gegensatz zu ASMLwt-Zellen keine Zellaggregate aus, was auf den Verlust
von CD44 und Hyaluronsäure vermittelten Zell-Zell-Kontakten zurückgeführt werden
konnte. Zusätzlich ist die Matrixproduktion dieser Zellen beeinträchtigt, da CD44v höchst
wahrscheinlich eine Rolle bei der ‘Matrixmontage’ zukommt. Als Bestandteile der Matrix
konnten Hyaluronsäure, sowie Laminin und Kollagen identifiziert werden. Die Matrix
ermöglicht ASML-Zellen eine rasche über β1-Integrin vermittelte Adhäsion und trägt
darüberhinaus zum Überleben der Zellen bei. Schließlich geht der Verlust von CD44v mit
einer deutlichen Abnahme der Apoptoseresistenz einher. Als Ursache für diesen Defekt
konnte eine beeinträchtigte PI3K-Akt-Signaltransduktion identifiziert werden, die durch
CD44v aktiviert wird.
Zusammenfassend konnte gezeigt werden, daß variantes CD44 maßgeblich zum
metastasierenden Phänotyp von ASML Zellen als multifunktionales Molekül beiträgt, indem
es mit der Mikroumgebung auf verschiedene Art und Weise interagiert. Zunächst als ZellZell
Adhäsionsmolekül, das die Bildung der Matrix organisiert, und schließlich als
Signaltransduktionsmolekül, das zum Überleben der Zelle beiträgt. Diese Ergebnisse heben
die Rolle von variantem CD44 für die Metastasenbildung von Krebszellen durch komplexe
Interaktionen mit der Tumor-Mikroumgebung hervor und unterstreichen die entscheidende
Bedeutung von komplex regulierten Wechselwirkungen zwischen Tumorzellen und ihrer
direkten Umgebung für die Metastasierung.
30 years of intensive research in the field of tumour biology produced a huge body of
knowledge and the basic mechanisms underlying the onset and progression of cancer have
been identified. Nonetheless, after cardiovascular diseases cancer is still the 2nd leading cause
of death in the western world (WHO, 2003).
Cancer arises from a single cell that underwent genomic alterations leading to gainof-function of so called oncogenes or loss-of-function of tumour suppressor genes enabling
uncontrolled growth and evading the bodies defence system to eliminate cells with
dysfunctions. Our understanding of the mechanisms underlying early tumour progression is
steadily growing, concomitantly with remarkable advances in the diagnosis and treatment of
early tumours. However, there is still little understanding of the late steps in tumour
progression leading to metastasis formation, which causes 90% of human cancer deaths
(Storm, 1996). Although there is a growing number of genes being identified to take part, the
underlying mechanisms that enable cancer cells to disseminate from the primary tumour mass
and settle at distant sites in the body to form metastases is still poorly understood and deeper
insights are needed for future therapeutic strategies to treat metastatic cancers.
Cancer evolves as a multistep process
During tumourigenesis the transformation of a normal cell into a malignant cancer
follows a multistep process, which can be understood as an evolutionary event following the
Darwinian concept. In order to develop into a life threatening invasive tumour a cell has to
acquire certain characteristics, reflecting genetic alterations, which confer a growth
advantage and drive the progressive transformation of the cell (Foulds, 1954; Nowell, 1976).
Hanahan and Weinberg proposed six essential capabilities required for metastatic cancer
formation (Hanahan and Weinberg, 2000), namely self sufficiency in growth signals,
insensitivity to antigrowth signals, evasion of programmed cell death (apoptosis), limitless
replicative potential, sustained angiogenesis and tissue invasion and metastasis.
Self sufficiency in growth signals
Normal cells need growth stimuli in order to proliferate. These signals can be
diffusible growth factors, extracellular matrix components or cell-cell stimulations. Usually
secreted by other cell types (heterotypic signalling) these signals are sensed mainly by
transmembrane receptors (binding to diffusible growth factors) and integrins (binding to
components of the ECM), which translate the outside stimulus into an inside signal. Many
cancer cell lines are independent on such exogenous growth stimulation because the activity
of oncogenes mimics these growth signals by modulating the underlying stimulatory
machinery at different levels. Tumour cells can either secrete their own growth factors
(autocrine stimulation) or modify the corresponding signals within the cell, by modulating the
receptors itself or the downstream signalling circuits. Alteration of growth factor receptor
signalling can be achieved at the expression level leading to hyper-responsiveness to a given
extracellular signal or by modulation of the signalling ability of the receptor, like expressing
constitutively active versions of the receptor. A prominent example for this is the truncated
version of the EGF-receptor (Fedi, 1997). Alternatively, the underlying growth signalling
circuits itself can be modulated. The Ras-Raf-MAP kinase pathway for example is altered in
25% of human tumours leading to mitogenic signals without ongoing upstream stimulation
(Medema and Bos, 1993). Cells are also dependent on growth stimuli from the underlying
ECM. Many cell-matrix interactions are regulated by integrins, which are heterodimeric cell
surface adhesion molecules composed of one α and one β subunit. To date, 18 different α
and 8 different β subunits have been identified, which form at least 24 heterodimers with
different characteristics (Hynes, 2002; Shimaoka and Springer, 2003). Cancer cells can
change their integrin repertoire by varying the combination of α and β subunits, for instance
to favour expression to integrins eliciting progrowth signals at the expense of integrins with
antiproliferative effects. In addition to this enhanced autonomy from their surrounding,
tumour cells are able to modulate the behaviour of their neighbourhood for their own benefit.
For example cancer cells can induce excess release of growth factors by neighbouring cells
(Skobe and Fusenig, 1998) or stimulate inflammatory cells that should rather eliminate
tumour cells, to promote their growth instead (Cordon-Cardo and Prives, 1999; Coussens et
al., 1999; Hudson et al., 1999).
Insensitivity to antigrowth signals
Just as normal cells display dependence on progrowth signals they are sensitive to
antigrowth signals. This ensures tissue homeostasis and cellular quiescence. Antiproliferative
signals can, like their positively acting counterparts, be soluble growth inhibitors and
immobilized inhibitors present in the ECM or on neighbouring cells that are sensed by
transmembrane receptors. Such inhibitory stimuli may lead to cell cycle arrest or induce a
permanent mitotic stop by driving the differentiation of the cell. Most antiproliferative signals
are relayed via the ‘retinoblastoma protein’ (pRb) pathway. PRb is a classical tumour
suppressor gene, which controls progression from G1 to S phase of the cell cycle (Weinberg,
1995). Cancer cells often fail to respond properly to antigrowth signals due to disruption of
this pathway. Inactivation of the pRb protein leads to cell cycle arrest, either by mutation of
the gene itself or by interfering with ‘transforming growth factor β’ (TGFβ) signalling, which
normally blocks phosphorylation and thereby inactivation of pRb (Hannon and Beach, 1994;
Datto et al., 1997). Blocking TGFβ signalling is achieved by cancer cells through downregulation of the corresponding receptor, by expression of dysfunctional receptors (Fynan
and Reiss, 1993; Markowitz et al., 1995) or by modulating downstream signalling events
(Schutte et al., 1996; Chin et al., 1998). Integrins can elicit antigrowth signals as well, and
their expression pattern exerting proliferative or antiproliferative effects can be modulated as
described above. Another way to avoid growth arrest is to circumvent differentiation. For
example, antagonizing c-Myc function can lead to cellular differentiation (Dang et al., 1999)
and recently a role for pRb in this context was shown as well (Goodrich, 2006).
Evading apoptosis
The third aspect limiting the expansion of a cell population is controlled by cross talk
between the cell and the environment, deciding if a cell should live or die. Programmed cell
death or apoptosis is the major event regulating this subject and it is evident that resistance to
apoptosis is a key feature of cancer cells and for metastasis formation. Programmed cell death
is a mechanism common in virtually all cell types and follows the same routes once it is
induced by a variety of different triggers. Cell membranes are disrupted, the cytoskeleton and
chromosomes are degraded and the cell usually dies within 24 hours (Wyllie et al., 1980).
Apoptosis can be induced via two distinct routes, depending on the source of the apoptotic
stimulus. The extrinsic or death receptor pathway is induced by surface receptors like Fas and
‘tumour necrosis factor receptor’ (TNFR) that bind their corresponding ligands and induce a
cellular signalling cascade eventually leading to apoptosis. On the other hand there are
several intracellular sensors detecting abnormalities within the cell, such as DNA damage,
survival factor insufficiency or other stress signals like hypoxia (Evan and Littlewood, 1998).
Such triggers induce the intrinsic or mitochondrial pathway, usually by activating pro- or
inhibiting antiapoptotic members of the Bcl-2 protein family. These signals lead to
breakdown of the mitochondrial membrane potential, release of cytochrome c and finally to
apoptosis (Green and Reed, 1998). The p53 tumour suppressor gene for example, which is
inactivated in more than 50% of human cancers, exerts its apoptotic effects via the intrinsic
pathway. In response to DNA damage p53 upregulates the proapoptotic Bcl-2 protein Bax,
which leads to cytochrome c release from the mitochondria (Harris, 1996). Both the external
and intrinsic pathway finally converge on the ultimate effectors of apoptosis, intracellular
proteases called caspases (Thornberry and Lazebnik, 1998). Two initiator caspases 8 and 9
are activated either by death receptor signalling or by cytochrome c release, which then in
turn activate several effector caspases, executing downstream steps of the apoptotic program
and eventually leading to degradation of cellular structures, organelles and the genomic
material (Fig. 1).
Fig.1: The two main death pathways of apoptosis
Death receptors on the surface of the cell bind their cognate ligands, leading to engagement of the
‘death inducing signalling complex’ (DISC) and to activation of caspase 8. Internal stimuli, such as
genotoxic stress induce apoptotic molecules, finally leading to inhibition of antiapoptotic molecules
of the Bcl-2 family, subsequent cytochrome c release from the mitochondria and caspase 9 activation.
Ultimately, both routes lead to activation of effector caspases and cell death. IAP inhibitor of
apoptosis proteins. Adapted by permission from Macmillan Publishers Ltd: Nature Reviews Cancer
(Mehlen and Puisieux., 2006).
Cancer cells escape from apoptosis either by deregulating the expression of death
receptors on their surface or by survival signals, counteracting apoptotic stimuli through
induction of antiapoptotic molecules. One of the best-studied pathways in this regard is the
PI3kinase-Akt pathway. PI3K-Akt can be influenced by external stimuli, for example IGF or
IL-3 (Evan and Littlewood, 1998), or by intracellular signals, e.g. the ras oncogene pathway
(Downward, 1998) or the loss of the tumour suppressor pTEN, which normally attenuates
Akt survival signals (Cantley and Neel, 1999). Likewise, cell-cell and cell-matrix interactions
are necessary for cell survival. Apoptosis due to loss of adhesion to the substratum is called
anoikis and provides another important mechanism to control cellular behaviour.
In summary, apoptosis is regulated by an extremely complex machinery of pro- and
antiapoptotic signals, which are balanced in a healthy cell and shifted out of this equilibrium
in cancer cells.
Limitless replicative potential
In addition to signals regulating the growth properties of a cell in response to its
environment, the life span of a normal cell underlies an intrinsic program limiting its
replicative potential. Non-transformed cells in culture have the capacity to divide only about
60-70 times. With increasing cell divisions they undergo a process termed senescence and
stop growing. This process can be circumvented for example by blockage of the pRb and the
p53 tumour suppressor genes, enabling the cell to further multiply until a second state called
crisis, is reached. This state is characterized by massive cell death and increased karyotypic
rearrangements, like chromosome fusions, and eventually enables a cell to replicate without
limit (Wright et al., 1989). Most isolated tumour cells demonstrate this immortalization in
culture, arguing for immortalization as a prerequisite for cancer to evolve. However,
theoretically 60-70 doublings are more than enough for a cell to expand to a life-threatening
tumour mass. On the other hand, the apoptotic rate within tumours is high and the actual cell
number of a tumour greatly under represents the cell generations required to produce it. As a
result of this, limitless replicative potential would be a prerequisite for cancerous
The molecular mechanism underlying senescence is based on the replication of the
genomic material. During each cell cycle the whole genome is replicated but the ends of the
chromosomes, called telomers are shortened through each round of replication, due to the
inability of the appropriate polymerase to replicate the chromosome completely. Telomers do
not carry any genomic information, but protect the ends of the chromosomal DNA. They
prevent end to end fusions and karyotypic misarrangements, which are associated with crisis
and result in the death of the cell (Counter et al., 1992). Telomer stability is observed in
nearly all types of cancer and cancer cells maintain the integrity of their telomers by overexpressing an enzyme called telomerase, which elongates the telomeric DNA and
compensates for the shortening during replication (Bryan and Cech, 1999). To a lesser extent,
telomer maintenance in cancer cells can also be achieved by recombination-based
interchromosomal exchanges of sequences termed ‘alternative lengthening of telomers’
(ALT) (Bryan et al., 1995).
Sustained angiogenesis
Every growing tissue is dependent on supply with nutrition and oxygen, which is
guaranteed in healthy tissue by a dense network of blood vessels. Tumours also depend on
the vasculature to be able to grow beyond a certain size. During tissue development the
formation of blood vessels from pre-existing vessels (angiogenesis) and the de novo
formation of new blood vessels (neovascularization) is regulated by complex interactions
between the tissue and the endothelial cells of the vasculature. Positive and negative
regulators of angiogenesis are tightly balanced and transiently regulated in order to induce
blood vessel outgrowth. Mainly soluble factors, like ‘vascular endothelial growth factors’
(VEGFs) and ‘fibroblast growth factors’ (FGFs), which attract endothelial cells expressing
the cognate surface receptors (Veikkola and Alitalo, 1999) or inhibitors of angiogenesis like
thrombospondin (Bull et al., 1994) regulate the process of neovascularization. In order to
grow out into a macroscopic tumour, cancer cells have to interfere with this programme by
shifting the balance to the angiogenic inducers and counteracting the inhibitors (Hanahan and
Folkman, 1996). Indeed many tumours reveal enhanced expression of VEGFs and FGFs and
impaired expression of thrombospondin (Singh et al., 1995; Volpert et al., 1997). Activators
or inhibitors of angiogenesis can also be stored in the ECM and released by proteases, that
are expressed by tumour cells (Whitelock et al., 1996). The importance of sustained
angiogenesis for tumour growth has been shown in several studies (Bouck et al., 1996;
Hanahan and Folkman, 1996; Folkman, 1997) and is a target for therapeutic intervention.
Tissue invasion and metastasis
90% of human cancer deaths are due to the formation of secondary tumours (Sporn,
1996) arising from cells, that moved out of the primary tumour mass and settled at distant
sites in the body. The metastatic spread of tumour cells can be understood as a multistep
process itself. In order to settle at a secondary site and to form a new tumour, a cancer cell
has to acquire the ability to break down cell-cell and cell-matrix connections, invade adjacent
tissues by overcoming tissue borders and acquire a migratory phenotype. Then, metastatic
cells have to enter the blood circulation or the lymphatic system to be transported to distant
sites, where they have to exit the circulation again in order to form secondary tumours
Fig. 2: Two routes of metastasis formation
Tumour cells disseminate from the primary tumour and invade adjacent tissues. After intravasation
they use the blood or the lymphatic system for transportation. Via lymphatic dissemination the cells
are transported to lymph nodes, where metastases are formed. Subsequently, tumour cells can enter
the blood system through this route as well. Circulating tumour cells in the blood have to attach to the
endothelial vessel wall, extravasate and settle at a secondary site to form distant metastases. Adapted
by permission from Macmillan Publishers Ltd: Nature Reviews Cancer (Pantel and Brakenhoff,
The first step during the metastatic process is the dissemination from the primary
tumour mass. Adhesion to neighbouring cells and to the ECM is lowered by metastasising
cells through modulating the functions of cell adhesion molecules (CAMs) of the
immunoglobulin, cadherin and integrin families and other CAMs. For example, the function
of E-cadherin, a homotypic cell-cell interaction molecule ubiquitously expressed on all
epithelial cells, is altered in many carcinomas during progression towards malignancy
(Christofori and Semb, 1999). The loss of E-cadherin mediated cell-cell adhesion has been
proposed to be a prerequisite for tumour cell invasion and metastasis (Birchmeier and
Behrens, 1994). Indeed re-establishing functional E-cadherin expression could reverse the
invasive phenotype of cultured tumour cells (Vleminckx et al., 1991; Birchmeier and
Behrens, 1994). The loss of E-cadherin expression often is accompanied by de novo
expression of motility promoting cadherins, like N-cadherin (Li and Herlyn, 2000; Tomita et
al., 2000). Such a ‘cadherin switch’ occurs during normal embryonic development, when
epithelial cells acquire a migratory phenotype (Hatta and Takeichi, 1986; Bendel-Stenzel et
al., 2000). Another prominent example is N-CAM, the expression of which is switched from
a highly to a poorly adhesive isoform in certain cancers (Johnson, 1991; Kaiser et al., 1996)
or is generally downregulated in others (Fogar et al., 1997). Finally, cancer cells are able to
adapt to a new environment and to modulate their interaction with matrix components, such
as collagen, laminin etc., e.g. by changing their integrin repertoire. It has been reported that
tumour cells can switch integrins in a way that attachment to proteolytically degraded ECM
components and migration is favoured over tight adhesion to normal epithelial matrix.
(Varner and Cheresh, 1996; Lukashev and Werb, 1998).
In order to invade adjacent tissues, tumour cells may express extracellular proteases
by themselves or induce their expression in neighbouring stromal cells (Werb, 1997).
Degradation of matrix components by proteases enables tumour cells to overcome restrictions
due to cell-cell contacts and to invade neighbouring tissues. Prominent examples are the
‘matrix metalloproteases’ (MMPs), a family of secreted or transmembrane proteases capable
of degrading various matrix components. Cancer cells can also deregulate MMP function by
modulating the expression of specific inhibitors (TIMPs) or of molecules that activate MMPs
through enzymatic cleavage.
Cancer cells use the blood circulation or the lymphatic system for transportation to
secondary sites. To enter a blood vessel a cancer cell has to pass the endothelial cell layer, a
process called intravasation. The majority of cells entering the blood stream will die due to
mechanical stress or elimination by the immune system, a process called immune
surveillance (Jakobisiak et al., 2003). Because lymphatic vessels consist of more loosely
associated cells the lymphatic system is easier to enter (Alitalo and Carmeliet, 2002).
Although vascular and lymphatic spread of tumour cells share the basic features, the
molecular mechanisms might differ. Once transported to a secondary site, a cancer cell has to
rest and extravasate out of the circulation, which means that the cell has to build up adhesive
properties again to be able to settle. In order to grow out and form a new tumour mass a
cancer cell has to adapt to the new environment, build up cell-cell and cell-matrix contacts
again and fulfil all the described capabilities to form a massive tumour.
In conclusion, the process of metastatic outgrowth above all requires dynamic
adhesive properties of cancer cells, switching between a sessile and a motile state when
needed. Recent evidence highlights that the crosstalk of a cancer cell and the
microenvironment at both the primary and the secondary site plays a key role for the
metastatic success of cancer cells (reviewed by Schedin and Elias, 2004).
The BSp73 cell system
In order to study late steps in tumour progression cell systems have been established,
comprising two related cell lines, which exhibit differences in their metastatic potential.
These cell lines may originate from the primary tumour and a metastatic lesion or may
present two subclones derived from the same primary tumour. The BSp73 tumour cell system
represents such a model. A stable cell line BSp73 was established from a spontaneous
pancreatic adenocarcinoma of a BDX rat (Zoller et al., 1978). Several in vivo passages after
subcutaneous application gave rise to tumours with different metastatic potential and finally
led to the establishment of two subclones of BSp73. One clone, BSp73AS (AS), displaying
only weak metastatic growth, whereas the other, BSp73ASML (ASML), exhibits a very high
metastatic potential (Matzku et al., 1983). When injected into the footpad of syngenic rats,
AS cells show strong local tumour growth but only reach the draining lymph nodes. In
contrast, ASML cells display only very limited local tumour growth, rapidly spread through
the lymphatic system and form miliary metastases in the lung, which will finally kill the
animal. Different to AS cells which show a spread out epithelial morphology with long
filopodia, ASML cells display a rounded cell shape without any visible spreading on
substrate and a very limited ability to bind to matrix components, such as laminin, fibronectin
and collagen. Moreover, these cells do not form a closed monolayer but rather detach from
the substrate before reaching full confluency. (Matzku et al., 1985; Ben-Ze’ev et al., 1986;
Raz et al., 1986).
Generation of monoclonal antibodies against membrane preparations of the two cell
lines (Matzku et al., 1989) led to the identification of 5 differentially expressed surface
molecules, namely a variant isoform of the cell adhesion molecule CD44 (Gunthert et al.,
1991), the α6β4 integrin (Herlevsen et al., 2003), C4.4A, a molecule showing high homology
to the uPA receptor (Rosel et al., 1998), the ephithelial cell adhesion molecule EpCAM
(Wuerfel et al., 1999) and the tetraspanin D6.1A (Claas et al., 1998). Subsequent expression
profiling of two sublines revealed several hundred differentially expressed genes (Nestl et a.,
2001; Tarbe et al., 2002).
The cell-cell and cell-matrix adhesion molecule CD44
One of the first molecules being described to be involved in metastasis formation is
CD44. The described BSp73 cell system was used to show that a single CD44 splice variant
conferred metastatic potential to the otherwise locally growing BSp73AS tumour (Gunthert et
al., 1991). CD44, first assigned as ‘lymphocyte homing receptor’ (Gallatin et al., 1983), is a
broadly distributed single pass transmembrane glycoprotein involved in several physiological
and pathological processes including development, wound healing, inflammation,
haematopoiesis, immune response and tumour progression (Herrlich et al., 1998; Naor et al.,
1997; Ponta et al, 2003). Posttranslational modification and excessive alternative splicing
give rise to a diverse pool of proteins ranging between 80 and 200 kDa in size.
1.3.1 Structural properties of CD44
The CD44 gene locus spans about 50 kb of genomic DNA and is highly conserved
among vertebrates (Naor et al., 1997). The corresponding pre mRNA consists of 20 exons, 12
of which can be regulated by alternative splicing (Gunthert et al., 1991; Screaton et al., 1992;
Tolg et al., 1993; reviewed by Naor et al., 1997; Lesley et al., 1998). Theoretically, about
1000 putative different splice products can be generated in this way, but apparently not all
combinations are expressed (Naor et al., 1997). The shortest CD44 isoform, called standard
isoform (CD44s), with all variant exons excised, is expressed in nearly all vertebrate cells
(Naor et al., 1997), while variant isoforms, named after the variant exons contained, show a
highly restricted expression pattern during embryonic development, in pathological
backgrounds or in a cell type specific manner. Alternative splicing can be dynamically
regulated depending on the activation state of the cell (Arch et al., 1992). Examples for cell
type specific variants are CD44v8-v10, which is present on epithelial cells, while CD44v3v10 is expressed by keratinocytes (Fig. 3).
Fig. 3: Exon map of CD44 and examples for variant isoforms
A. Exon map of CD44 B. Examples for variant isoforms, named by the variant exons they contain.
EC extracellular domain, TM transmembrane domain, CP cytoplasmic domain.
The CD44 protein consists of a large extracellular domain, made up of the aminoterminal part and a short membrane proximal stem structure, the transmembrane region and a
cytoplasmic tail. Up to ten variant exon products can be inserted in the stem structure. In
addition to this, a short version carrying a short cytoplasmic tail exists, which is only very
rarely expressed (Goldstein and Butcher, 1990) (Fig. 4).
The extracellular domain of CD44 can be modified by N- and O-linked glycosylation
and contains binding sites for hyaluronic acid (HA) and other glycosaminoglycans (GAGs)
(Naor et al., 1997). Insertion of variant exons can lead to additional modifications. For
example, variant exon 3 (v3) carries a site for heparan sulfate (HS) or chondroitin sulfate
(CS) modifications (Bennett et al., 1995). Through its GAG binding sites CD44 can bind to
GAG modified proteolycans such as versican (Kawashima et al., 2000), aggrecan (Fujimoto
et al., 2001) and serglycin (Toyama-Sorimachi et al., 1995). However, the functional
relevance of this property is unclear. The extracellular domain of CD44 can be shed from the
surface by proteolytic cleavage within the stem structure (Okamoto et al., 1999).
Fig. 4: Structure of the CD44 molecule
CD44 is a single pass transmembrane protein. The large extracellular domain carries binding sites for
glycosaminoglycans, such as hyaluronic acid and sites for posttranslational modifications. 10 variable
exons can be inserted into the stalk domain by alternative splicing. The short cytoplasmic tail can be
linked to the cytoskeleton via ERM proteins and Ankyrin.
The transmembrane domain, encoded by exon 18 is supposed to be involved in CD44
oligomerization and localization in raft like membrane microdomains (Liu and Sy, 1997;
Neame et al., 1995; Perschl et al., 1995), which are known to serve as signalling platforms.
The cytoplasmic domain was shown to interact with a multitude of molecules and is
important for linking CD44 to the cytoskeleton. The cytoplasmic tail of CD44 can be
phosphorylated by protein kinase C (PKC), which influences its ability to interact with other
proteins, as shown for ezrin (Legg et al., 2002). Upon shedding of the extracellular domain,
the cytoplasmic part is cleaved and translocates to the nucleus where an effect on
transcriptional regulation was demonstrated (Okamoto et al., 2001; reviewed by Nagano et
al., 2004).
