Mdwrt/ar urld C‘c~//dur Gzdocrinolo. (’ lYY2 Elsevier Scientific Publishers
MQLCEL
Xh ( 1092) Y3- IO I Ireland. Ltd. 0~03.7207/~~2/$05.00
Y3
027X0
Signal transduction mechanisms involved in carbachol-induced secretion from bovine adrenal glomerulosa cells
Kq
words:
Glomeruiosa
cell: Aldosterone
secretion:
Diacylglycerol;
Carbachol:
Angiotensin
aldosterone
II
Summary In cultured bovine adrenal glomerulosa cells, diacylglycerol content remains elevated for up to 75 min following the removal of angiotensin II. This maintained increase could provide a mechanism by which angiotensin I1 pretreatment may prime cells to secrete aldosteronc in response to the calcium channel agonist Bay K 8644. In the present study we find that carbachol failed both to produce this persistent diacylglycerol elevation and to exert a priming effect. In addition, because carbachol was also a less potent activator of phospholipase D than angiotensin II, our results implicate phospholipase D in the maintained increase in diacylglycerol content observed following stimulation with and removal of angiotensin II. Carbachol also elicited changes in the radiolabeled levels of both myristateand arachidonatc-containing diacylglycerol. However, the rapid decline in diacylglycerol content following carbachol removal resembled the rapid fall in arachidonate-diacylglycerol; we therefore proposed that the diacylglycerol species generated with carbachol stimulation contains predominantly arachidonic acid. In summary, our results suggest that prolonged elevations in diacylglycerol content following removal of hormones such as angiotensin II, as well as the identity of the diacylglycerol species itself, may be important in the regulation of cellular responses.
Introduction Correspondence to: Dr. floward Rasmussen, Department of Internal Medicine, Yale University School of Medicine. P.O. Box 3333, New Haven, CT 06510, USA. This work was supported in part by grants from the National Institutes of Health Nos. DKI9Xl3 (H.R.), HL36977 (P.Q.B.) and GM07527 (W.B.B.). W.B.B. was the recipient of a predoctoral fellowship from the National Science Foundation. M.L. was the recipient of a travel grant from the Yale University-Weizmann Institute Collaboration Program. This work was performed in partial fulfillment of the requirements for a Doctorate of Philosophy (W.B.B.).
Acetylcholine, or its stable analog carbachol, stimulates aldosterone secretion from adrenal glomerulosa (AGI cells by eliciting phosphoinositide turnover (Kojima et al., 1986; Apfeldorf and Rasmussen, 1988). Thus, like the binding of angiotensin II (Ang II) to its receptor, the binding of carbachol to the cholinergic receptor activates phospholipase C (PLCI to increase inositol 1,4,5-
trisphosphatc (Ins- 1,4,5-P,) levels and trigger a transient rise in cytosolic calcium (Kojima et al., 1086) and “‘Ca cfflux from prclabeled cells (Apfeldorf and Rasmussen, 1988). Although mcasurements arc lacking, by analogy with Ang II, the other signal presumably generated by carbachol-induced PLC activation is diacylglyccrol (DAG). However, recent evidence suggests that an increase in PLC activity may not be the sole pathway by which hormones increase membrane DAG content (Loffelholz, 1989; Exton, 1990). In a number of systems phospholipase D (PLD) has also been implicated in the agonist-induced elevation of DAG (Bucckino et al.. 1987; Pai et al., IYXXa, b; Rubin, 198X; Agwu et al., 1989; Billah et al.. 1089; Liscovitch and Amsterdam, 1080; Martinson ct al., lY89). The phosphatidic acid (PA) produced by this enzyme may be converted to DAG via the action of phosphatidate hydrolasts. Indeed, we have recently demonstrated that Ang II activates PLD in bovine AG cells and that this activation, under certain conditions, can presumably contribute to the hormone-induced elevation in DAG content and stimulation of aldosterone secretion (Bollag et al., 1990). Several studies have demonstrated the importance of DAG to the aldosterone secretory response (Kojima et al., 1984, 1Y85; Barrett et al.. lY86a, b; Bollag et al., lY90. 1991). Together with the Ang II-elicited increase in calcium influx (Kojima et al.. 1985), DAG is thought to function by stimulating protein kinase C (PKC), a phospholipid-dependent, calcium-sensitive, DAGactivated protein kinase (Nishizuka, 1238). It has been proposed that a DAG-elicited increase in PKC activity is an important feature of the signailing process during sustained aldosterone secretion from bovine AG cells (Kojima et al., lY84, lY85; Barrett et al., 1986a, b; Lang and Valloton, 1987). DAG and PKC may also be involved in potentiating the secretory response of bovine adrenal glomerulosa cells with Ang II pretreatment. This hormone has the ability to prime freshly-isolated and cultured bovine glomerulosa cells to a subsequent challenge with the calcium channel agonist, Bay K 8644. Thus, Bay K 8644, which is ineffective in naive untreated cells despite increasing
calcium influx (Barrett et al., lY861, is able to induce aldosterone secretion from cells which have been pretreated with Ang II (Barrett et al., 1986a). Similarly, Bay K 8644 also maintains aldosterone secretion in Ang II-treated cells following the addition of an Ang II antagonist. We postulated that this priming effect of Ang II was related to the induction of a sustained association of PKC with the plasma membrane after Ang II removal (Barrett et al., 1986a). At this location PKC would be able to ‘read’ the enhanced calcium influx elicited by Bay K 8644 and thus activate the secretory machinery (Barrett et al., 1986a. 1988, 1989). We recently demonstrated that in cultured bovine AG cells, DAG content remains elevated for up to 75 min following the removal of Ang II and proposed that this maintained increase is the mechanism by which PKC remains associated with the plasma membrane (Bollag et al., 1SYl). Because carbachol, like Ang II, induces phosphoinositide turnover (Kojima et al., 1986; Apfeldorf and Rasmussen, 1988) and presumably DAG production and PKC activation, we investigated the ability of this agonist to prime cultured bovine AG cells to respond to a subsequent challenge with Bay K 8644. We also investigated the effect of this agonist on DAG content and PLD activation and compared its effect to that of Ang II. Our evidence suggests that the maintained elevation in DAG content following hormonal removal may be related to the priming effect of an agonist. Materials Borine
and methods
adrenul
glomerulosa
cell isolation and cul-
ture
Bovine AG cells were isolated as described previously (Foster et al., 1981). Briefly, glomeruloss cell slices were prepared from adrenal glands obtained from a local slaughterhouse, and AG cells were dispersed from collagenase-digested slices by mechanical agitation. Freshly-isolated AG cells were used immediately or, following purification on a 56% Percoll gradient, were cultured overnight in Falcon Primaria dishes (Becton Dickinson Labware, Lincoln Park, NJ, USA) in a Dulbecco’s modified Eagle’s medium
95
(DMEM)/Ham’s F12 medium (1: 1) containing 10% horse serum (v/v>, 2% fetal bovine serum (v/v), ascorbate (100 PM), a-tocopherol (1.2 PM), Na,SeO, (0.05 PM), butylated hydroxyanisole (50 FM), metyrapone (5 FM), penicillin (100 U/ml>, streptomycin (100 pg/ml), gentamicin (30 pg/ml) and amphotericin B (3 pg/ml), as described (McCarthy et al., 1990). After replacement of the serum-containing medium with serum-free medium ( + 0.2% bovine serum albumin (BSA) f [“Hloleic acid), the cells were incubated for an additional 20-24 h before use. In some experiments ticarcillin and amikacin (200 pug/ml) were substituted for gentamicin. Secretion experiments in cultured adrenal glomerulosa cells Prior to stimulation, cells were incubated at 37°C in Krebs-Ringer buffer (KRB) with 5% CO, (containing also 2.5 mM Na acetate and 0.002% phenol red) for a control period of 30 min. The supernatants were collected, and KRB containing the appropriate agents was added. Thereafter, at the indicated times (every 10 or 20 min) supernatants were collected and replaced with KRB containing the indicated agent. All supernatants were stored at -20°C until assayed. At the conclusion of each experiment, cells were solubilized in 0.4 N NaOH and stored at -20°C for protein determination. Determination of membrane diacylglycerol content Cells were preincubated at 37°C in KRB (+ 2.5 mM Na acetate + 0.002% phenol red) for 30 min in 5% CO,. The appropriate agents were then added at staggered time intervals such that, after the indicated incubation periods, all samples were terminated at the same time by the addition of 0.2% sodium dodecyl sulfate (SDS). Neutral lipids were then extracted as described previously (Bollag et al., 1991) and DAG content was determined using a slight modification of the DAG kinase method of Preiss et al. (1986). Measurements of phosphatidic acid and phosphatidylethanol production Cells were labeled in serum-free medium containing 0.2% BSA (fraction V or in some experiments fatty-acid free) and 5 pCi/ml [3H]oleic
