| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Pharmacological and Physiological Sciences, The University of Chicago, Chicago, Illinois 60637, USA
1Correspondence: Department of Pharmacological and Physiological Sciences, The University of Chicago, 947 East 58th St., MC-0926, Chicago, Illinois 60637, USA. E-mail:rjmx{at}midway.uchicago.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Key Words: DMS • DHS • epidermal growth factor • PDGF
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
, 2)
. Cells then require growth factors to re-enter the cell cycle
(3
, 4)
. A crucial step during the course of the cell cycle
is the restriction point, located at the end of the
G1 phase. This represents the last opportunity
for a cell to decide whether to enter the S phase or not, a decision
that depends on growth factor availability (5
, 6)
. After
cells traverse the restriction point, they advance through the
G1/S boundary and enter S phase independently of
proliferative signals (7
, 8)
.
There has been great interest in trying to characterize the
physiological basis for the different susceptibility to growth factors
observed during the course of the cell cycle. Furthermore, an
increasing amount of evidence demonstrates that growth factors may
exert opposite effects on cellular fate, such as proliferation and
programmed cell death (9
10
11)
. Stimulation of
platelet-derived growth factor receptors
(PDGFRs)2
increases intracellular free Ca2+
([Ca2+]i) and activates
at least one sphingomyelinase, leading to the production of ceramide,
sphingosine (SPH), and sphingosine-1-phosphate (SPP)
(12
13
14
15)
. Sphingolipids and their metabolites represent an
emerging class of bioactive compounds that are involved in cellular
growth and death (16)
. Moreover, SPH and SPP can release
Ca2+ from intracellular stores independently from
myo-inositol-1,4,5-trisphosphate (IP3)
production (17
, 18)
. Ca2+ is a
second messenger of critical importance during cell cycle progression.
Indeed, blocking Ca2+ influx through plasma
membrane channels or mobilization from intracellular stores stops cells
from entering G1 and proceeding toward the
G1/S transition (19
20
21)
. It has
been recently shown that different patterns of
Ca2+ signals may be responsible for transmitting
distinct messages to the nucleus (22)
. This prompted us to
investigate whether the characteristics of Ca2+
signaling after PDGFR stimulation were dependent on the cell cycle.
Here we show that the kinetics of Ca2+ responses to platelet-derived growth factor BB (PDGF-BB) in CEINGE cl3 and HEK 293 cells are related to the cell cycle. We also obtained evidence for a role for sphingolipid metabolism in this phenomenon.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
. The HEK 293 cell line was from
ATCC (Rockville, Md.) (CRL1573). PDGF-BB, PDGF-AA, epidermal growth factor (EGF), and bradykinin (BK) were from Calbiochem-Novabiochem (San Diego, Calif.). Hoechst 33342 and Fura-2/acetoxy-methylester (Fura-2/AM) were from Molecular Probes (Eugene, Oreg.). DL-threo-dihydrosphingosine (DHS), N,N-dimethylsphingosine (DMS), endothelin-1 (ET-1), adenosine 5'-triphosphate (ATP), DL-sphingosine, bisindolylmaleimide I-HCl, and poly-L-lysine were from Sigma (St. Louis, Mo.). Aphidicolin and mimosine were from Biomol Research Laboratories (Plymouth Meeting, Pa.) and lovastatin was kindly provided by Alfred W. Alberts (Merck, Sharp & Dohme Research Laboratories, Rahway, N.J.).
Cell culture and synchronization
Cells were grown in Dulbecco's modified Eagle's medium (CEINGE
cl3) or minimum essential medium (HEK 293) (Life Sciences, Grand
Island, N.Y.) containing 10% fetal bovine serum (FBS, Hyclone, Logan
Utah) and 50 µg/ml gentamicin. G0 synchronized
CEINGE cl3 and HEK 293 cells were obtained by removing serum from the
culture medium for 48 and 72 h, respectively (24)
.
Cells synchronized in early G1 were obtained by
supplying cells in G0 with culture medium
containing 5% FBS and 10 µM lovastatin for 20 h. This drug
reversibly inhibits the synthesis of mevalonic acid, which is essential
for isoprenylation of p21ras, a small GTP binding
protein crucial for G0/G1
transition (25)
.
