Naunyn-Schmiedeberg'sArch Pharmacol (1995) 351:194-201

© Springer-Verlag 1995

P~r Gerwins • Bertil B. Fredholm

Activation of phospholipase C and phospholipase D by stimulation of adenosine A1, bradykinin or P2u receptors does not correlate well with protein kinase C activation

Received: 5 April 1994/Accepted: 4 October 1994

Activation of adenosine AI-, bradykinin- or Pzu-receptors on DDT~ MF-2 smooth muscle cells all increased the formation of inositol 1,4,5-trisphosphate and the mobilization of intracellular calcium. All three types of agents could increase [Ca2+]i in the same cell. Activation of the Pzu receptor with ATP or UTP produced larger responses than activation of bradykinin- and adenosine Al-receptors, with bradykinin and N6-cyclopen tyladenosine. When agonist-stimulated levels of diacylglycerol were determined, all agonists caused biphasic changes of similar magnitudes. If anything, ATP and UTP tended to give larger increases in the second phase of stimulation. Phospholipase D, measured as the formation of phosphatidylethanol in cells labeled with [3H]palmitic acid and activated in the presence of ethanol, was activated similarly as phospholipase C, i.e. ATP or UTP caused the largest increase in phosphatidylethanol formation, followed by N6-cyclopentyladenosine and bradykinin which caused weaker responses. Activation of PLD by Pzu receptors was pertussis toxin insensitive. The activation of PLD by the agonists was only weakly affected by a PKC inhibitor, Ro 31-7549 (3-[l-(3aminopropanyl)-3-indolyl]-4-(1-methyl-3-indolyl)-1Hpyrrole-2,5-dione). In contrast, ATP or UTP did not activate protein kinase C, determined in a permeabilized cell assay using two specific protein kinase C substrates, whereas N6-cyclopentyladenosine and bradykinin caused a substantial activation. In summary, the present study shows that the magnitude of the activation of protein kinase C by receptor agonists cannot be predicted from the degree of phospholipase C and phospholipase D activation, accumulation of diacylglycerol or rise in intracellular calcium, and suggests that additional factors are important in the activation of protein kinase C. Abstract

P. Gerwins - B.B. Fredholm (~) Department of Physiologyand Pharmacology, Division of Pharmacology,KarolinskaInstitutet, S-17177 Stockholm, Sweden

D i a c y l g l y c e r o l • C a l c i u m • Protein kinase C subforms • Purinoeeptors • ATP • UTP Key words

Introduction Activation of phospholipase C (PLC) is a crucially important step in the signaling via several types of receptors (Berridge 1993; Nishizuka 1992). This enzyme hydrolyzes membrane phosphatidylinositolbisphosphate to inositol 1,4,5-trisphosphate (Ins(1,4,5)P3) that releases calcium from intracellular stores via specific receptors (Berridge 1993; Furuichi et al. 1989) and diacylglycerol that activates protein kinase C (PKC) (Nishizuka 1992). There is a family of PKC isoenzymes that mediates a variety of cellular responses (Nishizuka 1986; Stabel and Parker 1991). All members of the family are serine/threonine kinases, but they differ in tissue distribution (Kosaka et al. 1988), subcellular distribution and e.g. in calcium-dependence for activation (Nishizuka 1992; Stabel and Parker 1991). It is commonly believed that there is a close correlation between the activation of PLC and the activation of PKC. However, there are reports that in some systems activation of PLC, with increases in Ins(l,4,5)P3 and intracellular calcium, does not cause any activation of PKC (Gonzalez et al. 1989). Also, in some instances there is an accumulation of diacylglycerol without the formation of inositolphosphates and increases in intracellular calcium (Doglio et al. 1989; F~llman et al. 1989; Rosoff et al. 1988) in agreement with the observation that in some cells phosphatidylinositolbisphosphate is quantitatively a minor source for the formation of diacylglycerol (Augert et al. 1989). There are other pathways to generate diacylglycerol for PKC activation, e.g. through phospholipase D (PLD) (Billah and Anthes 1990; Cockeroft 1992; Exton 1990; Thompson et al. 1991). PLD splits membrane phosphatidylcholine into choline and phosphatidic acid, and phosphatidic acid can be converted to diacylglycerol by phosphatidic acid phosphohydrolase (Brindley 1984).

195 The m e c h a n i s m whereby agonists activate P L D is n o t k n o w n , b u t a direct activation via a n u n i d e n t i f i e d G-protein has been suggested, as well as actions via PKC or calcium (Billah a n d A n t h e s 1990; Cockcroft 1992; E x t o n 1990; T h o m p s o n et al. 1991). We have previously reported that pertussis toxin-sensitive a d e n o s i n e Aa receptors, partially pertussis toxinsensitive P2u receptors, a n d pertussis toxin-insensitive b r a d y k i n i n receptors activate P L C a n d increase intracellular levels of Ca 2+ (Gerwins a n d F r e d h o l m 1992a, b). The p u r p o s e of the present study was to investigate whether these different receptors, coupled to different G proteins, n o t only activated P L C b u t also P K C a n d PLD. P K C activity was d e t e r m i n e d in a permeabilized cell system (Heasly a n d J o h n s o n 1989) with a specific peptide substrate for P K C based o n a sequence from myelin basic protein (MBP4_~4) (Yasuda et al. 1990) or PKC-e pseudosubstrate peptide (peptide e) (Kazanietz et al. 1993). P L D activity was m e a s u r e d using [3H]palmitic acid-labeled cells that were activated in the presence o f ethanol. U n d e r these conditions, P L D catalyzes a specific reaction where the p h o s p h a t i d y l m o i e t y from p h o s p h a t i d y l choline is transferred to p r i m a r y alcohols, such as ethanol, a n d p h o s p h a t i d y l a l c o h o l s are formed (Dawson 1967; Kobayashi a n d Kanfer 1987; Yang et al. 1967). The surprising f i n d i n g was that A T P a n d UTP, despite being very active in activating p h o s p h o l i p a s e C a n d D, a n d in raising intracellular calcium a n d diacylglycerol, were u n a b l e to stimulate protein kinase C.

