JOURNAL OF PHYSIOLOGY AND PHARMACOLOGY 2003, 54, 4, 533–551
www.jpp.krakow.pl
A. SLOMIANY, B.L. SLOMIANY
LIPIDOMIC PROCESSES IN HOMEOSTATIC AND LPS-MODIFIED CELL
RENEWAL CYCLE. ROLE OF PHOSPHATIDYLINOSITOL 3-KINASE
PATHWAY IN BIOMEMBRANE SYNTHESIS AND RESTITUTION OF
APICAL EPITHELIAL MEMBRANE.
Research Center University of Medicine and Dentistry of New Jersey, Newark, NY, USA
Background:
Nuclear
transcriptome
initiates
specific
proteome
that
facilitates
metabolic events culminating in restitution of cell components and reproduction of
the discrete cellular function, but the magnitude of various genes induction and
following proteomic, lipidomic, and glycomic processes provide distinctness to the
final product and its function. In homeostasis, the challenged cell responds to stimuli
in defined and predictable mode but in the disease such as ulcerative erosions the
ablation of cell survival signals and cell apoptosis is enhanced. Therefore, to uncover
the
discreteness
and
dissimilarity
of
the
pathological
processes
induced
by
Helicobacter pylori (H. pylori) lipopolysaccharide (LPS), not only measurement of
the genomic events is crucial, but a complete cycle of events reproducing the cell
specific
proteins,
lipids,
and
cell-specific
environment
created
in
situ
require
thorough investigation. Methods: An impact of H. pylori LPS-induced processes on
posttranslational lipidomic activity in endoplasmic reticulum (ER), Golgi and apical
membrane was evaluated in the in vitro paradigm assembled with components of the
rat gastric mucosal epithelial cells. Results: In ER, the signals commanding synthesis
of biomembrane in the presence of control, the LPS-derived or LPS-admixed cytosol
was identical. The assembled vesicles contained the same amount of apoprotein and
had the same lipid composition. Their biomembrane contained the same amount of
sphingolipids in form of ceramide, which is determining factor of the ER-transport
vesicle completion. The transport of apoprotein in ER vesicles to Golgi was also not
changed. In Golgi, LPS-derived cytosol affected two distinct and concurrent with
assembly of Golgi transport vesicles processes. The LPS-derived cytosol affected
formation
of
Golgi
transport
vesicles
destined
to
apical
membrane
and
the
incorporation (fusion) of Golgi vesicles with apical epithelial membrane. The LPSderived cytosol decreased the production of Golgi vesicles by 15% and their fusion
with the apical epithelial membrane by 83%. In contrast with wortmannin, the LPS-
derived cytosol had no impact on Golgi transport vesicles association with the
epithelial membrane. Conclusions: We concluded that LPS interferes with MAPK-
dependent activation of cytosolic PLA2 since MAPKs immunoprecipitate added to
the LPS-cytosol restored activation of cytosolic PLA2-specific fusion of the Golgi
534
transport vesicles with apical mucosal cell membrane. On the other hand, wortmannin
that inhibited the association of Golgi transport vesicles with apical membrane,
interferes with cytosolic activity that controls association of PI3K-containing Golgi
vesicles with the apical membrane. Together, our studies present evidence that allow
to
conclude
that
LPS
affects
MAPK-specific
phosphorylation
and
PLA2-assisted
membranes' fusion, whereas wortmannin affects association of PI3K- and PI3P-
containing Golgi-derived transport vesicles with the membrane. In the final outcome,
both actions result in a diminished or inhibited restitution of apical membrane.
Key
w o r d s : Lipidomics, phosphatidylinositol 3-kinase, H. pylori LPS, apical biomembrane
restitution, vesicular transport, mucosal epithelium
INTRODUCTION
Information storage, information processing, and the execution of various
cellular
programs
reside
in
distinct
levels
of
cellular
organization,
the
cell's
genome, transcriptome, proteome, and metabolome (1,2). Hence, the distinctness
of the information retrieval is strongly influenced by the stimuli and by the state
of metabolome. One group of cycles that maps out the lipid-protein-metabolite
interaction involves biogenesis of cell membranes, and systematic and specific
restitution of the cell membrane compartments (3,4). The process depends on the
complex system of enzymes and substrates engaged in membrane biogenesis,
which is controlled by the milieu of the cell substructures and compartments.
Similarly, as the structural confines of endoplasmic reticulum (ER) membrane
determine translation of mRNA, the same site procures from cytosol the enzymes
and substrates for lipid synthesis (4). This epigenetic process generates new
segments of biomembrane, assures posttranslational transit of the protein to site
of its modification or function, and renews the intra- and extracellular membranes
(3-5). Together, ER and the surrounding cytosol control initial complex process
of the specific intracellular transfer and regeneration of the cell via transport
vesicles that deliver new protein inserted in the new segment of the cell-specific
membrane.
Such
components,
and
cycle
is
responsible
provision
of
the
for
the
precise
environment
that
restoration
propagates
of
the
cell
cell-specific
stimuli in homeostasis, or may also initiate cellular changes that are provoked by
the modified environment (6). The cell environment is created and dependent on
the components of cell biomembrane, and the milieu specific for the cell. If these
components are modified or the milieu is displaced or substituted, the cell is
either not able to respond to the own environment or is liable to interact with
nonspecific foreign stimuli (7,8).
Evidence on the initiation of membrane biogenesis shows that synthesis of
biomembrane is accomplished in stages. In ER, biomembrane synthesis leads to
the production of membrane structures in form of ER transport vesicles consisting
of phosphoglycerides and ceramides (3). Further complexity and specificity of
535
the
new
biomembrane
is
attained
in
Golgi
and
the
cell
membrane
(3-5,
9).
Ceramide-dependent synthesis of sphingomyelin and glycosphingolipids, and the
formation of phosphatidylinositol 3-phosphate (PI3P) is fueled by the delivery of
ER transport vesicles to Golgi and association of cytosolic p85 PI3K with Golgi.
The metabolites in cytosol determine Golgi-specific membrane augmentation and
Golgi vesicles completion (9). Finally, Golgi vesicles that contain PI3P and
PI3PK associate with the apical cell membrane (9). In the presence of active
cytosol, the membrane-associated Golgi vesicles fuse with the apical plasma
membrane with the aid of apical membrane-bound phospholipase A2 (PLA2) (9).