Cells were reversibly synchronized in late G1 by
supplying cells in G0 for 20 h with culture
medium containing 5% FBS and 200 µM mimosine, a drug that can block
cells at a point that precedes the G1/S boundary
of ~2 h (24)
. Synchronization at the
G1/S boundary was obtained using the reversible
inhibitor of the 
-DNA polymerase aphidicolin, added at a 5 µM
concentration to asynchronously growing cultures for 20 h
(24)
.
A population enriched in S phase cells was obtained by removing
aphidicolin from cells blocked at the G1/S
boundary and supplying them with fresh culture medium containing 5%
FBS (24)
. The percentage of cells entering the S phase
after G1/S release was monitored by
immunocytochemistry using an antibody against 5-bromo-2'-deoxyuridine
(BrdU), a thymidine analog that is incorporated into the nucleus of
cells actively synthesizing DNA. Using this approach, the entry into S
phase was observed as early as 30 min after aphidicolin removal, with
36 ± 3% and 48 ± 6% of BrdU-positive cells stained at the
first and second hour, respectively. The last time point has been used
for all experiments relative to the Ca2+ response
to PDGF-BB during the S phase.
A population enriched in cells at the G2/M transition was obtained by monitoring the expression of cyclin B1 after the release from late G1 block and during the progression through the S phase.
Serum withdrawal as well as pharmacological cell cycle synchronization might cause variable degrees of apoptotic death in different cell types. Therefore, the percentage of apoptotic cells induced by different synchronizing treatments was assessed by performing a cell viability test using Hoechst 33342 on a representative coverslip before each series of experiments and proved to be negligible for both cell lines used in this study.
Western blot analysis
Nuclear and cytoplasmic extracts for cyclin expression and total
cell lysates for PDGFR expression were obtained as described (26
, 27)
. The protein concentration in the clarified lysate was
determined by BCA protein assay (Pierce, Rockford, Ill.). Cell lysates
containing equal amount of proteins were resolved on a sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (28)
using 10%
and 7.5% polyacrylamide gel for cyclins and PDGFRs, respectively. The
separated proteins were electroblotted to Hybond-ECL nitrocellulose
(Amersham, Arlington Heights, Ill.) or immobilon-P PVDF membrane
(Millipore Corporation, Bedford, Mass.). Membranes were then incubated
for 1 h in phosphate-buffered saline containing 4% dry milk
powder, 1% bovine serum albumin and 0.1% Tween-20. The antibodies for
the immunoblotting were used at the following dilutions: anti-cyclin A
(Upstate, Lake Placid, N.Y.) at 1:1000; anti-cyclin E, anti-cyclin B1,
and anti-cyclin D1 (Santa-Cruz Biotechnology, Santa Cruz, Calif.) at
1:2000, 1:1000, and 1:100, respectively. Anti-ß-PDGFR (Santa Cruz
Biotechnology) and anti--PDGFR (Upstate) rabbit polyclonal
antibodies were used at 1:1000 and 1:4000 dilution, respectively.
Anti-rabbit and anti-mouse HRP-conjugated secondary antibodies
(Promega, Madison, Wis.) were used at 1:20.000 dilution. The binding of
the antibodies to the membrane was detected using the SuperSignal Ultra
(Pierce) or ECL (Amersham) chemiluminescent substrates. Densitometric
analysis was performed using the UN-SCAN-IT software (Silk Scientific,
Inc., Utah).
Immunocytochemistry and immunofluorescence
For the immunocytochemistry of BrdU incorporation, cells were
incubated for 30 min with 10 µM BrdU added to the culture medium and
fixed with 70% ethanol/50 mM glycine buffer for 30 min at -20°C.
Primary antibody directed against BrdU, secondary
alkaline-phosphatase-conjugated antibody, and color substrate solutions
were from BrdU labeling and detection Kit II (Boehringer Mannheim,
Indianapolis, Ind.), and used according to the manufacturer's
instructions.