1.3.2 Different modes of interactions for CD44
A multitude of functions has been attributed to CD44 and its variants in different
functional contexts, such as morphogenesis and organogenesis (reviewed by Knudson and
Knudson, 1993), haematopoiesis (Ghaffari et al., 1999) and various immune functions,
including homing and migration of lymphocytes and leukocyte activation and effector
functions (reviewed in Naor et al., 1997; Pure and Cuff, 2001). The diversity of cellular
processes influenced by CD44, e.g. growth regulation, survival, differentiation, adhesion and
motility, raises the question how this can be achieved by a single molecule. Of course, one
reason lies in the heterogeneity of the CD44 protein family with different variants each
having unique characteristics. A second explanation might be the ability of CD44 to function
in different ways. First as a ligand binding surface receptor, mainly by interacting with its
principle ligand hyaluronic acid, second as a co-receptor for other surface molecules,
modulating for example the signalling of associated growth factor receptors, and third
through interactions with cytoplasmic molecules and as organizer of the actin cytoskeleton.
CD44 as a cell surface receptor
CD44 functions as the main hyaluronic acid receptor (Culty et al., 1990) but binding
was also demonstrated for other components of the ECM, namely fibronectin, laminins and
collagens (Turley and Moore, 1984) and also for cytokines like osteopontin and RANTES
(Weber et al., 1996; Wolff et al., 1999). The corresponding binding site for HA is located in
the standard part of the protein, but the affinity might be influenced by insertion of variant
exons (Sleeman et al., 1996b) or by the state of glycosylation (Skelton et al., 1998).
Moreover, binding to HA is not a constitutive ability of CD44 expressing cells but can be
regulated from within the cell.
Hyaluronic acid belongs to the family of glycosaminoglycans, but different from all
other GAGs, HA does not possess a protein component. Instead it displays a simple structure
as a large polysaccharide, exclusively composed of repeating disaccharides of glucoronic
acid and N-acetylglucosamine. Under physiological conditions HA consists of 2000 to 2500
disaccharides, corresponding to a molecular mass of 106-107 Da and a polymer length of 225µm. HA is a major component of the extracellular and pericellular matrix (Lee and Spicer,
2000) and due to its hygroscopic characteristic it plays an important role in tissue
homeostasis and biomechanical integrity. The role of CD44 in many physiological and
pathological processes is based on its interaction with HA. By binding to its ligand, CD44
can mediate adhesion to and migration on HA rich matrices and these processes can be
regulated by modulation of the binding affinity or by enzymatic cleavage of the extracellular
portion of CD44 (Okamoto et al., 1999-2). Moreover, CD44 expression can influence the
synthesis and endocytosis of HA (Culty et al., 1992; Hua et al., 1993).
CD44 was shown to recruit and regulate the activity of proteases on the cell surface.
For example, MMP9 (Bourguignon et al., 1998) and MMP7 (Yu et al., 2002) were
demonstrated to associate with CD44 and their function was dependent on this colocalization. In addition, growth factors and cytokines are captured by CD44, which can be
variant specific as it is discussed for osteopontin (Katagiri et al., 1999). Growth factors
described to bind to CD44 are ‘hepatocyte growth factor’/‘scatter factor’ (HGF/SF), ‘basic
fibroblast growth factor’ (bFGF) and ‘heparin-binding factor’ (Sherman et al., 1998; van der
Voort et al., 1999; Jones et al., 2000). The enrichment of soluble molecules on the surface by
CD44 can influence outside in signalling either directly via binding to CD44 or indirectly by
regulating the binding to other receptors.
CD44 as a ‘co-receptor’ for other transmembrane proteins
Lacking a catalytic domain itself, CD44 can act as a co-receptor for protein tyrosine
kinases (PTKs) like growth factor receptors. For example, CD44v6 has been demonstrated in
several cell lines to be essential for proper binding of HGF to its cognate receptor cMet/HGF-R through complex formation (Orian-Rousseau et al., 2002). A second prominent
example is the ErbB family of receptor tyrosine kinases, with some members showing
dependence on complex formation with CD44 for proper activation (Bourguignon et al.,
1997; Sherman et al., 2000). The nature of this co-receptor function might be due to
clustering of receptor subunits and stabilisation of receptor dimers or, as in the case of ErbB4,
by supporting activation of the ligand via proteolytic cleavage through associated MMPs as
mentioned above (Yu et al., 2002). Given the heterogeneity of the CD44 protein family,
functioning as a co-receptor could explain the ability of CD44 to modulate several different
signalling circuits, without any direct signal transfer through CD44 itself ever being
demonstrated. CD44 also associates with other transmembrane proteins without catalytic
activity, for instance tetraspanins and other adhesion molecules such as integrins or EpCAM
(Schmidt et al., 2004; Ladwein et al., 2005).
CD44 as an associating molecule with cytoplasmic proteins and as organizer of the actin
Several intracellular molecules were shown to associate with the cytoplasmic tail of
CD44. Importantly, CD44 can be crosslinked to the actin cytoskeleton via binding to ankyrin
and members of the ERM proteins (for ezrin, radixin and moesin) (Tsukita et al., 1994;
Bourguignon and Jin 1995). ERM proteins are involved in the regulation of cell shape, cell
migration and protein resorting in the plasma membrane (Bretscher et al., 2002; Gautreau et
al., 2002) and CD44 was demonstrated to influence these processes through interaction with
ERM proteins. Even though, the precise mechanism, e.g. if this leads to actin contraction,
polymerization or depolymerization is not clear. The binding affinity for ERM proteins is
tightly regulated and seems to be higher for variant CD44 than for the standard isoform
(Tsukita et al, 1994). Other associated molecules include cytoplasmic kinases like Src, PKC,
LCK and Fyn and the guanine nucleotide exchange factors TIAM1 and VAV2 (reviewed by
Naor et al., 1997; Bourguignon et al., 2000; Bourguignon et al., 2001a+b).
1.3.3 Physiological and pathological functions ascribed to CD44
As mentioned above, the physiological roles of CD44 are surprisingly diverse. The
fact that CD44 knock-out (k.o.) mice are viable and show only a very mild phenotype with
regard to haematopoiesis as well as lymphocyte activation and migration (Schmits et al.,
1997; Protin et al., 1999) argues for other molecules being able to compensate for the loss of
CD44 during embryonic development. However, antibody blockade led to a retardation of
development (Zoller et al., 1997). This corresponds to the observation that the k.o. phenotype
became more obvious, when CD44-/- animals were challenged, for example by infection with
pathogens or by artificial induction of autoimmune diseases and proved a role for CD44 in
the immune system. The same holds true in variant specific k.o. mice (Wittig et al., 2000).
CD44 in development
Expression of CD44 during embryonic development has been investigated in several
studies. However a defined role for CD44 was demonstrated only in a few cases, for example
in axon guidance during the formation of the optic chiasm (Stretavan et al., 1994, 1995),
during limb bud development (Sherman et al., 1998), and in uteric bud and mammary gland
development (Pohl et al., 2000). Expression of variant CD44 during embryogenesis is seen
on most epithelial and haematopoietic cells (Wirth et al., 1993; Terpe et al, 1994; Weber et
al., 1996), while the expression in adult animals is mainly restricted to the skin, the
epithelium of the gut, some glands and subpopulations of the haematopoietic cells (Kennel et
al., 1993; Wirth et al., 1993; Fox et al., 1994; Hirano et al., 1994). Several studies
demonstrated an important function of CD44 for differentiation and proliferation of
haematopoietic progenitor cells (Ghaffari et al., 1999).
CD44 in inflammation and leukocyte extravasation
Physiological functions of CD44 are best explored in haematopoietic cells and
important roles for CD44 and HA were shown in inflammatory processes (Pure et al., 2001).
The physiological importance of CD44 in this context was shown for example in autoimmune
diseases like rheumatoid arthritis and many others (Brennan et al., 1997; Seiter et al., 1999;
Stoop et al., 2001). During inflammation, leukocytes exit the circulation to enter into
different tissues. The first step of this process, a loose attachment of the leukocytes to the
vessel wall, termed ‘rolling’, is mediated by selectins on leukocytes binding to carbohydrate
ligands on the endothelium. This primary interaction is followed by a secondary ‘firm
adhesion’, mediated by integrins and subsequent extravasation (Albelda et al., 1994). There is
increasing evidence that CD44 contributes to both, rolling and firm adhesion. CD44 on the
surface of leukocytes was shown to mediate rolling on the vessel wall via interaction with
endothelial HA (De Grendele et al., 1996/1997).
CD44 in pericellular matrix assembly
HA is a key player in the assembly of pericellular matrices through interactions with
proteoglycans and other extracellular macromolecules. CD44 was shown to be the main HA
anchoring molecule for these processes in chondrocytes, the main cartilage cell type
(Knudson et al. 1996). Thereby, the presence of CD44 has implications for organizing the
structure of the cartilage. Other cell types like fibroblasts are also able to form such a
pericellular matrix (Hedman et al., 1979), which may be important for their locomotion
(Turley et al., 1989). Although tumour cells usually do not synthesize their own pericellular
matrix, they often have the ability to assemble one in the presence of exogenously added HA
and aggregating proteoglycans (Knudson and Knudson, 1991).
1.3.4 CD44 in tumour progression
A role of CD44 for tumour progression is well documented (reviewed by Ponta et al.,
1998; Naor et al., 2002; Marhaba and Zoller, 2004). The observation that variant CD44 is
overexpressed in several metastatic tumours (Matsumura et al., 1992) raised the question
whether CD44 might be predictive for metastasis formation. Several studies tried to define
CD44 as prognostic marker, but correlations vary between different kinds of cancer, states of
disease and the CD44 isoform being examined (Wielenga et al., 1993; Pals et al., 1997; Dall
et al., 1994). However numerous studies have demonstrated that CD44s and especially
variant CD44 is implicated in different aspects of tumour progression and particularly in
metastatic spread. A direct evidence for this was demonstrated in a CD44 k.o. mouse model,
showing unaltered primary tumour growth but inhibited sarcoma metastasis formation
(Weber et al., 2002). Recently, CD44 could be identified as a key regulator for leukemic stem
cell fate by blocking homing to the bone marrow (Jin et al., 2006). However, it has to be
mentioned, that in prostate cancer cells overexpression of CD44s suppressed their metastatic
capacity (Gao et al., 1998).
Deregulation of variant CD44 in several cancers is not surprising, taken into account,
that aberrant alternative splicing is frequently seen in cancer cells (reviewed by Kalnina et al.,
2005) and CD44 being most profoundly subjected to alternative splicing. This and the
remarkable functional diversity of CD44 provide an explanation as to how one molecule can
fulfil several different tasks of tumour progression. In fact, CD44 is described to influence
most of the proposed acquired capabilities needed for successful tumour progression.
The interaction between CD44 and HA has been described to trigger proliferation of
several tumour cell lines, such as melanoma cells (Ahrens et al., 2001), mammary carcinoma
(Peterson et al., 2000), glioma (Akiyama et al., 2001) and malignant mesothelioma cells
(Nasreen et al., 2002). On a molecular level CD44 was also demonstrated to promote tumour
cell proliferation through its co-receptor function as mentioned above, by activating members
of the ErbB receptor family (Bourguignon et al, 2001, Ghatak et al., 2005) and c-Met (OrianRousseau et al., 2002, Recio et al., 2003). In addition CD44v6 is able to directly induce
proliferation by activating the MAP kinase pathway (Marhaba et al., 2005). In contrast, CD44
was also found to act as a tumour suppressor. Binding of CD44 to HA has been reported to
inhibit cell growth during contact inhibition (Morrison et al., 2001) and induce terminal
differentiation of myeloid leukemia cell lines (Charrad et al., 2002).
The influence of CD44 on apoptosis is even more controversial than on proliferation.
In lymphoma and thymocytes CD44 engagement was shown to induce apoptosis through upregulation of the proapoptotic molecule Bax and down-regulation of the antiapoptotic
molecule Bcl-XL (Guy et al., 2002). Proapoptotic effects of CD44 were also shown for
dendritic cells and neutrophils (Yang et al., 2002; Takazoe et al., 2000). On the other hand,
there are several examples for CD44 induced survival by suppressing the induction of
apoptosis (Bates et al., 1998; Allouche et al., 2000). Survival-promoting functions of CD44
can be mediated by the PI3K-Akt pathway (Bates et al., 2001; Ghatak et al., 2002), or
through upregulation of antiapoptotic molecules like Bcl-2 and Bcl-XL (Khan et al., 2002;
Marhaba et al., 2003). Moreover CD44 can also downregulate Fas expression, thereby
inhibiting apoptosis in lung cancer cells (Yasuda et al., 2001), or inhibit Fas signalling
through interaction with the receptor (Mielgo et al., 2006). Notably, these prosurvival effects
are often attributed to variant CD44, like CD44v6 in colon carcinoma cells (Bates et al.,
1998) or CD44v7 in lymphocytes (Wittig et al., 2000; Marhaba et al., 2003).
Factors known to stimulate vascularization are ‘basic fibroblast growth factor’
(bFGF), ‘vascular endothelial growth factor’ (VEGF) and cytokines such as ‘transforming
growth factor β (TGFβ) (Pepper, 1997). MMP2 and MMP9 associated with CD44 were
demonstrated to cleave the pro-form of TGFβ, thereby releasing the active cytokine and
inducing angiogenesis (Yu and Stamenkovic, 1999, 2000). Other described angiogenic
effects of CD44 are mainly due to its expression on endothelial cells rather than on tumour
cells (Griffioen et al., 1997).
Finally, CD44 might influence the formation of metastatic lesions in several different
ways. CD44-HA binding can influence the adhesive properties of tumour cells to the ECM.
Modulation of this interaction can enhance the mobility of cancer cells during the metastatic
process. For example HA-CD44 binding was shown to be important for glioma cell invasion
and migration (Okada et al., 1996). Additionally, the loss of ECM contacts by cleavage of
CD44 has been demonstrated (Okamoto et al., 1999). The loss of cell-matrix adhesion
usually leads to growth arrest and cell death and CD44 was shown to be able to promote
anchorage-independent growth, a prerequisite for invasion (Peterson et al., 2000; Ghatak et
al., 2002).
The activation of cell surface MMPs by CD44 favours invasiveness of tumour cells,
enabling them to migrate into adjacent tissues as demonstrated for MMP9 (Yu and
Stamenkovic 1999, 2000). Migration of cells requires reorganization of the actin
cytoskeleton, which is under the control of Rho GTPases, like RhoA, Rac1 and Cdc42. HA
binding to CD44 can activate Rac1 and induce lamellipodia formation and migration
enabling invasive behaviour (Oliferenko et al., 2000; Bourguignon et al., 2000).
As already described in the context of inflammation, CD44v was shown to facilitate
attachment of lymphoma cells to vessel walls through interaction with endothelial surface
HA (Wallach-Dayan et al., 2001), facilitating subsequent firm adhesion and transmigration
into the underlying tissue and favouring settlement of tumour cells at a secondary site.
Indeed, a soluble form of CD44 competing with cell surface CD44 for HA binding reduced
metastasis formation in vivo (Yu et al., 1997; Peterson et al., 2000).
In summary, published data clearly show that CD44 and especially variant CD44 play
a crucial role in the course of tumour progression and metastasis formation.
RNA interference as a tool to study isoform specific gene
RNA interference (RNAi) depicts an evolutionary conserved mechanism, occurring in
most eukaryotic organisms (Hannon, 2002), which may have evolved as defense system
against viral and genetic parasites. Double stranded (ds) RNA molecules are processed into
short RNA duplexes, directing sequence specific cleavage or translational repression of
complementary messenger RNAs (reviewed by Meister and Tuschl, 2004) and are implicated
in chromatin remodeling and transcriptional regulation (reviewed by Lippman and
Martienssen, 2004). In addition to the growing importance of RNAi as a tool for
manipulating gene expression, the endogenous triggers of the RNAi pathway, called
microRNAs (miRNAs) recently attracted great interest and were shown to play a role during
embryonic development and to be implicated in tumourigenesis as well (reviewed by
O’Rourke et al., 2006; Esquela-Kerscher and Slack, 2006).
Upon the original observation that ds RNA triggers are far more efficient in gene
silencing than single stranded (ss) antisense RNA molecules (Fire et al., 1998), RNAi became
a popular tool for modulating gene expression in model systems like C.elegans and
Drosophila. However the subsequent adaptation of the RNAi technique to the mammalian
system (Elbashir et al., 2001) made RNAi one of the most profound discoveries of the last
decade and one of the most powerful techniques for studying gene function in a cell culture
based system and probably even as therapeutic tool in vivo. Compared to other techniques,
such as antisense technology or morpholino oligos, only RNAi enables splice variant specific
down regulation of gene products even on a stable basis. Isoform specificity can be achieved
by choosing exon specific target sequences and stable downregulation of the gene of interest
can be realized through integration of expression cassettes, driving transcription of self
complementary short hairpinRNAs (shRNAs), that are processed by the endogenous RNAi
machinery into active small interfering RNAs (siRNAs) (Paddison et al., 2002;
Brummelkamp et al., 2002). Thus, RNAi allows stable and isoform-specific loss-of-function
studies, which were rarely practical by conventional k.o., due to the hyperploidity of many
tumour cells.
Aims of the thesis
The aim of this work was to investigate the contribution of CD44v on the metastatic
behaviour of ASML cells and to gain a better understanding of the role of variant CD44
during the lymphatic spread of tumour cells in general.
Due to its remarkable structural and functional heterogeneity, CD44 plays a pivotal
role in several steps of tumour progression. First evidence for the importance of CD44
variants in metastasis formation was obtained in the rat pancreatic adenocarcinoma model
BSp73, where two major splice variants expressed only by the highly metastatic ASML
subline were identified as v4-v7 and v6/v7. Introduction of these isoforms conferred a
metastatic phenotype to the otherwise only locally growing AS subline, exclusively
expressing the standard CD44 isoform (Gunthert et al., 1991; Rudy et al., 1993). Moreover,
application of antibodies targeting variant exon 6 retarded metastasis formation of ASML
cells, providing a hint, that CD44v also plays a crucial role in the metastatic spread of these
cells (Seiter et al., 1993). On the other hand, compared to ASML cells, AS cells display
entirely different adhesion and migration characteristics. In addition to this, expression
profiling revealed several hundred differentially expressed genes between these two sublines
(Tarbe et al., 2002), including molecules known to play a role in cancer progression, such as
c-Met and c-Myc.
In order to investigate specific functions of CD44v and the underlying molecular
mechanisms in metastasis formation, a plasmid based RNAi system was used for creating
stable and variant specific CD44 knock-down cells of the highly metastatic carcinoma cell
line BSp73ASML. The resulting phenotype was studied in vitro and in vivo, particularly
emphasising interactions of CD44v with the microenvironment and the stimuli the tumour
cell may receive by this crosstalk.
A plasmid-based RNAi system (pSuper) was used to create stable and variant
specific knock-down cells of the highly metastatic tumour cell line BSp73ASML
(ASML), known to express CD44 variants at high levels. The plasmid drives expression
of small hairpin RNAs, which are processed into functional siRNAs and eventually lead
to sequence-specific down-regulation of the gene of interest. In addition, it carries a
GFP reporter and a neomycin resistance for detection of transfected cells and selection
of stable clones (Fig. 5). Constructs were designed on the premise to target the two
most abundant CD44 variants expressed by ASML, namely CD44v4-v7 (meta1 isoform)
and CD44v6/v7 (meta2 isoform), previously described to confer metastatic potential to
the otherwise weakly metastatic sister subline BSp73AS (Rudy et al., 1993). Target
sequences therefore lie within the variant exons contained in both isoforms. Construct
‘v6’ targets variant exon 6, ‘v7’ targets variant exon 7 and ‘v6/v7’ targets the border of
exon v6 and v7. Constructs were cloned and verified by sequencing.
Fig. 5: Expression cassettes of ‘pSuperGFPneo’ carrying construct ‘v7’
The self-complementary sequence of the transcript folds into a hairpin structure, which is processed
by the endogenous RNAi machinery into an active siRNA. A GFP-reporter and a neomycin resistance
are included in a second expression cassette.
Establishment of stable CD44vk.d. cell lines and rescue clones
2.1.1 RNAi construct evaluation by FACS and fluorescence microscopy
Efficiency in down-regulation of variant CD44 was monitored by FACS staining after
transient transfection of ASML cells with three different pSuper-constructs and the empty
vector (mock). Three days after transfection, cells were analyzed for CD44v expression by
FACS staining using an antibody specific to variant exon 6 (A2.6). The mean intensity of
GFP positive (transfected) cells was compared to GFP negative cells of the same pool and
revealed functionality of the ‘v7’ and the ‘v6/v7’ constructs in down-regulating CD44variant
expression on ASML cells. Table 1 shows the relative mean values of three experiments.
While ‘mock’ transfection did not change CD44v expression and pSuper-v6 showed only
weak down-regulation, transfection with pSuper-v7 reduced the mean intensity of CD44v by
55% and transfection with pSuper-v6/v7 resulted in a mean intensity reduced by 51%.
Tab. 1: Evaluation of pSuper constructs by FACS staining
relative mean intensity for CD44v
108 ± 6%
86 ± 5%
45 ± 4%
49 ± 6%
FACS staining for CD44v6 (A2.6) after transient transfection with different pSuper constructs and the
empty vector. Mean intensities of transfected (GFP positive) relative to untransfected cells are shown
in percent.
Results were confirmed by immunofluorescence microscopy. ASML cells grown on
coverslips were transfected with the three pSuper constructs and the empty vector and stained
after three days for variant CD44 expression. GFP positive cells in pSuper-v7 and pSuperv6/v7 transfections show reduced CD44 expression, while expression of other surface
molecules like EpCAM is not affected (shown for pSuper-v7 in Fig. 6A). PSuper-v6 or mock
transfected cells did not show a difference in CD44 expression (Fig. 6A, pSuper-v6 is not
2.1.2 Establishment of stable CD44vk.d. clones by selection and recloning
By selection with G418 and two rounds of recloning three stable clones of pSuper-v7
transfected ASML cells were established that revealed strong down-regulation of variant
CD44 on the protein level as shown by western blotting (Fig. 6B). CD44v levels are hardly
detectable in the CD44v ‘knock-down’ (k.d.) clones, but unaffected in a stable ‘mock’ clone.
EpCAM was used as an internal loading control. For unknown reasons no stable clones could
be obtained with the pSuper-v6/7 construct.
Fig. 6: CD44v expression of ASML cells is reduced after transfection with pSuper-v7 or
the empty vector
A. Immunofluorescence staining for CD44v6 with A2.6 (upper panels) or EpCAM with D5.7 (lower
panel). GFP positive cells display reduced CD44v expression only in the pSuper-v7 transfections
(middle panel), while mock transfection has no influence on the expression level (upper panel).
EpCAM staining is not affected (lower panel). B. Western blot analysis of stable clones. Upper panel:
the two major CD44 variants are down-regulated in the three knock-down clones, but unaffected in
the mock clone. EpCAM is used as an internal loading control (lower panel).
2.1.3 Restoring CD44 expression by introduction of mutated cDNAs
In order to control specificity of any phenotype arising in the k.d cells, ‘rescue’ clones
with restored expression of one of the dominant CD44 variants were established. This was
achieved by introduction of cDNAs coding for either CD44v4-v7 or CD44v6/v7, carrying
four silent point mutations in the v7 target sequence, which should protect them from
degradation (Fig.7A). Indeed co-transfecting ‘HEK 293T’ cells with the ‘rescue’ cDNAs
together with the pSuper-v7 construct showed high CD44 expression, while expression of a
wt CD44 cDNA was significantly affected by pSuper-v7 co-transfection (data not shown).
One of the k.d. clones, ASMLv71-14, was used for transfections with the ‘rescue’ cDNAs
and after selection and two rounds of recloning stable ‘rescue’ clones were established.
Western blot analysis for CD44variant expression shows successful restoration of either
CD44v4-v7 or CD44v6/v7 in the knock-down clone, although CD44v4-v7 expression was
not restored to levels comparable to the wt situation. One mock clone is shown as control.
EpCAM serves as a loading control (Fig. 7B).
v7 target sequence
v7 rescue sequence
Fig. 7: Restoring CD44 expression by transfection with mutated cDNAs
Four silent mutations were introduced in the v7 target site of CD44v4-v7 and CD44v6/v7 cDNAs by
PCR. The v71-14 clone was transfected with the mutated cDNAs and stable clones were established.
A. v7 target and rescue sequence. Silent mutations are highlighted in red. B. Western blot analysis of
stable clones, showing restored CD44v expression by A2.6 staining (upper panel). EpCAM is used as
an internal loading control by D5.7 staining (lower panel).