In our interpretation, this process restores a portion of the apical cell membrane
including the receptors that delimit the spectrum of possible signaling responses
of the cell, replaces the protein that determine the boundaries of exposure of the
membrane
to
the
lumenal
environment,
and
provides
PI3P
and
PI3PK
for
generation of the new sites for association of PX-domain-containing cytosolic
proteins (10).
In gastric epithelium, the process of apical membrane restoration is associated
with
provision
of
the
immediate
environment,
which
consists
of
a
newly
synthesized mucin delivered to apical site in the transport vesicles (3,6). The
described
cycle
represents
a
complete
round
of
homeostatic
processes
that
culminate in restitution of the functional apical epithelial surface. In gastric
diseases
associated
lipopolysaccharide
environment
with
(LPS)
restitution
Helicobacter
disturbs
(11,12).
the
pylori
orderly
Renewal
of
(H.
process
mucin
pylori)
of
cover
infection,
biomembrane
is
diminished,
its
and
and
epithelial cells' apoptosis is increased (11-13). We hypothesize that in the LPSexposed mucosal epithelial cells, the dramatic increase in apoptosis is attributed
to
down-regulation
or
interference
with
lipidomic
processes
that
control
the
homeostatic housekeeping of the cell membranes and its specific mucin-created
environment.
MATERIALS AND METHODS
Preparation of the LPS-cytosol from gastric mucosal cells.
Gastric mucosal cells preparation and the conditions of incubation with H. pylori LPS were as
described (6,8,12). The controls consisted of the preparations of the subcellular fractions from
normal
gastric
mucosa
or
where
indicated
admixed
with
the
LPS.
The
intracellular
transport
components consisting of active cytosol, ER, Golgi, the transport vesicles biogenesis, and their
fusion with the apical epithelial membranes were prepared as described earlier (3-5,9), unless
modifications are indicated. In some experiments the active cytosol was prepared from hepatocytes
and from gastric mucosal epithelial cells. Over the course of investigation of the intracellular
membrane biogenesis and the transformations from initial synthesis of phosphoglycerides in ER
and formation of ER transport vesicles, the experiments with subcellular organelles, membranes
and cytosol from hepatocytes and gastric mucosa were performed at least 10 times.
536
Preparation of transport-active cytosol from gastric mucosal epithelial cells.
Active cytosol for the in vitro synthesis of ER and Golgi transport vesicles and the experiments
on intracellular transport from ER to Golgi and from Golgi to apical epithelial membrane was
prepared as described previously (3-5,9). To remove mucin from apical mucosal surfaces, the tissue
was rinsed with homogenization buffer containing 1% Triton X-100 and subjected to subcellular
fractionation to separate ER membranes, ER-Golgi, and apical membranes (9). For clarity, the
nuclei were removed, ribosomes dissociated from ER, and translation inhibited (4).
Synthesis of PI3P in Golgi, and apical cell membranes
The
[ H]arachidonoyl-[ H]-inositol-labeled
3
ER
3
membranes,
Golgi
membranes
and
apical
membranes were incubated in the presence of active cytosol containing radiolabeled [ P]-ATP. Each
3
preparation
was
then
analyzed
for
membranes. The incorporation of the
the
label
incorporation
into
transport
vesicles
and
apical
P was monitored in PI of ER, PIP, PIP2 and PIP3 in Golgi,
32
Golgi vesicles and apical epithelial membranes. The fusion of Golgi vesicles labeled in PI3P with
[ H]-arachidonate
3
and
[ H]inositol
3
with
apical
membranes
was
determined
by
analysis
of
the
vesicles subjected to incubation under conditions allowing reversible association of the vesicles
(ATP-depleted cytosol) and after incubation in the presence of the active ATP-containing cytosol.
All details of the incubation and the media were described previously (9).
Assay of membrane fusion activity
The assay measured the release of [ H]arachidonic acid from the PI3P incorporated with Golgi
3
transport
vesicles
to
apical
membranes
(9).
The
reaction
was
initiated
by
addition
of
apical
membrane fractions adjusted to protein concentration of 1mg/ml and cytosol at 10mg/ml to Golgi
transport vesicles labeled with [ H]arachidonate and [ H]inositol. After 15 min at 37°C the reaction
3
3
was terminated by cooling mixture in ice water bath, and the vesicles, and the apical membranes
separated on sucrose gradient. The recovered, separated apical membranes and the remaining
Golgi vesicles were extracted with chloroform/methanol (2:1, v/v) containing 2% of 1N HCl. The
lipid extracts were then spotted onto silica gel thin layer chromatography plates and separated in
solvent mixtures detailed in (9). The radioactivity was determined using TLC Linear Analyzer and
scintillation counter (3-5,9). The impulses/min corresponding to 2.0-2.5 % disintegrations/min
were counted for 1-16 h. The linearity was checked using precalibrated standards. Formation of
lysoPI3P
was
determined
as
described
previously
(9).
The
spots
corresponding
to
PI3P
and
lysoPI3P were scraped from the plates and the radioactivity measured by liquid scintillation
counting. Formation of PIP2 and PIP3 was tentatively established by comparing the TLC of the
samples labeled with
3
H and
32
P.
Depletion of MAPKs from cytosol
Depletion of MAPKs from cytosol was performed by repeated incubation of 1ml of cytosol (810mg/ml
protein)
with
anti
MAPK
antibody.
The
immunoprecipitates,
and
the
cytosol
were
subjected to SDS-PAGE and immunoblot analysis using the same antibody at 1:1000 dilution.
H. pylori LPS, preparation of LPS-derived mucosal cell cytosol
H. pylori LPS was prepared from ATCC No. 4350 clinical isolate (11). The bacterium was
washed
with
water,
treated
with
ethanol
and
acetone,
dried,
and
homogenized
with
liquid
phenol-chloroform-petroleum ether. The resulting suspension was centrifuged, and the LPS
contained in supernatant was precipitated with water, washed with 80% phenol solution and
537
dried with ether. The dry residue was dissolved in small amount of water at 45°C, centrifuged
at 100,000g for 4 h, and the resulting sediment subjected to lyophilization. Mucosal preparations
were subjected to incubation in the presence of 400 ng/ml LPS for 16 to 20h. The cells recovered
by centrifugation at 300g for 3 min, were rinsed with saline and used for preparation the cytosol
used in transport vesicles biogenesis and transport experiments. This cytosol is referred to as
LPS-derived
or
LPS-cytosol.