Materials and methods Cell culture media, fetal calf serum and cell culture flasks were from NordCell, (Bromma, Sweden). D-myo-[2-3H]inositol 1,4,5-trisphosphate (51.4 Ci/mmol) and [TYP]adenosine 5'-triphosphate (3000 Ci/mmol) were from Amersham and [9,10-3H(N)]palmitic acid (52.4 Ci/mmol) from DuPont Scandinavia AB, (Stockholm, Sweden). AcMBP 4_14 (Ac-QKRPSQRSKYL), a protein kinase C substrate peptide, was from Gibco BRL (Grand Island, NY, USA). Protein kinase C pseudosubstrate (RFARKGALRQKNVHEVKN) and protein kinase C-e pseudosubstrate peptide (peptide a: ERMRPRKRQGSVRRRV) were obtained from Bachem (Torrance, CA, USA). N-cyclopentyladenosine, phosphatidylserine, Triton X-100, imidazole, diethylenetriamine-pentaacetic acid, BAPTAAM, bradykinin, uridine 5'-triphosphate, Hank's balanced salt solution, Fura 2-AM, digitonin, fl-glycerophosphate and HEPES were all from Sigma (St Louis, MO, USA). Inositol 1,4,5-trisphosphate and adenosine 5'-triphosphate were purchased from Boehringer, Mannheim, Germany. 1,3-dipropyl-8-cyclopentyladenosine was from Research Biochemicals Inc. (Natick, MA, USA), and P-81 ion exchange chromatography paper from Whatman International, Ltd., (Maidstone). TLC plates (silica gel 60), ethylacetate, isooctane, acetic acid, chloroform and methanol were from Merck (Darmstadt, Germany). sn-l,2-diacylglycerol kinase from Escherichia coli was from Lipidex, Inc., (Westfield, N J, USA). Phosphatidylethanol, prepared as described (Eibl and Kovatchev 1981), was kindly provided bv Dr. Lena Gustavsson. De!oartment of Psvchiatrv and Neurochemistry, University of Lurid, Sweden. Ro 31-7549 (3-[1-(3aminopropanyl)-3-indolyl]-4-(1-methyl-3-indolyl)-lH-pyrrole-2,5dione) was a gift from Astra Draco (Lund, Sweden) and R 59022 (6- [2-(4-{[4-fluorophenyll-phenylmethylene}-i-piperidinyl)-ethyl]7-methyl-5H-thiazolo[3,2-a]pyrimidin-5-one) a gift from Janssen (Beerse, Belgium).

Cell culture. DDT I MF-2 smooth muscle cells, originally isolated from a steroid-induced leiomyosarcoma of Syrian hamster vas deferens (Norris et al. i974), were obtained from the American Type Culture Collection and grown in suspension as described (Gerwins and Fredholm 1995). Determination ofIns(1,4,5)Ps. Cells were treated and stimulated as described (Gerwins and Fredholm 1995) and Ins(l,4,5)P3 quantitated as described (Gerwins i993). Measurement of intracellular concentrations of free calcium. Cells were washed and resuspended in Hank's balanced salt solution (HBSS, 1.2 mM CaC12 supplemented with 0.1% BSA and 20 mM HEPES, pH 7.4) to a concentration of 10 6 cells/ml and loaded with 5 ~M Fura 2-AM for 40 rain at 37 °C. After the loading period, cells were washed twice in HBSS and resuspended to a concentration of 106 cells/ml. Prior to the measurements, cells were washed once more and then placed in a cuvette (106 cells in 2 ml of HBSS) and the intracellular calcium concentration determined at 30 °C in a dual-wavelength Sigma ZFP22 fluorometer by using the ratio of excitation wavelengths 334/366 nm with emission cut off at 500 nm. Free calcium concentration was calculated as previously described (Grynkiewicz et al. 1985). Measurement of intracellular concentrations of calcium in single cells. DDT 1 MF-2 cells were grown in the same medium as above on washed cover slips (25 mm ;~) for two days until 90% confluency was obtained. They were loaded with 2 ~tM Fura 2-AM (Sigma) for 45 rain and then washed three times in Krebs buffer. The cells were then incubated at room temperature in the same buffer for 25 rain before they were mounted in a temperature-controlled (37 °C) chamber, placed on the stage of a Zeiss Axiovert 35 microscope, and superfused with the same buffer at the rate of 0.5 ml/min. The cells were stimulated with ATP (0.1 mM), bradykinin (1 ~M) and N6-cyclopentyladenosine (1 - 100 nM). The fluorescence intensity was determined using a Zeiss microfluorometer. The aperture was set to cover a single cell. Intensities were recorded at 510 nm and the excitation wavelengths were 340 and 380 nm. One ratio was calculated every 2.5 s and displayed on a computer. The ratio values were converted to Ca 2+ levels essentially as described by Grynkiewicz et al. (1985). To calibrate, the following protocol was used: EGTA was added to a final concentration of 4 raM; thereafter i0 ~M ionomycin was added to achieve a zero [Ca2+]i value. 10 mM Ca2+ was added for I rain (this plateau value gives the max-value). Finally, 5 mM MnC12 was added to quench the Fura-2 fluorescence and to establish the level of autofluorescence. Determination of protein kinase C activity. PKC activity was measured as described (Heasly and Johnson 1989; see also Gerwins and Fredholm 1995) using a permeabilized cell assay and two synthetic peptide substrates based on the amino acid sequence 4 - 1 4 of myelin basic protein (MBP4_14) (Yasuda et al. 1990) and peptide e (Kazanietz et al. 1993). Analysis of phospholipase D activity. PLD activity was measured in DDT a MF-2 cells as the formation of [3H]patmitoyl phosphatidylethanol as described (Gerwins and Fredholm 1995). Diacylglycerol determinations. The procedure was exactly as described (Gerwins and Fredholm 1995) using a diacylglycerol kinase assay (Bonser and Thompson 1992). Data analysis. Dose-response curves were generated by using the GraphPad (ISI Software) program. Statistical comparisons between different drug treatments were made using Wilcoxon signed-rank test or Mann-Whitney two sample test. Data are expressed as mean_+SEM.

