Transport Machineries
B1-1
Osteopontin expression in polarized MDCK cells
N. Tascene
1
and Z. S. Isguder
2
1
Ankara University Faculty of Veterinary Medicine, Ankara,
TURKEY,
2
University of Veterinary Medicine, Department of
Physiological Chemistry, Hannover, GERMANY
Osteopontin(OPN) is an arginine-glycine-aspartate (RGD)-contain-
ing adhesive glycoprotein that was first identified as a major sialo-
protein in bone and subsequently found to be expressed in kidney,
brain, macrophages, vascular smooth muscle cells and many cells
of epithelial linings. In this study we examined the expression of
OPN in MDCK cells in different cellular confluences. The cells
were examined at very low density, half confluence or complete
confluence. OPN levels were investigated by western blotting and
RT-PCR. An increase in OPN expression was observed due to the
increasing confluency and subsequent initiation of polarization.
Expression profiles of flotillin-2 in the same cells was used as a
control, since this protein is ubiquitously produced in MDCK cells
and its expression rates are independent of confluence and/or
polarization. Intracellular distribution of OPN was also monitored
by confocal microscopy on preparations immunolabeled with anti-
OPN antibodies. Staining patterns have also confirmed increased
OPN expression during confluency in MDCK cells.
We conclude that the expression of OPN in polarized MDCK cells

is induced by increasing confluency and polarization.
Key words: Osteopontin, polarization, MDCK cells.
B1-2
Biogenesis of VGF se cretory granules in
epithelial thyroid cells
F. Gentile
1
, A. Corteggio
2
, S. Avitabile
2
, G. Calı
`
1
and L. Nitsch
2
1
IEOS, CNR, Napoli, ITALY,
2
University Federico II, Napoli,
ITALY
VGF is a granin that is sorted to a regulated and polarized path-
way of secretion when expressed in FRT epithelial cells. We are
investigating the role of aggregation and interaction with lipid rafts
in VGF protein sorting. We demonstrated that a fraction of the
intracellular protein is associated to GM1 rafts and that in a low
pH and high calcium buffer VGF aggregates. Since only the intra-
cellular protein, not the one secreted in the culture medium, is cap-
able to aggregate, we hypothesized that VGF aggregation might be
due to its interaction with some other molecule. We have excluded

that the interaction of VGF to GM1 rafts is relevant for VGF
aggregation since it is not affected by the inhibition of glyco-
sphyngolipid synthesis. We have then looked for proteins capable
to interact with VGF by gel filtration and by SDS-PAGE analysis
of proteins that co-aggregated with VGF. We found a 40 kDa pro-
tein that is co-eluted with VGF and we demonstrated that it is not
flotillin 2, which also forms aggregates but has an intracellular dis-
tribution distinct from that of VGF. FRT-VGF cells were treated
with bafilomycin A1 that alters the proton gradient of the secre-
tory compartment. A dramatic reduction in VGF secretion and
granule formation, and lack of response to PMA were observed.
The secretion of thyroglobulin, that we have stably expressed in
FRT cells and that is also secreted through the apical cell domain
but by a constitutive pathway, was unaffected.
Overall our data support the hypothesis that VGF aggregation
does not depend on rafts association and possibly occurs in the
presence of a 40 kDa protein. Proton gradient perturbation affects
granule formation and VGF secretion.
B1-3
Cellular trafficking of cell penetrating peptides
and protein transport over plasma membrane
A. Lorents
1
, K. Padari
2
, M. Hansen
3
,U
¨
. Langel

3
and M. Pooga
4
1
Institute of Molecular and Cell Biology, University of Tartu, Tartu,
ESTONIA,
2
Institute of Zoology and Hydrobiology, University of
Tartu, Tartu, ESTONIA,
3
Institute of Neurochemistry, Stockholm
University, Stockholm, SWEDEN,
4
Estonian Biocentre, Tartu,
ESTONIA
Cell penetrating peptides (CPP) consisting of up to 30 amino acids
are capable of translocation into cells and delivering different car-
goes. Despite numerous efforts the internalization mechanism of
CPPs has not been fully understood. The cellular uptake of CPPs is
dependent on the properties and concentration of peptide, cellular
energy and lipid-rafts of plasma membrane. Binding to the cell sur-
face proteoglycans, followed by endocytic uptake is considered as
the main mechanism of highly cationic CPPs’ cell-entry. However,
the type of endocytosis induced by CPPs, further intracellular target-
ing, and liberation from vesicular structures are still under debate.
We assessed the protein delivery by CPPs in suspension cells K562
by flow cytometry and fluorescence microscopy. Among the studied
peptides Tat and transportan showed highest protein transduction
efficiency. The CPP-protein complexes localized in K562 cells both
in vesicles and diffusely in the cytoplasm as demonstrated by fluores-

cence microscopy. The inhibitors of macropinocytosis and cellular
metabolism decreased the uptake of complexes by K562 cells in a
concentration dependent manner, indicating that CPP-avidin con-
structs are taken up mostly by energy-dependent mechanisms, invol-
ving different types of endocytosis. The cellular trafficking of CPP-s
in HeLa cells was mapped on ultra-structural level by transmission
electron microscopy by using nanogold-labelled peptides.
B1-4
Cellular uptake and subcellular distribution of
Ap
n
A family of signaling molecules
R. Pe˛cherzewska, A. D. Krakowiak, J. Kaz´ mierczak,
M. Maszewska and W. J. Stec
CMMS PAS, Lodz, POLAND
Diadenosine polyphosphates are recognized as intra- and extracel-
lular signaling molecules and considered as the class of new second
messengers. Biochemical function of dinucleoside tri- and tetra-
phosphates has not been fully elucidated yet [1]. It is known that
Ap
3
A is cleaved by specific enzymes of hydrolases family. One of
them is the tumor suppressor Fhit protein. It is assumed that for-
mation of the FhitAp
3
A complex, recognized as a signaling mole-
cule induces cell apoptosis [2].
The aim of presented studies is to determine the cellular uptake of the
Fhit protein substrates. For our experiments we used fluorescent ana-
logs of Ap

n
A. Thus, ApppBODIPY and GpppBODIPY containing
the fluorophore residue (BODIPY) instead of one adenosine moiety
as well as an analog of inhibitor of Fhit protein [3] conjugated with
Alexa Fluor 350 have been used. Transfections were done with Lipo-
fectamine 2000. Intracellular distribution of fluorescent analogs of
diadenosine triphosphate was visualized by fluorescence microscopy.
It has been found that both substrates with fluorescent residue,
ApppBODIPY and GpppBODIPY, accumulate in cytoplasmic
compartments and in the nucleoli of the transfected cells. Detailed
results will be presented.
Acknowledgement: This project is financially assisted by the
grant 2PO4A07929 (to A.K.).
References
1. Kisselev L. et al., FEBS Lett. 427: 157–63 (1998).
2. Brenner C. Biochemistry 41: 9003–9014 (2002).
3. Varnum J.M. et al., BMC Chem Biol, 1(1):3, (2001).
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 97
B1-5
Characterization of heparan sulfate as a cell
surface attachment molecule for human
eosinophil ribonucleases
S. C. Lin, T. C. Fan and M. D. T. Chang
Institute of Molecular and Cellular Biology & Department of Life
Science, National Tsing Hua University, Hsinchu, TAIWAN
The eosinophil appears to be the primary leukocyte responsible for
tissue damage in bronchial asthma, which occurs when the granule
proteins, including eosinophil cationic protein (ECP) and eosino-
phil derived neurotoxin (EDN), are released into the extracellular

space. ECP and EDN belong to the ribonuclease A superfamily.
Both have been suggested as factors in allergic respiratory diseases,
and thus used as clinical biomarkers for detecting the severity of
asthma. Interestingly, ECP and EDN showed cytotoxicity toward
several cell lines, including HL60 and A431 cells. The cytotoxicity
was positively correlated with the internalization properties of the
RNases, but the exact route by which eosinophil RNases enter
cells has not been clarified. Recently, we hypothesized that ECP
and EDN may internalize target cells through binding to the cell
surface heparan sulfate (HS). In this study, we have investigated
the interaction between ECP/EDN and HS using surface plasmon
resonance instrument (SPR) technology. On the other hand, amino
acid sequence analysis reveals that EDN contains one putative
heparin-binding motif of the type XBBXBX, whose effects on HS
binding has been investigated by site-directed mutagenesis and
binding assays. Our results indicate that the heparin-binding region
plays a critical role in recognition and internalization of eosinophil
RNases.
B1-6
A heparan sulfate-facilitated and
raft-dependent macropinocytosis of eosinophil
cationic protein
T. Fan, S. Lin and M. D. T. Chang
National Tsing Hua University, Hsin-chu, TAIWAN
Eosinophils secrete several basic proteins, among which the eosino-
phil cationic protein (ECP) has been used as bio-markers for the
severity of asthma. ECP, which belongs to the human RNaseA su-
perfamily, is released from activated eosinophils during the inflam-
mation process and contributes to eosinophils’ helminthotoxic,
bactericidal and antiviral activities. The cytotoxicity of ECP is clo-

sely associated with its efficient endocytosis into target cells. We
have previously discovered that CPE functions as a cellular receptor
for ECP binding to cells. The exact endocytic pathway taken by
ECP has not yet been deciphered. In particular, whether proteins of
the RNase family are internalized by a raft-associated pathway is
unknown. In this study, some key elements regulating the early
events of internalization of ECP were discovered by characteriza-
tion of binding, uptake and intracellular trafficking of ECP in
human bronchial epithelial cells. Uptake and cytotoxicity of ECP
in glycosaminoglycan-deficient cells were significantly reduced.
After associating with cell surface heparan sulfate, ECP was found
to rapidly internalize through detergent-resistant lipid rafts in a
clathrin- and caveolin-independent fashion, followed by trafficking
from early endosomes to late endosomes.
B1-7
Activation of the ubiquitin ligase Itch following
treatment with the epidermal growth factor
B. A. Azakir and A. Angers
University of Montreal, Montreal, PQ, CANADA
Itch is an ubiquitin ligase localized at the endosomes and implica-
ted in the regulation of endocytosis. Many substrates of Itch have
been identified, including the endocytic proteins endophilin and
cbl, p73, Smad-7 and others. Here we show that stimulation with
EGF increases cbl, endophilin and Smad-7 ubiquitination mediated
by Itch and that this ubiquitination is sufficient for the degradation
of these proteins in the proteasomes. Moreover, Itch itself is regu-
lated by EGF treatment, Itch ubiquitination increasing transiently
after treatment. However, Itch is not degraded in the proteasomes,
but deubiquitinated by the ubiquitin protease FAM. It has recently
been reported that Itch can be phosphorylated by the c-Jun N-ter-

minal kinase (JNK), and that this resulted in a significant increase
in the ligase activity. It is known that JNK can also be activated
by EGF treatment. Likely, we have found that Itch phosphoryla-
tion is increased after treatment with EGF and that inhibition of
JNK blocks this effect. Remarkably, Itch substrate ubiquitination
after EGF treatment is also reduced after JNK inhibition. As Itch
phosphorylation influences its capacity to interact with its sub-
strates, we examined the interactions of Itch with endophilin, cbl,
and FAM after EGF treatment. We show that Itch interaction
with Cbl and endophilin increases after treatment with EGF, but
interaction with FAM decreases. These results explain the transient
ubiquitination of Itch and the sustained ubiquitination of its sub-
strates after treatment with EGF. We conclude that Itch activity
and stability is regulated by EGFR via JNK and that this regula-
tion affects the ubiquitination and the degradation of Itch sub-
strates, especially cbl and endophilin, which could in turn regulate
receptor internalisation.
B1-8
Identification of a novel Musk binding protein
with a potential role in intracellular trafficking
processes
B. Woller and R. Herbst
Medical University Vienna, Vienna, AUSTRIA
The muscle-specific kinase MuSK plays an essential role during the
formation of the neuromuscular junction. MuSK is activated by
the nerve-derived protein agrin, a process that is essential for all
known aspects of postsynaptic and presynaptic differentiation. The
steps that follow MuSK activation and lead to acetylcholine recep-
tor clustering are not fully understood. Therefore we have a high
interest in the identification and characterization of new MuSK

binding partners.
Using a mouse muscle cDNA library we performed a yeast-two-
hybrid screen with the cytoplasmic region of MuSK. Thereby we
have identified a protein that specifically interacts with MuSK and
represents a novel gene, termed Cl-II6. Current data suggest that
the newly identified protein is a new member of the RIN protein
family, that functions as GEFs for Rab5. We therefore propose
that the novel protein is able to link MuSK to the endocytic
machinery via Rab5. When a GFP-tagged form of Cl-II6 is
expressed in heterologous cells, a punctuated intracellular staining
is observed that indicates localization in vesicles or endosomes.
Co-expression experiments revealed a precise co-localization of
MuSK and Cl-II6. In addition we could show that the protein is
localized at the neuromuscular junction. Ongoing experiments con-
centrate on the functional analysis of the newly identified MuSK
binding protein with particular emphasis on its potential role as a
protein involved in MuSK trafficking.
Abstracts Transport Machineries
98 ª 2007 The Authors Journal compilation ª 2007 FEBS
B1-9
Motor protein KIFC5A and its interacting
proteins Nubp1&Nubp2 are involved in the
regulation of centrosome duplication
A. Christodoulou
1
, C. W. Lederer
1
, T. Surrey
2
, I. Vernos

3
and
N. Santama
1
1
University of Cyprus, Nicosia, CYPRUS,
2
EMBL, Heidelberg,
GERMANY,
3
Centre of Genomic Regulation, Barcelona, SPAIN
Inhibition of motor protein activity is linked with defects in the for-
mation of poles in the mitotic spindle but the molecular mecha-
nisms underlying the functional relationship between motor activity
and centrosome dynamics are unclear. We characterised KIFC5A,
a minus-end-directed mouse kinesin-like protein, highly expressed
in dividing cells. It is nuclear in interphase, localises to the centre of
spindle asters at the beginning of mitosis, and to spindle MTs in
later phases. Overexpression of KIFC5A in mouse cells causes the
formation of non-separated MT asters and mitotic arrest in a pro-
metaphase-like state. KIFC5A knockdown partly rescues the phe-
notype caused by monastrol inhibition of plus-end-directed motor
Eg5, indicating that it is involved in the balance of forces determin-
ing bipolar spindle assembly and integrity. Silencing of KIFC5A
results in centrosome amplification detectable throughout the cell
cycle. Supernumerary centrosomes arise primarily by reduplication
and partly from cytokinesis defects. They contain duplicated centri-
oles and have the ability to organise MT asters, resulting in the for-
mation of multipolar spindles. KIFC5A interacts with nucleotide-
binding proteins 1 and 2 (Nubp1, Nubp2), which have extensive

sequence similarity to prokaryotic division-site-determining protein
MinD. Nubp1&Nubp2 also interact with each other. Knockdown
of Nubp1 or double knockdown of Nubp1&Nubp2 both pheno-
copy the KIFC5A silencing effect, implicating KIFC5A and the
Nubps in a regulatory pathway involved in the control of centro-
some duplication.
B1-10
Which structural determinants make kinesin
processive?
S. Adio
1,2
, B. Ebbing
2
, J. Jaud
2
, M. Rief
2
and G. Woehlke
2
1
National Institute for Medical Research, London NW7 1AA, UK,
2
Technical University of Munich, Munich, GERMANY
Kinesin motor proteins transport cargo through the cell. Conven-
tional Kinesin-1 motors travel processively along microtubules by
taking hundreds of successive ‘hand-over-hand’ steps. They are
composed of two identical subunits, each consisting of a motor
domain and a neck/stalk. Despite extensive studies, structural ele-
ments responsible for the accurate coordination between the two
subunits remain unknown. To identify these elements, we have

constructed a chimeric motor with the motor domain of Kinesin-1
and a neck/stalk of a non-processive Kinesin-3. We found that this
novel kinesin behaves qualitatively as conventional Kinesin-1. It
moves processively in single molecule fluorescence assays and
exerts up to 3 pN force in optical trapping experiments. However,
reduced velocities and run lengths at a variety of loads indicate
that the unconventional neck domain hinders the diffusive search
for the next binding site. In the reverse chimera (motor domain of
non-processive Kinesin-3 and neck/stalk of Kinesin-1) the Kinesin-
1 neck allows sequential ADP release from the partner heads upon
microtubule binding. But this motor was unable to make successive
steps.
Our observations suggest that kinesin processivity requires two
independent elements. One, located in the neck/stalk region,
ensures coordinated ADP release from the motor domains. The
other, located in the motor domain itself, allows successive ‘hand-
over-hand’ stepping by coordinating the timing of events between
the two motor domains.
B1-11
Effects of kinesin motor domain structure on
kinesin activity
N. Kalchishkova, E. Unger and K. J. Bo
¨
hm
Leibniz Institute for Age Research - Fritz Lipmann Institute, Jena,
GERMANY
Kinesin-microtubule interaction results in generation of motility,
essential for organelle transport and cell division. Each kinesin
molecule reveals a distinct globular motor domain, a neck-linker
and a neck, a stalk and a cargo-binding tail domain. The force-

generating motor domain comprises both the microtubule-binding
site and the ATP-hydrolysis centre.
Kinesin binding to microtubules, known to promote kinesin AT-
Pase activity, induces the nucleotide-binding pocket to be closed
and initializes ATP hydrolysis followed by subsequent conforma-
tional changes, triggering the directed movement of the motor pro-
tein (Naber et al. Science 30, 798). So far, the molecular
mechanisms of kinesin-microtubule interaction cannot be com-
pletely understood. The a5-helices of kinesin-1 and Eg5 and their
neighbouring loops are important surfaces undergoing nucleotide-
dependent rearrangements to interact with microtubules (Turner
et al. J. Biol. Chem. 276, 25496). Therefore, we expressed both
human neuronal kinesin-1 and mitotic Eg5 constructs of different
length with changes mainly in sequences following the a5 helix to
find the minimal length of the motor domain still allowing binding
to microtubules, ATP hydrolysis, and finally motility generation. In
addition, chimeric constructs of fast (kinesin-1) and slow (Eg5)
motors have been expressed and characterized to understand velo-
city regulation.
We demonstrated that kinesin-1 and Eg5 constructs terminated at
a5 helix and its neighbouring loops do not reveal a microtubule-
promoted ATPase activity. Thus, structures next to these regions
are considered to have an important function in conformational
modulation of motor domain during the ATP hydrolysis cycles
and motility generation.
B1-12
Quality control during the asse mbly of myosin-V
dependent translocation complexes
A. Heuck
1

,T.Du
2
, R. Jansen
2
and D. Niessing
1
1
GSF & Gene Center Munich University, Munich, GERMANY,
2
Gene Center Munich University, Munich, GERMANY
Motor protein-dependent transport of cargo is a basic cellular
function. To exert their motile activity, the molecular motors kine-
sin and myosin must dimerize. Surprisingly little is known about
how such molecular motors assemble with other proteins into func-
tional translocation complexes. Recent studies on a minus-end
directed myosin, i.e. type VI myosin, show that this motor only di-
merizes upon complex assembly. However, for plus-end directed
myosins, which constitute the vast majority of myosin motors, we
know very little about the mechanisms controlling the assembly of
their translocation complexes. Available data suggest for these my-
osins a constitutive and unregulated interaction of their C-terminal
tails with cargo complexes.
We report a study involving biochemical, biophysical, and in vivo
experiments on the assembly of a yeast translocation complex with
plus-end directed myosin type V. Our results establish that myosin
dimerization significantly increases its affinity for its interaction
partner of the cargo complex and stabilizes the resulting complex.
Thus, myosin dimerization is required for efficient binding to the
core complex. Since only dimeric myosins move processively along
actin filaments, these observations reflect a quality control step for

the selective assembly of functional translocation complexes. Our
finding contrasts the previously suggested simple interaction of the
myosin C-terminal tail with cargo complexes.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 99
B1-13
A functional multivesicular body pathway is
required for clearance of protein aggregates by
autophagy
Filmonenko M.
1
, Raiborg C.
1
, Stuffers S.
1
, Yamamoto A.
2
, Brech
A.
1
, Stenmark H.
1
and Simonsen A.
1
1
Dept. of Biochemistry, Rikshospitalet-Radiumhospitalet HF,
N-0310 Oslo, NORWAY
2
Dept. of Physiology and Cellular
Biophysics, Columbia University, New York, NY 10032, USA

