Signal Transduction 2004 Signal transduction pathways as therapeutic by ouu11658

VIEWS: 554 PAGES: 269

                                 January 25-28th 2004

                           European Conference Center
                              Kirchberg - Luxembourg

                           Signal Transduction 2004
Signal transduction pathways as therapeutic targets


                                      Marc Diederich

                                        Organized by
                Recherches Scientifiques Luxembourg

          Printing of the Proceedings sponsored by the
 Fondation de Recherche Cancer et Sang (Luxembourg)


In 1998, we organized the first specialized meeting in the field of signal tranduction and
gene expression in Luxembourg. This type of meeting was originally thought to teach
doctoral students involved in the cellular and molecular biology teaching program (DEA de
Pharmacologie moléculaire) of the University of Nancy I (France).

From 2000 to 2003, more then 2500 fundamental and clincical researchers were gathering
in Luxembourg for meetings to discuss therapeutic applications in the field of signal
transduction, transcription and translation. These meetings allowed new insights into a
rapidly moving field. Novel antibodies against receptors, protein kinase inhibitors,
antisense oligonucleotides targeting both signal transduction and gene expression will
certainly predomine the therapeutic approaches for the next century.

For our Signal Transduction 2004 edition of meeting, more than 400 participants will be on
site, I am convinced that this meeting will be a great success.

Welcome to Luxembourg !

Marc Diederich


This meeting has been realized under the aegis of :

The Fondation de "Recherche Cancer et Sang"
Recherches Scientifiques Luxembourg asbl

We have the pleasure to acknowledge support from :

Kuwait Petroleum SA - Luxembourg
The City of Luxembourg
The Fondation de Recherche "Cancer et Sang"

Your onsite organization team :

Franck Morceau, Michael Schnekenburger, Sylvis Delhalle, Romain Blasius, Estelle
Henry, Marjorie Fougère

General Informations
Meeting Venue
All meeting sessions are held at the European Congress Center.
European Congress Center - Quartier Européen Sud,
1, rue du Fort Thüngen
L-1499 Luxembourg-Kirchberg

Registration will take place in the area outside of the "Hémicycle" (Level 0) at the
registration desk open daily (8H30-19H30).

Coffee breaks will be served in the exhibit area daily :
10H30-11H00 (morning break)
13H30-14H30 (after lunch)

For the participants that pre-paid lunch, lunch is served from 12h00 - 15h00 at the Bar de
l'Hémicycle (Level -1)

Additional tickets are NOT available at the registration desk.

Exhibits are open from Janaury 25-28th, 2004

Lecture rooms (see expo map for locations) :
-   Main lecture room : Hemicycle : Level 0

There will be two poster sessions :
I : Monday, January 26, 2004 (14h00 - 16h00) : Sessions I, II and III
II : Tuesday, January 27, 2004 (14h00 - 16h00) : Sessions IV, V, VI and VII
The organizers are not responsible for lost or damaged posters.

Welcome reception (expo area)
On Sunday January 25, 2004 from 20h00-20H30.

Gala Diner
The Gala Diner will take place at the Hotel Le Royal (12, Boulevard Royal, L-2449
Luxembourg) on Tuesday, Janury 27 th starting at 20H30. Additional tickets are NOT
available at the registration desk.

Our bus shuttles will bring the participants from the hotels to the meeting center in the
morning and back to the hotels or city center in the evening. Please note that the busses
leave between around 7h00 from your hotel depending on the distance to the meeting
center. The busses are red, orange and yellow and are from the bus company "Demy Cars".

Taxis can be called from the registration desk.

How to reach the congress center ?
- by taxi :
       from the airport (Luxembourg - Findel) in about 15 minutes.
       from the railroad station (about 15 minutes)
- by bus : take bus line number 16 (every 20 minutes from the city center)
- by car :
       from France : Highway A4 from Metz, take Highway A1 (E44) direction Trier
Plateau de Kirchberg Aéroport, choose exit number "8", orient towards "Quartier Europeen
Sud Luxembourg-Centre"
       from Belgium : Highway A411 from Brussels, take Highway A6 (E44) direction
Trier Plateau de Kirchberg Aéroport, choose exit number "8", orient towards "Quartier
Europeen Sud Luxembourg Centre"
       from Germany : Highway A6 (E44) from Trier, direction Luxembourg, choose exit
number "8", orient towards "Quartier Europeen Sud Luxembourg Centre".


                                    Scientific program
                   Signal transduction pathways as therapeutic targets
                           (January 25th - January 28th, 2004)

Sunday, January, 25th, 2004 (Afternoon)
Keynote session :
19h00 – 20h00 : Sara Courtneidge (Grand Rapids, USA) : Src family kinases in cellular
signal transduction pathways

Monday, January, 26th, 2004 (Morning)
Session I: Protein kinases as targets for novel treatments
8h30 – 9h00 : Roger Davis (Worcester, USA) : Targeting JNK for therapeutic benefit
9h00 – 9h30 : Roy Golsteyn (Croissy-sur-Seine, France) : Pharmacological inhibitors of
cyclin-dependent kinases (CDKs)
9h30 – 10h00 : Jacques Pouysségur (Nice, France) : Tumor angiogenesis and nutritional-
sensing mechanisms
10h00-10h15 : Abraham Amsterdam (Revohot, Israel) : Phosphodiesterase inhibitors as
anticancer drugs
10h15-10h30 : Katherine E. Kaempchen (Ulm, Germany) : The role of PAK2 in the development of
schwannoma; a target for tumour therapy?

10h30 – 11h00 : Coffee break

11h00 – 11h30 : Ying Xia (Cincinnati, USA) : Eye development and signaling through
11h30 – 12h00 : Matthias Wymann (Fribourg, Switzerland) : The Phosphoinositide 3-kinase
family: drug targets in inflammation and cancer
12h00 – 12h15 : Catherine Alexia (Paris, France) : Role of constitutively activated and of
Insulin-like Growth Factors-stimulated Erk-1/-2 signalling in human hepatoma cell
proliferation and apoptosis : evidence for heterogeneity of tumor cell lines
12h15 - 12h30 : Thomas Strömberg (Uppsala, Sweden) : Targeting the insulin-like growth
factor-I receptor (IGF-IR) in multiple myeloma cells using selective IGF-IR tyrosine kinase

13h00 – 14h00 : Workshop IBA : Thomas Schmidt : Applications of the Strep-tag
technology in proteomics. Special focus: purification of functional kinases.

14h00 - 15h00 : Workshop Becton-Dickinson : RNAi and gene function

15h00 - 16h00 : Workshop Becton-Dickinson : New technologies to study signal

Monday, January, 26th, 2004 (Afternoon)
Session II: Angiogenesis
16h00 – 16h30 : Andreas Bikfalvi (Bordeaux, France) : New molecules,strategies and
models to inhibit tumor angiogenesis and growth
16h30 – 17h00 : Raghu Kalluri (Boston, USA) : Extracellular Matrix Degradome as Novel
Tumor Suppressors
17h00 – 17h30 : Randolph Watnick (Cambridge, USA) : Ras modulates Myc activity to
repress thrombospondin-1 expression and increase tumor angiogenesis

17h30 – 17h45 : Juana Wietzerbin. (Paris, France) : Down regulation of angiogenic factors
in Ewing tumor xenografts by combination of human interferon-alpha or interferon-beta with
17h45 - 18h00 : Czeslaw Cierniewski (Lodz, Poland) : b1 integrins as targets to inhibit
tumor angiogenesis and carcinoma cell invasiveness
18h00 - 18h15 : Frédéric Kesteloot (Liège, Belgium) : Evaluation of the function of
ADAMTS-2, a metalloproteinase containing a disintegrin domain and thrombospondin type I
repeats, during angiogenesis in vitro and in vivo
18h15 - 18h30 : Levon M. Khachigian (Sydney NSW, Australia) : Catalytic DNA Targeting
Immediate-Early Genes as Novel Inhibitors of Angiogenesis

Tuesday, January, 27th (Morning)
Session III: Apoptosis
8h30 – 9h00 : Craig Thompson (Philadelphia, USA) : The proto-oncogenes AKT and Pim-2
are components of redundant signalling pathways that suppress apoptosis.
9h00 – 9h30 : Jean-Laurent Casanova (Paris, France) : The main Toll-like receptor signal-
transduction pathway in humans is not essential for survival and has a much narrower role
than expected
9h30 – 10h00 : D. Lambrechts (Leuven, Belgium) : VEGF is a modifier of amyotrophic
lateral sclerosis in mice and humans and protects motoneurons against ischemic death
10h00-10h15 : Christian Widmann (Lausanne, Switzerland) : Partial cleavage of RasGAP
by caspase-3 is required for survival in mild stress conditions
10h15-10h30 : Alain Chariot (Liège, Belgium) : Regulation of the oncogenic potential of
NF-kB proteins by phosphorylation

10h30 – 11h00 : Coffee break

11h00 –11h30 : Jo Trapani (Melburne, Australia) : Results indicate the importance of the
mitochondrial pathway in granzyme-B-mediated cell death.
11h30-12h00 : Paul Clarke (Dundee, UK) : Inhibition of caspase-9 through phosphorylation
at Thr 125 by ERK MAPK
12h00 – 12h30 : Kaspar Hoebe (La Jolla, USA) : Identification of Lps2 as a key transducer
of MyD88-independent TIR signaling
12h30 – 13h00 : Ivana Scovassi (Pavia, Italy) : Modulation of nuclear Poly(ADP-
ribosylation) in apoptotic cells

13h00 – 14h00 : Workshop CST

14h00 - 15h00 : Workshop nanoTools

15h00 - 16h00 : Workshop Ambion

Tuesday, January, 27th (Afternoon)
Session IV : Cancer specific signaling pathways as therapeutic targets
16h00 – 16h30 : Dean Felsher (Stanford, USA) : Cancer revoked: oncogenes as therapeutic
16h30 – 17h00 : Evan Keller (Ann Arbor, USA) : RAF kinase inhibitor protein (RKIP)
suppresses prostate cancer metastasis.

                                          - 10 -
17h00 – 17h30 : John Benson (Cambridge, USA) : Targeting of proteins to membranes
through hedgehog auto-processing
17h30 – 18h00 : Neil Watkins (Baltimore, USA) : Hedgehog signalling within airway
epithelial progenitors and in small-cell lung cancer.
18h00 – 18h30 : Stefano Piccolo (Padoa, Italy) : Convergence of the p53 and Smad signaling
18h30 – 19h00 : Jong-Wan Park (Seoul, Korea) : YC-1: A Potential Anticancer Drug
Targeting Hypoxia-Inducible Factor 1

Wednesday, January 28th, 2004 (Morning)
Session V: Chemopreventive agents
8h30 – 9h00 : Bharat Aggarwal (Houston, USA) : Chemopreventive agents as therapeutic
9h00 – 9h30 : Young-Joon Surh (Seoul, Korea) : Intracellular Signaling Molecules as Prime
Targets for Chemopreventive Phytochemicals
9h30 – 9h45 : Marc Diederich (Luxembourg, Luxembourg) : Effects of chemopreventive
agents on stress and drug resistance genes
9h45 –10h00 : Norbert Latruffe (Dijon, France) : Transport of resveratrol , a cancer
chemopreventive agent : plasmatic protein binding, uptake and cellular targets
10h00- 10h30 : Ho-Young Lee (Houston, USA) : Effects of Deguelin on the
Phosphatidylinositol 3-Kinase/Akt Pathway and Apoptosis in Premalignant Human Bronchial
Epithelial Cells

10h30 – 11h00 : Coffee break

11h00 – 11h30 : Hee-Gu Lee (Daejeon, Korea) : TetramethoxyFlavone p7F Inhibits Collagen-
Induced Arthritis by Suppressing Inflammatory Mediators

11h30-11h45 : Michael Naim (Rehovot, Israel) : Some sweet and bitter tastants inhibit GRK2,
GRK5 and PKA in vitro: possible effects on signal termination
11h45-12h00 : Jung-Ae Kim (Kyongsan, South Korea) : Anticancer effect of asiatic acid on
skin cancer: Induction of apoptosis in SK-Mel-2 human melanoma cells and
antitumorigenesis in DMBA/TPA-induced mouse skin tumor
12h00-12h15 : Richard Eckert (Cleveland, USA) : The antioxidants (-)-epigallochatechin-3-
gallate and curcumin inversly regulate human involucrin promoter activity via opposing
effects on CCAAT/Enhancer binding protein action
12h15-12h30 : Carmela Fimognari (Bologna, Italy) : 4-(methylthio)butylisothiocyanate as a
new chemopreventive agent: molecular pathway for cell-cycle inhibition and apoptosis of
human T lymphoblastoid cell line.
12h30-12h45 : Do-Hee Kim (Seoul, Korea) : Growth Inhibition and Cell Cycle
Dysregulation by Eupatilin, a Pharmacologically Active Flavone Derived From Artemisia
Plants, in H-Ras-Transformed Human Mammary Epithelial Cells

13h00 – 14h00 : Workshop Biosoucre

Session VI : siRNA and gene regulation
(Session organizer : Ch. Branlant, Nancy, France)
14h00 – 14h30 : W. Filipowicz (Basel, Switzerland) : Function of Dicer in RNAi and
microRNA processing in mammalian cells

                                          - 11 -
14h30 – 15h00 : Christophe F. Deroanne (Liège, Belgium) : MMP-1 expression in human
skin fibroblasts is negatively regulated by Cdc42.
15h00 - 15h30 : Natalia Martinez (Tübingen, Germany) : The role of the leukemic fusion
protein AML1/MTG8 in the proliferation of leukemic cells

End of the meeting

                                        - 12 -
Abstract Listing : Oral presentations

        - 13 -
Keynote session

Src family kinases in cellular signal transduction pathways

Darren F. Seals1, Clare L. Abram2, Ian Pass1, Eduardo Azucena1, Rebecca Gordon1, Daniel E.
Salinsky1 and Sara A. Courtneidge1
    Van Andel Research Institute, 333 Bostwick Avenue, Grand Rapids, MI 45903
    SUGEN, Inc., 230 East Grand Avenue, South San Francisco, CA 94080

We recently described a new method for isolating substrates for tyrosine kinases, that
involves phosphorylating proteins expressed from phage cDNA expression libraries with
purified kinase, and detection of phosphorylated plaque proteins with phosphotyrosine-
specific antibodies. Using this method we have isolated a number of novel proteins, including
Fish, an adaptor protein with 5 SH3 domains, a PX domain, and several motifs that can
promote association with SH2 and SH3 domain-containing proteins. The PX domain of Fish
associates with phosphatidylinositol phosphates, and the fifth SH3 domain associates with the
cytoplasmic tails of several members of the ADAMs family of metalloproteases. ADAMs
family proteins are implicated in shedding of growth factors from cell surfaces, as well as
invasion and motility.
Fish is primarily distributed in the cytosol of normal mouse fibroblasts, but translocates to
specialized, cell surface structures called podosomes in fibroblasts transformed with Src.
Podosomes, or invadopodia, are F-actin enriched structures common to Src-transformed cells
and some highly invasive carcinomas. We find that Fish is highly expressed in certain
invasive breast carcinomas and that Src, Fish, and ADAMs are co-localized at podosomes in
both Src-transformed fibroblasts and in Hs578T breast carcinoma cells. We are currently
investigating these cell lines for the mechanism behind this co-localization and whether Src,
Fish, and ADAMs are necessary for the extracellular matrix-remodeling properties of these
unique cell structures.

                                            - 14 -
Session I : Chair : Roger Davis

  - 15 -
Session I: Protein kinases as targets for novel treatments



Howard Hughes Medical Institute & University of Massachusetts Medical School, Worcester,
MA 01605, E-mail : Roger.Davis@Umassmed.Edu

The JNK group of stress-activated MAP kinases consists of ten protein kinases that
phosphorylate the NH2-terminal activation domain of c-Jun on Ser-63 and Ser-73 causing
increased transcriptional activity. JNK protein kinase activity is increased in response to
treatment of cells with pro-inflammatory cytokines or exposure to environmental stress.
Activated JNK is phosphorylated on Thr and Tyr within the tripeptide motif Thr-Pro-Tyr
located in kinase sub-domain VIII. Mutational analysis demonstrates that JNK activation
requires the phosphorylation of both Thr and Tyr within this motif. This phosphorylation is
mediated by dual specificity protein kinases, including MKK4 and MKK7.

The function of the JNK signaling pathway has been studied using a combination of
biochemical and genetic approaches. Genetic analysis of JNK signaling in Drosophila
demonstrates that JNK is required for early embryonic morphogenesis. Similarly, disruption
of the JNK signaling pathway in mice using homologous recombination demonstrates that
JNK is required for embryonic viability. In contrast, mice with genetically engineered
selective defects in JNK signaling are viable, but exhibit changes in stress-induced gene
expression and apoptosis. These studies provide insight into the role of the JNK stress-
activated MAP kinase pathway in the cellular response to environmental stress, including
apoptosis and cell survival.

                                              - 16 -
Session I: Protein kinases as targets for novel treatments

Analysis of cyclin-dependent kinases (CDKs) in apoptosis using inhibitors and siRNAs

Isabella Versteege, Annie Borgne, Séverine Ugnon-Café, Stéphane Léonce, Laurent Meijer,
J.A. Hickman , Roy M. Golsteyn

Institut de Recherches Servier, Cancer Research Division, 125 ch. de Ronde
78290 Croissy-sur-Seine, France

In a number of cell culture models in vitro and in vivo, cyclin dependent kinase (Cdk) activity
is correlated with apoptosis. By contrast to their role in the cell division cycle, the role of
Cdks in apoptosis is poorly understood. We examined cyclins E1, A2, B1 and the kinase
subunits Cdk2, Cdk1 in human HT29 cells that were undergoing apoptosis. Camptothecin
(CPT), a topoisomerase I inhibitor initiates apoptosis during a period of approximately 50-70
hours before cell death, which enabled us to examine Cdk/cyclin function in early stages of
apoptosis. Expression levels of cyclin B1, cyclin E1, cyclin A, Cdk1 and Cdk2 increased at
16h after CPT treatment. Analysis of treated cells by immunofluoresence microscopy
revealed striking co-expression of cyclin E1 and B1 in cells. Subsequent analysis of DNA
content and cyclin expression showed cyclin B1 was expressed in the G1 phase and cyclin E1
was present in the G2/M phase. A peak of cyclin B1 kinase activity was detected at 38h after
CPT treatment, whereas cyclin E1 associated activity was elevated by 16 hours and then
remained constant. The phosphorylation status of Cdk1 corresponded with protein levels of
WEE1 kinase and the activity of cyclin B1 complexes. Roscovitine, a chemical inhibitor of
Cdk1 and Cdk2 reduced the number of apoptotic cells was by 50%. Treatment of cells with
siRNAs directed against cyclin E1 or Cdk1 did not affect apoptosis, whereas Cdk2
knockdown by RNAi reduced the number apoptotic cells as measured by annexin V staining.
The high levels of cyclin B1 activity, which is normally associated with entry into mitosis,
prompted us to examine if the treated cells were entering mitosis prior to apoptosis. Mitotic
markers such as MPM2, phospho-histone H3 and phospho-nucleolin proteins were not
detected in cells committed to apoptosis whereas these proteins were easily detected in
control mitotic cells. We propose that in this apoptosis model, Cdk complexes are activated,
but only Cdk2 complexes are essential for completion of the apoptotic program.

                                              - 17 -
Session I: Protein kinases as targets for novel treatments

HYPOXIA signaling and Angiogenesis

Edurne Berra, Emmanuel Benizri, Frédéric Dayan, Amandine Ginouvès, Nathalie Mazure,
Danièle Roux and Jacques Pouysségur

Institute of Signaling, Developmental Biology and Cancer Research, CNRS UMR 6543,
Centre Antoine Lacassagne, 33 Avenue Valombrose, 06189, Nice, France

Angiogenic factors modulate basic endothelial cell functions such as migration and
proliferation, but how these factors are regulated in response to various nutritional stresses, at
a molecular level, is not yet fully understood. A remarkable signaling system devised for
rapid adaptation to and survival in a low oxygen environment (hypoxia) has been conserved
throughout evolution. The hypoxia inducible transcription factor HIF, found in worms, flies
and vertebrates is central to this adaptation and as such, hif-1 represents a ‘master’ gene in
oxygen homeostasis. HIF-1 is a transcriptional complex that plays a pivotal role in cellular
adaptation to low oxygen availability, inducing many genes encoding for VEGF, EPO,
GLUT1 and other essential glycolytic enzymes. In the presence of oxygen, the HIF-a
subunits are targeted for destruction by proline hydroxylation, a specific modification that
provides recognition for the E3 ubiquitin ligase complex containing the von Hippel-Lindau
tumour suppressor protein (pVHL). HIF-1a degradation takes place equally well in nucleus
and cytoplasm, suggesting that proline hydroxylation, pVHL-mediated ubiquitination and
proteasomal destruction are competent in both cellular compartments. Three mammalian HIF
prolyl-hydroxylases (PHD1, 2, 3), homologous to the HIF prolyl-hydroxylase of C. elegans
(EGL-9), were recently identified and shown, at least in vitro, to down regulate HIF-a
subunits. These enzymes, together with the HIF asparaginyl hydroxylase (FIH), which
‘represses’ the activity of HIF-1a, belong to a large family of non-haem iron oxygenases that
require O2 and 2-oxoglutarate for their function. In this presentation we will show that
specific ‘silencing’ of HIF prolyl-hydroxylase 2 (PHD2) with short interfering RNAs
(siRNA) is sufficient to stabilize and activate HIF-1a, in normoxia, in all the human cells
investigated. A remarkable synergy in the HIF-1a activation process is observed in normoxia
by ‘co-silencing’ both, PHD2 and FIH. This action measured by a HIF-dependent reporter
gene recapitulates a full hypoxic response. Surprisingly, ‘silencing’ of the other two HIF
prolyl-hydroxylases, PHD1 and PHD3, had no effect on the stability of HIF-1a either in
normoxia or upon re-oxygenation of hypoxic cells. We conclude that PHD2, a cytoplasmic
enzyme capable of shuttling between cytoplasm and nucleus, is the critical oxygen sensor
setting low steady-state levels of HIF-1a, in normoxic conditions. Interestingly, PHD2 is up-
regulated by HIF-1 providing an auto-regulatory mechanism driven by the oxygen tension, a
retro-control mechanism that we previously proposed.
The stabilisation of HIF-1a in normoxic cells, however, following PHD2 ablation, is
progressively “desensitized” within 4 to 5 days in culture uncovering a more complex
mechanism than originally thought. Central to this action appears to be a HIF-1 dependent
activation process, implicating an interplay between oxygen sensors PHD2 and PHD1, a
model of which will be presented.

                                              - 18 -
Session I: Protein kinases as targets for novel treatments

Phosphodiesterase inhibitors as anticancer drugs.

L. Hirsha, O. Merimskyb, A. Dantesa , A. Land-Brachaa and A. Amsterdama
    Molecular Cell Biology, Weizmann Institute of Science, Rehovot, Israel , 76100
    Tel-Aviv Medical Center, Section of Lung and Hard Tissue Cancer, Israel

It is well known that high intracellular levels of cAMP can effectively kill cancer cells in
vitro. Unfortunately substances elevating cAMP such as forskolin, 8bromo-cAMP, 8-chloro-
cAMP, mono- and di butiryl cAMP are not recommended to be used as anti- cancer drugs
because of their high cytotoxicity. In contrast blockers of phosphodieterases such as
theophylline and aminophylline, which could elevate intracellular cAMP, are commonly used
as anti-asthma drugs reaching concentrations in the blood of 20-25mg/ml. We tested the
effectiveness of theophylline and aminophylline to induce cell death alone or in combination
with common anti-cancer drugs such as cisplatin and gemcitabine (gemzar). We tested for the
induction of cell death on a variety of carcinoma cell lines derived from human ovarian,
prostate and lung cancer. While theophylline could induce moderate cell death alone, at 20-
25 mg/ml concentrations, aminophylline was ineffective at this concentration. Theophylline
was found in all three representative cell lines to synergize with gemcitabine and cisplatin, to
induce of programmed cell death, which permits a reduction in the effective doses of cisplatin
and gemcitabine by 2-3 fold. The effect of theophylline in induction of apoptosis, involved
reduction of intracellular levels of BCL 2. Such a reduction was proportional to the extent of
apoptosis induced by theophylline as well as by the combined drug treatments. Therefore we
propose that theophylline should be considered as a potential anti-cancer drug in combination
with other chemotherapeutic drugs. Screening of other phosphodiesterase blockers which are
not severely toxic could open a possibility to improved chemotherapeutic cancer treatments
with reduced undesired side-effects. A clinical trial using theofylline as anti-cancer drug is
currently conducted in lung cancer patients.

                                              - 19 -
Session I: Protein kinases as targets for novel treatments

The role of PAK2 in the development of schwannoma; a target for tumour therapy?

Katherine E. Kaempchen, Tamara Utermark, C. Oliver Hanemann.

Department of Neurology, Zentrum fuer klinische Forschung, Helmholtzstrasse 8/1, 89081
Ulm, Germany. E-mail:

Schwannoma or neurinoma are benign tumours of the nervous system which develop when
Schwann cells lack the tumour suppressor protein merlin (schwannomin). Research into the
mechanism by which merlin acts as a tumour suppressor has led to the discovery that this
protein is phosphorylated by the small GTPases Rac and Cdc42 via the P21 activated kinases
(PAK) (Kissil et al., 2002; Xiao et al., 2002). It has also been suggested that merlin can in
turn inactivate PAK1 by binding to it, and therefore prevent it from migrating to the
membrane at the site of focal adhesions (Kissil et al., 2003; Hirokawa et al., 2003). Patients
suffering from Neurofibromatosis type 2 (NF2) present with multiple schwannoma as they are
carriers of a somatic “first hit” mutation on one allele of the NF2 gene coding for merlin,
tumour cells deriving from such patients therefore provide an excellent model for studying the
development of neurinomas in man. We show that in primary human schwannoma cells
higher levels of the small GTPase Rac1 can be found at the membrane of these cells, where
activated Rac co-localises with the phosphorylated form of PAK. Moreover we have been
able to show by Western blotting that specific PAK isoforms are found at identical levels in
Schwann and schwannoma cells. Primarily PAK2, but also PAK1 can be detected under their
phosphorylated forms in these cells, suggesting a particular importance for PAK isoforms in
schwannoma development. Immunocytological assays have permitted to show that, following
the theory supported by Kissil et al., in schwannoma but not Schwann cells specific PAKs
migrate to the membrane of non confluent cells and to supposed focal adhesion points in
confluent cells, thus suggesting that the tumoural development of schwannoma cells is
dependent on the up-regulation of Rac-PAK activity. Specific PAK inhibitors are therefore of
potential therapeutic interest for neurinoma.

                                              - 20 -
Session I: Protein kinases as targets for novel treatments

MEK kinase 1 controls epithelial cell migration and tissue closure

Ying Xia and Lin Zhang

Department of Environmental Health and Center of Environmental Genetics, University of
Cincinnati Medical Center

The mitogen-activated protein kinases (MAPKs) are a group of intracellular signaling kinases,
consisting of the extracellular signal regulated kinase (ERKs), the c-Jun N-terminal kinases
(JNKs) and the p38s as the major subgroups. In any given biological setting, members of
each subgroup play unique roles in signal transduction and regulation of cell functions.
Often, alterations in signaling pathways are associated with disease states and investigation of
the mechanisms underlying MAPK function and specificity can provide novel avenues of
therapeutic intervention. MEK kinase 1 (MEKK1) is a MAPK kinase kinase that controls
MAPK activity in a cell type- and stimulus-specific fashion. Genetic ablation in mice has
demonstrated a role for MEKK1 in movement and sealing of embryonic eyelid epidermis by a
process known as eyelid closure. MEKK1-null mice show a specific eye-open at birth
phenotype as a result of failure of the embryonic eyelid to migrate and close. In the
developing eyelid epithelium, MEKK1 controls JNK-mediated c-Jun N-terminal
phosphorylation and F-actin formation, two functions that are likely to be responsible for the
epithelial cell motility that is required for eyelid closure. In cultured keratinocytes, MEKK1
is also required for cell migration induced by TGFb and activin, but not by TGFa. The TGFb
signal is transduced through RhoA to activate MEKK1, which in turn leads to the induction of
JNK phosphorylation in a pathway distinct from Smad activation by TGFb. The MEKK1-
driven JNK activation controls actin stress fiber formation through a transcription-
independent mechanism, while its role in epithelial cell migration is transcription dependent.
Furthermore, infection of keratinocyte with an adenovirus vector expressing active MEKK1
results in keratinocyte migration without affecting Smad4 nuclear translocation. Hence,
MEKK1-mediated JNK activation is a unique pathway that transduces TGFb signals in the
control of epithelial cell migration and mouse embryonic eyelid closure. Given the
controversial role of TGFb in tissue morphogenesis and skin wound repair, dissecting the
downstream events of TGFb signaling that depend on MEKK1 activation might reveal useful
targets for therapeutic intervention in tissue injury repair.

                                              - 21 -
Session I: Protein kinases as targets for novel treatments

The Phosphoinositide 3-kinase family: drug targets in inflammation and cancer

Matthias P. Wymann

Div. Biochemistry, Dept. Medicine, University of Fribourg, CH-1700 Fribourg

Phosphoinositide 3-kinases (PI3Ks) play a major role in the control of cell growth, proliferation and
survival. This is mediated by downstream effectors like phosphoinositide-dependent kinase (PDK),
protein kinase B (PKB) and the target of rapamycin (TOR). In other contexts PI3Ks link to cell
motility, secretory processes and the modulation of cardiovascular events. Present data implies that
PI3Ks have a potential as drug targets for the treatment of proliferative, inflammatory, allergic and
cardiovascular disease. Therapy is complicated by the involvement of PI3Ks in vital processes, e.g.
metabolic control and nutrient uptake.
The PI3K family: PI3Ks can be sorted into three families, class I PI3Ks produce PtdIns(3,4,5)P3 in
vivo, class II PI3Ks accept PtdIns 4-P and PtdIns in vitro, while class III PI3K converts PtdIns to
PtdIns 3-P. Of these PI3Ks, class IA members are mainly activated by translocation to protein tyrosine
kinase receptors or their substrates, as their regulatory subunit contains two src-homology 2 (SH2)
domains binding to phosphorylated TyrXXMet motifs. The class IB (PI3Kg), on the other hand, is
activated by bg subunits of trimeric G proteins, and was shown to be associated with a p101 protein in
neutrophils (reviews see 1,2).
Pharmacological tools: presently available PI3K inhibitors (wortmannin [3] and LY294002 [4]) are
potent, but not isoform specific. While novel compounds are not represented in academic journals,
isoform specific inhibitors start to appear in the patent literature (for a review see 2).
Insights from PI3Kg null mice: PI3Kg has been shown to be the sole PI3K isoform to directly relay
chemokine receptor signals, and is thus crucial for pathfinding in leukocytes in vitro and in vivo (5).
Moreover, PI3Kg plays an important role in the activation of mast cells, which results in a protection
of PI3Kg null mice from passive systemic anaphylaxis (6). Mast cells modulate inflammation through
the release of histamine, cytokines and lipid mediators. During antigen/IgE stimulation,
phosphoinositide 3-kinase g (PI3Kg) relays paracrine and autocrine inflammatory signals and thus
plays a central role in mast cell function (6). Half of the PI3Kg-dependent activation is mediated by
adenosine and other Gi-protein coupled receptor (GPCR) ligands. The rest is controlled by a novel,
Ca2+-dependent pathway, which can be triggered separately by Ca2+-ionophores or thapsigargin and is
resistant to B. pertussis toxin and adenosine deaminase. This, and the fact that also phorbol ester
(PMA) triggered protein kinase B (PKB/Akt) phosphorylation in a PI3Kg-dependent way, suggested
that diacylglycerol-binding molecules could be involved. As such, guanine nucleotide exchange factor
RasGRP4 activates Ras in mast cells. In 32D cells expressing RasGRP4, however, PI3Kg/PKB was
not activated by PMA, while MAPK responded. Using a panel of protein kinase C (PKC) inhibitors
we could dissect classical and novel PKC members, which are upstream of PI3Kg. PI3Kg thus
integrates signals from GPCRs via G protein bg subunits and relays a Ca2+-triggered, PKC-dependent
pathway, which plays an important role in allergen-mediated mast cell stimulation.
1. Wymann, M. P., and Pirola, L. (1998). Biochim. Biophys. Acta 1436:127.
2. Wymann M.P., Zvelebil M., Laffargue M. (2003) Trends Pharmacol. Sci. 24:366.
3. Arcaro, A., and Wymann, M. P. (1993). Biochem. J. 296, 297-301.
4. Vlahos, C. J. et al. (1994). J. Biol. Chem. 269, 5241-5248.
5. Hirsch, E et al. (2000) Science 287:1049
6. Laffargue et al. (2002). Immunity, 16:441.

                                                - 22 -
Session I: Protein kinases as targets for novel treatments

Role of constitutively activated and of Insulin-like Growth Factors-stimulated Erk-1/-2
signalling in human hepatoma cell proliferation and apoptosis : evidence for
heterogeneity of tumor cell lines.

Catherine Alexia & André Groyer

Inserm U.481, Faculté de Médecine Xavier Bichat, 75870 Paris Cédex 18, France

Several observations (e.g. enhanced IGF-II and IGF-type I receptor -IGF-IR- genes
expression in liver tumors, development of liver tumors in transgenic mice overexpressing
IGF-II in the liver) suggest that IGF-I and/or -II and underlying signalling cascades may play
an auto/paracrine role in the control of hepatocarcinoma (HCC) cell proliferation and in their
protection against apoptosis. Objectives. We have focused on the role of MAP kinase (Erk-1/-
2) signalling on human HepG2 and Huh-7 hepatoma cells proliferation and on the protection
of these cells against drug-induced apoptosis. The effects of both constitutive and IGF-I-
stimulated Erk-1/-2 activities were studied. Results. Physiological concentrations of IGF-I (a
high affinity ligand for IGF-IR but not for type A insulin receptor) stimulated HepG2 cells
DNA replication by 1.6-fold. By contrast, Huh-7 cells were unresponsive to the proliferative
effect of IGF-I. In both cell lines, doxorubicin or cisplatin treatment induced apoptosis, as
mirrored by caspase-dependent PARP cleavage. Partial and dose-dependent reversion of
drug-induced apoptosis (84-57%) was obtained in the presence of IGF-I in HepG2 but not in
Huh-7 cells. The effects of IGF-I on cell proliferation and drug-induced apoptosis in HepG2
cells were partially inhibited (50% and 30%, respectively) by treatment with PD98059 (MEK-
1 inhibitor). Unresponsiveness of Huh-7 cells to both the proliferative and antiapoptotic
effects of IGF-I is probably due to the very low level of IGF-IR at the plasma membrane . We
have shown that Erk-1/-2 are constitutively activated in cultured HepG2 and Huh-7 cells and
that Erk-1/-2 activity is enhanced by drug treatment (17-fold) in HepG2 cells and only slightly
increased (1.5-fold) in Huh-7 cells. In both cell lines, inhibition of constitutive and drug-
induced Erk-1/-2 activity by PD98059 yielded a complete inhibition of drug-induced
apoptosis. Conclusions. Our data clearly show that hepatoma cells are heterogeneous with
regard to the response to an IGF stimulus. They suggest that an auto/paracrine effect of liver-
secreted IGF-I/-II might contribute to the proliferation of HCC cells in vivo and to the
protection of these cells against apoptosis (HepG2). Moreover, the results obtained with Huh7
cells suggest that HCC cells may become insensitive to the IGFs. Finally, our findings suggest
that constitutive activity (Huh-7) and/or drug-induced activation (HepG2) of Erk-1/-2
activation plays an active role in drug-induced apoptosis in human hepatoma cell lines.

                                              - 23 -
Session I: Protein kinases as targets for novel treatments

Targeting the insulin-like growth factor-I receptor (IGF-IR) in multiple myeloma cells
using selective IGF-IR tyrosine kinase inhibitors

Thomas Strömberg1, Olle Larsson2, Magnus Axelson3, Kristina Carlson4, Karin
Vanderkerken5, Kenneth Nilsson1 and Helena Jernberg-Wiklund1
  Department of Genetics and Pathology, Rudbeck Laboratory, Uppsala University, Uppsala,
Sweden E-mail:
  Cancer Center Karolinska, Karolinska Institute, Stockholm, Sweden
  Department of Clinical Chemistry, Karolinska Hospital, Huddinge University Hospital,
Karolinska Institute, Stockholm, Sweden
  Department of Medical Sciences, University Hospital, Uppsala, Sweden
  Department of Hematology and Immunology, Vrije Universiteit Brussel (VUB), Brussels,

In multiple myeloma (MM) emerging evidence suggest the IGF-IR as an important mediator
of tumor cell survival and thus resistence to cytotoxic therapy. Since IGF-IR is not an
absolute requirement for maintenance of normal cell homeostasis, interfering with IGF-IR
signaling at the receptor tyrosine kinase (RTK) level represents an attractive strategy to
improve anti-cancer treatment. However, most IGF-I RTK inhibitors have the disadvantage
that they do not fully discriminate between the IGF-I RTK and the insulin RTK, i.e are
diabetogenic. Recently, members of the cyclolignan family have been evaluated for inhibitory
effects on the IGF-IR. Some of these compounds were shown to potently inhibit the IGF-I
RTK activity without downregulating the RTK activity of the insulin R. In a panel of nine
MM cell lines the IGF-I RTK inhibitors effectively inhibited growth providing increased
apoptosis as well as cell cycle arrest in the G2/M-phase. Notably, the two drug resistant
subclones of the MM cell line RPMI 8226, RPMI 8226/Dox40 (doxorubicin) and RPMI
8226/LR5 (melphalan), were also highly sensitive to the IGF-I RTK inhibitors. In addition,
the compounds showed inhibitory effects on freshly isolated MM cells cultured with or
without bone marrow stromal cell feeder layers. Thus, we show for the first time that
treatment of MM cells with selective IGF-I RTK inhibitors potently decreases both survival
and proliferation, thus emphasizing the pivotal role for IGF-IR signaling in MM.

                                              - 24 -
Session II : Chair : Andreas Bikfalvi

        - 25 -
Session II: Angiogenesis

Recent insights into anti-angiogenesis : New tumor angiogenesis inhibitors, mechanisms and

Andreas Bikfalvi, Molecular Angiogenesis Laboratory (INSERM E 0113), University Bordeaux I,

Inhibition of angiogenesis is an important strategy to block tumor growth and invasion. Results from
our current studies on platelet factor-4, on a new vascular endothelial growth inhibitor and on a new
tumor-angiogenesis model will be presented :

(1)Platelet factor-4 (PF-4) and derived molecules
We have studied the structure and anti-angiogenic activities of a C-terminal fragment of PF-4 named
PF-4 CTF. It contains two clusters of basic amino acids and one DLQ motif. PF4 CTF shows a far-UV
circular dichroism spectrum typical of polypeptides in which a-helix is the only defined element in
their secondary structure. The three-dimensional structure is consistent of two helices connected by a
spacer. PF-4 CTF completely inhibits endothelial cell proliferation, angiogenesis in vitro, ex vivo
migration and vessel assembly. PF-4 CTF also significantly blocks vessel ingrowth into collagen
sponges in vivo. When injected in immunodeficient mice implanted intracranially with U87 glioma,
systemically or locally administered PF-4 CTF inhibits tumor growth and recurrence. We have also
attempted to study the mechanisms of action of the fragment. Our results indicate an extracellular
mechanism for PF-4 CTF by directly interacting with angiogenesis factors. PF-4 CTF may be
improved by mutation or domain-swapping and seems, therefore, to be a good candidate for further
development. The results on one such molecule, PF-4 DLR will be presented. We will also present
results on the combinatory administration of this molecule with other inhibitors.

Cyclic vascular endothelial growth inhibitor (Cyclo-VEGI)
We have developed a novel vascular endothelial growth inhibitor designated cyclic vascular
endothelial growth inhibitor (cyclo-VEGI) and determined its structure and biological activity. Cyclo-
VEGI is a 17 amino-acid molecule which encompasses residues 79-93 of VEGF which are involved in
the interaction with VEGF receptor-2. In aqueous solution, cyclo-VEGI presents a propensity to adopt
an a helix conformation that was largely unexpected since only b-sheet structures or random coil
conformations have been observed for macrocyclic peptides. Cyclo-VEGI inhibits binding of
iodinated VEGF165 to endothelial cells, endothelial cells proliferation, migration and signaling induced
by VEGF165. This peptide also exhibits anti-angiogenic activity in vivo on the differentiated chicken
chorioallantoic membrane. Furthermore, cyclo-VEGI significantly blocks the growth of established
intracranial glioma in nude and syngeneic mice and improves survival without side effects.
Combinatory administration with other inhibitors further improves efficacy. These results suggest that
cyclo-VEGI is an attractive candidate for the development of novel angiogenesis inhibitor molecules
useful for the treatment of cancer and other angiogenesis-related diseases.

New tumor angiogenesis models
We have developed a new tumor angiogenesis model using the chicken chorioallantoide membrane in
which invasion and angiogenesis-dependent-growth are the cardinal features. We have characterized
extensively this model by immunohistochemistry and confocal microscopy. The model is also suitable
for investigating the effect of anti-angiogenesis molecules at a cellular and molecular level

                                                 - 26 -
Session II: Angiogenesis

Extracellular Matrix Degradome can Serve as Novel Tumor Suppressors

Rhagu Kalluri

Center for Matrix Biology, Beth Israel Deaconess Medical Center 330 Brookline Ave -Dana
512 Boston, MA 02215

Progression of cancer is dependent on angiogenesis. Relative levels of pro- and anti-
angiogenic factors, the 'angiogenic balance', likely govern tumor progression. While the role
of pro angiogenic factors such as VEGF and FGF is well established, emerging evidence
suggests that anti-angiogenic factors such as thrombospondin-1, tumstatin, endostatin and
canstatin are important for maintenance of the angiogenic balance. Conversion of dormant in
situ carcinoma into an invasive malignant phenotype is considered to involve a shift in favor
of enhanced 'angiogenesis potential.' Influenced by oncogenes and tumor suppressor genes,
disruption of the 'angiogenic check point' via increase in angiogenic factors such as VEGF or
decrease in the physiological levels of endogenous inhibitors of angiogenesis like
thrombospondin-1 and tumstatin could represent an important lethal step in the progression of
cancer. Recent evidence suggests that vascular integrins are critical mediators of the action of
such matrix derived degradome. Thus, genetic control of the physiological levels of
endogenous inhibitors of angiogenesis and their vascular integrin targets might constitute a
critical last line of defense against conversion of neoplastic events into a malignant phenotype
of cancer.

                                             - 27 -
Session II: Angiogenesis

Ras modulates Myc activity to repress thrombospondin-1 expression and increase tumor

Randolph S. Watnick, Ph.D.

Children’s Hospital, Harvard Medical School Boston, MA 02115

Tumor angiogenesis is postulated to be regulated by the balance between pro- and anti-
angiogenic factors. We demonstrate that the critical step in establishing the angiogenic
capability of human cells is the repression of a critical anti-angiogenic factor,
thrombospondin-1 (Tsp-1). This repression is essential for tumor formation by mammary
epithelial cells and kidney cells engineered to express SV40 early region proteins, hTERT,
and H-RasV12. We have uncovered the signaling pathway leading from Ras to Tsp-1
repression. Ras induces the sequential activation of PI3 kinase, Rho and ROCK, leading to
activation of Myc through phosphorylation; this phosphorylation enables Myc to repress Tsp-
1 expression. We thus describe a novel mechanism by which the activity of the oncogenes,
ras and myc, leads directly to angiogenesis and tumor formation.

                                           - 28 -
Session II: Angiogenesis

Down regulation of angiogenic factors in Ewing tumor xenografts by combination of
human interferon-alpha or interferon-beta with ifosfamide.

Josiane Sancéau and Juana Wietzerbin.

INSERM U365, Institut Curie, Section Recherche 26, rue d'Ulm, 75 248 - Paris 05, France.

Ewing sarcoma is the second most common bone tumor in childhood, characterized by a t
(11/22 chromosomal translocation which results in the production of the chimeric EWS/Fli-1
transcription factor. Despite very aggressive chemotherapy and radiotherapy strategies, the
prognosis of patients with metastatic disease remains poor. We have recently shown in a nude
mice model of Ewing tumor xenografts that human type I IFNs display an anti-growth effect
toward established xenografts. Combined therapy with Hu-IFNs and ifosfamide (IFO), an
alkylating agent widely used in high dose chemotherapy of Ewing tumors, results in a strong
synergistic antitumor effect. (Sanceau et al Oncogene 2002). We have investigated the effect
of IFNs and IFO treatment on tumor xenografts generated either from a primary tumor (EW7)
or from a metastatic tumor (COH) on the expression of VEGF, MMP9 and uPAR, three key
mediators of tumor growth and angiogenesis. Determination of VEGF protein concentration
in extracts from EW7 or COH fresh tumors xenographs showed that COH tumors expressed
much higher levels of VEGF (400ng/mg of proteins) than EW7 tumors (90ng/mg of proteins).
In EW7 tumors, treatment with IFNs, or IFO alone lead to a reduction of VEGF levels of 25,
% and 20% respectively. A reduction of 20% was obtained in COH tumors treated with IFO
alone while interferons had no effect on VEGF level. Interestingly, combined treatment with
IFNs/IFO reduced by more than 70% the amount of VEGF detected in COH and EW7
tumors.. The metastasis-derived COH tumor, expressed high levels (2500 ng/mg of proteins)
of active MMP9. Although the total amount of MMP9 remained unchanged, a strong
reduction of the active MMP-9 (50 to 70%) was observed in IFNs/IFO treated COH tumors.,
We did not detected constitutive MMP9 activity in EW7 tumors; treatment with IFNs/IFO
result in a modest increase of total MMP9 protein without concomitant increase of the active
MMP-9. Since generation of the active enzyme is dependent of the proteolytic cleavage of
the proMMP9, these results suggest that the cleavage step is directly or indirectly the target of
IFNs/IFO combined therapy. Finally, IFN/IFO treatment triggered in both COH and EW7
tumors the down regulation of uPAR expression, a molecule which was recently shown to be
involved in vascularization and endothelial cell migration. Thus, our results explain at least in
part the mechanism involved in tumor growth inhibition by IFNs /IFO therapy and provide a
rational foundation for a promising therapeutic approach to Ewing sarcoma.

                                              - 29 -
Session II: Angiogenesis

b1 integrins as targets to inhibit tumor angiogenesis and carcinoma cell invasiveness

Czeslaw Cierniewski, Marcin Cieslak, Magda Nawrot, Izabela Papiewska, Jolanta

Department of Molecular and Medical Biophysics, Medical University of Lodz, Center of
Medical Biology and Microbiology, Polish Academy of Sciences, E-mail :

Integrins are known to mediate a variety of cellular events including adhesion, migration,
proliferation, invasiveness, and cellular survival that are crucial for tumor angiogenesis, tumor
growth and metastasis. It is becoming increasingly clear that particularly members of the b1
integrin subfamily play a critical role in all these processes. Recently, we developed a novel
approach to downregulate expression of b1 integrins in endothelial and carcinoma cells
(Cieslak et al., J. Biol. Chem. 277, 6779-87, 2002). We engineered DNAzyme (b1DE) to
cleave a specific site in b1 mRNA and thus inhibit expression of b1 integrin subunit at the
level of mRNA, protein synthesis, and at the cell surface. Its enzymatic efficiency was
compared with ribozyme to b1 mRNA constructed also for the same purpose. b1DE, in the
presence of Mg2+, specifically cleaved its substrate, a synthetic b1 mRNA fragment. b1DE
doubly labeled with fluorescein and rhodamine was used to study the cellular uptake,
intracellular distribution and stability. There was the punctate fluorescence distribution of
Fluo-DE-Rhod observed even after 24 hours indicating that endosomal vesicles are the
primary targets of the probes. In functional tests, b1DE markedly reduced adhesion of cells to
proteins (fibronectin, vitronectin, and collagen) and abolished microvascular endothelial cell
capillary tube formation in fibrin and Matrigel. Since the DE catalytic activity in vitro highly
depends on Mg2+ concentration, we analyzed its folding transitions induced by cations. For
this purpose spectroscopic analyses such as fluorescence resonance energy transfer,
fluorescence anisotropy, circular dichroism, and surface plasmon resonance measurements
were performed. The global geometry of the DNAzyme in the absence of added Mg2+ seems
to be essentially extended, it has no catalytic activity and shows a very low binding affinity to
its RNA substrate. The folding of the DNAzyme induced by binding of Mg2+ may occur in
several distinct stages. The first stage, observed at 0.5 mM Mg2+, corresponds to the
formation of a compact structure with limited binding properties and without catalytic
activity. Then, at 5 mM Mg2+ flanking arms are projected at right position and angles to bind
RNA. DNAzyme in such a state shows substantial binding to its substrate and significant
catalytic activity. Finally, the transition occurring at 15 mM Mg2+ leads to the formation of
the catalytic domain and DNAzyme shows high binding affinity towards substrate and
efficient catalytic activity. In several tests, the DNAzyme and its analogues appeared to be
strong inhibitors of adhesion and migration of malignant prostate carcinoma cells (PC3) and
various human colon adenocarcinoma cells (HT29, CX1.1, SW620, LS180). Furthermore, it
blocked the invasiveness of malignant cells through the barrier of matrigel matrix in in vitro

                                              - 30 -
Session II: Angiogenesis

Evaluation of the function of ADAMTS-2, a metalloproteinase containing a disintegrin
domain and thrombospondin type I repeats, during angiogenesis in vitro and in vivo

Frédéric Kesteloot, Vincent Lambert, Jean-Marie Rakic, Charles M. Lapière, Betty V.
Nusgens and Alain Colige.

Laboratory of Connective Tissues Biology - GIGA/CRCE - Tour de Pathologie B23/3,
University of Liège, B-4000 Sart Tilman, Belgium. E-mail:

Formation of new blood vessels (angiogenesis) is a key step during the development of
various pathologies, including cancer, explaining the extensive efforts made to identify and/or
develop angiogenesis inhibitors. Enzymes of the ADAMTS family are closely related to
MMPs and ADAMs. They further contain specific domains, such as the "ThromboSpondin
type I" (TSP1) repeats, that are able to strongly repress angiogenesis, as described for
ADAMTS-1 and –8. Although the primary function of ADAMTS-2 is the maturation of
collagen type I, II and III precursors by excising the amino-propeptide, we hypothesized that
it could also modulate angiogenesis through its TSP1 repeats. This was investigated in
different in vitro models such as cell proliferation and formation of capillary structures by
human endothelial cells. Mice and rat aorta rings were also used as an ex vivo angiogenesis
model. Outgrowth of capillaries was slightly increased from aortas of ADAMTS-2 KO mice
(TS2-/-) as compared to aortas from control animals (TS2+/+), while addition of full size
recombinant ADAMTS-2 reduced the formation of capillary structures from rat aortas,
suggesting its anti-angiogenic activity. Choroidal neovascularization induced in TS2+/+ or
TS2-/- mice by LASER burns was used as in vivo model to confirm the in vitro and ex vivo
results. Angiogenesis, visualized by confocal microscopy after FITC-conjugated dextran
injection, was significantly increased (p<0,05) in TS2-/- mice. Preliminary data indicate that
several genes involved in the healing and angiogenesis processes (collagens, VEGF, TGFb,
CTGF, …) are not differently regulated in TS2+/+ and TS2-/- mice at 5 days after LASER
impact. Additional investigations are being performed in order to determine which domain(s)
of ADAMTS-2 is (are) anti-angiogenic and to identify the underlying mechanism(s).

                                             - 31 -
Session II: Angiogenesis

Catalytic DNA Targeting Immediate-Early Genes as Novel Inhibitors of Angiogenesis

Levon M. Khachigian

Centre for Vascular Research, The University of New South Wales, Sydney NSW, Australia.

Nuclear regulatory factors that control the expression of genes whose products stimulate
vascular cell growth have become attractive targets for interventional strategies. DNAzymes
are RNA-cleaving phosphodiester-linked DNA-based enzymes that seek out and cleave their
target mRNA in a gene-specific fashion. DNAzymes targeting the immediate-early gene
product, early growth response-1 (Egr-1), for example, inhibit intimal thickening in rat carotid
arteries following balloon angioplasty (1), permanent ligation (2), and in-stent restenosis in
pigs after coronary stenting (3). Similarly, DNAzymes targeting c-Jun, a prototypic member
of the basic region-leucine zipper family of nuclear proteins inhibit inducible c-Jun expression
in vascular smooth muscle cells, block smooth muscle cell proliferation and attenuate intimal
thickening in injured rat carotid arteries (4). We have recently demonstrated that Egr-1 and c-
Jun DNAzymes are also capable of inhibiting their respective target genes in microvascular
endothelial cells. These agents block endothelial cell replication, migration and microtubule
network formation in vitro. Egr-1 DNAzymes block angiogenesis in subcutaneous matrigel
plugs in mice, observations independently confirmed by plug analysis in Egr-1-deficient
animals. Moreover, these molecules inhibit neovascularization of rat cornea and block solid
human tumor growth in nude mice (5). Thus, microvascular endothelial cell growth,
neovascularization, tumor angiogenesis and tumor growth are processes that may be
controlled by DNAzyme strategies targeting key (nuclear) factors. DNAzymes therefore
represent a new, efficient and versatile class of therapeutic agent.
Refs: (1) Santiago et al. 1999. Nature Med. 5:1264-1269; (2) Lowe et al. 2002. Thromb.
Haemost. 87:134-140; (3) Lowe et al. 2001. Circ. Res. 89:670-677; (4) Khachigian et al.
2002. J. Biol. Chem. 277:22985-22991; (5) Fahmy et al. 2003. Nature Med. 9:1026-1032.

                                             - 32 -
Session II: Angiogenesis

The human model: a genetic dissection of immunity to infection in natural conditions

Jean-Laurent Casanova and Laurent Abel

Laboratory of Human Genetics of Infectious Diseases, Necker Medical School, 75015 Paris,
France, EU; Tel 33 1 40 61 53 81; Fax 33 1 40 61 56 88; Email

Tremendous progress has been achieved in developmental, cellular, and molecular
immunology in the last twenty years, largely thanks to studies using the mouse as a model
system and the arrival of molecular genetics. Immunology is now faced with a difficult
challenge. What are the functions of the individual cells and molecules in achieving immunity
to infection? Renewed interest in animal models of disease has provided considerable insight
in this area, but such models of infection suffer from the inherent limitation of being
experimental. In man, the complex host-environment interaction occurs in natural, as opposed
to experimental, conditions. The human model is thus an indispensable complement to animal
models, as it allows an observational genetic dissection of immunity to infection.

                                            - 33 -
Session II: Angiogenesis


Lambrechts D1*, Storkebaum E1, Del-Favero J2, Marklund SL3, H4Shaw C5, Morrison KE4,
Robberecht W1, Van Broeckhoven C2, Collen D1, Andersen PM3 and Carmeliet P1
*e-mail correspondence:

The Center for Transgene Technology and Gene Therapy, Flanders Interuniversity Institute
for Biotechnology and Departments of Neurology, Gasthuisberg, KU Leuven, Leuven, B-
3000, Belgium; (2) Department of Molecular Genetics, Flanders Interuniversity Institute for
Biotechnology, University of Antwerp, B-2610 Antwerpen, Belgium; (3) Department of
Neurology, Umeå University Hospital, and Institute of Clinical Neuroscience and Medical
Genetics, Umeå University, S-901 85 Umeå, Sweden; (4) Department of Neurology, The
Medical School, University of Birmingham, Edgbaston, B15 2TT, United Kingdom; (5)
Departments of Neurology and Medical and Molecular Genetics; Guy's King's and St Thomas
School of Medicine and Institute of Psychiatry, King's College London, London SE5 8AF.
We recently reported (Nat Genet 28, 131-8, 2001) that Vegf ∂/∂ mice – which have reduced
Vegf levels due to a subtle targeted mutation in the promoter – develop adult-onset motor
neuron degeneration (MND) reminiscent of ALS. This was entirely unexpected as Vegf is the
prototype angiogenic factor and was never implicated in MND. Our study suggested that
reduced Vegf levels caused MND by reducing perfusion and/or insufficient Vegf-dependent
To extrapolate our findings that abnormal regulation of Vegf in Vegf ∂/∂ mice, contributes to
MND in human, we examined whether of single nucleotide polymorphisms (SNPs) in the
VEGF gene might confer an increased risk for MND in ALS patients. In a meta-analysis of
over 900 individuals from Sweden and, as a replication, over 1000 individuals from Belgium
and England, we now report that individuals homozygous for the at risk genotypes in the
promoter and leader sequence of the VEGF gene had a 1.8-fold increased risk of ALS
(P=0.00004). The following results provide evidence that the at-risk VEGF genotypes are
functionally affecting VEGF expression: (i) Plasma VEGF levels in ALS cases from Sweden,
and serum VEGF levels in control individuals from Belgium, carrying the at-risk genotypes
were lower compared to other genotypes; (ii) transfection studies using VEGF/reporter
plasmids revealed that at-risk genotypes lowered VEGF transcription by almost 50% under
normoxic and hypoxic conditions; (iii) the at-risk SNPs in the VEGF promoter reduced
transcriptional activity, while the at risk SNP in the VEGF leader sequence impaired the
IRES-mediated enhancement of VEGF translation and, most intriguingly, impaired the
production of a novel Large-VEGF iso-form (L-VEGF). In addition, VEGF plasma levels
were also lower in ALS patients than in healthy spouses from Sweden. Moreover, SOD1G93A
mice crossbred with Vegfd/d mice died earlier due to more severe motoneuron degeneration.
Vegf∂/∂ mice were unusually susceptible to persistent paralysis after spinal cord ischemia, and
treatment with Vegf protected mice against ischemic motoneuron death.
These findings indicate that VEGF is a modifier of MND in humans and mice and unveil a
therapeutic potential of VEGF for stressed motoneurons. Moreover, abnormal regulation of
VEGF expression has now been implicated in several other neurodegenerative disorders. This
has stimulated an increasing interest in assessing the therapeutic potential of VEGF as a
neuroprotective agent for neurodegenerative disorders.

                                             - 34 -
Session III : Chair : Abraham Amsterdam

             - 35 -
Session III: Apoptosis

Partial cleavage of RasGAP by caspase-3 is required for survival in mild stress

Jiang-Yan Yang*, David Michod*, Joël Walicki*, Brona M. Murphy†, Shailaja Kasibhatla#,
Seamus J. Martin†, and Christian Widmann*
   Institut de biologie cellulaire et de morphologie (IBCM) [http://www-], Biology and Medicine Faculty, Lausanne University,
Switzerland; †Molecular Cell Biology Laboratory, Department of Genetics, The Smurfit
Institute, Trinity College, Dublin, Ireland; # Maxim Pharmaceuticals, San Diego, California,

Ectopic expression of the N-terminal fragment resulting from the partial cleavage of RasGAP
by caspases results in the activation of the anti-apoptotic Ras-PI3K-Akt pathway. Here we
demonstrate that this fragment is generated in a caspase-3-dependent manner in stressed cells
that do not undergo apoptosis. Subjected to the same low stress conditions, cells, in which the
wild-type RasGAP protein was replaced with an uncleavable mutant, could not activate Akt,
could not prevent an amplification of the caspase-3 activity, and eventually underwent
apoptosis. Formation of fragment N in response to low caspase-3 activation is therefore
critical for survival of stressed cells. Our results indicate that executioner caspases control the
extent of their own activation by a feedback regulatory mechanism initiated by the partial
cleavage of RasGAP.

                                               - 36 -
Session III: Apoptosis


Joseph A. Trapani, Michelle Wowk, Kylie A. Browne, Kevin Thia, Vivien R. Sutton and
Nigel Waterhouse.

Cancer Immunology Program, Peter MacCallum Cancer Centre, Locked Bag 1, A’Beckett
Street, Melbourne 8006, AUSTRALIA

Granzyme B is a serine protease secreted from cytotoxic T cells and natural killer cells, cells
of the immune system that eliminate virus-infected and transformed cells. Granzyme B is
released with a battery of other toxins with which it synergises to bring about target cell
death. In particular, granzyme B’s potency is completely dependent on a second molecule, a
membrane disrupting toxin named perforin which is necessary for granzyme B’s trafficking to
the cytosol of the target cell. Granzyme B is a pro-apoptotic protease in that it cleaves
substrates adjacent to acidic residues, particularly aspartic acid. It is therefore an exogenous
serine protease that is introduced into target cells and mimics the activity of endogenous
caspases (ubiquitous cysteine proteases involved in cell suicide).

Many aspects of granzyme B function and biology have remained controversial including the
nature of its uptake into target cells, the molecular and cellular basis for its interactions with
perforin, the precise pathways to apoptosis activated within the target cell and its
physiological relevance to virus infection and tumor immunity. In this presentation, recent
advances in several of these areas of research will be presented. Firstly, we have recently
shown that granzyme B enters cells through two distinct endocytic pathways, the first
mediated by binding to the mannose-6-phosphate receptor (MPR), and the second through
fluid phase uptake. Thus, cells deficient in MPR expression remain susceptible to granzyme
B, and transplants of MPR-deficient tumors are still avidly rejected in vivo. Once in the cell
cytosol, it has long been assumed that granzyme B processes and directly activates caspases.
However recent evidence from several groups indicates that granzyme B is very inefficient in
directly activating caspases in intact cells. Rather, granzyme B induces death through the
mitochondrial apoptosis pathway by efficiently processing the pro-apoptotic molecule Bid.
Truncated Bid then activates the release from mitochondria of other pro-apoptotic molecules
such as Smac/Diablo and HtrA2/Omi, leading to full-blown caspase activation and cell death.
Neither mitochondrial outer membrane permeabilization resulting in release of apoptotgenic
mediators nor cell death required activated caspases. However, the sustained loss of
mitochondrial membrane potential did require active caspase-3. The full processing of
procaspase-3 and other caspases involves the displacement of inhibitor of apoptosis molecules
(IAPs) from partially processed procaspases, permitting their auto-activation and attainment
of full caspase function.

                                              - 37 -
Session III: Apoptosis

Phosphorylation and inhibition of caspase-9 by protein kinase signalling pathways

Paul. R. Clarke, Lindsey A. Allan, Suzanne C. Brady and Morag C. Martin

Biomedical Research Centre, Ninewells Hospital and Medical School, University of Dundee,
Dundee DD1 9SY, Scotland, UK

Many apoptotic stimuli act through release of cytochrome c from mitochondria and activation
of caspase-9, an initiator protease, by Apaf-1. Caspase-9 in turn activates caspase-3 and other
downstream caspases to bring about cell death. This apoptotic response can be influenced by
a variety of signalling pathways. Activation of the Ras-Raf-MEK-ERK mitogen activated
protein kinase (MAPK) pathway is associated with protection of cells from apoptosis and
inhibition of caspase-3 activation. We have found that the ERK-MAPK pathway inhibits
caspase activation by directly phosphorylating caspase-9 (Allan et al., 2003). In a human cell-
free system, the protein phosphatase inhibitor okadaic acid inhibits activation of caspase-3 in
response to exogenous cytochrome c. Under these conditions, caspase-9 is phosphorylated at
Thr 125, a MAPK consensus site. This site is also phosphorylated efficiently by ERK-2 in
vitro. Using a phospho-specific antibody, we show that caspase-9 is phosphorylated at Thr125
in cells when the ERK MAPK pathway is stimulated by EGF or TPA. Phosphorylation of
caspase-9 Thr125 is sufficient to inhibit both the activation of caspase-9 and the subsequent
activation of caspase-3. We suggest that phosphorylation and inhibition of caspase-9 by ERK
MAPK promotes cell survival during development and tissue homeostasis. This mechanism
may also contribute to tumorigenesis when the ERK MAPK pathway is constitutively
activated. Phosphorylation of caspase-9 at additional sites in response to activation of distinct
kinase signalling pathways suggests that caspase-9 is a focal point for diverse signals that
regulate apoptosis and determine cell survival.

Allan et al. (2003) Nature Cell Biol. 5, 647-654.

This work was supported by Cancer Research UK and the Medical Research Council. PRC is
a Cancer Research UK Senior Research Fellow and a Royal Society-Wolfson Research Merit
Award holder.

                                              - 38 -
Session III: Apoptosis

Positional cloning of Lps2, a key transducer of MyD88-independent TIR signaling

K. Hoebe, X. Du, P. Georgel, E. Janssen†, K. Tabeta, S.O. Kim, J. Goode, N. Mann, S. Mudd,
K. Crozat, S. Sovath, J. Han and B. Beutler*

        Induced by N-ethyl-N-nitrosourea mutagenesis, the Lps2 phenotype is a MyD88-
independent TLR signaling defect, characterized by absence of responses to dsRNA and
severe impairment of responses to LPS. The existence of Lps2 suggested that TLR3 and
TLR4 might share a proximal transducer. We have identified the Lps2 mutation as being a
distal frameshift error in a TIR adapter protein known as TRIF or TICAM-1. In response to
LPS and/or dsRNA, macrophages derived from TrifLps2 homozygotes show an early activation
of MAP kinases and NFk-B and fail to activate IRF3 or secrete IFN-b. Furthermore, TrifLps2
homozygotes are markedly resistant to the toxic effects of LPS, and hypersusceptible to
mouse cytomegalovirus. Compound homozygosity for mutations at Trif and MyD88 loci
ablates all responses to LPS, indicating that two and only two signaling pathways emanate
from the LPS receptor.