For the immunofluorescence of cyclin B1, cells were fixed and permeabilized in cold methanol for 20 min, blocked in 5% normal donkey serum, and incubated with anti-cyclin B1 (Santa Cruz) 1:1000 in blocker for 1 h. The biotin-conjugated secondary antibody (1:500 for 1 h) was followed by streptavidin-Cy3 at 1:3000 for 15 min. Coverslips were mounted with glycerol/phosphate-buffered saline (9:1) containing 1,4 diazabicyclo[2.2.2]octane (2.5% W/V) as anti-fading agent.
Intracellullar Ca2+ measurement and videoimaging
analysis
Cells were grown on 25 mm N° 1 glass coverslips coated with
poly-L-lysine (100 µg/ml). Cell loading with 2 µM Fura-2/AM,
single-cell videoimaging of as many as 20–30 cells/field, and
calibration of fluorescent signals were performed as described
previously (27)
.
Thin-layer chromatography (TLC) analysis of lipids
Labeling with [3H]serine (20 µCi/ml,
Amersham), a precursor of cellular sphingolipids, and harvesting of
cells exposed to 10 ng/ml PDGF-BB for various time were performed as
reported (29)
. Lipid extraction was obtained using
chloroform/methanol/concentrated HCl (100:200:1 v/v) and the phases
were separated as described (30
, 31)
. The lipids, dried
under nitrogen, were analyzed by 2-dimensional thin-layer
chromatography performed on silica gel 60 G plates (Alltech, Deerfield,
Ill.). The plates were developed using chloroform/methanol/ammonium
hydroxide (13:7:1) as solvent for the first dimension and
chloroform/methanol/water/acetic acid (30:30:2:5) as solvent for the
second dimension (31)
. Lipids were located using iodine
vapors and phospholipids were visualized using molybdenum blue spray
(Sigma). SPH (Avanti Polar Lipids, Alabaster, Ala.) and SPP (Biomol)
were used as standards to identify the silica gel areas to scrape off
and count by liquid scintillation spectometry.
Statistical analysis
Data values are expressed as mean value ±SE.
Student's t test or the Mann-Whitney nonparametric test
were used and statistical significance was defined as a P
value of 0.01 or less.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
). These two types of Ca2+ responses were evenly
distributed in asynchronously proliferating cell cultures, but never
observed in the same cell. We therefore decided to investigate whether
the type of Ca2+ signal observed in response to
PDGF-BB was related to the position of the cells in the cell cycle.
|
A first series of experiments was performed using CEINGE cl3 cells
synchronized in G0 and early
G1 phase. Cell synchronization was assessed by
monitoring the nuclear levels of cyclin D1, whose expression is
down-regulated in resting cells. Furthermore, the percentage of cells
incorporating BrdU was evaluated (Fig. 2
A). When G0 arrested CEINGE cl3 cells were exposed to
PDGF-BB, the majority of cells showed non-oscillatory
Ca2+ responses (92 ±3%, 107 cells analyzed
in eight experiments). In contrast, 74 ± 7% of cells blocked in
early G1 responded to PDGF-BB in an oscillatory
fashion (84 cells analyzed in six experiments) (Fig. 2B
).
|
Activation of G-protein-coupled receptors (GPCRs) for BK,
ET-1, and ATP in asynchronous CEINGE cl3 cells exclusively induced a
non-oscillatory Ca2+ response, as we previously
reported (32)
. Cell synchronization in early
G1 did not modify the pattern of
Ca2+ signals induced by these agonists (data not
shown).
Drugs commonly used as synchronizing agents may induce a certain degree
of metabolic perturbation in cells (33
, 34)
. Consequently,
it seemed important to verify that the changes in the types of
Ca2+ signaling we observed were really connected
to the cell cycle and not attributable to the methods used to
synchronize the cells per se. We therefore analyzed the
changes in Ca2+ signaling in response to PDGF-BB
at fixed time intervals during exit from the cycle as well as during
re-entry and progression through G1 phase in the
absence of synchronizing drugs. As shown in Fig. 2C
, serum
withdrawal produced a progressive disappearance of the oscillatory
Ca2+ signaling in response to PDGF-BB. Although
after 12 h of serum removal the percentage of oscillatory cells
was still comparable to that observed in asynchronous cultures, after
24 and 48 h Ca2+ oscillations were observed
in only 31 ± 6% and 8 ± 3% of cells responsive to
PDGF-BB, respectively. The oscillatory Ca2+
signaling in response to PDGF-BB gradually reappeared when
G0 arrested cells were induced to re-enter the
cycle by serum addition, as shown by the re-expression of cyclin D1
(see insert in Fig. 2C
). For instance, 2, 4, and 6 h
after serum addition, 17 ± 8%, 29 ± 11%, and 45 ± 7
of responding cells exhibited an oscillatory Ca2+
response (20, 43, and 58 cells analyzed in three experiments for each
set of data). Finally, after 12 h, the percentage of oscillatory
Ca2+ responses to PDGF-BB reached 57 ± 7%
(23 cells analyzed in three experiments). The reappearance of
oscillatory Ca2+ signaling in response to PDGF-BB
during the transition from G0 to
G1 agreed with the results obtained in cells
pharmacologically synchronized in early G1 (Fig. 2B
).