Identical
preparations
without
LPS
treatment
were
used
for
preparation of control cytosol.
RESULTS
Synthesis of ER transport vesicles
The synthesis of ER transport vesicles was accomplished in vitro using
purified ER membranes, active cytosol derived from untreated control gastric
mucosa,
LPS-treated
gastric
mucosa
or
LPS-supplemented
cytosol
in
the
presence of [ H]serine or [ H]inositol and [ H]arachidonic acid The samples
3
incubated
for
3
indicated
3
period
were
separated
into
fractions
containing
transport vesicles, ER membranes and cytosol. In Figures 1and 2, the quantity
of transport vesicles formed and the incorporation of radiolabel serine into ERsynthesized
ceramide
phosphoglycerides
(Fig.1),
(Fig.2)
are
and
inositol
demonstrated.
The
and
arachidonate
results
indicate
that
into
ER
transport vesicles synthesis was not modified by the presence of LPS-cytosolinduced
signaling
vesicles
were
processes.
observed
and
Minimal
assigned
differences
to
in
experimental
the
yield
errors,
of
transport
rather
than
to
inducement by the investigated agents.
Fusion of ER-transport vesicles with Golgi
The next stage that contributes to membrane biogenesis, intracellular transport
and apical cell membrane restoration is reflected in the fusion of ER-transport
vesicles
with
Golgi
(4,5).
As
demonstrated
previously,
the
process
is
cytosol-
dependent and requires ATP-generating system. In our experiments, the fusion of ER
transport vesicles with Golgi was determined in the system containing ER-Golgi
membrane network. This approach eliminated the effects of the separation of ER
from Golgi and disabling activity in mechanically disrupted membrane network on
the overall fusion process with Golgi membranes. The fusion of ER-transport
vesicles
with
Golgi
was
followed
by
quantitation
of
[ H]serine-labeled
3
sphingomyelin and glycosphingolipids derived from the sphingoid base synthesized
in ER (as demonstrated in Fig.1) and transferred from ER to Golgi in ER transport
vesicles. Figure 3 illustrates the process of transport vesicles fusion with Golgi and
Golgi-specific maturation of the membranes reflected in the processing of the
ceramides from ER transport vesicles into sphingomyelin and glycosphingolipids.
Also, it demonstrates the effect of the LPS-cytosol on the synthesis and separation
of Golgi transport vesicles from Golgi membranes. While synthesis of sphingoid
538
Fig.
1.
LPS-induced
signaling
has
minimal
impact
on
initiation
of
biomembrane
synthesis,
ceramide synthesis in ER and formation of ER transport vesicles.
[3H]serine incorporation into ER membrane lipids was determined in the systems consisting of the
purified, ribosome-free ER membranes and the active cell cytosol from untreated controls, the cell
cytosol from mucosal tissue exposed to LPS, and the control cytosol admixed with LPS. Following
30 min incubation, the transport vesicles were separated from ER and the lipids analyzed (3-5). The
bars represent corresponding fractions from the one of the representative experiments in which the
ER
transport
vesicles
were
separated
on
sucrose
gradient,
the
corresponding
0.1
ml
aliquots
collected and the incorporation of the radiolabel into lipids of the ER and ER-transport vesicles
determined. For clarity, the fractions containing ER membranes, which remain in the sample aliquot
deposited at the bottom of the tube, are not shown.
lipids
in
Golgi
was
not
affected
by
LPS-derived
cytosol,
somewhat
more
radiolabeled lipids remained with Golgi membranes suggesting that LPS-cytosol
inhibited either completion of the Golgi transport vesicles or their scission from
Golgi membrane (Fig.3).
539
Fig. 2. LPS-induced signaling has minimal impact on the synthesis of phosphoinositides in ER and
their incorporation into ER-transport vesicles.
[3H]inositol incorporation into phosphoinositides of the ER transport vesicles was determined
under the conditions described in Figure 1. The material depicted in fractions 5-12 represents
radiolabeled lipids in purified vesicles. The material remaining in the aliquots of the ER-Golgi
membrane (fraction 1-4) was separated further into ER and Golgi and used for the determination of
the radiolabeled lipids that remained in the respective fractions.
Examination
of
the
Golgi
membrane
radiolabeled
with
inositol
and
arachidonate demonstrated the synthesis of PI3P in Golgi, the presence of PI3P
in
[
Golgi
P]ATP
32
vesicles
and
for
incubation
the
its
transfer
with
with
vesicles
cytosol
that
to
apical
afforded
membranes.
formation
of
Using
Golgi
transport vesicles, it was determined that the majority PI3P synthesis occurred in
Golgi
(Fig.4).
Although
p85PI3K
readily
associated
with
Golgi,
and
apical
membranes (Fig.5), the synthesis of PI3P was only apparent in Golgi. The Golgi
vesicles membranes contained PI3P and PI3K, but were not involved in the
synthesis
of
PI3P
incorporation of [
or
other
phosphoinositides.
In
apical
membranes,
the
P] was evident but reflected rather synthesis of PI3P2 or
32
PI4P2, since pattern of the radiolabeling was somewhat different than that in
Golgi and the vesicles. Also, it is possible that to some degree, the appearance of
PI4P was due to membrane cross-contamination.
540
Fig. 3. LPS impairs formation
of Golgi transport vesicles.
The figure depicts the amount
of
labeled
sphingolipids
remaining in Golgi membrane
after
incubation
Golgi
transport
that
affords
vesicles.
abbreviation
polyglycosylceramides
GSL),
sphingomyelin
(poly(SPM),
triglycosylceramides
GSL),
The
denote
(Tri-
diglycosylceramides
(Di-GSL),
ceramides
monoglycosylMono-GSL)
and
ceramides (Cer). The numbers
in the table reflect the amount
of
the
radiolabeled
lipids
recovered from 4 mg of Golgi
membranes
(30,000
cpm/sample).
Association and fusion of Golgi transport vesicles with apical membrane.
Impact of LPS-derived cytosol.