acid for 20-24 h. After washing and a 30 min preincubation period, cells were stimulated with the appropriate agonists and incubated an additional 30 min in the presence of 0.5% ethanol. Incubations were terminated using ice-cold methanol and the methanol-precipitated cells scraped from the dishes. Lipids were extracted into chloroform/methanol to which a solution of 0.1 N HCI (fl mM EDTA) was added (1:2:1 v/v/v final ratio). Aliquots of the lower phases were collected and dried under N,. Samples were resuspended in chloroform/methanol (containing also 20 pg phosphatidylethanol (PEt) and phosphatidic acid (PA) per sample) and spotted onto silica gel 60 or LK6 Whatman (prerun in 1% oxalate and heat-activated) thin-layer chromatography plates. Plates were developed in a solvent system of ethyl acetate/ isooctane/ acetic acid/ water (13 : 2: 3 : 10) and were visualized with iodine vapor and with fluorography using En’Hance. Spots corresponding to PA and PEt were quantified as above. Determination of radiolabeled DAG lel’els and distribution of radiolabel in phospholipids Cells were washed and placed in KRB (+2.5 mM Na acetate + 0.002% phenol red) containing 5 pCi/ml [“Hlarachidonic acid and [ ‘“Clmyristic acid. After a 2 h labeling period, the appropriate agents were added at staggered time intervals such that, after the indicated time periods, all samples were terminated at the same time by the addition of 0.2% SDS. Small aliquots of the SDS-solubilized cells were saved for protein determination and the remainder was extracted for l-2 h with ice-cold chloroform/methanol. Additional chloroform and 0.2 M NaCl were added to break phase, and the lower phases collected and evaporated under N,. Samples were resuspended in chloroform/methanol and spotted onto heatactivated silica gel 60 thin-layer chromatography plates (0.25 mm thickness with concentrating zone; Merck, Darmstadt, Germany). Plates were developed in a solvent system of benzene/ethyl acetate (70 : 30) for separation of 1,2-DAG and in chloroform/ methanol/ acetic acid/water (65 : 43 : 1 : 3) for separation of phospholipids. Lipids were visualized with autoradiography using En”Hance (New England Nuclear, Boston, MA,
USA). Spots corresponding to 1,2-DAG or various phospholipids were identified by comigration with authentic standards, and radioactivity was quantified using a double-label protocol. Materials Collagenase was purchased from BoehringerMannheim Biochemicals (Indianapolis, IN, USA). [9,10(n)-3H]Oleic acid, [5,6,8,9,1 l,12,14,1S-3H]arachidonic acid and adenosine 5’-[ -$‘P]triphosphate were obtained from Amersham (Arlington Heights, IL, USA). [l-“ClMyristic acid was from New England Nuclear (Boston, MA, USA). Diacylglycerol kinase was purchased from Lipidex (Middletown, WI, USA); Percoll from Pharmacia (Piscataway, NJ, USA). Fetal bovine serum, horse serum, penicillin, and streptomycin were obtained from Gibco (Grand Island, NY, USA); all other tissue culture reagents were from Sigma (St. Louis, MO, USA). Bay K 8644 was a generous gift of Dr. Alexander Scriabine, Miles Institute for Preclinical Pharmacology (West Haven, CT, USA). Aldosterone was assayed using a solid-phase radioimmunoassay (Diagnostic Products, Los Angeles, CA, USA). Phosphatidylethanol standard was prepared according to the method of Eibl and Kovatchev (1981). Angiotensin II and carbachol were from Sigma (St. Louis, MO, USA). Protein was determined by the method of Lowry et al. (19511 using BSA as standard. Results Carbachol exerts no priming effect on cultured holine AG cells In freshly-isolated and cultured bovine AG cells pretreated with Ang II, the calcium channel agonist Bay K 8644 slows the rate at which aldosterone secretion decays following the removal of the hormone (Barrett et al., 1986a; Bollag et al., 1991). Thus, the time required for the secretory rate to attain a half-maximal value after Ang II removal is approximately 2.5-fold greater in the presence of Bay K 8644 than in its absence (Barrett et al., 1986a; Bollag et al., 1991). Because Ang II and carbachol utilize similar signal transduction pathways in adrenal cells (Hadjian et al., 1984; Kojima et al., 1986; Apfeldorf and Rasmusscn, 19881, the effect of Bay K 8644 in carba-
3oo1
Carbac;,l
$
‘gBSaqYqK +
---+
200-
100 -
04
,
,
20
40
60
1
,
80
(
100
120
Time (minutes) Fig.
1.
Bay
secretory
K X634
is unable
lation for SO min with 100 PM repeatedly
to prolong
response in carbachol-pretreated carhachol.
the
aldosteronc
cells. After cultured
ctimu-
cells wet-c
washed with KRB in the absence (open squares) or
presence (closed circles) of IO0 nM Bay K 8h-14. Aldoatzronc secretory
rates
were
measured
and
arc
expressed
as the
means (k SE.) of a minimum of five determinations
from three
separate
relative
experiments:
* p < i).Ol.
* * p 5 0.0002
to
control.
chol-treated cultured AG cells was examined. As shown in Fig. 1, the addition of carbachol (100 PM) to cultured bovine AG cells resulted in a peak increase (at 20 min of stimulation) in aldosterone secretion to a rate of 244 + 40 pg/min/ mg protein, a value approximately 14-fold over the control secretory rate of 18 k 2 pg/min/mg protein. Subsequently, the aldosterone secretory rate declined to a value of 170 k 32 pg/min/mg protein over the next 20 min. When the agonist was then removed, aldosterone secretion rapidly declined to control values within approximately 20 min both in the absence and in the prcscncc of 100 nM Bay K 8644. Similar results were obtained using tither a greater concentration of carbachol (1 mM), of Bay K 8644 (1 PM), or of both carbachol and Bay K 8644 (data not shown). Thus, in contrast to Ang II, carbachol appears unable to prime cultured bovine AG cells. Carbachol induces a sustained eler.ation in DAG content It should be noted that aldosterone secretion from cells responding to carbachol. unlike from
those responding to Ang II, was not completely sustained, an observation which has been made previously (Kojima et al., 1986). Whereas Ang II induces a monotonic rise in the secretory rate, the increase induced by carbachol was biphasic, consisting of a peak at approximately 20 min of stimulation which subsequently fell to a somewhat reduced value (Fig. 1). The biphasic nature of the carbachol-elicited aldosterone secretory response has previously been attributed to a smaller enhancement of calcium influx induced by carbachol (Kojima et al., 1986). However, alternatively elevated DAG levels might not be sustained. We therefore investigated the effect of carbachol on DAG content in the cultured cells. The effect of carbachol on cellular DAG content is shown in Fig. 2. Carbachol elicited an approximate 75% increase in DAG content which was sustained in the continued presence of the hormone. Thus, elevated DAG levels are present throughout the 60 min stimulation period despite declining aldosterone secretory rates. Neverthe-
20
1
ff
ff
PA
less, the 75% increase in DAG content carbachol is less than the approximate rise induced by Ang II.
0
20
40
Time
60
(minutes)
Fig. 2. Carbachol increases DAG content. Cultured cells were incubated for the indicated times in the presence of 100 PM carbachol and the stimulation was terminated at the indicated times by the addition of SDS. Neutral lipids were extracted and DAG was measured by the DAG kinase method. Values are expressed as -fold over control and represent the mean (*SE) of at least four determinations from two separate experiments. Control DAG content was 0.49kO.04 nmol/mg protein for the 15 and 30 min time points and 0.44+0.01 nmol/mg protein for the 60 min time point; * p < 0.02, * * p 5 0.0002 relative to control.