A second series of experiments was performed with CEINGE cl3 cells
pharmacologically synchronized at the G1/S border
as well as on cultures enriched in cells in S phase. Cell
synchronization was evaluated by measuring the nuclear expression of
cyclin A and cyclin E, whose protein levels peak at the
G1/S border and are thought to be involved in
controlling DNA replication (35)
(Fig. 3
A). Cultures maximally enriched in cells in S phase
(48 ±6% of cells incorporating BrdU) were obtained at the second
hour after the release from the pharmacological block at the
G1/S border. Cells blocked at the
G1/S border responded to PDGF-BB mostly in a
non-oscillatory fashion, with a percentage of oscillatory
Ca2+ responses of only 12 ± 6% (53 cells
analyzed in five experiments). However, their entry into the
S phase was followed by a significant increase in the percentage of
oscillatory Ca2+ responses (48 ±10%;43
cells analyzed in three experiments) (Fig. 3B
).
|
As previously done for
G0/G1 transition, we
decided to analyze the Ca2+ responses to PDGF-BB
in CEINGE cl3 cells at fixed time intervals after the release from late
G1 block and during the progression through the
G1/S border and S phase. When cells blocked in
late G1 phase were exposed to PDGF-BB, the
Ca2+ signaling observed was mostly oscillatory
and comparable to that seen in early G1 cells
(69 ±18%, 19 cells analyzed in three experiments) (Fig. 3C
). Cells were released from late G1
block by the addition of fresh serum-containing medium and entered the
S phase at the third hour, as assessed by BrdU staining (data not
shown). At the first and second hour postrelease, probably
corresponding to the G1/S transition, a
significant reduction in the percentage of oscillatory
Ca2+ responses was observed (20 ±16% and
19 ±10%, respectively; 24 and 21 cells analyzed in three
experiments for each set of data) (Fig. 3C
). This was in
agreement with the results obtained in cells pharmacologically
synchronized at the G1/S boundary (Fig. 3B
). At the third and fourth hour postrelease (a time
interval corresponding to the S phase), the percentage of oscillatory
Ca2+ responses increased (49 ±10% and
65 ±22%, respectively; 22 and 15 cells analyzed in three
experiments for each set of data). This was in accord with the results
obtained in cells released from the pharmacological arrest in
G1/S (Fig. 3B
).
A third group of experiments was performed using cultures enriched in
CEINGE cl3 cells arrested at the G2/M transition.
This was achieved by monitoring the cytoplasmic and nuclear expression
of cyclin B1 in cells released from the block in late
G1 phase induced by mimosine. Cyclin B1 is first
synthesized during S phase, maximally expressed at the
G2/M transition, and then degraded during
anaphase (36)
. Western blot and densitometric analysis
revealed that cyclin B1 levels gradually increased starting from the
second hour after the release from late G1 block.