The incorporation of newly synthesized Golgi-derived biomembrane into the
apical
plasma
membrane
[ H]arachidonate-,
3
was
[ H]inositol3
followed
and
by
tracing
[ H]arachidonate-,
3
the
and
incorporation
[ P]
32
of
ATP-labeled
phosphatidylinositol phosphate (PI3P) of Golgi vesicles. Identification of PLA2-
specific products was accomplished by analysis of the vesicles labeled with inositol
and arachidonate. In the previous studies we have demonstrated that Golgi attracts
from cytosol p85 PI3K and that in the presence of active cytosol ensues synthesis
of transport vesicles containing phosphatidylinositol 3-phosphate (PI3P) (9). Thus
completed Golgi transport vesicles dissociate from Golgi, and in the presence of
active cytosol react with the apical membranes. The reaction consists of two stages,
first the vesicles associate with the membrane and then a fraction of the associated
vesicles fuses with the membranes. Based on the analysis of the [ H]inositol- and
3
[ H]arachidonate-labeled Golgi vesicles, about 39 % of the vesicles associate with
3
apical membrane within 15 min incubation period, of which 47% in the presence of
ATP-containing active cytosol undergoes fusion with the apical membrane. In the
presence of the LPS-cytosol, the fusion is inhibited by an 83% (Fig.6).
In addition, the analysis of lipids in the transport vesicles that only associate
with the apical membranes in the presence of active cytosol and those which fuse
and their radiolabel incorporates into apical membranes revealed presence of
PIP2 (Fig.7). The comparison revealed that following fusion of Golgi vesicles
with apical membranes 59.3 % increase in free arachidonate, 53.4 % decrease in
PI3P and a 44.5 % increase in PIP2 occurred. Since the increase in arachidonate
541
Fig.
Phosphatidylinositol
4.
3-phosphate
(PI3P)
and
phosphatidylinositol 4-phosphate (PI4P) are synthesized in Golgi.
The
incubation
with
[
32
P]ATP
in
active
control
cytosol
was
performed with Golgi membranes (lane 1), Golgi vesicles (lane 2)
and with the apical membranes (3). The lipids of the incubated
membranes were extracted with the solvent described previously and
separated
on
HPTLC
phosphoinositides.
Prior
plates
to
prepared
separation
of
separation
of
phosphoinositides,
for
the
samples were subjected to separation of phosphatidic acid (PA). As
shown labeled PA was found in Golgi membranes only. The labeled
lipids in Golgi transport vesicles consisted mainly of PI3P, whereas
apical
membrane
contained
PIP2
and
PIP3.
Although
PI4P
was
synthesized in Golgi and found in Golgi transport vesicles (lane 2),
the apical membranes did not contain detectable amount of PI4P
(lane 3).
could only result from action of PLA2, we concluded that the final step in the
fusion
of
the
Moreover,
we
vesicles
also
into
apical
noticed
that
membrane
during
requires
membrane
participation
incorporation
of
and
PLA2.
PI3P
hydrolysis, there was a prominent increase in polyinosities. On the average a 44.5
%
increase
in
the
formation
of
polyinositides
was
determined.
It
shoud
be
remembered that the determined changes in lipid profiles relate to Golgi transport
vesicles only, since apical membranes were not labeled.
As shown in Fig.6, the incubation of Golgi transport vesicles with apical
membranes in the presence of cytosol afforded three fractions of Golgi vesicles:
the free, unreacted vesicles (A), the apical membrane associated vesicles (B) and
the apical membrane-fused vesicles (C) The results of the experiments with the
LPS-derived,
the
LPS-admixed
with
MAPKs
and
MAPK-depleted
cytosol
demonstrated that fusion of the transport vesicles with apical membrane was
inhibited
by
the
LPS-derived
cytosol,
and
enhanced
by
ATP-
MAPKs-
supplemented cytosol (Fig.6). The LPS-cytosol generated similar effect as the
incubation under chelating conditions (EGTA) and ATP-depleted cytosol (9). In
Fig.
5.
p85
PI3K
associates
with apical membranes, Golgi
membranes and Golgi vesicles,
lane
after
2,3,4
before,
incubation
lane
with
5,6,7
active
cytosol containing p85 PI3K,
respectively.
depicts
markers.
Lane
prestained
1
and
8
molecular
542
Fig. 6. LPS-induced processes
inhibit
with
Golgi
vesicles
apical
fusion
membrane.
The
fraction A shows the portion of
free
transport
vesicles
recovered with cytosol, fraction
B depicts portion of the vesicles
that were recovered with apical
membrane but were not a part
of
the
apical
membrane
fusion
stage)
and,
depicts
Golgi
(pre-
fraction
vesicles
C
which
fused apical membrane and the
radiolabeled
lipids
permanently
incorporated
were
into
the apical membrane.
both, the transport vesicles could associate with the apical membrane but were
easily separated and their radiolabeled PI3P lipid profile remained unchanged (9).
Thus, it appears that permanent incorporation of transport vesicles into apical
membrane
activation
(fusion)
of
PLA2.
which
Based
is
on
ATPthe
and
MAPK-dependent,
evidence
provided
in
is
this
controlled
study,
and
by
the
experiments with specific inhibitors of MAPKs (8), we conclude that the cytosolic
MAPK-dependent PLA2 phosphorylation is affected by LPS-derived cytosol, and
that in turn inhibits fusion of Golgi transport vesicles with apical membrane.
Impact of worthmannin on Golgi vesicles fusion with apical epithelial
membrane
Wortmannin admixed cytosol was not affecting Golgi transport vesicles PI3K
(Fig.8).
However,
decreased
Golgi
preincubation
vesicles
of
association
apical
with
membranes
apical
with
membrane
wortmannin
and
inhibited
association of the cytosolic p85 PI3K with the membrane. This demonstrated to
us that wortmannin interfered with the process of PI3K binding and hence the
processes in apical membrane that initiate PI3K-specific association of the PI3Kcontaining Golgi vesicles.
Thus,
the
reversible
association
of
Golgi
transport
vesicles
with
the
membrane, which is PI3K-specific, reflects the first stage in transport vesicles
543
acquisition by an apical membrane, the stage that is not associated with PI3P
synthesis
or
PI3P
hydrolysis.
The
second
stage,
the
fusion
of
Golgi
vesicle
membrane with the apical membrane is ATP- and MAPKs-dependent, and PLA2-
specific, and its completion is evident when the Golgi vesicle-derived membrane
is incorporated permanently into apical membrane. In this concluding step an
arachidonate is released from the portion of PI3P and the portion of PI3P appears
to be phosphorylated further. Thus, the cycle of membrane biogenesis, appends to
the
apical
membrane
a
new
segment
of
biomembrane
consisting
of
new
membrane with new receptor proteins and PI3P substrates.
