JOURNAL OF CELLULAR PHYSIOLOGY 176:412–423 (1998)

Different Signaling Pathway Between

Sphingosine-1-Phosphate and
Lysophosphatidic Acid in Xenopus Oocytes:
Functional Coupling of the Sphingosine-1-
Phosphate Receptor to PLC-xb in
Xenopus Oocytes
SEUNG-JAE NOH, MYUNG-JUN KIM, SANGWOO SHIM, AND JIN-KWAN HAN*
Department of Life Science, Pohang University of Science and Technology,
Pohang, South Korea

In Xenopus oocytes, both sphingosine-1-phosphate (S1P) and lysophosphatidic

acid (LPA) activate Ca2/-dependent oscillatory Cl0 currents by acting through
membrane-bound receptors. External application of 50 mM S1P elicited a long-
lasting oscillatory current that continued over 30 min from the beginning of
oscillation, with 300 nA (n Å 11) as a usual maximum peak of current, whereas
1-mM LPA treatment showed only transiently oscillating but more vigorous current
responses, with 2,800 nA (n Å 18) as a maximum peak amplitude. Both phospho-
lipid-induced Ca2/-dependent Cl0 currents were observed in the absence of extra-
cellular Ca2/, were blocked by intracellular injection of the Ca2/ chelator, EGTA,
and could not be elicited by treatment with thapsigargin, an inhibitor of endoplas-
mic reticulum (ER) Ca2/ ATPase. Intracellular Ca2/ release appeared to be from
inositol 1,4,5-trisphosphate (IP3)-sensitive Ca2/ store, because Cl0 currents were
blocked by heparin injection. Pretreatment with the aminosteroid, U-73122, an
inhibitor of G protein-mediated phospholipase C (PLC) activation, to oocytes
inhibited the current responses evoked both by S1P and LPA. However, when
they were injected with 10 ng of antisense oligonucleotide (AS-ODN) against
Xenopus phospholipase C (PLC-xb), oocytes could not respond to S1P applica-
tion, whereas they responded normally to LPA, indicating that the S1P signal-
ing pathway goes through PLC-xb, whereas LPA signaling goes through another
unknown PLC. To determine the types of G proteins involved, we introduced
AS-ODNs against four types of G-protein a subunits that were identified in Xeno-
pus laevis; Gqa, G11a, G0a, and Gi1a. Among AS-ODNs against the Gas tested,
AS-Gqa and AS-Gi1a to S1P and AS-Gqa and AS-G11a to LPA specifically reduced
current responses, respectively, to about 20–30% of controls. These results dem-
onstrate that LPA and S1P, although they have similar structural features, release
intracellular Ca2/ from the IP3-sensitive pool, use different components in their
signal transduction pathways in Xenopus oocytes. J. Cell. Physiol. 176:412–
423, 1998. q 1998 Wiley-Liss, Inc.

Sphingosine-1-phosphate (S1P), the simplest form of has the potential role of a signaling molecule acting
sphingophospholipids, has emerged recently as a candi- through a membrane-bound receptor. It has been re-
date molecule of signal transduction involved in vari- ported that S1P, when it is added extracellularly to Xeno-
ous biological functions (Spiegel and Milstein, 1995). pus oocytes, leads to a transient, inward current that is
Endogenous S1P has been implicated in such diverse
activities as a putative second messenger, mobilizing Contract grant sponsor: Korea Science and Engineering Founda-
Ca2/ from internal stores via an Inositol 1,4,5-trisphos- tion; Contract grant number: 971-0504-021-2; Contract grant
phate (IP3)-independent manner (Ghosh et al., 1990; sponsor: Ministry of Science and Technology (The MOST Genome
Mattie et al., 1994), and modulating growth factor-reg- Project).
ulated microfilament reorganization and cell migration *Correspondence to: Jin-Kwan Han, Department of Life Sci-
(Bronfeldt et al., 1995), and it is involved in cellular ence, Pohang University of Science and Technology, San 31
proliferation induced by mitogens (Zhang et al., 1991; Hyoja Dong, Pohang 790-784, South Korea. E-mail: jkh@vision.
Oliveira and Spiegel, 1993). postech.ac.kr
Several lines of evidence have demonstrated that S1P Received 2 September 1997; Accepted 10 February 1998
q 1998 WILEY-LISS, INC.

