Biochimica et Biophysica Acta, 1092 (1991) 265-272 © 1991 Elsevier Science Publishers B.V. 0167-4889/91/$03.50 ADONIS 016748899100152M
265
BBAMCR 12908
Hydrolysis of phosphatidylcholine by phospholipase D is a common response to mitogens which stimulate inositol lipid hydrolysis in Swiss 3T3 fibroblasts S i m o n J. C o o k a n d M i c h a e l J.O. W a k e l a m Molecular Pharmacology Group, Institute of Biochemistry, University of Glasgow, Glasgow ( U.K.) (Received 12 September 1990)
Key words: Phospholipase D; Mitogen; Inositol lipid hydrolysis; Second messenger; (Swiss 3T3 fibroblasts)
The stimulated hydrolysis of inositol ]ipids and phosphatidylcholine (PtdCho) by bombesin, [ArgS]vasopressin ([ArgSlVp) and prostaglandin F2a (PGF2~) was analy~d in Swiss 3T3 cells pre-labeiled to isotopic equilibrium with either [methyl.3H]choline, myo.[2.3H]inositol or [9,10 (n)-3H]palmitic acid. AH three agonists activated the phospholipase D-catalysed hydrolysis of PtdCho as determined by the release of [3H]choline (Cho) and the formation of [3H]phosphatidylbutanol (PtdBut). The release of [3H]choline by each agonist exhibited similar sensitivity to prolonged pre-exposure to the phorbol ester 12-O-tetradecanoyiphorbol-13-acetate (I,~A). The release of [3H]choline exhibited the same dose dependency as the production of total inositoi phosphates for each mitogen suggesting that the two responses might be mediated through identical receptors. Acute pre-treatment with TPA allowed the dissociation of inositol lipid hydrolysis from PtdCho breakdown, since it inhibited inositol phosphate accumulation but stimulated choline generation. The loss of mitogen stimu|ated choline release in cells pre-treated with the phorbol ester for 48 h was not due to loss of stimulated inositoi phosphate production which was re?roducibly enhanced in these 'down-regulated' cells.
Introduction The rapid hydrolysis of phosphatidylinositol 4,5-bisphosphate (Ptdlns(4,5)P2) to generate the second messengers inositol-l,4,5-trisphosphate (Ins(1,4,5)P3 and
Abbreviations: [ArgSlVp, [arglnineS]vasopressin; Bomb, bombesin; Cho, choline; ChoP, phosphocholine; bAG, sn-l,2-diacylglycerol; DMBH, serum-free DMEM containing 20 mM Hepes (pH 7.4) and 1% (w/v) bovine serum albumin; DMEM, Dulbecco's modified Eagle's medium; DMSO, dimethyl sulphoxide; ECs0, concentration giving half of maximum response; GroPCho, glycerophosphocholine; HBG, Hanks' buffered saline solution containing 10 mM glucose and 1% (w/v) bovine serum albumin (pH 7.4); HBG.Li, HBG containing 10 mM LiCI; IC50, concentration giving half maximal inhibition of response; lns(1,4,5)Pa, inositol-l,4,5-trisphosphate; lnsPt, total inositol phosphates; lnsP, InsP2 and InsP3, total inositol mono-, hisand trisphosphates; PGF2o, prostaglandin F2a; PIC, phosphoinositidase C; PLC, phospholipase C; PLD, phospholipase D; PtdBut, phosphatidylbutanol; PtdCho, phosphatidylcholine; PtdOH, phosphatidic acid; TPA, 12-O-tetradecanoylphorbol-13-acetate. Correspondence: M.J.O. Wakelam, Molecular Pharmacology Group, Institute of Biochemistry, University of Glasgow, Glasgow, G12 8QQ, U.K.
sn-l,2-diacylglycerol (DAG) is now accepted as the means by which a number of mitogenic growth factors initiate a cascade of intracellular signals leading to exit from Go/G t, DNA synthesis and cell division (for review see Ref. 1). However, a number of recent reports have suggested that lipids other than the phospho;nositides may be in~oived in signalling pathways leading to the generation of DAG (reviewed in Refs. 2, 3). These include observations that mitogen-stimulated elevation of DAG is sustained long after the increase in Ins(1,4,5)Pa has declined [4,5], and that some growth factors appear to elevate DAG without concomitant hydrolysis of inositol lipids [6]. A major candidate for this alternative source of DAG is stimulated hydrolysis of phosphatidylcholine (PtdCho) which has now been reported in a variety of mitogenic and non-mitogenic sytems [2,3]. PtdCho hydrolysis has been proposed to occur by both phospholipase C and D catalysed processes [7,8], the former generating phosphorylcholine (ChoP) and DAG, the latter choline (Cho) and phosphatidic acid (PtdOH) which may then be converted to DAG by the action of phosphatidic acid phosphohydrolase [9].
266 We have previously shown that in Swiss 3T3 cells pre-labelled with [all]choline both bombesin, a potent mitogenic peptide, and the tumour promoting phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) stimulate the hydrolysis of PtdCho, apparently by a protein kinase C-dependent, phospholipase D catalysed mechanism [10]. Furthermore, we have shown that bombesin stimulates a biphasic increase in DAG mass; the first corresponding precisely with an increase in lns(1A,5)P3, whilst the second is sustained and divorced from any further elevation of Ins(1,4,5)P3 [51. The kinetics of bombesin-stimulated PtdCho hydrolysis and choline release are entirely consistent with it being the source of the second phase of DAG elevation. To determine if this effect upon PtdCho hydrolysis was a general response to mitogenic stimuli we have extended our studies to other peptide and non-peptide factors reported to be mitogenic in the Swiss 3T3 cell line. We report here that bombesin, [ArgSlVp and PGP2, all stimulate the hydrolysis of PtdCho by similar mechanisms which suggest a requirement for prior inositol lipid hydrolysis. Furthermore, we confirm that this release of choline represents activation of PLD by demonstrating formation of PtdBut, catalysed by the transferase activity of PLD, for each agonist. Materials and Methods
Cell cui:,re Swiss 3T3 cells were cultured in DMEM containing 10% calf serum in 24-weU plates and allowed to become quiescent by serum depletion and density arrest as described previously [101. Labelling of cells to isotopic equilibrium was achieved by allowing them to quiesce in medium containing either 1.5/~Ci ml-i [3H]choline or 1 /~Ci ml-~ myo.[3H]inositol and 1% calf serum for 48 h the latter utilising inositol.free DMEM. For assay of PtdBut formation Swiss 3T3 cells were grown in 6 well plates until 80% confluent; medium was then replaced with 2 ml of DMEM containing 2% calf serum and 2 ~Ci ml -I [9,10 (n)-3Hlpalmitic acid for 48 h. For experiments in which protein kinase C was 'down-regulated' cells were treated for 48 h during the period of isotopic labelling with 400 nM TPA or with the vehicle control, DMSO ( < 0.1% final concentration), or 400 nM p-phorbol. Such treatment has been extensively shown to deplete all cellular protein kinase C activity in Swiss 3T3 cells [17,34] and in our own laboratory to abolish [3H]phorbol-12,13-dibutyrate binding to Swiss 3T3 monolayer cultures (Carrie, S. and Wakelam, M.J.O., unpublished results).
Measurement of total inositol phosphates (InsPD Quiescent Swiss 3T3 cells labelled as above were washed in Hanks" buffered saline solution (pH 7.4) containing 10 mM glucose and 1% (w/v) bovine serum
albumin (HBG) for 5 min at 37°C. They were then washed in HBG containing 10 mM LiCI (HBG.Li) for a further 15 min at 37°C prior to treatment with the appropriate agent. In some experiments cells were pretreated during this 15-min period with varying concentrations of TPA during which the concentration of vehicle (DMSO) did not exceed 0.1% (v/v), which control experiments showed was without effect upon the parameters studied. After this pre-incubation period, cells were treated with the given concentration of mitogen for the times indicated in HBG.Li at 37°C. Incubations were terminated by addition of 0.5 ml of ice-cold methanol and the cell debris was scraped into plastic screw top poly-tubes. Following a second wash with 0.2 ml of methanol, the samples were extracted with 0.31 ml of chloroform for 1 h on ice. Phases were resolved by addition of 0.5 ml of distilled water and after centrifugation at 1 4 0 0 0 x g for 3 rain, 1 ml of the upper aqueous/methanolic phase was taken into 5 ml plastic insert vials. Total inositol phosphates in this aliquot were assayed as described previously [11,15].
Measurement of phosphatidylcholine hydrolysis products Quiescent Swiss 3T3 cells labelled with [3H] choline as above were washed at 37°C with Hanks' buffered salt solution and HBG as described previously. [10]. Cells were stimulated with the appropriate concentration of agonist for the times indicated in HBG at 37 °C and incubations were terminated by addition of 0.5 ml of ice-cold methanol without aspiration of the incubation medium. After scraping, washing with methanol and scraping again as described above, samples were extracted with 0.31 ml of chloroform for 1 h on ice. Following resolution of phases as above, aqueous, methanolic samples were applied to 1 ml Dowex-50WH + ion exchange columns prepared in glass wool plugged pasteur pipettes, and glycerophosphorylcholine (GroPCho), phosphocholine (ChoP) and choline (Cho) were eluted as described previously [10]. Aliquots from each fraction were analysed for radioactivity by liquid scintillation counting in Optiphase Hi-safe scintillant. An aliquot of the lipid phase was also counted for radioactivity allowing normalisation for variations in sample size.