Characterization of the knock-down cell lines in vivo
2.2.1 CD44vk.d. cells exhibit a reduced metastatic capacity in vivo
Intra-footpad injections (ifp) of BDX rats with ASMLwt, one mock and three k.d.
clones were performed and the metastatic spread was monitored. The metastatic growth of
ASML cells has been studied before (Matzku et al., 1983) and displays only little local
tumour growth, but rapid spread through the whole lymphatic system with massive tumour
burden in the proximal lymph nodes and miliar outgrowth of micro metastases in the lung,
which eventually lead to the death of the animal. Animals were injected ifp with 106 cells and
sacrificed after 50 days. Diameters of the primary tumour and lymph node metastases were
measured and lungs were photographed and weighed. In addition, lung samples were
analyzed by immunohistological staining. ASMLwt cells as well as the mock clone
resembled the expected metastatic behavior, but the CD44vk.d. clones showed clearly less
metastatic growth. Two k.d. clones (v71-14 and v72-1-17) displayed a greatly reduced over
all tumour burden in vivo, while one clone (v71-16) grew as fast as the wt and the mock
controls in the lymph nodes, while the metastatic settlement in the lung was clearly
diminished in all three clones, with most animals revealing tumour free lungs and some only
few metastatic nodules. In comparison, the wt- and mock-injected animals all displayed
entirely metastatic lungs.
The metastatic burden of each lung was examined macroscopically (Fig. 8). The
immense tumour burden in the lung of wt and mock treated animals is obvious by the size of
the lungs that do not collapse, but are completely filled with tumour cells. Compared to this,
the lungs of the CD44vk.d. clones appear normal or in the case of an v71-16-injected animal
only moderately enlarged. The weight of the lungs gives a good indication for tumour
burden (Tab. 2). Two animals injected with ASMLwt died before the end of the experiment,
obviously due to the tumour burden of the lung. In addition, immunohistochemical analyses
of sectioned lung samples were performed. C4.4A was used as a tumour marker in this case
and again demonstrates the immense tumour burden in the lungs of wt and mock injected
rats, while lungs of the k.d.-treated animals were tumour free or exhibited only few nodules
as shown for a v71-16 lung. In this case also the stability of the RNAi effect throughout the in
vivo experiment is demonstrated, as the v71-16 lung was unstained for CD44v6, while wt and
mock tumour tissue is strongly stained (Fig. 8).
Tab. 2: CD44k.d. clones exhibit a reduced metastatic growth in vivo
injected construct(animal#)
untreated control-(1-1)
untreated control-(1-2)
ASMLwt-(1-1)* †day38
ASMLwt-(1-2)* †day45
tumour burden
prim. tumour + lymph nodes
tumour mass in
in g
6.9±2.1 miliary, confluent
miliary, confluent
miliary, confluent
miliary, confluent
miliary, confluent
6.9±2.6 miliary, confluent
miliary, confluent
miliary, confluent
miliary, confluent
miliary, confluent
miliary, confluent
3.5±1.0 none
2.3±1.4 miliary, few
6.8±2.5 miliary, few
miliary, few
miliary, multiple
In vivo metastasis assay #1. Ifp injections of BDX rats with ASMLwt, mock and CD44vk.d. clones.
Animals were sacrificed and dissected after 50 days. The tumour burden is given as sum of primary
tumour and lymph nodes. The mean for each group is shown. Lungs were macroscopically examined
and the weight is given as an indication for tumour burden of the lung. Two ASMLwt animals marked
with asterisks (*) died within the course of the experiment.
The experiment was repeated with the rescue clones, displaying restored CD44v
expression. One mock clone, one knock-down clone (v7 1-14) and one clone of each rescue
construct were injected as before and animals were killed after 60 days and analyzed as
described above. Again the mock-treated animals displayed massive tumour burden with
miliary lungs, while the k.d. clone showed less metastasis formation without any tumour
growth in the lungs. For unknown reasons the v4-v7 rescue clone did not grow in vivo, but
four out of six animals injected with the CD44v6/v7 rescue clone displayed enhanced
metastatic growth in the lymph nodes and five of six animals revealed also settlement in the
lung, without reaching the massive tumour burden of the wt (Tab. 3). Immunohistology on
lung samples were carried out as before and confirmed the macroscopical examination (Fig.
Fig. 8: CD44vk.d. cells exhibit a reduced settlement in the lung
Ifp injections of BDX rats. Animals were killed, dissected and lungs were analyzed. Macroscopic
photographs of whole lungs and imunohistological analysis of sectioned lung samples, stained for
CD44v6 (A2.6) and C4.4A (C4.4) as a tumour marker are shown. A. In vivo metastasis assay #1,
showing ASMLwt, mock and CD44vk.d. clones. B. In vivo metastasis assay #2, showing one rescue
clone (v71-14rescv6/7).
The stability of the k.d. and the restored CD44v expression was reconfirmed by recultivation of tumour cells from the lungs or lymph nodes of the injected clones. After lysis,
SDS-PAGE and western blotting, blots were stained for CD44v6 (A2.6) and EpCAM (D5.7)
as loading control. This assay clearly demonstrates the stable down-regulation of variant
CD44 in the established clones and the restored CD44v6/v7 expression in the rescue clone
(Fig. 9).
Tab. 3: Restored CD44variant expression is able to rescue the metastatic capacity of
ASML in part
injected construct-(animal)
tumour burden
prim. tumour + lymph nodes
tumour mass in
in g
7.0±1.9 miliary, confluent
miliary, confluent
miliary, confluent
miliary, confluent
miliary, confluent
miliary, confluent
3.3±1.0 none
6.2±4.0 miliary, confluent
miliary, multiple
miliary, few
miliary, few
miliary, few
In vivo metastasis assay #2. Ifp injections of BDX rats with mock, one CD44vk.d. clone (v71-14) and
two v71-14 rescue clones. Animals were killed and dissected after 60 days and tumour burden is
given as sum of primary tumour and lymph nodes. The mean value for each group is shown. Lungs
were macroscopically examined and the weight is given as an indication for tumour burden of the
lung. The second rescue clone (v71-14rescv4-v7#8) did not grow in vivo and is not included in the
Fig. 9: Knocked down and restored expression of CD44v remained stable during the
animal experiments
Western blot analysis on re-cultivated tumour cells from injected animals stained for CD44v6 (A2.6)
and EpCAM (D5.7) as loading control. A. All tested k.d. tumour samples still show the same level of
down-regulated CD44v. Cultured k.d. (v71-16) and mock cells are shown as control. B. All tested
tumour samples of the rescue clones still show restored CD44v6/v7 expression. LN lymph node.
Characterization of knock-down cell lines in vitro
2.3.1 CD44vk.d. cells show no phenotypic changes
Cells were grown for 24 hours in 6-well plates. Microscopic analysis revealed no changes in
cell shape or growth behavior. All clones display the same rounded cell shape without visible
spreading. ASMLwt and CD44vk.d. cells do not form a dense monolayer with tight cell-cell
contacts, but rather detach from the ground before reaching complete confluency (Fig. 10).
The same accounts for the rescue clones. One rescue clone is shown as representative
Fig. 10: The phenotype of CD44vk.d. clones is not changed
Microscopic analysis of wt ASML, mock, CD44vk.d. clones and one rescue clone, cultured for 24
2.3.2 CD44vk.d. cells show no altered growth behaviour
As the k.d. clones revealed different growth in vivo, the proliferative capacity was compared
in vitro. Proliferation of wt and knock-down cells was monitored for three days, using either
H thymidine incorporation (data not shown) or staining with crystal violet. Both assays
revealed no significant changes in proliferation rates between the wt and knock-down cells,
irrespective of cells being grown in the presence of 10% FCS or under low serum conditions
(0.5% FCS) (Fig. 11). Moreover, colony formation in soft agar, reflecting the ability of
anchorage-independent growth, a hallmark of metastatic cells, did not show any significant
changes. All clones revealed a high colony-forming efficacy of 90-95% (Tab. 4).
Fig. 11: Proliferation of CD44vk.d. clones is not altered
Proliferation was monitored for 48h and quantified by crystal violet staining. A. Proliferation in
medium supplemented with 10% FCS. B. Proliferation in medium supplemented with 0.5% FCS.
Tab. 4: Soft agar colony formation of wt ASML, mock and CD44vk.d. clones
colony forming efficacy
mock #9
v7 1-14
v7 1-16
v7 2-1-17
91% +/- 4%
93% +/- 3%
90% +/- 5%
95% +/- 4%
89% +/- 6%
100 and 1000 cells of ASMLwt, mock and CD44vk.d. clones were seeded in RPMI/10%FCS
containing 0.5% agar. Colonies were counted after 4 weeks. The mean colony-forming efficacy is
2.3.3 CD44vk.d. cells do not differ in MMP2 and MMP9 expression
In order to invade adjacent tissues, tumour cells have to degrade extracellular matrix barriers,
which is often achieved by up-regulation of degrading enzymes like MMPs. To test if downregulation of CD44 variants is accompanied by decreased MMP expression, MMP2 and
MMP9 secretion into the supernatant was tested by a Gelatinease assay. Ten times
concentrated conditioned cell culture supernatant of wt, mock and k.d. cells was subjected to
SDS-PAGE, containing Gelatine as a substrate. After completion of the run gels were stained
with Coomassie (Fig. 12). Enzymatic activity was visualized as unstained bands that reflect
the pro-forms of MMP2 and MMP9. No significant differences were observed between the
Fig. 12: MMP2 and MMP9 expression of wt ASML, mock and CD44vk.d. clones
Gelatine zymography of concentrated cell culture supernatants. Unstained bands correspond to ProMMP2 and Pro-MMP9 expression.
2.3.4 ASMLwt cells but not CD44vk.d. cells aggregate in stromal cell culture
As lymph nodal spread is significantly reduced in CD44vk.d. cells, it was tested, if ASML
cells interact with lymph node stromal cells and if this interaction is impaired in the k.d. cells.
Adhesion to monolayers of the lymph node stromal cell lines ‘ST-A4’ and ‘ST-B12’ or an
immortalized lung fibroblast cell line did not reveal any affinity of ASML (data not shown).
Instead it was noticed that ASML cells clumped and formed cell agglomerates when added to
the stromal cells. This observation was tested subsequently with conditioned cell culture
supernatant of the three mentioned stromal cell lines. Only wt and mock cells formed huge
agglomerates in the supernatant within minutes, while CD44vk.d. cells did not clump at all or
formed only very small cell clusters. The ability to form agglomerates was completely
reestablished by the restored CD44v expression in the rescue clones (Fig. 13).
Fig. 13: Agglomerate formation in conditioned cell culture supernatant of stromal cells
Microscopic analysis of agglomeration of ASMLwt, mock, k.d. and rescue clones in conditioned cell
culture supernatant of immortalized lung fibroblasts (A) and the lymph node stromal cell line ST-A4
(B). ASMLwt cells in RPMI medium are shown as control.
As hyaluronic acid (HA) is known to crosslink CD44 on the surface of cells and stromal cells
are known to produce HA in large amounts, it was tested if the cell clumping was due to
CD44 cross-bridging by HA in the supernatant. Supernatant was treated with hyaluronidase
(1mg/ml) for 2 or 5 hours and agglomeration was monitored as before, heat-inactivated
hyaluronidase served as a control. Indeed cell clumping was abrogated by hyaluronidase
treatment in a time dependent manner (Fig. 14B). Likewise, addition of 1mg/ml HA to
normal RPMI medium was able to induce agglomerate formation of ASML cells (Fig. 14A),
demonstrating that the observed clustering is likely due to bridging of cell surface CD44 by
Fig. 14: Influence of hyaluronic acid on agglomeration of ASML cells
A. Addition of HA to normal RPMI medium is able to induce agglomeration of ASMLwt cells. B.
Stromal cell culture supernatant was treated with hyaluronidase for 2 or 5 hours, which abrogates
aggregation. Heat inactivated hyaluronidase was used as a control.
2.3.5 ASMLwt cells produce an adhesive matrix, which is impaired in the
ASML cells show a very limited ability to bind to matrix components, like
fibronectin, laminin or collagen, while cultivated cells strongly attach to the plastic flasks.
This feature is reduced in CD44vk.d. cells as noted by different sensitivities to trypsin
treatment. To test for differences in adhesion, conditioned cell culture supernatant of
ASMLwt cells was coated to plastic and used for adhesion assays. ASMLwt and CD44vk.d.
cells show rapid adhesion to the coated supernatant, indicating that adhesion is not impaired
in the k.d. (data not shown).
The supernatant of the knock-down cells was tested for adhesive properties after
coating. Interestingly, a dramatic reduction was observed. Figure 15A shows reciprocal
crisscross adhesion to matrices of wt cells, a mock clone and the three knock-down clones.
All 5 clones adhere rapidly to the wt and mock derived matrix, while adhesion to the matrices
produced by either of the knock-down clones is strongly impaired for all cells. To test if
adhesion is induced by a soluble or deposited factor, cells were cultured in 24-well plates for
24 hours and detached by either trypsin or EDTA treatment. The plates were used for
adhesion assays as described above. Only EDTA treated plates were able to promote
adhesion (data not shown). Using this approach wt and knock-down cells were compared for
their ability to deposit an adhesion promoting matrix. Indeed, the knock-down matrix was
clearly less capable of promoting adhesion than the wt matrix. However, the difference was
less pronounced than for the conditioned supernatant (Fig. 15B). These results show that the
loss of variant CD44 in ASML cells leads to an impaired ability to produce an adhesive
Fig. 15: Adhesion to coated supernatant and deposited matrix of wt ASML and k.d.
A. Adhesion assay on coated supernatant of wt ASML and k.d. clones. Relative adhesion of each line
to the different matrices is shown. Adhesion of wt ASML to wt matrix was set to 100%. B. Adhesion
of ASMLwt to deposited matrices of the different clones. Deposited matrix was prepared by removal
of cells through EDTA treatment.
In order to investigate, whether this matrix defect is indeed due to the loss of CD44,
the rescue clones displaying restored expression of CD44v4-v7 or CD44v6/v7 were used for
matrix production and compared to the parental knock-down clone. As expected, restoration
of either CD44 isoform was able to restore the matrix production significantly, but without
reproducing the full adhesive properties of the wt matrix (Fig. 16). In summary, loss of
variant CD44 in ASML cells leads to an impaired ability to generate an adhesion-promoting
matrix, which is deposited on the plastic and secreted into the supernatant. Restoring
CD44variant expression is able to restore this matrix production in part.
Fig. 16: Restoring CD44v expression rescues matrix production
Adhesion assay of ASMLwt cells to coated supernatant of ASMLwt, v71-14 and v71-14rescue
clones. Adhesion of ASMLwt cells to wt matrix is set to 100%. Adhesion promoting components are secreted
To verify that the components mediating adhesion are secreted factors, and not present on
membrane particles, like exosomes or membrane fragments, ultracentrifugation (100000g)
was performed, which should precipitate all membrane particles present in the supernatant.
The resuspended pellet and the supernatant were tested for adhesive properties after coating
to plastic. Only the supernatant was able to promote adhesion of ASML cells (Fig. 17A). The
tetraspanin D6.1A was used as exosomal marker (Wubbods et al., 2003) and was completely
removed from the supernatant by centrifugation, while shed CD44 was present in the
supernatant, demonstrating that the separation procedure was successful (Fig. 17B).
Fig. 17: Ultracentrifugation of cell culture supernatant
100000g centrifugation from concentrated supernatant of wt ASML A. Adhesion assay on coated
supernatant and the resuspended pellet. B. Western blot control for shed CD44v6 (A2.6) and for
D6.1A (D6.1) as an exosomal marker. The secreted matrix contains HA, collagen and laminin
As CD44 is known to be involved in the assembly of pericellular matrices through
anchoring hyaluronic acid, the matrix was tested for sensitivity to hyaluronidase treatment as
well as for collagenase treatment. Conditioned supernatant was treated with different
concentrations of hyaluronidase or collagenase either before or after coating to plastic and
adhesion assays were performed as described. Heat-inactivated enzymes were used as
controls (Fig. 18). While hyaluronidase treatment after coating did not alter the adhesive
properties, it did destroy the ability to promote adhesion, when the supernatant was treated
prior to coating (Fig. 18B). This indicates that HA is used as a scaffold for other matrix
components rather than as adhesive substrate and implicates that adhesion itself is not
mediated by CD44, which goes along with the finding that the k.d. clones do not display an
impaired adhesion to the wt matrix. In contrast, collagenase treatment did abrogate the
adhesive properties of the matrix irrespective if the supernatant was treated before or after
coating (Fig. 18A). In summary, these results indicate that HA is needed for the assembly of
other matrix components and that the rapid adhesion of ASML cells is mediated via binding
to collagen.
Fig. 18: Hyaluronidase and collagenase treatment disrupts the adhesive properties of
the supernatant
Supernatant of wt ASML was treated with different concentrations of collagenase after coating (A)
and hyaluronidase before or after coating (B). Adhesion assays were performed with ASMLwt cells
as described earlier.
The adhesive portion of the secreted matrix is between 600 and 4000kDa in size
To further characterize the composition of the secreted matrix, conditioned
supernatant was collected and concentrated 40 times through a ‘vivaspin’ column. The
concentrate was size separated by gelfiltration using different pore sizes. Fractions were
collected and used for coating to 24-well plates. The void volume of the column was defined
by blue dextrane. Adhesion assays were performed and revealed that the fractions promoting
adhesion were still within the void volume, when ‘Superdex200’ beads were used (Fig. 19A),
but were within the separation range of ‘CL6B’ beads (Fig. 19B). This indicates that the
components promoting adhesion are larger in size than roughly 600 kDa, (size exclusion of
Superdex200) but smaller than 4000 kDa (size exclusion of CL6B). The adhesive fractions of
wt cells after chromatography were compared to the corresponding fractions of the knockdown cells using silver staining. However, protein content was low in general and no visible
differences could be observed (data not shown).
Fig. 19: Size exclusion chromatography of wt and k.d. supernatant
Concentrated supernatants of wt ASML and v71-16 k.d. cells were size fractionated by
chromatography on Superdex200 (A) and CL6B (B) sepharose columns. Fractions were coated to
plastic and used for adhesion assays. The void volume was defined by blue dextrane.
In order to identify components of the matrix mediating adhesion, CL6B fractions
were subjected to SDS-PAGE and western blotting and probed with different antibodies.
Shed CD44 was not present in the fractions promoting adhesion. In contrast, laminin was
present, although not exclusively in the adhesive fractions and no difference in the laminin
distribution was observed between the wt and the knock-down fractions (Fig. 20).
Fig. 20: Western blot analysis for CL6B-fractionated conditioned supernatant
Western blot analysis of fractionated supernatants. Concentrated supernatants of wt ASML and a
CD44vk.d. clone (v71-16) were size fractionated by chromatography on CL6B sepharose columns.
Fractions including the adhesive fractions (see Fig. 19) were subjected to SDS-PAGE and stained for
CD44v6 (A2.6) and laminin (polyclonal serum recognizing several laminins). Adhesion to the matrix is mediated by β1 integrin
The observation that trypsin treatment destroyed the deposited matrix but EDTA
treatment left the matrix unaltered, together with the finding that collagen and not HA seems
to be involved in the adhesion process suggests that adhesion of ASML to their own matrix
could be mediated by integrins. In line with this, addition of EDTA during adhesion
abrogated attachment of the cells, arguing for Ca2+ dependent adhesion (data not shown). To
further test this hypothesis different antibodies were tested for their ability to block adhesion.
For this purpose cells were pre-incubated with different integrin antibodies and adhesion was
performed as before. Figure 21 shows that only a β1 integrin antibody blocked adhesion in a
concentration dependent manner, while all other antibodies, including anti CD44, had no
influence on adhesion. Anti α6β1 induced cell aggregation and could therefore not be used
for blocking experiments. ASML cells do not express α1, α4 and α5 integrin chains (data not
shown). Pre-incubation of the coated matrix with an anti laminin polyclonal serum (reacting
with several laminins) did not interfere with adhesion. Therefore it seems likely that initial
adhesion of ASML cells to their own matrix is mediated by a β1 integrin binding to matrix
bound collagen, even though the specific type of collagen and the integrin α chain could not
be identified.
Fig. 21: Interfering with β1-integrin blocks adhesion to the secreted matrix
ASMLwt cells were pre-incubated with different antibodies and used for adhesion assays to wt ASML
matrix. For laminin blocking, the coated matrix was pre-incubated with anti laminin polyclonal serum
2.3.6 CD44vk.d. cells lack a secreted 180 kDa protein
To identify components present in the supernatant that might be responsible for the
matrix defect, silver staining of concentrated conditioned supernatant from wt and k.d. cells
was performed and revealed a 180 kDa protein, which was greatly reduced in all three k.d.
clones (Fig. 22A). The protein was subjected to mass spectrometric analysis and was
identified as ‘complement component 3’ (C3), an element of the innate immune system. The
mass spectrometry result was verified using a specific antibody and confirmed, that wt and
mock ASML cells secrete complement component 3 and that this is greatly reduced in the
k.d. cells. Concentrated conditioned supernatant was subjected to SDS-PAGE under reducing
conditions, which leads to separation of C3 into two subunits C3α and C3β (Fig. 22B).
However, the restored CD44 expression in the rescue clones failed to restore the secretion of
complement component 3, therefore, it can not be ruled out that this is an unspecific offtarget effect of the RNAi approach.
Fig. 22: Differential protein expression of wt and k.d. clones
A. Silver staining of concentrated supernatant of wt and k.d. clones reveals a differentially expressed
180 kDa protein, marked by an arrow. B. Western blot analysis after SDS-PAGE under reducing
conditions confirms differential expression of complement component 3, which is not restored in the
v71-14 rescue clones.
2.3.7 CD44vk.d. cells exhibit a reduced resistance to apoptotic triggers
Apoptosis resistance is a hallmark of metastatic tumour cells and ASML cells are
highly resistant to induction of apoptosis (Matzku et al., 1985). In order to compare
susceptibility to apoptosis of wt and CD44vk.d. cells, resistance to the chemotherapeutic drug
cisplatin and to γ-irradiation was evaluated. Cells were treated with different concentrations
of cisplatin for three days and survival was monitored by MTT staining. For γ-irradiation
adherent cells were irradiated with different doses. ASML cells display high drug and
radiation resistance and mock transfectants showed comparable levels, while all three
CD44vk.d. clones displayed significantly higher susceptibility to both kinds of apoptotic
triggers (Fig. 23). The IC50 for cisplatin was about 45µg/ml for wt and mock and about 510µg for all three CD44v k.d clones. For γ-irradiation the IC50 for the k.d. clones was at 250
Gy, while 600Gy killed only about 40% of wt and mock cells. The restored CD44 expression
in the rescue clones was not able to reestablish apoptosis resistance (Fig. 24).
Fig. 23: CD44vk.d. clones are more susceptible to apoptotic triggers
Wt ASML and k.d. clones were treated with different concentrations of cisplatin (A) or subjected to
different doses of γ-irradiation (B). Survival was monitored after three days by MTT staining.
Fig. 24: Restoring CD44 expression does not rescue the apoptosis resistance of
CD44vk.d. clones
Mock, k.d. and rescue clones were treated with different concentrations of cisplatin (A) or subjected
to different doses of γ-irradiation B). Survival was monitored after three days by MTT staining. Apoptosis resistance is increased by elongated pre-cultivation prior to irradiation
In order to test an influence of the matrix produced by ASML cells on apoptosis
resistance, cells were seeded and pre-incubated for either 15 or 48 h before irradiation. The
longer cultivation clearly leads to an enhanced resistance. Although the k.d. cells also display
higher resistance, they do not reach the level of the wt cells (Fig. 25).
Fig. 25: Influence of cultivation period on resistance to radiation
Cells were cultivated for 15 (A) or 48 hours (B) prior to γ-irradiation. Survival was monitored after
three days by MTT staining.
Higher susceptibility to apoptosis of CD44vk.d. cells is not reversible by wt matrix
To test a contribution of the produced matrix to apoptosis resistance, wt, k.d. and
rescue cells were seeded onto wt matrix, which had been prepared by EDTA removal of wt
cells. Only the wt cells showed a slight increase in resistance, while neither the k.d. clones
nor the rescue clones were able to make use of the wt matrix in terms of enhanced apoptosis
resistance (Fig. 26).
Fig. 26: Wt matrix does not rescue apoptosis resistance of CD44vk.d. cells
ASMLwt, k.d. and rescue clones were seeded on wt matrix and subjected to different doses of γirradiation. Survival was monitored after three days by MTT staining.
Results PI3K-Akt, rather than MAPK signalling is involved in apoptosis resistance of
ASML cells
The two main pathways influencing apoptosis resistance described to be influenced
by CD44 are the MAPK pathway and the PI3K-Akt pathway. Therefore, the influence of
specific inhibitors to these pathways was tested for γ-irradiation-induced apoptosis. ASMLwt
and CD44vk.d. cells revealed a marked increase in susceptibility to γ-irradiation when treated
with the PI3K specific inhibitor LY294002, while an inhibitor of the MAPK pathway (MEK
1/2 inhibitor) had no influence at the applied dose (Fig. 27A, Fig. 28A).
PI3k-Akt signalling is impaired in CD44vk.d. cells
PI3K and Akt inhibitors are known to induce apoptosis in some cancer cells on their
own, this was also apparent in the LY294002 treated, but not irradiated controls (Fig. 28A).
Therefore, ASMLwt and CD44vk.d. cells were tested for their tolerance for these inhibitors
at high concentrations. Using the same survival assay as for cisplatin treatment, adherent cells
were treated with different concentrations of either LY294002 or Akt II inhibitor and tested
for survival after three days. Wt and k.d. cells display high tolerance for both inhibitors.