The ESCRT-III subunit CHMP2B/Vps2b was recently found to be
mutated in patients with frontotemporal dementia (FTD) and
amyotrophic lateral sclerosis (ALS). These diseases are characterized
by abnormal intracellular ubiquitin positive protein deposits in
affected neurons. The ESCRTs, first identified in yeast as vps class E
mutants, have proven important for recognition of ubiquitinated
endocytosed integral membrane proteins and their sorting into the
intralumenal vesicles of the multi-vesicular body (MVB) and subse-
quent degradation in the lysosome/vacuole. We here present evi-
dence that siRNA-mediated depletion of the ESCRT III subunit
Vps24 leads to accumulation of large ubiquitin-positive cytoplasmic
protein aggregates, which also contain p62/Sequestosome-1 and Alfy
(autophagy-linked FYVE protein), but are devoid of the early endo-
some marker EEA1. We show that p62-containing amphisomes
(positive for GFP-LC3 and CD63) do form in Vps24 depleted cells,
but that lysosomal degradation of p62 and LC3 is inhibited/delayed.
Vps24 is also required for efficient clearance of HttQ103 aggregates
in a cell-based model of Huntingtons disease. Taken together, our
data indicate that efficient autophagic clearance of ubiquitinated
proteins requires functional MVBs and provide a possible explan-
ation to the observed neurodegenerative phenotype seen in patients
with CHMP2B mutations.
B1-14
Screen for new factors affecting function of
ubiquitin ligases, human Nedd4 and yeast Rsp5,
in yeast S. cerevisiae
P. Kaliszewski
1
, M. Stawiecka-Mirota
1

, D. S. Haines
2
and
T. Zoladek
1
1
Institute of Biochemistry and Biophysics PAS, Warsaw, POLAND,
2
Fels Institute for Cancer Research and Molecular Biology, Temple
University School of Medicine, Philadelphia, PA, USA
Human Nedd4 and yeast Rsp5 are homologous ubiquitin ligases
that contain a catalytic Hect domain, a C2 domain and multiple
WW domains interacting with proteins. Previous studies indicated
that NEDD4 gene can be efficiently expressed in yeast, however,
can not replace RSP5. Expression of NEDD4w4 with mutation
inactivating WW4 domain of Nedd4 ligase is toxic for yeast cells.
In this study we analyzed the survival of cells overproducing
Nedd4w4 and showed that it causes inhibition of cell growth, not
the lethality. Multicopy genomic library was screened for genes
which suppress this growth defect. Five genes were isolated among
them ATG2 which is involved in autophagy. Using tester strain
pho8D60 we showed that overexpression of NEDD4w4 or ATG2
genes results in increase of autophagy in growth conditions and
the effect is additive. These results indicate that expression of
nedd4w4 possibly results in defects of proteasomal protein degra-
dation and induction of autophagy may help to overcome this
problem.
B1-15
The lysosomal targeting of the membrane
protein p40 is mediat ed by a dileucine signal

situated in its C-terminal tail
M. Boonen, G. Cuvelier, R. Castro, I. Hamer and M. Jadot
URPHYM, FUNDP, Namur, BELGIUM
Transport of newly synthesized lysosomal membrane proteins from
the TGN to the lysosomes is due to the presence in their cytoplas-
mic domains of specific signals which are recognized by clathrin-
associated adaptor complexes. p40, a predicted multispanning pro-
tein of 372 amino acids localized to the lysosomal membrane, con-
tains four putative lysosomal sorting motifs in its sequence: three
of the YXXF-type (Y
6
QLF, Y
106
VAL, Y
333
NGL) and one of the
dileucine-type (EQERL
360
L
361
). Mutations of the critical residues
of these motifs, tyrosine or leucine, revealed that the EQERLL
motif was implicated in the sorting of p40. Confocal microscopy
analyses, performed in transfected HeLa cells, showed that p40-
GFP was mis-targeted to the plasma membrane when its dileucine
motif was disrupted. This result was confirmed by cell surface bio-
tinylation studies which allowed us to estimate that approximately
50% of the p40 dileucine mutants (without GFP) were associated
with the plasma membrane. We also observed that, in self-forming
Percoll density gradients, the dileucine mutants were excluded from

fractions containing mature lysosomes. Taken together, our results
show that the sorting of p40 from the TGN to the lysosomes is
directed by the dileucine EQERL
360
L
361
motif situated in its C-ter-
minal tail.
B1-16
Elongation arrest activity of SRP is essential for
efficient translocation of proteins into the ER in
mammalian cells
C. Mary
1
*, A. K. K. Lakkaraju
1
*, A. Scherrer
1
, A. E. Johnson
2
and K. Strub
1
1
University of Geneva, Geneva, SWITZERLAND,
2
A&M University
System Health Science Center, College Station, TX, USA
The signal recognition particle (SRP) is responsible for co-transla-
tional translocation of proteins bearing signal sequence into the
endoplasmic reticulum (ER). In wheat germ translation system,

SRP causes an arrest in the elongation (EA) of nascent chains pre-
senting signal sequence and efficiency of their translocation is
dependent on this arrest. We have identified a five amino acid-long
motif in SRP14 required to confer EA activity to SRP. Particles
comprising the defective SRP14 protein (A5), no longer arrest elon-
gation of the nascent chain and have reduced translocation effi-
ciency. To examine the significance of this function for protein
secretion in mammalian cells, we depleted endogenous SRP14 using
RNAi and replaced it with the expression of wild type or mutant
GFP-14. Although the expression of both, GFP-14 and GFP-14A5,
restored cellular SRP levels to normal, the translocation efficiency
of reporter proteins was specifically diminished in cells expressing
GFP-14A5. The defect could be rescued by the addition of aniso-
mycin, which slows down nascent chain elongation. These results
demonstrate the physiological relevance of EA for efficient target-
ing in mammalian cells. Protein translocation into the ER is ineffi-
cient at normal cellular translation elongation rates presumably,
because nascent chains become too long during the time span
required for SRP to target nascent chains successfully to the trans-
locon. The selective slow down of translation elongation by SRP is
therefore required to optimize protein translocation efficiency in
mammalian cells without hampering continued rapid translation of
other cellular proteins.
*equal contribution.
Abstracts Transport Machineries
100 ª 2007 The Authors Journal compilation ª 2007 FEBS
B1-17
Regulation of vesicle trafficking by protein
S-nitrosylation
Y. Daaka

Medical College of Georgia, Augusta, GA, USA
The guanosine triphosphatase (GTPase) dynamin regulates endocy-
tic vesicle budding from the plasma membrane, but the molecular
mechanisms involved remain incompletely understood. We repor-
ted that dynamin, which interacts with nitric oxide (NO) synthase,
is S-nitrosylated at a single cysteine residue. C607. S-nitrosylation
increases dynamin self-assembly and GTPase activity, and facili-
tates its redistribution to the membrane. A mutant protein bearing
a C607A substitution does not self-assemble properly or increase
its enzymatic activity in response to NO. In NO generating cells,
expression of dynamin C607A, like the GTPase-deficient domin-
ant-negative K44A dynamin, inhibits active receptor internalizat-
ion. More recent results implicate the dynamin S-nitrosylation in
infectious particle uptake. Thus, NO regulates endocytic vesicle
budding by S-nitrosylation of dynamin. We will discuss the general
NO-dependent mechanism by which the trafficking of plasma
membrane vesicles may be regulated, and the specific idea that
pathogenic microbes and viruses may induce S-nitrosylation of
dynamin to facilitate cellular entry.
B1-18
The role of signaling in control of bi-directional
ER to Golgi trafficking
V. Gupta and D. Stephens
Dept. of Biochemistry, University of Bristol, UK
Our research is aimed at understanding the function of the early
secretory pathway in mammalian cells. While we now have a signi-
ficant amount of knowledge regarding the individual molecules
involved, we now need to develop this to an understanding of the
system as a whole. We have developed live cell imaging assays to
analyze the individual movements of structures in transit between

the ER and the Golgi, the first two compartments of the secretory
pathway. Movement of these transport carriers occurs in a stop-
start fashion with frequent bidirectional oscillatory, as well as lon-
ger range, movements. This behaviour is due to regulation of the
binding to, and/or regulated activity of, opposing motor proteins-
in general terms, dynein (along with dynactin) moves cargo
towards the Golgi, while kinesin moves cargo away from it. The
function of these motors is also controlled by protein phosphoryla-
tion, acting at least in part through a protein kinase A-dependent
pathway. We are studying the association of dynactin with carriers
in transit to the Golgi and the regulation of bidirectional motility
by protein phosphorylation by imaging GFP-tagged forms of var-
ious ER-Golgi markers. We have established an experimental sys-
tem that allows us to image the movement of ER-to-Golgi
transport carriers with extremely high spatial and temporal resolu-
tion. Application of 2D-Gaussian fitting models to the data allows
precise determination of object position. This should ultimately
allow us to determine motor step size, which will give key informa-
tion regarding the type of motors associated with these structures.
It will also lead to the detailed analysis of the dynamics of trans-
port carriers operating between early secretory pathway and help
in deciphering the mechanism of their regulation.
B1-19
Molecular analysis of YIP1 family protein
function
C. Chen and R. N. Collins
Cornell University, Ithaca, NY, USA
The YIP1 family is a group of integral membrane proteins with
the capability of binding to dual prenylated Rab GTPases. These
proteins are functionally conserved from yeast to humans, and, in

S. cerevisiae, comprise two essential proteins Yip1p and Yif1p, and
two non-essential proteins Yip4p and Yip5p. Although their mode
of action remains unclear at the molecular level, phenotypic analy-
sis of yip1 and yif1 mutant cells reveal blockage in early stages of
membrane traffic. We have previously shown that Yip1p is neces-
sary for biogenesis of ER derived COPII coated vesicles both in
vivo and in vitro (1). In our current study, we utilized serial trunca-
tion and site-directed mutagenesis to isolate alleles of Yif1p and
Yip1p which abrogate their Rab-interacting capabilities. Surpris-
ingly, our results suggest that the YIP1 family proteins might have
a function independent of Rab proteins, and present data to sug-
gest a possible role in organelle shape.
Reference
1. Heidtman M, Chen CZ, Collins RN, Barlowe C. A role for
Yip1p in COPII vesicle biogenesis. J Cell Biol. 2003 Oct 13;163
(1) :57–69.
B1-20
GTRAP3-18 serves as a negative regulator of
Rab1 in protein transport and neuronal
differentiation
S. Maier
1
, V. Reiterer
1
, A. M. Ruggiero
2
, J. D. Rothstein
3
,
H. H. Sitte

1
and H. Farhan
1
1
Institute of Pharmacology, Medical University, Vienna, AUSTRIA,
2
Department of Pharmacology, Vanderbilt University, Nashville,
TN, USA,
3
Departments of Neuroscience and Neurology, Johns
Hopkins University, Baltimore, MD, USA
GTRAP3-18 is an ER localized protein belonging to the prenylat-
ed rab-acceptor-family interacting with small Rab GTPases. Its
impact on secretory trafficking has not been investigated. We
report here that GTRAP3-18 inhibits Rab1, which is involved in
ER-to-Golgi trafficking. The effects on the early secretory pathway
were: reduction of the rate of ER-to-Golgi transport in HEK293
cells, inhibition of cargo concentration of the EAAC1 glutamate
transporter into transport complexes in HEK293 cells, an effect
that could be completely reversed in the presence of an excess of
Rab1. In accordance with the role of Rab1 in neurite formation,
overexpression of GTRAP3-18 significantly inhibited neurite out-
growth in CAD cells. Finally, we hypothesized that expression of
GTRAP3-18 in the brain should be lower at stages of active syn-
aptogenesis compared to early developmental stages. This was the
case as expression of GTRAP3-18 was lower in P0 rat brains com-
pared to E17 rat brains.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 101
B1-21

The signal peptide of LCMV GP-C is required for
glycoprotein maturation and virus infectivity
S. Schrempf
1
, M. Froeschke
1
, T. Giroglou
2
, D. von Laer
2
and
B. Dobberstein
1
1
Zentrum fu
¨
r Molekulare Biologie Heidelberg (ZMBH),
Heidelberg, GERMANY,
2
Georg-Speyer-Haus (GSH), Frankfurt,
GERMANY
Signal peptides direct nascent secretory and membrane proteins to
the endoplasmic reticulum (ER) and are usually cotranslationally
cleaved off from the preprotein by signal peptidase. Insertion of
the lymphocytic choriomeningitis virus (LCMV) precursor glyco-
protein C (pGP-C) into the ER membrane is mediated by an unu-
sual signal peptide (SP
GP-C
). It is longer than average SPs
comprising an extended hydrophilic N-terminal (n) region inclu-

ding a myristoylation site and two hydrophobic (h) regions. After
cleavage by signal peptidase SP
GP-C
was found to accumulate in
cells and virus particles.
In the present study, we identified the LCMV SP
GP-C
as an
essential component of the viral GP complex. By using several
deletion and point mutants we investigated signal sequence require-
ments for GP-C biosynthesis, transport to the cell surface and viral
infectivity. Our results show, that only one SP
GP-C
h-region is nee-
ded for membrane insertion of pGP-C, whereas both h-regions are
required for GP-C processing and cell surface expression. Further-
more, the n-region of the SP
GP-C
and the myristoylation are found
to be essential for viral infectivity, most likely for pH-dependent
fusion with the host cell membrane. Thus different regions of
LCMV SP
GP-C
fulfil essential functions in GP-C maturation and
virus infection.
B1-22
Different domains of GOPC regulate its Golgi
localization
I. Todros
1,2

, E. Kurzejamska
1,3
, A. Matysiak
1,3
and
M. Milewski
1
1
Institute of Mother and Child, Warsaw, POLAND,
2
Agricultural
University SGGW, Warsaw, POLAND,
3
Warsaw University,
Warsaw, POLAND
Golgi-associated PDZ and coiled coil motif-containing protein
(GOPC) is a peripheral Golgi protein involved in vesicular traffick-
ing of many integral membrane proteins, including CFTR, a pro-
tein defective in cystic fibrosis. The exact role of GOPC in post-
Golgi trafficking is not known, although it has been suggested that
it may function as an adaptor protein, participating in transport
from Golgi to lysosomes. To identify the amino acids sequences
responsible for subcellular localization of GOPC, a series of dele-
tions encompassing different structural domains was introduced
into the protein sequence. Immunofluorescence analysis of the sub-
cellular distribution of these GOPC variants in transfected epithe-
lial cells showed that the second coiled-coil domain (CC2) is
required for Golgi localization, with adjacent second conserved
region (CR2) also contributing to the efficiency of Golgi targeting.
Additionally, increased Golgi localization of the GOPC variants

devoid of the C-terminal PDZ domain suggested that Golgi distri-
bution of GOPC is negatively regulated by the PDZ-mediated
interactions. This was further supported by the fact that overex-
pression of CFTR, a protein interacting with GOPC through its
PDZ domain, substantially reduced Golgi localization of GOPC.
Thus, different domains of GOPC seem to contribute to its subcel-
lular distribution. While the coiled-coil-mediated interactions
ensure efficient targeting to the Golgi, the PDZ-based mechanism,
being responsible for binding a cargo protein, contributes to the
subsequent detachment of the whole complex from the trans-Golgi
membrane.
Acknowledgement: Supported by grant PBZ/KBN/122/P05/01-
06.
B1-23
Role of the multimolecular conserved
oligomeric Golgi complex in intra-Golgi
trafficking and glycosylation disorders
V. Lupashin, A. Shestakova, E. Suvorova, R. D. Smith and
O. Pavliv
University of Arkansas for Medical Sciences, Little Rock, AR, USA
Large oligomeric multiprotein complexes direct membrane trafficking
in eukaryotic cell. One class of traffic regulators, vesicle tethering fac-
tors, mediate the initial loose tethering of transport vesicles to their
target membranes. The peripheral membrane
conserved oligomeric
Golgi (COG) complex directs retrograde intra-Golgi membrane traf-
ficking and ensures proper Golgi functions. Mutations in COG sub-
units cause congenital glycosylation disorders in humans, suggesting
that COG is necessary for the correct sorting of glycosyltransferases,
which create a variety of developmentally important glycans. The

COG complex consists of two lobes, Lobe A (COG1-4) and Lobe B
(COG5-8). Rapid knock-down of Lobe A subunits induces accumu-
lation of intra-Golgi vesicles and Golgi ribbon breakdown, while the
knock-down of Lobe B subunits COG6 and COG8 does not change
Golgi morphology. The COG complex functionally communicates
with a subset of Golgi SNAREs and Rab proteins. Yeast COG com-
plex interacts with the SNARE domain of Sed5p and preferentially
binds to the Sed5p-containing SNARE complex. Yeast two hybrid
assay identifies mammalian Cog4p and Cog6p subunits as binding
partners of the mammalian Golgi t-SNARE Syntaxin5. FRET assay
reveals direct in vivo interaction between Syntaxin5 and the COG
complex. Knock-down of the COG complex results in decreased
Golgi SNARE mobility, accumulation of free uncomplexed form of
Syntaxin5 and decrease in steady-state level of intra-Golgi SNARE
complexes. In summary, the COG complex acts to promote forma-
tion and/or stabilization of the intra-Golgi SNARE complex.
Acknowledgements: Supported by grants from the NSF (MCB-
0234822) and Mizutani Foundation for Glycoscience.
B1-24
Cargo passage through the golgi apparatus
induces golgi lumenal Ca
2+
fluctuations
M. Micaroni
1
, K. Bianchi
2
, R. Rizzuto
2
, G. Perinetti

1
, A. Spaar
1
,
D. Di Giandomenico
1
, A. Luini
1
and A. A. Mironov
1
1
Consorzio ‘‘Mario Negri Sud’’, Santa Maria Imbaro, ITALY,
2
Uni-
versity of Ferrara, Ferrara, ITALY
Recent evidence has highlighted the functional importance of Ca
2+
in
intracellular trafficking, and Ca
2+
increases could act downstream of
the SNAREs, proteins that are involved in the fission/fusion processes
that occur during cargo progression through the Golgi apparatus.
Indeed, C a
2+
increases the rate of SNARE-mediated membrane fusion
via C a
2+
ion channel association with the SNARE complex. Our work-
ing hypothesis is based on the fact that the fusion events, that occur

during cargo progression from the Golgi apparatus to the trans-
Golgi network (TGN), involve redistribution of the Ca
2+
stored in
the lumen of the Golgi cisternae that mediates the recruitment of reg-
ulatory proteins involved in the final steps of SNARE-regulated
fusion. Here we have examined whether the Ca
2+
concentrations
near the Golgi apparatus and inside the trans-Golgi lumen change
during cargo progression through the Golgi. Using fluorescence res-
onance energy transfer (FRET) with HeLa cells stably transfected
with galactosyl transferase-Yellow Chameleon 3.3 (GT-YC3.3)
in vivo and in fixed samples, we have seen Ca
2+
leakage in the trans-
Golgi region from the lumen of Golgi cisternae in VSV-infected cells
during the passage of the viral VSVG protein. At the same time, in
HeLa cells transfected with sialyl transferase-aequorin (ST-Aeq), we
have confirmed this release of Ca
2+
from the lumen of the Golgi
cisternae. Moreover, we have seen an increased frequency of cyto-
solic Ca
2+
spikes using FURA-2. These data suggest that the redis-
tribution of Ca
2+
during the movement of cargo proteins through
the Golgi apparatus has a significant role in intra-Golgi transport.