PEt
Fig. 3. Carbachol increases the levels of radiolabeled phosphatidic acid (PA) and phosphatidylethanol (PEt). Cultured bovine AG cells were prelabeled for 20-24 h with IO PCi [‘Hloleic acid and stimulated for 30 min with 100 PM carbachol or 10 nM Ang II in the presence of 0.5% ethanol. Phospholipids were extracted and chromatographed as described in Materials and methods. Open bars represent control conditions; striped bars represent stimulation with carbachol; and hatched bars represent stimulation with Ang II. Values are the means ( f SE) of five determinations from two separate experiments; * p 5 0.02. * * p I 0.0005 versus control values: o p I 0.0001 relative to Ang II-stimulated values, Control PA and PEt levels were 8lO+ 129 cpm/dish and 624 k 73 cpm/dish. respectively.
elicited by 200-250%
Carbachol activates phospholipase D Because Ang II activates PLD (in addition to PLC) and this activation can presumably contribute to hormone-elicited elevations in DAG content (Bollag et al., 19901, the ability of carbachol to stimulate PLD activity was subsequently examined. PLD has the unique characteristic not only of catalyzing the hydrolysis of phospholipids but also, in the presence of small amounts of ethanol, of effecting the ethanolysis of these compounds. As shown in Fig. 3, treatment of [ 3H]oleate-prelabeled cells with carbachol induced an elevation in the levels of both PA and PEt, suggesting that carbachol does indeed activate PLD. However, the carbachol-elicited increase in radiolabeled PEt levels (to 1.48- * 0.09-
fold over control) represented only approximately 60% of the change induced by Ang II (l.SO-+ 0. l-fold over control); whereas carbachol elevated PA levels to only about 24% (to I. 16- + 0.04-fold over control) of those elicited by Ang II stimulation (1.67-i 0.04-fold over control). Thus, although carbachol appears to stimulate PLD activity, this hormone is less potent than Ang II in inducing increases in both DAG content and PA levels. DAG content declines rapidly upon carbachol remoi,al The rate of decline in DAG content after carbachol removal was then investigated. Again carbachol induced an approximate 80% rise in DAG content at 60 min of stimulation (Fig. 4A). The subsequent removal of the hormone resulted in a rapid decline in DAG content to attain a value of 1.36-f 0.04-fold over control by 30 min after carbachol removal. The effect of the removal of Ang II (in the presence or absence of Ang II antagonists) on DAG content is shown in Fig. 4B for comparison. Ang II elicited a larger elevation in DAG content than carbachol, raising DAG content to 2.43-k 0.12-fold over control after 60 min of stimulation. Upon removal of the hormone, DAG content decreased slowly, declining only approximately 24% (to a value of 2.09- f 0.13-fold over control) within the 30 min washout period; similar results have been reported previously (Bollag et al., 1991). Thus, these results support a prior suggestion that the priming effect may be mediated by a prolonged elevation in DAG content following hormonal removal. Carbachol elelsates rnyristate-containing DAG ler?els The maintenance of the Ang II-induced elevation in DAG content may be related to the production of a species of DAG which persists upon the inhibition of Ang II action (Bollag et al., 1991). Thus, in cells labeled with both [‘Hlarachidonic acid and [ “Clmyristic acid, myristatecontaining DAG levels remain elevated while the levels of DAG containing arachidonate decline rapidly upon the addition of an Ang II antagonist (Bollag et al., 1991). The effect of carbachol upon the production of these two species of DAG was
A
0
B
Time 3
15
after
Carbachol (minutes)
30
Removal
.. E
I
T
kme
after &List (minutes)
Rem02
Fig. 4. DAG content declines rapidly following the removal of carbachol. After stimulation for 60 min with (A) 100 FM carbachol (striped bars) or (R) IO nM Ang II (hatched bars). cultured cells were repeatedly washed and incubated for the indicated times with KRB or in (El with KRB+ IO PM Ang II antagonist. Values are expressed as -fold over control and represent the mean of at least four determinations from two separate experiments; (A I * p < 0.005, * * p 5 0.0001 relative to control values; (El * p < 0.005, * * p 5 0.0001 versus control; l p < 0.02, c p s 0.0002 relative to Ang II-stimulated values. Control DAG content was 0.74kO.05 nmol/mg protein.
examined. As shown in Table 1, carbachol creased the levels of both arachidonatemyristate-containing DAG by approximately
inand 70
99
TABLE
1
ANGIOTENSIN II AND CARBACHOL ELEVATE [‘H]ARACHIDONATEAND [“C]MYRISTATE-CONTAINING DAG LEVELS ”
Control Ang II (10 nM) Carbachol c100 mM)
[‘HlArachidonate DAG
[ ‘“ClMyristate
1.OOi 0.04 h 3.26 & 0.04 *
1.oo + 0.04 2.7hk0.2 *
l.h8*0.07
I .79 +o.on *,O
*.O
DAG
” Cultured cells were labeled for 2 h with [‘Hlarachidonate and [‘“Clmyristate. In the continued presence of radiolabel, cells were incubated for 30 min with 10 nM Ang II or 100 mM carbachol. Samples were terminated by the addition of SDS and processed as in Materials and methods. ” Data are expressed as -fold over control with a control value of 12.510+4266 ‘H dpm/mg of protein (l.OO-?0.04-fold over control) and 75.099k 18,538 “C dpm/mg protein t I .OO-i 0.04.fold over control). Values represent the means ( +SE) of four determinations from two separate experiments: * p < 0.0005 relative to control values; Op < 0.0005 relative to Any II-stimulated values.
and SO%, respectively. Nevertheless, carbachol elevated radiolabeled DAG levels of both species to a lesser extent than did Ang II, which increased arachidonateand myristate-containing DAG levels approximately 3.3-fold and 2.X-fold, respectively. In addition, it should be noted that, as with Ang II and an Ang II antagonist (Bollag et al., 19911, when atropine was added to carbachol-treated cells, arachidonate-containing DAG levels declined while myristate-containing DAG levels remained elevated (data not shown). Discussion The results of this study have confirmed a previous report (Kojima et al., 1986) that in AG cells carbachol induces a biphasic increase in the secretory rate, with an initial peak of aldosterone secretion followed by a subsequent decline to a reduced plateau level (Fig. 1). Kojima et al. (1986) attributed the biphasic nature of the carbacholelicited aldosterone secretory response to the smaller enhancement of calcium influx induced by carbachol. We have demonstrated here that the fall in aldosterone secretion is not the result of a decrease in DAG levels, as the carbacholelicited elevation in DAG content was sustained
in the continued presence of the hormone (Fig. 2). Therefore, as suggested by Kojima et al. (1986), the biphasic aldosterone secretory response induced by carbachol is likely the result of an insufficient carbachol-elicited elevation of calcium influx. Although an elevated DAG content persisted in the continued presence of carbachol, upon removal of the hormone DAG content declined quite rapidly to a near-control value within 30 min (Fig. 4A). This result is in contrast to our observation that following pretreatment with and removal of Ang II, DAG content remains elevated for a prolonged time period (Bollag et al., 1991; and Fig. 4B). As discussed previously, the persistent elevation in DAG after Ang II removal is likely the result of differences in the species of DAG, i.e. related to the fatty-acid composition, generated in response to the hormone (Bollag et al., 1991). Thus, we demonstrated that the species which remains elevated following Ang II removal contains myristate, while the levels of arachidonate-containing DAG decline rapidly upon removal of the hormone (Bollag et al., 1991). We found that carbachol also elevated levels of both radiolabeled myristateand arachidonate-containing DAG, although not to as great an extent as Ang II (Table 1). Because the labeling protocol used does not result in isotopic equilibrium, no conclusion concerning the absolute levels of either species can be reached. However, the carbachol-induced increase in DAG content declined quite rapidly upon agonist removal, with a time course similar to that obtained for [“Hlarachidonate-DAG, suggesting that the majority of DAG produced upon carbachol stimulation is of the rapidly turning-over arachidonate-containing type rather than the persistent myristate-DAG. Nevertheless, it should be noted that due to the relative affinity of the two ligands for their receptors, the unbinding of carbachol from its receptor should be more rapid than that of Ang II. Thus, the disappearance of the second messengers generated by Ang II might be expected to occur at a lower rate than those produced in response to carbachol. In order to verify our hypothesis, therefore, further experiments are necessary to determine the absolute amounts of the slowlyand rapidly-turning over DAG pool.