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

 S1P AND LPA SIGNALING IN XENOPUS OOCYTES 413

carried mainly by Cl0 ion through the endogenous Ca2/- Laboratories, Inc. (Plymouth Meeting, PA). LPA (1-
dependent Cl0 channel activity (Durieux et al., 1993). acyl-sn-glycerol-3-phosphate), niflumic acid, heparin
This responsiveness to S1P in Xenopus oocytes has been (MW 6,000), de-N-sulfated heparin, thapsigargin, and
considered to be conducted through a putative mem- ruthenium red (RuRed) were from Sigma (St. Louis,
brane-bound receptor, presumably through lysophospha- MO). S1P was dissolved by sonication in aqueous solu-
tidic acid (LPA) receptor, inferred from the similar pat- tion containing 1% fatty acid-free bovine serum albu-
tern of response between S1P and LPA. In addition, van min (BSA; Boehringer-Mannheim, Mannheim, Ger-
Koppen et al. (1996) demonstrated that S1P potently and many) to a concentration of 2.5 mM. LPA was also dis-
rapidly increased intracellular Ca2/ ([Ca2/]i) in many dif- solved in aqueous 1% fatty acid-free BSA to a stock
ferent types of cells, especially in human embryonic kid- concentration of 1 mM. The stock solutions of phospho-
ney (HEK) cells, in a pertussis toxin (PTX)-sensitive lipids were kept at 0207C in a small aliquot. U-73122
manner, implying the existence of an S1P-responsive, and its nonfunctional analog, U-73343, were aliquoted
membrane-bound receptor. Recently, a study has re- and saved at room temperature in dried form and were
vealed that S1P-induced Rho-dependent neurite retrac- dissolved in dimethyl sulfoxide (DMSO) with stock con-
tion in NIE-115 neuronal cells was through a distinct centration of 5 mM just before use. Niflumic acid was
G protein-coupled cell surface receptor (Postma et al., solubilized with 95% ethanol. Heparin, de-N-sulfated
1996). heparin, and RuRed were dissolved in deionized water.
Another simple phospholipid, LPA (1-acyl-sn-glyc- Thapsigargin was solubilized in DMSO as a 1-mM
erol-3-phosphate), acts as a signaling molecule. LPA, a stock solution, stored in 0207C as small aliquots. The
platelet-derived mediator that acts through its cognate bath concentrations of 1 mM of LPA and 50 mM of S1P
G protein-coupled receptor, plays a significant role in were used throughout the experiment, because these
platelet aggregation (Schumacher et al., 1979) and in- concentrations elicited maximal current response.
duces proliferation of rat-1 cells and human foreskin
fibroblasts (van Corven et al., 1989). It has been specu- Oocyte preparation
lated that S1P and LPA might signal through a single Mature wild type female Xenopus laevis purchased
receptor (Durieux et al., 1993). Recently, however, a from Xenopus I (Ann Arbor, MI) were maintained at
putative high-affinity LPA receptor was cloned in Xeno- 18 – 247C. Individual oocytes at stage VI (Dumont,
pus laevis oocytes (Guo et al., 1996). The cloned LPA 1972) were dissected manually from their outer follicles
receptor with common features of the seven transmem- by using watchmaker’s forceps and were stored in mod-
brane receptors responds only to LPA with high affinity ified Barth saline [MBS: 88 mM NaCl, 1 mM KCl, 2.4
and is not responsive to S1P or to other phospholipid mM NaHCO3 , 0.33 mM Ca(NO3)2 , 0.41 mM CaCl2 , 10
mediators. Given these facts, it seems clear that S1P mM Na-HEPES (Sigma), pH 7.4] containing 5 mg/ml
and LPA each has its own receptor, although the puta- penicillin, 5 mg/ml streptomycin, and 2.5 mM pyruvate
tive S1P receptor has not been identified yet in Xenopus at 19 { 17C and were used in 1 – 4 days after isolation.
oocytes.
Despite the fact that both S1P and LPA induced in- Voltage-clamp recording
tracellular Ca2/ release in Xenopus oocytes, the compo- Currents were recorded by using the standard dou-
nents of the signal-transduction pathway involved in ble-electrode, voltage-clamp technique, held at 070
the response of these phospholipids have not been de- mV. A single oocyte was placed in recording chamber
termined. In this study, we have examined the compo- filled with 1.5 ml MBS. Microelectrodes were pulled in
nents of the signal pathway involved in S1P response one stage from capillary glass (Borosilicate glass capil-
in Xenopus oocytes and compared them with those of laries with inner filament; catalog no. GT12; Warner
the LPA signal pathway. We have hypothesized that Corp., New Haven, CT) on a micropipette puller (model
intracellular Ca2/ release, especially from the IP3-sen- 700C; David Kopf Instruments, Tujunga, CA), and the
sitive pool, might be the main event of these two phos- tips were broken to a diameter of Ç10 mm. They were
pholipid responses and the corresponding IP3 , presum- filled with 3 M KCl, and tip resistances were usually
ably generated by phospholipase C (PLC). Because re- 1 – 5 MV. The cell was voltage clamped by using a two-
ceptor activation linked to IP3 production usually acts microelectrode voltage clamp amplifier (oocyte clamp
through G protein and, subsequently, to its coupled OC725A; Warner Corp.) connected to a data acquisition
PLC partner, we reasoned that the putative S1P recep- system (MacLab/4e; AD Instruments Pty Ltd., Castle
tor also coupled to distinct G proteins, and so did LPA. Hill, Australia) running on a Power Macintosh com-
In this study, we show that a difference exists in the puter. Membrane currents were sampled at 4 Hz. Com-
initial part of the signal-transduction pathway between pounds were delivered in 15-ml aliquots over 1 – 2 sec
LPA and S1P. The use of antisense oligonucleotides by using a hand-held micropipette positioned Ç5 mm
(AS-ODNs) demonstrates that S1P-induced signaling from the oocyte. Treatment of vehicle, 1% BSA (fatty
utilizes G-protein subunit Gq/Gi1 , whereas LPA-in- acid-free) solution in water or DMSO, could not induce
duced signal pathway is conducted through Gq/G11 , and any current responses. To minimize the variations
PLC-xb is involved in S1P signaling but not in LPA among oocytes from different females, the comparisons
signaling. of the current responses were made in oocytes from the
same animal. All experiments were performed at room
MATERIALS AND METHODS temperature.
Materials
S1P, aminosteroid (1-[6-[17b-3-methoxyestra-1,3,5(10)- Injection of oocytes
trien-17-yl]amino]hexyl]-1-H-pyrole-2,5-dione (U-73122), Microinjection was performed by using a Nanoliter
and U-73343 were purchased from BIOMOL Research Injector (WPI, Sarasota, FL). The final concentration

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

414 NOH ET AL.

of the injection solution in the oocyte was calculated

based on the assumption that the actual accessible cy-
toplasmic volume of the 1.2-mm oocyte is 450 nl, be-
cause half of the oocyte volume is estimated to be occu-
pied by membrane-bound yolk platelets. Ten to thirty
nanograms of AS-ODNs (summarized in Table 1), mod-
ified its internucleoside linkages at both the 5* and the
3* ends with phosphorothioate bond to provide protec-
tion against exonucleases (Dagle et al., 1990), were in-
jected into individual oocytes and cultured for 2 days
in MBS containing 5 mg/ml penicillin, 5 mg/ml strepto-
mycin, 400 mg/ml BSA, and 2.5 mM pyruvate at 19 {
17C. Poor impalements caused leakage of the cyto-
plasm, and these oocytes were discarded.
Quantitative reverse transcriptase-polymerase
chain reaction assay
Oocytes were injected with oligonucleotides and cul-
tured for 4 – 24 hr. Total RNA was prepared from oo-
cytes by using the acid guanidinium thiocyanate-phe-
nol-chloroform extraction method (Chomczynski and
Sacchi, 1987) and treated with RNase-free DNase. Five
micrograms of total RNA were reverse transcribed with
MMLV reverse transcriptase (RT) and random hex-
amer as a primer. For quantitative analysis, polymer-
ase chain reaction (PCR) was performed as described
(Wilson and Melton, 1994), except that DNA amplifica-
tion was carried out nonradioactively. PCR products
were analyzed on 2% agarose gels with ethidium bro-
mide, and fluorescent DNA fragments were digitized
by using a fluorescent image analyzer (FLA-2000; Fuji
Film, Tokyo, Japan). PCR amplifications were in the
linear range and yielded results indistinguishable from
radioactive PCR. Elongation factor 1-a (226 bp) was
used as internal control as in Kengaku and Okamoto
(1995). The other gene-specific primers used were Gqa
(Shapira et al., 1994; upstream: TATCTCAACGAC-
GTGGAC; downstream: TTGCTGCTTGTGCATCAC;
438 bp); G11a (Shapira et al., 1994; upstream: CAC-
GCAACAAGATGTGCT; downstream: CACACGTGA-
AGTGCGAAT; 469 bp); Gi1a (Olate et al., 1990; up-
stream: TTGATGAACAGCGGACCA; downstream: TTG-
GGGTGTTGTGCAACT; 399 bp); and G0a (Olate et al., Fig. 1. Dose-response relationship of Cl0 currents induced by sphin-
1989; upstream: TGCCAACACGCCTTTTCA; down- gosine-1-phosphate (S1P; A), and lysophosphatitic acid (LPA; B) in
Xenopus oocytes. The S1P and LPA elicited oscillatory Cl0 currents
stream: CACCTGTAGCCTGAGAAA; 492 bp). in a dose-dependent manner with half-maximal effective concentra-
tion of 35 mM and 0.2 mM, respectively. Each point represents mean {
RESULTS S.E. of four to six oocytes from one animal.
S1P induces long-lasting oscillatory chloride
currents in Xenopus oocytes: Distinctive
current patterns exhibited by S1P and LPA maximum peak of 300 nA and duration of 5 – 30 min
S1P and LPA induced concentration-dependent Cl0 (Fig. 2A). These were quite slow, and, long-lasting oscil-
currents in Xenopus oocytes (Fig. 1). The estimated lations, such as the initial peak of oscillation, usually
half-effective concentrations of S1P and LPA were 35 appeared from 1 min to 10 min after S1P application.
mM and 0.2 mM, respectively. The bath concentrations A few cells (8 of 41 cells) exhibited quite a rapid initial
of 50 mM of S1P and 1 mM of LPA were used throughout current on which oscillations were superimposed, then
the experiment, because, these concentrations elicited declined to a near-basal level followed by a secondary
maximal current response. phase of a series of sustained oscillations (Fig. 2B). The
Application of S1P (50 mM) induced a slow, transient secondary phase is more or less broader than the first
increase in [Ca2/]i over several minutes (Fig. 2). In ad- phase and exhibits a smaller peak intensity. Other cells
dition, higher frequency oscillations ride on top of the (11 of 41 cells) responded in a way similar to only the
transient increase. Oscillatory responses have varia- first part of the case shown in Figure 2C. Such cells
tions of their patterns (Fig. 2A – C) among oocytes from generally did not show a clear secondary oscillating
different animals and even from the same female. The current. For a variety of G protein-coupled receptors,
common response (22 of 41 cells) consisted of sustained it has been demonstrated that submaximal concentra-
oscillatory, inward current with an average range of tions of agonists can give rise to oscillatory currents,