Abstracts Transport Machineries
102 ª 2007 The Authors Journal compilation ª 2007 FEBS
B1-25
Direct interaction of the C-terminus of TRPC6
with Rab9 targets TRPC6 to the trans-Golgi
network
S. Cayouette and G. Boulay
Universite
´
de Sherbrooke, Sherbrooke, PQ, CANADA
TRPC proteins are implicated in Ca
2+
entry following activation of
Gq-protein coupled receptors. We previously showed that the inser-
tion of TRPC6 into the plasma membrane (PM) is regulated by hor-
monal stimulation. Since small G proteins of the Rab family are
involved in vesicle trafficking and fusion, we assessed the intracellu-
lar trafficking of TRPC6 with different Rab dominant negative
(DN) mutants. TRPC6 activation was measured in HEK293 cells
transiently co-expressing either Rab11DN (blocks transport from
the TGN and recycling endosome to the PM) or Rab9DN (blocks
fusion of late endosomes with the TGN). Rab9DN significantly
enhanced carbachol-induced Ca
2+
entry through TRPC6 activation,
whereas Rab11DN completely abolished its activity. Rab11 and
Rab9 have no significant effect on TRPC6 activity. GST-pulldown
assay showed that Rab9 and Rab9DN interact directly with the C-
terminus of TRPC6. These interactions occur also in vivo, since
Rab9 and Rab9DN co-immunoprecipitate with TRPC6. However,

neither Rab11 nor Rab11DN interact in vivo with TRPC6. Immuno-
fluorescence microscopy and cell-surface biotinylation assays
showed that Rab9DN enhances the amount of TRPC6 at the cell
surface and Rab11DN causes an intracellular retention of TRPC6.
We conclude that TRPC6 transits through the TGN and that Rab9
is directly involved in the process. Moreover, TRPC6 is redirected to
the PM in a Rab11-dependent fashion without involving an interac-
tion with TRPC6. Supported by CIHR and FMCQ.
B1-26
Role of Casein Kinase 2 in the early steps of
CFTR biogenesis
C. M. Farinha
1,2
, L. Pissarra
1
and M. D. Amaral
1,2
1
Faculty of Sciences, University of Lisboa, Lisboa, PORTUGAL,
2
Cent Hum Genet, Nat Inst Health, Lisboa, PORTUGAL
Cystic Fibrosis (CF) transmembrane conductance regulator
(CFTR), the product of the gene which is mutated in CF, func-
tions as a Cl
-
channel at the apical membrane of epithelial cells.
The most frequent mutation (F508del) causes retention of its
protein product in the endoplasmic reticulum as a core-glycosyl-
ated intermediate that is rapidly degraded. Thus, F508del-CFTR
fails to traffic to the plasma membrane. It was recently shown

that Casein Kinase 2 (CK2) binds wt-CFTR near the F508 resi-
due and phosphorylates serine residue at position 511. Deletion
of F508 abrogates this residue-dependent interaction [1]. Our
aim here was to identify whether CK2 interaction affects CFTR
biogenesis, turnover and processing. BHK cells were stably
transfected with wt- or F508del-CFTR in which S511 was sub-
stituted by alanine (S511A) or aspartate (S511D). Pulse-chase
experiments followed by CFTR immunoprecipitation were per-
formed in these lines, showing that substitution of S511 does
not affect the turnover or processing of either wt- or F508del-
CFTR. The effect of CK2 inhibition on the turnover and pro-
cessing of CFTR showed that: 1) steady-state levels of CFTR
are reduced. 2) turnover of wt-CFTR (but not F508del-CFTR)
is increased and 3) processing of wt-CFTR is decreased. Our
data suggest a putative stabilizing role for CK2 upon wt-CFTR,
not dependent on the charge that is added by the kinase, so it
is probably indirect.
Acknowledgement: Work supported by CIGMH and POCTI/
SAU/MMO/58425/2004 grant (FCT, Portugal).
Reference
1. Treharne et al. (2007) J Biol Chem (epub).
B1-27
Checkpoints in the traffic pathways of
Tyrosinase Related Proteins in melanoma cells
G. Negroiu
Institute of Biochemistry, Bucharest, ROMANIA
Tyrosinase Related Proteins-(TRPs) are the main regulators of
melanin synthesis in normal and pathological pigmentation and
immune targets in melanoma, glioma and autoimmune depigmen-
tation. TRP-2 has a significant role in intrinsic drug resistance of

tumor phenotypes. TRPs are all type I transmembrane N-glyco-
proteins which follow the common biosynthetic secretory pathway.
TRP-polypeptide folding and early N-glycan processing are accom-
plished in ER. TRPs move than via the Golgi complex to the
TGN, where are sorted and transported to the membrane of mel-
anosomes, the organelles for melanin synthesis and temporal
deposition. Abnormalities in the ER or post-ER events of the
intracellular processing of tyrosinase and TRP-1 result in their mis-
routing or/and degradation and have been investigated in different
pigmentary diseases and types of malignancy. TRP-2 biosynthesis
was analyzed for the first time in our studies using pharmacologi-
cal agents as modulators of intraluminal pH together with TRP-1
as a reporter molecule. We found that TRP-2 is sorted and traf-
ficked in the early secretory pathway with a cargo which does not
include TRP-1, supporting the concept of a selective ER-Golgi
transport; post-Golgi, unlike TRP-1, TRP-2 intersects the endocy-
tic pathway following a route via early endosomes, possibly by
rapid recycling from the plasma membrane. These findings demon-
strate that, in spite of the high degree of structural homology, the
traffic pathways of TRPs are individually regulated. TRPs face dis-
tinct ‘‘check points’’ along their biosynthetic journey in order to
become functional proteins.
B1-28
Impact of progesterone on cytokine-stimulated
NF-jB signaling
A. C. Vidaeff
University of Texas Houston Medical School, Houston, TX, USA
Objective: A key event in the pathways leading to preterm labor
may be the activation of nuclear factor-jB (NF-jB). Anti-inflam-
matory agents, such as the corticosteroids, inhibit the activation of

NF-jB. We proposed to investigate the effects of progesterone (P)
pretreatment on cytokine-stimulated activation of NF-jB.
Study design: HeLa cells were pretreated with 10
-7
M progester-
one for 24 h and exposed to 1 ng/ml IL-1b for 1 hour. Nuclear
and cytosolic extracts were subjected to Western analysis using
anti-p65 and anti-IjBa antibodies. Densitometric data (n =5)
were compared using Kruskal–Wallis test.
Results: Pretreatment with P interfered with IL-1b-induced inhibi-
tory protein-jBa (IjBa) degradation. However, P pretreatment
resulted in a significant decrease in p65 in the cytoplasm. Pretreat-
ment with P did not reduce the amount of nuclear p65 and did not
interfere with nuclear translocation of p65.
Conclusion: Our observations suggest that any possible role
played by P in preterm labor prevention is not exerted through
anti-inflammatory mechanisms of NF-jB downregulation.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 103
B1-29
Loss of alpha-tubulin polyglutamylation in
ROSA22 mice is associated with abnormal
targeting of KIF1A and modulated synaptic
function
M. Setou
NIPS and MITILS, Machida, JAPAN
Microtubules function as molecular tracks along which motor pro-
teins transport a variety of cargo to discrete destinations within the
cell. The carboxyl termini of alpha- and beta-tubulin can undergo
different posttranslational modifications, including polyglutamyla-

tion, which is particularly abundant within the mammalian nervous
system. Thus, this modification could serve as a molecular "traffic
sign" for motor proteins in neuronal cells. To investigate whether
polyglutamylated alpha-tubulin could perform this function, we ana-
lyzed ROSA22 mice that lack functional PGs1, a subunit of alpha-
tubulin-selective polyglutamylase. In wild-type mice, polyglutamyl-
ated alpha-tubulin is abundant in both axonal and dendritic neurites.
ROSA22 mutants display a striking loss of polyglutamylated alpha-
tubulin within neurons, including their neurites, which is associated
with decreased binding affinity of certain structural microtubule-
associated proteins and motor proteins, including kinesins, to micro-
tubules purified from ROSA22-mutant brain. Of the kinesins exam-
ined, KIF1A was less abundant in neurites from ROSA22 mutants
in vitro and in vivo. The density of synaptic vesicles, a cargo of
KIF1A, was decreased in synaptic terminals in the CA1 region of
hippocampus in ROSA22 mutants. Consistent with this finding,
ROSA22 mutants displayed more rapid depletion of synaptic vesi-
cles than wild-type littermates after high-frequency stimulation.
These data provide evidence for a role of polyglutamylation of
alpha-tubulin in vivo, as a molecular traffic sign for targeting of
KIF1 kinesin required for continuous synaptic transmission. Further
investigation of the role of polyglutamylation in vivo will be dis-
cussed in this presentation.
B1-30
Yif1B is implicated in the dendritic localisation
of the serotonin 5-HT1A receptor
J. Masson
1,2
, S. Al Awabdh
1,2

, D. Carrel
1,2
, C. Borg-Capra
3
,
J. Laine
´
2,4
, M. Hamon
1,2
, M. B. Emerit
1,2
and M. Darmon
1,2
1
INSERM U677, Paris, FRANCE,
2
Universite
´
Pierre et Marie
Curie, Paris, FRANCE,
3
Hybrigenics SA, Paris, FRANCE,
4
Dept Physiologie, UPMC, Paris, FRANCE
The 5-HT1A receptor is a dendritic receptor implicated in the con-
trol of brain (5-HT) neurotransmission. Evidence has been reported
that the short C-terminal domain of the rat 5-HT1AR plays a crucial
role in receptor targeting from the endoplasmic reticulum to the
plasma membrane. We used this 17 amino acid region as a bait in a

yeast two-hybrid screen and identified Yif1B as a new partner of 5-
HT1AR. The physical interaction between the receptor and Yif1B
was fully confirmed in GST pull down experiments with 5-HT1AR
C-tail on rat brain extracts. Yif1B is the ortholog of the yeast Yif1p
protein (
Yip1 Interacting Protein, implicated in the vesicular traffic
between the endoplasmic reticulum and the Golgi apparatus, Matern
et al. 2000). Yif1B is highly expressed in the rat brain as shown by
northern and western blot analyses. Immunolabeling with specific
polyclonal antibodies showed that in transfected cell lines Yif1B was
confined to an intracellular vesicular compartment partially overlap-
ping the Golgi apparatus, in concordance with the role of its yeast
Yif1p ortholog. Electron microscopy visualization of endogenous
Yif1B protein in rat brain tissue sections confirmed this Golgi locali-
sation. Finally, siRNA-induced inhibition of endogenous Yif1B
expression in neuronal primary cultures prevented the addressing of
5-HT1AR in distal portions of dendrites. All these results support
the hypothesis that Yif1B plays a role in the dendritic localisation of
the 5-HT1A receptor in brain neurons.
B1-31
Alpha-Syntrophin interacts with Kidins220 and
Protein Kinase D in neural cells
L. Sa
´
nchez Ruiloba, A. Higuero, R. Martı
´
n Jean-Mairet,
M. Rodrı
´
guez-Martı

´
nez, F. Portillo and T. Iglesias
Insituto de Investigaciones Biomedicas, Madrid, SPAIN
Protein kinase D1 (PKD1) belongs to a novel family of diacylglyc-
erol (DAG)-stimulated Ser/Thr kinases, constituted by two more
members, PKD2 and PKD3. We have previously identified the C
terminus of the different PKDs that constitutes a PSD-95, Dlg,
ZO-1 (PDZ)-binding motif in PKD1 and PKD2, but not in PKD3,
to be responsible for the differential control of Kinase D-interact-
ing substrate of 220-kDa (Kidins220) surface localization. Ki-
dins220 is a neural membrane protein identified as the first
substrate of PKD1. PKD1 controls Kidins220 transport by the
interaction with a PDZ protein. In an effort to identify that PDZ
protein we performed yeast two hybrid screenings using as baits
the PDZ-binding motifs of Kidins220 and PKD1. We found a pos-
itive clone containing the complete coding sequence of the PDZ
protein a-syntrophin that strongly interacted with both baits. a-
syntrophin has been described as a Kidins220 binding partner that
plays an important role in the regulation of Kidins220 localization
during neuromuscular junction differentiation. Our results show
that PKD1 phosphorylation of its PDZ-binding motif regulates the
interaction with a-syntrophin in yeast and that Kidins220, PKD1
and a-syntrophin form ternary complexes in neural cells. These
data suggest that the interaction of a-syntrophin with Kidins220
and PKD1 could be relevant for the control the PKD1 exerts on
Kidins220 transport.
B1-32
Cellular control of trafficking and activity of
perforin, a key regulator of immune
homeostasis

I. Voskoboinik
Peter MacCallum Cancer Centre, East Melbourne, AUSTRALIA
Perforin (PRF) is an essential cytolytic pore-forming protein stored
in the secretory granules of cytotoxic lymphocytes (CL). CLs kill
virus infected and transformed cells through the granule-mediated
pathway, where PRF synergises with serine proteases, granzymes,
to deliver the lethal hit. In humans, congenital PRF deficiency cau-
ses a fatal disorder of immune system, FHL or, in milder cases, it
affects tumour immune surveillance, which results in haematologi-
cal cancers (1–3). Despite the essential role of PRF in immune sys-
tem, its mechanism of action and the cell biology remain largely
unknown. Activated CLs synthesize a large amount of PRF, and a
key question in CL biology is how these cells protect themself from
the self-lysis by endogenous PRF. We have found that the extreme
C-terminal region of PRF encodes structural and trafficking motifs
which, acting in concert, are responsible for the inhibition of its
cytotoxic activity and, at the same time, allow PRF trafficking to
cytotoxic granules. There, PRF acquires an activated state through
modifications of its C-terminal tail. However, the acidic pH in the
lumen of the organelle prevents PRF from exerting its cytotoxicity
prior to its release outside the cell into the immune synapse. Thus,
we unravelled the interplay between finely tuned trafficking path-
ways and post-translational activation mechanisms of PRF, which
ensures its safe delivery and storage in the secretory granules.
References
1. Voskoboinik et al. (2006) Nature Rev Immunol 6, 940.
2. Voskoboinik et al. (2004) J Exp Med 200, 811.
3. Voskoboinik et al. (2005) Blood 105, 4700.
Abstracts Transport Machineries
104 ª 2007 The Authors Journal compilation ª 2007 FEBS

B1-33
Regulation of intracellular trafficking plays a
pivotal role in tumor pathogenesis
J. Go
¨
ttert
Max-Delbru
¨
ck-Centrum fu
¨
r Molekulare Medizin, Berlin,
GERMANY
Intracellular trafficking as a fundamental cellular process needs
to be tightly regulated, otherwise disturbances can lead to a
variety of disorders and diseases. Glycans play a pivotal role in
protein sorting, there synthesis proceeds in a hierarchical manner
along the secretory pathway. Mechanisms that alter the intracel-
lular localization of glycosyltransferases and glycosidases may
lead to an alternative glycan repertoire of cells. In addition, gly-
can linkage defects are associated with pathogenetic events in
hereditary and acquired human diseases. Here, we refer to
EBAG9 that has been identified as a primary estrogen respon-
sive gene. High expression levels of the gene product have been
recorded in several carcinomas, suggesting a tumor-promoting
role of the protein. EBAG9 is a ubiquitously expressed Golgi
protein with a peripheral membrane attachment. It was shown
to induce expression of the truncated O-linked-glycans Tn and
TF at the plasma membrane of non-secretory epithelial cell lines.
In general, Tn- and TF are extensively expressed on the cell sur-
faces of diverse tumor tissues and they are thought to be

involved in tumor cell adhesion, invasion and metastasis. How-
ever, the mechanism underlying their generation remains to be
elucidated. Here, we localize the EBAG9-imposed alterations to
the early biosynthetic pathway and provide a functional link
between hormone-dependent gene expression, tumor-associated
O-linked glycan deposition, vesicle transport, and tumor patho-
genesis.
B1-34
Enhancement of the fungicidal activity of
amphotericin B by allicin, an allyl sulfur
compound from garlic
A. Ogita, M. Ogita, K. Fujita and T. Tanaka
Osaka City University, Osaka, JAPAN
Amphotericin B (AmB) is a representative antibiotic for the control
of serious fungal infections, and its fungicidal activity was greatly
enhanced by allicin, an allyl sulfur compound from garlic. In Sac-
charomyces cerevisiae, allicin was only slightly effective or ineffective
in enhancing the fungicidal activities of other antibiotics causing
plasma membrane damage. AmB increased plasma membrane per-
meability in favor of potassium efflux from intact cells, but AmB-
induced plasma membrane damage was unaffected or attenuated in
the presence of allicin. At a lethal concentration, AmB additionally
induced vacuole membrane damage so that the organelles were vis-
ible as small discrete particles. AmB caused a serious structural dam-
age to the yeast vacuole even at a nonlethal concentration in
combination with allicin that interferes with ergosterol trafficking
from plasma membrane to the organelles. Allicin could also decrease
the minimum fungicidal concentration (MFC) of AmB against the
pathogenic fungus Candida albicans to less than 12.5%, and against
Aspergillus fumigatus to less than 4% of that detected with AmB

alone. In contrast, allicin did not enhance the cytotoxic activity of
AmB against cells of human promyelocytic leukemia (HL-60), a
vacuole-less organism.
B1-35
Hereditary Spastic Paraplegia caused by
Kinesin-1 mutants
B. Ebbing
1
, K. Mann
1
, R. Schu
¨
le
2
and G. Woehlke
1
1
Institute for Cell Biology LMU, Munich, GERMANY,
2
Hertie-Institute for Clinical Brain Research, Tu
¨
bingen, GERMANY
Hereditary Spastic Paraplegia is a neurodegenerative motor neuron
disease characterized by lower limb spasticity and weakness. This
autosomal dominant disease is linked to mutation of at least 20
genes, among them several point mutations in the neuronal Kinesin-
1 (KIF5A) gene. Here, for the first time, we investigate the in vitro
properties of the wild-type KIF5A and three of the point mutations,
which are located in the motor domain. The N256S mutant had a
lower gliding velocity, whereas the R280S mutant exhibited low

affinity to microtubules. The K253N mutant showed both a microtu-
bule-affinity and a velocity defect. In laser trapping assays, none of
the mutants moved more than a few steps on microtubules. Mixed
gliding assays were utilized to show whether these defects can domin-
ate the wild-type protein. Only the N256S mutant decreased the
overall gliding velocity, when mixed one to one with the wild-type
kinesin. Potentially only the N256S mutant can slow down the cargo
transport in neurons. Assuming equal gene expression, patients are
expected to have 25% wild type, 50%heterodimeric and 25% ho-
modimeric kinesin. To simulate neuronal cargo transport we
attached such mixtures to quantum dots and measured their veloci-
ties. Our data indicate that a significant portion of cargo is transpor-
ted at slower rates.
B1-36
New cationic lipids as promising gene delivery
carriers: effect of the structure on biological
activity
D. Medvedeva
1
, D. Rapoport
1
, A. Vladimirova
1
, N. Mironova
1
,
M. Zenkova
1
, V. Vlassov
1

, M. Maslov
2
and G. Serebrennikova
2
1
Institute of Chemical Biology and Fundamental Medicine SB RAS,
Novosibirsk, RUSSIAN FEDERATION,
2
Lomonosov Moscow State
Academy of Fine Chemical Technology, Moskow, RUSSIAN
FEDERATION
Cationic lipids are extensively used as non-viral vectors for the
gene transfer but a number of the drawbacks limit their use in vitro
and in vivo. One of the ways to solve these problems is creating the
new biodegradable non-toxic cationic lipids based on the natural
compounds and studying their properties. We synthesized and tes-
ted the new biodegradable cationic lipids based on the cholesterol
and glycerol derivatives with the heterocyclic cationic head groups
and various linkers and studied them as gene delivery carriers. It
was found that the new cationic lipids are low-toxic at the concen-
trations below 10 lM for the various cell lines (293, HeLa, BHK,
BHK IR-780). By fluorescent microscopy and FACS we showed
that the presence of the cationic lipids enhanced the delivery of
various nucleic acids (oligonucleotides, plasmid DNA and siRNA)
into cells. The transfection efficiency and toxicity of the new cati-
onic lipids were compared with the commercially available cationic
agent Lipofectamine. Hence, the obtained results suggest that the
new cationic lipids based on the natural compounds could be the
promising agents for the transfection of human cells.
Acknowledgements: This work was supported by RAS pro-

grams ‘‘Molecular and Cellular Biology’’ and ‘‘Science to Medi-
cine’’, RFBR grant 05-04-48985.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 105
B1-37
The LC8 family of dynein light chains:
multifunctional chaperon-like proteins
Z. Ho
´
di
1
, P. Rapali
1
, L. Radnai
1
, T. Molna
´
r
1
,A
´
. Szenes
1
,
J. Kardos
1
, L. Buday
2
, W. F. Stafford
3

and L. Nyitray
1
1
Eo
¨
tvo
¨
s Lora
´
nd University, Budapest, HUNGARY,
2
Semmelweis
University, Budapest, HUNGARY,
3
Boston Biomedical Research
Institute, Boston, MA, USA
DYNLL1 and DYNLL2 are two mammalian paralogs of the con-
served LC8 family of dynein light chains that were described as tail
subunits of dynein and myosin Va motor proteins. Moreover, they
were shown to interact with a wide variety of other polypeptides.
They may serve as adaptors to bind cargos to the transport motors
however, they also have other functions in the cell unrelated to their
role in motor complexes. We have localized the binding site of DYN-
LL2 on myosin Va tail within an uncoiled region, between two
coiled-coil domains. DYNLL2 binds to this site with a Kd of 40 nM
and stabilizes both flanking coiled-coils. By studying DYNLL bind-
ing to their targets, we have found that PAK1, an important regula-
tor of cell motility, regulates activity of DYNLL1, but not
DYNLL2, by phosphorylating Ser88 and dissociating the homo-
dimer protein into two monomers. The monomeric DYNLL (mim-

icked by Ser88Glu mutation) is unable to bind to their known targets
however, it may bind to yet undiscovered proteins. By analyzing the
sequences of all known partners (>60), we have noticed that the only
common feature of the DYNLL targets is that they are either intrin-
sically disordered proteins (e.g. the proaptototic proteins Bim and
Bmf) or have their DLC binding sites within a disordered domain
(e.g. myosin Va and the dynein intermediate chain). Thus, DYNLL
isoforms are now recognized as chaperon-like ‘‘hub proteins’’ with
well-defined structure that could promiscuously bind to diverse inter-
actors that contain a short binding motif within a disordered protein
or domain and contribute to their folding and/or regulation.
B1-38
Modified supramolecular concatemeric
complexes as a novel system for
oligonucleotide delivery into mammalian cells
O. Gusachenko, D. Pyshnyi, M. Zenkova and V. Vlassov
Institute of Chemical Biology and Fundamental Medicine SB RAS,
Novosibirsk, RUSSIAN FEDERATION
Application of nucleic acids for the inhibition of gene function by
turning off genes with antisense oligonucleotides or double-stran-
ded small interfering RNA has confidently gained recognition as a
promising tool in the gene therapy. One of the main hurdles in the
use of oligonucleotide-based therapeutics is their poor uptake by
target cells. Recently formation of concatemeric complexes by
oligonucleotides was shown to promote their binding to several
mammalian cell lines [Simonova et al., 2006]. We proposed to use
this phenomenon for the development of new oligonucleotides
delivery system. To improve the efficiency of complexes penetration
through the cellular membrane we attached lipophilic cholesterol
molecules to different components of the concatemers. Uptake,