196

Results Activation of phospholipase C and the mobilization of intracellular calcium As previously reported (Gerwins and Fredholm 1992a, b) activation of adenosine A~ receptors with the selective ligand N6-cyclopentyladenosine (CPA), and bradykinin and Pzu receptors with their natural ligands bradykinin and ATP, increases intracellular Ca 2+ levels via the formation o f Ins(l,4,5)P 3 (Fig. 1). The level of intracellular Ca 2+ shows an initial, Ins(l,4,5)P3 mediated peak, followed by a sustained plateau phase due to influx of extracellular calcium (Gerwins and Fredholm 1992a, b). The response to adenosine A1 receptor activation was completely blocked by treatment of cells with pertussis toxin; the response to ATP was only partially sensitive and the bradykinin response completely insensitive to pertussis toxin treatment. ATP caused the largest increase in Ins(1,4,5)P3 and intracellular calcium; the Al-selective agonist CPA and bradykinin caused smaller increases (Fig. 1). UTP, a selective agonist at the Pzu receptor, gave effects similar to those of ATP on both Ins(1,4,5)P3 (Fig. 1) and Ca 2+ (data not shown). The concentrations used are those that produce maximal responses (Gerwins and Fredholm 1992a, b). The Ins(1,4,5)P3 responses to ATP and U T P were slightly, but not significantly, modified by increasing the Mg 2+ concentration to 10 m M (47+13 and 55+_8 pmol Ins(1,4,5)P3/106 cells without

(n

Mg 2+ and 34+ 5 and 4 2 + 7 pmol Ins0,4,5)P3/106 cells in the presence of t0 m M MgC12. Data from three experiments performed in triplicate).

P2u receptor-mediated increase in diacylglycerol formation Since ATP and UTP, via an action on Pzu receptors, activate phospholipase D (Gerwins and Fredholm 1992a) it was expected that the two agents should be able to raise the level of diacylglycerol in much the same way as CPA and bradykinin (Gerwins and Fredholm 1995). Indeed, as seen in Fig. 2, there was a clearcut, biphasie increase in diacylglycerol formation. The initial peak represented an increase of approximately 3 0 - 7 0 pmol/106 cells in the three experiments for both agonists. The second peak was much more protracted and thus probably represents a much larger overall generation of diacylglycerol. Interestingly, the magnitude of the second peak appeared much larger for U T P and ATP than for CPA and bradykinin (cf data in Fig. 2 with data in Fig. 1 of Gerwins and Fredholm 1995).

Activation of phospholipase D The formation of phosphatidylethanol in cells labeled with [3H]palmitic acid and activated in the presence of ethanol, was used as a specific marker for P L D activity (Dawson 1967; Kobayashi and Kanfer 1987; Yang et al.

A 6o

40

300

i

35

2 ,~

2O

•

ATP

•

UTP

100 o .~ 25-

o

0

T

.Z.