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

 S1P AND LPA SIGNALING IN XENOPUS OOCYTES 415

Fig. 2. A – C: Patterns of currents evoked by S1P application. Inward

current is depicted as downward. S1P (2.5 mM) was delivered in 15-ml
aliquots over 1 – 2 sec by holding a hand-held pipette to the recording
chamber filled with 1.0 mL modified Barth saline (MBS; bath concen-
tration; 50 mM). Arrowheads indicate the points where reagent was
applied. Each trace is representative of at least three experiments.

Fig. 3. Effects of sequential application of S1P and LPA (A) or LPA

whereas a maximally effective concentration may and S1P (B) in Xenopus oocytes. There was no cross desensitization
evoke a single, large Ca2/ transient. However, this may between the two agonists. The bath concentrations of S1P and LPA
not be the case in the S1P-induced oscillations, because were 50 mM and 1 mM, respectively. Arrowheads indicate the points
where reagent was applied. Each trace is representative of at least
treatment with a higher dose of S1P elicited a similar three experiments.
oscillatory current pattern (data not shown).
In contrast, LPA (1 mM) caused rapid and transient
oscillating currents. The typical LPA response was usu-
ally more vigorous than S1P response (Figs. 3, 5A, 7A, S1P- and LPA-evoked Cl0 currents require
9A), such as maximum peak amplitude of about 2,800 [Ca2/]; release from IP3-sensitive Ca2/ pool
nA (n Å 18), which is ten times larger than that of To further verify the involvement of [Ca2/]; store in
S1P-induced currents (300 nA; n Å 11). The number of the phospholipid-induced current, we tested the oocyte
spikes was variable, from one to about ten, according responses in the following two conditions; first, the zero
to the concurrence of firing. The larger spike number, Ca2/ condition in the bath of the recording chamber by
the broader band, and the smaller peak amplitude had perfusion with Ca2/-free Oocyte Ringer-2 (OR-2) solu-
been obtained. Vehicle (1% BSA) alone did not induce tion (82.5 mM NaCl, 2.5 mM KCl, 1.0 mM MgCl2 , 1
these responses (data not shown). To demonstrate fur- mM Na2HPO4 , 5 mM HEPES, pH 7.6) containing 1 mM
ther that S1P and LPA signal through distinctive re- EGTA; second, the removal of free [Ca2/], by microin-
ceptors and downstream signaling pathways, we jection of EGTA into the oocytes (23 nl injection of 5
sought to determine whether there is a cross desensiti- mM EGTA; final concentration 300 mM). In the former
zation between LPA and S1P. Figure 3 shows that there condition, both S1P and LPA responses were unaffected
is no apparent heterologous desensitization of the cur- (Figs. 4B, 5B), whereas the currents completely disap-
rents between the two. In addition, there are no alter- peared in the later case (Figs. 4C, 5C). We also found
ations in the characteristic patterns (such as shape and that, when we depleted the internal Ca2/ store by incu-
amplitude) of each current. bation with thapsigargin (1 mM for 1 hr), a known inhib-

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

416 NOH ET AL.

Fig. 5. A: Normal response to 1 mM LPA of oocyte. LPA (0.1 mM)

was delivered in 10-ml aliquots over 1 – 2 sec by holding a hand-held
pipette to the recording chamber, which was filled with 1.0 ml MBS.
B: Effect of Ca2/ removal in the bath on the LPA current. After a 2-
min perfusion in Ca2/-free OR-2/1 mM EGTA solution, 1 mM LPA was
introduced. C: Effect of Ca2/ removal in cytosol on LPA response.
EGTA (300 mM) was injected; 30 min later, 1 mM LPA was applied to
the recording chamber. D: Oocyte treated with 1 mM thapsigargin for
1 hr in Ca2/-free OR-2 followed by voltage-clamp recording. E:
Niflumic acid was applied as a bath with a 500-mM concentration for
2 min before LPA application. Arrowheads represent the points where
reagent was applied. Each trace is representative of at least three
separate experiments.

Fig. 4. A: Control response to 50 mM S1P. B: Effect of Ca2/ removal

in the bath on the S1P current. After a 2-min perfusion in Ca2/-free
Oocyte Ringer-2 (OR-2)/1 mM EGTA solution, 50 mM S1P were applied. 1990), did not show any oscillatory currents (Figs. 4E,
There was no apparent difference from control in current responses. C: 5E), we could postulate that both LPA and S1P induced
Effect of Ca2/ removal in cytosol on S1P response. Twenty-three nanoli- a current mainly carried by Cl0 ions through Ca2/-
ters of 5 mM EGTA were injected (intracellular concentration, 300 mM) activated Cl0 channels.
in to Xenopus oocytes, and, 30 min later, 50 mM S1P were applied to
the recording chamber. D: Oocyte treated with 1 mM thapsigargin for To determine whether the source of [Ca2/], release
1 hr in Ca2/-free OR-2 followed by voltage-clamp recording. Intracellular is IP3-sensitive, we injected heparin, a well-known
Ca2/ store depletion abolished the current response to S1P completely. competitive inhibitor of IP3 receptor (Mak and
E: Niflumic acid, an inhibitor of Ca2/-activated Cl0 channels, was ap- Foskett, 1994), into the Xenopus oocytes. Heparin in-
plied as a bath with a 500-mM concentration for 2 min before S1P
application. The oscillatory current response did not appear. Arrow- jection (23 nl of 1 mM per single oocyte; intracellular
heads represent the points where reagent was applied. Each trace is concentration, 50 mM) completely abolished the cur-
representative of at least three independent experiments. rent evoked by S1P (Fig. 6), demonstrating a key role
for IP3-induced Ca2/ release. However, microinjec-
tion of 50 mM of ruthenium red (RuRed), a competitive
itor of the ER Ca2/ ATPase (Thastrup et al., 1987, inhibitor of ryanodine receptor that is responsible
1990), oocytes could not respond to S1P (Fig. 4D) or to for Ca2/-induced Ca2/ release (Phillippe and Basa,
LPA (Fig. 5D). Such results indicate [Ca2/], and its 1996), could not extinguish the S1P response, al-
release from that store is important for S1P and LPA though it brought about the partial decrease of peak
response in Xenopus oocytes. Because the oocytes pre- amplitude (Fig. 6). For a control of heparin injections,
treated with niflumic acid (500 mM for 2 min), a Ca2/- de-N-sulfated heparin, which is not specific to the
activated Cl0 channel blocker (White and Aylwin, IP3 receptor, was injected (23 nl of 1 mM injection;