However, the IC50 for wt and mock cells is 125µM for the PI3K inhibitor and 40µM for the
AKT II inhibitor, while k.d. cells show the same degree of apoptosis induction already at
50µM and 15-20µM respectively (Fig. 28B+C). This clearly demonstrates an impaired PI3KAkt signalling in the CD44vk.d. clones compared to wt cells. However, the rescue clones
displayed the same reduced tolerance as the k.d. clones. One rescue clone is shown as
representative example. The MEK1/2 inhibitor did not induce apoptosis even at very high
doses in none of the clones, which excludes a CD44 mediated involvement in MAPK
signalling (Fig. 27B).
Fig. 27: Interfering with MAPK signalling has no influence on apoptosis resistance of
ASML cells
A. ASMLwt (left panel) and CD44vk.d. cells v72-1-17 (right panel) were treated with a MEK1/2
inhibitor at 10µM and irradiated with 300 Gy. Survival was monitored after 3 days by MTT staining.
B. ASMLwt, mock and CD44vk.d. cells were treated with different concentrations of the MEK1/2
inhibitor. Survival was monitored after three days by MTT staining.
Fig. 28: Interfering with the PI3k-Akt pathway leads to decreased apoptosis resistance
and PI3kinase-Akt signalling is impaired in CD44vk.d. cells
A. ASMLwt and CD44vk.d. cells (v71-14) were treated with the PI3K specific inhibitor LY294002 at
50µM and irradiated with 300 Gy. Survival was monitored after 3 days by MTT staining. B. + C.
ASMLwt, mock, CD44vk.d. and rescue cells were treated with different concentrations of LY294002
(B) or an Akt II specific inhibitor (C). Survival was monitored after three days by MTT staining.
In order to identify differences in downstream signalling between wt and CD44vk.d.
cells, several anti- and proapoptotic molecules were evaluated. Upon cisplatin treatment with
10µg/ml for 24h, cells were lysed, subjected to SDS-PAGE, blotted and tested for expression
levels. Phosphorylation of Akt is lowered in the k.d. cells upon drug treatment, while levels
stay unaltered in the wt and mock cells. The same was observed for the antiapoptotic
molecule Bcl-2, which is only down-regulated in the CD44vk.d. clones after treatment. In
correlation with the inhibitor data, phosphoERK levels stay unchanged upon cisplatin
treatment in all clones (Fig. 29A+B).
CD44v are able to trigger activation of Akt
Restoring CD44v expression in the rescue clones did not compensate for the impaired PI3KAkt signalling of CD44vk.d. cells. Therefore, the influence of CD44v on the activation of this
pathway was tested by crosslinking surface CD44v on ASMLwt cells. Cells were seeded on
plates, coated with anti CD44v6 (A2.6) or BSA as control and lysed after 2h of incubation.
After SDS-PAGE and western blotting, phosphorylation of Akt was evaluated. An increase
upon crosslinking of CD44v was observed (Fig. 29C). This demonstrates that CD44v can
trigger activation of Akt in ASML cells and promote survival in this way.
Fig. 29: Reduced survival signalling in CD44vk.d. cells and activation of Akt by
crosslinking CD44v
A.+ B. Western blot analysis for untreated and cisplatin (10mg/ml) treated wt and k.d. cells. Cells
were lysed 24h after treatment. A. Akt becomes dephosphorylated in the CD44vk.d., but not in the wt
cells after drug treatment. Total Akt is used as loading control. Bcl-2 becomes down-regulated only in
the CD44vk.d. clones, but remains unaltered in ASMLwt and mock cells upon cisplatin treatment.
EpCAM (D5.7) is used as loading control. B. pERK staining does not change upon drug treatment.
EpCAM staining (D5.7) is used as loading control. C. Phosphorylation of Akt can be induced by
CD44v crosslinking. ASMLwt cells were seeded on plates coated with anti CD44v6 (A2.6) or BSA as
control and lysed after 2h. Total Akt is used as control.
The formation of metastases is the final stage of tumour progression and treatment is
still inefficient. Only very recently new therapeutic approaches gained access into clinical
application. For example monoclonal antibody therapies, such as ‘Herceptin’ targeting the
HER2 receptor in order to prevent breast cancer metastasis. The basis of these new strategies
is a molecular understanding of the mechanisms underlying the process of tumour
progression. However, many aspects of metastasis formation remain poorly understood.
The involvement of CD44 in tumour progression and particularly in the development
of metastases has been studied for many years, and multiple functions for CD44 and its
variant isoforms have been identified (Naor et al., 1997; Marhaba and Zoller, 2004). In this
work the role of variant CD44 in different aspects of the metastatic process was investigated
in ASML cells, a highly metastatic pancreatic adenocarcinoma. The pSuper RNAi system
was used to create stable and variant specific CD44 k.d. cells, which were characterized for
their metastatic capacities in vivo and in vitro. The contribution of CD44v as a cell-cell and
cell-matrix adhesion molecule was studied during the multistep process of metastasis
formation with special emphasis on interactions with the surrounding. In this respect, several
CD44v-mediated features supporting the settlement and survival of tumour cells during
lymphatic spread were identified. This highlights the role of CD44 as a multi functional
player during the course of metastasis formation through interactions with neighbouring cells
and the microenvironment, but also by actively organizing the ECM and finally functioning
as signalling molecule, supporting cell survival as well.
Loss of CD44 by stable and variant specific knock-down results in reduced metastatic
capacity of ASML cells
The BSp73 cell system comprises of two sublines of the same primary tumour, which
display different metastatatic potential. The two most abundant CD44 variants expressed by
the highly metastatic ASML cells are v4-v7 and v6/v7, and these isoforms were demonstrated
to confer metastatic capacity to otherwise only locally growing AS cells. In order to confirm
an essential contribution of these isoforms on the metastatic growth of ASML cells, RNAi
constructs were designed to target both isoforms. Two out of three constructs proved to be
efficient in down-regulating CD44v expression on ASML cells. Selection and recloning
yielded stable clones displaying only hardly detectable residual CD44v expression. Three
clones were established with ‘pSuper-v7’ (subsequently denoted ‘knock-down’ or ‘k.d.’
cells), while no stable clones could be raised with the ‘pSuper-v6/v7’ construct, which would
have allowed to easily control construct-specific off-target effects. To confirm any
phenotype, arising in the knock-down cells, clones with restored expression of one of the
dominant CD44v isoforms were successfully established from one of the k.d. clones (denoted
rescue cells). However, the expression level of the CD44v4-v7 rescue did not achieve the wt
Upon intra-footpad application, ASML cells spread exclusively through the
lymphatics. They display only little local tumour growth, but readily settle in the lymph
nodes and the lung, where they form miliary metastastic lesions (Matzku et al., 1983). All
three CD44v knock-down clones displayed reduced metastatic growth in vivo with severely
impaired settlement in the lung. For unknown reasons the injected CD44v4-v7 rescue clone
did not grow in vivo, probably due to a defect, acquired during in vitro culture, which might
have allowed the immune system to eliminate the tumour cells. However, five out of six
animals, injected with the CD44v6/v7 rescue clone displayed enhanced metastasis formation.
Especially when compared to the parental k.d. clone ‘v71-14’, this clearly demonstrates a
restored metastatic activity, although the massive tumour burden of the wt was not achieved.
The knock-down and the rescued CD44v expression remained stable throughout the in vivo
experiment, as confirmed by western blotting on re-cultivated tumour cells and by
immunohistology on lung sections. In summary, stable and variant-specific loss of CD44
expression in ASML does interfere with the lymphatic spread and lung settlement of these
cells. However, more rescue clones will be required to ensure statistical significance of the
restored metastatic capacity. It has to be mentioned in this respect, that complement
component 3 (C3) was found to be differentially expressed between ASMLwt and CD44vk.d.
cells, but could not be restored in the rescue clones, which might argue for an unspecific offtarget effect. As C3 is an important molecule of the innate immune system, it cannot be ruled
out that this might have affected the in vivo experiment. On the other hand, this does not
seem very likely, taken into account that the described rescue clone did restore the metastatic
capacity of the k.d., irrespective of the unrestored C3 expression.
CD44v knock-down cells display no phenotypic differences or altered growth characteristics
In order to identify defects, which could be causally related to the reduced metastatic
growth of ASML CD44vk.d. cells in vivo, the cells were studied in vitro. The impaired in
vivo growth is not due to a generally reduced proliferation rate, which was demonstrated by
proliferation assays under high and low serum conditions, nor to a reduced anchorageindependent growth, as shown in soft agar assays. Both, ASMLwt and CD44vk.d. cells are
not significantly affected by low serum conditions, as the proliferation rate was only very
slightly reduced. Both revealed a very high colony-forming efficacy of 90-95% in soft agar,
demonstrating high anchorage-independence. Another way, how CD44 could promote
metastasis formation is by up-regulating matrix degrading enzymes to promote invasion.
MMP2 and MMP9 are capable of degrading collagen IV, the major collagen of basement
membranes, and several studies link hyaluronic acid (HA) and CD44 to MMP2 and MMP9
secretion or activation (Zhang et al., 2002; Isnard et al., 2003; Murray et al., 2004). CD44 can
recruit MMP9 to the cell surface, leading to collagen IV degradation and invasion (Yu et al.,
1999). Gelatine zymography demonstrated low amounts of secreted MMP2 and MMP9 by
ASML, as enzymatic activity was only detectable after concentration of the supernatant. This
is in line with the relatively low ‘aggressive’ growth of ASML cells in vivo, which rather
grow by displacement than by destruction of host tissue (Matzku et al., 1983, 1985; Raz et
al., 1986). No significant differences in MMP2 and MMP9 secretion could be observed
between wt and CD44vk.d. cells. It can’t be ruled out though, that other MMPs are
deregulated in the k.d. cells, which might favour invasive growth. However, the lymphatic
spread of tumour cells might require different features and degrading basement membranes
might be less mandatory for entrance to the lymphatic vasculature than into blood vessels.
The lymphatic spread of tumour cells is still incompletely understood, but an interesting
aspect is the involvement of members of the VEGF growth factor family and chemokines in
this process. Tumour cells express chemokine receptors on their surface, enabling them to
migrate in a chemotactic manner to lymphatic vessels expressing the appropriate ligands. For
example, the chemokine receptors CXCR4 and CCR7 are involved in metastasis formation of
breast cancer (Muller et al., 2001) and melanoma cells (Wiley et al., 2001). A possible role
for chemokine receptors in the lymphatic spread of ASML cells was tested by RT-PCR for
CXCR4 and CCR7 expression. However, neither ASML nor AS cells express CXCR4 or
CCR7 (data not shown). Therefore, an implication of chemokine receptor expression on the
metastatic spread of ASML cells seems rather unlikely.
Agglomeration in HA-rich medium is abrogated in CD44vk.d. cells
ASML cells metastasize via the lymphatics to the lung. As the proliferative capacities
of CD44vk.d. cells are unaltered in vitro, the stromal surrounding of the lymph nodes or the
lung might promote tumour growth. When potential influences of stromal cell lines on
ASML cells were evaluated, it was noted, that conditioned cell culture supernatant of lymph
node stromal cells or of lung fibroblasts was able to induce agglomeration of wt, but not of
k.d. cells. This feature was completely rescued by restoring CD44v expression. Cell
clumping was due to HA-mediated cross-bridging of CD44 molecules on the cells, because it
could be induced by high concentrations of HA and was abrogated by hyaluronidase
treatment. CD44vk.d. cells are therefore unable to aggregate in stromal supernatant due to the
lack of receptor expression. CD44-mediated cell aggregation has been shown to enhance lung
metastasis after i.v. injection (Birch et al., 1991, Weber at al., 1996). Thus, it is interesting to
note that stromal cells, present at places of major tumour growth of ASML, are capable of
producing HA in amounts suitable to induce cell clumping, which could lead to reduced
motility and enhanced settlement and thereby promote tumour cell growth. Tumour cell
agglomeration has been described to facilitate settlement of tumour cells (Mansury et al.,
2002; Glinsky et al., 2003) and the lack of cell aggregation of CD44vk.d. cells may account
for reduced metastatic growth.
Impaired matrix production by CD44vk.d. cells
ASML cells adhere very slowly to plastic or matrix components and do not spread,
irrespective of the substrate (BenZe’ev et al., 1986). Instead, they display a rounded cell
shape, attach very tightly during cultivation and can only be detached by harsh trypsin
treatment or very long EDTA exposure. This attachment was reduced in the k.d. cells. When
conditioned supernatant of wt and k.d. cells was coated to plastic, both, wt and k.d. cells
adhered rapidly to the wt and mock matrix, while the k.d. matrix was clearly less adhesive.
Thus it seems likely, that matrixproduction is impaired in the CD44vk.d. cells. This was true
for matrix components secreted into the supernatant and for matrix deposited onto the plastic.
The rescue clones showed a partially restored matrix production, but could not fully
reproduce the adhesive properties of the wt matrix. The components of the matrix are
secreted, which could be demonstrated by ultracentrifugation. Hyaluronic acid is most likely
needed for proper assembly of the matrix, as hyaluronidase treatment prior to coating disrupts
the adhesive properties, but treatment after coating had no influence. This indicates that
adhesion of ASML cells is not mediated through CD44-HA interactions, which is confirmed
by the finding that CD44vk.d. clones do not display impaired adhesion to the wt-matrix.
Instead, collagenase treatment disrupts the adhesive properties of the matrix in a
concentration dependent manner, indicating that collagens function as ligands for adhesion.
When cells were removed by trypsin or EDTA treatment and the deposited matrix was used
for adhesion, only EDTA detachment of the cells left the matrix intact, while trypsin
treatment destroyed the adhesive properties. Adhesion could also be blocked by addition of
EDTA during the adhesion assay (data not shown), arguing for Ca2+-dependent adhesion.
Antibody blockade demonstrated that β1 integrins are involved, as pre-incubating the cells
with anti β1 antibody completely blocked adhesion in a concentration dependent manner.
ASML cells express different α chains complementing to β1 at high levels, but anti α2 and
anti α3 were not inhibiting adhesion. Anti α6β1 could not be used for antibody blocking, as
it induced cell clumping and hindered adhesion in this way (data not shown). Integrins
usually do not show high ligand specificity, but α2β1 and α3β1 are known to bind collagen
and other matrix components, while α6β1 shows a higher affinity to laminin. However,
additional α subunits that were not tested, such as α10β1 and α11β1 were shown to bind
collagens as well and might be involved in matrix binding of ASML cells.
Size chromatography under neutral conditions revealed, that the adhesion-mediating
components of the matrix are between 600 and 4000kDa in size. This corresponds to a
collagen ligand, when organized as multimer or with other matrix components, or present
within a HA-lattice. Altered matrix assembly in the k.d. cells may result in a differential
distribution of matrix components after size fractionation. The fractions were analysed by
western blotting revealing laminin as a component of the adhesive fractions. However, the
distribution did not differ in the k.d. matrix. It is possible though, that the resolution of the
applied chromatography was not sufficient for detecting differences in size distribution, or
that only the order of matrix components is affected, which does not necessarily alter the size
of matrix aggregates. However, antibody blocking with a polyclonal serum against laminin
did not affect adhesion. Therefore, laminin does not seem to mediate the rapid adhesion, but
is present in the matrix. Shed CD44 is not present within the adhesion-promoting fractions,
which excludes a mechanism of shedding CD44 to release assembled matrix components.
The specific collagen involved in adhesion could not be identified, as the available antibodies
did not function in western blotting. Tumour-stroma interactions are important for the
pathogenesis of pancreatic cancer, although the matrix production is usually attributed to
fibroblastic deposition of ECM (Gress et al., 1995), pancreatic cancer cells have also been
described to produce matrix components including collagens (Lohr et al., 1994). CD44 is
known to be important for pericellular matrix assembly in chondrocytes, where it functions as
a HA-anchor (Knudson et al., 2003; Jiang et al., 2002). Particle exclusion assays with
erythrocytes did not reveal any pericellular accumulation of matrix material by ASML cells
grown in vitro (data not shown). It might be possible however, that these cells are capable to
form a pericellular matrix, when aggregating proteoglycans are added in addition, as shown
for other cancer cells (Knudson and Knudson, 1991). The matrix of ASML cells is also
deposited on the culture dish and was even more adhesive when prepared in this way than the
coated supernatant. Whether the matrix material is actively deposited on the culture dish or
just passively bound to the plastic remains open. In conclusion, it seems that cell surface
CD44v is needed for proper assembly of matrix components in ASML cells, which most
likely is mediated through formation of a HA scaffold. In addition to a HA-anchoring
function of CD44, collagens might integrate into the matrix directly via binding to CD44, as
CD44-mediated adhesion to collagen containing matrices was described before (Knutson et
al., 1996). In chondrocytes, interfering with CD44 cell-matrix interactions results in a ‘matrix
remodelling’ response, which leads to chondrolysis via up-regulation of proteases and
enhanced biosynthesis of proteoglycans and HA (Knudson et al., 2000). In this view, it is
feasible, that also in ASML cells, CD44 interaction with matrix HA is not only important for
assembly, but might also affect matrix biosynthesis.
The ability of ASML cells to secrete their own adhesive matrix could promote
metastatic growth at a secondary site in two ways. First, adhesion is supported, enhancing
settlement of tumour cells in the lymph nodes and the lung, and second, the matrix could
support growth or survival. As metastases have to adapt to the new surrounding at a
secondary site, establishing their own matrix should be supportive. Metastasis formation of
CD44vk.d. cells may thus be reduced through an impaired ability to generate an adhesive
Impaired apoptosis resistance of CD44vk.d. cells
Pancreatic adenocarcinomas are aggressive cancers, characterized by invasiveness,
rapid progression and high resistance to chemo- and radiation therapy (reviewed by Bardeesy
and DePinho, 2002). Similarly, ASML cells exhibit low susceptibility to apoptotic triggers.
CD44 is described to be involved in apoptosis resistance and some reports link this feature to
variant isoform expression (Bates et al., 1998; Wittig et al., 2000; Marhaba et al., 2003).
Metastasis formation and resistance to apoptosis is closely related to the ECM and the
microenvironment, known to trigger survival signals. Susceptibility to apoptotic triggers was
greatly enhanced in ASML CD44vk.d. cells, using the chemotherapeutic drug cisplatin and γirradiation. Because CD44-HA interaction has been described to be involved in apoptosis
resistance (Toole, 2004) an impact of the secreted ASML-matrix was tested by enabling
matrix formation for different periods of time prior to irradiation. Indeed, prolonged
cultivation increased the apoptosis resistance of both, wt and k.d. cells significantly.
Moreover, wt cells could gain further protection by seeding on a preformed matrix,
confirming survival supporting functions of the matrix. However, CD44vk.d. cells did not
display enhanced resistance on a preformed wt-matrix, arguing for a direct influence of CD44
on survival signalling upon ECM mediated triggers. Yet, the rescue cells were not able to
compensate for this, which could have several reasons that will be discussed later.
Cells often react to the substrate by changing the expression of adhesion molecules
and alteration of the expressed integrin repertoire may contribute to tumour progression and
metastasis formation (Schwartz et al., 1993, Maschler et al., 2005). In pancreatic cancer for
example, α6β1 was described to influence metastatic behaviour (Vogelmann et al., 1999).
Integrin signalling is also known to affect survival (Lewis et al., 2002). For example, β1
integrin binding to ECM components promotes survival by activating PI3K in small-cell lung
cancer cells (Hodkinson et al., 2006). Islets of Langerhans cells are protected from anoikis
through β1 ligation with antibodies or by cultivation on collagen IV, which is accompanied
by an increase of Akt phosphorylation (Pinkse et al., 2006). However, evaluation of
expression levels of α6β1, α6β4 and the subunits α2, α3 and β1 in ASML cells revealed no
significant differences, neither between wt and k.d. cells nor upon prolonged cultivation (data
not shown). However, not all integrin dimers could be tested due to a lack of suitable
antibodies. Besides integrins, also other molecules involved in cell-matrix interactions and
contributing to survival could be affected by the matrix defect.
Irrespective of the protective feature of the produced matrix, CD44 is described to
interfere with apoptosis in different ways. As ASML CD44vk.d. cells display a reduced
resistance to drug treatment and irradiation, it seems unlikely that receptor-mediated
apoptosis is the reason. An involvement of multidrugresistance (MDR) cannot account for the
observed resistance to radiation as well. Nevertheless, CD44 was shown to influence the
expression of MDR genes (Misra et al., 2005; Tsujimura et al., 2006), prompting the analysis
of MDR genes in ASML cells. MRP2 and MRP5 are described to be involved in cisplatin
transport (Suzuki et al., 2001; Nomura et al., 2005; Oguri et al., 2000), but ASML cells were
tested negative for both transporters (data not shown). Instead, CD44v may promote
resistance to apoptosis by inducing survival signals. CD44 can signal through the PI3K-Akt
and the MAPK pathway to support survival. For example, CD44v6 crosslinking was shown
to protect from apoptosis in a thymoma cell line and this was accompanied by persisting
activation of MAPK signalling (Marhaba et al., 2005). In addition, in ASML cells CD44v6 is
necessary for c-Met activation by mediating ligand binding through complex formation,
resulting in phosphorylation of ERK (Orian-Rousseau et al., 2002). However, activation of
MAPK-ERK signalling upon apoptotic triggers was not observed in ASML cells and does not
seem to be influenced by CD44v expression, as wt and k.d. cells display comparable pERK
levels. In addition, a MEK1/2 specific inhibitor failed to interfere with apoptosis resistance
even at concentrations above MEK1/2 specificity inhibiting other members of the MAPK
pathway, like ERK1 and MKK3/p38 as well. Instead, inhibition of PI3K enhanced
susceptibility to apoptosis in wt and k.d. cells significantly. As inhibitors of PI3K-Akt are
known to induce apoptosis at high concentrations on their own, the resistance to this
treatment was tested without any additional apoptotic trigger.
Indeed, CD44vk.d. cells
displayed reduced tolerance to high doses of PI3K and Akt inhibition compared to wt cells,
phosphorylation of Akt is reduced in the k.d. cells but not in wt cells upon ciplatin treatment
and the anti-apoptotic molecule Bcl-2 is down-regulated only in the k.d. cells as well. In
addition, antibody-crosslinking of surface CD44v induced phosphorylation of Akt in ASML
cells. These findings clearly demonstrate that in CD44vk.d. cells one of the major survival
pathways is impaired, and that PI3K-Akt signalling contributes, probably in conjunction with
other triggers, to the high apoptosis resistance observed in ASMLwt cells. Although the
rescue clones could not reverse this defect, an involvement of CD44 was confirmed by
CD44v-mediated activation of survival signals through crosslinking.
Activation of PI3K leads to phosphorylation and activation of Akt. Active Akt
interferes with the apoptotic machinery by phosphorylating and inactivating the proapoptotic
molecule BAD and by inhibiting transcription of pro- and inducing transcription of
antiapoptotic molecules. In the active conformation BAD inhibits antiapoptotic members of
the Bcl-2 family, which stabilize the mitochondrial membrane and support survival (Igney
and Krammer, 2002). Because differences in pAkt levels become apparent after apoptotic
triggering, PI3K signalling is not constitutively altered in CD44vk.d. cells, which is observed
in many tumour cells through over-expression of PI3K subunits and Akt or lost expression or
mutation of the PI3K antagonist PTEN (Igney and Krammer, 2002). Instead, impaired
activation of PI3K by CD44 upon apoptotic triggering seems to account for the observed
lowered tolerance of CD44vk.d. cells to cisplatin and γ-irradiation. Soluble HA oligomers,
competing for HA binding to CD44 can inhibit anchorage-independent growth by
suppressing PI3K-Akt (Ghatak et al., 2002). Ligation of CD44 could enhance resistance to
drug treatment by activating the protein tyrosine kinase FAK leading to PI3K association and
activation of downstream targets of FAK, such as MAPKs (Fujita et al., 2002). Only
interfering with PI3K signalling, but not with the MAPK pathway decreased resistance to
apoptosis in this study as well. How CD44v induces PI3K activation remains to be explored
in detail, but it seems likely, that CD44v associated protein kinases are involved. However,
no associations with FAK or PI3K could be demonstrated by immunoprecipitations in ASML
cells (data not shown). Another possibility might be an inhibiting activity of CD44v on PI3K
antagonists, as soluble HA oligomers were demonstrated to be capable to stimulate the
expression of the PI3K antagonist PTEN (Ghatak et al., 2002). An involvement of the
described matrix production in apoptosis resistance seems likely as well. Activation of Akt
upon crosslinking of CD44v argues for CD44 acting directly as receptor for matrix
components, but, because crosslinking induces clustering as well, CD44v could also
cooperate with other molecules, such as integrins to activate survival signalling in ASML
cells. As β1 integrin was identified to mediate adhesion of ASML cells and ligation of β1
integrins is known to support survival, both possibilities seem feasible and might even act
CD44v-mediated apoptosis resistance and adhesiveness as crucial contributions during the
lymphatic spread of ASML cells
The importance of apoptosis resistance for metastatic cancer cells and especially
survival induced by the microenvironment is becoming more and more apparent. As
reviewed by Mehlen and Puisieux (Mehlen and Piusieux, 2006), one way to explain the very
rare incidence of metastatic outgrowth (Liotta et al., 1978; Varani et al., 1980) is to look at
apoptosis acting as a multistep barrier to metastasis at three crucial steps. During the initial
step of metastasis, cells detach from the underlying ECM and the actin cytoskeleton is
disrupted leading to cell rounding. These early events usually induce apoptotic processes
called anoikis and amorphosis (Streuli et al., 1999, Martin et al., 2004) and cancer cells must
become resistant to these apoptotic stimuli. The second step during which apoptosis can be
induced is the process of intravasation and during residence in the circulation. Cell death may
be induced by mechanical stress associated with entrance into the blood stream (Weiss et al.,
1993; Ziegler et al., 1998) or be mediated by the immune system, known as immune
surveillance (Jakobisiak et al., 2003). Indeed, cancer cells display a high frequency of
apoptosis when injected into the circulation, while over-expression of antiapoptotic
molecules like Bcl-2 can increase the number of metastatic lesions at secondary sites (Wong
et al., 2001). Moreover, the tumour suppressor p53 was shown to facilitate experimental
metastasis by promoting the survival of tumour cells in the circulation (Nikiforov et al.,
1996). Finally, after settlement cancer cells have to survive at the secondary site to form
metastases, the third phase, when cancer cells display high frequency of apoptosis. Enhanced
antiapoptotic signalling is supporting metastasis formation at this step as well (Wong et al.,
2001; Luzzi et al., 1998).