Assay of PtdBut formation [9,10 (n)-3H]Palmitate labelled Swiss 3T3 cells were washed in DMBH for 30 rain at 37°C before incubating with DMBH containing agonist in the presence or absence of 30 mM butan-l-ol (0.3%, v/v) for 10 rain. Reactions were terminated by aspiration and addition of 0.7 ml ice-cold methanol and cell debris was harvested into glass vials. Following extraction of phospholipids by addition of 0.7 ml of chloroform phases were resolved by addition of 0.585 ml of water. The lower
267 chloroform phase was dried down in vacuo, re-dissolved in 75/tl of chloroform/methanol (19/1, v/v) and applied to Whatman LK5DF t.l.c, plates which were than fully developed in the solvent system B described by Bonser et al. [39]. PtdBut, identified by its co-migration with a [~4C]PtdBut standard and novel appearance in the presence of butan-l-ol, was excised from the t.l.c. plate and the associated radioactivity determined by scintillation counting. In each case the results shown represent the mean + S.D. for a single experiment typical of three or four, where for each experimental point n - 3. Significance was estimated using an unpaired Student's t-test, whilst ECs0 values were calculated from dose-response curves fitted to a logistic equation as defined in Ref. 35. Bombesin and [ArgS]Vp were obtained from Cambridge Research Biochemicals, Cambridge, U.K. TPA, PGF2o and Dowex resins were from Sigma, Poole, Dorset, U.K. myo-[2-~H]inositol, [methyl-3H]choline chloride and [9,10 (n)-3Hlpalmitic acid were from Amersham International, Amersham, U.K. All other chemicals were of the highest grade commercially available as reported previously [5,10,11,15]. Results
We have previously shown that bombesin stimulates the rapid hydrolysis of PtdCho in Swiss 3T3 cells and suggested that this was catalysed by a phospholipase D on the basis that Cho generation was always observed before any rise in Chop [10]. Experiments were performed to see if [Arga]Vp and PGF2~, both mitogenic for Swiss 3T3 cells [12,13] and both simulating the hydrolysis of inositol lipids [14,11], could also stimulate PtdCho hydrolysis. Fig. 1 shows the time-course of increase in [3H]Cho generation in Swiss 3T3 cells stimulated with [ArgS]Vp. The peptide caused a rapid elevation of [3H]Cho which was maximally above the control at 5 min and sustained for the duration of the time-course up to 15 min (Fig. 1) and 30 rain (data not shown). Changes in the radioactivity associated with the ChoP fractions were less reproducible, occurred over a longer time scale and were smaller in fold terms (data not shown). No significant changes in the radioactivity associated with the GroPCho fraction were observed (data not shown). These results, and particularly the kinetics of Cho elevation, are very similar to those obtained in bombesinstimulated Swiss 3T3 cells [10] and also in other systems recently reported [23,24]. Elevation of choline levels was also observed using 1/~M PGF2,~ (results not shown). Experiments were performed to investigate the dosedependency of the effects of bombesin, [ArgS]Vp and PGF2~ upon choline levels (Fig. 2). The results show that all three agonists stimulated increases in Cho in a dose-dependent manner with ECs0 values of 1.44 4- 1.02
5°t
4500
4000" 3500'
E
3000' 2500' 2000' 1500
.
.
.
.
i
0
.
.
.
.
|
5
.
.
.
.
10
i
15
Time (min) Fig. I. Time-course of [ArgS]Vp-stimulated increases in Cho. [3H]choline labelled Swiss 3T3 cells were stimulated with 100 nM [ArgS]Vp in HBG (m) or HBG alone (El) for the times indicated. Following termination of the reactions [3H]Cho was isolated by ion-exchange chromatography as described in Materials and Methods. The results are means+_S.D (n = 3) from a single experiment representative of three.
nM (mean of four experiments) for bombesin (data not shown, see Ref. 10), 0.92 4-0.57 nM (mean of three experiments) for [ArgS]Vp and 0.51 4- 0.32 #M (mean of three experiments) for PGF2~. The elevation of Cho was generally greatest with bombesin lower with [Arg~:]Vp and smallest with PGF2o. 200 180
ii
160' 140' 120' 100'
80
-13
i
|
- 11
-9
-7
i
[Mitogen]
M
-5
-3
Fig. 2. Dose-dependence of [ArgS]Vp and PGF2,~-stimulated [3H]Cho generation. [3H]Choline labelled Swiss 3T3 cells were stimulated with HBG containing [ArgS]Vp (m) or PGF2a (rl) at the final concentrations indicated for 5 min. Reactions were terminated and the radioactivity associated with the Cho fraction determined as described in Materials and Methods section. The results are means+ S.D. (n = 3) from a single experiment representative of three and are expressed as a percentage of the control. The control values were 3290+ 151 dpm for [ArgS]Vp and 5992+368 dpm for PGF2,~.
268 TABLE 1
800
Stimulated [~H]phosphatidylbutanol formation in Swiss 3T3 cells [3H]Palmitate-labelled Swiss 3T3 cells were stimulated with either bombesin (100 nM), [ArgS]Vp (100 nM) or PGF2a (2 pM) for 10 rain in the presence of 0.3~ (v/v) butan-l-ol, [3H]PtdBut was resolved by t.l.c, as described in Materials and Methods. Results are means :t: S.D. (n = 3) from a single representative experiment.
600'
E~ •_
Control Bombesin [ArgS]Vp PGF2~
2OO
-12
-14
.10
.s
- butan-l-ol
+ butan-l-ol
658 + 682+ 614+ 618+
933 + 255 6320+ 350 5847+1008 4788+ 357
125 77 85 131
-6
bombesin, [Arga]Vp or PGF2, resulted in the formation of PtdBut only in the presence of butan-l-ol (Table I). To compare the dose-dependency of increases in Cho with inositol lipid breakdown, dose response curves for the production of total inositol phosphates were performed. The results shown in Fig. 3 (a and b) show that the three agonists stimulated the accumulation of total inositol phosphates with a very similar dose-dependency to the corresponding Cho response (ECs0 values: 0.58 + 0.34 nM (mean of three experiments) for bombesin, 0.48 + 0.15 nM (mean of three experiments) for [ArgS]Vp and 0.16 + 0.14 pM (mean of four experiments) for PGF2~). In addition, the accumulation of
[Bemhesin] nM
500
400
300
m ~ mg=
D.p.m. in [3H]PtdBut
Incubation
~tm
t.
200 E~
100
30000' 0
- 13
i
!
i
-i I
-9
-7
w
-5
-3
[Mitogen] M
Fig. 3. (a) Dose-dependence of bombesin-stimulated InsPt accumulation with or without acute TPA pre-treatment. [3Hllnositol-labelled Swiss 3T3 cells were washed with HBG.Li containing 400 nM TPA (ra) or DMSO (11) as described in Materials and Methods before stimulating with the given concentration of bombesin in HBG.Li for 20 rain. (b) Dose-dependence of [Ar88]Vp and PGF2~-stimulated InsP, accumulation. [3Hllnositol labelled Swiss 3T3 cells were washed in HBG.Li befoi.~, stimulating with HBG.Li containing the given final concentrations of [ArgSlVp (11) or PGF2,, (ra) for 20 rain. In all cases InsPt were assayed as described in Materials and Methods. The results, expressed as percentage of control to allow comparisons, are means+S.D (n ffi 3) of a single experiment representative of three. Control values were 2716+93 dpm for bombesin, 2586:i:400 dpm for [ArsSlVp and 4997 + 1072 dpm for PGF2,~.
To confirm that this release of Cho did represent PLD-catalysed PtdCho hydrolysis, the ability of all three agonists to form [3H]PtdBut in [3H]palmitatelabelled cells in the presence of butan-l-ol was determined. In Swiss 3T3 cells the majority of [3H]palmitare incorporated into phospholipids is found in PtdCho (53~$) with negligible labelling of PtdIns (< 3~) (data not shown). Under these conditions stimulation with
25000'
.ffi
20000'
E eL
15000 ~
10000
.......
0.001
I
0,01
........
|
O. 1
........
!
........
I
g
I0
........
|
I00
.......
1(300
[TPA] nM Fig. 4. Dose-dependence of TPA-mediated inhibition of [ArgS]Vp stimulated InsPt accumulation. [3Hllnositol labelled Swiss 3T3 cells were washed for 15 rain in HBG.Li containing the indicated final concentrations of TPA (I) before stimulating with 100 nM [ArgS]Vp for 20 min. InsPt accumulation was assayed as described. The results are means+S.D. ( n = 3 ) of a single experiment representative of three. In this particular example the unstimnlated level of InsPt was 2566+829 dpm and 100 nM TPA gave a maximum 57% inhibition of the response to [ArgS]Vp. Similar dose response curves for the effects of TPA upon bombesin and PGF2,rstimulated lnsPt accumulation were also obtained (data not shown).