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

 S1P AND LPA SIGNALING IN XENOPUS OOCYTES 417

Fig. 6. Responses to 50 mM S1P in oocytes injected with heparin or

ruthenium red (RuRed). A: H2O (23 nl)-injected control oocyte re-
sponse to 50 mM S1P (top). Inhibitory effect of heparin on S1P-acti-
vated current (middle). The oocyte was injected with 23 nl of 1 mM
heparin (MW, 6,000; intracellular concentration, 50 mM), and, 30 min
later, 50 mM S1P were applied. The oocyte was injected with 23 nl of
1 mM RuRed (intracellular concentration; 50 mM), and, 30 min later,
50 mM S1P were applied (bottom). A is the result of oocytes from one
animal. Arrowheads represent the points where reagent was applied.
Fig. 7. Responses to 1 mM LPA in oocytes injected with heparin or
B: Maximum peak of currents was evaluated and averaged. De-N-
ruthenium red (RuRed). A: H2O (23 nl)-injected control oocyte re-
sulfated heparin (de-NS-H) was tested as a control of heparin.
sponse to 1 mM LPA. Inhibitory effect of heparin on LPA-activated
Twenty-three nanoliter injection of 1 mM de-NS-H (intracellular con-
current. The oocyte was injected with 23 nl heparin (50 mM), and, 30
centration, 50 mM) did not affect the current response evoked by S1P
min later, 1 mM LPA was applied. The oocyte was injected with 50
application. RuRed did not affect the pattern of current significantly,
mM RuRed, and, 30 min later, 1 mM LPA was applied. A is the result
although it reduced the amplitude of the maximum peak compare with
of oocytes from one animal. Arrowheads represent the points where
that of the control. Heparin abolished the S1P current completely. The
reagent was applied. B: Maximum peak of currents was evaluated
data represent mean { S.E. of five or six oocytes per condition from
and averaged. RuRed did not affect the pattern of current. However,
one animal, and similar results were obtained from another two inde-
heparin abolished the LPA current completely, whereas the same
pendent experiments from different animals.
concentration of control de-N-sulfated heparin had no effect. The data
represent mean { S.E. of six or seven oocytes per condition from one
animal. The same results were obtained from another two indepen-
intracellular concentration, 50 mM). This had no ef- dent experiments.
fect on S1P-induced current (Fig. 6B). We previously
demonstrated that the range of concentrations of
these reagents is comparatively effective in Xenopus 1 mM LPA, whereas 50 mM RuRed or 50 mM de-N-
oocytes (Han and Lee, 1995). LPA response was also sulfated heparin have no effect in the response, sug-
examined in the same way. Figure 7 shows that 50 gesting the involvement of IP3 in LPA-induced cur-
mM heparin inhibits the current response evoked by rent response for [Ca2/]; release from the store.

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

418 NOH ET AL.

S1P response is mediated through PLC-xb,

whereas LPA signaling is through another
new PLC
In view of the fact that IP3 is usually generated by
PLC from phosphatidylinositol 4,5-bisphosphate (PIP2)
hydrolysis, we questioned whether a PLC would be in-
volved in the responsiveness of oocytes to S1P and LPA.
To demonstrate the possible involvement of the G pro-
tein-linked PLC (PLCb) in the process of phospholipid
response, we investigated the effect of the aminosteroid
U-73122, a relatively specific inhibitor of G protein-
mediated PLC activation in various cells (Smith et al.,
1990; Thompson et al., 1991; Smallridge et al., 1992).
The pretreatment with U-73122 (10 mM) for 30 min of
Xenopus oocytes inhibited the current responses
evoked by S1P (Fig. 8) and LPA (Fig. 9) in a dose-
dependent manner (Fig. 10). Recently, we demon-
strated that 10 mM U-73122 greatly reduced the in-
crease of IP3 mass in Xenopus oocytes in response to
progesterone stimulation (Noh and Han, 1997). U-
73343, a nonfunctional analog of U-73122, did not affect
the current responses evoked by S1P and LPA.
To further determine the involvement of PLC, we
injected AS-ODNs corresponding to nucleotide se-
quence 928 – 958 of PLC-xb (Ma et al., 1993; Table 1).
PLC-xb is the only PLCb isotype isolated from Xenopus
oocyte (Ma et al., 1993). The AS-ODN we used for this
study has already been shown to result in the degrada-
tion of PLC-xb (Ma et al., 1993; Quick et al., 1994) after
24 hr in the oocyte. A 10-ng injection of the AS-ODN
resulted in a 83.4% decrease in the response to S1P
(Fig. 8). The water or nonsense controls did not affect
the pattern or peak intensity of S1P-evoked Cl0 cur-
rent. The inhibitory effect of the AS-ODN was more
significant than that of U-73122 treatment, which is
an approximately 60% decrease (Fig. 8). These results
suggest that S1P responses are mediated through en-
dogenous Xenopus PLC, especially PLC-xb, in oocytes.
In contrast, LPA response was not diminished by AS-
ODN application against PLC-xb but labile to U-73122
treatment (Fig. 9). U-73122 reduced responsiveness to Fig. 8. Effects of U-73122 treatment and Xenopus phospholipase C
(PLC-xb) antisense oligonucleotide application to Cl0 currents induced
LPA upto 80%; however, oocytes injected with AS-PLC- by 50 mM S1P in Xenopus oocytes. A: Representative control response
xb did not show any difference in current pattern or (top). Oocyte was treated with 10 mM U-73122 for 30 min and then
peak intensity from nonsense-injected control oocytes. washed with MBS five times followed by clamping to 070 mV holding
These results indicate that LPA signaling pathways potential applied with 50 mM S1P (middle). Twenty-three nanoliters
of 30 mM antisense oligonucleotide, 31-mer, against Xenopus PLCb
are linked not to a PLC-xb but to another, unidentified (AS-PLCxb; final concentration 1.5 mM; Ç10 ng) were injected into
PLC isotype. Because the only PLC isotype cloned in oocytes (bottom). After 2 days in culture, oocytes current responses
Xenopus oocytes was PLC-xb, the use of antisense ap- were recorded. Arrowheads represent the points where reagent was
proaches to other types of PLC were limited. Nonethe- applied. B: Maximum peak of currents was evaluated and averaged.
Twenty nanograms of nonsense oligomer (39-mer; Table 1) were in-
less, we designed degenerate antisense oligomers jected as a control for the AS-PLC-xb injection. The sequence of non-
against PLCb and PLCg types, referring to several al- sense oligonucleotide was from Stehno-Bittel et al. (1995). U-73343
ready identified PLCb or g types from human, rat, and (10 mM for 30 min), a nonfunctional analog of U-73122, did not affect
the current response evoked by S1P. The data represent mean { S.E.
bovine. An AS-PLCb against a well-conserved region of of four or five oocytes per conditions from two animals.
the Y domain among PLCb1, b2, b3, and b4 types was
designed as a 24-mer degenerate oligodeoxynucleotide
containing five inosine bases conservative for PLC-xb
(2012 Ç 2035; (5*-TCCAGAAIAICTGIGGCATITAIT- G-proteins coupling with LPA or S1P induced
3*). An AS-PLCg, a 21-mer containing 1 inosine base, signal-transduction pathways in Xenopus
was designed to include a conserved region of X domain oocytes
among various PLCg types as template for rat PLCg The preferential coupling of the S1P responses to a
1319 Ç 1339 region (5*-GTCCTCIATGGACAGGAT- native PLC, presumably PLC-xb, prompted us to ask
GAC-3*). These AS-ODNs, did not abolish LPA and S1P which of the endogenous G proteins in Xenopus oocytes
responses (data not shown). are coupled with this PLC isoform in response to S1P.