Fig. 30: Apoptosis at three crucial steps during the metastatic spread of tumour cells
a. Detachment from the primary tumour mass induces anoikis. b. Cell death in the circulation through
immune surveillance or mechanical stress. c. Apoptosis after extravasation during micrometastasis
formation at the secondary site. ECM extracellular matrix, NK natural killer cell. Adapted by
permission from Macmillan Publishers Ltd: Nature Reviews Cancer (Mehlen and Puisieux., 2006).
The observed CD44-mediated apoptosis resistance could protect ASML cells at
different points in the metastatic process. ASML cells do not form tight cell-cell contacts
under in vitro culture conditions and display only little local tumour growth in vivo.
CD44vk.d. cells do not display reduced anchorage-independence. Thus, the initial step of
dissemination from the primary tumour is probably not limiting for ASML metastasis
formation. ASML cells spread exclusively via the lymphatics, therefore they do not need to
invade blood vessels, and because lymphatics lack the tight inter-endothelial junctions of
blood vessels (Alitalo and Carmeliet, 2002), intravasation should be less stressful for tumour
cells and probably even allow small cell aggregates to enter. The growth in the draining
lymph nodes was already reduced in CD44vk.d. cells, which argues for a growth and/or
survival advantage in this environment, but the settlement in the lung seems to be most
severely impaired in the CD44vk.d. clones. While cells from connective tissue tumours, like
fibrosarcomas usually migrate individually, carcinomas were described to often migrate
collectively as small aggregates (Friedl and Wolf, 2003). It seems likely that ASML cells
start to migrate as individual cells, enter the lymphatics and aggregate only once they reach
the draining lymph node and are confronted with a HA-rich environment. This might be
favouring growth and survival within the lymph node and the lymphatic vessels and probably
facilitate further travel within the lymphatic system. The physiology of lymph nodes and the
relatively low shear flow of lymphatic fluid was discussed in this context to favour the
concentration of tumour cell aggregates in the lymph nodes, which then may support the
growth of local metastases that could serve as ‘bridgeheads’ for further dissemination
(Sleeman, 2000). In comparison, the metastatic progression in the lung is highly inefficient,
which could be explained in this respect by the large capillary bed, leading to dispersal of
individual tumour cells (Chambers et al., 2002). In addition, the lung seems to be harder to
colonize than the lymph nodes also for ASML cells, as even the rapidly growing ‘v71-16’
clone exhibited impaired settlement in the lung. This could be due to the lack of CD44vmediated aggregation or adhesion or again to increased apoptosis. Aggregation may facilitate
arrest in the capillary bed, which in turn would enable matrix deposition and subsequent firm
adhesion. Adhesion to matrix possibly then promotes survival at the secondary site.
Strong CD44-HA interaction can promote certain steps of the metastatic process,
while weak CD44-HA interaction might favour others. For example, release of cells from the
primary tumour is accompanied by reduced expression of CD44 in endometrial carcinoma
(Fujita et al., 1994). CD44 was shown to mediate attachment of circulating cancer cells to the
endothelial vessel wall through interaction with HA (De Grendele et al., 1996/1997).
Moreover, survival during micrometastasis formation can be mediated by CD44-HA binding.
This was shown in a mammary carcinoma cell line transfected with a soluble form of CD44,
which competed for HA-binding. The cells were able to infiltrate lung tissue but underwent
apoptosis thereafter and failed to form lung tumours (Yu et al., 1997). In addition, an
influence of CD44 glycosylation on HA binding was demonstrated (Skelton et al., 1998;
English et al., 1998). Because it is well known that glycosylation patterns are changed in
cancers and that changes can increase with tumour progression (Alhadeff, 1989), it seems
likely that changes in the HA binding ability of CD44 can by this means support the
metastatic process at different steps. The loss of CD44v may therefore not only suppress
metastasis formation by the inability to bind HA, but probably also because of the lost ability
to modify this interaction in a dynamic manner. This could still be the case in the rescue
clones and might be a reason for the limited ability of these cells to restore all observed
phenotypes of the k.d. cells. For instance the untranslated regions (UTRs) of the CD44
transcript could be functionally involved, as the rescue cDNAs did not contain the
endogenous UTRs. Recently the 3’ UTR of CD44 was described to be involved in
translational control by stabilizing the transcript through bound IMPs, a family of
ribonucleoproteins (RNPs). In addition, different CD44 variants exhibited differences in their
3’UTR sequence and regulation by IMPs was isoform specific. Moreover, interfering with
IMP function led to abrogation of invadopodia formation, which was attributed to the
deregulation of CD44 (Vikesaa et al., 2006). IMPs are described to be involved in mRNA
localization, with implications on coordinated spatio-temporal protein expression and
overexpression of IMPs is implicated in cancer progression (Ioannidis et al, 2001; Tessier at
al., 2004). This mechanism provides an additional level of complexity to CD44 regulation
and it cannot be ruled out that similar mechanisms are responsible for the failure of the
restored CD44 expression to rescue all observed phenotypes. The fact that the rescue clones
restored the metastatic ability as well as the matrix production in part, but failed to restore
apoptosis resistance is most easily explained by the lower overall CD44 expression level
compared to the wt situation. Specifically, the initiation of signal transduction might depend
on the expression level. It is also possible, that ASML cells require both isoforms, which
were affected by the knock-down or that additional isoforms were targeted, but not restored.
Future perspectives and conclusions
Differential contributions of untranslated sequences during the course of metastasis
formation seem very interesting with respect to CD44 regulation. Rescue constructs, carrying
endogenous UTRs could be applied to the established k.d. cell system, to study influences on
the metastatic growth in detail. As CD44 is reported to take part in the regulation of gene
expression, it might be worthy to look for deregulated gene or protein expression in the k.d.
cells, for example by microarray and CHIP analysis. With respect to the described matrix
production, changes in the expression of matrix components or molecules involved in matrix
remodelling, such as proteases could be addressed. In addition, the role of CD44v in matrix
organization could be analysed by studying morphological differences by electron or
fluorescence microscopy on deposited matrix material.
As the rescue clones did not revert all CD44vk.d. phenotypes, e.g. the reduced
expression of ‘complement component 3’, the CD44 specificity of these observations has to
be critically judged. The only alternative way to control any phenotype arising by RNAi is to
reproduce the same phenotype using a second construct. However, no stable clones could be
established with the second functional construct. Due to a low transfection efficiency of
ASML cells, transient transfection is not practical either. However, the use of an inducible
RNAi system could solve this problem, as generation of stable clones should be less
This work demonstrates an essential contribution of CD44v to the metastatic capacity
of ASML cells. However, considering the differences of AS and ASML cells, CD44v seems
to support metastasis formation in different ways depending on the cellular background. AS
cells adhere to and spread on different matrices, while ASML cells hardly adhere to any
substrate except their own matrix. Therefore matrix generation might be a crucial feature of
ASML cells but not for AS. Abrogation of HA binding had no influence on the metastatic
capacity of AS cells transfected with CD44v (Sleeman et al., 1996) and hence, CD44vmediated aggregation in lymph nodes or the lung might not be limiting for AS cells, as these
cells tend to build up tight cell-cell contacts anyhow. Still, the induction of agglomeration
might be an essential contribution of CD44v on metastasis formation of ASML cells. In
addition, upon transfection with CD44v, AS cells did not gain apoptosis resistance (data not
shown). Accordingly, the multitude of differentially expressed genes between AS and ASML
seems to influence the way CD44v contributes to metastatic progression of these cells. This
strengthens the idea that actions of CD44v depend on the cellular background, which is not
surprising given the heterogenous nature of CD44 functions and interactions.
In summary, variant-specific down-regulation of CD44 in a highly metastatic
pancreatic adenocarcinoma is accompanied by a markedly reduced metastatic capacity and
settlement in the lung. Defects in proliferative or anchorage-independent growth were ruled
out. Likewise, a change in the level of MMP2 and MMP9 secretion was not observed, which
would have argued for differences in invasive capacities. On the other hand, several
differences were examined and characterized, that could account for the observed metastatic
defect. First, CD44vk.d. cells lost the ability to aggregate in a stromal surrounding due to
their inability to crosslink surface CD44 through hyaluronic acid. Second, ASML cells
secrete a highly adhesive matrix, containing HA, collagen and laminin, to which they adhere
rapidly via β1 integrins and which might contribute to apoptosis resistance. CD44v are most
likely involved in matrix production by assembly of a HA-rich scaffold and therefore
CD44vk.d. cells display an impaired matrix generation. Finally, CD44vk.d. cells are clearly
less resistant to apoptotic triggers, as demonstrated for drug resistance and γ-irradiation,
which seems to be the cause of impaired PI3K-Akt signalling due to the loss of CD44vmediated activation.
These results support the idea of CD44 as a molecule with multiple features that, due
to its compositional and functional heterogeneity influences tumour progression and
metastasis formation at several different steps. Even in the studied cell system, which reflects
only the advanced steps of tumour progression, CD44v contributes to the process of
metastasis formation in several ways: as a cell-cell adhesion molecule, as organizer of
extracellular matrix components and as signalling molecule influencing survival. Importantly,
all the described mechanisms are based on complex interactions of CD44v on the tumour cell
with its surrounding in diverse ways. This underlines the importance of communication
between cancer cells and their microenvironment for the metastatic cascade, which is a major
subject of recent investigation and possesses growing importance for future therapeutic
Materials and Methods
Materials and Methods
4.1.1 Chemicals
3-Amino-9-ethyl-carbazol (AEC)
Ammoniumpersulfate (APS)
Brilliant Blue G-Colloidal Concentrate
Cisplatin (cis-diammineplatimun(II)dichloride)
Ethylendiamintetraessigsäure (EDTA)
Fetal calf serum (FCS)
Formaldehyde 37%
Yeast extract
Hyaluronic acid (from rooster comb)
Lubrol WX (17A17)
Milk powder
Mayer’s Hämalaun
β -Mercaptoethanol
Mowiol (4-88)
Pepton 140
Phenylmethylsulfonylfluorid (PMSF)
Protease Inhibitor Cocktail
Protein G Sepharose 4 Fast Flow
Rotiphorese Gel 30 (Acrylamid-Mix)
Sepharose CL-6B
Sepharose Superdex 200
Fluka, Buchs, Schweiz
Sigma, Seelze
Sigma, Seelze
Sigma, Seelze
Sigma, Seelze
Calbiochem, Darmstadt
Fluka, Buchs, Schweiz
Sigma, Seelze
Merck, Darmstadt
Sigma, Seelze
Merck, Darmstadt
Sigma, Seelze
Sigma, Seelze
Merck, Darmstadt
Merck, Darmstadt
Life Technologies, Karlsruhe
Roth, Karlsruhe
Roth, Karlsruhe
Gibco BRL, Eggenstein
Sigma, Seelze
Sigma, Seelze
Calbiochem, Darmstadt
Serva, Heidelberg
Roth, Karlsruhe
AppliChem, Darmstadt
Sigma, Seelze
Calbiochem, Darmstadt
Sigma, Seelze
Sigma, Seelze
Sigma, Seelze
Gibco BRL, Eggenstein
Sigma, Seelze
Roche Diagnostics, Mannheim
Amersham Pharmacia, Freiburg
Roth, Karlsruhe
Amersham Biosc., Freiburg
Amersham Biosc., Freiburg
Sigma, Seelze
Sigma, Seelze
Materials and Methods
Triton X-100
Tween 20
Amersham Biosc., Freiburg
Sigma, Seelze
Sigma, Seelze
Serva, Heidelberg
All other chemicals, not listed were analytical grade and purchased from Sigma (Seelze),
Calbiochem (Darmstadt), Serva (Heidelberg) or Applichem (Darmstadt).
4.1.2 Enzymes
resctriction enzymes
T4-Polynucleotidkinase (T4-PNK)
Klenow fragment
ImProm II reverse transcriptase
calf intestinal alkaline phosphatase (CIAP)
hyaluronidase type IV-S from bovine testis
collagenase, type 2
MBI Fermentas, St. Leon Rot
MBI Fermentas, St. Leon Rot
Promega, Mannheim
Promega, Mannheim
Promega, Mannheim
Promega, Mannheim
Promega, Mannheim
Promega, Mannheim
Sigma, Seelze
PAA, Coelbe
4.1.3 Chemical inhibitors
Akt II inhibitor
Calbiochem, Darmstadt
Calbiochem, Darmstadt
Calbiochem, Darmstadt
4.1.4 Nucleotide and protein standards
100bp Gene Ruler
1 kb Gene Ruler
Prestained Protein ladder
MBI Fermentas, St. Leon Rot
MBI Fermentas, St. Leon Rot
MBI Fermentas, St. Leon Rot
4.1.5 Kits
Qiaprep Spin miniprep kit
Qiaquick Gel Extraction Kit
Qiaquick midi prep kit
ECL Western Blotting Detection Reagents
Vectastain ABC kit
TRI Reagent
QIAGEN, Hilden
QIAGEN, Hilden
QIAGEN, Hilden
Amersham Biosc., Freiburg
Vector Laboratories, Burlingame, USA
Sigma, Seelze
Materials and Methods
4.1.6 Vectors
eukaryotic expression vector for
eukaryotic expression vector
eukaryotic expression vector
Oligoengine, Seattle,
Invitrogen, Karlsruhe
Invitrogen, Karlsruhe
4.1.7 Primers and oligos
Oligonucleotides were purchased from Operon Biotechnologies (Koeln)
Materials and Methods
4.1.8 cDNAs and constructs
CD44v4-v7 cDNA and CD44v6/v7 cDNA (Gunthert et al., 1991; Rudy et al., 1993) were
used as template for PCR amplification. All constructs were verified by sequencing.
cloning sites
4.1.9 Antibodies Primary antibodies
antibody (clone)
mouse anti rat CD44pan (Ox50)
mouse anti ratCD44v6 (A2.6)
mouse anti rat EpCAM (D5.7)
mouse anti rat D6.1A (D6.1)
mouse anti rat C4.4A (C4.4)
mouse anti rat transferrin
receptor (Ox26)
mouse anti rat α3β1 integrin
mouse anti rat α6β4 integrin
mouse anti rat β1 integrin
rabbit anti human α3 integrin
hamster anti rat α2 integrin
mouse anti rat α6β1 integrin
mouse anti rat α4 integrin
FACS, IP, blocking exp.
FACS, IP, blocking exp.,
immunohistology, WB,
FACS, immunohistology
FACS, IP, blocking exp.
FACS, blocking exp.
FACS, IP, blocking exp.
FACS, blocking exp.
Paterson et al. 1987
Matzku et al. 1989
Wurfel et al., 1999
Claas et al., 1998
Rosel et al., 1998
European Collection of
Animal Cell Culture
Developmental Studies,
Hybvridoma Bank
Herlevsen et al., 2003
BD Biosciences,
Chemicon, Temecula,
BD Biosciences,
Chemicon, Temecula,
BD Biosciences,
Materials and Methods
hamster anti rat α1 integrin
hamster anti mouse α5 integrin
mouse anti rat β2 integrin
mouse anti rat β3 integrin (F11)
mouse anti human ERK (16)
mouse anti human pERK1/2
mouse anti human Akt (2)
mouse anti human pAkt
mouse anti human Bcl-2(7)
rabbit anti human laminin
(polyclon. #600-401-116-05)
goat anti rat complement C3
WB, blocking exp.
BD Biosciences,
BD Biosciences,
BD Biosciences,
BD Biosciences,
BD Biosciences,
BD Biosciences,
BD Biosciences,
BD Biosciences,
BD Biosciences,
Rockland, Gilbertsville,
MP Biomedicals, Aurora,
Ohio, USA Secondary antibodies/ reagents
anti mouse-IgG-TxRed
anti mouse-IgG-biotin
anti mouse-IgG-HRP
anti goat-IgG-HRP
anti rabbit-IgG-HRP
anti mouse-IgG-PE
anti mouse-IgG-APC
anti mouse-IgG-FITC
anti rabbit-IgG-PE
anti hamster-IgM-FITC
dianova, Hamburg
dianova, Hamburg
Rockland, Gilbertsville, USA
Rockland, Gilbertsville, USA
Rockland, Gilbertsville, USA
Sigma, Seelze
dianova, Hamburg
BD Biosciences, Heidelberg
dianova, Hamburg
dianova, Hamburg
dianova, Hamburg
Materials and Methods
4.1.10 Cell lines
Cell line
HEK 293T
Rattus norvegicus, pancreas
Rattus norvegicus, pancreas
Rattus norvegicus, pancreas
adenocarcinoma BSp73AS, transfected
with cDNA for CD44v4-v7
Homo sapiens, renal cell carcinoma
Rattus norvegicus, immortalized lung
Rattus norvegicus, immortalized
lymph node stromal cell
Rattus norvegicus, immortalized
lymph node stromal cell
Matzku et al., 1983
Matzku et al., 1983
Gunthert et al., 1991
Graham et al., 1977
Weth, 2000
LeBedis et al., 2002
LeBedis et al., 2002
4.1.11 Animals
BDX rats were bred at the animal facilities of the German Cancer Research Center (DKFZ),
kept under pathogen free conditions and fed with sterilized food and water.
4.2.1 Molecular biology Bacteria
For all bacterial work DH5α were used. DH5α were cultured in liquid Luria Bertani
(LB) medium (10g peptone, 5g yeast extract, 10g NaCl per l) or on plates containing 1%
(w/v) bacto agar and either 60µg/ml ampicillin or 50µg/ml kanamycin for selection.
Transformations were carried out with an ‘EasyJect’ electroporator (Eurogentec, Seraing,
Belgium) using standard protocols for electro-transformation.
DH5α genotype:
Φ80dlacZΔM15, recA1, endA1, gyrA96, thi-1, hsdR17(rK-, mK+), supE44,
relA1, deoR, Δ(lacZYA-argF)U169;
Materials and Methods Plasmid preparation
All plasmid preparations were carried out using either the ‘QIAprep spin Miniprep
kit’ or ‘plasmid midi prep kit’ (both QIAGEN, Hilden) according to manufacturer’s
instructions. RNAinterference (RNAi) construct design and cloning
In order to create stable CD44 knock-down clones the pSuper plasmid based RNAi
system (oligoengine, Seattle, USA) was used. The pSuper plamids facilitate expression of
‘small hairpin RNAs’ (shRNAs) under the control of the human RNA polymerase III
promoter H1, which were shown to enter the endogenous RNAi pathway and induce
degradation of the target mRNA. A GFP reporter and a neomycin resistance are included in
the vector for detection and selection of transfected cells.
Fig. 31: pSuper.gfp/neo inserts and predicted haipin structure
A. ds inserts for cloning into pSuper.gfp/neo. The target sequences lie within exon v6, v7 or span the
border of exon v6/v7. Due to the self-complementary sequence, transcripts fold into a
hairpinstructure, which is processed by the endogenous RNAi machinery to a functional siRNA
leading to target mRNA degradation. B. Predicted hairpin structure for construct ‘v7’.
The target sites were chosen on the premise to target the two most abundant CD44
variant isoforms, expressed by ASML, which are v4-v7(meta1) and v6/v7(meta2). Target
sites for constructs ‘v6’ and ‘v7’ were chosen randomly, except that stretches of more than 3
A-residues were avoided. Homologies to other genes were excluded by ‘BLAST’ search. For
Materials and Methods
the ‘v6/v7’ construct, which spans the border of the v6-v7 exons, the RNAi target site
validation program ‘Sfold’ was used, which is provided on the homepage of Wadsworth
Center-NYS, Department Of Health (; Ding et al., 2004; Ding and
Lawrence, 2001/2003) and includes several parameters like internal stability of the siRNAdublex and target site accessibility. The constructs consist of a self complementary sequence
stretch comprising the 21-28 bp target sequence and the corresponding reverse complement,
separated by a 10 bp loop and followed by a stretch of 5 T-residues leading to termination of
transcription. Sense and antisense oligos were designed to elicit sticky ends upon annealing.
200pmol of each ss oligo were diluted in ligation buffer (Promega, Mannheim), boiled for
2min and slowly cooled down to room temperature to ensure proper annealing. Because
oligos were initially designed for XhoI-XbaI cloning, the ds inserts were partially filled up
and ligated into the, as well partially filled up, Bgl II/Hind III sites of the pSuper.gfp/neo
plasmid. Positive clones were verified by sequencing and used for transfections. PCR-based mutagenesis for rescue constructs
In order to be able to control specificity of any phenotype arising in the knock-down
cells, ‘rescue’ clones were established, in which expression of one of the dominant variants
was restored. This was achieved by transfection of cDNAs for either CD44v4-v7 or
CD44v6/v7, protected from degradation by four silent point mutations (highlighted in red) in
the v7 target sequence, which were introduced by PCR.
v7 target site:
v7 rescue site:
The amplificates were cloned into pCDNA3.1Hygro (Invitrogen, Karlsruhe) and
positive clones were varified by sequencing. One of the stable knock-down clones,
ASMLv71-14 was used for transfections with the ‘rescue’ constructs. RNA-isolation and reverse transcription-PCR (RT-PCR)
RNA preparations were carried out with ‘TRI Reagent’ (Sigma, Seelze). About 5x106
cells were lysed directly into 1ml of TRI Reagent by scraping. After addition of 0.2ml
chloroform and centrifugation (12000g, 15min, 4°C) the aqueous phase was transferred and
Materials and Methods
RNA was precipitated with isopropanol. After centrifugation the pellet was washed with
ethanol, air-dried and resuspended in DEPC treated H2Od. Integrity of the isolated RNA was
controlled by gelelecrophoresis.
cDNA was generated using the ‘ImProm II’ system (Promega, Mannheim) with oligo
(dT) primers following the manufacturer’s instructions. 2µl of the reverse transcription
reaction was used as template for the PCR.
4.2.2 Cell biology Cell culture
Eukaryotic cells were kept in RPMI 1640-medium, containing 10% heat inactivated
fetal calf serum (FCS), 100U/ml penicillin, 100µg/ml streptomycin and cultured at 37°C,
95% humidity, 5% CO2. For passaging, cells were trypsinized with 0.25% trypsin (w/v)/5mM
EDTA in PBS (137mM NaCl, 8,1mM Na2HPO4, 2,7mM KCl, 1,5mM KH2PO4, pH 7.4). Cryo-conservation of eukaryotic cells
1x107 cells were trypsinized, washed once with fresh medium and resuspended in icecold FCS/10% DMSO. Cells were kept over night at -80°C and transferred to liquid nitrogen. Transfection of eukaryotic cells
ASML cells were seeded the day before transfection to 70% confluency. Transfection
was carried out with the ‘ExGene 500’ reagent (MBI Fermentas, St.Leon Rot) following the
instructions of the manufacturer. HEK293T cells were transfected with ‘PolyFect
Transfection reagent’ (QIAGEN, Hilden) at 50% confluency according to the manufacturer’s
instructions. Recloning of transfected cells by limiting dilution
Transfected cells were selected for drug resistance and checked by FACS for
expression of the transgene. Limiting dilutions of 1 or 5 cells per well were carried out in 24well plates. Cells were grown in the presence of 1x106 freshly prepared rat thymocytes as
Materials and Methods
growth support. Clones were checked by FACS and used for a second round of dilution to
ensure clonality. Collection of conditioned cell culture supernatant
90% confluent cells were grown in serum free medium for 24h, the supernatant was
harvested and centrifuged to remove cells (300g, 10min). Cell free supernatant was filtered
through a 0.2µm filter and either coated to 24-well plates for adhesion assays, or concentrated
through a ‘Vivaspin 6 column’ (50.000 MWCO) (Sartorius, Goettingen) and used for SDSPAGE or gel filtration.
Deposited matrix was prepared by cultivating confluent cells for 24 hours and
removing cells by EDTA treatment (5mM in PBS, pH8.0), followed by intensive washing
with PBS. Coating of plastic surfaces
Proteins bind under alkaline conditions to plastic surfaces. 50mM Tris (pH 9.5) was
used as a binding buffer for matrix components, which were coated at a concentration of
10µg/ml. Collagens were pre-incubated at 37°C for 4h prior to coating. Hyaluronic acid was
coated at very high concentrations (1mg/ml) under neutral pH in PBS. Conditioned cell
culture supernatant was coated at pH 7.4. All coatings were carried out over night at 4°C,
wells were washed once with PBS, followed by blocking with BSA (3mg/ml in PBS) for 2h
at RT and three washes with PBS. Coated plates were used for adhesion assays.