The highest cytoplasmic levels were reached at the fourth hour whereas
the nuclear levels reached a peak at the sixth and then declined at the
tenth hour (Fig. 4
A, B). The nuclear translocation of cyclin B1 was also
assessed by immunofluorescence performed using an anti-cyclin B1
specific antibody on cells arrested at late G1
phase and at the sixth hour after the release from the block (Fig. 4C, D
). In the latter cell population the percentage of
oscillatory Ca2+ responses to PDGF-BB was similar
to that observed in cells traversing the S phase (54 ±6%; 21
responsive cells out of 84 total cells analyzed in three separate
experiments). These data therefore indicate that the cell cycle does
not exert any modulatory activity on Ca2+
signaling induced by PDGF-BB in cells at the G2/M
transition.
|
It seemed important to ascertain whether such a cell cycle-dependent
modulation of Ca2+ responses to PDGF-BB was
exclusive to CEINGE cl3 cells or whether it could also be observed in
other cell types. We therefore decided to apply the same experimental
paradigm used for CEINGE cl3 cells to HEK 293 cells. In these cells,
PDGF-BB induces membrane ruffling and activation of the PTP2C protein
phosphatase (37)
, but Ca2+ signaling
after PDGFR activation has never previously been characterized. When
asynchronously growing HEK 293 cells were exposed to PDGF-BB,
exclusively oscillatory Ca2+ signals were
observed (79 cells analyzed in three experiments; Fig. 5
A). However, synchronization in G0
phase of HEK 293 caused the appearance of non-oscillatory
Ca2+ responses to PDGF-BB (Fig. 5B
).
As observed in CEINGE cl3 cells, the decrease in oscillatory
Ca2+ responses was time dependent but required a
prolonged period of serum withdrawal. For instance, although 48 h
after serum removal no changes in the percentage of oscillatory
Ca2+ responses were observed, after
60 h oscillatory Ca2+ responses to the
growth factor were 51 ± 2%. Finally, after 4 days of starvation,
only 26 ± 7% of HEK 293 cells responded in an oscillatory
fashion (Fig. 5C
; 60, 43, and 53 cells analyzed in three
separate experiments for each data point). Synchronization in
G0 of HEK 293 cells was confirmed by the
progressive decrease in the percentage of cells stained for BrdU and by
the reduction of cyclin D1 expression (data not shown). As shown for
CEINGE cl3 cells, quiescent HEK 293 cells could be induced to re-enter
the cell cycle by serum readdition. Indeed, this caused a rapid
restoration of the oscillatory Ca2+ responses
(Fig. 5C
). Synchronization of HEK 293 cells at the
G1/S border was also able to induce the
appearance of non-oscillatory Ca2+ responses in
20 ± 8% of cells (56 cells analyzed in three experiments).
|
To understand whether the control of Ca2+
signaling operated by the cell cycle was limited to PDGFRs, we decided
to test a different growth factor such as EGF (50 ng/ml). CEINGE cl3
cells are unresponsive to EGF, whereas in HEK 293 cells this growth
factor induces stimulation of adenylyl cyclase (38)
,
mitogen-activated tyrosine kinases, and trans-activation of
Elk-1 (39)
.
When asynchronously growing HEK 293 cells were exposed to EGF, both
oscillatory and non-oscillatory Ca2+ responses
were observed (61 ±19% of oscillatory responses; 75 cells
analyzed in three experiments) (Fig. 5D
). However, cell
synchronization in G0 was unable to modify the
relative percentage of occurrence of these two types of
Ca2+ response (68 ±10% of oscillatory
responses observed in 31 cells analyzed in two experiments).
These results demonstrate that in the two cell lines we tested, there is a correlation between the position of a cell in the cell cycle and the type of Ca2+ signaling produced by PDGF-BB. They also indicate that Ca2+ responses induced by GPCRs and by the EGF receptor are not influenced by the same mechanism.
Recent studies have shown that in BALB/c-3T3 fibroblasts, expression of
the ß isoform of PDGFR is up-regulated when cells undergo growth
arrest on serum deprivation (40)
, whereas terminal
differentiation of 3T3-L1 fibroblasts induces a down-regulation of both
- and ß-PDGFRs (41)
. In the light of these results,
it has been proposed that changes in PDGFR levels may provide a
mechanism for regulating cell responsiveness to this growth factor
during the cell cycle (41)
. Since CEINGE cl3 cells express
both - and ß-PDGFRs (27)
, we investigated whether
cell cycle-dependent changes in the expression of these two receptors
could account for the differences in the Ca2+
response to PDGF-BB we observed. Western blot analysis performed using
cells synchronized in G0,
G1, and at G1/S boundary
showed that the protein levels of both - and ß-PDGFR did not
change significantly in all phases of the cell cycle examined
(Fig. 6
A). This idea was also confirmed by the fact that similar
patterns of and ß receptor expression were observed in cells
responding with opposite kinetics to PDGF-BB, such as
G1 and G1/S synchronized
cells (Fig. 6B
).