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

 S1P AND LPA SIGNALING IN XENOPUS OOCYTES 419

Fig. 9. Effects of U-73122 treatment and PLC-xb antisense oligonu-

cleotide (AS-PLC-xb) application to LPA-evoked Cl0 currents in Xeno-
pus oocytes. A: Representative control response (top left). Oocyte
treated with 10 mM U-73122 for 30 min and then washed with MBS
five times, followed by clamping to 070 mV holding potential applied
with 1 mM LPA (bottom left). AS-PLC-xb (1.5 mM) was injected into Fig. 10. Dose-dependent inhibitory effect of U-73122 on S1P- and
oocytes (top right). After 2 days in culture, oocyte current responses LPA-induced Cl0 currents. Maximum peak of currents elicited by S1P
were recorded. Arrowheads represent the points where reagent was (A) and LPA (B) was inhibited by increasing doses of U-73122. Each
applied. B: Maximum peak of currents was evaluated and averaged. point represents mean { S.E. of four to six oocytes from one animal.
Twenty nanograms of nonsense oligomer were injected, as a control
for the AS-PLC-xb injection. U-73343 (10 mM for 30 min) did not affect
LPA current. The data represent mean { S.E. of 9 – 12 oocytes per
condition from one animal, and the same results were obtained from against Gas, AS-Gqa and AS-G11a specifically reduced
another two independent experiments. LPA-induced current up to 22% and 30% of control,
respectively (Fig. 12A), whereas, in S1P, AS-G1a and
AS-Gi1a inhibited current responses by 27% and 20%
To address this question, we also have utilized AS- of nonsense controls, respectively (Fig. 12B). In the case
ODNs against various G protein a subunits. We used of S1P, a partial decrease of current amplitude, al-
four kinds of AS-ODNs (Table 1), which have been re- though not as significant as that with AS-Gqa and AS-
ported previously in other papers; AS-Gqa and AS-G11a Gi1a, was observed with AS-G11a and AS-G0a. These
(from Stehno-Bittel et al., 1995) and AS-G0a and AS- data strongly indicate that S1P-induced signaling uti-
Gi1a (from Chen et al., 1994). We introduced 10 – 30 ng lizes Gq/Gi1 , whereas the LPA-induced signal pathway
of these AS-ODNs into oocytes by microinjection and is conducted via Gq/G11 , and PLC-xb is coupled prefer-
incubated them for 1 – 2 days. This should have been entially with S1P signaling but is not involved in LPA
sufficient to remove endogenously existing G proteins, signaling.
in which the turnover time is no longer than 2 days
(Stehno-Bittel et al., 1995). Indeed, quantitative RT- DISCUSSION
PCR analysis demonstrated that these AS-ODNs sig- S1P and LPA, as simple phospholipid molecules, in-
nificantly degraded the corresponding Ga transcripts duce Ca2/-activated Cl0 current in Xenopus laevis oo-
present in the oocytes (Fig. 11). Among AS-ODNs cytes. However, the components of the signal transduc-

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

420 NOH ET AL.

TABLE 1. Oligodeoxynucleotide sequences

Name Sequences Target
1
AS-PLCxb CGACATCTGGTCTCTGTCCAGGAACTGCCTGA PLCxb (928 – 958)
AS-Gq GCCTCCTTGACTAGTGGGGAGTG xGqa (15 – 38)
AS-G11 CATACCTTCCCCCCGGCACCGTGG xG11a (14 – 37)
AS-Go CATCGGCCCTTCTTTCT xGoa (421 – 437)
AS-Gi1 AACTCCAGCTAGTTCTGCA xGi1a (544 – 562)
Nonsense GCCGTGCTGAACGCCAAGGCCCTCGCCACCGGCACGCTC Nothing
1
All of the oligonucleotides were modified at their internucleoside linkages at both the 5* and the 3* ends with phosphorothioate bond to provide protection against
exonucleases. AS, antisense; PLCxb, Xenopus phospholipase C; Gq , G11 , Go , Gi1 , G-protein family.

Fig. 11. Quantitative analysis of mRNAs for G-protein alpha sub- as described in Materials and Methods. Elongation factor 1-a (EF1-
units (Gas) in oocytes injected with either antisense (AS) or nonsense a) was coamplified as an internal control. G0a transcripts appear to
(NS) oligonucleotides. Thirty oocytes per group were injected with be the least abundant compared with other Gas. Injection of antisense
10 – 30 ng of the oligonucleotides. Total RNA was extracted from the oligonucleotides against each Gas specifically reduced corresponding
oocytes, and cDNA was synthesized by reverse transcription. mRNA Ga transcripts in oocytes. Molecular markers (M) are indicated on
levels of a subunits of different types of G proteins were assayed by the right. bp, Base pair.
using quantitative reverse transcriptase-polymerase chain reaction,