IO0
Although the mechanism for the persistence of myristate-containing DAG is unknown, we have proposed an involvement of PLD activity in the maintained increase in myristate-containing DAG via a positive feedback effect (Exton, 1990). Such a feed-forward mechanism could be initiated as a result of the activation of PLD by protein kinasc C (PKC), presumably stimulated via phosphoinositide turnover and DAG production (Liscovitch et al., 1987; Cabot et al., 1988, 1989; Billah et al., 1989; Liscovitch, 1989; Liscovitch and Amsterdam, 1989), as described previously (Bollag et al., 1991). Because in some systems PLD activity has been reported to hydrolyze preferentially (myristate-containing) phosphatidylcholine (Martin, 19881, the synthesis of myristate-DAG could persist, through the functioning of PLD, despite an immediate termination of phosphoinositidespecific PLC activation and the production of arachidonate-containing DAG. Martinson et al. ( 1990) also found a persistent, delayed elevation in DAG content in 1321Nl astrocytoma cells following stimulation with carbachol and inhibition with atropine. These investigators proposed that the lag associated with DAG formation was the result of slow conversion of PLD-generated PA to DAG by phosphatidate hydrolases (Martinson et al., 1990). We report in the present study that carbachol activated PLD with a considerably reduced potency relative to Ang II (Fig. 31. In particular, the elevation in radiolabeled PA levels elicited by carbachol is only approximately 24%, while the increase in PEt levels is 60%, of that induced by Ang II. Therefore, carbachol may be incapable of stimulating PLD sufficiently to engage the feedforward response such that PLD-induced DAG production cannot be maintained without the continued generation of DAG from the phosphoinositides. It should be noted as well that PEt is a novel phospholipid which is not metabolized by the cellular machinery (Billah et al., 1989; Liscovitch, 1989). Thus, the differential carbachol-induced increase in PA and PEt levels, relative to Ang II, may be the result of transient, rather than sustained, activation of PLD, with subsequent metabolism of the generated PA. However, further studies arc needed to explore this possibility, especially in view of the recent finding that tran-
sient PLD activation can result in a maintained DAG production (Martinson et al., 1990). Ang II may also activate a signalling pathway in addition to those activated by carbachol, e.g. a PC-specific PLC as in astrocytoma cells (Martinson et al., 1989); this possibility too requires further investigation. Alternatively, the maintained elevation in myristate-containing DAG levels following Ang II removal may be the result of the slower metabolism of this species of DAG versus arachidonate-containing DAG, as discussed previously (Bollag et al., 1991) Thus in this scenario as well, if carbachol induces the production of predominantly the rapidly metabolized arachidonateDAG, DAG levels would decline quickly upon removal of the hormone. In addition, it is possible that the rate of DAG metabolism is maximal at that level of production induced by carbachol. Therefore, because Ang II elicits a greater DAG production, DAG levels might remain elevated for longer periods of time following Ang II rcmoval. The rapid decline of DAG content following the inhibition of carbachol action (Fig. 4) may explain the inability of this hormone to prime cells to respond to the calcium channel agonist Bay K 8644 with aldosterone secretion (Fig. I). According to our hypothesis, priming is the result of a persistent elevation in DAG content (Bollag et al., 1991); this DAG provides one of the critical signals required for aldosterone secretion, the other of which is an enhancement of calcium influx which can be elicited by Bay K 8644 (Barrett et al., 1986a). Ang II, because it generates such persistent DAG, is thus able to prime cells to respond to the second signal provided by Bay K 8644. Carbachol, however, elicits no such maintained elevation in DAG levels and thus is unable to prime cells to secrete aldosterone in response to the calcium channel agonist. It is possible, in fact, that a threshold level of persistent DAG is required for the stimulation of aldosterone production. In this case, the inability of carbachol to prime glomerulosa cells is the result of the rapidity with which DAG content declines below this threshold following carbachol removal. However, further research is necessary to determine if such a threshold DAG level exists.
101
Thus, the results of the present study together with our previous data suggest that the priming effect of Ang II may be related to its ability to elicit a persistent elevation in DAG content following removal of the hormone. Carbachol fails either to produce this persistent DAG elevation or to exert a priming effect. In turn, this priming effect could provide a mechanism by which, with repeated Ang II stimulation, a second sustained aldosterone secretory rate could be initiated more effectively by a different agonist, or perhaps more rapidly or by a lower concentration of Ang II than might othcrwisc be necessary. Thus, the priming ability of Ang II may be physiologically relevant and contribute to its effectiveness as a primary regulator of aldosterone secretion. Acknowledgements We would like to express our appreciation for the expert technical assistance of Laura Kiernan and John Nee. References Agwu. D.E.. McPhail. L.C.. Chabot, M.C., Daniel. L.W., Wykle, R.L. and McCall. C.E. (IYXY) J. Biol. Chem. 764, 1405~1413. Apl’eldorf, W.J. and Rasmussen. H. (IYXX) Cell Calcium 9, 71-80. Barrett, P.Q.. Kojima. I.. Kojima, K.. Zawalich, K., Isales, C.M. and Rasmussen, H. (19X6a) Biochem. J. 23X,905-912. Barrett, P.Q.. Kojima. I.. Kojima, K.. Zawalich. K., Isales, C.M. and Rasmussen, H. (198hb) Biochem. J. 238. XY3~903. Barrett, P.Q.. Bollag, W.B. and Rasmussen. H. (19881 Hormones and their Actions (Cooke. B.A.. King, R.J.B. and van der Molen, H.J.. Part 2, pp. 21 l-229, eds.). Elsevier. Amsterdam. Barrett. P.Q., Bollag. W.B.. Isales. C.M., McCarthy, R.T. and Rasmussen. H. (19x01 Endocr. Rev. IO, 496-518. Billah. M.M.. Pai. J.-K., Mullman. T.J., Egan, R.W. and Siegel, M.I. (19X91 J. Biol. Chem. 264, OtlnY-9076.
Bocckino. S.B., Blackmore, P.F., Wilson, P.B. and Exton. J.II. (19871 J. Biol. Chem. 262, 153OY-15315. Bollag. W.B., Barrett, P.Q., Isales, C.M. and Rasmussen, H. (19901 Endocrinology 127, 14361443. Bollag. W.B.. Barrett, P.Q.. Isales, C.M. and Rasmussen. II. (IYYI) Endocrinology 12X. 231-241. Cabot. M.C.. Welsh. C.J.. Cao, H.-t. and Chabbott, H. (IYXXl FEBS Lett. 233. 153-157. Cabot. M.C.. Welsh. C.J., Zhang. Z.-c. and Cao, H.-t. (19X91 FEBS Lett. 245. X5-90. Eibl. H. and Kovatchev, S. (IYXI) Methods Enzymol. 72, h32~h3Y. Exton, J.H. (19901 J. Biol. Chem. 265, l-4. Foster. R.. Lobo. M.V., Rasmussen, H. and Marusic, E. t 19X1) Endocrinology 109. 2 lYh~220 I, Hadjian, A.J., Culty. M. and Chambaz. E.M. (1984) Biochim. Biophys. Acta X04, 427-433. Kojima. 1.. Kojima. K.. Kreutter, D. and Rasmussen. H. (19841 J. Biol. Chem. 259, 1444X-14457. Kojima. 1.. Kojima. K. and Rasmussen. H. (1985) J. Biol. Chem. 200, 9177-9184. Kojima. I., Kojima, K.. Shibata, II. and Ogata, E. (19861 Endocrinology 1 IY. 2X4-291. Lang, U. and Valloton, M.B. (19x71 J. Biol. Chem. 262, x047-X050. Liscovitch, M. (19XYl J. Biol. Chem. 264. 14SO-1456. Liacovitch, M. and Amsterdam. A. (19891 J. Biol. Chem. 264. 117h2-11767. Liacovitch, M.. Blusztajn. J.. Freeae. A. and Wurtman. R.J. tlYX71 Biochem. J. 241. Xl-Xh. Loffelholz. K. (19XYl Biochem. Pharmacol. 38, 1543-1540. Lowry. O.H.. Rosebrough. N.J.. Farr, A.L. and Randall, R.J. (19511 J. Biol. Chem. 193, 265-275. Martin, T.W. tlYX81 Biochim. Biophys. Acta Y62, 2X2-296. Martinson, E.A.. Goldstein, D. and Brown, J.H. (19891 J. Biol. Chem. 264, 14748X 14754. Martinson. E.A.. Trilivas. I. and Brown, J.H. (1990) J. Biol. Chem. 76.5, 222X2-22287. McCarthy. R., Isales. C., Bollag. W., Rasmussen, H. and Barett. P. (IYYO) Am. J. Physiol. 25X, F473F478. Nishizuka. Y. tlYXX1 Nature 334, 66ll665. Pai. J.-K.. Siegel. M.I., Egan, R.W. and Billah, M.M. (IYXXal Biochem. Biophys. Res. Commun. 150, 355-364. Pai, J.-K.. Siegel, M.I., Egan, R.W. and Billah. M.M. (19XXb) J. Biol. Chem. 263, 12472-12477. Rubin. R. (198X) Biochem. Biophys. Res. Commun. 156, 1090~ 1096.