cellular distribution and biological activity of the complexes formed
by delivered oligonucleotides and cholesterol-modified carrier-oligo-
nucleotides were studied. We show that incorporation of the desired
antisense oligonucleotide into the self-assembling concatemeric sys-
tem significantly promotes its delivery into cells without the addi-
tion of any supplementary transfection agents and allows to achieve
specific inhibition of the target gene. In contrast to the use of con-
ventional delivery agents, concatemeric oligonucleotide system was
shown to posses no cytotoxic and non-specific intercellular activit-
ies. In conclusion, delivery of pharmacologically active oligonucleo-
tides into the target cells in concatemeric form provides perspective
alternative to the use of conventional transfection methods.
Acknowledgements: This work was supported by RAS programs
‘‘Molecular and cellular biology’’ and ‘‘Science to Medicine’’.
B1-39
The role of cytoplasmic networks in plasmid
DNA intracellular trafficking and intranuclear
movement of plasmid DNA
V. Ondrej, E. Lukasova, S. Kozubek and M. Falk
Institute of Biophysics Academy of Sciences of the Czech Republic,
v.v.i., Brno, CZECH REPUBLIC
The role of microtubule and actin networks in transport and pro-
cessing of transfected plasmid DNA in human cells was studied.
We observed strong binding activity of plasmid DNA to both
networks, their immobilization at actin filaments and directional
motion along the microtubules with average velocity
v = 1.12 lm/min. In addition, actin filaments play an important
role in dissociation of plasmid DNA aggregates at the cell periph-
ery. Disruption of one of these networks led to the abortion of
plasmid transport and accumulation of plasmid DNA in huge

aggregates at the cell periphery. The microtubule network consti-
tutes the high-ways for long-distance transport of plasmid DNA
as a cargo towards the cell nucleus. Our results show also the
functional role of actin in plasmid movement and targeting in the
cell nucleus. The most frequent sites of plasmid targeting were
those in which the genome integrity was impaired. Plasmid DNA
frequently localized in double stranded DNA breaks (DSB)
induced spontaneously or by ionizing radiation detected as sites
with phosphorylated H2AX. Inhibition of actin polymerization by
latrunculin B perturbed plasmid transport not only throughout
cytoplasm but also inside the nucleus and resulted in stopping of
plasmid co-localization with newly induced DSBs.
B1-40
Cellular localization and trafficking of
endogenous ATP7B and COMMD1 proteins
S. Donadio, B. De Keukeleire, J. Micoud, F. Piccarreta and
M. Benharouga
CEA, Grenoble, FRANCE
Copper (Cu), an essential trace element for cellular function, is
highly cytotoxic when in excess. In the liver, a Cu-transporter
ATPase (ATP7B) excretes the excess of Cu into the bile. A defect
in ATP7B is associated with Wilson disease, a Cu-overload disor-
der. Recently, COMMD1, a cytoplasmic protein responsible for
liver’s Cu toxicosis, was shown to interact, in vitro, with the N-
terminal region of ATP7B suggesting a cooperation for hepatic
Cu detoxification. Today, the intracellular localization of ATP7B
and COMMD1 is still a mater of controversy. In this study, we
investigated in vivo interaction and endogenous localization of
ATP7B and COMMD1 in human hepatoma cell line (HepG2).
ATP7B and COMMD1 were co-immunoprecipitated in vivo inde-

pendently of the cellular Cu content. Co-immunolocalization with
intracellular compartment markers (endoplasmic reticulum, ER;
Golgi apparatus, GA; early endosomes, EE; late endosomes/lyso-
somes, LE/Ly and plasma membrane, PM) showed that 200 lM
CuSO
4
stimulated trafficking of COMMD1 from ER, GA and
EE to LE/Ly, as well as trafficking of ATP7B from ER and EE
to the LE/Ly compartment. No GA and PM localization was
observed. These results were sustained by those obtained after
differential centrifugation and separation on sucrose gradient.
Taken together these results suggested that the mechanism of bil-
iary Cu excretion involves the intracellular trafficking of ATP7B/
COMMD1 and their interaction in the endosomes.
Abstracts Transport Machineries
106 ª 2007 The Authors Journal compilation ª 2007 FEBS
B1-41
Conformational changes in actin-binding
proteins, revealed by single particle electron
microscopy
O. Sokolova
1,2
, S. Maiti
2
, N. Grigorieff
3
, P. Lappalainen
4
and
B. L. Goode

2
1
Moscow University, Moscow, RUSSIAN FEDERATION,
2
Brand-
eis University, Waltham, MA, USA,
3
HHMI, Waltham, MA, USA,
4
Institute of Biotechnology, Helsinki, FINLAND
Cell locomotion, endocytosis, and intracellular motility of vesicles,
organelles and pathogens all rely on rapid assembly of actin net-
works. At the heart of these processes are actin nucleators that,
upon activation by nucleation promoting factors, stimulate actin
assembly. Formins are thought to processively cap the fast-growing
ends of actin filaments, while Arp2/3 complex seeds actin polymer-
ization by forming a pseudo-actin trimer of its two actin-related
subunits bound to WASP. The Arp2/3 dependent filament branch-
ing stimulates the formation of membrane protrusions. The recent
studies demonstrated that formation of such protrusions depends
on controlled interplay between direct membrane deformation by
IRSp53/MIM family proteins and the actin cytoskeleton. Here we
show using electron microscopy and single particle averaging that
actin binding proteins undergo major conformational changes
upon their activation. To identify the domain rearrangement we
used docking of the known crystal structures of domains into the
reconstructed 3D volume. We demonstrated that wild-type Arp2/3
complex exists in solution in three distinct conformations: open,
intermediate and closed. Activation of the Arp2/3 complex closes
the structure. On the other hand, autoinhibited formin exists in

closed conformation the number of open molecules increases dra-
matically after activation. Next, we studied the structure of actin
binding proteins MIM and IPRs53 during their activation. The
obtained results explained ligand-induced conformational changes
inside the MIM/IRSp53 proteins.
B1-42
Fluorescence Correlation Spectroscopy (FCS)
study of l-opioid receptor dynamics in live cells
V. Vukojevic, Y. Ming, C. D’Addario, B. Johansson, R. Rigler
and L. Terenius
Karolinska Institute, Stockholm, SWEDEN
Cellular dynamics of G protein-coupled receptors (GPCRs) is
rather complex and not yet fully elucidated, but it is widely accep-
ted that it is an integral part of their function - the trafficking steps
are regarded to be dynamical regulators of GPCRs density at the
cellular surface. Hence, the surface density of GPCRs can be con-
trolled without modifying the overall number of receptors in the
cell through receptor synthesis (up- or down-regulation) [1, 2].
Recent advances in molecular imaging offer the possibility to study
protein interactions and intracellular trafficking in real time and
with single-molecule sensitivity in living cells. We present results on
nondestructive observation of l-opioid receptor (MOR) inter-
actions with selected ligands using Fluorescence Correlation Spectr-
oscopy (FCS) integrated with Confocal Laser Scanning Microscopy
(CLSM) [3, 4]. With this approach we were able to monitor the
number of MOR exposed at the plasma membrane, determine their
lateral motility at the plasma membrane and evaluate how these
parameters are affected by interaction with selected ligands.
References
1. Gainetdinov, R.R., Premont, R.T., Bohn, L.M., Lefkowitz,

R.J., Caron, M.G. (2004) Annu. Rev. Neurosci. 27, 107–144.
2. Beaulieu, J.M., Sotnikova, T.D., Marion, S., Lefkowitz, R.J.,
Gainetdinov, R.R., Caron, M.G. (2005) Cell 122, 261–273.
3. Eigen, M., Rigler, R. Proc. Natl Acad. Sci. USA (1994) 91,
5740–5747.
4. Vukojevic, V., Pramanik, A., Yakovleva, T., Rigler, R., Teren-
ius, L., Bakalkin, G. (2005) Cell. Mol. Life Sci. 62, 535–555.
B1-43
Characterization of the nuclear import pathway
for the human hypoxia inducible factor HIF-
1alpha
G. Chachami
1
, E. Paraskeva
1
, J. M. Mingot
2
,D.Go
¨
rlich
2
and
G. Simos
1
1
Department of Medicine, University of Thessaly, Larissa,
GREECE,
2
ZMBH, Heidelberg, GERMANY
The hypoxia inducible factor 1 (HIF-1) is the key mediator of the

cellular response to hypoxia. Its regulated alpha subunit (HIF-
1alpha) can be induced, in addition to hypoxia, by transition
metals, iron chelators and various growth factors, thus responding
to the activation of several signaling pathways. Regulation of
expression and activity of HIF-1alpha also involves several post-
translational modifications. We have recently shown that MAPK-
dependent phosphorylation of HIF-1alpha blocks its nuclear
export by CRM1 and facilitates its accumulation inside the nucleus
(1). To further understand the mechanism regulating the intracellu-
lar distribution of HIF-1alpha, we have investigated its nuclear
import using recombinant GFP-HIF-1alpha in digitonin-permeabi-
lized HeLa cells. We show that nuclear entry of HIF-1alpha is not
mediated by the classical NLS receptor importin alpha/beta but by
two other monomeric receptors of the importin-beta family. Bind-
ing assays confirm that these two importins can physically interact
with recombinant GST-HIF-1alpha. Taken together, our results
show that shuttling of HIF-1alpha between cytoplasm and nucleus
is a complex and regulated process involving several members of
the nuclear transport receptor family.
Reference
1. Mylonis I. et al., (2006), J. Biol. Chem. 281(44):33095.
B1-44
Searching for NLS and NES signals in the
Methoprene-tolerant protein
B. Greb-Markiewicz
1
, J. Dutko-Gwo
´
z´ dz´
1

, T. Gwo
´
z´ dz´
1
,
M. Orowski
1
, J. Dobrucki
2
and A. O
_
zyhar
1
1
Wroclaw University of Technology, Wroclaw, POLAND,
2
Jagiellonian University, Krakow, POLAND
Juvenile hormone (JH) regulates insect development by up to now
not completely characterized mechanism. The Juvenile Hormone
Resistance Protein (Methoprene-tolerant, Met), belonging to the
family of the bHLH-PAS transcription factors, is one of the candi-
dates for being a component of the Juvenile Hormone Receptor. It
is actually proposed that interdependence of JH and 20E (20-
Hydroksyecdysone) signaling pathways in insects comes from inter-
action between Broad-Complex (BR-C) and Met. Functional activ-
ity of the bHLH-PAS transcription factors is connected with
differential usage of the nuclear import (NLS) and nuclear export
(NES) signals. For this reason we have decided to perform a set of
experiments to identify regions of Met responsible for its subcellu-
lar localization. We have used pEYFP-C1 vector containing

enhanced yellow fluorescence protein for cloning full length and
some fragments of Met cDNA. In the next step, vectors were used
for COS-7 cells transfection and finally localization of expressed
fusion proteins was visualized with Confocal Microscopy System.
We will present results suggesting presence of more than one single
NLS and NES signal in the Methopren-tolerant, existing in differ-
ent fragments of protein.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 107
B1-45
Characterisation of NLS in human PHD1 and
PHD3
A. Hilz
1,2
, T. Schillinger
1
, S. G. Schindler
1
, M. Koehler
2
and
R. Depping
1
1
University of Luebeck, Luebeck, GERMANY,
2
Reha Clinic Damp
GmbH, Damp, GERMANY
HIF (hypoxia inducible factor) family members act as transcrip-
tional master regulators of genes involved in cellular and systemic

oxygen homeostasis. The prototypical member of the family HIF-1
consists of two subunits, HIF-1 alpha and HIF-1 beta. The alpha
subunit is hydroxylated at specific proline residues (Pro564 and
Pro402) by HIF-1 alpha prolyl hydroxylases (PHD-1, PHD-2, and
PHD-3) in an oygen dependent manner and thus targeted for poly-
ubiquitination and proteasomal degradation. From this it follows
that PHDs act as cellular oxygen sensors. It has been shown previ-
ously that PHD-1 is exclusively present in the nucleus, PHD2 is
mainly located in the cytoplasm and PHD3 is homogeneously dis-
tributed in the cytoplasm and the nucleus, indicating that the
PHDs act in different cellular compartments [Metzen, E, et al.,
(2003), J Cell Sci. Apr 1; 116, 1319]. To date less is known about
the mechanism of nuclear translocation of PHD-1 and PHD-3.
Here we report on the determination and characterisation of nuc-
lear localisation signals of PHD-1 and PHD-3. This study show
direct protein-protein interaction of both, PHD1 and PHD3, with
various classical nuclear import factors (importin alpha and beta).
In silico analysis of PHD1 predicted a nuclear localisation signal,
which we verified by mutant analysis.
B2-1
Incorporation of alkali-extractable proteins in
the Saccharomyces cerevisiae cell wall
I. Stuparevic, R. Teparic and V. Mrsa
Faculty of Food Technology and Biotechnology, Zagreb, CROATIA
Yeasts have evolved three different ways of attaching proteins to
cell wall glucan. Some proteins are bound to b-1,3-glucan noncova-
lently, while others are attached covalently through GPI-anchor
and b-1,6-glucan, or directly to b-1,3-glucan by alkali labile ester
linkage between the c-carboxyl groups of glutamic acid and hydro-
xyl groups of glucoses (Pir-proteins). A multiple disruption of genes

coding for the Pir family proteins (ccw5 ccw6 ccw7 ccw8) was per-
formed in order to investigate their potential role. After disruption
of the genes coding for all Pir family members, 67 kDa protein
band still remained in the NaOH extract. Disruption of SCW4
resulted in apparent disappearance of the 67 kDa band, indicating
that Scw4p could also be covalently linked to the cell wall. In order
to investigate the role of the Scw4p in the construction of the cell
wall, wilde type yeast was transformed with a high copy number
plasmid containing SCW4 gene. Phenotypes of this strain as well as
scw4, ccw5ccw6ccw7ccw8 and ccw5ccw6ccw7ccw8scw4 were exam-
ined. In order to get further insight in the binding mechanism a
novel, simple binding assay for Pir family proteins, was developed.
It has been shown that pir, as well as scw4 and scw10 mutant cells,
can bind externally added Ccw5p to their cell wall. A study of
appropriate binding conditions revealed the requirement of the
native conformation of Ccw5p. The presence of EDTA blocked the
binding of Ccw5p, indicating the cation dependence of the reaction.
Both wild type and mutant cells showed enhanced binding in 0.6 M
KCl. Further experiments should provide answers to question
regarding exact conditions required, as well as reveal if the binding
occurs autocatalytically or require a particular enzyme in the wall.
B2-2
The haemolysin A system of E.coli: from single
subunits to the whole transport machinery
S. Jenewein
1
, B. Holland
2
and L. Schmitt
1

1
Institute of Biochemistry, University of Duesseldorf, Duesseldorf,
GERMANY,
2
Institute de Ge
´
ne
´
tique et Microbiologie, Universite
´
de
Paris XI, Orsay, FRANCE
In Gram-negative bacteria, Type I protein secretion is a system
which exports proteins in one step across the inner and outer mem-
brane to the extracellular medium without any periplasmic interme-
diates. This machinery is able to transport large proteins, for
example the 108 kD RTX toxin haemolysin A (HlyA). In contrast to
Sec-dependent processes these proteins do not carry a classical N-
terminal signal sequence, but rather a signal sequence located at the
extreme C-terminus. The Escherichia coli HlyA transport machinery
is composed of haemolysin B (HlyB), a member of the ABC trans-
porter family, haemolysin D (HlyD) a membrane fusion protein,
both of which are located in the inner membrane and the outer mem-
brane protein TolC. HlyB is a so-called ‘‘halfsize’’ ABC transporter
consisting of a TMD and the cytosolic nucleotide binding domain
(NDB) fused together. It also contains a N-terminal C39 peptidase
domain which seems to be specific for Type I secretion systems. As a
first step towards characterization of this multicomponent system we
solved the crystal structure of the HlyB-NBD in different nucleotide
bound conformations, hence being able to describe the whole cata-

lytic cycle of the NBD in detail. However, in order to understand the
process of protein secretion in detail we overexpressed other compo-
nents such as the N-terminal C39 domain, full-length HlyB and dif-
ferent domains of HlyD. Furthermore using the haemolysin system
as a tool for heterologous protein overexpression we designed a vec-
tor being able to secrete large amounts of proteins into the media.
B2-3
Interaction analysis of HrpE with other TTSS
components of Pseudomonas syringae pv.
phaseolicola injectisome
M. N. Bastaki
1
, N. J. Panopoulos
1,2
and A. P. Tampakaki
3
1
Department of Biology, University of Crete, Heraklion-Crete,
GREECE,
2
Institute of Molecular Biology and Biotechnology
(IMBB), Heraklion-Crete, GREECE,
3
Department of Agricultural
Biotechnology, Agricultural University of Athens, Athens, GREECE
Pseudomonas syringae pathovars utilize a specialized protein secre-
tion pathway called «type III secretion system» (TTSS) to inject viru-
lence-related proteins into host cells. TTSSs are also used by
mammalian pathogens such as Yersinia, Salmonella and Shigella as
well as by plant and insect symbiotic bacteria. The TTSS injectisome

is made up of over 20 proteins which in the phytopathogens are
encoded by hrp/hrc genes. About half of these proteins are conserved
in most TTSSs and in the flagellum, implying that the assembly pro-
cess of the TTSS core is likely to be conserved. In this study, we
investigated pairwise interactions among proteins involved in TTSS
complex assembly by employing the yeast two-hybrid system (Y2H).
We studied several bait-prey combinations between TTSS compo-
nents of the phytopathogen P. syringae pv. phaseolicola and identi-
fied new protein-protein interactions. When used as bait, the full
length HrpE protein was found to interact with HrpO, HrcN as well
as with itself. The importance of these interactions is strengthened
by previous findings that the HrpE analogue of the flagellum, FliH,
interacts with and negatively regulates the FliI ATPase, the flagellar
homologue of HrcN. Y2H analysis with a series of HrpE deletion
mutants allowed us to map the regions of HrpE involved in the
above interactions in the C-terminal region of HrpE. Cell fraction-
ation analysis further indicated that these proteins co-localize in the
bacterial cell.
Abstracts Transport Machineries
108 ª 2007 The Authors Journal compilation ª 2007 FEBS
B2-4
Structural studies of AlgK: a lipoprotein
required for alginate biosynthesis in
Pseudomonas aeruginosa
C. Keiski
1,2
, P. Yip
1
, J. Koo
1,2