tion pathway involved in these phospholipids responses lack of inhibition by removal of external free calcium
in Xenopus oocyte are not known. In this study, compar- ions establishes that these signals are not mediated
ative examination of currents evoked by S1P and LPA by extracellular Ca2/ but by intracellular Ca2/. These
was performed. In addition, the signal-transduction results are consistent with earlier demonstrations (Du-
pathways induced by these molecules were addressed rieux et al., 1992, 1993; Ferguson and Hanley 1992).
in their initial steps by using AS-OND injection experi- In addition, the need of intracellular Ca2/ store for S1P-
ments. We show here that differences exist in their and LPA-induced Cl0 channel activity was further veri-
initial part of signaling cascades between LPA and S1P. fied by Ca2/-depletion experiments using thapsigargin.
From the results of AS-ODN experiments, it is demon- Furthermore, heparin/RuRed injection experiments re-
strated that S1P-induced signaling utilizes Gq/Gi1 , vealed that an IP3-sensitive Ca2/ pool is responsible for
whereas the LPA-induced signal pathway is conducted both S1P- and LPA-induced Ca2/ release in Xenopus
via Gq/G11 , and, as a more interesting divergence, PLC- oocytes.
xb was engaged in S1P signaling but was not involved What causes the different patterns of current elicited
in LPA signaling. These results indicate that signal by S1P and LPA is difficult to assess. In response to
pathways evoked by S1P and LPA differ from each S1P, three types of current patterns were observed. This
other in G protein levels as well as PLC levels. Our variability in the response pattern is likely to be caused
data also strongly suggest that there may be an addi- by different physiological conditions among oocytes. Sev-
tional PLC yet to be identified that couples selectively eral sequential signaling components link receptor acti-
with LPA signaling in Xenopus oocytes. vation to the Cl0 channel opening. If these components
The responses evoked by these two molecules share are expressed to various degrees in different oocytes,
common features. Both of them need intracellular Ca2/ then such variability will be expected. Nonetheless, the
store and its release from the store. These currents are current patterns exhibited between S1P and LPA are
carried mainly by Cl0 ions through the Ca2/-activated quite distinctive. It may be possible that preferential
Cl0 channel. Increases in cytosolic Ca2/ were detected coupling of different signaling components (such as Gq/
indirectly by activation of an endogenous Ca2/-acti- Gi1 , PLC-xb in S1P, and Gq/G11 , or an unknown PLC
vated Cl0 channel downstream as a result of a common in LPA) to each receptor activation would cause such
mechanism of S1P and LPA responses. Identification unique current patterns. Different PLCs may have dif-
of S1P- and LPA-stimulated signals as a Cl0 current ferent enzymatic activities and regulation mechanisms;
was accomplished by its inhibition with niflumic acid. thus, although IP3/Ca2/ is involved in both pathways, it
Inhibition of currents with injected EGTA demon- could be somewhat different in its fine features, contrib-
strates the Ca2/ dependence of these currents, and a uting a characteristic Cl0 current pattern.

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

 S1P AND LPA SIGNALING IN XENOPUS OOCYTES 421

The observation of S1P responses to both the pertus-

sis toxin (PTX)-sensitive (Gi1) and insensitive G pro-
teins (Gq) for coupling with PLC is unique. G proteins
of the G1 family are involved mainly in the coupling of
membrane-bound receptors to PLC in mammalian cells
(Smrcka et al., 1991; Wilkie et al., 1991). In addition,
several lines of evidence have been suggested showing
that G0 and Gi proteins may be also involved in the
coupling of PLC in native and exogenously expressed
receptors in Xenopus oocytes (Blizter et al., 1993; Chen
et al., 1994; Karahasa and Sugiyama, 1994; Quick et
al., 1994). This includes those of receptors that use a
PTX-insensitive pathway to activate PLC in its native
environment (Aragay et al., 1992; Hsieh and Tashijian,
1992; Wu et al., 1992; Blizter et al., 1993). For example,
both thyrotropin-releasing hormone (TRH) and a1B-ad-
renergic receptors have been shown to activate PLC
via Gq in many types of cells (Aragay et al., 1992; Hsieh
and Tashijian 1992; Wu et al., 1992), such that it could
be expected that they would also use the same trans-
ducer in their oocyte system. However, in the Xenopus
oocyte, the Goa subunit is involved in the a1B-adrener-
gic response in oocytes (Blizter et al., 1993), whereas
Gs , but not Go , is transducing the TRH response (de la
Pena et al., 1995). In the case of the M1 muscarinic
receptor, which exhibits PTX-insensitive PLC activa-
tion in mammalian cells, it has been reported that Gi1
could participate in the PLC signal-transduction path-
way in Xenopus oocytes (Karahasa and Sugiyama,
1994). A similar case was shown in 5-HT (5-hydroxy-
tryptamine)2C receptor-mediated responses, in which
Go in oocytes activates PLC (Kaneko et al., 1992).
Meanwhile, Stehno-Bittel et al. (1995) have reported
that PTX-sensitive coupling of PLC with exogenously
expressed muscarinic receptor could be modulated by
bg subunits of G-protein (Gbg) in Xenopus oocytes. This
presents the possibility of bg-dependent activation of
PLCb and may account for the PTX-sensitive regula-
tion of PLC. In S1P, responses in Xenopus oocyte, Gq
and Gi1 appeared to be involved in the coupling of PLC-
xb. It may be possible that PLC-xb is activated by the
a subunit of Gq and the Gbg subunit of Gi1. PLC-xb, at
the time of identification, has been reported to couple
with PTX-sensitive G protein (Ma et al., 1993). How-
ever, a recent report of an in vitro reconstitution experi-
ment (Filtz et al., 1996) using signal components, in-
cluding PLC-xb itself, a subunits of several types of G
proteins and bg subunits of G proteins demonstrated a
possibility that the activity of PLC-xb can be modulated
by the a subunit of the Gq family as well as by the Gbg
Fig. 12. A, B: Effects of antisense oligonucleotides (AS-ODN) appli-
cation against four kinds of Gas on LPA- or S1P-induced current subunit. These data further indicate that PLC-xb could
responses. Ten to thirty nanograms of AS-ODNs to a subunits of be coupled with both of PTX-sensitive and insensitive
different types of G proteins were microinjected into oocytes. After a G protein systems. In human embryonic kidney (HEK)
1-hr holding period for wound healing, oocytes were transferred to cells, S1P potently and rapidly increased cytosolic Ca2/
oocyte culture media (MBS containing 5 mg/ml penicillin, 5 mg/ml
streptomycin, 400 mg/ml bovine serum albumin, and 2.5 mM pyruvate) in a PTX-sensitive manner (van Koppen et al., 1996),
and incubated for 2 days at 19 { 17C. A: LPA response. AS-ODNs indicating the possible involvement of Gi protein(s) in
against Gqa or Gi1a specifically diminished the peak amplitude of coupling of IP3/Ca2/ in S1P signaling. Together with
LPA-induced current, whereas other types of AS-ODNs, those to G0a these findings, our observation that Gq and Gi1 play a
or Gi1a, did not affect it significantly. Values are mean { S.E. of 7 –
12 oocytes per condition from one animal, and similar results were role in the coupling of PLC-xb can be explained. The
obtained from another two independent experiments. B: S1P response. significance of the apparently unusual dependence of
AS-ODNs against Gqa or Gi1a, not G11a or G0a, greatly reduced cur- S1P responses on both the PTX-sensitive and insensi-
rent responses evoked by 50-mM S1P application. Nonsense oligonu- tive G proteins, Gq and Gi1 , for coupling with PLC is
cleotide was used as a general control of AS-ODN injection experi-
ments. Values are mean { S.E. of four to seven oocytes per conditions not known. It is also unclear whether both a subunits
from one animal, and results similar were obtained from another two of Gq and Gbg coregulate PLC-xb at the same time.
independent experiments. In a previous report, LPA induced Cl0 current in a

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

422 NOH ET AL.

PTX-sensitive manner in the Xenopus oocyte, sug- Filtz, T.M., Paterson, A., and Harden, T.K. (1996) Purification and G
protein subunit regulation of a phospholipase C-beta from Xenopus
gesting that Gi protein would be linked in that pathway laevis oocytes. J. Biol. Chem., 271:31121 – 31126.
(Durieux et al., 1992). That result is not consistent with Ghosh, T.K., Bian, J., and Gill, D.L. (1990) Intracellular calcium re-
our data in which Gq and G11, which are categorized lease mediated by sphingosine derivatives generated in cells. Sci-
within Gq family, are responsible elements. In fact, we ence, 248:1653 – 1658.
were not able to observe consistent PTX sensitivity in Guo, Z., Liliom, K., Fischer, D.J., Bathurst, I.C., Tomei, L.D., Kiefer,
M.C., and Tigyi, G. (1996) Molecular cloning of a high-affinity recep-
either LPA or S1P response (data not shown). One ex- tor for the growth factor-like lipid mediator lysophosphatidic acid
planation could be that Gi may not be the only G protein from Xenopus oocytes. Proc. Natl. Acad. Sci. USA, 93:14367 – 14372.
that is susceptible to the toxin. Furthermore, because Han, J.K., and Lee, S.K. (1995) Reducing PIP2 hydrolysis, Ins(1,4,5)P3
we tested only the Gi1 type among the Gi family, other receptor availability, or calcium gradients inhibits progesterone-
stimulated Xenopus oocyte maturation. Biochem. Biophys. Res.
untested members of the Gi family, e.g., Gi2, Gi3, etc., Comm., 217:931 – 939.
might be involved naturally as additional elements. Hsieh, K.P., and Tashijian, A.H., Jr. (1992) Thyrotropin-releasing hor-
Whether the LPA response is PTX-sensitive or PTX- mone and gonadotropin-releasing hormone receptors activate phos-
insensitive, our data provide an important speculation pholipase C by coupling to the guanosine triphosphate-binding pro-
teins Gq and G11. Mol. Endocrinol., 6:1673 – 1681.
that, along with PLC-xb, additional, unidentified Kaneko, S., Takahashi, H., and Satoh, M. (1992) Metabotropic re-
PLC(s) is present in the oocyte that mediates LPA re- sponses to acetylcholine and serotonin of Xenopus oocytes injected
sponse, presumably, preferentially coupled with PTX- with rat brain mRNA are transduced by different G-protein sub-
insensitive G proteins. types. FEBS Lett., 299:179 – 182.
In conclusion, this study demonstrates that LPA and Karahasa, J., and Sugiyama, H. (1994) Inositol phospholipid metabo-
lism in Xenopus oocytes mediated by endogenous Go and Gi proteins.
S1P both release intracellular Ca2/ from the IP3-sensi- FEBS Lett., 355:41 – 44.
tive Ca2/ store through membrane receptors in Xeno- Kengaku, M., and Okamoto, H. (1995) bFGF as a possible morphogen
pus oocytes. They utilize distinctive signal components for the anteroposterior axis of the central nervous system in Xeno-
to generate intracellular second messengers. pus. Development, 121:3121 – 3130.
Ma, H.W., Blitzer, R.D., Healy, E.C., Premont, R.T., Landau, E.M.,
and Iyengar, R. (1993) Receptor-evoked Cl0 current in Xenopus oo-
ACKNOWLEDGMENT cytes is mediated through a b-type phospholipase C. J. Biol. Chem.,
We thank Dr. Richard Nuccitelli for critical reading 268:19915 – 19918.
Mak, D.O., and Foskett, J.K. (1994) Single-channel inositol 1,4,5-tris-
of the paper. phosphate receptor currents revealed by patch clamp of isolated
Xenopus oocyte nuclei. J. Biol. Chem., 269:29375 – 29378.
LITERATURE CITED Mattie, M., Brooker, G., and Spiegel, S. (1994) Sphingosine-1-phos-
phate, a putative second messenger, mobilized calcium from inter-
Aragay, A.M., Katz, A., and Simon, M.I. (1992) The G alpha q and G nal stores via an inositol trisphosphate-independent pathway. J.
alpha 11 proteins couple the thyrotropin-releasing hormone recep- Biol. Chem., 269:3181 – 3188.
tor to phospholipase C in GH3 rat pituitary cells. J. Biol. Chem., Noh, S.J., and Han, J.K. (1998) Inhibition of the adenylyl cyclase and
267:24983 – 24988. activation of the phosphatidylinositol pathway in oocytes through
Blizter, R.D., Omri, G., Vivo, M.D., Carty, D.J., Premont, R.T., Codina, expression of serotonin receptors does not induce oocyte maturation.
J., Birnbaumer, L., Cotecchia, S., Caron, M.G., Lefkowitz, R.J., Lan- J. Exp. Zool., 280:45 – 56.
dau, E.M., and Iyengar, R. (1993) Coupling of the expressed a1B- Olate, J., Jorquera, H., Purcell, P., Codina, J., Birnbaumer, L., and
adrenergic receptor to the phospholipase C pathway in Xenopus Allende, J. (1989) Molecular cloning and sequence determination of
oocytes. J. Biol. Chem., 268:7532 – 7537. a cDNA coding for the a-subunit of a G0-type protein of Xenopus
Bronfeldt, K.E., Graves, L.M., Raines, E.W., Igarashi, U., Wayman, laevis oocytes. FEBS Lett., 244:198 – 192.
G., Yamamura, S., Yatomi, Y., Sidhu, J.S., Krebs, E.G., Hokomori,
Olate, J., Martinez, S., Purcell, P., Jorquera, H., Codina, J., Birn-
S., and Ross, R. (1995) Sphingosine-1-phosphate inhibits PDGF-
baumer, L., and Allende, J. (1990) Molecular cloning and sequence
induced chemotaxis of human arterial smooth muscle cells: spatial
determination of four different cDNA species coding for the a-sub-
and temporal modulation of PDGF chemotactic signal transduction.
unit of G proteins form Xenopus laevis oocytes. FEBS Lett., 268:27 –
J. Cell Biol., 130:193 – 206.
31.
Chen, Y., Baez, M., and Yu, L. (1994) Functional coupling of the 5-
HT2C serotonin receptor to G proteins in Xenopus oocytes. Neurosci. Oliveira, A., and Spiegel, S. (1993) Sphingosine-1-phosphate as second
Lett., 179:100 – 102. messenger in cell proliferation induced by PDGF and FCS mitogens.
Chomczynski, P., and Sacchi, N. (1987) Single-step method of isolation Nature, 365:557 – 560.
by acid guanidinium thiocynate-phenol-chloroform extraction. Anal. Phillippe, M., and Basa, A. (1996) The effects of ruthenium red, an
Biochem., 162:156 – 159. inhibitor of calcium-induced calcium release, on phasic myometrial
Dagle, J.M., Walder, J.A., and Weeks, D.L. (1990) Targeted degrada- contractions. Biochem. Biophys. Res. Comm., 221:656 – 661.
tion of mRNA in Xenopus oocytes and embryos directed by modified Postma, F.R., Jalink, K., Hengeveld, T., and Moolenaar, W.H. (1996)
oligonucleotides: Studies of An2 and cyclin in embryogenesis. Nu- Sphingosine-1-phosphate rapidly induces Rho-dependent neurite
cleic Acids Res., 18:4751 – 4757. retraction: Action through a specific cell surface receptor. EMBO
de la Pena, P., del Camino, D., Pardo, L.A., Dominguez, P., and Barros, J., 15:2388 – 2395.
F. (1995) GS couples thyrotropin-releasing hormone receptors ex- Quick, M.W., Simon, M.I., Davidson, N., Lester, H.A., and Aragay,
pressed in Xenopus oocytes to phospholipase C. J. Biol. Chem., A.M. (1994) Differential coupling of G protein a subunits to seven-
270:3554 – 3559. helix receptors expressed in Xenopus oocytes. J. Biol. Chem.,
Dumont, J.N. (1972) Oogenesis in Xenopus laevis (Daudin) I. Stages of 269:30164 – 30172.
oocyte development in laboratory maintained animals. J. Morphol., Schumacher, K.A., Classen, H.G., and Spath, M. (1979) Platelet aggre-
136:153 – 180. gation evoked in vitro and in vivo by phosphatidic acids and lysoder-
Durieux, M.E., Salafranca, M.N., Lynch, K.R., and Moorman, J.R. ivatives: Identity with substances in aged serum (DAS). Thromb.
(1992) Lysophosphatidic acid induces a pertussis toxin-sensitive Haemost., 42:631 – 640.
Ca2/-activated Cl0 current in Xenopus oocytes. Am. J. Physiol., Shapira, H., Way, J., Lipinsky, D., Oron, Y., and Battey, J.F. (1994)
263:C896 – C900. Neuromedin B receptor, expressed in Xenopus laevis oocytes, selec-
Durieux, M.E., Carlisle, S.J., Salafranca, M.N., and Lynch, K.R. tively couples to Gqa and not G11a. FEBS Lett., 348:88 – 92.
(1993) Endogenous responses to sphingosine-1-phosphate in X. Smallridge, R.C., Kiang, J.G., Gist, I.D., Fein, H.G., and Galloway,
laevis oocytes: Similarities with lysophosphatidic acid signaling. R.J. (1992) U-73122, an aminosteroid phospholipase C antagonist,
Am. J. Physiol., 264:C1360 – 1364. noncompetitively inhibits thyrotropin-releasing hormone effects in
Ferguson, J.E., and Hanley, M.R. (1992) Phosphatidic acid and lyso- GH3 rat pituitary cells. Endocrinology, 131:1883 – 1888.
phosphatidic acid stimulate receptor-regulated membrane currents Smith, R.J., Sam, L.M., Justen, J.M., Bundy, G.L., Bala, G.A., and
in the Xenopus laevis oocyte. Arch. Biochem. Biophys., 297:388 – Bleasdale, J.E. (1990) Receptor-coupled signal transduction in hu-
392. man polymorphonuclear neutrophils: Effects of a novel inhibitor