DISCUSSION
Cell survival pathways are invaluable for the design of rational approach to
determine the processes that establish homeostatic renewal of the cell, as well as
those causing pathological changes observed in disease such as gastric ulcer or
cancer. Cellular activities expressed in homeostatic restitution of the cell reflect
the processes that take place in tissue exposed to its normal environment. In
Fig. 7. Fusion of Golgi transport vesicles with apical membrane is reflected in PLA2-specific release
of arachidonate from PI3P and an increase in DAG.
[3H]inositol-
and
[3H]arachidonate-labeled
Golgi
transport
vesicles
and
unlabeled
apical
membranes were subjected to incubation in the presence of active cytosol (9). After 15min at 37°C,
the incubation mixtures were subjected to separation; the apical membrane fractions were treated
with 2M urea to remove associated Golgi transport vesicles. The released vesicles and the apical
membrane
fractions
were
subjected
to
lipid
extraction
and
HPTLC
(9).
Each
analysis
was
performed with the extracts containing up to 20,000cpm per sample. The bars represent the quantity
of
radiolabel
recovered
phosphatidylethanolamine,
PIP2-phosphatidylinositol
with
AA
arachidonic
PC-phosphatidylcholine,
bisphosphate.
The
acid-,
DAG-
diacylglycerol,
PIP-phosphatidylinositol
figure
shows
the
phosphate,
representative
data
of
PEand
six
independent experiments, which also served to identify the individual lipids and PLA2-specific lipid
products.
544
Fig. 8. LPS-induced changes in cytosol prevent fusion of Golgi transport vesicles with apical
membrane, whereas wortmannin inhibits association of Golgi vesicles with apical membrane.
Effect
of
preincubation
of
apical
membranes
with
wortmannin
or
LPS
on
the
association
of
transport vesicles with apical membrane. The association-fusion of Golgi transport vesicles with
apical membranes is expressed in the appearance of p85PI3K in the recovered apical membranes.
Golgi transport vesicles and apical membranes were incubated in the presence of active cytosol for
15
min
at
37°
preincubated
experiment.
C.
with
The
Where
10
nM
western
indicated,
the
wortmannin
blot
of
the
apical
or
100
p85PI3K
membranes
ng
LPS,
was
or
or
the
performed
Golgi
transport
LPS-cytosol
on
apical
vesicles
was
used
membranes
were
in
the
fraction
recovered from incubation mixtures. The analysis was performed as described previously (9). Lane
1,10-prestained molecular weight markers, 5,9-control mixture before incubation consisting of
apical membranes, Golgi transport vesicles and active cytosol, or LPS-cytosol, respectively, 2,6same mixtures as in lanes 5,9 but after 15 min incubation, 3,7-apical membranes preincubated with
10nM wortmannin, or 100ng LPS, respectively, and then with Golgi transport vesicles and active
cytosol, 4,8-apical membranes incubated with active cytosol and with Golgi transport vesicles that
were preincubated with wortmannin or LPS, respectively.
contrast,
the
processes
evoked
in
mucosal
cells
by
H.
pylori
LPS
reflect
disturbances in mucin synthesis and secretion, and increased apoptosis (8,11-13)
and hence demonstrate the impact of the LPS on intracellular signals evoked by
LPS-modified cell environment.
The major mechanism for communicating environmental signals is inositol
signaling pathway; activation of phosphatidylinositol phospholipase C at the cell
membrane leads to cleavage of phosphatidylinositol 4,5P2, generating secondary
messengers inositol 1,4,5-P3 and diacylglycerol, an activator of protein kinase C.
Recent studies documented the link between inositol polyphosphates, chromatin
remodeling and gene expression (14-16). It was found that IP4 (1,4,5,6) and IP5
consistently stimulated nucleosome mobilization whereas IP4 (1,3,4,5) isomer
was
inhibitory
(16).
Consequently,
the
results
on
LPS-induced
inhibition
of
mucin synthesis may be connected to the LPS and IP4 (1,3,4,5) isomer signaling
that reduces mucin gene expression (8), while the alteration in the LPS-induced
signaling that is mimicked by specific MAPK inhibitors (8,12) could reflect on
the LPS-induced changes in postgenomic and metabolomic processes in cytosol
(12,13).
While the study on protein phosphorylation and metabolic status of cytosol
allow to determine the precise site of phosphorylation and its impact on signaling,
the lipidomic approach allows to visualize the consequences of the changes on the
performance of the cell or entire organ. The previous study with gastric mucosal
545
cells showed that H. pylori LPS induced divergent signals observed in mucin
synthesis (8), and generated with the LPS and MAPK-specific inhibitors (8,12).
The evidence presented here shows that later effects may be linked to the LPSinduced postgenomic changes associated with membrane biogenesis, intracellular
transport and cell restitution processes.
Earlier, we have demonstrated that association of p85 PI3K with Golgi, is
essential for the PI3P synthesis, final modification of Golgi transport vesicle
membrane and the association with apical plasma membrane (9). Alone, the p85
PI3K can associate with apical membrane but it does not produce PI3P in the
apical
membrane.
Therefore,
the
availability
of
the
substrate,
PI3P,
for
the
recruitment of inactive Alt and PX-domain requiring cytosolic proteins is linked
to
transport
biomembrane
synthesis
in
Golgi
and
the
transport
to
apical
membrane. The inclusion of the Golgi vesicular membrane into apical membrane
consists
of
two
specific
reactions;
first
produces
specific
association
of
the
vesicles with apical membrane and depends on the recognition of the PI3K and
the second reaction consists of the fusion of the vesicular membrane with the
apical membrane. The later depends on PLA2-specific reaction in the apical
membrane, since the membrane becomes enriched in PI3P, arachidonate and
lysoPI3P (9).
The evidence gathered suggests that inhibitors of MAPK (8,12,13), depletion
of ATP or depletion of MAPK from cytosol (9) inhibit the final step in the
incorporation of the vesicle into apical membrane (fusion). Hence, decrease in
ATP-dependent phosphorylation of cytosolic proteins including PLA2 appears as
the major effect of LPS on postgenomic cytosol-driven processes. The decrease
in PI3K activity in the membrane reflects not the inhibition of the enzyme itself
but decrease in the transport of PI3K-containing vesicles to the cell membrane,
decrease
in
their
fusion
with
apical
membrane,
or
decrease
in
intracellular
membrane biogenesis. With LPS-cytosol, the membrane biogenesis in ER is
practically undisturbed, in Golgi the reduction in lipid synthesis is not greater
than 15%, but the vesicles fusion with apical membranes is reduced by 83%.