MQLCEL
Xh ( 1092) Y3- IO I Ireland. Ltd. 0~03.7207/~~2/$05.00
Y3
027X0
Signal transduction mechanisms involved in carbachol-induced secretion from bovine adrenal glomerulosa cells
Kq
words:
Glomeruiosa
cell: Aldosterone
secretion:
Diacylglycerol;
Carbachol:
Angiotensin
aldosterone
II
Summary In cultured bovine adrenal glomerulosa cells, diacylglycerol content remains elevated for up to 75 min following the removal of angiotensin II. This maintained increase could provide a mechanism by which angiotensin I1 pretreatment may prime cells to secrete aldosteronc in response to the calcium channel agonist Bay K 8644. In the present study we find that carbachol failed both to produce this persistent diacylglycerol elevation and to exert a priming effect. In addition, because carbachol was also a less potent activator of phospholipase D than angiotensin II, our results implicate phospholipase D in the maintained increase in diacylglycerol content observed following stimulation with and removal of angiotensin II. Carbachol also elicited changes in the radiolabeled levels of both myristateand arachidonatc-containing diacylglycerol. However, the rapid decline in diacylglycerol content following carbachol removal resembled the rapid fall in arachidonate-diacylglycerol; we therefore proposed that the diacylglycerol species generated with carbachol stimulation contains predominantly arachidonic acid. In summary, our results suggest that prolonged elevations in diacylglycerol content following removal of hormones such as angiotensin II, as well as the identity of the diacylglycerol species itself, may be important in the regulation of cellular responses.
Introduction Correspondence to: Dr. floward Rasmussen, Department of Internal Medicine, Yale University School of Medicine. P.O. Box 3333, New Haven, CT 06510, USA. This work was supported in part by grants from the National Institutes of Health Nos. DKI9Xl3 (H.R.), HL36977 (P.Q.B.) and GM07527 (W.B.B.). W.B.B. was the recipient of a predoctoral fellowship from the National Science Foundation. M.L. was the recipient of a travel grant from the Yale University-Weizmann Institute Collaboration Program. This work was performed in partial fulfillment of the requirements for a Doctorate of Philosophy (W.B.B.).
Acetylcholine, or its stable analog carbachol, stimulates aldosterone secretion from adrenal glomerulosa (AGI cells by eliciting phosphoinositide turnover (Kojima et al., 1986; Apfeldorf and Rasmussen, 1988). Thus, like the binding of angiotensin II (Ang II) to its receptor, the binding of carbachol to the cholinergic receptor activates phospholipase C (PLCI to increase inositol 1,4,5-
trisphosphatc (Ins- 1,4,5-P,) levels and trigger a transient rise in cytosolic calcium (Kojima et al., 1086) and “‘Ca cfflux from prclabeled cells (Apfeldorf and Rasmussen, 1988). Although mcasurements arc lacking, by analogy with Ang II, the other signal presumably generated by carbachol-induced PLC activation is diacylglyccrol (DAG). However, recent evidence suggests that an increase in PLC activity may not be the sole pathway by which hormones increase membrane DAG content (Loffelholz, 1989; Exton, 1990). In a number of systems phospholipase D (PLD) has also been implicated in the agonist-induced elevation of DAG (Bucckino et al.. 1987; Pai et al., IYXXa, b; Rubin, 198X; Agwu et al., 1989; Billah et al.. 1089; Liscovitch and Amsterdam, 1080; Martinson ct al., lY89). The phosphatidic acid (PA) produced by this enzyme may be converted to DAG via the action of phosphatidate hydrolasts. Indeed, we have recently demonstrated that Ang II activates PLD in bovine AG cells and that this activation, under certain conditions, can presumably contribute to the hormone-induced elevation in DAG content and stimulation of aldosterone secretion (Bollag et al., 1990). Several studies have demonstrated the importance of DAG to the aldosterone secretory response (Kojima et al., 1984, 1Y85; Barrett et al.. lY86a, b; Bollag et al., lY90. 1991). Together with the Ang II-elicited increase in calcium influx (Kojima et al.. 1985), DAG is thought to function by stimulating protein kinase C (PKC), a phospholipid-dependent, calcium-sensitive, DAGactivated protein kinase (Nishizuka, 1238). It has been proposed that a DAG-elicited increase in PKC activity is an important feature of the signailing process during sustained aldosterone secretion from bovine AG cells (Kojima et al., lY84, lY85; Barrett et al., 1986a, b; Lang and Valloton, 1987). DAG and PKC may also be involved in potentiating the secretory response of bovine adrenal glomerulosa cells with Ang II pretreatment. This hormone has the ability to prime freshly-isolated and cultured bovine glomerulosa cells to a subsequent challenge with the calcium channel agonist, Bay K 8644. Thus, Bay K 8644, which is ineffective in naive untreated cells despite increasing
calcium influx (Barrett et al., lY861, is able to induce aldosterone secretion from cells which have been pretreated with Ang II (Barrett et al., 1986a). Similarly, Bay K 8644 also maintains aldosterone secretion in Ang II-treated cells following the addition of an Ang II antagonist. We postulated that this priming effect of Ang II was related to the induction of a sustained association of PKC with the plasma membrane after Ang II removal (Barrett et al., 1986a). At this location PKC would be able to ‘read’ the enhanced calcium influx elicited by Bay K 8644 and thus activate the secretory machinery (Barrett et al., 1986a. 1988, 1989). We recently demonstrated that in cultured bovine AG cells, DAG content remains elevated for up to 75 min following the removal of Ang II and proposed that this maintained increase is the mechanism by which PKC remains associated with the plasma membrane (Bollag et al., 1SYl). Because carbachol, like Ang II, induces phosphoinositide turnover (Kojima et al., 1986; Apfeldorf and Rasmussen, 1988) and presumably DAG production and PKC activation, we investigated the ability of this agonist to prime cultured bovine AG cells to respond to a subsequent challenge with Bay K 8644. We also investigated the effect of this agonist on DAG content and PLD activation and compared its effect to that of Ang II. Our evidence suggests that the maintained elevation in DAG content following hormonal removal may be related to the priming effect of an agonist. Materials Borine
and methods
adrenul
glomerulosa
cell isolation and cul-
ture
Bovine AG cells were isolated as described previously (Foster et al., 1981). Briefly, glomeruloss cell slices were prepared from adrenal glands obtained from a local slaughterhouse, and AG cells were dispersed from collagenase-digested slices by mechanical agitation. Freshly-isolated AG cells were used immediately or, following purification on a 56% Percoll gradient, were cultured overnight in Falcon Primaria dishes (Becton Dickinson Labware, Lincoln Park, NJ, USA) in a Dulbecco’s modified Eagle’s medium
95
(DMEM)/Ham’s F12 medium (1: 1) containing 10% horse serum (v/v>, 2% fetal bovine serum (v/v), ascorbate (100 PM), a-tocopherol (1.2 PM), Na,SeO, (0.05 PM), butylated hydroxyanisole (50 FM), metyrapone (5 FM), penicillin (100 U/ml>, streptomycin (100 pg/ml), gentamicin (30 pg/ml) and amphotericin B (3 pg/ml), as described (McCarthy et al., 1990). After replacement of the serum-containing medium with serum-free medium ( + 0.2% bovine serum albumin (BSA) f [“Hloleic acid), the cells were incubated for an additional 20-24 h before use. In some experiments ticarcillin and amikacin (200 pug/ml) were substituted for gentamicin. Secretion experiments in cultured adrenal glomerulosa cells Prior to stimulation, cells were incubated at 37°C in Krebs-Ringer buffer (KRB) with 5% CO, (containing also 2.5 mM Na acetate and 0.002% phenol red) for a control period of 30 min. The supernatants were collected, and KRB containing the appropriate agents was added. Thereafter, at the indicated times (every 10 or 20 min) supernatants were collected and replaced with KRB containing the indicated agent. All supernatants were stored at -20°C until assayed. At the conclusion of each experiment, cells were solubilized in 0.4 N NaOH and stored at -20°C for protein determination. Determination of membrane diacylglycerol content Cells were preincubated at 37°C in KRB (+ 2.5 mM Na acetate + 0.002% phenol red) for 30 min in 5% CO,. The appropriate agents were then added at staggered time intervals such that, after the indicated incubation periods, all samples were terminated at the same time by the addition of 0.2% sodium dodecyl sulfate (SDS). Neutral lipids were then extracted as described previously (Bollag et al., 1991) and DAG content was determined using a slight modification of the DAG kinase method of Preiss et al. (1986). Measurements of phosphatidic acid and phosphatidylethanol production Cells were labeled in serum-free medium containing 0.2% BSA (fraction V or in some experiments fatty-acid free) and 5 pCi/ml [3H]oleic