, H. Robinson
3
, L. L. Burrows
4
and
P. L. Howell
1,2
1
Hospital for Sick Children, Toronto, ON, CANADA,
2
University of
Toronto, Toronto, ON, CANADA,
3
Brookhaven National Laborat-
ories, Upton, NY, USA,
4
McMaster University, Hamilton, ON,
CANADA
The exopolysaccharide alginate is the major component of P. aerugi-
nosa biofilms infecting the lungs of cystic fibrosis patients. Ten pro-
teins, located on the algD operon, are required for alginate
polymerization and export. Since polymerization requires protein
components in both the inner and outer membranes, these proteins
are believed to form a large multi-protein complex that spans the cell
envelope and facilitates export of the polymer. To date, our struc-
tural and functional characterization of the alginate secretion com-
plex has focused on the outer membrane lipoprotein, AlgK. AlgK is
essential for alginate production and is hypothesized to interact with
the outer membrane secretin, AlgE. Recombinant AlgK from P. flu-
orescens has been expressed, purified and crystallized, and native

data collected to 2.5 A
˚
resolution at beamline X29 at the National
Synchrotron Light Source. The structure was solved using the Se-
SAD technique to 2.8 A
˚
resolution and an initial model of AlgK has
been built into the experimental map. The refinement of the structure
against the native data is in progress. Our model reveals that AlgK is
a solenoid protein with 9 a/a repeat motifs arranged in a right-han-
ded super helix, a structure that is consistent with the protein’s role
as part of the alginate secretion complex.
B2-5
Multi-resolution spatio-temporal analysis of
mammalian cells reconstructed in 3D by
electron microscope tomography
B. J. Marsh
Institute for Molecular Bioscience, St Lucia, AUSTRALIA
The beta cell – the sole source of the hormone insulin in mam-
mals – resides within the ‘islets of Langerhans’ in the pancreas.
We are focused on understanding the basic mechanisms that
underpin normal beta cell function, so that we can elucidate the
steps that lead to beta cell/islet dysfunction and ultimately,
diabetes. To this end, we combine fast-freezing with electron
microscope tomography (ET) to conduct comparative structure-
function studies of pancreatic islets isolated from mice and
humans. To complement insights from these high-resolution 3D
reconstructions (tomograms) of parts of cells, and to move
toward a more integrated or ’holistic’ approach to understanding
the mammalian cell as a unitary example of an ordered complex

system, we have undertaken a multi-scale/multi-resolution
approach to reconstructing mammalian (beta) cells in toto in 3D
at the EM level. By providing complete sets of 3D spatio-tem-
poral coordinates for cells at a range of resolutions that will
uniquely inform advanced in silico studies of 3D cell and
molecular organization, we are working to develop the world’s
first navigable ’Visible Cell atlas’. Such an interactive high-reso-
lution map of the 3D landscape of an entire mammalian cell
imaged and reconstructed by ET at £5 nm resolution will serve
as a unique international resource for protein and organelle
annotation, database integration and 3D visualization, and as a
framework for 4D animations of cells at pseudo-molecular reso-
lution.
B2-6
Direct evidence for the involvemen t of
VAMP-8/endobrevin in constitutive exocytosis
T. Takuma, M. Okayama, T. Arakawa and I. Mizoguchi
Health Sciences University of Hokkaido, Hokkaido, JAPAN
Studies with gene knockout mice have established that VAMP-8 is a
v-SNARE for the regulated exocytosis of various exocrine glands.
Although VAMP-8 is also believed to play an important role in con-
stitutive exocytosis and endosomal vesicle fusion, these processes
were normal in the mutant mice, probably because other v-SNAREs
can compensate the role of VAMP-8. In this study we have directly
examined the role of VAMP-2, -7, -8 in constitutive exocytosis by
TIRF microscopy. The plasmids of hGH and VAMPs with C-ter-
minal GFP-tags, prepared as described previously (Histochem Cell
Biol 125: 273–281, 2006), were transfected into HeLa cells and PC12
cells. Cells were incubated at 37 °C in a portable CO
2

incubator for
microscopy and TIRF images were obtained with a Nikon micro-
scope equipped with a CFI Plan Apo TIRF objective (100x/1.45 oil)
and EM-CCD camera, which were controlled by AQACOSMOS
(HAMAMATSU). Human GH-GFP was expressed in rapidly mov-
ing small vesicles in HeLa cells, and the most vesicles were abruptly
disappeared without increase in light intensity. VAMP-2-GFP
showed explosive exocytotic images in PC12 cells that were stimula-
ted with ionomycin, but it was very difficult to observe similar ima-
ges in HeLa cells. VAMP-7-GFP expressed in slowly moving vesicles
with various sizes, and that these vesicles were sticky each other. Ves-
icles containing VAMP-8-GFP showed clear explosive images of
constitutive exocytosis in HeLa cells.
B2-7
Docking and fusion of synaptic vesicles in
cell - free model system of exocytosis
G. Volynets
Palladin Institute of Biochemistry, Kyiv, UKRAINE
The study presents testing of docking and fusion of synaptic vesi-
cles in cell - free model system. The membrane fusion of synaptic
vesicles was measured using the fluorescent marker R18. The
method of photon correlation spectroscopy for determination of
particle sizes was used. We have shown that in the medium with
cytoplasmic proteins the sizes of membrane particles were
increased. But the fusion reaction of membrane components was
not detected in calcium-free medium. Thus, cytosolic proteins pro-
mote the synaptic vesicles into close proximity and they become
stably bound or docked. Therefore, the docking process didn’t lead
to membrane mixing, but changed the particle’s size. Fusion reac-
tion was shown to require Ca

2+
in micromolar range concentra-
tions. This fusion step was separated from the docked state in our
model. The specific action of synaptosomal cytosolic proteins on
synaptic vesicles was demonstrated using cytosolic proteins from
different rat tissues and bovine serum albumin. Our results indicate
that incubation synaptic vesicles with liver cytosol proteins or in
the bovine serum albumin solution do not lead to the enlargement
of the vesicles size. Studies of exocytosis in vitro have shown that
this process can be divided into Ca
2+
- independent step, termed
docking followed by rapid fusion step that is triggered by Ca
2+
.
Soluble proteins of synaptosomes have critical roles in the exocy-
totic pathway, perhaps involving modulation of docking site at the
targeting membrane.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 109
B2-8
The proteins participate in binding of
cell-surface extracellular nucleic acids
V. S. Mal’shakova, P. P. Laktionov and V. V. Vlassov
Institute of Chemical Biology and Fundamental Medicine,
Novosibirsk, RUSSIAN FEDERATION
Extracellular nucleic acids are always presented in biological fluids
and at the cell surface. It has been demonstrated that not only apop-
tosis and necrosis, but active secretion by cells input into their gen-
eration. Although the data on mechanisms of secretion of nucleic

acids by cells are absent to date. One way to justify the mechanisms
of secretion of nucleic acids is investigation of the compounds co-
transported with nucleic acids. Here we attempt to isolate proteins
bound with endogeneous extracellular nucleic acids. We have devel-
oped two methods of isolation of native nucleoprotein complexes
bound with the surface of eukaryotic cells. First method was based
on isolation of deoxyribonucleoprotein complexes using specific
binding of nucleic acids with glass surface. Second method was based
on isolation of complexes of proteins with biotinilated amplicons
(360 bp) using avidin-Sepharose. We have isolated 71, 69, 65, 46, 38
and 27-kDa nucleic acid-binding proteins from PBS/EDTA eluate
and trypsin supernatant of human endothelial cells. Proteins were se-
quenced by MS/MS mass spectroscopy and identified as zinc finger
proteins 585A and 586A, vesicular membrane protein p24, cytokera-
tin 8, myoneurin, kinesin-like protein KIF12, activator of G protein
signaling, MCTP 1 protein, inositol 1,3,4,6-tetrakisphosphate 5-kin-
ase, tubulin. Vesicular membrane protein p24 was shown to be impli-
cated in COPI and COPII vesicle formation, suggesting input of
similar mechanisms in extracellular secretion of nucleic acids.
Acknowledgement: The present work was supported by the
Russian Foundation for Basic Research (06-04-49485-a), Integra-
tion Project Siberian Division of Russian Academy of Sciences No
5.1.
B2-9
Regulation of insulin release in CAPS-deficient
mice
D. Speidel
1
, A. Salehi
1

, S. Obermueller
1
, I. Lundquist
1
,
N. Brose
2
, E. Renstro
¨
m
1
and P. Rorsman
1,3
1
Lund University, Malmo
¨
, SWEDEN,
2
Max-Planck Institute for
Experimental Medicine, Go
¨
ttingen, GERMANY,
3
Oxford Center for
Diabetes, Endocrinology and Metabolism, Oxford, UK
CAPS1 is thought to play an essential role in mediating exocytosis
from transmitter storing large dense-core vesicles in neuroendocrine
cells. Whereas antibody inhibition studies suggested a role for
CAPS1 in late stages of secretion, studies on CAPS1-knockout mice
pointed to a role in vesicular storage of transmitters. The function of

the second mammalian isoform, CAPS2, has been only basically
characterized. Recently, CAPS function in pancreatic beta-cells of
CAPS2-KO and CAPS1+/-; CAPS2-/- mice (DKO) has been stud-
ied in order to investigate involvement in peptide-hormone storage
and secretion, and relevance for diabetes. CAPS2-KO and DKO-
mice display reduced plasma insulin content and glucose intolerance
after intraperitoneal glucose challenge. In vitro insulin-release was
decreased in KO- and DKO mice. Membrane capacitance measure-
ments on single beta-cells indicated reduced exocytosis in KO- and
DKO-mice. Less vesicular fusion events were observed after perfu-
sion of KO-islets with 20 mM Glucose. Furthermore KO- and
DKO-beta-cells contain less docked granules. Total granule number
per cell and insulin content is reduced in DKO- but not in KO mice.
We conclude that CAPS1 and CAPS2 proteins regulate recruitment
and exocytosis of insulin storing granules. CAPS1 proteins have an
additional effect on granule numbers by promoting either granule
synthesis or stabilization.
B2-10
Cholesterol regulates prostasome release in
human prostate cancer PC-3 cells
A. Llorente
1
, B. van Deurs
2
and K. Sandvig
3
1
Dept. of Biochemistry, The Norwegian Radium Hospital, Oslo,
NORWAY,
2

Dept. of Medical Anatomy, Panum Institute,
University of Copenhagen, Copenhagen, DENMARK,
3
The
Norwegian Radium Hospital, Dept. of Biochemistry, Oslo,
NORWAY
Prostasomes are vesicles secreted by epithelial cells of the prostate
gland. The prostasome membrane is enriched in cholesterol and
sphingomyelin, and a recent proteomic study revealed that prosta-
somes contain at least 139 proteins. We have investigated whether
cholesterol plays a role in prostasome release using the human pros-
tate cancer cell line PC-3 as a model. Prostasomes released to the sup-
ernatants of PC-3 cells treated with cholesterol-depleting drugs were
isolated and analyzed by western blot. Caveolin-1 and LAMP-1, two
proteins found in PC-3 cells prostasomes, were used as markers of
prostasome release. Interestingly, methyl-beta-cyclodextrin (MBCD),
that in our experimental conditions reduced the cholesterol levels of
PC-3 cells by 40%, increased the amounts of both caveolin-1 and
LAMP-1 in prostasomes. Furthermore, the release of prostasomes
from PC-3 cells was also increased when the cellular cholesterol levels
were metabolically reduced by lovastatin and mevalonate. Moreover,
to investigate whether the total protein composition of PC-3 cells
and prostasomes was altered by MBCD, PC-3 cells were labelled
with [35S]methionine. The protein profiles of both cell lysates and
prostasomes were similar in control cells and in MBCD-treated cells,
but as expected, more radioactivity was associated with prostasomes
isolated from MBCD-treated cells. Finally, electron microscopy stud-
ies revealed that the morphology of prostasomes was not changed by
MBCD. In conclusion, these results suggest that cholesterol plays an
important role in the release of prostasomes.

B2-11
Transmembrane anchor may act as a signal for
endoplasmic reticulum secretion of type I
membrane glycoproteins
S. M. Petrescu, Sr.
1,2
, C. I. Popescu
1
, C. Paduraru
1
, A. Mares
1
,L.
Zdrentu
1
and R. Dwek
2
1
Institute of Biochemistry, Bucharest, ROMANIA,
2
Oxford
Glycobiology Institute, Oxford, UK
Membrane and soluble proteins are co-translationally translocated
into or across the endoplasmic reticulum(ER) membrane. Specific
signals determine the maturation and sorting events that allow the
ER export of the newly synthesized proteins. Transmembrane
domains (TMDs) are usually known as structural elements required
for the membrane insertion of integral membrane proteins. Are the
TMDs required simply to anchor proteins to the membrane or do
they also play an active role in the early events during protein mat-

uration? We provided here an example showing that the presence
of the TMD is compulsory for the productive folding and ER
secretion of a membrane glycoprotein. Tyrosinase is a type I mem-
brane protein regulating the pigmentation process in humans. We
have shown that soluble tyrosinase lacking its TMD is retained in
the ER becoming an ERAD substrate (1). We further used con-
structs of tyrosinase ectodomain fused with chimeric TMDs or
GPI anchor and found that all chimeras have been secreted (2).
The TMD controls tyrosinase folding by regulating the chaperone
recruitment at the translocon. This is a novel role for the TMD in
the folding and secretion of membrane bound proteins that adds to
our understanding of the multiple roles of TMDs and of the post-
translational regulation of protein expression.
References
1. Popescu CI, Mares A, Zdrentu L, Zitzmann N, Dwek RA, Pet-
rescu SM. J Biol Chem. 281, 21682–9, 2006.
2. Popescu CI, Paduraru C, Dwek RA, Petrescu SM, J Biol Chem.
280, 13833–40, 2005.
Abstracts Transport Machineries
110 ª 2007 The Authors Journal compilation ª 2007 FEBS
B2-12
Characterization of processing and secretion of
the N-terminal domain of a-dystroglycan
K. Matsumura, Y. Arai, F. Saito and T. Shimizu
Teikyo University School of Medicine, Tokyo, JAPAN
Purpose: Dystroglycan (DG) complex, composed of aDG and
bDG, provides a tight link between the extracellular matrix (ECM)
and intracellular cytoskeleton. aDG is an extracellular heavily gly-
cosylated mucin-type glycoprotein which is involved in the binding
of ECM proteins laminin and agrin. During the glycosylation of

aDG, its N-terminal domain is cleaved by a protein convertase
(PC) called furin. At present, the physiological function of the
cleaved N-terminal domain of aDG (aDGN) remains unknown.
Methods: Monoclonal antibody (1D9) and rabbit polyclonal anti-
body (Ry1528) were raised using the synthetic peptide of aDGN.
Fusion protein construct of aDGN was prepared. Ry1528 was
affinity-purified using the aDGN fusion protein (AP1528). Expres-
sion of aDGN in tissues, body fluids and cultured cells was ana-
lyzed by immunoblotting using 1D9 and AP1528.
Results: aDGN was detected as multiple bands in the range of
40 kDa in the culture medium of most of the cultured cells, but
not in the cells themselves. The amount of aDGN in the cell cul-
ture medium increased with the progression of culture time, while
it was undetectable in the cell culture medium when the cells were
cultured in the presence of CMK, an inhibitor of PC. aDGN was
most abundantly expressed in the brain and peripheral nerve, while
it was far less abundant in the other tissues including skeletal mus-
cle. aDGN was detected in not only the serum but also the cere-
brospinal fluid (CSF).
Conclusion: aDGN is cleaved by a PC and secreted into the cul-
ture medium of cultured cells. aDGN is most abundantly
expressed in the nervous tissues and secreted into not only serum
but also CSF in vivo.
B2-13
Novel role for ER calcium pump SERCA2b in
biogenesis of G protein-coupled receptors
J. Tuusa, P. Markkanen, P. Apaja, A. Hakalahti and
U. E. Peta
¨
ja

¨
-Repo
University of Oulu, Oulu, FINLAND
The endoplasmic reticulum (ER) is the major intracellular storage
for calcium (Ca
2+
) in animal cells. The high [Ca
2+
]ER /
[Ca
2+
]cytoplasm ratio is established by the Sarco(endo)plasmic ret-
iculum Ca
2+
ATPase 2b (SERCA2b). High [Ca
2+
]ER enables effi-
cient Ca
2+
signaling but is also required for synthesis of several
secreted and membrane proteins, because several ER chaperones
and folding catalysts require Ca
2+
. G protein-coupled receptors
(GPCRs) are polytopic membrane proteins whose function requires
highly dynamic structure - the fact that makes them excellent but
challenging models for studying biogenesis and quality control of
membrane proteins. We found that SERCA2b interacts physically
with newly synthesized incompletely folded precursors of the
human delta opioid receptor (hDOR) as well as other GPCRs in

the ER. Inhibition of SERCA2b by thapsigargin did not prevent
co-translational translocation of the hDOR, but did reduce associ-
ation with SERCA2b and increased targeting for degradation,
causing a reduction in the expression of mature receptors at cell
surface. Surprisingly, the lowered expression of mature receptors
correlated with decreased SERCA2b-hDOR interaction and not
with disturbed Ca
2+
homeostasis, because Ca
2+
ionophore
A23187 neither decreased interaction nor impaired receptor matur-
ation. The potential roles and mechanisms of SERCA2b in the
biogenesis of GPCRs will be discussed.
B2-14
Subcellular targeting of acyl-CoA synthetases
M. Becker, J. Wunsch, B. Rudolph, W. Stremmel and
J. Fu
¨
llekrug
University of Heidelberg, Heidelberg, GERMANY
Long chain fatty acids are important metabolites for the synthe-
sis of lipids and ATP, and the molecular mechanism of uptake
and utilization is of high medical relevance. Intracellular activa-
tion of fatty acids is mediated by enzymes catalyzing the esterifi-
cation with Coenzym A. These acyl-CoA synthetases have also
been implied in regulating fatty acid uptake. Mammalian long
chain acyl-CoA synthetases are subdivided into ACSL, bubble-
gum and Slc27/ FATP family proteins. Recently we localized
FATP4 to the endoplasmic reticulum (ER) and ACSL1 to mito-

chondria, and showed that the necessary and sufficient targeting
information was contained in the respective N-termini. We have
now systematically analyzed the subcellular sorting of eleven
acyl-CoA synthetases. GFP reporter proteins were compared by
confocal microscopy to organelle specific markers. While many
fusion proteins were restricted to the ER, we also observed local-
ization to the Golgi apparatus, lipid droplets and the cytosol.
We believe that this remarkable diverse subcellular localization
within one enzyme family has functional implications. From
expression data it is obvious that multiple different acyl-CoA
synthetases are present at the same time in the same cell type.
Substrate specificities likely account for some of this apparent
redundancy. Our tentative hypothesis implies that the spatial
organization of acyl-CoA synthetase activity is also a key factor
in channeling fatty acids towards a particular metabolic fate.
B2-15
Thermodynamic studies on SNARE complex
assembly
K. Wiederhold and D. Fasshauer
Research Group Structural Biochemistry, Department of
Neurobiology, Max Planck Institute of Biophysical Chemistry,
Go
¨
ttingen, GERMANY
The proteins syntaxin 1A, SNAP-25 and synaptobrevin play an
essential role during Ca
2+
-dependent neurotransmitter release.
Their stepwise assembly into a stable SNARE complex between
synaptic vesicle and plasma membrane brings the two membranes

into close apposition, eventually leading to membrane fusion. It is
thought that SNARE complex assembly proceeds in a zipper-like
process towards the C-terminal membrane anchors. Yet, this pro-
cess probably proceeds in an ordered fashion involving structurally
defined intermediates. In a first step, the two plasma membrane
proteins SNAP-25 and syntaxin form a transient 1:1 heterodimer,
which offers the binding site for the vesicle protein synaptobrevin.
However, a second syntaxin readily competes with synaptobrevin
for this binding site, leading to an unproductive 2:1 complex. This
off-pathway reaction renders it difficult to investigate the rate, spe-
cificity, and thermodynamics of synaptobrevin binding. To gain
deeper insights into this step, we now made use of the fact that the
heterodimer can be stabilized by a C-terminal synaptobrevin pep-
tide. This complex offers a free N-terminal binding site that allows
for rapid binding of synaptobrevin. In a subsequent step, the syna-
ptobrevin peptide is displaced. Here, we have applied fluorescence
spectroscopy and isothermal titration calorimetry to investigate
binding process of synaptobrevin to the stabilized intermediate.
Aim of this work is to understand the thermodynamic properties
of SNARE complex assembly.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 111
B2-16
Disrupting SNARE complexity – biophysical
investigations of NSF/SNAP-mediated SNARE
complex disassembly
U. Winter and D. Fasshauer
MPI for Biophysical Chemistry, Go
¨
ttingen, GERMANY

The formation of ternary SNARE-complexes is known to draw
apposing biological membranes close to each other and eventually
enable them to fuse. After fusion these extremely stable complexes
are dissociated into their single SNARE-components to allow fur-
ther rounds of fusion to occur. The disassembly reaction is endo-
thermic and therefore needs to be fuelled by the AAA-ATPase
NSF (N-ethylmaleimide sensitive factor) and its cofactors a-, b-
and c-SNAP. However, only little is known about how the disas-
sembly enzymes attack four-helix bundle SNARE complexes. In
order to investigate the reaction in molecular detail we established
several in vitro FRET and fluorescence anisotropy readouts for the
neuronal SNARE complex consisting of syntaxin1a, synaptobr-
evin2, and SNAP25. For all three neuronal SNAREs several single
cysteine variants exist to which fluorescent dyes can be specifically
attached. The fluorescence intensities of the dyes change upon
interaction of the respective proteins and allow for recording of
the speed of the disassembly process. By comparison of disassem-
bly kinetics of SNARE complexes in solution to that of complexes
incorporated into liposomes we attempt to determine possible
influences of membrane-anchorage on the efficacy of SNARE dis-
assembly. To characterize possible differences between the mammal
machinery to that of other organisms we have also performed
measurements using the yeast homologues of NSF/aSNAP called
sec18 and sec17 respectively.
B2-17
Studies on the molecular mechanism of the
acceleration of SNARE-mediated membrane
fusion by synaptotagmin 1
A. Stein, A. Radhakrishnan, D. Fasshauer and R. Jahn
MPI for Biophysical Chemistry, Go

¨
ttingen, GERMANY
Synaptotagmin 1 (Syt) is the calcium-sensor for fast, synchronous
release of neurotransmitter from synaptic vesicles. However, the
molecular mechanism of its actions and its interplay with soluble
N-ethylmaleimide-sensitive factor attachment protein receptors
(SNAREs) is not fully understood. The liposome fusion assay has
been widely used before to reconstitute the molecular events during
exocytosis. Indeed, it has been demonstrated that the formation of
tight SNARE complexes between two liposomes drives their slow
fusion. Addition of the soluble portion of Syt has been found to
moderately accelerate this fusion reaction. But does this effect
indeed reconstitute the very fast molecular action of Syt in secre-
tion? Here, we reinvestigated in detail the assay’s applicability to
study synaptotagmin’s function. We show that Ca
2+
-dependent
binding of synaptotagmin to membranes, both in cis- and trans-
configuration, as well as Ca
2+
-independent interaction with neur-
onal Q-SNAREs contribute to the accelerating effect of synapto-
tagmin in liposome fusion. However, we found that a fast fusion
reaction driven by a highly reactive Q-SNARE complex is not
accelerated further by Syt. Our results question the hypothesis that
synaptotagmin’s cellular function is truly mirrored in the liposome
fusion assay.
B2-18
A common binding mode for Sec1/Munc18-
proteins