                                          - 39 -
Session III: Apoptosis

Inhibition of poly(ADP-ribosylation): a strategy for modulating the apoptotic response
in tumoral cells

A. Ivana Scovassi

Istituto di Genetica Molecolare CNR, 27100 Pavia, Italy. E-mail:

Poly(ADP-ribosylation) is a post-translational modification of proteins playing a crucial role
in many processes, including transcription, DNA repair and replication. The best known
poly(ADP-ribosylating) enzime, PARP-1, is a DNA nick-sensor and uses NAD+ to form
poly(ADP-ribose), which is further degraded by the enzyme poly(ADP-ribose)
glycohydrolase (PARG). During apoptosis, PARP-1 is responsible for the rapid and early
synthesis of polymers of (ADP-ribose), which may allow cellular NAD depletion. To prevent
intracellular energy loss, PARP-1 is cleaved by caspases into an 89 kDa fragment, with a
reduced catalytic activity, and a 24 kDa peptide, which retains the DNA binding domains.
The p89 PARP-1 fragment migrates from the nucleus into the cytoplasm of cells with
advanced chromatin condensation and DNA fragmentation, possibly playing a role in
autoimmune reactivity.
Although poly(ADP-ribosylation) may be beneficial as a cellular emergency reaction, a high
level of PARP-1 activation under damage conditions can be detrimental to the cells because
of energy depletion and consequent occurrence of necrosis. Cell death due to severe NAD+
depletion plays a major pathogenetic role for many diseases in animal models.
Pharmacological inhibition of PARP-1 exerts a protective effect toward a number of
pathological conditions, including cerebral ischemia, neurodegenerative diseases, myocardial
injury, and diabetes, and could represent a novel therapeutical strategy also against tumors.
The potential applications of PARP-1 and PARG inhibitors as
chemosensitizing/radiosensitizing agents in cancer therapy will be discussed.

                                            - 40 -
Session IV : Chair : Dean Felsher

     - 41 -
Session IV : Cancer specific signaling pathways as therapeutic targets

Cancer revoked: oncogenes as therapeutic targets.

Dean W. Felsher, MD PhD

Division of Oncology, Department of Medicine and Pathology, Stanford University

Cancer can largely be conceived as a consequence of genomic catastrophes resulting in
genetic events that usurp physiologic function of a normal cell. These genetic events mediate
their pathologic effects by activating oncogenes or inactivating tumor suppressor genes. The
targeted repair or inactivation of these damaged gene products may counteract the effects of
these genetic events, reversing tumorigenesis and thereby serve as an effective therapy for
cancer. However, since cancers are caused by many genetic events, the inactivation of no
single mutant gene product may be sufficient to reverse cancer. Despite this caveat,
compelling recent evidence suggests that there are circumstances when even the brief
interruption of activation of a single oncogene can be sufficient to reverse tumorigenesis.
Utilizing conditional transgenic model systems, we have defined circumstances when the
inactivation of the MYC oncogene is sufficient to induce sustained regression of tumors.
Moreover, we have defined genetic events that are associated with the ability of these tumors
to become independent of MYC to sustain their neoplsatic properties. Using these model
systems, we should be able to determine how and when oncogene inactivation reverses
cancer, which will be important in both defining the molecular pathogenesis of cancer as well
as developing new molecularly based treatments.

                                            - 42 -
Session IV : Cancer specific signaling pathways as therapeutic targets

Raf Kinase Inhibitor Protein (RKIP) is a prostate cancer metastasis suppressor gene.

Evan T. Keller, DVM, PhD

University of Michigan, Ann Arbor, MI 48109-0940

Background: Gene array analysis revealed that expression of Raf kinase inhibitor protein
(RKIP), an inhibitor of Raf-mediated MEK activation, is downregulated in a metastatic (C4-
2B) compared to its non-metastatic prostate cancer (CaP) parental cell line (LNCaP). In this
study, we determined if RKIP is a metastasis suppressor gene.                       Methods:
Immunohistochemistry was used to identify RKIP expression in clinical primary CaP and CaP
metastases. LNCaP and C4-2B cells were stably transfected with antisense and sense RKIP
cDNA, respectively. Cell proliferation, soft agar colony formation, and in vitro invasion
assays were performed to evaluate the transfected cells’ malignant phenotypes. The effect of
restoring RKIP expression in C4-2B on the development of spontaneous metastasis was
evaluated in an orthotopic murine model. In vitro invasion assay was performed in the
presence or absence of a MEK inhibitor. Results: RKIP was detectable in primary CaP but
not metastases. Overexpression of RKIP in C4-2B cells had no effect on in vitro
proliferation, colony formation, or primary tumor growth in vivo, but diminished in vitro
invasion and inhibited development of lung metastases in vivo, in association with decreased
vascular invasion in the primary tumor. Chemical inhibition of MEK diminished in vitro
invasion. Conclusions: These results demonstrate that RKIP is a clinically relevant novel CaP
metastasis suppressor gene that suppresses metastasis possibly through decreasing vascular
invasion. Implications: Diminished expression of a signal transduction pathway inhibitor is a
novel mechanism for promotion of metastasis. This study suggests that targeting the
MEK/ERK will have beneficial effects to prevent CaP metastasis.

                                            - 43 -
Session IV : Cancer specific signaling pathways as therapeutic targets

Targeting of proteins to membranes through hedgehog auto-processing

John Benson

Novartis Institutes for Biomedical Research, Cambridge, MA, US

Hedgehog proteins utilize a novel autoprocessing strategy to generate cholesterol- conjugated
peptide products that act as extracellular ligands in a number developmental signaling
pathways. We have developed and implemented a strategy for use of the hedgehog
autoprocessing reaction to carry out intracellular modification of heterologous peptides and
proteins, resulting in their localization to cell membranes. Such processing occurs
spontaneously, without the need of accessory proteins or modification by other enzymes.
Hedgehog processing of GFP results in its localization to lipid rafts, cholesterol-rich
subdomains of the plasma membrane that are rich in signaling proteins and are also involved
in viral entry and other processes. As further proof-of-principle, this system was also used to
process a constitutively active form of H-Ras (Val12) that lacked its naturally occurring lipid
modification signals restored its fibroblast transforming activity. This system can also
potentially be used to generate and phenotypically screen peptide libraries that are directed to
lipid rafts through this processing reaction. Potential strategies for these applications will be

                                              - 44 -
Session IV : Cancer specific signaling pathways as therapeutic targets

Convergence of the p53 and Smad signaling networks

Stefano Piccolo

Padoa, Italy

The p53 tumor suppressor belongs to a family of proteins that sense multiple cellular inputs to
regulate cell proliferation, apoptosis and differentiation. Whether and how these functions of
p53 intersect with the activity of extracellular growth factors is not understood. Here we
report that key cellular responses to TGF-b signals rely on p53 family members. During
Xenopus embryonic development, p53 promotes the activation of multiple TGF-b target
genes. Moreover, mesoderm differentiation is inhibited in p53-depleted embryos. In
mammalian cells, the full transcriptional activation of the CDK inhibitor p21WAF1 by TGF-b
requires p53. p53-deficient cells display an impaired cytostatic response to TGF-b signals.
Smad and p53 protein complexes converge on separate cis-binding elements on a target
promoter and synergistically activate TGF-b induced transcription. p53 can physically interact
in vivo with Smad2 in a TGF-b-dependent fashion. The results unveil a previously
unrecognized link between two primary tumor suppressor pathways in vertebrates.

                                             - 45 -
Session IV : Cancer specific signaling pathways as therapeutic targets

YC-1: a potential anticancer drug targeting hypoxia-inducible factor 1

Eun-Jin Yeo, Yang-Sook Chun, Young-Suk Cho, Jinho Kim, June-Chul Lee,
Myung-Suk Kim, Jong-Wan Park

Department of Pharmacology, BK21 Human Life Sciences, Seoul National University
College of Medicine, 28 Yongon-dong, Chongno-gu, Seoul 110-799, Korea. E-mail :

Hypoxia-inducible factor 1alpha (HIF-1a), a component of HIF-1, is expressed in human
tumors, and under hypoxic conditions, renders cells with the ability to survive and stimulates
endothelial cell growth. YC-1 inhibits platelet aggregation and vascular contraction and also
inhibits HIF-1 activity in vitro. Here, we tested whether YC-1 inhibits HIF-1 and tumor
growth in vivo. Hep3B hepatoma, NCI-H87 stomach carcinoma, Caki-1 renal carcinoma,
SiHa cervical carcinoma, and SK-N-MC neuroblastoma cells were grown as xenografts in the
flanks of immunodeficient mice. After the tumors were 100–150 mm3, mice received daily
intraperitoneal injections of the vehicle or YC-1 (30 mg/g) for 2 weeks. HIF-1a protein and
vascularity in tumors were assessed by immunohistochemistry and HIF-1-inducible gene
expression by reverse-transcription polymerase chain reaction. All statistical tests were two-
sided. Tumors from YC-1-treated mice were statistically significantly smaller, expressed
lower levels of HIF-1a, were less vascular, and expressed lower levels of HIF-1-inducible
genes than tumors from vehicle-treated mice. The inhibition of HIF-1a in tumors from YC-1-
treated mice is associated with blocked angiogenesis and an inhibition of tumor growth. YC-1
has the potential to become the first antiangiogenic, anticancer agent to target HIF-1a.

                                            - 46 -
Session V : Chair : Bharat Aggarwal

        - 47 -
Session V: Chemopreventive agent

"Targeting Transcription Factors for Prevention and Therapy of Cancer By

Bharat B. Aggarwal, Ph.D.

Cytokine Research Laboratory, Department of Bioimmunotherapy, The University of Texas
M. D. Anderson Cancer Center, Houston, Texas 77030 , U.S.A.

        NF-kB, a transcription factor, is present normally in the cytoplasm as an inactive
heterotrimer consisting of p50, p65 and IkBa subunits. When activated, NF-kB translocates
to the as a p50-p65 heterodimer. This factor regulates the expression of various genes that
control apoptosis, viral replication, tumorigenesis, various autoimmune diseases, and
inflammation. NF-kB has been linked to the development of carcinogenesis for several
reasons. First, various carcinogens and tumor promoters have been shown to activate NF-kB.
Second, activation of NF-kB has been shown to block apoptosis and promote proliferation.
Third, the tumor microenvironment can induce NF-kB activation. Fourth, constitutive
expression of NF-kB is frequently found in tumor cells. Fifth, NF-kB activation induces
resistance to chemotherapeutic agents. Fifth, several genes involved in tumor initiation,
promotion, and metastasis are regulated by NF-kB. Sixth, various chemopreventive agents
have been found to downregulate the NF-kB activation. All these observation suggest that
NF-kB could mediate tumorigenesis and thus can be used as a target for chemoprevention and
for the treatment of cancer. Besides NF-kB, we have also targeted AP-1 and STAT3, other
transcription factors that mediate tumorigenesis. We will present the data which shows that
phyochemicals are important inhibitors of NF-kB, AP-1 and STAT3 activation, and can
suppress the expression of genes involved in carcinogenesis and tumorigenesis in vivo.

Estrov Z., Shishodia S., Faderl S., Harris D., Van Q., Kantarjian H.M., Talpaz M. and Aggarwal B. B.
Reserveratrol blocks interlekin-1[[beta]]-induced activation of the nuclear transcription factor NF-[[kappa]]B,
inhibits proliferation, causes S-phase arrest, and induces apoptosis of acuate myeloid leukemia cells, BLOOD,
2003 Aug 1;102(3):987-995.
Bharti AC, Aggarwal BB. Chemopreventive Agents Induce Suppression of Nuclear Factor-kappaB Leading to
Chemosensitization. Ann N Y Acad Sci. 2002 Nov; 973:392-395.
Shishodia S., Potdar, P., Gairola, C. G., and Aggarwal B. B., Curcumin (Diferuloylmethane) Downregulates
Cigarette Smoke -Induced NF-kB Activation Through Inhibition of IkBa Kinase in Human Lung Epithelial
Cells: Correlation with Suppression of COX-2, MMP-9 and Cyclin D1. Carcinogenesis Vol. 24, No. 7, 1269-
1279, 2003
Bharti A.C., Donato N., and Aggarwal B.B., Curcumin (Diferuloylmethane) Inhibits Constitutive and
Interleukin-6-Inducible STAT3 Phosphorylation in Human Multiple Myeloma Cells. Journal of Immunology,
2003, 171: pp3863-3871
Shishodia S., Majumdar S., Banerjee S., and Aggarwal B. B. , Ursolic Acid Inhibits Nuclear Factor-kB
Activation Induced by Carcinogenic Agents Through Suppression of IkBa Kinase and p65 Phosphorylation:
Correlation with Downregulation of COX2, MMP-9 and CyclinD1, Cancer Research 2003 Aug 1;63(15):4375-
Bharti AC and Aggarwal B. B. Nuclear Factor-kappa B and Cancer: Its Role in Prevention and Therapy. 2002
Biochemical Pharmacology 64, 2002, 883-888
Garg A. and Aggarwal B.B. Nuclear Transcription Factor-kB as a Target for Cancer Drug Development. 2002.
Leukemia 16, 1053-1068
Takada Y. and Aggarwal B. B., Betulinic acid Suppresses Carcinogen-Induced Nuclear Factor-kB Activation
Though Inhibition of IkBa Kinase and p65 Phosphorylation: Abrogation of Cyclooxygenase-2 and Matrix
Metalloprotease-9, Journal of Immunology, 2003, 171, 3728-3286

                                                    - 48 -
Aggarwal B.B. Kumar A., Bhart A.C. Anticancer Potential of Curcumin: Preclinical and Clinical Studies.
Anticancer Research 23:363-398, 2003
Manna S.K., Bueso-Ramos, C., Francisco Alvarado F., and Aggarwal B.B. Calagualine Inhibits Nuclear
Transcription Factors-kB Activated by Various Inflammatory and Tumor Promoting Agents. Cancer Lett 2003,
190, 171-182

Bharti, A. C., Donato, N., Singh, S. and Aggarwal, B.B. Curcumin (diferuloylmethane) downregulates the
constitutive activation of nuclear factor-kB and IkBa kinase in human multiple myeloma cells leading to
suppression of proliferation and induction of apoptosis BLOOD 101, 2003, 1053-1062

Mukhopadhyay, A., Banerjee, S., Stafford, L.J., Xia, Chunzhi, X., Liu, M. and Aggarwal, B.B. Curcumin-
induced suppression of cell proliferation correlates with donregulation of cyclin D1 expression and CDK4-
mediated retinoblastoma protein phosphorylation. ONCOGENE. 2002 Dec 12; 21 (57):8852-61.

Ashikawa K, Majumdar S, Banerjee S, Bharti AC, Shishodia S, Aggarwal BB. Piceatannol Inhibits TNF-
Induced NF-kappaB Activation and NF-kappaB-Mediated Gene Expression Through Suppression of
IkappaBalpha Kinase and p65 Phosphorylation. J Immunol. 2002 Dec 1;169(11):6490-7.

Bharti AC, Aggarwal BB. Chemopreventive Agents Induce Suppression of Nuclear Factor-kappaB Leading to
Chemosensitization. Ann N Y Acad Sci. 2002 Nov; 973:392-395.

Garg, A.K. and Aggarwal, B.B. Reactive oxygen intermediates in TNF signaling. Molecular Immunology
Volume 39, Issue 9, December 2002, Pages 509-517.

Aggarwal, B. B., Shishodia, S., Ashikawa, K., and Bharti, A. C.. The Role of TNF and Its Family Members in
Inflammation and Cancer: Lessons from Gene Deletion; Current Drug Targets in Inflammation and Allergy . 1,
2002, 327-342

Anto R. J., Mukhopadhyay A. , Gairola C. G. and Aggarwal B. B. Cigarette Smoke Condensate Activates
Nuclear Transcription Factor-kB Through Phosphorylation and Degradation of IkBa: Its Role in Induction of
Cyclooxygenase-2, Carcinogenesis. 23, 1511-1518, 2002
Bharti AC and Aggarwal B. B. Nuclear Factor-kappa B and Cancer: Its Role in Prevention and Therapy. 2002
Biochemical Pharmacology 64, 2002, 883-888
Garg A. and Aggarwal B.B. Nuclear Transcription Factor-kB as a Target for Cancer Drug Development. 2002.
Leukemia 16, 1053-1068

Banerjee S., Bueso-Ramos C.and Aggarwal B.B., Suppression of 7, 12 Dimethyl-Benz(a)anthracene-Induced
Mammary Carcinogenesis in Rats by Resveratrol: Role of Nuclear Factor-kB, Cyclooxygenease 2 and Matrix
Metalloprotease-9. Cancer Research 2002 Sep 1;62 (17):4945-54.
Anto R. J., Mukhopadhyay A., Denning K., and Aggarwal B.B., Curcumin (Diferuloylmethane) Induces
Apoptosis Through Activation of Caspase-8, BID cleavage and Cytochrome C Release: Its suppression by
Ectopic Expression of Bcl-2 and Bcl-xL, Carcinogenesis ; 23 (1):143-50, 2002
Mukhopadhyay A, Bueso-Ramos C, Chatterjee D, Pantazis P and Aggarwal BB: Curcumin Downregulates Cell
Survival Mechanisms in Human Prostate Cancer Cell Lines. ONCOGENE 20, 7597-7609, 2001)
Aggarwal B. B., Ahmed N. And Mukhtar H., Spices as potent antioxidants and their therapeutic potential; in
Handbook of Antioxidants, (eds. Cadena E and Packer L.); Marcel Dekker Inc. 2nd (in press)

Singh S. and Aggarwal B.B., Activation of transcription factor NF-kB is suppressed by curcumin
(Diferulolylmethane) J. Biol. Chem. 270: 24995-25000, 1995

Chainy GBN, Manna SK, Chaturvedi MM, and Aggarwal BB. Anethole Blocks Both Early and Late Cellular
Responses Transduced by Tumor Necrosis Factor: Effect on NF kB, AP-1, JNK, MAPKK and Apoptosis.
Oncogene 19, 2943-2950, 2000.

Manna SK, Sah NK, Newman RA, Cisneros A, and Aggarwal BB., Oleandrin Suppresses Activation of
Nuclear Transcription Factor-kB, Activator Protein-1, and c Jun N-Terminal Kinase, Cancer Research 60, 3838-
3847, 2000.

                                                   - 49 -
Manna SK, and Aggarwal BB., All-trans-Retinoic Acid Upregulates TNF Receptors and Potentiates TNF-
Induced Activation of Nuclear Factors-kB, activated protein-1, and apoptosis in human lung cancer cells
Oncogene 19, 2110-2119, 2000

Manna SK, Mukhopadhyay A., and Aggarwal BB., Resveratrol Suppresses TNF Induced Activation of Nuclear
Transcription Factors NF- kB, Activator Protein-1, and Apoptosis: Potential role of reactive oxygen
intermediates and lipid peroxidation J. Immunol. 164, 6509-6519, 2000

S. K., Mukhopadhyay A. Van N. T., and Aggarwal B. B. Silymarin Suppresses TNF-Induced Activation of
Nuclear Transcription Factor-k B , c-Jun N-Terminal Kinase and Apoptosis J. Immunol. 163, 6800-6809, 1999

Manna S., Gad Y., Mukhopadhyay A. and Aggarwal B. B., Suppression of Tumor Necrosis Factor-activated
Nuclear Transcription Factor-kB, Activator Protein-1, c-Jun N-terminal kinase and Apoptosis by b-Lapachone
Biochem. Pharmacol. 57, 763-774, 1999.

Kumar K., Dhawan S., and Aggarwal B. B., Emodin (3-methyl-1,6,8 trihydroxyanthraquinone) inhibits the TNF-
induced NF- kB activation, IkB degradation and expression cell surface adhesion protein in human vascular
endothelial cells. Oncogene 17, 913-918, 1998.

Kumar A., Dhawan S., Hardegen N. J. and Aggarwal B. B. Curcumin (Diferuloylmethane) inhibition of TNF-
mediated adhesion of monocytes to endothelial cells by suppression of cell surface expression of adhesion
molecules and of nuclear factor-kB activation, Biochem. Pharmacol. 55, 775-783, 1998.

Chaturvedi M., Kumar A., Darnay B., Chainy G. B. N., Agarwal S. and Aggarwal B. B. Sanguinarine
(Pseudochelerythrine) is a potent inhibitor of NF-kB activation, IkBa phosphorylation, and degradation J. Biol.
Chem. 272, 30129-30134, 1997.

Mehta K., Pantazis P., McQueen T. and Aggarwal B. B. Curcumin (Diferuloylmethane) is a potent
antiproliferative agent against human breast tumor cell lines. Anti-Cancer Drugs. 8, 470-481, 1997.

Singh, S., Natarajan, K., and Aggarwal, B. B. Capsaicin (8-methyl-N-vanillyl-6 nonenamide) Is a Potent
Inhibitor of Transcription Factor NF-kB Activation by Diverse Agents. J. Immunol. 157: 4412-4420, 1996.

Natarajan K., Singh S., Burke Jr. T.R., Grunberger D. and Aggarwal B. B. Caffeic acid phenethyl ester (CAPE)
is a potent and specific inhibitor of activation of nuclear transcription factor NF-kB. Proc. Natl. Acad. Sci.
U.S.A. 93: 9090-9095, 1996.

Reddy, S. and Aggarwal, B.B. Curcumin is a non-competitive and selective inhibitor of phosphorylase kinase.
FEBS Lett, 341(1):19-22, 1994.

                                                    - 50 -
Session V: Chemopreventive agent


Young-Joon Surh

College of Pharmacy, Seoul National University
Seoul 151-742, South Korea. E-mail:

Chemoprevention refers to the use of agents to inhibit, reverse, or retard tumorigenesis.
Numerous phytochemicals present in edible plants have been reported to interfere with a
specific stage of the carcinogenic process. Some antioxidative and anti-inflammatory
substances derived from dietary or medicinal plants exert chemopreventive properties by
targeting intracellular signaling molecules or events. Curcumin, a yellow colouring agent
contained in turmeric (Curcuma longa L., Zingiberaceae), has been reported to possess strong
anti-tumor promotional as well as anti-inflammatory and antioxidant activities. Recent
studies from this laboratory have revealed that curcumin inhibits expression of
cyclooxygenase-2 (COX-2) in mouse skin treated with the tumor promoter 12-O-
tetradecanoylphorbol-13-acetate (TPA) through inactivation of the redox-sensitive eukaryotic
transcription factor NF-kB. Inhibition of NF-kB by curcumin appears to be mediated by
blocking ERK1/2 and p38 MAP kinases. [6]-Gingerol, a pungent ingredient present in ginger
(Zingiber officinale Roscoe, Zingiberaceae), inhibited TPA-induced tumor necrosis facror-
alpha production, ornithine decarboxylase activity, and skin tumor promotion in female ICR
mice. Its anti-tumor promoting effects appears to be associated with inhibition of p38 MAP
kinase and of subsequent phosphorylation of p65 subunit of NF-kB activation. Capsaicin, a
major pungent priniciple of hot chili pepper (Capsicum annuum L., Solanaceae) also
suppressed TPA-induced activation of NF-kB and AP-1 as well as tumor promotion in mouse
skin in vivo. The soy isoflavone genistein inhibits COX-2 induction in TPA- and TNF-alpha-
stimulated human mammary epithelial cells by inactivating ERK1/2 and NF-kB. Resveratrol,
a phytoalexin present in grapes and red wine, attenuated TPA-induced expression of COX-2
and activation of NF-kB in mouse skin. The green tea polyphenol epigallocatechin 3-gallate
(EGCG) inhibited activation of NF-kB and AP-1 in the TPA-stimulated human mammary
epithelial cell line. Under the same experimental conditions, EGCG suppressed COX-2
induction, while it upregulated heme oxygenase-1 (HO-1). The molecular basis of reciprocal
regulation of COX-2 and HO-1 by EGCG is under investigation.

                                           - 51 -
Session V: Chemopreventive agents

Induction of apoptosis by curcumin: mediation by glutathione S-transferase P1-1

Annelyse Duvoix*, Franck Morceau*, Sylvie Delhalle*, Martine Schmitz*, Michaël
Schnekenburger*, Marie-Madeleine Galteau†, Mario Dicato* and Marc Diederich*°

* Laboratoire de Recherche sur le Cancer et les Maladies du Sang (RCMS), Centre
Universitaire de Luxembourg, 162A, avenue de la Faïencerie, L-1511 Luxembourg,
† "Thiols et fonctions cellulaires" Faculté de Pharmacie 30, rue Lionnois - F-54000 NANCY,

Expression of glutathione S-transferase P1-1 (GSTP1-1) is correlated to carcinogenesis and
resistance of cancer cells against chemotherapeutic agents. Curcumin, a natural compound
extracted from Curcuma longa, has shown strong antioxidant and anticancer properties and
also the ability to regulate a wide variety of genes that require Activating Protein 1 (AP-1)
and Nuclear Factor kB (NF-kB) activation. In the present study, we examined the inhibitory
effect of curcumin on the expression of GSTP1-1 mRNA as well as protein and we correlated
this inhibition with the apoptotic effect of curcumin on K562 leukemia cells. Curcumin
efficiently inhibited the Tumor Necrosis Factor a (TNFa)- and phorbol ester-induced binding
of AP-1 and NF-kB transcription factors to sites located on the GSTP1-1 gene promoter.
TNFa-induced GSTP1-1 promoter activity was also inhibited by curcumin as shown by
reporter gene assay. In parallel, curcumin induced pro-caspases 8 and 9 as well as Poly ADP
Ribose Polymerase (PARP) cleavage and thus leading to apoptosis in K562 cells. Our results
overall add a novel role for curcumin as this chemoprotective compound could contribute to
induce apoptosis by its ability to inhibit the GSTP1-1 expression at the level of transcription.

                                             - 52 -
Session V: Chemopreventive agents

Transport of resveratrol , a cancer chemopreventive agent : plasmatic protein binding,
uptake and cellular targets

Brigitte JANNIN, Dominique DELMAS, Allan LANCON, Matthias MENZEL, Jean-Pierre

Laboratoire de Biologie Moléculaire et Cellulaire, Université de Bourgogne, 6
boulevardGabriel, 21000 Dijon, France; E-mail,,,, mamenzel@rhrk.uni-,,

Resveratrol chemopreventive properties towards cardiovascular and neurodegenerative
diseases and also cancer are well documented but the understanding of its
preventive/therapeutic action mechanism imply to elucidate the molecular steps involving : I)
binding to plasmatic proteins , II) cell uptake and III) targets. Concerning the step I, we have
studied the binding of resveratrol in particular by exaltation of resveratrol fluorescence and by
quenching of protein intrinsic fluorescence. It appears that resveratrol interacts with calf
serum proteins. It is bound especially to albumin, this binding being enhanced in the presence
of fatty acids. These results support a role played by albumin in the plasmatic transport of
resveratrol for delivery to the tissues. For the step II, the hepatoblastoma cell uptake of
resveratrol (a fluorescent polyphenol) was followed by fluorescence microscopy. Resveratrol
appeared to be distributed among whole cell excepted nucleus but visible in nucleoles. By
incubation with tritiated resveratrol, we found that uptake is maximal after 10 minutes; the
temperature- and dose- dependencies of this uptake suggest that it involves both passive
diffusion and carrier mediated transport. For step III, data concerning cell targets have shown
that resveratrol exerts a potent inhibition of cell proliferation of Hep G2 cells with a cell cycle
arrest to G2-M transition (Delmas et al., Oncol. Rep. 2000, 7: 847-852) and of human
colorectal tumor cells SW 480 (Delmas et al, Int. J. Mol. Med. 2002, 10: 193-199), with also
an apoptotic effect, associated with Fas redistribution in the rafts and the formation of a death
inducing signaling complex (Delmas et al, J. Biol. Chem. 2003, 278: 41481-41489).
        These results need to be discussed in a physiological context.

 Supported by the “Conseil Régional de Bourgogne”, BIVB, ONIVINS and the”Ligue
Bourguignonne contre le Cancer.

                                               - 53 -
Session V: Chemopreventive agent

Effects of Deguelin on the Phosphatidylinositol 3-Kinase/Akt Pathway and Apoptosis in
Premalignant Human Bronchial Epithelial Cells

Ho-Young Lee

Thoracic Head and Neck Medical Oncology, M.D.Anderson Cancer Center, Houston, US

Background: In the United States, lung cancer leads all other cancers in both incidence and
mortality rate. The lack of effective and safe agents for therapy of lung cancer, the high
proportion of patients with advanced disease at the time of diagnosis, and the rapidity of
tumor progression are major contributors to lung cancer mortality. Cancer chemoprevention
is, therefore, a logical and obvious strategy to help alleviate this disease. This study evaluated
the potential of deguelin, a natural product isolated from Mundulea serica (Leguminosae), as
a lung cancer chemopreventive agent. In addition, the mechanism of action of deguelin in
premalignant and malignant HBE cells was studied. Methods: The effects of deguelin on
proliferation and apoptosis were assessed in in vitro lung carcinogenesis model, which is
composed of normal, premalignant, and malignant HBE cell lines, by using the MTT assay,
flow cytometry, DNA fragmentation, western blot analysis, and immune complex kinase
assay. The specific effect of PI3K/Akt on deguelin-induced apoptosis was examined using an
adenovirus expressing constitutively active Akt. Results: Deguelin inhibited the growth of
premalignant and malignant HBE cells by causing a G2/M arrest and apoptosis, whereas
normal HBE cells were not affected. The phospho-Akt (pAkt) level in premalignant and
malignant HBE cells was higher than in normal HBE cells. Over-expression of constitutively
active Akt after adenoviral vector infection rescued premalignant and malignant HBE cells
from deguelin-mediated apoptosis. The MAPK activity was not affected by deguelin.
Conclusions: Deguelin inhibits premalignant and malignant HBE cell proliferation without a
detectable cytotoxicity on NHBE cells. The ability of deguelin in diminishing the signal
transduction pathway involving PI3K/Akt may contribute to the potency and specificity of
this novel drug. The specific sensitivity of premalignant and malignant HBE cells to deguelin
raises the possibility of its potential to be used in the clinic as a chemopreventive agent for the
early stages of lung carcinogenesis as well as a therapeutic agent against lung cancer.

                                               - 54 -
Session : Chemopreventive agents

TetramethoxyFlavone p7F Inhibits Collagen-Induced Arthritis by Suppressing
Inflammatory Mediators*

Hee-Gu Lee, Jong-Wan Kim, Jin-Tae Hong§ and Do-Young Yoon‡

Laboratory of Cell Biology, Korea Research Institute of Bioscience and Biotechnology,
Yuseong P. O. Box 115, Daejeon 305-600, Korea, §College of Pharmacy, Chung-buk National
University, Cheongju 361-763, Korea.

Artemisia has been traditionally used in Korean herbal medicine for clearing damp heat and
for treating uteritis and jaundice. Flavonoids isolated from the Artemisia are also known to
possess anti-inflammatory activities. In this study, 5, 6, 3’, 5’-tetramethoxy 7, 4’-hydroxy
flavone, called p7F, was isolated from Artemisia absinthium. We examined in vitro and in
vivo regulatory function of p7F on production of NO, PGE2 and TNF-a as well as expressions
of inducible NO synthase (iNOS), COX-2 and on collagen-induced arthritis. p7F inhibited the
expression or production of proinflammatory mediators such as COX-2/PGE2 and iNOS/NO
in LPS–stimulated RAW654.7 cells. p7F also suppressed the serum level of TNF-a in mice
treated with collagen, and inhibited NF-kB activation as well as NF-kB promoter activity in
RAW264.7 cells stimulated with LPS. This compound directly inhibited the intracellular
accumulation of reactive oxygen species in hydrogen peroxide-stimulated RAW654.7 cells.
Collagen-induced arthritis (CIA) was induced in DBA/1J mice by the injection of type II
collagen. The mice were then orally injected with p7F (10 mg/kg). The CIA scores in the
control group were 3.1 + 0.65 at the fifth week, and maximum levels of 9.1+1.35 at the ninth
week. However, mice treated with p7F developed arthritis at much lower CIA scores
(1.6+0.25 at the fifth week, and maximum values of 3.0+1.0 at the ninth week). These effects
appear to be due to both its antioxidant activity and the inhibition of NF-kB activation. Taken
together, these results suggest that p7F can be clinically applied to the treatment of
inflammatory diseases such as rheumatoid arthritis (This research was supported by a grant
(M103KB010022-03K0201-02210, M103KD010034-03K0401-03410) from 21st Century
Frontier Research Program and from Biodiscovery Program (M1-0106-00-0078-01-A20-00-
021-0-0) by Ministry of Science and Technology of Korean government.

                                             - 55 -
Session V: Chemopreventive agents

Some sweet and bitter tastants inhibit GRK2, GRK5 and PKA in vitro: possible effects
on signal termination

Michael Naim, Meirav Zubare-Samuelov, Irena Peri, Merav Shaul and Alexander Aliluiko

Institute of Biochemistry, Food Science and Nutrition, The Hebrew University of Jerusalem,
POB 12, Rehovot 76100, Israel. E-mail:

Some non-sugar sweeteners and bitter tastants produce a delay in taste termination, termed
"lingering aftertaste", the molecular basis of which is unknown. Although sugars and non-
sugar sweeteners appear to stimulate the same taste G-protein-coupled receptors (GPCRs), the
delay in taste termination is uniquely related to the latter and to some bitter tastants.
Furthermore, some of these tastants are agonists of melatonin or alpha2-adrenergic receptors
(Zubare-Samuelov et al., Am. J. Physiol. Cell Physiol., 285: C1255-C1262, 2003) and due to
their amphipathic properties some can also permeate taste and other cells under physiological
conditions. We now report that some non-sugar sweeteners (e.g., cyclamate, saccharin,
neohesperidin dihydrochalcone) and bitter tastants (e.g., caffeine, quinine, limonin, naringin),
inhibit GPCR-kinases, e.g., GRK2- and GRK5-phosphorylated rhodopsin (a GPCR) and
protein kinase A-phosphorylated casein in vitro. Concentration-dependence for certain
tastants was evident, revealing 80 to 100% kinase inhibition. Because these tastants inhibit
GRKs (and perhaps other kinases) directly, they may inhibit taste GPCRs desensitization and
taste termination. Furthermore, since they are components of our daily diets, these tastants
have access to other tissues along the gastrointestinal tract with implications to cellular
signaling in other tissues. The studied tastants, although not as potent as some other kinase
inhibitors, are membrane permeant, and would be desirable for research purposes.
This research was supported by grant no. IS-3366 from BARD, The US-Israel Binational
Agricultural Research and Development Fund. We thank Dr. Robert J. Lefkowitz of Duke
University, NC, USA, for kindly providing us with purified GRKs and rhodopsin.

                                             - 56 -
Session V: Chemopreventive agents

Anticancer effect of asiatic acid on skin cancer: Induction of apoptosis in SK-Mel-2
human melanoma cells and antitumorigenesis in DMBA/TPA-induced mouse skin

Byung Chul Park1, Kefa O. Bosire2, Eung-Seok Lee3, Jung-Ae Kim4

College of Pharmacy, Yeungnam University, Kyongsan 712-749, South Korea. E-mail :

Asiatic acid is a pentacyclic triterpene compound found in Centella asiatica which has been
traditionally used for skin diseases and leprosy. Previously, we reported asiatic acid induced
apoptosis through increased intracellular Ca2+, which, in turn, enhanced p53 expression in
HepG2 human hepatoma cells. In the present study, we investigated whether asiatic acid
induces apoptosis in SK-Mel-2 human melanoma cells and what signaling pathway is
involved in the cells. Asiatic acid induced a significant cytotoxicity starting at the
concentration of 30 mM and apoptosis of SK-Mel-2 cells, which was confirmed by FACS
analysis on stained cells with annexin-V FITC and propium iodide. Asiatic acid also enhanced
the expression of Bax and p53 but not Bcl-2 protein in the cells. In addition, asiatic acid
induced activation of caspase-3 activity in a dose-dependent manner. Pretreatment with Ac-
DEVD-CHO, a specific caspase-3 inhibitor, and Z-VAD-fmk, non-selective caspase inhibitor,
prevented asiatic acid-induced cell death. Furthermore, local application of asiatic acid
suppressed the development of DMBA/TPA-induced mouse skin tumor, in a concentration-
dependent manner. These results suggest that proapoptotic Bax protein, p53 and caspase-3
may play important roles in asiatic acid-induced apoptosis of SK-Mel-2 cells. These results
further suggest that asiatic acid may be a valuable agent for the therapeutic intervention of
human skin cancer.

                                            - 57 -
Session V: Chemopreventive agents


Sivaprakasam Balasubramanian1, James F. Crish1, and Richard L. Eckert1,
     Departments of Physiology and Biophysics1, Dermatology2, Biochemistry3,
Reproductive Biology4, and Oncology5, Case Western Reserve University School of
Medicine, Cleveland, Ohio 44106-4970.

Antioxidants are important candidate agents for the prevention of disease. However, the
possibility that different antioxidants may produce opposing effects in tissues has not been
adequately explored. We have previously reported that (-)-epigallocatechin-3-gallate
(EGCG), a green tea polyphenol antioxidant, stimulates expression of the keratinocyte
differentiation marker, involucrin (hINV), via a RAS, MEKK1, MEK3, p38d signaling
cascade (Balasubramanian et al., J Biol Chem. 277, 828-36, 2002). We now demonstrate a
role for CCAAT/enhancer-binding protein (C/EBP) transcription factors in the EGCG-
dependent regulation of hINV gene expression. EGCG treatment increases C/EBPa and
C/EBPb level, and C/EBP factor binding to hINV promoter C/EBP DNA binding site. This
binding is associated with increased promoter activity. Mutation of the hINV promoter
C/EBP binding site eliminates the regulation as does expression of GADD153, a dominant-
negative C/EBP factor. These results suggest that the EGCG-activated p38d signaling
cascade targets C/EBP transcription factors to enhance hINV gene expression. In contrast,
the antioxidant, curcumin, inhibits the EGCG-dependent promoter activation. This is
associated with inhibition of the EGCG-dependent increase in C/EBP factor level and C/EBP
factor binding to the hINV promoter. Curcumin treatment also inhibits the EGCG-dependent
increase in endogenous hINV levels. We conclude that curcumin and EGCG produce
opposing effects on involucrin gene expression, indicating an antagonistic action of EGCG
and curcumin in normal human keratinocytes. These results indicate that possible opposing
actions of antioxidants must be considered when designing therapy.

                                           - 58 -
Session V: Chemopreventive agents

4-(methylthio)butylisothiocyanate as a new chemopreventive agent: molecular pathway
for cell-cycle inhibition and apoptosis of human T lymphoblastoid cell line.

Carmela Fimognari1, Michael Nüsse2, Fausto Berti1, Renato Iori3, Giorgio Cantelli-Forti1,
Patrizia Hrelia1
 Department of Pharmacology, University of Bologna, via Irnerio 48, 40126 Bologna, Italy.
patrizia.hrelia@; 2GSF-Flow Cytometry Group, 85758 Neuherberg, Germany. E-
mail:; 3Research Institute for Industrial Crops, MiPAF, Via di Corticella 133,
40129 Bologna, Italy. E-mail:

Chemoprevention can be defined as the use of non-cytotoxic drugs and natural agents to block
the progression to invasive cancer. Recently, isothiocyanates, natural products found in the
diet of humans, has been shown to function as cancer chemopreventive agents. They are
strong inhibitors of phase I enzymes and inducers of phase II enzymes. They can also induce
apoptosis and modulate cell-cycle progression of highly proliferating cancer cells. We
investigated the response to the new isothiocyanate 4-(methylthio)butylisothiocyanate
(MTBITC) of a human T leukemia cell line and detected some of the molecular pathways that
are triggered by it. We demonstrated that the effects of MTBITC consist in cell-cycle
derangements in which G2/M phase inhibition is a key event, associated with a decrease in
cyclin B1, but not cyclin-dependent kinase 1, protein expression. Moreover, MTBITC induces
apoptosis, associated with an increase in p53 and bax, but not bcl-2, protein expression. To
help elucidate whether the effects of MTBITC are specific for cancer cells, we tested it on
freshly isolated, non-transformed human peripheral T lymphocytes. No effects of MTBITC
were demonstrated on non-transformed T lymphocytes. Taking into account its in vitro
antineoplastic activity and selectivity toward leukemia cells, MTBITC can be viewed as a
conceptually promising agent in cancer therapy.

                                           - 59 -
Session V: Chemopreventive agents

Growth Inhibition and Cell Cycle Dysregulation by Eupatilin, a Pharmacologically
Active Flavone Derived From Artemisia Plants, in H-Ras-Transformed Human
Mammary Epithelial Cells
Do-Hee Kim1, Hye-Kyung Na1, Tae-Young Oh2 and Young-Joon Surh1
 College of Pharmacy, Seoul National University, Seoul 151-742 and 2Dong-A
Pharmaceutical Co., Kyunggi-do, South Korea

Previous studies from this laboratory have shown that extracts of Artemisia asiatica Nakai
(Asteraceae) possess anti-inflammatory and anti-tumor promoting activities (Seo et al., Int. J.
Cancer, 100, 456-462, 2002). Eupatilin (5,7-dihydroxy-3,4,6-tri-methoxy-flavone), one of the
pharmacologically active ingredients derived from Artemisia asiatica, was shown to induce
apoptosis in human promyelocytic leukemia (HL-60) cells (H.-J. Seo and Y.-J. Surh, Mutat.
Res., 496, 191-198, 2001). In the present work, we found that eupatilin inhibited the growth
of H-ras-transformed human breast epithelial (MCF10A-ras) cells in a concentration and
time-related manner. Eupatilin inhibited the activation of ERK1/2 as well as expression of
Raf-1 and Ras in MCF10A-ras cells. To determine whether the antiproliferative effects of
eupatilin are mediated through disruption of the cell cycle progression in MCF 10A-ras, DNA
contents were analyzed by the flow cytometry. Eupatilin (100 mM) didn’t change the area of
the peak corresponding to hypodiploid or apoptotic DNA, but blocked the cell cycle
progression in both G1/S and G2/M phases. Moreover, eupatilin inhibited the expression of
Cdk2, Cdc2, cyclin B1 and cyclin D1, while it increased the expression of cyclin-dependent
kinase inhibitors, such as p53 and p27kip1. While up-regulation of p21waf/Cip1 generally
accompanies reduced cyclin D1 expression during cell cycle arrest, we note that the
expression of p21waf/Cip1 is decreased at both protein and mRNA levels in eupatilin-treated
MCF10A-ras cells. In addition, treatment with the ultrapotent MEK-specific inhibitor U0126
suppressed the growth of MCF10A-ras cells. Levels of cyclin D1 expression decreased in
U1026-treated MCF10A-ras cells. Eupatilin failed to inhibit activation of Akt, an important
component of survival signaling pathways, that is frequently activated in transformed cells
harboring mutated ras. In conclusion, the anti-proliferative effect of eupatilin is associated
with its inhibition of ERK1/2 activation and subsequent blocking of both G1/S and G2/M
phases of cell cycle progression in MCF10A-ras cells.

                                             - 60 -
Session VI : Chair : Christiane Branlant

            - 61 -
Session VI : siRNA and gene regulation

Function of Dicer in RNAi and microRNA processing in mammalian cells

Witold Filipowicz, Haidi Zhang, Fabrice Kolb, and Kaifu Tang.

Friedrich Miescher Institute for Biomedical Research, 4002 Basel, Switzerland

In eukaryotes, dsRNA induces sequence-specific inhibition of gene expression at the level of
mRNA degradation known as RNAi. During RNAi, dsRNA is first processed into 21-nt
dsRNA fragments, referred to as siRNAs. This reaction is catalyzed by the RNase III-like
nuclease called Dicer. siRNAs are then incorporated into RNA Induced Silencing Complex
(RISC) to guide sequence specific cleavage of mRNA. Some proteins associated with RISC
have been identified, among them Ago proteins, members of the PPD (PAZ-Piwi Domain)
family. MiRNAs are ~21-nt RNA regulators interacting with 3’-UTRs of mRNAs and
arresting translation by unknown mechanism. ~200 miRNAs are predicted to function in
mammals. miRNAs are processed from precursor hairpins in the reaction catalysed by Dicer.
Our research is focused on the biochemistry of Dicer and identification of miRNA targets in
mammals. Human Dicer is a ~220 kD protein containing an RNA helicase, DUF and PAZ
signatures, two RNase III domains, and a dsRNA-binding domain. We have overexpressed
and purified differently tagged human proteins. Recombinant Dicer cleaves dsRNAs into ~22-
nt siRNAs, processing dsRNAs preferentially at their termini. In contrast, Dicer excises let-7
miRNA from the inner region of its hairpin precursor. In contrast to the findings obtained
with Drosophila and C. elegans extracts, cleavage of dsRNA by the human Dicer is ATP
independent. Recombinant Dicer turns over very slowly but its activity is strongly stimulated
by proteolysis. Low catalytic efficiency of Dicer is at least partially due to siRNA products
remaining associated with the enzyme. In the course of the RNAi reaction, siRNAs have to be
handed over to the downstream RNAi component, the nuclease RISC. We have attempted to
reconstitute this putative “hand-over” step. We purified human Ago2 (hAgo2), a component
of the RISC complex. Rather unexpectedly, addition of hAgo2 to the Dicer reaction inhibits
the enzyme. In collaboration with Tom Hobman’s group (University of Alberta), we have
mapped domains responsible for the interactions between Dicer and hAgo2 and Hiwi
proteins, members of two distinct subfamilies of the PPD (PAZ and Piwi Domain) family of
proteins involved in RNAi and miRNA function.
mRNA targets of mammalian miRNAs remain largly unknown. Limited complementarity of
miRNAs to mRNAs makes target prediction by bioinformatics difficult. We are developing
affinity-based strategies, involving the use of biotinylated miRNAs, to identify miRNA
targets. We are also looking into consequences if inhibiting Dicer activity in mammalian

                                            - 62 -
Session VI : siRNA and gene regulation

MMP-1 expression in human skin fibroblasts is negatively regulated by Cdc42.

Christophe F. Deroanne, Delphine Hamelrijckx, Charles A. Lambert, Charles M. Lapière and
Betty V. Nusgens

Laboratory of Connective Tissues Biology, University of Liège, Belgium. E-mail :

In fibroblasts, matrix metalloproteinase (MMP)-1 expression is modulated by integrin-
mediated changes in cell shape and by disruption of the cytoskeleton. The small GTPases of
the Rho family are key intermediates in cellular signaling triggered by clustered cell adhesion
receptors. The implication of these GTPases in the control of MMP-1 expression has been
tested by transfection of constructs coding for constitutively active or dominant inactive forms
of these molecular switches. However, this procedure is precluded with hard-to-transfect
primary cells. The siRNA technology is an alternative for investigating the involvement of
RhoA, Rac1 and Cdc42 in the control of MMP-1 expression since efficient transfection of
siRNA can be obtained in primary cells like human skin fibroblasts (HSF). Transient
transfection of HSF with the respective specific siRNA induced a 90% reduction in the RhoA
and Cdc42 protein level and an 80% reduction in the Rac1 protein. A similar repression of
their GTP-bound forms was measured with a pull-down assay. This silencing was maintained
for up to 7 days post-transfection. Ablation of RhoA did not induce morphological alteration
while ablation of Rac1 decreased lamellipodia formation and ablation of Cdc42 induced a
“dendritic” morphology. None of these siRNA was able to repress MMP-1 overexpression
induced by actin cytoskeleton disruption. At the opposite, the reduction of Cdc42 in HSF
cultured on plastic induced a 5 fold overexpression of MMP-1. This up-regulation was
observed at both mRNA and protein level, and was detected for up to 7 days post-transfection
with 2 different siRNA targeting Cdc42. Ablation of Cdc42, but not RhoA nor Rac1, also
induced a constitutive activation of p42/44MAPK. The MEK inhibitor U0126 completely
suppressed the MMP-1 expression strongly supporting the involvement of the p42/44MAPK
pathway in this phenomenon. On the other hand, cytokine proteoarray, ELISA and RT-PCR
measurements revealed an overexpression of IL-8 and MCP-1 induced by specific ablation of
Cdc42. Investigations are in progress to evaluate their involvement in MMP-1

                                             - 63 -
Sesion VI: si RNA and gene regulation

The role of the leukemic fusion protein AML1/MTG8 in the proliferation of leukemic

Natalia Martinez1, Jürgen Krauter2, and Olaf Heidenreich1.
  Department of Molecular Biology, Institute for Cell Biology, University of Tübingen, Auf
der Morgenstelle 15, 72076 Tübingen, Germany.
  Department of Hematology, Hemostaseology and Oncology, Hannover Medical School,
30625 Hannover, Germany.

The fusion protein AML1/MTG8, which is the result of the t(8;21) translocation, is associated
with the M2 subtype of acute myeloid leukemia (AML), and accounts for 10-15% of all AML
cases. The leukemic fusion gene was targeted with siRNAs to analyse its function in cell
cycle progression and apoptosis. Electroporated t(8;21) positive Kasumi-1 cells with
AML1/MTG8 siRNA have a lower tendency towards apoptosis than control cells, indicating a
proapoptotic role of AML1/MTG8. Continuing AML1/MTG8 depletion results in an
inhibition of Kasumi-1 cell proliferation, and an increase of the doubling time by 90%, and
the inhibition of G1-S cell cycle transition. These data strongly suggest that AML1/MTG8
support proliferation of Kasumi-1 cells. The fact that the growth factors G-CSF and GM-CSF
restore proliferation in AML1/MTG8 depleted cells indicates that AML1/MTG8 decrease the
dependency on growth factors.

                                            - 64 -
Abstract Listing : Poster presentations

          - 65 -
Session I: Protein kinases as targets for
novel treatments

                   - 66 -
Session I: Protein kinases as targets for novel treatments       Poster I, 1

Inhibition of PKA and MEK enhances genomic and non-genomic up-regulation of Bcl-2
by estradiol

Marisa Cañadas, Christian Lachaud, Francisco Martinez, Jan Tesarik and Carmen Mendoza.

Departamento de Bioquimica y Biologia Molecular, Facultad de Ciencias, Universidad de
Granada, 18071 Granada, Spain. E-mail:

Estrogen signalling is controlled by the nuclear alpha and beta estrogen receptors (ER) and by
membrane receptors (non-genomic signalling). One mechanism by which estrogens may
affect apoptosis is increasing expression of Bcl-2, which has been shown to suppress
apoptosis of MCF-7 cells. Recently, estradiol anti-apoptotic effect has been shown to involve
rapid stimulation of cytoplasmic signalling cascades such as Erk, a member of the MAPK
family, the anti-apoptotic AKT protein, regulation of the c-Jun N-terminal kinase (JNK) and
p38 components of the MAPK pathway. In this study, free estradiol (E2) or estradiol-BSA
(E2-BSA) conjugate, a membrane-impermeant conjugate that does not activate nuclear ER,
were added to MCF-7 cells pre-treated with genistein, a protein tyrosin kinase (PTK)
inhibitor, H-89, a PKA inhibitor, PD98059, a MEK inhibitor, chelerythrine chloride, a PKC
inhibitor or wortmannin, a PI3-K inhibitor. Exposure of MCF-7 cells to estradiol for 24 hours
increased levels of Bcl-2 protein. Genistein inhibited Bcl-2 expression but this effect was
counteracted by addition of free E2 or E2-BSA conjugate. PD98059 and H-89 did not cause
significant changes in Bcl-2 expression when administered alone, but enhanced Bcl-2
stimulation by E2 when administered with the hormone. These results suggest the inhibition
of the PKA and MEK pathways enhance Bcl-2 up-regulation by E2; inhibition of PI3-K does
not cause any significant changes in Bcl-2 expression when compared with the effect
produced by the hormones, and PKC pathway is not involved in genomic up-regulation of
Bcl-2 by E2.

This work has been partly supported by Organon Española, S.A.

                                              - 67 -
Session I: Protein kinases as targets for novel treatments         Poster I, 2

Cooperative effects between PKA and p44/p42 Mitogen-Activated Protein Kinase
(ERK1/2) to promote CREB activation following beta cell stimulation by glucose and its
alteration due to glucotoxicity

Safia Costes1, Christine Longuet1, Christophe Broca2, Omar Faruque1, El Habib Hani1,
Dominique Bataille1 and Stephane Dalle1.
INSERM U376, 34295 cedex 5 Montpellier, France ; 2UMR 5160 CNRS, 34060 cedex 1
Montpellier, France. E-mail :,

It is well established from animal studies and clinical observations that the long term
hyperglycemia characteristic of diabetic state contributes to the deterioration of beta cell
function, a concept known as beta cell glucotoxicity. Therefore, it is essential to identify the
mechanism by which the detrimental effect of chronic excessive circulating glucose leads to
failure of beta cell function. In this study, we used the MIN6 beta cell line and isolated rat
islets 1) to clarify the signaling mechanism used by glucose to activate CREB, a transcription
factor crucial for beta cell survival and glucose-competent phenotype; 2) to evaluate the
possible down-regulation of this mechanism mediated by long term hyperglycemia. We report
that glucose (10 mM, 5 min) induces a rise in [Ca2+]i which leads to cAMP-induced Protein
Kinase A (PKA) activation that act upstream of MEK, the ERK1/2-Kinase, to mediate
ERK1/2 phosphorylation. By immunofluorescence staining, we observed a nuclear
translocation of ERK1/2 induced by glucose, strongly suggesting an access to the beta cell
transcriptional machinery. Glucose (10 mM, 5 min) induced CREB phosphorylation which
was totally inhibited by the PKA inhibitor H89 (2 mM) and reduced by 50% with the ERK1/2
inhibitor PD98059 (20 mM), indicating that ERK1/2, located downstream of PKA, cooperates
with PKA and is responsible for half of the PKA-mediated CREB phosphorylation elicited by
glucose in beta cells. We also found that exposure of beta cells for 24 h to high glucose (25
mM) induces a 70-80% decrease in cellular ERK1, ERK2 and CREB content. In high
glucose-treated, ERK1/2 and CREB-down regulated beta cells, there was a marked inhibition
of glucose (10 mM, 5 min)-stimulated ERK1/2 and CREB phosphorylation. By using nuclear
staining, this was associated with apoptotic characteristics. As we have shown that ERK1/2 is
crucial for CREB phosphorylation, a loss of ERK1/2-CREB signaling pathway in the beta cell
due to long term hyperglycemia is likely to exacerbate beta cell failure in diabetes by
affecting normal gene expression and by contributing to pro-apoptotic profiles.

                                              - 68 -
Session I: Protein kinases as targets for novel treatments         Poster I, 3

Ligand Quality And Quantity Dictate The Requirement For Lck In Signal Transduction
Through The Antigen Receptor On T Lymphocytes

Gabriel Criado, Miren L. Baroja and Joaquín Madrenas

FOCIS Centre for Clinical Immunology and Immunotherapeutics, Robarts Research Institute,
and Departments of Microbiology and Immunology and Medicine, The University of Western
Ontario, London, Ontario, Canada, N6A 5K8. E-mail:

Activation of the Src-related tyrosine kinases Lck and Fyn is one of the earliest events
following engagement of the T cell antigen receptor complex (TCR). These kinases promote
phosphorylation of the immune receptor tyrosine-based activation motifs in the CD3 chains of
the TCR complex, triggering a signalling network that culminates in T cell cycling and
differentiation. Although mature T cells express both Lck and Fyn, the former seems to play a
predominant role under most physiological circumstances as illustrated by the phenotype of T
cell lines, mice and humans deficient for Lck expression. The factors that determine the
involvement of either Lck or Fyn remain unknown. We hypothesized that intrinsic features of
the TCR ligand dictate the requirement for Lck in TCR-mediated signaling.
Under conditions of different amounts of TCR engagement (‘quantity’) and oligomerization
(‘quality’) on primary human T cells and on Jurkat T cell lines, we observed that ‘low quality
ligands’ (e.g., soluble or plate-bound anti-CD3 antibodies) required the activity of Lck to
induce IL-2 production, despite Fyn expression. In contrast, ‘high quality’ ligands such as
anti-CD3-coated beads or bacterial superantigens were able to induce IL-2 production in the
absence of Lck, and this correlated with sustained ERK activation and little or no
phosphorylation of the transmembrane adapter LAT. More importantly, genetic and
pharmacological experiments showed that not only was Lck dispensable for T cell activation
by high quality ligands but this kinase had a negative regulatory role on T cell activation by
these ligands.
Therefore, the intrinsic features of the TCR ligand dictate the differential requirement for Lck
or Fyn in early TCR signaling and T cell activation. These features should be considered
when addressing the manipulation of signal transduction pathways on T cells for therapeutic

                                              - 69 -
Session I: Protein kinases as targets for novel treatments          Poster I, 4

PP1 Binding Proteins as therapeutic targets for male contraception and infertility

Margarida Fardilha, Wu Wenjuan, Odete A. B. da Cruz e Silva and Edgar F. da Cruz e Silva

Centre for Cell Biology, Department of Biology, University of Aveiro, 3810-193 Aveiro,
Portugal. E-mail:

Many diseases and dysfunctional states are associated with abnormal phosphorylation of key
proteins (e.g. cancer, diabetes, Alzheimer’s disease, etc.). Thus, protein phosphorylation
systems represent attractive targets for diagnostics and therapeutics. Protein phosphatase 1
(PP1) is an ubiquitously expressed serine/threonine-specific protein phosphatase whose
activity towards different substrates appears to be mediated via binding to specific regulatory
proteins. The increasing diversity of such PP1 regulatory subunits and their tissue specificity
make them attractive pharmacological targets. Of the three mammalian PP1 genes, only
PP1gamma is known to undergo alternative splicing to yield the ubiquitous PP1gamma1 and
the testis-specific PP1gamma2 isoforms. Previous work has also shown that PP1gamma2 is
specifically expressed in sperm and likely to play a key role in the control of sperm motility.
Particular attention is being devoted in our laboratory to the identification and
characterization of testis-specific PP1 interacting proteins that may be involved in the control
of sperm motility and thus would represent attractive therapeutic targets. Recently, we have
undertaken an in-depth survey using the yeast two-hybrid approach to identify proteins
expressed in human testis capable of interacting specifically with the alternatively spliced
isoforms of PP1gamma. The results obtained were validated by a variety of criteria, including
the identification of “bona fide” PP1 binding proteins and the presence of a consensus PP1
binding motif in the identified proteins. Thus, PP1 regulatory proteins that interact
specifically with either or with both splice variants were identified. We will present the
results obtained and discuss their implication for signal transduction therapeutics, in particular
for the treatment of male infertility and male contraception.

Supported by the EU V Framework Program, FCT and CBC

                                              - 70 -
Session 1: Protein Kinases as targets for novel treatments         Poster I, 5

PET Imaging of [18F]Iressa in Mouse Xenograft Tumors

Onofre T. DeJesus1,4, D. Murali1, L.G. Flores1, A.K. Converse2, Rachel M. Bartlett1, D.W.
Dick1, T.R. Oakes2, Eric A. Armstrong3, R.J. Nickles1,4 and Paul M. Harari3,4.
  Department of Medical Physics, University of Wisconsin Medical School, Madison, WI
53706, U.S.A., 2W.M. Keck Laboratory for Functional Brain Imaging, Waisman Center,
University of Wisconsin, Madison, WI 53705, US.A., 3Department of Human Oncology and
  Comprehensive Cancer Center, University of Wisconsin Medical School, Madison, WI
53792, U.S.A. E mail:

Epidermal growth factor receptor (EGFR) is an epithelial cell membrane receptor with an
intracellular tyrosine kinase (TK) involved in signal transduction crucial to proliferation,
apoptosis, repair and angiogenesis. More than two thirds of human cancers derive from
epithelial tissues and EGFR-TK is overexpressed in the majority of these. In recent years
several selective EGFR-TK inhibitors with nanomolar affinities have been developed as
potential anti-cancer agents. One potent inhibitor approved for clinical use is Iressa (ZD1839,
gefitinib). Iressa is a fluorine-containing anilinoquinazoline which we have labelled
isotopically with [18F] to test its use as a PET imaging agent. PET imaging of 18F-Iressa may
be useful in identifying patients who may benefit from Iressa treatment and in monitoring
their therapy.

[18F]-Iressa was prepared via the standard Kryptofix-K2CO3-mediated nucleophilic [18F]
exchange reaction with a trimethylammonium triflate precursor to give 4-[18F]fluoro-3-chloro-
nitrobenzene. Reduction to aniline followed by condensation with 7-methoxy-6-(3-
morpholinopropoxy)-4-chloro-quinazoline in DMF at 145 oC gave [18F]-Iressa.

Biodistribution studies in normal HSD-ICR mice showed high uptake of [18F]-Iressa in the GI
tract, liver, kidneys and bladder. Plasma metabolite analysis in several mice up to 3 hr showed
temporally decreasing percentage of unchanged [18F]-Iressa declining to 52% at 3 hr
postinjection. Dynamic PET scans of these mice using a Concorde microPET P4 scanner gave
the time course of distribution and confirmed the ex vivo results.

Athymic mice with human squamous cell carcinoma tumor xenografts, previously shown to
overexpress EGFR, were injected i.v. with [18F]-Iressa and immediately imaged with the
microPET scanner. Regions of interests (ROI) delineating the liver, small intestines, kidney
and tumor were drawn in the PET images and time-activity curves showed continuous
increase of [18F] in these areas. At the end of the 3 hr PET scan, one animal was sacrificed,
dissected and ex vivo counts in several organs were assayed using an automatic gamma
counter. Tissue to blood [18F] ratios were found to be spleen > kidney > liver ~ small
intestines > lungs ~ stomach > skin ~ tumor > muscle. Tumor to blood ratio was 4.5. The
spleen to blood ratio was about 10-fold this value while brain to blood ratio was less than one.
These results suggest that persistence of [18F]-Iressa in non-tumor regions may limit its use as
a PET imaging agent for EGFR-rich tumors.

                                             - 71 -
Session I: Protein kinases as targets for novel treatments         Poster I, 6

Glycogen synthase kinase-3 inhibitors- novel therapeutic compounds

Hagit Eldar-Finkelman and Oksana Kaidanovich-Beilin
Department of Human Genetics and Molecular Medicine, Sackler Faculty of Medicine, Tel
Aviv University, Israel
Glycogen synthase kinase-3 (GSK-3) is a ubiquitous cytosolic serine/threonine protein kinase
that has recently emerged as a novel and promising target for drug discovery. Its elevated
activity has been linked with the pathogenesis of several chronic diseases, in particular type 2
diabetes. We recently generated a novel class of substrate competitive peptide inhibitors of
GSK-3, and examined their therapeutic implications in insulin resistance and type 2 diabetes.
The GSK-3 inhibitors L803-mts provoked glucose uptake in primary adipocytes and
increased glycogen synthase activity in HEK 293 cells. Subsequently, administration of the
GSK-3 inhibitor to normal or diabetic obese mice such as high fat diet-fed mice or ob/ob mice
improved their performance on glucose tolerance tests (GTT) and decreased blood glucose
peak by 20 -30%. Long-term treatment of ob/ob mice with L803-mts (4 weeks) reduced blood
glucose levels (by 20%), improved their performance on GTT (30% reduction in blood
glucose peak) and reduced their food consumption (by 15%). Tissue distribution analysis
confirmed the presence of L803-mts in muscle liver and fat tissues.
We present here a novel rational strategy for developing specific GSK-3 inhibitors, and point
toward GSK-3 as a promising therapeutic target in insulin resistance and type 2 diabetes.

                                              - 72 -
Session: I Protein kinases as targets for novel treatments         Poster I, 7

p14ARF activates p44/42 MAPK and Chk1/2 kinases pathways to induce G2 arrest
independently of p53

Beatrice Eymin, Caroline Salon, Camille Leduc, Paule Claverie, Elisabeth Brambilla and
Sylvie Gazzeri.
Lung Research Group, INSERM U578, Institut Albert Bonniot, Domaine de la Merci, 38000

We had previously demonstrated the ability of the p14ARF tumor suppressor protein to induce
G2 arrest in cell lines deprived of functional p53 (Eymin et al., Oncogene, 2003). The aim of
this study was to investigate the cellular signaling pathway(s) involved in this effect. We
demonstrate that pharmacological inhibition of MEK1/2 MAPK using U0126 and PD98059
and Chk1/2 kinases using caffeine can prevent p14ARF-induced G2 arrest and reverse the
ability of p14ARF to induce p21WAF1 accumulation and inactive phosphorylation of cdc25c and
p34cdc2 proteins. Consistent with an implication of Chk1/2 and MEK1/2 signaling pathways in
the antiproliferative effect of p14ARF, we show that p14ARF activates the MAP kinases p44/42
and Chk1 and –2 kinases by inducing their phosphorylation on specific residues. Furthermore,
we also show using RNA interference that activation of both Chk1 and Chk2 is required for
p14ARF-induced G2 arrest and that p44/42 MAPK and Chk1/2 pathways are closely connected.
Finally, we show that p14ARF induces the accumulation of phosphorylated ATM/ATR
substrates as well as the phosphorylation and the accumulation into nuclear foci of histone
H2A.X and Rad17 proteins, both involved in the response to DNA damage. Moreover,
inhibition of endogenous ATM or ATR proteins by siRNA prevents the activation of Chk1/2
kinases by p14ARF. Altogether, these results indicate that p14ARF is able to activate pathway(s)
involved in the DNA damage response independently of p53 by a mechanism(s) that remains
to be clarified.

                                              - 73 -
Session: I Protein kinases as targets for novel treatments       Poster I, 8

Oxidative stress stimulates multiple MAPK signalling pathways and the
phosphorylation of the small HSP27 in the perfused amphibian heart

Catherine Gaitanaki, Constantina Stathopoulou, Chrysa Stavridou and Isidoros Beis
Department of Animal & Human Physiology, School of Biology, Faculty of Sciences
University of Athens, Panepistimioupolis, Athens 157 84, GREECE, e-mail:;

We investigated the activation of three subfamilies of MAPKs (ERK, JNKs and p38-MAPK),
by oxidative stress in the isolated perfused amphibian heart. Activation of p43-ERK by 100
uM H2O2 was maximally observed within 5 min, remained elevated for 30 min and was
comparable with the effect of 1 uM PMA. p43-ERK activation by H2O2 was inhibited by
PD98059, but not by SB203580. The p46 and p52 species of JNKs were maximally activated
by 2.5- and 2.1-fold, respectively by 100 uM H2O2 within 2 min. JNKs activation was still
detectable after 15 min, reaching control values at 30 min of treatment. p38-MAPK was
maximally activated by 100 uM H2O2 by 9.75-fold at 2 min and this activation progressively
declined thereafter, reaching control values within 45 min of treatment. The dose-dependent
profile of p38-MAPK activation by H2O2 observed revealed that 30 uM of H2O2 induced the
maximal phosphorylation, whereas 100-300 uM H2O2 induced a considerable activation of the
kinase. Our studies also showed that the phosphorylation of MAPKAPK2 by H2O2 followed a
parallel time-dependent pattern and that SB203580 abolished this phosphorylation.
Furthermore, our experiments clearly showed that 30 uM of H2O2 induced a strong, specific
phosphorylation of HSP27. Our immunohistochemical studies showed that immune
complexes of phosphorylated forms of both, p38-MAPK and HSP27 were strongly enhanced
by 30 uM H2O2 in the perinuclear region as well as dispersedly in the cytoplasm of ventricular
cells and that SB203580 abolished this phosphorylation. These data indicate that oxidative
stress is a powerful activator of all three MAPK subfamilies in the amphibian heart.
Stimulation of p38-MAPK and the consequent phosphorylation of HSP27 may be important
in cardioprotection under such conditions.