Fig. 1 Agonist-induced formation of Ins(l,4,5)P 3 and mobilization of intracellular Ca2+ in DDT 1 MF-2 smooth muscle cells. A Cells were incubated with vehicle, CPA (100 nM), bradykinin (1 gM), ATP (100 gM) or UTP (100 gM) for 30 s and the amount of Ins(1,4,5)P3 formed analyzed with a competitive protein binding assay as described in Materials and methods. The levels of Ins(1,4,5)P 3 increased from a basal level of 18_+2 pmol/106 cells to 30+_2 for CPA, 33±3 for bradykinin, 67±4 for ATP, and 60±3.1 for UTP. Mean_+SEM from four experiments each performed in triplicate. B DDT 1 MF-2 smooth muscle ceils were loaded with the fluorescent dye Fura 2-AM and intracellular levels of free calcium determined as described in Materials and methods. The maximal increases (at the top of the initial peak; basal level 107_+14 nM) were: for CPA 233±12nM, bradykinin 213_+10nM and ATP 326_+13 nM. UTP-stimulated increases in Ca2+ were investigated in only a few cases but were similar to ATP induced responses (data not shown). Concentrations of agonists used are the same as in A. Mean_+SEM from four experiments

~" ~5

10-

50

4

+

0

~

,

2

4

6

,

,

8

10

12

Time (rain)

Fig. 2 ATP- and UTP-induced increase in diacylglycerol. The agonist was added at a supramaximal level (0.1 raM) to cultured DDTj MF-2 cells (0.15-1.4x 106 cells/point). At indicated times the reaction was stopped and diacylglycerol was measured as described (Gerwins and Fredholm 1995). To minimize the variations due to differences in cell number and in specific activity of ATP, the results are expressed as per cent of the basal level in unstimulated cells at each time point. Mean ± SEM of three separate experiments

197

..-. 15°/I[

-,,,

L TII

1/

g b .E

1-7///

~J (8

c t00 0 O_ 03

®

75

.qE

5O

i

~

I

O PKC

L ~ k

100 1:3

E

50

25

Z

n

7///

0

CPA BRAD ATP

I

I

T

I

oo 7 6 5 -log[to51 -7549](M)

UTP

Fig. 3 The formation of phosphatidylethanol as a marker for phospholipase D activation. DDT 1 MF-2 smooth muscle cells were labeled with [3Hlpalmitic acid and activated for 10 min with 100 nM CPA, 1 ~tM bradykinin, 100 gM ATP or 100 laM UTP in the presence of 0.5% ethanol. The amount of phosphatidylethanol formed was used as a marker for phospholipase D activity, determined as described in Materials and methods. In addition, cells were treated with pertussis toxin (200 ng/ml, 4 h) and activated as described above. Data are from 4 - 5 experiments and are expressed as increases over basal levels. Statistical analysis was done using Mann-Whitney two sample test (*p < 0.05)

Fig. 5 Effects of the protein kinase C inhibitor Ro 31-7549 on protein kinase C and phospholipase D activity in DDT 1 MF-2 cells. Cells were activated with 100 nM phorbol 12-myristate 13-acetate for 10 min in the presence of increasing concentrations of Ro 31-7549. Protein kinase C (O) and phospholipase D ( • ) activities were determined as described in Materials and methods. Data are expressed as % of the phorbol 12-myristate 13-acetate-induced response in the absence of inhibitor, n = 3, mean_+SEM. PLD was stimulated, in the average, 3.2-fold and PKC 4.8-fold by TPA. The basal activity was subtracted and the inhibition of the TPA-induced increase calculated

1967). We have shown t h a t this m e t h o d can be used to assess P L D a c t i v a t i o n in DDT~ M F - 2 cells (Gerwins a n d F r e d h o l m 1995). These e x p e r i m e n t s showed t h a t at m a x i m a l l y effective c o n c e n t r a t i o n s o f agonists, A T P a n d U T P caused the largest increase, whereas C P A a n d b r a d y k i n i n c a u s e d smaller increases (Fig. 3). T h e re-

sponse to C P A was sensitive to pertussis toxin t r e a t m e n t o f cells, whereas b o t h b r a d y k i n i n - a n d ATP- or U T P - i n d u c e d activation o f P L D was c o m p l e t e l y insensitive to toxin t r e a t m e n t (Fig. 3). T h e i m p o r t a n c e o f C a 2+ in the a c t i v a t i o n o f P L D in DDT~ MF-2 was evaluated by c h e l a t i o n o f intracellular C a 2÷ with B A P T A - A M (30 g M , 30 m i n p r e t r e a t m e n t o f cells) or extracellular C a 2+ with 2 m M EGTA. There was a tendency, a l t h o u g h n o t always statistically significant, t h a t e l i m i n a t i o n o f intra- or extracellular c a l c i u m r e d u c e d a g o n i s t - i n d u c e d P L D a c t i v a t i o n (Fig. 4). P h o r b o l 12-myristate 13-acetate caused a 3 - 4 fold increase in P L D activity ( d a t a n o t shown), a response t h a t was a n t a g o n i z e d by the selective P K C i n h i b i t o r Ro 31-7549 with an ICs0 o f 2 . 4 + 0 . 4 ~tM, n = 3 (Fig. 5). A g o n i s t - i n d u c e d increases in P L D activity were n o t affected by 5 g M o f the P K C i n h i b i t o r Ro 31-7549 (Fig. 4).

[

]

9O O

= ,m

60

Ill 3 0

13.