|
HEK 293 cells lack ß-PDFGR (39)
; therefore, the
Ca2+ signaling induced by PDGF-BB in these cells
must be due to stimulation of -PDGFR, which binds the growth factor
with an affinity equivalent to that of the ß subunit
(12)
. The expression levels of -PDGFR in asynchronously
growing and G0 synchronized HEK 293 cells did not
show significant differences (data not shown), suggesting that
-PDGFR is subjected to the same modulation by the cell cycle as the
ß isoform.
These results therefore indicated that the target(s) involved in the cell cycle control of PDGF Ca2+ signaling was located downstream of the plasma membrane receptor.
It is known that PDGF induces an increase in cellular levels of SPH and
activation of sphingosine kinase, the enzyme responsible for the
conversion of SPH in SPP (29)
. These two sphingolipids,
among other cellular effects, mobilize Ca2+ from
thapsigargin-sensitive intracellular pools without interacting with the
IP3 binding site of the IP3
receptor (42
, 43)
.
We have previously shown that in asynchronously growing CEINGE cl3
cells, exposure to PDGF-BB in the presence of DHS, a potent inhibitor
of sphingosine kinase (44)
, produced an increase in the
percentage of oscillatory Ca2+ responses.
Furthermore, the addition of 10 µM exogenous SPH, which can be taken
up by intact cells (31)
, induced a
Ca2+ response with similar oscillatory kinetics
and delay to that induced by PDGF-BB itself (32)
. This led
to the hypothesis that SPH and SPP were responsible for the oscillatory
and non-oscillatory Ca2+ response, respectively.
We therefore considered the possibility that sphingosine kinase activity, and thus the intracellular levels of SPH and SPP, were regulated in a cell cycle specific fashion, thereby determining the type of Ca2+ signaling induced by PDGF in a particular phase of the cell cycle.
The ability of DHS to block the sphingosine kinase was investigated in
CEINGE cl3 cells by TLC analysis. As shown in Fig. 7
A,
though DHS was unable to affect the production of SPH or SPP per
se under basal conditions, it significantly increased the levels
of SPH and reduced the levels of SPP when these cells were stimulated
with PDGF-BB.
|
Preincubation with 10 µM DHS induced a significant increase in
oscillatory Ca2+ responses to PDGF-BB both in
G0 (50 ±7%; 75 cells analyzed in four
experiments) and G1/S synchronized CEINGE cl3
cells (47 ±11%; 64 cells analyzed in five experiments) (Fig. 7B
).
In parallel experiments, we replaced PDGF-BB with exogenous SPH
in the presence of DHS. Under these conditions, oscillatory
Ca2+ responses were observed in 59 ± 5% of
G0 and 89 ± 7% of
G1/S synchronized CEINGE cl3 cells, respectively
(17 and 31 cells analyzed in two separate experiments for each set of
data) (Fig. 7B
). DHS alone was without effect on
intracellular Ca2+, as described previously
(32)
.
Similarly, 78 ± 7% of G0 arrested HEK 293 cells responded to PDGF-BB in an oscillatory manner when the production of SPP was inhibited compared with 26 ± 10% of cells not treated with DHS (41 cells analyzed in two experiments, data not shown).
Thus, both in G0 and G1/S synchronized cells, the ability to respond with Ca2+ oscillations to PDGF-BB could be restored when SPH levels were increased.
A putative inhibitory activity of sphingosine kinase antagonists on
protein kinase C (PKC) has been reported (45)
. However,
incubation of G0 synchronized CEINGE cl3 cells
for 30 min with the specific inhibitor of PKC bisindolylmaleimide I-HCl
(200 nM) (46)
was unable to reproduce the increase in the
percentage of oscillatory Ca2+ responses to
PDGF-BB observed when cells were treated with DHS (18 cells studied in
3 separate experiments). An effect similar to that caused by DHS was
also produced by 10 µM DMS, another inhibitor of sphingosine kinase,
which at this concentration is ineffective on PKC (47)
.