                                              - 74 -
Session I: Protein kinases as targets for novel treatments        Poster I, 9

Sensitivity changes of enteric cholinergic neurons after chronic sympathetic
denervation: role of PKC

Cristina Giaroni, Elena Zanetti, Luca Canciani, Daniela Giuliani, Sergio Lecchini, Gianmario

Clinical and Applied Pharmacology Centre, University of Insubria and University of Pavia,
via O Rossi 9, I-21100 Varese, Italy. E-mail:

Rearrangements of enteric neuronal circuitries after extrinsic denervation may play a role in
development of functional abnormalities/functional recovery after injuries or surgical
procedures in the gut. In the guinea-pig distal colon chronic sympathetic denervation entails
subsensitivity to inhibitory a2-adrenoceptor agonists and supersensitivity to inhibitory m- and
k-opioid receptor agonists on acetylcholine (ACh) release. These observations are highly
indicative for the occurrence of a functional interplay among different inhibitory pathways in
this model, whose mechanism/s remain largely to be unravelled. In the present study, we
evaluated the possible involvement of protein kinase C (PKC). The PKC activator, PMA,
dose-dependently enhanced spontaneous endogenous ACh overflow from the guinea-pig
colon with an EC50 of 0.12 (0.01-1.13)mM. The maximal effect, obtained at 1 mM, was lower
in sympathetically-denervated preparations than in normal preparations. In normal
preparations, the inhibitory effect of U69593 (k agonist, 100 nM) and DAMGO (m agonist,
100 nM) on ACh overflow increased in the presence of PKC inhibitors calphostin C (100 nM)
and cheleritrin (1 mM). In sympathetically-denervated preparations, the effect of U69593 and
DAMGO on ACh overflow was higher than in normal preparations and was not modified by
PKC inhibitors. Expression levels of PKC, evaluated by Western Blot, decreased after chronic
sympathetic denervation. The present data indicate that activation of PKC enhances ACh
overflow from the myenteric plexus of the guinea-pig distal colon. At this level, m- and k-
opioid receptors activate PKC, and this might represent a feed-back mechanism on inhibition
of ACh release. Chronic sympathetic denervation entails a reduced efficiency of PKC and this
change might contribute to development of supersensitivity to opioid agonists. On the whole
these data indicate that PKC might play a role in the rearrangements of intrinsic enteric
circuitries after extrinsic denervation.

                                              - 75 -
Session I : Protein kinases as targets for novel treatments      Poster I, 10

The protein kinase C signalling pathway regulates a molecular switch between
transactivation and transrepression activity of the Peroxisome Proliferator-activated
Receptor alpha (PPARalpha)

Christophe Blanquart, Roxanne Mansouri, Réjane Paumelle, Jean-Charles Fruchart, Bart
Staels and Corine Glineur.

 INSERM UR 545, Département d'Athérosclérose, Institut Pasteur de Lille, 1 rue du Pr.
Calmette 59019 Lille, and Faculté de Pharmacie, Université de Lille II, 59000 Lille, France.
E-mail :

Peroxisome proliferator-activated receptor alpha (PPARalpha) is a nuclear receptor implicated
in several physiological processes such as lipid and lipoprotein metabolism, glucose
homeostasis and the inflammatory response. PPARalpha is activated by natural fatty acids
and synthetic compounds like fibrates. PPARalpha activity has been shown to be modulated
by its phosphorylation status. PPARalpha is phosphorylated by kinases such as the mitogen-
activated protein kinases (MAPK) and cAMP-activated protein kinase (PKA). However,
whether PKC control PPARalpha activity is not known. In this report, we show that PKC
inhibition impairs ligand-activated PPARalpha transcriptional activity. Furthermore, PKC
inhibition decreases PPARalpha ligand-induction of its target genes including PPARalpha
itself and CPT-I. By contrast, PKC inhibition enhances PPARalpha transrepression properties
as demonstrated using the fibrinogen-beta gene as model system. Finally, PKC inhibition
decreases PPARalpha phosphorylation activity of hepatocyte cell extracts. In addition,
PPARalpha purified protein is phosphorylated in vitro by the classical recombinant PKC.
Taken together, these results demonstrate that PPARalpha activity is controlled by the PKC
pathway. Our data indicate that the PKC signalling pathway acts as a molecular switch
dissociating the transactivation and transrepression functions of PPARalpha.

                                              - 76 -
Session I: Protein kinases as targets for novel treatments          Poster I, 11

Efficient proteomics methods to identify the cellular targets of protein kinase inhibitors

Klaus Godl, Dirk Brehmer, Alexander Kurtenbach, Peter Habenberger, Stephanie Blencke,
Heidrun Gutbrod, Kostadinos Salassidis, Matthias Stein-Gerlach, Andrea Missio, Matt
Cotten, Josef Wissing and Henrik Daub.

Axxima Pharmaceuticals AG, Max-Lebsche-Platz 32, 81377 München, Germany
Email :

Small-molecule inhibitors of protein kinases are widely used in signal transduction research
and are emerging as a major class of drugs for therapeutic intervention in a variety of
diseases. Interpretation of biological results obtained with protein kinase inhibitors critically
depends on compound selectivity, however a proteome-wide analysis of the various enzymes
targeted by these inhibitors has not been performed yet.
To address this important issue, we have immobilised suitable analogues of various protein
kinase inhibitors for affinity purification of their cellular targets. Optimised adsorption and
elution conditions permitted the dramatic enrichment of small subsets of cellular proteins,
which were resolved by one- or two-dimensional SDS-PAGE prior to identification by mass
This proteomics approach led to the identification of several protein kinases, which had not
previously been described as targets of some well known protein kinase inhibitors. Receptor-
interacting kinase (RICK), for instance, was even more sensitive to the widely used p38
kinase inhibitor SB 203580, than p38alpha itself. Also subtle chemical modifications of the
immobilised compounds significantly altered the set of proteins isolated by affinity
chromatography. Moreover, tyrosine kinase inhibitors originally developed for cancer therapy
were shown to inhibit serine/threonine kinases implicated as targets of anti-inflammatory
Our proteomics approach serves as an efficient tool to identify cellular targets of protein
kinase inhibitors and thereby addresses critical selectivity issues, which are of high relevance
for both signal transduction research and drug development.
Reference:      Godl, K. et. al. (2003) An efficient proteomics method to identify the cellular
targets of protein kinase inhibitors. Proc. Natl. Acad. Sci. U.S.A. 100, 15434-15439.

                                              - 77 -
Session I: Protein kinases as targets for novel treatments       Poster I, 12

Contribution of MAPK and p53 tumor suppressor to apoptosis of glioma cells

Bozena Kaminska, Beata Pyrzynska, Agata Zupanska, Aneta Master, Dorota Owczarek

Laboratory of Transcription Regulation, Nencki Institute, Warsaw, Poland, E-mail:

 p53 tumor suppressor induces growth arrest and cell death via apoptosis in response to a
number of cellular stresses. MAP kinases are among the known regulators of p53 and after
activation they phosphorylate and activate p53 in response to stress signals. We have
previously shown cyclosporin A (CsA) treatment of glioma cells results in apoptosis
associated with accumulation of p53 protein, its transcriptional activation and expression of
target genes encoding p21/Waf1 and Bax proteins. Induction of cell death by CsA is
associated with a prolonged activation of Sek1/JNK and MKK3/p38 MAP kinase signaling
pathways. Accumulation of p53 is blocked by SB202190, a pharmacological inhibitor of p38
MAPK pathway and overexpression of a dominant negative form of MKK3, an upstream
activator of p38, abrogates phosphorylation of p38 MAPK and p53 accumulation induced by
CsA. These results demonstrate the contribution of p38 MAPK signaling pathway to
increasing of p53 protein levels and induction of p53-dependent apoptotic program activated
by CsA. Moreover, prolonged activation of JNK and p38MAPK leads to phosphorylation of
c-Jun and ATF2 proteins, induction of AP-1 complex and transcriptional activation of FasL
expression. Induction of JNK/cJun signaling pathway and FasL expression in CsA-induced
apoptosis are inhibited in glioma cells overexpressing p53Val135 mutant, which suggests
these are secondary events, dependent upon prior p53 activation. Similar defects in activation
of JNK/c-Jun pathway were observed in glioma cells exposed to toxic doses of UVC and
adriamycin that correlates with decreased susceptibility (UVC) or complete resistance of
(ADR) to the treatment. p53 is ofter mutated in many human brain tumors and lack of JNK/c-
Jun signaling pathway activation may contribute to tumor resistance to conventional
therapies. Supported by PBZ- MIN/001/P05/10.

                                              - 78 -
Session I: Protein kinases as targets for novel treatments        Poster I, 13

Lack of the N1L gene expression results in a significant decrease of vaccinia virus
replication in mouse brain

Barry Billings1, Scott A. Smith1, Zhouning Zhang1, Debomoy K. Lahiri2, Girish J. Kotwal1,3
  Department of Microbiology and Immunology, University of Louisville School of Medicine,
Louisville, KY, 40202, USA. 2Department of Psychiatry, Indiana University School of
Medicine, Indianapolis, IN, 46202, USA. 3Division of Medical Virology, University of Cape
Town, Medical School, Observatory, Cape Town, South Africa.

Vaccinia virus encodes secretory proteins termed virokines. One of the major virokines is the
13.8 kDa protein. A recombinant virus, termed vGK5 lacking this protein when injected
intracranially into mice, has one of the highest level of in vivo attenuation achieved by
deletion of any single open reading frame of vaccinia virus. Here we show that the 13.8 kDa
protein significantly enhances viral replication within brain tissue, however, analysis of
histology, neutrophil infiltrate, and nitric oxide synthase activity shows no significant
differences between wild type vaccinia virus and vGK5. Since there is poor growth of vGK5
virus in the brain, the possibility of post-vaccinial encephalitis is significantly diminished.
Structurally, the 13.8 kDa protein is similar to the family of aba proteins, which includes
adenylate kinase, an enzyme implicated in increasing energy generation in the brain.

                                              - 79 -
Session I: Protein kinases as targets for novel treatments         Poster I, 14

p42/p44 MAP kinase signal transduction pathway regulates IL-6 expression in PC3 cells,
a line of hormone refractory prostate cancer cells

Sweaty Koul, Mei Huang, Lakshmi Chaturvedi, Randall B Meacham, Roger Davis and Hari
K Koul*
University of Colorado School of Medicine

Introduction and Objective: Prostate cancer is the second leading cause of cancer related
deaths in males. In our recent studies we demonstrated that p42/44 Mitogen activated Protein
Kinases (p42/p44 MAP kinase) signal transduction pathways are critical for clonogenic
activity, cell migration as well as invasion of PC3 cells. In the present study we demonstrate
that p42/p44 MAP kinase pathway is required for IL-6 synthesis and secretion by hormone
refractory prostate cancer cells.
Methods: PC3 cells were grown in DMEM/F12 medium supplemented with FCS (10%) and
antibiotics in multi-well plates, in the presence or the absence of PD 098059, a specific
inhibitor of MEK (the upstream activator of p42/p44 MAPK) or cells were transfected with
p42/p44 phosphorothioate antisense nucleotides or dominant negative p42 and p44 MAP
kinase vectors. The conditioned mediums were collected and an IL-6 protein level was
measured by enzyme-linked immunosorbent assay (ELISA). The total cellular RNA was
isolated from the cells and subjected to relative quantitative RT-PCR to determine the
expression of IL-6 mRNA. The statistical analysis of the results was carried out using the
Student’s t test.
 Results: Treatment of PC3 cells with PD 098059 decreased secretion of IL-6 protein. PD
098059 treatment of PC3 cells also induced transcriptional down-regulation of IL-6 gene as
determined by decreased level IL-6 mRNA expression following PD treatment. Treatment of
PC3 cells with PD 098059 inhibited p42/p44 MAP kinase activity in a dose dependent fashion
with over 80% inhibition at 50_M PD. Transfection of the cells with p42/p44
phosphorothioate antisense nucleotides inhibited expression of p42/p44 as well kinase activity
by 60-70% and decreased IL-6 transcription by ~60%. Thus inhibition of p42/p44 MAP
kinase activity co-related with the inhibition of IL-6 transcription and secretion by PC3 cells.
This is the first report demonstrating regulation of IL-6 in PC3 cells by p42/p44 MAP kinase
signal transduction pathway. These data demonstrate that p42/p44 MAP kinase signal
transduction pathway plays a significant role in IL-6 synthesis and secretion in PC3 cells.
Conclusions: These results suggest that p42/p44 MAP kinase signal transduction pathway is
essential for IL-6 expression in hormone independent prostate cancer cells.

                                              - 80 -
Session I: Protein kinases as targets for novel treatments         Poster I, 15

p42/p44 MAP kinase Signal Transduction Pathway a Novel Target for the Treatment of
Hormone Resistant Prostate Cancer?

Sweaty Koul, Randall B Meacham, David Crawford, Roger Davis1 and Hari K Koul*
Division of Urology and Department of Surgery, University of Colorado School of Medicine,
Denver, CO 80262, USA and 1Howard Hughes Medical Institute and Program in Molecular
Medicine, UMASS Medical School, Worcester, MA 01605, USA

        Prostate cancer is the second leading cause of cancer related deaths in males. Most of
the prostate cancer deaths result from emergence of an androgen resistant phenotype of
prostate cancer. To date no satisfactory treatment options are available for these androgen
resistant prostate cancer patients. These facts underline the need to develop new therapies that
will improve outlook for hormone-independent prostate cancer.
        Several lines of evidence suggest a role of p42/44 Mitogen activated Protein Kinases
(p42/p44 MAP kinase) signal transduction pathways in hormone-independent prostate cancer.
This talk will provide an overview of the role played by p42/p44 MAP kinase signal
transduction in androgen-independent prostate cancer. We will also present recent data from
our laboratory demonstrating critical requirement of p42/p44 MAP kinase signal transduction
pathway for clonogenic activity, cell migration as well as invasion of PC3 cells. Finally we
will present arguments in favor of targeting p42/p44 MAP kinase signal transduction pathway
for treatment of prostate cancer.

* Presenting and corresponding author

                                              - 81 -
Session I : Protein kinases as targets for novel treatments      Poster I, 16

Ras-Recruitment System: high throughput system for the identification and validation
of drug targets

Claudia Kruse, Michael Udelhoven, Sergey Vasiliev and Hanjo Hennemann

Protein Interaction Analysis, center of advanced european studies and research (caesar),
Ludwig Erhard Allee 2, 53175 Bonn, Germany, E-mail:

We are investigating disease relevant kinases as potential drug targets by a novel in vivo
protein interaction screening system, the Ras-Recruitment-System (RRS). The RRS uses a
selection mechanism in the cytoplasm of yeast cells, which activates the mitogenic RAS
pathway. In comparison to the 2-hybrid systems the RRS has a very tight selection
mechanism. This results in the identification of a high percentage of significant interaction
partners and allows also the analysis of membrane proteins. Currently we are developing a
high-throughput version of the RRS (HT-RRS) which significantly increases the capabilities
to perform many RRS-screenings of expression libraries in parallel. By the integration of
pipetting- and picking robots we can increase the screening capacity to more than 200
complete library screens in 3 months. Vital part of the HT-RRS is a proprietary laboratory
information system (LIMS), which was programmed inhouse (object oriented database) and
tracks all experimental data.
 We could show that our system is suitable to analyse different classes of proteins, which
failed in conventional 2-hybrid screens, like the Pim-1 kinase or the c-Myc transactivating
region. In addition we found many novel interaction partners with proteins, which had been
intensively used with other screening methods before (e.g. JNK-1). Our aim is to elucidate
protein kinase networks by identifying specific substrates and novel regulatory proteins of
disease relevant kinases. In addition we are developing technologies which interfere with
protein kinase functions. These novel tools could be used for drug target validation and the
identification of novel modes of kinase inhibition.

                                              - 82 -
Session I: Protein kinases as targets for novel treatments       Poster I, 17

Novel Therapies Exploiting Signal Plasticity of Cell Surface Receptors: Ligand-
Dependent Conversion of CTLA-4 from Inhibitor to Activator of T Cells

Luan A. Chau, Wendy A. Teft, Beatriz M. Carreno*, Vincent Ling*, and Joaquín Madrenas

FOCIS Centre for Clinical Immunology and Immunotherapeutics, Robarts Research Institute,
and Departments of Microbiology and Immunology and Medicine, The University of Western
Ontario, London, Ontario, Canada, N6A 5K8, and *Wyeth Research, Cambridge MA, USA.

Antibodies or their recombinant fragments against surface receptors of the immunoglobulin
superfamily can induce or block the receptors’ native function depending on whether they
induce or prevent the assembly of signalosomes on their cytoplasmic tails. Here, we propose a
novel paradigm in which different ligands of a receptor induce the formation of different
signalosomes on the cytoplasmic tail of that receptor leading to opposite functions.
CTLA-4 is an activation-induced receptor that inhibits T cell activation when co-ligated with
the antigen receptor (TCR). We generated a recombinant ligand against CTLA-4 that by
itself converts this inhibitory receptor into an activating receptor for primary human T
lymphocytes. This reversal of function results from increased recruitment of PP2A
phosphatase to the cytoplasmic tail of CTLA-4, and assembly of a distinct signalosome that
activates an lck-dependent signaling cascade and induces IL-2 production. Therefore, the
cytoplasmic domains of some cell surface receptors, such as CTLA-4, may have an inherent
plasticity illustrated by the capacity to assemble different signalosomes leading to opposite
functional outcomes. Such plasticity may be exploited therapeutically with recombinant
ligands for those receptors.

                                              - 83 -
Session I: Protein kinases as targets for novel treatments           Poster I, 18


Alessandro Massa1, Federica Barbieri2, Cinzia Aiello2, Rodolfo Iuliano3, Alfredo Fusco4,
Gianluigi Zona5, Renato Spaziante5, Gennaro Schettini1,2 and Tullio Florio1,2.
  Section of Pharmacology, Dept. Oncology Biology and Genetics, University of Genova,
  Pharmacology and Neuroscience, National Institute for Cancer Research (IST) Genova,
  Dept. of Clinical and Experimental Medicine, University of Catanzaro, 4Endocrinology and
Experimental Oncology Center, National Research Council and Dept. of Molecular and
cellular pathology, University of Naples, 5Division of Neurosurgery, Dept Neuroscience,
Ophtalmology and Genetics, University of Genova.

Somatostatin (SST) is an endogenous regulator of proliferation in a variety of epithelial and
endocrine cells. The effects of SST are mediated by a family of five G protein-coupled
receptors (named SSTR1 through 5) which are variably expressed in both normal tissues and
tumors. SST inhibition of cell proliferation can be mediated by both direct and indirect
mechanisms: one of the main intracellular pathways responsible for the direct inhibition of
cell growth by SST is the modulation of phosphotyrosine phosphatase (PTP) activity. We
reported that SST-dependent cytostatic activity is mediated by the activation of a receptor-like
PTP, named PTPh (Mol Endocrinol 2001, 15: 1838) which is expressed ubiquitously,
showing high levels in the brain, liver and spleen. Several studies suggested that PTPh plays a
role in the inhibition of cell growth and in the cell differentiation process.
In this study we characterized the intracellular effectors of the antiproliferative activity of SST
in 5 glioma cell lines (C6, U87MG, U373MG, DBTRG 05MG, CAS1) and in 7 post-surgical
specimens. The treatment with SST caused an inhibition of cell growth only in C6 and U87
cells, which expressed the PTPh In C6 cells, SST effects were reverted by pretreatment with
pertussis toxin (PTX) and vanadate indicating the involvement of G proteins and PTPs in the
antiproliferative effects. To demonstrate the involvement of PTPh in the SST antiproliferative
effects we tested the activity of PTPh by time-course experiments: its activity was increased
by SST treatment reaching the maximum after 60 min and a correspondent reduction of
ERK1/2 activation was observed. Since bFGF-dependent MEK phosphorylation was not
affected by SST exposure, it can be suggested a direct effect of SST-activated PTPh on
ERK1/2 phosphorylation. In vitro experiments showed that a recombinant PTPh
dephosphorylated ERK activated by bFGF. Moreover, SST reduced cell proliferation trough
the increase of p27kip1 mediated by the ERK1/2 dephosphorylation.
Finally, we evaluated the requirement for PTPh in SST inhibition of cell proliferation in
postsurgical specimens derived from different grade human gliomas. We identified SST
receptors mRNA in all the gliomas analyzed, while the expression of PTPh was limited to one
third of the tumors. Culturing seven gliomas, the same correlation between the expression of
PTPh and the somatostatin antiproliferative effects was identified. In conclusion we propose
that, in glioma cells, SST-mediated inhibition of cell proliferation requires the expression and
activation of PTPh and that this PTP was able to directly dephosphorylate ERK1/2 .

                                               - 84 -
Session I: Protein kinases as targets for novel treatments        Poster I, 19

Molecular mechanisms involved to maintain a mitochondrial biogenesis in mtDNA
depleted cells

L.Mercy, C. Gilquin, A. Houbion, C. Demazy, P. Renard, J. Remacle, M. Raes
and T. Arnould.

Laboratoire de Biochimie et Biologie Cellulaire, University of Namur (F.U.N.D.P), 61 rue de
Bruxelles, 5000 Namur, Belgium
e-mail :
Cells depleted in mitochondrial DNA (mtDNA) or rho0 cells are commonly used for studying
the cellular responses to mitochondrial dysfunction. Despite their chronic energy depletion,
rho0 cells still survive and display interesting features. Indeed, mitochondrial structures are
still observed, and one can measure a mitochondrial membrane potential in these cells.
Therefore, an active mitochondrial biogenesis must be maintained in mtDNA-depleted cells.
Here, we tried to better understand how this complex process that involves both nuclear and
mitochondrial factors, still occur in L929 mtDNA-depleted cells and 143B rho0 cells. First,
we studied different transcription factors and co-activators that are well known or suspected
to play a crucial role in the biogenesis of mitochondria. We showed that among the different
factors studied so far (PPARg, Sp1, NRF-1, NRF-2, YY1 and CREB), only CREB (cAMP-
Responsive Element Binding protein) seems to be activated. Using CAT reporter systems
driven by the authentic promoters of different nuclear genes encoding mitochondrial markers
such as cytochrome c, ß subunit of the F1-ATPase, and mitochondrial transcription Factor A
(mtTFA), we showed that the transcription of these genes are up-regulated in 143B rho0 cells.
The over-expression of two dominant negative mutants of CREB (K-CREB and M1-CREB)
inhibits dramatically the expression of cytochrome c. Mitochondrial protein import is also
very important in mitobiogenesis to maintain a mitochondrial structure. We thus next assessed
mitochondrial protein import activity in isolated mitochondria of mtDNA-depleted cells. For
this purpose, we used radiolabeled mitochondrial matrix-targeted fusion proteins and showed
that mitochondrial protein import activity is reduced in 143B rho0 cells. ATP content and
Dym, the two main forces involved in mitochondrial matrix protein import, are also
dramatically reduced in 143B rho0 cells. However, endogenous cytochrome c is more
abundant in mitochondria of mtDNA-depleted cells meaning that other mechanisms than
classical TOM / TIM pathway are involved. These preliminary results evidenced a possible
role for CREB in preserving mitochondrial biogenesis in mtDNA-depleted cells.

                                              - 85 -
Session I: Protein kinases as targets for novel treatments       Poster I, 20

CK2 controls multiple signal-transducing protein kinases by phosphorylating a kinase-
targeting molecular chaperone Cdc37

Yoshihiko Miyata and Eisuke Nishida

Department of Cell and Developmental Biology, Graduate School of Biostudies, Kyoto
University. Kitashirakawa Oiwake-cho, Sakyo-ku, Kyoto 606-8502, Japan.

Cdc37 is a kinase-associated molecular chaperone whose function in concert with Hsp90 is
essential for many signaling protein kinases. Here we report that mammalian Cdc37 is a
pivotal substrate of CK2 (casein kinase II). Purified Cdc37 was phosphorylated in vitro on a
conserved serine residue, Ser13, by CK2. Moreover, Ser13 was the unique phosphorylation
site of Cdc37 in vivo. Crucially, the CK2 phosphorylation of Cdc37 on Ser13 was essential
for the optimal binding activity of Cdc37 toward various kinases examined, including Raf1,
Akt, AuroraB, Cdk4, Src, MOK, MAK, and MRK. In addition, non-phosphorylatable
mutants of Cdc37 significantly suppressed the association of Hsp90 with protein kinases,
while the HSP90-binding activity of the mutants was unchanged. Treatment of cells with a
specific CK2 inhibitor suppressed the phosphorylation of Cdc37 in vivo and reduced the
levels of Cdc37-target kinases. These results unveil a regulatory mechanism of Cdc37 and
identify a novel molecular link between CK2 and many crucial protein kinases via Cdc37, and
reveal the molecular basis for the ability of CK2 to regulate pleiotropic cellular functions.
Altogether, we suggest that CK2-Cdc37 system can be a novel and efficient pharmacological
target for cancer chemotherapeutic agents.

                                              - 86 -
Session I: Protein kinases as targets for novel treatments           Poster I, 21

Rottlerin dependent inhibition of protein kinase C delta activity is modulated by its
scavenger effects.

Mariapaola Nitti, Barbara Marengo, Patrizio Odetti*, Damiano Cottalasso, Maria A. Pronzato,
Umberto M. Marinari, Cinzia Domenicotti.

Department of Experimental Medicine and *Department of Internal Medicine, University of
Genoa, 16132 Genoa, Italy. E-mail:

Protein kinase C (PKC) delta is a critical target of oxidative stress that modulating its activity
mediates proliferative and apoptotic cell signalling pathways.
To determine the biological roles of PKC delta many studies have been carried out in the
presence of rottlerin, an inhibitor that exhibits a specificity for this isoform and that acts as a
mitochondrial uncoupling agent indirectly limiting cellular reactions.
Our previous studies in GSH-depleted neuroblastoma cells have suggested that the biological
action of rottlerin is not only directly related to inactivation of delta isoform but also to an
inhibitory effect on oxyradical production comparable to that induced by classic antioxidants.
In this context, we have clearly demonstrated that GSH-dependent increase in PKC delta
activity and ROS production are two related events during stress-induced cellular responses.
Recently, we have found that a two-fold stimulation of PKC delta activity and ROS
generation (50% increase) triggered by glycoxidative stress in human NTera 2 neurons are
modulated by cell pre-treatment with rottlerin and vitamin E.
These results suggest that scavenger properties of rottlerin, well known as a specific
enzymatic inhibitor, might be utilized to modulate PKC delta activity, a critical target in many
oxidative stress-related cellular pathologies.
(Grants by COFIN 2000, Molecular signalling by GSH depletion, COFIN 2002, Intracellular
signals induced by modulation of glutathione levels, FIRB 2001 Poli Giuseppe).

                                               - 87 -
Session I: Protein kinases as targets for novel treatments          Poster I, 22

Sodium ascorbate modulates insulin signaling affected by reactive oxygen/nitrogen
species in L6 myoblasts

Arkadiusz Orzechowski, Ma_gorzata _okociejewska, Patrycja Muras

Department of Physiological Sciences, Faculty of Veterinary Medicine, Warsaw Agricultural
University, Nowoursynowska 159, 02-776 Warsaw, Poland, E-mail:

Muscle cell proliferation is stimulated by insulin but inhibited by oxidative/nitrosative stress.
One-day pretreatment with sodium ascorbate (1 mM) protected insulin-mediated mitogenesis
in L6 rat myoblasts affected by the reactive oxygen/nitrogen species (ROS/RNS). We
investigated the effect of sodium ascorbate on insulin signaling affected by ROS/RNS during
short-term (0, 5, 10, 15, 20, 25, 30, 45 min) and long-term (24/48 h) studies in L6 myoblasts.
We observed elevated level of Serine473 phospho-PKB/Akt kinase (phospho-PKB/Akt Ser473
- a hallmark of PKB activation) after 24 h treatment with 0.1 uM insulin combined with
ROS/RNS but not with ROS/RNS alone (0.1, 0.5 mM). One-day pretreatment with sodium
ascorbate (1 mM) elevated the level of phospho-PKB/Akt Ser473 in response to insulin
regardless of ROS/RNS presence. Short-term studies revealed that either insulin or hydrogen
peroxide (H2O2) increased PKB/Akt Ser473 level, however, the effect of ROS/RNS peaked
earlier (5 min) than insulin (45 min). Not surprisingly, phospho-PKB/Akt Ser473 level in
response to insulin was reduced by concomitant treatment with H2O2 (0.1 or 0.5 mM) in a
dose-dependent fashion. 4-hour sodium ascorbate pretreatment accelerated and augmented the
signal from phospho-PKB/Akt Ser473 after insulin and H2O2 co-treatment. Furthermore,
immunobloting of p70s6k and c-Jun mirrored the results obtained from phospho-PKB/Akt
Ser473 protein determination. Our studies indicate that insulin-mediated PKB/Akt Ser473
phosphorylation which is initially inhibited by H2O2 (0-45 min) in turn is amplified by
ROS/RNS at the end of the long-term study (24th hour). In either case, sodium ascorbate
pretreatment led to the increased level of phospho-PKB/Akt Ser473. These results explain how
antioxidant pretreatment protected insulin-stimulated mitogenicity and give prospect to
antioxidant preconditioning of muscle cells.

                                              - 88 -
Session I: Protein kinases as targets for novel treatments       Poster I, 23

Elevated NF-kB and Bcl-2 levels in C2C12 myotubes during myogenesis affected by
PD98059, LY294002, or SB 203580 treatment

Arkadiusz Orzechowski, Paulina Kwieci_ska, Bart_omiej Roszkiewicz, Ma_gorzata

Department of Physiological Sciences, Faculty of Veterinary Medicine, Warsaw Agricultural
University, Nowoursynowska 159, 02-776 Warsaw, Poland, E-mail:

Myogenesis is featured by acquisition of an apoptosis-resistant phenotype by myotubes which
are formed in response to the activation of several transcription factors including muscle
regulatory factors (MRFs). On the other hand the initial phase of muscle differentiation is
dependent on the activity of some protein kinases. Among them the phosphatidyl-inositol-3-
kinase (PI-3K), as well as the extracellular signal regulated kinases ERK1/2 (p42 and p44)
and p38 seem to be the most important. The effect of the above-mentioned protein kinase
cascade inhibitors on myogenesis from C2C12 mouse satellite cells was examined
individually during 5 subsequent days. The negative effect of PD98059 (5, 25, 50 uM),
LY294002 (1, 5, 10 uM) or SB203580 (1, 5, 10 uM) on cell viability was evident at initial
period of myogenesis (up to 3rd day). Additionally, 3rd day appeared critical for terminal
differentiation and extensive formation of C2C12 myotubes. On 3rd day the nuclear expression
of myogenin was suppressed by SB203580 in a dose-dependent fashion. In contrast,
decreased cytoplasmic but elevated nuclear expression of myogenin was observed in
myotubes treated with PD98059 or LY294002, respectively. In turn, the consequences of
SB203580 treatment indicate that p38 kinase is involved in the onset of myogenesis. The
cytoplasmic and nuclear expression of NF-kB was elevated after treatment with the above-
mentioned protein kinase inhibitors. Likewise, Bcl-2 expression in the cytosol increased. The
latter observations suggest an existing switch from the MEK/ERK1/2- and PI-3K/PDK1/Akt-
dependent into NF-kB- and Bcl-2-dependent systems which sustain cell viability.
Furthermore, enhanced NF-kB and Bcl-2 protein expression correlated positively with
myogenesis. These studies might shed more light on the safety of clinical use of kinase
inhibitors in the treatment of proliferative diseases.

                                              - 89 -
Session I: Signal transduction pathways as therapeutic targets               Poster I, 24

Role of post-translational modifications on cell distribution and FUNCTIONS of protein
kinase C theta

Mario Passalacqua1, Marco Pedrazzi1, Sabina Ledda1, Bianca Sparatore1, Mauro Patrone2,
Debora Gaggero1, Edon Melloni1 and Sandro Pontremoli1
Department of Experimental Medicine-Biochemistry Section-University of Genoa-Viale
Benedetto XV, 1-16132 Genoa-Italy; 2 Department of Advanced Science and Technology-
University of Eastern Piedmont “Amedeo Avogadro”, Corso Borsalino, 54-15100

The novel protein kinase C (PKC) theta isozyme has been recently proposed as a drug target
for human leukemias. In single cell types we have identified multiple PKC theta forms,
characterised by differential post-translational modifications. Aim of this work was to
establish if these molecular modifications affect PKC theta intracellular localisation and
catalytic competence.
A critical role in the control of both PKC theta cell distribution and catalytic activity is played
by the phosphorylation state of the kinase activation loop, at the Thr538 residue. PKC theta
molecules showing a Mr of 85 kDa (named theta-85) and dephosphorylated at Thr538 have
been found specifically associated with the Golgi complex and catalytically active. In
contrast, PKC theta molecules with a Mr of 76 kDa (named theta-76), partially
phosphorylated at Thr538, are localised in the detergent-soluble cell fraction. Only theta-76
kinase forms not phosphorylated at Thr538 can undergo conversion to theta-85 by an
autophosphorylation process. Cell treatment with calyculin A, a protein phosphatase 1 and 2A
inhibitor or with LY294002, an inhibitor of Thr538 phosphorylation via PI3K/PDK1, results in
a significant increase in the amount of protein kinase PKC theta associated with the Golgi
complex. Moreover, as demonstrated by the behaviour of Thr538ÆAla and Thr538 ÆGlu PKC
theta mutants, the absence of pThr538 is sufficient for the recruitment of PKC theta to the
Golgi complex. Moreover, the Thr538ÆAla PKC theta mutant also results modified by N-
glycosylation. This kinase form binds wheat germ agglutinin and acquires a lower Mr by
treatment with N-glycosydase F. These findings suggest that different PKC theta forms might
be involved in distinct cell functions. New pharmacological strategies, aimed specifically to
control PKC theta activities that promote malignant cell proliferation, should be designed and
screened on the basis of their effect on individual PKC theta forms.

                                               - 90 -
Session I: Signal transduction pathways as therapeutic targets.            Poster I, 25

Investigating the signalling pathways in tumour cells mediated by the sigma-1 receptor.

Evon S. Poon, Steve Hobbs, Will Court, Suzanne Eccles.

Cancer Research UK Centre for Cancer Therapeutics, Institute of Cancer Research, Sutton,
Surrey, SM2 5NG, UK. Email:

The little understood sigma-1 receptor was originally thought to be a member of the
opioid/cannabinoid receptor family. However, it is overexpressed in many cancers and was
found to play a role in sustaining the immortality of tumour cells by providing a powerful
anti-apoptotic signal during proliferation. The sigma-1 receptor has recently been suggested
as a target for therapy. Sigma-1 receptor contains two putative transmembrane domains, one
at the N-terminus and one in the centre of the sequence. Attempts to express the full sequence
in bacteria using different vectors and host strains yielded very little full-length recombinant
protein. Studies with a deletion mutant showed that removal of the N-terminal
transmembrane domain increased expression of the protein. The truncated protein is now
being purified to raise monoclonal antibodies for analytical studies and to act as “bait” to seek
the endogenous ligand for sigma-1, which is unknown. We have also expressed a
fluorescently-tagged sigma-1 protein, which has a GFP sequence at the C-terminus, in a
variety of human tumour cell lines with differentially activated signalling pathways. Using
confocal microscopy and double labelling techniques we have found that the sigma-1 fusion
protein is localised outside the nucleus and appeared to be more concentrated in certain areas
within the cytoplasm. However, further work is required to determine whether it is primarily
associated within specific cellular organelles or substructures. Dynamic image analysis
showed that the synthetic ligands, Pentazocine and Rimcazole, apparently caused
translocation of the sigma-1 receptor towards the plasma membrane. These results point the
way to future studies with this construct which may shed light on the function of the sigma-1
receptor in cancer cells. Future work will explore co-localisation with potential signalling
partners and differences between normal and cancer cells.

                                              - 91 -
Session I : Signal transduction pathways as therapeutic targets           Poster I, 26

Convergence of the protein Kinase C alpha and calpain networks in muscular cells

Dulong Sandrine, Goudenege Sebastien, Poussard Sylvie and Cottin Patrick

USC INRA-429, Université Bordeaux I, Avenue des Facultés, 33405 Talence cedex, France.
E-mail :

Protein MARCKS (Myristoylated Alanine-Rich C Kinase Substrate) is a major cytoskeletal
substrate of protein kinase C (PKC) whose cellular functions are still not defined. However
numerous studies seem to implicate MARCKS in processes linked to the regulation of
cytoskeletal structures during cell adhesion, migration and differentiation. Such events being
also highly controlled by PKC and calpains. Basically, our study concerns the potential role of
Ca2+-dependent proteinases (calpains) during myogenesis via proteolysis of MARCKS. We
have examined in situ, during myogenesis of embryonic myoblasts in culture, the expression
of MARCKS in subcellular compartments. In such conditions we have observed a significant
decrease of MARCKS associated with the appearance of a proteolytic fragment of MARCKS
at the time of the burst of fusion : the fusion being a major event during muscular
differentiation. The addition of CS peptide, a specific calpain inhibitor, in the culture medium
induces a significant decrease of this fragment. We have also shown that endogenous Ca2+ -
proteolysis is dependent of MARCKS phosphorylation by the conventional PKC alpha, this
kinase inducing the cytosolic translocation of MARCKS and interacts in signalling complex
with MARCKS. Furthermore over-expression of MARCKS or addition of antisense
oligonucleotides against MARCKS induces respectively important decreases and increases of
myoblast fusion. Such results suggest that calpain mediated MARCKS proteolysis and protein
kinase C alpha signalling interconnect in muscular cells and are necessary for myoblast
fusion. Current proteomic studies on C2C12 muscular cell lines and rhabdomyosarcoma-
derived human muscle cell lines (do not undergo myoblast fusion) would allow us to precise a
more complete signaling network involved in this cancer and other muscle diseases.
Aknowledgments : We are grateful to Dr. S. Manenti for the gift of MARCKS expression
vectors and for stimulating discussions.

                                             - 92 -
Session I: Protein kinases as targets for novel treatments         Poster I, 27


Ricote M, García-Tuñon I, Fraile B, Paniagua R and Royuela M

Department of Cell Biology and Genetics, Uniersity of Alcalá. Alcalá de Henares. Madrid.
Spain. E-mail:

Tumor necrosis factor-alpha exerts several functions as apoptosis or proliferation throughout
an intracellular transduction pathway that involves different mitogen-activated protein kinases
as JNK and p38. Via JNK might be inhibited by several anti-apoptotic factors (bcl-2 or p21)
and TNF-alpha signal is displaced towards p38 activation at MEK-4 level. Furthermore, p38
is activated by stress, cytokine and hormone stimulus throughtout PAK-1 and MEK-6
transduction pathway. This pathway has been involved in apoptosis but its activation pathway
in vivo, especially in prostate, has not yet been addressed enough. The aim of this study was
to investigate, by means of immunohistochemistry and western blot, the expression of PAK-1,
MEK-6 and p38 in normal prostate (NP), benign prostatic hyperplasia (BPH) and prostate
cancer (PC).
Biopsies from 15 NP, 35 BPH and 27 PC were processed by immunohistochemistry and
western blot to study the expression and distribution of PAK-1 and MEK-6 and p38.
In NP, no immunoreaction was found to PAK-1, MEK-6. P38 was found in basal epithelial
cells in NP. In BPH, 70.3% of samples to PAK-1, 59.5% to MEK-6 and 90% to p38, showed
immunoreaction epithelial cells. In PC, the percentages and the intensity of immunoreaction
increased to all proteins.
These results suggests that this transduction pathway is over stimulated when the
malignization increase by TNF-alpha or other pro-apoptotic stimulus; perhaps as an attempt to
maintain the apoptosis/proliferation balance to conunteract proliferative stimulus.
Furthermore, p38 has been studied as potential mechanism of chemotherapy, since, lower or
lack of p38 activation correlates with a more resistant phenotype to these types of treatment in
other cancers, so its presence may be well progression therapy marker and the potentiation of
this p38 transduction pathway may be an important target to cancer prostate therapy.
ACKNOWLEDGEMENTS : This work was supported by grants from Fondo de
Investigaciones Sanitarias (PI020383) and the University of Alcalá.

                                              - 93 -
Session I: Protein kinases as therapeutic targets       Poster I, 28

Control of acute inflammation by C-terminal Src kinase (Csk)

Richard M. Thomas, Christian Schmedt1, Marco Novelli, Jane Skok, Alexander Tarakhovsky1
and Jurgen Roes*

University College London, Department of Immunology and Molecular Pathology, The
Windeyer Institute of Medical Sciences, 46 Cleveland Street, London W1T 4JF, UK
  Rockefeller University, Laboratory of Lymphocyte Signaling, 1230 York Avenue, 10021
New York

Swift and robust inflammatory responses are needed to restrict the spread of microbial
pathogens at the early stages of an infection. Polymorphonuclear granulocytes (PMN,
neutrophils, granulocytes) are rapidly recruited from the blood stream. The recruitment
process is coupled with successive stages of activation ultimately enabling efficient
phagocytosis and killing of ingested microbes. While potent microbicidal enzymes released
into the phagolysosome are necessary for efficient neutralisation of the pathogen, they are
also implicated in immuno-pathogenesis. To establish whether the widely expressed regulator
of Src family kinases, Csk, contributes to the regulation of acute inflammation in vivo, we
inactivated csk in granulocytes by conditional mutagenesis (Cre/loxP). Mutant mice (Csk-
GEcre) developed acute multifocal inflammation in skin and lung. Animals were protected
from the disease in a microbiologically controlled environment, but remained hypersensitive
to LPS-induced shock. Csk-deficient granulocytes showed enhanced spontaneous and ligand
induced degranulation with hyper-induction of integrins. The hyper-responsiveness was
associated with hyper-adhesion and impaired migratory responses in vitro. Hyper-
phosphorylation of key signaling proteins such as Syk and Paxillin in mutant granulocytes
further supported break-down of the activation threshold set by Csk. By enforcing the need
for ligand engagement Csk thus prevents premature granulocyte recruitment while supporting
the motility of stimulated cells through negative regulation of cell-substratum adhesion.

                                              - 94 -
Session I: Protein kinases as targets for novel treatments         Poster I, 29

Genome-wide Functional Annotation of the Sch9/PKB protein kinase in Saccharomyces

Johnny Roosen, Bart Smets and Joris Winderickx

Laboratory of Functional Biology, Institute of Plant Physiology, Katholieke Univerisiteit
Leuven, Kasteelpark Arenberg 31, 3001 Heverlee-Leuven, Belgium, E-Mail :

The Sch9 protein kinase in Saccharomyces cerevisiae is an AGC serine/threonine protein
kinase and is functionally related to the mammalian PKB/Akt (Geyskens et al., 2000). Similar
to PKB, Sch9 has been implicated in the regulation of cell size (Jorgenson et al., 2002) and
longevity (Fabrizio et al., 2001). Futhermore, Sch9 is required for proper amino acid- and
nitrogen-induced signalling (Crauwels et al., 1997). We recently demonstrated that Sch9
regulates the intracellular localization of the protein kinase Rim15 (Pedruzzi et al., 2003).
However, it is rather unclear how the activity of Sch9 is altered in response to nutritional
stimuli and how its signal is transmitted to other downstream targets.
Genome-wide expression analysis during growth on glucose and upon entry into diauxic shift
show that Sch9 is mainly involved in the regulation of amino acid biosynthesis, ribosome
biogenesis, transcriptional control, cell cycle and DNA processing, lipid, fatty acid and
isoprenoid metabolism and cell rescue. Most interestingly, Sch9 exerts mainly a repressive
function in logarithmic phase whereas it switches to activation of transcription upon entry into
diauxic shift. Application of the compendium approach (Hughes et al., 2000) revealed that
Sch9 is most closely related to components involved in oxidative stress and the calcineurin
pathway as well as conditions of nutrient deprivation such as rapamycin treatment. Promotor
analysis using REDUCE (Bussemaker et al., 2001) identified Sch9-dependent cis-acting DNA
elements involved in oxidative stress, amino acid biosynthesis and cell cycle regulation.
Together, we identified new Sch9-dependent transcriptional targets and could reveal some of
the underlying mechanisms involved in different physiological processes. Knowledge of the
cis-acting DNA elements and their corresponding transcription factors will reveal new
insights in the function of Sch9 and mammalian PKB/Akt and might lead to new therapeutics
in PKB-related diseases by application of for instance the ‘Transcription Factor Decoy’

                                              - 95 -
Session I: Protein kinases as targets for novel treatments          Poster I, 30

Down-regulation of Akt/PKB signaling pathway by inhibition of CK2 activity in PTEN-
null cells

Maria Ruzzene, Giovanni Di Maira, Mauro Salvi and Lorenzo A. Pinna

Venetian Institut of Molecular Medicine (VIMM) and Department of Biological Chemistry,
University of Padova, 35121Padova, Italy. E-mail:

Akt/PKB and CK2 are two Ser/Thr protein kinases playing prominent roles in cell survival.
Akt/PKB is linked to the PI3K/PDK1 pathway, being activated in response to specific signals,
and promotes cell survival by phosphorylating a number of proteins with evident functions in
cell decision between death and life. On the other hand, CK2 is constitutively active,
ubiquitous and highly expressed in tumors, but, despite the huge number of its substrates, the
molecular mechanism by which it exerts an antiapoptotic function is still largely unknown.
We used two different approaches to reduce CK2 activity in cells, namely, pharmacological
inhibition with a panel of chemically different CK2-specific inhibitors, and knock-down of
the CK2 catalytic subunits by the RNA interference technique. In both cases, we observed a
down-regulation of Akt signaling, as judged by reduced phosphorylation of the activation
sites in Akt itself (Thr308 and Ser473) and of the Akt-dependent phosphorylation site Ser9 in
GSK3. The effect seems to be played directly at the level of Akt protein, since PDK1, the
upstream kinase, is not affected. When CK2 is inhibited, we also found a reduced association
of Akt with Hsp90, a chaperone protein known to be a substrate of CK2, but also a partner of
Akt responsible for its protection from dephosphorylation.
In vitro experiments show that CK2 can directly phosphorylate Akt (at sites different from
T308 and Ser473), inducing a hyperactivation of Akt.
We therefore suggest that CK2 can provide an additional mechanism for Akt regulation, both
acting directly on the enzyme and controlling its dephosphorylation at the canonical sites; our
data also indicate that the anti-apoptotic role of CK2 is played, at least in part, by exploiting
the survival signals of Akt.

                                              - 96 -
Session I: Protein kinases as targets for novel treatments          Poster I, 31

DYRKs and CK2: development of specific inhibitors that discriminate between these
two classes of protein kinases.
    * Stefania Sarno, # Flavio Meggio, $ Zygmunt Kazimierczuk and #* Lorenzo A. Pinna.
Department of Biological Chemistry, *Venetian Institute for Molecular Medicine (VIMM)
University of Padova, Italy; $Institute of Chemistry, Agricultural University, Warsaw, Poland.

Dyrk1A is a member of a new family of protein kinases comprising at least seven mammalian
isoforms. It has been defined as a dual specificity kinase, phosphorylating Ser/Thr residues in
protein substrates and Tyr residues in its catalytic domain. In particular the phosphorylation of
two tyrosines in the activation loop is a requirement for its kinase activity. Increased
expression of Dyrk1A kinase appears to play a significant role in neuropathology of Down
syndrome, and the disruption of fully differentiated neurons in trisomy 21 seems to be
Protein kinase CK2 is the most pleiotropic Ser/Thr protein kinase with more than 300 protein
substrates identified to date. It is implicated in a wide variety of cellular functions, with
special reference to gene expression, signal transduction, RNA and protein synthesis. The
catalytic alpha subunits are constitutively active either alone/or in combination with the
regulatory beta subunits to give the heterotetrameric holoenzyme, and its activity is essential
to cell viability. In a wide variety of tumors an elevated level of CK2 activity was observed as
compared to normal tissues, supporting the implication of CK2 in aberrant proliferation.
Given these premises considerable efforts have been recently made toward the development
of selective CK2 inhibitors. This led to the observation that most CK2 inhibitors, with special
reference to TBB (tetrabromo-benzotriazole), strongly inhibit besides CK2, only Dyrk1A
amid a panel of > 30 protein kinases tested, supporting the view that the ATP binding site of
the two kinases might share some common features. A systematic analysis of TBB derivatives
was therefore performed on CK2 and Dyrk1A to identify derivative(s) that could discriminate
between the two kinases and lead on one side to the optimization of CK2 inhibitors, on the
other to the development of compounds capable to specifically inhibit DYRKs but not CK2.

                                              - 97 -
Session I: Protein kinases as targets for novel treatments       Poster I, 32

Alterations in the activation pathways of Protein Kinase B and Erk in the progression to
metastasis of human prostate cancer

R Michael Sharrard, Jennifer C Spalton and Norman J Maitland.

YCR Cancer Research Unit, Biology Dept, University of York, York YO10 5YW, UK

     Progression to invasiveness and metastasis in epithelial tumours is characterised by cell
growth and survival independent of growth factors (GFs) and cell-cell and cell-matrix contact.
These functions are modulated by Protein Kinase B (PKB/Akt), activated by
phosphatidylinositol 3-kinase (PI3K) via PIP3, and by MAP kinases erk1/2 activated through
ras, raf and MEK1/2. Enhanced survival of metastatic prostate tumour cells has been linked
to PKB deregulation through loss of PTEN, which antagonises PI3K by degrading PIP3.
      PNT2 and PNT1a (non-tumour epithelial), P4E6 (early tumour) and LNCaP and PC3
(metastatic) prostate cell lines were grown overnight without GFs. Kinase inhibitors
AG1478 (EGFR), LY294002 (PI3K), PP2 (src) or PD98059 (MEK) were added singly or in
pairs for 2 hours, followed by 5 ng/ml EGF or IGF1. Phosphorylation of PKB/Akt at ser473
and of erk1/2 at thr202/tyr204 was assayed by Western blotting and immunostaining. IGF1
receptor (IGF1R) and its downstream adaptor protein IRS1 were assayed by Western blotting.
      Prostate tumour cells showed reduced dependence of PKB/Akt activation on GFs and
enhanced sensitivity to the PI3K inhibitor LY294002. LNCaP and PC3 cells express reduced
levels of IGF1R, while PNT2 and LNCaP lack expression of IRS1. In non-tumour cells EGF
activates PI3K/PKB via src and MEK/erk independently of src, while IGF1 activates
PI3K/PKB (but not MEK/erk) via src-dependent and -independent pathways. MEK inhibitor
PD98059 enhances PI3K/PKB activation, possibly via interactions of MEK with the PI3K
pathway independent of its erk-kinase activity. In P4E6 tumour and PC3 metastatic cells,
alterations to pathway effectors between the GF receptors and src lead to GF-independent
PKB activation, while in LNCaP PKB is constitutively activated independent of src.
Immunocytochemistry shows that phosphorylation of both PKB and erk is linked to the cell
cycle. Thus GFs may affect PKB and erk activation both through their respective signalling
pathways and indirectly through effects on cell proliferation.
     We conclude that elucidation of the molecular lesions and altered patterns of cross-talk
between mechanisms activating PKB and erk1/2 in prostate tumours will allow formulation of
combinations of specific protein kinase inhibitors to target these cancers and their metastases.

                                              - 98 -
Session I: Protein kinases as targets for novel treatments         Poster I, 33

Expression of ion channels and their role in intracellular pH regulation
in normal human nasal epithelial cells

Ji-Hyun Shin_, Eun-Jin Yang_ , Joo-Heon Yoon, MD__
_Department of Otorhinolayngology, Yonsei University College Medicine, Seoul; and
_ Department of Medical Science, Brain Korea 21 Project for Medical Sciences, Yonsei
University College of Medicine, 134 Shinchon-dong, Seodaemun-gu, Seoul , 120-752, Korea
, E-mail:

Intracellular pH (pHi) regulation is important in airway pathophysiology. Acid-base channels
exert their effects on pHi regulation of airway epithelial cells. The aim of this study was to
investigate the expression of the ion channel isoforms in normal human nasal epithelial cells
according to the culture duration and to examine their functional role in the regulation of pHi.
In anion exchangers, AE2, bAE3 and AE4 mRNA were expressed. In sodium proton
exchangers, NHE1- 5 and 7 mRNA were expressed. In sodium bicarbonate cotransporters,
NBC1-4 mRNA were expressed. Fluorescence intensity of NHNE cells were monitored using
the pH-sensitive fluorescent dye, BCECF-AM. 1mM diphenylamine-2-carboxylate, an
inhibitor of Cl- channel, and 500_M 4,4'-diisothiocyanatostilbene-2,2'-disulfonate, an inhibitor
of HCO3-coupled channel, significantly suppressed AE activity in the basolateral membrane.
NHE was inhibited by 1_M 3-methylsulphonyl-4-piperidinobenzoyl guanidine
methanesulphonate, an NHE1 selective inhibitor, in both luminal and basolateral membranes.
In addition, 30_M 5-(N-ethyl-N-isopropyl)-amiloride, an inhibitor of Cl- channel, inhibited
Na+ dependent pHi recovery. NBC pHi recovery was more effectively inhibited by both
DIDS 500_M and EIPA 30_M than by only EIPA30_M. Taken together, these results suggest
that all ion channels may interact with each other and regulate intracellular pH. supported by
the BK21 Project for Medical Science, Yonsei University.

                                              - 99 -
Session I: Protein kinases as targets for novel treatments        Poster I, 34

IL-4 enhances wound rate closure in lung epithelial cells that is EGF-receptor dependent
and induces activation of ERK1/2.

Sandra van Wetering, Klaus F. Rabe and Pieter S. Hiemstra.

Dept of Pulmonology, LUMC, P.O. Box 9600, 2300 RC, Leiden, The Netherlands. E-mail.

Asthma is a chronic inflammatory disease characterised by variable airflow obstruction and
airway hyperresponsiveness. During the development of asthma structural changes in the
airways occur. These changes include epithelial desquamation, goblet cell hyperplasia and
mucus hypersecretion. Various studies have indicated that these features are, at least in part,
regulated by the Epidermal Growth Factor Receptor (EGFR). It has been suggested that the
epithelial repair response in asthma is impaired, despite the relative abundant epithelial
expression of the EGFR. Increasing evidence obtained from in vivo and cell culture studies
points to a role of the Th2 cytokines IL-4, IL-9 and IL-13 in goblet cell hyperplasia and
mucus hypersecretion. However, the effect of Th2 cytokines on epithelial repair is unclear.
Therefore the aim of this study was to examine the effect of IL-4 on airway epithelial wound
closure and the involvement of the EGFR. Using NCI-H292 and 16HBE bronchial epithelial
cells, we observed that IL-4 induced a time- and dose-dependent enhancement of wound
closure. The stimulatory effects of IL-4 were observed at concentrations of 10 ng/ml and
higher. A significant difference in wound closure was already observed after 24 hours, but
was most prominent between 48 and 72 hours where the closure rate in the presence of IL-4
was 1.5 to 2 fold higher as compared to control-treated cells. In the presence of an antibody
against the EGFR these effects were completely abolished. Following activation of the EGFR,
stimulation of various signalling pathways occurs, including activation of ERK1/2. Therefore
we also studied the effect of IL-4 on ERK1/2 activation. We observed that IL-4 induced
ERK1/2 activation within 5 min, and that this persisted up to 20 min. ERK1/2 activation was
prevented by the EGFR inhibitor AG1478 and the MEK inhibitor U0126. These results
indicate that the Th2 cytokine IL-4 promotes epithelial restitution, and that this process
involves activation of the EGFR and downstream signalling pathways.
Supported by a grant of the Netherlands Asthma Foundation.

                                             - 100 -
Session I: Protein kinases as targets for novel treatments        Poster I, 35

MNNG induces p53 serine 15 phosphorylation by ATM and ATR kinases:
activation of PAI-1 gene expression.

Berta Vidal, Maribel Parra, Mercè Jardí and Pura Muñoz-Cánoves
Centre de Regulació Genòmica (CRG), Programa de Diferenciació i Cancer, Barcelona

The alkylating agent MNNG is an environmental carcinogen that causes DNA lesions leading
to cell killing; MNNG also triggers a protective cellular response characterized by the
induction of DNA repair/transcription-related genes. We recently demonstrated that the
expression of the extracellular serine protease inhibitor PAI-1 was induced by MNNG in a
p53-dependent manner. However, the mechanism(s) linking external MNNG stimulation and
PAI-1 gene induction remained to be elucidated. Here we show that MNNG induces
phosphorylation of p53 at serine 15, resulting in stabilization and nuclear accumulation of the
protein. We further demonstrate that ATM and ATR kinases, which participate in DNA
damage-activated checkpoints, regulate phosphorylation of serine 15 in MNNG-treated cells.
Using ATM-deficient cells, ATM was shown to be required for the early phosphorylation of
serine 15 in response to MNNG, whereas overexpression of catalytically inactive ATR
selectively interfered with late phase serine 15 phosphorylation. In agreement with this, the
sequential activation of both kinases was also required for the adequate induction of PAI-1
gene transcription in response to MNNG. Since PAI-1 is involved in the control of tumor
invasiveness, being a prognostic factor of many metastatic cancers, our finding that a
genotoxic alkylating agent induces PAI-1 gene expression via ATM/ATR-mediated
phosphorylation of p53 provides a new feature for the role of these DNA damage-induced cell
cycle checkpoint kinases. Moreover, these results suggest that alkylating carcinogens may
contribute to tumor metastasis by inducing PAI-1 gene without involving genetic alterations.

                                             - 101 -
Session I: Protein kinases as targets for novel treatments     Poster I, 36

Characterization of signal transduction pathways leading to the activation of Dyrk1A
during the neuronal differentiation and cell death in hippocampal neuroprogenitor cells

Eun Jin Yang1,3, Joo Heon Yoon2,3, and Kwang Chul Chung4
 Dept of Medical Science, 2Otorhinolaryngology, 3Brain Korea 21 Project for Medical
Sciences, Yonsei Univ. Coll. Med., Seoul 120-752; 4Dept of Biology, Yonsei Univ Coll. Sci.,
Seoul 120-749, Korea E-mail:

Dual specificity protein kinase Dyrk1A is involved in normal embryogenesis and brain
development. Defects in this kinase have been suggested to play an important role in the
mental retardation of Down syndrome patients. In the present study we examined the possible
upstream signal transduction pathways leading to Dyrk1A activation during neuronal
differentiation in in immortalized hippocampal progenitor H19-7 cells. The tyrosine
phosphorylation and the kinase activity of Dyrk1A appeared to be increased by MEK and PI-
3K-Akt under neuronal differentiation in H19-7 cells, whereas MEK and PI-3K suppressed
Dyrk1A activity in quiescent H19-7 cells. Furthermore, MEK1 could directly binds to and
phosphorylate Dyrk1A during bFGF-induced neuronal differentiation of H19-7 cells.
Interestingly, JNK activity was also markedly induced by bFGF, which subsequently led to
the phosphorylation of Dyrk1A. Meanwhile, the addition of etoposide resulted in the
activation of JNK, which then selectively phosphorylates Dyrk1A. These findings suggest
that Dyrk1A activity is differentially modulated through MAP kinases as well as PI-3K in
hippocampal neuronal cells.
supported by the BK21 Project for Medical Science, Yonsei University.

                                             - 102 -
Session I: Protein kinases as targets for novel treatments       Poster I, 37

Rat preadipocyte differentiation is regulated by glucocorticoid and TGF-b

Young Yang, Sun Mi Shin, Kun-yong Kim, Jae Kwang Kim, Inpyo Choi

Laboratory of Immunology, Korea Research Institute of Bioscience and Biotechnology,
Daejeon 305-333, Korea.

Dexamethasone and TGF-b show contrary effects on differentiation of adipocytes.
Dexamethasone stimulates adipocyte differentiation whereas TGF-b inhibits it. In the present
study, we investigated whether dexamethasone could reverse the TGF-b-mediated inhibition
of preadipocyte differentiation. Primary rat preadipocytes, obtained from SD rats, were
pretreated with dexamethasone in the presence or absence of TGF-b, prior to the induction of
differentiation. Co-treatment of dexamethasone and TGF-b before inducing differentiation
reversed the TGF-b-mediated inhibition of preadipocyte differentiation. In order to elucidate
the mechanism by which dexamethasone reversed the effect of TGF-b on the inhibition of
preadipocyte differentiation, the expressions of C/EBPa and PPAR-g were examined.
Dexamethasone increased the C/EBPa and PPAR-g expressions in the absence of TGF-b and
also recovered the TGF-b-mediated suppression of C/EBPa expression in preadipocytes. Its
effect was sustained in differentiated adipocytes as well. However, those effects were not
observed in 3T3-L1 preadipocytes or differentiated adipocytes. These results indicate that
dexamethasone reverses the TGF-b-mediated suppression of the adipocyte differentiation by
regulating the expressions of C/EBPa and PPAR-g, which is dependent on the cellular
contexts (This research was supported by a grant (CBM1-B114-001-1-0-0) from the Center
for Biological Modulators of the 21st Century Frontier R&D Program, the Ministry of
Science and Technology, Korea.)

                                             - 103 -
Session I: Protein kinases as targets for novel treatments       Poster I, 38

Th2 cytokine-induced eotaxin release by human airway smooth muscle cells:
involvement of the MAPkinase ERK

Suzanne Zuyderduyn, Pieter S. Hiemstra, Klaus F. Rabe

Department of Pulmonology, Leiden University Medical Center, Leiden, PO Box 9600, 2300
RC Leiden, The Netherlands. E-mail:

Human airway smooth muscle cells (HASM) are thought to play an important role in the
pathogenesis of asthma, based on their contractile properties and their more recently
recognized function as producers of inflammatory mediators. HASM produce eotaxin in
response to the Th2 cytokines IL-4 and IL-13, which are involved in airway inflammation, but
may also cause structural changes in the airways (airway remodelling). It has been shown that
activation of Mitogen-Activated Protein kinases (MAPkinases) leads to enhanced cell
proliferation and differentiation, which in turn may cause remodelling. Extracellular signal-
regulated kinase-1 and -2 (ERK-1/-2) are MAPkinases know to be involved in proliferation.
The ERK pathway can be activated by receptors such as the Epidermal Growth Factor
Receptor (EGFR). We have studied whether the EGFR and the ERK pathway are involved in
the eotaxin and eotaxin-3 production by HASM induced by both IL-4 and IL-13.
Both IL-4 and IL-13 induced release of eotaxin and eotaxin-3 by HASM. The Th2 cytokines
induced activation of the MAPkinases ERK-1/-2, which was inhibited by preincubation with
U0126 and PD98059 (both inhibitors of the MAPkinase kinase MEK). In addition, these
compounds inhibited eotaxin and eotaxin-3 release by HASM. Blocking the EGFR (by adding
a blocking antibody or by inhibiting the tyrosine kinase) did not inhibit ERK-1/-2 activation
or release of chemokines after stimulation with IL-4 and IL-13, suggesting that the release of
the eotaxins was independent of transactivation of the EGFR.
We conclude that IL-4 and IL-13 induced release of eotaxin and eotaxin-3 is dependent on
activation of ERK-1/-2 but independent of transactivation of the EGFR.
This study was supported by grants from the Netherlands Asthma Foundation (AF00.17) and
AstraZeneca, Sweden

                                             - 104 -
Session II: Angiogenesis

                    - 105 -
Session II: Angiogenesis             Poster II, 1


Gianluca Baldanzi§ #, Stefania Mitola§¥, Santina Cutrupi#, Nicoletta Filigheddu#, Wim J van
Blitterswijk$, Fabiola Sinigaglia#, Federico Bussolino*¥, Andrea Graziani ¶*#

From the # Department of Medical Sciences, University Amedeo Avogadro of Piemonte
Orientale, Novara, Italy; ¥ Institute for Cancer Research and Treatment (I.R.C.C.) and
Department of Oncological Sciences, University of Torino, Italy, $ Division of Cellular
Biochemistry at the Netherlands Cancer Institute, 1066 CX Amsterdam, The Netherlands.
§,* These authors contributed equally to this work.

Vascular endothelial growth factor A (VEGF-A) promotes angiogenesis by stimulating
migration and proliferation of endothelial cells in a coordinate manner, resulting in the
formation of tubule-like structures. The signaling pathways by which VEGF-A stimulates
angiogenesis upon activation of its tyrosine kinase receptor, VEGFR-2, require the activation
of Src tyrosine kinase and of PI 3-kinase activities. We had previously shown that Src-
mediated activation of Diacylglycerol kinase-a , an enzyme which phosphorylates
diacylglycerol to phosphatidic acid, is required for HGF-stimulated chemotaxis in endothelial
cells. Thus we have investigated the role of Diacylglycerol kinase-a in VEGF-A signaling in
endothelial cells. We demonstrate that VEGF-A stimulates the enzymatic activity of
Diacylglycerol kinase-a in a Src-tyrosine kinase dependent manner and induces the formation
of a Src/Diacylglycerol kinase-a complex. We also demonstrate that stable expression of a
dominant negative form of Diacylglycerol kinase-a impairs chemotaxis and proliferation
induced by VEGF-A in PAE-KDR cells. In addition we show that either pharmacological
inhibition of Diacylglycerol kinase-a or its downregulation by transfection with specific
siRNA, impairs VEGF-A-induced chemotaxis, proliferation and in vitro angiogenesis of both
PAE-KDR and HUVEC endothelial cells. These results by demonstrating that the activation
of Diacylglycerol kinase-a is required for VEGF-A signaling, suggest that it may generate a
signal which is essential for both the proliferative and migratory response of endothelial cells
to VEGF-A, and indicate that it may constitute a novel pharmacological target for
angiogenesis control.

                                             - 106 -
Session II: Angiogenesis             Poster II, 2

Phosphodiesterase 4 inhibition synergizes with relaxin signaling to promote
decidualization of human endometrial stromal cells.

Olaf Bartsch, Bettina Bartlick, Richard Ivell.