acid for 20-24 h. After washing and a 30 min preincubation period, cells were stimulated with the appropriate agonists and incubated an additional 30 min in the presence of 0.5% ethanol. Incubations were terminated using ice-cold methanol and the methanol-precipitated cells scraped from the dishes. Lipids were extracted into chloroform/methanol to which a solution of 0.1 N HCI (fl mM EDTA) was added (1:2:1 v/v/v final ratio). Aliquots of the lower phases were collected and dried under N,. Samples were resuspended in chloroform/methanol (containing also 20 pg phosphatidylethanol (PEt) and phosphatidic acid (PA) per sample) and spotted onto silica gel 60 or LK6 Whatman (prerun in 1% oxalate and heat-activated) thin-layer chromatography plates. Plates were developed in a solvent system of ethyl acetate/ isooctane/ acetic acid/ water (13 : 2: 3 : 10) and were visualized with iodine vapor and with fluorography using En’Hance. Spots corresponding to PA and PEt were quantified as above. Determination of radiolabeled DAG lel’els and distribution of radiolabel in phospholipids Cells were washed and placed in KRB (+2.5 mM Na acetate + 0.002% phenol red) containing 5 pCi/ml [“Hlarachidonic acid and [ ‘“Clmyristic acid. After a 2 h labeling period, the appropriate agents were added at staggered time intervals such that, after the indicated time periods, all samples were terminated at the same time by the addition of 0.2% SDS. Small aliquots of the SDS-solubilized cells were saved for protein determination and the remainder was extracted for l-2 h with ice-cold chloroform/methanol. Additional chloroform and 0.2 M NaCl were added to break phase, and the lower phases collected and evaporated under N,. Samples were resuspended in chloroform/methanol and spotted onto heatactivated silica gel 60 thin-layer chromatography plates (0.25 mm thickness with concentrating zone; Merck, Darmstadt, Germany). Plates were developed in a solvent system of benzene/ethyl acetate (70 : 30) for separation of 1,2-DAG and in chloroform/ methanol/ acetic acid/water (65 : 43 : 1 : 3) for separation of phospholipids. Lipids were visualized with autoradiography using En”Hance (New England Nuclear, Boston, MA,
USA). Spots corresponding to 1,2-DAG or various phospholipids were identified by comigration with authentic standards, and radioactivity was quantified using a double-label protocol. Materials Collagenase was purchased from BoehringerMannheim Biochemicals (Indianapolis, IN, USA). [9,10(n)-3H]Oleic acid, [5,6,8,9,1 l,12,14,1S-3H]arachidonic acid and adenosine 5’-[ -$‘P]triphosphate were obtained from Amersham (Arlington Heights, IL, USA). [l-“ClMyristic acid was from New England Nuclear (Boston, MA, USA). Diacylglycerol kinase was purchased from Lipidex (Middletown, WI, USA); Percoll from Pharmacia (Piscataway, NJ, USA). Fetal bovine serum, horse serum, penicillin, and streptomycin were obtained from Gibco (Grand Island, NY, USA); all other tissue culture reagents were from Sigma (St. Louis, MO, USA). Bay K 8644 was a generous gift of Dr. Alexander Scriabine, Miles Institute for Preclinical Pharmacology (West Haven, CT, USA). Aldosterone was assayed using a solid-phase radioimmunoassay (Diagnostic Products, Los Angeles, CA, USA). Phosphatidylethanol standard was prepared according to the method of Eibl and Kovatchev (1981). Angiotensin II and carbachol were from Sigma (St. Louis, MO, USA). Protein was determined by the method of Lowry et al. (19511 using BSA as standard. Results Carbachol exerts no priming effect on cultured holine AG cells In freshly-isolated and cultured bovine AG cells pretreated with Ang II, the calcium channel agonist Bay K 8644 slows the rate at which aldosterone secretion decays following the removal of the hormone (Barrett et al., 1986a; Bollag et al., 1991). Thus, the time required for the secretory rate to attain a half-maximal value after Ang II removal is approximately 2.5-fold greater in the presence of Bay K 8644 than in its absence (Barrett et al., 1986a; Bollag et al., 1991). Because Ang II and carbachol utilize similar signal transduction pathways in adrenal cells (Hadjian et al., 1984; Kojima et al., 1986; Apfeldorf and Rasmusscn, 19881, the effect of Bay K 8644 in carba-
3oo1
Carbac;,l
$
‘gBSaqYqK +
---+
200-
100 -
04
,
,
20
40
60
1
,
80
(
100
120
Time (minutes) Fig.
1.
Bay
secretory
K X634
is unable
lation for SO min with 100 PM repeatedly
to prolong
response in carbachol-pretreated carhachol.
the
aldosteronc
cells. After cultured
ctimu-
cells wet-c
washed with KRB in the absence (open squares) or
presence (closed circles) of IO0 nM Bay K 8h-14. Aldoatzronc secretory
rates
were
measured
and
arc
expressed
as the
means (k SE.) of a minimum of five determinations
from three
separate
relative
experiments:
* p < i).Ol.
* * p 5 0.0002
to
control.
chol-treated cultured AG cells was examined. As shown in Fig. 1, the addition of carbachol (100 PM) to cultured bovine AG cells resulted in a peak increase (at 20 min of stimulation) in aldosterone secretion to a rate of 244 + 40 pg/min/ mg protein, a value approximately 14-fold over the control secretory rate of 18 k 2 pg/min/mg protein. Subsequently, the aldosterone secretory rate declined to a value of 170 k 32 pg/min/mg protein over the next 20 min. When the agonist was then removed, aldosterone secretion rapidly declined to control values within approximately 20 min both in the absence and in the prcscncc of 100 nM Bay K 8644. Similar results were obtained using tither a greater concentration of carbachol (1 mM), of Bay K 8644 (1 PM), or of both carbachol and Bay K 8644 (data not shown). Thus, in contrast to Ang II, carbachol appears unable to prime cultured bovine AG cells. Carbachol induces a sustained eler.ation in DAG content It should be noted that aldosterone secretion from cells responding to carbachol. unlike from
those responding to Ang II, was not completely sustained, an observation which has been made previously (Kojima et al., 1986). Whereas Ang II induces a monotonic rise in the secretory rate, the increase induced by carbachol was biphasic, consisting of a peak at approximately 20 min of stimulation which subsequently fell to a somewhat reduced value (Fig. 1). The biphasic nature of the carbachol-elicited aldosterone secretory response has previously been attributed to a smaller enhancement of calcium influx induced by carbachol (Kojima et al., 1986). However, alternatively elevated DAG levels might not be sustained. We therefore investigated the effect of carbachol on DAG content in the cultured cells. The effect of carbachol on cellular DAG content is shown in Fig. 2. Carbachol elicited an approximate 75% increase in DAG content which was sustained in the continued presence of the hormone. Thus, elevated DAG levels are present throughout the 60 min stimulation period despite declining aldosterone secretory rates. Neverthe-
20
1
ff
ff
PA
less, the 75% increase in DAG content carbachol is less than the approximate rise induced by Ang II.
0
20
40
Time
60
(minutes)
Fig. 2. Carbachol increases DAG content. Cultured cells were incubated for the indicated times in the presence of 100 PM carbachol and the stimulation was terminated at the indicated times by the addition of SDS. Neutral lipids were extracted and DAG was measured by the DAG kinase method. Values are expressed as -fold over control and represent the mean (*SE) of at least four determinations from two separate experiments. Control DAG content was 0.49kO.04 nmol/mg protein for the 15 and 30 min time points and 0.44+0.01 nmol/mg protein for the 60 min time point; * p < 0.02, * * p 5 0.0002 relative to control.