Therefore, we concluded that LPS-cytosol affected processes that require ATPdependent phosphorylation and activation of cytosol-derived protein involved in
final processing of the vesicular membrane reflected in PLA2-specific scission of
PI3P into lysoPI3P and arachidonate. Perhaps, the fact that lysophospholipids
regulate
Schwann
cells
survival
through
activation
of
protein
kinase
Act
demonstrate the same path that links active intracellular biomembrane genesis,
transport and cell membrane restitution and thus rescue cell from apoptosis (17).
Our studies imply that apical membranes draw their supply of active PI3K and
PI3P from Golgi transport vesicles and provide new insight to the findings on
proteins with PX domain implicated in protein trafficking, signaling and sorting
(18-20). With consensus that sequential enrichment in specific lipids along the
secretory
path
generates
specific
biomembrane
that
serves
as
an
anchor
to
functional protein, the puzzling findings on distribution of PX domain-containing
546
proteins are justifiable. PI3P binding motif of Vam PX domain which functions
on vacuolar membrane, Mvp1 that functions in the anterograde transport from
Golgi to the endosome (21), or Vps17, Vps27 that are implicated in endosomal
trafficking all may have common derivation (20). Membranes of Golgi vesicles
containing PI3P in outer membrane and directed to apical membrane are further
specialized
by
binding
cytosolic
proteins
with
PX
domain
(19).
In
turn,
exocytosis generates enlarged cell membrane and thus the concurrent endocytotic
process of membrane cropping is inevitable. With this metabolic turnover in
place,
the
PX
synthesized
domain-containing
Golgi
membrane,
proteins
will
be
will
present
be
in
associated
the
cell
with
newly
membrane
and
transferred to the endosomes.
Investigation of biomembrane synthesis and tracing specific lipid synthesis to
the intracellular organelle allows us to speculate with more certainty about the
involvement of PI3P in membrane specialization and turnover (18-20). The tSNARE (target-soluble NSF attachment protein receptor) Vam7 also contain a
PX domain and functions in the docking of transport vesicles (19,21). Is it then
that its interaction with v-SNARE is dependent on PI3P of transport vesicle?
Indeed,
the
determined
protein-lipid
that
PX
binding
domains
from
assays
Vam7
and
membrane
(22),
p40
phox
localization
(18)
and
study
SNX3
(23)
specifically bind PI3P, thus providing further documentation that Golgi transport
vesicles attachment to apical membrane is through its PI3P affinity. Thence, since
the apical membranes are not involved in synthesis of PI3P, this would imply that
formation of PX and plecstrin-specific sites is initiated when the Golgi vesicles
fuse with cell membrane (24). Would this represent yet another and final stage in
membrane maturation where acquisition of cytosolic proteins with PX domain
provide membrane with site-specific function and complete formation of fully
assembled biomembrane prepared to respond to the membrane-specific signals?
This is yet another question that could be answer by lipidomic approach to the
cellular processes induced by signaling.
Ultimately, since the association of p85 PI3K subunit with apical membrane
does not generate PI3P, it would appear that formation of the cell membrane
polyinositides
is
accomplished
through
PI4K
and
PI5K,
rather
than
PI3K-
specific phosphorylation of PI4P, and PI4,5P2. Indeed, with green fluorescent
protein (GFP) it was established in cell membrane PI4,5 P2 and PI3,4,5 P3
synthesis but not of the PI3P (23,24). Also, in contrast to our studies which
were performed on highly purified subcellular organelles and place synthesis of
PI3P in Golgi membrane, the studies suggesting PI3P synthesis in the cell
membrane
were
performed
with
the
mixture
of
cellular
and
subcellular
membranes (25,26). Therefore, strong possibility exists that formation of the
PI3K-specfic products assigned to cell membrane reflects contribution of the
processes taking place in Golgi and in apical membrane. The fact that external
stimulation with various growth factors does not increase PI3P level in the
membrane
(27)
provides
yet
another
support
to
the
contention
that
active
547
synthesis of the PI3P is not associated with initiation of signaling in apical
membrane,
rather
demonstrates
that
the
assembly
of
PI3,4P2
and
PI3P3
is
taking place when the signaling cycle and the entire cell membrane biogenesis
cycle is completed. Because Act is activated when PI3,4,5 P3 is generated, the
resulting Act activity is dictated by the import of the new membrane containing
PI3P, which in turn is involved in the downstream signaling involving PDK1
and PKBs (28).
Together, it appears that signals generated by LPS impact ATP-dependent
genomic and postgenomic processes (29,30); formation of secondary messengers
modulating
processes
mucin
that
gene
impair
expression
membrane
(15,16)
restitution
and
ATP-dependent
processes.
Both,
cytosolic
however,
the
genomic effect of LPS, and the postgenomic effect of LPS-induced signaling
relate to metabolic status of the cytosol are ATP-dependent (9,15).
In the presence of LPS-derived cytosol, or ATP-depleted cytosol, the PI3Kspecific association of Golgi transport vesicles with apical membranes is not
affected. But, in the presence of wortmannin, the association of the vesicles with
cell membrane is inhibited. Therefore, the action of LPS-cytosol and wortmannin
(31) is inhibitory toward different sites. Since LPS affects the processes that
follow vesicle association with apical membrane, namely the process of PLA2-
specific processes in the apical membrane, it appears that wortmannin inhibits Alt
kinase PI3,4,5 P3-containing sites, whereas LPS inhibits phosphorylation and
acquisition of PLA2.