Where indicated, conditioned cell culture supernatant was treated with hyaluronidase
(Sigma, Seelze) or collagenase (Sigma, Seelze) for 4 h at 37°C prior to or after coating. Heat
inactivated enzymes were used as controls. Adhesion assay
Cells were trypsinized and recovered for 1-2h in RPMI/10% FCS. Adhesion assays
were carried out in 24-well plates. Cells were washed with PBS, counted and 1x106 cells
were resuspended in serum free medium or PBS with or without additives. Cells were seeded
and incubated at 37°C for 15min. Adherent cells were stained with crystal violet (see section
Materials and Methods Agglomeration assay
Cells were trypsinized and recovered for 1-2h and resuspended in conditioned cell
culture supernatant of different origin. Where indicated, cell culture supernatant was treated
with hyaluronidase (1mg/ml) at 37°C for 2 or 5h before use, or cells were seeded in RPMI
medium, supplemented with 1mg/ml hyaluronic acid. Agglomeration was monitored with a
Leica ‘DM-IL’ inverse microscope (Leica, Solms) and documented with a SPOT CCD
camera using the SPOT 2.1.2 software. Proliferation assay
5x104 cells were seeded in 96-well plates in RPMI supplemented with 10% or 0.5%
FCS. 3H-thymidin was added for 24h at different time points. Cells were harvested and H3
incorporation was counted in a liquid scintillation counter. Alternatively 5x104 cells were
seeded as before and quantified by crystal violet staining after different time intervals (see
section Soft agar assay
Tumour cells were suspended in RPMI/0.5% agar and either 100 or 1000 cells were
seeded on a pre-poured RPMI/3% agar layer in the presence of 10% FCS. Colonies were
counted after 4 weeks. Drug treatment
1x105 cells per well were seeded in 96-well plates and grown overnight. Serial
dilutions of a 50µg/ml starting concentration of cisplatin in RPMI medium were carried out
in 1:2 steps. For chemical inhibitors starting concentrations were 100mM for the MEK1/2
and Akt II inhibitor and 200mM for the PI3K specific inhibitor LY294002 (Calbiochem,
Darmstadt). Cells were treated for 3 days and surviving cells were stained either by MTT
staining or crystal violet staining.
Materials and Methods γ-irradiation of adherent cells
1x106 cells were seeded in 35mm petridishes, grown for 15, 24 or 48h, fresh medium
was added and monolayers were subjected to different doses of γ-irradiation in a
‘GAMMACELL 1000D’ unit (AECL, Ontario, Canada). Survival was monitored by MTT
staining after 72h. Where indicated, chemical inhibitors were added 30min prior to
irradiation. MTT staining of respiratory active cells
A 5mg/ml stock solution of MTT in PBS was diluted 1:10 directly in the growth
medium of the cells and incubated for 30-40min at 37°C. Non-adherent cells were
precipitated by centrifugation, the supernatant was aspirated and cells were resuspended in
DMSO and measured at 550nm in an ELISA reader. Crystal violet staining of adherent cells
Adherent cells were washed and fixed with 4% formalin in PBS for 4min at RT. The
solution was changed to 1% crystal violet (in 10% EtOH) and incubated for 4min at RT. The
plates were washed extensively with H2Od, dried and cells were resuspended in 10% acetic
acid. Absorbance was measured at 595nm in an ELISA reader. FACS analysis
For FACS staining, cells were trypsinized and recovered in complete medium for 12h. Staining was performed in U-shaped 96-well plates with 106 cells per well. Incubation
with primary and secondary antibodies in PBS was carried out at 4°C for 30min in the dark.
After each step cells were washed three times with PBS by centrifugation at 300g for 4min.
FACS analysis was performed using a ‘FACSCalibur’ (Becton Dickinson, Heidelberg) and
analyzed with the ‘CellQuest Pro’ software. Immunofluorescence staining of cells grown on coverslips
Cells were grown on coverslips for 1-2 days. Adherent cells were washed three times
with PBS and fixed with 4% paraformaldehyde (PFA) in PBS for 25min on ice, followed by
Materials and Methods
three washes with PBS/200mM glycine and three washes with PBG (PBS, 0.2% (v/v)
gelatine, 0.5% (w/v) BSA). Primary antibody incubation was carried out for 1h at 4°C in
PBG, followed by three washes with PBG. After incubation with Texas-Red conjugated
secondary antibody for another hour at 4°C and three more washes with PBG, cells were
rinsed with H2Od, air-dried and mounted in Elvanol (20% (w/v) Mowiol in 2/3 PBS, pH 8.0
and 1/3 glycerine). Fluorescence microscopy was done using a Leica DMRBE microscope,
equipped with a SPOT CCD camera using the SPOT 2.1.2 software for documentation. Cryo-sectioning of tumour tissue
Tissues were embedded in frozen section medium ‘Neg-50’ (Richard-Allan Scientific,
Kalamazoo, USA) and frozen in liquid nitrogen. Cryo-sectioning was carried out with a
Reichert Jung ‘2800-FRIGOCUT E’ to sections of 5µm thickness and transferred to
chromalaune-gelatine coated glass slides for staining. Immunohistological staining of cryo-sections
Sections were blocked with PBS/2% FCS for 30min, fixed with acetone/methanol
(1:1) for 4min and washed with PBS. All incubation steps were carried out at 37°C in a
humidity chamber. Sections were stained with primary antibodies or mouse IgG for 1h, after
washing and incubation with biotinylated secondary antibody, detection was carried out using
the ‘Vectastain ABC kit’ according to the manufacturer’s instructions. Briefly, Vectastain
AB-complex (containing avidin and biotinylated-peroxidase) was added for 30min, followed
by incubation for 5-20min at RT with freshly prepared AEC mix [AEC-solution-1 (2.1ml
acetic acid (0.1M), 7.9ml sodiumacetate (0.1M)) + AEC-solution-2 (4mg 3-amino-9-ethylcarbazole (AEC) in dimethylformamide (DMF)) + 5µl H2O2 (30%)]. After washing with
PBS, sections were counterstained with Mayer’s Haemalaun and mounted in ‘Kaisers
glycerine-gelatine’. Microscopy was done using a Leica DMRBE microscope, equipped with
a SPOT CCD camera using the SPOT 2.1.2 software for documentation.
Materials and Methods
4.2.3 Animal experiments In vivo metastasis assay
1x106 tumour cells were suspended in PBS and injected into the footpad of 10-14
week old female BDX rats. Animals were sacrificed on day 50 in experiment 1 and on day 60
in experiment 2. Rats were dissected and diameters of primary and lymph node tumours were
measured. Lungs were photographed and weighed. Samples of infiltrated tissues were
embedded for cryo-sectioning and immunohistology or recultivated. For recultivation,
tumour tissue was meshed through a sterile gauze and seeded in RPMI-medium.
4.2.4 Protein biochemistry Surface biotinylation of molecules
Adherent cells were washed twice in PBS and incubated 30min at RT with 100500µg/ml Biotin-X-NHS (Calbiochem, Darmstadt) in 25mM HEPES/150mM NaCl/5mM
MgCl2 on a shaking platform. Cells were washed three times with ice-cold PBS/200mM
glycine and suspended in lysis buffer by scraping. Immunoprecipitation (IP)
6x gel loading buffer: 300mM Tris pH6.8, 12% (w/v) SDS, 0.6% (w/v), bromophenolblue,
20% (v/v) glycerine
Lysis of cells was performed on ice or at 4°C and lysates were kept cold during the
whole procedure. Cells were washed twice with PBS and scraped into ice-cold lysis buffer
(25mM HEPES, 150mM NaCl, 5mM MgCl2, 2mM PMSF, 1x proteinase inhibitor mix
(Roche, Mannheim), 1% (v/v) detergent). Lysis was performed on a rotating platform for 1h.
Unsolubilized material and cell nuclei were pelleted by centrifugation (15min, 15000g) and
the cleared lysate was used for IP. Either 5µg purified antibody or 200µl hybridoma
supernatant was used per 1ml cell lysate. Antibody binding was carried out for 1h on a
rotating platform. For precipitations, 0.1 volumes protein G Sepharose was added to the
antibody complexes and samples were rotated for another hour. Complexes were washed 4
Materials and Methods
times with lysis buffer. After the last washing step all liquid was removed through a 35gneedle attached to a vacuum line to ensure minimal background. Complexes were
resuspended in gel loading buffer and boiled for 5min at 95°C. Sepharose beads were
pelleted by short centrifugation and the supernatant was subjecetd to SDS-PAGE. For Re-IPs,
the sepharose complexes were resuspended in lysis buffer containing a detergent of higher
stringency (usually TX-100) and extracted for 1h at 37°C on a shaker. After removal of the
sepharose, the supernatant containing the extracted antigens was subjected to another round
of IP as described above. Lysis of intact cells for SDS-PAGE
For western blotting of cell lysates from complete cells, cells were washed twice with
PBS and scraped directly into gel loading buffer. Cell lysates were sonicated (5 impulses,
5sec each). Lysates were boiled for 4min and used for western blot analysis. SDS-polyacrylamide gel electrophoreses (SDS-PAGE)
For electrophoretic separation of protein samples the ‘Mini-Protean II’ system from
Biorad (Munich) was used for discontinuous SDS-PAGE. 5ml of separating gel (375mM Tris
pH 8.8, 0.1% (w/v) SDS, 6-12% acrylamid-bisacrylamid, 0.1% (v/v) TEMED, 0.1% (w/v)
ammoniumpersulphate) were overlaid with 2ml of stacking gel (375mM Tris pH 6.8, 0.1%
(w/v) SDS, 5% acrylamid-bisacrylamid, 0.1% (v/v) TEMED, 0.1% (w/v) ammoniumpersulphate). After complete polymerization, gels were loaded and run in gel running buffer
(25mM Tris, 192mM glycine, 0.1% (w/v) SDS) at a constant voltage of 200V. Gels were
either stained with Colloidal Coomassie or silver or subjected to western blot analysis. Western blotting (modified after Towbin et al., 1979)
After SDS-PAGE, protein gels were equilibrated for 10min in transfer buffer (25mM
Tris, 192mM glycine, 0.02% (w/v) SDS, 20% (v/v) methanol). Nitrocellulose membranes
(Amersham, Braunschweig) and 3MM whatman paper were equilibrated as well. For protein
transfer, the gel was placed on whatman paper, followed by nitrocellulose and another
whatman paper. The wet transfer was carried out in transfer buffer at a constant voltage of
30V over night at 4°C.
Materials and Methods
After transfer had been completed, the membranes were blocked for 1h at room
temperature with 5% (w/v) fat free milk in PBST (PBS/0.1% (v/v) TWEEN 20) or, for
detection with phosphospecific-antibodies, with 5% (w/v) BSA in TBST (TBS/0.1% (v/v)
TWEEN 20). Antibody incubations were carried out for 1h at RT with hybridoma
supernatant or purified antibody in PBST or TBST, respectively. Membranes were washed
three times for 5min in PBST or TBST and incubated with secondary antibody conjugated to
horseradish peroxidase (HRP) (diluted 1:5000 in PBST or TBST) for 1 h at RT, followed by
additional three washing steps.
Biotinylated proteins were detected with ExtrAvidin-peroxidase (Sigma, Seelze).
Detection was done by chemiluminescence using the ‘ECL Western blotting detection
reagents’ and ‘ECL radiography films’ (both Amersham, Braunschweig). Colloidal Coomassie staining of protein gels
After electrophoretic separation, proteins were fixed for 1h in 7% acetic acid/40%
(v/v) methanol. Gels were stained over night in staining solution (4 volumes Colloidal
Coomassie staining solution (Sigma, Seelze) + 1 volume methanol). Gels were destained for
30s in 10% acetic acid/25% (v/v) methanol and kept in 25% (v/v) methanol. Silver staining of protein gels
After separation of proteins by SDS-PAGE, gels were fixed over night in 30%
ethanol/10% acetic acid and sensitized for 45min (in 0.3% potassium tetrathionate, 0.5M
potassium acetate, 30% ethanol), followed by 6 washes a 10min with H2Od. Gels were
stained with 0.2% silver nitrate for 1-2 h, rinsed with H2Od and developed for up to 40min in
developer (3% potassium carbonate, 31µl Na2S2O3-5H2O (10%), 75µl formalin (37%) per
250ml). The reaction was stopped by adding 330mM TRIS/2% acetic acid and gels were kept
in H2Od. Gelatine zymography for detection of MMP activity
Conditioned cell culture supernatant was collected as described and concentrated 10
times through a ‘Vivaspin’ column (50.000 MWCO) (Sartorius, Goettingen). The supernatant
was mixed with Laemmli buffer, incubated for 15min at 37°C and subjected to SDS-PAGE
Materials and Methods
in an 8% acrylamide gel containing 1mg/ml gelatine as substrate. After electrophoresis the
gel was incubated three times for 40min in 2.5% (v/v) TX-100, washed in developing
solution (50mM Tris pH 7.4, 10mM CaCl2, 150mM NaCl2) and incubated for 24h in
developing solution at 37°C and subsequently subjected to Coomassie staining. Gel-filtration
Superdex 200/CL6B sepharose beads were washed in PBS/0.1% NaN3, degased and
packed into a column of 1.5cm diameter and 60cm length. The column material was
equilibrated with PBS/0.1% NaN3 over night. The void volume was calculated by blue
dextrane. 30x concentrated cell culture supernatant was mixed with glycerine (900µl
supernatant + 100µl glycerine) and loaded onto the column. Fractions of 3ml each were
collected, of which 2ml were used for adhesion assays and 1ml was concentrated by TCAprecipitation and subjected to SDS-PAGE analysis. Ultracentrifugation of cell culture supernatant
Cell culture supernatants were centrifuged at 100000g over night at 4°C in a Beckman
Coulter ‘Optima LE-80K’ ultracentrifuge using a SW-41 rotor. The supernatant was
transferred and precipitated material was washed with RPMI, centrifuged again for 1h and
resuspended in RPMI medium. Supernatant and resuspended pellet was used for coating 24well plates or subjected to SDS-PAGE. TCA-precipitation of proteins
To concentrate protein in solution, proteins were precipitated with trichloric acid
(TCA) at a final concentration of 5% (w/v) for 5min at 65°C and cooled on ice. After
centrifugation (12000g, 30min, 4°C) the pellet was resuspended in gel loading buffer and
subjected to SDS-PAGE. Analysis of proteins by mass spectrometry
Protein gels were stained with Colloidal Coomassie as described. Proteins of interest
were cut out with a scalpel. Subsequent preparations and mass spectrometrical analysis was
Materials and Methods
carried out at the central service of the DKFZ by MALDI-analysis (matrix assisted laser
desorption/ionisation) using a ‘Reflex II time-of-flight’ mass spectrometer (Bruker-Daltonics
GmbH, Bremen).
Adhikary, S. and M. Eilers (2005). "Transcriptional regulation and transformation by Myc
proteins." Nat Rev Mol Cell Biol 6(8): 635-45.
Ahrens, T., J. P. Sleeman, C. M. Schempp, N. Howells, M. Hofmann, H. Ponta, P. Herrlich
and J. C. Simon (2001). "Soluble CD44 inhibits melanoma tumor growth by blocking
cell surface CD44 binding to hyaluronic acid." Oncogene 20(26): 3399-408.
Akiyama, Y., S. Jung, B. Salhia, S. Lee, S. Hubbard, M. Taylor, T. Mainprize, K. Akaishi,
W. van Furth and J. T. Rutka (2001). "Hyaluronate receptors mediating glioma cell
migration and proliferation." J Neurooncol 53(2): 115-27.
Albelda, S. M., C. W. Smith and P. A. Ward (1994). "Adhesion molecules and inflammatory
injury." Faseb J 8(8): 504-12.
Alhadeff, J. A. (1989). "Malignant cell glycoproteins and glycolipids." Crit Rev Oncol
Hematol 9(1): 37-107.
Alitalo, K. and P. Carmeliet (2002). "Molecular mechanisms of lymphangiogenesis in health
and disease." Cancer Cell 1(3): 219-27.
Allouche, M., R. S. Charrad, A. Bettaieb, C. Greenland, C. Grignon and F. Smadja-Joffe
(2000). "Ligation of the CD44 adhesion molecule inhibits drug-induced apoptosis in
human myeloid leukemia cells." Blood 96(3): 1187-90.
Arch, R., K. Wirth, M. Hofmann, H. Ponta, S. Matzku, P. Herrlich and M. Zoller (1992).
"Participation in normal immune responses of a metastasis-inducing splice variant of
CD44." Science 257(5070): 682-5.
Bardeesy, N. and R. A. DePinho (2002). "Pancreatic cancer biology and genetics." Nat Rev
Cancer 2(12): 897-909.
Bates, R. C., C. A. Elith, R. F. Thorne and G. F. Burns (1998). "Engagement of variant CD44
confers resistance to anti-integrin antibody-mediated apoptosis in a colon carcinoma
cell line." Cell Adhes Commun 6(1): 21-38.
Bates, R. C., N. S. Edwards, G. F. Burns and D. E. Fisher (2001). "A CD44 survival pathway
triggers chemoresistance via lyn kinase and phosphoinositide 3-kinase/Akt in colon
carcinoma cells." Cancer Res 61(13): 5275-83.
Ben-Ze'ev, A., M. Zoller and A. Raz (1986). "Differential expression of intermediate filament
proteins in metastatic and nonmetastatic variants of the BSp73 tumor." Cancer Res
46(2): 785-90.
Bendel-Stenzel, M. R., M. Gomperts, R. Anderson, J. Heasman and C. Wylie (2000). "The
role of cadherins during primordial germ cell migration and early gonad formation in
the mouse." Mech Dev 91(1-2): 143-52.
Bennett, K. L., D. G. Jackson, J. C. Simon, E. Tanczos, R. Peach, B. Modrell, I.
Stamenkovic, G. Plowman and A. Aruffo (1995). "CD44 isoforms containing exon
V3 are responsible for the presentation of heparin-binding growth factor." J Cell Biol
128(4): 687-98.
Birch, M., S. Mitchell and I. R. Hart (1991). "Isolation and characterization of human
melanoma cell variants expressing high and low levels of CD44." Cancer Res 51(24):
Birchmeier, W. and J. Behrens (1994). "Cadherin expression in carcinomas: role in the
formation of cell junctions and the prevention of invasiveness." Biochim Biophys
Acta 1198(1): 11-26.
Bouck, N., V. Stellmach and S. C. Hsu (1996). "How tumors become angiogenic." Adv
Cancer Res 69: 135-74.
Bourguignon, L. Y. and H. Jin (1995). "Identification of the ankyrin-binding domain of the
mouse T-lymphoma cell inositol 1,4,5-trisphosphate (IP3) receptor and its role in the
regulation of IP3-mediated internal Ca2+ release." J Biol Chem 270(13): 7257-60.
Bourguignon, L. Y., H. Zhu, A. Chu, N. Iida, L. Zhang and M. C. Hung (1997). "Interaction
between the adhesion receptor, CD44, and the oncogene product, p185HER2,
promotes human ovarian tumor cell activation." J Biol Chem 272(44): 27913-8.
Bourguignon, L. Y., Z. Gunja-Smith, N. Iida, H. B. Zhu, L. J. Young, W. J. Muller and R. D.
Cardiff (1998). "CD44v(3,8-10) is involved in cytoskeleton-mediated tumor cell
migration and matrix metalloproteinase (MMP-9) association in metastatic breast
cancer cells." J Cell Physiol 176(1): 206-15.
Bourguignon, L. Y., H. Zhu, L. Shao and Y. W. Chen (2000). "CD44 interaction with tiam1
promotes Rac1 signaling and hyaluronic acid-mediated breast tumor cell migration." J
Biol Chem 275(3): 1829-38.
Bourguignon, L. Y., H. Zhu, B. Zhou, F. Diedrich, P. A. Singleton and M. C. Hung (2001).
"Hyaluronan promotes CD44v3-Vav2 interaction with Grb2-p185(HER2) and
induces Rac1 and Ras signaling during ovarian tumor cell migration and growth." J
Biol Chem 276(52): 48679-92.
Bourguignon, L. Y., H. Zhu, L. Shao and Y. W. Chen (2001). "CD44 interaction with c-Src
kinase promotes cortactin-mediated cytoskeleton function and hyaluronic aciddependent ovarian tumor cell migration." J Biol Chem 276(10): 7327-36.
Brennan, F. R., K. Mikecz, T. T. Glant, P. Jobanputra, S. Pinder, C. Bavington, P. Morrison
and G. Nuki (1997). "CD44 expression by leucocytes in rheumatoid arthritis and
modulation by specific antibody: implications for lymphocyte adhesion to endothelial
cells and synoviocytes in vitro." Scand J Immunol 45(2): 213-20.
Bretscher, A., K. Edwards and R. G. Fehon (2002). "ERM proteins and merlin: integrators at
the cell cortex." Nat Rev Mol Cell Biol 3(8): 586-99.
Brummelkamp, T. R., R. Bernards and R. Agami (2002). "A system for stable expression of
short interfering RNAs in mammalian cells." Science 296(5567): 550-3.
Bryan, T. M., A. Englezou, J. Gupta, S. Bacchetti and R. R. Reddel (1995). "Telomere
elongation in immortal human cells without detectable telomerase activity." Embo J
14(17): 4240-8.
Bryan, T. M. and T. R. Cech (1999). "Telomerase and the maintenance of chromosome
ends." Curr Opin Cell Biol 11(3): 318-24.
Bull, H. A., P. M. Brickell and P. M. Dowd (1994). "Src-related protein tyrosine kinases are
physically associated with the surface antigen CD36 in human dermal microvascular
endothelial cells." FEBS Lett 351(1): 41-4.
Cantley, L. C. and B. G. Neel (1999). "New insights into tumor suppression: PTEN
suppresses tumor formation by restraining the phosphoinositide 3-kinase/AKT
pathway." Proc Natl Acad Sci U S A 96(8): 4240-5.
Chambers, A. F., A. C. Groom and I. C. MacDonald (2002). "Dissemination and growth of
cancer cells in metastatic sites." Nat Rev Cancer 2(8): 563-72.
Charrad, R. S., Z. Gadhoum, J. Qi, A. Glachant, M. Allouche, C. Jasmin, C. Chomienne and
F. Smadja-Joffe (2002). "Effects of anti-CD44 monoclonal antibodies on
differentiation and apoptosis of human myeloid leukemia cell lines." Blood 99(1):
Chin, L., J. Pomerantz and R. A. DePinho (1998). "The INK4a/ARF tumor suppressor: one
gene--two products--two pathways." Trends Biochem Sci 23(8): 291-6.
Christofori, G. and H. Semb (1999). "The role of the cell-adhesion molecule E-cadherin as a
tumour-suppressor gene." Trends Biochem Sci 24(2): 73-6.
Claas, C., S. Seiter, A. Claas, L. Savelyeva, M. Schwab and M. Zoller (1998). "Association
between the rat homologue of CO-029, a metastasis-associated tetraspanin molecule
and consumption coagulopathy." J Cell Biol 141(1): 267-80.
Cordon-Cardo, C. and C. Prives (1999). "At the crossroads of inflammation and
tumorigenesis." J Exp Med 190(10): 1367-70.
Counter, C. M., A. A. Avilion, C. E. LeFeuvre, N. G. Stewart, C. W. Greider, C. B. Harley
and S. Bacchetti (1992). "Telomere shortening associated with chromosome
instability is arrested in immortal cells which express telomerase activity." Embo J
11(5): 1921-9.
Coussens, L. M., W. W. Raymond, G. Bergers, M. Laig-Webster, O. Behrendtsen, Z. Werb,
G. H. Caughey and D. Hanahan (1999). "Inflammatory mast cells up-regulate
angiogenesis during squamous epithelial carcinogenesis." Genes Dev 13(11): 138297.
Culty, M., K. Miyake, P. W. Kincade, E. Sikorski, E. C. Butcher and C. Underhill (1990).
"The hyaluronate receptor is a member of the CD44 (H-CAM) family of cell surface
glycoproteins." J Cell Biol 111(6 Pt 1): 2765-74.
Culty, M., H. A. Nguyen and C. B. Underhill (1992). "The hyaluronan receptor (CD44)
participates in the uptake and degradation of hyaluronan." J Cell Biol 116(4): 105562.
Dall, P., K. H. Heider, A. Hekele, G. von Minckwitz, M. Kaufmann, H. Ponta and P. Herrlich
(1994). "Surface protein expression and messenger RNA-splicing analysis of CD44 in
uterine cervical cancer and normal cervical epithelium." Cancer Res 54(13): 3337-41.
Dang, C. V., L. M. Resar, E. Emison, S. Kim, Q. Li, J. E. Prescott, D. Wonsey and K. Zeller
(1999). "Function of the c-Myc oncogenic transcription factor." Exp Cell Res 253(1):
Datto, M. B., P. P. Hu, T. F. Kowalik, J. Yingling and X. F. Wang (1997). "The viral
oncoprotein E1A blocks transforming growth factor beta-mediated induction of
p21/WAF1/Cip1 and p15/INK4B." Mol Cell Biol 17(4): 2030-7.