PEt
Fig. 3. Carbachol increases the levels of radiolabeled phosphatidic acid (PA) and phosphatidylethanol (PEt). Cultured bovine AG cells were prelabeled for 20-24 h with IO PCi [‘Hloleic acid and stimulated for 30 min with 100 PM carbachol or 10 nM Ang II in the presence of 0.5% ethanol. Phospholipids were extracted and chromatographed as described in Materials and methods. Open bars represent control conditions; striped bars represent stimulation with carbachol; and hatched bars represent stimulation with Ang II. Values are the means ( f SE) of five determinations from two separate experiments; * p 5 0.02. * * p I 0.0005 versus control values: o p I 0.0001 relative to Ang II-stimulated values, Control PA and PEt levels were 8lO+ 129 cpm/dish and 624 k 73 cpm/dish. respectively.
elicited by 200-250%
Carbachol activates phospholipase D Because Ang II activates PLD (in addition to PLC) and this activation can presumably contribute to hormone-elicited elevations in DAG content (Bollag et al., 19901, the ability of carbachol to stimulate PLD activity was subsequently examined. PLD has the unique characteristic not only of catalyzing the hydrolysis of phospholipids but also, in the presence of small amounts of ethanol, of effecting the ethanolysis of these compounds. As shown in Fig. 3, treatment of [ 3H]oleate-prelabeled cells with carbachol induced an elevation in the levels of both PA and PEt, suggesting that carbachol does indeed activate PLD. However, the carbachol-elicited increase in radiolabeled PEt levels (to 1.48- * 0.09-
fold over control) represented only approximately 60% of the change induced by Ang II (l.SO-+ 0. l-fold over control); whereas carbachol elevated PA levels to only about 24% (to I. 16- + 0.04-fold over control) of those elicited by Ang II stimulation (1.67-i 0.04-fold over control). Thus, although carbachol appears to stimulate PLD activity, this hormone is less potent than Ang II in inducing increases in both DAG content and PA levels. DAG content declines rapidly upon carbachol remoi,al The rate of decline in DAG content after carbachol removal was then investigated. Again carbachol induced an approximate 80% rise in DAG content at 60 min of stimulation (Fig. 4A). The subsequent removal of the hormone resulted in a rapid decline in DAG content to attain a value of 1.36-f 0.04-fold over control by 30 min after carbachol removal. The effect of the removal of Ang II (in the presence or absence of Ang II antagonists) on DAG content is shown in Fig. 4B for comparison. Ang II elicited a larger elevation in DAG content than carbachol, raising DAG content to 2.43-k 0.12-fold over control after 60 min of stimulation. Upon removal of the hormone, DAG content decreased slowly, declining only approximately 24% (to a value of 2.09- f 0.13-fold over control) within the 30 min washout period; similar results have been reported previously (Bollag et al., 1991). Thus, these results support a prior suggestion that the priming effect may be mediated by a prolonged elevation in DAG content following hormonal removal. Carbachol elelsates rnyristate-containing DAG ler?els The maintenance of the Ang II-induced elevation in DAG content may be related to the production of a species of DAG which persists upon the inhibition of Ang II action (Bollag et al., 1991). Thus, in cells labeled with both [‘Hlarachidonic acid and [ “Clmyristic acid, myristatecontaining DAG levels remain elevated while the levels of DAG containing arachidonate decline rapidly upon the addition of an Ang II antagonist (Bollag et al., 1991). The effect of carbachol upon the production of these two species of DAG was
A
0
B
Time 3
15
after
Carbachol (minutes)
30
Removal
.. E
I
T
kme
after &List (minutes)
Rem02
Fig. 4. DAG content declines rapidly following the removal of carbachol. After stimulation for 60 min with (A) 100 FM carbachol (striped bars) or (R) IO nM Ang II (hatched bars). cultured cells were repeatedly washed and incubated for the indicated times with KRB or in (El with KRB+ IO PM Ang II antagonist. Values are expressed as -fold over control and represent the mean of at least four determinations from two separate experiments; (A I * p < 0.005, * * p 5 0.0001 relative to control values; (El * p < 0.005, * * p 5 0.0001 versus control; l p < 0.02, c p s 0.0002 relative to Ang II-stimulated values. Control DAG content was 0.74kO.05 nmol/mg protein.
examined. As shown in Table 1, carbachol creased the levels of both arachidonatemyristate-containing DAG by approximately
inand 70
99
TABLE
1
ANGIOTENSIN II AND CARBACHOL ELEVATE [‘H]ARACHIDONATEAND [“C]MYRISTATE-CONTAINING DAG LEVELS ”
Control Ang II (10 nM) Carbachol c100 mM)
[‘HlArachidonate DAG
[ ‘“ClMyristate
1.OOi 0.04 h 3.26 & 0.04 *
1.oo + 0.04 2.7hk0.2 *
l.h8*0.07
I .79 +o.on *,O
*.O
DAG
” Cultured cells were labeled for 2 h with [‘Hlarachidonate and [‘“Clmyristate. In the continued presence of radiolabel, cells were incubated for 30 min with 10 nM Ang II or 100 mM carbachol. Samples were terminated by the addition of SDS and processed as in Materials and methods. ” Data are expressed as -fold over control with a control value of 12.510+4266 ‘H dpm/mg of protein (l.OO-?0.04-fold over control) and 75.099k 18,538 “C dpm/mg protein t I .OO-i 0.04.fold over control). Values represent the means ( +SE) of four determinations from two separate experiments: * p < 0.0005 relative to control values; Op < 0.0005 relative to Any II-stimulated values.
and SO%, respectively. Nevertheless, carbachol elevated radiolabeled DAG levels of both species to a lesser extent than did Ang II, which increased arachidonateand myristate-containing DAG levels approximately 3.3-fold and 2.X-fold, respectively. In addition, it should be noted that, as with Ang II and an Ang II antagonist (Bollag et al., 19911, when atropine was added to carbachol-treated cells, arachidonate-containing DAG levels declined while myristate-containing DAG levels remained elevated (data not shown). Discussion The results of this study have confirmed a previous report (Kojima et al., 1986) that in AG cells carbachol induces a biphasic increase in the secretory rate, with an initial peak of aldosterone secretion followed by a subsequent decline to a reduced plateau level (Fig. 1). Kojima et al. (1986) attributed the biphasic nature of the carbacholelicited aldosterone secretory response to the smaller enhancement of calcium influx induced by carbachol. We have demonstrated here that the fall in aldosterone secretion is not the result of a decrease in DAG levels, as the carbacholelicited elevation in DAG content was sustained
in the continued presence of the hormone (Fig. 2). Therefore, as suggested by Kojima et al. (1986), the biphasic aldosterone secretory response induced by carbachol is likely the result of an insufficient carbachol-elicited elevation of calcium influx. Although an elevated DAG content persisted in the continued presence of carbachol, upon removal of the hormone DAG content declined quite rapidly to a near-control value within 30 min (Fig. 4A). This result is in contrast to our observation that following pretreatment with and removal of Ang II, DAG content remains elevated for a prolonged time period (Bollag et al., 1991; and Fig. 4B). As discussed previously, the persistent elevation in DAG after Ang II removal is likely the result of differences in the species of DAG, i.e. related to the fatty-acid composition, generated in response to the hormone (Bollag et al., 1991). Thus, we demonstrated that the species which remains elevated following Ang II removal contains myristate, while the levels of arachidonate-containing DAG decline rapidly upon removal of the hormone (Bollag et al., 1991). We found that carbachol also elevated levels of both radiolabeled myristateand arachidonate-containing DAG, although not to as great an extent as Ang II (Table 1). Because the labeling protocol used does not result in isotopic equilibrium, no conclusion concerning the absolute levels of either species can be reached. However, the carbachol-induced increase in DAG content declined quite rapidly upon agonist removal, with a time course similar to that obtained for [“Hlarachidonate-DAG, suggesting that the majority of DAG produced upon carbachol stimulation is of the rapidly turning-over arachidonate-containing type rather than the persistent myristate-DAG. Nevertheless, it should be noted that due to the relative affinity of the two ligands for their receptors, the unbinding of carbachol from its receptor should be more rapid than that of Ang II. Thus, the disappearance of the second messengers generated by Ang II might be expected to occur at a lower rate than those produced in response to carbachol. In order to verify our hypothesis, therefore, further experiments are necessary to determine the absolute amounts of the slowlyand rapidly-turning over DAG pool.