269 total InsP, by each agonist was inhibited by approx. 50~ following a 15 min preincubation with TPA (Fig. 3a and Fig. 4). In the case of bombesin and [ArgS]vasopressin this pre-treatment resulted in a shift to the right in the dose-response curve for InsP t accumulation giving new ECso values of 4.4 +_ 4.5 nM and 1.12 _ 0.52 nM, respectively. This TPA-mediated inhibition of inositol phosphate production was both time- and dose-dependent. The onset of TPA-stimulated inhibition of a subsequent bombesin-stimulated inositol phosphate accumulation occurred with a short lag time of approx. 1-1.5 min after addition of the phorbol ester (data not shown). Fig. 4. shows the dose-response curve for the TPAstimulated inhibition of [ArgS]Vp InsPt accumulation. Similar dose-response curves were obtained for the inhibition of bombesin and PGF2o stimulated InsPt accumulation (data not shown) and the ICs0 values for TPA were 5.58 + 3.84 nM against bombesin, 2.36 + 1.68 nM against [Arga]Vp and 2.16 _+0.52 nM against PGF2o. In cells treated with 400 nM TPA for 48 h, a treatment which down-regulates protein kinase C activity [17], the production of inositol phosphates in response to each agonist was reproducibly and significantly enhanced (Fig. 5) (the significance of enhancement was for bombesin P = 0.005, [Arga]Vp, P < 0.001 and for PGF2~, P < 0.001). Interestingly, the degree of enhancement of InsPt accumulation in 'down-regulated' cells varied according to the mitogen used. Thus, the enhancement of responses over the control cells was typically 1.2-1.4-fold for bombesin, 2.2-2.5-fold for [ArgS]Vp and 2.5-3.2-fold for PGF2~. Furthermore, in
160000 •
120000 •
"=" =[
4 8OOOO' OOOO'
0r-m-Control
r-I Bb
AVp
PGF2¢x
TPA
Fig. 5. Effect of chronic TPA pro-treatment upon agonist stimulated InsPt accumulation. Swiss 3T3 cells were labelled with [3H]inositol for 48 h in the presence of 400 nM TPA (hatched bars) or DMSO (open bars). After washing in HBG.Li the cells were stimulated with either a control buffer (HBG.Li), 100 nM bombesin, 100 nM [ArgS]Vp, 1 ~M PGF2, or 100 nM TPA in HBG.Li as indicated for 20 rain. InsPt was assayed as described. The results are means± S.D. (n = 3) for a single experiment representative of three.
12000"
80oo
6OOO
4OOO Con
TPA
Bomb
AVp
PGF2¢z
Fig. 6. Effect of chronic pre-treatment with TPA upon agonist-stimulated [3H]Cho generation. Swiss 3T3 cells were labelled with [3H]choline for 48 h in the presence of 400 nM TPA (hatched bars) or DMSO (open bars). After washing in HBG, cells were stimulated with a control buffer (HBG), 100 nM TPA, 100 nM bombesin, 100 nM [ArgS]Vp or 1/~M PGF2,, in HBG for 5 rain. The amount of [3H]Cho generated was normalised to the labelling in total choline-containing lipids. Results are means +_S.D. (n = 3 ) from a single experiment representative of four.
'down-regulated' Swiss 3T3 cells acute pre-treatment with TPA no longer inhibited InsPt accumulation in response to bombesin, [ArgS]vasopressin or PGF2,. For example, in 'down-regulated' cells InsPt accumulation after 20 min stimulation in the presence of 10 mM LiCI was 80451 + 2716 dpm for bombesin, 80112 + 3134 dpm for [ArgS]vasopressin and 60255 + 1703 dpm for PGF2~, whilst in "down-regulated" cells pre-treated with 400 nM TPA the values were 82619 + 3334 dpm, 83002 + 2620 dpm and 61 603 + 4482 dpm, respectively; the basal dpm was 9436 + 488 (values from a single, representative experiment). We have previously reported that long term treatment of Swiss 3T3 cells with TPA, as described above, results in the abolition of bombesin- and TPA-stimulated generation of choline [10]. In Fig. 6 we confirm these findings and also show that the ability of both [ArgS]Vp and PGF2~ to stimulate Cho release is also greatly reduced in 'down-regulated' cells. In 'control' cells agonist-stimulated Cho elevation was significantly above the buffer control ( P < 0.05) and the corresponding 'down-regulated' response ( P < 0.02) in all cases, whilst there was no significant elevation of Cho in 'down-regulated' cells. When sub-maximal concentrations of bombesin and [ArgS]vasopressin are combined the elevation [3Hlcholine is additive rather than synergistic when compared to the two agonists alone, whilst combination of maximal doses of bombesin and [Arga]vasopressin or TPA does not enhance the reponse of either agent alone (data not shown).
270
Discussion Agonist-stimulated hydrolysis of phosphatidylcholine has now been reported in a variety of call types including NG-108 cells [18,19], neutrophils [20], Swiss 3T3 fibroblasts [10,21,22], pulmonary endothelial cells [231. and spermatazoa stimulated to undergo the acrosome reaction by a fucose sulphate glycoconjugate [24]. Agonist-stimulated turnover of PtdCho is thought to have a role in providing an alternative source of DAG in response to extracellular stimuli and a PtdCh," ~.3"cle analagous to the inositol lipid cycle has been proposed [3]. Both phospholipase D and phospholipase C catalysed pathways of PtdCho hydrolysis have been reported [20,71 and the products of each reaction seem to have potent biological effects. Thus, DAG is known to be the endogenous activator of protein kinase C (reviewed in Ref. 25) and PtdOH has been reported to be a growth factor [27] and a precursor of DAG [9]. In Swiss 3T3 cells stimulated with bombesin or TPA, a protein kinase C-dependent, PtdCho phospholipase D activity is stimulated [10] and it has been suggested that this may contribute to the sustained elevation of DAG mass in Swiss 3T3 cells stimulated by the mitogenic peptide bombesin [5]. Here we show that this phenomenon is not peculiar to bombesin but also occurs in response to stimulation by two other calcium-mobilizing, mitogenic agents, the peptide [ArgSlVp and the eicosanoid PGF2,. Like bombesin, [ArgS]Vp stimulated the rapid generation of [3H]Cho from cells labelled with [3H]choline prior to any change in [3H]ChoP, suggesting activation of a phospholipase D activity (Fig. 1). The slower and relatively modest increases in ChoP (in fold terms) may represent a phospholipase C activity, but may also be explained in terms of phosphorylation of the liberated choline by choline kinase [31,32]. The elevation of choline by all three mitogens was dose-dependent (Fig. 2) and the ECs0 values are identical to those for the stimulated accumulation of total inositol phosphates in each case (Fig. 3). This suggests that for each mitogen the two responses are stimulated through common receptors. Furthermore, the ECs0 values for both [ArgSlVp and PGF2° were of a similar order to those quoted for stimulation of DNA synthesis by each agonist [12,13]. The results in Table I confirm that the release of Cho by the three agonists is associated with the activation of PLD since all three stimulated the formation of Ptdt6ut, in the presence of butan-l-ol, catalysed by the transferase activity of PLD [20]. The much greater magnitude of this response compared with Cho release reflects the virtual absence of PtdBut in control cells. This indicates that PLD activity is very low in the absence of agonist. The ability of all three mitogens to stimulate accumu-
lation of total inositol phosphates exhibited similar sensitivities to pre-treatment with TPA (Figs. 3a and 4). Thus, a 15 min pre-treatment with TPA resulted in a dose-dependent inhibition of a subsequent response to bombesin (Fig. 3a), [ArgS]Vp (Fig. 4) or PGF2, (data not shown). The IC5o for this effect of TPA was similar for all three receptor agonists (2-6 nM); this value is of a similar order to that previously quoted for inhibition of bombesin-stimulated inositol phosphate accumulation [16], activation of C-kinase [33] and the generation of choline by TPA in Swiss 3T3 cells [10]. In addition, the TPA-mediated inhibition of agonist-stimulated InsPt accumulation was also time-dependent with a lag of approx. 1 rain from addition of TPA to the onset of inhibition. We have previously observed a similar lag in the time-course of TPA stimulated increases in choline levels [10] and this, together with the similarities in dose-dependency of the TPA effects, clearly suggests that activation of protein kinase C, inhibition of agonist stimulated inositol phosphate accumulation and activation of PtdCho hydrolysis are related. Indeed, acute treatment with TPA effectively dissociated choline generation from inositol phosphate accumulation since it stimulated the former at the same concentration as it inhibited the latter. In this respect TPA is also able to dissociate DAG generation from inositol lipid hydrolysis, since addition of the phorbol ester to Swiss 3T3 cells results in an elevation of DAG mass in the absence of InsP t accumulation but with similar kinetics to [3H]choline release (Cook, S.J. and Wakelam, M.J.O., unpublished results, Ref. 29). To further investigate the relationship between inositol phosphate accumulation, choline release and protein kinase C, the enzyme was 'down-regnlated' by pre-treating Swiss 3T3 cells with 400 nM TPA for 48 h [17]. In agreement with the results of Brown et ai. [16] we found that under such conditions bombesin-stimulated accumulation of lnsP t was no longer inhibited by a short pre-treatment with TPA and that responses were reproducibly enhanced (Fig. 5). Similar results were obtained for [ArgSlvasopressin and PGF2, and are consistent with the hypothesis that protein kinase C activates a feedback inhibitory pathway to limit inositol lipid hydrolysis, presumably through the production of DAG [16]. The degree of enhancement of InsPt accumulation in 'down-regulated' cells was different for each agonist, with PGF2, always giving the greatest enhancement, then [ArgS]Vp and bombesin always the smallest. Similar results were recently reported by Brown et al. [34] and may represent subtle differences in the C-kinase sensitivity and perhaps mechanism of coupling of the receptors to PIC. However, a simpler alternative interpretation may lie in the fact that in 'control' cells PGF2,, gives the smallest accumulation of InsPt (Fig. 3b) and the smallest depletion of [3H]inositol lipids over a 20 min stimulation in the presence of 10 mM LiCI. Re-