DeGrendele, H. C., P. Estess, L. J. Picker and M. H. Siegelman (1996). "CD44 and its ligand
hyaluronate mediate rolling under physiologic flow: a novel lymphocyte-endothelial
cell primary adhesion pathway." J Exp Med 183(3): 1119-30.
DeGrendele, H. C., P. Estess and M. H. Siegelman (1997). "Requirement for CD44 in
activated T cell extravasation into an inflammatory site." Science 278(5338): 672-5.
Ding, Y. and C. E. Lawrence (2001). "Statistical prediction of single-stranded regions in
RNA secondary structure and application to predicting effective antisense target sites
and beyond." Nucleic Acids Res 29(5): 1034-46.
Ding, Y. and C. E. Lawrence (2003). "A statistical sampling algorithm for RNA secondary
structure prediction." Nucleic Acids Res 31(24): 7280-301.
Ding, Y., C. Y. Chan and C. E. Lawrence (2004). "Sfold web server for statistical folding and
rational design of nucleic acids." Nucleic Acids Res 32(Web Server issue): W135-41.
Elbashir, S. M., J. Harborth, W. Lendeckel, A. Yalcin, K. Weber and T. Tuschl (2001).
"Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian
cells." Nature 411(6836): 494-8.
English, N. M., J. F. Lesley and R. Hyman (1998). "Site-specific de-N-glycosylation of
CD44 can activate hyaluronan binding, and CD44 activation states show distinct
threshold densities for hyaluronan binding." Cancer Res 58(16): 3736-42.
Esquela-Kerscher, A. and F. J. Slack (2006). "Oncomirs - microRNAs with a role in cancer."
Nat Rev Cancer 6(4): 259-69.
Evan, G. and T. Littlewood (1998). "A matter of life and cell death." Science 281(5381):
Fire, A., S. Xu, M. K. Montgomery, S. A. Kostas, S. E. Driver and C. C. Mello (1998).
"Potent and specific genetic interference by double-stranded RNA in Caenorhabditis
elegans." Nature 391(6669): 806-11.
Fogar, P., D. Basso, C. Pasquali, M. De Paoli, C. Sperti, G. Roveroni, S. Pedrazzoli and M.
Plebani (1997). "Neural cell adhesion molecule (N-CAM) in gastrointestinal
neoplasias." Anticancer Res 17(2B): 1227-30.
Folkman, J. (1997). "Angiogenesis and angiogenesis inhibition: an overview." Exs 79: 1-8.
Foulds, L. (1954). "The experimental study of tumor progression: a review." Cancer Res
14(5): 327-39.
Fox, S. B., J. Fawcett, D. G. Jackson, I. Collins, K. C. Gatter, A. L. Harris, A. Gearing and D.
L. Simmons (1994). "Normal human tissues, in addition to some tumors, express
multiple different CD44 isoforms." Cancer Res 54(16): 4539-46.
Friedl, P. and K. Wolf (2003). "Tumour-cell invasion and migration: diversity and escape
mechanisms." Nat Rev Cancer 3(5): 362-74.
Fujimoto, T., H. Kawashima, T. Tanaka, M. Hirose, N. Toyama-Sorimachi, Y. Matsuzawa
and M. Miyasaka (2001). "CD44 binds a chondroitin sulfate proteoglycan, aggrecan."
Int Immunol 13(3): 359-66.
Fujita, N., N. Yaegashi, Y. Ide, S. Sato, M. Nakamura, I. Ishiwata and A. Yajima (1994).
"Expression of CD44 in normal human versus tumor endometrial tissues: possible
implication of reduced expression of CD44 in lymph-vascular space involvement of
cancer cells." Cancer Res 54(14): 3922-8.
Fujita, Y., M. Kitagawa, S. Nakamura, K. Azuma, G. Ishii, M. Higashi, H. Kishi, T. Hiwasa,
K. Koda, N. Nakajima, et al. (2002). "CD44 signaling through focal adhesion kinase
and its anti-apoptotic effect." FEBS Lett 528(1-3): 101-8.
Fynan, T. M. and M. Reiss (1993). "Resistance to inhibition of cell growth by transforming
growth factor-beta and its role in oncogenesis." Crit Rev Oncog 4(5): 493-540.
Gallatin, W. M., I. L. Weissman and E. C. Butcher (1983). "A cell-surface molecule involved
in organ-specific homing of lymphocytes." Nature 304(5921): 30-4.
Gao, A. C., W. Lou, J. P. Sleeman and J. T. Isaacs (1998). "Metastasis suppression by the
standard CD44 isoform does not require the binding of prostate cancer cells to
hyaluronate." Cancer Res 58(11): 2350-2.
Gautreau, A., D. Louvard and M. Arpin (2002). "ERM proteins and NF2 tumor suppressor:
the Yin and Yang of cortical actin organization and cell growth signaling." Curr Opin
Cell Biol 14(1): 104-9.
Ghaffari, S., F. Smadja-Joffe, R. Oostendorp, J. P. Levesque, G. Dougherty, A. Eaves and C.
Eaves (1999). "CD44 isoforms in normal and leukemic hematopoiesis." Exp Hematol
27(6): 978-93.
Ghatak, S., S. Misra and B. P. Toole (2002). "Hyaluronan oligosaccharides inhibit
anchorage-independent growth of tumor cells by suppressing the phosphoinositide 3kinase/Akt cell survival pathway." J Biol Chem 277(41): 38013-20.
Ghatak, S., S. Misra and B. P. Toole (2005). "Hyaluronan constitutively regulates ErbB2
phosphorylation and signaling complex formation in carcinoma cells." J Biol Chem
280(10): 8875-83.
Glinsky, V. V., G. V. Glinsky, O. V. Glinskii, V. H. Huxley, J. R. Turk, V. V. Mossine, S. L.
Deutscher, K. J. Pienta and T. P. Quinn (2003). "Intravascular metastatic cancer cell
homotypic aggregation at the sites of primary attachment to the endothelium." Cancer
Res 63(13): 3805-11.
Goldstein, L. A. and E. C. Butcher (1990). "Identification of mRNA that encodes an
alternative form of H-CAM(CD44) in lymphoid and nonlymphoid tissues."
Immunogenetics 32(6): 389-97.
Goodrich, D. W. (2006). "The retinoblastoma tumor-suppressor gene, the exception that
proves the rule." Oncogene 25(38): 5233-43.
Graham, F. L., J. Smiley, W. C. Russell and R. Nairn (1977). "Characteristics of a human cell
line transformed by DNA from human adenovirus type 5." J Gen Virol 36(1): 59-74.
Green, D. R. and J. C. Reed (1998). "Mitochondria and apoptosis." Science 281(5381): 130912.
Gress, T. M., F. Muller-Pillasch, M. M. Lerch, H. Friess, M. Buchler and G. Adler (1995).
"Expression and in-situ localization of genes coding for extracellular matrix proteins
and extracellular matrix degrading proteases in pancreatic cancer." Int J Cancer 62(4):
Griffioen, A. W., M. J. Coenen, C. A. Damen, S. M. Hellwig, D. H. van Weering, W. Vooys,
G. H. Blijham and G. Groenewegen (1997). "CD44 is involved in tumor
angiogenesis; an activation antigen on human endothelial cells." Blood 90(3): 1150-9.
Gunthert, U., M. Hofmann, W. Rudy, S. Reber, M. Zoller, I. Haussmann, S. Matzku, A.
Wenzel, H. Ponta and P. Herrlich (1991). "A new variant of glycoprotein CD44
confers metastatic potential to rat carcinoma cells." Cell 65(1): 13-24.
Guy, R., E. Yefenof, D. Naor, A. Dorogin and Y. Zilberman (2002). "CD44 co-stimulates
apoptosis in thymic lymphomas and T cell hybridomas." Cell Immunol 216(1-2): 8292.
Hanahan, D. and J. Folkman (1996). "Patterns and emerging mechanisms of the angiogenic
switch during tumorigenesis." Cell 86(3): 353-64.
Hanahan, D. and R. A. Weinberg (2000). "The hallmarks of cancer." Cell 100(1): 57-70.
Hannon, G. J. and D. Beach (1994). "p15INK4B is a potential effector of TGF-beta-induced
cell cycle arrest." Nature 371(6494): 257-61.
Hannon, G. J. (2002). "RNA interference." Nature 418(6894): 244-51.
Harris, C. C. (1996). "p53 tumor suppressor gene: from the basic research laboratory to the
clinic--an abridged historical perspective." Carcinogenesis 17(6): 1187-98.
Hatta, K. and M. Takeichi (1986). "Expression of N-cadherin adhesion molecules associated
with early morphogenetic events in chick development." Nature 320(6061): 447-9.
Hedman, K., M. Kurkinen, K. Alitalo, A. Vaheri, S. Johansson and M. Hook (1979).
"Isolation of the pericellular matrix of human fibroblast cultures." J Cell Biol 81(1):
Herlevsen, M., D. S. Schmidt, K. Miyazaki and M. Zoller (2003). "The association of the
tetraspanin D6.1A with the alpha6beta4 integrin supports cell motility and liver
metastasis formation." J Cell Sci 116(Pt 21): 4373-90.
Herrlich, P., J. Sleeman, D. Wainwright, H. Konig, L. Sherman, F. Hilberg and H. Ponta
(1998). "How tumor cells make use of CD44." Cell Adhes Commun 6(2-3): 141-7.
Hirano, H., G. R. Screaton, M. V. Bell, D. G. Jackson, J. I. Bell and R. J. Hodes (1994).
"CD44 isoform expression mediated by alternative splicing: tissue-specific regulation
in mice." Int Immunol 6(1): 49-59.
Hodkinson, P. S., T. Elliott, W. S. Wong, R. C. Rintoul, A. C. Mackinnon, C. Haslett and T.
Sethi (2006). "ECM overrides DNA damage-induced cell cycle arrest and apoptosis in
small-cell lung cancer cells through beta1 integrin-dependent activation of PI3kinase." Cell Death Differ 13(10): 1776-88.
Hua, Q., C. B. Knudson and W. Knudson (1993). "Internalization of hyaluronan by
chondrocytes occurs via receptor-mediated endocytosis." J Cell Sci 106 ( Pt 1): 36575.
Hudson, J. D., M. A. Shoaibi, R. Maestro, A. Carnero, G. J. Hannon and D. H. Beach (1999).
"A proinflammatory cytokine inhibits p53 tumor suppressor activity." J Exp Med
190(10): 1375-82.
Hynes, R. O. (2002). "Integrins: bidirectional, allosteric signaling machines." Cell 110(6):
Igney, F. H. and P. H. Krammer (2002). "Death and anti-death: tumour resistance to
apoptosis." Nat Rev Cancer 2(4): 277-88.
Ioannidis, P., T. Trangas, E. Dimitriadis, M. Samiotaki, I. Kyriazoglou, C. M. Tsiapalis, C.
Kittas, N. Agnantis, F. C. Nielsen, J. Nielsen, et al. (2001). "C-MYC and IGF-II
mRNA-binding protein (CRD-BP/IMP-1) in benign and malignant mesenchymal
tumors." Int J Cancer 94(4): 480-4.
Isnard, N., L. Robert and G. Renard (2003). "Effect of sulfated GAGs on the expression and
activation of MMP-2 and MMP-9 in corneal and dermal explant cultures." Cell Biol
Int 27(9): 779-84.
Jakobisiak, M., W. Lasek and J. Golab (2003). "Natural mechanisms protecting against
cancer." Immunol Lett 90(2-3): 103-22.
Jiang, H., R. S. Peterson, W. Wang, E. Bartnik, C. B. Knudson and W. Knudson (2002). "A
requirement for the CD44 cytoplasmic domain for hyaluronan binding, pericellular
matrix assembly, and receptor-mediated endocytosis in COS-7 cells." J Biol Chem
277(12): 10531-8.
Jin, L., K. J. Hope, Q. Zhai, F. Smadja-Joffe and J. E. Dick (2006). "Targeting of CD44
eradicates human acute myeloid leukemic stem cells." Nat Med 12(10): 1167-74.
Johnson, J. P. (1991). "Cell adhesion molecules of the immunoglobulin supergene family and
their role in malignant transformation and progression to metastatic disease." Cancer
Metastasis Rev 10(1): 11-22.
Jones, M., L. Tussey, N. Athanasou and D. G. Jackson (2000). "Heparan sulfate proteoglycan
isoforms of the CD44 hyaluronan receptor induced in human inflammatory
macrophages can function as paracrine regulators of fibroblast growth factor action."
J Biol Chem 275(11): 7964-74.
Kaiser, U., B. Auerbach and M. Oldenburg (1996). "The neural cell adhesion molecule
NCAM in multiple myeloma." Leuk Lymphoma 20(5-6): 389-95.
Kalnina, Z., P. Zayakin, K. Silina and A. Line (2005). "Alterations of pre-mRNA splicing in
cancer." Genes Chromosomes Cancer 42(4): 342-57.
Katagiri, Y. U., J. Sleeman, H. Fujii, P. Herrlich, H. Hotta, K. Tanaka, S. Chikuma, H.
Yagita, K. Okumura, M. Murakami, et al. (1999). "CD44 variants but not CD44s
cooperate with beta1-containing integrins to permit cells to bind to osteopontin
independently of arginine-glycine-aspartic acid, thereby stimulating cell motility and
chemotaxis." Cancer Res 59(1): 219-26.
Kawashima, H., M. Hirose, J. Hirose, D. Nagakubo, A. H. Plaas and M. Miyasaka (2000).
"Binding of a large chondroitin sulfate/dermatan sulfate proteoglycan, versican, to Lselectin, P-selectin, and CD44." J Biol Chem 275(45): 35448-56.
Kennel, S. J., T. K. Lankford, L. J. Foote, S. G. Shinpock and C. Stringer (1993). "CD44
expression on murine tissues." J Cell Sci 104 ( Pt 2): 373-82.
Khan, S. A., C. A. Lopez-Chua, J. Zhang, L. W. Fisher, E. S. Sorensen and D. T. Denhardt
(2002). "Soluble osteopontin inhibits apoptosis of adherent endothelial cells deprived
of growth factors." J Cell Biochem 85(4): 728-36.
Knudson, W. and C. B. Knudson (1991). "Assembly of a chondrocyte-like pericellular matrix
on non-chondrogenic cells. Role of the cell surface hyaluronan receptors in the
assembly of a pericellular matrix." J Cell Sci 99 ( Pt 2): 227-35.
Knudson, C. B. and W. Knudson (1993). "Hyaluronan-binding proteins in development,
tissue homeostasis, and disease." Faseb J 7(13): 1233-41.
Knudson, W. (1996). "Tumor-associated hyaluronan. Providing an extracellular matrix that
facilitates invasion." Am J Pathol 148(6): 1721-6.
Knudson, W., D. J. Aguiar, Q. Hua and C. B. Knudson (1996). "CD44-anchored hyaluronan89
rich pericellular matrices: an ultrastructural and biochemical analysis." Exp Cell Res
228(2): 216-28.
Knudson, W., B. Casey, Y. Nishida, W. Eger, K. E. Kuettner and C. B. Knudson (2000).
"Hyaluronan oligosaccharides perturb cartilage matrix homeostasis and induce
chondrocytic chondrolysis." Arthritis Rheum 43(5): 1165-74.
Knudson, C. B. (2003). "Hyaluronan and CD44: strategic players for cell-matrix interactions
during chondrogenesis and matrix assembly." Birth Defects Res C Embryo Today
69(2): 174-96.
Knutson, J. R., J. Iida, G. B. Fields and J. B. McCarthy (1996). "CD44/chondroitin sulfate
proteoglycan and alpha 2 beta 1 integrin mediate human melanoma cell migration on
type IV collagen and invasion of basement membranes." Mol Biol Cell 7(3): 383-96.
Ladwein, M., U. F. Pape, D. S. Schmidt, M. Schnolzer, S. Fiedler, L. Langbein, W. W.
Franke, G. Moldenhauer and M. Zoller (2005). "The cell-cell adhesion molecule
EpCAM interacts directly with the tight junction protein claudin-7." Exp Cell Res
309(2): 345-57.
LeBedis, C., K. Chen, L. Fallavollita, T. Boutros and P. Brodt (2002). "Peripheral lymph
node stromal cells can promote growth and tumorigenicity of breast carcinoma cells
through the release of IGF-I and EGF." Int J Cancer 100(1): 2-8.
Lee, J. Y. and A. P. Spicer (2000). "Hyaluronan: a multifunctional, megaDalton, stealth
molecule." Curr Opin Cell Biol 12(5): 581-6.
Legg, J. W., C. A. Lewis, M. Parsons, T. Ng and C. M. Isacke (2002). "A novel PKCregulated mechanism controls CD44 ezrin association and directional cell motility."
Nat Cell Biol 4(6): 399-407.
Lesley, J. and R. Hyman (1998). "CD44 structure and function." Front Biosci 3: d616-30.
Lewis, J. M., T. N. Truong and M. A. Schwartz (2002). "Integrins regulate the apoptotic
response to DNA damage through modulation of p53." Proc Natl Acad Sci U S A
99(6): 3627-32.
Li, G. and M. Herlyn (2000). "Dynamics of intercellular communication during melanoma
development." Mol Med Today 6(4): 163-9.
Liotta, L. A., D. Vembu, R. K. Saini and C. Boone (1978). "In vivo monitoring of the death
rate of artificial murine pulmonary micrometastases." Cancer Res 38(5): 1231-6.
Lippman, Z. and R. Martienssen (2004). "The role of RNA interference in heterochromatic
silencing." Nature 431(7006): 364-70.
Liu, D. and M. S. Sy (1997). "Phorbol myristate acetate stimulates the dimerization of CD44
involving a cysteine in the transmembrane domain." J Immunol 159(6): 2702-11.
Lohr, M., B. Trautmann, M. Gottler, S. Peters, I. Zauner, B. Maillet and G. Kloppel (1994).
"Human ductal adenocarcinomas of the pancreas express extracellular matrix
proteins." Br J Cancer 69(1): 144-51.
Lukashev, M. E. and Z. Werb (1998). "ECM signalling: orchestrating cell behaviour and
misbehaviour." Trends Cell Biol 8(11): 437-41.
Luzzi, K. J., I. C. MacDonald, E. E. Schmidt, N. Kerkvliet, V. L. Morris, A. F. Chambers and
A. C. Groom (1998). "Multistep nature of metastatic inefficiency: dormancy of
solitary cells after successful extravasation and limited survival of early
micrometastases." Am J Pathol 153(3): 865-73.
Mansury, Y., M. Kimura, J. Lobo and T. S. Deisboeck (2002). "Emerging patterns in tumor
systems: simulating the dynamics of multicellular clusters with an agent-based spatial
agglomeration model." J Theor Biol 219(3): 343-70.
Marhaba, R., M. Bourouba and M. Zoller (2003). "CD44v7 interferes with activation-induced
cell death by up-regulation of anti-apoptotic gene expression." J Leukoc Biol 74(1):
Marhaba, R. and M. Zoller (2004). "CD44 in cancer progression: adhesion, migration and
growth regulation." J Mol Histol 35(3): 211-31.
Marhaba, R., M. Bourouba and M. Zoller (2005). "CD44v6 promotes proliferation by
persisting activation of MAP kinases." Cell Signal 17(8): 961-73.
Markowitz, S., J. Wang, L. Myeroff, R. Parsons, L. Sun, J. Lutterbaugh, R. S. Fan, E.
Zborowska, K. W. Kinzler, B. Vogelstein, et al. (1995). "Inactivation of the type II
TGF-beta receptor in colon cancer cells with microsatellite instability." Science
268(5215): 1336-8.
Martin, S. S. and K. Vuori (2004). "Regulation of Bcl-2 proteins during anoikis and
amorphosis." Biochim Biophys Acta 1692(2-3): 145-57.
Maschler, S., G. Wirl, H. Spring, D. V. Bredow, I. Sordat, H. Beug and E. Reichmann
(2005). "Tumor cell invasiveness correlates with changes in integrin expression and
localization." Oncogene 24(12): 2032-41.
Matsumura, Y. and D. Tarin (1992). "Significance of CD44 gene products for cancer
diagnosis and disease evaluation." Lancet 340(8827): 1053-8.
Matzku, S., D. Komitowski, M. Mildenberger and M. Zoller (1983). "Characterization of
BSp73, a spontaneous rat tumor and its in vivo selected variants showing different
metastasizing capacities." Invasion Metastasis 3(2): 109-23.
Matzku, S., H. O. Werling, C. Waller, B. Schmalenberger and H. Zankl (1985). "Clonal
analysis of diversity in the BSp73 rat tumor." Invasion Metastasis 5(6): 356-70.
Matzku, S., A. Wenzel, S. Liu and M. Zoller (1989). "Antigenic differences between
metastatic and nonmetastatic BSp73 rat tumor variants characterized by monoclonal
antibodies." Cancer Res 49(5): 1294-9.
Medema, R. H. and J. L. Bos (1993). "The role of p21ras in receptor tyrosine kinase
signaling." Crit Rev Oncog 4(6): 615-61.
Mehlen, P. and A. Puisieux (2006). "Metastasis: a question of life or death." Nat Rev Cancer
6(6): 449-58.
Meister, G. and T. Tuschl (2004). "Mechanisms of gene silencing by double-stranded RNA."
Nature 431(7006): 343-9.
Menzel, E. J. and C. Farr (1998). "Hyaluronidase and its substrate hyaluronan: biochemistry,
biological activities and therapeutic uses." Cancer Lett 131(1): 3-11.
Mielgo, A., M. van Driel, A. Bloem, L. Landmann and U. Gunthert (2006). "A novel
antiapoptotic mechanism based on interference of Fas signaling by CD44 variant
isoforms." Cell Death Differ 13(3): 465-77.
Misra, S., S. Ghatak and B. P. Toole (2005). "Regulation of MDR1 expression and drug
resistance by a positive feedback loop involving hyaluronan, phosphoinositide 3kinase, and ErbB2." J Biol Chem 280(21): 20310-5.
Morrison, H., L. S. Sherman, J. Legg, F. Banine, C. Isacke, C. A. Haipek, D. H. Gutmann, H.
Ponta and P. Herrlich (2001). "The NF2 tumor suppressor gene product, merlin,
mediates contact inhibition of growth through interactions with CD44." Genes Dev
15(8): 968-80.
Muller, A., B. Homey, H. Soto, N. Ge, D. Catron, M. E. Buchanan, T. McClanahan, E.
Murphy, W. Yuan, S. N. Wagner, et al. (2001). "Involvement of chemokine receptors
in breast cancer metastasis." Nature 410(6824): 50-6.
Murray, D., M. Morrin and S. McDonnell (2004). "Increased invasion and expression of
MMP-9 in human colorectal cell lines by a CD44-dependent mechanism." Anticancer
Res 24(2A): 489-94.
Nagano, O. and H. Saya (2004). "Mechanism and biological significance of CD44 cleavage."
Cancer Sci 95(12): 930-5.
Naor, D., R. V. Sionov and D. Ish-Shalom (1997). "CD44: structure, function, and
association with the malignant process." Adv Cancer Res 71: 241-319.
Naor, D., S. Nedvetzki, I. Golan, L. Melnik and Y. Faitelson (2002). "CD44 in cancer." Crit
Rev Clin Lab Sci 39(6): 527-79.
Nasreen, N., K. A. Mohammed, J. Hardwick, R. D. Van Horn, K. Sanders, H. Kathuria, F.
Loghmani and V. B. Antony (2002). "Low molecular weight hyaluronan induces
malignant mesothelioma cell (MMC) proliferation and haptotaxis: role of CD44
receptor in MMC proliferation and haptotaxis." Oncol Res 13(2): 71-8.
Neame, S. J., C. R. Uff, H. Sheikh, S. C. Wheatley and C. M. Isacke (1995). "CD44 exhibits
a cell type dependent interaction with triton X-100 insoluble, lipid rich, plasma
membrane domains." J Cell Sci 108 ( Pt 9): 3127-35.
Nestl, A., O. D. Von Stein, K. Zatloukal, W. G. Thies, P. Herrlich, M. Hofmann and J. P.
Sleeman (2001). "Gene expression patterns associated with the metastatic phenotype
in rodent and human tumors." Cancer Res 61(4): 1569-77.
Nikiforov, M. A., K. Hagen, V. S. Ossovskaya, T. M. Connor, S. W. Lowe, G. I. Deichman
and A. V. Gudkov (1996). "p53 modulation of anchorage independent growth and
experimental metastasis." Oncogene 13(8): 1709-19.
Nomura, M., T. Matsunami, K. Kobayashi, T. Uchibayashi, K. Koshida, M. Tanaka, M.
Namiki, Y. Mizuhara, T. Akiba, K. Yokogawa, et al. (2005). "Involvement of ABC
transporters in chemosensitivity of human renal cell carcinoma, and regulation of
MRP2 expression by conjugated bilirubin." Anticancer Res 25(4): 2729-35.
Nowell, P. C. (1976). "The clonal evolution of tumor cell populations." Science 194(4260):
O'Rourke, J. R., M. S. Swanson and B. D. Harfe (2006). "MicroRNAs in mammalian
development and tumorigenesis." Birth Defects Res C Embryo Today 78(2): 172-9.
Oguri, T., T. Isobe, T. Suzuki, K. Nishio, Y. Fujiwara, O. Katoh and M. Yamakido (2000).