These results suggest that the modulatory role of SPH and SPP on
PDGF-BB-induced Ca2+ signaling is independent to
PKC inhibition.
A clear correlation between the cell cycle and production of SPH and
SPP induced by PDGF-BB was established by measuring the levels of these
two sphingolipids in G0 and early
G1 synchronized CEINGE cl3 cells, which (as shown
above) responded with opposite Ca2+ signals to
the growth factor stimulation (Fig. 2C
).
Indeed, in G0 synchronized cells, exposure to
PDGF-BB produced higher levels of SPP and lower levels of SPH compared
with cells synchronized in early G1 (Fig. 7C
). Thus, opposite SPH/SPP ratios were observed in
G0 and in early G1
synchronized cells (Fig. 7D
), suggesting that the relative
rather than absolute levels of the two sphingolipids appear to be
critical in the modulation of Ca2+ signaling
induced by
PDGF-BB.
The onset of changes in SPH/SPP ratios was observed at 3 min in
G0 synchronized and at 6 min in early
G1 synchronized cells after PDGF-BB exposure.
This correlates with the time delays preceding the appearance of
non-oscillatory and oscillatory Ca2+ responses
(170 ±7 and 285 ±20 s, respectively in 188 cells
examined; see also ref 32
).
In conclusion, this study provides the first evidence for a cell
cycle-dependent control of the Ca2+ signaling and
the production of SPH and SPP induced by PDGF-BB. It also strongly
suggests that these two sphingolipids determine the type of
Ca2+ signals transduced in response to this
growth factor. The intracellular target(s) for the
Ca2+ mobilizing action of SPH and SPP has not
been identified yet. Modulation or binding to distinct subtypes of
IP3 receptors that have different
Ca2+ releasing properties (48)
or
interaction with additional signaling pathways could explain the
differences in the kinetics of the Ca2+ responses
induced by these two sphingolipids.
It has been shown that PDGF-BB activates sphingomyelinase, ceramidase,
and sphingosine kinase, whereas EGF has no effect on sphingolipid
metabolism (15
, 49)
. GPCRs have been shown to activate
sphingosine kinase. However, this appears to be a receptor-specific
phenomenon, also dependent on the cell type (50
, 51)
.
Indeed, the lack of activation of the sphingolipid pathway in CEINGE
cl3 and HEK 293 cells by EGF receptors and the GPCRs tested could
explain why Ca2+ responses induced by these
receptors were not modulated by the cell cycle.
It is interesting that in the cells we tested, it seems that distinct
Ca2+ signals are required in specific phases of
the cell cycle. It is widely recognized that the activity of several
intracellular targets can be modulated by both the frequency and
amplitude of Ca2+ signals (52
53
54)
and that sphingolipids are involved in cell proliferation
(29)
and programmed cell death (55)
.
Therefore, differences in PDGF-induced signaling during cell cycle
progression could participate in regulating the occurrence and/or
progression of these events.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication June 23, 1998.
Revision received March 15, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
S transition of mammalian cells. J. Cell. Biochem. 54,379-386[Medline]
This article has been cited by other articles:
|
|
 |
|
  S. A. Shulby, N. G. Dolloff, M. E. Stearns, O. Meucci, and A. Fatatis CX3CR1-Fractalkine Expression Regulates Cellular Mechanisms Involved in Adhesion, Migration, and Survival of Human Prostate Cancer Cells Cancer Res., July 15, 2004; 64(14): 4693 - 4698. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. D. Batheja, D. J. Uhlinger, J. M. Carton, G. Ho, and M. R. D'Andrea Characterization of Serine Palmitoyltransferase in Normal Human Tissues J. Histochem. Cytochem., May 1, 2003; 51(5): 687 - 696. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Wei, V. G. Henke, C. Strubing, E. B. Brown, and D. E. Clapham Real-Time Imaging of Nuclear Permeation by EGFP in Single Intact Cells Biophys. J., February 1, 2003; 84(2): 1317 - 1327. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Soliven, L. Ma, H. Bae, B. Attali, A. Sobko, and T. Iwase PDGF upregulates delayed rectifier via Src family kinases and sphingosine kinase in oligodendroglial progenitors Am J Physiol Cell Physiol, January 1, 2003; 284(1): C85 - C93. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