A c t i v a t i o n o f p r o t e i n kinase C CPA BRAD ATP

UTP

Fig. 4 The influence of PKC and intra- and extracellular calcium on agonist-induced PLD activation. Cells were treated with 30 gM BAPTA-AM for 30min, 2 m M EGTA for 5rain or 5~tM Ro 31-7549 for 5 rain before activation of cells with indicated agonists for 10 min, and phospholipase D activity determined as described in Materials and methods. In each set of experiments the first set of bars (diagonal cross-hatching) represent control, the second set (horizontal lines) cells treated with EGTA, the third set (vertical lines) cells treated with BAPTA-AM, and the fourth set (vertical cross-hatching) cells treated with the PKC inhibitor Ro 31-7549. Concentrations of agonists were the same as in Fig. 3. Data from 5 experiments, mean+SEM, are expressed as % increase over nonstimulated levels. Statistical analysis was done using Wilcoxon signed-rank test (*p < 0.05)

In o r d e r to m e a s u r e a c t i v a t i o n o f P K C we have used a p e r m e a b i l i z e d cell assay ( H e a s l y a n d J o h n s o n 1989) (see also Gerwins a n d F r e d h o l m 1995). A f t e r a c t i v a t i o n with agonists, cells were p e r m e a b i l i z e d a n d the p h o s p h o r y l a tion o f specific p e p t i d e substrates (Kazanietz et al. 1993; Yasuda et al. 1990) was used to m e a s u r e a g o n i s t - i n d u c e d P K C activity. D a t a presented in Gerwins a n d F r e d h o l m 1995, demonstrate t h a t there is n o g e n e r a t i o n o f messengers after p e r m e a b i l i z a t i o n . I n DDT~ M F - 2 s m o o t h m u s cle cells, a c t i v a t i o n o f a d e n o s i n e A~ receptors with 100 n M C P A or b r a d y k i n i n receptors with 1 ~ M b r a d y kinin increased P K C activity, whereas a c t i v a t i o n o f P2u receptors with 100 g M A T P c a u s e d no significant activa-

198

tion of the enzyme (Fig. 6). UTP (100 gM), the selective Pzu receptor agonist, also failed to significantly activate PKC (Fig. 6). Examination of the time course for PKC activation showed that both bradykinin and CPA transiently increased PKC activity with maximal responses at 1 - 2 min, after which activity returned to nonstimulated levels within 10 min (Gerwins and Fredholm 1995). In contrast, neither ATP nor UTP caused significant activation of PKC at any time examined (results not shown). Over the same time period, ATP and UTP failed to stimulate the phosphorylation of another substrate peptide, peptide e. CPA and bradykinin, however, stimulated phosphorylation also of this peptide (Fig. 6). The response to CPA was completely sensitive to treatment of cells with pertussis toxin while that to bradykinin was insensitive (Fig. 6). Previous data suggest that ATP activates more than one G-protein (Gerwins and Fredholm 1992a) and hence there is a possibility that ATP activates a pathway that inhibits PKC activity. Another possibility is that ATP, which is a nonselective P2 agonist, activates some other subtype(s) of P2 receptors that might inhibit PKC. However, ATP did not inhibit PKC activity induced by bradykinin, CPA or phorbol esters (data not shown). Phorbol 12-myristate 13-acetate increased PKC activity approximately 14-fold over basal when cells were stimulated for 10 rain (data not shown). Ro 31-7549, a rather selective PKC antagonist (Nixon et al. 1992), inhibited the phorbol 12-myristate 13-acetate-induced PKC activation with half maximal inhibition at 100 nM (Fig. 5) and completely abolished agonist-induced PKC activity at i gM (Fig. 6).

A 60

~- ~45 ~. O ~ ~30 15

MBP4.14

B peptide E

~

~ U

~ PTX 41~ I~= RO31-7549

L ~

ctrl

ll~/ ....

FTf

l

/ 60

-j45

F 1 9 1 ~

0

30 15

0 CPA BRAD ATP UTP

320 3OO 280 260 24o

220

200 160 1600

240

480

720

960

1200

Time (see)

Fig. 7 The effect of ATP (100 gM) bradykinin (1 gM) and CPA (0.1 gM) on intracellular Ca2+ in a single DDT~ MF-2 cell. Cells were grown on round cover slips (25 ram) and loaded with 2 gM Fura 2-AM for 45 min, washed with buffer and then mounted in an open chamber, attached to the Zeiss inverted microscope (Axiovert 35) and superfused with buffer. The drugs were added to the buffer and the cells exposed to the drug for 1 min. The results are from a single experiment which is representative of five similar experiments

ATP, bradykinin and adenosine act on the same cells It could be argued that the reason for the failure of ATP (or UTP) to activate PKC, whereas bradykinin and adenosine analogues do, is that different cells are involved. In order to address this question, we examined the ability of ATP, bradykinin and N6-cyclopentyladenosine to stimulate the accumulation of Ca 2+ in single cells as a measure of the phospholipase C activation. In Fig. 7 are shown the results from one such experiment. They show that receptors for ATP, CPA and bradykinin coexist in a single cell (Fig. 7). Using visual inspection of the fluorescence changes after excitation at 380 nm we see no clear differences in the population of cells activated by one or the other of the agonists, even though the magnitude of the response and the time course shows considerable variation between cells.

~ ~: .~ o nan

Fig. 6A, B Agonist-induced activation of protein kinase C in DDT1 MF-2 smooth muscle cells. Cells were activated with indicated agonist for 45 s and PKC activity determined by measuring the phosphorylation of MBP4_I4 (A) or peptide e (B) in a permeabilized cell system as described in Materials and methods. The phosphorylation of MBP4_14 was 0.55_+0.03 pmol/10 min in unstimulated cells. Phosphorylation was increased 48_+16% when cells were stimulated with 100nM CPA, 52+12% with I gM bradykinin, 14_+4.5% with 0.1 mM ATP and 8_+19% with 0.1 mM UTP. Peptide e phosphorylation was increased 36_+4% with CPA, and 22_+7% with bradykinin, while ATP and UTP caused no increases at all. Mean_+SEM from three to four experiments. Pertussis toxin significantly blocked the CPA-inducedphosphorylation of MBP4 14 and 1 gM Ro 31-7549 blocked CPA- and bradykinininduced responses (A) (* = p < 0.05, Wilcoxon signed-rank test)

Discussion The important finding of the present study is that there is a discrepancy between the activation of PLC or PLD and of PKC. Thus, activation of the P2u receptor in DDT~ MF-2 smooth muscle cells with ATP or UTP increases both the formation of Ins(1,4,5)P 3 and the mobilization of intracellular calcium without causing significant activation of PKC. In contrast, activation of adenosine A 1- or bradykinin receptors increased PLC and PLD as well as PKC activity. Furthermore, we present evidence that all three types of agonist can activate the same cell.

199

Thus, the differences in the ability to activate PKC by Pzu agonists cannot be explained by marked differences in the affected cell populations. Although it is generally assumed that activation of PLC leads to the activation of PKC there is one previous report showing that stimulation of cells with ATP activates PLC without activating PKC (Gonzalez et al. 1989). Here we show that this was not due to a different ability of the agonists to increase diacylglycerol, the endogenous activator of PKC. Instead, all agonists studied, both t h n ~ o t h a t nnt~xTnto P l ~

~nr~ t h n ~ o t h a t d n n n t

r n ' ~ o r l r~i-

acylglycerol in a similar, biphasic manner. It has been shown in several other systems that the first phase of the increase probably reflects diacylglycerol generated from phosphatidylinositolbisphosphate through PLC, while the second peak originates from hydrolysis of phosphatidylcholine, presumably catalyzed by PLD (Billah 1993; Billah and Anthes 1990; Exton 1990). Furthermore, all agonists studied increased PLD activity: P2u receptor activation gave the largest increase, followed by adenosine A 1- and bradykinin-receptor activation, which caused smaller increases. The agonists studied appeared to increase intracellular levels of diacylglycerol to the same extent, but differed in their ability to increase Ins(1,4,5)P 3 and intracellular Ca 2+. Thus, there appeared to be no absolute correlation between PLC activation and diacylglycerol production. Since increases in diacylglycerol were very transient, the intracellular levels of diacylglycerol must be under strict control of metabolizing enzymes, particularly the diacylglycerol kinase (Kanoh et al. 1990). There were no indications that the agonists used in this study influenced diacylglycerol kinase activity since the effect of the diacylglycerol kinase inhibitor R 59022 (Nunn and Watson 1987) on PKC activity was not altered by the agonists (data not shown). R 59022 by itself increased PKC activity to the same extent as did phorbol 12-myristate 13-acetate (data not shown). However, the fact that the first peak of diacylglycerol formation was very transient indeed and only one time point was used to determine the magnitude of the peak means that the above conclusions are only provisional. The notion that diacylglycerol is rapidly metabolized implies that agonist-induced increases in intracellular levels of diacylglycerol might be strictly localized to the site of production and hence show compartmentalization. Thus the discrepancy in PLC, PLD, and PKC activation might be explained by compartmentalization which separates the increases in Ins(1,4,5)P3, intracellular calcium, and diacylglycerol from PKC. It has been shown that diacylglycerol generated from PLD-mediated phosphatidylcholine hydrolysis in thrombin-stimulated fibroblasts is unable to activate PKC and a compartmentalization of diacylglycerol was therefore suggested (Leach et al. 1991). Furthermore, effects on diacylglycerol without activation of PKC have been shown for PDGF receptor-mediated signaling (Larrodera et al. 1991). Another possibility is that ATP induces the specific activation of an isoform of PKC for which MBP4_14 is

not a substrate: it has been shown that MBP 4_ 14 is a substrate only for the a, fl and y isoforms of PKC (Kazanietz et al. 1993; Yasuda et al. 1990). DDT 1 MF-2 cells express a, e and ( isoforms of PKC (Assender et al. 1994) and since e and ( PKC do not require calcium for activation, an increase in Ins(1,4,5)P3 and intracellular calcium is not necessarily a good predictor of PKC activation in all situations. The implication of this might be that whereas stimulation of adenosine A~- and bradykinin-receptors activates PKC a, the stimulation of Pzu receptors activates e and/or ( isoforms of PKC. This possibility is made unlikely by the finding that another substrate, peptide e, which is phosphorylated well by e and ~ PKC (Kazanietz et al. 1993), was also not phosphorylated when cells were treated with ATP or UTP. CPA, bradykinin and phorbol 12-myristate 13-acetate were, however, reasonably good stimulators. Furthermore, in another study ATP was found to increase intracellular Ca 2+ without activation of PKC, as measured by phosphorylation of an 80-kDa acidic protein or transmodulation of the receptor for epidermal growth factor (Gonzalez et al. 1989). The exact mechanism whereby agonists activate PLD is unknown. Direct activation through an as yet unidentified G-protein as well as direct activation by Ca 2+ or PKC has been suggested (Billah and Anthes 1990; Cockcroft 1992; Exton 1990; Thompson et al. 1991). If PKC is an obligatory component of PLD activation, then the findings that ATP can stimulate PLD but not PKC would be difficult to reconcile. The mechanism of activation was therefore examined further. The protein kinase C inhibitor Ro 31-7549 blocked both phorbol 12-myristate 13-acetate-induced PKC and PLD activity; it was about 20 times more potent inhibiting the former. This is probably not due to differences in assay procedures, since in both the PKC and the PLD assay, the inhibitor is administered to the intact cell, whereafter cells are assayed for PKC and PLD activity. A possibility is that the phorbol 12-myristate 13-acetate-induced activation of these effectors is brought about by different PKC isoforms that might differ in their sensitivity to Ro 31-7549. Also, it is possible that in order to maximally activate PLD, only a small fraction of the PKC needs to be activated: hence a shift to the right compared to PKC activity would be expected. The receptor agonists produced a much smaller activation of PKC, and a somewhat smaller activation of PLD, than did phorbol 12-myristate 13-acetate. Therefore even an incomplete inactivation of PKC would be expected to reduce PLD if PKC is indeed an obligatory coupling factor. This was, however, not observed. Thus, our data suggest that PKC is required for maximal activation of PLD, as shown in several studies (Billah 1993; Bonser and Thompson 1992; Nixon et al. 1992), but is not the only means of its activation in DDT~ MF-2 cells. The fact that PLD was activated by Pzu receptor stimulation without any apparent activation of PKC further strengthens the latter conclusion. In DDT~ MF-2 cells, we find that reduction of intra- or extracellular calcium, as well as inhi-

200

bition of PKC, tended to reduce PLD activation, al- Doglio A, Dani C, Grimaldi P, Ailhaud G (1989) Growth hormone stimulates c-fos gene expression by means of protein kinase C though statistical significance was reached only in a few without increasing inositol lipid turnover. Proc Natl Acad Sci situations. This to us suggests that PLD in DDT 1 MF-2 USA 86:1148-1152 cells is activated via the combined action of Ca 2+ , PKC Eibl H, Kovatchev S (1981) Preparation of phospholipids and their and G-proteins. It has previously been stated that to analogs by phospholipase D. Methods Enzymol 72:632-639 achieve maximal activation of PLD, a synergistic activa- Exton JH (1990) Signaling through phosphatidyleholine breakdown. J Biol Chem 265:1-4 tion of the enzyme by G-proteins, Ca 2+ and PKC is reFuruiehi T, Yoshikawa S, Miyawaki A, Wada K, Maeda N, quired (Billah and Anthes 1990; Exton 1990). Mikoshiba K (1989) Primary structure and functional expresAn interesting finding in the present study is that the sion of the inositol 1,4,5-trisphosphate-binding protein P400" Pzu receptor-mediated activation of PLD was insensitive Nature 342:32-38 to pertussis toxin treatment. This is in contrast with the F~llman M, Lew DP, Stendahl O, Andersson T (1989) Receptor-mediated phagocytosis in human neutrophils is associated with inactivation of PLC via these receptors, where a substantial creased formation of inositol phosphates and diacylglycerol. Elinhibition was seen following pertussis toxin treatment. evation in cytosolic free calcium and formation of inositol phosThis reinforces the idea that P2u receptors couple via phates can be dissociated from accumulation of diacylglycerol. more than one G protein. J Clin Invest 84:886-891 In summary, the present study indicates that there is Gerwins P (1993) Modification of a competitive protein binding assay for determination of inositol 1,4,5-trisphosphate. Anal no direct correlation between agonist-stimulated activaBiochem 210:45-49 tion of PLC, PLD, increases in C a 2+ or diacylglycerol, Gerwins P, Fredholm BB (1992a) ATP and its metabolite adenosine and the activation of PKC. Thus, activation of P2u react synergistically to mobilize intracellular calcium via the formation of inositol 1,4,5-trisphosphate in a smooth muscle cell ceptors does not activate PKC, whereas activation of line. J Biol Chem 267:16081-16087 bradykinin and adenosine receptors does cause PKC activation. This is despite the fact that Pzu receptor activa- Gerwins P, Fredholm BB (1992b) Stimulation of adenosine A 1 receptors and bradykinin receptors, which act via different G-protion, if anything, causes larger activation of PLC and teins, synergistically raises inositol 1,4,5-trisphosphate and inPLD, as well as larger increases in C a 2+ and diacyltracellular free calcium in DDT i MF-2 smooth muscle cells. Proc Natl Aead Sci USA 89:7330-7334 glycerol, than does activation of the other two receptors. One explanation of this apparent discrepancy is that re- Gerwins P, Fredholm BB (1995) Activation of adenosine A 1 and bradykinin receptors increases protein kinase C and phosphoceptor activation of PKC is highly compartmentalized. lipase D activity in smooth muscle cells. Naunyn-SchmiedeAcknowledgements We want to thank Mrs. Janet Holmdn for excellent secretarial assistance and Mrs. Eva Irenius for assistance with the single cell calcium measurements. These studies were supported by the Swedish Medical Research Council (project no. 2553 and K93-04P-09717-03), the Swedish Cancer Association, the Swedish Association for Medical Research, Gustaf V's 80 year fund and by Karolinska Institutet.

References Assender JW, Kontny E, Fredholm BB (1994) Expression of protein kinase C isoforms varies with state of differentiation in smooth muscle cells. FEBS Lett 342:76-80 Augert G, Blackmore PF, Exton JH (1989) Changes in the concentration and fatty acid composition of phosphoinositides induced by hormones in hepatocytes. J Biol Chem 264:2574-2580 Berridge MJ (1993) Inositol trisphosphate and calcium signalling. Nature 361:315-325 Billah MM (1993) Phospholipase D and cell signaling. Curr Opin Immunol 5:114-123 Billah MM, Anthes JC (1990) The regulation and cellular functions of phosphatidyleholine hydrolysis. Biochem J 269:281-291 Bonser RW, Thompson NT (1992) Phosphatidylcholine hydrolysis by phospholipase C and D. In: Milligan G (ed) Signal transduction. A practical approach. IRL Press, Oxford, pp 123-151 Brindley DN (1984) Intracellular translocation of phosphatidate phosphohydrolase and its possible role in the control of glycerolipid synthesis. Prog Lipid Res 23:115 - 133 Cockcroft S (1992) G-protein-regulated phospholipases C, D and A2-mediated signalling in neutrophils. Biochim Biophys Acta 1113:135-160 Dawson RMC (1967) The formation of phosphatidylglycerol and other phospholipids by the transferase activity of phospholipase D. Biochem J 102:205-210

berg's Arch Pharmacol 351:186-193 Gonzalez FA, Rozengurt E, Heppel LA (1989) Extracellular ATP induces the release of calcium from intracellular stores without the activation of protein kinase C in Swiss 3T6 mouse fibroblasts. Proc Natl Acad Sci USA 86:4530-4534 Grynkiewicz G, Poenie M, Tsien RY (1985) A new generation of Ca z+ indicators with greatly improved fluorescence properties. J Biol Chem 260:3440-3450 Heasly LE, Johnson GL (1989) Regulation by protein kinase C of nerve growth factor, epidermal growth factor and phorbol esters in PC12 pheochromocytoma cells. J Biol Chem 264:8646-8652 Kanoh H, Yamada K, Sakane F (1990) Diacylglycerol kinase: a key modulator of signal transduction? Trends Biochem Sci 15:47-50 Kazanietz MG, Areees LB, Bahador A, Mischak H, Goodnight J, Mushinski JE Blumberg PM (1993) Characterization of ligand and substrate specificity for the calcium-dependent and calcium-independent protein kinase C isozymes. Mol Pharmacol 44:298-307 Kobayashi M, Kanfer JN (1987) Phosphatidylethanol formation via transphosphatidylation of rat brain synaptosomal phospholipase D. J Neurochem 48:1597-1603 Kosaka Y, Ogita K, Ase K, Nomura H, Kikkawa U, Nishizuka Y (1988) The heterogeneity of protein kinase C in various rat tissues. Biochem Biophys Res Commnn 151:973-981 Larrodera, P, Cornet ME, Diaz-Meco MT, Lopez-Barahona M, Diaz-Laviada I, Guddal PH, Johansen T, Moscat J (1991) Phospholipase C-mediated hydrolysis of phosphatidyleholine is an important step in PDGF-stimulated DNA synthesis. Cell 61:1113-1120 Leach KL, Ruff VA, Wright TM, Pessin MS, Raben DM (1991) Dissociation of protein kinase C activation and sn-l,2-diacylglycerol formation. Comparison of phosphatidylinositol- and phosphatidylcholine-derived diglycerides in a-thrombin-stimulated fibroblasts. J Biol Chem 266:3215-3221 Nishizuka Y (1986) Studies and perspectives of protein kinase C. Science 233:305-312 Nishizuka Y (1992) Intracellular signaling by hydrolysis of phos-

201 pholipids and activation of protein kinase C. Science 258: 607-614 Nixon JS, Bishop J, Bradshaw D, Davis PD, Hill CH, Elliott LH, Kumar H, Lawton G, Lewis E J, Mulqueen M, Westmacott D, Wadsworth J, Wilkinson SE (1992) The design and biological properties of potent and selective inhibitors of protein kinase C. Biochem Soc Trans 20:419-425 Norris JS, Gorski J, Kohler PO (1974) Androgen receptors in a Syrian hamster ductus deferens tumor cell line. Nature 248:422-424 Nunn DL, Watson SP (1987) A diacylglycerol inhibitor, R59022, potentiates secretion by and aggregation of thrombin-stimulated human platelets. Biochem J 243:809-813 Rosoff PM, Savage N, Dinarello CA (1988) Interleukin-i stimulates

diacylglycerol production in T-lymphocytes by a novel mechanism. Cell 54:73-81 Stabel S, Parker PJ (1991) Protein kinase C. Pharmacol Ther 51:71-95 Thompson NT, Bonser RW, Garland LG (1991) Receptor-coupled phospholipase D and its inhibition. Trends Pharmacol Sci 12:404- 408 Yang SF, Freer S, Benson AA (1967) Transphosphatidylation by phospholipase D. J Biol Chem 242:477-484 Yasuda I, Kishimoto A, Tanaka S, Tominaga M, Sakurai A, Nishizuka Y (1990) A synthetic peptide substrate for selective assay of protein kinase C. Biochem Biophys Res Commun 166:1220-1227