271 moval of C-kinase mediated-inhibition by 'down-regulation' would then allow large potentiation of the response. Such a mechanism would operate for [ArgS]Vp and bombesin, but since they give successively greater responses in 'control" cells and much greater depletion of [3H]incqtol lipids, potentiation of InsPt accumulation in 'down-regulated' cells may be limited by the amount of [3H]inositol lipids available as substrate for PIC (data from a typical experiment: dpm in total inositol lipids; control 20 min = 133714 + 8774 dpm, bombesin 20 rain = 70120 + 3239 dpm, [ArgS]Vp 20 min = 107748 + 10859, PGF2~ 20 min = 136826 + 8123 dpm). In protein kinase C-'down-regulated' cells the ability of TPA and bombesin to stimulate choline release was severely attenuated (Fig. 6) as previously reported [101. Furthermore, the ability of [ArgS]vasopressin and PGF2~ to stimulate choline release was also abolished in the down-regulated cells indicating a similar mechanism for activation of PtdCho hydrolysis, presumably in:olving protein kinase C activation. Since in the 'down-regulated' cells stimulated InsPt accumulation was enhanced (Fig. 5) the abolition of mitogen-stimulated PtdCho hydrolysis was not due to loss of inositol lipid hydrolysis and was thus distal to this event. A common mechanism of activation of PtdCho hydrolysis is supported by the finding that maximal concentrations of bombesin and [ArgS]Vp do not stimulate Cho generation in an additive manner, whilst sub-maximal concentrations are indeed additive. Our results differ from those of Price et al. [22] who were unable to observe PGF2,-stimulated PtdCho hydrolysis in Swiss 3T3 cells. We can only speculate that this represents variation between different clones of Swiss 3T3 cells, a proposal supported by the fact that Price et al. were only able to observe a very small accumulation of InsPt in response to PGFz,~ (less than 2-fold). Indeed, this may explain why they were unable to see stimulated PtdCho hydrolysis since such a small inositol lipid signal would result in a much smaller C-kinase activation. If protein kinase C is a major link between inositol lipid hydrolysis and PtdCho hydrolysis, the magnitude of PKC activtion will be a major determinant of the magnitude of PtdCho hydrolysis. Such a model is supported by the observation that the clear rank order of InsPt accumulation (bombesin > [ArgS]Vp > PGF2,~) is broadly reflected in the rank order of Cho generation (bombesin > [ArgS]Vp > PGF2,) and to a lesser extent PtdBut formation (Table I). Based on the observations reported here and previously by us [5,10] and others [21,22] we suggest that in Swiss 3T3 cells those mitogens which stimulate the rapid hydrolysis of inositol lipids may also stimulate hydrolysis of PtdCho by a C-kinase mediated, phospholipase D-catalysed mechanism. Activation of PLD will result in the formation of PtdOH. Early reports of a
Ca2+-ionophore role for PtdOH [26] have remained unproven but it is clear that both PtdOH [2'7] and lyso-PtdOH [28] are potent mitogens when added exogenously to cultured call lines. Indeed, PtdOH and lysoPtdOH exhibit many of the characteristics of classical receptor-coupled mitogens causing hydrolysis of Ptdlns(4,5)P 2 (Plevin, R., MacNulty, E., Palmer, S. and Wakelam, M.J.O., unpublished results) inhibition of adenylate cyclase activity, calciuni mobilization and DNA synthesis [28]. In addition to agonist-like properties, PtdOH also exhibits second messenger-like properties including replacing PtdSer in supporting Ca2+-independent PKC activity [37]. Furthermore, arachidonoylcontaining PtdOH is able to inhibit the catalytic activity of the p21 r~s GTPase activating protein (GAP) in vitro resulting in the accumulation of the active GTP-bound form of p2V a~ [38] which is a potent mitogenic signal. Alternatively, a functional link between inositol lipid hydrolysis and PtdCho hydrolysis may serve to provide a sustained production of DAG [5] as part of a long-term mitogenic signal when inositol lipid hydrolysis has been rapidly desensitized [5,16], Ins(1,4,5)P3 levels have declined and intraeellular Ca 2+ levels have returned towards control levels. Since the DAG produced from hydrolysis of PtdCho may have a different fatty acid composition compared to that from Ptdlns(4,5)P2 [36] and will be produced in the absence of elevated intracellular calcium concentrations, the recent description of protein kinase C-~ [30], a novel calcium-independent PKC isozyme, may be of relevance to these observations. Since for each mitogen the generation of choline exhibited the same dose-dependency as inositol phosphate accumulation and DNA synthesis in Swiss 3T3 cells it is possible that it is a normal component of the mitogenic signal resulting in sustained elevation of DAG and activation of perhaps novel PKC species.
Acknowledgements This study was supported by grants from the Medical Research Council (U.K.) and the Cancer Research Campaign. In addition, S.J.C. received an S.E.R.C. CASE award studentship in association with the Wellcome Foundation Ltd.
References 1 Berridge, M.J. (1987) Biochim. Biophys. Acta 907, 33-45. 2 Exton, J.H. (1990) J. Biol. Chem. 265, l-4. 3 Pelech, S.L. and Vance, D.E. (1989) Trends Biochem. Sci. 14, 28-30. 4 Wright, T.M., Rangan, L.A., Shin, H.S. and Raben, D.M. (1988) J. Biol. Chem. 263, 9374-9380. 5 Cook, S.J., Palmer, S., Plevin, R. and Wakelam, M.J.O. (1990) Biochem. J. 265, 617-620. 6 Wright, T.D., Shin, H.S. and Raben. D.M. (1990) Biochem. J. 267, 501-507.
272 7 Diaz-Meco, M.T., Larrodera, P., Lopez-Barahona, M., Cornet, M.E., Barreno, P.G. and Moscat, J. (1989) Biochem. J. 263, 115-120. 8 Billah, M.M., Pal, J.K., Mullmann, T.J., Egan, R.W. and Siegel, M.I. (1989) J. Biol. Chem. 264, 9069-9076. 9 Martin, T.W. (1988) Biochim. Biophys. Acta 962, 282-296. 10 Cook, SJ. and Wakelam, M.J.O. (1989) Biochem. J. 263, 581-587. 11 Black, F.M. and Wakelam, MJ.O. (1990) Biochem.J. 266, 661-667. 12 Corps, A.N., Rees, L.H. and Brown, K.D. (1985) Biochem. J. 231, 781-784, 13 Macphce, C.H., Drummond, A.H., Otto, A.M. and Jimenez de Asua, L, (1984) J. Cell. Physiol. 119, 35-40. 14 Nanberg, E, and Rozengurt, E. (1988) E M B O J. 7, 2741-2747. 15 PIevin,R,, Palmer, S., Gardner, S.D, and Wakelam, MJ.O. (1990) Biochem, J. 268, 605-610, 16 Brown, K.D,, Blakeley, D.M.. Hamon, M.H,, Laurie. M.S, and Corps, A.N, (1987) Biochem. J, 245, 631-639, 17 Rodriguez.Pena, A. and Rozengurt. E. (1984) Biochem. Biophys. ges, Comman. 120, 1053-1059. 18 Liscovitch, M., Blusztajn, J,K., Freese. A. and Wurtman, R.J.
(1987) Biochem. J. 241, 81-86, 19 Liscovitch. M, (1989) J, Biol. Chem, 264, 1450-1456. 20 Pal, J.K,, Siegel, M.I,, Egan, R.W. and Billah, M.M. (1988) J. Biol. Chem. 263, 12472-12477. 21 Muir, J,G, and Murray, A.W. (1987) J. Cell, Physiol. 130, 382-391. 22 Price, B.D,, Morris, J.D.H. and Hall, A. (1989) Biochem. J. 264, 509-515, 23 Martin, T.W. and Michaelis, K. (1989) J. Biol. Chem. 264, 88478856, 24 Domino, S.E., Bocckino, S.B. and Garbers, D.L. (1989) J. Biol. Chem. 264, 9412-9419.
25 Parker, P.J., Kour, G., Marais, R.M., Mitchell, F., Pears, C., Schaap, D., Stabel, S. and Webster, C. (1989) Mol. Cell. Endocrinol. 65, 1-11. 26 Putney, J.W. Jr., Weiss, S.J., Van de Walle, C.M. and Haddas, R.A. (1980) Nature (London) 284, 345-347. 27 Moolenaar, W.E., Kruijer, W., Tilley, B.C., Verlaan, 1., Bierman, A.J. and De Latt, S.W. (1986) Nature (London) 323, 171-173. 28 Van Corven, E.J., Groanink, A., Jalink, K., Eichholtz, T. and Moolenaar, W.H. (1989) Cell 59, 45-54. 29 Takuwa, N., Takuwa, Y. and Rasmussen, H. (1987) Biochem. J. 243, 647-653. 30 Schaap, D. and Parker, P.J. (1990) J. Biol. Chem. 265, 7301-7307. 31 Warden, C.H. and Friedkin, M. (1984) Biochim. Biophys. Acta 792, 270-280. 32 Warden, C.H. and Friedkin, M. (1985) J. Biol. Chem. 260, 60066011. 33 Castagna, M., Takai, Y., Sano, K., Kikkaw, U. and Nishizuka, Y. (1982) J. Biol. Chem. 257, 7847-7851. 34 Brown, K.D., Littlewood, C.J. and Blakeley, D.M. (1990) Biochem. J. 270, 557-560. 35 DeLean, A., Munson, P.J.and Rodbard, D. 0980) Am. J. Physiol. 235, E97-EI02. 36 Pessin, M.S. and Raben, D.M. (1989) J. Biol. Chem. 264, 87298738. 37 Epand, R.M. and Stafford, A.R. (1990) Biochem. Biophys, Res. Commun. 171,487-490. 38 Tsai, M.H., Yu, C.L. and Stacey, D.W. (1989) Science 243, 522526. 39 Bonser, R.W., Thomson, N.T., Randall, R.W. and Garland, L.G. (1989) Biochem. J. 264, 617-620.
265
BBAMCR 12908
Hydrolysis of phosphatidylcholine by phospholipase D is a common response to mitogens which stimulate inositol lipid hydrolysis in Swiss 3T3 fibroblasts S i m o n J. C o o k a n d M i c h a e l J.O. W a k e l a m Molecular Pharmacology Group, Institute of Biochemistry, University of Glasgow, Glasgow ( U.K.) (Received 12 September 1990)
Key words: Phospholipase D; Mitogen; Inositol lipid hydrolysis; Second messenger; (Swiss 3T3 fibroblasts)
The stimulated hydrolysis of inositol ]ipids and phosphatidylcholine (PtdCho) by bombesin, [ArgS]vasopressin ([ArgSlVp) and prostaglandin F2a (PGF2~) was analy~d in Swiss 3T3 cells pre-labeiled to isotopic equilibrium with either [methyl.3H]choline, myo.[2.3H]inositol or [9,10 (n)-3H]palmitic acid. AH three agonists activated the phospholipase D-catalysed hydrolysis of PtdCho as determined by the release of [3H]choline (Cho) and the formation of [3H]phosphatidylbutanol (PtdBut). The release of [3H]choline by each agonist exhibited similar sensitivity to prolonged pre-exposure to the phorbol ester 12-O-tetradecanoyiphorbol-13-acetate (I,~A). The release of [3H]choline exhibited the same dose dependency as the production of total inositoi phosphates for each mitogen suggesting that the two responses might be mediated through identical receptors. Acute pre-treatment with TPA allowed the dissociation of inositol lipid hydrolysis from PtdCho breakdown, since it inhibited inositol phosphate accumulation but stimulated choline generation. The loss of mitogen stimu|ated choline release in cells pre-treated with the phorbol ester for 48 h was not due to loss of stimulated inositoi phosphate production which was re?roducibly enhanced in these 'down-regulated' cells.
Introduction The rapid hydrolysis of phosphatidylinositol 4,5-bisphosphate (Ptdlns(4,5)P2) to generate the second messengers inositol-l,4,5-trisphosphate (Ins(1,4,5)P3 and
Abbreviations: [ArgSlVp, [arglnineS]vasopressin; Bomb, bombesin; Cho, choline; ChoP, phosphocholine; bAG, sn-l,2-diacylglycerol; DMBH, serum-free DMEM containing 20 mM Hepes (pH 7.4) and 1% (w/v) bovine serum albumin; DMEM, Dulbecco's modified Eagle's medium; DMSO, dimethyl sulphoxide; ECs0, concentration giving half of maximum response; GroPCho, glycerophosphocholine; HBG, Hanks' buffered saline solution containing 10 mM glucose and 1% (w/v) bovine serum albumin (pH 7.4); HBG.Li, HBG containing 10 mM LiCI; IC50, concentration giving half maximal inhibition of response; lns(1,4,5)Pa, inositol-l,4,5-trisphosphate; lnsPt, total inositol phosphates; lnsP, InsP2 and InsP3, total inositol mono-, hisand trisphosphates; PGF2o, prostaglandin F2a; PIC, phosphoinositidase C; PLC, phospholipase C; PLD, phospholipase D; PtdBut, phosphatidylbutanol; PtdCho, phosphatidylcholine; PtdOH, phosphatidic acid; TPA, 12-O-tetradecanoylphorbol-13-acetate. Correspondence: M.J.O. Wakelam, Molecular Pharmacology Group, Institute of Biochemistry, University of Glasgow, Glasgow, G12 8QQ, U.K.
sn-l,2-diacylglycerol (DAG) is now accepted as the means by which a number of mitogenic growth factors initiate a cascade of intracellular signals leading to exit from Go/G t, DNA synthesis and cell division (for review see Ref. 1). However, a number of recent reports have suggested that lipids other than the phospho;nositides may be in~oived in signalling pathways leading to the generation of DAG (reviewed in Refs. 2, 3). These include observations that mitogen-stimulated elevation of DAG is sustained long after the increase in Ins(1,4,5)Pa has declined [4,5], and that some growth factors appear to elevate DAG without concomitant hydrolysis of inositol lipids [6]. A major candidate for this alternative source of DAG is stimulated hydrolysis of phosphatidylcholine (PtdCho) which has now been reported in a variety of mitogenic and non-mitogenic sytems [2,3]. PtdCho hydrolysis has been proposed to occur by both phospholipase C and D catalysed processes [7,8], the former generating phosphorylcholine (ChoP) and DAG, the latter choline (Cho) and phosphatidic acid (PtdOH) which may then be converted to DAG by the action of phosphatidic acid phosphohydrolase [9].
266 We have previously shown that in Swiss 3T3 cells pre-labelled with [all]choline both bombesin, a potent mitogenic peptide, and the tumour promoting phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) stimulate the hydrolysis of PtdCho, apparently by a protein kinase C-dependent, phospholipase D catalysed mechanism [10]. Furthermore, we have shown that bombesin stimulates a biphasic increase in DAG mass; the first corresponding precisely with an increase in lns(1A,5)P3, whilst the second is sustained and divorced from any further elevation of Ins(1,4,5)P3 [51. The kinetics of bombesin-stimulated PtdCho hydrolysis and choline release are entirely consistent with it being the source of the second phase of DAG elevation. To determine if this effect upon PtdCho hydrolysis was a general response to mitogenic stimuli we have extended our studies to other peptide and non-peptide factors reported to be mitogenic in the Swiss 3T3 cell line. We report here that bombesin, [ArgSlVp and PGP2, all stimulate the hydrolysis of PtdCho by similar mechanisms which suggest a requirement for prior inositol lipid hydrolysis. Furthermore, we confirm that this release of choline represents activation of PLD by demonstrating formation of PtdBut, catalysed by the transferase activity of PLD, for each agonist. Materials and Methods
Cell cui:,re Swiss 3T3 cells were cultured in DMEM containing 10% calf serum in 24-weU plates and allowed to become quiescent by serum depletion and density arrest as described previously [101. Labelling of cells to isotopic equilibrium was achieved by allowing them to quiesce in medium containing either 1.5/~Ci ml-i [3H]choline or 1 /~Ci ml-~ myo.[3H]inositol and 1% calf serum for 48 h the latter utilising inositol.free DMEM. For assay of PtdBut formation Swiss 3T3 cells were grown in 6 well plates until 80% confluent; medium was then replaced with 2 ml of DMEM containing 2% calf serum and 2 ~Ci ml -I [9,10 (n)-3Hlpalmitic acid for 48 h. For experiments in which protein kinase C was 'down-regulated' cells were treated for 48 h during the period of isotopic labelling with 400 nM TPA or with the vehicle control, DMSO ( < 0.1% final concentration), or 400 nM p-phorbol. Such treatment has been extensively shown to deplete all cellular protein kinase C activity in Swiss 3T3 cells [17,34] and in our own laboratory to abolish [3H]phorbol-12,13-dibutyrate binding to Swiss 3T3 monolayer cultures (Carrie, S. and Wakelam, M.J.O., unpublished results).
Measurement of total inositol phosphates (InsPD Quiescent Swiss 3T3 cells labelled as above were washed in Hanks" buffered saline solution (pH 7.4) containing 10 mM glucose and 1% (w/v) bovine serum
albumin (HBG) for 5 min at 37°C. They were then washed in HBG containing 10 mM LiCI (HBG.Li) for a further 15 min at 37°C prior to treatment with the appropriate agent. In some experiments cells were pretreated during this 15-min period with varying concentrations of TPA during which the concentration of vehicle (DMSO) did not exceed 0.1% (v/v), which control experiments showed was without effect upon the parameters studied. After this pre-incubation period, cells were treated with the given concentration of mitogen for the times indicated in HBG.Li at 37°C. Incubations were terminated by addition of 0.5 ml of ice-cold methanol and the cell debris was scraped into plastic screw top poly-tubes. Following a second wash with 0.2 ml of methanol, the samples were extracted with 0.31 ml of chloroform for 1 h on ice. Phases were resolved by addition of 0.5 ml of distilled water and after centrifugation at 1 4 0 0 0 x g for 3 rain, 1 ml of the upper aqueous/methanolic phase was taken into 5 ml plastic insert vials. Total inositol phosphates in this aliquot were assayed as described previously [11,15].
Measurement of phosphatidylcholine hydrolysis products Quiescent Swiss 3T3 cells labelled with [3H] choline as above were washed at 37°C with Hanks' buffered salt solution and HBG as described previously. [10]. Cells were stimulated with the appropriate concentration of agonist for the times indicated in HBG at 37 °C and incubations were terminated by addition of 0.5 ml of ice-cold methanol without aspiration of the incubation medium. After scraping, washing with methanol and scraping again as described above, samples were extracted with 0.31 ml of chloroform for 1 h on ice. Following resolution of phases as above, aqueous, methanolic samples were applied to 1 ml Dowex-50WH + ion exchange columns prepared in glass wool plugged pasteur pipettes, and glycerophosphorylcholine (GroPCho), phosphocholine (ChoP) and choline (Cho) were eluted as described previously [10]. Aliquots from each fraction were analysed for radioactivity by liquid scintillation counting in Optiphase Hi-safe scintillant. An aliquot of the lipid phase was also counted for radioactivity allowing normalisation for variations in sample size.
Assay of PtdBut formation [9,10 (n)-3H]Palmitate labelled Swiss 3T3 cells were washed in DMBH for 30 rain at 37°C before incubating with DMBH containing agonist in the presence or absence of 30 mM butan-l-ol (0.3%, v/v) for 10 rain. Reactions were terminated by aspiration and addition of 0.7 ml ice-cold methanol and cell debris was harvested into glass vials. Following extraction of phospholipids by addition of 0.7 ml of chloroform phases were resolved by addition of 0.585 ml of water. The lower
267 chloroform phase was dried down in vacuo, re-dissolved in 75/tl of chloroform/methanol (19/1, v/v) and applied to Whatman LK5DF t.l.c, plates which were than fully developed in the solvent system B described by Bonser et al. [39]. PtdBut, identified by its co-migration with a [~4C]PtdBut standard and novel appearance in the presence of butan-l-ol, was excised from the t.l.c. plate and the associated radioactivity determined by scintillation counting. In each case the results shown represent the mean + S.D. for a single experiment typical of three or four, where for each experimental point n - 3. Significance was estimated using an unpaired Student's t-test, whilst ECs0 values were calculated from dose-response curves fitted to a logistic equation as defined in Ref. 35. Bombesin and [ArgS]Vp were obtained from Cambridge Research Biochemicals, Cambridge, U.K. TPA, PGF2o and Dowex resins were from Sigma, Poole, Dorset, U.K. myo-[2-~H]inositol, [methyl-3H]choline chloride and [9,10 (n)-3Hlpalmitic acid were from Amersham International, Amersham, U.K. All other chemicals were of the highest grade commercially available as reported previously [5,10,11,15]. Results
We have previously shown that bombesin stimulates the rapid hydrolysis of PtdCho in Swiss 3T3 cells and suggested that this was catalysed by a phospholipase D on the basis that Cho generation was always observed before any rise in Chop [10]. Experiments were performed to see if [Arga]Vp and PGF2~, both mitogenic for Swiss 3T3 cells [12,13] and both simulating the hydrolysis of inositol lipids [14,11], could also stimulate PtdCho hydrolysis. Fig. 1 shows the time-course of increase in [3H]Cho generation in Swiss 3T3 cells stimulated with [ArgS]Vp. The peptide caused a rapid elevation of [3H]Cho which was maximally above the control at 5 min and sustained for the duration of the time-course up to 15 min (Fig. 1) and 30 rain (data not shown). Changes in the radioactivity associated with the ChoP fractions were less reproducible, occurred over a longer time scale and were smaller in fold terms (data not shown). No significant changes in the radioactivity associated with the GroPCho fraction were observed (data not shown). These results, and particularly the kinetics of Cho elevation, are very similar to those obtained in bombesinstimulated Swiss 3T3 cells [10] and also in other systems recently reported [23,24]. Elevation of choline levels was also observed using 1/~M PGF2,~ (results not shown). Experiments were performed to investigate the dosedependency of the effects of bombesin, [ArgS]Vp and PGF2~ upon choline levels (Fig. 2). The results show that all three agonists stimulated increases in Cho in a dose-dependent manner with ECs0 values of 1.44 4- 1.02
5°t
4500
4000" 3500'
E
3000' 2500' 2000' 1500
.
.
.
.
i
0
.
.
.
.
|
5
.
.
.
.
10
i
15
Time (min) Fig. I. Time-course of [ArgS]Vp-stimulated increases in Cho. [3H]choline labelled Swiss 3T3 cells were stimulated with 100 nM [ArgS]Vp in HBG (m) or HBG alone (El) for the times indicated. Following termination of the reactions [3H]Cho was isolated by ion-exchange chromatography as described in Materials and Methods. The results are means+_S.D (n = 3) from a single experiment representative of three.
nM (mean of four experiments) for bombesin (data not shown, see Ref. 10), 0.92 4-0.57 nM (mean of three experiments) for [ArgS]Vp and 0.51 4- 0.32 #M (mean of three experiments) for PGF2~. The elevation of Cho was generally greatest with bombesin lower with [Arg~:]Vp and smallest with PGF2o. 200 180
ii
160' 140' 120' 100'
80
-13
i
|
- 11
-9
-7
i
[Mitogen]
M
-5
-3
Fig. 2. Dose-dependence of [ArgS]Vp and PGF2,~-stimulated [3H]Cho generation. [3H]Choline labelled Swiss 3T3 cells were stimulated with HBG containing [ArgS]Vp (m) or PGF2a (rl) at the final concentrations indicated for 5 min. Reactions were terminated and the radioactivity associated with the Cho fraction determined as described in Materials and Methods section. The results are means+ S.D. (n = 3) from a single experiment representative of three and are expressed as a percentage of the control. The control values were 3290+ 151 dpm for [ArgS]Vp and 5992+368 dpm for PGF2,~.
268 TABLE 1
800
Stimulated [~H]phosphatidylbutanol formation in Swiss 3T3 cells [3H]Palmitate-labelled Swiss 3T3 cells were stimulated with either bombesin (100 nM), [ArgS]Vp (100 nM) or PGF2a (2 pM) for 10 rain in the presence of 0.3~ (v/v) butan-l-ol, [3H]PtdBut was resolved by t.l.c, as described in Materials and Methods. Results are means :t: S.D. (n = 3) from a single representative experiment.
600'
E~ •_
Control Bombesin [ArgS]Vp PGF2~
2OO
-12
-14
.10
.s
- butan-l-ol
+ butan-l-ol
658 + 682+ 614+ 618+
933 + 255 6320+ 350 5847+1008 4788+ 357
125 77 85 131
-6
bombesin, [Arga]Vp or PGF2, resulted in the formation of PtdBut only in the presence of butan-l-ol (Table I). To compare the dose-dependency of increases in Cho with inositol lipid breakdown, dose response curves for the production of total inositol phosphates were performed. The results shown in Fig. 3 (a and b) show that the three agonists stimulated the accumulation of total inositol phosphates with a very similar dose-dependency to the corresponding Cho response (ECs0 values: 0.58 + 0.34 nM (mean of three experiments) for bombesin, 0.48 + 0.15 nM (mean of three experiments) for [ArgS]Vp and 0.16 + 0.14 pM (mean of four experiments) for PGF2~). In addition, the accumulation of
[Bemhesin] nM
500
400
300
m ~ mg=
D.p.m. in [3H]PtdBut
Incubation
~tm
t.
200 E~
100
30000' 0
- 13
i
!
i
-i I
-9
-7
w
-5
-3
[Mitogen] M
Fig. 3. (a) Dose-dependence of bombesin-stimulated InsPt accumulation with or without acute TPA pre-treatment. [3Hllnositol-labelled Swiss 3T3 cells were washed with HBG.Li containing 400 nM TPA (ra) or DMSO (11) as described in Materials and Methods before stimulating with the given concentration of bombesin in HBG.Li for 20 rain. (b) Dose-dependence of [Ar88]Vp and PGF2~-stimulated InsP, accumulation. [3Hllnositol labelled Swiss 3T3 cells were washed in HBG.Li befoi.~, stimulating with HBG.Li containing the given final concentrations of [ArgSlVp (11) or PGF2,, (ra) for 20 rain. In all cases InsPt were assayed as described in Materials and Methods. The results, expressed as percentage of control to allow comparisons, are means+S.D (n ffi 3) of a single experiment representative of three. Control values were 2716+93 dpm for bombesin, 2586:i:400 dpm for [ArsSlVp and 4997 + 1072 dpm for PGF2,~.
To confirm that this release of Cho did represent PLD-catalysed PtdCho hydrolysis, the ability of all three agonists to form [3H]PtdBut in [3H]palmitatelabelled cells in the presence of butan-l-ol was determined. In Swiss 3T3 cells the majority of [3H]palmitare incorporated into phospholipids is found in PtdCho (53~$) with negligible labelling of PtdIns (< 3~) (data not shown). Under these conditions stimulation with
25000'
.ffi
20000'
E eL
15000 ~
10000
.......
0.001
I
0,01
........
|
O. 1
........
!
........
I
g
I0
........
|
I00
.......
1(300
[TPA] nM Fig. 4. Dose-dependence of TPA-mediated inhibition of [ArgS]Vp stimulated InsPt accumulation. [3Hllnositol labelled Swiss 3T3 cells were washed for 15 rain in HBG.Li containing the indicated final concentrations of TPA (I) before stimulating with 100 nM [ArgS]Vp for 20 min. InsPt accumulation was assayed as described. The results are means+S.D. ( n = 3 ) of a single experiment representative of three. In this particular example the unstimnlated level of InsPt was 2566+829 dpm and 100 nM TPA gave a maximum 57% inhibition of the response to [ArgS]Vp. Similar dose response curves for the effects of TPA upon bombesin and PGF2,rstimulated lnsPt accumulation were also obtained (data not shown).
269 total InsP, by each agonist was inhibited by approx. 50~ following a 15 min preincubation with TPA (Fig. 3a and Fig. 4). In the case of bombesin and [ArgS]vasopressin this pre-treatment resulted in a shift to the right in the dose-response curve for InsP t accumulation giving new ECso values of 4.4 +_ 4.5 nM and 1.12 _ 0.52 nM, respectively. This TPA-mediated inhibition of inositol phosphate production was both time- and dose-dependent. The onset of TPA-stimulated inhibition of a subsequent bombesin-stimulated inositol phosphate accumulation occurred with a short lag time of approx. 1-1.5 min after addition of the phorbol ester (data not shown). Fig. 4. shows the dose-response curve for the TPAstimulated inhibition of [ArgS]Vp InsPt accumulation. Similar dose-response curves were obtained for the inhibition of bombesin and PGF2o stimulated InsPt accumulation (data not shown) and the ICs0 values for TPA were 5.58 + 3.84 nM against bombesin, 2.36 + 1.68 nM against [Arga]Vp and 2.16 _+0.52 nM against PGF2o. In cells treated with 400 nM TPA for 48 h, a treatment which down-regulates protein kinase C activity [17], the production of inositol phosphates in response to each agonist was reproducibly and significantly enhanced (Fig. 5) (the significance of enhancement was for bombesin P = 0.005, [Arga]Vp, P < 0.001 and for PGF2~, P < 0.001). Interestingly, the degree of enhancement of InsPt accumulation in 'down-regulated' cells varied according to the mitogen used. Thus, the enhancement of responses over the control cells was typically 1.2-1.4-fold for bombesin, 2.2-2.5-fold for [ArgS]Vp and 2.5-3.2-fold for PGF2~. Furthermore, in
160000 •
120000 •
"=" =[
4 8OOOO' OOOO'
0r-m-Control
r-I Bb
AVp
PGF2¢x
TPA
Fig. 5. Effect of chronic TPA pro-treatment upon agonist stimulated InsPt accumulation. Swiss 3T3 cells were labelled with [3H]inositol for 48 h in the presence of 400 nM TPA (hatched bars) or DMSO (open bars). After washing in HBG.Li the cells were stimulated with either a control buffer (HBG.Li), 100 nM bombesin, 100 nM [ArgS]Vp, 1 ~M PGF2, or 100 nM TPA in HBG.Li as indicated for 20 rain. InsPt was assayed as described. The results are means± S.D. (n = 3) for a single experiment representative of three.
12000"
80oo
6OOO
4OOO Con
TPA
Bomb
AVp
PGF2¢z
Fig. 6. Effect of chronic pre-treatment with TPA upon agonist-stimulated [3H]Cho generation. Swiss 3T3 cells were labelled with [3H]choline for 48 h in the presence of 400 nM TPA (hatched bars) or DMSO (open bars). After washing in HBG, cells were stimulated with a control buffer (HBG), 100 nM TPA, 100 nM bombesin, 100 nM [ArgS]Vp or 1/~M PGF2,, in HBG for 5 rain. The amount of [3H]Cho generated was normalised to the labelling in total choline-containing lipids. Results are means +_S.D. (n = 3 ) from a single experiment representative of four.
'down-regulated' Swiss 3T3 cells acute pre-treatment with TPA no longer inhibited InsPt accumulation in response to bombesin, [ArgS]vasopressin or PGF2,. For example, in 'down-regulated' cells InsPt accumulation after 20 min stimulation in the presence of 10 mM LiCI was 80451 + 2716 dpm for bombesin, 80112 + 3134 dpm for [ArgS]vasopressin and 60255 + 1703 dpm for PGF2~, whilst in "down-regulated" cells pre-treated with 400 nM TPA the values were 82619 + 3334 dpm, 83002 + 2620 dpm and 61 603 + 4482 dpm, respectively; the basal dpm was 9436 + 488 (values from a single, representative experiment). We have previously reported that long term treatment of Swiss 3T3 cells with TPA, as described above, results in the abolition of bombesin- and TPA-stimulated generation of choline [10]. In Fig. 6 we confirm these findings and also show that the ability of both [ArgS]Vp and PGF2~ to stimulate Cho release is also greatly reduced in 'down-regulated' cells. In 'control' cells agonist-stimulated Cho elevation was significantly above the buffer control ( P < 0.05) and the corresponding 'down-regulated' response ( P < 0.02) in all cases, whilst there was no significant elevation of Cho in 'down-regulated' cells. When sub-maximal concentrations of bombesin and [ArgS]vasopressin are combined the elevation [3Hlcholine is additive rather than synergistic when compared to the two agonists alone, whilst combination of maximal doses of bombesin and [Arga]vasopressin or TPA does not enhance the reponse of either agent alone (data not shown).
270
Discussion Agonist-stimulated hydrolysis of phosphatidylcholine has now been reported in a variety of call types including NG-108 cells [18,19], neutrophils [20], Swiss 3T3 fibroblasts [10,21,22], pulmonary endothelial cells [231. and spermatazoa stimulated to undergo the acrosome reaction by a fucose sulphate glycoconjugate [24]. Agonist-stimulated turnover of PtdCho is thought to have a role in providing an alternative source of DAG in response to extracellular stimuli and a PtdCh," ~.3"cle analagous to the inositol lipid cycle has been proposed [3]. Both phospholipase D and phospholipase C catalysed pathways of PtdCho hydrolysis have been reported [20,71 and the products of each reaction seem to have potent biological effects. Thus, DAG is known to be the endogenous activator of protein kinase C (reviewed in Ref. 25) and PtdOH has been reported to be a growth factor [27] and a precursor of DAG [9]. In Swiss 3T3 cells stimulated with bombesin or TPA, a protein kinase C-dependent, PtdCho phospholipase D activity is stimulated [10] and it has been suggested that this may contribute to the sustained elevation of DAG mass in Swiss 3T3 cells stimulated by the mitogenic peptide bombesin [5]. Here we show that this phenomenon is not peculiar to bombesin but also occurs in response to stimulation by two other calcium-mobilizing, mitogenic agents, the peptide [ArgSlVp and the eicosanoid PGF2,. Like bombesin, [ArgS]Vp stimulated the rapid generation of [3H]Cho from cells labelled with [3H]choline prior to any change in [3H]ChoP, suggesting activation of a phospholipase D activity (Fig. 1). The slower and relatively modest increases in ChoP (in fold terms) may represent a phospholipase C activity, but may also be explained in terms of phosphorylation of the liberated choline by choline kinase [31,32]. The elevation of choline by all three mitogens was dose-dependent (Fig. 2) and the ECs0 values are identical to those for the stimulated accumulation of total inositol phosphates in each case (Fig. 3). This suggests that for each mitogen the two responses are stimulated through common receptors. Furthermore, the ECs0 values for both [ArgSlVp and PGF2° were of a similar order to those quoted for stimulation of DNA synthesis by each agonist [12,13]. The results in Table I confirm that the release of Cho by the three agonists is associated with the activation of PLD since all three stimulated the formation of Ptdt6ut, in the presence of butan-l-ol, catalysed by the transferase activity of PLD [20]. The much greater magnitude of this response compared with Cho release reflects the virtual absence of PtdBut in control cells. This indicates that PLD activity is very low in the absence of agonist. The ability of all three mitogens to stimulate accumu-
lation of total inositol phosphates exhibited similar sensitivities to pre-treatment with TPA (Figs. 3a and 4). Thus, a 15 min pre-treatment with TPA resulted in a dose-dependent inhibition of a subsequent response to bombesin (Fig. 3a), [ArgS]Vp (Fig. 4) or PGF2, (data not shown). The IC5o for this effect of TPA was similar for all three receptor agonists (2-6 nM); this value is of a similar order to that previously quoted for inhibition of bombesin-stimulated inositol phosphate accumulation [16], activation of C-kinase [33] and the generation of choline by TPA in Swiss 3T3 cells [10]. In addition, the TPA-mediated inhibition of agonist-stimulated InsPt accumulation was also time-dependent with a lag of approx. 1 rain from addition of TPA to the onset of inhibition. We have previously observed a similar lag in the time-course of TPA stimulated increases in choline levels [10] and this, together with the similarities in dose-dependency of the TPA effects, clearly suggests that activation of protein kinase C, inhibition of agonist stimulated inositol phosphate accumulation and activation of PtdCho hydrolysis are related. Indeed, acute treatment with TPA effectively dissociated choline generation from inositol phosphate accumulation since it stimulated the former at the same concentration as it inhibited the latter. In this respect TPA is also able to dissociate DAG generation from inositol lipid hydrolysis, since addition of the phorbol ester to Swiss 3T3 cells results in an elevation of DAG mass in the absence of InsP t accumulation but with similar kinetics to [3H]choline release (Cook, S.J. and Wakelam, M.J.O., unpublished results, Ref. 29). To further investigate the relationship between inositol phosphate accumulation, choline release and protein kinase C, the enzyme was 'down-regnlated' by pre-treating Swiss 3T3 cells with 400 nM TPA for 48 h [17]. In agreement with the results of Brown et ai. [16] we found that under such conditions bombesin-stimulated accumulation of lnsP t was no longer inhibited by a short pre-treatment with TPA and that responses were reproducibly enhanced (Fig. 5). Similar results were obtained for [ArgSlvasopressin and PGF2, and are consistent with the hypothesis that protein kinase C activates a feedback inhibitory pathway to limit inositol lipid hydrolysis, presumably through the production of DAG [16]. The degree of enhancement of InsPt accumulation in 'down-regulated' cells was different for each agonist, with PGF2, always giving the greatest enhancement, then [ArgS]Vp and bombesin always the smallest. Similar results were recently reported by Brown et al. [34] and may represent subtle differences in the C-kinase sensitivity and perhaps mechanism of coupling of the receptors to PIC. However, a simpler alternative interpretation may lie in the fact that in 'control' cells PGF2,, gives the smallest accumulation of InsPt (Fig. 3b) and the smallest depletion of [3H]inositol lipids over a 20 min stimulation in the presence of 10 mM LiCI. Re-