"Increased expression of the MRP5 gene is associated with exposure to platinum
drugs in lung cancer." Int J Cancer 86(1): 95-100.
Okada, H., J. Yoshida, M. Sokabe, T. Wakabayashi and M. Hagiwara (1996). "Suppression
of CD44 expression decreases migration and invasion of human glioma cells." Int J
Cancer 66(2): 255-60.
Okamoto, I., Y. Kawano, M. Matsumoto, M. Suga, K. Kaibuchi, M. Ando and H. Saya
(1999). "Regulated CD44 cleavage under the control of protein kinase C, calcium
influx, and the Rho family of small G proteins." J Biol Chem 274(36): 25525-34.
Okamoto, I., Y. Kawano, H. Tsuiki, J. Sasaki, M. Nakao, M. Matsumoto, M. Suga, M. Ando,
M. Nakajima and H. Saya (1999). "CD44 cleavage induced by a membraneassociated metalloprotease plays a critical role in tumor cell migration." Oncogene
18(7): 1435-46.
Okamoto, I., Y. Kawano, D. Murakami, T. Sasayama, N. Araki, T. Miki, A. J. Wong and H.
Saya (2001). "Proteolytic release of CD44 intracellular domain and its role in the
CD44 signaling pathway." J Cell Biol 155(5): 755-62.
Oliferenko, S., I. Kaverina, J. V. Small and L. A. Huber (2000). "Hyaluronic acid (HA)
binding to CD44 activates Rac1 and induces lamellipodia outgrowth." J Cell Biol
148(6): 1159-64.
Orian-Rousseau, V., L. Chen, J. P. Sleeman, P. Herrlich and H. Ponta (2002). "CD44 is
required for two consecutive steps in HGF/c-Met signaling." Genes Dev 16(23):
Paddison, P. J., A. A. Caudy, E. Bernstein, G. J. Hannon and D. S. Conklin (2002). "Short
hairpin RNAs (shRNAs) induce sequence-specific silencing in mammalian cells."
Genes Dev 16(8): 948-58.
Pals, S. T., P. Drillenburg, T. Radaszkiewicz and E. Manten-Horst (1997). "Adhesion
molecules in the dissemination of non-Hodgkin's lymphomas." Acta Haematol 97(12): 73-80.
Pantel, K. and R. H. Brakenhoff (2004). "Dissecting the metastatic cascade." Nat Rev Cancer
4(6): 448-56.
Paterson, D. J., W. A. Jefferies, J. R. Green, M. R. Brandon, P. Corthesy, M. Puklavec and A.
F. Williams (1987). "Antigens of activated rat T lymphocytes including a molecule of
50,000 Mr detected only on CD4 positive T blasts." Mol Immunol 24(12): 1281-90.
Pepper, M. S. (1997). "Transforming growth factor-beta: vasculogenesis, angiogenesis, and
vessel wall integrity." Cytokine Growth Factor Rev 8(1): 21-43.
Perschl, A., J. Lesley, N. English, R. Hyman and I. S. Trowbridge (1995). "Transmembrane
domain of CD44 is required for its detergent insolubility in fibroblasts." J Cell Sci
108 ( Pt 3): 1033-41.
Peterson, R. M., Q. Yu, I. Stamenkovic and B. P. Toole (2000). "Perturbation of hyaluronan
interactions by soluble CD44 inhibits growth of murine mammary carcinoma cells in
ascites." Am J Pathol 156(6): 2159-67.
Pinkse, G. G., W. P. Bouwman, R. Jiawan-Lalai, O. T. Terpstra, J. A. Bruijn and E. de Heer
(2006). "Integrin signaling via RGD peptides and anti-beta1 antibodies confers
resistance to apoptosis in islets of Langerhans." Diabetes 55(2): 312-7.
Pohl, M., H. Sakurai, R. O. Stuart and S. K. Nigam (2000). "Role of hyaluronan and CD44 in
in vitro branching morphogenesis of ureteric bud cells." Dev Biol 224(2): 312-25.
Ponta, H., D. Wainwright and P. Herrlich (1998). "The CD44 protein family." Int J Biochem
Cell Biol 30(3): 299-305.
Ponta, H., L. Sherman and P. A. Herrlich (2003). "CD44: from adhesion molecules to
signalling regulators." Nat Rev Mol Cell Biol 4(1): 33-45.
Protin, U., T. Schweighoffer, W. Jochum and F. Hilberg (1999). "CD44-deficient mice
develop normally with changes in subpopulations and recirculation of lymphocyte
subsets." J Immunol 163(9): 4917-23.
Pure, E. and C. A. Cuff (2001). "A crucial role for CD44 in inflammation." Trends Mol Med
7(5): 213-21.
Raz, A., M. Zoller and Z. e. Ben (1986). "Cell configuration and adhesive properties of
metastasizing and non-metastasizing BSp73 rat adenocarcinoma cells." Exp Cell Res
162(1): 127-41.
Recio, J. A. and G. Merlino (2003). "Hepatocyte growth factor/scatter factor induces
feedback up-regulation of CD44v6 in melanoma cells through Egr-1." Cancer Res
63(7): 1576-82.
Rosel, M., C. Claas, S. Seiter, M. Herlevsen and M. Zoller (1998). "Cloning and functional
characterization of a new phosphatidyl-inositol anchored molecule of a metastasizing
rat pancreatic tumor." Oncogene 17(15): 1989-2002.
Rudy, W., M. Hofmann, R. Schwartz-Albiez, M. Zoller, K. H. Heider, H. Ponta and P.
Herrlich (1993). "The two major CD44 proteins expressed on a metastatic rat tumor
cell line are derived from different splice variants: each one individually suffices to
confer metastatic behavior." Cancer Res 53(6): 1262-8.
Schedin, P. and A. Elias (2004). "Multistep tumorigenesis and the microenvironment." Breast
Cancer Res 6(2): 93-101.
Schmidt, D. S., P. Klingbeil, M. Schnolzer and M. Zoller (2004). "CD44 variant isoforms
associate with tetraspanins and EpCAM." Exp Cell Res 297(2): 329-47.
Schmits, R., J. Filmus, N. Gerwin, G. Senaldi, F. Kiefer, T. Kundig, A. Wakeham, A.
Shahinian, C. Catzavelos, J. Rak, et al. (1997). "CD44 regulates hematopoietic
progenitor distribution, granuloma formation, and tumorigenicity." Blood 90(6):
Schutte, M., R. H. Hruban, L. Hedrick, K. R. Cho, G. M. Nadasdy, C. L. Weinstein, G. S.
Bova, W. B. Isaacs, P. Cairns, H. Nawroz, et al. (1996). "DPC4 gene in various tumor
types." Cancer Res 56(11): 2527-30.
Schwartz, M. A. (1993). "Signaling by integrins: implications for tumorigenesis." Cancer Res
53(7): 1503-6.
Screaton, G. R., M. V. Bell, D. G. Jackson, F. B. Cornelis, U. Gerth and J. I. Bell (1992).
"Genomic structure of DNA encoding the lymphocyte homing receptor CD44 reveals
at least 12 alternatively spliced exons." Proc Natl Acad Sci U S A 89(24): 12160-4.
Seiter, S., R. Arch, S. Reber, D. Komitowski, M. Hofmann, H. Ponta, P. Herrlich, S. Matzku
and M. Zoller (1993). "Prevention of tumor metastasis formation by anti-variant
CD44." J Exp Med 177(2): 443-55.
Seiter, S., P. Engel, N. Fohr and M. Zoller (1999). "Mitigation of delayed-type
hypersensitivity reactions by a CD44 variant isoform v3-specific antibody: blockade
of leukocyte egress." J Invest Dermatol 113(1): 11-21.
Sherman, L., D. Wainwright, H. Ponta and P. Herrlich (1998). "A splice variant of CD44
expressed in the apical ectodermal ridge presents fibroblast growth factors to limb
mesenchyme and is required for limb outgrowth." Genes Dev 12(7): 1058-71.
Sherman, L. S., T. A. Rizvi, S. Karyala and N. Ratner (2000). "CD44 enhances neuregulin
signaling by Schwann cells." J Cell Biol 150(5): 1071-84.
Shimaoka, M. and T. A. Springer (2003). "Therapeutic antagonists and conformational
regulation of integrin function." Nat Rev Drug Discov 2(9): 703-16.
Singh, R. K., M. Gutman, C. D. Bucana, R. Sanchez, N. Llansa and I. J. Fidler (1995).
"Interferons alpha and beta down-regulate the expression of basic fibroblast growth
factor in human carcinomas." Proc Natl Acad Sci U S A 92(10): 4562-6.
Skelton, T. P., C. Zeng, A. Nocks and I. Stamenkovic (1998). "Glycosylation provides both
stimulatory and inhibitory effects on cell surface and soluble CD44 binding to
hyaluronan." J Cell Biol 140(2): 431-46.
Skobe, M. and N. E. Fusenig (1998). "Tumorigenic conversion of immortal human
keratinocytes through stromal cell activation." Proc Natl Acad Sci U S A 95(3): 10505.
Sleeman, J. P., S. Arming, J. F. Moll, A. Hekele, W. Rudy, L. S. Sherman, G. Kreil, H. Ponta
and P. Herrlich (1996). "Hyaluronate-independent metastatic behavior of CD44
variant-expressing pancreatic carcinoma cells." Cancer Res 56(13): 3134-41.
Sleeman, J., W. Rudy, M. Hofmann, J. Moll, P. Herrlich and H. Ponta (1996). "Regulated
clustering of variant CD44 proteins increases their hyaluronate binding capacity." J
Cell Biol 135(4): 1139-50.
Sleeman, J. P. (2000). "The lymph node as a bridgehead in the metastatic dissemination of
tumors." Recent Results Cancer Res 157: 55-81.
Sporn, M. B. (1996). "The war on cancer." Lancet 347(9012): 1377-81.
Sretavan, D. W., L. Feng, E. Pure and L. F. Reichardt (1994). "Embryonic neurons of the
developing optic chiasm express L1 and CD44, cell surface molecules with opposing
effects on retinal axon growth." Neuron 12(5): 957-75.
Stoop, R., H. Kotani, J. D. McNeish, I. G. Otterness and K. Mikecz (2001). "Increased
resistance to collagen-induced arthritis in CD44-deficient DBA/1 mice." Arthritis
Rheum 44(12): 2922-31.
Streuli, C. H. and A. P. Gilmore (1999). "Adhesion-mediated signaling in the regulation of
mammary epithelial cell survival." J Mammary Gland Biol Neoplasia 4(2): 183-91.
Suzuki, T., K. Nishio and S. Tanabe (2001). "The MRP family and anticancer drug
metabolism." Curr Drug Metab 2(4): 367-77.
Takazoe, K., G. H. Tesch, P. A. Hill, L. A. Hurst, Z. Jun, H. Y. Lan, R. C. Atkins and D. J.
Nikolic-Paterson (2000). "CD44-mediated neutrophil apoptosis in the rat." Kidney Int
58(5): 1920-30.
Tarbe, N., S. Losch, H. Burtscher, M. Jarsch and U. H. Weidle (2002). "Identification of rat
pancreatic carcinoma genes associated with lymphogenous metastasis." Anticancer
Res 22(4): 2015-27.
Terpe, H. J., H. Stark, P. Prehm and U. Gunthert (1994). "CD44 variant isoforms are
preferentially expressed in basal epithelial of non-malignant human fetal and adult
tissues." Histochemistry 101(2): 79-89.
Tessier, C. R., G. A. Doyle, B. A. Clark, H. C. Pitot and J. Ross (2004). "Mammary tumor
induction in transgenic mice expressing an RNA-binding protein." Cancer Res 64(1):
Thornberry, N. A. and Y. Lazebnik (1998). "Caspases: enemies within." Science 281(5381):
Tolg, C., M. Hofmann, P. Herrlich and H. Ponta (1993). "Splicing choice from ten variant
exons establishes CD44 variability." Nucleic Acids Res 21(5): 1225-9.
Tomita, K., A. van Bokhoven, G. J. van Leenders, E. T. Ruijter, C. F. Jansen, M. J.
Bussemakers and J. A. Schalken (2000). "Cadherin switching in human prostate
cancer progression." Cancer Res 60(13): 3650-4.
Toole, B. P. (2004). "Hyaluronan: from extracellular glue to pericellular cue." Nat Rev
Cancer 4(7): 528-39.
Towbin, H., T. Staehelin and J. Gordon (1979). "Electrophoretic transfer of proteins from
polyacrylamide gels to nitrocellulose sheets: procedure and some applications." Proc
Natl Acad Sci U S A 76(9): 4350-4.
Toyama-Sorimachi, N., H. Sorimachi, Y. Tobita, F. Kitamura, H. Yagita, K. Suzuki and M.
Miyasaka (1995). "A novel ligand for CD44 is serglycin, a hematopoietic cell lineagespecific proteoglycan. Possible involvement in lymphoid cell adherence and
activation." J Biol Chem 270(13): 7437-44.
Tsujimura, S., K. Saito, K. Kohno and Y. Tanaka (2006). "Fragmented hyaluronan induces
transcriptional up-regulation of the multidrug resistance-1 gene in CD4+ T cells." J
Biol Chem.
Tsukita, S., K. Oishi, N. Sato, J. Sagara and A. Kawai (1994). "ERM family members as
molecular linkers between the cell surface glycoprotein CD44 and actin-based
cytoskeletons." J Cell Biol 126(2): 391-401.
Turley, E. and D. Moore (1984). "Hyaluronate binding proteins also bind to fibronectin,
laminin and collagen." Biochem Biophys Res Commun 121(3): 808-14.
Turley, E. A. (1989). "The role of a cell-associated hyaluronan-binding protein in fibroblast
behaviour." Ciba Found Symp 143: 121-33; discussion 133-7, 281-5.
van der Voort, R., T. E. Taher, V. J. Wielenga, M. Spaargaren, R. Prevo, L. Smit, G. David,
G. Hartmann, E. Gherardi and S. T. Pals (1999). "Heparan sulfate-modified CD44
promotes hepatocyte growth factor/scatter factor-induced signal transduction through
the receptor tyrosine kinase c-Met." J Biol Chem 274(10): 6499-506.
Varani, J., E. J. Lovett, S. Elgebaly, J. Lundy and P. A. Ward (1980). "In vitro and in vivo
adherence of tumor cell variants correlated with tumor formation." Am J Pathol
101(2): 345-52.
Varner, J. A. and D. A. Cheresh (1996). "Integrins and cancer." Curr Opin Cell Biol 8(5):
Veikkola, T. and K. Alitalo (1999). "VEGFs, receptors and angiogenesis." Semin Cancer
Biol 9(3): 211-20.
Vikesaa, J., T. V. Hansen, L. Jonson, R. Borup, U. M. Wewer, J. Christiansen and F. C.
Nielsen (2006). "RNA-binding IMPs promote cell adhesion and invadopodia
formation." Embo J 25(7): 1456-68.
Vleminckx, K., L. Vakaet, Jr., M. Mareel, W. Fiers and F. van Roy (1991). "Genetic
manipulation of E-cadherin expression by epithelial tumor cells reveals an invasion
suppressor role." Cell 66(1): 107-19.
Vogelmann, R., E. D. Kreuser, G. Adler and M. P. Lutz (1999). "Integrin alpha6beta1 role in
metastatic behavior of human pancreatic carcinoma cells." Int J Cancer 80(5): 791-5.
Volpert, O. V., K. M. Dameron and N. Bouck (1997). "Sequential development of an
angiogenic phenotype by human fibroblasts progressing to tumorigenicity." Oncogene
14(12): 1495-502.
Wallach-Dayan, S. B., V. Grabovsky, J. Moll, J. Sleeman, P. Herrlich, R. Alon and D. Naor
(2001). "CD44-dependent lymphoma cell dissemination: a cell surface CD44 variant,
rather than standard CD44, supports in vitro lymphoma cell rolling on hyaluronic acid
substrate and its in vivo accumulation in the peripheral lymph nodes." J Cell Sci
114(Pt 19): 3463-77.
Weber, G. F., S. Ashkar, M. J. Glimcher and H. Cantor (1996). "Receptor-ligand interaction
between CD44 and osteopontin (Eta-1)." Science 271(5248): 509-12.
Weber, B., M. Rosel, R. Arch, P. Moller and M. Zoller (1996). "Transient expression of
CD44 variant isoforms in the ontogeny of the rat: ectoderm-, endoderm- and
mesoderm-derived cells express different exon combinations." Differentiation 60(1):
Weber, G. F., S. Ashkar, M. J. Glimcher and H. Cantor (1996). "Receptor-ligand interaction
between CD44 and osteopontin (Eta-1)." Science 271(5248): 509-12.
Weber, G. F., R. T. Bronson, J. Ilagan, H. Cantor, R. Schmits and T. W. Mak (2002).
"Absence of the CD44 gene prevents sarcoma metastasis." Cancer Res 62(8): 2281-6.
Weinberg, R. A. (1995). "The retinoblastoma protein and cell cycle control." Cell 81(3): 32330.
Weiss, L. (1993). "Deformation-driven destruction of cancer cells in the microvasculature."
Clin Exp Metastasis 11(5): 430-6.
Werb, Z. (1997). "ECM and cell surface proteolysis: regulating cellular ecology." Cell 91(4):
Weth, R. (2000). “Vakzinierungsstrategien für die Therapie von Tumoren,” Dissertation,
University of Karlsruhe.
Whitelock, J. M., A. D. Murdoch, R. V. Iozzo and P. A. Underwood (1996). "The
degradation of human endothelial cell-derived perlecan and release of bound basic
fibroblast growth factor by stromelysin, collagenase, plasmin, and heparanases." J
Biol Chem 271(17): 10079-86.
WHO (2003). World Cancer Report. Lyon, IARC Press.
Wielenga, V. J., K. H. Heider, G. J. Offerhaus, G. R. Adolf, F. M. van den Berg, H. Ponta, P.
Herrlich and S. T. Pals (1993). "Expression of CD44 variant proteins in human
colorectal cancer is related to tumor progression." Cancer Res 53(20): 4754-6.
Wiley, H. E., E. B. Gonzalez, W. Maki, M. T. Wu and S. T. Hwang (2001). "Expression of
CC chemokine receptor-7 and regional lymph node metastasis of B16 murine
melanoma." J Natl Cancer Inst 93(21): 1638-43.
Wirth, K., R. Arch, C. Somasundaram, M. Hofmann, B. Weber, P. Herrlich, S. Matzku and
M. Zoller (1993). "Expression of CD44 isoforms carrying metastasis-associated
sequences in newborn and adult rats." Eur J Cancer 29A(8): 1172-7.
Wittig, B. M., B. Johansson, M. Zoller, C. Schwarzler and U. Gunthert (2000). "Abrogation
of experimental colitis correlates with increased apoptosis in mice deficient for CD44
variant exon 7 (CD44v7)." J Exp Med 191(12): 2053-64.
Wolff, E. A., B. Greenfield, D. D. Taub, W. J. Murphy, K. L. Bennett and A. Aruffo (1999).
"Generation of artificial proteoglycans containing glycosaminoglycan-modified
CD44. Demonstration of the interaction between rantes and chondroitin sulfate." J
Biol Chem 274(4): 2518-24.
Wong, C. W., A. Lee, L. Shientag, J. Yu, Y. Dong, G. Kao, A. B. Al-Mehdi, E. J. Bernhard
and R. J. Muschel (2001). "Apoptosis: an early event in metastatic inefficiency."
Cancer Res 61(1): 333-8.
Wright, W. E., O. M. Pereira-Smith and J. W. Shay (1989). "Reversible cellular senescence:
implications for immortalization of normal human diploid fibroblasts." Mol Cell Biol
9(7): 3088-92.
Wubbolts, R., R. S. Leckie, P. T. Veenhuizen, G. Schwarzmann, W. Mobius, J.
Hoernschemeyer, J. W. Slot, H. J. Geuze and W. Stoorvogel (2003). "Proteomic and
biochemical analyses of human B cell-derived exosomes. Potential implications for
their function and multivesicular body formation." J Biol Chem 278(13): 10963-72.
Wurfel, J., M. Rosel, S. Seiter, C. Claas, M. Herlevsen, R. Weth and M. Zoller (1999).
"Metastasis-association of the rat ortholog of the human epithelial glycoprotein
antigen EGP314." Oncogene 18(14): 2323-34.
Wyllie, A. H., J. F. Kerr and A. R. Currie (1980). "Cell death: the significance of apoptosis."
Int Rev Cytol 68: 251-306.
Yang, T., T. F. Witham, L. Villa, M. Erff, J. Attanucci, S. Watkins, D. Kondziolka, H.
Okada, I. F. Pollack and W. H. Chambers (2002). "Glioma-associated hyaluronan
induces apoptosis in dendritic cells via inducible nitric oxide synthase: implications
for the use of dendritic cells for therapy of gliomas." Cancer Res 62(9): 2583-91.
Yasuda, M., Y. Tanaka, K. Fujii and K. Yasumoto (2001). "CD44 stimulation down-regulates
Fas expression and Fas-mediated apoptosis of lung cancer cells." Int Immunol 13(10):
Yu, Q., B. P. Toole and I. Stamenkovic (1997). "Induction of apoptosis of metastatic
mammary carcinoma cells in vivo by disruption of tumor cell surface CD44
function." J Exp Med 186(12): 1985-96.
Yu, Q. and I. Stamenkovic (1999). "Localization of matrix metalloproteinase 9 to the cell
surface provides a mechanism for CD44-mediated tumor invasion." Genes Dev 13(1):
Yu, Q. and I. Stamenkovic (2000). "Cell surface-localized matrix metalloproteinase-9
proteolytically activates TGF-beta and promotes tumor invasion and angiogenesis."
Genes Dev 14(2): 163-76.
Yu, W. H., J. F. Woessner, Jr., J. D. McNeish and I. Stamenkovic (2002). "CD44 anchors the
assembly of matrilysin/MMP-7 with heparin-binding epidermal growth factor
precursor and ErbB4 and regulates female reproductive organ remodeling." Genes
Dev 16(3): 307-23.
Zhang, Y., A. A. Thant, K. Machida, Y. Ichigotani, Y. Naito, Y. Hiraiwa, T. Senga, Y.
Sohara, S. Matsuda and M. Hamaguchi (2002). "Hyaluronan-CD44s signaling
regulates matrix metalloproteinase-2 secretion in a human lung carcinoma cell line
QG90." Cancer Res 62(14): 3962-5.
Ziegler, T., P. Silacci, V. J. Harrison and D. Hayoz (1998). "Nitric oxide synthase expression
in endothelial cells exposed to mechanical forces." Hypertension 32(2): 351-5.
Zoller, M., S. Matzku and K. Goerttler (1978). "High incidence of spontaneous transplantable
tumours in BDX rats." Br J Cancer 37(1): 61-6.
Zoller, M., K. Herrmann, S. Buchner, S. Seiter, C. Claas, C. B. Underhill and P. Moller
(1997). "Transient absence of CD44 expression and delay in development by antiCD44 treatment during ontogeny: a surrogate of an inducible knockout?" Cell Growth
Differ 8(11): 1211-23.
I want to thank Margot Zöller for the opportunity to do this thesis in her group and for
everything I learned in her lab. She was always there to discuss and help. Many thanks as
well to Jochen Wittbrodt for being my thesis-referee. I would like to thank Christoph for
sharing his experimental skills and for proof-reading. I will remember quiet some evenings
with him, Joachim, Markus and Sebastian with hot discussions - not only on scientific topics.
Thanks for being great lab mates. Many thanks to all other present and former members of
the lab, particularly to Susanne who was a big help with the animals and to Rachid, Pooja,
Mehdi and Frank.
I want to thank Clemens for english corrections, for kicking me when necessary and
for the good times. I’m grateful to Björn for his support during the writing and for reminding
me by his interest how fascinating this topic still is to me. And I would like to thank my
mother and my brother for their mental support during all this time.
Thanks a lot!
E. coli
BSp73AS, rat pancreatic adenocarcinoma
BSp73ASML, rat pancreatic adenocarcinoma
basic fibroblast growth factor
base pair
bovine serum albumin
cell adhesion molecule
CD44- standard isoform
CD44- variant isoform
complementary DNA
death inducing signaling complex
‚Deutsches Krebsforschungszentrum’
desoxyribonucleic acid
Escherichia coli
enhanced chemiluminescence
ethylendiamintetra acetic acid
epidermal growth factor
epidermal growth factor receptor
epithelial cell adhesion molecule
extra cellular matrix
extracellular regulated kinase
phosphorylated ERK
fluorescence-activated cell sorter
fetal calve serum
fluorescein isothiocyanate
gravitational acceleration
green fluorescence protein
distilled water
hyaluronic acid
N-2-hydroxyethylpiperazine-N´-2-ethanesulfonic acid
hepatocyte growth factor
hepatocyte growth factor receptor (c-met)
horseradish peroxidase
immune globulin
immuno precipitation
knock down
mitogen activated protein kinase
neural cell adhesion molecule
phosphate buffered saline
polymerase chain reaction
phosphatidylinositol 3-kinase
protein kinase C
ribonucleic acid
room temperature
single stranded
double stranded
sodium dodecyl sulfate
small interfering RNA
short hairpin RNA
trichloric acid
transforming growth factor-β
tissue inhibitor of metalloproteases
Texas Red
urokinase-type plasminogen activator
untranslated region
vascular endothelial growth factor
western blot
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF