JOURNAL OF CELLULAR PHYSIOLOGY 150:57&585 (1992)

Endothelin Stimulates Phosphatidic Acid Formation in Cultured Rat Mesangial Cells: Role of a Protein Kinase C-Regulated Phospholipase D M A R K KESTER,* M I C H A E L S. S I M O N S O N , R. GUY M C D E R M O T T , ELISABETTA BALDI, AND M I C H A E L J. DUNN Departments of Medicine and PhysiologylBiophysics, Case Western Reserve University, University Hospitals of Cleveland, Cleveland, Ohio 44 106 We have previously reported that endothelin-1 stimulates phospholipase C-induced hydrolysis of phosphatidylinositol-4,5-bisphosphate, Other signal transduction pathways that hydrolyze alternative phospholipids through phospholipase D may also mediate endothelin-stimulated cellular responses. We initially evaluated endothelin-dependent generation of "P-phosphatidic acid as an indirect indication of phospholipase D activity in rat mesangial cells. Endothelin (lO-'M) induced an elevation of phosphatidic acid that was maximal at 1 5 min and persisted upward of 60 min. Pretreatment with the diacylglycerol-kinase inhibitor, R59022, did not reduce formation of endothelin-stimulated 32P-phosphatidic acid, demonstrating that the sequential actions of phospholipase C/diacylglycerol kinase do not contribute to endothelin-stimulated phosphatidic acid formation. We next conclusively identified a role for phospholipase D in the generation of phosphatidic acid by assessing the formation of 'H-phosphatidylethanol from 'H-alkyl lyso glycerophosphocholine and exogenous ethanol. Endothelin stimulated 3H-alkyl phosphatidylethanolformation in the presence but not the absence of 0.5% ethanol. Also, endothelin induced a concomitant elevation of 'H-alkyl-phosphatidic acid that was significantly reduced when the cells were exposed to exogenous ethanol, reflecting the formation of phosphatidylethanol. In addition, endothelin stimulated the release of 3H-choline and 3H-ethanolamine, demonstrating that additional phospholipids may serve as substrates for phospholipase D. Phorbol esters and synthetic diglycerides mimicked the effects of endothelin to stimulate phospholipase D and inhibitors of protein kinase C significantly reduced endothelin-stimulated phospholipase D. In addition, endothelin did not stimulate phosphatidylethanol formation in protein kinase C down-regulated cells. The calcium ionophore, ionomycin, did not stimulate phospholipase D and mesangial cells pretreated with BAPTA to chelate cytosolic calcium did not show a diminished endothelin-stimulated phospholipase D. Thus these data demonstrate that mesangial cells possess a protein kinase C-regulated phospholipase D activity that can be stimulated with endothelin.

Endothelin (ET) is a recently characterized vasoconstrictor hormone produced by vascular endothelial cells as well as by many other cell types (Yanagisawa et al., 1988; Simonson and Dunn, 1990a). There have been four isopeptides characterized in human and rat tissues, and the ET-1 isoform is preferentially released from vascular endothelium and bovine glomerular endothelial cells (Marsden e t al., 1989; Simonson and Dunn, 1990a). It is postulated that released endothelin induces adjacent vascular smooth muscle or mesangial cells to constrict and/or proliferate (Marsden et al., 1989; Simonson et al., 1989). Mesangial cell contraction regulates glomerular capillary surface area and thereby glomerular filtration rate by altering the ultrafiltration coefficient, whereas mesangial cell proliferation is associated with matrix accumulation and 0 1992 WILEY-LISS. INC.

sclerotic lesions within the mesangium (Mene' et al., 1989). Recent studies in several tissues, including mesangial cells, suggest that endothelin elevates intracellular calcium through a phospholipase C-dependent generation of Ins 1,4,5 P, (Mitsushashi et al., 1989; Simonson et al., 1989). Endothelin also increases diacylglycerol generation, providing corroborating evidence for a phospholipase C signal transduction pathway (Griendling et al., 1989). Finally, endothelin stimulates choline production in fibroblasts, a n indirect marker for phospholipase D activity (MacNulty

Received June 6,1991; accepted October 7, 1991.

*To whom reprint requestsicorrespondence should be addressed.

ENDOTHELIN AND PHOSPHOLIPASE

et al., 1990). It is hypothesized that a n additional signal transduction pathway that hydrolyzes phosphatidylcholine via phospholipase D may also mediate ET-stimulated cellular responses. Phospholipase D activity has been conclusively demonstrated in platelets (Wang and Tai, 19881, smooth muscle cells (Welsh et al., 19901, endothelial cells (Martin, 1988), hepatocytes (Bocckino et al., 1987), HL-60 granulocytes (Billah et al., 1988) and mesangial cells (Pfeilschifter, 1990, unpublished data). Phospholipase D hydrolyzes phospholipid substrates to yield phosphatidic acid, which, like ET-1, stimulates proliferation and increases intracellular calcium (Moolenar et al., 1986). The mechanics of the phospholipase D reaction involve a phosphatidyl-enzyme intermediate (base-exchange) that accepts water as the phosphatidate acceptor (Billah and Anthes, 1990). This process can be exploited as alcohol can substitute for water to form phosphatidylalcohol derivatives (transphosphatidylation). Thus the formation of phosphatidylethanol in the presence of exogenous ethanol is conclusive proof of phospholipase D activity, as transphosphatidylation is unique to phospholipase D. Therefore, we set out to characterize a n ET-1-induced phospholipase D activity in mesangial cells and to identify if transmembrane signals induced by ET-1 activation of phospholipase C are necessary cofactors for ET-1 stimulable phospholipase D activity.

METHODS Materials ET-1 was obtained from the Peptide Institute (Osaka, Japan). All radiolabelled materials, including 3H-alkyl lyso glycero-3-phosphocholine, were obtained from Amersham Corp. (Chicago). Phosphatidylethanol was purchased from Avanti Biochemical (Birmingham, AL), whereas all other phospholipid standards were purchased from Serdary Biochemicals (London, Ontario, Canada). Sangivamycin was a generous gift of the National Cancer Institute. All other materials were purchased from either Sigma Chemicals (St. Louis) or Calbiochem (La Jolla, CAI.

Glomerular mesangial cell isolation and culture Mesangial cells were grown from collagenase-digested glomeruli obtained from 100 g male Sprague Dawley rats by a sequential sieving technique (Striker et al., 1980; Simonson and Dunn, 1986).Mesangial cells were grown in RPMI 1640 culture medium supplemented with 17% fetal bovine serum, 100 u/ml penicillin, 100 pg/ml streptomycin, 5 pg/ml insulin, 5 pgiml transferrin, and 5 ng/ml selenium at 37°C in 5% CO,. Cells were used in their third-to-eleventh passage. We have previously verified that the mesangial cell cultures are devoid of epithelial, endothelial, macrophage, and fibroblast contamination (Simonson and Dunn, 1986; Werber et al., 1987). Radiolabelling and extraction of membrane phospholipids PA can be formed directly by phospholipase D hydrolysis of membrane phospholipids or alternatively through a combined phospholipase C/DAG kinase pathway. PA can also be formed through de novo phospho-

579

lipid synthesis as glycerol-3-phosphate (derived from glycolysis or by ATP-induced phosphorylation of glycerol) can be sequentially acylated by glycerol phosphate acyltransferase and monoacyl glycerol phosphate (lyso PA) acyltransferase. 32P-P04=labeling of cultured cells yields 32PA derived from all pathways, whereas 3H alkyl lyso glycero-3-phosphocholine labeling in the presence of ethanol is a reflection primarily of phospholipase D-derived PA. Twelve well-cluster dishes were used for "P-orthophosphate, 3H-choline, and 3H-ethanolamine experiments; 60 mm Petri dishes were used for 3H-alkyl lyso GPC ex eriments. Cells were labelled for 3 h r with 50 yCi/ml '2P-P04= or 2 h r with 1 pCi/ml 3H-alkyl lyso GPC or 36 h r with 1 pCi/ml 3H-choline or "-ethanolamine. Maximum 32P-P0, labelling was obtained in low phosphate KHH buffer, whereas maximal 3H-alkyl lyso GPC labelling was obtained in calcium-free KHH. RPMI 1640 was used as the labelling media for choline or ethanolamine. Radioisotope was removed by extensive washing on ice, and then radiolabel-free, ice-cold complete KHH buffer was added and the cells were rewarmed to 37°C. 0.1 pM ET-1 dissolved in KHH or corresponding KHH control was added for the designated times. We have previously verified that 10-7M ET maximally stimulated [Ca2+],release, inositol phosphate formation, thymidine incorporation, cellular alkalinization (Simonson et al., 1989) and, in data not published, 1,Z-DAG generation (E. Baldi, M. Kester, unpublished observations). The alkyl lyso GPC experiments were run in the presence or absence of 0.5% EtOH. The incubations were terminated after various intervals with ice-cold acidified methanol and the cells scraped and transferred into a n equal volume of chloroform. Unless otherwise indicated, the alkyl lyso GPC experiments were stopped after 15 min, the time point corresponding to maximal ET-stimulated 32P-PA formation. A second methanol wash was added to yield a chloroform/ methanol/H,O ratio of 1/2/0.8. After sitting for 30 min at 4"C, the methanol and chloroform extracts were adjusted to yield two phases (chloroform/methanol/water, 1/1/0.9, v/v/v) (Bligh and Dyer, 1959). For the lipid experiments, the chloroform lipid extracts were separated and the methanol/water extracts were rewashed with chloroform. The combined chloroform extracts were dried under N2 and resuspended in 90% chloroform/ 10%methanol. The samples were spiked with authentic synthetic PA and/or phosphatidylethanol (PEt) and spotted on silica gel 60 TLC plates (250 pM thickness, EM Science, Cherry Hill, NJ, heat activated to 110°C). Phospholipids were eluted from the origin with a mobile phase consisting of chloroform/methanol/acetic acid, 65/15/6, v/v/v. Using the above elution system, PEt was shown to be well resolved from several phospholipid contaminants including bis-phosphatidic acid, cardiolipin, and phosphatidyl glycerol. The lipids were visualized by toluidino-2-napthalene sulfonic acid spray and UV light. The radiolabelled lipids that comigrate with the internal standards were scraped and radioactivity determined in a liquid scintillation counter. Preliminary studies assessing the labelling efficiency indicate that, after a 2-hr incubation period, 44 k 2% of the "H alkyl lyso

580

KESTER ET AL

GPC was reacylated into alkyl PtdCho and that this was the only phospholipid labelled (n = 16). In addition, the increase in PEt induced by ET represents 1.6 0.1%of the labelled PtdCho pool. To identify water soluble phospholipid-derived bases, the methanol/ H,O extracts were separated and the chloroform extract was re-extracted. The combined watersoluble extracts were evaporated utilizing a Speed-Vac concentrator (Savant, MA) and then redissolved in 50% ethanol. The samples were spiked with appropriate standards and separated by TLC with methanol/0.5% NaClI ammonia (100:100:2, vol/vol/vol) (Yavin, 1976). Choline was visualized by iodine or Dragendorff reagent, whereas ethanolamine was identified by spraying with ninhydrin. The phosphorylated and unphosphorylated forms of the bases were well separated from each other and comigrated with authentic standards. Data were expressed as cpmlpg prot for triplicate determinations in a t least n = 4 experiments on separate subcultures. Four separate primary cultures were utilized to establish the individual cell lines. The variability in basal 3H-PEt values (0.4 to 2.5 cpm/pg prot/l5 min) between experiments reflects the age of the subculture (older passages have higher basal values) and, for this reason, all ET-stimulated cultures were run against age-matched controls from the same primary culture. Statistical analysis for time course data consisted of 2-way analysis of variance followed by independent t-tests. Protein concentrations were determined by the methods of Lowry et al. (1951) on similarly plated corresponding 12-well cluster or petri dishes.

*

RESULTS Initially, ET-dependent generation of PA was evaluated as a n indication of phospholipase D activity. Mesangial cells (MC) were labelled for 3 h r a t 37°C with 32P-P04', and then the phospholipids were extracted and separated by TLC to assess ET-induced 32P-PA generation. ET (10-7M) increased PA formation a t 5 min, which peaked a t 15 min, and this elevation persisted through 60 min (Fig. 1). When MC were pretreated with the DAG-kinase inhibitor R59022, ETinduced PA synthesis was unaffected a t all time points suggesting that ET-stimulated PA was not a result of phospholipase C-derived DAG that is subsequently phosphorylated. The dose of R59022 utilized has previously been shown to augment IL-1-stimulated DAG in mesangial cells (Kester et al., 1989). Conclusive evidence demonstrating a role for ET-stimulable phospholipase D in the generation of PA was obtained by assessing the formation of 3H-phosphatidylethanol (PEt) from 3H-alkyl lyso glycero-3-phosphocholine and exogenous ethanol. Phospholipase D has the unique property (transphosphatidylation) to transfer alcoholic moieties to PA, and these products can be separated by TLC. ET stimulated alkyl-PEt generation approximately 230% above control in the presence but not the absence of 0.5% ethanol a t 15 rnin (Fig. 2). Also, ET induced a concomitant elevation of 3H-alkyl-PA a t 15 rnin in the presence or absence of ethanol. This elevation of 3H-alkyl-PA after ET treatment was significantly reduced when the cells were exposed to exogenous ethanol, which reflects the formation of 3H-alkyl

A

Endothelin

+R59022

Ti me (min)

Fig. 1. The effects ofendothelin (10-7M) upon "P-PA as a function of time in the presence ( A , A) or absence (o,.) or R59022, a DAG kinase inhibitor. Mesangial cells were prelabeled with 50 kCi/ml 32P-orthophosphate at 37°C and then extensively washed. Cells were either pretreated with lO-'M R59022 in 0.005% HC1 or control for 15 min and then treated with ET or control for the specified times. The data represent the x i SEM of replicate determination of n = 4 experiments. Two-way ANOVA established a significant (p < 0.011 difference between ET-1-treated and control groups

ncontrol

+ ETOH

- ETOH I

-ETOH I

PEt

+ETOH I

1

PA

The effects of endothelin (10-7M)upon phospholipase D activit< Mesangial cells in 60 mm petri dishes were labelled for 2 hr with 1 pCi/ml 3H alkyl lyso GPC and then extensively washed. Cells were then treated with ET for 15 min in the presence or absence of 0.5% ethanol. Lipids were extracted as described and PEt and PA were separated by TLC. n = 4 x 5 SEM, p** < 0.001, p* < 0.05.

PEt. Finally, we assessed 3H-PEt formation as a function of ET dose (Fig. 3). Threshold responses were seen with concentrations of ET as low as lo-' M with maximal stimulation observed at M. These results are similar to previously reported data for ET-stimulated Ptd Ins-4,5-P2-specificphospholipase C activity (Simonson et al., 1989). We next explored the role of early transmembrane signals (ICa2+liand PKC) generated by ET-stimulated

ENDOTHELIN AND PHOSPHOLIPASE

***

3.0

.E 2.0E

9

-

\

1.6-

h

581 control

PMA ( 1 0 - 7 ~ )

-

0

3 1.2Z L -

2 0.8&

n

n

-

- ETOH

OI.', Con

12

I1

10

9

8

7

6

+ ETOH

+ETOH

-ETOH

u u PEt PA

-log [Endothelin] M Fig. 3. The effects of ET upon PEt formation as a function of dose. All experiments contained 0.5% ETOH and were run for 15 min. n = 3, x 5 SEM, p* < 0.01.

phospholipase C on subsequent phospholipase D activity. Figure 4 demonstrates that PMA mimicked the effect of ET to stimulate phospholipase D. Again, P E t was formed only in the presence of agonist (PMA) and exogenous ethanol, whereas PA was generated in the presence or absence of ethanol. A subtle mechanistic difference between ET and PMA-stimulated phospholipase D is that PMA stimulated preferentially more PEt, whereas ET stimulated slightly more PA. Utilizing subcultured mesangial cells from another primary culture, the effects of acute PMA treatment upon subsequent ET-stimulated PEt formation were investigated. Individually, PMA (lO-'M) and ET (10-7M) elevated PEt formation to 1.38 2 .06 and 1.52 i .22 cpmlpg prot/l5 min, respectively, from a basal value of 0.81 & .07 cpmlpg protll5 min. However, when this submaximal dose of PMA (10-'M) was added 3 rnin before a subsequent addition of 10-7M ET, 3H-PEt Values were further increased to 1.89 .09 cpmlpg protll5 min (n = 4, p < 0.01). To rule out a direct effect of PMA on phospholipase D, we utilized synthetic diradylglycerols. Octadecyl acetyl glycerol (10-7M) and DiC, (10-7M) also elevated 3H-alkyl PEt 130 and 234%) respectively (n = 3) in experiments where PMA (lOP7M) elevated PEt 244% and ET (10-'M) stimulated PEt 135%. The inactive phorbol ester, 4aPMA, had no effect upon either 3H-alkyl-PEt or -PA levels (84% and 71% of control, respectively, where control was 0.38 cpmlpg prot.). We also investigated the role of protein kinase C inhibitors (H7, 10-6M, Fig. 5A, and sangivamycin, lO-'M, Fig. 5B) to modulate ET-stimulated phospholipase D. Both inhibitors had no effects upon basal P E t or PA concentrations. A 15-min preincubation with H, diminished the ET stimulation of phospholipase D by nearly loo%, whereas sangivamycin reduced ET-stimulated PEt concentration by approximately 70% and PA by approximately 33%. We next explored the effects of chronic PMA exposure (10T7M,18 hr) upon ET-stimulated PEt formation. We have previously demonstrated by both immunocytochemistry as well as a decreased number of 3H-PDBU binding sites that longterm exposure to PMA down-regulates PKC activity in mesangial cells (Simonson and Dunn, 1991) without

*

Fig. 4. The effects of PMA (10~.7M, 15 min) upon phospholipase D activity. PMA was dissolved in DMSO and suitable controls were utilized. n = 4 x ? SEM, p*** < 0.001, p** < 0.005, p* < 0.01.

influencing ET-binding kinetics in mesangial cells (Baldi and Dunn, 1990). In Figure 6, ET (10-7M) significantly increased PEt formation in the absence but not the presence of chronic PMA exposure. However, chronic PMA pretreatment also elevated basal phospholipase D activity. To document the effects of [Ca2+Ii upon phospholipase D, we stimulated mesangial cells with a dose of ionomycin that we have previously shown to raise [Ca2+lito levels seen with 10-7M E t (-600 nM) (Simonson and Dunn, 1990b). Ionomycin a t either 30 sec or 15 min in the presence of ethanol did not elevate 3H-alkyl PEt and, in fact, actually slightly diminished PEt formation (Fig. 7). ET-stimulated phospholipase D served as a positive control for these mesangial cell subcultures. To confirm these results, mesangial cells were preincubated with 10 pM BAPTA-AM for 25 min and then the BAPTA buffer was removed. This dose and time of BAPTA preincubation has previously been shown to abolish ET-induced elevations in [Ca2+li, clamping [Ca2+Iiat approximately 125 nM (Simonson and Dunn, 1991). The cells remained viable during the BAPTA-AM treatment a s the cells maintained a constant [Ca2+li(data not shown). ET elevated 3H-alkyl PEt and PA to a similar extent in the presence or absence of BAPTA (Fig. 8). Again, all experiments included 0.5% ethanol. Similar results were obtained (data not shown) utilizing a different protocol in which MC were pretreated with BAPTA for 5 min and then subsequently treated with ET or control for a n additional 15 min in the presence of BAPTA. 3H-alkyl lyso-glycero-3-phosphocholinelabelling of MC identifies a specific pool of phospholipase D hydrolyzable substrate. To confirm other phospholipid substrates for phospholipase D, we labelled MC with 3Hcholine or 3H-ethanolamine. After 36 hr, the cells were then extensively washed and treated with 0.1 pM ET or control for various time periods. Elevations in TLCseparated 3H-choline or -ethanolamine reflects acyl as well as alkyl PtdCho or PtdEth substrate pools for phospholipase D. Figure 9 shows th a t 3H-ethanolamine is the redominant watersoluble base th a t is elevated by H-choline is also released at 15 rnin and reinforces ET. ! the alkyl-PEt data that uses 3H-alkyl lyso glycero phos-

582

KESTER ET AL.

phocholine as substrate label. Significant increases in these watersoluble bases were only noted after 5 min and again reflect activation of phospholipase D subsequent to PtdIns 4,5-P2-specific phospholipase C.

DISCUSSION We have demonstrated in cultured mesangial cells that ET stimulates a PKC-regulated phospholipase D activity that generates the putative second messenger phosphatidic acid. To establish a n ET-1-coupled phospholipase D activity, the following unambiguous criteria were met: (1)ET-1 increased 3H PEt in the presence but not the absence of exogenous ethanol, and (2) ET-1induced 3H-PA formation decreased in the presence of exogenous ethanol compared to ET-1-stimulated controls in the absence of ethanol. Even though ethanol modestly elevated PEt formation, this increase was nonsignificant, a phenomenon also observed in HL-60 granulocytes (Pai et al., 1988). This suggests that basal phospholipase D activity is minimal in mesangial cells, a t least when the alkyl PtdCho pool serves as substrate for the reaction. 3H alkyl lyso glycero 3 phosphocholine is relatively permeable across the cell membrane and is rapidly reacylated. The tritium label is in the non-metabolized sn-1 position of the molecule and thus can be used to follow transphosphatidylation. Yet this label may underestimate the extent of phospholipase D activity, especially if diacyl pools of PtdCho or, for that matter, other membrane lipids are also substrates for phospholipase D. Thus the release of 3H-choline and 3H-ethanolamine was also utilized as a n indirect measure of phospholipase D activity. These data suggest that PA formation by phospholipase D utilizes several phospholipid substrates including PtdEth. The molecular species of phospholipid species that serve as substrates for receptor-mediated phospholipase D have not been thoroughly characterized. Even though PtdCho may be the preferred substrate for phospholipase D in several tissues, it is not the only substrate (Billah and Anthes, 1990). A phospholipase D-mediated hydrolysis of PtdEth but not PtdCho stimulated by ethanol was recently demonstrated in 3T3 fibroblasts (Kiss and Anderson, 1989). In addition, ATP and GTP yS were found to stimulate a phospholipase D-dependent hydrolysis of PtdEth and PtdCho in this same cell line (Kiss and Anderson, 1990). Several studies have attempted to identify if transmembrane signals induced by agonist-activation of phospholipase C are necessary cofactors for agoniststimulable phospholipase D and the conclusions reached are inconsistent. The role of physiological [Ca2+lito regulate phospholipase D is still controversial as the PtdCho-dependent phospholipase D has been shown to be both [Ca2'li-sensitive (Takai and Kanfer, 1979; Tettenhorn and Mueller, 1988; Anthes et al., 1989) and -insensitive (Martin 1988; Domino et al., 1989; Martinson et al., 1989). Equally as controversial, phospholipase D has been shown to be PKC-dependent in endothelial cells (Martin et al., 1989) and granulocytes (Billah et al., 1988), but not in platelets (Huang et al., 1991). In addition, in MDCK and A10 cell lines, PMA-stimulated phospholipase D is not regulated by external calcium concentrations (Huang and Cabot,

-H7

-HI

+H7

I

I

I

PA

PEt

B.

4.5[

447

1

*

0control

40t

Endothelin

+Song

L

PEt

, ,

- Song

+

Song

I

PA

Fig. 5. The effects of H, (A) (10-"M, 15 min) or sangivamycin (B) (10-"M, 15 min) upon ET-1 (10-,M, 15 min)-stimulated phospholipase D. All experiments contained 0.5% ETOH. n = 6 x 2 SEM, p* < 0.01 for A and p* < 0.005 for B.

1990). Several Ca2+-insensitivesubtypes of PKC have recently been identified and this may, in part, explain PMA-induced PKC-activation of phospholipase D a t physiological calcium levels (Bell and Burns, 1991). Very few studies have investigated the role of calcium and PKC upon agonist-stimulated phospholipase D in the same tissue. To assess ET-1-stimulated phospholipase D as either a [Ca2+li-orPKC-dependent or -independent event, it was necessary to assess 3H-PEt formation from 'H-alkyl lyso glycero-phosphocholine and exogenous ethanol in the presence of PMA andlor ionomycin as well as to assess ET-1-stimulated phospholipase D as a function of agents that antagonize or down-regulate PKC activity or clamp (Ca2+Iiat physiological concentrations. In mesangial cells, ET-stimulated phospholipase D was found to be PKC-dependent and [Ca2+1,-independent. The combined results with acute PMA administration, PKC down-regulation and PKC inhibitors strongly support a role for PKC to regulate ET-stimulated phospholipase D. However, in contrast to H,, a maximal concentration of sangivamycin did not totally reduce PEt or PA formation suggesting the possibility of a PKC-independent mechanism in the mesangial cell. This phenomenon has been observed before a s Billah et al. (1988) note that phorbol esters as

ENDOTHELIN AND PHOSPHOLIPASE

*

583

0control

n

I

-e

W

Untreated

s PKC- deple ted

Fig. 6. The effects of long-term PMA exposure upon ET-stimulated PEt formation. Mesangial cells were incubated with PMA (10-7M) or DMSO control for 18 hr at 37°C under 5% C O , and the cells were subsequently washed. Then either ET (10-7M) or control was added for 15 min a t 37°C in the presence of 0.5% ETOH. The reactions were stopped on ice with acidified methanol. n = 4,x i SEM, p* < 0.005.

lonomycin (IO,uM) Endothelm (lo-' M)

u 30 sec

15 rnin

I

PEt

Fig. 7. The effects of ionomycin (10-5M) upon phospholipase D. Ionomycin in DMSO was added for either 30 sec or 15 min ET (10 -7M) was utilized as a positive control. All experiments contained 0.5%ETOH. Alkyl PEt was used as a measure of phospholipase D activity. In data not shown, ionomycin did not significantly influence alkyl PA concentration at either time point. n = 4 x 2 SEM, p* < 0.01.

unique ET-stimulated phospholipase D activity from well as synthetic DAG species activate phospholipase phospholipase C activation. Specifically, we have previD, even though the PKC inhibitor K252A fails to in- ously demonstrated that acute PMA treatment (lO-'M, hibit this activity. Similar discrepancies were noted in 5 min) blocked upward of 90% of the ET-stimulated ovarian granulosa cells, as PKC-inhibitors diminished [Ca2+li response (Simonson and Dunn, 1991) even PMA but not gonadotropin-releasing hormone-stimu- though, in the present study, PMA still stimulated both lated phospholipase D even though PKC-down regula- basal and ET-stimulated phospholipase D. These data tion abolished agonist-stimulated phospholipase D argue for but do not prove a unique phospholipase D (Liscovitch and Amsterdam, 1989). Thus to firmly es- activity that can be dissociated from PtdIns-4,5-P2-detablish a role for protein kinase C to facilitate ET-stim- pendent phospholipase C activity. Similar results were ulated phospholipase D, we used chronic PMA treat- noted in PKC-down-regulated endothelial cells as ment to downregulate PKC activity and noted that ET bradykinin-stimulated phospholipase D was inhibited, was no longer able to stimulate PEt. Surprisingly, whereas PtdIns hydrolysis was enhanced (Martin et al., chronic PMA treatment by itself elevated basal PEt 1989). Thus, PKC may exert a dual role in regulating formation, a phenomenon observed before in PKC- ET signal transduction. PKC-mediated phosphoryladownregulated endothelial cells and granulosa cells tion may complete a negative feedback loop, inhibiting (Liscovitch and Amsterdam, 1989; Martin et al., 1989). the initial ET response (PtdIns-4,5-P2hydrolysis) while We were concerned that the elevation of basal PEt may sustaining a positive regulatory loop through the actimask E T stimulation of PEt in PKC-downregulated vation of phospholipase D. In addition, ET may also cells. Thus we confirmed these observations in a perme- stimulate a phospholipase C that hydrolyzes PtdCho, abilized (saponin-treated) mesangial cell preparation which, although not the primary signal to activate and in this PKC down-regulated preparation, GTPyS, PKC, may contribute to long-term activation of the kibut not ET, stimulates phospholipase D (M. Kester, nase and thus sustain phospholipase D activity even unpublished observation). This suggests that chronic after the initial P t d I n ~ 4 , t i - Presponse ~ has abated. In PMA exposure does not maximally stimulate basal this regard, Martinson et al. (1989) note that in astrocyphospholipase D activity and thus mask the effects of toma cells, PKC activity is required for acetylcholineET. Thus it can be argued th at PKC may, in part, min- stimulated phospholipase D, whereas calcium mobiliimize receptor-independent activation of phospholipase zation but not PKC primarily serves to regulate D. Alternatively, acute or chronic PMA exposure may acetylcholine-stimulated PtdCho-specific phospholidirectly activate basal phospholipase D activity (Cao pase C. The regulation and physiological significance of reet al., 1990). However, as the inactive phorbol ester-, 4a phorbol, does not activate phospholipase D and several ceptor-mediated phospholipase D and subsequent PA distinct activators of protein kinase C mimic the effects accumulation have not been satisfactorily described. of PMA or ET to induce phospholipase D activity, it is Phospholipase D production of PA may serve, in part, likely but not conclusively proven that the effects of as a lipid signal augmenting or sustaining ET-signal PMA administration are mediated through PKC and transduction. Exogenous application of PA or lyso PA, are probably not mediated via a direct effect upon phos- like ET, elevates [Ca2+l,,increases inositol phosphates, and induces proliferation (Kester et al., 1989; Simonson pholipase D. Acute short-term phorbol ester treatment of mesan- et al., 1989; Jalink et al., 1990; Knauss e t al., 1990). gial cells was also utilized as a tool to dissociate a However, even though phosphatidate derivatives are

KESTER ET AL.

584

401

*

*T

control

I

I80

-

Choline

H

* I

+ BAPTA

- BAPTA PEt

- BAPTA

*op

w Ethonolomine

I

PA

The of BAPTA-AM wM) upon ET (10-7M)-stimulated phospholipase D, After cell labelling and subsequent wash BAPTA-AM or DMSO control was added for 25 min and the cells then washed again, ET-1 or control in 0.5% ETOH was then added for SEM, 15 min and the lipids were extracted as described, = p** < 0.01, p* < 0.05. Fig,

Fie. 9. The effects of ET (10-7M) upon the release of watersoluble bases. Mesangial cellswere labelled with either 3H-cholineor -ethanolamine for 36 hr. After extensive washing, ET or control was added for variable times. The cells were extracted utilizing the procedure of Bligh and Dyer (1959) and the aqueous phase was evaporated and applied to silica gel 60-W TLC plates. An elution system consisting of methanol/0.5% NaCliammonia, 100:100:2, viviv separated free base from uhosuhorvlated base as well as from elvcero-DhosDhorvlbase. n = 4, x _t SEM, p* < 0.05, nonparametric-K&skal-WalliH test 1

1

"

ACKNOWLEDGMENTS predominantly the only phospholipids that can induce these events, the concentrations needed to stimulate This work was supported by National Institutes of these biological responses are certainly quite high (of- Health grants HL22563, DK41684 and AR40225 and ten 50 pg/ml and higher). Thus a more physiological the Kidney Foundation of Ohio. We thank Angela Mufunction for phospholipase D-induced PA may be to sial and Robert Ondash for excellent technical assisserve as a n additional source of endogenous diglycer- tance, and Norma Minear for preparing the manuides via a PA-phosphohydrolase activity (Martin, 1988; script. Martinson et al., 1990; Qian and Drews, 1990). However, the exact role of these unique diglyceride species LITERATURE CITED is still undefined, as PtdCho-derived diglycerides have Anthes S.C., Eckel, S., Siegel, M.I., Egan, R.W., and Billah, M.M. been shown to be effective (Slivka et al., 1988) or inef(1989) Phospholipase D in homogenates from HL-60 granulocytes: fective (Martin et al., 1990) activators of PKC. Implications of calcium and G protein control. Biochem. Biophys. Res. Comm., 163:657-664. In conclusion, ET-1 has been shown to stimulate mulE., and Dunn, M.J. (1991) Endothelin binding and receptor tiple signaling pathways in cultured rat mesangial Baldi, down regulation in rat glomerular mesangial cells. J. Pharm. Exp. cells. In addition to ET stimulation of a PtdIns-4,5Therap., 256:581-586. bisphosphate-specific phospholipase C, we have shown Bell, R.M., and Burns, D.J. (1991) Lipid activation of PKC. J. Biol. Chem., 266:46614664. that ET stimulates a PKC-regulated phospholipase D Billah, M.M., and Anthes, J.C. (1990) The regulation and cellular that hydrolyses PtdCho and PtdEth. functions of PtdCho hydrolysis. Biochem. J.,269:281-291.

Abbreviations alkyl-lyso GPC l-0-hexadecyl-2-lyso-sn-glycero-3-phosphocholine BAPTNAM bis-(o-aminophenoxyl)-ethane-N,N,N,N'-tetra-acetic acid, tetra acetoxy methyl ester [Ca"], cytosolic free calcium concentration 1,2-diacylglycerol DAG Endothelin ET 1-(5-iso-guinoIine sulfonyl)-2 methyl piperazine H7 Krebs Henseleit-4-(2-hydroxy-ethyl)-l-piperazineKHH ethane-sulfonic acid (HEPES) LPA lysophosphatidic acid mesangial cells MC phosphatidic acid PA phosphatidylethanol PEt protein kinase C PKC phorbol myristate acetate PMA phosphatidylcholine PtdCho phosphatidyl ethanolamine PtdEth phosphatidylinositol PtdIns

Billah, M.M., Pai, J.K., Mullmann, T.J., Egan, R.W., and Siegel, M.I. (1989)Regulation of phospholipase D in HL 60 granulocytes: activation by phorbol esters, diglycerides, and calcium ionophores via PKC-independent mechanisms. J. Biol. Chem., 264:9069-9076. Bligh, E.G., and Dyer, W.J. (1959)A rapid method oftotal lipid extraction and purification. Can. J. Biochem. Physiol., 37:911-917. Bocckino, S.P., Blackmore, P.F., Wilson, P.B., and Exton, J.H. (1987) Phosphatidate accumulation in hormone-treated hepatocytes via a phospholipase D mechanism. J . Biol. Chem., 262:15309-15315, Cao, Y.Z., Reddy, C., and Mastro, A.M. (1990) Evidence for PKCindependent activation of phospholipase D by phorbol esters in lymphocytes. Biochem. Biophys. Res. Commun., 171r955-962. Domino, S.E., Bocckino, S.B., and Garbers, D.L. (1989) Activation of phospholipase D by the fucose-sulfate glycoconjugate induces a n acrosome reaction in spermatozoa. J. Biol. Chem., 264:9412-9419. Griendling, K.K., Tsuda, T., and Alexander, R.W. (1989) Endothelin stimulates diacyl glycerol accumulation and activates protein kinase C in cultured vascular smooth muscle cells. J. Biol. Chem., 264:8237-8240. Huang, C., and Cabot, M.C. (1990) Phorbol esters stimulate the accumulation of PA, PEt and diglyceride in three cell types. J. Biol. Chem., 265r14858-14863. Huana, R., Kucera. G.L., and Rittenhouse. S.E. (1991) Elevated cvtosohi Ca2+ activates phospholipase D in human platelets. J . Biol. Chem., 266t1652-1655.

ENDOTHELIN AND PHOSPHOLIPASE Jalink, K., Van Corven, E.J., and Moolenaar, W.H. (1990) Lyso-PA, but not PA, is a potent Ca2’ mobilizing stimulus for fibroblasts. J. Biol. Chem., 265:12232-12239. Kester, M., Simonson, M.S., Mene’, P., and Sedor, J.R. (1989)Interleukin-1 generates transmembrane signals from phospholipids through novel pathways in cultured rat mesangial cells. J. Clin. Invest., 83t7 18-723, Kiss, Z., and Anderson, W.B. (1989) Alcohols selectively stimulate phospholipase D-mediated hydrolysis of PtdEth in NIH 3T3 cells. FEBS Lett., 257:45-48. Kiss, Z., and Anderson, W.B. (1990) ATP stimulates the hydrolysis of PtdEth in NiH 3T3 cells. J. Biol. Chem., 265t7345-7350. Knauss, T.C., Jaffer, F.E., and Abboud, H.E. (1990) PA modulates DNA synthesis, phospholipase C and PDGF mRNAs in cultured mesangial cells. J. Biol. Chem., 265t14457-14463. Liscovitch, M., and Amsterdam, A. (1989) Gonadotropin-releasing hormone activates phospholipase D in ovarian granulosa cells. J. Biol. Chem., 264:11762-11767. Lawry, O H . , Rasebrough, N.S., Farr, A.L., and Randall, R. (1981) Protein measurement with the folin phenol reagent. J . Biol. Chem., 193t265-275. MacNulty, E.E., Plevin, R., and Wakelam, M.J.O. (1990) Stimulation of the hydrolysis of PtdIns 4,5, P, and PtdCho by endothelin, a complete mitogen for Rat-1 fibroblasts. Biochem. J.,272r761-766. Marsden, P. A., Martin, E.R., Dorfman, D., Brock, T.A., Brenner, B.M., Collins, T., and Ballermann, B.J. (1989) Endothelin: gene expression, release and action in cultured cells of the renal glomerulus. Am. J. Hypertension, 2r49A. Martin, T.W. (1988) Formation of diacylglycerol by a phospholipase D-phosphatidate phosphatase pathway specific to PtdCho in endothelial cells. Biochim. Biophys. Acta, 962r282-296. Martin, T.W., Feldman, D.R., Goldstein, K.E., and Wagner, J.R. (1989)Long-term phorbol ester treatment dissociates phospholipase D activation from phosphoinositide hydrolysis and prostacyclin synthesis. Biochem. Biophys. Res. Comm., 165:319-326. Martin, T.W., Hsieh, K.P., and Porter, B.W. (1990) The sustained second phase of hormone-stimulated diacylglycerol accumulation does not activate PKC in GH, cells. J. Biol. Chem., 265~7623-7631. Martinson, E.A., Goldstein, D., and Brown, J.H. (1989) Muscarinic receptor activation of PtdCho hydrolysis. J. Biol. Chem., 264:1474% 14754. Martinson, E.A., Trilivas, I., and Brown, J.H. (1990) Rapid PKC-dependent activation of phospholipase D leads to delayed 1,2-diglyceride accumulation. J . Biol. Chem., 265r22282-22287. Mene’, P., Simonson, M.S., and Dunn, M.J. (1989) Physiology of the mesangial cell. Physiol. Rev., 69r1347-1424. Mitsushashi, T., Morris, R.C., and Ives, H.E. (1989) Endothelin-induced increases in vascular smooth muscle Ca‘+ do not depend on dihydropyridine-sensitive Ca2 channels. J . Clin. Invest., 84t635639. Moolenar, W.H., Kruiger, W., Tilly, B.C., Verlaan, I., Bierman, A.J., +

585

and de Laat, S.A. (1986) Growth factor-like action of PA. Nature, 323:171-173. Pai, S.K., Siegel, M.I., Egan, R.W., and Billah, M.M. (1988)Activation of phospholipase D by chemotactic peptide in HL-SO granulocytes, Biochem. Biophys. Res. Commun., 150:355-364. Pfeilschifter, J. (1990) Regulatory function of PKC in glomerular mesangial cells. Klin. Wochenscher, 68r1134-1137. Qian, Z., and Drews, L.R. (1990) A novel mechanism for acetylcholine to generate diacylglycerol in brain. J. Biol. Chem., 265r3607-3610. Simonson, M.S., and Dunn, M.J. (1986) Leukotriene C, and D, contract rat glomerular mesangial cells. Kidney Int., 30524-531. Simonson, M.S., and Dunn, M.J. (1990a) Cellular signaling by peptides of the endothelin gene family. FASEB J., 4:2989-3000. Simonson, M.S., and Dunn, M.J. (1990b) Endothelin-1 stimulates contraction of rat glomerular mesangial cells and potentiates p-adrenergic-mediated cyclic adenosine monophosphate accumulation. J. Clin. Invest., 85r790-797. Simonson, M.S., and Dunn, M.J. (1991) Ca” signaling by distinct endothelin peptides in glomerular mesangial cells. Exp. Cell Res., 192:148-152. Simonson, M.S., Wann, S., Mene’, P., Dubyak, G.R., Kester, M., Nakazato, Y., Sedor, J.R., and Dunn, M.J. (1989) Endothelin stimulates phospholipase C, Nai/H+ exchange, c-fos expression, and mitogenesis in rat mesangial cells. J . Clin. Invest., 83t708-712. Slivka, S.R., Meier, K.E., and Insel, P.A. (1988) 01, Adrenergic receptors promote PtdCho hydrolysis in MDCK-D, cells: A mechanism for rapid activation of PKC. J. Biol. Chem., 263r12242-12246. Striker, G.E., Killen, P.O., and Farrin, F.M. (1980)Human glomerular cells in vitro. Transplant Proc., 12 (Suppl. 1):88-99. Takai, T., and Kanfer, J.N. (1979)Partial purification and properties of a rat brain phospholipase D. J . Biol. Chem., 254:9761-9765. Tettenhorn, C.S., and Mueller, G.C. (1988) TPA activates PEt and Ptdglycerol synthesis by phospholipase D in cell Iysates. Biochem. Biophys. Res. Commun., 155:249-255. Wang, C.J., and Tai, H.H. (1988) Receptors of platelet-activating factor are coupled to PtdCho-related phospholipases C and D in rabbit platelets. FASEB J . , 2:A1575. Welsh, C.J., Schmeichel, K., Cao, H., and Chabbet, H. (1990) Vasopressin stimulates phospholipase D activity against PtdCho in vascular smooth muscle cells. Lipids, 25r675-684. Werber, H.I., Emancipator, S.N., Tykocinski, M.L., and Sedor, J.R. (1987) The interleukin-1 gene is expressed by rat glomerular mesangial cells and is augmented in immune complex glomerulonephritis. J . Immunol. 138r3207-3212. Yanagisawa,M.,Kurihara,H.,Kimura, S., Tombe, Y.,Kobayashi,M., Mitsui, Y., Yazaki, Y,, Goto, K., and Masaki, T. (1988) A novel uotent vasoconstrictor Deutide uroduced bv vascular endothelial cells. Nature, 332:411-&5’. Yavin, E. (1976) Regulation of phospholipid metabolism in differentiating cells from rat brain cerebral hemispheres in culture. J. Biol. Chem., 251t1392-1397.