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209

 S1P AND LPA SIGNALING IN XENOPUS OOCYTES 423
of phospholipase C-dependent processes on cell responsiveness. J. N-SH neuroblastoma cells. A role for Gq in receptor compartmenta-
Pharmacol. Exp. Ther., 253:688 – 697. tion. J. Biol. Chem., 266:23856 – 23862.
Smrcka, A.V., Hepler, J.R., Brown, K.O., and Sternweis, P.C. (1991) van Corven, E.J., Groenink, A., Jalink, K., Eichholtz, T., and Moole-
Regulation of polyphosphoinositide-specific phospholipase C activ- naar, W.H. (1989) Lysophosphatidate-induced cell proliferation:
ity by purified Gq . Science, 251:804 – 807. identification and dissection of signaling pathways mediated by G
proteins. Cell, 59:45 – 54.
Spiegel, S., and Milstein, S. (1995) Sphingolipid metabolites: Members
van Koppen, C.J., Heringdorf, D.M. zu, Laser, K.T., Zhang, C., and
of a new class of lipid second messengers. J. Membr. Biol., 146:225 – Jakobs, K.H. (1996) Activation of a high affinity Gi protein-coupled
237. plasma membrane receptor by sphingosine-1-phosphate. J. Biol.
Stehno-Bittel, L., Krapivinsky, G., Krapivinsky, L., Perez-Terzic, C., Chem., 271:2082 – 2087.
and Clapham, D.E. (1995) The G protein bg subunit transduces the White, M.M., and Aylwin, M. (1990) Niflumic and flufenamic acids
muscarinic receptor signal for Ca2/ release in Xenopus oocytes. J. are potent reversible blockers of Ca2/-activated Cl0 channels in
Biol. Chem., 270:30068 – 30074. Xenopus oocytes. Mol. Pharmacol., 37:720 – 724.
Thastrup, O., Linnebjerg, H., Bjerrum, P.J., Knudsen, J.B., and Chris- Wilkie, T.M., Scherle, P.A., Strathmann, M.P., Slepak, V.Z., and Si-
tensen, S.B. (1987) The inflammatory and tumor-promoting sesqui- mon, M.I. (1991) Characterization of G-protein alpha subunits in
terpene lactone, thapsigargin, activates platelets by selective mobi- the Gq class: Expression in murine tissues and in stromal and hema-
lization of calcium as shown by protein phosphorylations. Biochim. topoitic cell lines. Proc. Natl. Acad. Sci. USA, 88:10049 – 10053.
Biophys. Acta, 927:65 – 73. Wilson, P.A., and Melton, D.A. (1994) Mesodermal patterning by an
inducer gradient depends on secondary cell-cell communication.
Thastrup, O., Cullen, P.J., Drobak, B.K., Hanley, M.R., and Dawson,
Curr. Biol., 4:676 – 686.
A.P. (1990) Thapsigargin, a tumor promoter, discharges intracellu- Wu, K., Katz, A., Lee, C.H., and Simon, M.I. (1992) Activation of
lar calcium stores by specific inhibition of the endoplasmic reticulum phospholipase C by alpha 1-adrenergic receptors is mediated by the
calcium ATPase. Proc. Natl. Acad. Sci. USA, 87:2466 – 2470. alpha subunits of Gq family. J. Biol. Chem., 267:25798 – 25802.
Thompson, A.K., Mostafapour, S.P., Denlinger, L.C., Bleasdale, J.E., Zhang, H., Desai, N.N., Oliveira, A., Seki, I., Brooker, G., and Spiegel,
and Fisher, S.K. (1991) The aminosteroid U-73122 inhibits musca- S. (1991) Sphingosine-1-phosphate, a novel lipid involved in cellular
rinic receptor sequestration and phosphoinositide hydrolysis in SK- proliferation. J. Cell Biol., 114:155 – 167.

893D / 893D$$0209 06-04-98 11:22:05 wlcpa W Liss: JCP 0209