P. Burkhardt and D. Fasshauer
Research Group Structural Biochemistry, Department of
Neurobiology, Max Planck Institute for Biophysical Chemistry, Go-
ettingen, GERMANY
The SNARE proteins syntaxin1, SNAP-25, and synaptobrevin play
a central role during Ca
2+
-dependent neurosecretion. Syntaxin1
and SNAP-25 are located in the plasma membrane, whereas syna-
ptobrevin resides on the synaptic vesicles. Their assembly into a
membrane-bridging ternary SNARE complex is believed to drive
membrane fusion. Sec1/Munc18-like (SM) proteins functionally
interact with SNARE proteins in vesicular fusion. Yet, structurally
and functionally dissimilar binding modes for SM proteins with
their cognate syntaxins have been uncovered: only a short N-ter-
minal peptide-finger of the syntaxin Sed5 binds to an outer surface
of the SM protein Sly1, whereas Munc18 tightly clasps around a
closed conformation of syntaxin 1, blocking its SNARE complex
association. We performed a detailed kinetic and thermodynamic
analysis of the Munc18/syntaxin1-interaction using a combination
of isothermal titration calorimetry (ITC) and fluorescence spectros-
copy. Furthermore, we investigated the influence of Munc18 on
the assembly pathway of the synaptic SNAREs. We now show that
the homologous peptide-finger of syntaxin1 and the closed confor-
mation of syntaxin1 simultaneously participate in binding to
Munc18. Consequently, the peptide-finger
´
s binding status deter-
mines the accessibility of Munc18-bound syntaxin for its partner
SNAREs, suggesting a common mechanism of how SM proteins

control SNARE complex formation.
B2-19
Biochemical analysis of toposome, a protein
mediating membrane-membrane interactions in
the sea urchin egg and embryo
J. J. Robinson, M. Hayley, P. J. Davis and M. R. Morrow
Memorial University, St John’s, NF, CANADA
Toposome is a protein found in the sea urchin egg and embryo
localized to both the yolk granule organelle and the plasma mem-
brane. In earlier studies, toposome has been shown to facilitate
cell-cell adhesion in the egg and embryo. In addition, toposome
may participate in yolk granule-plasma membrane fusion events
which restore the structural integrity of damaged membranes. We
are defining the mechanistic basis for toposome function. We have
previously shown that Ca
2+
induced secondary (apparent kd,
25 lM) and tertiary (apparent kd, 240 lM) structural changes in
toposome which facilitate the binding of toposome to membranes
as well as toposome-driven, membrane-membrane interactions.
Ongoing experiments, investigating the effects of Ca
2+
on thermal
stability and proteolytic digestion profiles also define toposome as
aCa
2+
regulated membrane binding protein. Dissociation and
reconstitution experiments confirm that toposome binding to the
membrane is reversible and regulated by Ca
2+

. The peripheral
association of toposome with the bilayer has been confirmed in 2
H-NMR experiments using liposomes composed of perdeuterated
dimyristoyl phosphatidyl serine. Chain orientational order was lar-
gely uneffected by toposome while quadropole echo decay times
were sensitive to toposome in the liquid crystalline but not the gel
phase. Collectively, our data support a model of toposome func-
tion in which Ca
2+
regulates both the association of toposome
with the membrane and subsequent toposome-driven, membrane-
membrane juxtaposition, perhaps as a prelude to a fusion event.
Acknowledgements: This work was supported by grants from
the Natural Sciences and Engineering Research Council of Canada
to JJR and MMR.
Abstracts Transport Machineries
112 ª 2007 The Authors Journal compilation ª 2007 FEBS
B3-1
The yeast proteins Ccz1p, Mon1p and Ypt7p are
involved in the movement and positioning of
vacuoles and nuclei
M. Hoffman-Sommer and J. Rytka
Institute of Biochemistry and Biophysics, Warsaw, POLAND
The Ccz1p protein of Saccharomyces cerevisiae is known to func-
tion in intracellular vesicular transport. It is necessary for efficient
trafficking of proteins to the vacuole and for proper vacuolar bio-
genesis. In its N-terminal part, Ccz1p shows homology to the
human protein HPS4, which is mutated in type 4 Hermansky-Pud-
lak syndrome. HPS4 is involved in intracellular transport and in
the biogenesis of lysosomes (the mammalian counterpart of yeast

vacuoles) and lysosome-related organelles, and also in lysosomal
movement and positioning inside cells. This prompted us to look
at the involvement of Ccz1p in the movement of yeast vacuoles.
Analysis of the transmission of vacuoles into buds revealed a
defect in vacuolar inheritance in ccz1 strains. Similar results were
obtained for cells lacking Mon1p or Ypt7p, proteins known to
cooperate with Ccz1p in protein trafficking. Since CCZ1 and
MON1 were known to display synthetic genetic interactions with
NUM1 (Tong et al. 2004) – a gene engaged in nuclear transfer dur-
ing cell division – we tested also their involvement in nuclear inher-
itance. Genetic analysis confirmed the synthetic interactions of
CCZ1 and MON1, as well as YPT7, with NUM1, and demonstra-
ted their specificity. Microscopic observation revealed a large pro-
portion of multinuclear cells in cultures of all three double
mutants, indicating strong enhancement of the nuclear transfer
defect of num1 cells. These results show that the proteins Ccz1p,
Mon1p and Ypt7p are necessary for efficient inheritance of vacu-
oles and nuclei.
B3-2
A novel mitochondrial parvulin specific for
great apes
J. W. Mueller
1
, D. Kessler
1
, P. Papatheodorou
2
, J. Rassow
2
and

P. Bayer
1
1
University of Duisburg-Essen, Essen, GERMANY,
2
Ruhr-Universita
¨
t Bochum, Bochum, GERMANY
The Parvulin Par14 is highly conserved in all metazoans and is
assumed to play a role in cell cycle and chromatin remodeling.
Recently, we confirmed the existence of a longer Parvulin mRNA
species expressed in all human tissues examined so far. This elon-
gated mRNA encodes a novel N-terminal domain whose expres-
sion was shown in human cell lysates. This novel Parvulin was
denoted as Par17 (Mueller et al., BMC Mol. Biol. 2006;7:9).
In contrast to Par14, endogenous Par17 was found in mitochond-
rial and membrane fractions of human cell lysates. Fluorescence of
EGFP fusions of Par17, but not Par14, co-localized with mitoch-
ondrial staining and the Par17 N-terminal domain was a prerequis-
ite for mitochondrial targeting. In vitro translated Par14 and Par17
associated with isolated human, rat and yeast mitochondria at low
salt concentrations, but only the Par17 mitochondrial association
was resistant to higher salt concentrations. Par17 was imported
into yeast mitochondria in a time and membrane potential-depend-
ent manner, where it reached the mitochondrial matrix. Moreover,
Par17 was shown to bind to double-stranded DNA under physio-
logical salt conditions. Based on the observation that the human
genome encodes Par17, but bovine and rodent genomes do not,
Par17 exon sequences from ten different primate species were
cloned and sequenced. Par17 is only encoded in the genomes of

Hominidae species including humans. Taken together, the DNA
binding Parvulin Par17 is targeted to the mitochondrial matrix by
the most recently evolved mitochondrial prepeptide known to date,
thus, adding a novel protein constituent to the mitochondrial prot-
eome of Hominidae (Kessler et al., BMC Biol. under review).
B3-3
Mitochondrial protein import: biogenesis of the
carrier translocase
N. Bolender
1
, N. Wiedemann
1
, B. Guiard
2
and N. Pfanner
1
1
Institut fu
¨
r Biochemie, Freiburg, GERMANY,
2
Centre de Ge
´
ne
´
tique
Mole
´
culaire, Centre National de la Recherche Scientifique, Gif-sur-
Yvette, FRANCE

Carrier proteins of the inner mitochondrial membrane are required
for transport of metabolites, nucleotides and cofactors across the
inner membrane. Mitochondrial carrier deficiency in humans leads
to lactic acidosis, cardiomyopathy, muscular hypotonia and causes
death in the first months of life. Like most mitochondrial proteins,
carrier proteins are synthesized as precursors in the cytosol and
imported into mitochondria in a post-translational manner. On the
way to their functional destination, carrier preproteins are first
translocated through the translocase of the outer membrane (TOM
complex). They are transferred across the intermembrane space
with the help of the soluble Tim9/Tim10 chaperone complex to the
inner membrane machinery, the carrier translocase (TIM22 com-
plex). The TIM22 complex consists of the integral components
Tim18, Tim22 and Tim54 and the peripheral components Tim12,
Tim9 and Tim10. Little is known about the function of Tim12 and
its role on the TIM22 complex. We analyzed the import and
assembly of Tim12. We show that newly imported Tim12 requires
endogenous IMS proteins to be assembled into the mature TIM22
complex. Tim12 import assembly is affected in Tim9 and Tim10
mutants, showing the requirement of these proteins for the Tim12
assembly. We purified the TIM22 complex and observed specific
interactions between components of the carrier translocase. Using
novel mutants of Tim12, we show that this protein play a dual role
in maintaining the stability of the TIM22 complex and in the bio-
genesis of mitochondrial carrier proteins.
B3-4
Assembly of TOM core proteins into the
mitochondrial outer membrane
D. Stojanovski, N. Pfanner and C. Meisinger
University of Freiburg, Freiburg, GERMANY

In Saccharomyces cerevisae the majority of mitochondrial proteins
are encoded by the nuclear genome and must be imported into the
organelle. A central junction in the import of all nuclear encoded
precursors is at the outer membrane where they encounter the
Translocase of the Outer Membrane (TOM complex). The TOM
complex consists of seven different subunits: the pore forming unit
Tom40, the small Toms, Tom5, Tom6 and Tom7 and the receptors
Tom22, Tom20 and Tom70. All Tom precursors are encoded by
the nuclear genome and must also be imported into mitochondria.
The pore forming unit of the TOM complex, Tom40, adopts a I
ˆ
2
-
barrel structural fold and requires the action of the Sorting and
Assembly Machinery (SAM complex) for integration and assem-
bly. The remaining components of the TOM complex are structur-
ally distinct from that of Tom40 and the mechanism by which
these constituents are integrated and assembled remains undefined.
The SAM complex is composed of three proteins, Sam50, Sam37
and Sam35. It is assumed that the SAM complex plays a specific
role in the biogenesis of I
ˆ
2
-barrel precursors. However, given that
I
ˆ
2
-barrel precursors engage with the SAM complex on the IMS
side, it is unclear why Sam37 and Sam35 expose domains on the
cytosolic face of the outer membrane. This feature led us to ana-

lyse the role of the SAM complex in the import and assembly of
alternative outer membrane proteins. This has revealed a role for
SAM components in the biogenesis of alternative Tom components
pointing to a novel and currently undefined function for the SAM
complex.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 113
B3-5
Biogenesis of mitochondrial intermembrane
space proteins – the MIA pathway
D. Milenkovic
1
, K. Gabriel
1
, B. Guiard
2
, N. Pfanner
1
and
A. Chacinska
1
1
Institute for Biochemistry and Molecularbiology, Freiburg,
GERMANY,
2
Gentre de Ge
´
ne
´
tique Mole

´
culaire, CNRS,
Gif-sur-Yvette, FRANCE
Mitochondria fulfill many important biological functions and have
also been implicated in apoptosis, ageing and a number of diseases.
The majority of mitochondrial proteins are encoded in the nucleus,
synthesized in the cytosol and have to be imported into mitochon-
dria. Import of these nuclear-encoded proteins is accomplished by
complex import machineries in the mitochondrial membranes and
aqueous compartments. The intermembrane space (IMS) of mito-
chondria came into focus recently with the discovery of a new
import and assembly pathway situated in this compartment, the
MIA (Mitochondrial Intermembrane space Assembly) machinery.
This machinery consists of two essential proteins, Mia40 and the
sulfhydryl oxidase Erv1, and is required for the import and assem-
bly of small IMS proteins with characteristic cysteine motifs. Fol-
lowing translocation across the outer mitochondrial membrane, the
IMS precursor proteins are initially recognized by the key compo-
nent of the pathway, Mia40. Mia40 forms an intermediate complex
with the incoming precursors and this interaction is based on the
formation of a disulfide bond. Mia40 is essential for the completion
of translocation across the outer membrane. Moreover, it dictates
the specificity of substrate entry into the MIA pathway as it is the
first step in the pathway and all subsequent biogenesis events of the
IMS precursors proceed from the Mia40-bound state. Therefore,
Mia40 functions as a receptor on the trans side of the outer mitoch-
ondrial membrane.
B3-6
Motion at the inner membrane: the
mitochondrial presequence translocase

M. Bohnert
1
, P. Rehling
1
, B. Guiard
2
, N. Pfanner
1
and
M. van der Laan
1
1
University of Freiburg, Freiburg, GERMANY,
2
Centre de
Ge
´
ne
´
tique Mole
´
culaire, CNRS, Gif-sur-Yvette, FRANCE
The majority of mitochondrial proteins are synthesized on cytoso-
lic ribosomes. A sophisticated import machinery catalyzes the
transport of these proteins to one of the four mitochondrial com-
partments, the outer membrane, the inner membrane, the inter-
membrane space and the matrix. Import of one subgroup of
mitochondrial proteins which is characterized by a positively
charged, cleavable, N-terminal signal sequence, is mediated by the
presequence translocase (TIM23 complex). The presequence tran-

slocase switches between two forms with distinct subunit composi-
tion and function: One form consists of the membrane-integrated
core proteins Tim23, Tim17 and Tim50 which are associated with
the presequence translocase-associated motor complex (PAM).
This molecular machine catalyzes the import of soluble matrix
components. The second recently described form of the prese-
quence translocase contains another membrane protein, Tim21,
but is devoid of the PAM complex. This so called sorting form
mediates integration of proteins into the inner membrane. The
sorting competent presequence translocase has recently been des-
cribed to be coupled to respiratory chain complexes. We investi-
gate the function and dynamic behaviour of the presequence
translocase with special interest on the bioenergetic requirements
of import processes.
B3-7
Lon protease in yeas t and mammalia
H. Adamuskova
1
, G. Ondrovicova
1
, J. Nosek
2
, J. Janata
3
,
C. Suzuki
4
and E. Kutejova
1
1

Institute of Molecular Biology, Bratislava, SLOVAKIA,
2
Department of Biochemistry, Faculty of Natural Sciences,
Komenius University, Bratislava, SLOVAKIA,
3
Institute of
Microbiology, Academy of Sciences of Czech Republic, Prague,
CZECH REPUBLIC,
4
Department of Biochemistry and Molecular
Biology, UMDNJ-New Jersey Medical School, Newark, NJ, USA
Lon protease is important for degradation of misfolded, unassem-
bled, oxidatively damaged and short-lived proteins. In yeast and
mammalian cells the Lon protease is located exclusively in mito-
chondria, were the vast majority of mitochondrial matrix proteins
function as multi-subunit complexes. We founded that human as
well as yeast Saccharomyces cerevisiae Lon protease recognizes and
initially cleaves of folded subunits of mitochondrial processing
peptidase based on their surface determinants. No degradation was
observed in case of the functional complex of these subunits.
Moreover human mitochondrial Lon protease binds the mtDNA
and is responsible for degradation of some mtDNA binding pro-
teins. At present, Lon protease from C. parapsilosis has been
cloned and characterized with the aim to look for the possible role
in mtDNA nucleoid function.
B3-8
Reversible phosphorylation of the
mitochondrial fission/fusion machinery as a
determinant for neuronal survival
J. T. Cribbs

1
, R. K. Dagda
1
, R. A. Merrill
1
, S. K. Nifoussi
1
,
M. Cleland
2
, R. J. Youle
2
and S. Strack
1
1
University of Iowa Carver College of Medicine, Iowa City, IA,
USA,
2
National Institutes of Health, Bethesda, MD, USA
Balanced fission and fusion reactions determine the shape and
interconnectivity of mitochondria. While mutations in mitochon-
dria-restructuring enzymes can cause neurodegenerative disorders,
the impact of death and survival signaling cascades on mitochond-
rial fission/fusion is largely unexplored. We show here that a neu-
ron-specific protein phosphatase 2A (PP2A) holoenzyme, PP2A/
Bb2, translocates to the outer mitochondrial membrane to promote
mitochondrial fragmentation, which in turn sensitizes hippocampal
neurons to oxidative and excitotoxic injury. Conversely, cAMP-
dependent protein kinase (PKA) tethered to mitochondria via A
kinase anchoring protein AKAP1 enhances neuronal survival by

maintaining the integrity of the mitochondrial network. In a search
for relevant substrates, we identified critical phosphorylation sites
in two of the large GTPases that mediate mitochondrial morpho-
genesis. PKA-phosphorylation of dynamin-related protein 1 (Drp1)
sequestered Drp1 in the cytosol and inhibited its mitochondria-
fragmenting activity. PC12 cells in which endogenous Drp1 was
replaced by the constitutively active (Ala) mutant displayed
increased sensitivity to several apoptotic stimuli, while the hypoac-
tive, phospho-mimetic (Asp) Drp1 mutant interfered with caspase
activation and cell death. Phosphorylation at a second site in the
Charcot-Marie-Tooth disease protein mitofusin-2 (Mfn2) promoted
mitochondrial fusion. Thus, reversible phosphorylation of Drp1
and/or Mfn2 by PP2A/Bb2 and PKA/AKAP1 controls the mitoch-
ondrial fission/fusion equilibrium as the pivot point for neuronal
life and death decisions.
Abstracts Transport Machineries
114 ª 2007 The Authors Journal compilation ª 2007 FEBS
B3-9
The peroxisomal serine protease Deg15 from
Arabidopsis thaliana processes the PTS2 signal
H. Schuhmann
1
, P. F. Huesgen
1
, C. Gietl
2
and I. Adamska
1
1
University of Konstanz, Konstanz, GERMANY,

2
Technical
University of Munich, Freising, GERMANY
We identified a serine protease from the Deg/HtrA family, named
Deg15, which is targeted to microbodies and is responsible for the
N-terminally processing of glyoxysomal malate dehydrogenase
(gMDH) in vitro and in vivo. A GFP-Deg15 fusion protein colocal-
ized in planta with a gMDH-CFP marker for microbodies, indica-
ting the peroxisomal localization of the protease. The residues of
the catalytic triad were identified by sequence similarity analysis of
Deg15 and homologs from other organisms. They were confirmed
experimentally by assaying the proteolytic activity of recombinant
Deg15 and mutants with replaced histidine-392, aspartate-491, and
serine-580 against the model substrate b-casein and recombinant
gMDH. A homozygous deg15 knock-out mutant line showed no
processing of gMDH in vivo, but processing was restored in com-
plementation lines carrying DEG15 cDNA in a knock-out back-
ground. This indicates that the Deg15 is the protease responsible
for cleavage of the peroxisomal targeting signal 2 (PTS2) in
A. thaliana.
B3-10
PTS2-dependent protein import into
peroxisomes
S. Grunau, W. Girzalsky and R. Erdmann
Institut fu
¨
r Physiologische Chemie, Ruhr-Universita
¨
t Bochum,
Bochum, GERMANY

The matrix protein import into peroxisomes uses either one of the
two well-characterized peroxisomal targeting signals, PTS1 or
PTS2 that are recognized and bound by the receptor proteins
Pex5p and Pex7p. This recognition event is supposed to take place
in the cytosol, which however, has not been confirmed. Whereas
Pex5p is capable to perform its role in PTS1 protein targeting on
its own, the PTS2-receptor Pex7p needs the auxiliary proteins
Pex18p and Pex21p. After its formation, the receptor-cargo com-
plex docks to distinct proteins at the peroxisomal membrane, prob-
ably Pex13p and Pex14p. These two proteins together with Pex17p
form the receptor docking complex, responsible for the initial bind-
ing of the receptor-cargo complex at the membrane. We focussed
on the PTS2-dependent protein import into peroxisomes with
Fox3p as target protein. In order to isolate and characterize
Fox3p-complexes, we constructed strains expressing Fox3p fused
to protein A. This fusion-protein, which turned out to be fully bio-
logical active, was used to isolate protein complexes from either a
soluble cell fraction or from Digitonin-solubilized membranes by
affinity chromatography. Soluble Fox3p was associated with the
PTS2-receptor Pex7p and the auxiliary protein Pex18p. Membrane-
bound Fox3p in addition contains components of the docking
complex. Furthermore, we analyzed the size of the protein-com-
plexes by BN-PAGE and gel filtration. From these experiments,
we estimated an approximate size of 200 and 300 kD for the sol-
uble and membrane associated Fox3p-complex. Our data demon-
strate that the receptor-cargo recognition in PTS2-dependent
protein import takes place in the cytosol and that it does not
depend on components of the docking complex.
B3-11
Functional characterization of the co-receptor

Pex18p in the protein import into peroxisomes
A. Korneli, H. W. Platta, F. El Magraoui, W. Girzalsky and
R. Erdmann
Institut fu
¨
r Physiologische Chemie, Ruhr-Universita
¨
t Bochum,
Bochum, GERMANY
The peroxisomal matrix protein import is facilitated by either the
PTS1-receptor Pex5p or the PTS2-receptor Pex7p. Both receptors
bind their cargo proteins in the cytosol, target them to the peroxi-
somal membrane and shuttle back to the cytosol after cargo
release. In contrast to Pex5p, Pex7p of S. cerevisiae requires in
addition the auxiliary proteins Pex18p and Pex21p. Interestingly,
Pex18p and Pex5p exhibit striking sequence similarities in the
N-terminal part of the proteins. In fact, it has been demonstrated
that the N-terminal half of Pex5p can be functionally replaced by
Pex18p. For Pex5p, conserved amino-acid residues that are present
in both proteins were shown to be crucial for its ubiquitination
and release back to the cytosol. Here we investigated the import-
ance of these conserved amino acids for Pex18p-function. Con-
served lysine- and cysteine-residues of Pex18p were replaced by
arginine or alanine, respectively. The mutated Pex18p was unable
to restore the growth defect of the corresponding deletion. The
mutations did not influence the binding of Pex18p to the PTS2-
receptor Pex7p and the docking of this receptor-cargo-complex to
the peroxisomal membrane. However, from the fact that the
mutated Pex18p is stabilized and that it accumulates at the peroxi-
somal membrane, we conclude that these residues within Pex18p

are essential for its dislocation from the peroxisomal membrane to
the cytosol as well as for its subsequent proteasomal degradation.
B3-12
Ubiquitination is a prerequisite for the receptor
cycle of the peroxisomal import receptor Pex5p
H. W. Platta, F. El Magraoui, S. Grunau, D. Schlee,
W. Girzalsky and R. Erdmann
Institut fu
¨
r Physiologische Chemie, Ruhr Universa
¨
t Bochum,
GERMANY
The peroxisomal import receptor Pex5p binds its cargo proteins in
the cytosol, targets them to the peroxisomal membrane and shut-
tles back to the cytosol after cargo release. Dislocation of Pex5p
from the peroxisomal membrane back to the cytosol is performed
by the AAA-proteins Pex1p and Pex6p. Pex5p can be modified by
mono- or polyubiquitination at the peroxisomal membrane but the
functional role of this modification was not known. We demon-
strate that Pex5p is polyubiquitinated by Ubc4p when its export is
blocked. In contrast, monoubiquitination is present under wild-
type conditions, most likely as a physiological modification of
Pex5p. We demonstrate by in vivo and in vitro ubiquitination
experiments that Pex4p/Ubc10p facilitates Pex5p monoubiquitina-
tion. Pex4p is essential for the biogenesis of peroxisomes but its
molecular target has been a mystery for 13 years. The functional
role of ubiquitination was tested by in vitro export assays. When
both mono- and polyubiquitination of Pex5p were blocked, release
of the receptor was completely inhibited. From this we conclude

that ubiquitination of Pex5p is a prerequisite for its dislocation
from the peroxisomal membrane by the AAA-peroxins. As ubiqui-
tination is essential for the receptor cycle, we have expanded the
energy-requirement of the peroxisomal import pathway by a sec-
ond ATP-dependent step, namely receptor-ubiquitination.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 115
B3-13
Identification of novel peroxisomal proteins: a
proteomic approach
C. Brocard, K. Gro
ß
schopf, V. De Wever, I. Dohnal and
A. Hartig
Max F. Perutz Laboratories, University of Vienna, Center of
Molecular Biology, Vienna, AUSTRIA
Peroxisomes are highly dynamic organelles present in virtually all
eukaryotic cells. These organelles play an important role in the cel-
lular metabolism, which is illustrated by the occurrence of severe
human diseases correlated with defects in peroxisome biogenesis
and metabolism such as Zellweger syndrome and Adrenoleukodys-
trophy. Their size, number and protein content greatly vary with
the environmental and physiological state of the cells. In yeast a
similar flexibility has been demonstrated upon growth on different
carbon sources. Nevertheless, knowledge about the enzymatic con-
tent of peroxisomes remains incomplete. To fully appreciate the
vital role of peroxisomes it is essential to identify all proteins asso-
ciated with this organelle. Likewise we seek to analyze peroxisomal
precursors, which are suggested to originate as buds from distinct
regions of the Endoplasmic Reticulum (ER). The yeast Saccharo-

myces cerevisiae was used as a model organism to apply an organ-
ellar proteomic approach. We established a novel isolation
procedure for peroxisomes based on a combination of density gra-
dient centrifugation and affinity purification. Proteins from highly
purified and intact organelles were fractionated and analyzed by
mass spectrometry. This procedure enabled us to identify a number
of proteins previously not known to be associated with peroxi-
somes. Here we present potential candidates, their involvement in
oleate utilization and their localization.
B3-14
Molecular recognition of dysferlin-containing
proteins at the pero xisome membrane
S. Merich
1
, A. Korneli
2
, C. David
2
, R. Erdmann
2
and
C. Brocard
1
1
University of Vienna, Vienna, AUSTRIA,
2
Ruhr-Universita
¨
t
Bochum, Bochum, GERMANY

Peroxisomes are vital organelles involved in lipid metabolism and
cellular detoxification. Among the proteins involved in peroxisome
biogenesis the dysferlin-containing proteins PEX30, PEX31 and
PEX32 were identified in the yeast S. cerevisiae to act as negative
regulators of peroxisome size and number. In order to establish
the function and spatiotemporal dynamics of PEX30 both in yeast
and human cells we analyzed its molecular interaction at the per-
oxisomal membrane.
We have isolated membrane protein complexes using TAP-tagged
forms of ScPEX30, ScPEX31 and ScPEX32. We could show that
ScPEX31 and ScPEX32 co-precipitate with ScPEX30. To decipher
whether these proteins build a macromolecular complex at the per-
oxisomal membrane we analyzed pull downs using blue-native elec-
trophoresis. The visualized complexes will be analyzed by mass
spectrometry to identify the complex building partners. In parallel,
we have exploited homology probing bioinformatics methods to
identify human homologs of ScPEX30. Here, we report the identifi-
cation and cloning of a putative human PEX30. We established sta-
ble cell lines that express various domains of this protein in frame
with a protein A tag in human embryonic kidney cells. Membrane
complexes will be isolated using this tag to identify interaction part-
ners of HsPEX30. We plan to evaluate variations in protein complex
formation by comparing cells grown under various culture condi-
tions e.g. applying oxidative stress.
B3-15
Analysis of PEX11 proteins
A. Huber, S. Merich, A. Hartig and C. Brocard
Max F. Perutz Laboratories, University of Vienna, Center of
Molecular Biology, Vienna, AUSTRIA
Size, number and protein content of peroxisomes can vary depend-

ing on the cell and tissue type and on the environmental conditions
of the cells. For example, yeast cells possess one or two peroxisomes
when grown on glucose as carbon source, but they contain up to
twenty when grown on oleate. Therefore, a molecular mechanism for
proliferation of peroxisomes must exist. Among the proteins known
to be directly involved in peroxisome proliferation are members of
the PEX11 protein family, small and basic proteins containing at
least one predicted transmembrane domain. In Saccharomyces cere-
visiae this family consists of PEX11, PEX25 and PEX27, the human
counterparts are PEX11A, PEX11B and PEX11G. Most likely, they
exert their function not alone but in a protein complex, and our aim
is the elucidation of the composition and function of these mem-
brane protein complexes.
We complemented pPex11-deleted yeast cells with plasmids expres-
sing either HsPEX11A or HsPEX11B proteins fused to GFP under
the control of the yeast PEX11 promotor, and followed the co-
localisation with the marker protein DsRed-SKL in the fluores-
cence microscope. In addition we generated constructs to express
yeast/human hybrid proteins. We aim to analyse and compare the
composition of the various protein complexes and the function of
their members in peroxisome proliferation.
B3-16
The role of EGY1 in plastid biogenesis
N. Li
The Hong Kong University of Science and Technology, Hong Kong,
CHINA
Egy1 was isolated as an ethylene-dependent gravitropism-deficient
and yellow green mutant. Molecular studies reveal that EGY1 gene
encodes a 59-kDa plastid-targeted metalloprotease and is up-regu-
lated by ethylene and light mainly at post-translational level in

light-grown hypocotyls. EGY1 is expressed constitutively in Ara-
bidopsis including endodermis and it is localized to chloroplast
and endodermal amyloplasts but not columella amyloplast. EGY1
plays a direct regulatory role in the light-induced replication of
endodermal amyloplast. It is also involved in regulation of lipid
biosynthesis.
Abstracts Transport Machineries
116 ª 2007 The Authors Journal compilation ª 2007 FEBS
B3-17
Characterization of chloroplast Hsp70s
R. M. U. Ratnayake
1
, H. Nonami
2
and M. Akita
2
1
The United Graduate School of Agricultural Sciences, Ehime
University, Matsuyama 790-8566, JAPAN,
2
Faculty of Agriculture,
Ehime University, Matsuyama 790-8566, JAPAN
Members of the HSP70 family proteins (Hsp70s) are reported asso-
ciated with membranes and protein aggregates at all subcellular
compartments of the cell with various functions. In chloroplasts,
presence of the Hsp70s has been reported at the stroma but with
unknown functions so far. Moreover, they are proposed to be
involved in chloroplastic protein import mechanism at the inter-
membrane space of the envelopes (known as Hsp70-Import Associ-
ated Protein) as motor molecules.

According to the recently released data on Arabidopsis thaliana
genome there are at least three Athsp70 genes which could be pre-
dicted to be localized at the plastids. To identify the suborganeller
localization of the putative AtHsp70s, we have constructed cDNAs
of these three Athsp70 genes into expression vectors. Afterwards,
the respective recombinant precursor proteins were overexpressed
in Escherichia coli and were in vitro imported into intact chloro-
plasts. Currently, positive sub localizations of two of the putative
AtHsp70s have been identified.
We are further experimenting to characterize and to investigate the
possible functions of these proteins at the specific subcompart-
ments and/or their relation to the components of the protein trans-
location pathways of the chloroplasts.
B3-18
Regulation of protein translocation across the
chloroplast envelope
E. Schleiff
LMU Mu
¨
nchen, Mu
¨
nchen, GERMANY
Protein translocation across and into membranes is a fundamental
process as up to 50% of all proteins synthesized in the cytosol need
to traverse at least one membrane to reach their place of function.
The basic principles of protein translocation have been established,
although the extent to which the molecular mechanism of each sys-
tem has been determined varies. Proteins destined for chloroplasts
are synthesised in the cytosol as precursors containing a targeting
signal. They are subsequently delivered to the surface of the organ-

elle by guiding complexes. At the surface, a membrane located multi-
subunit complex, the translocation machinery named Toc and Tic
complex (Translocon on the outer/inner envelope of chloroplasts
recognises the precursor proteins and facilitates their translocation.
Protein targeting toward and translocation into chloroplasts is a
highly ordered process involving the action of protein kinases and
phosphatases as well as GTP-dependent and redox-regulated
machines at the membranes. The recognition at the surface
requires the Toc complex. This complex is composed of so far five
identified subunits, which select proteins to be imported and facili-
tate their translocation across the outer envelope. Phosphorylation,
GTP hydrolysis and redox-sensing tightly regulate this molecular
ensemble as well as the dynamic assembly of this complex. We
were able to reconstitute this complex and its function in vitro.
Furthermore, initial structural information about components are
obtained. The mechanistics of protein translocation and its regula-
tion will be discussed.
B3-19
Lipid particle variants from the yeast
Saccharomyces cerevisiae
T. Czabany
1
, A. Wagner
1
, D. Zweytick
2
, E. Ingolic
3
,
M. Spanova

1
, I. Hapala
4
and G. Daum
1
1
Institute of Biochemistry, Graz University of Technology, Graz,
AUSTRIA,
2
Institute for Biophysics and Nanosystem Research,
Academy of Sciences, Graz, AUSTRIA,
3
Research Institute for
Electron Microscopy; TU Graz, Graz, AUSTRIA,
4
Institute of Ani-
mal Biochemistry and Genetics, Slovak Academy of Sciences, Ivanka
pri Dunaji, SLOVAKIA
In Saccharomyces cerevisiae the neutral lipids TAG (triacylglyce-
rols) and SE (steryl esters) are stored in so-called lipid particles.
Major yeast enzymes catalyzing synthesis of these neutral lipids,
namely the TAG synthases Dga1p and Lro1p, and the SE synthas-
es Are1p and Are2p, have been identified and partially character-
ized. To study the coordinate process of neutral lipid synthesis and
the specific contribution of each of the four acyltransferases to
lipid particle biogenesis we made use of triple mutants with only
one of the gene products active. All triple mutants form lipid parti-
cles, although at different rate, lipid composition, size and struc-
ture. These experiments also showed that TAG or SE alone are
sufficient to form lipid particles. Moreover, we studied yeast lipid

particle formation under anaerobic conditions. Interestingly, lipid
particles from anaerobic cells contain substantial amounts of squa-
lene which accumulates under these conditions. Similar results were
obtained with heme deficient cell grown under aerobic conditions.
Thus, yeast lipid particle properties vary as a result of growth con-
ditions and cellular properties. Supported by FWF projects 18857
and W901-B05 to G.D.
B4-1
Role of CaGAP permeases
L. Kraidlova
1
, M. Maidan
2
, H. Sychrova
1
and P. van Dijck
2
1
Institute of Physiology, Prague, CZECH REPUBLIC,
2
Institute of
Botany and Mikrobiology, Leuven, BELGIUM
The majority of fungal infections are caused by Candida species,
mainly Candida albicans. C. albicans can proliferate in many differ-
ent niches within the host and must be able to sence its environ-
ment in order to express only those genes to proliferate in that
area. In this case sensing and uptake of amino acids is very
important for C. albicans cell growth as well as for virulence. In S.
cerevisiae, one general amino acids permease Gap1 was found. It is
not only required for amino acids transport, but also for sensing

the presence of amino acids and activating signal transduction
pathways that induce many intracellular changes. In C. albicans,a
family of Gap1 homologues exists. In our work, we focused on
investigation the role of CaGAP genes. We would like to elucidate
the role of individual CaGap1 permeases in the virulence and path-
ogenecity of this species, together with the characterization of their
transport properties. For this we employ deletion of GAP genes in
the C. albicans SC5314 strain, using the URA3-based deletion cas-
sette, and heterologous expression of CaGAP genes in a S. cerevisi-
ae gap1D mutant to characterize substrate speficity and kinetic
parameters of individual CaGap permeases. This work was suppor-
ted by the GA CR 204/03/H066, MSMT LC531, and Czech-Fla-
mish bilateral project.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 117
B4-2
Topological analysis of a haloacid permease in
a Burkholderia species
Y. Tse, M. Yu and J. Tsang
University of Hong Kong, Pokfulam Road, HONG KONG
There are many computer programs and databases used for the
prediction of protein structures and functions. Despite their useful-
ness, predictions are often incorrect and require experimental veri-
fication. Such limitation is exemplified by the study of a haloacid
permease in the bacterium Burkholderia sp. MBA4.
The haloacid permease Deh4p is a membrane protein of 552 resi-
dues. Analysis with Pfam, ScanProsite and BLAST against TCDB
suggested that Deh4p belongs to the major facilitator superfamily
(MFS). Proteins in the MFS have been proposed to contain 12 or
14 transmembrane regions. Prediction of Deh4p using programs

like SOSUI, PHDhtm, SPLIT4 and TMHMM2.0 gave topology
models with 12 or 11 transmembrane domains. We have study the
topology of Deh4p using a dual-reporter system containing a
recombinant alkaline phosphatase-beta-galactosidase. A series of
Deh4p derivatives with progressively truncated C-terminal was
fused in-frame with the dual-reporter. The relative activities of the
two reporter enzymes indicate the location of the ends of the
Deh4p derivatives. More than 20 Deh4p-reporter proteins were
constructed and the enzyme activities measured. The results sug-
gested that Deh4p has no more than 10 transmembrane domains.
This indicates the need for further refinements of topology predic-
tion method and shows diversity in MFS transporter structures.
B4-3
Isolation and characterization of a novel
haloacid permease from Burkholderia cepacia
J. S. H. Tsang, M. Yu, Y. Faan and W. Y. K. Chung
The University of Hong Kong, Hong Kong, HONG KONG
Burkholderia cepacia MBA4 is a bacterium that can utilize 2-halo-
acids as carbon and energy sources for growth. It has been pro-
posed that dehalogenase associated permease mediates the uptake
of haloacid. Here, we report the first cloning and characterization
of such a novel permease. The structural gene, deh4p, was found
353 bases downstream of the dehalogenase gene deh4a. The nucleo-
tide sequence of deh4p was determined and an open reading frame
of 1656 bp encoding for a putative peptide of 552 amino acids was
identified. Deh4p has the signatures of sugar transport proteins
and integral membrane proteins of the major facilitator superfami-
ly. Uptake of
14
C-monochloroacetate (MCA) into the cell was

Deh4p dependent. Deh4p has an apparent K
m
of 5.5 and 8.9 M
and a V
max
of 9.1 and 23.1 nmol/mg/min for acetate and MCA,
respectively. A mutant with a transposon inactivated haloacid
operon failed to grow on MCA even when deh4a was provided
in trans.
B4-4
Influence of epigallocatechin gallate on artificial
lipid membranes
M. Dima, A. Iftime, A. Popescu and C. Ganea
Faculty of Medicine, Bucharest, ROMANIA
Epigallocatechin gallate (EGCG) is a natural flavonoid from the
tea plant (C. sinensis), with reported antioxidant and other health
beneficial effects. The Black Lipid Membranes (BLM) method has
been used to examine the interaction between EGCG and artificial
lipid bilayers, as reflected by recorded changes of the membrane
electrical parameters (capacitance and conductance). Experiments
were performed at constant EGCG concentrations in the BLM cu-
vettes and at increasingly higher concentrations, applied on both
sides of the membrane. At constant EGCG concentrations, a sta-
tistically significant difference (P<0.05) has been noticed between
insertion slopes at 5 lM and 20 lM, but not between the 20 lM
and 50 lM EGCG concentrations. This might suggests saturation
of insertion at higher EGCG concentrations.
In the presence of increasing EGCG concentrations the conduct-
ance showed non-significant statistical fluctuations, a possible clue
to a non-disruptive insertion of EGCG in the lipid bilayer (differ-

ent from results reported for other flavonoids). The biphasic capa-
city curve had an initial decrease followed by a slight increase and
a plateau. While the descendent phase is probably more dependent
on the membrane geometry, the ascendent phase could reflect com-
petition between antagonic effects on capacitance of the increasing
membrane thickness and changes of the dielectric constant. Our
results suggest concentration-dependent effects of EGCG on lipid
membranes, to be accounted for the usage of EGCG as a dietary
supplement.
B4-5
Phenanthrene uptake and membrane
adaptations of the PAH degrading Arthrobacter
sp. strain Sphe3
A. Kallimanis, S. Frillingos, C. Drainas and A. I. Koukkou
University of Ioannina, Ioannina, GREECE
This report describes phenanthrene uptake and the effect of phen-
anthrene on the membrane phospholipid and fatty acid composition
in a newly isolated bacterial strain, Sphe3, that we taxonomically
identified as Arthrobacter sp. strain Sphe3 is able to utilize phenanth-
rene as a carbon source and appears to internalize phenanthrene
with two mechanisms: a passive diffusion when cells are grown on
glucose, and an inducible active transport system, when cells are
grown on phenanthrene. Active transport followed Michaelis-Men-
ten kinetics with a Kt value of 0.19 ± 0.06 lM and it was amenable
to inhibition by 2,4-dinitrophenol and sodium azide. Moreover, the
saturation kinetics of the uptake implies specific binding of phen-
anthrene to the PAH-transporting system, with an apparent Kd of
0.48 lM. Evidence provided here indicates that, apart from inducing
an active PAH uptake, the presence of phenanthrene elicits signifi-
cant changes in membrane fluidity most likely due to decrease of the

ratio of the iso-anteiso fatty acids.
Acknowledgements: This project is co-funded by the European
Union-European Social Fund (ESF) and National Sources, in the
framework of the program ‘Pythagoras II’ of the ‘Operational Pro-
gram for Education and Initial Vocational Training’ of the 3rd
Community Support Framework of the Hellenic Ministry of Edu-
cation
Abstracts Transport Machineries
118 ª 2007 The Authors Journal compilation ª 2007 FEBS
B4-6
The role of multidrug resistance gene 1 (MDR1)
in breast carcinoma resistance
R. Vaclavikova
1
, G. Alnaes
2
, M. Hubackova
1
, E. Kubala
3
,
R. Kodet
3
, M. Mrhalova
3
, J. Novotny
4
, I. Gut
1
,