Institute for Hormone and Fertility Research, University of Hamburg,
Falkenried 88, 20251 Hamburg, Germany. E-mail:

From being one of the earliest hormones described with a very specific function in parturition,
recent research identidfied relaxin as a vasoactive hormon which is involved in tumor
progression. Relaxin upregulates vascular endothelial growth factor to promote
neovascularization and activates matrix metalloproteinases to fascilate tumor invasion. Our
findings indicated that a tyrosine kinase-dependent pathway couples relaxin signaling to the
cAMP degrading activities of phosphodiesterases (PDE). We have chosen a pharmacological
approach to test whether relaxin binding and PDE inhibition cooperate to induce
decidualization of human endometrial stromal cells. This profound differentiation process
during the menstrual cycle is an essential prerequisite for the implantation of a blastocyst.
Decidualization is typified by a marked development of spiral arteries and accompaignied by
sustained elevated intracellular cAMP concentrations in vivo. The PDE4 inhibitor rolipram
was most effective in elevating intracellular cAMP concentrations and synergizing with
relaxin to achieve maximal in vitro decidualization, as determined by measurement of the
expression of the decidual marker genes for prolactin and IGF-BP1, by measurement of
prolactin secretion and by detection of the typical cobble-stone-morphology of decidualized
endometrial stromal cells. Gene expression for PDE4D and PDE4C was significantly
upregulated during in vitro decidualization. Treatment of cell cultures with the PKA inhibitor
H89 revealed a minor role for PKA-mediated positive feedback control of PDE4 activity,
consistent with sustained elevated cAMP essential for decidualization in vitro. These findings
introduce the new idea of clinically applying the combination of a specific PDE4 inhibitor
with an effector such as relaxin, thereby offering an alternative non-steroidal luteal phase
support for the endometrium to encourage endometrial development and implantation in
subfertile women undergoing ART (assisted reproductive technology) procedures.

                                             - 107 -
Session II: Angiogenesis             Poster II, 3

Radiation induced increase of bFGF in HNSCC is abolished by the addition of the Cox-
inhibitor Fluriprofen

Jürgen Brieger, Petra Schroeder, Wolf J. Mann

Laboratory of Molecular Tumour, Department of Otorhinolaryngology, University Hospital
of Mainz, Germany, 55101 Mainz, Germany, eMail:

Operation and radiation therapy are the common treatment of head and neck squamous cell
carcinoma. It has been shown that radiation might induce by unknown mechanisms the
expression and secretion of VEGF (vascular endothelial growth factor), one of the most
potent endothelial growth factors known. In this study we analyzed the induction of VEGF
and another strong angiogenic growth factor, bFGF (basic fibroblast growth factor), after
radiation in HNSCC (Head and Neck Squamous Cell Carcinoma) cell lines in vitro. We
observed a time and dosage dependant strong increase of VEGF- and bFGF-secretion into the
culture media. In further experiments we analyzed the potential of the Cox (cyclooxygenase)
–inhibitor Fluriprofen to reduce this radiation mediated release. We observed a slight decrease
of VEGF-secretion and a strong decrease (about –80% compared to the culture without
Fluriprofen) of bFGF. We suggest that Fluriprofen should be further analyzed for its potential
as radiosensetizer in radiation therapy of HNSCC. The targeted pathways have to be analyzed
in further studies.

Meeting: “Signal transduction pathways as therapeutic targets.”
Luxembourgh 25.1. – 28.1.2004

                                            - 108 -
Session II: Chromatin structure in health and disease              Poster II, 4

Modulation of nuclear processes by new analogues of anthracycline antibiotics

Agata Szu_awska, Marek Gniazdowski & Malgorzata Czyz

Department of Medicinal Chemistry, Medical University of Lodz, 6/8 Mazowiecka Street, 90-
131 Lodz, Poland, e-mail:

DNA-binding drugs appear to affect cellular processes such as activity of DNA and RNA
polymerases, topoisomerases, nucleases, binding of transcription factors and other nuclear
proteins. Therapeutic strategies for tumors are directed to induce differentiation and/or growth
inhibition of malignant cells. Anthracyclines such as doxorubicin and daunorubicin have a
broad spectrum of activity in human tumors and are widely used in chemotherapy. New
derivatives of daunorubicin and doxorubicin with a morpholine ring were synthesized.
Several methods have been applied in order to characterize the mode of action of these
compounds in comparison with parental drugs. The DNA sequence-specific binding ability of
anthracycline derivatives has been investigated using both DNase I footprinting and
Restriction Endonuclease Protection. The combined results of DNA footprinting and
restriction endonuclease analysis show that the G.C pair is required for covalent binding of
anthracycline derivatives since different sequences containing G.C pair are protected from
endonuclease digestion. The 5’-GC-3’, 5’-CG-3’ and 5’-TC-3’ are inhibited most commonly
by parental compounds and their morpholine derivatives. Some increased protection of 5’-
TC-3’ and 5'-CT-3' sequences is observed for morpholine analogues comparing to the
parental drugs. Since most of the anthracycline derivatives display selectivity towards G.C
pair, their effect on the interaction between transcription factor Sp1 and its doubled binding
site has been analyzed by means of EMSA. 50% of inhibition is observed when 10
micromolar concentrations of the drugs is used. New anthracycline derivatives have been also
used at nanomolar concentrations to induce erythroid differentiation in the human leukemia
cell line K562. To determine if K562 cell line is capable of expression of erythroid genes
including gamma-globin and transcription factor GATA-1 following exposure to the
anthracycline derivatives, their RNA levels have been determined by real time PCR. GATA-1
protein level has been monitored by its DNA-binding activity in EMSA. Results of the
experiments showing the differences between parental drugs, doxorubicin and daunorubicin,
and their morpholine derivatives will be presented.

                                             - 109 -
Session II, Angiogenesis            Poster II, 5

HTLV-I infected cells extravasate through the endothelial barrier using a local
angiogenesis-like mechanism: implications for invasion and metastasis

Ali Bazarbachi1, Raghida Abou Merhi1*, Antoine Gessain2*, Hilda El-Khoury1*, Rihab Nasr1*,
Olivier Gout3*, Fadia Homaidan1*, Hugues de Thé4*, Olivier Hermine5, and Marwan E. El-
Sabban1*. 1American University of Beirut, Lebanon; 2Institut Pasteur, Paris, France;
  Fondation Rotschild, Paris, France; 4UPR 9051 CNRS, Paris, France; 5Necker Hospital,
Paris, France.

Extravasation of tumor cells through the endothelial barrier is a critical step in cancer
metastasis. HTLV-I associated adult T-cell leukemia/lymphoma (ATL) is an aggressive
disease characterized by frequent visceral invasion. We have shown that ATL cells, in vitro,
synthesize biologically active angiogenic factors. In this report we show that ATL and HTLV-
I associated myelopathy (TSP/HAM) patients exhibit very high plasma levels of vascular
endothelial growth factor (VEGF) and basic fibroblast growth factor. Accordingly, plasma
from ATL and TSP/HAM patients induce angiogenesis in vitro, a phenomenon inhibited by
anti-VEGF antibodies. We also show that the HTLV-I oncoprotein Tax trans-activates the
promoter of the gap junction protein connexin-43 and enhances gap junction-mediated
heterocellular communication with endothelial cells. The interaction of HTLV-I transformed
cells with endothelial cells induces matrix metalloproteinases MMP-2 and MMP-9 gelatinase
activity and downregulates the tissue inhibitor of MMP, TIMP-1, in endothelial cells. This
results in the degradation of sub-endothelial basement membrane followed by endothelial cell
retraction, reminiscent of angiogenesis, hence allowing neoplastic lymphocytes extravasation.
This local angiogenesis-like sequence induced by HTLV-I infected cells may facilitate central
nervous system invasion in TSP/HAM or visceral invasion in ATL. Using ATL as a model,
we propose that following specific adhesion to endothelia of target organs, tumor cells induce
a local and transient angiogenesis-like mechanism through paracrine stimulation and direct
cell-cell communication with endothelial cells. This culminates in a breach of the endothelial
barrier function allowing cancer cell invasion.

                                            - 110 -
Session II: Angiogenesis            Poster II, 6

FGF8b mediates angiogenesis through repression of TSP1 in S115 mouse mammary
tumor cells

Mirjami MT. Mattila*, Kati M. Tarkkonen*, Jani A. Seppänen, Johanna K. Ruohola, Eeva M.
Valve, Pirkko L. Härkönen, Institute of Biomedicine, Department of Anatomy and Medicity
Research Laboratory, University of Turku, Kiinamyllynkatu 10, 20520 Turku, Finland,
E-mail:, *equal contribution

Increased expression of fibroblast growth factor 8 (FGF-8) has been observed in several
forms of hormonal cancer including human breast cancer, ovarian cancer and prostate cancer.
In MCF-7 human breast cancer cells over-expression of FGF-8 leads to stimulated
proliferation in vitro and increased growth and angiogenesis in vivo in nude mouse tumours.
In S115 mouse breast cancer cells FGF-8 is expressed in an androgen-controlled manner and
it has been suggested to mediate most if not all of the features of androgen-induced
transformed-like phenotype in these cells. In order to look for androgen-regulated genes
responsible for malignant phenotype in S115 mouse breast cancer cells we performed a
cDNA microarray analysis of the cells treated with or without testosterone for various time
periods (1h -74 h). TSP-1 mRNA was shown to be expressed at a high level in the cells
growing without testosterone but it was rapidly (in 6 h) downregulated after addition of the
hormone. Similar changes were observed in TSP-1 at the protein level. The experiments with
cycloheximide suggested that testosterone repression of TSP-1 was dependent on de novo
protein synthesis. The time course analysis showed that TSP-1 repression by testosterone was
preceded by induction of FGF-8. Therefore, we studied the possibility that it was FGF-8 that
repressed TSP-1 expression in testosterone-treated S115 cells. We found that addition of
FGF-8b to S115 cultures grown in the absence of testosterone markedly decreased TSP-1
mRNA expression. Furthermore, TSP-1 expression was suppressed in FGF8b-overexpressing
S115 cells and tumors grown without additional testosterone. And TSP-1 over-expression in
S115 cells was able to decrease FGF8b-overexpression induced growth of S115 tumors in
vivo. In conclusion, our experiments suggest that testosterone repression of TSP-1 expression
is mediated by FGF-8 induction and that TSP-1 is a target gene for FGF-8. TSP-1 repression
could thus be one of the mechanisms by which FGF-8 was previously found to stimulate
angiogenesis not only in S115 tumours but also in MCF-7 tumors in nude mice.

                                           - 111 -
Session II: Angiogenesis             Poster II, 7

VEGF and IL-8 Release in Cultured Monkey Choroidal-Retinal Endothelial Cells
Subjected to Hypoxia

Eileen Rojo, Anna Ottlecz, Helena Bühler-Nurmi, and George N. Lambrou; Novartis
Institutes for BioMedical Research, DA Ophthalmology, Basel, Switzerland

        The role of oxygen-sensitive transcription factors (TFs) in activating genes involved in
retinal neovascularization (NV) was investigated. RF/6A rhesus monkey choroid-retinal
endothelial cells were incubated under hypoxic conditions for 8 hours and treated with the
phosphatidylinositol-3 (PI-3) kinase inhibitor LY294002 (Calbiochem) or vehicle. Cells and
culture media were harvested at 0, 24, 48, and 72 hours after hypoxic challenge. Nuclear
extracts were used in 1) electrophoretic mobility shift assays (EMSA) with hypoxia-inducible
factor-1 (HIF-1) and NF-kB elements in the vascular endothelial growth factor (VEGF) and
interleukin-8 (IL-8) promoters and 2) Western Blots for relevant TF detection (HIF-1, NF-kB,
and p53). VEGF and IL-8 release was determined by enzyme-linked immunoabsorbent assay
        Our results indicate that in addition to HIF-1, a larger protein species interacted with
the HIF-1 binding site of the VEGF promoter during 24-72 hours post-hypoxia. LY294002
(10 mM) inhibited the binding of this larger species but left HIF-1 binding unaffected. At later
post-hypoxic time points (24, 48, and 72 hours), VEGF release in LY294002-treated cells was
abolished, while IL-8 secretion was up to sevenfold higher. NF-kB binding on the IL-8
promoter was observed at 0-48 hours post-hypoxia but was abrogated in LY294002-treated
cells. HIF-1/NF-kB expression was observed at 48-72 hours post-hypoxia, while p53
expression was seen up to 48 hours post-hypoxia.
        Our results suggest that HIF-1 activation and DNA binding may exert a stronger
inducing effect on VEGF expression than NF-kB-mediated activation of IL-8 release in the
ocular NV phenotype. It needs to be clarified whether the IL-8 stimulation is related to PI-3
kinase inhibition or if it is an independent effect of LY294002. VEGF activation in ocular
endothelial cells may be due to the HIF-1-mediated recruitment of a larger protein species to
the VEGF promoter.

                                             - 112 -
Session II: Angiogenesis            Poster II, 8

A TcPTP Interacting Protein (TcPTPIP51) is expressed in endothelial
cells of tumor induced vasculature.

Albrecht Stenzinger, Claudia Tag, Monika Wimmer

Institute of Anatomy and Cell Biology, Justus-Liebig-University, 35835Giessen, Germany. E-

TcPTP interacting protein (TcPTPIP51) had been identified as a partner of TcPTP by a yeast
two hybrid system. TcPTPIP51 is an evolutionary conserved protein occurring in two forms
of different molecular weight (32 kD and 45 kD), with a trend to dimersisation. Our
immunoblotting experiments suggest both a tissue-dependent binding of the two subunits as a
hetero- or homodimer and their presence as single forms. By immunofluorescence studies we
have detected TcPTPIP51 in defined compartments of various tissues that undergo a complex
process of differentiation such as epidermis, seminiferous epithelium and embryonic tissue.
Furthermore, we have shown colocalization of this protein and keratins, especially in
Tumor-induced angiogenesis consists of multiple, sequential differentiation steps leading to
maturation and remodelling of newly formed microvessels. Therefore, we started
investigation of different human tumor specimens (adrenal cortical carcinoma, Wilm’s tumor,
duodenal carcinoma, bone sarcoma, carotid body tumor) in order to describe the localization
of TcPTPIP51 in tumor vessels. So far we have detected the protein by immunofluorescence
predominantly in endothelial cells of tumor vasculature whereas tumor cells were unreactive.
In controls, i.e. normal corresponding tissues, TcPTPIP51 in endothelial cells has only been
found scarcely whereas the parenchyma remained unstained.
Probably TcPTPIP51 is involved in a process of endothelial differentiation under pathological
conditions, such as in tumor-induced angiogenesis. Our preliminary studies have shown the
presence of a phosphorylated form of this protein in normal tissue, assuming a role in
intracellular signalling pathways.
Further experiments are projected in order to explain the specific role of TcPTPIP51 in the
process of tumor-derived angiogenesis.

                                           - 113 -
Session II: Angiogenesis             Poster II, 9

Disruption of gap-junctions by VEGF only attenuates the intercellular communication
in coronary capillary endothelium

Dominique Thuringer

N2C, INSERM EMI0211, 2 Bd Tonnellé, 37032 Tours Cedex, France.

Changes in cytosolic Ca2+ concentration are a common theme in initiating various cellular
processes, such as microvascular permeability and angiogenesis. Currently, there is little
known concerning if and how coordination of Ca2+ signaling occurs to aid in capillary
endothelial cell (CEC) responses. Previous reports support the major involvement for gap
junction channels in Ca2+ signaling propagation. However, the mechanism of cell-to-cell
communication may not be straightforward, especially if we consider the participation of
active molecules released by stimulated CEC. In this study, spatio-temporal effects of
vascular endothelial growth factor (VEGF-165) were firstly compared to those of bradykinin
(BK) on gap junction coupling (GJC) and remodeling of connexin (Cx43), then secondary
analyzed on intercellular Ca2+ signal in primary cultures of coronary CEC. Dye-coupling
experiments revealed that BK or VEGF completely blocked GJC. These early effects
correlated with the internalization of Cx43 and its tyrosine phosphorylation at least in part via
the PI3K/Akt pathway. GJC slowly recovered with BK but not with VEGF in the following
hour. Focal mechanical stimulation of a single cell trigged a cytosolic Ca2+ wave that
propagated to a few neighboring cells. Cell treatment with VEGF did not prevent propagation
of waves, although both amplitude and number of cells communicating were reduced.
Blocking gap junctions with heptanol severely restricted propagation which was totally
suppressed by the supplementary addition of apyrase, an ATP/ADPase that acts as a
scavenger of extracellular ATP. Thus, VEGF-induced disruption of GJC via Cx43 remodeling
is relayed by an autocrine communication via ATP secretion to preserve communication
between in CEC and probably periendothelial cells expressing purinergic receptors. How cell-
to-cell communication controls endothelium functions remains unknown, but aberrations in
Cx expression are involved in ischemia injury and tumor angiogenesis.

                                             - 114 -
Session II: Angiogenesis                   Poster II, 10

RNA as Cytokine [Ribokine] in Angio-Morphogenesis and Vascularization of Tissue: Redox and
Metalloregulated Extracellular eRNA as Bioswitch and Therapeutic Target from Inflammation to Wound
Healing or Tumor Vascularization.

Josef H. Wissler and Enno Logemann.

ARCONS Applied Research Institute, Postfach 1327, D-61231 Bad Nauheim and
University of D-79111 Freiburg, Germany. E-Mail:

Inflammation and healing are balanced superimposed and countercurrent processes aimed at regeneration and
repair of injured tissue. Inflammation aims at destruction and removal of fo-reign and self debris. Healing is
bioconstruction of new development fields and morpho-genesis of new organoid patterns in recapitulation of
processes of development. We have in-vestigated biomolecular switch reactions from oxidative stress in
inflammation to tissue mor-phogenesis in healing which may direct the balance in favour of turnover of
inflammatory to healing processes. Reactive oxygen species [ROS] are prominent components in inflammato-ry
reactions. In general, ROS are considered only constraint to destructive processes. Since misleading, this
viewpoint deserves reconsideration:

A major issue of healing processes is angio-morphogenesis which was investigated by angio-tropins [AT]. These
are leukocyte-/ macrophage-derived non-mitogenic angiomorphogens for endothelial cell vascularization active
in vitro and in vivo [Wissler et al., Materialwiss. Werk-stofftech. (Mat. Sci. Eng. Technol.) 32:984-1008, 2001;
Ann. N.Y. Acad. Sci. 991,333-338, 2003 & 961:292-297, 2002; Biol. Chem. 381:S234 & S246, 2000; FASEB J.
12:A1463, 1998; Mol. Biol. Cell Suppl. 8:231a, 1997; Protides Biol. Fluids 34:517-536, 1986]. We found that
AT are copper-ribonucleoprotein cytokines [CuRNP ribokines]. They consist of metalloregu-lated AT-related
protein [ARP] and small extracellular eRNA bioaptamers [ARNA]. ARP and ARNA are complexed together by
copper ions. The sequence of ARP is homologous to S100- A12-EF-hand proteins [calgranulins, hippocampal
neurite differentiation factor] which bind to receptors for advanced glycosylation end products [RAGE]. ARNA
are OH* radical redox re-action-sensitive, highly modified and edited 5'-end-phosphorylated eRNA which could
be de-fined by nucleotide sequence. Isoguanosines as modified nucleotides are formed in ARNA precursor RNA
[9-10 kb] by OH*radical redox reactions on adenosines. It is considered a key reaction to switch on the angio-
morphogenesis process. Thus, radicals are requisite to mor-phogen synthesis and action. It suggests ROS and
transition metal ions in constructive func-tions of a biomolecular switch directed to eRNA in flipping on the
angiomorphogen-promoted wound healing. A physiologic role of endothelial cell membrane RAGE is indicated
in angio-morphogenesis by transduction of signals of eRNA bioaptamers complexed by copper ions to calcium
ion-regulated ARP. A new method has been used to visualize this process by use of structural data of
macromolecules based on NMR or crystallography, called 3D-rapid prototy-ping of accurate molecular image
models [Materialwiss. Werkstofftech. (Laub et al., Mat. Sci. Eng. Technol.) 32:926-930, 2001; Wissler, Ann.
N.Y. Acad. Sci. 991,333-338, 2003]. The re-sults suggest that eRNA is presented to endothelial cells as
reformed ssRNA via extra- and in-tracellular RNA-binding RAGE domains. These are considered to function as
lid of RNA cha-perone-shaped ARP hexamer assemblies. Steps in this process are: Non-mitogenic activation of
cells upon Ca/Cu ion-mediated multivalent binding reactions of ARP to RAGE, Ca ion-de-pendent ARP ligand
and RAGE receptor oligomerization to RNA chaperone-shaped ARP he-xamer assemblies at oligomerized
RAGE membrane sites, concomitant formation of cellular asymmetry and polarity centers. By its intrinsic
metallo-chelating and self-assembling iono-phore molecular properties, isoguanosine is considered essential in
metal ion-dependent ss-RNA-reforming of ARNA. Several novel therapeutic targets with emphasis to redox- and
me-talloregulated eRNA as cytokine [ribokine] can now be suggested in bioswitsches of signal transduction
pathways from inflammation to wound healing or tumor vascularization.

                                                    - 115 -
Session III: Apoptosis

                    - 116 -
Session III: Apoptosis              Poster III, 1

Changes in mammalian chromatin structure as a function of protein-poly(ADP-
ribosyl)ation. Susceptibility of interphase chromatin to enzymatic digestion with
Deoxyribonuclease I (DNase I) and Microccocal Nuclease (MNase)

Maria A. Pérez-Lamigueiro and Rafael Álvarez-González *

Department of Molecular Biology and Immunology
 University of North Texas Health Science Center
3500 Camp Bowie Blvd. Fort Worth, Texas 76107

In this research project, we evaluated the contribution of protein-poly(ADP-ribose)
metabolism to the susceptibility of interphase chromatin (highly condensed) to step-wise
relaxation, as determined by DNase I and MNase endonuclease digestion. The susceptibility
to treatment with either endonuclease was followed as function of the time of incubation as
well as the enzyme concentration. To validate our methodology, we performed the same
experiments with calf thymus naked DNA as negative chromatin controls. Experiments were
carried out in the absence or presence of exogenous _NAD+, the substrate for poly(ADP-
ribosyl)ation, to examine the role of this pathway on the structural state of chromatin.
Enzymatic utilization of _NAD+ as a substrate for chromatin was stopped with benzamide, a
competitive inhibitor of protein-poly(ADP-ribosyl)ation. Endonuclease enzymatic reactions
were stopped with electrophoresis loading buffer containing EDTA. Samples were run in low
concentration agarose gels and the oligonucleosomal electrophoretic migration patterns were
obtained by staining of the DNA with Ethidium Bromide. Our initial results with MNase
show a faster and increased degradation of chromatin upon protein-poly(ADP-ribosyl)ation.
Thus, our results support the hypothesis that, the covalent poly(ADP-ribosyl)ation of
chromatin proteins, particularly histones, favors a more relaxed or open structure, rendering
chromatin more sensitive to endonuclease digestion. Overall, our data are consistent with the
conclusion that, the poly(ADP-ribosyl)ation of histones and non-histone proteins modulates
chromatin condensation-relaxation cycles by both covalent and non-covalent interactions.

Key words:
Chromatin structure, poly(ADP-ribosyl)ation, endonucleases, PARP, and _NAD+ .

                                           - 117 -
Session III: Apoptosis               Poster III, 2

Does direct mitochondrial targeting of APP influence apoptotic regulators in
Alzheimer´s disease?

Astrid Bonert, Celio Marques, Uta Keil, Walter E. Müller, Anne Eckert
Department of Pharmacology, Biocentre, University of Frankfurt, Germany

Still today it is unknown wether intracellular or extracellular A beta is responsible for the
oxidative stress-induced neurotoxic effects on mitochondria. In addition, it might be possible
that an altered distribution of APP in subcellular compartments plays a crucial role for
enhanced oxidative stress in the AD brain. We investigated the effects of the Swedish double
mutation in the beta amyloid precursor protein (APPsw) on oxidative stress-induced cell death
mechanisms in PC12 and HEK cells. At first, it was of interest to determine APP expression,
A beta production and secretion into the extracellular compartment under basal conditions.
APPwild type and APPsw PC12 cells express the same amount of APP, but A beta production
and secretion is 3-5 fold increased in APPsw PC12 cells due to the mutation. In contrast to
this, APPsw HEK cells express more APP than HEK APPwt cells and A beta production is 30
fold enhanced compared to APPsw PC12 cells. Characterization of A beta production by
Western Blot in PC12 cells was not possible probably due to the insufficient sensitivity of the
antibody, since in APPsw HEK cells, A beta was detectable. Then, we investigated
mitochondrial factors released during apoptosis. Here, we used APPsw PC12 cells as model,
because Aß and APP levels mimick physiological conditions that are present in AD brain.
After treatment with hydrogen peroxide, a time-dependent release of cytochrome c was
observed, reaching a maximum after 6 hours, with no differences between the cell types. After
four hours of treatment with hydrogen peroxide, there was an enhanced release of Smac
observed in APPsw cells. AIF was not released by mitochondria before cytochrome c and
Smac, but after 24 hours of hydrogen peroxide exposure. Furthermore, we detected an
increased amount of AIF in mitochondria of APPsw PC12 cells after 6 hours of exposure. Our
findings support the hypothesis that there might be a direct effect of APP or Aß on
mitochondria, which enhances the vulnerability of the cells for oxidative stress.

                                            - 118 -
Session III: Apoptosis             Poster III, 3

Targeted Vpr-derived Peptides Reach Mitochondria to Induce Apoptosis of aVb3-
Expressing Endothelial Cells
                         1             1              1                 1                1
Annie Borgne-Sanchez , Sylvie Dupont , Ludwig Baux , Alain Langonné , Hervé Lecoeur ,
               1              1                   2                       3           4
David Chauvier , Olivier Déas , Aurélien Deniaud , Jean-Jacques Brières , Pascal Roux ,
                 5                 6                   2               3              1
Christine Longin , Jean-Paul Briand , Catherine Brenner , Pierre Rustin , Léna Edelman ,
                     1                   1
Dominique Rebouillat and Etienne Jacotot
  Theraptosis Research Laboratory, THERAPTOSIS S.A., Institut Pasteur/Biotop, 75015
Paris, France
  CNRS FRE 2445 Université de Versailles St-Quentin, 78035 Versailles, France
  Unité de Recherches sur les Handicaps Génétiques de l’Enfant, INSERM U393, Hôpital
Necker - Enfants Malades, 75015 Paris, France
  Centre d’Imagerie Dynamique – Institut Pasteur, 75015 Paris, France
  INRA de Jouy-en-Josas, 78352, Jouy-en-Josas, France
  Immunologie et Chimie Thérapeutiques CNRS UPR9021, 76084 Strasbourg, France

Viral protein R (Vpr), an apoptogenic accessory protein encoded by HIV-1, induces
mitochondrial membrane permeabilisation (MMP) via a specific interaction with the
permeability transition pore complex. We have designed and synthetised peptides composed
of two functional domains, one a tumor blood vessel RGD-like ‘homing’ motif and the other a
MMP-inducing sequence covering the minimal mitochondriotoxic domain of Vpr (Vpr67-82).
The cyclic RGD domain was designed to specifically recognize aVb3 integrins expressed at
the surface of the endothelial cells and allow peptide endocytosis. Once internalized, the
chimeric peptide transitory localise to lysosomes and progressively co-distribute with the
mitochondrial compartment. Human primary endothelial cells treated by micromolar doses of
targeted Vpr(67-82) peptide undergo a dissipation of the mitochondrial transmembrane
potential (DYm), the mitochondrial release of cytochrome c, as well as phosphatidyl serine
exposure and nuclear chromatin condensation. Hence, our prototypes specifically recognize
aVb3 integrins expressing cells, internalise and reach mitochondria to selectively induce
MMP and apoptosis of angiogenic endothelial cells.

                                           - 119 -
Session III: Apoptosis               Poster III, 4

Thiol-mediated redox regulation of mitochondrial permeability conditions by
modulation of thioredoxin reductase activity

M. Bragadina, G. Scutarib, A. Bindolic, M.P. Rigobellob
Dipartimento di Scienze Ambientali, Università di Venezia, DD2137 30123 Venezia, Italy.
 Dipartimento di Chimica Biologica, Università di Padova, viale G. Colombo 3, 35121
Padova, Italy. cIstituto di Neuroscienze (CNR), Sezione di Biomembrane, c/o Università di
Padova, viale G. Colombo 3, 35121 Padova, Italy

A large part of the mitochondrial processes including respiratory chain functions, membrane
potential, permeability transition and apoptosis are modulated by the redox state of protein
and non-protein thiols. In fact, in mitochondria, together with the well-known system
glutathione/glutathione reductase/glutathione peroxidase, another system is also present and is
based on thioredoxin/ thioredoxin reductase and thioredoxin peroxidase. Both systems depend
on NADPH as electron donor. Thioredoxin reductase has been purified and characterized in
our laboratory and, it was shown to create a connection between pyridine nucleotides and
membrane protein thiols. It is also apparent that both systems cooperate to the removal of
hydrogen peroxide that is largely produced by mitochondria. In addition to the isolation,
purification and characterization of thioredoxin reductase several inhibitors or substrate of the
enzyme were tested. Most of the inhibitors appear to act at the level of the selenium moiety
present at the active site. Besides, several chemicals and drugs act also as acceptors of
electrons, therefore creating a diversion of electrons from thioredoxin that is the natural
substrate. Among the various inhibitors tested we found that some metal complexes and
particularly gold complexes are extremely effective and constitute a class of inhibitors more
efficient than other compounds such as cisplatin, cadmium ions and tributyltin. Another
effective inhibitor of thioredoxin reductase is Zn pyrithione that is also able to inhibit
mitochondrial respiration and ATP synthase activity. The inhibition of thioredoxin reductase
determines several consequences for the mitochondrial physiology and the most relevant are
the induction of permeability transition and the decrease of membrane potential. A small,
although significant decrease of total thiol groups was also observed together with the
production of hydrogen peroxide. A major consequence deriving from the inhibition of
thioredoxin reductase is the increase of permeability of both mitochondrial membranes that
determine the release of apoptogenic factors such as cytochrome c that ultimately lead to cell
In conclusion, in mitochondria a strict correlation appears to involve thioredoxin reductase,
membrane permeability and apoptosis process. The latter feature makes mitochondria
attractive targets for drugs potentially acting as antitumor agents.

                                             - 120 -
Session III: Apoptosis               Poster III, 5

Docosahexaenoic acid modulation of apoptotic pathways enhances 5-fluorouracil
antineoplastic effects in human colorectal cancer cell lines.
    Gabriella Calviello, 1Fiorella Di Nicuolo, 1Simona Serini, 1Elisabetta Piccioni, Alma
    Boninsegna 2Nicola Maggiano, 3Franco O. Ranelletti and 1Paola Palozza.
 Institute of General Pathology, 2Pathology, and 3Histology, Catholic University. L.go F. Vito,
1. 00168 – Rome, Italy.

Various experimental studies have shown that n-3 polyunsaturated fatty acids (PUFAs) inhibit
growth of colon cancer cells. Recently it has been shown that n-3 PUFA may sensitize several
kinds of tumors (breast cancers, sarcomas and leukemias) growing in animals to different
anticancer drugs. Similarly, various strains of human tumor cells cultured in vitro (mammary,
glioblastoma, lung or leukemic cells) and treated with n-3 PUFA resulted more sensitive to
different antitumoral agents. In this study we investigated the ability of docosahexaenoic acid
(DHA) to augment the antineoplastic efficacy of 5-fluorouracil (5-FU) against human colon
cancer cell lines. The concentration of 5-FU used (0.1-1.0mM) were much lower than those
currently found in plasma patients after infusion of this drug, and those used for DHA
(≤10mM) in combination with 5-FU were inferior than those generally used in vitro and
known to cause peroxidative effects in vivo. Since it has been shown both in vitro and in vivo
that 5-FU efficacy is often related to the p53 status of human colon cancer cells, we analyzed
the effects both in p53-wild type (LS-174 and Colo 320) and p53-mutant colon cancer cells
(HT-29 and Colo 205). Whereas the cells showed different sensitivity to the growth-inhibiting
action of 5-FU, DHA reduced cell growth independently of p53 cellular status. DHA
increased 5-FU-efficacy against colon cancer cell growth and its effect was related to the
ability to enhance the pro-apoptotic effect of 5-FU. DHA markedly augmented the inhibitory
effect of 5-FU on the expression of the anti-apoptotic proteins BCL-2 and BCL-XL and
increased the expression of c-MYC, known to promote apoptosis and sensitize cancer cells to
the action of various proapoptotic agents. These results indicate that the combined treatment
of colon cancer cells lines with low concentrations of DHA and 5-FU synergistically promote
apoptosis and inhibit cell growth. Moreover, they suggest the possible combined application
of low doses of the two compounds as a chemotherapeutic strategy to reduce the adverse
health effects of 5-FU.

                                            - 121 -
Session III: Apoptosis               Poster III, 6

Control of death-receptor and mitochondrial-dependent apoptosis by c-Jun NH2
–terminal kinase in a model of global ischemia in gerbils

Sonia Carboni, Bruno Antonsson, Pascale Gaillard, Jean-Pierre Gotteland, Jean-Yves Gillon
and Pierre-Alain Vitte

Serono Pharmaceutical Research Institute, 14 Chemin des Aulx, CH-1228 Plan-les-Ouates/
Geneva, Switzerland. E-mail:

c-Jun NH2 –terminal kinase (JNK) is a member of the mitogen-activated protein kinase
(MAPK), that is activated in response to a number of extracellular stimuli, including
The purpose of this study was to examine the role of the JNK signalling pathway on caspase
control during cerebral ischemia using AS601245, a selective JNK inhibitor, known to protect
neurons from ischemic insults. Cerebral ischemia was induced in gerbils by 5 min of bilateral
arteries occlusion. Two days after the insult, a strong caspase-3 activation was found followed
by a second peak of activation 7 days after the occlusion. Cleavage studies of caspase
substrate revealed that the first peak was the result of activation of the death receptor pathway
via caspase-8. The second peak was due to the activation of the mitochondrial pathway
through release of cytochrome c from the mitochondria and caspase-9 activation.
Administration of AS601245 (80 mg/kg, i.p), 15 min and 24 hours after the reperfusion,
significantly prevented neuronal death, inhibited both caspase-3 and caspase-8 activity and
prevented cytochrome c release.
Together these data indicate that JNK is required for both receptor-mediated and
mitochondrial-dependent apoptosis pathways after global ischemia in gerbils. Thus, inhibition
of the JNK signalling pathway could represent an efficient way to prevent apoptotic cell death
following ischemic injuries in the brain.

                                             - 122 -
Session III: Apoptosis              Poster III, 7

The role of the Fas receptor in ethanol-induced apoptosis in HepG2 cells

Francisco Castañeda and Rolf K.H. Kinne

Pathophysiology Laboratory, Department of Epithelial Cell Physiology, Max Planck Institute
of Molecular Physiology, Dortmund, Germany.

Despite numerous studies on the treatment of hepatocellular carcinoma (HCC), there is
controversy about the best therapeutic approach to improve the survival rate of the patients.
HCC is characterized by reduced sensitivity to drugs and often chemoresistance. Thus
eradication of the malignant tumor by injection of absolute ethanol has been used as an
alternative in the treatment of HCC. However, due to the high concentrations used severe side
effects are observed. We previously found that millimolar concentrations of ethanol induce
apoptosis in HepG2 cells but not in normal human hepatocytes. The aim of the present study
was to determine whether the Fas pathway plays a role in ethanol-induced apoptosis in
HepG2 cells. HepG2 cells were incubated with 1 mM ethanol for 24 hours. Apoptosis was
assessed by DNA fragmentation and caspase-8 activity. Caspase-8 activity increased
significantly 3 fold (p<0.005) after 12 hours incubation of HepG2 cells with 1 mM ethanol
whereas no change was observed in control cells. Incubation with a caspase-8 inhibitor
completely prevented apoptosis induction by ethanol (p<0.001). Neutralization of Fas-
receptors by Fas fusion proteins completely attenuated ethanol-induced apoptosis in HepG2
cells treated with ethanol. These findings show that apoptosis induced by low concentrations
of ethanol in human HepG2 cells is associated with Fas-receptor activation and subsequent
caspase-8 activation. Triggering of apoptosis through Fas-receptors represents a mechanism
of action different from that observed with high concentrations of ethanol, thus, reducing the
complications that follows the inflammatory reaction after necrosis.

                                            - 123 -
Session III: Apoptosis             Poster III, 8

Tissue inhibitor of metalloproteinase-1 protects vascular endothelial cells from apoptosis
through PI3-kinase/ Akt activation and eNOS phosphorylation

Gwénola Boulday, Juliette Fitau, Stéphanie Coupel, Jean-Paul Soulillou and Béatrice

INSERM U437 and ITERT, C.H.U. Hôtel-Dieu, 30, bd Jean Monnet, 44093 Nantes cedex 01,
France.Email :

Background - In addition to inhibiting matrix metalloproteinase activity, recent studies
suggest that tissue inhibitor of metalloproteinase (TIMP)-1 may inhibit apoptosis in various
cell lines. In previous studies, we showed that TIMP-1 was strongly upregulated at mRNA
and protein levels in vascular endothelial cells (ECs) upon activation. The present work
examines the possibility that TIMP-1 belongs to a protective pathway via anti-apoptotic
properties and investigates the signaling pathway mediated by TIMP-1 in human ECs.
Methods - Since TIMP-1 is mostly distributed as a secreted factor, experiments were
performed by incubating primary vascular renal ECs with exogenous, recombinant, TIMP-1.
We have used viability assays, nuclear staining and DNA content analysis to ask whether
exogenous TIMP-1 could prevent apoptosis induced by TNFa in cyclohexamide-sensitized
ECs. Effects of TIMP-1 on PI3-kinase and NFkB signaling pathways and phosphorylation of
dowstream targets was examined by western blot analysis. Results – Our data show that the
antiapoptotic effect of TIMP-1 was dose dependent and a maximal effect of TIMP-1 (30%
protection) was reached using 250 ng/ml of the TIMP-1. We present evidence that TIMP-1
induces activation of PI3-Kinase but not NFkB pathway in ECs. Our findings further indicate
that TIMP-1-induced EC survival is mediated through Akt activation and the downstream
phosphorylation of eNOS. Blocking NO with L-NAME restaures TNFa-mediated EC death.
Conclusion - Collectively, our results support the hypothesis that exogenous TIMP-1
efficiently protect ECs from apoptosis by activating PI-3K pathway and identify NO synthesis
and Bad inactivation as effector mechanisms for this protection.

                                           - 124 -
Session III: Apoptosis              Poster III, 9

Alpha-fetoprotein activates apoptosis in tumor cells via induction of the cytochrome c
release and positive regulation of cytochrome c-depended caspase activation.

Elena Dudich, Lidia Semenkova, Igor Dudich, Natasha Tokhtamisheva, *Edward Tatulov.

Department of Molecular & Cell Biology, Institute of Immunological Engineering,
Lyubuchany, Moscow Region, Russia, 142380, *JSC Bio-Sistema, Moscow, Ostozhenka-16,
Russia. E-mail:

Alpha-Fetoprotein (AFP) is an oncoembryonic marker with multiple cell growth regulative,
differentiating and immunosuppressive activities. We have demonstrated that AFP is able to
induce apoptosis in tumor cells through activation of caspase-3 bypassing Fas and TNFR-
dependent signaling. Caspase-3 inhibitor Ac-DEVD-cho significantly inhibited AFP-induced
apoptosis, whereas a general caspase inhibitor z-VAD-fmk had no inhibitory effect, showing
independence of upstream caspase-8. AFP treatment of Raji cells increased Bcl-2 protein
level, showing that AFP-induced apoptosis is not explained by downregulation of Bcl-2 gene.
Incubation of AFP-sensitive tumor cells (HepG2, Raji) with neutralizing anti-Fas, anti-
TNFR1 or anti-TNFR2 antibodies did not prevent AFP-induced apoptosis, demonstrating its
independence on Fas- and TNFR-dependent signaling. It was shown that AFP-treatment of
tumor cells is accompanied by cytosolic translocation of mitochondrial cytochrome c. In a cell
free system, AFP mediated processing and activation of caspases-3 and -9 by synergistic
enhancing of the low dose cytochrome c-mediated signals. AFP was unable to regulate
activity of caspase-3 in cell extracts depleted from cytochrome c or caspase-9. Using a high-
resolution chromatography, we showed that AFP positively regulates cytochrome c/dATP-
mediated apoptosome complex formation, enhances recruitment of caspases and Apaf-1 into
the complex and stimulated release of the active caspases-3 and –9 from the apoptosome. By
using direct protein-protein interaction assay we demonstrated that pure human AFP
practically completely disrupts the association between processed caspases-3 and -9 and
cIAP-2 demonstrating its release from the complex. Our data suggest that AFP might regulate
cell death by displacing of cIAP-2, the inhibitor-of-apoptosis-protein, from the apoptosome
resulting in promoting of caspase-3 activation and its release from the complex.

                                            - 125 -
Session III: Apoptosis               Poster III, 10

Alpha-fetoprotein confers estradiol-mediated mammary carcinoma cell growth in vitro
in synergy with tamoxifen. Direct association of the AFP/E2 complexes with estrogen

Elena Dudich, Olga Goncharova, Lidia Semenkova, Igor Dudich and *Edward Tatoulov.
Institute of Engineering Immunology, Lyubuchany, Moscow Region, Russia, *JSC Bio-
Sistema, Moscow, Ostozhenka-16, Russia. E-mail:

Human a-fetoprotein (AFP), isolated from the cord sera, was shown to interact with estradiol
(E2) that significantly affects its tertiary structure conformation. It has been demonstrated
earlier, that human AFP could induce apoptosis in various tumor cells in a strongly dose-
dependent manner. We studied tumor suppressive effects of AFP/E2 complexes and the role of
estrogen receptors in this effect, using ER-expressing human hepatoma HepG2 and mammary
carcinoma MCF-7 tumor cell lines. To prepare AFP/E2 complexes AFP was incubated with
molar excess of E2 for 1 h at room temperature and then serially diluted complex was added to
the cell culture for 24 h. Both types of cells revealed significant enhance of the total growth-
suppressive response to AFP/E2 in comparison with AFP alone. Our data demonstrated that
the most significant tumor suppression was observed at low concentration of E2 in AFP/E2
complex (molar ratio 1:1), whereas significant molar excess of E2 in AFP/E2 led to the
decrease of the total tumor suppressive effect. At the contrast, simultaneous addition of AFP
and E2 to the MCF-7 cells without pre-incubation showed no effect. However, cell
preincubation with E2 with subsequent treatment with AFP/E2 led to the complete abrogation
of apoptotic effect, whereas E2 did not confer apoptosis induced by AFP alone in the same
cells. Preincubation of MCF-7 cells with tamoxifen to block nuclear estrogen receptors did
not abrogate AFP/E2 or AFP-mediated apoptosis. Moreover, it was observed synergistic
enhance of the total tumor-suppressive effect when cells were treated with suboptimal doses
of AFP/tamoxifen mixture. Physical interaction of AFP and ER-complex was demonstrated
by coimmunoprecipitation. It was shown that pure AFP interacted with ER and Hsp90 by
formation of heterooligomeric complexes. Taken together, our data show that: (i) AFP/E2
complexes use distinct intracellular molecular pathways in apoptosis signaling, in comparison
with AFP alone; (ii) AFP and AFP/E2 are involved in physical interaction with nuclear ER
complex; (iii) AFP induces apoptosis independently on ER; (iv) AFP and tamoxifen operate
synergistically in tumor suppression by engaging of distinct intracellular pathways; (v) E2
excess prevents AFP/E2-mediated tumor suppression via concurrence for the binding with
intracellular ERs.

                                             - 126 -
Session III: Apoptosis               Poster III, 11

Theoretical model for the general structure of the human alpha-fetoprotein: structural
requirements for functional activity.

Igor Dudich, Elena Dudich, Alexander Denesyuk, Lidia Semenkova.
Department of Molecular & Cell Biology, Institute of Immunological Engineering,
Lyubuchany, Moscow Region, Russia, 142380. E-mail:

Theoretical modeling of the general structure of the human a-fetoprotein molecule (AFP) was
performed basing on the experimental studies of its secondary and tertiary structure
parameters by using of circular dichroism, fluorescence spectroscopy, scanning
microcalorimetry, fluorescence spectroscopy and analytical ultracentrifugation techniques.
The secondary structure of the AFP molecule was generally stable and did not depend on the
presence of noncovalently-bound ligands or protein homodimerization. Microcalorimetry
data, which characterize the tertiary structure of the macromolecule, demonstrated that AFP
heat denaturation could be considered as a two-transition process, showing existence of two
thermodynamically independent cooperative units, which evidently correspond to separate
structural domains. These data allow to characterize AFP molecule as a three-domain
molecule, in which two compact rigid C- and N-terminal domains are connected by relatively
labile middle domain that does not induce significant input in the total enthalpy of
denaturation. Our data evidenced that tertiary structure of the AFP molecule undergoes
reversible conformational change induced by ligand removal or protein homodimerization
leading to the significant destabilization of the protein conformation. Various forms of the
AFP and its peptic fragments were studied for their ability to induce apoptosis in cancer cells
and showed significant dependence on conformational state of the molecule. Summarized
structural and functional experimental data indicated that functional active site of AFP
molecule that is responsible for apoptosis induction could be constructed via concentration-
dependent homodimerization occurring via formation of the heterodimeric complex of C- and
N-terminal domains. The hydrophobic fragments have been localized in the subdomains Ib
and IIIb and were predicted to form an indivisible and an extensive hydrophobic sheet, which
may be a good basis for AFP homodimerization. The theoretical model of the dimeric form of
the AFP molecule was proposed according to the structural and functional requirements
obtained in the course of experimental studies.

                                            - 127 -
Session III: Apoptosis               Poster III, 12

Zoledronic Acid inhibits the mitogenic effects of Insulin-like Growth Factor-1 on
prostate cancer cells.

J.C. Dumon, N. Kheddoumi, L. Lagneaux, F. Journé, J.J. Body.
Lab. of Endocrinology and Bone Diseases, Lab. of Hematology, Inst. J. Bordet, Free
University of Brussels, Belgium
The bisphosphonate Zoledronic Acid (Zol) reduces skeletal morbidity from metastatic bone
disease in patients with prostate cancer, presumably through inhibition of osteolysis.
Osteolysis disrupts skeletal integrity and releases IGF-1, thus stimulating prostate cancer cells
growth. We have here investigated the effects of the co-addition of Zol and IGF-1 on human
PC-3 prostate cancer cells. PC-3 cells were cultured in Ham’s-F12/RPMI supplemented with
2mM L-Gln and 10% inactivated FCS that was also depleted of growth factors. Cell growth
was evaluated at days 1, 2, 4 and 6 by crystal violet staining. We selected optimal stimulatory
concentrations of IGF-1 (100 ng/ml) and three concentrations of Zol (10-6, 10-5 and 10-4 M).
IGF-1 increased PC-3 cells growth by 28±5% (mean±SEM) at day 2 and by 23±7% at day 6
(P<0.001). At 10-6 M, Zol did not affect cell growth but it completely abolished the mitogenic
effects of IGF-1. At 10-5 M, Zol inhibited cell proliferation whether IGF-1 was present or not.
At 10-4 M, Zole induced a marked decrease in cell number (at day 6, cell population fell to
15% in comparison with the day 1 value). We previously reported that Zol reduces prostate
cancer survival by exerting cytostatic and apoptotic effects (Eur Urol, in press). We assessed
apoptotic cell death by using annexin V/propidium iodide double staining method based on
apoptosis-related cell membrane modifications. Apoptosis was assessed when cells were co-
treated with 10-4 M Zol and 100 ng/ml IGF-1. Zol induced apoptosis in PC-3 cells and this
effect was enhanced if IGF-1 was simultaneously added, maybe because of an enhanced entry
of cells into the cell cycle. In conclusion, Zol can inhibit the mitogenic effects of IGF-1 on
prostate cancer cells, already at 10-6 M. Moreover, at higher concentrations, induction of
apoptosis by Zol was not affected by IGF-1. This could represent a new mechanism of action
of Zoledronic Acid whereby it exerts its protective effects against prostate cancer-induced
bone disease.

                                             - 128 -
Session III: Apoptosis              Poster III, 13

Increased mitochondrial and nuclear gene expression of cytochrome oxidase subunits I
and IV in neuronal aging

Patrizia Fattoretti, Carlo Bertoni-Freddari and Belinda Giorgetti
Neurobiology of Aging Laboratory, INRCA Research Department, Via Birarelli 8, 60121
Ancona, Italy. E-mail:

To asses the role of mitochondrial metabolic competence in neuronal aging, a quantitative
immunohistochemical study has been carried out on cytochrome oxidase (COX) subunits I
(mitochondrial-encoded) and IV (nuclear-encoded) in the cerebellar cortex of adult and old
rats. Following anaesthesia and perfusion with 4% Sorensen buffered paraformaldehyde, the
cerebellar tissue samples were prepared from rats of 12 and 24 months of age. By applying a
semiautomatic systematic random sampling procedure, the optical density (OD) values of
COX subunits I and IV were measured on an overall area of 75,000 square microns in the
granular and molecular layers of the cerebellar cortex of each animal. In old animals, OD
values of subunit I were increased by 35.5 and 34.2% in the molecular and granular layers,
respectively, but only the difference found in the latter cerebellar zone was statistically
significant (p< 0.05%). As regards subunit IV, old animals showed higher, not significant,
densitometric values in the molecular (+20.6%) and granular (+26.8%) layers. The present
findings sustain that gene expression of COX subunit I and IV appears not to be involved in
the well documented time-related mitochondrial decay. The proper functioning of COX
depends on several factors which can affect the mitochondrial metabolic competence in the
aging cell. In the fully assembled holoenzyme, both the subunits investigated in the present
study span the inner mitochondrial membrane. On the basis of these molecular biology data, it
is reasonable to suppose that any alteration of the physico-chemical features and chemical
composition of the mitochondrial membranes reported to occur in aging (e.g. decreased
membrane fluidity and cardiolipin content, increased cholesterol/phospholipid molar ratio and
free radical damage, etc.) may significantly affect the proper assembling of the enzyme and,
in turn, its activity. Considering the reported significant decline of COX activity with
advancing age, our findings further support that an adequate mitochondrial metabolic
competence, while including proper nuclear and mitochondrial gene expression of subunits of
the respiratory chain, relies on the overall balance among various determinants which can be
differently damaged by aging and represent critical causative events responsible of the age-
related functional decline of selected mitochondrial populations.

                                           - 129 -
Session III: Apoptosis               Poster III, 14

Effects of amyloid beta and metals on reactive oxygen species production and apoptotic
cell death in human lymphocytes

Claudia Frey1, Gunther Rexroth2, Wolfgang Rösch 2, Walter E. Müller1, Anne Eckert1
Department of Pharmacology, Biocentre, University of Frankfurt, Germany.
 medical clinic, Nordwest Hospital, Frankfurt, Germany

One hallmark of the Alzheimer´s disease (AD) are extracellular plaques containing
aggregated amyloid beta 1-42 (Abeta 1-42) which has neurotoxic potential. Furthermore, there
were found increased amounts of iron, copper, and zinc in senile plaques from AD patients.
As many in vitro studies show these elements are generators of reactive oxygen species
(ROS), especially in combination with Abeta. Hereby iron, copper, and zinc seem to induce
and/or accelerate aggregation of Abeta proteins. An increase of ROS levels may lead via
protein oxidation, lipid peroxidation, and DNA oxidation to apoptotic cell death.
In our present study, we treated lymphocytes with different Abeta protein fragments in the
presence or absence of iron, copper, and zinc. We determined ROS levels, transmembrane
potential, and apoptotic cell death in the whole population of these lymphocytes, as well as in
the CD4 T-cell subpopulation. The latter is the most sensitive to induction cell death. We
have found an elevation of ROS levels, transmembrane potential, and apoptotic cell death in
lymphocytes after addition of the long physiological form of Abeta (Abeta 1-42). In contrast,
after treatment of lymphocytes with the synthetic peptide Abeta 25-35, which contains the
cytotoxic functional sequence of the amyloid peptide, we detected an elevation in apoptotic
cell death, but not an increase in ROS levels. Notably, we did not see enhancing effects of
metal ions on Abeta toxicity nor effects of these metals per se on ROS production or cell
death, Moreover, production of nitric oxide was not involved in these processes.

                                            - 130 -
Session III: Apoptosis               Poster III, 15


García-Tuñon I, Ricote M, Ruiz A, Fraile B, Paniagua R and Royuela M.

Departament of Cell Biology and Genetics. University of Alcala. Alcala de Henares. Madrid
E-28871. Spain. E-mail:

Interleukin-6 (IL-6) is a multifunctional cytokine produced by non-malignant cells such as T
lymphocytes, fibroblasts, or monocytes. In addition a variety of tumor types are stimulated by
IL-6, including melanoma, renal cell carcinoma, Kaposi´s sarcoma and ovarian carcinoma. In
breast cancer cell lines, IL-6 inhibits the proliferation but enhances the cells motility,
increased cell-cell separation, and decreased adherens type junctions. Moreover recent studies
suggest IL-6 prevent apoptosis via regulating Bcl-2 family gene products. The aim of this
study was investigate, by means of immunohistochemistry and western blot, the expression of
IL-6, bcl-2 and bax in benign and tumor breast samples.
Biopsies from 13 benign pathologies and 52 carcinomatous woman breasts were processed by
western-blot and immunohistochemistry by ABC method for the detection of IL-6, bcl-2 and
bax. The carcinomatous cases included 35 infiltrative and 17 “in situ” tumors samples.
Non-malignant mammary glands showed a weak immunoreaction to IL-6 in the epithelial
cells cytoplasm. In tumor cases we found an increased expression of this cytokine, but the
highest proportion of positive cases was found in the infiltrating carcinomas (68,6%). Benign
breast lesions shown weak labeling to bcl-2 (46%) in epithelial cells cytoplasm. The 56,2% in
situ tumors was positive to bcl-2 and the immunoreaction was stronger than in bening
conditions. The highest intensity was found in the 92% of infiltrating tumors. In these tumors,
the most of bcl-2 positive samples (79,2%) were also positive to IL-6. The labeling to bax
appears with moderate intensity in the cytoplasm of the epithelial cells. In the 53,8% of
fibrocystic samples showed moderate intensity. This expression was similar in 100% of the in
situ tumors. The most intensity reaction to bax was found in the 82,5% of infiltrating tumors.
In our study the increase of IL-6 signal with the malignancy of the tumor seems to be
correlated with the increase of expression of bcl-2. Since other authors purchase that anti-
apoptotic effects of IL-6 was possibly attribute to up regulation of Bcl-2 family of genes, this
could be a mechanism through the IL-6 modify the equilibrium proliferation/apoptosis in
human breast cancer, and it could be one of the several responsible for infiltrating tumor
ACKNOWLEDGEMENTS: This work was supported by grants from Fondo de
Investigaciones Sanitarias (PI020383) and the University of Alcalá.

                                             - 131 -
Session III: Apoptosis              Poster III, 16

Distinct form of cell death induced by derivatives of sesquiterpens in breast cancer
MCF-7 cells. New experimental techniques in apoptosis research.

Micha_ M. Godlewski1, Magdalena Górka1, El_bieta _usakowska2, Barbara Gajkowska3,
Urszula Wojewódzka3, W _odzimierz Daniewski4, Tomasz Motyl1
  Department of Physiological Sciences, Faculty of Veterinary Medicine, Warsaw
Agricultural University, Nowoursynowska 159, 02-776 Warsaw, Poland, e-mail:; 2 Institute of Physics, Polish Academy of Science, Al. Lotników
32/46, 02-668 Warsaw; 3 Laboratory of Cell Ultrastructure, Medical Research Centre, Polish
Academy of Science, Pawi_skiego 5, 05-106 Warsaw; 4 Institute of Organic Chemistry,
Polish Academy of Science, Kasprzaka 44/52, 01-224 Warsaw; POLAND.

Taxoids are widely used as cytostatic drugs in cancer therapy. Their stabilizing effect on the
microtubule network prevents cells from undergoing mitosis. Apart from that their
mechanism of action remains unknown. The use of taxol is limited due to variety of its
negative side effects. We examined mechanism of action of two newly synthesized
derivatives of sesquiterpene alcohols, AGS 115 and EFDAC. Proapoptotic potential and
mechanisms of action were compared with camptothecin (CPT) – well-known and widely-
used inhibitor of DNA topoisomerase I. The laser scanning cytometry, homeostatic confocal
microscopy, atomic force microscopy and immunogold electron microscopy were employed
in our research.
Our observations revealed that cell death induced by new compounds is different from classic
CPT-induced form of apoptosis, and comprises of three characteristic events:
crater-forming dipping of nucleus,
destruction of actin microfilaments,
autofagolizosome formation.
Furthermore we observed Smac-DIABLO release from mitochondria and caspase-7 activation
but delayed in time and secondary event during EFDAC and AGS 115-induced cell death.
Therefore we assume that apoptotic action of newly-developed derivatives of sesquiterpene
alcohols is based on lizosome-mediated form of cell death.

                                            - 132 -
Session III: Apoptosis               Poster III, 17

Epithelial cell apoptosis in the gut of pig neonates and chicken – new application for
laser scanning cytometry

Micha_ M. Godlewski1, Marzena Biernat2, Jaros_aw Woli_ski2, Romuald Zabielski1, 2, Sylwia
Szyma_czyk3, Jose L. Valverde Piedra3, Tomasz Motyl1
 Department of Physiological Sciences, Warsaw Agricultural University, Nowoursynowska
159, 02-766 Warsaw, 2The Kielanowski Institute of Animal Physiology and Nutrition, PAS,
05-110 Jab_onna, and 3Department of Animal Physiology, Agricultural Academy in Lublin,
Akademicka 12, 20-934 Lublin, Poland

Just after birth, the small intestinal mucosa undergoes rapid and major tissue remodeling. The
aim of the present study was to investigate apoptosis in the small intestinal mucosa of pig
neonates and newly hatched chicken.
Samples were collected from the mid-jejunum in non suckling (0 d), suckling (1 and 7 d) and
weaned (12 w) piglets as well as in chickens at 4, 6 and 8 weeks posthatch. Light,
fluorescence and confocal microscopy and laser scanning cytometry were used.
The apoptotic cells were observed on the entire length of the villi as well as in and between
intestinal crypts as visualized by TUNEL and 7-AAD staining in piglets. Similar findings
were also done in the chicken gut mucosa. In neonatal piglets, no age-related changes in the
distribution of the apoptotic cells were observed. The apoptotic rates in pigs at day 0, 1 and 7,
and week 12, were, 21.8%, 15.8%, 21.1% and 21%, and in chicken at week 4, 6 and 8
posthatch, 17.8%, 17.2% and 15.0%, respectively. Caspase-3 synthesis in the enterocytes was
the highest at birth while that of caspase-8 was high at birth and in 7 d piglets. The
enterocytes showing caspase-8 and –3 fluorescence were packed in groups of several cells. In
chicken, the grouping pattern of enterocytes undergoing apoptosis was also evident in slides
stained with 7-AAD.
In conclusion, in neonatal piglets and newly hatched chicken enterocyte apoptosis was
observed on the entire villi and crypts region suggesting this pattern is related to a remodeling
of the gut mucosa during early stages of development. Laser scanning cytometry seems to be
a good tool for investigation of apoptotic processes in the intestinal mucosa.

                                             - 133 -
Session III: Apoptosis               Poster III, 18

Homeostatic confocal microscopy (HCM) as a new tool for analysis of minute kinetics of
proapoptotic proteins in living cells.

Micha_ M. Godlewski, Magdalena Górka, Monika Lamparska-Przybysz, Tomasz Motyl

Department of Physiological Sciences, Faculty of Veterinary Medicine, Warsaw Agricultural
University, Nowoursynowska 159, 02-776 Warsaw, Poland, e-mail:

Mitochondria are major organelles involved in activation of apoptosis in many physiological
and pathological conditions. Proapoptotic factors create several alterations in structure and
functionality of mitochondrial membrane changing its permeability by formation and opening
of megachannels, consisting of VDAC-1, ANT, BAX and probably other proteins. That leads
to decrease in mitochondrial potential and ATP synthesis, and distortion of calcium
distribution within the cell. Increased level of calcium in cytozol allows amplification of BAX
activation and efflux of cytochrome c, Smac/DIABLO, Omi/HtrA2 and AIF from
mitochondria. Observations conducted with the use of traditional methods (Western blot,
fluorescence and immunoelectron microscopy, and simple confocal imaging) allowed barely
indirect time-sequence analysis based on fixed cell populations treated with the drug for
different times. HCM gives a possibility to analyze a precise minute pattern of changes
occurring in the same cell during whole experimental period. HCM was combined with
Microimage system for semiquantitative analysis of fluorescence IOD. In the present study
2D/time and 4D HCM were applied for analysis of minute pattern of changes occurring at the
mitochondria. Cells were double-transfected with either BAX-GFP or Smac/DIABLO-GFP
and mito-vector-RFP. Conducted experiments suggest that Bax aggregation on mitochondria,
as well as Smac/DIABLO release from mitochondria in tumor cells under the apoptogenic
stimulus, consist of two phases: the first immediately after drug administration, and the major
second one after 15 min. Interestingly, the time of the second phase is similar, regardless of
types of tumor cell lines and apoptogenic stimuli used. It indicates conservatism in dynamics
of megachannel creation in outer mitochondrial layer, and release of mediators of apoptosis
from intermembrane space in tumor cells stimulated to apoptosis by anticancer drugs.

                                            - 134 -
Session III: Apoptosis             Poster III, 20

Wee1 inhibitor abrogates TGFb1-induced G2 arrested cells and induced apoptosis

Osamu Hashimoto, Hironori Koga, Ryuichiro Sakata, Toru Nakamura, Masaharu Sakamoto,
Takuji Torimura, Michio Sata and Takato Ueno*

Second Department of Medicine, Kurume University School of Medicine
*Research Center for Innovative Cancer Therapy, Kurume University,
Kurume City, Fukuoka Pref., Japan. E-mail:

Transforming growth factor b1 (TGFb1)-induced G1 cell cycle arrest was observed when the
proliferation inhibitory effect of Rb protein was compromised. However, the mechanism
underlying TGFb1-induced G2 arrest was poorly characterized. We reported that G2 arrest
was induced by TGFb1 (1ng/ml) in the Rb-negative hepatoma cell line (Hep3B) through the
stabilization of Wee1 and inhibition of Cdc2 activity (Hashimoto et al, Molecular
Carcinogenesis; 2003; 36; 171-182). In the present study, we investigated the possibility of
Wee1 as a targeting molecule for anti-tumor cell proliferation. We incubated TGFb1-induced
G2 arrested cells with the Wee1 inhibitor, PD0166285 (0.25uM), for 4h. The Wee1 inhibitor
dephosphorylated Tyr-15 of Cdc2, and abrogated the G2 arrest in Hep3B. However, these
cells were unable to enter S phase, and apoptosis was subsequently induced. Wee1 inhibitors
may accelerate the unscheduled mitosis and induce apoptosis in G2 arrested cells. It is
suggested that Wee1 may be useful as a targeting molecular agent.

                                           - 135 -
Session III. Apoptosis               Poster III, 21

Translational control in apoptosis – a novel role for IRES.

Martin Holcik

Apoptosis Research Center, Children’s Hospital of Eastern Ontario, 401 Smyth Road, Ottawa,
Ontario, K1H 8L1 Canada

Components of the cellular translation machinery, including the initiation factor eIF4G, are
targets of caspase-mediated cleavage during apoptosis that correlates with the inhibition of
protein synthesis that accompanies apoptosis. Paradoxically, however, protein synthesis is
required for apoptosis to occur in many experimental settings. Previous studies showed that
two proteins that regulate apoptosis by controlling caspase activity, XIAP and Apaf-1, are
translated by a unique, cap-independent mechanism. This cap-independent translation is
mediated by an IRES (Internal Ribosome Entry Site) elements that are found in the 5’ UTR of
XIAP and Apaf-1 and are used preferentially under conditions in which normal cap-
dependent translation is repressed, such as during cell cycle, apoptosis or a broad range of
cellular stresses.
The present study investigated the ability of UV irradiation, as an apoptotic trigger, to
modulate activity of XIAP and Apaf-1 IRES and the mechanisms behind the IRES activation.
We show that UV irradiation leads to the inhibition of proteins synthesis and cell death. In
contrast, the cell survival is greatly enhanced by pre-treatment of cells with protein-synthesis
inhibitor cycloheximide suggesting that protein synthesis is required for apoptosis to occur.
We further show that the IRES-mediated translation of Apaf-1, but not XIAP, is enhanced by
UV irradiation and that this increase in Apaf-1 translation is necessary for apoptosis. In
contrast, down-regulation of Apaf-1 levels by siRNA or by Apaf-1 IRES decoy increases
resistance of cells to UV-induced cell death. Unlike in etoposide-treated cells, the enhanced
Apaf-1 IRES translation is independent of caspase-activation but instead depends on PERK
kinase. These data suggests that progression of UV-induced apoptosis requires IRES-
mediated translation of Apaf-1 that ensures continuous levels of Apaf-1 despite the overall
suppression of protein synthesis. Furthermore, our data suggest that IRES-mediated
translation of genes critically involved in the regulation of apoptosis could be specifically
targeted as a therapeutic modality.

                                             - 136 -
Session III: Apoptosis              Poster III, 22

Induction of apoptosis by essential oils from Zanthoxylum schinifolium pericarpium
through reactive oxygen species and p53 in HepG2 human hepatoma cells

Sung-Mok Beak1,a, Kefa O. Bosire2,a , Kyung-Hee Koh3,b, Soon-Young Paik4,c, Jung-Ae Kim5,a
  College of Pharmacy, Yeungnam Univerity, Kyongsan 712-749, Korea, E-mail :
2, bDepartment of Food Science and nutrition and
  College of Medicine, The Catholic University of Korea, Seoul, 137-701, Korea, E-mail :

The essential oils derived from dried pericarp of aromatic plant Zanthoxylum schinifolium,
has been used as a flavoring condiment for centuries and as a medicinal herb for epigastric
pain accompanied by vomiting, diarrhea and abdominal pain due to intestinal parasitosis,
ascariasis. Since the anticarcinogenic properties of some essential oil have been well known
in onions, garlic and oranges, we have tested the anticarcinogenic effects and signaling
molecules of volatile components derived from the Zanthoxylum schinifolium pericarpium
using HepG2 cells in vitro.
The essential oils were extracted from the pericarpium by simultaneous distillation with
dichloromethane and water and analyzed by gas chromatography. The essential oils induced
apoptotic cell death in HepG2 human hepatoma cells in a time-related manner assessed by
annexin-V binding and flow cytometry by determining hypodiploid DNA content. The drug
effect on HepG2 cells was more potent than that in Chang liver cells. In addition, the oils
increased the production of reactive oxygen species and p53 expression in a dose-dependent
manner in HepG2 cells. Pretreatment of the cells with Trolox, a well-known antioxidant,
significantly suppressed the essential oil-induced generation of reactive oxygen species, p53
expression and cell death. However, caspase-3 activity was not changed in the oil-treated
cells, suggesting no involvement of caspase-3 in the oil-induced HepG2 cell death.
Furthermore, in nude mice inoculated with Huh-7 human hepatoma cells, the oils significantly
inhibited tumor development. These results suggest that volatile components of Zanthoxylum
schinifolium pericarpium is a good drug candidate for cancer therapy and that reactive oxygen
species are the key signaling molecules in the oil-induced cell death in HepG2 cells.

                                           - 137 -
Session III: Apoptosis               Poster III, 23

Insulin-like Growth Factor-I Promotes Interleukin-8 Production and Survival of
Human Neutrophils.

Ron Kooijman, Astrid Coppens, Eddie Himpe and Elisabeth Hooghe-Peters

Laboratory for Neuroendocrine Immunology, Department of Pharmacology, Medical School,
Free University of Brussels (VUB), B-1090 Brussels, Belgium

To address possible therapeutic strategies in immune disorders by targeting the IGF-system,
we studied the in vitro effects of IGF-I on the production of inflammatory cytokines and on
neutrophil apoptosis. We found that IGF-I stimulates the secretion of IL-8 in resting and LPS-
stimulated human peripheral blood mononuclear cells. IL-8 serves as a major chemoattractant
for neutrophils, and IL-8 regulation by IGF-I may influence the inflammatory process. In the
present study, we show that IGF-I stimulates IL-8 mRNA expression and IL-8 secretion in the
promyelocytic leukemic cell line HL-60 through binding to the IGF-I receptor. The
expression of IL-8 mRNA was up-regulated at the level of transcription through stimulation
of the MEK-ERK pathway. In addition, inhibition experiments using SP600125 as a JNK
inhibitor suggested that basal activity of JNK is also required for the effect of IGF-I. In
contrast, inhibitors of p38 MAPK and PI3K did not abrogate the effect of IGF-I. Collectively,
our results suggest that basal JNK activity and activation of the MEK-ERK pathway are
required for up-regulation of IL-8 by IGF-I in HL-60 cells.
We have also investigated the effects of IGF-I on neutrophil apoptosis. Neutrophils are short-
lived cells which are programmed to undergo spontaneous apoptosis. This process can
however be inhibited by cytokines and growth factors, or can be accelerated by stimulation
through FAS/APO-1 (CD95). Regulation of apoptosis is crucial for the resolution of
inflammation. We show that IGF-I inhibits both spontaneous apoptosis and anti-FAS-induced
apoptosis in human neutrophils through stimulation of the IGF-I receptor.
We believe that the relation between IGF-I and the immune system is not only of academic
interest. More insight in this relation may contribute to the design of clinical strategies for
modulation of immune responses through components of the IGF system, e.g. IGF-I receptor
antagonists, IGF-binding proteins or MAPKs activated by IGF-I.

                                            - 138 -
Session III: Apoptosis              Poster III, 24

The effects of 8-chloroadenosine-3¢, 5¢-monophosphate and tiazofurin on melanoma cell

Lela B. Kori_anac1, Danijela V. Todorovi_1, Miroslav A. Demajo1, Sabera D. Ru_diji_2,
Aleksandra M. Risti_-Fira1
  Laboratory for Molecular Biology and Endocrinology, Vin_a Institute of Nuclear Sciences,
POBox522, 11001Belgrade, Serbia and Montenegro
  Institute for Biological Research, 29.Novembra142,11000Belgrade, Serbia and Montenegro

Novel antineoplastic agents, 8-chloroadenosine 3¢, 5¢-monophosphate (8-Cl-cAMP) and
tiazofurin (TR), have been shown to be effective against different malignant cells. Through
specific mechanisms of action they modulate the cellular signal transduction pathway, thereby
causing growth inhibition, cell differentiation and apoptosis. The aim of this work was the in
vitro study of either 8-Cl-cAMP or TR effects on B16/F10 and B16/C3 mouse melanoma cell
growth, cell cycle and cell death. Significant cell growth inhibition was obtained after
application of 8-Cl-cAMP or TR. The analysis of genomic DNA isolated from B16/F10 and
B16/C3 samples after application of 8-Cl-cAMP or TR (6h, 24h) has shown the “smear”
pattern on agarose gel electrophoresis. The number of apoptotic nuclei, evaluated by flow
cytometry, after treatment with antineoplastic agents did not significantly change in B16/F10
cells, but they significantly increased in B16/C3 cells. The tested agents induced an increase
of the cell number in G2/M phase (6h after treatment) or in G0/G1 phase (24h after treatment)
in B16/F10 samples. The changes observed in cell cycle distribution in B16/C3 samples after
treatment did not show any regularity. Expression of c-myc did not significantly increase in
B16/F10 cells after treatment with 8-Cl-cAMP or TR and in B16/C3 cells after treatment with
8-Cl-cAMP. C-myc expression significantly increased in B16/C3 cells after treatment with
TR. Concerning the effects the analyzed agents exhibit on melanoma cells and other cancer
cells, further pre-clinical studies of these drugs will potentially lead to more efficient
therapeutic approaches of malignant diseases.

                                            - 139 -
Session III : Apoptosis             Poster III, 25

Platelet-activating factor receptor signals in rat osteoblasts during spaceflight

Yasuhiro Kumei, Sadao Morita, Hiroshi Nakamura, Hideo Akiyama, Masahiko Hirano
Hitoyata Shimokawa, and Kei-ichi Ohya

Graduate School of Tokyo Medical and Dental University, Tokyo 113-8549, Japan.
Toray Research Center, Kamakura 248-0036, Japan

The platelet-activating factor (PAF) is a lipid mediator. The G-protein-coupled receptor of
PAF (PAF-R) is activated by inflammation and pathological/stressful conditions in numerous
cell types. Recent reports show that PAF/PAF-R is involved in apoptotic or anti-apoptotic
process. The PAF-R signals are transmitted through systems including arachidonic acid,
phospholipase C, protein kinase C (PKC), and mitogen-activated protein kinase. PAF-R gene
expression was increased by mechanical stretch in smooth muscle cells, and counteracted by
PKC inhibitors. We examined the effects of microgravity on PAF-R gene expression and
PKC family in rat osteoblasts that were cultured aboard Space Shuttle for 4 days. PAF-R gene
expression was increased 3.5-fold by microgravity, as compared to the ground control.
Microgravity also increased gene expression of the conventional PKCa and the novel PKCd,
PKCe, and PKCJ to 3- to 5-fold of the ground controls. However, PKA was not enhanced by
microgravity. PKCd, PKCe, and PKCJ are target substrates of caspases. PAF-R gene
expression is mediated by PKC and the transcription factor NF-kB to protect cells from
apoptosis, coincidently with IAP (inhibitor of apoptosis protein). PAF-R gene may act as a
mechano-sensitive gene that is involved in apoptotic or anti-apoptotic process during
spaceflight. Data suggest molecular features underlying osteopenia that is induced by gravity

                                           - 140 -
Session III: Apoptosis               Poster III, 26

TIMP-1 signalling pathway leading to hematopoietic cell survival

Elise Lambert, Emilie Dassé, Marie-Line Sowa, Laurent Martiny and Emmanuelle Petitfrère.

Laboratoire de Biochimie, CNRS FRE 2534, IFR53, UFR Sciences Exactes et Naturelles, BP
1039, Université de Reims Champagne-Ardenne, F 51687 Reims Cedex 2, France. E-mail :

The Tissue Inhibitors of Metalloproteinases (TIMPs), specific inhibitors of matrix
metalloproteinases (MMPs) are a central point in the control of extracellular remodeling. Four
TIMPs have been characterized in a variety of species and designated as TIMP-1, -2, -3 and -
4. TIMPs exhibit different properties, some of which seem to be independent of MMP
inhibition, and are so considered as multifunctionnal proteins. First described for its erythroid
potentiating activity, TIMP-1 was since demonstrated to be a growth factor for non-erythroid
cells such as fibroblasts and keratinocytes. In a previous study, we showed TIMP-1 induces
survival of two hematopoietic cell lines. Using specific kinases inhibitors, we showed that
TIMP-1-mediated cell survival is dependent on JAK2 and PI 3-kinase activation. By transient
transfection with dominant-negative Akt, we demonstrated that this kinase is crucial for the
TIMP-1 anti-apoptotic effect. Moreover, TIMP-1 controls the level of the anti-apoptotic
protein Bcl-XL in a PI 3-kinase dependent manner [Lambert et al., 2003].
Structure-function studies have distinguished the MMP-inhibitory activity from the growth-
promoting effect of TIMP-1. We show here that TIMP-1 effect is totally abolished by anti-
MMP-9 antibodies suggesting that MMP-9 could be involved in TIMP-1 signalling. By
different techniques, we evidence the presence of the pro-MMP-9 at UT-7 cell surface and we
show that TIMP-1 associates with pro-MMP-9 localized at the cell surface. Other proteins
immunoprecipitate with pro-MMP-9 and TIMP-1. Their identification as well as molecular
mechanisms underlying TIMP-1 binding will require further investigations.
We conclude that TIMP-1 induces hematopoietic cell survival through JAK-2/PI 3-
kinase/Akt/Bad pathway and that a pro-MMP-9 localized at UT-7 cell surface is involved in
TIMP-1 signalling.

                                             - 141 -
Session III: Apoptosis              Poster III, 27

Transduction pathway of apoptosis induced by the chemokine SDF-1/CXCL12 in
human neuroblastoma SK-N-SH

Alain Lombet*, Audrey Segret* and France Haour‡

*CNRS UMR 8078, Hôpital Marie Lannelongue, 92350 Le Plessis-Robinson, France.
  INSERM EMI 03-50, Hôpital St-Antoine, 75012 Paris, France. E-mail: fhaour@st-

The chemokine stromal cell-derived factor 1 (SDF-1/CXCL12) is the unique biological ligand
for CXCR4, a G Protein Coupled Receptor (GPCR) used as coreceptor for the entry of T-
tropic strain of HIV-1 into CD4+ cells. SDF-1 receptors were demonstrated and characterized
in the human neuroblastoma cell line SK-N-SH. Binding studies with [125I]-SDF-1 on living
cells indicated an affinity constant of 3 nM. Various cytokines (Il-1, TNF-alpha) and
chemokines (RANTES, MCP-1, IL-8) did not compete with the binding of [125I]-SDF-1.
Conversely, the gp120 protein involved in the entry of the HIV-1 virus into target cells,
competed efficiently for the binding of the ligand. SDF-1 (10nM) induced a rapid and
transient increase in intracellular free calcium, detectable in Fura-2-AM treated cells. This
stimulation indicated a functional coupling between the CXCR4 receptor and intracellular
calcium mobilisation. Continuous SDF-1 (10nM) treatment induced an apoptotic process after
48 hrs and a massive cell death at 72 hrs. A short SDF-1 treatment (2 hrs), followed by
washing, did not lead to cell apoptosis after 3 days. At lower concentration, (2 nM SDF-1 for
72 hours) cell death was not total and was reversed by Na+/orthovanadate, an inhibitor of
protein tyrosine phosphatase. A MEK-1 inhibitor (PD 98059) as well as inhibitors of other
types of phosphatases, were unable to antagonize the apoptotic process induced by SDF-1.
Activation of tyrosine phosphatase seems therefore to be implicated in the long-term effect of
SDF-1 on neuroblastoma cells apoptosis. The toxicity that appears after a long-term
exposition to SDF-1 might be due to the generation of truncated ligand induced by SK-N-SH
metabolism. These results demonstrate the potential role of SDF-1 in human neuronal cell
death and open up new ways of research on the CXCR4 receptor as regulator of brain
function in health and disease.

                                            - 142 -
Session III: Apoptosis               Poster III, 28

A natural defence molecule in cancer control. Tracking betaGBP pathways to apoptosis.

Livio Mallucci and Valerie Wells

Cell Signalling and Growth Laboratory, Department of Life Sciences,
King’s College London, UK. E-mail:

Much attention has been placed for more than a decade on identifying molecular alterations in
cancer cells which may serve as probable therapeutic targets for specially designed drugs.
Expectedly, the success of these therapeutic approaches is often impaired by the drugs’
inability to discriminate between normal and tumour cells, by toxicity, resistance, and an
implicit inability, because of their design, to overcome the broad spectrum of genetic changes
which protect cancer cells from apoptotic death. Recent studies have shown that betaGBP, an
antiproliferative cytokine produced by activated T cells (Blaser et al (1998) Eur J Immunol
28,2311-2319), can selectively enforce programmed cell death in cancer cells (Novelli et al
(1999) J Cell Physiol 178,102-108; Wells et al (1999) Eur J Cancer 35,978-983; Mallucci et
al Biochem Pharmacol 66,1563-1569), via downregulation of signalling and by forcing
persistant ectopic E2F1 transactivation, by suppressing akt gene expression and by impairing
cytoskeletal function. Studies on different panels of cancer cells demonstrate that either one or
all of these routes can be used by betaGBP to selectively induce apoptotic death in cancer
cells. Such a natural ability by a physiological molecule provides a rationale to assign to
betaGBP a role in cancer surveillance and a justification to consider its use in the clinic.

                                             - 143 -
Session III: Apoptosis              Poster III, 29

The 40 kDa Carboxy-terminal Domain of Poly(ADP-ribose)
Polymerase-1 Forms Catalytically Competent
Homo- and Hetero-dimers in the Absence of DNA

Hilda Mendoza-Alvarez and Rafael Alvarez-Gonzalez

The Department of Molecular Biology and Immunology, University of North Texas
Health Science Center at Fort Worth, 3500 Camp Bowie Boulevard, Fort Worth, TX
76107-2699, USA

The 40 kDa carboxy-terminal catalytic domain (CD) of avian poly(ADP-ribose) polymerase
(PARP-1) was cloned, expressed in a baculovirus expression system, and purified to
homogeneity by affinity chromatography. The purified polypeptide synthesized covalent CD-
poly(ADP-ribose) conjugates in the absence of DNA. Electrophoretic analysis of the ADP-
ribose chain length distribution generated indicated that recombinant CD was able to catalyze
the initiation, elongation, and branching reactions of poly(ADP-ribose) synthesis, although at
a 500-fold lower efficiency than wild-type PARP-1. Kinetic evaluation of poly(ADP-ribose)
synthesis showed that the enzymatic activities of CD increased for up to 60 minutes in a time-
dependent manner. Moreover, the rates of CD auto-poly(ADP-ribosyl)ation increased with
second order kinetics as a function of the protein concentration with either betaNAD+ or 3’-
deoxyNAD+ as a substrate. Furthermore, the formation of catalytically competent CD-[PARP-
1] heterodimers was also observed in specific ultrafiltration experiments. Thus, we conclude
that the 40 kDa carboxy terminus of PARP-1 forms a competent catalytic dimer in the
absence of DNA, and that its automodification reaction is intermolecular.

Keywords: poly(ADP-ribose) polymerase-1; catalytic domain; molecular mechanism;
dimerization; ultrafiltration.

                                            - 144 -
Session III: Apoptosis                Poster III, 30

Hypoxia protects HepG2 cells against apoptosis: a possible anti-apoptotic role for HIF-1

Michiels C., Piret J-P., Ninane N., Raes M.

Laboratoire de Biochimie et Biologie Cellulaire, University of Namur (F.U.N.D.P), 61 rue de
Bruxelles, 5000 Namur, Belgium
e-mail :
HIF-1 is the main transcription factor activated by hypoxia. Beside the well-described role
assigned to HIF-1 in the adaptation of cells to hypoxia, recent data describe a possible role for
HIF-1 in the modulation of apoptosis. However, this precise role is not yet clearly understood.
In this study, chemical (3,4 dihydroxybenzoate, a prolyl hydroxylase inhibitor, or CoCl2) and
physiological hypoxia which were shown to induce HIF-1a stabilization and HIF-1 activation,
were able to inhibit apoptosis induced in HepG2 cells by two different pro-apoptotic
conditions, serum deprivation and t-BHP-induced oxidative stress. Indeed, hypoxia reduced
DNA fragmentation, caspase activation and PARP cleavage induced by these two apoptotic
In order to investigate by which mechanism hypoxia could protect against apoptosis, we
studied the expression of some pro- and anti-apoptotic protein in hypoxia by ribonuclease
protection assay. Mcl-1, an anti-apoptotic protein of the Bcl2 family, was shown to be
overexpressed in hypoxia already after 3h incubation. This result has been confirmed by real
time-PCR and immunofluorescence studies. A fragment of 173bp from the human Mcl-1
promoter which contains a potential HRE site has been cloned into the pGL3-enhancer
plasmid upstream of the luciferase reporter gene. When transiently transfected in HepG2, a
basal luciferase activity was detected with the pGL3-173bp in normoxia which increased in
hypoxia or when HIF-1a was overexpressed. Two specific mutations of the HRE decreased
the luciferase activity in comparison to the wild type pGL3-173bp in hypoxia. Using a probe
corresponding to the Mcl-1 HRE, direct HIF-1 binding was demonstrated. These results
indicate that the anti-apoptotic protein Mcl-1 is overexpressed in hypoxia and seems to be
regulated by HIF-1.
These results are very interesting because it is the first demonstration of a direct regulation of
an anti-apoptotic protein by HIF-1. HIF-1-induced Mcl-1 expression probably explains how
hypoxia could exert its anti-apoptotic effect. This observation is an important data in
understanding how tumor growth can occur in challenging environmental conditions.

                                              - 145 -
Session III: Apoptosis               Poster III, 31

The anti-apoptotic protein ICBP90 is target for protein kinases.

Christian Bronner1, Marie-Aline Trotzier1, Cécile Rochette-Egly2, Odile Filhol3, Claude
Cochet3, Marie Scholler-Guinard1, Jean-Paul Klein1 and Marc Mousli1.
 Inserm, UMR S392, Faculté de Pharmacie, B.P. 60024, 67401 Illkirch cedex, 2Inserm UMRS
596 1, rue Laurent Fries, B.P. 10142, 67404 Illkirch Cedex, 3EMI 104, TS, DRDC, CEA, 17
rue des Martyrs, 38054, Grenoble Cedex 09. France. E-mail: Marc.Mousli@pharma.u-

We have recently described a new human protein called ICBP90 that is able to regulate the
topoisomerase II alpha gene expression, essential for chromosome segregation and target for
chemo-therapy. Based on their amino-acid sequences homologies, we have recently proposed
that ICBP90 belongs to a new family of proteins that includes Np95 and NIRF whose
expression is deregulated in cancer cells. We have suggested that this family may exert anti-
apoptotic properties. As phosphorylation processes are essential in the regulation of
transcription factors and as several putative phosphorylation sites have been identified in the
primary sequence of ICBP90, we attempted to identify protein kinases being capable of
phosphorylating ICBP90. We showed that ICBP90 is constitutively phosphorylated in COS-1
cells and that forskolin is able to induce ICBP90 phosphorylation. Protein kinase A (PKA)
was found to be a phosphotransfer enzyme for ICBP90 leading to the in vitro phosphorylation
of multiple serine residues. Protein kinase CK2 is a protein serine/threonine kinase that has
been implicated in cell growth and proliferation whose expression is dysregulated in tumors
that may influence apoptotic activity. We observed that ICBP90 was phosphorylated by the
heterotetrameric CK2 (alpha2, beta2) and more efficiently phosphorylated by the free
CK2alpha subunit. Together our results suggest that PKA and CK2 are important regulators
of the trancriptional activity of ICBP90 and therefore of the anti-apoptotic properties of

                                            - 146 -
Session III: Apoptosis               Poster III, 32

Effects of ZVAD.FMK treatment and VP16-induced apoptosis in human promyelocytic
HL-60 cells

Ruta Navakauskiene1, Grazina Treigyte1, Jurate Savickiene1, Arunas Gineitis1 and Karl-Eric
  Department of Developmental Biology, Institute of Biochemistry, 2600 Vilnius, Lithuania.
  Division of Medical Microbiology, Department of Molecular and Clinical Medicine,
Linköping University, S-581 85 Linköping, Sweden. E-mail:

Many biologic and molecular targeting agents are now under study. Many of these are quite
promising in cancer therapy, i.e. in the treatment of acute promyelocytic leukemia, where new
chemotherapeutic agents with different mechanisms of action are tested. In the present study,
we chose to compare apoptosis in HL-60 cells, treated with the chemotherapeutic drug
etoposide (VP16) or the broad caspase inhibitor ZVAD.FMK. Apoptosis was followed by cell
morphology and agarose gel electrophoresis of extracted cell DNA. We found that
ZVAD.FMK prevents VP16-induced DNA fragmentation and the appearance of increased
number of apoptotic cells in the culture. We compared also the effect of etoposide-induced
apoptosis alone or together with pan-caspase inhibitor ZVAD.FMK on Bcl-2, PCNA and
actin expressions in human promyelocytic leukemia HL-60 cells. Moreover, we assessed
changes in cytoplasmic and nuclear proteins during cell treatment for apoptosis with VP16
alone or together with ZVAD.FMK. Indeed, some proteins were strongly induced in the
cytoplasm and subsequently accumulated in the nuclei following etoposide treatment. During
induction of apoptosis, the increased level of cytosolic proteins and the nuclear translocation
of proteins coincided with the activation of apoptosis. The accumulation of proteins in the
cytoplasm and the nucleus was slightly inhibited by the caspase inhibitor ZVAD-FMK. We
suggest that these proteins are associated with induction of specific signaling cascades that
execute an apoptotic cell death process.

                                            - 147 -
Session III: Apoptosis              Poster III, 33

MAP Kinases and apoptosis during human fibroblast aging

Elise R. Nielsen, Y.E.G. Eskildsen-Helmond, B.F.C. Clark & S.I.S. Rattan

Danish Centre for Molecular Gerontology, Department of Molecular Biology, University of
Aarhus, Denmark. E-mail:

The focus of our laboratory is to understand the molecular basis of human cellular ageing and
to find novel methods of intervention in human fibroblasts. Ageing is a progressive
impairment in functional ability making us more prone to diseases and death. Of various
genetic pathways influencing ageing and longevity, molecular maintenance, repair and stress
response pathways have been shown to play a crucial role. We have shown that induction of
defence and repair pathways by mild stress has beneficial effects on cells and organisms. This
effect is known as hormesis. Furthermore it has been shown that repeated mild heat shock
(RMHS) given to long term cultures of human fibroblasts have many anti-ageing effects.
Although the beneficial effect of RMHS on human cells have been well documented, it is still
not clear how cells sense stress and which signalling mechanisms are involved.
Our focus is on the balance between proliferation and apoptosis in cells. It is generally
accepted that activation of ERK leads to cell proliferation, ERK delivers a cell surviaval
signal counteracting the proapoptotic signals from JNK and p38. But recently it has been
shown that ERK activation is responsible for oxidative stress-induced apoptosis. Earlier JNK
has been shown to be responsible for the heat-induced apoptosis. If apoptosis is increased in
old cells it can lead to cell loss and loss of apoptosis can lead to cancer because of loss of
phenotypic fidelity. This leads us to our current aim of study; whether changes in the balance
between various signalling intermediates change with age and thereby change the balance
between proliferation and apoptosis. If this is the case we will try to counteract it with

                                            - 148 -
Session III: Apoptosis               Poster III, 34

Glutamate induced apoptosis in SH-SY5Y human neuroblastoma cells: role played by
plasma membrane NADPH oxidase complex and intracellular calcium ions.

Sevdalina Nikolova1 and Jung-Ae Kim2
 Department of Biotechnology and 2College of Pharmacy, Yeungnam University, Kyongsan
712-749, South Korea, E-mail :,

Oxidative stress has been extensively studied in various neurologic diseases. Since the brain is
especially susceptible to oxidative stress and subsequent damage to the cells, in a disease such
as Alzheimer’s disease and stroke/ischemia, oxidative stress is felt to play a key role in the
loss of neurons and progression to the disease. In a previous study, we found that SH-SY5Y
human neuroblastoma cells are expressing all membrane-associated (cytochrome b558) and
cytosolic components (Rac1/2, p67phox, p47phox, p40phox) of NADPH oxidase system and
that reactive oxygen species (ROS) production through NADPH oxidase play an important
role in glutamate-induced apoptosis of the cells. In the present study, we thus examined the
signaling pathway of glutamate-induced activation of plasma membrane NADPH oxidase.
Treatment with glutamate in serum-supplemented or serum-free conditions induced apoptotic
cell death at IC50 of 100 mM. We could not detect any alterations in intracellular free calcium
levels in glutamate-treated cells. However, we found that pretreatment with BAPTA/AM (2
µM), dantrolene (40 µM) and TMB-8 (4µM) prevented glutamate-induced cell death. Also,
combination of glutamate and calcium ionophore A23187 (0.5 µM) further suppressed the
neuronal cell viability compared to glutamate alone. These results indicate the pivotal role of
calcium ions in the process of glutamate neurotoxicity. The use of N-methyl-D-aspartate
(NMDA), receptor agonist (50mM) and (±)-a-amino-3-hydroxy-5-methylisoxazole-4-
propionic acid hydrate (AMPA) receptor agonist (20mM) potentiated glutamate-induced cell
death. In addition, MK 801 (20 µM), a selective NMDA-receptor antagonist (+), abolished the
lethal effect of glutamate in the cells. Collectively, our results suggest that glutamate acting
on ionotropic glutamate receptor may induce dysregulation of calcium homeostasis, activation
of NADPH oxidase complex and ROS overloading, which eventually leads to cell death in
SH-SY5Y human neuroblastoma cells.

                                             - 149 -
Session III: Apoptosis               Poster III, 35

Survival pathways and apoptosis regulatory proteins expression pattern in activated
human hepatic stellate cells: a mechanism to sustain development of liver fibrosis.
    Erica Novo, 1Elena Zamara, 1Lorenzo Valfrè di Bonzo, 2Ilaria Petrai, 2Andrea Bonacchi,
    Sebastiano Colombatto, 2Massimo Pinzani, 2Fabio Marra, 1Maurizio Parola.
Dip. Medicina e Oncologia Sperimentale, University of Torino and 2Dip. Medicina Interna,
University of Florence, Italy. e-mail

Activated hepatic stellate cells (HSC) in their myofibroblast-like phenotype (HSC/MF) are
crucial pro-fibrogenic cells able to sustain fibrotic progression of chronic liver disease
towards the end-stage of cirrhosis. Human HSC/MF in primary culture express an antigenic
profile identical to that of interface myofibroblasts detected in biopsies of fibrotic/cirrhotic
human livers, do not undergo spontaneous apoptosis and are able to survive to several pro-
apoptotic stimuli or conditions like serum deprivation, FasL, NGF, TNFalfa, doxorubicin,
etoposide and the oxidative stress mediator 4-hydroxynonenal. Inhibition of protein synthesis
by cycloheximide or of transcription by Actinomycin D is an absolute requirement for
induction of classic apoptosis in HSC/MF. The present study was performed to investigate
possible mechanisms that may favour survival of activated human HSC/MF to apoptotic cell
death. Quiescent (i.e. freshly isolated) HSC and activated HSC/MF in primary culture,
obtained from human livers unsuitable for transplantation, have been used in these
experiments. Our results indicate that activated human HSC/MF express very high protein
levels of Bcl-2, IkBalfa, Akt and phospho-Akt, that are not expressed or expressed at very
low levels by quiescent HSC. These cells also express low levels of bax and co-express p75
and TrkA, with NGF-stimulated, TrkA-dependent activeRaf-1/Erk signaling, a feature known
to favour survival in cells exposed to NGF. We conclude that activation of human HSC/MF
is associated with the development of an expression pattern favouring survival to the
induction of apoptosis and that such a feature may be relevant to sustain fibrotic development
of chronic liver disease.

                                             - 150 -
Session III: Apoptosis                  Poster III, 36

Apoptotis signal transduction of ejaculated human spermatozoa in response to
physiological and pathological stimuli

Uwe Paasch1, Sonja Grunewald1, Tamer S. Mahmout2, Ashok Agarwal2; Rakesh K. Sharma2, Hans-
Juergen Glander1
  EAA Center University of Leipzig, Stephanstrasse 11, 04109 Leipzig, Germany, andro@medizin.uni-, 2 Cleveland Clinic Foundation, Cleveland, Ohio, USA

Introduction: In spermatozoa the presence of CD95 and caspase (CP) 8 activation (type I apoptosis),
activation of CP9 (type II apoptosis), the activation of their shared effector CP3 and CP1 have been
reported. Up to know it remains unclear which pysiological or pathological factors lead to an activation
of those signaling pathways. The objective of our study was to examine the role of receptor and
mitochondrial mediated as well as CP1 derived apoptosis signaling in human spermatozoa in response to
inducers of type I and II apoptosis (CD95, Betulinic acid, BA), oxidative stress (HOCI) and capacitation.
Methods: Semen specimens collected from 15 healthy donors were separated into 7 aliquots. Three
pellets was re-suspended in PBS (10 min, 1h and 3h) and served as a control. Two aliquots were
incubated with 2 µg/mL CD95 antibody and 60 µg/mL Betulinic acid for 1 h and 10 min respectively.
Another aliquot was subjected to incubation with HOCI (10-3 mol/L, 1h). The remaining pellet was re-
suspended in 1 mL of BWW + 3% BSA at 5% CO2 to induce capacitation (3h). Using
carboxyfluorescein derivatives, the levels of active caspase 1, 3, 8 and 9 were estimated via flow
Results: CD95 and HOCI treatment did not result in activation of any CP. BA resulted in a significant
increase in activation of CP9 and 3. Capacitation process, a prerequisite for fertilization, lead to
significantly higher CP 1, 9 and 3 activation.
Conclusions: Human spermatozoa do not display the complete signaling pathways of apoptosis in
response to physiological and pathological apoptotic stimuli. The mitochondria derived type II pathways
are predominant in both conditions and should be the target of intervention.

                                                 - 151 -
Session III: Apoptosis               Poster III, 37

Proapoptotic potentials of genistein under growth stimulation by estrogen

Ock Jin Park and Jang-In Shin

Department of Food and Nutrition, Hannam University, Daejeon 306-791, Korea, E-mail:

In mammary carcinogenesis, hormonal effects have been reported to be important factors.
Estrogens are known to regulate the proliferation of breast cancer cells. Whereas genistein has
been shown to induce apoptosis in mammary tumor cells. This study examined genistein-
induced apoptosis through the regulation of bcl-2 and bax expressions in the presence of
estrogen. MCF-7 cells were treated with either genistein (25, 50 and 100 uM) or in the
presence of 17-beta estradiol (12.5, 25 and 50 nM) for 48 hr. DNA ladder analysis and
western blot analysis of bax, bcl-2, cyclin B1, p21 and p53 were carried out. For comparison,
in vivo system was employed using estrogen-deficient and estrogen-sufficient female rats at
the two different concentrations of genistein. In MCF-7 cells, DNA fragmentation was
evident by the treatment of genistein in the absence or in the presence of estrogen.
Downregulation of bcl-2 and upregulation of bax by genistein were observed. However,
genistein could not show proapoptotic properties in the presence of estrogen except with the
lowest concentration of estrogen. However, p21 and p53 protein expressions were
upregulated by high concentrations of genistein in the presence of estrogen. Bcl-2/bax ratios
were lowered by genistein treatment at the presence or absence of estrogen in female rats.
These results demonstrate that proapoptotic property of genistein might be influenced greatly
by the concentration of estrogen in vitro system, and this influence by estrogen was not
evident in vivo system.

                                            - 152 -
Session III: Apoptosis               Poster III, 38

Endogenous oxidative stress in melanoma cells induces distinct redox forms of TNFR1
receptor with distinct ligand binding and apoptotic signaling ability

Silvia Dominici, Aldo Paolicchi, Maria Franzini, Vincenzo De Tata, Alfonso Pompella

Dipart. di Patologia Sperimentale BMIE, Università di Pisa – Via Roma 55, 56126 Pisa, Italy.

Oxidation/reduction reactions are a mechanism for regulation of important functions of the
cell, including the proliferation/apoptosis balance. Several biomolecules involved in signal
transduction and regulation of gene expression are sensitive to prooxidants, e.g. reactive
oxygen species (ROS), and a true “redox regulation” has been described for many of them.
Low, basal levels of prooxidants are produced in the cell by several sources, e.g. respiration,
NADPH oxidase, xantine oxidase. Our previous work allowed to identify membrane gamma-
glutamyltransferase (GGT) as one additional source (Bioch. Pharm. 64, 1029, 2002). ROS
and other free radicals are in fact originated the cell surface during the GGT-mediated
metabolism of extracellular glutathione (GSH). Cell surface receptors of the TNFR
superfamily are known to possess cysteine-rich regions in their extracellular domains. This
study was aimed thus to verify whether different conditions of GGT activity might be
reflected in redox changes of TNFR1 redox status and function.
Me665/2 human melanoma clones, expressing varying levels of GGT activity, were studied,
in conditions of GGT stimulation (addition of substrates) vs. specific inhibition with the
boronate analogue L-2-amine-4-boronobutanoic acid (ABBA). Thiol redox status of TNFR1
protein was analyzed by immunoblot after derivatization of cell surface reduced thiols with
the thiol reagent maleimide-polyethylene glycole (MalPEG), by analyzing the molecular
weigth shifts produced with MalPEG. Ligand binding capacity of TNFR1 was determined by
using 125I-labeled recombinant TNF-a. Triggering of apoptosis was investigated in cells
exposed to TNF-alpha and cycloheximide, by determining caspase activation with an ELISA
At least five distinct redox forms of TNFR1 could be identified with MalPEG. Forms
containing more reduced thiols were prevalent in the 2/21 clone, expressing low levels of
GGT activity, as compared to GGT-rich 2/60 cells. Accordingly, GGT stimulation in 2/21
clone resulted in a shift from reduced to oxidized TNFR1 forms. At the same time, GGT
stimulation resulted in a significant increase in 125I-TNF-a binding to the cells, while the
opposite was observed after GGT inhibition. A higher resistance of cells with ‘oxidized’
TNFR1 to TNF-a-induced apoptosis was also observed.
Thus, prooxidant reactions initiated at the cell surface during GGT-mediated GSH catabolism
appear able to modulate cell sensitivity to TNF-a challenge, by altering the binding affinity of
TNFR1 through a redox modulation of its thiol residues. As expression of GGT is frequent in
human neoplasms, and is often increased at more advanced stages of progression, the
processes described might concur to the onset of drug resistance, particularly in the case of
(prooxidant) drugs acting through the cellular apoptotic machinery.

Acknowledgments. The present study was supported by the Associazione Italiana Ricerca sul
Cancro, (A.I.R.C., Italy) and by the Italian Ministry for Education, University and Research
(‘FIRB-2001’ Funds).

                                             - 153 -
Session III : Apoptosis              Poster III, 39

Bace overxpression regulates amyloid precursor protein cleavage and interaction with
ShcA adapter: role in intracellular signaling and apoptotic death.

Emanuela Repetto, Claudio Russo, Valentina Venezia, Serena Salis, Virginia Dolcini,
Francesca Genova, Mario Nizzari, Elisabetta Violani, Pia Carlo and Gennaro Schettini.

Farmacologia e Neuroscienze, DOBiG Università di Genova, c/o IST Largo R. Benzi 10,
16132 Genova, Italy.

There is evidence that BACE1 is the enzyme responsible for beta-site cleavage of the amyloid
precursor protein (APP) and for the formation of C-terminal fragments of APP or CTFs. The
proteolytic processing of CTFs by a gamma-secretase then leads to the formation of the
amyloid beta peptide (Abeta)) that is thought to be the culprit for the neurodegeneration in
Alzheimer's disease (AD).
APP has a structure similar to a type one transmembrane receptor with a cytosolic domain
that can be involved in different protein-protein interactions that are likely linked to signal
transduction events. We have recently shown in human brain that APP and CTFs
phosphorylated in tyrosine residues are substrates for adaptor proteins like ShcA and Grb2.
To investigate the interaction between ShcA, APP and CTFs and to clarify the significance
for such interactions in cultured cells, we have utilised an in vitro model that overexpress
We have stably transfected a human neuroglioma cell line H4 with BACE and we have
analysed the interaction between APP/CTFs and ShcA through immunoprecipitation with
specific antibodies and Western Blot analysis.
Co-immunoprecipitation studies in H4wt and H4 BACE revealed that APP, CTFs and ShcA
interaction is differently modulated when BACE is overexpressed and that the composition of
CTFs between 16 KDa and 8 KDa that are coimmunoprecipitated with ShcA and Grb2 has a
different electrophoretic pattern in H4 compared to H4 BACE, consistently with BACE
activity. Moreover the overexpression of BACE1 in H4 cells influence the apoptotic status of
the cells and the signalling activity of ShcA adaptor.
Therefore, our data suggest that BACE1 activity influences APPs’ processing and its
intracellular signaling through ShcA adaptor protein, rendering the cells more prone to
apoptotic death..
Alzheimer Association Grant 2002 IIRG-02-3976, Telethon Grant E1144 to G.S.

                                            - 154 -
Session III: Apoptosis              Poster III, 40

Expression of cell proliferation and apoptosis markers in papillomas and cancers of
conjunctiva and eyelid

Joanna Reszec, Mariola Sulkowska, Mariusz Koda, Luiza Kanczuga-Koda, Stanislaw

Department of Clinical and General Pathology Medical University of Bialystok, Waszyngtona
13, Poland, e-mail:

Cell proliferation and programmed cell death are considered to be important events in
carcinogenesis. The object of our study was to evaluate the expression of Bcl-2 protein family
(Bcl-2, Bak, Bax), P53 protein expression, PCNA and Ki-67 protein immunoexpression as
well as the correlation between examined markers and some clinicopathological features in
papillomas and cancers of conjunctiva and eyelid. 45 squamous cell papillomas (SCP), 11
squamous cell cancers (SCC) and 27 basal cell cancers (BCC) were estimated. In SCP group
P53 protein expression was observed in 30 cases (66,6%), Ki-67 in 14 cases (31,1%), PCNA
in 44 cases (97,7%), Bcl-2 in 24 (53,3%), Bak in 28 (62,2%) and Bax in 31 cases (68,8%). In
SCC group P53 protein expression was evaluated in 8 cases (72,7%), Ki-67 in 2 (18,1%),
PCNA in 8 (72,2%), Bcl-2 in 5 (45,4%), Bax and Bak both in 10 cases (90,9%). In BCC
group P53 protein expression was estimated in 23 cases (85,1%), Ki-67 in 13 (48,1%), PCNA
in 26 cases (96,2%), Bcl-2 in 13 (48,1%), Bak in 21 (77,7%) and Bax in 22 cases (81,4%).
There was observed a correlation between some clinicopathological features and examined
markers of apoptosis and cell proliferation, which seemed to be important events in cancer

                                            - 155 -
Session III: Apoptosis               Poster III, 41

Monocytic differentiation and apoptosis of promyelocytic leukemia: alterations in Sp1
and NF-kB transcription factors activity

Jurate Savickiene1, Grazina Treigyte1, Ruta Navakauskiene1 and Karl-Eric Magnusson2.
  Department of Developmental Biology, Institute of Biochemistry, LT-2600 Vilnius,
Lithuania. E-mail:      
  Division of Medical Microbiology, Department of Molecular and Clinical Medicine,
Linköping University, SE-58185 Linköping, Sweden. E-mail:

The induction of leukemic blasts differentiation represents an attractive strategy for the
treatment of leukemias. Differentiation of promyelocytic leukemia cells toward granulocytes
or monocytes has been shown to induce apoptotic cell death. However, the relationship
between terminal differentiation and apoptosis remains unclear. We examined whether cell
cycle regulating p21 (Waf1/Cip1) and cell death (FasL) genes are controlled by Sp1 and NF-
kappaB transcription factors during monocytic differentiation-related apoptotic process in
promyelocytic leukemia NB4 cell line induced by PMA. Using electrophoretic mobility shift
assays (EMSA) we observed that 8h-PMA treatment caused an early response on Sp1 binding
to the p21 and the FasL promoter. The firmly adherent cell phenotype, characteristic for
differentiated cells, retained Sp1 binding activity to both promoters, but it was lost completely
or partially in secondarily detached, apoptotic cells. The binding capacities of Sp1 to the p21
promoter correlated with the levels of expressed protein detected by immunoblotting. NF-
kappaB binding intensity to consensus sites and to the FasL promoter was associated
positively with NB4 cell maturation and negatively with apoptosis. The data speak for the
requirement of p21 and FasL gene activation for leukemic cell survival and maturation, but
not for death through the involvement of Sp1 and NF-kappaB, as regulators of these genes.

                                             - 156 -
Session III: Apoptosis               Poster III, 42

Mitochondrial transmembrane potential, ATP and ROS levels in transgenic mouse
model of alzheimer`s disease

Isabel Scherping1, Christian Czech2,3, Laurent Pradier2, Walter E. Mueller1, Anne Eckert1
  Department of Pharmcology, Biocentre, University of Frankfurt, Germany
  Aventis Pharma, Research and Development, F-94403 Vitry-sur-Seine, France
  F. Hoffmann-La Roche AG, CNS Research, CH-4070 Basel, Switzerland

Mitochondrial dysfunction has been identified in a large proportion of neurodegenerative
disorders including Alzheimers`s disease (AD). It was reported that accumulation of amyloid
beta (A beta) and oxidative stress also play central roles in the pathogenesis of this disease by
probably directly leading to mitochondrial dysfunction.
We investigated two different groups of transgenic (tg) mice, the swedisch mutant amyloid
precursor protein (APP) tg mice and double transgenic mice for APP and presenilin1
(tgAPP/PS1) at the age of 3 months. We determined the mitochondrial membrane potential,
reactive oxygen species (ROS) and ATP levels in acutely dissociated brain cells of non-tg
littermate control animals, tgAPP mice and tgAPP/PS1 mice. TgAPP is one mouse model,
which exhibits onset of A (beta) plaques at an age of 6 months, but intracellular A (beta) load
is already increased at the age of 3 months. TgAPP/PS1 mice posses A (beta) plaques already
at an age of 2.5-3 months.
We detected decreased basal levels of mitochondrial membrane potential (Psim) in tgAPP
mice compared to littermate non-tg controls mice. A further loss of Psim was measured in
tgAPP/PS1 mice compared to tgAPP mice. Additionally, basal ATP levels were decreased in
the same pattern. Hydrogen peroxide and the nitric oxide donor sodium nitroprussid damaged
the cells by decreasing Psim in a dose-dependent manner. Complementary, the intracellular
accumulation of ROS induced by Fe(III) was increased in cells of tg APP/PS1 mice and
tgAPP mice compared to littermate non-tg controls mice.
Our results further emphazise the important role of mitochondrial dysfunction in the
pathogenesis of AD. Moreover, they indicate that already soluble A (beta) is involved in these
neurotoxic process.

                                             - 157 -
Session III: Apoptosis               Poster III, 43


Lidia Semenkova , Elena Dudich, Igor Dudich, Natasha Tokhtamisheva

Department of Molecular & Cell Biology, Institute of Immunological Engineering,
Lyubuchany, Moscow Region, Russia, 142380. E-mail:

Mitochondrial apoptosis pathway is mediated via cytochrome c release with subsequent
formation of the Apaf-1/cyt-c/dATP/procaspase-9 complex, leading to activation of caspase-9
and downstream effector caspase-3. Our recent data demonstrated that oncoembryonic marker
alpha-fetoprotein (AFP) could synergistically enhance caspase-9 and caspase-3 activation in
cell-free cytosolic extracts of tumor cells in the presence of the suboptimal dose of
endogenous or exogenous cyt-c. We therefore proposed that AFP might physically associate
with certain members of the apoptosome complex regulating thus its activity. High-resolution
gel chromatography of the cyt-c/dATP-activated cell-free cytosolic extract of HepG2 cells
demonstrated that addition of AFP was resulted in the notable changes in distribution of the
caspase-3 activity along the gel-filtration elution pattern. It was observed that AFP induced
notable enhance of the caspase-3 activity in the region of the biologically active Mr ~ 700 kDa
apoptosome complex and stimulated release of the active caspase-3 from the complex.
Distribution of the activity of caspase-9 was not affected by AFP treatment. To determine
possible mechanisms of the AFP-mediated regulation of the apoptosome complex, we
monitored the distribution of the Apaf-1 and cIAP-2 along the chromatographic patterns of
the apoptosome assembly, which had been formed with and without AFP. Our data
demonstrated that AFP positively regulated cyt-c/dATP-mediated formation of the active Mr
~ 700 kDa apoptosome by recruitment of Apaf-1 into the complex. It was shown that cIAP-2
distribution was also affected by AFP addition. In the absence of AFP, cIAP-2-specific
material was presented in the Mr ~ 700 kDa apoptosome, whereas after AFP addition, the
cIAP-2 specificity was significantly reduced in this region. By using direct protein-protein
interaction assay we demonstrated that pure human AFP practically completely disrupted the
association between processed caspases-3 and -9 and cIAP-2 demonstrating its release from
the complex. Our data suggest that AFP may regulate cell death by displacing of cIAP-2, the
inhibitor-of-apoptosis-protein, from the apoptosome resulting in promoting of caspase-3
activation and its release from the complex.

                                            - 158 -
Session III : Apoptosis             Poster III, 44

Red ox–signaling by ionizing rad iation in mouse liver

Jeu ng Hee An, Jiyoung Kim, Jinsil Seong ,

Department of Radiation Oncology, Brain Korea 21 Project for Medical Science, Yonsei
University Medical College, Shinchon-dong 134, Seodamun-Ku, Seoul, Korea, E-mail:

Since radiation treatment has been reappraised in the treatment of hepatic tumors, radiation
response in the liver is emerging as an interesting new area of investigation. In this study,
identification of the repertoire of signaling proteins was performed by a proteomics approach
involving cellular responses of liver tissue to ionizing radiation. Approximately 800 protein
spots were detected. Among them, a least 28 proteins showed a significant quantitative
alteration following radiation. The significantly altered proteins were categorized to ones
related to reactive oxygen species (ROS) metabolism, metabolic pathway, and G-type
proteins. Particularly, the expression level of proteins related to ROS metabolism including
cytochrome c, glutathione S transferase Pi (GSTP), NADH dehydrogenase and peroxiredoxin
VI (Prx VI) were increased after radiation. It is suggested that while radiation initiates its
cytotoxic effects, it can also induce radioprotective antioxidant system.

                                            - 159 -
Session III: Apoptosis               Poster III, 45

Tetrandrine-induced apoptosis is mediated by activation of caspases and PKC-delta in
U937 cells.

Byeong-Churl Jang, Ki-Jo Lim, Taek Kyu Kwon, Jong-Wook Park, Seong-Il Suh

Chronic Disease Research Center and Institute for Medical Science, School of Medicine
Keimyung University, Daegu, South Korea 700-712. E-mail :

Tetrandrine, which is isolated from Chinese herb Stephania tetrandrae, possesses anti-
inflammatory, immunosuppressive, and cytoprotective properties. Though it was previously
shown that tetrandrine causes a G1 blockade and apoptosis in various cell types, however, the
mechanism by which tetrandrine initiates apoptosis remains poorly understood. In present
study, we investigated the mechanisms of apoptosis induced by tetrandrine in U937 cells.
Treatment of U937 cells with tetrandrine (10 uM) for 24 hr induced chromatin fragmentation,
cytochrome c release, and caspase activation. Tetrandrine also induced early oxidative stress,
which results in activation of JNK, but not ERK and p38 MAPK. A broad-spectrum caspase
inhibitor and anti-oxidants significantly blocked tetrandrine-induced caspase-3 activation.
However, inhibition of the JNK activity with SP600125 did not block tetrandrine-induced
apoptosis. Tetrandrine-induced apoptosis of U937 cells also required activity of PKC-delta,
because pretreatment with a specific PKC-delta inhibitor greatly blocked tetrandrine-induced
caspase-3 activation and the apoptotic response to tetrandrine was significantly attenuated in
dominant-negative PKC-delta transfected MCF-7 cells, suggesting that PKC-delta plays an
important role in tetrandrine-induced apoptosis. These results suggest that tetrandrine induces
oxidative stress, JNK activation, and caspase activation. However, JNK activation by ROS is
not involved in the tetrandrine-induced apoptosis. In addition, tetrandrine induces caspase-
dependent generation of a catalytically active fragment of PKC-delta, and this fragment also
appears to play a role in the activation of caspases.

                                            - 160 -
Session III: Apoptosis               Poster III, 46

Alpha-fetoprotein confers estradiol-mediated mammary carcinoma cell growth in vitro
in synergy with tamoxifen. Direct association of the AFP/E2 complexes with estrogen

Elena Dudich, Olga Goncharova, Lidia Semenkova, Igor Dudich and *Edward Tatoulov.
Institute of Engineering Immunology, Lyubuchany, Moscow Region, Russia, *JSC Bio-
Sistema, Moscow, Ostozhenka-16, Russia. E-mail:

Human a-fetoprotein (AFP), isolated from the cord sera, was shown to interact with estradiol
(E2) that significantly affects its tertiary structure conformation. It has been demonstrated
earlier, that human AFP could induce apoptosis in various tumor cells in a strongly dose-
dependent manner. We studied tumor suppressive effects of AFP/E2 complexes and the role of
estrogen receptors in this effect, using ER-expressing human hepatoma HepG2 and mammary
carcinoma MCF-7 tumor cell lines. To prepare AFP/E2 complexes AFP was incubated with
molar excess of E2 for 1 h at room temperature and then serially diluted complex was added to
the cell culture for 24 h. Both types of cells revealed significant enhance of the total growth-
suppressive response to AFP/E2 in comparison with AFP alone. Our data demonstrated that
the most significant tumor suppression was observed at low concentration of E2 in AFP/E2
complex (molar ratio 1:1), whereas significant molar excess of E2 in AFP/E2 led to the
decrease of the total tumor suppressive effect. At the contrast, simultaneous addition of AFP
and E2 to the MCF-7 cells without pre-incubation showed no effect. However, cell
preincubation with E2 with subsequent treatment with AFP/E2 led to the complete abrogation
of apoptotic effect, whereas E2 did not confer apoptosis induced by AFP alone in the same
cells. Preincubation of MCF-7 cells with tamoxifen to block nuclear estrogen receptors did
not abrogate AFP/E2 or AFP-mediated apoptosis. Moreover, it was observed synergistic
enhance of the total tumor-suppressive effect when cells were treated with suboptimal doses
of AFP/tamoxifen mixture. Physical interaction of AFP and ER-complex was demonstrated
by coimmunoprecipitation. It was shown that pure AFP interacted with ER and Hsp90 by
formation of heterooligomeric complexes. Taken together, our data show that: (i) AFP/E2
complexes use distinct intracellular molecular pathways in apoptosis signaling, in comparison
with AFP alone; (ii) AFP and AFP/E2 are involved in physical interaction with nuclear ER
complex; (iii) AFP induces apoptosis independently on ER; (iv) AFP and tamoxifen operate
synergistically in tumor suppression by engaging of distinct intracellular pathways; (v) E2
excess prevents AFP/E2-mediated tumor suppression via concurrence for the binding with
intracellular ERs.

                                             - 161 -
Session III: Apoptosis                Poster III, 47

Apoptosis modulates the signalling of APP in SH-SY5Y neuroblastoma cells.

Valentina Venezia, Claudio Russo, Emanuela Repetto, Virginia Dolcini, Pia Carlo, Elisabetta
Violani,         Mario         Nizzari        and         Gennaro           Schettini.

Pharmacology and Neurosciences, DOBiG University of Genova and IST, c/o CBA. Largo R.
Benzi 10, 16132 Genova, Italy. E-mail :

The amyloid precursor protein (APP) is a cell surface protein with a short cytoplasmic tail. Its
location and structural features are characteristic of a receptor for signal transduction. Yet, the
physiological function of APP is unclear, although it is well documented that APP’s
proteolytic processing, through the formation of membrane-bound C-terminal fragments
(CTFs) and of beta-amyloid peptides (Ab), likely influences the development of Alzheimer’s
disease (AD). We have recently shown that tyrosine-phosphorylated APP (and CTFs) may
interact with ShcA and Grb2 adaptor proteins and that the activation of these pathways may
be pathogenically related to AD. Here we show that in normally proliferating SH-SY5Y
neuroblastoma cells APP is complexed with Grb2 adaptor, while upon induction of apoptosis
APP and its complex with Grb2 is degraded. On the contrary, in apoptotic cells tyrosine-
phosphorylated CTFs are complexed with ShcA and Grb2. Inhibitors of BACE1 (the enzyme
responsible for the formation of some CTFs) and caspase’s inhibitors may partially block the
degradation of APP and the co-precipitation of CTFs with Grb2-ShcA adaptors, restoring the
complexes with APP. Moreover, as caspase’s inhibitor partially rescues SY5Y cells from
death, BACE1 inhibitors also may partially protect neuroblastoma cells from apoptosis,
suggesting a role for CTFs in maintaining the apoptotic stimulus.
In summary our data suggest that in SH-SY5Y tyrosine-phosphorylated APP is involved in a
complex with Grb2-ShcA adaptors that is disrupted during apoptosis. The abnormal
degradation of APP and consequent increased levels of CTFs (like in Alzheimer Disease or in
Down’s syndrome) generate a complex between tyrosine-phosphorylated CTFs and
intracellular adaptors. The signalling through APP and its CTFs may have a significant
relevance for apoptotic cell death in AD.

Alzheimer Association Grant 2002 IIRG-02-3976, Telethon Grant E1144, MURST PRIN
2000, to G.S.

                                              - 162 -
Session III: Apoptosis               Poster III, 48

Sodium ascorbate enhances menadione redox cycling killing cancer cells by autoschizis,
a caspase-3 independent cell death. A key role of hydrogen peroxide.

Julien Verrax, Marianne Delvaux, Isabelle Blave, Henryk Taper and Pedro Buc Calderon

Unité de Pharmacocinetique, Metabolisme, Nutrition et Toxicologie, Département de sciences
pharmaceutiques, Faculté de médecine, Université Catholique de Louvain, e-

We have shown that a simultaneous administration of ascorbate and menadione produced
tumour growth inhibition and increased the life-span of tumour-bearing mice. The molecular
mechanism underlying such a process is, however, still unknown. Several arguments support
the idea that hydrogen peroxide generated during ascorbate oxidation by menadione thus
generating a redox cycling, surpasses cancer cellular defence systems and results in cell death.
The ability of several quinones (with different half-redox potentials) to oxidize ascorbate was
tested by recording both oxygen uptake and cell death. A strong relationship was observed
between the half-redox potential of quinones and the oxygen uptake and the cytolytic effect
they induced when associated with ascorbate.
With regards to the type of cell death, our results exclude an apoptotic process, considering
the absence of DEVDase activity, the lack of procaspase-3 and PARP cleavage as well as the
severe ATP depletion after treatment. In agreement to previous results (morphological
examination and flow cytometry analysis) it appears that cancer cells exposed to the
association of ascorbate and menadione dye mainly by autoschizis, a novel form of cell death
that is caspase-3 independent
A better knowledge about the mechanism by which this association kills cancer cells may
argue for their use in cancer treatment, taking advantage of the poor antioxidant status of
cancer cells.

                                             - 163 -
Session III: apoptosis        Poster III, 49

A RasGAP-derived cell permeable peptide potently enhances genotoxin-induced
cytotoxicity in tumor cells.

David Michod§, Jiang-Yan Yang§, and Christian Widmann§
   Institut de biologie cellulaire et de morphologie (IBCM) [http://www-], Biology and Medicine Faculty, Lausanne University,

Treatment of many cancers relies on the combined action of several genotoxins but the
detrimental effect of these drugs on normal cells can cause severe side effects. One major
challenge in anti-cancer therapy is therefore to increase the selectivity of current treatments
toward cancer cells in order to spare normal cells. We have recently demonstrated that a
RasGAP caspase cleavage fragment is able to sensitize HeLa cells towards cisplatin-induced
apoptosis. Here we extend this observation by showing that this fragment also enhances cell
death induced by adriamycin and mitoxantrone, two other widely used genotoxins.
Furthermore, we have delineated a short sequence within this fragment that still bears the
genotoxin-sensitization property. The peptide encoded by this sequence, when fused to the
TAT cell permeation sequence, potently sensitized a number of tumors cells, but not normal
cells, towards apoptosis induced by cisplatin, adriamycin and mitoxantrone. Our results
demonstrate the feasibility in enhancing the efficacy of currently used drugs to selectively kill
cancer cells using peptides derived from pro-apoptotic caspase substrate fragments.

                                               - 164 -
Session III : Apoptosis             Poster III, 50

Artifactual generation of oxygen free radicals in the mixture of cyanide and glycerol

Eun-Jin Yeo, Yang-Sook Chun, Hwa-Jin Suh, and Jong-Wan Park

Cancer Research Institute and Department of Pharmacology, College of Medicine, Seoul
National University, 28 Yongon-dong, Chongno-gu, Seoul 110-799, Korea. E-mail :

Reactive oxygen species (ROS) are involved in tumor promotion or apoptosis. On the other
hand, cyanide and glycerol have been commonly used as a hypoxia inducer and a stabilizer of
NAD(P)H oxidases, respectively. Since both compounds are relatively inert, they have been
used concomitantly regardless of any mutual interference. In this study, we demonstrate that a
mixture of glycerol and cyanide reduced cytochrome C and nitroblue tetrazolium, which are
superoxide anion indicators. As a mixture, they also enhanced the production of superoxide
anion in the presence of redox-cycling compounds. Superoxide production by the mixture was
confirmed by electron spin resonance spectra. Moreover, the mixture induced lipid
peroxidation and hemolysis in human erythrocytes. These results suggest that cyanide and
glycerol should be used carefully in reaction systems used for measuring ROS production or
antioxidant activity. In contrast, sucrose and sodium azide in combination do not produce
such artifacts, and thus, may be used as an alternative.

                                            - 165 -
Session III: Apoptosis              Poster III, 51

Wei Li (M.D., Ph.D.) and Xi-Ming Yuan (M.D., Ph.D.)
Division of Pathology II, Faculty of Health Sciences,
Linköping University, Linköping, Sweden
Objective: It has been recently noticed that atherosclerotic lesions in both human and mice
over-express several lysosomal proteases including cathepsins B, D, L, and S, which are
potent elastases and may participate in the remodelling of extra cellular matrix associated
with the plaque stability. However, the mechanisms responsible for the overexpression of
lysosomal cathepsins and related atherogenic implications remain unknown. Our study aims
were to test if (i) oxidized LDL and oxysterols may cause overexpression of cathepsins in
macrophages, and (ii) lysosomal cathepsins can function extra-lysosomally. Methods: Normal
human arterial segments from mammary (n = 4), carotid (n = 4), coronary arteries (n = 5), and
atherosclerotic coronary (n = 5), and carotid arteries (n = 16) were collected for cathepsins
immunohistochemistry and computerized image analysis. U937, THP-1, J774 cells and
human macrophages were used for assays of cell viability, apoptosis, and necrosis after
exposure to oxidized LDL and oxysterols. The distribution of cathepsins B, D and L were
examined by cytochemistry, immunocytochemistry, and western blot. The mRNA levels of
cathepsins B, D, L were studied by RT-PCR. Results: In human atheroma, over-expression of
cathepsins B and L is mainly in macrophage-infiltrated areas in a lesion dependent manner.
Over-expressed cathepsins B and L, normally confined to the lysosomal compartment, are
present in the cytoplasm and nuclei of apoptotic macrophages. In vitro macrophages exposed
to oxLDL or oxysterols initially transformed into foam cells, and then assumed apoptotic-type
morphology with TUNEL-positive nuclei. This is secondary to lysosomal destabilization,
with leakage to the cytosol of lysosomal enzymes and caspase-3 activation. After oxLDL
exposure the cytosolic activities of lysosomal enzymes NAbetaGase and cathepsins B & L
were significantly enhanced. Relocation to the cytosol of cathepsin D, as estimated by light
and electron microscopic immunocytochemistry is also noticed. Exposure to AcLDL resulted
in its uptake with enlargement of the lysosomal apparatus, but the stability of the lysosomal
membranes was not changed. Individual 7ß-oxysterol induced over-expression of cathepsins
B and D as detected by immunocytochemistry and RTPCR. Like macrophages within
atheroma, intralysosomal cathepsins B and L are translocated to the cytoplasm and nuclei of
7-oxysterol-exposed cells. Data from computerized image analysis reveal that 7-oxysterol up-
regulates cathepsin synthesis and causes the translocation of the enzymes to the cytosol and
nuclei areas. Conclusion: The results indicate that endocytosed oxLDL and oxysterols not
only destabilizes the acidic vacuolar compartment but also cause relocation and up-regulation
of lysosomal enzymes. Lysosomal cathepsins may act as cleaving enzymes during DNA
damage and macrophage apoptosis induced by oxidized lipids.

                                           - 166 -
Session III: Apoptosis              Poster III, 52

Wei Li (M.D., Ph.D.) and Xi-Ming Yuan (M.D., Ph.D.)
Division of Pathology II, Faculty of Health Sciences, Linköping University, Linköping,
Objective: We earlier proposed that erythrophagocytosis and iron metabolism by
macrophages may contribute to iron-driven oxidative stress in atherogenesis. Recent studies
indicate the macrophage hemoglobin scavenger receptor (HbSR/CD163) is a key molecule in
the process of removing hemoglobin released from ruptured erythrocytes. In this study we
aimed to investigate erythrophagocytosis and its relation to ferritin accumulation in human
atherosclerotic lesions and if CD163 is involved in ferritin induction in human atheroma
lesions. Methods: Normal human arterial segments from mammary (n = 4), carotid (n = 4),
coronary arteries (n = 5), and atherosclerotic coronary (n = 5), and carotid arteries (n = 16)
were collected for Fe histochemistry, immunohistochemistry of CD68, hemoglobin, CD163
and ferritin, and computerized image analysis. Results: The lesion-dependent accumulation of
ferritin and hemoglobin were seen in atherosclerotic carotid and coronary arteries. The
immunoreactivity of hemoglobin was significantly corresponding to the same regions of
ferritin immunoreactivity on serial sections of carotid and coronary atherosclerotic arteries.
The staining intensity of hemoglobin and ferritin also significantly correlated. Hemoglobin
deposition is often associated with micro vessels surrounding the lipid core areas in advanced
lesions, where there was mainly CD68 positive macrophages. CD163 expression was noted in
both early and advanced lesions. Accumulation of tissue iron and ferritin also frequently
occur at vascular rich regions and shoulder areas of the advanced atheroma, both of which
were significantly correlated. Conclusion: Erythrophagocytosis and hemoglobin catabolism
by macrophages contribute to iron deposition and ferritin induction in human atheroma. The
involvement of CD163 during ferritin induction may play an important role in modulating
inflammatory processes in atherogenesis.

                                            - 167 -
Session III: Apoptosis              Poster III, 53

Wei Li (M.D., Ph.D.) and Xi-Ming Yuan (M.D., Ph.D.)
Division of Pathology II, Faculty of Health Sciences,
Linköping University, Linköping, Sweden
Objective: The natural resistance associated macrophage proteins (NARMPs) have been
reported to modulate inflammatory reactions. Nramp 1 together with Nramp 2 is not only
responsible for intracellular divalent metal transport, but may also determine macrophage
functions in inflammation. In the present study we study if NARMP 1 is involved in
macrophage apoptosis in vivo and in vitro in atherogenesis. Methods: Arterial segments of
watanabe heritable hyperlipidaemic (WHHL) rabbits were used for examination of Nramp1
mRNA by in-situ PCR, macrophage immunohistochemistry, and terminal deoxynucleotidyl
transferse-mediated dUTP nick end labeling of apoptotic cells (TUNEL). U937 cells were
treated with desferrioxamine (DFO) or 7bhydroxycholesterol (7b-OH) and then were
examined with in situ RT-PCR to see if there is 7b-OH induced alteration of Nramp1 mRNA.
Results: We find that in the macrophage-rich areas (positive to RMA-11) of the rabbits there
is a lesion-dependent increase in Nramp 1 mRNA, which are often apoptotic macrophages
and positive to TUNEL staining. Interestingly DFO can decrease mRNA level of Nramp 1,
while both iron compounds and 7b-OH dramatically enhance Nramp 1 mRNA, particularly in
apoptotic cells. Conclusion: Enhanced expression of NARMP 1 in macrophage regions of
atherosclerotic lesions may be associated with oxidized lipids-induced apoptosis. This finding
may reveal a new novel function of Nramp 1 in atherogenesis.

                                            - 168 -
Session III: Apoptosis               Poster III, 54

The adenosine-activated signal transduction pathway conferring protection against
ischemia-reperfusion injury in primary rat neuronal cultures

Esther Zoref-Shani, Ayelet Reshef, Noam Di Capua and Oded Sperling

Department of Clinical Biochemistry, Sackler Faculty of Medicine, Tel-Aviv University, Tel-
Aviv 69978, Israel. E-mail:

Ischemic preconditioning in the neurons activates an endogenous protective mechanism,
aimed at conferring transient resistance to a subsequent, otherwise lethal ischemia-reperfusion
insult. This mechanism is initiated through the release of adenosine and several other
diffusible stress signaling molecules, which induce protective signal transduction pathways.
We studied the neuronal adenosine- activated signal transduction pathway. Employing an
experimental model of primary rat neuronal cultures subjected to chemical ischemia
(iodoacetic acid)-reperfusion, we demonstrated that binding of adenosine to its receptors,
activation of PKC-epsilon and opening of KATP channels are vital steps in the neuronal
adenosine-induced protective signal transduction pathway. Activation of each of these steps
conferred to the neurons a relatively wide time window of protection (lasting several hours up
to several days), representing probably combination of mechanisms leading to early
(immediate) and delayed windows of protection, and/or “memory” of the protection signal by
one or more of the participating components. Inhibition of the binding of adenosine, of the
activation of PKC-epsilon and of the opening of the KATP channels, each abrogated the
protection induced by R-PIA, an agonist for A1 adenosine receptors. Opening of the KATP
channels appears to play a major and mandatory role in the protective mechanism, since
blocking the opening of the channels abolished the protection conferred by R-PIA and by
activation of PKC, throughout the entire duration of the acquired time window of protection.
Apparently, opening of the KATP channels is downstream to activation of PKC-epsilon.

                                            - 169 -
Session IV : Cancer specific signaling
pathways as therapeutic targets

                 - 170 -
Session IV : Cancer specific signaling pathways as therapeutic targets   Poster IV, 1

Diagnostic and therapeutic value of the PDGFR-alpha and -beta expression in KIT-
negative gastrointestinal stromal tumors and other mimicking neoplasms

Bigiani N1, Rossi G1, Sartori G1, Migaldi M1, Valli R1, Bertolini F2, Mucciarini C2, Maiorana
A1, Marchioni A1, Federico M2, Trentini GP1
 Department of Pathologic Anatomy and Legal Medicine, Section of Pathologic Anatomy,
and 2Department of Oncology, University of Modena and Reggio Emilia, Modena, Italy. E-

Gastrointestinal stromal tumors (GIST) are the most common mesenchymal tumors of the
gastrointestinal tract and are characterized by the expression of KIT (CD117), a type III
tyrosine kinase (TK) encoded by the proto-oncogene c-kit. In GIST, KIT is usually
autophosphorylated as a result of a constitutive mutation of c-kit and selectively blocked by
the TK-inhibitor STI571. However, a small number of GIST does not stain for CD117, since
they present a mutually exclusive mutation of plateled-derived growth factor receptor-alpha
(PDGFR-A), another type of III TK inhibited by STI571. Among 118 GIST, we found 16
KIT-negative cases (13.5%). All these 16 cases, 102 KIT-positive GIST and 38 commonly
KIT-negative primary soft-tissue tumors of the gastrointestinal tract mimicking GIST (12
leiomyomas, 15 desmoids, 8 leiomyosarcomas and 3 schwannomas) constituted the present
case series. All cases were investigated by immunohistochemistry (IHC) for PDGFR-A and
–beta (PDGFR-B) (Santa Cruz Biotechnology, Santa Cruz, CA, USA; 1:100 dilution) using
an automated immunostainer (Ventana, Strasbourg, France). All the 102 KIT-positive GIST
resulted unstained for PDGFR-A and –B, as well as the 38 spindle cell lesions that at
morphology mimicked GIST but clinically do not respond to STI571. Conversely, 14 out of
16 KIT-negative GIST showed strong expression only for PDGFR-A, while the remaining
two co-expressed PDGRF-A and –B. Thus, practically speaking, our results first highlight
that PDGFR-A is an helpful marker in discriminating between KIT-negative GIST and other
mimicking tumors. Moreover, we confirm that, even at IHC, GIST are characterized by a
mutually exclusive expression for KIT and PDGFR-A. Of note, 2 out of 16 KIT-negative
GIST were immunoreactive for both PDGFR-A and –B. Since PDGFR may exists as
homodimer (AA or BB) or heterodimer (AB), it could be that some GIST present the PDGFR
in form of heterodimer on cell surface, but whether also PDGFR-B might be mutated clearly
requires further molecular investigations.

                                            - 171 -
Session IV: Cancer specific signaling pathways as therapeutic targets      Poster IV, 2

How Feedback Loops Govern the Dynamics of Mammalian Signaling Cascades

Nils Blüthgen and Hanspeter Herzel

Institute for Theoretical Biology, Humboldt University Berlin and Charite, Invalidenstr. 43,
10115 Berlin, E-mail: and

One of the best studied and characterized signaling cascades in eukaryotic cells is the mitogen
activated protein kinase (MAPK) cascade containing RAS/Raf/Mek/Erk. It is involved in
many central processes such as proliferation and differentiation and can be activated by a
variety of extracellular stimuli. Aberrant activity in this signaling cascade is observed in many
Since the cascade is incorporated into a complex network of other pathways by cross-talk and
since it is also regulated by many feedback loops, the response of the cascade to stimulations
can be highly dynamic and non-intuitive. By mathematical modelling we investigate how
different feedback loops can lead to complex responses even for simple stimuli:
A positive feedback can result in a pronounced all-or-none response and hence accounts for
switch-like functions. Here, a threshold can be established which the signal has to exceed in
order to activate a process. Once this threshold has been passed, a positive feedback loop can
keep the process activated. Thus a positive feedback loop can create a checkpoint.
A negative feedback loop can lead to adaptation and can reset the response if the stimulus still
persists. In case of a very strong feedback the response can be oscillatory.
Therefore, embedded in feedback loops, the MAPK-cascade is capable of performing many
dynamic functions depending on the presence and absence of feedback loops. And indeed
there is experimental evidence that shows the multipotency of the MAPK cascade. We present
relevant previously published results to illustrate these phenomena.
Moreover, we compare the effect of feedback loops on the level of single cells and on the
population level. We argue that some effects, like all-or-none responses, can only be
measured in single-cell experiments like FACS while experiments like Western-blots, which
measure the activity of a cell pool, fail to show these effects.

                                             - 172 -
Session IV:Cancer specific signalling pathways as therapeutic targets        Poster IV, 3


Carla Testorelli°,Isabella Faraoni*,Lorena Rossi* and Enzo Bonmassar*

°Dpt.of Pharmacology,School of Medicine,University of Milan,Italy
*Department of Neuroscience,University of Rome Tor Vergata,Rome,Italy

Telomerase is considered an ideal target for tumor therapy..Oligonucleotides that selectively
inhibit thre enzyme are likely to be used in association with traditional chemotherapy. We
report here the results of preliminary experiments set up in order to evaluate wether the “in
vitro” cytotoxic activity of a panel of chemotheraputic agents could be correlated to the level
of telomerase in cultured tumor cells. To this end , an experimental model consisting in a
telomerase negative VA13 tumor line that uses an alternative way to maintain the length of
telomeres was used,as well as VA13/htert (telomerase negative) and VA13/tel (telomerase
positive) cell lines ,obtained following two successive transfection of the parental line.
Moreover, in order to obtain cell lines expressing different level of telomerase, a number of
clones were derived from VA13/tel line .After detecting the telomerase expression by TRAP
assay,clones A12,A13,C4 and A7 ,showing increasing levels of telomerase,were selected.A
further characterization showed that all lines used in this study shared a similar patternr of cell
cycle (FACS analyses) ,as well as similar clonogenicity.So far this model resulted suitable for
investigating the cytotoxic activity of cisPlatin (cisDP), oxalilplatin (oxDP), carmoustine
(BCNU) and cyclophosphamide as active form (MAF). Cell lines were exposed “in vitro” for
72 hrs to different concentrations of antitumor drugs. The cytotoxicity was evaluated by
MTT assay The results showed an increased chemoresistance to the cytotoxic agents in
telomerase positive VA13/tel cell line and related clones compared to telomerase
negativeVA13 cell line.Namely the difference of IC50 detected for cisDP and MAF in
control and treated cells resulted statistically higher in VA13/tel compared to VA13.
Surprisingly VA13/htert cell lines ,although telomerase negative, resulted more resistant to
the cytotoxic agents than VA13 cell lines.

                                              - 173 -
Session IV : Cancer specific signaling pathways as therapeutic targets    Poster IV, 4

Modulation of VEGF induction pathways by n-3 PUFAs in colon cancer cells cultured
in vitro and transplanted in nude mice.
Gabriella Calviello, 1Fiorella Di Nicuolo, 1Simona Serini, 1Elisabetta Piccioni, 2Nicola
Maggiano, 3Giuseppe Tringali, 3Pierluigi Navarra, 4Franco O. Ranelletti and 1Paola Palozza.
Institute of General Pathology, 2Pathology, 3Pharmacology and 4Histology, Catholic
University. L.go F. Vito, 1. 00168 – Rome, Italy.

n-3 Polyunsaturated fatty acids (PUFAs) have been shown to possess many beneficial effects
in neoplastic pathology, ranging from the decrease of tumor growth, inhibition of metastasis,
reduction of cachexia and attenuation of various side effects of different chemotherapeutic
agents. Moreover, recently it was shown that dietary docosahexaenoic acid (DHA) reduce the
development of microvessels in human mammary tumors growing in nude mice and
eicosapentaenoic acid (EPA) suppress the “tube forming capacity” of endothelial cells
cultured in vitro. Since most colorectal tumors express autonomously the vascular endothelial
growth factor (VEGF) and their prognosis is directly related to the level of this pro-
angiogenic factors and to the tumor microvessel density, in the present study we investigated
the effects of EPA and DHA on the expression of VEGF in human colon cancer cells cultured
in vitro and transplanted in nude mice. We found that both fatty acids reduced the constitutive
VEGF expression in colon cancer cells cultured in vitro and this effect was related to the
inhibition of the COX-2/PGE(2) signalling pathway of VEGF induction, active in colon
cancer cells. The growth of the tumors originated from colon cancer cells inoculated in nude
mice was markedly inhibited by the dietary treatment of animals with EPA or DHA.
Moreover, the reduced expression of VEGF, COX-2 and PGE(2) in the tumors grown in EPA
or DHA treated rats substantiated the hypothesis that these fatty acids may act as anti-
angiogenic factors in colon cancer modulating the COX-2/PGE(2) pathway of VEGF
induction. Overall, the results suggest a possible clinical application of these fatty acids as
anti-angiogenic compounds in colon cancer therapy.

                                            - 174 -
Session IV : Cancer specific signaling pathways as therapeutic targets         Poster IV, 5

Is Notch1 involved in the deregulation of molecular processes in T-ALL?

R.Chiaramonte, V.Cecchinato, A.Basile , E.Calzavara, E.Tassi, G.V.Sherbet* and P.Comi.

Dept. Biomedical Science & Technology, University of Milano, via Fratelli Cervi 93, 20090
Segrate, Milano, Italy. E-mail:
*Dept. Electrical & Electronic Engineering, Merz Court,University of Newcastle upon Tyne,
Newcastle upon Tyne, NE1 6RU [U.K.]. E-mail:

Notch1 is involved in the genesis of T acute lymphoblastic leukemia (T-ALL) carrying the
translocation t(7;9)(q43;q34.3).This leads to the overexpression of a constitutively active portion of
Notch1 which has been proven to induce selectively T-ALL among hematopoietic neoplasms.
Our previous results made evident that, whereas only 1% T-ALL patients has been reported to carry
Notch1 translocated form, Notch1 hyperactive signaling is a common feature of T-ALL if compared
to AML and B-ALL. Notch1 deregulation may be a convergence point of different genetic
alterations leading to the raising of leukemia and in virtue of this possible central role it might
become a potential therapeutic target in T-ALL.
This work aims to find the possible influence of Notch pathway on the cellular processes that may
be altered during leukemogenesis; indeed recent evidences suggests that the improper activation of
the Notch pathway could be involved in maintaining the tumour cells in an undifferentiated state,
promoting self renewal while safeguarding them from apoptosis.
At this purpose Notch1 expression inhibition has been performed by treatment of several T-ALL
cell lines with the HMBA, a hybrid polar compound, and suggests a role for Notch1 signalling in
maintaining a tumoral phenotype by inducing proliferation and suppressing apoptosis. Our attention
has been directed to deepen our knowledge of the effects of Notch1 signalling on molecular
pathways involved in the regulation of cell cycle, differentiation and apoptosis. This study has been
performed on two selected T-ALL cell lines Molt4 and SupT1 (the last carries Notch1 translocated
form). Here we show that Notch1 inhibition by HMBA treatment is accompanied by an increase of
cells in G1/Go phase possibly mediated by a parallel increase of p53 and p21 followed by a down
regulation of proteins involved in the S phase entry such as CDK4 and CycD1. The increase in
apoptosis rate evident in Molt4 cells is accompanied by a down regulation of Bcl2.

                                               - 175 -
Session IV : Cancer specific signaling pathways as therapeutic targets   Poster IV, 6

Quantitative RT-PCR expression of glycosyltransferases in human melanoma cell lines.

Dorota Cio_czyk-Wierzbicka, Marek Bodzioch, Danuta _mudzi_ska,
Aldona Dembi_ska-Kie_, Piotr Laidler.

Many studies suggested the contribution of N-linked oligosaccharides of proteins in cancer
metastasis During carcinogenesis aberrant glycosylation leads to the development of
subpopulations tumor with different adhesion properties. Biosynthesis of carbohydrate
structures is tissue –specific and developmentally regulated by glycosyltransferases like
fucosyl-, sialyl- and N-acetylglucosaminyltransferases..
We analyzed the expression of glycosyltransferases: fucosyltransferase FUT-1 and FUT-4,
sialyltransferase SIAT4C and beta 1,6-N-acetylglucosaminyltransferase V (MGAT-5).
Enzyme expression was determined by quantitative analysis of the RT-PCR real-time (on the
basis of SYBR Green were performed with ABI PRISM 7700 PCR DNA Engine Opticon).
We analyzed the expression of glycosyltransferases in human melanoma cell lines; WM35
from primary tumor site and WM239, WM9, A375 from metastatic sites. Examined cell lines
showed expression of fucosyltransferases, sialyltransferase and
N-acetylglucosaminyltransferase V. The mRNA expression of fucosyltransferases was
significantly higher in melanoma cell lines from metastatic site than primary cell line. The
mRNA expression of FUT-1 was about 100 times higher in melanoma cell line from
metastatic side -A375 (solid tumor) than primary cell line-WM35. The expression of FUT-4
from metastatic side: in WM9 (lymph node) was 80 times higher and, in WM239 (skin) was
37 times higher than primary cell line. In all melanoma cell lines observed very low
expression of MGAT-5 and high expression of SIAT4C.
Higher mRNA expression of fucosyltransferases ( FUT-1) may be an initial event for the
formation of surface structures that facilitate metastases melanoma cell line.

                                            - 176 -
Session IV : Cancer specific signaling pathways as therapeutic targets   Poster IV, 7

Phorbol ester stimulates non-hypoxic induction of a novel HIF-1alpha isoform:
Implication of tumor promotion

Yang-Sook Chun, Kyoung-Hwa Lee, Eunjoo Choi, Soo-Young Bae, Eun-Jin Yeo, L. Eric
Huang, Myung-Suk Kim, Jong-Wan Park

Human Genome Research Institute and Cancer Research Institute, Seoul National University
College of Medicine, 28 Yongon-dong, Chongno-gu, Seoul 110-799, Korea. E-mail :

Hypoxia-inducible factor-1 (HIF-1), composed of HIF-1alpha and HIF-1beta, plays important
roles in tumor promotion. Its activity depends on HIF-1alpha, which is tightly controlled by
the oxygen tension. Besides hypoxia, various non-hypoxic stimuli can stabilize HIF-1alpha in
tumor cells, implying that both hypoxia and non-hypoxic stimuli contribute to over-
expression of HIF-1alpha in tumor. On the other hand, phorbol esters such as PMA are well
known as potent tumor promoters. Here, we have identified a novel HIF-1alpha isoform,
which is regulated primarily by PMA. The variant mRNA lacks exon 11 and produces HIF-
1a785 having C-terminal transcriptional activity. HIF-1a785 is induced markedly by PMA and
heat-shock which is also a HIF-1alpha inducer. It escaped from lysine acetylation due to loss
of Lys532, and was stabilized under normoxic conditions. Its expression was blocked by
reducing agents and a MEK-1 inhibitor, and enhanced by hydrogen peroxide. In addition,
HIF-1a785 over-expression strikingly enhanced tumor growth in vivo. These results suggest
that HIF-1a785 is induced by PMA under normoxic conditions via a redox-dependent MEK-1
pathway, and seems to play important roles in tumor promotion.

                                            - 177 -
Session IV : Cancer specific signalling pathways as therapeutic targets    Poster IV, 8

Design and characterisation of a new class of inhibitors of Ras activation

Sonia Colombo, Francesco. Peri, Renata Tisi, Francesco Nicotra and Enzo Martegani

Dipartimento di Biotecnologie e Bioscienze, Università di Milano-Bicocca, Piazza della
Scienza 2, 20126, Milano, Italy. E-mail:

Normal RAS genes, which are present in several eukaryotes including yeast, are critical
regulators of many physiological functions, including cell growth and proliferation. Ras
proteins are small GTPase which are active in a GTP-bound form and inactive in a GDP-
bound form. This cycling is influenced by different classes of proteins that either positively or
negatively regulate. An high percentage of human tumours contains an altered oncogenic Ras;
therefore inhibitors of GDP-GTP nucleotide exchange of mutated, tumorigenic Ras, or of Ras
mediated signalling, could prevent uncontrolled growth of cancer cells. The aim of this work
is to study a new class of inhibitors of Ras activation (1) and to synthesise new Ras nucleotide
exchange inhibitors, to characterise them in vitro and in vivo and eventually to determine the
structure of the drug-protein complex. The molecules SCH 53870 (1) and other six new
molecules prepared using similar synthetic strategies were characterised in vitro using the
assay reported in (2) with some modifications. This assay measures the extent of exchange of
GDP with a fluorescent analogue of GTP on the Ras proteins. The exchange is activated by
the addition of a purified mammalian GEF (CDC25Mm) (3). Under these conditions both SCH
53870 and all the new inhibitors, although at a different extent, were found to inhibit the
CDC25Mm mediated nucleotide exchange on h-Ras and on yeast Ras2. Two of them inhibited
also the EDTA-induced exchange, suggesting that they primarily interact with the Ras protein
and not with the GEF. In addition we developed new yeast strains useful for a high
throughput screening of potential Ras inhibitors.
1. Taveras A. G. et al. (1997) Bioorganic & Medicinal Chemistry 5, 125-133
Lenzen C. et al. (1995) Methods in Enzymology 255, 95-109
Coccetti et al.(1995) Biochem. Biophys. Res. Commun. 206, 253-259

                                             - 178 -
Session IV: Cancer specific signaling pathways as therapeutic targets    Poster IV, 9

Inhibition of p53 Function Diminishes Androgen Receptor-Mediated Signaling in
Prostate Cancer Cell Lines

Marcus V Cronauer, Wolfgang A Schulz, Rolf Ackermann, and Martin Burchardt

Department of Urology, Heinrich-Heine University, D-40225 Düsseldorf, Germany

Introduction: Current therapy for advanced prostate cancer (PCa) is mainly based on androgen
deprivation, although most patients relapse to androgen-insensitive disease. Several
mechanisms contributing to androgen-independent growth of prostate cancer cells include
alterations in structure or expression of the androgen receptor (AR) or its co-factors. Recent
evidence suggests that the tumor suppressor p53 which is often mutated in advanced PCa is
also directly involved in androgen signaling.
Results: Experiments were performed in 22Rv1 and LNCaP prostate cancer cells which
express both p53 and AR. Overexpression of p53 diminished AR-induced transactivation of
probasin and PSA promoters in both 22Rv1 and LNCaP cells. Similar results were obtained
following induction of endogenous p53 with cisplatin in 22Rv1 cells. Conversely, inhibition
of endogenous p53 by 3 different inhibitors, Pifithrin-1alpha (PFT), an inhibitor of p53-
dependent transactivation; MDM2, a regulator of p53 expression, and a dominant-negative
truncated p53 dramatically reduced androgen-dependent reporter gene transactivation.
Moreover, inactivation of p53 signaling by PFT decreased AR-protein expression in androgen
stimulated 22Rv1 and LNCaP cells.
Conclusions: Overexpression of wild-type p53 decreases AR function. Conversely,
physiological levels of p53 are necessary for AR-signaling and AR-expression. Our findings
suggest a balance of AR and p53 expression during the androgen-dependent growth of
prostate cancer which is obliterated during progression of the disease.

Supported by: Forschungskomission der Medizinischen Fakultät der Heinrich-Heine
Universität and Action Lions - Vaincre le Cancer, Luxembourg.

                                            - 179 -
Session IV : Cancer specific signaling pathways as therapeutic targets   Poster IV, 10

EGFR overexpression in anaplastic thyroid carcinomas

Ensinger Christian1, Moser Patrizia1, Tschoerner Ines1, Schmid Kurt Werner2

Institutes of Pathology, Universities of Innsbruck1, Austria and Essen2, Germany
Muellerstrasse 44, A-6020 Innsbruck, Austria; e-mail:

Anaplastic thyroid carcinoma is one of the most aggressive human malignancies with a
median survival up to 6 months. Such a bad prognosis under the present treatment procedures
suggests a need for novel approaches in the management of this disease. Since some EGFR
inhibitors are now in clinical trials and only few data are available concerning EGFR
expression in anaplastic thyroid carcinomas, we tried to estimate a possible overexpression of
this receptor in a larger tumor series.
Therefore we used 25 anaplastic thyroid carcinomas, including 3 poorly differentiated thyroid
carcinomas carcinomas with large anaplastic tumor parts. Immunohistochemistry was
performed with a mouse monoclonal antibody directed against EGFR (EGFR pharmDXTM kit:
DAKO, Denmark).
The incubated tumor cells revealed a distinct membraneous staining pattern. Differences were
found in the amount of positivly stained tumor areas, partly reflecting overexpression only in
isolated cell clusters. In a few tumor cells an additionally cytoplasmic reaction could be
observed. The anaplastic carcinomas presented with 5/22 (22,7%) without EGFR reaction,
10/22 (45,5%) with reactivity and 7/22 (31,8%) with overexpression of the receptor. In
contrast, the poorly differentiated carcinomas developing to anaplastic carcinomas revealed in
all 3/3 (100%) cases an EGFR overexpression.
Due to our results and an increasing number of antagonists and inhibitors, for at least one
third of all anaplastic thyroid carcinomas EGFR seems to be a promising agent for the
targeted molecular therapy of these extraordinary aggressive tumors.

                                            - 180 -
Session IV: Cancer specific signalling pathways as therapeutic targets    Poster IV, 11

Ethanol signalling pathway in breast cancer: involvement of aromatase

Nicolas Etique1, Dominique Chardard1, Amand Chesnel1, Jean-Louis Merlin2, Stephane
Flament1 and Isabelle Grillier-Vuissoz1.

1 EA 3442 Génétique, Signalisation, Différenciation. Université Henri Poincaré – Nancy I.
Faculté des Sciences – BP 239 - 54506 Vandoeuvre-lès-Nancy Cedex France. E-mail : and 2 Centre Alexis Vautrin, Laboratoire de
Recherche en Oncologie, 54511 Vandoeuvre-lès-Nancy Cedex, France.

It is well documented that alcohol is associated with an increased risk factor for breast
carcinogenesis although the underlying mechanisms are not clearly understood. It has been
reported that in vitro, the culture of estrogen receptor (ER) expressing breast cancer cells in
ethanol containing medium was associated with an increase in the proliferation rate, in the
ERa content as well as in ER transcriptional activity. Since these changes are not observed in
ER negative breast cancer cells, and since alcohol intake has been associated to an increased
level of circulating estrogens, we have postulated that aromatase expression could be
increased following ethanol exposure. The results of our studies show a 1.3-fold increase in
cell proliferation after 6-days of culture of MCF-7 cells in the presence of 0.1% ethanol. This
enhanced proliferation is confirmed by the use of clonogenic assays which show a 1.5-fold
increase in clonal growth in the presence of 0.1 % ethanol. No statistically significant changes
are observed in the presence of higher ethanol concentration (0.3%). After a 6-day exposure
to 0.1 % ethanol, RT-PCR analyses reveal a 1.7-fold increase in ERa mRNA, whereas
western blot analyses show a significant 3.3-fold increase in ERa content. At the same stage,
RT-PCR studies demonstrate a 2.4-fold increase in aromatase mRNA level which is
confirmed at the protein level by western-blots performed after immuno-precipitation of the
enzyme. Taken together, these results are in agreement with the involvement of ER signalling
in ethanol-induced stimulation of breast cancer cells proliferation. Experiments are currently
performed to further analyse ethanol-signalling pathway in breast cancer cells.

                                             - 181 -
Session IV : Cancer specific signaling pathways as therapeutic targets   Poster IV, 12

Constitutive STAT3 activity in colorectal cancer coincides with loss of cell attachment,
enhanced proliferation and autocrine cytokine production

Florian Corvinus1, Svetlana A. Tsareva1, Carina Orth1, Richard Moriggl2, Edith Pfitzner3,
Lukas A. Huber4, Kurt Zatloukal5 and Karlheinz Friedrich1
  Friedrich-Schiller University Jena Medical School, Institute of Biochemistry I, Nonnenplan
2, D-07743 Jena, Germany, E-Mail:
  Institute of Molecular Pathology (IMP), Dr. Bohrgasse 7, A-1030 Vienna, Austria
   Georg-Speyer-Haus, Institute for Biomedical Research, D-60596 Frankfurt am Main,
  University of Innsbruck, Department of Histology and Molecular Cell Biology, A-6020
Innsbruck, Austria
  University of Graz, Institute of Pathology, A-8036 Graz, Austria

We have shown for the first time that aberrant activity of signal transducer and activator of
transcription STAT3 actively contributes to colorectal carcinoma and, thus, is a potential
therapeutical target for this malignancy. Constitutive STAT3 activity was found abundant in
de-differentiated cancer cells and infiltrating lymphocytes of CRC samples, but not in non-
neoplastic colon epithelium. Unexpectedly, cell lines derived from malignant colorectal
tumors lost persistent STAT3 activity in culture. However, xenograft tumors arising from
implantation of colon carcinoma cells into nude mice showed restoration of STAT3 activity.
Evidence was obtained for the involvement of an extracellular stimulus within the tumor
microenvironment as a trigger for STAT activation. In vitro, growth conditions forcing
detachment of CRC cancer cells induced the secretion of a STAT3-activating autocrine factor
whose abundance correlated with the level of STAT3 phosphorylation. STAT3 activity was
shown to functionally contribute to CRC development by abrogation of cell contact inhibition
as well as by promotion of cell proliferation and invasiveness.

                                            - 182 -
Session IV: Cancer specific signaling pathways as therapeutic targets    Poster IV, 13

Role of Nrf2 in 15-deoxy-_12,14-prostaglandin J2 mediated-induction of
heme oxygenase-1 in human breast cancer cells.

EunHee Kim, Hye-Kyung Na, and Young-Joon Surh.
Seoul National University, Seoul, Korea

15-Deoxy-delta12,14-prostaglandin J2 (15d-PGJ2), an endogenous peroxisome proliferator-
activated receptor-gamma ligand, has been shown to inhibit cell growth and induces apoptosis
in various types of malignant cells. However, this cyclopentenone prostaglandin can also
exert a cytoprotective effects at relatively low concentrations. 15d-PGJ2 has been shown to
induce the expression of heme oxygenase-1 (HO-1), a ubiquitous stress-responsive enzyme
that catalyzes oxidative cleavage of heme to form iron, carbon monoxide, and biliverdin.
However, the molecular mechanism underlying 15d-PGJ2-mediated HO-1 induction and its
physiological/toxicological implications remain poorly understood. In the present work, we
examined the signaling cascade regulating the HO-1 induction by 15d-PGJ2. Treatment of
human breast cancer (MCF-7) cells with 15d-PGJ2 resulted in a concentration- and time-
dependent increase in the expression and activity of HO-1. Reverse transcription-polymerase
chain reaction (RT-PCR) revealed that the expression of ho-1 mRNA was induced in 12 h by
15d-PGJ2 at a concentration as low as 3 _M. 15d-PGJ2 induced phosphorylation of Akt/PKB
in 6 h. Up-regulation of HO-1 by 15d-PGJ2 was abolished in cells pretreated with the
phosphatidyl inositol-3 kinase (PI3K) inhibitor LY294002. Additionally, ERK1/2 and JNK
were activated by 15d-PGJ2 after 12 h. Nrf2, a basic-leucine zipper transcription factor, has
been reported to positively regulate the SRE-mediated expression of various cellular
defensive/antioxidant enzyme genes including ho-1. The nuclear tranlocation of Nrf2 peaked
at 3 h after the 15d-PGJ2 treatment, leading to increased binding to the stress-response
element (SRE) as assessed by the electrophorectic mobility gel shift assay. A dominant-
negative mutant form of Nrf2 abrogated 15d-PGJ2-induced HO-1 expression. The promoter
activity conferred by the luciferase reporter constructs containing SRE sequences was up-
regulated by 15d-PGJ2 in transiently transfected MCF-7 cells. Mutation of the GC box in the
SRE core sequence obliterated the 15d-PGJ2-dependent transcriptional activity of Nrf2. These
findings, taken together, suggest that induction of HO-1 expression by 15d-PGJ2 is mediated
by the transcription factor Nrf2.

                                            - 183 -
Session IV: Cancer specific signaling pathways as therapeutic targets     Poster IV, 14

Enhancement of Tumor Response by Farnesyltransferase Inhibitor (FTI) in C3H/HeJ

Jiyoung Kim, Jinsil Seong and Sung Hee Kim.

Department of Radiation Oncology, Brain Korea 21 Project for Medical Science, Yonsei
University Medical College, Seoul 120-752, Korea, E-mail:

Farnesyltransferase inhibitor (FTI) acts on ras, which ultimately can enhance radiosensitivity.
The objective of this study was to explore whether FTI could potentiate the antitumor efficacy
of radiation in vivo, particularly in radioresistant hepatocarcinoma (HCa-I), syngeneic to
C3H/HeJ. The presence of ras mutations was examined by polymerase chain reaction (PCR)
and DNA sequencing. Mice bearing HCa-I were treated with a novel FTI, LB42907, and 25
Gy radiation. LB42907 was administered 60 mg/kg twice daily per oral for 30 days. The
expression of regulating molecules was analyzed by western blotting for p53, p21WAF1/CIP1, and
Bcl-2 family. In HCa-I, ras mutations were not detected. Downregulation of ras by LB42907
was most prominent at 4 h after the treatment. In tumor growth delay assay, LB42907
increased the effect of tumor radioresponse with an enhancement factor of 1.32. Combination
treatment increased radiation induced apoptosis; peak apoptotic index was 3.6% in radiation
alone and drug alone, respectively, and 7.1% in the combination treatment group. Analysis of
apoptosis regulating molecules with western blotting showed upregulation of p53 and
p21WAF1/CIP1 in the combination treatment group comparing to those in either treatment alone
group, but Bcl-2 family remained less changed. FTI in combination with radiation therapy
may have potential benefit in cancer treatment even if there are no ras mutations. FTI could
inhibit ras activity, but may also affect any protein that requires farnesylation for activity.

                                            - 184 -
Session IV: Cancer specific signaling pathways as therapeutic targets   Poster IV, 15

Expression of the insulin-like growth factor-I receptor and proapoptotic Bax and Bak
proteins in human colorectal cancer.

Mariusz Koda, Joanna Reszec, Mariola Sulkowska, Luiza Kanczuga-Koda, Stanislaw

Department of Clinical and General Pathology, Medical University of Bialystok,
Waszyngtona 13, 15-269 Bialystok, Poland. E-mail:

Deregulation of growth factor receptors as well as apoptosis contribute to tumor development.
Insulin-like growth factor-I receptor (IGF-IR) and proapoptotic proteins Bax and Bak play an
important role in tumorigenesis of human colorectal cancer. However, data on IGF-IR
expression in human tissues and its relationship with proapoptotic proteins, are limited. A
total number of 155 cases of colorectal cancer were examined by immunohistochemistry,
using the avidin-biotin-peroxidase method, for the expression of IGF-IR, Bax and Bak
proteins. The results were correlated with selected clinicopathological features. The
expression of IGF-IR, Bax and Bak was noted in 48%, 54% and 47% of the cases,
respectively. Increased IGF-IR expression we observed in colorectal adenocarcinoma
compared with adenocarcinoma mucinosum (p=0.053). We observed the positive correlation
between IGF-IR and Bax (p<0.002, r=0.302), between IGF-IR and Bak (p<0.0001, r=0.407)
as well as between Bax and Bak (p<0.0001, r=0.474). No relationship was noted between
IGF-IR expression and tumor grade, stage and lymph node status. We observed negative
association between Bax, Bak and tumor grade (p<0.01, p<0.003, respectively), but no
relationship between Bax, Bak expression and tumor stage and between Bax, Bak and lymph
node status. Our results demonstrate that, in addition to overexpressed IGF-IR, there are
relationships between IGF-IR and proapoptotic proteins in colorectal carcinomas. The results
suggest that signaling pathways could become beneficial because of development of effective
anticancer therapies.

                                            - 185 -
Session IV : Cancer specific signaling pathways as therapeutic targets    Poster IV, 16

Overexpression of dominant-negative phosphotyrosine phosphatase PTP1B mediates
resistance of Bcr-Abl positive cells to the Abl tyrosine kinase inhibitor STI571

Noriko Koyama, Steffen Koschmieder, Sandhya Tyagi, Silke Myloch, Wolf K Hofmann,
Dieter Hoelzer, and Oliver G Ottmann

Division of Hematology, Department of Internal Medicine III, Johann Wolfgang Goethe
University, 60590 Frankfurt, Germany

The enhanced activation of Abl protein tyrosine kinase of the fusion protein Bcr-Abl is
implicated in the pathogenesis of chronic myelogenous leukaemia (CML). Protein tyrosine
phosphatase-1B (PTP1B) recognizes Bcr-Abl and is thought to be an endgeneous negative
regulator for Bcr-Abl. Since STI571, a potent inhibitor of the Bcr-Abl, requires inactivation of
Bcr-Abl protein tyrosine kinase prior to binding, we hypothesized that PTP1B can modulate
STI57-induced apoptosis, cell growth, and differentiation of Bcr-Abl positive leukaemia cells.
To assess the effects of PTP1B, a full-length (PTP1B), a substrate-trapping dominant-
negative mutant (DN-PTP1B) or a control vector (mock) was stably transfected in Bcr-Abl
positive K562 cells. Cell cycle analysis and annexin-V analysis showed that DN-PTP1B
transfected cells were resistant to STI571-induced cell death as compared with mock or
PTP1B transfected cells. This was associated with reduced processing of caspase-9, -3, and -8
in DN-PTP1B transfected cells. Further, STI571 induced growth inhibition and
phosphorylation of ERK1/2 and increased the level of glycophorin-A, a marker of erythroid
cells, in both of PTP1B and mock transfected cells. Both cells differentiated morphologically
into erythroid precursor cells after expose to STI571. In contrast, STI571 had no significant
effect on proliferation of the DN-PTP1B transfected cells. These cells failed to differentiate
morphologically in the presence of STI571.
Our results indicate that PTP1B could confer resistance to STI571 in Bcr-Abl positive
leukemia cells.

                                             - 186 -
Session IV : Cancer specific signaling pathways as therapeutic targets     Poster IV, 17

Molecular modeling of cell cycle regulation in the signal transduction database

Mathias Krull1, Claudia Choi1, Susanne Pistor1, Anatolij P. Potapov1, Nico Voss1, Edgar
BIOBASE GmbH, Halchtersche Str. 33, D-38304 Wolfenbüttel, Germany; 2Department of
Bioinformatics, UKG, University of Göttingen, Goldschmidtstr. 1, D-37077 Göttingen,
Germany; E-mail:

The passage of eukaryotic cells through the cell cycle is a highly regulated process that is
under balanced control of growth-inducing and growth-inhibiting factors and their
corresponding pathways. Checkpoints at different phases of the cycle ensure that the genetic
material is passed on without damage. If not, signals are emitted that stop the progression
through the cycle and, in case of severe damage, induce apoptosis. Nevertheless, mutations in
one of several cell cycle regulators such as pRb, p53, ATM, Chk1/2 can lead to abnormal cell
proliferation and are a central part of cancer development in many cases.
To help understanding the role of each component in the cell cycle control system and the
complex regulatory network they constitute, a database is needed that stores the current
knowledge and adds facilities for network visualization and analysis. The signal transduction
database TRANSPATH“ contains information on molecules, genes and their reactions in
vertebrate cells. Networks of the manually curated data can be visualized as a bipartite graph
on-the-fly. The integration with TRANSFAC®, a database about transcription factors and their
DNA binding sites, makes it possible to obtain complete signaling pathways from ligand to
target genes and their products, which may be involved in regulatory processes themselves.
Currently, the database contains about 1,000 reactions that belong to the cell cycle regulation
network. By using integrated bioinformatic tools (PathwayBuilderTM, ArrayAnalyzerTM) this
data can be analyzed to identify potential therapeutic targets in cancerous cells with respect to
pathway cross-talk and possible side effects.

                                             - 187 -
Session IV : Cancer specific signaling pathways as therapeutic targets   Poster IV, 18

The role of the SHH/PTCH/SMO pathway in oncogenesis

Sonja Levanat, Ph.D; Arijana Komar, B.Sc; Vesna Musani, B.Sc
Division of Molecular Medicine, Rudjer Boskovic Institute, Zagreb, Croatia

Constitutive inactivation of the Hedgehog signaling pathway can result in tumorigenesis.
Inactivating mutations in the tumor suppressor gene patched (PTCH) have been shown in
basal cell carcinoma, medulloblastoma, ovarian fibromas and dermoids.
We propose to compare the expression of target genes of hedgehog signaling in basal cell
carcinoma with ovarian fibromas and ovarian dermoids to identify tissue specific and
common patterns characteristic of tumorigenesis by deregulated Hedgehog signaling.
The levels of expression of particular genes of the pathway we determined from extracted
RNA which was previously reverally transcribed into cDNA from basal cell carcinomas,
ovarian fibromas and dermoids. The levels of expression of some of the pathway genes was
enhanced in all cases. This results indicate that the SHH/PTCH/SMO pathway is aberrant in
these pathological conditions.
Extensions of such analyses will lead to new insights into the mechanism of tumour
development and the identification of potential targets and therapies.

                                            - 188 -
Session IV : Cancer specific signaling pathways as therapeutic targets                               Poster IV, 19

Effects of Genistein on EGFR mediated signaling transduction pathway in human
ovarian carcinoma cells lines SKOV3 and its xenograft in nude mice

LI Yu , MI Can , WANG Chaojie , YANG Zhengqin
Keywords: Genistein ·ovarian carcinoma· proliferation· signaling transduction pathway

Department of pathology, Chongqing Medical University, Chongqing 400016, China ( Li Y,
Mi C, Wang CJ, Yang ZQ)
Correspondence to: Dr. Li Yu, Department of pathology, Chongqing Medical University,
Chongqing 400016, China ( Tel: 86-023-68485789. E-mail: )

Objective To investigate the effects of Genistein on EGFR mediated signaling transduction
pathway and explore the mechanisms of proliferation inhibition in human ovarian carcinoma
cell line SKOV3 and its xenograft in nude mice.
Methods The expression of C-erbB2 protein was determined using immunocytochemistry.
The expressions of C-jun and C-fos protein were determined using western blotting method.
The expression of c-erbB2, c-raf-1, c-jun and c-fos mRNA were tested by reverse
transcription-polymerase chain reaction (RT-PCR).
Results The expression of c-erbB2, c-raf-1 and its downstream gene c-jun and c-fos were
decreased at mRNA level in 20µmol/L Genistein group. The expression of C-erbB2 protein
were decreased, its average optical density were decreased after treated SKOV3 with
20µmol/L Genistein for 48h(P<0.05).Western blotting demonstrated the expression of C-jun
and C-fos protein were decreased gradually after treated with 20µmol/L Genistein for
Conclusion Genistein could down-regulate the expression of two key genes---c-erbB2 and c-
raf-1 at protein and mRNA level in EGFR mediated signaling transduction pathway, and
down-regulate the expression of its downstream nuclear transcription factors c-jun and c-fos
at protein and mRNA level. It is suggested that interference the expression of main signal
molecule in EGFR mediated signal transduction system by Genistein may be its molecular
foundation of proliferation inhibition in ovarian carcinoma.


                                                             - 189 -
Session IV : Cancer specific signaling pathways as therapeutic targets   Poster IV, 20

The TP53 gene alterations in Tunisian breast cancer


The integrity of the human genome is essential for the normal growth and cell development.
The TP53 is at the same time the sensor of almost the whole cell stress signals and the
guardian of the genome integrity.
In this report, we studied the TP53 genetic alteration in 30 cases of Tunisian breast
carcinoma. We used the SNP genotype identification method at codon 72 polymorphism to
analyse the loss of heterozygoty (L.O.H.). The p53 protein status was analysed by IHC using
3 antibodies recognized the p53 protein (DO7, 1801 for the wild and mutant type, 240 for the
mutant one) and 3 antibodies recognized the TP53 target gene products: MDM2, BAX, BCL2
proteins. The direct sequencing of the exon4-exon9 region encompassing the domain binding
was used as tool for looking for the eventual associated mutation either with L.O.H. or with
The p53 protein status.
Our results showed that, three cases with L.O.H. were associated with point mutation, one as
hotspot in the codon 175 (Arg ‡ His) and two were cited for the first time: Asp ‡ Glu in the
codon 256 and Asn ‡ Ser in the codon 268. A mutation at codon 151 (Pro ‡ Ser) was found
in a case showing a particular profile in IHC and which the L.O.H. was non informative.
Although the sampling number was limited, the results might suggest that the TP53 gene
altered expression was mainly involved in the breast tumour process genesis than TP53 point
mutation, which highlights the potential role of p53-protein interactions in the p53

                                            - 190 -
Session IV : Cancer specific signaling pathways as therapeutic targets    Poster IV, 21

Identification of new targets for melanoma therapy by cDNA microarray technology

Alireza Mirmohammadsadegh, Annett Baer, Sandeep Nambiar, Walter Bardenheuer and
Ulrich Remigius Hengge

Department of Dermatology, Heinrich-Heine University Duesseldorf, D-40225 Duesseldorf

One important application of DNA microarray technology is the simultaneous analysis of
gene expression of different mRNA species. Comparison of mRNA patterns of diseased and
healthy tissue may help to understand the pathogenesis of a given disorder. Identified
dysregulated genes for example in cancer tissue may function as new molecular markers for
diagnosis or prognosis or may ideally serve as a new target for therapy.
Using membrane cDNA arrays technology we analyse gene expression of human melanoma,
one of the most agrgressive types of cancer with a high metastaic potential and with markedly
increased incidence worldwide.
To account for the heterogeneity of different tumors we collected total RNA from 10 different
melanoma metastases from 10 different patients. The RNA pattern of the pooled specimens
was compared to gene expression of primary human melanocytes using 5 different cDNA
membrane arrays from ClontechR with appromximaly 1940 genes.
An abudance of genes were dysregulated (up/down) which involved for examlpe in apoptosis
like FAST, TFAR 15, GRB10, or in angiogenesis like VEGF.
Here, we focus our description on the JAK/STAT signaling pathway which is involved in cell
proliferation, cell differentiation and apoptosis. Stat-induced inhibitor 2 (SSI-2) is
upregulated in human primary melanocytes in comparison to human malignant melanoma
metastases. SSI-2 is a member of a protein family, that inhibits cytokine responses and
activation of ´signal transducer and activator of transcription´ (STAT).
Therefore, we investigated the protein expression of Stat3 and Stat5 via Western blot analysis
and observed a constitutive level of these proteins in primary melanocytes as well as in
malignant melanoma metastasis. In contrast, the phospho-Stat3 and phospho-Stat5 was
weakly expressed in primary melanocytes and upregulated in melanoma metastases. The
upregulation of SSI-2 in primary melanocytes seemed to regulate cytokine signal transduction
and Stat3, Stat5 activation negatively. This leads to normal proliferation and differentation of
melanocytes. In contrast to this lack of SSI-2 and the upregulation of phospho-Stat3, phosph-
Stat5 in melanoma metastses seem to be involved in dysegulation of proliferation. Therefore,
targeting of Stat3, (e.g. by dominant negative Stat3ß) and Stat5 signaling may provide a
potential therapeutic strategy in malignant melanoma.

                                             - 191 -
Session IV : Cancer specific signaling pathways as therapeutic targets     Poster IV, 22

Spontaneous apoptosis induction following wild type p53 gene transfer using
photochemical internalisation in cells bearing p53 mutation or deletion.

Alioune Ndoye 1, Jean-Louis Merlin 1, Agnès Leroux 1, Gilles Dolivet 1, Patrick Erbacher 2,
Jean-Paul Behr 3, Kristian Berg 4, François Guillemin 1.
 Centre Alexis Vautrin, Vandoeuvre-les-Nancy, France; 2PolyPlus Transfection, Strasbourg,
France; 3Faculté de Pharmacie CNRS UMR 7514, Illkirch, France; 4Norwegian Radium
Hospital, Oslo, Norway.

This study evaluates spontaneous apoptosis induction after external wild-type (wt) p53
nonviral gene transfer using polyethylenimine derivatives (PEI) alone or coupled with
photochemical internalisation (PCI), in two human cancer cell lines bearing p53-mutation or
deletion and consequently unable to initiate apoptosis. PCI is a recent technology for site-
specific delivery of macromolecules such as DNA. This method is based on membrane
photosensitisation of cells treated with Lumitrans® and exposed to Lumisource® light (PCI
Biotech, Norway) leading to massive release of DNA complexes into the cytosol. The
influence of coupling PCI with linear 22kDa (PEI-22k) and tetraglucosylated (PEI-Glu4) PEI
derivatives (PolyPlus Transfection, France) on apoptosis induction following wt-type p53
gene transfer was investigated.
When PCI was coupled with PEI, transgene expression was detected in approximately 50% of
the cells and 48 h after transfection subsequent recovery and increase in p53 mRNA and
protein expression was observed.
Spontaneous induction of apoptosis following wt-p53 gene transfer was investigated
qualitatively using Hoescht 33342 / propidium iodide double labelling fluorescence
microscopy. The apoptotic cell fraction was then using annexin V-FITC / propidium iodide
double labelling flow cytometry. Following wt-p53 gene transfer, chromatin condensation and
nuclear fragmentation patterns, illustrating spontaneous apoptosis induction, were observed in
both cell lines. Quantitatively, PCI was found to significantly potentiate apoptosis induction
in the two cell lines with 2 to 3-fold-increase in apoptotic cell population. Apoptosis induction
was similar to that achieved using conventional 5-fluorouracil chemotherapy at the 50%
growth inhibiting concentration in wt-p53 cells. The present results demonstrate the strong
potency and the benefits of using PCI for wt-p53 gene therapy and warrant further in vivo
investigation. Supported by the French “Ligue contre le Cancer, Comités lorrains”.

                                             - 192 -
Session IV: Cancer specific signalling pathways as therapeutic targets    Poster IV, 23

Oxytocin and vasopressin mitogen signaling in human small-cell lung carcinoma

Christel Péqueux1, Brendan P. Keegan2, Marie-Thèrèse Hagelstein1, Vincent Geenen1,
William G. North2 and Jean-Jacques Legros1.
Neuroendocrine Unit, Centre of Immunology, Pathology Institute, CHU-B23, University of
Liège, B-4000 Liège, Belgium, e-mail:; 2Département of Physiology,
Dartmouth Medical School, Lebanon, N-H, 03756 USA.

Small cell lung carcinoma (SCLC) is an aggressive form of lung cancer and presently
accounts for 20-25% of all lung cancer cases. It is characterised by ectopic secretion of
various neuropeptides. Among these neuropeptides, oxytocin (OT) as well as vasopressin
(VP), two neurohypophysial hormones, are synthesised and secreted by these tumours.
Moreover, the expression of the oxytocin receptor (OTR) has been demonstrated in SCLC cell
lines that also express the V1aR, V1bR/V3R and V2R vasopressin receptors. Both hormones,
from concentrations as low as 10-9M, increase SCLC cell growth. When cells are incubated
with a mix of OT (10-9M) and an OT antagonist (OVTA, 10-9M), the mitogenic effect of OT is
inhibited. OVTA (10-9M) used alone induces a decrease of SCLC cell growth. The objective
of the present study was to characterize intracellular signaling events after OT and VP binding
to their receptors in SCLC cells. Agonists for OTR ([Thr4, Gly7]OT) and V1aR (F180) were
able to elicit increases in cytosolic Ca++, demonstrating the functionality of these receptors.
OT and VP were also shown to increase cytosolic Ca++ levels, and this effect could be blocked
using an OTR antagonist (OVTA) or a V1aR antagonist (SR49059) respectively. There was
no activation of the cAMP pathway detected after VP, dDAVP (V2R agonist), or OT
treatment. Stimulation of SCLC cells with OT and VP led to an increase of ERK1/2
phosphorylation, which peaked at 5 min, and of the phosphorylation of its downstream target
p90RSK. These data prompt us to conclude that OT- and VP-induced mitogen effects on SCLC
are mediated respectively by the OTR and the V1aR and involve the phosphorylation of
ERK1/2 and p90RSK.

                                            - 193 -
Session IV : Cancer specific signaling pathways as therapeutic targets   Poster IV, 24

Stromal cell-derived factor-1a stimulates ovarian cancer cell growth through the
activation of Erk1/2 and Egf receptor.

Carola Porcile, Adriana Bajetto, Simone Barbero, Paolo Pirani, and Gennaro Schettini.
Pharmacology and Neuroscience DOBiG, University of Genova and IST c/o CBA. Largo R.
Benzi, 10 -16132- Genova, Italy. E-mail:

Ovarian cancer (OC) is the leading cause of death in gynecologic diseases and is one
epithelial malignancy in which there is evidence for a complex cytokine and chemokine
network. A number of autocrine and paracrine cytokine loops in the biology of this tumor has
been described. Chemokines are a family of proteins important in the immune surveillance,
that play also an important role in tumor progression influencing tumor cell proliferation,
regulating the angiogenic/angiostatic process, directing cell migration and metastasis, and
finally regulating the immune cells recruitment into the tumor mass. We previously
demonstrated that in astrocytes and glioblastoma cells, which express both the chemokine
receptors CXCR4 and the chemokine stromal cell-derived factor-1 (SDF1), SDF-1a induces
cell proliferation through ERK1/2 activation supporting an important role of chemokines in
the growth of the tumor cells in vitro.
We investigated the role of the CXCR4 activation in human ovarian cancer showing that both
CXCR4 and SDF-1 mRNAs are expressed in OC cell lines. We demonstrate that SDF-1a
induces a dose-dependent proliferation in OC cells through the activation of ERK1/2 and Akt.
Our results further indicate, for the first time, that the CXCR4 activation induces EGFR
phosphorylation that was in turn linked to the downstream intracellular kinases activation,
ERK1/2 and Akt, and the SDF-1a induced mitogenic activity. In addition, we provide
evidence for Src involvement in the CXCR4-EGFR transactivation. These results suggest a
possible important “cross-talk” between CXCR4 and EGF receptor intracellular pathways that
may link signals of cell proliferation in ovarian cancer.

                                            - 194 -
Session IV : Cancer specific signaling pathways as therapeutic targets    Poster IV, 25

Site-specific expression of Polycomb-group genes in clinically defined primary nodal and
cutaneous large B cell lymphomas

Frank M. Raaphorst1, Danny Dukers1, Arie P. Otte2, Rein Willemze3, and Chris J.L.M. Meijer1
  VU Medical Center, Dept. Pathology, Amsterdam, The Netherlands. Email: /
  Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, The
Netherlands. Email:
  Leiden University Medical Center, Dept. Dermatology, Leiden, The Netherlands. Email:

Polycomb-group (PcG) genes preserve cell identity by gene silencing, and contribute to
regulation of embryogenesis, the cell cycle, and lymphopoiesis. Experimental model systems
have indicated that altered PcG expression is strongly linked to malignant transformation. We
studied human PcG expression in clinically defined subclasses of large B cell lymphomas
(LBCL), and found that primary nodal LBCL, and secondary cutaneous deposits from such
lymphomas, abnormally express the BMI-1, RING1, and HPH1 PcG genes in cycling
neoplastic cells. By contrast, tumor cells in primary cutaneous LBCL lacked BMI-1
expression, whereas RING1 was variably detected. Lack of BMI-1 expression was
characteristic for primary cutaneous LBCL, because other primary extranodal LBCL
originating from brain, testes and stomach, were BMI-1 positive. Expression of HPH1 was
rarely detected in primary cutaneous LBCL of the head or trunk and abundant in primary
cutaneous LBCL of the legs, which fits well with its earlier recognition as a distinct clinical
pathologic entity with different clinical behavior. We conclude that clinically defined sub-
classes of primary LBCL display site-specific abnormal expression patterns of PcG genes of
the HPC-HPH/PRC1 PcG complex. Some of these patterns (such as the expression profile of
BMI-1) may be diagnostically relevant. We propose that distinct expression profiles of PcG
genes results in abnormal formation of HPC-HPH/PRC1 PcG complexes, and that this
contributes to lymphomagenesis and different clinical behavior of clinically defined LBCL.

                                            - 195 -
Session IV : Cancer specific signaling pathways as therapeutic targets     Poster IV, 26

Modulation of the B cell signalling network in TGF-b induced immune homeostasis

Jurgen Roes, B. Ken Choi and Balthazar Cazac

University College London, Department of Immunology and Molecular Pathology, Windeyer
Building, 46 Cleveland Street, London W1T 4JF,

By conditional mutagenesis (cre/loxP), we inactivated the TGF-beta receptor (TbR)
selectively in B cells and demonstrated the essential role in the negative control of B cell
activation and homeostasis in vivo. To establish whether the well known anti-proliferative
effects of TGF-b on lymphocytes alone account for this inhibition, we determined TbR
dependent transcriptome changes by comparative gene expression profiling using TbR-
deficient primary B cells as a negative control. Considering 6500 known genes, the analysis
revealed a comprehensive, coherent transcriptome shift indicating profound redirection of B
cell responsiveness, due to suppression of activating pathways and concurrent enhancement of
survival and homeostatic signals. Induction of inhibitors of antigen receptor (Ship-1, CD72),
Jak/Stat and Toll-like receptor (SOCS1, 3) signaling was complemented by induction of
antiproliferative transcription factors. Conversely, induction of G protein-coupled receptors
such as CXCR4 and agonists mediating Ca2+ flux (ITPR2) indicated enhancement of the
Ca2+ storage/ release system and chemotactic responses. Suppression of proapoptotic genes
suggested support of cell survival. Confirming the shift in B cell responsiveness, antigen-
receptor-mediated activation of Syk and phospholipase C-gamma2, as well as Stat6
phosphorylation, is inhibited, whereas chemotaxis, Ca2+ release, and cell survival are
enhanced in TGF-beta sensitive B cells. The data provide a molecular basis for TbetaR-
mediated inhibition of B cell responses and indicate that TbetaR maintains homeostasis by
delivering a coherent instructive signal that not only inhibits but also redirects responsiveness
to micro-environmental cues. By extending these studies to characterise genes and ORFs with
unkonwn function we are working towards an integrated picture of homeostatic gene
expression changes in B cells and hope to identify novel regulatory genes controlling
lymphocyte homeostasis.

                                             - 196 -
Session IV : Cancer specific signaling pathways as therapeutic targets    Poster IV, 27

Evaluation of microsatellite instability and cell-cycle regulatory proteins expression in
young patients with superficial papillary bladder cancer

G. Sartori1, N. Bigiani1, L. Garagnani1, G. Rossi1, M. Migaldi1, C. Curatola1, B. Faraglia2, A.
Sgambato2, GP. Trentini1
  Department of Pathologic Anatomy and Legal Medicine, Section of Pathologic Anatomy,
University of Modena and Reggio Emilia, Modena, Italy
  Centro Ricerche Oncologiche Giovanni XXIII, Institute of Pathology, Catholic University,
Rome, Italy. E-mail:

Microsatellite alteration (MA) has been widely used as a tool for the detection of loss of
heterozygosity (LOH) and microsatellite instability (MI) in bladder cancer, as well as the
expression of cell-cycle regulatory proteins for cancer prognostic evaluation. However, only
limited data are available on the genetic characteristics and the pathway of cell-cycle
regulatory markers of this tumor in young patients. Twenty-seven superficial (17 pTa, 10
pT1) papillary bladder cancers arisen in young patients (less than 45 years of age) were
examined for the immunohistochemical expression of cyclinD1, Ki67, p27Kip1 and p53 and
analyzed by 19 microsatellite markers (at 9 chromosomes). Five micron thick paraffin-
embedded sections were obtained for immunohistochemistry (ABC complex technique) and
microdissection of neoplastic and control normal tissues (lymphocytes). Microsatellites were
examined by PCR amplification and analyzed using a ABI-Prism 310 automatic sequencer
(Applied Biosystem, Monza, Italy). MI at one or more loci was detected in all cases (100%),
with a mean of 5.2 MI for each case (range, 1 to 9). LOH was observed in 22 (81%) tumors,
with a mean of 2 LOH for case (range, 0 to 4). At immunohistochemistry, only cyclinD1
showed a significantly higher expression in tumors associated with frequent MA (p=0.04, chi
square test). Surprisingly, the occurrence of MA was more frequent in low grade (p=0.0197)
and low stage (p=0.0127) tumors at statistical analysis (chi-square test). Of note, in a
univariate analysis MA (especially at D4S243) predicted a longer disease-free survival than
the low tumor stage, and this finding was in agreement with nuclear cyclinD1 expression,
resulting lower in high-grade tumors. In conclusion, we demonstrated that MA, in form of MI
or LOH, is a frequent and early event in young patients with bladder cancer and might be
related with a more favorable prognosis, as happen in other solid tumors such as colorectal

                                            - 197 -
Session IV: Cancer specific signaling pathways as therapeutic targets   Poster IV, 28

Radiation-induced alteration of pain-related signals in an animal model with bone
invasion from cancer
Jinsil Seong, 2Hee Chul Park, 1Jiyoung Kim, 3Un Jung Kim, 3Bae Whan Lee
Dept of Radiation Oncology, 3Yonsei Medical Research Center, Brain Korea 21 Project for
Medicince,       Yonsei      University,        Hallym     University,  Korea,       E-

While radiotherapy is highly effective in relieving bone pain from cancer invasion, its
mechanism remains unclear. To explore the mechanism, we have developed an animal model
with bone pain from cancer invasion. Using that animal model system, radiation-induced pain
response and pain-related signals in the spinal cord were analyzed. Hind paw model of cancer
pain was developed by transplanting hepatocellular carcinoma, HCa-1, into the periosteal
membrane of foot dorsum in C3H/HeJ mice. Bone invasion from HCa-1 was confirmed by
histopathological examinations. We also measured the development of pain-associated
behaviors. In the histopathological examinations, bone invasion from HCa-1 was seen from
day 7 and evident at day 14 after transplantation. Measurable pain-associated behaviors were
developed from day 7. In this model, mice given radiation (25 Gy) on tumors showed
decrease of objective level of pain with a higher threshold to mechanical stimulation than in
control from day 3 after irradiation. After irradiation of tumor, significant decrease in the
expression of CGRP was shown in the spinal cord, while neither substance P nor c-fos
showed any alteration. In conclusion, we developed a novel model of bone pain from cancer
invasion, confirmed by histopathological examination and measurable pain-associated
behaviors. Radiotherapy decreased the objective level of pain and the underlying mechanism
involved the alteration of pain-related host signal, CGRP, in the spinal cord.

                                            - 198 -
Session IV: Cancer specific signaling pathways as therapeutic targets    Poster IV, 29

Alterations in expression of MYC and MYC-associated genes in prostate cancer
identified by high throughput analysis (DNA microarrays): Coagulation factors.

Rona J. Strawbridge1, Jacques Lapointe3,4,5, James D. Brooks5, Patrick O. Brown4, Jonathon R.
Pollack3, Anders Zetterberg1, Sten Nilsson1, Chunde Li2*

Department of Oncology and Pathology1 and Surgical Science2, Karolinska Institute,
Stockholm, Sweden; Department of Pathology3, Biochemistry4, Urology5, Stanford
University, Stanford, California, USA. *to whom correspondence should be addressed at:
Urological Research Laboratory at CCK/Department of Surgical Science, CCK R8:04,
Karolinska Hospital, SE-171 76 Stockholm, Sweden. Email:

Gene expression profiling has highlighted gene expression variations between normal and
cancerous prostate samples. One area of interest is the expression variation of the MYC gene
and factors involved in the coagulation cascade. c-Myc is crucial for tumourigenesis and often
deregulated in prostate cancer. c-Myc induces the expression of proangiogenic factors such
as Vascular Endothelial Growth Factor and eukaryotic Initiation Factor 4E, and further tips
the equilibrium towards neovascularization by repression of Thrombospondin 1. Expression
of 18,180 cDNA sequences in 42 prostate samples (including normal tissue, primary cancer
and lymph node metastases) were measured by DNA micro arrays and were visualized by
Cluster and Treeview platforms to enable comparison of gene expression patterns between
normal and malignant samples. Up regulation of the MYC gene was observed in 85% primary
tumours as well as a subset of lymph node metastases. Other genes with altered expression
profiles included up regulation of coagulation Factor V, Tissue Factor and the Factor II
receptor. Genes which function as tumour suppressors such as Thrombospondin 1 and TP53
showed a reduction in transcription. Of important note, the human prostate cancer in our study
shares a significant gene expression signature with mouse MYC-driven prostate cancer in a
recent publication. Data mining (Bioinformatics) based on previous publications as well as
our results gave a list of MYC-associated genes, including the above coagulation factors, and
formed a MYC-driven oncogenic pathway in prostate cancer. This is of interest for two
reasons, firstly it has implications for progression of tumours with respect to angiogenesis,
and secondly, elucidation of important alterations in the cellular signaling connected to the
coagulation pathway has potential for developing novel therapeutic approaches to tumour
progression control, in particular targeting tumour angiogenesis.

                                            - 199 -
Session IV : Cancer specific signaling pathways as therapeutic targets   Poster IV, 30

Effects of the overexpression of the transcription factor Ets2 and a dominant negative
allele in the osteoblast differentiation of C2C12 cells mediated by BMP-2

Antonio R. G. Susperregui, 1Raquel Fernández-Lloris, 1Cristina Gamell-Fullà, 2Kim E.

Boulukos, 1Francesc Viñals and 1Francesc Ventura.

  Departament de Ciències Fisiològiques II, Campus de Bellvitge, Universitat de Barcelona,
Feixa Llarga s/n, 08907 L´Hospitalet de Llobregat, Spain. E-mail:
  CNRS-UMR 6548 Batiment Sciences Naturelles, Universite de Nice, Parc Valrose 06108
Nice cedex 2, France

Transcription factors of the ets family have been implicated in important processes of
organogenesis and differentiation. As Ets2 is the only member of this family that is highly
expressed during osteogenesis of the osteoblat-like MC3T3-E1 cell line, the purpose of this
study was to define the role of this transcription factor in the osteoblast differentiation of
C2C12 cells stimulated with BMP-2. We have analyzed the signal transduction in response to
BMP-2 as well as differents osteoblasts markers in C2C12 cells that overexpress Ets2 and
cells that express a dominant negative allele. A slight acceleration of the differentiation
process has been observed when overexpressing Ets2. Expression of the dominant negative
allele does not stop or delay the osteoblast phenotype althought disregulate the alkaline
phosphatase activity.

                                            - 200 -
Session IV: Cancer specific signaling pathways as therapeutic targets     Poster IV, 31

The Sonic Hedgehog (Shh) signal transduction pathway regulates gastric epithelial cells
differentiation and proliferation.

Andrea Todisco, Vinzenz Stepan, Saravanan Ramamoorthy, Hildegard Nitsche, Jung Park.

Division of Gastroenterology, University of Michigan Medical School, Ann Arbor, MI, USA.

The Sonic Hedgehog (Shh) signal transduction pathway regulates cellular growth and
differentiation. Shh binds to the receptor protein patched (Ptc), relieving the tonic inhibitory
effect of Ptc on the transmembrane protein Smoothened (Smo). This allows the activation of
downstream targets through the Gli family of transcription factors. We previously reported
that prolonged incubation (> 72 h) of gastric epithelial cells expressing markers of parietal
cell differentiation with EGF, induces cellular proliferation, morphological transformation
and inhibition of H+/K+-ATP-ase gene expression, a marker of parietal cell differentiation.
We explored the role of Shh/Ptc/Smo into these actions of EGF. EGF suppressed the
expression of Shh and Ptc but not that of Smo in the cells after 72 h of incubation. Shh, Ptc
and Smo expression were measured by western blots and immunohistochemistry. We
examined the consequences of loss of expression of the inhibitory protein Ptc. Cyclopamine
an inhibitor of Smo signaling, reversed EGF-induced morphological transformation and it
inhibited EGF induction of both cyclin D1 and TGF-alfa expression in the cells.
Cyclopamine also inhibited MAPK activation and it blocked EGF-stimulated cellular
proliferation. To explore if deregulated Smo signaling affects TGF-alfa expression, we co-
transfected the cells with a luciferase reporter plasmid containing 2813 bases of the TGF-alfa
gene promoter together with a dominant negative Gli2 gene. EGF induced TGF-alfa-
luciferase activity two-fold, and dominant negative Gli2 inhibited this effect. Since Akt
promotes parietal cell differentiation and maturation, we examined the effect of Akt on Shh
and Ptc expression in the EGF-treated cells. Adenovirus mediated gene transfer of
constitutively active Akt into the EGF-treated, morphologically transformed cells, restored the
expression of both Shh and Ptc. Accordingly, prolonged exposure of gastric epithelial cells
expressing markers of parietal cells differentiation to EGF leads to cell proliferation and
dedifferentiation through loss of Ptc and deregulated signaling from Smo. Akt is a crucial
switch for the activation of programs of parietal cell differentiation.

                                             - 201 -
Session IV : Cancer specific signaling pathways as therapeutic targets    Poster IV, 32

STAT3 drives proliferation, malignant transformation and invasive behavior of colon
carcinoma cells

Svetlana A. Tsareva1, Florian Corvinus1, Bernd Wiederanders1, Anja Meissner1,             Edith
Pfitzner2, Richard Moriggl3 and Karlheinz Friedrich1
  Friedrich-Schiller University Jena Medical School, Institute of Biochemistry I, Nonnenplan
2, D-07743 Jena, Germany, E-Mail:
   Georg-Speyer-Haus, Institute for Biomedical Research, D-60596 Frankfurt am Main,
  Institute of Molecular Pathology (IMP), Dr. Bohrgasse 7, A-1030 Vienna, Austria

Signal Transducer and Activator of Transcription STAT3 is a regulator of fundamental
cellular processes such as proliferation, differentiation and cell death. It possesses oncogenic
properties and was found aberrantly active in various malignant tumors. We have observed
constitutive STAT3 activity in more than 90% of surgical tumor specimens from colorectal
carcinoma patients. Potential functional involvement of STAT3 in the malignant
transformation of colon epithelium cells was addressed using colon carcinoma cell lines
transfected with different derivatives of STAT3. Heterologous overexpression of STAT3 and
activation via the interleukin-6 receptor resulted in accelerated cell proliferation, enhanced
colony formation in soft agar and elevated invasiveness in vitro. All these effect were
pronounced upon transfection with a constitutively active STAT3 mutant and suppressed by a
dominant negative STAT3 variant. To obtain information on mechanistical connections
between STAT3 activity and invasiveness, we analyzed mRNA from cancer samples by
means of cDNA arrays for protease expression patterns. A striking correlation of strong
STAT3 activity with upregulation of matrix metalloproteinases MMP-1, MMP-3, MMP-7 and
MMP-9 was observed. STAT3 was able to positively regulate transcription from the MMP-1
promoter in colon carcinoma cell lines as shown by luciferase reporter gene experiments.
These results support the notion of dysregulated STAT3 as a contributor to malignancy in
colorectal cancer.

                                             - 202 -
Session IV: Cancer specific signaling pathways as therapeutic targets     Poster IV, 33

ERK MAPK regulation of tumour cell motility

Emmanuel Vial1, Erik Sahai2, Jacques Pouysségur1 and Chris J. Marshall2.
  Institute of Signaling Developmental Biology and Cancer, UMR CNRS 6543, 33 avenue de
Valombrose, 06189 Nice, France. Email:
  Institute of Cancer Research, London, UK, SW3 6JB.

Point mutations that activate Ras GTPases or their effector BRAF are present in more than 30
% of human cancers, making the ERK MAPK signalling pathway one of the most commonly
activated in cancer. In addition to its role in tumour proliferation and survival, an increasing
number of studies indicate that the ERK MAPK pathway might be involved in the invasive
progression of cancer and the metastatic process. One key element in the invasive and
metastatic progression is the acquisition by carcinoma cells of cell motility. We have analysed
the contribution of the ERK MAPK pathway to carcinoma cell motility. Using siRNA and
pharmacological agents we have shown that the ERK activity is required for colon carcinoma
cell motility in 2D and invasion in 3D. We show that ERK promotes a mesenchymal type of
cell movement through the regulation of membrane protrusive activity. We demonstrate that
ERK regulate the activity/expression of membrane associated receptors (beta1 integrin,
uPAR) leading to the up-regulation of Rac1 activity and the down-regulation of RhoA activity
necessary for the lamella extension and the formation of protrusions. Our results reveal a
mechanism by which the ERK MAPK signaling pathway downstream of Ras and BRAF
oncogenes might participate in tumour invasion and metastasis.

This work is supported by Cancer Research UK, CNRS, ARC and Marie Curie fellowships.

                                             - 203 -
Session V: Chemopreventive agents

                   - 204 -
Session V: Chemopreventive agents                      Poster V, 1

Prevention of Photosensitization-Induced oxidative Stress on Antioxidant Status of
Human Erythrocytes by Natural Antioxidant Substance

Mosaad Abou-Seif

Faculty of Science, Mansoura University; Al-Gomhoria, El-Mansoura

Generation of reactive oxygen species (ROS) by photosensitization is the corner stone of
photodynamic therapy of tumors. Cell damage may be mediated by free radical species and
lipid peroxidation of their membranes. Reactive oxygen species bring about a wide variety of
biological effects in the living system. Therefore, the effects of oxygen free radicals (.OH and
O2.-) photogenerated by the novel photosenstizer m-chloroperbenzoic acid (m-CPBA) on
antioxidant status of human erythrocytes were studied. The biological toxicity of reactive
oxygen species on human erythrocytes was evident by increased osmotic fragility,
spherocytosis and haemoloysis. The haemolysis was increased in concentration and time
dependent manner. The lipid peroxidation product thiobarbituric acid reactive substance
(TBARS) and depletion of erythrocyte levels of glutathione (GSH) and ascorbic acid (AsA),
as well, as the activities of ceruloplasmin, lactate dehydrogernase and gluathione reductase in
m-CPBA-photosensitized RBC indicating increased oxidative stress. In addition, the
antioxidant enzyme activities such as superoxide dismutase, catalase, glutathione peroxidase
and glucose-6-phosphate dehydrogenase were elevated in m-CPBA-treated erythrocytes, were
considered other markers of oxidative stress. The level of nitric oxide was elevated, while the
activity of arginase was decreased in m-CPBA-treated RBC. These effects were blunted by
gallic acid which has superoxide dismutase-like and catalase-like activities. These results
suggested that the oxygen free radicals photogenerated by m-CPBA may disturb the
antioxidants and cause an oxidative haemolysis of membrane lipids of human erythrocytes.
The presence of gallic acid ameliorates the oxidative stress of photogenerated oxygen free
radicals. Furthermore, this study provides an investigational promising data for photodynamic
therapy by using safe fluorescent lamp and novel photosensitizer m –CPBA and gallic acid of
antioxidant capacity.

Keywords: Antioxidants, gallic acid as antioxidant substance, photosensitization, erythroctes,
m-chloroperbenzoic acid, fluorescent lamp.

                                             - 205 -
Session V: Chemopreventive agents                      Poster V, 2

c-Jun protein expression did not correlate with anti-proliferative effects of arachidonic
acid metabolism inhibitors in immortalised keratinocytes HaCaT

Zdenek Andrysik1, Karel Soucek1,2, Jiri Pachernik1, Viktor Horvath1, Jirina Hofmanova1,
Alois Kozubik1
 Institute of Biophysics, Academy of Sciences of the Czech Republic, Kralovopolska 135,
612 65 Brno, Czech Republic, 2Department of Internal Medicine, Division of Nephrology,
One Shields Avenue, TB 136, University of California, Davis, California 95616 USA, E-mail:

The transcription factor AP-1 has been postulated to be involved in mechanisms modulating
cellular proliferation and differentiation, as well as cell death. c-Jun protein, the most potent
transactivator in a group of proteins forming the AP-1 transcription factor, is actively
involved in proliferation control. Both arachidonic acid (AA) itself and its metabolites can
activate c-Jun and AP-1. The aim of our study was to investigate the role of AA in stimulation
of c-Jun expression after release from contact inhibition and induction of proliferation.
We showed that selected inhibitors of AA metabolism (MK-886, eicosatetraynoic acid and
esculetin) did not decrease c-Jun expression. However nordihydroguaiaretic acid (NDGA)
was identified as a strong c-Jun inducer and inhibitor of JunB expression. AA was found to
slightly stimulate c-Jun expression, however, NDGA did not induce AA release from HaCaT
cells. Moreover, an increased c-Jun expression induced by NDGA occured simultaneously
with suppression of AP-1 activity. To explain the elevated c-Jun level after NDGA treatment,
we analysed oxidative status of HaCaT cells. NDGA induced a massive intracellular
production of reactive oxygen species (ROS) and glutathione depletion. These effects could
be reversed by using ROS scavengers, such as N-acetylcysteine and ascorbic acid.
Based on the fact that inhibitors of AA did generally not affect c-Jun level we concluded that
AA and its metabolites play only a minor role in stimulation of c-Jun expression after release
of HaCaT cells from contact inhibition. We showed that effect of NDGA on c-Jun expression
is mediated through changes in oxidative metabolism. The elevated c-Jun expression after
NDGA treatment is in contrast to suppression of AP-1 transcription factor activity.
Financial support: FRVS 2003/532 and Grant Agency of the Czech Rep. 524/03/0766.

                                             - 206 -
Session V: Chemopreventive agents                     Poster V, 3

alpha-Methylene-gamma-butyrolactone derivatives block TNF-alpha-induced ICAM-1
gene expression, monocyte adhesion and tumor cell Invasion by targeting IkB kinase

Wei-Chien Huang, Shu-Ting Chan, Tzu-Lin Yang, Cherng-Chyi Tzeng, and Ching-Chow

Department of Pharmacology, College of Medicine, National Taiwan University, Taipei,
10018, Taiwan. E-mail:

        The transcription factor NF-kB is a regulator related to cellular inflammation, immune
responses and carcinogenesis. Therefore, components of the NF-kB-activating singnaling
pathways are frequent targets for anti-inflammatory and anti-cancer agents. In this study,
CYL-19s and CYL-26z, two synthetic alpha-methylene-gamma-butyrolactone derivatives,
were shown to inhibit TNF-alpha-induced ICAM-1 expression in human A549 alveolar
epithelial cells and adhesion of U937 cells to these cells. RT-PCR analysis also demonstrated
their inhibitory effects on TNF-alpha-induced ICAM-1 mRNA expression. ICAM-1 and NF-
kB-dependent promoter activities induced by TNF-alpha were attenuated by CYL-19s and
CYL-26z. Overexpression of wild-type NF-kB-inducing kinase (NIK)- and IkB kinase beta
(IKKbeta)-induced ICAM-1 promoter activities were also inhibited by both compounds.
Furthermore, CYL-19s and CYL-26z inhibited the TNF-alpha-induced IKK activity,
phosphorylation and degradation of IkBalpha, and NF-_B specific DNA-protein binding
activity. In addition to ICAM-1 expression, CYL-19s and CYL-26z also suppressed TNF-
alpha-induced COX-2 expression in NCI-H292 cells, another alveolar carcinoma cell line. In
Matrigel assays, ICAM-1 and COX-2 expression induced by TNF-alpha elicited A549 and
NCI-H292 cell invasion, respectively, and these effects were inhibited by both compounds.
Taken together, our data demonstrate that CYL-19s and CYL-26z down-regulate TNF-alpha-
induced inflammatory genes expression through suppression of IKK activity and NF-kB
activation. These agents may be effective in anti-inflammatory and anti-cancer therapy.

                                            - 207 -
Session V : Chemopreventive agents                     Poster V, 4

Docosahexaenoic acid from the enriched microalga Crypthecodinium cohnii induces
apoptosis in human breast cancer MCF-7 cells by down-regulating anti-apoptotic Bcl-2
protein expression

Lawrence C.-M. Chiu, Elaine Y.-L. Wong and Vincent E.C. Ooi

Department of Biology, The Chinese University of Hong Kong, Hong Kong, China. E-mail:

        Dietary docosahexaenoic acid (DHA) in fish oil has been shown previously to control
growth and development of different cancers. However, safety issue has been raised
repeatedly about contamination of toxins in fish oil that makes it no longer a clean source of
the fatty acid. We investigated in this study the cell-growth inhibition of DHA from the
enriched alga Crypthecodinium cohnii (ADHA) in human breast cancer MCF-7 cells. ADHA
inhibited the growth of the cancer cells dose-dependently by 16 to 59 percents of the control
level, after incubations with 40 to 160 micromolars of the fatty acid for 72 h. Flow cytometry
analysis shows that ADHA did not significantly affect the cell cycle of MCF-7 cells but
induced sub-G1 cells, or apoptotic cells, by 64 to 171 percents of their controls, after
incubations with 80 micromolars of the fatty acid for 24, 48 and 72 h. Western-blot studies
show that ADHA affected neither cell cycle-related cyclin D1 and Rb nor pro-apoptotic Bax
and Bid protein expressions in the breast cancer cells. However, ADHA induced down-
regulation of anti-apoptotic Bcl-2 protein time-dependently, causing increases of Bax/Bcl-2
ratio by 303 and 386 percents respectively after 48 and 72 h of incubations with the fatty acid.
Results from this study show that the algal DHA retards the growth of MCF-7 cells by
inducing apoptosis, and down-regulation of Bcl-2 protein plays an important function during
the process.

                                             - 208 -
Session V : Chemopreventive agents                    Poster V, 5

Protein Kinase C regulates the activity and stability of serotonin N-acetyltransferase

Bo-Hwa Choi‡, Hee-Don Chae‡, Tae-Ju Park‡, Jisun Oh                 , Jinkyu Lim   , Shin-Sung
Kang    , Hyunjung Ha   , and Kyong-Tai Kim‡

‡Department of Life Science, Division of Molecular and Life Science, Pohang University of
Science and Technology, Pohang, 790-784, Korea, E-mail :,,,, §Department of Animal
Science and Biotechnology, College of Natural Science, Kyungpook National University,
Taegu, 702-701, Korea, E-mail :,,_
Department of Biology, College of Natural Science, Kyungpook National University, Taegu,
702-701, Korea, E-mail :, and       Department of Biochemistry, School
of Life Sciences, Chungbuk National University, Cheonju, Korea, E-mail : .

Effects of protein kinase C on protein stability and activity of rat AANAT were investigated
in in vitro and in vivo. When COS-7 cells tranfected with AANAT cDNA were treated with
phorbol 12-myristate 13-acetate (PMA), both the activity and protein level of AANAT were
increased. These effects of PMA were blocked by GF109203X, a specific inhibitor of PKC.
Moreover, PMA increased the phosphorylation of AANAT and induced the formation of
AANAT/14-3-3z complex. PMA did not affect the basal level of cAMP and did not involve
the potentiation of the cAMP production by forskolin, indicating that PKC-dependent
activation of adenylyl cyclase was excluded in transfected COS-7 cells. To identify which
amino acids were phosphorylated by PKC, several conserved Thr and Ser residues in AANAT
were targeted for site-directed mutagenesis. Mutations on Thr29 and Ser203 prevented the
increase of enzymatic activity and protein level mediated by PMA. Mutations on Thr127 and
Ser192 were not effective in altering the protein level, although it slightly reduced the PMA-
induced increase of enzymatic activity. These data, therefore, suggest that the residues Thr29
and Ser203 are more effective in regulating the stability of AANAT protein by PKC. To
explore the nature of AANAT phosphorylation, purified rat AANAT was subjected to in vitro
PKC kinase assay. PKC directly phosphorylated the rat recombinant AANAT. The resulting
phosphopeptides were identified by mass spectrometric analysis and Western blotting,
indicating that Thr29 is one of target sites for PKC. To confirm the effects of physiological
activation of PKC, rat pineal glands were treated with a 1-adrenergic specific agonist
phenylephrine. Phenylephrine caused phosphorylation of endogenous AANAT and
GF109203X significantly inhibited its phosphorylation. These results suggest that AANAT at
Thr29 was phosphorylated by PKC activation through a1-adrenergic receptor in the rat pineal
glands, and its phosphorylation might contribute the stability and the activity of AANAT.

                                            - 209 -
Session V: Chemopreventive agents                     Poster V, 6


Sivaprakasam Balasubramanian1, Tatiana Efimova1, and Richard L. Eckert1, 2,3,4,5

Departments of Physiology and Biophysics1, Dermatology2, Biochemistry3, Reproductive
Biology4, and Oncology5, Case Western Reserve University School of Medicine, Cleveland,
Ohio 44106-4970.

(-)-Epigallocatechin-3-gallate (EGCG) is an important bioactive constituent of green tea that
efficiently reduces epidermal cancer cell proliferation. This inhibition is associated with a
reduction in activator protein 1 (AP1) transcription factor level and activity. However, its
effects on AP1 function in normal epidermal cells have not been extensively explored. Our
present studies show that EGCG regulates normal keratinocyte function. To understand the
mechanism of action, we examined the effects of EGCG on AP1 factor activity, MAPK signal
transduction, and expression of the AP1 factor-regulated human involucrin (hINV) gene.
EGCG increases hINV promoter activity in a concentration-dependent manner that requires
the presence of an intact hINV promoter AP1 factor binding site. This response appears to be
physiologic, as endogenous hINV gene expression is also increased. Fra-1, Fra-2, fosB, junB,
junD, c-jun and c-fos levels are increased by EGCG treatment, as is AP1 factor binding to
hINV promoter AP1 site. Gel mobility shift studies show that this complex contains Fra-1
and junD. Signal transduction analysis indicates that the EGCG response requires Ras,
MEKK1, MEK3 and p38 kinases. Kinase assays and inhibitor studies suggest that p38d is the
p38 isoform responsible for the regulation. These changes are also associated with a
cessation of cell proliferation and enhanced cornified envelope formation. These studies
show that in normal human keratinocytes EGCG markedly increases, via a MAPK signaling
mechanism, AP1 factor-associated responses.

                                            - 210 -
Session V: Chemopreventive agents                    Poster V, 7

Marine natural products as potential inhibitors of transcription factor NF-kB

Florence J. Folmer*, Marc Diederich^, Marcel Jaspars*.

* Marine Natural Products Chemistry Laboratory, Department of Chemistry, University of
Aberdeen, Aberdeen, AB24 3UE, Scotland. E-mail:

^ Laboratoire de Biologie Moleculaire et Cellulaire du Cancer, Centre Universitaire de
Luxembourg, L-2540 Luxembourg, Luxembourg. E-mail:

Nuclear factor-kB (NF-kB) is a family of transcription factors involved in a wide variety of
diseases including cancer, AIDS, Alzheimer’s disease, asthma, atherosclerosis, and
rheumatoid arthritis, and it has hence become one of the major targets for drug discovery.
During the past twenty-five years, marine natural products chemists have discovered that the
oceans are a very rich source of highly promising bioactive compounds with tremendous
potentials in the field of drug discovery. Sessile soft-bodied marine invertebrates, such as
sponges, that lack obvious physical defense are known to often possess very strong bioactive
metabolites, since they depend merely on chemical defense mechanisms for protection against
their predators. Numerous marine natural products have already been reported as highly
promising anticancer, antibiotic, or anti-inflammatory compounds and several of them are
currently in advanced stages of clinical trials. Recently, marine bacteria have also been
discovered to be very rich in novel highly promising metabolites. Despite the abundance of
marine natural products reported to have pharmaceutical activity, very few marine compounds
have been reported to date as NF-kB inhibitors, hence the strong interest in screening marine
natural products for their NF-kB inhibitory activity in the present study.

                                           - 211 -
Session V: Chemopreventive agents                    Poster V, 8

Antioxidant Signaling Cascades as Molecular Targets for Some Edible Phytochemicals
Against Oxidative and Inflammatory Cell Death
  Jung-Hee Jang, 1Mei-Hua Li, 1So-Young Lim, 1Ji-Woo Kim, 1Chuyue Chen, 2Buxiang Sun
and 1Young-Joon Surh
  College of Pharmacy, Seoul National University, Seoul 151-742, South Korea and 2Amino
Up Chemical Co., Sapporo, Japan

beta-Amyloid (Ab) is considered to be responsible for the formation of senile plagues that
accumulate in the brains of patients with Alzheimer’s disease (AD). There has been
compelling evidence supporting that Ab-induced cytotoxicity is mediated through generation
of reactive oxygen/nitrogen species (ROS/RNS). In this study, we have investigated the
effects of resveratrol, a phytoalexin found in the skin of grapes, L-ergothioneine (EGT), a
naturally occurring antioxidant thiol compound, and grape seed extract on Ab-induced
oxidative/nitrosative cell death in cultured rat pheochromocytoma (PC12) cells. PC12 cells
treated with Ab exhibited increased accumulation of intracellular ROS/RNS and underwent
apoptotic death as determined by characteristic morphological alterations, positive in situ
terminal end-labeling, perturbation of mitochondrial membrane potential, an increase in the
Bax/Bcl-XL ratio, activation of c-Jun N-terminal kinase, elevated caspase-3 activity and the
cleavage of poly(ADP-ribose)polymerase. Resveratrol and grape seed extract attenuated Ab-
induced cytotoxicity, apoptotic features, and intracellular ROS accumulation and increased
cellular GSH pool. EGT pretreatment attenuated Ab-induced cytotoxicity, apoptosis and lipid
peroxidation, which appears to be attributable to its ability to scavenge RNS, particularly
peroxynitrite. In another experiment, we have investigated the molecular mechanisms
underlying inflammatory cell death induced by Ab. Exposure of PC12 cells to Ab resulted in
time-dependent activation of COX-2. Ab-induced cytotoxicity was aggravated by COX-2
products, such as prostaglandin E2 and 15-deoxy-D12,14 prostaglandin J2 (15d-PGJ2). In
contrast, pretreatment of PC12 cells with relatively low concentration (< 10 mM) of 15d-PGJ2
attemuated Ab-induced cell death. At concentrations lower than 10 mM, 15d-PGJ2 induced
transient activation of Akt/protein kinase B as well as ERK1/2 and expression of heme
oxygenase-1 and g-glutamate cysteine ligase, which appears to be mediated by binding of Nrf-
2 to the antioxidant response element located in the promoter region of genes encoding these
antioxidant enzymes.

Key words : apoptosis, beta-amyloid, cyclooxygenase-2, 15-deoxy-D12,14 prostaglandin J2,
ergothioneine, grape seed extract, PC12 cells, reactive oxygen/nitrogen species, resveratrol

                                           - 212 -
Session V: Chemopreventive agents                     Poster V, 9

Effect of Daunomycin and Daunomycin-metal complex (cadmium and nickel) on the
respiratory chain and cell morphology in the yeast Candida utilis

Jacqueline Keyhani(1), Shokoofeh Golkhoo(2), Manijeh Sabokdast(3) and Ezzatollah Keyhani(1,3)
 Lab. for Life Sciences, 19979 Tehran; (2)Azzahra Univ., Fac. of Sci., Tehran;         (3)
Biochem. Biophys., Univ. of Tehran, 13145 Tehran. E-mail:

The importance of electrostatic interactions on changes in DNA morphology such as
deformation, twisting, groove-width variation, bending and condensation, is now well
documented [1]. Binding of metals cations such as Cd2+ and Ni2+ produces DNA structure
breakdown by Fenton reaction, in addition to electrostatic effects. Daunomycin (DAU), an
efficient anthracycline chemotherapeutic agent, intercalates into DNA. In this study, the
effect of DAU and Cd2+ or DAU and Ni2+ combined, as well as that of DAU alone, on the
respiratory chain and the morphology of the yeast Candida utilis was investigated. Cells
grown for 24 h at 28ºC under constant aeration in the presence of increasing concentrations of
either DAU, Cd2+, Ni2+, DAU-Cd2+, or DAU-Ni2+ were analyzed spectrophotometrically for
their cytochrome content and cell yield was measured. Cell growth as well as cytochromes
synthesis were stimulated in up to 10 _g/ml DAU but were progressively inhibited as DAU
concentration increased. In 200 _g/ml DAU, cell growth was reduced by 50% while the
amount (ng/mg prot) of cytochromes c oxidase, b and c were reduced respectively to 11%,
26% and 36% of the control. This effect was greatly enhanced when cells were exposed to
DAU-Cd2+ or DAU-Ni2+ complexes. In the presence of only 20 _g/ml DAU with 0.06 mM
Cd2+, cytochrome c oxidase, b, and c were reduced respectively to 6%, 20% and 26% of the
control; 20 _g/ml DAU with 1 mM Ni2+ reduced cytochrome c oxidase to 29%.
Morphological studies showed that DAU brought various changes to the uniformly elliptic
form of normal yeast cells, ranging from elongated cells to triangular cells and cell
aggregation and disruption. DAU-Cd2+ or DAU-Ni2+ produced a large and centralized
vacuole within cells, cell fragmentation and the formation of microcells. Data showed that
the morphological alterations due to DAU alone were also greatly enhanced by Cd2+ and Ni2+.
[1] L. McFail-Isom et al. (1999) Current Opinion in Structural Biology 9:298-304.

                                            - 213 -
Session V: Chemopreventive agents                     Poster V, 10

Hypoxia/anoxia as signaling for increased alcohol dehydrogenase activity in saffron
(Crocus sativus L.) corm

Ezzatollah Keyhani

Laboratory for Life Sciences, 19979 Tehran, and Institute of Biochemistry and Biophysics,
University of Tehran, 13145 Tehran, Iran. E-mail: and

Saffron has been shown a good therapeutic and preventive agent for various cancers; thus a
better knowledge of the plant metabolic pathways is mandatory. Flooding or transfer of
plants from aerobiosis to hypoxia/anoxia (H/A) produces a switch from aerobic respiration to
glycolysis with concomitant induction of alcohol dehydrogenase (ADH). Briefly: flooding à
O2 depletion à cytochrome c oxidase blocked à ATP production inhibited à new gene
transcription à glycolysis enhancement à production of pyruvate, lactic acid, NADH à
stimulation of ADH production and activity. However, so far the effect of H/A on corms or on
saffron remain unreported. H/A was produced in saffron (Crocus sativus L.) corms that were
either dormant (no root or shoot) or that had been cultivated for 3 days (15-mm roots and 40-
mm shoots), by flooding them. Flooded corms were withdrawn after 1, 2, 3, 4, 8 and 14 days
in H/A. They were depleted from their roots and shoots (when present), then washed and
homogenized in phosphate buffer. After 2 cycles of centrifugation at 3,000 g and 35,000 g,
respectively, extracts were obtained and used for enzymatic measurements. ADH as well as
L-lactate dehydrogenase (LDH) activity was measured spectrophotometrically. In dormant
corms LDH activity was not significantly affected by H/A compared to the control (0.6 u/mg
prot). But for cultured corms the activity dropped to less than half (0.24 u/mg prot) after 14
days H/A. ADH activity remained roughly similar to the control (0.008 u/mg prot) in dormant
corms except that it doubled after 14 days H/A. In contrast, corms that were cultivated for 3
days prior to H/A showed a 4-times increase in ADH activity after one to ten days of H/A
(0.034 u/mg prot); a new surge (6 times the control) of ADH activity was found after 14 days
in H/A. Thus while H/A did not greatly affect LDH activity, it was a signal for an increase in
ADH activity that was dependent on the development stage of the corm. The effect of H/A on
shoot and root development will be reported.

                                            - 214 -
Session V: Chemopreventive agents                      Poster V, 11

Multiple-interaction between doxorubicin and yeast Candida utilis leading to apoptosis

Ezzatollah Keyhani and Jacqueline Keyhani

Laboratory for Life Sciences, 19979 Tehran, Iran. E-mail :

Doxorubicin ((DOX, adriamycin) has proven an extremely valuable tool in various cancer
chemotherapies. However, its clinical use is limited by dose-dependent cardiotoxicity.
Ceramide generation via plasma membrane, mitochondrial impairement, free radical
formation, DNA dysfunction and other metabolic alterations have been proposed to account
for DOX cardiotoxicity. In this research the effect of DOX on structural and biochemical
parameters in the yeast Candida utilis was investigated. C. utilis was cultured for 16 h in the
presence of 10, 25, 50, 100, 500 and 1000 _g DOX/ml. O2 uptake as measured by Clarke O2
electrode showed a stimulation in up to 100 _g/ml DOX, the maximum being at 25 _g
DOX/ml with 32 ng atom O2 min-110-8 cells compared to 15 ng atom O2 min-110-8 cells for the
control. As DOX concentration increased, O2 uptake decreased, reaching quasi zero in 1000
_g DOX/ml. Isolated mitochondria from cells cultured in 25 _g DOX showed 27%, 33% and
67% increase in oxidation for NADH, succinate and TMPD-ascorbate, respectively.
Substrate oxidation was zero when cells were cultured in 500 _g DOX/ml.
Spectrophotometry of the respiratory chain showed an initial stimulation of all cytochromes
(a, b, aa3) in up to 25 _g DOX/ml; a severe depression of all cytochromes was progressively
reached with increasing DOX concentrations. 3H-Leu and 3H-U incorporation into
mitochondrial proteins and RNAs increased also in up to 25 _g DOX/ml. Thereafter the
incorporation decreased to reach zero in 1000 _g DOK/ml. Thin section electron microscopy
showed no noticeable alteration in the fine structure of C. utilis and its subcellular organelles
in up to 25 _g DOX/ml. But culture in 50 _g DOX/ml led to alterations in the fine structure
of mitochondria, such as formation of ribbon-like mitochondria, very elongated mitochondria.
Image of cell apoptosis was observed for all DOX concentrations (50 to 500 _g/ml) although
with different proportions. Freeze-fracture EM showed that even at very low DOX
concentrations, severe alterations of the plasma membrane characterized by deep depressions
in PF and protuberances at the EF were seen. Data showed that there was a multiple site of
interaction and alteration between DOX and yeast cells which led ultimately to apoptosis.

                                             - 215 -
Session V: Chemopreventive agents                      Poster V, 12

Chlorophyllin derived from traditional Chinese medicine Cansha induces apoptosis in
human breast carcinoma MCF-7 cells by down-regulating Bcl-2 expression

Carrie K.L. Kong, Lawrence C.M. Chiu and Vincent E.C. Ooi

Department of Biology, The Chinese University of Hong Kong, Shatin, Hong Kong SAR, PR
China. Email:

Chlorophyllin is a mixture of sodium-copper derivatives of chlorophyll, which has recently
been shown as a highly effective chemopreventive agent against liver cancer development in
the high-risk population exposed unavoidably to aflatoxin B1 in diet. Chlorophyll metabolites
in the traditional Chinese medicine Cansha, which is the excreta of silkworm Bombyx mori, is
the source of chlorophyllin interested in this study. We have previously reported that Cansha
chlorophyllin (CHL) induces growth inhibition in human breast carcinoma MCF-7 cells after
72-h incubation, with 50-% inhibitory concentration (IC50) as 200 ug/ml. In the present
study, we pursued the mechanisms of apoptosis induction in this breast cancer cell line.
Results from flow cytometry show that CHL induced apoptosis in the breast cancer cells in
both time- and dose-dependent manners. CHL at 400 ug/ml induced sub-G1 cells, or
apoptotic cells, after 24-h incubation. The apoptotic cells increased by 5- to 29-folds of their
control levels after incubations with 200 and 400 ug/ml of CHL for different periods. CHL
induced apoptosis significantly after 72-h incubation, with down-regulation of the anti-
apoptotic bcl-2 expression at both molecular and protein levels. Results from semi-
quantitative RT-PCR and western blot analyses show that CHL treatments at 200 and 400
ug/ml with 72-h incubation reduced the expression of bcl-2 mRNA transcript. At the same
time, bcl-2 protein expression exhibited remarkable inhibitions by 10 % and 16 %,
respectively. The current findings suggest that CHL triggers apoptosis in MCF-7 cells through
down-regulation of the anti-apoptotic bcl-2 expression. To further examine the role of CHL as
a chemopreventive agent against aberrant growth of human breast cancer cells, its effects on
other functional genes and proteins in different apoptotic pathways demand further

                                             - 216 -
Session V: Chemopreventive agents                      Poster V, 13

Effects of oxalate ON IMCD cells, a line of mouse Inner Medullary Collecting Duct cells

Paul D. Maroni, Sweaty Koul, Paramjit S. Chandhoke, Randall B. Meacham and Hari K

Introduction and Objective: Oxalate, a metabolic end product and a major constituent of
majority of renal stones has been shown to be toxic to renal epithelial cells of cortical origin.
However, cells of inner medullary collecting duct (IMCD), which are physiologically exposed
to higher levels of oxalate, have not been evaluated. In the present study, we examined the
effects of oxalate on IMCD cells
Materials and Methods: IMCD cells were maintained in DMEM: F12 media supplemented
with FBS and antibiotics. Cells were exposed to oxalate (0.2-10 mM) for various time points.
Trypan blue exclusion criteria was used to assess membrane integrity, cell morphology was
assessed by H&E staining, while crystal violet staining was used to measure cell density.
Results: Exposure of IMCD cells to oxalate produced time and concentration dependent
changes in the light microscopic appearance of the cells. The long-term exposure to oxalate
resulted in alterations in cell viability, with net cell loss following exposure to high oxalate
concentrations. The effects of oxalate were time and concentration dependent. However no
significant cell damage was observed up to an oxalate concentration of up to 2mM. In
comparison to IMCD cells, LLC-PK1 cells as well as HK2 cells, showed significant toxicity
starting at lower oxalate concentrations (0.4 mM and above). These results demonstrate the
IMCD cells are resistant to oxalate toxicity as compared to renal epithelial cells of cortical
Conclusions: The results provide first direct demonstration of toxic effects of oxalate in
IMCD cells, a line of renal epithelial cells of inner medullary collecting duct; and suggest that
the cells lining the collecting duct are relatively resistant to oxalate toxicity.

                                             - 217 -
Session V: Chemopreventive agents                     Poster V, 14

Different effect of low doses of beta carotene on proliferation of androgen-dependent
and androgen-independent prostate cancer cells.

Piotr Laidler1 , Joanna Duli_ska1, Aldona Dembi_ska-Kie_2, Marek Bodzioch2, Gerd
Schmitz3, Thomas Langmann3
 Institute of Medical Biochemistry and 2Department of Clinical Biochemistry, 1,2 Jagiellonian
University Medical College, Kraków, Poland; 3Department of Clinical Chemistry and
Laboratory Medicine, University of Regensburg, Germany; 1 ul. Kopernika 7, 31-034 Kraków,
Poland; E-mail:

Carotenoids, precursors of various retinals and antioxidants, are recently often recognized as
potential anti-carcinogenic agents. In humans, beta carotene (BC) may also be converted to
retinoic acids such as eg. all trans retinoic acid (ATRA) and 9-cis retinoic acid (RX) that are
important regulators of cell differentiation and proliferation and can also act as potent anti-
tumor factors. It has been proposed that BC may positively regulate the expression of some
apoptotic genes and that way induce apoptosis of treated cells. It was also shown that high
doses of BC (> 30 _M) decrease proliferation of prostate cancer cells in vitro. However, it is
rather doubtful whether such high doses of BC are really accessible in vivo.
We studied the effect of low doses of BC (3, 10 _M) on proliferation (ELISA BrdU and
Crystal Violet Test) and gene expression (Human High-density Microarray), in LNCaP and
PC-3 prostate cancer cell lines.
The applied doses of (BC) as well as ATRA and to lesser extent RX significantly increased
proliferation of LNCaP prostate cancer cells while had very weak or no effect on PC-3
prostate cancer cells. Initial global analysis of expression of genes in both types of prostate
cancer cells treated with 3 and 10 _M BC (Affymetrix HG-U133A) - showed remarkable
differences in number of responsive genes and direction of changes in their expression. In
case of LNCaP the expression of many genes of steroid metabolism was affected while no
such effect were noticed in PC-3, androgen independent cell line. This indicates strong links
between BC and androgen metabolism and supports view on possible coordinate modulation
of expression of important prostate specific proteins and nuclear transcription factors by
growth factors and steroid hormones.
This work was financially supported by grants: DLARFID, QRLT-2001-00183 (EU 5thFP)
and 6 P05A 074 21 (State Committee for Science – KBN, Poland).

                                            - 218 -
Session V: Chemopreventive agents                     Poster V, 15

H-ras induces motility and invasive phenotype through the increase of COX-2
expression and MMP-9 activity via activation of NF-_B in WB-F344 rat liver epithelial

Ki Won Lee1,2, Kyung-Soo Chun2, Jeong-Sang Lee2, Young-Joon Surh2, and
Hyong Joo Lee1
  Department of Food Science and Technology
  College of Pharmacy
 Seoul National University, Seoul 151-742, South Korea

H-ras expression has been suggested as a marker for carcinogenic process. There is
accumulating evidence that cyclooxygenase-2 (COX-2) induction, matrix metalloproteinases
(MMPs) activation, and abnormal cell proliferation are strongly related to carcinogenic
The present study investigated the mechanism of H-ras induced carcinogenesis. H-ras induces
motility and invasive phenotype in WB-F344 rat liver epithelial cells (WB cells). MMP-9
activation and COX-2 expression also were founded in H-ras WB cells. To study the
mechanisms of H-ras induced carcinogenesis, we investigated activation of NF-_B and AP-1.
H-ras induced the activation of NF-_B, but not AP-1, in WB cells. Caffeic acid phenethyl
ester (CAPE), a phenolic NF-_B inhibitor in honey bee, has shown to induce apoptosis,
inhibit COX-2 expression and MMP activity by blocking improper NF-_B DNA binding
activity in H-ras WB cells. Furthermore, it also suppressed the mobility and invasion of H-ras
WB cells. Our results suggest the mobility and invasive phenotype induced by H-ras may be
strongly related to COX-2 induction and MMP activation via the activation of NF-_B. Thus,
NF-_B can consider as a molecular target for cancer chemoprevention.

                                            - 219 -
Session V: Chemopreventive agents                      Poster V, 16

Xanthine oxidase inhibitory activity of North Algeria plant remedies used for gout

K Madani*, M. Chibane , A. Mouhoubi

Biophysics biomathematics Biochemistry and Scientometry laboratory. Faculté des Sciences
de la Nature et de la Vie. Universite de Bejaia. 06000. Bejaia . Algeria.
(*)Email :, fax : 0021334216098

Xanthine oxidase activity was tested from 31 species belonging to 8 families traditionally
used for the treatment of symptoms related to gout by autochthons peoples of north Algeria.
The variation of activity was determined by measuring the increase in absorbance at 295 nm
associated with urate formation. Ninety three percent of the assayed plants were found to have
inhibitory activity at 165mg/ml, with 35% having greater than 50% inhibition. Ficus carica
exhibited the highest activity with an inhibition of 72.88%.
The inverse representation namely Lineweaver-Burk plots showed that the inhibition mode of
the species with the highest activity was a combination of a linear mixed-type, uncompetitive
and incompetitive. Inhibitory activity of the plants correlated significantly with their phenolic
content (r = 0.68 P<0.01) and tannin content (r = 0.83 P<0.001)

Keywords: Xanthine oxidase, inhibitor, Gout, North Africa, Phenols, Medicinal plants

                                             - 220 -
Session V: Chemopreventive agents                    Poster V, 17

Algal docosahexaenoic acid induces apoptosis in human leukemia HL-60 cells by
modulating bax and bcl-2 expressions

K.F. Tong, Lawrence C.M. Chiu, Y.L. Wong, Vincent E.C. Ooi

Department of Biology, The Chinese University of Hong Kong, Hong Kong SAR, PR China.
E - m a i l :                   k i t t o n g @ a l u m n i . c u h k . n e t

Docosahexaenoic acid (DHA; 22:6 omega-3) is a dietary polyunsaturated fatty acid (PUFA),
which is present in high concentration in fish and algal oils. A number of previous studies
have shown that fish-derived DHA has anti-proliferative properties on a variety of cancers.
However, environmental pollution causes contamination and accumulation of toxins in fish
oil makes it no longer a clean and safe source. Cultured microalgae have recently been
regarded as another cleaner and safer source of PUFAs. This study was to investigate the
effect of DHA from the enriched microalga, Crypthecodinium cohnii, on the proliferation of
human leukemia HL-60 cells, and to elucidate its mechanisms. Trypan blue-exclusion
analysis shows that the algal DHA inhibited the cell growth dose-dependently after 72-h
incubations. The 50-% inhibitory concentration (IC50) was estimated as 74 micromolars. The
fatty acid, at 80 micromolars, induced cell death by 9.6 % of the control level, and all the
leukemia cells were killed at 160 micromolars. The algal DHA modulates cell-cycle
progression and induces apoptotic cell death in HL-60 cells. DNA flow cytometry shows that
the fatty acid arrested cells in G1 phase after 24-h incubation, which was accompanied by the
decrease of S phase cells. The algal DHA induced sub-G1 cells, or apoptotic cells, after 24-h
incubation, which were then increased in a time-dependent manner. Immunoblots analysis
further shows that the algal DHA increased pro-apoptotic bax and reduced anti-apoptotic bcl-
2 protein expression in the leukemia cells, suggesting that modulation of these Bcl-2 family
proteins play important functions in the apoptosis induced by this fatty acid.

                                           - 221 -
Session V: Chemopreventive agents                    Poster V, 18

Algal docosahexaenoic acid induces apoptosis in human leukemia HL-60 cells by
modulating bax and bcl-2 expressions

K.F. Tong, Lawrence C.M. Chiu, Y.L. Wong, Vincent E.C. Ooi

Department of Biology, The Chinese University of Hong Kong, Hong Kong SAR, PR China.
E - m a i l :                   k i t t o n g @ a l u m n i . c u h k . n e t

Docosahexaenoic acid (DHA; 22:6 omega-3) is a dietary polyunsaturated fatty acid (PUFA),
which is present in high concentration in fish and algal oils. A number of previous studies
have shown that fish-derived DHA has anti-proliferative properties on a variety of cancers.
However, environmental pollution causes contamination and accumulation of toxins in fish
oil makes it no longer a clean and safe source. Cultured microalgae have recently been
regarded as another cleaner and safer source of PUFAs. This study was to investigate the
effect of DHA from the enriched microalga, Crypthecodinium cohnii, on the proliferation of
human leukemia HL-60 cells, and to elucidate its mechanisms. Trypan blue-exclusion
analysis shows that the algal DHA inhibited the cell growth dose-dependently after 72-h
incubations. The 50-% inhibitory concentration (IC50) was estimated as 74 micromolars. The
fatty acid, at 80 micromolars, induced cell death by 9.6 % of the control level, and all the
leukemia cells were killed at 160 micromolars. The algal DHA modulates cell-cycle
progression and induces apoptotic cell death in HL-60 cells. DNA flow cytometry shows that
the fatty acid arrested cells in G1 phase after 24-h incubation, which was accompanied by the
decrease of S phase cells. The algal DHA induced sub-G1 cells, or apoptotic cells, after 24-h
incubation, which were then increased in a time-dependent manner. Immunoblots analysis
further shows that the algal DHA increased pro-apoptotic bax and reduced anti-apoptotic bcl-
2 protein expression in the leukemia cells, suggesting that modulation of these Bcl-2 family
proteins play important functions in the apoptosis induced by this fatty acid.

                                           - 222 -
Session V: Chemopreventive agents                       Poster V, 19

Chemoprotective effect by total Uncaria tomentosa extract against Mitomycin-C
induced cytotoxicity on bladder transitional carcinoma cells

Tzvetomira A. Tzanova, Niko J. Bembassat, Spiro M. Konstantinov, Margarita H.

Lab. of Experimental Chemotherapy and Molecular Pharmacology, Dept. of Pharmacology
and Toxicology, Dept. of Pharmacognosy, Faculty of Pharmacy, Medical University of Sofia,
2 Dunav str., Sofia 1000, Bulgaria, e-mail:

Uncaria tomentosa, Rubiaceae (Cat’s Claw) has been used in cancer therapy because of its
antiproliferative, antimutagenic, cytotoxic activities. It has also antioxidant and cytoprotective
properties. In our study we investigated if the cytotoxic efficacy of Mitomycin-C might be
influenced through the concomitant application of the total methanolic extract or pentacyclic
oxindole alkaloid fraction from the inner bark of U. tomentosa. Our panel of tumor cell lines
included EJ and 5637 bladder transitional carcinoma cells. The MTT-dye reduction assay was
used (Mosmann, 1983) in order to measure the cytotoxic efficacy of the combinations applied
after 72 h pre-treatement. The concomitant Mitomycin-C treatement with total U. tomentosa
extract led to nearly significant loss of cytotoxic efficacy. 28% increase of cell vitality at 15
microM Mitomycin-C concentration on EJ cells and 21% increase of cell vitality at 30
microM and 60 microM Mitomycin-C concentrations on 5637 cells. The measured GSH
concentration trough the method described by Sedlak et al. showed that after 24 h pre-
incubation with total Cat’s Claw extract a significant increase over the untreated control. Our
data indicate that the total U. tomentosa extract exerts chemoprotective activity on EJ and
5637 cells against the Mitomycin-C cell death. In contrast unlike the pentacyclic oxindole
alkaloid fraction from U. tomentosa showed own concentration dependent cytotoxicity. These
findings could be explained by the presence of different biologically active compounds in the
inner bark of U. tomentosa.

                                              - 223 -
Session V : Chemopreventive agents                    Poster V, 20

Exogenous HIV-1 Vpr protein activates AP-1 and JNK in promonocytic cells: a
potential target for antiretroviral therapy.

Audrey Varin 1, Anne-Zélie Decrion 1, Bernard P. Roques2, Bharat B. Aggarwal3, Georges
Herbein 1
  Department of Virology and Institut d’Etude et de Tansfert de Gènes, Franche-Comté
University, F-25030 Besançon, France; 2Department de Pharmacochimie Moléculaire et
Structurale, U266 INSERM, UMR 8600 CNRS, F-75270 Paris, France ; 3Department of
Bioimmunotherapy, Cytokine Research Section, The University of Texas M.D. Anderson
Cancer Center, Houston, TX 77030, USA. E-mail:
The HIV-1 Vpr accessory gene product is a 14 kDa virion-associated protein which is
involved in cell cycle and apoptosis control, nuclear import of preintegration complexes and
transactivation of cellular and viral promoters. Most of the data obtained so far have been in
experiments with endogenous Vpr protein, so far overlooking the effects of exogenous
soluble Vpr protein. Using synthetic Vpr protein, we studied the activation of nuclear
transcription factors NF-kB, AP-1 and JNK by EMSA in the promonocytic cell line U937.
Exogenous Vpr activated AP-1 in a dose- and time- dependent manner in U937 cell lines. The
stimulation of AP-1 by exogenous Vpr was specific since it was totally inhibited by treatment
with a neutralizing anti-Vpr antibody. Exogenous Vpr did not significantly activate NF-kB in
U937 cells. Activation of JNK is early event initiated by many stress stimuli and is required
for AP-1 activation. The N-terminus of exogenous Vpr activated JNK in a time-dependent
manner in U937 cell lines. Since exogenous Vpr activated AP-1 which binds to the HIV-1
Long Terminal Repeat (LTR), we investigated the effect of exogenous Vpr on HIV-1
transcription. Exogenous Vpr stimulated the HIV-1 LTR in U937 cells. The N-terminus of
exogenous Vpr enhanced viral replication in the chronically HIV-1-infected monocytic cells
U1. Overall our results indicate that exogenous Vpr activates HIV-1 replication in
promonocytic cells and therefore might be a potential target for future antiretroviral

                                            - 224 -
Session V: Chemopreventive agents                           Poster V, 21

Chiral Epimer Inositols and Inososes in Signal Transduction and Therapeutic Targets
in Insulin Resistance and Diabetes Mellitus: Redox Products of Inositol Dehydrogenases and
NADP-Dependent Epimerase [D-Xylose:NADP 1-oxidoreductase, E.C.].

Josef H. Wissler and Enno Logemann.

ARCONS Applied Research Institute, Postfach 1327, D-61231 Bad Nauheim and
University of D-79111 Freiburg, Germany. E-Mail:

Free and bound natural epimers of (vitaminoid) inositol and redox-related [ascorbate-similar]
reductone structures [inososes, inosenoles] are rare but ubiquitously occuring isomers. They are
formed from myo-inositol by inositol dehydrogenases and epimerases and were shown up to play
pivotal biological roles. Thus, e.g., beside numerous forms of their phosphorylated de-rivatives,
inositols are known constituents of proteinaceous inositol phosphoglycan membrane anchors. These
anchors are of relevance in protein structure and function and several signal transduction mechanisms.
Examples are membrane-anchored prion proteins and insulin recep-tors. In the latter, a chiral epimer,
D-chiro-inositol [DcI], is a mediator of insulin action. Thus, altered DcI metabolism in non-insulin-
dependent diabetes mellitus [type II, NIDDM] and di-rect relationships of urinary DcI excretion
[hypochiro-inositoluria] to insulin sensitivity sug-gested that DcI levels are useful diagnostic indices in
insulin resistance and NIDDM. Oral DcI administration as therapeuric measures improved deficiencies
in hormonal control of cells [Larner et al, BBRC 151:1416-1426, 1988; Kennington et al, NEJM
323:373- 378, 1990; Suzuki et al, Diabet. Care 17:1465-1468, 1994].

We evaluated chiral and epimer inositols and inososes in signal transduction and free radical and
antioxidant defense functions [Logemann & Wissler, Biol. Chem. 379:S98, 1998]. Two alternatives
were assessed for functional food reducing risks of DcI and inosose deficiencies: Access to DcI and
inososes preformed naturally in plantal / herbal dietary supplements, food and non-toxic organisms
(meat) and by enzyme technology. For the first approach, different grains, beans, peas and their germs
[oats, wheat, spelt, rape, barley, maize, soy bean, climbing and bush beans, sweet pea], mushrooms,
cauliflower, broccoli, carrot, cress, parsley, linseed, camomile, maple, tomato, potato, spinach, apple,
orange, lemon, pineapple, cherry, nut, algae were analyzed for free and bound epimers by GC-MS.
Only few of them at all contained the rare compounds in significant, useful amounts (e.g. >1µg DcI/g).
These results suggest that in modern nutrition behaviour food does not supply enough DcI for
equilibrating metabolic defi-ciencies. Yet, selected blends as plantal / herbal dietary supplements rich
in natural DcI and inososes can be orchestrated. For biotechnology with enzymes of epimer or chiral
substrate selectivity, inositol dehydrogenases from mammalian eye lens and bacterial membranes that
have been purified to homogeneity and defined in amino acid sequence, were put into play [Wissler et
al, Meth. Enzym. Anal. 6: 449-465, 1984; Fresenius Z. Anal. Chem. 330:367-368, 1988; FASEB J.
9:A1482, 1995 & 13:A271, 1999]. It could be shown that in contrast to the bacterial enzyme,
mammalian dehydrogenase is a NADP-dependent inositol epimerase which is identical in structure
and function to D-xylose:NADP 1-dehydrogenase [E.C.] [Wissler et al, Hoppe-Seylers
Z.Physiol. Chem. 358:1300-1301, 1977; Meth. Enzym. Anal. 6: 449-465, 1984]. The different
properties of enzymes in use allow multiple approaches to epi-mers of inositols and inososes. Thus,
DcI can be obtained starting from grain phytin, myo- or scyllo-inositol via myo-inosose-2, enolization
and reduction of myo-inosose-1 [D-chiro-ino-sose]. The results suggest access to valuable blends and
functional food forms to aim at redu-cing health risks by DcI deficiencies. Furthermore, as further
novel targets for therapeutic interventions, this investigation could disclose reciprocal dependences of
enzymatic epimeri-zation on D-glucose levels. It suggests a role of reductone-structured inososes in
cellular an-tioxidant protection and its impairment by some metal ions and in [pre-]diabetic situations.

                                                  - 225 -
Session IV: siRNA and gene regulation

                   - 226 -
Session IV: siRNA and gene regulation                     Poster VI, 1

Efficient siRNA Delivery for Accurate Gene Knockdown Studies

Jörg Dennig, Peter Hahn, Ute Krüger, Anja Grewe, Silvia Magyar, Cornelia Schmidt, Claudia
Ferfers, Jie Kang, and Wolfgang Bielke

QIAGEN GmbH, QIAGEN Strasse 1, 40724 Hilden, Germany

RNA interference (RNAi) is a cellular defense mechanism that controls the expression of
ectopic genes in a wide range of different species. The function of RNAi is highly conserved
through evolution, and parallels the immune system by protecting the genome against the
invasion by mobile genetic elements, such as viruses or transposons. This process, also known
as post-transcriptional gene silencing (PTGS) in plants, is driven by double-stranded RNA
(dsRNA) and leads to the specific degradation of homologous mRNA.
In higher eukaryotic cells, transfection with dsRNAs longer than 30 base pairs leads to
induction of the interferon response. However, it could be shown that the use of 21-nucleotide
small dsRNAs (so called siRNAs) allows efficient RNA interference without non-sequence-
specific suppression of gene expression (1). Therefore, RNAi started to evolve to a widely
used tool for the generation of loss-of-function phenotypes in the biotechnology and
pharmaceutical industries. Based on RNAi, times for screening and validation for drug targets
can be reduced significantly.
The process of degradation of the target mRNA, its kinetics and mechanism is yet not
understood completely. We will present data addressing degradation patterns of mammalian
mRNAs that are targeted specifically for destruction by synthetic siRNA after transient
transfection of mammalian cells. Our findings outline some critical aspects that need to be
taken into consideration when establishing a new mRNA “knockdown” system to avoid e.g.
misleading results in quantitative analysis of mRNA or verification of the “knockdown” of
the targeted mRNA. In addition, we will present an integrated and automated system for
RNAi using the Biorobot Gene Expression System. This system allows efficient, high-
throughput testing of gene silencing using a large number of siRNAs.

(1) Elbashir et al., 2001 Nature 411, 494.

                                             - 227 -
Session VI: siRNA and gene regulation                     Poster VI, 2

Regulation of IL-1beta and IL-6 mRNA stability by their 3'-untranslated region.

Sandrine Servotte, Alain Colige, Charles A. Lambert, Charles M. Lapière and Betty V.

Laboratory of Connective Tissues Biology, University of Liège, Belgium.

Selective destabilization of short-lived mRNAs as those of cytokines is facilitated by an AU-
rich element (AURE) located in their 3'-untranslated region (3'UTR). This AURE contains
one or more AUUUA motifs, which may be overlapping or dispersed. Our work aimed at
studying the regulation operated by the AURE-containing 3'UTR of IL-1beta and IL-6. WI-
26, a human embryonic lung fibroblastic cell line, were transfected using constructions
containing the Secreted Alkaline Phosphatase (SEAP) as reporter gene under the control of
the SV40 promoter (pSV40) and followed by the IL-6 (3'IL6), the IL-1beta (3'IL1) or the
SV40 (3'SV40) 3'-untranslated region. A maximal secretion of SEAP was measured in the
WI-26/pSV40-SEAP-3'SV40 reflecting the strong activity of the SV40 promoter and the lack
of AURE in the 3'SV40. By replacing the 3'SV40 by the 3'IL6 or the 3'IL1, an almost
complete suppression of SEAP production was observed at both protein and mRNA levels,
pointing to the efficacy of the mRNA destabilization activity of the 3'UTR of both cytokines.
Similar data were obtained in HEK293, a human embryonic kidney cell line. To determine the
importance of the various AUUUA motifs of the 3'UTR of IL-1beta and IL-6 in the
destabilization of their mRNA, plasmids containing the SEAP gene under the control of an
inducible promoter (Ind) and followed by native or truncated 3'UTR were constructed. They
were transiently transfected in HEK293 cells stably expressing a receptor (VgRXR) enabling
activation of the Ind promoter by ecdysone or analogs as ponasterone A. Transcription of
SEAP was induced by ponasterone A for 24 hours and switched off after washing out of the
inducer, and the SEAP activity was measured after another 24 hours. SEAP activity was
largely expressed in pInd-SEAP-3'SV40 transfected cells while it disappeared when all the
AUUUA motifs of the IL-1beta or IL-6 3'UTR are present. Progressive deletion of the
AUUUA motifs resulted in progressive recovery of SEAP expression, showing that the
various motifs cooperate to control the mRNA destabilization.

                                           - 228 -
SessionVI: siRNA and gene regulation                    Poster VI, 3

Adenoviral gene transfer of siRNA into primary neurons

Sabine Sturany, Marc Niere, Thomas Pruss, Frank Weise and Hansjürgen Volkmer

NMI-Natural and Medical Sciences Institute, University of Tübingen, Department of
Molecular Biology, Markwiesenstraße 55, D-72770 Reutlingen, Germany, E-mail:

We established the adenoviral expression of siRNA under the control of the H1RNA
promoter. The functionality of this system was validated by siRNA-mediated knock-down of
both EGFP (Enhanced Green Fluorescent Protein) and neurofascin – a cell adhesion molecule
being important for neuronal differentiation.
Before the construction of recombinant adenoviruses, candidate siRNAs were evaluated by
transient transfections into HEK 293 cells expressing EGFP and neurofascin, respectively..
Thus, the successful knock-down of EGFP and neurofascin could be shown both on the level
of mRNA and protein.
Finally, recombinant adenovirus expressing the validated neurofascin-specific siRNA was
used to transduce primary neurons being difficult to transfect. In a functional assay, the
expression of neurofascin was then shown to be vital for the formation of neurites.
Furthermore, recombinant adenoviruses were immobilized on a microarray and were shown
to be capable of infecting cells being trapped on adenoviral spots. Thus, immobilized
recombinant adenoviruses expressing siRNAs should allow the study of multiple loss-of-
function phenotypes in a miniaturized and parallelized manner.

                                          - 229 -
Session VI : siRNA and Gene regulation                    Poster VI, 4

Depletion of MLL1-AF4 fusion gene by RNAi inhibits proliferation and induces
apoptosis in t(4;11)-positive SEM cells.

Maria Thomas1,2, Christian Huenefeld2, Johann Greil2 and Olaf Heidenreich1.
  Department of Molecular Biology, Institute for Cell Biology, University of Tuebingen,
Auf der Morgenstelle 15, 72076 Tuebingen, Germany; E-mails:;
  Department of Pediatric Hematology and Oncology, University Children's Hospital,
Hoppe-Seyler-Str.1, 72076 Tuebingen, Germany; E-mails: johann.greil@med.uni-;

Reciprocal chromosomal translocation t(4;11)(q21;q23) creates the fusion genes MLL-AF4
and AF4-MLL located on derivative chromosome 11 or derivative chromosome 4,
respectively. The t(4;11)(q21;23) translocation is diagnosed in more than 50% of infant high-
risk acute lymphoblastic leukemias and is associated with very poor prognosis. Here, using
sequence-specific mechanism of RNAi, we have silenced the expression of the MLL/AF4
fusion gene in the SEM cells carrying t(4;11)(q21;23) translocation. We performed a scanning
of the fusion site with 14 different siRNAs, of which two, designated siMLL1 and siMLL7,
caused a more than two-fold decrease in MLL1-AF4 mRNA levels in SEM cells after
transfection by electroporation. Down-regulation of MLL-AF4 mRNA expression with
specific siRNAs against MLL-AF4 suppressed MLL-AF4 mRNA expression transiently for
four days. This treatment reduced the clonogenic capacity of MLL1 or MLL7 -treated SEM
cells by factor 10. The specificity of the applied siRNAs was also tested with the t(8;21)-
positive Kasumi-1 cells. We observed, that one of two active MLL1-AF4 siRNAs reduced
two-fold the clonogenicity in Kasumi-1, whereas the other one did not. Furthermore, two
repetitive transfections of SEM cells with the active MLL1-AF4 siRNAs, but not with control
siRNAs, reduced not only the clonogenic capacity, but also the proliferation, inhibited G1-S
transition during the cell cycle and induced apoptosis. Therefore, we conclude that siRNA
with specificity against MLL-AF4 are promising agents for novel treatment concept of
t(4;11)(q21;q23) leukemias.
This work was supported by José-Carreras Foundation.

                                           - 230 -
Session VI: siRNA and gene regulation                        Poster VI, 5

The mechanism of pathogenic bacteria activation in host body.

Alexandra A. Tomova, Yulia M. Romanova, Alexandr L. Gintsburg

Laboratory of Genetic Engineering of Pathogenic Microorganisms, The Gamaleya Research
Institute of Epidemiology and Mycrobiology, Russian Academy of Medical sciences, 18
Gamaley street, Moscou 123098, Russia; e-mail:

The problem of interaction micro- and macroorganisms is one of the important problems in
infectious pathology process. The investigation of mechanisms of reproduction and growth
activation of culturable and nonculturable forms of pathogenic bacteria in host body and
following then infectious process is very important and actual. Besides, it may be a model
system for studying mechanisms of latent form infection transfer into acute form. One of the
first step of investigation there is search of natural inductors of such activation in environment
and host body. To reveal mechanism of pathogenic bacteria activation in host body we
investigated the growth rates of vegetative and resuscitation of nonculturable forms of
Salmonellae after effect of different inductors. There was shown that addition of cytokine
TNF-alfa stimulated the growth of vegetative cells as resuscitation of its nonculturable forms
in vitro as well as in vivo. Mutation of pqi gene, controlling of nonculturable forms
generation in Salmonellae prevented the effect of TNF-alfa. Concerning the important
function of pqi gene in mechanism of interaction bacteria and cytokine, the sequence of whole
gene was fulfilled. The genes with enhanced level of expression during incubation of cells
with cytokine were revealed by method of molecular display. A list of genes highly up
regulated in the salmonella cells treated with TNF-alfa in vitro. The products of the genes up
regulated in these conditions are associated with active replication and metabolism of
bacteria. The most interesting genes are the ones associated with a regulatory mechanism
called “quorum sensing”. There was shown that antioxydant compound (oxyrase enzyme) and
component of “quorum sensing” regulatory system provides more active resuscitation.

                                              - 231 -
Session VII : Signal transduction
mechanisms in health and disease

               - 232 -
Session VII : Signal transduction mechanisms in health and disease      Poster VII, 1

New strategies for the development of anti-cancer drugs in gonadotropin responsive
cancers based on signal transduction pathways evoked through the gonadotropin

Abraham Amsterdam and Abigail Land-Bracha

Weizmann Institute of Science, Department of Molecular Cell Biology, Rehovot, Israel

Gonadotropins play a crucial role in ovarian homeostasis and fertilization through the
activation of the cAMP cascade. However, hypergonadotropin stimulation may be associated
with elevated risk of ovarian cancer. Some correlation between high gonadotropic levels in
peritoneal and ovarian cyst fluids of patients suffering from ovarian cancer has been
suggested (1, 2). Moreover, we have recently discovered that gonadotropin stimulation can
activate the MAPK cascade in target cells (3-5). Using DNA microarray technology and RNA
from human granulosa cells, which comprise the main bulk of the ovarian follicular somatic
cells, we discovered that stimulation with saturating doses of gonadotropins dramatically
elevates activity of genes coding for epiregulin and amphiregulin. These gene products can
bind and activate the EGF receptor and ERB-4 which are associated with the development of
ovarian, breast, endometrial and other non-gynecological cancers. Since gonadotropin
receptors are expressed not only in the gonads, but also in non-gonadal tissue and cancer
cells, a novel anti-cancer therapy is being considered: 1. specific interference with
gonadotropic hormone receptor gene expression, 2. blocking the expression of epiregulin and
amphiregulin, which belong to the EGF family, in cancerous tissue. This can be achieved
using anti-sense and or SiRNA techniques targeted against the expression of the gonadotropin
receptors and/or the growth factors of the EGF family. Thus, the discovery that gondaotropins
activate certain mitogenic signal transduction pathways, may serve as a guide for the
treatment of specific cancers by blocking the gonadotropic response, either on the receptor
level or downstream on the mitogenic signals generated by these hormones.

1. Halperin, R., E. Hadas, R. Langer, I. Bukovsky, and D. Schneider. (1999) Peritoneal fluid
gonadotropins and ovarian hormones in patients with ovarian cancer. Int J Gynecol Cancer
2. Halperin, R., M. Pansky, Z. Vaknin, S. Zehavi, I. Bukovsky, and D. Schneider. (2003)
Luteinizing hormone in peritoneal and ovarian cyst fluids: a predictor of ovarian carcinoma.
Eur J Obstet Gynecol Reprod Biol 110:207-10.
3. Seger, R., T. Hanoch, R. Rosenberg, A. Dantes, W. E. Merz, J. F. Strauss, 3rd, and A.
Amsterdam. (2001) The ERK signaling cascade inhibits gonadotropin-stimulated
steroidogenesis. J Biol Chem 276:13957-64.
4. Tajima, K., A. Dantes, Z. Yao, K. Sorokina, F. Kotsuji, R. Seger, and A. Amsterdam.
(2003) Down-regulation of steroidogenic response to gonadotropins in human and rat
preovulatory granulosa cells involves mitogen-activated protein kinase activation and
modulation of DAX-1 and steroidogenic factor-1. J Clin Endocrinol Metab 88:2288-99.
5. Amsterdam, A., Tajima, K., Frajese V., Seger, R. (2003) Analysis of signal transduction
stimulated by gonadotropins in granulosa cells. Molecular Cell Endocrinology 202, 77-80.

                                           - 233 -
Session VII : Signal transduction mechanisms in health and disease       Poster VII, 2

Red ox–signaling by ionizing rad iation in mouse liver

Jeu ng Hee An, Jiyoung Kim, Jinsil Seong ,

Department of Radiation Oncology, Brain Korea 21 Project for Medical Science, Yonsei
University Medical College, Shinchon-dong 134, Seodamun-Ku, Seoul, Korea Email adress:

Since radiation treatment has been reappraised in the treatment of hepatic tumors, radiation
response in the liver is emerging as an interesting new area of investigation. In this study,
identification of the repertoire of signaling proteins was performed by a proteomics approach
involving cellular responses of liver tissue to ionizing radiation. Approximately 800 protein
spots were detected. Among them, a least 28 proteins showed a significant quantitative
alteration following radiation. The significantly altered proteins were categorized to ones
related to reactive oxygen species (ROS) metabolism, metabolic pathway, and G-type
proteins. Particularly, the expression level of proteins related to ROS metabolism including
cytochrome c, glutathione S transferase Pi (GSTP), NADH dehydrogenase and peroxiredoxin
VI (Prx VI) were increased after radiation. It is suggested that while radiation initiates its
cytotoxic effects, it can also induce radioprotective antioxidant system.

                                            - 234 -
Session VII : Signal transduction mechanisms in health and disease         Poster VII, 3

Modulation of arachidonic and oleic acids incorporation into signaling phospholipids by
1-methyl, 2-mercaptoimidazol in liver

Nataliya A. Babenko, Oksan A. Krasilnikova

Department of Physiology of Ontogenesis, Institute of Biology, Karazin Kharkov National
University, Kharkov, 61077, Ukraine. E-mail:

Sensitivity of various cell types to stimulation by agonists depends on the animals thyroid
status. Previously it was reported that thyroid hormones are important physiological
modulators of diacylglycerol (DAG) level in rat liver cells. The effect of hyper- and
hypothyroidism on DAG/protein kinase C signaling in hepatocytes has been determined. Cell
response to calcium-mobilizing hormones dependent on acyl groups unsaturation of DAGs
accumulated in stimulated cells. As little is known about phosholipid DAG precursors acyl
group remodeling at different functional states of thyroid gland, the present work was carried
out to study the influence of 1-methyl, 2-mercaptoimidazol (MMI) and exogenous L-
thyroxine (L-T4) on fatty acids incorporation into hepatic lipids. The experiments were
performed on the liver of control, MMI- and MMI+ L-T4-treated 90-day-old Wistar mail rats.
To investigate the fatty acids incorporation into lipids, [3H]arachidonic and [14C]oleic acids
were used. Liver phosphatidylinositol (PtdIns), phosphatidylinositol 4-phosphate (PtdInsP),
phosphatidylinositol 4,5-bisphosphate (PtdInsP2) and total phospholipids (PL) were separated
by means of HPTLC. MMI administration to rats was accompanied by a marked decrease of
thyroid hormones (L-T4 and L-triiodothyronine) levels in blood serum and [3H]arachidonic
acid incorporation into total lipids and PL and increase of fatty acid incorporation into PtdIns,
PtdInsP and PtdInsP2 in liver homogenates. However, drug administration to rats increases
markedly the [14C]oleic acid incorporation into total liver lipids, decreases the PtdIns
acylation and has no effect on other lipids labeling. L-T4 injection to the MMI-treated rats
diminished the drug stimulatory effect on [3H]arachidonic acid incorporation into total lipids
and PL and did not change significantly phosphoinositides acylation. However, hormone
farther decreases the level of [14C]oleic acid labeled PtdIns in hypothyroid liver. The results
obtained provide the first demonstration of MMI regulation of phoshpoinositide acylation in
liver. MMI acts on phoshpoinositide acylation via thyroid hormone independent mechanism,
while MMI action on arachidonic acid incorporation into PL is mediated by its suppressive
action on thyroid gland.

                                             - 235 -
Session VII : Signal transduction mechanisms in health and disease       Poster VII, 4


Cristiana Bertolani, Stefania Gelmini*, Ilaria Petrai, Sara Aleffi, Claudio Orlando* and Fabio

Dipartimenti di Medicina Interna e di *Fisiopatologia Clinica, Università di Firenze

The molecular mechanisms of fibrogenesis during non-alcoholic steatohepatitis (NASH) are
only partially understood. Adiponectin is a hormone exclusively expressed by adipose tissue,
which circulates as a full-length 30kDa molecule or the isolated C-terminal globular domain.
Plasma adiponectin concentrations correlate with insulin sensitivity and are decreased in
patients with obesity and insulin resistance. Moreover, administration of adiponectin
ameliorates liver damage in a model of steatohepatitis. Two adiponectin receptors (AdipoR1
and AdipoR2) have been recently cloned. While AdipoR2 binds both full-length and globular
adiponectin with intermediate affinity, only the globular form is a high-affinity ligand for
AdipoR1. Aim of the present study was to investigate whether cultured human hepatic stellate
cells (HSC) express adiponectin receptors and to correlate receptor activation with the
biologic actions of the profibrogenic cytokine, platelet-derived growth factor (PDGF).
Analysis of different lines of myofibroblastic HSC by quantitative RT-PCR demonstrated the
expression of both AdipoR1 and AdipoR2, with higher levels of AdipoR1. Recombinant
globular adiponectin dose-dependently reduced the increase in DNA synthesis elicited by
PDGF, with approximately 70% inhibition at a concentration of 5 microg/ml. Analysis of
HSC migration in Boyden chambers showed a similar inhibitory action on PDGF-dependent
chemotaxis, with marked effects with 5 microg/ml adiponectin. Quantitative RT-PCR also
indicated that adiponectin reduces collagen I expression, while only modest effects were
observed on the expression of MMP2 or TGF-beta1.
In conclusion, this study shows for the first time that HSC express functionally active
adiponectin receptors, which modulate the effects of PDGF. Reduced adiponectin
concentrations may be involved in the pathogenesis of fibrosis during NASH.

                                            - 236 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 5

Regulated transcription in UVB induced premature senescence.

Céline Borlon1, Florence Chainiaux1, José Remacle1, Olivier Toussaint1
    Unit of research on Cellular Biology, University of Namur (FUNDP), Belgium

Repeated in vitro exposure to UVB (290–320 nm) induces stress-induced premature
senescence (SIPS) of skin human diploid fibroblasts (HDFs) (Chainiaux, et al. 2002), with
appearance of biomarkers of replicative senescence like an increase in the proportion of cell
positive for senescence-associated b galactosidase activity, a dramatic loss of replicative
potential and an increase of the steady-state level of the mRNA of several senescence-
associated genes (i.e. fibronectin, osteonectin, SM22 and apolipoprotein J). The stress model
is a “10 stress model” : twice a day for 5 days, cells are exposed to 250 mJ/cm2 of UVB. We
also know that many other genes are implicated in SIPS induced by UVB from a premiminary
study based on transcriptomics. In order to find transcription factors involved in these changes
of gene expression, we performed multi well colorimetric assays (Renard, et al. 2001) for
different transcription factor (ATF2, NFKB, AP1, sp1, p53, CREB, Hif, STAT1, STAT3,
STAT5). These transcription factors are known in litterature as being involved in stress
response. Furthermore, we studied evolution of the DNA-binding activity of these
transcription factors with time. The STAT’s, CREB and Hif did not show any change of
DNA-binding activity whereas AP1, ATF2, p53, NFKB and sp1 did.
The DNA-binding activity of ATF2 increased up to 1.5 fold in cells exposed to UVB while
the DNA-binding activity of NFKB was attenuated. AP1 and p53 DNA-binding activity
increased just after stress and then returned to basal level within 72h after stress.
These preliminary results are encouraging. Indeed gene expression studies can now be
performed after inhibition of the DNA-binding activity of selected transcription factor.

Chainiaux F., et al. (2002). "UVB-induced premature senescence of human diploid
fibroblasts. " Int. J. Biochem. Cell. Biol. 34, 1331-1339
Renard P., et al. (2001). "Development of a sensitive multi-well colorimetric assay for active
NFkappaB. " Nucleic Acids Res 29(4): E21.

C. Borlon is funded by the Région Wallone and FSE, « First Europe » , Arrayage (contract
EPH3310300R0472/215316) project. O. Toussaint is a Research Associate and F. Chainiaux a
Research Assistant of the Fonds National de la Recherche Scientifique, Belgium.

                                             - 237 -
Session VII: Transcription factors interaction with HATs and HDACs        Poster VII, 6


Florence CAMMAS1, Marielle HERZOG, Thierry LEROUGE, Pierre CHAMBON and
Régine LOSSON.

Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP/Collège
de France, BP10142, 67404 Illkirch-Cedex, France.

Regulation of chromatin structure is a highly dynamic process known to play essential roles in
cell physiology. The transcriptional Intermediary Factor 1 (TIF1) family protein TIF1b is a
co-repressor for Krüppel-Associated box (KRAB)-domain-containing zinc finger proteins and
is believed to regulate chromatin organization at specific loci through recruitment of
Heterochromatin Protein 1 (HP1) proteins. We have assessed the functional relevance of the
interaction between TIF1b and HP1. A specific mutation, TIF1bHP1box, leading to disruption of
the interaction between TIF1b and HP1 was introduced by homologous recombination in the
wild type TIF1b allele of embryonal carcinomal (EC) F9 cells carrying a TIF1b- null allele.
TIF1bHP1box/- cells show specific differentiation defects, being able to differentiate into
primitive endodermal-like (PrE) cells upon retinoic acid (RA) treatment, but not into parietal
endodermal-like (PE) differentiation upon further stimulation with cAMP. Ectopic expression
of TIF1b cDNA within TIF1bHP1box/- cells using a doxycyclin-inducible system of regulation
allowed us to demonstrate that, although the interaction between TIF1b and HP1 is not
required to establish PrE cells, it is essential in these cells for their progression towards PE
differentiation, whereas the interaction is dispensable during the cAMP dependent phase of
PE differentiation. Gene expression analysis showed that TIF1bHP1box/- cells display an
impaired response to RA treatment and of particular interest a downregulation of expression
of several genes known to play essential roles during endoderm differentiation. Collectively,
these data demonstrate that the interaction between TIF1b and HP1 is an essential factor for
commitment of PrE cells towards PE differentiation.

                                             - 238 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 7

Proteomic study discriminating proteins undergoing expression changes in the
appearance of the premature senescence of human fibroblasts induced dependently or
independently of TGF-b1.

Unit of Research on Cellular Biology , University of Namur (FUNDP), Belgium; 2Biovallée,
Gosselies, Belgium.

Premature senescence induced by subcytotoxic H2O2 stress in human diploid fibroblasts
(HDFs) constitutes an in vitro model to study the long term effects of this stress on cells. The
study of the molecular mechanisms of appareance of H2O2-induced senescence-like phenotype
reveals that an overexpression of TGF-b1 at 24 h after subcytotoxic H2O2 stress in HDFs
contributes to the appareance of this phenotype. This phenotype is also observed after
stimulation with TGF-b1 for 72 h. After neutralisation of TGF-b1 or of its receptor type II,
most of, but not all, these characteristics of SIPS disappear [1]. This suggests that TGF-b1 is
deeply involved in the appareance of SIPS, but is not the only factor involved. We wish to
discriminate genes/proteins wish undergo expression changes dependently or independently
of TGF-b1. For this purpose, we undertook a transcriptomic study based on a low-density
arrays representing genes undergoing expression changes in (premature) senescence and a
proteomic study. The proteomic study is based on two-dimension gel electrophoreses (2
DGE) after radiolabelling of proteins with [35S]-methionine from 48 h to 72 h after
subcytotoxic H2O2 stress ± incubation with antibodies neutralizing TGF-b1. Comparison to
the effect of stimulation with TGF-b1 was performed. Controls with cells incubated with
culture medium ± antibodies were also performed. A quantitative analysis of the 2 DGE
triplicate reveals that thirty-one spots undergo expression changes in one or several conditions
depending on TGF-b1 or not. This preliminary study suggests there are effectors of SIPS
acting independently of TGF-b1. Identification by mass spectrometry of the proteins
associated with the thirty-one spots of interest is ongoing.

Reference :
[1] Frippiat C, Chen QM et al., J Biol Chem 276 :2531-7, 2001
Acknowledgements :
O. Toussaint is a research associate of the belgican F.N.R.S.

                                             - 239 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 8

Artifactual generation of oxygen free radicals in the mixture of cyanide and glycerol

Yang-Sook Chun,1 Eun-Jin Yeo,1 Hwa-Jin Suh,2 and Jong-Wan Park1*
  Department of Pharmacology and Cancer Research Institute, College of Medicine, Seoul
National University, 28 Yongon-dong, Chongno-gu, Seoul 110-799, Korea.
  Department of Agricultural Chemistry, College of Agriculture and Life Sciences, Seoul
National University, Suwon, Korea

* Corresponding author : Jong-Wan Park, Department of Pharmacology, Seoul National
University, College of Medicine, 28 Yongon-dong, Chongno-gu, Seoul 110-799 Korea.
Telephone (2)740-8293, Fax (2)745-7996, E-mail

Cyanide has been commonly used as an inhibitor in the study hemoproteins, such as oxidases
and peroxidases. It is also useful for selectively inhibiting mitochondrial oxidases and Cu, Zn-
superoxide dismutase. On the other hand, glycerol has been widely used as a stabilizer of
various enzymes. In particular, glycerol is required to maintain the enzymatic activities of
membrane-bound NAD(P)H oxidases extracted from surrounding phospholipids. Since both
cyanide and glycerol are relatively inert, they have been used concomitantly regardless of any
mutual interference. In this study, we demonstrate that a mixture of glycerol and cyanide
reduced cytochrome C and nitroblue tetrazolium, which are superoxide anion indicators. As a
mixture, they also enhanced the production of superoxide anion in the presence of redox-
cycling compounds. Superoxide production by the mixture was confirmed by electron spin
resonance spectra. Moreover, the mixture induced lipid peroxidation and hemolysis in human
erythrocytes. These results suggest that cyanide and glycerol should be used carefully in
reaction systems used for measuring superoxide production or antioxidant activity. However,
sucrose and sodium azide in combination do not produce such artifacts, and thus, may be used
as an alternative.

                                             - 240 -
Session VII : Signal transduction mechanisms in health and disease             Poster VII, 9

Post-translational modification of sox6 transcriptional factor by SUMO

Fernández-Lloris, R.1, Shen, L.2, Jaffray, E.2, Hay, R.T.2 and Ventura, F. 1
   Departament de Ciències Fisiològiques II, Campus de Bellvitge, Universitat de Barcelona,
Feixa Llarga s/n, 08907 L’Hospitalet de Llobregat, Spain.
   University of St Andrews. Lab 2, 2nd floor. BMS Building, KY 169 AJ, North Haugh. St.
Andrews, Scotland.
Post-translational modification can target proteins to different cellular fates. SUMO (small
ubiquitin-like modifier) conjugation has diverse cellular functions, in both nuclear and
cytoplasmatic processes, as protein targeting and/or stability, regulation of transcriptional
processes and regulation of mitosis. To date, only a small proportion of substrates for SUMO
modification have been identified, but this work is ongoing to find new SUMO target
proteins. Sox6 and Sox9, two members of the Sox family with crucial functions as a
transcription activators in a great many developmental processes, including neurogenesis, sex
determination and skeleton formation, present some motifs that match perfectly with the
sumoylation consensus sequence (yKxE, where y represents L, I, V or F) . Our studies are
mainly focused on the determination of a putative modification by SUMO of these two
transcriptional factors (Sox6 and Sox9), the characterization of the ligases enzymes and the
consensus sites involved in this process and the study of the effects of this sumoylation.

                                             - 241 -
Session VII : Signal transduction mechanisms in health and disease         Poster VII, 10

Lipid rafts functioning in T cells according to T cell subsets and aging

Tamas FÜLÖP, Anis LARBI, Abdelouahed KHALIL, Gilles DUPUIS,

Research Center on Aging and Immunology program, Faculty of Medicine, University of
Sherbrooke, 1036, rue Belvedere sud, Sherbrooke, Qc, J1H 4C4, Canada. E-mail:

Aging is associated with a decline in T-cell activation and proliferation but the underlying
mechanisms are not fully understood. Recent findings suggest that cholesterol-enriched
microdomains called lipid rafts act as a platform in the initiation of T-cell activation by
formation of the initial complex of signal transduction. Since lipid rafts are the earlier events
in T cell signalling we tested the hypothesis that lipid raft properties are altered in T
lymphocytes from elderly individuals. Results showed that the cholesterol content of lipid
rafts derived from these cells was consistently higher in the case of elderly donors and that
membrane fluidity was decreased. In addition, lipid rafts coalescence and signalling
molecules recruitment to the site of TCR engagement was impaired in T-cells from elderly
donors. Studies in CD4+ and CD8+ subsets show differential alterations in relation to lipid
rafts functions. Contrary to CD4+ T cells, CD8+ did not show coalescence upon stimulation.
Our data suggest that some properties of lipid rafts are altered in aging and the differences
observed in T cells subsets suggest different roles of lipid rafts in lymphocytes signalling as
well as in immunosenescence.

                                             - 242 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 11

Functional Characterization of APP Genetic Regulatory Polymorphisms in Alzheimer’s

Y.-W. Ge1; B. Maloney1, F. Wavrant-De Vrièze2; J. Hardy2; D.K. Lahiri1*
 Department of Psychiatry, Institute of Psychiatric Research, Indiana University School of
Medicine, Indianapolis, Indiana-46202, USA; 2Laboratory of Neurogenetics, National
Institute on Aging, Bethesda, MD, USA

Abnormalities in regulation of the b-amyloid precursor protein (APP) gene might be an
important factor in Alzheimer’s disease (AD). Our aim is to understand biological variability
in APP regulatory elements (promoter) that control APP expression, and to test this variability
for association with AD. Sequencing of the APP gene revealed two polymorphisms in the
regulatory region at -1012 (T/C) and -3818 (C/T) from the transcription start site (+1).
Using the gel shift assay and synthetic oligonucleotides corresponding to the polymorphic
regions revealed a differential pattern of DNA-protein binding with these polymorphic
regions. This was tested with nuclear proteins isolated from neuronal cells and the mouse
brain. This differential pattern was also checked by Southwestern analysis. One of the protein
bands may correspond to a specific transcription factor, and other proteins do not directly
correspond to predicted transcription factors. Alterations in DNA protein interaction with
specific polymorphism are indicative of a regulatory role of the APP promoter region in FAD
subjects. DNA affinity purification between these polymorphic regions and nuclear proteins
from the neuronal cells is in progress. Taken together, these results suggest that certain
transcription factor(s) may differentially operate on these polymorphic regions to regulate
APP expression, and that has implication in AD pathogenesis. This work is supported by a
grant from the National Institute on Aging, NIH, USA.

                                            - 243 -
Session VII : Signal transduction mechanisms in health and disease          Poster VII, 12

Direct binding of STAT5A to multiple sites of the _-opioid receptor reveals novel
regulatory signaling pathways mediated upon activation of this receptor

Georgia Mazarakou and Zafiroula Georgoussi

Laboratory of Cellular Signalling and Molecular Pharmacology, Institute of Biology,
N.C.S.R. “Demokritos”, 153 10 Ag. Paraskevi-Attikis (Athens),Greece

Signal Transducers and Activators of Transcription (STATs) are transcriptional factors that
mediate many of the effects of the cytokines. However, many G protein coupled receptors
have recently been reported to activate not only conventional G protein signaling pathways,
but also to increase tyrosine phosphorylation of the STAT family members in various cell
lines. All three opioid receptor subtypes (_,_,_) contain in their C-terminal regions the highly
consensus motif YXXL known to be critical for STAT5A/5B binding. In this regard, we
demonstrate for the first time that glutathione S-transferase fusion proteins containing either
the 69 amino-acids of the rat _-opioid receptor carboxyl-terminal tail, or the third intracellular
loop bind STAT5A in an agonist independent manner in pull down experiments using cell
extracts from COS-7 cells expressing transiently the _-opioid receptor and STAT5A. Site
directed mutagenesis of the conserved Tyr336, showed that this association is independent of
the _-receptor motif YXXL (amino acids 336-339). Moreover, we demonstrate that acute
exposure to morphine of COS-7 cells expressing the rat _-opioid receptor and STAT5A
induces STAT5A tyrosine phosphorylation in a time dependent manner. Similar
phosphorylations were also observed with the _-opioid receptor agonist DAMGO. Responses
to morphine and DAMGO were blocked by naloxone. Treatment of COS-7 cells, transiently
expressing the _- opioid receptor and STAT5A with the Src specific inhibitor PP1, abolished
opioid agonist-mediated STAT5A phosphorylation. This finding is in accord with src kinase
phosphorylation observed in COS-7 cells upon _-opioid agonist stimulation. In addition we
also indicate that activation of HEK293 cells stably expressing the _- opioid receptor, with
morphine, stimulated the expression of a STAT5A responsive reporter gene. Our data reveal
novel signaling pathways through which _- opioid receptor regulate transcriptional activity,
probably altering gene expression in specific target neurons and thereby inducing tolerance
and dependence.

                                              - 244 -
Session VII : Signal transduction mechanisms in health and disease       Poster VII, 13

Progesterone Regulates Transcription of the p27Kip1 Cyclin-dependent Kinase Inhibitor
through Sp1.

Florence Gizard1, Romain Robillard1, Olivier Barbier1, Dean Hum2 and Bart Staels1
 UR545 INSERM, Département d'Athérosclérose, Institut Pasteur de Lille et Faculté de
Pharmacie, Université de Lille II, France; 2Genfit, Parc Eurasanté, Loos, France.
E-mail :

Progesterone is required for terminal differentiation and growth of the mammary gland. In
breast cancer cells, the predominant effect of treatment with progesterone is long-term
inhibition and arrest in G1 phase. The progesterone-mediated growth effects implicate the
induction of the cyclin-dependent kinase inhibitor p27Kip1, a major regulator of cell cycle
progression and differentiation. However, the elements required for progesterone-induced
transcriptional up-regulation of the p27Kip1 promoter have not yet been described. We show
here that progesterone activates the p27 Kip1 gene by a mechanism that involves interaction
between the progesterone receptors (PRs) with the transcription factor Sp1 through two Sp1
binding sites located at the -549/-511 region, which was previously shown to mediate the
vitamin D3 response. Deletion and mutational analysis of the promoter revealed that the
-545/-539 Sp1 site 1 is necessary for basal p27Kip1 transcription, and both Sp1 sites are
required for transcriptional up-regulation by progesterone. Chromatin immunoprecipitation
experiments further indicated that PR interacts with this Sp1 binding element in a
progesterone-dependent manner. Finally, gel shift analyses showed that PR does not bind
directly to the Sp1 DNA sequences, but instead stimulates the DNA binding activities of Sp1.
Identical progesterone-dependent transcriptional mechanisms have recently been shown for
other specific target genes, such as p21WAF1. Therefore, the present study provides an
additional evidence of the regulation by progesterone of certain crucial cell cycle regulating
genes through non-canonical progesterone responsive elements.

                                            - 245 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 14

Stimulation of aldosterone secretion by sphingosine 1-phosphate

Leyre Brizuela, Miriam Rábano, Ana Peña, José M. Macarulla, Miguel Trueba and Antonio

Department of Biochemistry and Molecular Biology. Faculty of Science and Technology.
University of the Basque Country. 48080 – Bilbao (Spain)

Sphingosine 1-phosphate (Sph 1-P) is a bioactive phospholipid that regulates critical
biological functions including cell proliferation, differentiation, angiogenesis, and cell
survival. It is formed by ATP-dependent phosphorylation of sphingosine by sphingosine
kinase, and can be metabolized by lyase or phosphatase activity. There is evidence of dual
messenger functions for Sph 1-P. First, it can be released into the blood stream by different
cell types thereupon interacting with specific G protein-coupled receptors of the endothelial
differentiation gene family. Second, it can be generated intracellularly by growth factor
stimulation of sphingosine kinase. We show here for the first time a new biological action of
Sph 1-P: stimulation of aldosterone secretion in glomerulosa cells of bovine adrenal glands.
This effect was inhibited by preincubating the cells overnight with 4-beta phorbol 12-
myristate 13-acetate (PMA), a condition known to cause downregulation of protein kinase C
(PKC), or by selective PKC inhibitors. In addition, chelation of extracellular calcium with
EGTA also blocked Sph 1-P-stimulated aldosterone secretion. Sph 1-P activated
phospholipase D (PLD) through a mechanism that was also dependent of PKC and
extracellular calcium. Of interest, primary alcohols, which attenuate the formation of
phosphatidic acid (the product of PLD), and cell-permeable ceramides, which inhibit PLD
activity potently, blocked the stimulation of aldosterone secretion by Sph 1-P. Taken together,
these results suggest that PKC, extracellular calcium, and PLD are all important components
in the cascade of events leading to the stimulation of aldosterone secretion by Sph 1-P.

Supported by grant PI99/8 from the “Departamento de Educación, Universidades e
Investigación del Gobierno Vasco (Basque Country).

                                            - 246 -
Session VII : Signal transduction mechanisms in health and disease       Poster VII, 15

Lysophosphatidic acid activates human endothelial cells cultured in vitro

    Laboratory of Cellular Biochemistry, University of Namur, Belgium
    EAT, Namur, Belgium

Lysophosphatidic acid (LPA) is a bioactive lipid mediator produced by platelets, found in
blood and atherosclerotic plaques (Siess et al., 2002), and recently shown to affect platelets
and smooth muscle cells. Hence, we wondered if LPA could also affect endothelial cells, and
in particular gene expression, using the EAhy926 cell line (graciously provided by Coran-
Jean S. Edgell of the University of North California). Cytotoxicity was first evaluated by an
acridine orange/ethidium bromide staining. No cell mortality was observed up to 200 µM. We
then measured the bioactivity of this molecule using preliminary tests. LPA, at 150 µM,
increased the phosphorylation of ERK 1/2 after 5 minutes (210 % and 230 %, respectively),
which is in agreement with the litterature (Lee et al., 2000). After 1 hour stimulation with
LPA, the transcription factor AP-1 was activated (216 %), whereas NF-kappaB was not
affected. Transcription factor activation was monitored by using a colorimetric assay
developed in our laboratory (Renard et al., 2001). At the protein level, LPA also induced a
reproducible increase of MCP-1 expression (315 %) in endothelial cells after 6 hours of
stimulation. MCP-1 was detected using an ELISA kit (R&D’s). To have a broader view of
the LPA-induced response, we have chosen the proteomic approach. Therefore, cells were
stimulated with LPA 150 µM during 6 hours. Gene expression profiling was achieved at the
protein level by 2D-PAGE, but also at the mRNA level using a low-density DNA array
(DualChip Human General - Eppendorf). The analysis of both proteomic and transcriptomic
data is presently underway. We hope with this study to unravel the effects of LPA on
endothelial cells, to identify the molecular mechanisms involved and to confirm the pro-
atherogenic role of LPA.
Reference :
Lee, H. et al., (2000) Am. J. Physiol. Cell Physiol. 278, C612-C618
Renard, P. et al., (2001) Nucleic. Acids. Res. 29, 1-5
Siess, W., (2002) Biochim Biophys Acta 1582, 204-15
Acknowledgements :
This poster presents research results of the Belgian Programme on IPA initiated by the
Belgian State, Primer Minister’s Office, Science Policy Programming. The scientific
responsibility is assumed by its authors. Sofia Dos Santos is a FNRS Research fellow. Cindy
Gustin is recipient of a FRIA fellowship.

                                             - 247 -
Session VII : Signal transduction mechanisms in health and disease       Poster VII, 16

6-hydroxydopamine (6-OHDA) causes death of dopaminergic neurons

Thomas Herdegen

Institute of Pharmacology, Hospitalstrasse 4, 24105 Kiel, Germany,

6-hydroxydopamine (6-OHDA) causes death of dopaminergic neurons by generation of
radical oxygen species and mitochondrial dysfunction. The underlying molecular
mechanisms, however, are poorly understood. Here we show that 6-OHDA induces death by
JNK2 mediated release of cytochrome c from mitochondria in PC12 cells, which express the
JNK1 and JNK2 isoforms.
Following 25 µM 6-OHDA, total JNK activity was enhanced at protracted kinetics in the
cytoplasm, nucleus and mitochondria. Inhibition of JNKs by 2 µM SP600125 rescued 68% of
the otherwise dying cells after 24 h. In contrast to persisting JNK1, the amount of JNK2
increased in the nucleus and mitochondria. The JNK inhibitor SP600125, which reduced the
pool of active JNKs in the nucleus and the activation of c-Jun, inhibited the translocation of
JNK2, but not JNK1, to the mitochondria and the release of cytochrome c. Similarly,
transfection of dominant negative JNK2 (dnJNK2) which did not translocate to the nucleus
and mitochondria, reduced the pool of activated JNK in the nucleus and prevented the release
of cytochrome c. Moreover, dnJNK2 abrogated the activation of MKK4 at the mitochondria,
but did not affect the presence of MKK4 and JIP-1; MKK7 could not be detected in
mitochondrial preparations. Finally, SP600125 and dnJNK2 were not protective either after
72 h or against 50 µM 6-OHDA.
Our data provide novel details on the pathological role of individual JNK isoforms, the
signalosome at the mitochondria and the mode of JNK-induced release of cytochrome c.
Alterations in the pathological context render the 6-OHDA induced death independent of

                                            - 248 -
Session VII : Signal transduction mechanisms in health and disease     Poster VII, 17

g-Glutamyltransferase is regulated through the Ras signal transduction pathway in
colon carcinoma cell lines

Seila Møller, Serhiy Pankiv, Ugo Moens, Nils-Erik Huseby

University of Tromsø, Norway

GGT is important in glutathione (GSH) degradation and homeostasis. The enzyme is
upregulated during oxidative stress of colon carcinoma cells, and may play an important role
in their defence against such stress. As the Ras pathway is central in regulating the
proliferation of tumor cells and their resistance towards oxidative stress, we investigated
whether GGT is regulated through the Ras pathway. We found that GGT activity was
upregulated after incubation of colon carcinoma cells with menadione. The increased activity
was blocked by addition of actinomycin D and also by addition of MAPK inhibitors; both
PI3K, MEK1/2 and p38 inhibitors reduced the elevation.

GGT is transcribed from five promoters into multiple mRNAs. The promoter II contains
putative binding sites for AP1, AP2, NFkB and SP1. Transient cotransfection of this promoter
in a luciferase reporter plasmid with active Ras resulted in increased luciferase activity,
whereas no activity was measured with dominant negativ Ras. Permanent tranfection of colon
carcinoma cells that express Ras oncogene with dominant negative Ras, and cells expressing
wildtype Ras with active Ras, resulted in cell clones with reduced GGT and increased GGT,

Culturing colon carcinoma cells with apocynin, to inhibit the cellular production of ROS
(reactive oxygen species) through the Ras regulated NADPH oxidase, downregulated the
enzyme level of GGT. Continuous culturing with menadione to create higher concentrations
of ROS, upregulated the GGT level.

We conclude that trancription of GGT in rat colon carcinoma cells is regulated through the
Ras pathway and that ROS is an important effector in this regulation.

                                           - 249 -
Session VII : Signal transduction mechanisms in health and disease       Poster VII, 18

Transcriptional intermediary factor 1b regulates the down-regulation of the human
S100A9 gene expression. Implications for the pathogenesis of psoriasis.

Claus Kerkhoff§, Clemens Sorg§, and Jacques Doussiere#
 Institute of Experimental Dermatology, Roentgenstr. 21, 48149 Muenster, Germany
 Laboratoire de Biochimie et Biophysique des Systèmes Intégrés (UMR 5092 CEA-CNRS-
UJF), Département Réponse et Dynamique Cellulaire, CEA-Grenoble, 17 Avenue des
Martyrs, 38054 Grenoble cedex 9, France

A protein complex formed by the two Ca2+-binding proteins S100A8 and S100A9 has been
shown to modulate the phagocyte NADPH oxidase activation. Beside their specific
expression in myeloid cells the other tissue type found to express the two S100 proteins is
mucosal epithelium and involved epidermis in conditions such as psoriasis and malignant
disorders. Due to growing evidence that reactive oxygen species (ROS) are involved in the
pathogenesis of psoriasis, we investigated the effect of S100 protein expression in otherwise
S100-negative epithelial cells. We found that epithelial cells overexpressing the two S100
protein have a drastically increased NADPH oxidase activity. However, the molecular
mechanisms underlying the S100 gene expression in abnormally differentiated keratinocytes
remains unclear.
By a detailed promoter analysis we identified the novel regulatory region MRE at position
-400 to -374 bp within the S100A9 promoter. Two distinct nuclear complexes were found to
bind to MRE, and their specific formation correlated with the S100A9 gene expression in a
cell type-specific, activation- and differentiation-dependent manner. One nuclear complex
consisting of a Kruppel-related zinc finger protein (ZNF) and the transcriptional intermediary
factor 1b (TIF1b) was found to suppress S100A9 gene expression while the other one drives
S100A9 gene expression. Further analysis revealed evidence that the the expression pattern of
TIF1b inversely correlated with S100A9 gene expression.
Because the dysregulated expression of the two S100 proteins is involved in the pathogenesis
of psoriasis, we conclude that ZNF/TIF1b might represent a novel target for pharmacological
intervention. Further studies need to be performed to elucidate whether S100 gene expression
and the concomitantly NADPH oxidase activity are involved in the pathogenesis of several
inflammatory disorders.

                                            - 250 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 19

Erucylphospho-N,N,N-trimethylpropylammonium (ErPC3): an antineoplastic
alkylphosphocholine with influence on signal transduction and lipid raft reorganisation

Spiro M. Konstantinov1,2, Sornica Nyagolova1,2, Mariya Shivkova1,2, Iana Tsoneva2,3, Hansjörg
Eibl4, Martin R. Berger2
  Lab for Experimental Chemotherapy, Dept. of Pharmacology, Medical University of Sofia, 2
Dunav St., 1000 Sofia, Bulgaria; e-mail:
  Unit of Toxicology and Chemotherapy, German Cancer Research Center, INF 280, 69120
Heidelberg, Germany; e-mail:
  Inst. of Biophysics, BAS, Acad. G. Bonchev St., bl 21, 1113 Sofia, Bulgaria
  Max Planck Inst. of Biophysical Chemistry, Am Faßberg, 37077, Göttingen, Germany

Alkylphosphocholines are membrane targeting agents that induce programmed cell death in
cancer cells. ErPC3 is a new generation agent of this group which is i.v. injectable and exerts
strong antileukemic activity without myelosuppression. In contrast, it has been reported to
stimulate normal hematopoiesis. This unusual biological activity of ErPC3 might be due to
specific influences on cellular mitogenic signaling and signal molecule distribution within the
plasma membrane, cytoplasm or nucleus. As evidenced by Western blot, breast cancer (MCF-
7) and leukemia cells (SKW-3 and AR-230) showed induction of Rb expression in response
to treatment with ErPC3. Rb hypophosphorylation was found in K-562, CML-T1 and DOHH-
2 cell lines. Rb dephosphorylation and fragmentation was induced in BV-173 and SKW-3
cells. The distribution of signal molecules such as Abl, PI-3K, Akt and ERK was altered by
treatment with ErPC3 in AML-derived HL-60 and CLL-derived SKW-3 cells. Interestingly, a
concentration dependent dephosphorylation of Rb was found together with the translocation
of protein kinases that control cell cycle progression into the nucleus. Specifically, ErPC3
treatment at different concentrations for shorter or longer exposure periods resulted in
biphasic changes of the expression and phosphorylation status of the tyrosin kinases Akt, PI-
3K and ERK. This is in line with the previously reported properties of ErPC3 causing
stimulation of proliferation as well as induction of apoptosis. Taken together, our data
indicate that ErPC3 exerts its antineoplastic and bone marrow stimulating activity by distinct
early changes in the lipid raft fraction of the plasma membrane and by interfering with certain
signal transduction pathways, especially the MAPK and PI-3K tyrosine kinase cascades.

                                            - 251 -
Session VII : Signal transduction mechanisms in health and disease           Poster VII, 20

Three dimensional bioreactor cultures: an useful dynamical model for studying of
cellular interactions

Spiro M. Konstantinov1, Mina M. Mindova2, Panayot T. Gospodinov2, Penka I. Genova2
  Lab for Experimental Chemotherapy, Dept. of Pharmacology, Faculty of Pharmacy, Medical
University of Sofia, 2 Dunav St., 1000 Sofia, Bulgaria E-mail:
  Dept. of Mechanics, Technical University Sofia, IPF Sliven, Bulgaria

The ex vivo expansion of hematopoietic cells is a rapidly developing area with special
emphasize on bioreactor systems for up-scale of culture conditions. It is believed that a
rational design of bioreactors, especially those allowing microgravity could allow the
production of stem cells and will offer new approaches for studying the mechanisms of
proliferation, differentiation and signal transduction of cultured cells. The efficacy of two
commercially available bioreactors (rotating vessel miniPERM and static INTEGRA CL 350)
to support long term bone marrow cultures (LTBMCC) and three-dimensional growth of
Hodgkin’s lymphoma HD-MY-Z cells was investigated. In the miniPERM system growth of
spheroids (containing 30-40 cells) was obtained. An essentially higher content of
hematopoietic precursor cells (CFU-GM) was registered in the rotating vessel system. In this
bioreactor a growth of large HD-MY-Z spheroids (containing about 100-200 cells) was
achieved. The composed mathematical models of the physico-mechanical behavior of
spheroids enabled the evaluation of the revolution frequency increase schedule. The
differential equations took into account all inertial effects caused by the production module
rotation movement, as well as those caused by the relative movement of the spheroid in the
fluid. The models aimed the optimization of the rotation frequency increase schedule for
different type of cellular spheroids in order to reduce the shear stress and to augment the
productivity and to allow the growth of large three-dimensional spheroids. The models were
numerically tested by the MATLAB-SIMULINK-software and trajectories of pre-stained HD-
MY-Z spheroids were filmed. The coincidence of the theoretical and experimental trajectories
was sufficient. In conclusion, rotating vessel bioreactors that produce microgravity provide
better conditions for growth and support of hematopoietic cells and may have special
significance for studying of cellular interactions that regulate proliferation and differentiation.

                                              - 252 -
Session VII : Signal transduction mechanisms in health and disease       Poster VII, 21

Tfs1p, a homologue of the Raf-kinase inhibitor protein in yeast, is an activator of Ras
which binds in vitro the GTPase activating protein Ira2p

Gunasekaran Krishnasamy, Hélène Chautard, Charles Zelwer and Hélène Bénédetti

Centre de biophysique moléculaire, Centre national de la recherche scientifique, rue Charles
Sadron, 45071 ORLEANS Cedex 2, France. E-mail:

Tfs1p belongs to the PEBP family, a member of which is RKIP a kinase and serine protease
inhibitor and a metastasis inhibitor in prostate cancer. TFS1 is a multicopy suppressor of the
cdc25-1 mutation in yeast. Chautard et al.1 demonstrated that Tfs1p is a physiological
inhibitor of Ira2p, a GTPase activating protein (GAP) which is antagonistic to Cdc25p. Ira2p
is very similar to the human NF1p which is responsible of a genetic disease: Type 1
neurofibromatosis. Prior to start crystallographic studies of the complex, we have prepared
different recombinant domains (including the GAP) of Ira2p, and shown that they bind to the
recombinant Tfs1p.
Hélène Chautard, Michel Jacquet, Françoise Schoentgen, Nicole Bureaud and Hélène
Bénédetti, submitted.

                                            - 253 -
Session VII : Signal transduction mechanisms in health and disease          Poster VII, 22


KYRIAKOPOULOS A., Hoppe B., Bukalis K., Graebert A., Behne D.

Hahn-Meitner-Institut, Dept. “Molecular Trace Element Research
in the Life Sciences” Glienicker Str. 100, D-14109 Berlin, Germany.

Since the discovery of the essential trace element selenium which is necessary for the
prevention of several diseases, studies on selenium were mainly concentrated on the
identification of new biological active selenium compounds. Some selenoenzymes such as
glutathione peroxidases (GPXs) which are involved in the process for the protection of
biological membranes were already characterized. Information of further biological active
selenoproteins was obtained by labelling of rats in vivo with Se-75 and separation of the
proteins in the prostate, lung, thyroid gland, brain and colon homogenates by SDS-PAGE. In
this way, a 15 kDa selenium-containing protein was present in all investigated homogenates
after its detection by means of autoradiography. In order to obtain more information of its
subcellular distribution differential centrifugation was performed in the homogenate of the
organs such as prostate, lung, thyroid gland, brain and colon which are susceptible to
carcinogenesis. In addition, they contain a high amount of selenium because they produce
H2O2 for oxidative hormone synthesis and have to protect themselves from oxidative damage
by the expression of selenoperoxidases. After subcellular fractionation the 15 kDa selenium-
containing protein was found enriched in the cytosolic fraction of the tissues
prostate>lung>brain>thyroid gland> colon. After co-electrophoresis of the separated 15 kDa
labelled band obtained from each cytosolic fraction the 15 kDa labelled proteins migrated in
the same way like the combined bands isolated from the five tissue cytosols. After proteolytic
cleveage in the gel of the 15 kDa labeled band which was obtained from each tissue cytosol
and re-electrophoresis the same labelled peptide pattern was found in each gel slice after
autoradiography. This finding indicates that the 15 kDa selenium-containing protein is the
same protein in the cytosols of the examined tissues. In this paper, some important
characteristics are presented with regard to the similarities of the 15 kDa selenium-containing
protein in the different tissues and its function which has a protective role in the redox process
like the already known selenium glutathione peroxidases.

                                              - 254 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 23

Effect of Phenserine on Levels of Nitric Oxide Synthase in the APP/ PS1 Double
Transgenic Mice

D.K. Lahiri 1, Y.-W. Ge 1, M.R. Farlow 1, and N.H. Greig 2
Departments of Psychiatry and Neurology, Institute of Psychiatric Research, Indiana
University School of Medicine, Indianapolis, IN; 2National Institute on Aging, GRC/ NIH,
Baltimore, USA

Alzheimer's disease (AD) is characterized by a reduction in the levels of the potentially toxic
amyloid b-peptide (Ab), which is derived from b-amyloid precursor protein (APP). Familial
cases of AD are caused by mutations in APP and presenilin-1 (PS1) genes. PS1 mutations
were shown to perturb neuronal calcium homeostasis, increase Ab production, and enhance
the vulnerability of neurons to synaptic dysfunction, excitotoxicity, and apoptosis. Astrocytes
with elevated levels of NOS were observed in direct association with Ab-deposits in AD and
in APP transgenic (Tg) mice. However it is not fully understood whether Ab generation
directly induces apoptotic cell death or triggers an alternative pathway, such as nitric oxide
synthase (NOS) that then leads to neurodegeneration. Also, it is not known whether
cholinesterase inhibitors, the most common drug prescribed for AD, have any effect on NOS.
To address these questions, we compared levels of NOS in brain extracts between Tg mice
(expressing human APP with the Swedish double mutation plus PS-1) and non-transgenic
(nTg) mice. We tested the effect of Phenserine, a highly selective acetylcholinesterase
inhibitor under Phase II clinical trials for AD, on NOS in vivo. Levels of neuronal NOS
(nNOS) and inducible NOS (iNOS) were measured by western blotting in brain extracts from
both APP/ PS-1 double Tg and nTg mice of similar age and gender. In both mouse groups,
cerebellar extracts possessed the highest levels of NOS. There was no significant difference in
levels of nNOS or iNOS between Tg and nTg brain extracts. In mice administered
Phenserine, which lowered brain Ab levels, a reduction in NOS was observed from the
control group. Thus, Ab deposition may not directly contribute to the NOS-mediated nitric
oxide release, and treatment with an Ab lowering cholinesterase inhibitor has a reducing
effect on NOS. These results have important implications in the development of a suitable
immunotherapy for AD. This work is supported by grants from Alzheimer Association and
the National Institute on Aging, NIH, USA.

                                            - 255 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 24

Role of APP Gene Promoter in Apoptosis and Alzheimer’s Disease

D.K. Lahiri, and Y.-W. Ge

Laboratory of Molecular Neurogenetics, Department of Psychiatry, Institute of Psychiatric
Research, Indiana University School of Medicine, Indianapolis, Indiana-46202, USA

Apoptosis and DNA damage has been recognized in Alzheimer’s disease (AD). One of the
characteristic features of AD is a substantial loss of neurons in the hippocampus and cerebral
cortex and the loss has been attributed to apoptotic neuronal cell death, which is promoted by
several factors, such as excess depositions of the amyloid b peptide (Ab). Ab is derived from
the large Ab precursor protein (APP). Our hypothesis is that abnormalities in APP gene
regulation might be a crucial factor in AD neuropathology. To study gene regulation, we have
cloned and characterized the APP regulatory regions, including the promoter (-7900 bp from
the transcription start site, +1). We examined the role of specific APP promoter elements in
apoptosis. DNA transfection studies with APP promoter deletion constructs (-1100/+1)
indicate that each cell type differently regulates promoter activity under apoptotic conditions.
Results from the gel electrophoretic mobility shift assay (EMSA) indicate that the proximal
promoter region (-76/+1) forms a distinct DNA-protein complex with nuclear protein (s) from
both neuronal and other cell types. This DNA-protein interaction is regulated by both serum
starvation and hypoxia. These results suggest that specific APP genetic regulatory elements
participate in enhanced apoptotic neuronal cell death. We propose a model of interaction
between APP promoter and nuclear proteins to explain apoptosis and neurodegeneration in
relation to AD. This work is supported by a grant from the National Institute on Aging, NIH,

                                             - 256 -
Session VII : Signal transduction mechanisms in health and disease      Poster VII, 25

Soluble factors secreted by invasive melanoma cells can activate microvascular
endothelial cells

S. Löffek, P. Zigrino, C. Mauch
Department of Dermatology, University of Cologne, Germany.

Tumour invasion and metastasis of malignant melanoma have been shown to require
proteolytic degradation of the extracellular environment, which is achieved primarily by
enzymes of the matrix metalloproteinases (MMP) family. We have previously shown, that
increased enzyme activity is localised at the border of tumour cells and the adjacent
peritumoral connective tissue emphasising the crucial role of tumour-stroma interactions in
the regulation of MMP activity.
Cell surface receptor-mediated interactions of tumour cells with the surrounding structural
and cellular components of the stroma and soluble factors released by the tumour cells are
likely to contribute to the activation of stromal cells, that is associated with increased
proteolysis of the matrix. To better understand the molecular mechanisms involved not only
in the invasion of the dermal compartment but also in the intravasation of human melanoma
leading to metastasis we have analysed the influence of melanoma cells on microvascular
endothelial cells in co-culture systems.
We have previously shown that in vitro cross-talk between invasive melanoma cells and
stromal fibroblasts is mediated by soluble factors such as IL1alpha and bFGF released by the
tumour cells in high amounts.
In response to these stimuli, dermal fibroblasts express and activate different matrix
metalloproteinases. In contrast, growth of melanoma cells in two dimensional co-cultures
with human dermal microvascular endothelial cells (HDMEC), either in direct or indirect
contact, failed to show any changes in the expression or activation of MMPs. Instead, when
endothelial cells were cultured in three-dimensional fibrin gels, addition of supernatants
collected from melanoma cells induced formation of tubular structures similar to those
observed upon treatment of endothelial cells with an angiogenic cocktail. These findings
indicate that soluble factors derived from melanoma cells mediate neovascularisation by an as
jet unclear mechanism. Alltogether, these data clearly demonstrate the importance of the
interaction of tumour cells with surrounding resident as one of the critical steps in tumour

                                           - 257 -
Session VII : Signal transduction mechanisms in health and disease       Poster VII, 26

NF-kB and Signalling Pathway Activated by Glycoprotein D of HSV-1 that Suppresses
Fas-mediated Apoptosis

M. Teresa Sciortino, M. Antonietta Medici, Donata Perri, Vincenza Valveri,
A. Pia Camuti, Sandro Grelli 1, Antonio Mastino

Dept. of Microbiological, Genetic and Molecular Sciences, University of Messina, Sperone
31, 98186 Messina, Italy; e-mail: 1Dept. of Exp. Medicine and
Bioch. Sciences, University of Rome “Tor Vergata”, 00133 Rome, Italy.

The signalling pathway utilised by herpes simplex virus 1 (HSV-1) to regulate cell viability
during the early phase of infection in semi-permissive U937 monocytoid cells, was
investigated. HSV-1 promptly activated nuclear factor kappa B (NF-kB) after infection. UV-
inactivated virus, cell-to-cell contact with stable transfectants expressing the envelope
glycoprotein D (gD) of HSV-1 on their surface as well as soluble gD by itself, exerted a
similar, NF-kB-activating, effect. Moreover, the same stimuli led to a protective effect on
Fas-induced apoptosis. Protection against apoptosis was abrogated in NF-kB-dominant-
negative stable transfectants (mIkBa)) but not in control transfectants. Exposure to UV-
inactivated viral particles induced the up-regulation of c-FLIP, c-IAP2 and Survivin in control
transfectants but not in mIkBa transfectants. In addition: a) antibodies able to specifically
interfere with interactions between gD and the cell receptor for gD HveA/HVEM/TNFRSF14,
which was fully expressed in U937 cells, neutralized inhibition of Fas-mediated apoptosis by
soluble forms of gD, b) cell-to-cell contacts with transfectants expressing on their surface
mutated forms of gD lacking the capability to interact with HveA, did not protect against Fas-
mediated apoptosis. Our results form a solid foundation for establishing a new pathway of
signals, starting from gD/HveA interaction, continuing with NF-kB activation and arriving at
inhibition of apoptosis through the action of anti-apoptotic NF-kB-responsive genes. The
definition of this signalling pathway contributes to the understanding of the complex
HSV/cell interactions in the very-early phase of infection. Moreover, applicative implications
could be expected.

                                            - 258 -
Session VII : Signal transduction mechanisms in health and disease    Poster VII, 27


Krista Rombouts, Benedetta Lottini, Alessandra Caligiuri, Maurizio Parola*, Fabio Marra,
Vinicio Carloni, Massimo Pinzani.

*Dipartimento Medicina e Oncologia Sperimentale, sez. Patologia Generale, Università di
Torino, Italy and Dipartimento di Medicina Interna, Università degli Studi di Firenze,
Florence, Italy.

Myristoylated Alanine-rich protein kinase C substrate (MARCKS), is one of the most
prominent intracellular substrates for PKC. It has been suggested that MARCKS is implicated
in cell motility, secretion, membrane trafficking and mitogenesis through regulation of the
membranous actin cytoskeletal structure. Previously, we have shown by using
immunofluorescence microscopy, co-immunoprecipitation and Western Blot analysis a cross-
talk of MARCKS with the PDGFb-R in human hepatic stellate cells (HHSC) a cell type
contributing to hepatic fibrogenesis. PDGF-BB is known as a strong
chemoattracttant/mitogen for HHSC.
Since PDGF-BB induced a translocation of MARCKS from the membrane to the cytosol and
the knowledge that PDGF-BB initiates a re-organization of the filamentous actin
cytoskeleton, we have started to explore molecular mechanisms involving MARCKS in the
PDGF-BB signalling pathway. AG1296, a PDGFb-R autophosphorylation inhibitor and two
PKC inhibitors, GO6850 and GO6976, were used to investigate the phosphorylation of
MARCKS at the protein level. PDGF-BB-induced chemotaxis was investigated by
transfecting HHSC with a peptide containing the phosphorylation site domain of MARCKS.
Furthermore, Bile Duct Ligation was performed in rats, to induce liver fibrosis in vivo and
thereafter MARCKS protein expression was investigated.
PDGF-BB did not induce MARCKS phosphorylation when intrinsic PDGFb-R kinase activity
and PKCe were inhibited; moreover, PDGF-BB-induced migration was inhibited by pre-
treating HHSC with the PKCe inhibitor. Further analysis indicated that Ser5 and Ser9 of
MARCKS are important in PDGF-BB-induced migration of HHSC. Finally, Western blot
analysis showed a strong up-regulation of MARCKS protein expression in animals in which
liver fibrosis was induced by Bile Duct Ligation.
In conclusion, we report that PDGF-BB-induced phosphorylation of MARCKS involves
autophosphorylation of the PDGFb-R and PKCe isoform. Furthermore, Ser5 and Ser9 of the
phosphorylation site domain of MARCKS are involved in the PDGF-BB-induced chemotaxis
of HHSC. These results further expand the knowledge on the link between PDGF signalling
and cytoskeletal re-organization indispensable for PDGF biologic effects in HHSC.

                                           - 259 -
Session VII : Signal transduction mechanisms in health and disease         Poster VII, 28

Analysis of the signal transduction pathway leading to IRF-1 up-regulation by HIV-1

Sgarbanti M., Marsili G., Remoli A. L., Ridolfi B., Borsetti A., Ensoli B. and Battistini A.

Laboratorio di Virologia, Istituto Superire di Sanita’, Viale Regina Elena, 299- 00161 Roma
Italy. E-mail:

Interferon (IFN) regulatory factors (IRFs) constitute a family of transcriptional activators and
repressors involved in the regulation of immune responses, host defence, and cell growth
regulation. All members share conserved DNA binding domains that recognize DNA
sequences termed IRF-binding elements/IFN-stimulated response elements (IRF-E/ISRE)
present on the promoter of IFN-a/b and IFN-stimulated genes (ISGs). Since to achieve an
efficient level of replication human immunodeficiency virus type 1 (HIV-1) need to stimulate
transcription from its 5’ long terminal repeat (LTR) before the viral trans-activator Tat is
produced, the virus has hijacked cellular transcription factors for this purpose. Our previous
data showed that among the IRF factors only IRF-1 is able to stimulate the HIV-1 LTR
transcription and that HIV-1 induces IRF-1 early upon infection and prior to expression of Tat
in both T-cell lines and primary CD4+ T cells. Key IRF-1 promoter elements which mediate
activation of transcription upon induction by inflammatory cytokines are IFN-g-activated
sequences (GAS) which binds members of the STAT family as well as binding sites for NF-
kB. In order to find signal transduction pathway components of either STAT or NF-kB as
therapeutic targets to block IRF-1 mediated HIV-1 LTR transcriptional activation, we tried to
define the pathways involved in the HIV-1 induced IRF-1 expression. For this purpose we
generated Jurkat T cells expressing an NF-kB super-repressor IkB-a 2ND4) that blocks
the ability of NF-kB to activate transcription. Preliminary results show a dramatic decrease in
HIV-1 replication upon de novo infection of Jurkat cells expressing the dominant negative
IkB-a 2ND4 compared to control cells, measured by the p24 HIV-1 protein accumulation.
We have previously shown that IRF-1 is able to bind the kB sites on the HIV-1 LTR and to
form a complex with NF-kB on this site resulting in an additive stimulation of transcription.
In this respect the phenotype of Jurkat cells expressing the NF-kB super-repressor may be at
least partially explained by the lack of both IRF-1 and NF-kB induction. Experiments are
ongoing in order to determine whether the blockade of the NF-kB pathway also results in the
inhibition of the HIV-1 mediated IRF-1 stimulation and of the IRF-1 mediated HIV-1

                                             - 260 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 29

Differential gene expression of regenerative and fibrotic pathways in canine hepatic
portosystemic shunt and portal vein hypoplasia.

Bart Spee1, Louis C. Penning1, Ted S.G.A.M. van den Ingh2, Brigitte Arends1, Jooske IJzer2,
Freek J. van Sluijs1, and Jan Rothuizen1.

From the 1Department of Clinical Sciences of Companion Animals, Faculty of Veterinary
Medicine, Utrecht University, The Netherlands; and the 2Department of Pathology, Faculty of
Veterinary Medicine, Utrecht University, The Netherlands.

Chronic liver disease may result in decreased regeneration and fibrosis. We have analyzed
two spontaneous dog models characterized by subnormal portal perfusion resulting in absence
of normal liver growth without fibrosis or other pathology (congenital portosystemic shunts,
CPSS) or in reduced growth in association with portal fibrosis (primary portal vein
hypoplasia, PPVH). In these pathologies, inflammation or cholestasis is not present so that
they represent simplified models to study liver growth and fibrosis. We used quantitative real-
time polymerase chain reaction (Q-PCR) to analyze the mRNA expression of hepatocyte
growth factor (HGF), transforming growth factor _1 (TGF-_1), and relevant down stream
mediators in liver tissues of dogs with CPSS or PPVH. CPSS and PPVH were associated with
a threefold decrease (P < 0.05) in mRNA expression of HGF, and two- and threefold decrease
in expression of c-MET, respectively, compared to control liver samples. These results show
that liver regeneration capacity is severely inhibited in CPSS and PPVH. The TGF-_1
pathway was unchanged in CPSS but 2-fold increased in PPVH indicating an active TGF-_1
pathway, consistent with the portal fibrosis, as seen in PPVH. Western blots on
phosphorylated Smad 2 confirm an activated fibrotic pathway in PPVH. In conclusion, our
results demonstrate that CPSS and PPVH in the dog are important models for unraveling the
pathogenic pathways of decreased hepatic growth, regeneration, as well as hepatic fibrosis.

                                            - 261 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 30


Tanner, T., Haelens, A., Callewaert, L., Christiaens, V., Schauwaers, K., Verrijdt, G. and
Claessens, F.

Androgens, the male sex hormones, are responsible for the development and maintenance of
the male reproductive system. They enter the cell by passive diffusion through the cell
membrane and bind to the androgen receptor (AR). The ligand-bound receptor becomes
activated and translocates to specific DNA motifs called androgen response elements (AREs)
in the regulatory regions of androgen-responsive genes, resulting in their transcription
activation. On the one hand, there are androgen-selective AREs, which can only be bound by
the AR and not by the other steroid receptors. On the other hand, there exist non-selective
AREs, which can be bound by all steroid receptors.
The AR is a member of the nuclear receptor family and has a modular structure. It is
composed of a long N-terminal region with a transactivation function AF-1, a central DNA-
binding domain, an intermediate hinge region (HR) and a C-terminal ligand-binding domain
with an additional trans-activation function AF-2.
Initially, the HR was considered as a non-functional linker between the DNA-binding domain
and the ligand-binding domain. However, recent results indicate that the HR is a
multifunctional region involved in DNA-binding, nuclear localisation and modulation of the
transactivation. The HR is also a phosphorylation and acetylation target-site and is an
interaction domain for several proteins. Two mutated amino acids residues in the HR were
found in prostate cancer patients.
In this study we investigated more in detail the role of the HR in the transactivation of the AR
by a functional mutation analysis. A deletion mutant of the AR was created by PCR and the
effect was evaluated in transient transfection experiments by comparing the transcriptional
activity of the mutated and wild type receptor.
Deletion of the C-terminal part of the HR between aa 628 and aa 646 (DHR) results in an
increased transcriptional activation when using non-selective ARE-reporter constructs in
COS-7 cells, but not with androgen-selective reporters. This effect is not due to a difference
in expression since both the wild type and mutant AR are expressed equally. A disturbed
DNA binding by DHR, as observed in band shifts, might be an explanation. However, in Hela
cells this superactivation by the DHR is also observed for the androgen-selective AREs.
We can conclude that the HR of the AR limits the potential transcriptional activity in a
promoter- and cell type-specific way. The discrepancy between the androgen-selective and
non-selective AREs is an indication for an alternative use of cofactors between these two
types of AREs.

                                             - 262 -
Session VII : Signal transduction mechanisms in health and disease          Poster VII, 31

The tumor suppressor gene Zac1 is a downstream target of somatostatin receptor
signaling in pituitary tumor cells.

Theodoropoulou M., Stalla G.K, Pagotto U.*

Max Planck Institute of Psychiatry, Munich, Germany; Endocrine Unit, and C.R.B.A.,
S.Orsola-Malpighi Hospital, Bologna, Italy *.

Somatostatin is a neuropeptide which inhibits multiple functions, such as, hormone secretion,
angiogenesis, cell proliferation and tumor progression. Octreotide is one of the most
commonly used somatostatin analogues in the treatment of neuroendocrine tumors. It acts
mainly through somatostatin receptor 2, which belongs to G-protein coupled receptor
superfamily. Although the major signaling events taking place after ligand-receptor binding
are under intensive investigation, not much is known about the gene-targets of octreotide’s
antiproliferative signaling. In this study, we used the rat pituitary tumor cell line GH3 to study
the effect of octreotide on Zac1 gene expression. ZAC/Zac1 is a tumor suppressor gene which
is highly expressed in normal pituitary, mammary, and ovarian glands, but downregulated in
pituitary, breast, and ovarian tumors. Octreotide induced Zac1 gene expression, in a
concentration dependent manner, 6 hours after stimulation and the effect remained for 24
hours thereafter. Pre-treatment with pertussis toxin or with the tyrosine phosphatase inhibitor
orthovanadate completely abolished this effect, indicating involvement of Gi(alpha) subunit
and a phosphotyrosine phosphatase, respectively. Both phosphotyrosine phosphatases that are
involved in somatostatin’s signaling, i.e. SHP-1 and SHP-2, were found in GH3 cells.
Inhibiting MEK with PD098059 had no effect on octreotide’s action. Treating GH3 cells with
the PI3K inhibitor wortmannin increased the basal Zac1 levels, implying that Zac1 may lay
downstream to the PI3K/Akt pathway. In conclusion, this study presents Zac1 as a
downstream target of octreotide’s signaling and implicates it as a mediator of octreotide’s
antiproliferative action.

                                              - 263 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 32

The Wnt/Frizzled/Dishevelled pathway is involved in development of cardiac
hypertrophy in mice

V. van de Schans1; A. Strzelecka1; A. Wynshaw-Boris2; J. Smits1; W.M. Blankesteijn1

CARIM, Maastricht University; The Netherlands; 2 HHMI, UCSD, USA

Cardiac hypertrophy is a cellular response to a number of stimuli that create an overload of
the heart. Prolongation of this process can lead to heart failure. The goal of this study was to
evaluate the role of the Wnt/Frizzled/dishevelled (Wnt/Fz/Dvl) pathway in cardiac
hypertrophy development. Wnt stimulation acts to stabilize _-catenin levels by binding to the
Fz2 receptor, thereby activating the Dvl protein which antagonizes the action of GSK-3_, a
potent suppressor of the hypertrophic response. Stabilized _-catenin translocates to the
nucleus, where it activates both Tcf/Lef signaling and transcription of _-catenin target genes,
including the proto-oncogenes c-fos and c-myc.
Mice lacking the Dvl-1 gene (Dvl-/-) and wild-type littermates (WT) were subjected to left
ventricular pressure overload by transverse aortic constriction (TAC). At day 7 after TAC,
there was an upregulation of Fz2 mRNA in both Dvl -/- and WT mice. Compared to WT
mice, Dvl-/- mice developed reduced ventricular hypertrophy, as measured by
echocardiography of the left ventricular posterior wall thickness in systole and in diastole.
This was accompanied by lower expression of ANF, a marker of cardiac hypertrophy, in Dvl-
/- vs. WT. Western blot analysis showed higer protein levels of GSK-3_ and Q-PCR analysis
showed a reduction in the expression of the proto-genes c-fos and c-myc in Dvl -/- mice
compared to WT.
In summary, disruption of the Dvl-1 gene reduces the development of cardiac hypertrophy
induced by pressure overload and thus demonstrates a role for the Wnt/Fz/Dvl pathway in
hypertrophy development. A higher GSK-3_ protein expression and reduced induction of c-
fos and c-myc suggests that these proteins play a role in this attenuated response.

                                             - 264 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 33

c-Jun N-terminal kinases (JNKs) participate in the activation of microglia

Vicki Waetzig (1), Ute Hidding (1), Karen Czeloth (1), Stephan Brecht (1), Ralph Lucius (2),
Uwe Hanisch (3), and Thomas Herdegen (1)

(1) Institute of Pharmacology, Hospitalstrasse 4, 24105 Kiel, Germany, (2) Institute of
Anatomy, Olshausenstrasse 40, 24098 Kiel, Germany, (3) Max-Delbrueck-Center, Robert-
Roessle-Strasse 10, 13122 Berlin, Germany

The activation and function of c-Jun N-terminal kinases (JNKs) were investigated in primary
microglia from neonatal rats. We have shown before that microglia expressed all JNK
isoforms including the so-called neuronal JNK3. Lipopolysaccharide (LPS), tumor necrosis
factor alpha (TNFa) or thrombin were potent activators of JNKs with subsequent nuclear N-
terminal phosphorylation of c-Jun. Surprisingly, untreated microglia revealed a substantial
pool of activated JNKs in the nucleus (indicating constitutively active JNK) which declined
following LPS stimulation, while the amount of JNK2 in particular increased. Incubation with
1-5 µM SP600125, a direct inhibitor of JNKs, and 1-2 µM CEP-11004, an inhibitor of MLKs
and an indirect inhibitor of JNKs, attenuated (i) the cellular enlargement via activation of c-
Jun and (ii) the proliferation of microglia independently of inhibition of MLKs and c-Jun.
Finally, inhibition of JNKs significantly reduced the expression of COX-2, interleukin-6 (Il-6)
and TNFa as well as the total amount of LPS-induced release of TNFa and interleukin-6.
Thus, JNKs are essential players of fundamental and pro-inflammatory functions in microglia.
In consequence, experimental or therapeutic antagonization of JNKs in the brain does not
merely affect neurons, but also interferes with important pathophysiological actions of

                                            - 265 -
Session VII : Signal transduction mechanisms in health and disease   Poster VII, 34

MyD88 as target sturcture for immune intervention in polymicrobial septic peritonitis

Heike Weighardt , Simone Kaiser-Moore , Gabi Jusek , and Bernhard Holzmann

Department of Surgery, Klinikum rechts der Isar, Technische Universität München,
Ismanninger Str. 22, 81675 Munich, Germany;

Toll-like receptors (TLRs) are important for the activation of innate immune cells upon
encounter of microbial pathogens. The present study investigated the role of the central
signaling adaptor protein MyD88 in polymicrobial septic peritonitis. MyD88-deficiency
resulted in improved survival of mice. MyD88-deficient mice showed a strongly attenuated
systemic hyperinflammatory response, whereas the antibacterial defense mechanisms were
not impaired. Gene expresssion profiling of spleens and livers of septic MyD88-deficient
mice revealed differences in the requirement of MyD88 for the local production of immune
mediators during polymicrobial sepsis. Whereas MyD88-deficiency strongy reduced the
sepsis induced gene expression in the liver, gene expression in the spleen was largely
independent of MyD88. Expression of the MyD88 independently regulated genes MCP-1,
MCP-5, IFN-b and IP-10 could be detected in both, livers and spleens of septic MyD88
deficient mice. Additionally, we observed a subset of immune response genes to be
selectively upregulated in MyD88-deficient but not in wildtype mice.
These results imply a central role of MyD88 for the systemic immune pathology of
polymicrobial sepsis. The regulation of MyD88-dependent and MyD88-independent signaling
pathways may provide new insights into the innate immune response to severe infection.
MyD88 may serve as target structure for the development of new drugs for reprogramming
pathophysiological immune responses during sepsis.

                                           - 266 -
Session VII : Signal transduction mechanisms in health and disease      Poster VII, 35

Rat preadipocyte differentiation is regulated by glucocorticoid and TGF-b

Young Yang, Sun Mi Shin, Kun-yong Kim, Jae Kwang Kim, Inpyo Choi

Laboratory of Immunology, Korea Research Institute of Bioscience and Biotechnology,
Daejeon 305-333, Korea.

Dexamethasone and TGF-b show contrary effects on differentiation of adipocytes.
Dexamethasone stimulates adipocyte differentiation whereas TGF-b inhibits it. In the present
study, we investigated whether dexamethasone could reverse the TGF-b-mediated inhibition
of preadipocyte differentiation. Primary rat preadipocytes, obtained from SD rats, were
pretreated with dexamethasone in the presence or absence of TGF-b, prior to the induction of
differentiation. Co-treatment of dexamethasone and TGF-b before inducing differentiation
reversed the TGF-b-mediated inhibition of preadipocyte differentiation. In order to elucidate
the mechanism by which dexamethasone reversed the effect of TGF-b on the inhibition of
preadipocyte differentiation, the expressions of C/EBPa and PPAR-g were examined.
Dexamethasone increased the C/EBPa and PPAR-g expressions in the absence of TGF-b and
also recovered the TGF-b-mediated suppression of C/EBPa expression in preadipocytes. Its
effect was sustained in differentiated adipocytes as well. However, those effects were not
observed in 3T3-L1 preadipocytes or differentiated adipocytes. These results indicate that
dexamethasone reverses the TGF-b-mediated suppression of the adipocyte differentiation by
regulating the expressions of C/EBPa and PPAR-g, which is dependent on the cellular
contexts (This research was supported by a grant (CBM1-B114-001-1-0-0) from the Center
for Biological Modulators of the 21st Century Frontier R&D Program, the Ministry of
Science and Technology, Korea.)

                                           - 267 -
Session VII : Signal transduction mechanisms in health and disease        Poster VII, 36

Computer Simulation of NGF Induced TrkA Signal Transduction Pathway in PC12

Sertan Yilmaz

Bogazici University, Institute of Biomedical Engineering, Bebek Istanbul 34342 Turkey. E-

Levent Kurnaz

Bogazici University, Department of Physics, Bebek Istanbul 34342 Turkey. E-mail:

NGF-stimulated TrkA activates a mitogenic response in non-neuronal cells. However, the
same combination results in differentiation in cells of neuronal lineage. Not only the
activation of the NGF-stimulated TrkA pathway, but also the duration of this activation is
crucial for the biological input generated. In PC12 cells it has been experimentally shown that
NGF-stimulated TrkA causes mitogenesis through the Shc/Grb2/SOS/Ras/Raf1 pathway and
leads to differentiation through the FRS2/SHP2/Crk/C3G/Rap1/B-Raf pathway. In the present
study, the effects of blocking/activation of the Shc/Grb2/SOS/Ras/Raf1 pathway and the
FRS2/SHP2/Crk/C3G/Rap1/B-Raf pathway on the cellular response are investigated through
computer simulations. Some kinetic parameters needed for these pathways are obtained from
the literature. However, many of the kinetic parameters needed for these simulations are
estimated to match the experimental results and their ranges of validity are determined.

                                            - 268 -
          Participant list

- 269 -

To top