In
situ,
the
LPS-exposed
mucosal
explants
exhibit
changes
that
are
in
agreement with the findings presented above and underline significance of the
restitution of the biomembrane and generation of the specific environment that is
provided by surface associated mucin which blocks LPS access. For instance,
previous
studies
with
LPS
promoted
EGF-
or
TNF-signaling
and
impelled
disparate signals that counteracted homeostatic cell restitution by promoting the
apoptotic elimination (32-34). Thus, we speculate that LPS-refracted homeostatic
cell renewal cycle is propagated by uncovering the cell receptors that normally in
mucin created environment, are not accessible, but may become accessible when
the cells fails to renew its apical membrane and its mucin cover. In keeping with
the concept, in situ the signal transduction pathways in gastric pit stem cells are
activated by the information provided from the cell environment and the extent of
signaling is established by cell itself. Thus, starting in gastric pit, epithelial cells
of
gastric
mucosa
contain
combination
of
the
exposed
receptors
whose
presentation to ligands changes with the migration of the cell ending along the
apical epithelial ridge. At the apical stage the cells are isolated from the numerous
ligands that contributed to changes in the stem cell they originated from, and
engage in production and secretion of mucin. Encoding and transmitting the
information by the migrating cells results in homeostatic restitution of the cell
components, activation of the processes that establish cell maturity and function,
or cause direct changes in cell behavior.
548
Consistent with our concept which in homeostasis gives emphasis to precise
reduplication of cell environment are the findings in other unrelated systems
(25,26,30). In keratinocytes, the survival and growth of the cells is associated
with adhesion to extracellular matrix which is dependent on PI3K activity (27).
With loss of PI3K activity keratinocytes differentiate into multilayered epidermis
(27). In our interpretation, loss of PI3K activity is equivalent to the inhibition
intracellular biomembrane synthesis and the capacity to restore biomembrane and
its PI3K and PI3P.
From
transport
the
experiments
vesicles
with
documenting
apical
epithelial
synthesis,
transport
membrane
under
and
control
fusion
and
of
LPS-
induced conditions, we conclude that metabolic status of cytosol impacts the
processes that affect the active restitution of cell membrane and its specific
environment. The active components of the cytosol are responsible for the
restitution of biomembranes and cell protection. It is this process that regulates
which growth factor is allowed to interact with membrane proteins (receptors)
and which factor is prevented from the interaction and signaling. In the instance
of LPS, the interaction of the bacterial product with mucin binding protein
(6,7,11)
misdirects
signaling
required
for
homeostatic
restitution
of
cell
membrane and its environment consisting of mucin. The misreading, or the
overwhelming mimicry of the LPS to mucin, causes interruption in the specific
MAPK-dependent signaling that activates PLA2 and the membrane restitution
cycle.
In our view PI3K activity in cell membrane demonstrates completion of
postgenomic
response
evoked
by
cell
external
environment.
Normally,
LPS
binding and induction of toll-like processes is not taking place when epithelial
cell is protected with mucin. The fact that LPS gains access to cell membrane
and
transmits
its
signals
indicates
that
the
apical
epithelial
surface
is
not
protected by mucin and may, in addition to LPS, be exposed to the lumenal
agents
that
signal
processes
initiating
differentiation,
proliferation
or
transformation into different cell.
The LPS-induced changes in cell restitution suggest that LPS causes inhibition
of
the
processes
that
maintain
restitution
of
the
cell
elemental
homeostatic
structure and that causes premature losses of epithelial cell. In our understanding,
the induction of specific signaling pathways is dependent on activation of the
specific receptors in the membrane, which are precisely controlled by the cell
specific environment.
Thus, the main contribution of this study is that it integrates the genomic and
postgenomic responses into biological cycle of cell restitution under normal
conditions and shows LPS-induced metabolic pathway that causes cell demise.
The above presented pathway shows the impact of lipidomic and metabolic
processes on cell renewal cycle and that together with genomic studies provides
vital information for elucidation of the full complexity of particular cell function.
The broad sense of the concept that induction of the processes is controlled by the
549
precise cell environment (hence precise signaling) is enabling us to interpret
findings as disparate as stem cells plasticity and patterns of cancer metastasis, and
adds direction to the search for the agent which effectively displace or protect the
cell from change-inducing environment (1,2,8).
REFERENCES
1.
Oltvai ZN, Barbasi AL. Life's complexity pyramid. Science 2002; 298: 763-764.
2.
Davidson EH, Rast JP, Olveri P, Ransick A, Calestani C et al. A genomic regulatory network
3.
Slomiany A, Grzelinska E, Kasinathan C, Yamaki K, Palecz D et al. Biogenesis of endoplasmic
for the development, Science 2002;295: 1669-1682.
reticulum transport vesicles transferring gastric apomucin from ER to Golgi. Exp Cell Res
1992;201: 321-329.
4.
Slomiany A, Grzelinska E, Grabska M, Yamaki K, Tamura, S. et al. Intracellular processes
associated with glycoprotein transport and processing. Arch Biochem Biophys 1992; 298: 167175.
5.
Slomiany A, Grabska M, Piotrowski E, Sengupta, S., Morita M. et al. Intracellular processes
associated
with
vesicular
transport
from
endoplasmic
reticulum
to
Golgi
and
exocytosis:
Ethanol-induced changes in membrane biogenesis. Arch Biochem Biophys 1994; 310: 247-255.
6.
Slomiany A, Grabska M, Slomiany BL. Essential components of antimicrobial gastrointestinal
epithelial barrier: Specific interaction of mucin with integral apical membrane protein of gastric
mucosa. Molecular Medicine 2001; 7: 1-10.
7.
Slomiany A, Nishikawa H, Slomiany BL Screening and modulation of extracellular signals by
mucous
barrier.
Serum
glycosylphosphatidylinositol
phospholipase
D
(GPI-PLD)
releases
protective mucous barrier from oral mucosa. J Physiol Pharm 2002; 53: 21-38.
8.
Slomiany
BL,
Slomiany
lipopolysaccharide
A.
involves
Disruption
ERK
and
in
gastric
p38
mucin
synthesis
mitogen-activated
by
protein
Helicobacter
kinase
pylori
participation.
Biochem. Biophys Res Commun 2002; 294: 220-224.
9.
Slomiany A, Nowak P, Piotrowski E, Slomiany BL. Effect of ethanol on intracellular vesicular
transport from Golgi to apical cell membrane: Role of i.c. phosphatidylinositol 3-kinase and
phospholipase
A2
in
Golgi
vesicles
association
and
fusion
with
the
apical
membrane.
Alcoholism: Clin Exp Res 1998; 22: 167-175.
10. Sato TK, Overduin M, Emr SD. Location, location, location: Membrane targeting directed by
PX domains. Science 2001; 294: 1881-1885.