V. N. Kristensen
2
and P. Soucek
1
1
National Institute of Public Health, Prague, CZECH REPUBLIC,
2
Institute for Cancer Research, Oslo, NORWAY,
3
2nd Medical Fac-
ulty CU, Prague, CZECH REPUBLIC,
4
General Teaching Hospital
and 1st Medical Faculty CU, Prague, CZECH REPUBLIC
A significant part of tumor multidrug resistance to chemotherapy
is caused by ABC transporter P-glycoprotein encoded by MDR1
gene. Six MDR1 SNPs were analyzed in 90 sets of samples from
breast cancer patients by novel Nanogen technology allowing an
assessment of multiple SNPs on a single microarray platform.
MDR1 expression was quantified in breast tumor and non-neoplas-
tic tissues of these patients by real-time PCR. MDR1 was
expressed in 98.9% of tumor and in 97.5% of non-neoplastic sam-
ples and down-regulated in 89.7% of all tumors. No significant
correlation was observed between MDR1 expression and clinico-
pathological data. We revealed a high frequency of variant alleles
in MDR1 exon 12 (C1236T, q = 38.3) and exon 26 (C3435T,
q = 54.0). Individuals with variant alleles in these SNPs had a sig-
nificantly lower MDR1 expression in their tumors than patients
with normal genotype. SNPs in exon 12 and exon 26 also co-segre-
gated with estrogen and progesterone status. These SNPs may

affect function of P-gp and modify the clinical outcome.
Acknowledgements: Supported by GACR 305/07/P347, NIPH
Young Scientist Program and the National Program FUGE no.:
151924/150 and 15204/150, funded by The Research Council in
Norway.
B4-7
Crystal structures of a bacterial multidrug
transporter reveal a functionally rotating
mechanism
S. Murakami
Osaka University, Osaka, JAPAN
The emergence of bacterial multidrug resistance is an increasing
problem in the treatment of infectious diseases. AcrB and its
homologues are the major multidrug efflux transporter in gram-
negative bacteria, which confer intrinsic drug tolerance and multi-
drug resistance when they are overproduced. AcrB exports a wide
variety of toxic compounds including anionic, cationic, zwitterion-
ic, and neutral compounds directly out of the cells bypassing the
periplasm driven by proton motive force. It cooperates with mem-
brane fusion protein AcrA and outer membrane channel TolC.
The X-ray crystal structure of AcrB was first solved by Murakami
et al.[1] in 2002.
Now we solve the crystal structures of AcrB with and without sub-
strates in the new crystal form [2]. The new crystal structure solved
with new crystal form is asymmetric. The AcrB-drug complex con-
sists of asymmetric three protomers, each of which has different
conformation corresponding to one of the three functional states
of the transport cycle. Bound substrate was found in the periplas-
mic domain of one of the three protomers. The voluminous bind-
ing pocket is aromatic and allows multi-site binding. The

structures show that drugs are presumably exported by a three-step
functionally rotating mechanism in which drugs undergo ordered
binding change.
References
1. Murakami, S et al. Nature 2002; 419: 587–593.
2. Murakami, S et al. Nature 2006; 443: 173–179.
B4-8
Purification of S. cerevisiae membrane ABC
transporters involved in multidrug resistance,
stress response and membrane lipid homeostasis
C. Klein
1
, R. Ernst
2
, L. Schmitt
2
and K. Kuchler
1
1
Medical University Vienna, Wien, AUSTRIA,
2
Heinrich Heine Uni-
versity Duesseldorf, Duesseldorf, GERMANY
ABC transporters constitute a ubiquitous superfamily of membrane
pumps containing an evolutionary conserved ATP-binding cassette
domain. ABC proteins are responsible for energy-driven efflux of a
variety of substrates including amino acids, ions, sugars, xenobiot-
ics and toxins. A subset of them plays a major role in pleiotropic
drug resistance (PDR), where overexpressed ABC transporters such
as yeast PDR5 confer resistance to a vast variety of drugs. PDR in

yeast is similar to MDR occurring in mammalian tumor cells.
Little is known about physiological substrates for ABC transporters
or their molecular mechanisms of function, mainly due to limited
structural information. Human ABC proteins have been implicated
in the transport of lipids and sterols. Likewise, Yor1p and Pdr5p in
the yeast S. cerevisiae may also be involved in maintaining or modu-
lating membrane lipid homeostasis, which is also true for their main
transcription factors Pdr1p and Pdr3p.
Our goal is to overexpress and purify the major ABC transporters
Pdr5p, Pdr10p, Pdr15p, Pdr12p and Snq2p from S. cerevisiae.We
use an approach as used successfully for Pdr5p, by genomically
tagging the genes with the 14-HIS tag at the N-terminus. This sys-
tem also allows for overexpression, as the PDR5 promoter drives
all genes, which is highly active in a genetic background harboring
the PDR1–3 gain-of-function alleles encoding a hyperactive tran-
scription factor. Purified ABC transporters will be subjected to
reconstitution experiments as well as structural studies to unravel
their molecular mechanism of function, which could perhaps lead
to a better understanding of clinically relevant ABC transporters.
B4-9
Localization and function of ABCC/MRP efflux
pumps and organic anion uptake transporters
in human brain
A. T. Nies
German Cancer Research Center, Heidelberg, GERMANY
Integral plasma membrane proteins decisively control cellular sub-
stance uptake and efflux. We analyzed the localization of multidrug
resistance proteins (MRPs) and of organic anion uptake transporters
(OATPs) in human brain. MRPs belong to branch C of the ATP-
binding cassette (ABC) transporters and mediate the ATP-driven

efflux of endogenous and xenobiotic lipophilic organic anions. We
immunolocalized ABCC4, ABCC5, and ABCC11 in different brain
cell types where they may function in exporting physiological sub-
strates (e.g. reduced glutathione, sulfated steroids, cyclic nucleo-
tides), but also several cytotoxic and antiviral drugs. Together with
our immunodetection of OATP1A2 and OATP2B1 in brain capillary
endothelial cells, these data provide a basis for assessing the role of
uptake transporters and efflux pumps in the accessibility of the
human brain for organic anions.
References
1. Nies AT et al. Expression and immunolocalization of the multi-
drug resistance proteins, MRP1-MRP6 (ABCC1-ABCC6), in
human brain. Neuroscience 2004; 129: 349.
2. Bronger H et al. ABCC drug efflux pumps and organic anion
uptake transporters in human gliomas and the blood-tumor bar-
rier. Cancer Res 2005; 65: 11419.
3. Bortfeld M et al. Human multidrug resistance protein ABCC11
(MRP8), an apical efflux pump for steroid sulfates, is an axonal
protein of the central and peripheral nervous system. Neuro-
science 2006; 137: 1247.
4. Nies AT (2007) The role of membrane transporters in drug
delivery to brain tumors. Cancer Lett in press. DOI:10.1016/
j.canlet.2006.12.023.
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 119
B4-10
Differential effect of verapamil enantiomers on
the MDR phenotype induced by MRP1
T. Perrotton, D. Trompier, A. Di Pietro and
H. Baubichon-Cortay

IBCP, Lyon, FRANCE
Multidrug resistance protein 1 (MRP1) is a member of the ATP-
binding cassette (ABC) superfamily of membrane transporters, some
of which, like MRP1, are involved in cancer cell Multidrug Resist-
ance (MDR) phenotype. It has been previously shown for
P-glycoprotein (Pgp) and MRP1 that verapamil could modulate
their function. Verapamil acts on Pgp as an inhibitor, whereas it acts
on MRP1 as a modulator. Verapamil activates glutathione transport
by MRP1, leading to cell apoptosis induced by a strong cellular
glutathione efflux. Verapamil exists under two enantiomeric forms,
and we therefore investigated the potential effects of verapamil e-
nantiomers on the MRP1 transport function. The results show that
S-verapamil is responsible for the strong cellular glutathione efflux,
while R-verapamil is not cytotoxic. However, S-verapamil alone has
a stronger effect than the racemic mixture and R-verapamil was
demonstrated to antagonize the activity of S-verapamil on MRP1.
Indeed, we showed that the R-enantiomer, as the S enantiomer,
inhibited calcein transport by MRP1. Furthermore, R-verapamil can
revert the cell growth resistance to vincristine. A molecular study on
purified MRP1 shows that, in correlation with cellular effects, both
enantiomers bind to MRP1 and the conformational changes of
MRP1 induced by the two enantiomers are quite different, mediating
two distinct and specific effects of the isomers on the transporter.
These results could be of great potential interest in therapy.
They show that verapamil enantiomers display opposite effects on
MRP1 function, S-verapamil acting as a modulator inducing gluta-
thione efflux, and R-verapamil acting as an inhibitor blocking the
transport activity.
B4-11
Contribution of non-apoAI:ABCA1 pathways for

cholesterol homeostasis
M. Hossain
1
, N. Akita
1
, F. Kobayashi
1
, A. T. Remaley
2
,
S. Yokoyama
1
and M. Tsujita
1
1
Nagoya City University Graduate School of Medical Sciences,
Nagoya, JAPAN,
2
Lipoprotein Metabolism Section, NHLBI, NIH,
Bethesda, MD, USA
To reveal the importance of cholesterol homeostasis via non-apo-
AI:ABCA1 pathways, the diffusion mediated ’efflux’ pathway which
accelerate by LCAT, LCAT genotype were compared in ABCA1
deficient back ground. Wild, LCAT(D), LCAT(-), ABCA1(-),
LCAT(D)/ABCA1(-) and LCAT(-)/ABCA1(-), of 22 week-old mice
were examined. Euthanatized mice were perfused by PBS with subse-
quent fixation by 4% formaldehyde. Cholesterol content of each tis-
sue was measured by enzymatic color detection method. TC of liver
(44.57 ± 6.6, 43.3 ± 8.7 lg/mg prot) of both female and male mice
were increased in LCAT(-) (60.12 ± 5.5, 134.77 ± 10.4), ABCA1(-)

(62.58 ± 24.8, 79.8 ± 7.44) and in LCAT(-)/ABCA1(-) (80.3 ±
15.4, 197.3 ± 16.34). CE content in steroidogenic tissue was signifi-
cantly decreased in LCAT(-)/ABCA1(-). FC level of female spleen
and male brain were also significantly decreased in ABCA1(-) back
ground mice. No significant differences of TC contents were
observed in kidney, thymus, heart, and lungs. These results demon-
strate that ABCA1 null mice showed significantly higher cholesterol
accumulation in the liver. The result corresponds to our previous
finding with the ABCA1 specific inhibitor, probucol, which demon-
strated the liver is the major organ for apoAI:ABCA1 mediated cel-
lular cholesterol release, production of nascent HDL particles.
Interestingly, the cholesterol deposition was accelerated in ABC/
LCAT double deficient mice. This indicates that one of cellular non-
apoAI:ABCA1 pathways must also highly contribute to cholesterol
homeostasis in liver. Involvement of ABCG1 transporter in this phe-
nomenon will also be examined in future studies.
B4-12
Lateral compartmentation of proteins and lipids
in the plasma membrane: involvement of the
membrane potential
M. Opekarova, Sr.
1
, G. Grossmann, Jr
2
, J. Malinsky
3
and
W. Tanner
4
1

Institue of Microbiology, Prague, CZECH REPUBLIC,
2
Universityof Regensburg, Regensburg, GERMANY,
3
Institue of
Exp. Med. Academy of Sciences, Prague, CZECH REPUBLIC,
4
Cell Biology, University of Regensburg, GERMANY
The membrane potential is mainly considered as the driving force
for ion and nutrient translocation. We have discovered a novel role
of the membrane potential in the organization of the plasma mem-
brane.
Within the yeast plasma membrane, two non-overlapping sub-com-
partements have been visualized (1). The first one, represented by a
network-like structure, is occupied by the proton ATPase, Pma1,
and the second one, forming 300 nm patches, houses a number of
proton symporters (Can1, Fur4, Tat2) and the Sur7-protein.
We show that sterols, the main lipid constituent of the plasma
membrane, also accumulate within the patchy compartment (2).
This compartmentation is highly dependent on energization. Mem-
brane depolarization causes reversible dispersion of the H+-sym-
porters, not however of the Sur7 protein. Mitochondrial mutants
affected in the plasma membrane energization, show a significantly
lower degree of membrane protein segregation.
References
1. Malinska K, Malinsky J, Opekarova M, Tanner W. Visualiza-
tion of protein compartmentation within the plasma membrane
of living yeast cells. Mol Biol Cell 2003; 14: 4427–4436.
2. Grossmann G, Opekarova
´

M, Malinsky J, Weig-Meckl I and
Tanner W. Membrane potential governs lateral segregation of
plasma membrane proteins and lipids in yeast. EMBO J 2007;
26: 1–8.
B4-13
The change of taurine transport by oxidative
stress in the retinal capillary endothelial cell
and syncytiotrophoblast cells
Y. Kang
Sookmyung Women’s University, Seoul, REPUBLIC OF KOREA
Taurine transport to fetus occurs through syncytiotrophoblast, the
layer of placenta villi between maternal blood and fetal blood. The
retina is formed by retinal capillary endothelial cells (inner BRB,
iBRB) and retinal pigment epithelial cells (outer BRB). The inner
BRB plays important role in supplying nutrients to the retina from
the circulating blood.
We examined taurine transport by measuring [
3
H]taurine under
various conditions inducing cytokines, bacterial endotoxin, a com-
pound inducing a depletion of antioxidant, hydrogen peroxide and
nitric oxide donor at the iBRB (TR-iBRB cells) and in placenta
(TR-TBT cells). Taurine uptakes through taurine transporter
(TauT) showed a time, sodium ion and chloride ion-dependency,
and taurine uptake was decreased by PKC activator in TR-TBT
cells. Also, taurine uptake was decreased in calcium free condition
in those cells. Taurine transport in TR-iBRB cells was inhibited by
calcium channel blockers. The taurine uptake was increased by
TNF-a, LPS and DEM stimulation but decreased by H
2

O
2
and
NO stimulation. In addition, the taurine efflux was regulated by
stimulation with TNF-a in TR-TBT cells. Moreover we investi-
gated change of taurine uptake in TR-iBRB cells by hyperglycemic
effect. TR-iBRB cells exposed to high glucose medium and the
[
3
H] taurine uptake was reduced at the condition. Also, we induced
hyperglycemic condition using cytochalasin B, glucose transport
inhibitor, and [
3
H] taurine uptake was reduced in TR-iBRB cells.
Abstracts Transport Machineries
120 ª 2007 The Authors Journal compilation ª 2007 FEBS
B4-14
Cationic lipid-based nanostructures for gene
delivery and imaging
F. Chang, C. Chen, H. Huang, K. Chang, Y. Lin, M. Lin and
C. Cheng
National Taiwan University, Taipei, TAIWAN
Lipid-based nanoparticles such as liposomes and micelles have
been used in the drug delivery systems for decade. For gene ther-
apy, cationic lipid vectors were developed and generally consisted
of two major classes. One is based on hydrocarbon chain as back-
bone; the other is the use of cholesterol as backbone. Beside cati-
onic lipids, some helper lipids, e.g. phosphatidylethanolamine or
cholesterol, are needed in the formulations. In the field of cationic
cholesterol derivatives, it usually composed of cholesterol core and

aliphatic side chain with the positively charged functional groups.
A guanidino-containing cholesterol derivative, GEC-Chol, was syn-
thesized to improve the transfection efficiency and reduce cytotox-
icity. The physical parameters of these gene complexes including
morphology, size, surface charge of were evaluated. The GEC-
Chol formed cationic liposomes or micelles with diameters less
than 100 nm, and with the positive surface charges of 20–35 mV.
The transfection efficiency of GEC-Chol micelles was two order of
magnitude higher than other cationic cholesterols such as DC-
Chol. In addition, the micellar forms of lipid nanoparticles after
further functionalization displayed the structural stability and func-
tional selectivity in the cell culture system. Finally, the cationic
lipid-based nanostructures that contain inorganic nanoparticles
such as magnetic iron oxide nanoparticles or quantum dots have
been formulated and employed for live cell imaging and in vivo cell
tracking analysis.
B4-15
Dissecting a purine transporter: intrinsic
topogenetic signals, substrate binding and a
selectivity filter
A. Vlanti, A. Pantazopoulou, S. Amillis, I. Papageorgiou,
N. Lemuh, C. Gournas and G. Diallinas
University of Athens, Athens, GREECE
UapA, a member of the NAT family, is a uric acid-xanthine/H
+
symporter in Aspergillus nidulans. Mutant, chimaeric and truncated
versions of UapA were analysed by physiological tests, fluorescent
microscopy and kinetic assays using a plethora of purine ana-
logues. Specific internal regions were found to be critical for
expression in the plasma membrane, substrate binding and sub-

strate selection while the N- and C-termini included determinants
for ER-exit and ammonium-induced endocytosis, respectively.
More specifically: (i) specific residues (Q408, N409, G411, T416,
R417) within a highly conserved loop upstream from TMS9 seem
to interact with purines; (ii) the presence of an aromatic residue
(F528, Y528) within TMS12 plays the role of a novel-type selectiv-
ity filter that acts synergistically with the purine-binding site to
define UapA-mediated substrate translocation; (iii) the highly con-
served residues Q85 and H86 within TMS1 affect either transport
catalysis per se (Q85), or expression in the plasma membrane
(H86). Second-site suppressors of H86 map in TMS3 indicating a
close association of TMS1 and TMS3 for UapA topogenesis; (iv)
Leu to Ala mutations in a Leu Repeat (L77, L84 and L91) within
TMS1 of UapA were shown to alter differentially the binding
affinities of xanthine and uric acid, in a temperature-sensitive man-
ner, suggesting that the Leu interphase is involved in the fine flexi-
bility of UapA substrate binding site; (v) a conserved sequence in
fungal NATs, which inludes a di-acidic motif, was necessary for
ER-exit, while deletions of the N-terminus totally blocked insertion
into the ER; and (vi) deletion of the C-terminus led to a functional
UapA version that, unlike the wild-type, is not endocytosed by
ammonium.
B4-16
The role of
324
QN
325
and flanking sequences of
the nucleobases-ascorbate transporters (NAT)
in purine:H

+
symport
E. Georgopoulou, G. Mermelekas, E. Karena, P. Karatza,
P. Panos and S. Frillingos
Laboratory of Biological Chemistry, University of Ioannina Medical
School, Ioannina, GREECE
To study structure-function relationships of the nucleobases-ascor-
bate transporter (NAT) family, we employ Cys-scanning mutagen-
esis of YgfO, the xanthine permease of E. coli. In this context, we
analyzed
324
QN
325
, a sequence motif conserved in all purine-trans-
porting NATs, and the flanking
304
DGLVSVIASAVG-
SLPLTTFA
323
and
326
NGVIQMTGVASRYVGR
341
. We found
that: (a) Q324 is essential for high-affinity xanthine transport;
(b) N325 is irreplaceable for active transport; (c) D304 is also
essential, as mutants D304E, D304N and D304C display strikingly
low uptake activity; (d) the environment of
324
QN

325
is highly sen-
sitive to alkylation by N-ethylmaleimide (NEM); substrate binding
affects conformation by enhancing (A323C) or decreasing (N325C)
sensitivity to NEM; (e) residues 329–339 form an alpha-helix with
an alpha-helical stripe of residues (T332, G333, S336, V339) being
highly sensitive to inactivation by NEM; these residues are also
important for selectivity against different purine analogues;
(f) although sequence 304–341 is not exposed to the periplasm and
residues L319C or M331C are inaccessible from outside, N325C
and I329C are sensitive to inactivation by the hydrophilic, mem-
brane-impermeant reagent MTSES
-
; (g) P318 and G340 are essen-
tial for the conformational integrity of the protein and folding/
stability in the membrane. Taken together, our findings support the
conclusion that conformation in and around
324
QN
325
is important
with respect to purine binding and selectivity and may represent a
major determinant of the xanthine-translocation pathway.
Acknowledgement: This research is funded by the Greek Secre-
tariat for Research and Technology (GSRT), program NONEU-05.
B4-17
Molecular identification and functional
characterization of the vitamin C transporters
expressed by sertoli cells
C. Angulo

1
, M. A. Castro
1
, R. Maldonado
1
, D. Segretain
2
,
A. J. Yan
˜
ez
1
, J. C. Slebe
1
, J. C. Vera
3
and I. I. Concha
1
1
Universidad Austral de Chile, Valdivia, CHILE,
2
Universite
´
Paris
V, Paris, FRANCE,
3
Universidad de Concepcio
´
n, Concepcio
´

n,
CHILE
Vitamin C is a micronutrient that is essential for the normal devel-
opment of the male germ cells. The deficiency of vitamin C affects
the normal development of the seminiferous epithelium, in a man-
ner similar to the effect observed after androgenic deprivation. In
the male gonad, the germ cells are isolated from the systemic circu-
lation by the hematotesticular-barrier, which consists of a basal
layer of Sertoli cells associated through an extensive array of tight
junctional complexes.
A central question for a better understanding of testicular function
is how germ cells obtain vitamin C, considering that the hemato-
testicular barrier limits the direct access of molecules from the
plasma into the adluminal compartment of the seminiferous tubule.
In this study we analyzed the molecular identity of the transporters
involved in the transport of vitamin C in isolated Sertoli cells and
42GPA9 Sertoli cell line. WT1 was used as Sertoli cells specific
marker to further validate the use of 42GPA9 cell line as a model
system. The kinetic assays revealed that both, ascorbic acid trans-
porters (SVCTs) and facilitative hexose transporters (GLUTs), are
functionally active in these cells. This data is consistent with the
concept that Sertoli cells have the ability to take up vitamin C, a
finding that may be of importance to the physiology of the germ
cells. (FONDECYT 1060135)
Transport Machineries Abstracts
ª 2007 The Authors Journal compilation ª 2007 FEBS 121