IO0
Although the mechanism for the persistence of myristate-containing DAG is unknown, we have proposed an involvement of PLD activity in the maintained increase in myristate-containing DAG via a positive feedback effect (Exton, 1990). Such a feed-forward mechanism could be initiated as a result of the activation of PLD by protein kinasc C (PKC), presumably stimulated via phosphoinositide turnover and DAG production (Liscovitch et al., 1987; Cabot et al., 1988, 1989; Billah et al., 1989; Liscovitch, 1989; Liscovitch and Amsterdam, 1989), as described previously (Bollag et al., 1991). Because in some systems PLD activity has been reported to hydrolyze preferentially (myristate-containing) phosphatidylcholine (Martin, 19881, the synthesis of myristate-DAG could persist, through the functioning of PLD, despite an immediate termination of phosphoinositidespecific PLC activation and the production of arachidonate-containing DAG. Martinson et al. ( 1990) also found a persistent, delayed elevation in DAG content in 1321Nl astrocytoma cells following stimulation with carbachol and inhibition with atropine. These investigators proposed that the lag associated with DAG formation was the result of slow conversion of PLD-generated PA to DAG by phosphatidate hydrolases (Martinson et al., 1990). We report in the present study that carbachol activated PLD with a considerably reduced potency relative to Ang II (Fig. 31. In particular, the elevation in radiolabeled PA levels elicited by carbachol is only approximately 24%, while the increase in PEt levels is 60%, of that induced by Ang II. Therefore, carbachol may be incapable of stimulating PLD sufficiently to engage the feedforward response such that PLD-induced DAG production cannot be maintained without the continued generation of DAG from the phosphoinositides. It should be noted as well that PEt is a novel phospholipid which is not metabolized by the cellular machinery (Billah et al., 1989; Liscovitch, 1989). Thus, the differential carbachol-induced increase in PA and PEt levels, relative to Ang II, may be the result of transient, rather than sustained, activation of PLD, with subsequent metabolism of the generated PA. However, further studies arc needed to explore this possibility, especially in view of the recent finding that tran-
sient PLD activation can result in a maintained DAG production (Martinson et al., 1990). Ang II may also activate a signalling pathway in addition to those activated by carbachol, e.g. a PC-specific PLC as in astrocytoma cells (Martinson et al., 1989); this possibility too requires further investigation. Alternatively, the maintained elevation in myristate-containing DAG levels following Ang II removal may be the result of the slower metabolism of this species of DAG versus arachidonate-containing DAG, as discussed previously (Bollag et al., 1991) Thus in this scenario as well, if carbachol induces the production of predominantly the rapidly metabolized arachidonateDAG, DAG levels would decline quickly upon removal of the hormone. In addition, it is possible that the rate of DAG metabolism is maximal at that level of production induced by carbachol. Therefore, because Ang II elicits a greater DAG production, DAG levels might remain elevated for longer periods of time following Ang II rcmoval. The rapid decline of DAG content following the inhibition of carbachol action (Fig. 4) may explain the inability of this hormone to prime cells to respond to the calcium channel agonist Bay K 8644 with aldosterone secretion (Fig. I). According to our hypothesis, priming is the result of a persistent elevation in DAG content (Bollag et al., 1991); this DAG provides one of the critical signals required for aldosterone secretion, the other of which is an enhancement of calcium influx which can be elicited by Bay K 8644 (Barrett et al., 1986a). Ang II, because it generates such persistent DAG, is thus able to prime cells to respond to the second signal provided by Bay K 8644. Carbachol, however, elicits no such maintained elevation in DAG levels and thus is unable to prime cells to secrete aldosterone in response to the calcium channel agonist. It is possible, in fact, that a threshold level of persistent DAG is required for the stimulation of aldosterone production. In this case, the inability of carbachol to prime glomerulosa cells is the result of the rapidity with which DAG content declines below this threshold following carbachol removal. However, further research is necessary to determine if such a threshold DAG level exists.
101
Thus, the results of the present study together with our previous data suggest that the priming effect of Ang II may be related to its ability to elicit a persistent elevation in DAG content following removal of the hormone. Carbachol fails either to produce this persistent DAG elevation or to exert a priming effect. In turn, this priming effect could provide a mechanism by which, with repeated Ang II stimulation, a second sustained aldosterone secretory rate could be initiated more effectively by a different agonist, or perhaps more rapidly or by a lower concentration of Ang II than might othcrwisc be necessary. Thus, the priming ability of Ang II may be physiologically relevant and contribute to its effectiveness as a primary regulator of aldosterone secretion. Acknowledgements We would like to express our appreciation for the expert technical assistance of Laura Kiernan and John Nee. References Agwu. D.E.. McPhail. L.C.. Chabot, M.C., Daniel. L.W., Wykle, R.L. and McCall. C.E. (IYXY) J. Biol. Chem. 764, 1405~1413. Apl’eldorf, W.J. and Rasmussen. H. (IYXX) Cell Calcium 9, 71-80. Barrett, P.Q.. Kojima. I.. Kojima, K.. Zawalich, K., Isales, C.M. and Rasmussen, H. (19X6a) Biochem. J. 23X,905-912. Barrett, P.Q.. Kojima. I.. Kojima, K.. Zawalich. K., Isales, C.M. and Rasmussen, H. (198hb) Biochem. J. 238. XY3~903. Barrett, P.Q.. Bollag, W.B. and Rasmussen. H. (19881 Hormones and their Actions (Cooke. B.A.. King, R.J.B. and van der Molen, H.J.. Part 2, pp. 21 l-229, eds.). Elsevier. Amsterdam. Barrett. P.Q., Bollag. W.B.. Isales. C.M., McCarthy, R.T. and Rasmussen. H. (19x01 Endocr. Rev. IO, 496-518. Billah. M.M.. Pai. J.-K., Mullman. T.J., Egan, R.W. and Siegel, M.I. (19X91 J. Biol. Chem. 264, OtlnY-9076.
Bocckino. S.B., Blackmore, P.F., Wilson, P.B. and Exton. J.II. (19871 J. Biol. Chem. 262, 153OY-15315. Bollag. W.B., Barrett, P.Q., Isales, C.M. and Rasmussen, H. (19901 Endocrinology 127, 14361443. Bollag. W.B.. Barrett, P.Q.. Isales, C.M. and Rasmussen. II. (IYYI) Endocrinology 12X. 231-241. Cabot. M.C.. Welsh. C.J.. Cao, H.-t. and Chabbott, H. (IYXXl FEBS Lett. 233. 153-157. Cabot. M.C.. Welsh. C.J., Zhang. Z.-c. and Cao, H.-t. (19X91 FEBS Lett. 245. X5-90. Eibl. H. and Kovatchev, S. (IYXI) Methods Enzymol. 72, h32~h3Y. Exton, J.H. (19901 J. Biol. Chem. 265, l-4. Foster. R.. Lobo. M.V., Rasmussen, H. and Marusic, E. t 19X1) Endocrinology 109. 2 lYh~220 I, Hadjian, A.J., Culty. M. and Chambaz. E.M. (1984) Biochim. Biophys. Acta X04, 427-433. Kojima. 1.. Kojima. K.. Kreutter, D. and Rasmussen. H. (19841 J. Biol. Chem. 259, 1444X-14457. Kojima. 1.. Kojima. K. and Rasmussen. H. (1985) J. Biol. Chem. 200, 9177-9184. Kojima. I., Kojima, K.. Shibata, II. and Ogata, E. (19861 Endocrinology 1 IY. 2X4-291. Lang, U. and Valloton, M.B. (19x71 J. Biol. Chem. 262, x047-X050. Liscovitch, M. (19XYl J. Biol. Chem. 264. 14SO-1456. Liacovitch, M. and Amsterdam. A. (19891 J. Biol. Chem. 264. 117h2-11767. Liacovitch, M.. Blusztajn. J.. Freeae. A. and Wurtman. R.J. tlYX71 Biochem. J. 241. Xl-Xh. Loffelholz. K. (19XYl Biochem. Pharmacol. 38, 1543-1540. Lowry. O.H.. Rosebrough. N.J.. Farr, A.L. and Randall, R.J. (19511 J. Biol. Chem. 193, 265-275. Martin, T.W. tlYX81 Biochim. Biophys. Acta Y62, 2X2-296. Martinson, E.A.. Goldstein, D. and Brown, J.H. (19891 J. Biol. Chem. 264, 14748X 14754. Martinson. E.A.. Trilivas. I. and Brown, J.H. (1990) J. Biol. Chem. 76.5, 222X2-22287. McCarthy. R., Isales. C., Bollag. W., Rasmussen, H. and Barett. P. (IYYO) Am. J. Physiol. 25X, F473F478. Nishizuka. Y. tlYXX1 Nature 334, 66ll665. Pai. J.-K.. Siegel. M.I., Egan, R.W. and Billah, M.M. (IYXXal Biochem. Biophys. Res. Commun. 150, 355-364. Pai, J.-K.. Siegel, M.I., Egan, R.W. and Billah. M.M. (19XXb) J. Biol. Chem. 263, 12472-12477. Rubin. R. (198X) Biochem. Biophys. Res. Commun. 156, 1090~ 1096.