11. Piotrowski J, Slomiany A, Slomiany BL. Inhibition of gastric mucosal mucin receptor by
Helicobacter pylori lipopolysaccharide. Biochem Mol Biol Int 1993; 31: 1051-1058.
12. Slomiany BL, Slomiany A. Role of ERK and p38 mitogen activated protein kinase cascades in
gastric inflammatory responses to Helicobacter pylori lipopolysaccharide. IUBMB Life 2001;
51: 315-320.
13. Watters JT, Sommer JA, Pheiffer ZA, Prabhu U, Guerra A N et al. A differential role for
mitogen-activated protein kinase in lipopolysaccharide signaling. J Biol Chem 2002; 277: 90779087.
14. Brivanlou AH, Darnell JE Jr. Signal transduction and the control of gene expresson. Science
2002;295: 813-818.
15. Shen X, Xiao H, Ranallo R, Wu W-H, Wu C. Modulation of ATP-dependent chromatinremodeling complexes by inositol polyphosphates. Science, 2003; 299: 112-114.
550
16. Steger DJ, Haswell ES, Miller AL, Wente SR, O'Shea EK. Regulation of chromatin remodeling
by inositol polyphosphates. Science 2003; 299: 114-116.
17. Hla T, Lee M-J, Ancellin N, Paik JH, Kluk MJ. Lysophospholipids-receptor revelations. Science
2001; 294: 1875-1878.
18. Bravo J, Karathanassis D, Pacold C.M, Pacold ME, Ellson CD. The crystal structure of the PX
domain from p40 (phox) bound to phosphatidylinositol 3-phosphate. Mol Cell 2001; 8: 829839.
19. Wishart MJ, Taylor GS, Dixon, JE. Phoxy lipids; revealing PX domains as phosphoinositide
binding modules. Cell 2001; 105: 817-820.
20. Sato TK, Overduin, M, Emr SD. Location, location, location: Membrane targeting directed by
PX domains. Science, 2001; 294: 1881-1885.
21. Sato TK, Darsow, T., Emr SD. Vam 7p, a SNAP-25-like molecule and Vam 3p, a syntaxin
homolog, function together in yeast vacuolar protein trafficking. Mol Cell Biol 1998; 18: 53085319.
22. Cheever M.L., Sato TK, deBeer, T, Kutateladze TG, Emr SD et al. Phox domain interaction
with PtdIns (3)P targets the Vam7 t-SNARE to vacuole membranes. Nature Cell Biol 2001; 3:
613-618.
23. Czech M.P. Dynamics of phosphoinositides in membrane retrieval and insertion. Annu Rev
Physiol 2003; 65: 791-815.
24. Varnai P, Rother KI, Balla T. Phosphatidylinositol 3-kinase-dependent membrane association
of the Bruton's tyrosine kinase pleckstrin homology domain visualized in a single living cells.
J Biol Chem 1999; 274: 10983-10989.
25. Madge LA, Pober JS. A phosphatidylinositol 3-kinase/Act pathway, activated by tumor necrosis
factor or interleukin-1, inhibits apoptosis but does not acivate NFkB in human endothelial cells.
J Biol Chem 2000; 275: 15458-15465.
26. Sayama K M, Yamasaki K, Hanakawa Y, Shirakata Y, Tokumaru S, et al. Phosphatidylinositol
3-kinase is a key regulator of early phase differentiation in keratinocytes. J Biol Chem 2002;
277: 40390-40396.
27. Kast R, Furstenberger G, Marks F. Activation of cytosolic phospholipase A2 by transforming
growth factor-
α in HEL-30 keratinocytes. J Biol Chem 1993; 268: 16795-16802.
28. Gray A., Van der Kaay J, Downes C.P. The pleckstrin homology domains of protein kinase B
and GRP1 (general receptor for phosphoinositides-1) are sensitive probes for cellular detection
of
phosphatidylinositol
3,4-bisphosphate
and
/or
phosphatidylinositol
3,4,5-triphosphate
in
vivo. Biochem J 1999; 344: 929-936.
29. Hirschberg CB, Robbins PW, Abeijan C. Transporters of nucleotide sugars, ATP, and nucleotide
sulfate in the endoplasmic reticulum and Golgi apparatus. Annu Rev Biochem 1998; 67: 49-69.
30. Shakibaei M, Schulze-tanzil G, deSouza P, John T, Rahmanzadeh M. et al. Inhibition of mitogen
-activated protein kinase induces apoptosis of human chondrocytes. J Biol Chem 2001; 276:
1389-1394.
31. Park
YC,
Lee
CH,
Kang
HS,
Chang
HT,
Kim
HD.
Wortmannin,
a
specific
inhibitor
of
phosphatidylinositol 3-kinase, enhances LPS-induced NO production from murine peritoneal
macrophages. Biochem Biophys Res Commun 1997; 240: 692-696.
32. Guha M, Mackman N. The phosphatidylinositol 3-kinase pathway limits lipopolysaccharide
activation
of
signaling
pathways
and
expression
of
inflammatory
mediators
in
human
monocytic cells. J Biol Chem 2002; 277: 32124-32132.
33. Takeda K, Ichiki T, Tokunou T, Iino N, Takeshita A. 15-deoxy-
∆
12,14
-prostaglandin J2 and
thiazolidinediones activate the MEK/ERK pathway through phosphatidylinositol 3-kinase in
vascular smooth muscle cells. J Biol Chem 2001; 276: 48950-48955.
551
34. Slomiany BL, Slomiany A. Impedance of Helicobacter pylori lipopolysaccharide interference
with
gastric
mucin
synthesis
by
Peroxisome
Proliferator-Activated
Receptor
γ
activation
involves phosphatidylinositol 3-kinase/ERK pathway. IUBMB Life 2003; 50: 97-102.
35. Kohler
H,
Rodrigues
SP,
McCormick
BA.
Shigella
flexnari
interactions
with
basolateral
membrane domain of polarized model of intestinal epithelium: Role of lipopolysaccharide in
cell invasion and in activation of the mitogen activated protein kinase ERK. Infection and
Immunity 2002; 70: 1150-1158.
Received:
July 21, 2003
Accepted:
November 18, 2003
Authors address: Amalia Slomiany, University of Medicine and Dentistry of New Jersey-NJDS
Research Center C-873, 110 Bergen Street, Newark, NJ 07103, PO Box1709
E-mail: [email protected]