271 moval of C-kinase mediated-inhibition by 'down-regulation' would then allow large potentiation of the response. Such a mechanism would operate for [ArgS]Vp and bombesin, but since they give successively greater responses in 'control" cells and much greater depletion of [3H]incqtol lipids, potentiation of InsPt accumulation in 'down-regulated' cells may be limited by the amount of [3H]inositol lipids available as substrate for PIC (data from a typical experiment: dpm in total inositol lipids; control 20 min = 133714 + 8774 dpm, bombesin 20 rain = 70120 + 3239 dpm, [ArgS]Vp 20 min = 107748 + 10859, PGF2~ 20 min = 136826 + 8123 dpm). In protein kinase C-'down-regulated' cells the ability of TPA and bombesin to stimulate choline release was severely attenuated (Fig. 6) as previously reported [101. Furthermore, the ability of [ArgS]vasopressin and PGF2~ to stimulate choline release was also abolished in the down-regulated cells indicating a similar mechanism for activation of PtdCho hydrolysis, presumably in:olving protein kinase C activation. Since in the 'down-regulated' cells stimulated InsPt accumulation was enhanced (Fig. 5) the abolition of mitogen-stimulated PtdCho hydrolysis was not due to loss of inositol lipid hydrolysis and was thus distal to this event. A common mechanism of activation of PtdCho hydrolysis is supported by the finding that maximal concentrations of bombesin and [ArgS]Vp do not stimulate Cho generation in an additive manner, whilst sub-maximal concentrations are indeed additive. Our results differ from those of Price et al. [22] who were unable to observe PGF2,-stimulated PtdCho hydrolysis in Swiss 3T3 cells. We can only speculate that this represents variation between different clones of Swiss 3T3 cells, a proposal supported by the fact that Price et al. were only able to observe a very small accumulation of InsPt in response to PGFz,~ (less than 2-fold). Indeed, this may explain why they were unable to see stimulated PtdCho hydrolysis since such a small inositol lipid signal would result in a much smaller C-kinase activation. If protein kinase C is a major link between inositol lipid hydrolysis and PtdCho hydrolysis, the magnitude of PKC activtion will be a major determinant of the magnitude of PtdCho hydrolysis. Such a model is supported by the observation that the clear rank order of InsPt accumulation (bombesin > [ArgS]Vp > PGF2,~) is broadly reflected in the rank order of Cho generation (bombesin > [ArgS]Vp > PGF2,) and to a lesser extent PtdBut formation (Table I). Based on the observations reported here and previously by us [5,10] and others [21,22] we suggest that in Swiss 3T3 cells those mitogens which stimulate the rapid hydrolysis of inositol lipids may also stimulate hydrolysis of PtdCho by a C-kinase mediated, phospholipase D-catalysed mechanism. Activation of PLD will result in the formation of PtdOH. Early reports of a
Ca2+-ionophore role for PtdOH [26] have remained unproven but it is clear that both PtdOH [2'7] and lyso-PtdOH [28] are potent mitogens when added exogenously to cultured call lines. Indeed, PtdOH and lysoPtdOH exhibit many of the characteristics of classical receptor-coupled mitogens causing hydrolysis of Ptdlns(4,5)P 2 (Plevin, R., MacNulty, E., Palmer, S. and Wakelam, M.J.O., unpublished results) inhibition of adenylate cyclase activity, calciuni mobilization and DNA synthesis [28]. In addition to agonist-like properties, PtdOH also exhibits second messenger-like properties including replacing PtdSer in supporting Ca2+-independent PKC activity [37]. Furthermore, arachidonoylcontaining PtdOH is able to inhibit the catalytic activity of the p21 r~s GTPase activating protein (GAP) in vitro resulting in the accumulation of the active GTP-bound form of p2V a~ [38] which is a potent mitogenic signal. Alternatively, a functional link between inositol lipid hydrolysis and PtdCho hydrolysis may serve to provide a sustained production of DAG [5] as part of a long-term mitogenic signal when inositol lipid hydrolysis has been rapidly desensitized [5,16], Ins(1,4,5)P3 levels have declined and intraeellular Ca 2+ levels have returned towards control levels. Since the DAG produced from hydrolysis of PtdCho may have a different fatty acid composition compared to that from Ptdlns(4,5)P2 [36] and will be produced in the absence of elevated intracellular calcium concentrations, the recent description of protein kinase C-~ [30], a novel calcium-independent PKC isozyme, may be of relevance to these observations. Since for each mitogen the generation of choline exhibited the same dose-dependency as inositol phosphate accumulation and DNA synthesis in Swiss 3T3 cells it is possible that it is a normal component of the mitogenic signal resulting in sustained elevation of DAG and activation of perhaps novel PKC species.
Acknowledgements This study was supported by grants from the Medical Research Council (U.K.) and the Cancer Research Campaign. In addition, S.J.C. received an S.E.R.C. CASE award studentship in association with the Wellcome Foundation Ltd.
References 1 Berridge, M.J. (1987) Biochim. Biophys. Acta 907, 33-45. 2 Exton, J.H. (1990) J. Biol. Chem. 265, l-4. 3 Pelech, S.L. and Vance, D.E. (1989) Trends Biochem. Sci. 14, 28-30. 4 Wright, T.M., Rangan, L.A., Shin, H.S. and Raben, D.M. (1988) J. Biol. Chem. 263, 9374-9380. 5 Cook, S.J., Palmer, S., Plevin, R. and Wakelam, M.J.O. (1990) Biochem. J. 265, 617-620. 6 Wright, T.D., Shin, H.S. and Raben. D.M. (1990) Biochem. J. 267, 501-507.
272 7 Diaz-Meco, M.T., Larrodera, P., Lopez-Barahona, M., Cornet, M.E., Barreno, P.G. and Moscat, J. (1989) Biochem. J. 263, 115-120. 8 Billah, M.M., Pal, J.K., Mullmann, T.J., Egan, R.W. and Siegel, M.I. (1989) J. Biol. Chem. 264, 9069-9076. 9 Martin, T.W. (1988) Biochim. Biophys. Acta 962, 282-296. 10 Cook, SJ. and Wakelam, M.J.O. (1989) Biochem. J. 263, 581-587. 11 Black, F.M. and Wakelam, MJ.O. (1990) Biochem.J. 266, 661-667. 12 Corps, A.N., Rees, L.H. and Brown, K.D. (1985) Biochem. J. 231, 781-784, 13 Macphce, C.H., Drummond, A.H., Otto, A.M. and Jimenez de Asua, L, (1984) J. Cell. Physiol. 119, 35-40. 14 Nanberg, E, and Rozengurt, E. (1988) E M B O J. 7, 2741-2747. 15 PIevin,R,, Palmer, S., Gardner, S.D, and Wakelam, MJ.O. (1990) Biochem, J. 268, 605-610, 16 Brown, K.D,, Blakeley, D.M.. Hamon, M.H,, Laurie. M.S, and Corps, A.N, (1987) Biochem. J, 245, 631-639, 17 Rodriguez.Pena, A. and Rozengurt. E. (1984) Biochem. Biophys. ges, Comman. 120, 1053-1059. 18 Liscovitch, M., Blusztajn, J,K., Freese. A. and Wurtman, R.J.
(1987) Biochem. J. 241, 81-86, 19 Liscovitch. M, (1989) J, Biol. Chem, 264, 1450-1456. 20 Pal, J.K,, Siegel, M.I,, Egan, R.W. and Billah, M.M. (1988) J. Biol. Chem. 263, 12472-12477. 21 Muir, J,G, and Murray, A.W. (1987) J. Cell, Physiol. 130, 382-391. 22 Price, B.D,, Morris, J.D.H. and Hall, A. (1989) Biochem. J. 264, 509-515, 23 Martin, T.W. and Michaelis, K. (1989) J. Biol. Chem. 264, 88478856, 24 Domino, S.E., Bocckino, S.B. and Garbers, D.L. (1989) J. Biol. Chem. 264, 9412-9419.
25 Parker, P.J., Kour, G., Marais, R.M., Mitchell, F., Pears, C., Schaap, D., Stabel, S. and Webster, C. (1989) Mol. Cell. Endocrinol. 65, 1-11. 26 Putney, J.W. Jr., Weiss, S.J., Van de Walle, C.M. and Haddas, R.A. (1980) Nature (London) 284, 345-347. 27 Moolenaar, W.E., Kruijer, W., Tilley, B.C., Verlaan, 1., Bierman, A.J. and De Latt, S.W. (1986) Nature (London) 323, 171-173. 28 Van Corven, E.J., Groanink, A., Jalink, K., Eichholtz, T. and Moolenaar, W.H. (1989) Cell 59, 45-54. 29 Takuwa, N., Takuwa, Y. and Rasmussen, H. (1987) Biochem. J. 243, 647-653. 30 Schaap, D. and Parker, P.J. (1990) J. Biol. Chem. 265, 7301-7307. 31 Warden, C.H. and Friedkin, M. (1984) Biochim. Biophys. Acta 792, 270-280. 32 Warden, C.H. and Friedkin, M. (1985) J. Biol. Chem. 260, 60066011. 33 Castagna, M., Takai, Y., Sano, K., Kikkaw, U. and Nishizuka, Y. (1982) J. Biol. Chem. 257, 7847-7851. 34 Brown, K.D., Littlewood, C.J. and Blakeley, D.M. (1990) Biochem. J. 270, 557-560. 35 DeLean, A., Munson, P.J.and Rodbard, D. 0980) Am. J. Physiol. 235, E97-EI02. 36 Pessin, M.S. and Raben, D.M. (1989) J. Biol. Chem. 264, 87298738. 37 Epand, R.M. and Stafford, A.R. (1990) Biochem. Biophys, Res. Commun. 171,487-490. 38 Tsai, M.H., Yu, C.L. and Stacey, D.W. (1989) Science 243, 522526. 39 Bonser, R.W., Thomson, N.T., Randall, R.W. and Garland, L.G. (1989) Biochem. J. 264, 617-620.
